Vol. 49, No. 2, 1996 World Health STATISTICS Quarterly Rapport trimestriel de STATISTIQUES sanitaires mondiales Health security for women Securite sanitaire pour Jes femmes World Health Organization Organisation mondiale de la Sante Geneve The World health statistics quarterlyreplaces (since 1978) the monthly World health statistics report(published since 1967) and its forerunner the Epidemiological and vital statistics report (published since 1947). It deals with the detailed analysis of selected health topics of current interest. Starting with Vol. 4 1 ( 1988). the Quarterly contains articles in either French or English with a summary in both languages. Annual subscription Sw. fr. 110.- Price per copy Sw. fr. 31 - Material from the Quarterly may be reproduced providing due acknowledgement is made. Symbols used in tables Preliminary. approximate or estimated data Data not available Nil or magnitude negligible Category not applicable T Total M Male F Female N Absolute numbers The World Health Organization welcomes requests for permission to reproduce or translate its publications. in part or 1n full Applications and enquiries should be addressed to the Office of Publications. World Health Organization. Geneva. Switzerland, which will be glad to provide the latest 1nformat1on on any changes made to the text. plans for new ed1t1ons. and reprints and translations already available © World Health Organization 1996 Publications of the World Health Organization enjoy copyright protection 1n accor- dance with the provisions of Protocol 2 of the Universal Copyright Convention All rights reserved. The designations employed and the presentation of the material in this publ1cat1on do not imply the expression of any opinion whatsoever on the part at the Secretariat of the World Health Organ12at1on concerning the legal status of any country territory, city or area or of its autonties. or concerning the del1m1tat1on of its frontiers or boundaries The mention of specific companies or of certain manufacturers' products does not imply that they are endorsed or recommended by the World Health Organization 1n preference to others of a similar nature that are not mentioned Errors and om1ss1ons excepted. the names of proprietary products are distinguished by 1nit1al capital letters Le Rapport trimestriel de statlstiques sanital(es mond,ales remplace (depuis 1978) le Rapport de stat1s/Jques sanita1res mond1ales (publie depuis 1967) et son precurseur le Rapport epidem1ologiqueet demogra- ph1que (publie depuis 1947) II presente des analyses detail lees sur des sujets specif1ques d'1nteret courant. A compter du Vol. 41 (1988). le Trimestriel presente des articles orig1naux en fran9ais ou en anglais, accompagnes d'un resume dans les deux langues. Prix de l'abonnement annuel Fr. s. 110.- Le numero Fr. s. 31 - La reproduction d'extraits duTrimestriel est autorisee. sous reserve d'indication de la source. Explication des signes Donnee preliminaire. approximative ou estimative Donnee non d1sponible Zero ou quant1te negligeable Categorie non applicable T Total M Masculin F Fem1nin N Nombres absolus L'Organisat1on rnond1ale de la Sante est tou1ours heureuse de recevo1r des deman- des d'autor1sat1on de reprodu1rc au de tradu1re ses pubhcat1ons, en partie au 1ntegralement Les demandes a cct effet et les demandes de rense1gnements do1vent etre adressees au Bureau des Publications. Organisation mondtale de la Sant€. Geneve. Suisse. Qu1 se fera un pla1s1r de fourn1r les renseignements les plus recents sur les changemcnts apportes au textc. les nouvelles editions prevues et les reimpress1ons et traduct1ons de1a d1sponibles © Organisation mondiale de la Sante 1996 Les publ1cat1ons de l'Organisat1on mond1ale de la Sante benefic1ent de la protection prevue par les d1spos1t1ons du Protocole No. 2 de la Convention un1verselle pour la Protection du Dro1t d'Auteur Taus dro1ts reserves. Les appellations employees dans cette publication et la presentation des donnees qui y f1gurent n'1mpllquent de la part du Secretariat de l'Organ1sat1on mond1alc de la Sante aucune prise de pos1t1on quant au statut 1und1que des pays. territo1res. v1lles ou zones. ou de leurs autontes, ni quant au trace de leurs front,eres ou hmites La mention de f1rmes et de produ1ts commerc1aux n'imphque pas que ces firmes et produ1ts commerc1aux sont agrees ou recommandes par l'Organisat1on mondiale de la Sante de preference a d'autres Sauf erreur au om1ss1on. unc ma1uscule 1nit1ale 1nd1que qu'II s'ag1t d'un nom depose. IX ISSN 0043-8510 Printed in Switzerland 96/11134 -Atar SA, Geneva-5100 World Health Statistics Quarterly Rapport trimestriel de statistiques sanitaires mondiales Vol. 49, N° 2, 1996 Health security for women Securite sanitaire pour Jes femmes Contents Sommaire Introduction: women and health security Introduction: women and health security Aleya El Bindari Hammad 74 [anglais seulement] Aleya El Bindari Hammad 74 Maternal mortality. Carla AbouZahr et al. 77 Mortalite maternelle [resume] Carla AbouZahr et al. 87 Contraceptive method mix menu: providing Assortiment de methodes contraceptives : healthy choices for women. offrir des choix sains aux femmes (•esume). K. Fern Greenwell 88 K. Fern Greenwell 92 Women and AIDS in South and South-East Les femmes et le SIDA en Asie du Sud Asia: the challenge and the response. et du Sud-Est [resume] Nafsiah Mboi 104 Nafsiah Mboi 94 Sexual transmission of HIV to women in Transmission du VIH par voi6 sexuelle industrialized countries. Catherine Hankins 106 aux femmes des pays industrialises [resume]. Catherine Hankins 112 Women and tuberculosis. Les femmes et la tuberculose [resume]. Maire Connolly & Paul Nunn 115 Maire Connolloy & Paul Nunn 118 Double jeopardy: women and leprosy in India. Double peril : les femmes et la lepre en lnde Carol Vlassoff et al. 120 (resume]. Carol Vlassoff et al. 125 Women and smoking: a global issue. Les femmes et le tabac : un probleme Amanda Amos 127 mondial [resume]. Amanda Amos 133 Older women in an aging world: achieving Les femmes plus a.gees dans un monde health across the life course. vieillissant : ccnserver la sante durant toute la Ruth Bonita & Anna L. Howe 134 vie [resume]. R{Jth Bonita & Anna L. Howe 140 Combating female genital mutilation: an Lutte contre les mutilations genitales agenda for the next decade. Efua Dorkenoo 142 feminines : ordre du jour pour la prochaine decennie [resume]. Efua Dorkenoo 146 Ethics and reproductive health: L'ethique et la sante reproductive sous l'angle a principled approach. Ruth Macklin 148 des principes [resume]. Ruth Macklin 153 Sexual abuse in the health professions - who's Abus sexuels dans les professions de sante - counting? Marilou McPhedran 154 qui compte? [resume). MtHilou McPhedran 157 Women and migration: a public health issue. Les femmes et les migrations : question de Manuel Carballo et al. 158 sante publique [resume]. Manuel Carballo et al. 162 Wld hlth statist. quart., 49 (1996) d:\ataroao\oms\evs3\49-2som.om6 563700 2196 Table des matieres Introduction: women and health security As this special issue of the World health statistics quarterly demonstrates, a great deal of information on all aspects of women's health is now available. This knowledge and information has been indis- pensable to the continual assessment of the situa- tion of women throughout the world and in all periods of their lives. Women's overall health and well-being has im- proved. Life expectancy indicators illustrate a gen- eral trend for women to live longer. This is espe- cially the case where women have enjoyed the fruits of development, have had access to services, have benefited from a physical and cultural envi- ronment containing few health risks and have ac- quired the information necessary to enable them to make informed choices. There is also no doubt that the greater involve- ment of women in the workforce and in the socio- economic and health arenas has contributed sub- stantially to a different vision of women's health, not only as care-providers, but also as decision- makers. Problems such as violence against women have existed throughout time, but it is women who have had the vision and commitment to bring to light the terrible health consequences of this vio- lence, and to ensure that the silence around gen- der-based violence is broken. In addition to ensuring that the knowledge base on women's health problems continues to grow, the unflagging commitment and work of individuals, institutions and organizations working in the area of women's health have provided partic- ularly useful lessons. In looking for and applying solutions, we are now able to draw upon the wealth of experience already gained, both positive and negative, in order to build on what has been achieved, guard against repeating past mistakes, and prevent wasting precious resources. Among the most significant lessons is the realiza- tion that health must be considered in a holistic way. This applies to the health of all human beings - men and women - at all ages. As far as women's health is concerned, there has been a tendency in the past to deal with health care needs through separate pro- grammes related to particular health issues. An ex- ample of this has been the overemphasis on the reproductive aspects of women's health, often with subsequent neglect of other areas. At both interna- tional and national levels there is now a movement which adopts a more humane approach and provides integrated services to women in a holistic fashion. 74 Another lesson learned is that, in looking at health issues affecting women, it is essential to take a lifespan perspective, since health conditions in one phase of a woman's life not only affect subse- quent phases of her own life, but also have an impact on future generations. This inter-genera- tional link is a characteristic unique to women. In spite of the overall improvement in women's health and the lessons learned the situation is still highly unsatisfactory. Today there is a certain rest- lessness among those who have been actively work- ing at all levels for so many years to improve women's health. The right words have been said in all the right places. General consensus has been reached on the steps which need to be taken to make a measurable difference in so many areas of women's health across age groups and borders. And yet action is still lagging and the measurable differences remain elusive. The following statistics are testimony to the fact that so much more needs to be done: In the area of violence against women - • Studies worldwide indicate that between 20 and 50% of women have been beaten by a male partner, with wide-ranging physical and mental health consequences. • Violence against women in conflict and refugee situations is also a serious health issue, with women and children representing the majority of civilian victims worldwide. Maternal morbidity and mortality continue to be unacceptably high - • 99% of the 585 OOO deaths from pregnancy- related causes in 1990 took place in developing countries. In other words, 1 woman in 50 in the developing world dies as a result of pregnancy- related complications compared to 1 woman in 8 OOO in the developed world. • The number of maternal deaths represents only a small fraction of the total burden of disease associated with pregnancy and child- birth. Conservatively estimated, the submerged fraction of the iceberg amounts to some 20 mil- lion cases of morbidity and disability each year, ranging from acute and devastating injuries such as obstetric fistula to chronic, debilitating conditions such as severe anaemia, reproduc- tive tract infections and uterine prolapse. Some one-third of all women in the developing world have suffered such problems at some time in their lives. Rapp. trimest. statist. sanit. mond., 49 (1996) With respect to HIV I AIDS - • While women were under-represented among those suffering from HIV infection in the 1980s, this trend is no longer discernible. In 1995, almost half of all newly-infected adults through- out the world were women. Women in develop- ing countries, notably in areas of the world where the pace of HIV infection is growing at an especially alarming rate, have been particu- larly affected. In 1995, 35 % of all new infections affecting women were found in South-East Asia. • Seven million women have already contracted HIV. By the year 2000 this number will have climbed to 13 million, and 4 million of these women will have already died. While statistics do not tell the whole story, when appropriately used, they are very revealing of the inequities that continue to take their toll on the health of women. How can we provide a guarantee that this scenario becomes a relic of the past in a brighter health future for today's infant girl? One answer is to spare no effort in supporting and moving forward the cause of women's health. Anocher complementary strategy is to ensure that women's health issues remain firmly on the politi- cal and developmental agenda at all levels. The World Health Organization is firmly com- mitted to both of these avenues. WHO has main- streamed women's health in all relevant pro- grammes and strengthened these components in order to address all aspects of women's health and well-being throughout their entire lives. In addi- tion, the Global Commission on Women's Health was created as an essential expression of the com- mitment of WHO to forging ahead with positive and effective measures at all levels for improving women's health, and carrying out international and national advocacy on behalf of women's health concerns. It has been said by many people that neither men nor women can enjoy their human rights until they enjoy the most fundamental hu- man right of all - the right to health. The right to health is best expressed in the notion of "health security" throughout the lifespan. Health security traces the entire lifespan of a woman from the time in utero through to old age. It encompasses all aspects of the basic human right to health: the right to freedom of choice and personal security; the right to food in sufficient quantity and of good quality; the right to live and work in environments where known health risks are controlled; and the right to education, information and decent hous- ing. Health security also encompasses the principle of universality in health care, so that all human beings - men or women - may live with the knowl- edge that they can seek and receive quality health care which is also accessible and affordable. Health security, therefore, seeks the empowerment of people through various forms of societal and eco- Wld hlth statist. quart., 49 (1996) nomic support and fuller knowledge and aware- ness, thus enabling people to make the right choices, cope with the changing patterns of vulner- ability and keep healthy. Women will only enjoy the right to health once progress is made in finally overcoming persistent barriers to equity, choice and participation. The World Health Organization and the Global Commission on Women's Health have adopted women's right to health and the enjoyment by women of health security throughout their lifespan as the platform for advocacy efforts and actions. In this way, the attention which is so necessary to improving women's health across age groups and borders will continue to occupy the spotlight in all arenas and at all levels. Every woman has the right: to know that she is free from the threat of gender-based violence both inside and outside her home; to reap the benefits of education in both formal and informal environ- ments which are themselves free from health risks; to know that everything that can be done will be done to ensure that she does not die, or become ill as a result of pregnancy and childbirth; and to the social, political and economic empowerment that will allow her to protect herself from HIV/ AIDS and other sexually transmitted diseases. Finally, every woman has the right to accessible and afford- able health care services and to gender-sensitive interventions delivered by providers who adhere to the highest ethical standards. We are all accountable for guaranteeing these health rights to all women throughout the world. We can fulfil our responsibilities by expanding our partnerships to strengthen our individual and collective actions and efforts. Unceasing ad- vocacy, concrete actions with measurable out- comes and solidarity across all barriers will make the difference in providing for a better health future for today's infant girl, tomorrow's girl and woman. This special issue of the World health statistics quarterly contains a series of articles which present the wide range of issues that are critical to achiev- ing health security for women. Women's reproduc- tive health is comprehensively addressed. The arti- cle on maternal mortality presents new data which reveal that the scale of the problem is much greater than earlier believed. Options for expanding the contraceptive choices available to underserved women who wish to regulate their fertility, and the adverse repercussions on health status resulting from the limited range of contraceptives made available in some contexts are presented in the article on the contraceptive method mix. In "Combating female genital mutilation: an agenda for the next decade", the author reports that this phenomenon continues to be a problem in many African countries and proposes actions to acceler- ate progress in eliminating it. 75 Another group of articles addresses women's differential vulnerability to infection, the differ- ential impact of diseases on women, and differ- ential treatment rates by sex. Articles on HIV I AIDs address women's increased vulnerability to HIV. Critical areas for action by policy makers and the health care profession are identified. The article "Women and tuberculosis" under- scores the fact that women's lower propensity to seek diagnosis and treatment has devastating con- sequences for their health. It has been found that gender plays a large role in the impact of leprosy. Gender-sensitive strategies for improving disease control are laid forth. "Women and smoking" makes it clear that, although smoking is seen largely as a male problem, it leads to a num- ber of killer diseases in women. Recommenda- tions for research, public policy and education are made. Several other articles examine the special health needs of specific groups of women. The article on aging women reinforces the importance of address- ing women's health throughout the entire lifespan in order to avoid certain health conditions in their 76 old age. One author points out that migration poses a major public health problem for women. Health security for women cannot be div- orced from broader ethical concerns. The article "Ethics and reproductive health: a principled ap- proach" points out that women must be treated as competent partners and participants when formu- lating medical decisions. The piece on sexual abuse among health practitioners furthers knowl- edge on the barely explored field of the sexual exploitation offemale patients by health care pro- fessionals in positions of trust. It is hoped that this special issue of the World health statistics quarterly will provide health profes- sionals and lay people in all regions of the world with new and critical data and information which will be of use in their continuing endeavours to improve women's health across age groups and political borders. A/,eya El Bindari Hammad Executive Administrator for Health Policy in Development, World Health Organization, Geneva Rapp. trimest. statist. sanit. mond., 49 (1996) Maternal mortality Carla Aboulahra, Tessa Wardlawb, Cynthia Stantonc & Kenneth HilJd Introduction Levels of maternal mortality in industrialized and developing countries show a greater disparity than any other public health indicator, far exceeding differences in infant mortality rates which are most often taken as the measure of comparative disad- vantage. Although there have been significant de- clines in infant mortality rates in recent years, the available evidence indicates that the same is not true for maternal mortality. As the end of the 20th century approaches, in the developing world 1 woman in 50 still dies as a result of pregnancy- related complications and the figure rises to 1 in 10 in many parts of Africa. By contrast, the figure for developed countries can be as low as 1 in 8 OOO. Maternal mortality as an indicator of development Maternal mortality is a particularly sensitive indica- tor of inequity; it offers a litmus test of the status of women, their access to health care and the adequa- cy of the health care system in responding to their needs. Information about the levels and trends of maternal mortality is needed, therefore, not only for what it tells us about the risks of pregnancy and childbirth, but also for what it implies about women's health in general and, by extension, their social and economic status. The importance of maternal mortality as an indicator of women's health and status, and by extension of human development, is reflected by its inclusion in the goals of international confer- ences, such as the Nairobi Safe Motherhood Con- ference in 1987, the World Summit for Children (WSC) in 1990, the International Conference on Population and Development (ICPD) in 1994, and the Fourth World Conference on Women (FWCW) in 1995. The goal common to all these conferences is reduction of maternal mortality to half the 1990 levels by the year 2000. Related goals include increasing access to antenatal care, ensur- a Technical Officer, Maternal Health and Safe Motherhood Programme, WHO, Geneva. h Project Officer - Statistics and Monitoring, Planning Office, UNICEF, New York. c Health Specialist, Demographic and Health Surveys Project, Macro International Inc. d Professor, Department of Population Dynamics, Johns Hopkins University & Director, Hopkins Population Center. Wld hlth statist. quart., 49 (1996) ing that all women deliver with a skilled birth atten- dant and improving the nutritional status of girls and women. Definition and causes of maternal mortality The Tenth revision of the international statistical dassi- fication of diseases (ICD-10) defines a maternal death as the death of a woman while pregnant or within 42 days of termination of pregnancy, irrespective of the duration and site of the pregnancy, from any cause related to or aggravated by the pregnancy or its management but not from accidental or inci- dental causes (1). Thus, like infant mortality, ma- ternal mortality is, in part, a time-of-death indica- tor. Unlike infant mortality, however, measuring maternal mortality requires knowledge not only of the timing of death but also of the cause of death. Maternal deaths should be divided into two groups: Direct obstetric deaths are those resulting from obstetric complications of the pregnant state (pregnancy, labour and the puerperium), from interventions, omissions, incorrect treatment, or from a chain of events resulting from any of the above. Indirect obstetric deaths are those resulting from previous existing disease or disease that devel- oped during pregnancy and which was not due to direct obstetric causes, but was aggravated by physiological effects of pregnancy. The direct causes of maternal mortality are the same around the world; haemorrhage, sepsis, eclampsia, obstructed labour and abortion compli- cations account for around 80% of all maternal deaths. Although the distribution of causes differs somewhat from region to region, globally the big- gest proportion of deaths (25%) is due to haemor- rhage, most of which occurs during the postpar- tum period. Sepsis accounts for around 15% of the total, obstructed labour for 7%, eclampsia for 8%, and unsafe abortion for 13%. Other direct obstet- ric causes, including ectopic pregnancy, embo- lisms and anaesthesia accidents, account for a fur- ther 8% of direct maternal deaths. Indirect causes of maternal death such as anaemia, malaria, car- diovascular diseases, hepatitis and diabetes, ac- count for some 20% of all maternal deaths. (Fig. 1) Deaths from "accidental or incidental" causes have historically been excluded from maternal mortality. However, in practice, the distinction be- tween incidental and indirect causes of death is 77 Fig. 1 Causes of maternal deaths Causes de mart maternelle Hypertensive disease of pregnancy, including eclampsia Maladies hypertensives de la grossesse, y compris l'eclampsie 12% Obstructed labour Dystocie 7% Sepsis Haemorrhage Indirect causes Causes indirectes 20% Avortements pratiques dans des conditions dangereuses 13% Other direct causes Autres causes directes 8% difficult to make. Some deaths from external causes-for example, suicide or homicide - may be attributable to the pregnancy itself (2) . This phe- nomenon is common to both developed and devel- oping countries. A study in the United States found that several deaths of pregnant or recently preg- nant women were the result of physical violence related to the fact of pregnancy and that had they been included as maternal deaths the overall maternal mortality ratio would have increased by 8% (3). The medical causes of maternal death repre- sent only the most visible dimension of a multi- layered problem. In reality it is often logistic or health service factors that determine whether a woman with pregnancy-related complications lives or dies. Lack of access to skilled health care for complicated pregnancies or for emergencies is of- ten the root cause of death. Barriers to such access can be physical (distance, lack of transport) , eco- nomic (lack of resources to pay for the needed transport, care or drugs), or sociocultural (low sta- tus which impedes women's ability to take deci- sions to seek care and which may impose restric- tions on their mobility). Even when a woman reaches a health facility a variety of health service factors can prevent her from receiving appropriate care in a timely man- ner. These include, for example, shortages of drugs and equipment, lack of safe blood supplies for blood transfusion, and shortages of skilled per- sonnel. In some cases, a woman may not receive care simply because providers fail to realize the gravity of her condition and act appropriately (2). Actions needed to reduce maternal mortality The actions needed to attain the goal of reducing maternal mortality have been known and under- stood for many years. Because women's status is often a determinant of their access to education, health care and nutrition, it is a foundation upon 78 which improvements in maternal health care must rest. However, improving women's status will necessarily remain a long-term objective. In the short term, there are three essential interventions that will help to reduce maternal deaths: • reducing the number of high risk and un- wanted pregnancies; • reducing the numbers of obstetric complica- tions; and • reducing deaths among women who develop complications. Ensuring access to family planning so as to avoid births that are too early or too late in a woman's life, too close together or unwanted, is an important pathway for the prevention of mater- nal deaths. Unwanted and mistimed pregnancies are associated with a higher risk of maternal death. It has been estimated that ready access to family planning for all women could reduce the numbers of maternal deaths by some 20%. Once a woman is pregnant, however, she needs care throughout pregnancy and delivery, in- cluding essential care for the management of ob- stetric complications. The number and severity of obstetric complications can be reduced by ensur- ing that all women have access to basic maternal care - good-quality antenatal, delivery (clean and safe) and postpartum care. Basic maternal care is an opportunity to provide women and their fami- lies with information on how to take care of them- selves throughout pregnancy and childbirth, to in- form them about danger signs and symptoms, and to advise on what to do if complications arise. Such care should include the prevention and manage- ment of conditions which can lead to complica- tions - such as anaemia or sexually transmitted diseases - or diseases aggravated by the physiologi- cal effects of pregnancy- such as malaria, diabetes, tuberculosis or cardiovascular diseases. Of particu- lar importance during basic maternal care is the early detection and prompt management or refer- Rapp. trimest. statist. sanit. mond., 49 (1996) ral of obstetric complications before they become life-threatening emergencies. Any woman, even if she is healthy and well- nourished, can unexpectedly develop obstetric complications. Such complications often happen suddenly and dramatically, requiring medical in- tervention quickly if death or serious disability is to be avoided. Providing access to essential care for obstetric complications will help to reduce case fatality rates among women experiencing compli- cations and thus reduce maternal mortality. Ser- vices for the management of complications should be available as close as possible to where women live. Whereas certain interventions such as caesare- an delivery or blood transfusion require the skills and equipment normally available only at the first referral or district hospital, many other life-saving interventions, such as management of eclampsia, haemorrhage, abortion and sepsis, can be deliv- ered at health centres staffed by people with appro- priate midwifery skills. A description of the inter- ventions needed at different levels of the health care system - community, health centre or district hospital - has recently been issued.e New estimates of maternal mortality Although the interventions needed are well-estab- lished, ascertaining progress in reducing maternal mortality is extremely difficult for two reasons: ma- ternal mortality is difficult to measure; and the infor- mation available at country level does not generally permit the establishment of good baseline data. In order to address these problems WHO and UNICEF worked with Cynthia Stanton and Ken- neth Hill of Johns Hopkins University to develop a new approach to estimating levels of maternal mor- tality in developing countries. The new approach has the dual objective of generating improved esti- mates for countries with inadequate or no national data on maternal mortality, while at the same time providing improved estimates for countries with data by adjusting for problems of under-reporting and misclassification. A full description of the strat- egy will be published separately.f The new WHO/UNICEF approach to estimating maternal mortality Assessing levels of maternal mortality at the nation- al level requires knowledge about deaths of women of reproductive age (15-49 years), the cause of death and also whether or not the woman was c World Health Organization. The mother-baby package: implementing safe motherhood in countries. Maternal Health and Safe Motherhood Programme, Geneva, 1994 (\VHO/FHE/ MSM/94.1 l). f World Health Organization and United Nations Children's Fund. Revised I 990 estimates of maternal mortality: a new approach by WHO and UNICEF. Geneva & New York, 1996 (\\!HO /FHE/ MSM/94.11- UNICEF/PLN/96.l). Wld hlth statist. quart., 49 (1996) pregnant at the time of death or had recently been so. Yet few countries count births and deaths; even fewer register the cause of death; and fewer still systematically note pregnancy status on the death form. Broadly speaking, countries fall into one of three categories: 1. Countries with no reliable system of vital regis- tration where maternal deaths - like other vital events - go unrecorded; 2. Countries with relatively complete vital registra- tion in terms of numbers of births and deaths but where cause of death is not adequately clas- sified; cause of death is routinely reported for only 78 countries or areas, covering approxi- mately 35% of the world's population.(4) 3. Countries with complete vital registration and good cause of death attribution -although even here, misclassification of maternal deaths can arise for a variety of reasons. Maternal mortality can be measured by incorpo- rating questions on pregnancy and deaths into large-scale household surveys. The disadvantage of such approaches is that they require large sample sizes and are extremely expensive and time consuming(5).g A more cost-effective approach is the Sisterhood Method which adds on to existing household surveys a few simple questions about whether or not the sisters of the respondent are still alive. Smaller sample sizes are needed because each respondent can provide information on a number of sisters. However, the results do not provide a current estimates but give an idea of the levels of maternal mortality roughly ten years earlier. There is some evidence that the method may underesti- mate pregnancy-related mortality (6, 7). The best way of measuring maternal mortality in the absence of vital registration is to identify and investigate the causes of all deaths of women of reproductive age: the R£productive Age Mortal- ity Survey (RAMOS). Multiple sources ofinformation -civil registers, health facility records, community leaders, religious authorities, undertakers-are used to identify all deaths (8). Subsequently, interviews with household members and health care providers and facility record reviews are used to identify ma- ternal deaths. RAMOS studies are time consuming and complex to undertake, particularly on a large scale and only ten developing countries have car- ried out RAMOS or household studies to estimate maternal morality at the national level. Methodology for the new estimates The new estimates were developed using a dual strategy: existing national maternal mortality esti- K For example, a sample of nearly 10,000 pregnancies in Addis Ababa, Ethiopia, yielded 43 deaths and an estimated maternal mortality ratio of 480. At the 95% level of significance this gives a sampling error of around 30%, that is, the ratio could lie between 370 and 660. (Source: Ref- Rif 7) 79 mates were adjusted to account for under-report- ing and misclassification; and a simple model was developed to predict values for countries with no data. The model uses two widely available indepen- dent variables - general fertility rates and pro- portion of births that are assisted by a trained person - to predict maternal mortality. Maternal mortality estimates for individual countries can be divided into 5 categories (these categories are used in Tabl,e 1): A. Deue!,oped countries with compkte vital registration systems and attribution of cause of death. For these countries the maternal mortality ratio is the reported number adjusted by a factor of 1.5 to account for misclassification of maternal deaths (9,10). 11 B. Deue!,oping countries with good death registration but poor or non-existent attribution of cause of death. The model is used to predict the proportion of deaths of women of reproductive age that are maternal. This proportion is then applied to the deaths of women of reproductive age actu- ally registered to obtain the number of mater- nal deaths and the maternal mortality ratio. C. Countries with RAMOS type estimates of maternal mortality. The maternal mortality ratio derived from the RAMOS study is used directly without any adjustments. D. Countries with Sisterhood estimates of maternal mor- tality. Several recent studies have found that the Sisterhood Method under-estimates total fe- male adult mortality, and presumably, maternal mortality as well (6, 7). However, the Sisterhood Method, in addition to providing an estimate of maternal mortality, also provides estimates of the proportion of all deaths of women of repro- ductive age that are maternal) Therefore, for these countries, this observed proportion was applied to the total number of deaths of women of women ofreproductive age generated by the United Nations Population Division's popula- tion projections (1994 revision)j for the year 1990 since these are believed to be better esti- mates offemale adult mortality. E. Countries with no estimates of maternal mortality. For countries without accurate information on numbers of deaths and without direct or indi- rect estimates of maternal mortality, the model is used to predict the proportion maternal of all h The 1.5 adjustment factor is based on evidence from several studies. See, for example, Rrf- Rif 9 & W. i In so far as the Sisterhood Method identifiesall pregnancy-related deaths which mav include some due to fortuitous or accidental causes, it may ov~r-estimate maternal mortality. Howe\'er, the method is likely to miss some early maternal deaths such as those related to abortion or ectopic pregnancy. It has been assumed that the two biases cancel out. i United Nations Population Division, World popuu,tion prospect,: the 1994 revision, New York, United Nations, 1993 (ST/ESA/ SER.A/145). 80 deaths of women of reproductive age and this proportion is applied to the 1990 United Nations projections of adult female deaths to derive the maternal mortality ratio. The results of the new approach indicate that globally, there are some 585 OOO maternal deaths, 99% of them in developing countries.f This is around 80 OOO deaths more than earlier estimates have suggested and indicates a substantial underes- timation of maternal mortality in the past. More- over, the number of deaths represents only a small fraction of the total burden of disease associated with pregnancy and childbirth. Conservatively esti- mated, the submerged fraction of the iceberg amounts to some 20 million morbidities and dis- abilities each year, ranging from acute and devas- tating injuries such as obstetric fistula to chronic, debilitating conditions such as severe anaemia, re- productive tract infections and uterine prolapse. Some one-third of all women in the developing world have suffered such problems at some time in their lives. In developing countries as a whole, maternal mortality ratios range from 190 per 100 OOO live births in Latin America and the Caribbean to 870 per 100 OOO in Africa. Extremely high ratios of over 1000 per 100 OOO live births are found in Eastern and Western Africa (Tabl,e 2). Differences between these revised estimates and previous estimates of maternal mortality The maternal mortality ratios derived from this new approach differ from earlier estimates, both in terms of global numbers of maternal deaths, and in terms of the regional breakdowns. In particular, estimates for Africa are generally much higher whereas those for Asia and Latin America as a whole are broadly comparable to the earlier figures (Tabl,e 3). These new estimates differ - in some cases considerably- from official figures or from figures derived from other sources such as Sisterhood studies. For example, the figures quoted for devel- oped countries are based on official figures in- flated by a factor of 1.5 to account for misclassifica- tion of maternal deaths. As already noted, this new approach results in systematically higher estimates of maternal mortality than Sisterhood studies due to the fact that the Sisterhood estimates appear to underestimate adult female mortality and have been adjusted accordingly. Using the new estimates The new WHO/UNICEF approach to estimating maternal mortality is primarily intended to be of use in countries with no estimates of maternal mortality or where there is concern about the adequacy of officially reported estimates. The intention was to draw attention to the existence and likely dimensions Rapp. trimest. statist. sanit. mond., 49 (1996) Table 1 Country estimates of maternal mortality, lifetime risk and numbers of maternal deaths (1990) Tableau 1 Estimations de la mortalite maternelle, par pays, risque sur la vie entiere et nombre de deces maternels (1990) Maternal Number of Lifetime risk Categoryb of mortality ratio maternal of maternal estimate-(Maternal deaths- death. Categorieb deaths per 100 OOO Nombre de 1 in:- d'estimation live births) - deces Risque de Taux de maternels deces mortalite maternel sur maternelle la vie entiere (deces maternels 1 sur: pour 100 OOO naissances vivantes) Afghanistan 1 700 13 OOO 7 E Albania - Albanie 65 50 430 A Algeria - Algerie 160 1 200 120 E Angola 1 500 7 200 8 E Antigua and Barbuda - Antigua-et-Barbudaa Argentina - Argentine 100 690 290 B Armenia - Armenie 50 40 640 A Australia - Australie 9 25 4 900 A Austria - Autriche 10 10 5 600 A Azerbaijan -Azerba'idjan 22 40 1 400 A Bahamas 100 5 400 E Bahrain - Bahre'in 60 10 360 E Bangladesh 850 33 OOO 21 E Barbados - Barbade 43 5 1100 E Belarus - Belarus 37 50 1 300 A Belgium - Belgique 10 10 5 200 A Belizea Benin - Benin 990 2 300 12 E Bhutan - Bhoutan 1 600 980 9 E Bolivia - Bolivie 650 1 600 26 D Bosnia and Herzegovina - Bosnie-Herzegovinea Botswana 250 120 65 E Brazil - Bresil 220 8 400 130 E British Virgin Islands - lies Vierges britanniquesa Brunei Darussalam - Brunei Darussalam 60 5 430 B Bulgaria - Bulgarie 27 30 1 800 A Burkina Faso 930 4 OOO 14 E Burundi 1 300 3 400 9 E Cambodia - Cambodge 900 3 600 17 E Cameroon - Cameroun 550 2 600 26 E Canada 6 25 7700 A Cape Verde - Cap-Verta Central African Republic - Republique centrafricaine 700 850 21 E Chad-Tchad 1 500 3 700 9 E Chile-Chili 65 200 490 B China - Chine 95 22 OOO 400 c Colombia - Colombie 100 800 300 E Comoros - Comores 950 260 12 E Congo 890 890 15 E Cook Islands - lies Cooka Costa Rica 55 45 420 B Cote d'Ivoire 810 4 900 14 E Croatia - Croatiea Cuba 95 170 490 B Cyprus - Chypre 5 5 6 900 E Czech Republic - Republique tcheque 15 20 2 900 A Democratic People's Republic of Korea - Republique populaire democratique de Coree 70 370 500 E Denmark - Danemark 9 5 5 800 A Djibouti 570 110 24 E Wld hlth statist. quart., 49 (1996) 81 Table 1 (continued) Tableau 1 (suite) Dominica - Dominiquea Dominican Republic - Republique dominicaine 110 220 230 E East Timar - Timar orienta1a Ecuador - Equateur 150 460 150 E Egypt - Egypte 170 3 100 120 c El Salvador 300 530 65 D Equatorial Guinea - Guinee equatoriale 820 130 17 E Eritrea - Erythree 1 400 1 900 10 E Estonia - Estonie 41 10 1100 A Ethiopia - Ethiopie 1 400 33 OOO 9 E Fiji-Fidji 90 15 300 E Finland - Finlande 11 5 4 200 A France 15 110 3100 A French Polynesia - Polynesie fran~aisea Gabon 500 210 32 E Gambia - Gambie 1 100 460 13 E Georgia - Georgie 33 30 1 100 A Germany - Allemagne 22 190 2 700 A Ghana 740 4 800 18 E Greece - Grece 10 10 5 600 A Grenada - Grenadea Guadaloupe - Guadeloupe a Guama Guatemala 200 730 75 E Guinea - Guinee 1 600 4 700 7 D Guinea-Bissau - Guinee-Bissau 910 380 16 c Guyanaa Haiti-Ha'iti 1 OOO 2 300 17 E Honduras 220 410 75 c Hong Kong 7 5 9 200 A Hungary - Hongrie 30 35 1 500 A Iceland - lslande 0 0 0 A India- lnde 570 147 OOO 37 E Indonesia - lndonesie 650 31 OOO 41 E Iran (Islamic Republic of) - Iran (Republique islamique d') 120 2 700 130 c Iraq 310 2 200 46 E Ireland - lrlande 10 5 3 800 A Israel - Israel 7 5 4 OOO A Italy- ltalie 12 65 5 300 A Jamaica - Jamai'que 120 65 280 c Japan - Japan 18 230 2 900 A Jordan - Jordanie 150 260 95 E Kazakstan - Kazakstan 80 300 370 A Kenya 650 7 OOO 20 E Kiribatia Kuwait - Kowert 29 15 820 E Kyrgyzstan - Kirghizistan 110 150 190 A Lao People's Democratic Republic - Republique democratique populaire lao 650 1 200 19 c Latvia - Lettonie 40 15 1100 A Lebanon - Liban 300 220 85 E Lesotho 610 420 26 E Liberia - Liberia 560 690 22 E Libyan Arab Jamahiriya - Jamahiriya arabe libyenne 220 430 55 E Lithuania - Lituanie 36 20 1 200 A Luxembourg 0 0 0 A Madagascar 490 2 800 27 D Malawi 560 2 700 20 D Malaysia - Malaisie 80 440 270 B Maldivesa 82 Rapp. trimest. statist. sanit. mond., 49 (1996) Table 1 (continued) Tableau 1 (suite) Mali 1 200 5 700 10 E Malta - Malte 0 0 0 A Marshall Islands - lies Marshalla Martiniquea Mauritania - Mauritanie 930 750 16 E Mauritius - Maurice 120 25 300 B Mexico - Mexique 110 2 700 220 B Micronesia Federal States - Micronesie (Etats federes de)a Mongolia - Mongolie 65 45 310 B Montserrata Morocco - Maroc 610 4 500 33 0 Mozambique 1 500 9 800 9 E Myanmar 580 8100 33 E Namibia - Namibie 370 190 42 0 Nepal - Nepal 1 500 11 OOO 10 E Netherlands - Pays-Bas 12 25 4 300 A Netherlands Antilles - Antilles neerlandaisesa New Caledonia - Nouvelle-Caledoniea New Zealand - Nouvelle-Zelande 26 15 1 600 A Nicaragua 160 250 100 c Niger 1 200 5100 9 0 Nigeria - Nigeria 1 OOO 44 OOO 13 E Norway - Norvege 6 5 7 300 A Oman 190 150 60 E Pakistan 340 18 OOO 38 E Palau - Palaosa Panama 55 35 510 B Papua New Guinea - Papouasie-Nouvelle-Guinee 930 1 200 17 E Paraguay 160 240 120 E Peru - Perou 280 1 700 85 E Philippines 280 5 400 75 0 Poland - Pologne 19 100 2 200 A Portugal 15 20 3 500 A Puerto Rico - Porto Ricca Qatara Republic of Korea - Republique de Goree 130 900 380 B Republic of Moldova - Republique de Moldova 60 50 580 A Reunion - Reuniona Romania - Roumanie 130 410 340 A Russian Federation - Federation de Aussie 75 1 500 620 A Rwanda 1 300 4 OOO 9 E Saint Kitts and Nevis - Saint-Kitts-et-Nevisa Saint Lucia - Sainte-Lucie a Saint Vincent and the Grenadines - Saint-Vincent-et-les-Grenadinesa Samoa 35 5 500 E Sao Tome and Principe - Sao Tome-et-Principea Saudi Arabia - Arabie saoudite 130 730 95 E Senegal - Senegal 1 200 3 900 11 0 Seychellesa Sierra Leone 1 800 3 600 7 E Singapore - Singapour 10 5 4 900 A Slovakia - Slovaquiea Slovenia - Slovenie 13 5 4 OOO A Solomon Islands - lies Salomona Somalia - Somalie 1 600 7 OOO 7 E South Africa - Afrique du Sud 230 2 700 85 E Spain - Espagne 7 30 9 200 A Sri Lanka 140 520 230 B Sudan - Soudan 660 6 600 21 E Wld hlth statist. quart., 49 (1996) 83 Table 1 (continued) Tableau 1 (suite) Surinamea Swaziland 560 160 29 E Sweden - Suede 7 10 6 OOO A Switzerland - Suisse 6 5 8 700 A Syrian Arab Republic - Republique arabe syrienne 180 950 75 c Tajikistan - Tadjikistan 130 270 120 A Former Yugoslav Republic of Macedonia - Ex-Republique yougoslave de Macedoinea Thailand - Tha'ilande 200 2 300 180 E Togo 640 1 OOO 20 E Tongaa Trinidad and Tobago - Trinite-et-Tobago 90 25 360 B Tunisia - Tunisie 170 380 140 E Turkey- Turquie 180 3 OOO 130 c Turkmenistan - Turkmenistan 55 70 350 A Turks and Caicos Islands - lies turques et Ca'iquesa Tuva1ua Uganda - Ouganda 1 200 11 OOO 10 E Ukraine 50 320 930 A United Arab Emirates - Emirats arabes unis 26 10 730 E United Kingdom of Great Britain and Northern Ireland - Royaume-Uni de Grande-Bretagne et d'lrlande du Nord 9 70 5 100 A United Republic of Tanzania - Republique-Unie de Tanzanie 770 8 700 18 E United States of America - Etats-Unis d'Amerique 12 480 3 500 A Uruguay 85 45 410 B Uzbekistan - Ouzbekistan 55 380 370 A Vanuatu 280 15 60 E Venezuela 120 680 200 B Viet Nam 160 3 300 130 E Yemen - Yemen 1 400 8100 8 E Yugoslavia - Yougoslaviea Zaire - Za'ire 870 16 OOO 14 E Zambia - Zambie 940 3 500 14 E Zimbabwe 570 2 300 28 E a For these countries it was not possible to calculate maternal mortality ratios using this methodology in the absence of independent variables. - Pour ces pays, ii n'a pas ete possible de calculer les rapports de mortalite maternelle a l'aide de cette methode a cause de !'absence de variables independantes. b Categories explained in text. - Les categories sont definies dans le texte. Sources: World Health Organization and United Nations Children's Fund. Revised 1990estimates of maternal mortality: a new approach by WHO and UNICEF. Geneva & New York, 1996 (WHO/FHE/MSM/94.11 - UNICEF/PLN/96.1 ). of the problem of maternal mortality. The estimates should be taken as indicating orders of magnitude rather than precise estimates and are not necessarily what governments consider most appropriate. The standard errors associated with the predicted mater- nal mortality ratios are very large. They cannot, therefore, be used for regular monitoring of trends. The figures pertain to the year 1990 and should be seen as a recalculation of the earlier 1991 revision rather than as indicative of trends since then. The results for each country should serve as a stimulus to action and to help mobilize national and external resources to this end. The nature of such action will be determined in large measure by the social and economic conditions of the country but must include increased access to high quality care during pregnancy and childbirth for all women. 84 Other ways of monitoring trends in maternal mortality Where current vital registration systems underesti- mate maternal mortality due to misclassification of maternal deaths, there is room for improvement through the establishment of a system of confiden- tial inquiries which not only result in better estima- tion of the dimensions of the problem but also, in so far as they identify the causes of misclassification and analyze the management of each case, lead directly to improvements in case management and reductions in substandard care ( 11). National-level confidential enquiries are pri- marily of value in settings where the majority of deaths of women of reproductive age are officially registered. Where this is not the case, similar ap- proaches can be used at the facility level to gener- Rapp. trimest. statist. sanit. mond., 49 (1996) Table 2 Revised estimates of maternal mortality by United Nations regions (1990) Tableau 2 Estimations revisees de la mortalite maternelle selon les regions des Nations Unies (1990) Maternal mortality ratio Number of Lifetime risk of (maternal deaths per maternal maternal death, 1 in: - 100 OOO live births)- deaths- Risque de deces Taux de mortalite Nombre de maternel sur maternelle ( deces deces la vie entiere, maternels pour maternels 1 sur: 1 00 OOOnaissances vivantes) World total - Total mondial 430 585 OOO 60 More developed regions - Regions plus developpeesa 27 4 OOO 1 800 Less developed regions - Regions mains developpees 480 582 OOO 48 Africa - Afrique 870 235 OOO 16 Eastern Africa - Afrique orientale 1 060 97 OOO 12 Middle Africa - Afrique centrale 950 31 OOO 14 Northern Africa - Afrique septentrionale 340 16 OOO 55 Southern Africa - Afrique australe 260 3 600 75 Western Africa - Afrique occidentale 1 020 87 OOO 12 Asia-Asiea 390 323 OOO 65 Eastern Asia - Asie orientale 95 24 OOO 410 South-central Asia - Asie meridionale et australe 560 227 OOO 35 South-eastern Asia - Asie du Sud-Est 440 56 OOO 55 Western Asia - Asie occidentale 320 16 OOO 55 Europe - Europe 36 3 200 1 400 Eastern Europe - Europe orientale 62 2 500 730 Northern Europe - Europe septentrionale 11 140 4 OOO Southern Europe - Europe meridionale 14 220 4 OOO Western Europe - Europe occidentale 17 350 3 200 Latin America & the Caribbean - Amerique latine et Caraibes 190 23 OOO 130 Caribbean - CaraTbes 400 3 200 75 Central America - Amerique centrale 140 4 700 170 South America -Amerique du Sud 200 15 OOO 140 North America - Amerique du Nord 11 500 3 700 Oceania - Oceaniea 680 1 400 26 Australia & New Zealand - Australie & Nouvelle-Zelande 10 40 3 600 Melanesia - Melanesie 810 1 400 21 • Australia, New Zealand and Japan have been excluded from the regional totals but are included in the total for developed countries. - L'Australie, la Nouvelle-Zelande et le Japan sont exclus des totaux regionaux mais inclus dans le total des pays developpes. Note: Figures may not add to totals due to rounding. - II se peut que le total ne corresponde pas tout a fail aux chiffres donnes, ceux-ci etant arrondis. Source: World Health Organization and United Nations Children's Fund. Revised 1990 estimates of maternal mortality: a new approach by WHO and UNICEF. Geneva & New York, 1996 (WHO/FHE/MSM/94.11 - UNICEF/PLN/96.1 ). ate valuable information about the causes and cir- cumstances of maternal deaths. Facility-level audits of maternal deaths can be used to identify health service factors associated with each death and to determine to what extent each death could have been avoided. Tracing the route taken by the de- ceased woman prior to arrival at the facility offers clues about possible physical, sociocultural and economic barriers that impede access to appropri- ate care in a timely manner. Such studies are essen- tially qualitative in nature and do not provide esti- mates of levels of maternal mortality. Nonetheless, W/d hlth statist. quart., 49 (1996) they can lead directly to improvements in service delivery and to efforts to remove barriers to care. WHO has developed a guide describing this pro- cess in some detail.k To permit countries without a vital registration system to monitor trends, UNICEF and WHO pro- pose process indicators which describe the causal k Ireland, J. &: Graham, W. Conducting a r.a1e review of maternal deaths (draft for fieM testing). World Health Organization, May 1996. 85 Table 3 Maternal mortality - New regional estimates compared with previous estimates Tableau 3 Mortalite maternelle - Nouvelles estimations regionales comparees aux estimations precedentes UN Region- Maternal Maternal Maternal Maternal Region des Nations Unies mortality ratio mortality ratio deaths deaths (Maternal (Maternal (OOOs)- (OOOs)- deaths per 100 OOO deaths per 100 OOO 01\ces Deces live births)- live births) - maternels maternels Taux de Taux de (pour 1 OOO) (pour 1 OOO) mortalite mortalite maternelle maternelle ( deces matemels (deces maternels pour 100 OOO pour 100 OOO naissances naissances vivantes) vivantes) Old estimates - New estimates - Old estimates - New estimates - Anciennes Nouvelles Anciennes Nouvelles estimations estimations estimations estimations WORLD - MONDE 370 430 509 585 MORE DEVELOPED REGIONS - REGIONS PLUS DEVELOPPEES 26 27 4 4 LESS DEVELOPED REGIONS - REGIONS MOINS DEVELOPPEES 420 480 505 582 AFRICA - AFRIQUE 630 870 169 235 Eastern Africa - Afrique orientale 680 1060 60 97 Middle Africa - Afrique centrale 710 950 21 31 Northern Africa - Afrique septentrionale 360 340 17 16 Southern Africa - Afrique australe 270 260 4 3.6 Western Africa - Afrique occidentale 760 1020 66 87 ASIA-ASIEa 380 390 310 323 Eastern Asia - Asie orienta1ea 120 95 30 24 South-central Asia - Asie meridionale et centrale (570)b 560 (224)b 227 South-eastern Asia - Asie du Sud-Est 340 440 42 56 Western Asia - Asie occidentale 280 320 12 16 EUROPE (23)b 36 (1 )b 3.2 LATIN AMERICA/ CARIBBEAN - AMERIQUE LATINE/CARAiBES 200 190 25 23 Caribbean - Carai"bes 260 400 2 3.2 Central America - Amerique centrale 160 140 6 4.7 South America - Amerique du Sud 220 200 17 15 NORTH AMERICA - AMERIQUE DU NORD 12 11 0.5 OCEANIA - OCEANIEc 600 680 1.4 • Excluding Japan. -A !'exclusion du Japan. b Direct comparisons are not possible because of the redistribution of parts of the former USSR between the two regions. - Les comparaisons directes sont impossibles a cause de la repartition de certaines parties de l'ancienne URSS entre les deux regions. c Excluding Australia and New Zealand -A !'exclusion de l'Australie et de la Nouvelle-Zelande. Note: Figures may not add to totals due to rounding. - II se peut que le total ne corresponde pas tout a fait aux chiffres donnes. ceux-ci etant arrondis. Sources: World Health Organization. Maternal mortality ratios and rates: a tabulation of available information. Third edition. WHO/MCH/MSM/91.6. World Health Organization and United Nations Children's Fund. Revised 1990 estimates of maternal mortality: a new approach by WHO and UNICEF. Geneva & New York, 1996 (WHO/FHE/MSM/94.11 - UNICEF/PLN/96.1 ). pathways leading to maternal deaths and examine the coverage and quality of services for the man- agement of obstetric complications.I Process indi- cators can help to identify the most appropriate I UNICEF &: World Health Organization. Maternal mmtality: guidelines for monitoring progress ( draft), 1996. 86 mix ofinterventions and to assess progress towards improved coverage and quality of care. The use of process indicators does not imply the abandonment of efforts to measure impact, that is, maternal mortality ratios. However, it is unrealistic to expect that all countries will be able to establish the kind of ongoing monitoring systems needed for aregularappraisalofmaternalmortality.Norwould it be appropriate to direct scarce resources to such Rapp. trimest. statist. sanit. mond., 49 (1996) an undertaking at the expense of programmes to deal with the problem at its source. Implications for the future Despite its limitations in terms of monitoring, this approach represents a substantial improvement on earlier efforts to estimate maternal mortality at regional and global levels, but more particularly at the national level. At regular intervals, WHO and UNICEF will update and expand the data set and re-estimate maternal mortality. The use of such strategies to estimate maternal mortality is a short-term solution to the problem of measurement. In the long term, accurate informa- tion about maternal mortality is dependent on im- provements in vital registration systems and their incorporation into all national health information systems. This must be the ultimate objective of all national authorities and of multilateral and bi- lateral development agencies. In the meantime, the information now available on probable levels of maternal mortality (Tab/,e 2) should serve to stimulate greater action on the part of national authorities and international assistance agencies to reduce the toll of disease, death and disability associated with pregnancy and childbirth. Summary A new approach to measuring maternal mortality indi- cates that there are some 585 OOO maternal deaths, 99% of them in developing countries. This is around 80 OOO deaths more than earlier estimates have suggested and indicates a substantial underestimation of maternal mortality in the past. There is a greater disparity in levels of maternal mortality between industrialized and devel- oping countries than in any other public health indicator While significant progress has been made in reducing infant mortality, the same is not true for maternal mortal- ity. Although the actions needed to reduce maternal mortality have long been known, 1 woman in 50 is still dying as a result of pregnancy-related complications and the figure rises to 1 in 10 in many parts of Africa. By contrast, the figure for developed countries can be as low as 1 in 8 OOO. Resume Mortalite maternelle Une nouvelle approche de la mesure de la mortalite maternelle indique qu'il y a environ 585 OOO deces ma- Wld hlth statist. quart., 49 (1996) ternels, dont 99% dans les pays en developpement. On compte done environ 80 OOO deces de plus que le nombre suggere par les estimations precedentes, ce qui indique que l'on a sous-estime considerablement la mortalite maternelle dans le passe. L'ecart est plus important dans les niveaux de mortalite maternelle entre les pays industrialises et les pays en developpement que dans tout autre indicateur de sante publique. On a realise des progres considerables concernant la reduc- tion de la mortalite infantile, mais cela n'est pas vrai en ce qui concerne la mortalite maternelle. Bien que l'on sache comment reduire la mortalite maternelle depuis longtemps, 1 femme sur 50 meurt encore de complica- tions liees a la grossesse, et ce chiffre passe a 1 sur 10 dans de nombreuses regions d'Afrique. Par contre, le chiffre relatif aux pays developpes est bas et n'est parfois que de 1 sur 8 OOO. References/References l. International statistical classification of diseases and related health problems. Tenth revision. Geneva, World Health Organization, 1992. 2. Fortney,J.A. et al. Ca.uses of death to women of reproductive age in Egypt. Michigan State University Working Paper N. 49, 1984. 3. Chavkin, W. & Allen, M. Questionable category of non- maternal death. American journal of obstetrics and g;·necology, 16(5): 1640-41 (1993). 4. Cause of death statistics and vital rates, civil registration systems and alternative sources of information. World health statistics annual 1993. Geneva, World Health Organization, 1994. 5. Kwast, B.E. et al. Epidemiology of maternal mortality in Addis Ababa: a community-based study. Ethiopian medical journal, 23: 7-16 (1985). 6. Shahidullah, M. The Sisterhood Method of estimating maternal mortality: the Matlab experience. Studies in family planning. 26:2: 101-106 ( 1995). 3. Stanton, C. et al. Modelling maternal mortality in the developing world (forthcoming, 1996). 8. Walker, G.J. et al. Maternal mortality in Jamaica. Lancet 1(8479): 486-488 (1986). 9. Bouvier-Colle, M.-H. et al. Reasons for the underreporting of maternal mortality in France, as indicated by a survey of all deaths of women of childbearing age. International journal of ejJidt,mio/,ogy, 20: 717-721 ( 1991). 10. Atrash, H.K. et al. Maternal mortality in developed countries: not just a concern of the past. Obstetrics and {fjneco/,og;•, 86:700-703 (199.'i). 11. Report on r.onfidential enquiries into maternal deaths in England and Wales 1982-1984. United Kingdom, Department of Health and Social Security. London, HMSO, 1989. 87 Contraceptive method mix menu: providing healthy choices for women K. Fern GreenwelJa. b Introduction Policy makers, programme managers and interna- tional aid agencies are faced with a dual challenge: to provide an array of family planning method options - the "contraceptive mix menu" - and to ensure that methods are safe, effective, acceptable and affordable. Their challenge is further compli- cated by the number of couples lacking an appro- priate method to space or terminate childbearing (fertility regulation) and the complex barriers to providing these methods. Meeting the challenge begins with knowing who the underserved are, and why they are under- served. As will be defined in more detail, under- served couples arc for the present purpose those who wish to regulate their fertility, but who cur- rently use no method. It also includes those who use either a supply method or a non-supply meth- od, but have not had a menu of options from which to choose. "Supply" methods are contraceptives which require supplies or medical services. "Non- supply" methods include fertility awareness meth- ods as well as traditional methods such as with- drawal and indigenous herbs. This article addresc;es the expansion of menu options, specifically in the context of underserved women who, like all women, expect to maintain their health status while successfully preventing an unintended pregnancy. It is a critical review of the health implications of the contraceptive methods most commonly included on a limited menu of choices. It includes fertility awareness methods as essential non-supply method options where bar- riers currently exist for supply methods. The underserved population Despite progress resulting from vigorous efforts to make contraception available in developing coun- tries, the method choices remain scarce or non- existent for many couples. Although the number of underserved couples who have an unmet desire to space births is at least 120 million ( 1), the actual number of underserved couples falling into two a Short-term professional, U:'.'JOP/UNFPA/WHO/World Bank Special Programme of Research, Development and Research Training in Human Reproduction, World Health Organiza- tion, 1211 Geneva 27, Switzerland. b The views expressed in this article are those of the author and do not necessarily reflect those of the Special Programme or of its co-sponsors. 88 categories may be significantly higher. The first category of underserved couples includes those wish to regulate their fertility but use neither a supply nor a non-supply method. One possible way of estimating the number of couples in this cate- gory would be to assume that if couples worldwide wished to achieve "replacement level" families (an average of 2.1 children per couple), then approxi- mately 74% of couples now of child-bearing age would regulate their fertility (2). Given this as- sumption, in 1990 there were still 21 % of couples in the less developed regions who, despite the de- sire to regulate fertility, used neither a supply nor a non-supply method to do so. The percentages of such couples vary greatly between regions: 56% in Africa; 16% in some countries in Asia and Oceania (the contraceptive prevalence rate in Eastern Asia - including China, Hong Kong and the Re- public of Korea - is 79%, suggesting that all cou- ples desiring to regulate their fertility are using some method, but in other countries in the region an average of32% still use no method to meet their fertility regulation desires); and 16% in Latin America and the Caribbean (3). The percentages in Africa may be overstated as couples in these regions frequently desire higher fertility than re- placement level (4). The second category of underserved couples include those who have a limited menu of options available. This number includes couples who are using some method (either supply or non-supply) but would switch to a more acceptable method given the opportunity. The percentage of under- served couples in this category is not available from global data as published, but descriptions from regional studies help identify this population. For example, in a study of 4 countries - Nigeria, Mexico, Bangladesh and Turkey- it was found that the use of a commonly available method does not necessarily reflect the woman's preference; the main reason that many women chose their initial method is that they knew of no other alternative at the time (5). Barriers to using a supply method Underserved couples experience three main bar- riers to using supply methods to exercise their "right to freely space and time their children" (6). In order of significance they are: fear of health risks; lack of good-quality services and products; and reasons of conscience or cultural impediments. Rapp. trimest. statist. sanit. mond., 49 (1996) Health concerns and side effects play a predom- inant role in non-use of supply methods. Although these are often unfounded misconceptions (7), they remain primary reasons for not using supply meth- ods. For instance, Bongaarts and Bruce (8) analysed 12 developing countries from the Demographic and Health Survey data and found that health con- cerns were expressed with greater frequency than any other problem. They estimate that health con- cerns reduce prevalence on average by 71 % for the pill, 86% for IUDs, and 52% for sterilization. The second most common reason is the lack of information and access to family planning ser- vices, especially in sub-Saharan Africa where 1 in 4 women is not using a method for this reason (9). Countries where the government or external do- nors do not commit substantial resources often lack services and available products. Furthermore, family planning services which are largely provided by the private sector generally cost more and may be cost-prohibitive for the client (4). The third and less well-documented reason for not using a supply method still remains an impor- tant one. Women whose cultures strongly adhere to traditions may not choose an available method because it is not in harmony with their cultural, traditional or religious customs. Some women choose not to use a method because their hus- bands or family members do not approve. Other women may not be able to use supply methods in good conscience if their cultural norms or reli- gious mores do not permit it ( 10). A menu of safe choices The contraceptive method mix menu must offer an array of safe choices; contraceptive "safety" refers to contraceptive security, including both protection against pregnancy (effectiveness) and consideration of short-term or long-term health risks. Protection against pregnancy depends on the method's effectiveness in perfect (ideal) use, and on how well each couple complies with perfect use. Likewise, the health risks associated with a method are determined through clinical trials, and also by each couple's physical and environmental circum- stances. That which is a "mere side effect" to a biomedical scientist may be a significant problem to a woman in a poor or rural area. For example, an already malnourished woman who must spend her day working in the fields may not tolerate the in- convenience of a headache or bleeding between menstrual periods. Scientists have concluded that the risk of death from contraceptives is lower than that from pregnancy ( 11). However, even when practising contraception is safer than experiencing pregnancy and childbirth, the complications and other disadvantages associated with each available method must be made clear to the women who will then weigh the risks and benefits in their personal and cultural milieu ( 12). Wld hlth statist. quart., 49 (1996) A woman's choice of a safe contraceptive op- tion is influenced by her understanding of how a contraceptive method might affect her short- and long-term quality of life. In order to reduce risks, all methods must be provided with appropriate quality of care and always with full information provided in an understandable manner, especially to women who are illiterate or infrequently ex- posed to modem technology and information dis- semination. According to ethical guidelines, the following information should be shared with any prospective client: mode of action of each contra- ceptive; long- and short-term health effects, includ- ing positive and negative; effectiveness; reversibili- ty; and protection against sexually transmitted diseases where there is increased risk of exposure. Information regarding the range of methods available, the medical support needed and the re- liability of supply must also be communicated. The standard menu The "standard menu" of contraceptives for wom- en consists of the methods most used in developing regions. It includes female sterilization, the IUD, the pill, condoms and non-supply methods (3). Hormonal injectables and implants are being of- fered more frequently as options, but conclusive data on their use are not yet widely available. A summary highlighting the comparative benefits and disadvantages of commonly-used methods fol- lows. Sterilization is the method most used by wom- en in less developed regions to terminate child- bearing; it is chosen by 20% of women using any contraceptive method. This method has the high- est rate of use in Eastern Asia where 33% of the women ofreproductive age are sterilized (3). Most of the recent growth in contraceptive prevalence in developing countries has been due to increases in sterilization ( 13). Sterilization is the most effec- tive way to prevent pregnancies, with failure rates of less than 1 % in the first year after surgery. Vol- untary sterilization may he an attractive option es- pecially for older women for whom childbirth is associated with increased risks of complications. It requires only one-time motivation and no on- going conscious effort to regulate fertility. It is generally available at a low, one-time cost to the client (14). Female sterilization is relatively free of risks out- side of the surgical procedure and shortly there- after. Complications can be reduced by proper medical screening and by having well-trained med- ical personnel perform the procedure. The mortal- ity rate for female sterilization in developing coun- tries ranges from 4 to 20 per 100 OOO; many rates are less than 10 per 100 OOO which is significantly lower than the maternal mortality rate in many developing countries ( 15). Although the incidence of pregnancy following sterilization is very low, 89 4-64% of these pregnancies may be ectopic com- pared to 1 % among women who are not sterilized ( 16). A potential disadvantage of sterilization is that it is an irreversible method of contraception and therefore not an appropriate option for couples who want to space their childbearing. The IUD is the second most widely used contra- ceptive after female sterilization in the less devel- oped regions. It accounts for 14% of all women who use a method. In the East Asian countries of China, Hong Kong and the Republic of Korea, 31 % of women use the IUD. High rates of use are also found in Egypt (28%), Viet Nam (33%) and Cuba (33%). By contrast, it is a method seldom used at all in Africa ( 4%), other Asian and Oceanic countries (5%), and in Latin America and Carib- bean countries (7%) (3). The IUD is a highly effec- tive, long-term means to avoid having children. First-year failure rates are only 0.5-3%, depending on whether the IUD is medicated or non-medicat- ed. It is effective for an average of 5-10 years, again depending on the type of IUD inserted. Other advantages which may attract users to this method are that it allows for spontaneous sex and does not rely on an ongoing supply of products or medical intervention. The costs of using this method are moderately low for the IUD client (14). Despite the high rates of use of this method in some countries, as well as continuation rates higher than those of oral contraceptives, the IUD in other countries holds an unfavourable reputa- tion with regard to safety. While an increased inci- dence of pelvic inflammatory disease (PIO) has been found in some women with IUDs, studies show that infection usually occurs within the first four months after insertion or to women at risk of being exposed to STDs (14). To minimize the risk of PIO and its subsequent complications, care must be taken to insert the IUD only in hygienic condi- tions, and to carefully screen out women with mul- tiple partners, or who have a partner with multiple partners. Other disadvantages may accompany use of the IUD. With a non-medicated IUD, many users expe- rience longer and more profuse menstrual bleed- ing, and some bleeding between menstruations. This may cause or worsen anaemia. Some women may experience pain due to uterine contractions. There is the risk of the uterus being perforated during insertion, especially in cases where muscles are weak after a cesarean section (14). The IUD should be inserted by trained health personnel who remain available to attend to any subsequent complications, such as IUD expulsion, or requests to remove the IUD. Combined oral contraceptives, known com- monly as "the pill", are used by about 38 million women in developing countries (17), or about 6% of women of childbearing age. The pill is most widely used in Latin America by about 16% of the 90 women, but relatively infrequent use is seen in most other less developed regions, with the ex- ception of certain countries, for example in Africa- Algeria (39%), Morocco (28%), Zimbabwe (31%) - and Asia- Iran (23%) and Kuwait (24%) (3). In theory, i.e., with perfect practice, the effica- cy of the pill is close to 100% in preventing unin- tended pregnancy. The method is reversible and highly convenient, with a good safety record which is dependent on the quality of health care. Despite efforts from some advocacy groups to deregulate pill provision, it remains highly regulated in most countries due to certain contraindications and the uncertainties about breast and cervical cancer in some women ( 18). Secondary health benefos of pill use include reduced menstrual flow which lowers the risk of iron deficiency anaemia; up to 50% reduction in pelvic inflammatory disease (PIO), probably due to an alteration in cervical mucus; and some protection against ectopic pregnancy probably related to the reduced risk of PIO. Pill use is associated with a reduction in the risk of certain cancers such as ovarian and uterine cancers pro- portional to the duration of use. Additionally, menstrual cycle difficulties such as painful periods and pre-menstrual syndrome are alleviated ·with pill use ( 19). The most serious adverse side effect attributable to oral contraceptive use is the increased risk of cardiovascular complications. Fortunately, this risk is reduced by the lower-dose pills currently on the market and by screening out high-risk patients. The pill is contraindicated for smokers, diabetics and women with hypertension, since blood pressure may be higher during the period of use. Oral contra- ceptive use has been associated with cervical and breast cancers. Cervical cancer is common among women in developing countries, and although there may be some association with use of oral contraceptives for more than 5 years, confounding factors related to a woman's sexual behaviour make it difficult to interpret the data ( 19). Current evi- dence remains contradictory on whether long-term oral contraceptive use by young women plays a role in breast cancer development (18). Women must also be made aware of possible side effects of pill use such as weight gain, nausea, headaches, increase in cervical mucus and de- crease in sex drive. This method is not appropriate for breast-feeding women since the quantity of breast milk may be decreased ( 14). There must also be the assurance of a regular supply of pills, ideally without having too many different brands available which could confuse providers and clients. For some individuals this is a costly method, especially in countries where medical screening and supervi- sion are required. There is widespread knowledge of the male condom as a contraceptive method, but it contin- ues to be the least used of all the "widely used" Rapp. trimest. statist. sanit. mond., 49 (1996) methods. Condoms are used by less than 10% of couples in less developed countries (3). Mainly be- cause of inconsistent or incorrect use, condoms are not as effective as other methods such as the oral contraceptive pill or IUD. Pregnancy rates vary widely between 3 and 15% although significantly lower rates are seen with perfect use. A significant advantage of condom use is that it does not require intervention by the medical community, and mini- mal training is needed for its use. In the present era of HIV and increased STD infection, the con- dom is getting increased attention as a safe method which provides partial protection against some STDs and reproductive tract infections. Studies have shown that there is a protective effect against gonorrhoea, syphilis and chlamydia (20-22); while more research is needed to determine the con- dom's actual protection, it may be increasingly pro- moted to help combat the growing STD and HIV epidemics in many countries. The biggest disadvantage to this barrier meth- od is that condoms interrupt sexual intercourse and are said to decrease sensitivity (23). Some users may find that condoms are cost-prohibitive espe- cially if used on a regular basis. Maintaining sup- plies and disposal after use are potential problems in rural areas. Fertility awareness methods are non-supply, natural fertility-regulating methods which are not commonly offered on the contraceptive mix menu. Fertility awareness methods are distin- guished from traditional non-supply methods, such as withdrawal or indigenous herbs, in that they require training to identify a woman's fertile period. Identification of the fertile period may be accomplished through noting subtle physiological changes in basal body temperature or cervical mu- cus ( the Billings method), by calculating the fertile period between menstrual cycles ( the calendar or rhythm method), or by using home-based indica- tors for detecting reproductive hormones, for ex- ample, in the urine (24). Abstinence or use of a back-up method during the fertile period is re- quired to avoid unintended pregnancy. It is diffi- cult to ascertain how many couples practice fertility awareness to avoid pregnancy. Of 77 million people worldwide who use a method but not a supply method, the International Medical Advisory Panel to the International Planned Parenthood Federation estimates that 10-15 million practise periodic abstinence.c Available data provide a wide range of effective- ness rates depending on the accuracy in identify- ing the fertile period, and the commitment to ab- stinence or a back-up method during this period. Overall use-effectiveness of these methods ranges c IMAP Highlights: Thirteenth meeting of IPPF International Medical Advisory Panel (IMAP), 11-13 October 1993, International Planned Parenthood Federation, London. Wld hlth statist. quart., 49 (1996) from less than 2% to over 30% during the first year. Methods based on observation of naturally occur- ring signs of the fertile period have a theoretical failure rate of between 1 % and 5%, depending on the method used. However, these methods have an actual failure rate of 10-30%, with lower rates seen when a method is being used to avoid pregnancy rather than to merely delay pregnancy (25). The calendar or rhythm method is most commonly used, although it has been shown to be the least effective in correctly identifying the fertile period. Other methods such as the cervical mucus method, the basal body temperature method, or a combina- tion of both (sympto-thermal method) are more accurate. Some over-the-counter fertile-period in- dicators are available but little information on their effectiveness has been published. There are no known side effects and no contra- indications to using fertility awareness methods. Medically qualified personnel are not required to teach or deliver services; the methods do not re- quire continuous supplies; and there is little cost attached to them. In areas where women have little opportunity for education, these methods provide an opportunity for empowerment which comes from knowing about their body's reproductive health and capabilities. In an era when men's in- volvement in family planning is being encouraged, these methods require male cooperation in practis- ing abstinence or agreeing to use another method during the fertile period. Finally, the ability of a woman to detect her fertile period may be useful in the diagnosis and treatment of infertility (25). The main problem with periodic abstinence is that it is ineffective if the women does not correctly identify her fertile period, or if the couple is not committed to abstinence or an effective back-up method during this period. Despite their high rate of ineffectiveness, fertility awareness methods are used by those who do not have continuous access to supply methods, who have health concerns re- garding supply methods, or who do not use supply methods for reason of conscience or culture. For example, Demographic and Health Survey data analysed for Bolivia showed that periodic absti- nence was the most popular method used by 26% of contracepting couples (26); and periodic absti- nence and other traditional methods were report- ed to have initial acceptance rates of 19% in Lagos, Nigeria, about 25% in Bangladesh, 15% in Mexico, and 26% in Turkey (mainly withdrawal) (5). The effectiveness of a fertility awareness method could be improved for these couples if better training in identifying the fertile period were incorporated into existing women's health services. Conclusions The cohort of contraceptive users is expected to grow rapidly in the developing world, increasing from 381 million couples in 1990 to 567 million 91 couples in the year 2000 (27). This presents a daunting task for the policy makers, programme managers and international aid agencies responsi- ble for providing a menu of safe, effective, accept- able and affordable contraceptive options. In light of this, two major conclusions can be drawn. Firstly, a menu of options is necessary because there is not a single method that is appropriate for all couples. Among the standard options, female sterilization, the IUD and the combined oral con- traceptive are all highly effective in preventing un- intended pregnancies. However, they are acknowl- edged to be associated with health risks or other disadvantages unacceptable to some women. On the other hand, the male condom and periodic abstinence are free from health risks associated directly with the contraceptive, but the rate of ef- fectiveness is unacceptably low for many couples. Secondly, barriers to the use of supply methods remain, mainly due to health concerns, service logistics and religious and cultural traditions. For underserved couples faced with these barriers, non-supply methods such as fertility awareness methods are an essential menu option to increase fertility regulation opportunities. For wider accept- ability, however, effectiveness must be improved through continuing research on indicators which help a woman identify the beginning and end of her fertile period, and education on these options should be integrated with existing family planning services. For couples who choose supply methods, research and efforts to improve options and avail- ability should continue. Information about the benefits and disadvan- tages of each menu option must be presented to couples so that they can make a fully informed decision about which method is most suitable for their fertility regulation needs. The outcome will be successful when all couples can exercise their "right to freely space and time their children" (6). Acknowledgements My appreciation is extended to Dr Paul Van Look and Dr Iqbal Shah for their helpful comments and suggestions. Summary This article addresses expansion of the contraceptive method mix, in the specific context of underserved women in developing countries who, like all women, expect to maintain their health status while successfully regulating their fertility. It is a critical review of the health implications of the contraceptive methods most com- monly included on a menu of options and includes fertility awareness methods as essential non-supply method options where barriers currently exist for supply methods. 92 Resume Assortiment de methodes contraceptives: offrir des choix sains aux femmes Cet article aborde le developpement du «menu» des choix de methodes contraceptives. plus precisement en ce qui concerne les femmes sous-desservies des pays en developpement qui, a l'instar de toutes les femmes. esperent maintenir leur etat de sante tout en regulant leur fecondite avec succes. II examine de maniere critique les repercussions sanitaires des me- thodes contraceptives qui figurent le plus couramment sur un menu d'options. II comprend des methodes qui permettent aux femmes de savoir a quel moment du cycle elles sont fecondes, ces methodes etant essen- tielles dans les pays qui ne peuvent s'approvisionner en d'autres moyens de contraception, pour des raisons culturelles ou financieres. References/References I. Robey, B. et al. The reproductive revolmion: t'\ew survey findings. Population re/,ort;, Series M, :-.lo. 11 ( 1992). 2. Weinberger, M.B. Recent trends in contraceptive behaviour. Prom•dings of the Dernograjlhic l lealth Surory.s World Conjl'rena, Washington, DC, I: ci:i3-,,74, 1991. 3. United Nations. H'orld conlmaptivl' ,hart 1994, New York, United Nations (1994) (Publication No. ST/ESA/SER.A/ 143). 4. Population Action International. Finrmring the future: meeting the demand for p1rni(l planning. 'Washington, DC, 1994. 3. Pullum, T. & Shah, I. Stmtegit!s of contmaptinn use dwing fertility transitions, International Populalion Conjirena, lvlontr('(l/, 1993, Volume I. Liege, lnlernalional Cnion for the Scientific Study of Population (IUSSP). 6. Pmgramm.e of ru:tirm, Cairo, International Conference on Population and De\'clopment (ICPD), Chapler 7.2 (1994). 7. Grubb, G. Women's perceptions of lhe safety of the pill: a survey in eight developing countries. Journal of biomedical scienw, 19('.1) (1987). 8. Bongaarts, J. & Bruce, J. The causes of unmet need for contraception and lhe social content of ser,,,ices. Studies in Jami(~ plrwniug, 26(2): 37-75 (199:i). 9. Shah, I. l'erspectivl'S on method., of fertility regultllion: setting a research ageuda, l!NDP /UNFPA/\\'HO/\\'orld Bank Special Programme of Research, De\'clopment and Research Training in Human Reproduction. Geneva (forthcoming, 1996). 10. Ryder, R.E. et al. :-.latural family planning: values and evaluations. R,f,rodtulive h.ealth matters. 3: 63-67 ( 1994). 11. Hatcher, R. et al. (eds.). Contmaptiue technology, !\'.cw York, Ir,:ington Publishing Inc., 1994. 12. Edelman, D. IUD complications in perspective. Contraceplion. 36(1): 159-167 (1987). 13. Benagiano, C. & Shah, I. The ernlution of corllraceptive methods and practices lo the year '.1000. In: Fertility and .,tnility. f'rocet!dings of the 15th World Congress on Fertility and Sterility. Mompellier, France, 1995. 14. World Health Organization. Contmr£/itine method mix: Guidt!lirws for policy and service delivery, Geneva, World Heallh Organizalion, 1994. 15. Ross,J.A. etal. Voluntmy sterilization: an internationalfart book, l\ew York, Association for Voluntary Sterilization, 1985. 16. World Health Organization. Fnnale sterilization: A guidl' to provision of servias. Geneva, WHO (1992). Rapp. trimest. statist. sanit. mond., 49 (1996) 17. Lower-dose pills. Population reports, Series A, No. 7 ( 1988). J 8. Rees, H. Acquiring the pill: safety issues. Reproducliue health matters, 3: 4 J -45 ( 1994). 19. Oral contrrU,eptiues and neoplasia. Report tJJ a WHO scientific group, Geneva, World Health Organization, 1992 (WHO Technical Report Series, No. 817). Contraception omlR et tumeurs. Rapport d 'un groupe scientifique de l'OMS, Gencve, Organisaton mondiale de la Santi:, 1992 (OMS Serie de RapporL~ kchniques, N° 817). 20. Update on condoms - products, protection, promotion. Population reports. Series H, No. 6 (1982). 21. Infertility and sexually transmiued diseases: a public health challenge. Population reports, Series L, No.4 (1983). 22. AIDS - A public health crisis. Population rej,mts, Series L, No. 6. (1986). Wld hlth statist. quart., 49 (1996) 23. World Health Organization. Barner contraceptiues and sf,ermicides: thdr rul.e in family planning care, Geneva, V.1{0, 1987. 24. Continence periodique. Poj,ulation rtt)()rts. Series I, No. 3 (l 982). 25. World Health Organization. Natural family plannin1r a guide tu the f,rouision of semices, Gene\'a, WHO, 1988. 26. Remez, L. Rhythm accounts for half of all contraceptive use in Bolivia, DHS re\'cals. lnterrwtio111il family planning penfwtiues, 17(1): 3&,;17 (1991). 27. Shah, I. The ad\·ance of the contracepti\'e re\'olution. Hnrld health .1tatistics qumll>rl;, 47(1): 9-15 (1994). Shah, I. Les progres clans la rernllllion contraceptive (resume] . lu1f1port trirnestriel d,, statistiques sanitaires mondial.es, 47(1): 15 (1994). 93 Women and AIDS in South and South-East Asia: the challenge and the response Nafsiah Mboia Introduction: What is special about women and AIDS? The impact of AIDS on society is disastrous. Women in South and South-East Asiab are particu- larly vulnerable to infection and particularly hard hit by the presence of HIV/ AIDS in the community (1-3). It is the purpose of this article to examine this phenomenon, to describe the most common con- ditions and attitudes which place women in the region at risk of infection and hamper efforts to reduce their vulnerability, and to suggest some actions which can be taken to reduce or eliminate some of the risk and/ or burden. In short, two concerns are addressed: breaking the chain of in- fection and reducing the suffering caused by HIV I AIDS. To do this effectively we must acknowledge and work to overcome traditional, persistent gender: imbalances and inequities which complicate ef- forts to halt HIV/ AIDS. At the same time, we sub- mit that this can be done only if we work with the whole community - with men and women, with parents and children, with families and community groups. But does infection of women really matter? Is the special effort that will be needed to reduce the numbers infected and impact of infection among women justified? We would answer with a resounding "yes". There are, in fact, a number of reasons why infection of women with HIV is an issue deserving special attention throughout the region. First: Health is a basic human right. HIV I AIDS is a direct threat to realization of that right. For • Indonesian paediaLrician and Member of Parliament, Vice Chair of the Global Commission on 'Women's HealLh, member of the Indonesian National Working group for AIDS. h For purposes of Lhis article "South and South-East A~ia" is defined to include the l i contiguous countries lying to Lhe south of China and stretching from PakisLan on the west to the Philippines on the easL - Pakistan, Nepal, Bhutan, lndia, Sri Lanka, Maldives, Bangladesh, Myanmar, Cambodia, Thailand, Lao Peoples Democanic Republic, Vietnam, Malaysia, Singapore, Brunei Darussalam, Indonesia and the Philippines. c The term "gender" is used in this article to refer Lo the roles, personal characteristics and behaviours which society assigns and reinforces for women and men in everyday life. Notions of gender, communicated consciously and unconsciously through training and experience from earliest childhood, arc of profound significance to men and women in many aspects of life, including their understanding and expression of sex and sexuality. This is, in turn, for a vast majority of women a determining factor influencing how they view and respond to the threat of AIDS in Lheir own lives, families and community. 94 everyone - women, men, children - it is a death sentence but one which is avoidable. Thus, every effort should be made to reduce the number of women affected and to make it possible for them to live with dignity and support if infected. However, a combination of social, cultural and economic factors limit women's access to the necessary infor- mation and services to enjoy those rights. This is an injustice. Second: Infected men and women both have the potential to spread infection to a sexual part- ner. Only a woman, however, can pass infection to her baby and thus the next generation. In most parts of the region, women are also likely to be the prime care-givers to people living with HIV I AIDS either in work settings (e.g., health care providers) or in home care. If they are not adequately in- formed and given the support needed, they run the risk of becoming infected themselves or uncon- sciously spreading infection further. Third: Women's role in the economies of the region is absolutely crucial. Throughout the re- gion they are increasingly well-represented in the work force. No country could sustain the economic dislocations that would come from serious reduc- tions in women's participation in the economy. In most countries of the region, women's contribu- tion to the family economy is equally important. Particularly among poor families, the woman's in- come often makes it possible for the family to stay above the poverty line and for their children to stay in school. Fourth: The objective of development through- out the region is improvement of the quality of life for all people. From the day a woman is infected with HIV the quality of her life is in decline. How- ever difficult her life was before, after infection it is worse. "Positive women "d talk with graphic clarity and passion about the combination of physical, psychological, and economic burdens they face once they know they have been infected. They worry about transmitting the virus. They worry about the costs of the illnesses which are part of the HIV I AIDS continuum. They worry about the situa- rl "Positive woman" means a woman who has been infected and is HIV-posith·e. "Woman living with AIDS", in our definiLion, means a woman who is nol HIV-positive herself, but is directly affected by HIV I AIDS, for example, Lhrough sickness of a spouse or other immediate family member, or a woman nurturing children or young people orphaned through AIDS. related deaLh of parents, etc. Rapp. trimest. statist. sanit. mond., 49 (1996) tion of their children. As one positive woman put it, "We don't want help to a dignified death. We want to be able to lead productive lives!" For all these reasons, then - human rights, control of infection, national and household econ- omy, quality of life - the issue of women and AIDS must be given high priority. In the remainder of this article we examine the situation in Asia and suggest some alternatives for action. The challenge The region and the epidemic The South and South-East Asia "region" (as we define it here) has a population of nearly 2 billion people, roughly 30% of the world's populatione (4). Politically and economically these countries are inextricably linked to one another. The rich flow of people and goods within and among them creates a seamless web of relations the importance of which cannot be underestimated in relation to the HIV I AIDS epidemic. Given the fertile environ- ment for the spread of HIV infection provided in the urban setting, it is worthy of note that urbaniza- tion is well advanced in some countries and rapid everywhere.f The total urban population in the region is just under 500 million people (4, 5). It is also important to recognize that well-developed public transportation systems (for example, in In- dia, Malaysia and Indonesia) mean that even if HIV I AIDS is initially concentrated in urban areas, it will quickly find its way to the villages, take root and spread with no further initiative needed from the city. Economically, there is great variety across the region butthe majority of countries (10outofl7)g are categorized as low income by the World Bank (6). Some economies have had 5, 10 or more years of sustained growth while others - for exam- ple, Vietnam, Cambodia or the Lao People's Dem- ocratic Republic - are only just beginning to recov- c Population estimates for countries vary somewhat from source to source. For purposes of this discussion we have taken data from the UN Publication The wor/,d's women 1995: trends and statistics (Ref 4), the most complete for the countries under discussion and the issues of concern. In cases where the basic data we sought were not found there, we have turned to 1995 report~ of the United Nations Population Fund (UNFPA) (RPJ 5, p. 68), World Bank (Ref 6) and United Nations Development Programme (UNDP) (Ref 7, p. 134). r In 1995, 3 countries of the region were less than 20% urban (Bhutan, Cambodia and Nepal); 7 were 20% to 30% urban (Bangladesh, India, Lao PDR, Myanmar, Sri Lanka, Thailand and Viet Nam); 3 were 31 %-35% urban (Indonesia, Maldives and Pakistan); 3 were 46-58% urban (Brunei, Malaysia and the Philippines); and Singapore was 100% urban. Urban growth rates ranged from a low of l % per year (Singapore) to a high of 6.9% (Nepal) with 3 countries near 6% per year (1990- 1995), a very high rate in both absolute and global terms. (RPJ 4, pp. 63-64). g Bangladesh, Bhutan, Cambodia, India, Lao PDR, Myanmar, Nepal, Pakistan, Sri Lanka and Viet Nam (Ref 6, p. 248). Wld hlth statist. quart., 49 (1996) er from decades of internal and international con- flict and their economies are struggling to achieve sustained growth. Notwithstanding the great progress which has been made across the region, poverty - absolute or relative - is a fact of life for hundreds of millions of women, men and children (Tabl.e 1). In 1994 UNDP estimated that 625 million people lived in absolute poverty across the region (8). In reality, the position of many more people is extremely precarious. They fall into the category which the World Bank referred to as the "near poor" in a study of Indonesia (7, 8). These people are not more than 10% above the official poverty line (in their respective countries) and thus ex- tremely vulnerable to the slightest economic change. Given the intimate, well-demonstrated re- lationship between poverty and vulnerability to HIV infection, particularly among women, these economic facts cannot be overlooked in address- ing what is sometimes mistakenly said to be "only" a health problem. According to data from UNAIDS in 1994 there were 884 million people between age 15 and 49 living in South-East Asia (9). Estimating the male and female populations to be roughly equal, we can figure that at least 442 million women are sexually mature and of an age where it can be assumed most are sexually active. Tragically, there are indications that estimates of sexually active females must be increased to include more younger girls - younger adolescents and children - who are neither sexually mature nor acting of their own free will. Although it is difficult to get reliable numbers, there is a consen- sus that child prostitution (girls and boys) and child sexual abuse, including domestic and international trafficking in children, are on the increase across the region (10-12), and personal communication with staff of the Centre for Prevention and Treatment of Child Sexual Abuse, Manila). In 1991 the UN Spe- cial Rapporteur on the Sale of Children, Child Prostitution and Child Pornography concluded that the root causes were "self-evident: poverty, com- pounded by family disintegration including incest and domestic violence, and migration from rural to urban areas and from one country to another in search of a livelihood" (13). Other studies have observed, significantly, that "relative rather than absolute poverty, may be more important ... " ( 10). Students of the global HIV I AIDS epidemic have documented, further, a rise in the demand for ever younger girls as sex partners, with a particularly high premium given for virgins, from patrons of the sex industry who believe that in this way they will avoid infection by older sex workers. Whatever the cause, the facts seem clear, response to the challenge of HIV I AIDS in the Asian region will be incomplete if the issue ofchild sexual exploitation and abuse is not taken up. Field studies also make it clear that howev- er important international influences may be - sex tours, military presence, international recruitment 95 Table 1 Some basic data on 17 countries of Asia Tableau 1 Donnees de base concernant 17 pays d'Asie Column Number No de colonne 2 3 1995 Life Expectancy - Population Esperance de vie (millions) F M (en millions) Bangladesh 128.3 53 53 Bhutan - Bhoutan 1.7 49 48 Brunei Darussalam - Brunei Darussalam 2.9 76 73 Cambodia- Cambodge 9.5 52 50 lndia-lnde 931.0 61 60 Indonesia - lndonesie 201.5 65 61 Lao People's Democratic Republic - Republique democratique populaire lao 4.9 53 50 Malaysia - Malaisie 20.1 73 69 Maldives 0.3 62 65 Myanmar 46.6 59 56 Nepal - Nepal 22.1 53 54 Pakistan 135.0 59 59 Philippines 69.3 67 63 Singapore - Singapour 2.9 77 72 Sri Lanka 18.3 74 70 Thailand - Tha'1lande 58.3 72 67 Viet Nam 73.8 66 62 Sources: Ref. ·Ref. 4 (Columns 1, 2, 4. 5, 6, 8) and Ref. ·Ref. 6. (Column 7). and sale of children - the existence of domestic sex industries and domestic systems of protection of illegal, commercial sex activity often encourage such activity and always make a profit. HIV I AIDS began to appear comparatively late in the Asian region, but it is "spreading now in many areas at a pace reminiscent of Africa 5-10 years ago" (1, 3, 14). Figures from the Global AIDS Policy Coalition are instructive (Tabl,e 2) (14). We find that as of 1 January 1995 the region made up only 14% of the cumulative global infections, but accounted for 34% of the previous year's new HIV infections. This pattern repeated itself in 1995. The region now accounts for 18% of the global total but 41 % of new infections during the previ- ous year! Infection among women showed a similar pattern. On 1 January 1996, the women of South- East Asia made up only 15% of the global total of infected women but they comprised 35% of all new infections among women during 1995. Asian women and the risks of infection Several factors - biological, sociocultural and economic, along with poor access to informa- tion - place women at high risk of infection and complicate efforts to reduce their vulnerability. 96 4 5 6 7 8 IMR- % Urban GNP-PNB %of TMI Urban- growth- (US$, 1993) population % en milieu Croissance < 15years- urbain urbaine % de la 1990·95 population < 15 ans 108 20 5.9 220 40 129 6 6.0 152 41 8 5.8 2.3 14 516 32 116 13 4.5 105 42 88 27 3.0 300 35 65 33 4.2 740 33 97 22 6.1 280 45 14 47 4.2 3 140 38 55 33 5.6 450 44 81 26 3.3 866 37 99 14 6.9 190 43 74 35 4.3 430 44 40 46 3.4 850 38 8 100 1.0 19 850 23 24 22 2.2 600 30 26 25 3.9 2 110 29 36 21 2.9 170 37 Basic data on women (Tabl,e 3) show generally low education, continuing high percentage of births before the mother is aged 20, and low literacy rates (both in absolute terms and as compared with men). All of these are indicators of women's poten- tially high vulnerability to infection from HIV. A prominent feature of the epidemic in the Asian region is the dominant role of sexual trans- mission. Worldwide, more than 80% of all infected women get the virus from a male sex partner. h Men have unprotected extramarital sex and become in- fected. They spread infection to their monoga- mous wives some of whom, in turn, pass infection to their babies. This pattern has been documented in Thailand and India with HIV infection showing up first among sex workers and their clients, subse- quently in prenatal clinics, and finally paediatric HIV begins to appear.i Most observers suggest that h UNAIDS. Points of 11ieiv: reducing women's 11ulnerability to HIV infection, 1996. i Brown & Xcnos report "l .8% of women going to prenatal care clinics in June 1993 tested H£V positive, and, at several maternity hospitals where infection levels are monitored, the percentage ofinfected mothers has been doubling every year." (RP/ J,p.3). Rapp. trimest. statist. sanit. mond., 49 (1996) Table 2 The global epidemic of HIV/AIDS in South and South-East Asia (Bangladesh Brunei, Burma, India, Indonesia, Malaysia, Maldives, Nepal, Philippines, Singapore, Sri Lanka and Thailand) Tableau 2 Epidemie mondiale de VIH/SIDA en Asiedu Sud et en Asiedu Sud-Est (Bangladesh, Brunei, Birmanie, lnde, lndonesie, Malaysie, Maldives, Nepal, Philippines, Singapour, Sri Lanka et Tha"ilande) World - Monde New infections, 1994 - Infections nouvelles, 1994 New infections, 1995 - Infections nouvelles, 1995 Cumulative, 1 Jan. 1996- Chiffres cumules au 1 er janv. 1996 South-East Asia (SEA) - Asie de Sud-Est (ASE) New infections, 1994 - Infections nouvelles, 1994 New infections, 1995 - Infections nouvelles, 1995 Cumulative, 1 Jan. 1996 - Chiffres cumules au 1 er janv. 1996 SEA as % of World - ASE en % du monde New infections, 1994 - Infections nouvelles, 1994 New infections, 1995 - Infections nouvelles, 1995 Cumulative, 1 Jan. 1995 - Chiffres cumules au 1 er janv. 1995 Cumulative, 1 Jan. 1996- Chiffres cumules au 1 er janv. 1996 Source: Ref-Ref. 3, 14. when testing and analysis are more widespread in Asia this will be confirmed as a tragically common pattern ( 1, 2). The insidious role of drugs is important in some areas, in particular, where the worlds of commercial sex and illicit drugs intermingle. For example in Thailand and Burma, HIV/ AIDS first showed up among sex workers who were drug users or had drug-using partners. Women, already more vulner- able to infection than men, are at dramatically much higher risk if they are in a drug-controlled environ- ment. Infection rates are high among drug users and girls "serving" them are frequently exposed to infected partners. More dangerous still, the sex workers themselves often become drug-dependent and lose their last vestiges of personal control and self-respect with the result that they accept, uncriti- cally, any treatment they receive. If they are intrave- nous drug users (in contrast to those who use non- injectable drugs) there is the additional possibility of infection from shared needles. Gender expectations/roles are crucial in deter- mining if or how a woman may protect herself, her sexual partner(s), even her unborn child from HIV infection. Within the region widely held stereo- types about what is "proper" and "normal" for men and women regarding sexual feeling and ex- pression severely limit the latitude most women have ( or will exercise) for action in the micro- Wld hlth statist. quart., 49 (1996) Men- Women- Children- Total Hammes Femmes Enfants 2 800 OOO 2 OOO OOO 700 OOO 5 500 OOO 3 367 OOO 2 311 OOO 919 OOO 6 597 OOO 20 072 OOO 15 402 OOO 5 549 OOO 41 022 OOO 1153 OOO 577 OOO 155 OOO 1 885 OOO 1 639 OOO 819 OOO 240 OOO 2 698 OOO 4 577 OOO 2 289 OOO 586 OOO 7 452 OOO 41.18% 28.85% 21.14% 34.27% 48.67% 35.43% 26.11% 40.90% 17.60% 11.30% 7.52% 13.86% 22.80% 14.86% 10.56% 18.17% settings where sexual decisions are made. In gener- al "knowledge", "pleasure", "rights" and "ini- tiative" belong to men, while "innocence", "ac- ceptance" and "duty" are portrayed as "normal" for women (15Jj "Pleasure and fulfilment" are reserved for a lucky few in the context of marriage. Women are expected to defer to their men, partic- ularly in sexual matters. In absolute terms and compared to men, women are more susceptible to HIV infection for biological reasons. k The shape of the vagina, larger mucosal area of women, and greater viral inoculum present in semen compared with vaginal secretions are all suggested as key factors. Younger girls are particu- A young Thai woman, having learned ahoUl HTV/ AIDS and the protection which could he provided by a condom, was asked about the likelihood of her suggesting condom use to her bo}friend. She indicated it would be very difficult for her to suggest such a thing lest he think she "had sexual experience". Kathleen Cash, commenting on this, observes that "for women there arc potentially negative social and moral consequences LO adopting AIDS prevention behaviours which may influence their sexual behaviour more significantly than the threatened health consequences of acquiring STDs or HIV" infection. (RP/ 15, p. i8). k Male-to-female transmission of HIV is 24 times more efficient than the reverse (female to male). Source: Women and ,\IDS: agenda for action, "\,\'orld Health Organization and the United Nations Development Programme in consultation with the United Nations Division for the Advancement of Women, 1994. 97 Table 3 Some basic data on women in South and South-East Asia Tableau 3 Ouelques donnees fondamentales sur les femmes en Asie du Sud et en Asie du Sud-Est Column number No de colonne 2 3 4 5 6 7 8 9 Women/ Women % women in % of birth % Literacy-% Average age Births/1000 Maternal 100 men- as% workforce- below age d'alphabetisation at first women age mortality femmes/ of adult % de femmes 20-% de F M marriage- 15·19- ratio/ 100 OOO 100 hommes work force: dans la naissances en Age moyen Naissances/ live births- 15+- population dessous de au premier 1000 femmes Taux de Bangladesh Bhutan Brunei Cambodia India Indonesia Laos Malaysia Maldives Myanmar Nepal Pakistan Philippines Singapore Sri Lanka Thailand Viet Nam 94 97 94 108 94 101 103 92 92 100 95 92 97 97 101 102 103 Femmes active en% de la population active: 15+ 41% 62% 32% 42% 33% 48% 41% 50% 24% 28% 37% 37% 45% 68% 36% 45% 22% 25% 36% 47% 32% 42% 13% 36% 31% 46% 36% 46% 27% 29% 65% 44% 69% 47% Sources: Ref.-Ref. 4 (Columns 1, 2, 3, 7. 8, 9) and Ref.-Ref. 6 (Column 4). larly vulnerable to infection because the immature cervix and lower vaginal mucus production present less of a barrier to HIV than in more mature women. Thus, very young entry into sex work and young marriage, both of which are still common in some areas, expose millions of young women to infection when they are particularly vulnerable. I As discussed above, evidence also shows an increase in the num- ber of younger girls being drawn into sexual rela- tions outside of marriage with older men. The presence of an untreated STD in either partner "multiplies the risk of HIV transmission by 300-400%''.h It is highly relevant, therefore, that women throughout Asia suffer more from untreated STDs than men (16).m The reasons are well known. I ln 4 of the 17 countries of the region the average age of first marriage is still well below 20 years (Bangladesh, 16.7; lndia, 18.7; Nepal and the Maldives, 17.9) and in many other countries, particularly in rural areas, adolescent marriage is still common. ln short, millions of young women under the age of 20 arc regularly exposed to infection. m Research across Asia among "ordinary" women attending pre-natal clinics - not sex workers, not the very poor, not women in refugee camps - again and again finds disturbing indications of STD infection. For example, a study of 1 900 blood samples from pregnant women in Jakarta, lndonesia found that 2.15% tested positive for syphilis.(&/ 16, p. 21.) 98 l'age de mariage de 15 a 19 mortalite 20ans ans maternelle/ 100 OOO naissances vivantes 20% 22% 47% 16.7 149 480 9% 25% 51% n.a. 65 1 310 n.a. n.a. n.a. 25 27 n.a. n.a. 22% 48% 21.3 108 900 11% 34% 62% 18.7 57 250 14% 62% 86% 21.2 43 450 6% n.a. n.a. n.a. 51 300 5% 70% 87% 23.5 29 20 n.a. n.a. n.a. 17.9 64 300 5% 72% 89% 22.4 32 90 14% 13% 38% 17.9 86 1 500 7% 21% 47% 21.7 64 400 17% 90% 90% 22.4 28 100 2% n.a. n.a. 26.1 11 7 8% 84% 93% 24.4 33 60 14% 90% 96% 22.7 20 20 3% 84% 92% 23.2 5 120 First of all, women with STDs are often asymptomat- ic. Secondly, although there may be symptoms, many women do not recognize them - "most women I know have the same problems ... they must be nor- mal" - and when symptoms are recognized, there is a lack ofaccessible services. Finally, because of shame and fear of stigmatization many women are reluctant to seek treatment even if they know they have a problem ( 17, 18). Economic factors: poverty and relative affluence Poverty is a direct cause of many women and girls being at high risk of HIV infection. For many women the overriding motive for sexual relations with a man upon whom she is dependent is the fear of being turned out without support. Poverty can have an immobilizing effect and cause a woman to stay in an abusive or exploitative family situation where she is frequently involved in unprotected sex. Rape, unwanted pregnancy, abandonment or widowhood may leave girls and women on their own and, in many cases, responsible for children. Women and girls trade sex for protection, an old hut now dangerous custom. In the same way, while sex work is a "last resort" for some girls, there are poor districts in most countries of Asia where there Rapp. trimest. statist. sanit. mond., 49 (1996) is some measure of acceptance of sex work as a temporary job for a young girl, and the girls go off to earn an income which supports parents, chil- dren and younger siblings. A desire for regular employment or career ad- vancement sometimes underlie a women's accep- tance of sexual harassment which may culminate in a demand for sexual relations. Millions of adoles- cent girls and young women all over Asia flock to the assembly lines of new factories, join the work force in export processing zones, clerk in opulent modem department stores, and fill increasingly well-paid junior white-collar positions in modem businesses throughout South and South-East Asia. Usually the managers who make employment deci- sions are men. Although most do not misuse their authority and many women will resist sexual ex- ploitation whatever the cost, coerced, unexpected sex does occur and this situation creates a favour- able environment for the spread of STDs/HIV in- fection. There are few, if any social, economic or legal sanctions if men force themselves on unwill- ing employees. On the other hand, there may be profound repercussions for a woman who tries to protect herself. Economic factors also promote and sustain both domestic and international migration of male and female workers. In either case, migration increases the likelihood of exposure to STDs and HIV infec- tion, for women in particular. Male migrants- fish- ermen, construction workers, skilled technicians, etc. - away from home for extended periods and with some money to spend, often enter into casual sexual relations (accepted as "normal" for a man) and one may assume that many, knowing nothing of STDs/ AIDS, make no effort to protect themselves. Thus, they bring the infection home. Women work- ers, on the other hand, are recruited by the thou- sands for international service as household ser- vants, custodial and cleaning staff, dancers, singers, etc. (19).n Some have reasonable contracts and a good experience but others are enticed or forced into sexual relations and are without information, friends or allies to protect them. There are already reports in several countries of the region of girls coming home infected. Finally, there is a rapidly growing prosperous middle class and a small number of very rich peo- ple in most countries of A,;ia. These people are able to live lives of comfort. For some there are better schools, homes and family recreation. For others, the extra money goes for alcohol, drugs n For example, in the 8 years from 1984 to 1992, 742 OOO [ndonesians went overseas for work through official channels: 75% of them were women and 93% of the women went as domestic ser:ants. The official flow of workers, however, pales by comparison with estimates of unofficial or illegal migrants. The [LO estimates the number, men and women, to be in the neighbourhood of l to 1.2 million a year going from Indonesia to Malaysia alone (RP/. 19'). Wld hlth statist. quart., 49 (1996) and promiscuous sex -in total disregard for their own personal safety and that of others. This lifes- tyle contributes to high-risk behaviour which opens the door for HIV transmission. While the numbers are comparatively small in each country, their total numbers throughout the region are sig- nificant. Equally important is the "multiplier ef- fect" of their activity, drawing other people into risky situations as the impact of their "recreation" works its way through society. In addition, this life- style becomes "glamorous" and promotes life- threatening aspirations particularly among urban young people, who want to "try it out". Unfortu- nately, as the incidence of HIV infection grows among sex workers the likelihood of infection also rises. 0 In this day and age, boys "sowing their wild oats" place their own lives and the lives of women, including their girlfriends and future wives, in jeopardy. Poor access to information Work with women and women's groups through- out the region over the past 10 years has high- lighted "poor access to information" as a crucial issue contributing to the persistence of "second class-citizenship" for women in many settings. This is particularly acute in the field of health and sexual matters. While this problem can be caused by physical isolation, equally important are the in- tellectual, social and economic isolation which are part of the tyranny of poverty and gender over the lives of hundreds of millions of women in the re- gion. In this regard, there is little difference between married and unmarried women ( 18, 20). Women have the right to information about HIV I AIDS and yet millions have never heard of the disease. In the next 12 months, how many young women will marry a "nice" young man and give no thought to the many years of casual sex he had, with the result that he brings disease to the marriage bed - "ordinary" STDs or fatal AIDS? Many prospective brides neither know nor think of this. In today's world it can be life-threatening for women to have so little information about repro- ductive health or STDs including HIV/ AIDS. Alternatives for action Millions of women and men across the region need to be reached and influenced if the impact of HIV I AIDS on women, and thus on all of society, is to be reduced. There will need to be many kinds of messages and activities to fit the reality of people's everyday lives and mobilize them to take appropri- ate, sustained, compassionate, nondiscriminatory O Narain reports that, "in Bombay, the HIV seroposili\ity rate among prostitutes increased from 2% 1988-89 to 30% in 1994," and that in Myanmar, increases in STDs and HIV have been overwhelming, STD rates among female prostitutes jumping from 8% to 13.9% in only one year (1990-91) (RP/. 2). 99 action. Many things need to be done by women themselves, much can be done by men, and even more by men and women together. Whoever does this work, four principles must guide their actions: 1. The women concerned - whatever their educa- tional level, HIV status, lifestyle or age - must have a key role in determining problems and priorities, selecting activities and evaluating their effectiveness. 2. HIV/ AIDS must be seen in the context of a woman's whole life (social, economic, legal, cultural, etc.). 3. Both men and women must participate actively if the issue of women and AIDS is to be effec- tively addressed. 4. Activities need to integrate technical and social concerns (medical, legal, economic, scientific, cultural, motivational, etc.). At least four broad categories of activity will need to be considered in all settings: 1. Education and information on STDs and HIV I AIDS. 2. Basic education and economic activity to re- duce gender inequities. 3. Change in policy and social environments. 4. Provision of gender-sensitive health and other services. Action: education and information High-profile, massive, basic HIV/AIDS public education and mobilization efforts. Enormous effort is needed to carry out basic HIV/ AIDS education on a mas- sive scale throughout the region. This will require mobilization of governments, the media, the pri- vate sector and communities, but one can begin anywhere there are people who are ready to act. Every campaign needs to include gender-specific information on the determinants of HIV infection and on the special vulnerabilities of women, the impact of HIV/ AIDS on children, and the benefit to all society in trying to help women avoid infec- tion. Specifically, the mutual responsibility of men and women needs to be emphasized and practical examples given of what can be done individually, within the family, in school, on the job, and in society, in order to achieve this objective. Messages related to women and AIDS should not be "add- on-when-you-want". Rather, they should be a stan- dard, well-developed component included in all basic AIDS training and informational activity. Some people would suggest that such mass cam- paigns are an inefficient way to reach women. In fact, for some, they may be an essential first step. They can alert women to their special vulnerability ( even within the confines of marriage) and some possible steps they can take to protect themselves or get further information. An informative cam- paign can be a trigger to start girls and women talking to each other and beginning to think how they might address this new threat in their lives and 100 it can help men be more responsible, compassion- ate partners. Audience-specific intensive education and training (a) on risks and risk avoidance and (b) to help address problems of gender bias which might affect HIV/AIDS activity with women. People to be reached might include people in health care; sex workers; control- lers/ owners of the sex industry; people with high- risk occupations or lifestyles such as sailors, truck/ bus drivers and their families; high-and mid-level business groups; international labour-recruiting agencies; adolescents and youth groups; formal and informal religious leaders; people in law en- forcement; policy makers and planners; factory owners; and leaders and workers at all levels in the hotel and tourism industry. Each group needs to be reached in their own style and with messages appropriate to their respective risk or task. Experience shows that work with health care providers is particularly important. Health care providers - women and men - must all know and always practise universal precautions if they and the patients they serve are to be protected from infec- tion. They must also be trained to overcome tradi- tional gender biases which might cause them, inten- tionally or not, to discriminate in favour of male patients, to the disadvantage of women whom they are serving. Training, retraining and refreshing are ongoing needs to assure availability and sustainabil- ity of appropriate, compassionate, non-discrimina- tory, gender-sensitive service and information. Activities with school teachers and youth workers (both peer educators and older people working with pre-adolescents and adolescents) are particularly important. Many need help learning to talk comfort- ably, clearly and appropriately about sex and sexual- ity, issues which, historically, have been taboo. How- ever, if well done, the benefits are enormous. Expe- rience in many parts of the world has shown again and again that if sex education is well done before young people become sexually active it can help reduce both premature and high-risk sexual behav- iour. Boys and girls need to be socialized in new ways: boys hearing more about sexual responsibility; girls learning they do not always have to comply with the sexual requests or demands of a man; and both learning more about mutual respect and dialogue in expression and fulfilment of their sexuality.P Action: basic education and economic activity to reduce gender inequities and increase personal autonomy As we have seen, millions of women throughout the region are highly vulnerable to HIV infection P Mboi, Nafsiah. Challenges of rrproductive health care. Remark.~ presented al Colloquium on Women and Health Security organized by the World Health Organization at the Fourth World Conference on Women, 5 September 1995, Beijing. Rapp. trimest. statist. sanit. mond., 49 (1996) Table 4 Median age for first sexual intercourse, Indonesia Tableau 4 Age moyen du premier rapport sexuel en lndonesie Education - Enseignement de type scolaire Age No education or incomplete primary - Pas d'enseignement ou ecole primaire incomplete Complete primary school - Ecole primaire complete 16.6-16.9 17.5 21.6 Some secondary education - Certaine education secondaire Source: Indonesian demographic health survey 1991, as quoted in Ref.·Ref. 19. because they are either poor or ignorant or both. An HIV "risk assessment" should be an essential part of the planning process at the micro level and should involve the women who are expected to benefit from programmes intended to reduce transmission of infection. Such an assessment will help determine which women and which environ- ments need to be reached first; and what content, skills and support systems need to be provided, both initially and in the long run, if the activity is to be effective and sustainable. To break the chain of transmission and help women, both those at risk and those who have already been infected, priority must be given to (i) improving access of women as well as adolescent and younger girls to basic education and (ii) ex- panding non-exploitative economic opportunity for women. (Many forms of activity might be ap- propriate - employment schemes, cooperatives, credit, skill training, etc. What is important is that such activities ensure increased incomes and that the woman controls the money she earns.) These two kinds of activities, individually and in combina- tion, can influence dramatically the quality of a girl or woman's day-to-day life, as well as her long-term potential to avoid coercion into high-risk sexual activity. Some years ago, when the world was first con- sumed with the challenge of lowering infant mor- tality rates, it was discovered that there was a corre- lation between a mother's education and infant mortality. There is now evidence that there may also be a correlation between a woman's education and her becoming sexually active, and thereby, perhaps, encountering her first exposure to poss- ible HIV infection. The Indonesian Demographic Health Survey of 1991, for example, found the median age for first sexual intercourse was 16.6- 16.9 years for women with no education or incom- plete primary education, 17.5 for those who had completed primary school, and 21.6 for those who had some secondary education (Tab"le 4). Further, at the time of the 1990 Census, the Singulate Mean Age at first Marriage (SMAM) was already 21 for rural women, 24 for urban women, 25 for rural men, and 27 for urban men (21). Continuation of basic education, then, is a good investment as part Wld hlth statist. quart., 49 (1996) of a comprehensive approach to reducing the threat of HIV/ AIDS to women. Action: improvements in policy and social environments Changes in policy and social environments are needed almost everywhere in the region to facili- tate and support a ·wide range of activity related to HIV/ AIDS as it affects women - those who are HIV+ themselves as well as those living with HIV I AIDS.q Priorities and approaches will have to be locally determined. The following key policy issues, however, will probably be important everywhere. The legal status of commercial sex. So long as sex work is officially "non-existent" it is almost impos- sible to have effective education programmes for sex workers and their clients. Illegality of commer- cial sex promotes its association with other illegal activities such as marketing and use of illegal drugs, trafficking in women and children, etc. This situa- tion fosters insurmountable power differences be- tween those promoting the sex industry and sex workers (female and male), making any initiative by sex workers to protect themselves virtually im- possible. Particular attention needs to be given to the review of existing policy, codes and statutes to ensure that severe penalties exist- and are applied - with regard to sexual abuse and exploitation of children, both girls and boys. What to do and how to reach the unmarried? With rising marriage age and rapid urbanization, there are increasing numbers of young, unmarried men and women between 15 and 30 years of age throughout the region.r In many countries, there is 'I Mboi, Nafsiah. H'<'llnen and AIDS in South A.,ia: some kiry issues of polir:y. Prepared for Regional Workshop on AIOS Policy in South A~ia organized by the Economic Development Institute of the World Bank in collaboration with the l'l\DP regional project on HIV/ AIDS, Kathmandu, Nepal, February 1996. r A\"ailable data for 13 of the 17 countries co\"ered in this discussion show 9 countries with singulatc mean age at mar- riage (S'.\(AM) above 21 years(&/ 4, pp. '.16-37). In Indonesia, for example, the 1990 Census found most of the adolescents and young adulL~ aged l:'>-24 were unmarried- 90.9% of the men and 73% of rhe women. (R,,f 21) Within the region, Singapore has the highest SMA'.\f: 27 years. 101 not yet a clear national consensus about how to handle reproductive health including STDs and HIV/ AIDS information (and supplies) for the un- married (particularly women) although it is clear that inequities in power and status between men and women, combined with ignorance and psycho- logical vulnerability put many young women at high risk of unwanted, unprotected, risky sex and HIV infection. Family and religion. It is important to indicate strong support for the role of family and religion in ad- dressing the challenge of HIV I AIDS. However, at the same time, policy-makers must find ways to keep the door open for many kinds of action in order to assure broad effectiveness of national mo- bilization against HIV/ AIDS. For example, grow- ing numbers of women ofreproductive age do not live in "conventional families", i.e., a family with husband, wife and children under one roof. There are women heading households with no spouse and young, urban, unmarried women, working or studying, and living on their own. There are also women and girls, in complete families, for whom dialogue on sexual matters is impossible. If dia- logue and action in the field of HIV I AIDS educa- tion and prevention is too narrowly focused on the traditional family, all these women will be missed, althoiugh there is no guarantee that their lives are without risk. Social environment. Public opinion must be pre- pared to sustain the social, economic and emotion- al support systems which will be needed to face the tragedy of HIV/ AIDS as the numbers of those af- fected by the epidemic grow and their needs be- come more diverse and more urgent. In the ab- sence of such systems, women, as society's depend- able and long-suffering "safety net", will be in- creasingly squeezed between the economic and care-giving demands of their families and commu- nities. The lesson of Africa is that under such cir- cumstances the impact will ultimately fall on the next generation, as children, because of insuffi- cient funds, drop out of school, and families, hard- pressed to buy medicine, sell their homes and land. In short, the unsupported and painful death of one generation will give birth to the renewal of poverty and ignorance in the next (22). Action: provision of health and other services Blood testing (men and women). Blood testing has two important functions in any comprehensive HIV/ AIDS programme: first, to establish the HIV status of the individual; and second, to ensure that only HIV-free blood is used in transfusions. Both are important for women. To gain the full benefit we must increase participation in voluntary blood testing by (i) strengthening understanding of its importance, (ii) providing pre-test and post-test 102 counselling, with assured confidentiality of results, and (iii) improving availability and accessibility (supplies, equipment, training for those who do the testing, provision of subsidies for those who cannot afford it). Blood testing for HIV/ AIDS in connection with transfusion must be universal and consistent. Women, who overall receive a signifi- cant proportion of all transfused blood in connec- tion with maternity and gynaecological proce- dures, will benefit Counselling services for women. Development of any STD/HIV testing programme must include devel- opment of a "user-friendly", gender-sensitive, pre- and post-test counselling system. While counselling is important in connection with blood testing, its role is far greater than that. It is essential to have accessible counselling services for those living with HIV I AIDS and their families as they struggle to restructure their lives and face issues of quality of life - daily family support, their own sexuality and uncertain future. For counselling to be effective with women, the quality of interaction between the counsellor/service provider and the woman is cru- cial. Effectiveness will be sharply limited if uncon- scious habits of gender inequity are not identified and counteracted among counsellors themselves. For "positive women" of reproductive age, one of the essential issues is exploration of the social, medical and psychological pros and cons of preg- nancy and childbirth. For women across Asia, where motherhood is looked forward to as a per- sonal pleasure, a contribution to society, and often as the ultimate route to "full status" in society, sympathetic counselling will be vital. Condom use for protection of women and men. Im- proved quality, increased availability and wider uti- lization of condoms are absolutely crucial. If the chain of infection is to be broken, boys and men must use condoms every time they have premarital or extramarital sex and with their wives if they are or may be infected. In some parts of Asia the con- dom has a considerable measure of acceptability. In other areas, however, achieving consistent use will not be easy. Men say, "What will my wife think if all of a sudden I suggest using condoms?" Women say, "My husband will be angry and think I don't trust him if I ask him to use a condom." Female sex workers consistently raise two issues related to con- doms: (i) how to get men to use them and (ii) the quality and availability of condoms. Addressing the problems of the sex workers is a good example of the need to look at the larger environment to identify the people, factors and systems which de- termine the efficacy of the individual. Unless there is help for sex workers in addressing problems of their work environment the chances are very slim that they will be successful in promoting consistent use of condoms. To increase condom use to pro- tect sex workers and the community at large, any Rapp. trimest. statist. sanit. mond., 49 (1996) programme must work with patrons of the sex in- dustry, the police, health care providers who serve people in the sex industry, managers and owners of institutions providing sexual services as well as insti- tutions (bars, nightclubs, massage parlours) which contribute to and benefit from the sex industry. STD education campaigns. Enabling women to take a more active role in their own protection requires them to know more about their own bodies and the symptoms of STD infection. This, in turn, re- quires organization of gender-sensitive, culturally acceptable programmes to reach women and pro- vide the necessary services. Experience with STD programmes indicates that women are most com- fortable if service is integrated with "regular" pro- grammes. Seeking information is not conspicuous and they can take care of various health matters at once. Whatever pattern is adopted, it is essential that women, as clients, be directly involved in on- going evaluation and reorientation of service, as needed. Effective care of STDs will not only avoid complications but it is also a fundamental element in the prevention of HIV transmission. Research to support action. Research will be needed at all levels from the international/regional level ( on issues such as international trafficking in women and children for sex work) to the commu- nity and family level. It will be needed in technical fields (for example development of female con- doms and other female-con trolled methods of pro- tection, women's response to specific therapies, etc.). There will be many crucial issues calling upon the social sciences ( exploration of motiva- tion for young people towards or away from high- risk behaviour, etc.). To gain maximum benefit from the national and international research ef- fort, attention must be given to networking and information exchange to minimize duplication of effort and wasted time, energy and resources. Partners and networking: who will do what needs to be done? Women can do wonderful things individually and in groups, but they do not make up the world alone and they cannot change it alone. Partnerships will be an essential key to reducing women's vulnera- bility to HIV I AIDS - partnerships between women and men, among institutions and nations - to help create environments which encourage mainte- nance of safe sexual behaviour, change high-risk behaviours, and ensure availability of appropriate service to people living with AIDS. General community groups of all sorts have the potential to play particularly important roles re- gardless of whether they have formal structure and status or are informal coalitions of concerned peo- ple. They can reach people who may not be reached by government programmes but who are Wld hlth statist. quart., 49 (1996) at high risk of infection - for example sex workers, street children, urban squatters and street dwellers. They can be a vital "bridge" between government (or large donors) and community people who need technical assistance or resources. They may take the lead in developing audience-specific mate- rials or programmes which will carry the message of prevention, compassion and action tailored to a particular lifestyle, a particular setting, or a particu- lar philosophy. Women's organizations/groups of all sorts (middle-class housewives, sex workers, religious youth groups, women farmers, etc.) will be impor- tant not only as a channel to make information and service available to women but also in helping identify and develop strategies and approaches which will make it possible for women to discuss sex, sexuality and sexual behaviour with their hus- bands or sexual partners. Resolving the issues of power and gender relations will be a long and complicated process. Experience in Asia and else- where suggests that women working in groups of their own making and on issues of their own choice have an important contribution to make in the process of empowerment of women. Women who experience efficacy gain new images of who wom- en are, their potential, and their place in society. They are empowered by the experience of solving their own problems (15, 18). Collaboration between groups of women work- ers and employers can be to the mutual advantage of both. For the employer a healthy, experienced and productive work force is more efficient and less expensive than a sickly and constantly chang- ing crew. For the first time, the women, working together, may begin to be able to deal effectively with issues of sexual harassment and personal safe- ty on the job. Inequities of power and decision- making between men and women are reduced and the workplace is more productive and safer. In time, organizations of women who are HIV positive ("positive women") will be crucial in the overall effort to deal in humane and effective ways with the impact of AIDS on women's lives. These organizations will be vital sources of information and non:iudgemental support for the women, themselves, their family and friends. They can also play an essential advocacy role identifying and ar- ticulating the special policy and programme needs of positive women. Governments, with their resources, authority and influence will obviously be key players. Official policy will be crucial in helping establish environ- ments which facilitate and encourage the activities of community groups specifically those working with women and working on gender-related issues with both men and women. Leadership by govern- ment will also be essential. Working harmoniously with community groups, standards of service can be developed and maintained which ensure equity, 103 confidentiality and high technical quality in test- ing, counselling, care coordination, delivery of ser- vice and other activities related to HIV I AIDS. Other institutions and the individuals within them must also be "recruited to the cause", in- cluding the print and electronic media, research institutions of all sorts, and a whole spectrum of national and international foundations and donor institutions. Conclusion Hundreds of thousands of women and children across South and South-East Asia are already in- fected with HIV and the numbers are rising. They will die prematurely and there is nothing we can do to stop it In our region, for social and economic reasons, there is also little we can do to mitigate the painful process of decline and ultimate death. What we can do is try to see that positive women are able to continue their productive lives as long as possible with optimal quality of life. They have the right to live free from stigmatization and isolation and we must make common cause with them to assure them of that right. For positive women, these are the core issues of women and AIDS - the right to be part of the family and community, the right to be productive. Another several hundred million women are at risk of infection. There is a great deal that can and must be done to reduce their vulnerability. Men and women, acting together, can make a differ- ence and save lives in this generation and the next. Inspired by our sisters for whom we are too late, we must join hands throughout the region and act. We must take up issues of public health and human rights, poverty alleviation and gender relations. New social environments must be created in which there is greater equity, women are empowered, and are thus less vulnerable to infection. Policy and master plans will be needed but, in the end, it is individual acts in the family, the neighbourhood and the workplace which will reduce and eliminate risks and make life bearable for those who are affected and their families. Summary South and South-East Asia are at the centre of the most aggressive advances of the AIDS epidemic today. The challenge this presents to the region is clear. While reported absolute numbers still lag behind the African region (11 160 900 in Africa; 3 081 235 in Asia) knowl- edgeable observers agree that the pace of infection and potential devastation in this region exceed what we have seen in Africa. Those concerned with the welfare of the people of Asia, therefore, must make serious efforts to break the chain of HIV transmission as quickly and effectively as possible and identify and care for the infected. Women are entitled to protection by right- the same as men. However, for anatomical reasons, women 104 are more vulnerable than men to infection by HIV. In addition, throughout the Asian region, women's "natu- ral" vulnerability is vastly magnified by poverty and generally low levels of education and personal autono- my which make it difficult for them to gain access to information and appropriate services. Because of women's multiple roles in the epidemic - potential "infectee", care-giver, transmitter of infection - if we are to be successful in halting the spread of HIV/AIDS we must give particular attention to reaching, working with, and serving women. Meeting this challenge requires involvement of men as well as women, individuals and institutions, governments and NGOs, in four broad ar- eas of activity: (i) HIV/AIDS education and information; (ii) basic education and economic activity to reduce gender inequities; (iii) improvements in policy and social environments; and (iv) provision of health and other services. Lack of commitment, skill, or persistence in meeting the challenge will cost lives across Asia. Resume Les femmes et le SIDA en Asiedu Sud et du Sud-Est C'est en Asiedu Sud et en Asiedu Sud-Est que l'on note la progression la plus rapide concernant l'epidemie de SIDA aujourd'hui. Le defi que cela represente pour la Region est clair. Si les chiffres absolus notifies demeu- rent inferieurs a ceux de la Region africaine ( 11 160 900 en Afrique; 3 081 235 en Asie), des observateurs com- petents admettent que le rythme de l'infection et les ravages qui menacent cette region depassent ce que nous avons vu en Afrique. Par consequent. ceux qui s'interessent au bien-etre de la population asiatique doivent tout tenter pour briser la cha1ne de la transmis- sion du VIH aussi rapidement et efficacement que possible et identifier et soigner les personnes infectees. Les femmes doivent avoir le droit d'etre protegees, tout autant que les hommes. Toutefois, pour des raisons anatomiques, les femmes sont plus vulnerables que les hommes a !'infection a VIH. Par ailleurs, dans !'ensemble de la Region asiatique, la vulnerabilite «naturelle» des femmes se trouve considerablement amplifiee par la pauvrete et des niveaux generalement faibles d'educa- tion et d'autonomie personnelle qui les empechent le plus souvent d'avoir acces a !'information et aux servi- ces appropries. Du fait des multiples roles qu'elles jouent dans l'epidemie - personne infectee potentielle, dispensatrice de soins, vectrice de l'infection - si nous voulons reussir a stopper la propagation du VIH/SIDA, nous devons nous efforcer plus particulierement d'at- teindre les femmes, de leur offrir des services et de travailler avec elles. Relever ce defi exige la participa- tion des hommes comme des femmes, des individus et des institutions, des gouvernements et des ONG, dans quatre grands domaines d'activite 1) education et infor- mation en matiere de VIH/SIDA; 2) education fondamen- tale et activite economique afin de reduire les inequites entre les sexes; 3) amelioration de la situation politique et sociale; et 4) fourniture de services de sante et autres. Dans !'ensemble de l'Asie, !'absence d'engagement, de competences ou de persistance dans la poursuite de la tache entreprise coOtera cher en vies. Rapp. trimest. statist. sanit. mond., 49 (1996) References/References I. Brown, Tim & Xenos, Peter. AIDS in Asia: tlu! gathering stonn. Analysis from the East-West Centre, No. 16, Honolulu, August 1994 2. HIV/AIDS in S&uth-Emt Asia: an update, ~ew Delhi, WHO- SEARO, 7 January 1995. 3. Global AIDS Policy Coalition. Status and trends of the HIV/ AIDS pandemir as of January, 1995, Cambridge, Harvard School of Public Health, 1995. 4. UN Department for Economic and Social Information and Policy Analysis, Statistical Di\'ision. The worl.d's women 1995 trends and statistics. Social statistics and indirators, series K, No. 12. (Publication no. ST/ESA/STAT/SER.K/12) New York, United Nations, 1995. 5. The .,Late of world population: I 995, New York, United Nations Population Fund (UNFPA), 1995. 6. World Bank. World devewpment report I 995 - workers in an integrating world, New York, Oxford University Press, 1995. 7. Human deuewpment report: 1994, New York, United Nations Development Programme (lJNDP). 8. Strategy for .sustairwd reduction of poverty, Washington, World Bank, 1990. 9. HIV/AIDS data. Geneva, UNAIDS, I February 1996. 10. Ireland, Kevin. v\·hh you werm 'there. Tlw sexual explnitation of children and the rnnnertion with t&urism and international travel, Working Paper No. 7. London, Save the Children, September 1993. 11. Life after the streets: I O Janner street d1ildren tell tlll'ir stories, Manila, Childhope Asia Philippines, 1995. 12. International Catholic Child Bureau. Chil.dren worldwide - Spi,cial isme on sexual exploitation of rhil.dren, 19(2) ( 1992). Wld hlth statist. quart., 49 (1996) 13. Muntarhhorn, Vi tit. Rights of the child: sak of children. Report of the SpecialR.app&rteurapj,ointed in accordance with Commission of Human Rights resolution 1990/68-E/CN/!992/55. (Cited in &f 10) 14. Global AIDS Policy Coalition. Status and trends of the HIV/ ,-\IDS pandemic as of January 1996, Cambridge, Harvard School of Public Health, l 9Y6. 15. Cash, Kathleen et al. E~perimental edurational interventions/or Aff)S fnevention among ,Vorthern Thai single migratory Jat:tory worlren, Women and AIDS Research Program, Research Report Series, No. 9. 'Washington DC, International Center for Research on Women, AugusL 1995. 16. Dharmaputra, Nick G. et al. Situation rma(nis on l/lV/AIDS and its impact on rhiulren, women and Jami lie.,· in Indonesia. R,j,ort fmpared for !he United Sations Childrm 's Fund, .Jakarta, UNICEF, November 1995. 17. WHOGlobalProgrammeonAIDS. TlwHIV/Aff)SJ'and£mic. 1994 Overview, Geneva, WHO, 1994. 18. Gupta, GeetaRao & Weiss, Ellen. Wom,mondAWS: devrloping a 111•w health .,'lrate{!J, \\'ashington DC, International Center for Research on Women, 1993. 19. International Labour Organisation. A comprehensive women's emplm1nent strategy for Indonesia: draft report of an ILO/UNDP TSSI mission., Bangkok, 11.0 Regional Office for Asia and the Pacific. May 1993. 20. Setiadi, Bernadette etal. Women anrl,l!DS in Indonesia: a repmt on qualitative rPsearch m,uiurted among 11wnied wo11wn in Jakmta, hulmu•sia, \-\'ashington DC, AJDSCOM, 1992. 2 I. Central Bureau of Statistics. lrulon,sirm population, Pnpuullion cen.ms 1990, S-Series,Jakana. CBS. 22. Black, Maggie. Ri•port on a mretiug about AIDS and orphan, in Afrim,. New York, UNICEF, I 991. 105 Sexual transmission of HIV to women in industrialized countries Catherine Hankinsa From being virtually absent from the human im- munodeficiency virus (HIV) epidemic in the 1980s, women living with HIV now number close to 7 million. In 1995 alone, one million women be- came infected with HIV - in fact, close to half of all newly-infected adults in the world today are women. Each day 3 OOO additional women are in- fectedb and by the year 2000, over 13 million women ·will have been infected and 4 million of them will have died. c Why is it that a virus which infects both men and women is affecting women in an increasingly dis- proportionate manner? Sex-related biological dif- ferences in susceptibility to HIV acquisition are important but they are greatly overshadowed by the roles played by the sexual and economic subor- dination of women. Societal forces and gender- based power inequalities create the sexual subor- dination that places women at risk. Socialized concepts of masculinity and femininity as well as gender and power relations limit the capacity of many women to negotiate the boundaries of sexual encounters so as to ensure both their safety and their satisfaction ( 1). This is true to varying degrees in all societies. But poverty in any setting further interacts with gender imbalances to prevent wom- en from protecting themselves against HIV. Lack of economic and educational opportunities may not only lead women to resort to entering into sexual relationships for economic reasons but may also reduce both their willingness and their capaci- ty to negotiate protection. This article will provide a brief overview of the roles played by biological vulnerability, social and economic subordination, and social contexts of HIV transmission in industrialized countries as well as the potential role to be played by woman-con- trolled prevention methods in the HIV epidemic. a Infectious Disease Unit, Montreal Regional Public Health Department and Department of Epidemiology and Biostatistics and the McGill AIDS Centre, McGill University. E- mail: md77@musica.mcgill.ca h Piot, P. The HIV I AIDS pandemic: global spread and response. Presented at the opening ceremony of the Third International Conference on AIDS in Asia and the Pacific/the fifth national AIDS seminar in Thailand. Chiang Mai (1995). c Women and AIDS: agmda for action. World Health Organization and the United Nations Development Programme in Consultation with the United Nations Division for the Advancement of Women. (1994) 106 Biological vulnerability to HIV in women HIV is essentially a sexually transmitted disease (STD) and it is now accepted that, as with other STDs, women are biologically more vulnerable than men to HIV infection. Studies in different settings suggest that male-to-female transmission of HIV appears to be 2 to 4 times more efficient than female-to-male transmission.c,d,e,f This is thought to be due to the larger mucosa! surface area ex- posed to the virus in women and the greater viral inoculum present in semen as compared with vagi- nal secretions. As well, while normal acidic vaginal secretions may be relatively virucidal to HIV, alka- line semen has a buffering capacity which can act to prolong HIV survival in the vagina (2). Young girls are particularly susceptible to HIV as a result of the lack of maturation of the cervix and their relatively low vaginal mucus production which pre- sents less of a barrier to HIV. Among biological co- factors which may increase a woman's risk of being infected by her male partner are sexually transmit- ted diseases, anal intercourse practices, viral load and stage of infection, and cervical ectopy. Sexually transmitted diseases (STDs) Without doubt the most important and potentially preventable biological risk factors associated with HIV infection in women are STDs. Although the association between the history of an STD and HIV acquisition was originally demonstrated for ulcer- ative STDs such as chancroid (3), syphilis (4, 5), and herpes (6-8), nonulcerative diseases such as gonor- rhoea, chlamydial infection, and trichomoniasis (9) have now been implicated. Overall an estimated 3- to 5-fold increased risk of HIV acquisition is associated with either an ulcerative or a discharge STD (10). This association between STDs and risk of HIV acquisition underscores the importance of <l Fisher, D.G. et al. Meta-analysis of heterosexual HIV transmission. [Abstract PC-Cll-2831]. IXth international conference on AlDS/IVth STD world congress, Berlin. ( 1993). c Haverkos, H.W. & Battjes, R. Pooled analysis of HIV transmission among heterosexual couples. [Abstract PoC 4157]. VI!Ith international conference on AIDS/I!Ird STD world congress, Amsterdam (1992) f Fontanet, A. & Piot, P. State of our knowledge: the epidemiologyofH[V/ AIDS. Presented at the seminar on AIDS impact and prevention in the developing world: the contribution of demography and social science. IUSSP Working Group on AIDS, Annecy, France (1993). Rapp. trimest. statist. sanit. mond., 49 (1996) early diagnosis and treatment of sexually-acquired genital infections in women through services of- fered in a variety of settings including community clinics, school- and university-based clinics, family- planning clinics, and maternal and child health clinics. Anal intercourse The practice of anal intercourse carries at least double the risk for women compared with vaginal intercourse (11) with one review of several studies indicating that the risk may be situated somewhere between 2- and &-fold ( 12). This is likely due in part to the potential for anal tears or rectal abrasions during anal intercourse which can facilitate entry of HIV and in part as a result of the poor protection provided by the columnar epithelium which con- stitutes the rectal lining. The prevalence of anal practices in heterosexuals is not well known but in industrialized societies from 20 to 25% of women report having experienced anal intercourse at least once in their lifetimeg (13, 14) and from 8 to 10% report anal intercourse as a regular sexual activity (15, 16). Contraception, preservation of vaginal vir- ginity, menstruation, partner request, personal pleasure, sexual assault and other factors underlie this practice. Viral load and stage of infection Although women are at risk of HIV acquisition even when male sexual partners have a low viral load ( 17), both the early, viraemic stage of HIV infection and the clinically-advanced HIV disease state, when immunity wanes and CD4+ counts de- cline in the male partner, have been associated with increased likelihood that men will infect their female partners.h However, there is considerable variation in viral load between individuals regard- less of the stage of HIV disease ( 18). As well, diver- sity in the biological properties of viral quasispecies could influence the likelihood of transmission ( 19) with some male partners being particularly effi- cient transmitters of HIV infection. Cervical ectopy The presence of cervical ectopy likely increases the risk of acquiring HIV infection (20) in the same manner that it has been shown to increase suscep- tibility to chlamydial infection (21-23). Cervical ectopy refers to an outgrowth of the columnar epithelium of the endocervix towards the vagina g Colon, H. et al. Hrv status and risk behaviours among women sexual parmers of [VDUs in Puerto Rico. [Abstract 3008]. Vlth international conference on AIDS, San Francisco (1990). h Goedert, JJ. et al. Heterosexual transmission of human immunodeficiency virus (HIV): Association with severe T4-cell depletion in male haemophiliacs. [Abstract 2.6]. lHrd international conference on AIDS, Washington ( 1987). Wld hlth statist. quart., 49 (1996) which creates a shift in the squamocolumnar junc- tion of the cervix. Columnar epithelium is more susceptible to some sexually transmitted diseases than is the more protective squamous epithelium of the vagina. Cervical ectopy is a normal finding in adolescent girls and among young women who have never been pregnant. Although an increased prevalence of cervical ectopy has been document- ed among oral contraceptive pill users, there is as yet no clear consensus concerning increased risk of HIV acquisitioni (3, 24) or lack thereof (25-27) related to oral contraceptive use (28). Social and economic subordination In virtually every society, women face discrimination in employment and social status, which may create economic vulnerability to HIV. This includes occu- pational segregation of women into low-paying cler- ical and service jobs, unequal pay and fewer promo- tions compared to men, fewer workplace benefits, and the concentration of women in the informal sector. Equal pay for equal work and recognition of the value of child care remain goals to be striven for in virtually all countries. Households headed by women - and estimates indicate that women are the sole earners in one- fourth to one-third of all the world's households (29) - are more likely to be financially poor than those in which there is a working resident male (30). Women's economic dependence on male partners in order to avoid poverty for themselves and for their children may make it difficult for them to negotiate safer sex practices to protect themselves from infection. Laws regarding mar- riage, divorce, child custody and child support in some countries can impede women's ability to leave relationships in which they or their children are physically or sexually abused or exposed to the risk of HIV infection. The proportion of women who engage in sex work is often directly related to the economy and the level of unemployment. In most parts of the world prostitution is illegal and underground, which means that prostitutes often lack adequate control over their working conditions. Coerced sex such as rape, sexual abuse and other non-consensual penetrative sex in and out- side the family can carry an increased risk of trans- mission of HIV and other STDs, in part due to the absence of lubrication and, but more importantly, because men who rape are not likely to use con- doms. The problems associated with rape and other forms of violence against women are often intensified in war situations, in which occupying or i Cates, W. HlVNET International cohorts: past accom- plishment~, present status, and future potential. Conference on advances in AIDS vaccine development. VIIIth annual meeting of the national cooperative vaccine development groups, Bethesda, Maryland, February ll-15, 1996. 107 invading armies may systematically rape women as part of a strategy to intimidate the local popula- tion. Such behaviour was widely reported in Bosnia in the 1990s. Rural-urban and international migra- tion resulting from unstable economic or environ- mental conditions increase the risk of HIV trans- mission to women by disrupting the normal mech- anisms of social control over male sexuality and by constraining women to engage in sexual barter to obtain entry or residence permits, in exchange for transport, or to obtain or hold onto jobs. Unequal access to education tor young girls Diminished educational opportunities for young girls related to differential social expectations for men and women and the tendency for girls to leave school for a variety of reasons encourage early part- nership formation and early sexual activity. Contin- uation of schooling appears to have a retardant effect on the onset of sexual activity in young wom- en, helping girls move beyond the heightened bio- logical vulnerability of adolescence and increasing the chances that healthy sexual choices will be made. In general, better-educated women marry later and start their families later. Improving the access of girls and women to formal education, not only helps to equalize the age of partnership for- mation which can reduce the risk of HIV, but also increases women's competitiveness in urban econ- omies which could positively affect the unequal gender-mix seen in some cities (31). The World devewpment report 1993 argues that lowering barri- ers to schooling for girls can be done through scholarships, by offering free textbooks or fee ex- emptions with safeguards to prevent diversions to males, and by siting schools close to people's homes so that parents are less worried about their daughters' safety (32). Education increases the chance that women will make good use of health services, increases their access to income, and en- ables them to make healthier choices. Investing in women through improved education is not simply a desirable end in itself - it is a key to higher productivity and growth for national economies and to reduced heterosexual and perinatal HIV transmission. In social contexts where control over sexual information and decision-making is not equally shared between the sexes, lack of access to sexuality education may compound cultural prohibitions against discussing sexual matters. Comprehensive, non-judgemental provision of public health infor- mation about sexual activity and it'i possible conse- quences encourages responsible, health-positive choices. An exhaustive review commissioned by the World Health Organization found no support for the contention that sexuality education encour- ages sexual experimentation or increased sexual activity. When effects are observed, they are toward postponed initiation of sexual intercourse and/ or 108 effective use of contraceptivesj Unplanned preg- nancies, sexually transmitted infections that can lead to infertility, and HIV acquisition are among the negative consequences for young girls who en- gage in early sexual activity in the absence of sexu- ality education. Economic opportunities for women Women currently account for half of the world's population and two-thirds of the hours worked. They receive one-tenth of the world's income and have one one-hundredth of the world's property registered in their name (29). Women face discrim- ination in employment and in economic opportu- nities in virtually every society. The result is the feminization of poverty and women's economic subordination to men. Economic crises and deficit control policies have tended to exacerbate poverty, particularly among single-parent families headed by women. As the economic crisis deepens, some women are finding it increasingly difficult to pro- vide for themselves and their children without en- gaging in unsafe sexual activities to help meet im- mediate needs for financial and other support. The lower status of women fuels the HIV epidemic with interactions between low literacy and educa- tion levels, disadvantaged economic status, and vul- nerability to HIV infection mediated by gender power imbalance and the lack of social autonomy for women. Social contexts of sexual transmission of HIV to women It is now generally accepted that the majority of women living with HIV infection in the world have acquired HIV heterosexually from their reguiar partners. For this reason any discussion of risk be- haviours must necessarily include a review of the behaviours of both men and women which place women at risk for HIV. The following sections dis- cuss the profiles of multi-partnered individuals, the roles of male bisexuality and of lesbian and bi- sexual behaviour, specific risks associated with sex work, and the influence of drug and alcohol use. Multiple partners and HIV risk In industrialized countries, some men and women have several sexual partners sequentially, a behav- iour often referred to as serial monogamy. Having only one partner at a time, such indi,.,iduals may not perceive themselves to be multi-partnered and may underestimate the risk associated with sexual activities. For others, extra-marital sexual activity with one or more partners, or pre-marital sexual i Grunseit, A. & Kippax, S. Efkcts of sex education on young people's sexual behaviour. U npublishcd rc\'iew commissioned by the Global Programme on AIDS, World Health Organization (1993). Rapp. trimest. statist. sanit. mond., 49 (1996) activity with two or more partners over a period of time, may constitute the multi-partnered state. Re- cent research suggests that concurrence of sexual partners, i.e., where two or more sexual relation- ships overlap in time, is likely to be associated with higher rates of HIV transmission than sequential but non-overlapping relationships. k In industrialized societies, the reasons underly- ing multi-partner behaviour in men and women can vary. In a study of 18 876 adulL'i broadly repre- sentative of the age, marilal slatus, and social class of the population of England, Scotland and Wales, the probability of reporting two or more partners in the past year was significantly increased among the unmarried, those who reported first sexual intercourse before the age of 16, and those in the two highest social classes (33). In the United States of America, a nationwide study of 2 559 heterosex- uals revealed that 5.8% of women and 11.6% of men aged 18 to 75 years reported having more than one sexual partner in the previous year (34). Multiple sexual relationships were strongly associ- ated in multivariate analysis with gender, age and marilal status: males, young adults or unmarried people were more likely to have multiple partners. In another study from the same survey 3 482 wom- en aged 18-49 living in 23 urban areas in the Unit- ed Slates were interviewed for information con- cerning sexual risk (35). Overall, 15% of women had engaged in sexual behaviour in the previous 12 months that might expose them to HIV. Predic- tors for having multiple partners and for having a main partner at high risk for HIV were found to be distinctly different and influenced by ethnicity. Single women were more likely than married wom- en to have multiple partners, regardless of race. Factors associated with having multiple partners were age under 30 years for Black and Hispanic women, and having more than 12 years of educa- tion for White women. Having a main sexual part- ner at high risk for HIV was associated with being married and having more than 12 years of educa- tion for Hispanic women, and having less than 12 years of education for White women. Knowl- edge of the social and demographic characteristics of women who are more likely to have a high-risk main partner can assist in the design of prevention programmes aimed at reducing risk for these women specifically. Male bisexuality Women who are sexual partners of bisexual men are at risk of HIV infection because of relatively high HIV prevalence levels among bisexual men in k Morris, M. et al. ConcurrcnL partnerships and Hrv transmission in Rakai District, Uganda. [Abstract TuC105]. Presented at the IXLh international conference on AIDS and STD in Africa, Kampala, Uganda (1995). Wld hlth statist. quart., 49 (1996) industrialized countries and practices of unpro- tected anal and vaginal sexual activity.I Although bisexuality may refer to self-identity, desire for or intention to have sex with members of both gen- ders, and/ or a lifestyle choice, behaviouri.al bisexual- ity, defined solely in terms of sexual behaviour within a given time frame (lifetime experience, last 5 years, previous 12 months, etc.), is the most use- ful definition of bisexuality for assessing risk of HIV transmission to women. The prevalence of men who have ever had part- ners of both sexes has been estimated in large scale surveys at 1.5% in England and Wales (36), 2.9% in Norway (37), and 3.2% in France (38). These find- ings contrast with those from an Australian survey using a geographically stratified sample which re- vealed that 6.4% of previously-married men and 4.2% of currently-married men had had a male partner in the previous year (39). Among 65 389 American men who reported having sex with men since 1977, 26% also reported having sex with women (40). In Mediterranean countries where homosexuality is highly stigmatized, bisexual be- haviour may be relatively more common among heterosexual men than it is in North America where bisexuality appears to be more common among homosexual men. I In a study of 467 men in the United Slates who had been behaviourially bi- sexual in the previous 3 years, 71 % reported hav- ing at least 1 female sexual partner in the previous 6 months. Of these men, 54% said none of the women were aware of their homosexual activity.m Of 91 men who had had unprotected sex with a steady female partner, 22% also reported unpro- tected anal sex with a male partner during the same time frame. In an Australian study, 58% of bisexual men revealed that all of their female part- ners knew that they had male partners and a fur- ther 10% said that some knew (41). The percent- age of women who are aware of having a bisexual partner during their lifetime varied from 19% among prostitutes in Sydney, Australia (42), 11.2% among STD patients in Montreal, Canada (16), to 4.2% and 1.2% among STD clinic attendees and family planning patients, respectively, in Harris- burg, Pennsylvania (43). Among 3 555 women with heterosexually-ac- quired AIDS in the United Slates, 11 % reported sexual contact with a bisexual man and no other risk factor. In 1989, the rate of AIDS in American women associated with sex with a bisexual man (and no other partner at risk) was 3 and 5 times I Boulton, M. & Weatherbum, P. Literature reviewofbisexualiLy and HIV transmssion. Report commissioned by the Social and Behavioural Resarch Unit, Global Programme on AIDS, World Health Organization. (June 1990). mStokes,J. et al. Female sexual partners of bisexual men: what Lhey don't know might hurt Lhcm. [AbstracL WS-D08-3]. IXLh intcrnaLional conference on AIDS, Berlin (1993). 109 higher among Hispanic and African-American women, respectively, than among White women (40). Lesbian and bisexual behaviour among women The risk of female-to-female HIV transmission through sexual behaviour has not been quantified and it has been argued that the lack of data on lesbians excludes them from access to resources and renders them essentially invisible in the HIV epidemic ( 44). There are report" of possible female-to-female transmission (45, 46); however, an analysis of AIDS cases in women who were ex- clusively lesbian found all cases to have had other risk factors such as unscreened blood transfusion or injection drug use (47). A study of incarcerated, injecting-drug-using women found that lesbian women were more likely to share needles than were heterosexual women, thus placing themselves at a higher risk of HIV acquisition (48). Conse- quently, although female-to-female sexual trans- mission is extremely rare, HIV acquisition through needle-sharing with female partners remains a risk. High-risk heterosexual behaviour was found in a study of 1 086 self-identified lesbian and bisexual women, with 13% of lesbian women and 42% of bisexual women having had sex with a man who they knew or believed to be gay or bisexual and 9% and 15% respectively having had sex with someone who used injection drugs (49). In another survey of 483 women of whom 77% self-identified as lesbian 19% as bisexual, and 3% as other or no sexuai orientation, 34% had had sex with a male in the previous 3 years (21 % of lesbians; 82% of bisexual women). Consistent partner condom use rates among lesbian and bisexual women who have sex with men were 52% and 44% respectively." Women who self-identify as lesbian may be behav- iourally bisexual, placing themselves at risk of HIV acquisition through heterosexual activity. Sex work Differences in the organization of work and in working conditions for various types of prostitutes can influence risk while working (50). Just as eco- nomic issues may play a major role in entry into the sex trade, so do they influence the ability of sex workers to refuse unsafe sex. Offers of financial reward for unprotected sex are reported by female prostitutes around the world in varying contexts. In industrialized societies, the temptation to accept financial inducements for unsafe sex appears to be greater for women with a drug-injecting habit. In fact, the pressure to earn money quickly can be further exacerbated when women must finance n Garcia, D. et al. Lesbian and bisexual women: indications of high-risk behaviours with men. [WS-C07-6]. IXth international conference on AIDS/IVth STD world congress, Berlin ( 1993). 110 their own and a partner's habit as was the case, for example, in Glasgow, Scotland where half of the sex workers studied were working to support dual habits (51). Furthermore, the sale of sexual services is not conducive to foreplay since clients are con- cerned primarily with their own pleasure. As a re- sult, sex workers are frequently obliged to engage in relatively dry vaginal intercourse, increasing bio- logical susceptibility for acquisition of HIV (52) and other STDs. Striking declines in the frequency of reliance by men on sex workers for initial sexual experiences have been documented in countries such as France where the percentage of French men indicating that their first sexual experience was with a prosti- tute declined from 21 % to 0% between the 1922- 1925 and 1972-1973 birth cohorts. In 1992, 16% of all professional and managerial men age 40 and over at the time of interview versus 5% of blue- collar workers of similar age reported having had their first sexual experience with a prostitute (53). HIV infection among female sex workers in industrialized countries is primarily associated with injection drug use (54). The link between sex work and drug use by women varies by setting. For exam- ple, over 50% of street prostitutes, who comprise only about 20% of all sex workers in the United States, inject drugs (55). Other prostitutes who work in brothels, outcall and escort services, bars and cafes, and massage parlours are less likely to be involved in injection drug use (56). In a multi- centre study of 1396 female prostitutes in the U:nited States, injecting-drug use was the best pre- dictor of HIV seropositivity (57). Sex work to fi- nance a drug habit and the use of drugs and alco- hol in association with sexual activity with clients are often accompanied by lower condom utiliza- tion rates and higher risk of HIV acquisition. In the context of sex work, condom use is high when condoms are available and social conven- tions reinforce their use. Working conditions and client expectations clearly influence condom utili- zation. In general, condom use rates are relatively higher for clients than for steady or regular partners.O·P Paradoxically, the increasing use of condoms in the sex trade has made promotion of condoms in the non-commercial relationships of sex workers more difficult (51, 58). In-depth interviews with prostitutes reveal barriers to con- dom use which are similar to those of other hetero- sexual women. In addition, female prostitutes make a clear distinction between their work con- O Alary, M. & Worm, A. M. Risk behaviours of female prostitutes from Copenhagen. [Abstract PO 009 3647]. !Xth inter- national conference on AIDS/IVth STD worid congress, Berlin (1993). P Lamothe, F. et al. Behaviours of male and female intra\·enous dru? user; (IVDU's) involved in prostimtion in Montreal, Quebec, Canada. [Abstract PO 009 3648]. IXth international conference on AIDS/IVth STD world congress, Berlin ( 1993). Rapp. trimest. statist. sanit. mond., 49 (1996) text, which is closely associated with condom use, and the intimate context with a private partner, which by their definition excludes condoms.q Mainstream gender norms and stereotypes thus influence sex workers to dichotomize their behav- iours between public and private worlds with the result that, for them, romantic sex, not commercial sex, is unsafe (59). Women as drug users Female drug users are vulnerable to HN either directly through injection practices or indirectly through sexual behaviours associated with drug and alcohol use (60-63). Differences between male and female injecting drug users in the rates of new HN infections sug- gest that heterosexual exposure may be contribut- ing significantly to HN spread to female injecting drug users (64). Compared to male injectors, female injectors have been described as having higher risk sexual behaviour, reporting more sexu- al contact with injectors and more unprotected receptive anal intercourse.r In fact, while male in- jecting drug users tend to be in sexual relation- ships with non-injecting women, women injecting drug users are more likely to report that the major- ity of their sex partners are other iajectors (65). Of 64 822 cumulative AIDS cases among women in the United States reported to June 1995, 47% were injecting drug users and an additional 18% were non-injecting partners of male iajecting drug users (66). Crack cocaine and alcohol Crack cocaine use in North America, and to a lesser extent in Europe, has been associated with high-risk sexual behaviours and is known to in- crease the risk of acquiring HN (67) and other SIDs (68). Female crack users generally report more high-risk sexual behaviours than do men (69). As seen for injecting drug users, female crack users are significantly more likely than males to trade sex for drugs and/ or money (70). Women may support a crack cocaine habit in a variety of ways including obtaining money or bonuses from selling drugs, exchanging stolen goods or other drugs, or as payment for sex (71). Although alcoholic women are at increased risk for HN acquisition associated with impaired judgement and unsafe sexual practices (60), the use of moderate amounts of alcohol by non-alco- q Gates, S. et al. Factors mitigating against partner condom use among women sex workers visiting Momreal's needle exchange. Fourth annual conference, Canadian Association for HP/Research, Toromo (1994). r Anderson, R. et al. Differences in slated behaviour of female and male IVDUs. [Abstract 3001]. VIth international conference on AIDS, San Francisco (1990). Wld hlth statist. quart., 49 (1996) holic women may also increase risk as a result of the disinhibiting effect of alcohol combined with lowered motivation to use condoms. Sexual transmission of HIV: moving toward woman-controlled prevention methods Male condom utilization rates are influenced by many factors including weak social norms, costs, previous experience, negative attitudes and poor accessibility. Prevailing attitudes towards condom use among men and women are negative in many cultural contexts despite expressed desires for self- protection. More knowledge is required of real and perceived barriers to partner condom use in comparison with obstacles associated with alterna- tive woman-controlled methods for HN preven- tion. Attempted negotiation concerning partner condom use can destabilize sexual power roles, potentially contributing to emotional coercion and physical abuse. Strong social norms clearly limit the ability of women, particularly monoga- mous women, to request partner condom use and also act to encourage their acceptance of male refusal. In the face of these constraints, consider- able effort is being directed at the development and evaluation of woman-controlled prevention methods. The female condom has been the subject of several acceptability studies which have shown high levels of satisfaction (72-74). The efficacy of the female condom in preventing HN transmission has not been measured and the effects of exposing the female condom to oil-based lubricants, spermi- cides and detergents are not well known. The im- pact of reuse of the female condom, which was developed for single usage, on its durability, non- infectiousness, and protective efficacy requires fur- ther exploration. Cost remains the most important barrier to its widespread use and acts as a strong incentive for reuse. There is an urgent need for negotiation of a public-sector price based on large volume purchases. Research concerning spermicides such as non- oxynol-9, benzalkonium chloride, chlorhexidine and mefengol indicates good anti-HN activity in the laboratory. Studies have suggested a protective effect against STDs (75, 76) or HN (77) for women who consistently use spermicides. Preliminary re- search indicates that the irritative effects which can increase the risk of HN acquisition reported in some studies of spermicides may be dose-related, as suggested in the case of nonoxynol-9 (78), or may depend on the vehicle, as suggested for mefengol.5 In order to reduce the possibility of inadvertently facilitating HN transmission during preliminary ' Laga, M. Intravaginal STD/HIV prevention technology comrollable by women. Session 173, VIIIth international conference on AIDS, Amsterdam, the Netherlands (1992). 111 research, toxicity studies evaluating dosage, appli- cation methods, and insertion frequencies are now performed in women at low risk of HIV exposure before efficacy studies are begun among high-risk women such as sex workers or those in discordant couples. Other products under development in- clude a vaginal ovule containing a non-nucleoside reverse transcriptase inhibitor-iodine mixture and sulphated polysaccharide microbicides such as dextran sulphate and carrageenans which inhibit HIV (79-81) and appear to block viral entry into vaginal mucosal cells (80). The diaphragm and the cervical cap have not been investigated for HIV prevention but have been shown to provide a 50-100% reduction in the transmission of other STDs, particularly those due to cervical pathogens, when used in cortjunction with a gel or foam spermicide. As with spermicide and microbicide research, efficacy trials of these woman-controlled methods should include the ad- vocacy of partner condom use. The lack of knowl- edge concerning the relative roles of the vaginal mucosa and the cervix in HIV acquisition has led to a focus on prevention methods such as abstinence and partner condom use which entail high theor- etical protection but potentially low user- effectiveness. Methods such as the diaphragm and cervical cap may have equal or superior overall protective efficacy in the end because higher real- life utilization rates may be achievable. Conclusion As the HIV I AIDS epidemic intensifies, women continue to be identified as reservoirs of infection or as vectors of transmission to their male partners and to their offspring. In fact, following the publi- cation of the findings from a French-American study on the effectiveness of AZT given prenatally, during labour and delivery, and to newborns (82), many countries appear more prepared to invest in programmes to prevent mother-to-child transmis- sion than in programmes to prevent transmission to women in the first place. This draws attention away from men's equal responsibility to prevent HIV and delays the development of research and intervention strategies designed to enable women to protect themselves. Increasingly it is being argued that short-term solutions to women's vulnerability to HIV infection and other sexually-transmitted diseases lie in the development of clandestine, woman-controlled methods of HIV prevention that do not require male compliance (83), a research agenda which has been slow to receive attention. Long-term solu- tions lie in the empowerment of women to achieve economic autonomy. The promotion of gender equity should receive high priority in the national AIDS plans of all countries. Labour laws and hu- man resources policies should be revised to guar- antee women the right to earn salaries on a par 112 with men and to have equal access to trammg. Women's ability to protect themselves should be strengthened by expanding access to education including sex education; by fighting reduced social expectations for girls; and by helping women to learn about their rights and organize on their own behalf. Until women become fully part of the dia- logue that establishes policy and distributes re- sources and until women share in power more equally with men - in both the public and the private spheres - they will remain at heightened risk for HIV (30). Summary Biological differences between men and women help explain why the human immunodeficiency virus (HIV) is infecting increasing numbers of young girls and women in industrialized countries through heterosexual activity. As well, lack of economic and educational opportunities may not only lead women to resort to entering into sexual relationships for economic reasons but may also reduce both their willingness and their capacity to negotiate protection. Although advocating partner condom use remains the primary prevention message, increasing effort is being directed at the development and evalua- tion of woman-controlled prevention methods. Long- term solutions to women's vulnerability lie in the empow- erment of women to achieve economic autonomy and full participation in decisions which affect their lives. Resume Transmission du VIH par voie sexuelle aux femmes des pays industrialises Les differences biologiques qui existent entre les hom- mes et les femmes expliquent en partie pourquoi le virus de l'immunodeficience humaine (VIH) infecte de plus en plus de jeunes lilies et de femmes des pays industriali- ses lors des rapports heterosexuels. Par ailleurs. !'ab- sence de facilites economiques et educatives risque non seulement de pousser les femmes a avoir des relations sexuelles pour des raisons economiques, mais egalement de restreindre a la fois leur volonte et leur capacite de negocier une protection. Bien que le plai- doyer en faveur de !'utilisation du preservatif par le partenaire demeure le premier message de prevention, on s'efforce de plus en plus de developper et d'evaluer les methodes de prevention controlees par les femmes elles-memes. On remediera a long terme a la vulnerabi- lite des femmes en leur donnant les moyens d'acquerir leur independance economique et de participer pleine- ment aux decisions qui affectent leur vie. Rapp. trimest. statist. sanit. mond., 49 (1996) References/References I. Holland,J. et al. Risk, power and the possibility of pleasure: young women and safer sex. AIDSrrm·, 4(3), 273-283 (1992). 2. Voeller, B. & Anderson, D.J. Heterosexual transmission of HIV (letter]. Journal of the American Afediwl Association, 267: 1917-1918 (1992). 3. Plummer, F.A. et al. Cofactors in male-female ~exual transmission of human immunodeficiency virus type I. Journal of infertirms diseases, 163: 233-239 ( 19Y I). 4. Castro, K.G. et al. Transmission of HIV in Belle Glade. Florida: Lessons for other communities in the Cnitcd States. Science, 239: 193-197 (1988). 3. Quinn, T.C. et al. The association of s1philis with risk of human immunodeficiency virus in patients auending sexually transmitted disease clinics. A.rchives of internal medicine, 150: 12Y7-1302 (1990). 6. Holmberg, S.D. et al. Prior he1 pes simplex virus type 2 infection as a risk factor for Hl\' infection . .founw/ of llie ,1 merican Mediral Associalion, 259, I 048-1030 ( l 988). 7. Hook III, E.W. etal. Herpes simplex virus infection as a risk factor for human immunodeficiency virus inkction in heterosexuals. Journal of infertinus diser,scs, 165: 231-233 (1992). 8. Keet, I. et al. Herpes simplex virus type 2 and other genital ulcerative infections as a risk factor for HI\'-1 acquisition. Genitourinary nwdicinr, 66: 330-'.tn (1990). 9. Laga, M. et al. Non-ulccrati1·c sexually transmitted diseases as risk factors for HIV-I transmission in women: results from a cohort study. All)_\ 7: !l:i-102 ( 1993). 10. Wasserheit, J.N. Epidemiological synergy: Inter- relationships between human immunodeficiency 1irus in- fection and other sexually transmitted diseases. St<xually tmnsmi//ed diseases, 19(2): 61-77 (1992). 11. Padian, N. et al. Male-to-female transmission of immunodeficiency virus. .founlfll of llw .-tmerican Afediral Association, 258: 788-790 ( 1987). 12. Van de Wijgert, J.H.H.M. & Padian, N. S. Heterosexual transmission of HIV. In: L. Sherr (ed.). AllJS and llie heterosexual popuullion. Switzerland, Harwood Academic Publishers, 1993, pp 1-19. 13. King, A.J.C. et al. Highliglus from the Canada Youth and AIDS Studv.Joumal of sdwol heallh, 59, 139-143 (1989). 14. Chetwynd,J.etal. Condom use in anal intercourse amongst people who identify as homosexual, heterosexual or bisexual. ,\'ew :lt.,a1and 11u•diraljoumal, 105, 262-264 (1992). 13. Bolling, D.R. Prevalence, goals and complications of heterosexual anal intercourse in a gynecologic population. Joumalofrejlroduclive medidne, 19(3), 120-124 (IY77). 16. Hankins, C. et al. Une etude sero-epidemiologique portant snr la clicntclc feminine de quatre cliniques de MTS a Montreal (Abstract]. .4.nnale{ de l'As.wcialion crmadienne- franraise pour l'avancemml des sciences, 57, 328 ( 1989). 17. European Study Group on Heterosexual Transmission of HIV. Comparison of female to male and male to female transmission of HIV in 363 stable couples. Brilislt mediral joumfll, 304(6830): 809-813 (1992). 18. Simmonds, P. Variation in HIV virus load of individuals at differen tstages in infection: possible relationship with risk of transmission. AIDS, 4(suppl. 1): S77-S83 (1990). 19. Anderson, R.M. et al. The spread of H[V-1 in Africa: sexual contact patterns and the predicted demographic impact of AIDS. Valure, 352: 581-389 ( 1991). 20. Moss, G.B. et al. Association of cervical ectopy with heterosexual transmission of human immunodeficiency virus: results of a swdy of couples in ~airobi, Kenya.Journal of infectious diseases, 164: 588-391 (1991). 21. Saltz, G.R.etal. Chlamydiatrachomatiscen"ical infections in female adolescents. Jmmwl of pediatrics, 98: 981-983 (1981). 22. Oriel, J.D. et al. Infection of the Ulcrine cen·ix with Chlamydia! trachomatis.Journal of lnferlious Diseases, 137 ( 4): 443-451 (1978). Wld hlth statist. quart., 49 (1996) 23. Rees, E. elal. Chlamydia in relation to cervical infection and pch·ic inflammatory disease. In: Hobson, D. & Holmes, K.K. (eds.). :\'ongonorrowl uretlaitis and relaled inferlions. Washington, DC: American Society for Microbiology, 1977, pp 67-76. 24. Plourde, P.J. et al. Human immunodeficiency virus type I infection in women attending a sexually transmitted diseases clinic in Kenya:fournalofi11fecliousdiseases, l66: 86-92 (1992). 23. Pepin,J. etal. HIV-2 infection among prostitutes working in The c;ambia: association with serological evidence of genital ulcer diseases and with generaliLed lymphadenopathy.A/DS, 5:69-75 (1991). 26. European Study Group. Ri.sk factors for male to female transmission of HIV. Brilish .Hedical Journal, 298: 411-,t 13 (1989). 27. Allen, S. et al. Human immunodeficiency virus infection in urban Rwanda. joumal of thl' American Mnliml Association, 266(12): 1657-166'.1 ( l Y9 l ). 28. Howe,J.E. et al. Contraceptives and HIV. MDS, 8: 861-871 (1994). 29. Commonwealth Expert Group. Engendl'ring adjustment for lhe 1990s (R,jwrt ofa Cm1111w11wmlth L'xpert Grm1f1 on H\mien and Slmctuml Adjustment). London, England: Commonwealth Secretariat, I Y89. 30. Heise, L.L. & Elias, C. Transforming AIDS prevention to meet women', needs: a focus on dc1-dopingco11ntries. Social .scimreand 11wdirine, 40(7): 931-94'.3 (1995). 31. DeCosas, J. & Pedneault, V. Women and AIDS in Africa: demographic implications for health promotion. Health poliry and/J/m111ing, 7(3): 227-233 (19Y2). 32. World Bank. World dn1t•l.ofmwnl r,1,011 I 993: invesling in h.ealllt. 1':ew York, Oxford l;nivcrsity Press (1993). 33. Johnson, A. M. et al. Sexual lifestyle, and HIV risk . .'Vatul'f', 360: 410-412 (1992). 3·1. Dolcini, M.M. et al. Demographic characteristics of heterosexuals with multiple partners: the national AIDS behavioral sun'cy. Family planning f1tcnjletli1""• 25: 208-214 (1993). 33. Grinstead, 0. A. et al. Sexual ri,k for human immuno- deficiency virus infection among women in high-1isk cities. Famil\• filanning f1ersf1eclive.,, 25: 252-256, 277 ( 1993). 36. Foreman, D. & Chilvers, C. Sexual behaviour ofvoung and middle-aged men in England and \,\'ales. Brilish 11wdical jm,mal, 298: 1137-1142 (1989). 37. Sundet,J. etal. Prevalence ofrisk-pronc sexual behaviour in the general population ofNonvay. In: Fleming et al. (eds.). Tiu, l(lobal impact of AIDS, New York, Allen R. Liss. 1988. 38. Messiah, A. & Mouret-Fourme, E. Homosexualite, bisexualite. Element, de socio-biop;raphie sexuelle. Pojlulation, 5: 1333-1380 (1993). 3Y. Ross, M.W. Prevalence ofriskfactors for HIV infection in the Australian population. The niediral journal o/"Austmlia, 149: 362-365 (1988). 40. Chu, S.Y. et al. AIDS in bisexual men in the United States: epidemiology and transmission co women. American journal offmblic hmllh, 82(2): 220-224 ( 1992). 41. Crawford, J. et al. Gay and bisexual: bisexual and gay. The sexual behaviour of men who ha1·e sex with men and women in an Australian sample of gay and bisexual men.In: Social tt5jlells oflhe J1revmlio11 oJA!DSproject. Macquarie l'ni1·ersity, Sydney, Australia. 1992. 42. Philpot, C.R. et al. A sun·ey of female prostitutes al risk of HIV infection and other sexually transmissible dise,Lses. Genitourinar,,• 11u,ditine, 67: 384-388 ( 1991). 43. Wasser, S.C. etal. Assessing behal"ioral risk for H l\' infection in family-planning and STD clinics: similarities and differences. S,xually transmitted diseases, 16(4): 178-183 (1989). 44. Stevens, P.E. Lesbians and HI\': clinical, research, and policv issues. ,lm.erican joumal of ortlwj,syrhiatry, 63(2): 289-294 (1993). 113 43. Monzon, O.T. & Capellan, J.M.B. Female-Lo-female transmission of human immunodeficiency virus. Lanai, 2: 40-41. (1987). 46. Perry, S. et al. Orogenital transmission of human immunodeficiency virus (HIV). Annals of internal medicine, 111: 931 (l 989). 47. Chu, S.Y. et al. Epidemiology of reported cases of AIDS in lesbians, United Slates 1980-89. A.meriran journal of public health, 80: 1380-1381 (1990). 48. Magura,S.etal. Women usuallytakecareofthcirgirlfriends: BisexualiLy and HIV tisk among female intravenous clrug uscrs.Journalofdrugissul's, 22(1): 179-190 (1992). 49. Einhorn, L. & Polgar, M. HlV-risk bcha\·iour among lesbians and bisexual women. AIDS education and jmvention, 6(6): 314-323 (1994). 30. Jackson, L. et al. Varied potential risks of HIV infection among prostitULes. Solial science and mediriw·, 35(3): 28 i-286 (1992). 31. McKeganey, N. & Barnard, M. Selling sex: female streeL prostitution and H[V dsk behaviour in Giasgow. AIDS rare, 4(4): 395-407 (1992). 32. Taylor, C.C. Condoms and cosmology: The "fractal" person and sexual risk in Rwanda. Social stiencr and rn,•dirine, 31 (9): 1023-1028 (1990), 33. Bozon, M. L'enLrec dans la sexualite adulLe: le premier rapport et scs suites. Poj,uw.tion, 5: 1317-13.32 (1993). 34. Plant, M.L. Sex work, alcohol, drugs and AIDS. In: Plant, M. (ed). AIDS, drugs and /irostitution. London: Ta\'istock. 1990. 33. Lyons, C. & Fahrner, R. HIV in women in the sex indusLry and/or injection drug users. Xationrtl As.ml'iation of Amnican College of Obstetricians and Oynecologists diniral i.mw in perinatal and women's health nursin1r 1(1): 3?,-40 (1990). 36. Cohen, J. et al. Prosliuttes and AIDS: public policy issues. AIDS and public poliC'j jrn1rnal, 3: I G-22 ( 1988). 37. Darrow, W. et al. Prostitution, inLra\·enous drug use and HIV-1 in the United States. [n Plant, '.\1. (ed.) .AIDS, drugsand prostitution. London, Tavistock. 1990. 38. Ward, H. et al. Prostitution and risk of HIV: female prosti- tutes in London. lfritish mt•diwl journal, 307: 356-338 ( 1993). 39. Osmond, M.W. et al. The multiple jeopardy of race, class, and gender for AIDS risk among women. Geruler & soriety, 7(1): 99-120 (1993). 60. Windle, M. High-risk behaviors for AIDS among heterosexual alcoholics: a pilot study. journal of studit<s on alcoho~ 50 ( 6): 303-307 ( 1989). 61. Quadagno, D. et al. Women at risk for human immunodeficiency \'irus. Jrnmwl of piyrhology & human sexuality, 4(3): 97-110 (1991). 62. Suffet, F. & Lifshitz, M. Women addicts and the threat of AIDS. Qualitative health mearch, l( I): 51-79 (1991 ). 63. Cohen,J.B. \'ihywoman partners of drug users will continue to be at high risk for HIVinfectionjrmmalnfaddirtivediseases, 10(4): 99-110 (1991). 64. Solomon, L. et al. Differences in risk factors for human immunodeficiency \'irus t)lle l serocotwersion among male and female intravenous drug users. Amerimn jourrwl of epidemiology, 137: 892-898 (1993). 63. Booth, R.E. et al. Intravenous drug users and AIDS: Risk behaviors. Amerimn journal nf drug and alrolwl almse, 17: 337-333 (1991). 114 66. Centers for Disease Control and Prevention. HIV/AIDS surveillance Report, 7 ( l): lO (1995). 67. Chiasson, M.A. et al. Heterosexual transmission of HIV-l associated with the use of smokable freebase cocaine (crack). AIDS, 5(9): 1121-1126 (1991). 68. Centers for Disease Control. Relationship of syphilis Lo drug use and prmtiLutinn - Connecticut and Philadelphia, Pennsylvania. ,vlorbidity and mortality wePk(\' re/inrt, 37: 755-764 (1988). 69. Siegal, H.A. et al. High-risk behaviors for transmission of syphilis and human immunodeficiency \irus among crack cocaine-using women. Sexuallv transmitted diseases, 19(3): 266-271 ( 1992). 70. McCoy, H.V. & Miles, C. A gender comparison of health sLatus among users of crack cocaine. Journal of /Js)·clwactive drugs, 24(4): 389-S\97 (1992). 71. Forney, M.A. et al. Exchanging sex for crack-cocaine: A comparison ofwomett from rural and urban communities. Juurnal 0Jam1munity hmllh, 17(2): 73-85 (1992). 72. Shenington, D.O. The acceptability of the female condom among low-income African-American women. Journal of the Nationtll Medical Assot:ialion, 85(S): '.Hl-347 (1993). 73. Golub, E.L. et al. Shon-term acceptability of the female condom among staff and patient~ata New York City hospital. Family planning pers/wtivt•.,, 27(4): l S3-l:i8 (1995). 74. Farr, G. et al. Contraceptive efficacy and acceptability of the female condom. Ameriamjournal of jmblic health, 84(12): 1960-1964. (1991). 73. Rosenberg, M.J. et al. Effect of contracepLi\·e sponge on chlamydia! infection, f\Onon-hea and candidiasis - a comparative trial. ]oumal tif the Ameiirrm 1Hedical Association, 257: 2308-2312 (1987). 76. Weir, S.S. et al. The use of nonoxynol-9 for protection against cervical gonorrhea. American journal of public health, 84: 910-911 ( !994). 77. Feldblum, P.J. & Weir, S.S. The protective eflect of nonoxynol-9 against Hl\' infection. American journal ofj1uhlic helllth, 84: 1032-10'.H (1994). 78. Krciss,J. et al. Efficacy of nonuxynol-9 contraceptive sponge use in preveminihctcrosexual acquisition of HIV in Nairobi ProsLituLes. jnurnlll ,if" th.e A11u•rimn .\1rdiral Associlltion, 268: 477-482 (1992). 79. Baba, M. et al. Sulfatetl polysaccharides are potent and selective inhibiLors of various enveloped \'intses, including herpes simplex virus, cvtomegalovirus, vesicular stomaLiLis \'irus, and human immunodeficiency \'ints. Antimicrolnal agents & d,emotheraf/'j, 32(11 ): I 742-174S (1988). 80. Elias, C.J. & Heise, L.L. Challenges for the development of female-controlled vaginal microbicidcs.1\fl)S, 8: 1-9 (1994). 81. Bagasra, 0. & Llschner, H.W. Acti\ity of dextran sulphate and other polyanionic polysaccharides against human immunodeficiency virus. Journal of infntim,s diseast'5, 158: 1084 (1988). 82. Connor, E.M. et al. Reduction of maternal-infanL transmission of human immunodeficiency virus Lype 1 wiLh zidomdinc treatment ;\',w England joumal of mi!dicine, 331: 1173-1180 ( 1994). 8,1. Stein, Z.A. HIV prevemion: the need for methods women can use. Amniran journal of public hea/Jh, 80, 460-462 (1990). Rapp. trimest. statist. sanit. mond., 49 (1996) Women and tuberculosis Maire ConnoJJya & Paul Nunna Introduction Tuberculosis kills over one million women each year ( 1 ). b It is the greatest single infectious cause of death in women worldwide. The consequences of tuberculosis in women are far-reaching, affecting women themselves, their children and society in general. Tuberculosis in women creates orphans, impoverishes families and reduces the economic development of society. From a global perspective, the magnitude of the tuberculosis epidemic is enormous. Approximately one-third of the world's population is infected with Mycobacterium tubercuw- sis. It has been estimated that, in 1995 alone, 8.8 million people would have developed tuberculosis and about 3 million would have died (2). The great- est burden of disease is in the age group 15-49 years which includes women and men in their most pro- ductive years. It is estimated that 80% of deaths due to tuberculosis occur in this age group, accounting for 26% of avoidable adult deaths in low-income countries (3,4). Tuberculosis is an important cause of morbidity in women, particularly for those in their reproduc- tive years. It is estimated that 646 million women and girls worldwide are already infected with the tubercle bacillus, and of these, an estimated 3.1 million become ill each year with tuberculosis. b In women aged 15-44 in low-income countries, tu- berculosis is the third cause of morbidity and mor- tality combined, leading to the loss of about 11 million disability-adjusted life years (DALYs) in 1990 (5). The disease leads to even more deaths in women annually than all causes of maternal mor- tality combined (5). This represents a major cause of preventable suffering and death in women that is largely unappreciated. Population growth, the HIV epidemic, increasing poverty and rising levels of drug resistance are increasing the burden of tuberculosis in women. Women and the epidemiology of tuberculosis Tuberculosis is a bacterial infection most common- ly affecting the lungs. Transmission can occur when an individual with pulmonary tuberculosis expels tubercle bacilli into the air by coughing or • Medical Officer, Tuberculosis Research and Sur;eillance Unit, Global Tuberculosis Programme, World Health Organization, Geneva. b Global Tuberculosis Programme, Groups at Risk, WHO repmt on the tubercuwsis epidemic in 1996, Geneva, (\VHO/TB/96.196). Wld hlth statist. quart., 49 (1996) sneezing. The risk of an uninfected individual be- coming infected depends on the duration and inti- macy of contact with an infectious case. Studies on gender differences in infection rates show that the risk and prevalence of tuberculosis infection are similar in males and females until adolescence, after which they are higher in males (6). The higher rates of infection in men are thought to be due to greater numbers of social contacts experi- enced by men who tend to work out'iide the home more than women. Once infected, an individual has a 10% lifetime risk of developing the clinical disease. A number of studies have shown that in high-prevalence coun- tries, women in their reproductive years (15 to 40 years old) have higher rates of progression to disease than men of the same age (7, 8). This may be due to the physiological changes associated with reproduction. The excess of disease observed in women of reproductive age contrasts with higher prevalence of infection in men of the same age (9). There is some evidence that postpartum women may be subject to higher rates of progression to disease than other women ( 10, 11). Speculation as to why this should be has been interesting but not, so far, enlightening. Theories include rapid hor- monal changes, postpartum descent of the dia- phragm and expansion of the lung, the nutritional strain oflactation, and stress associated with insuffi- cient sleep due to the demands of the new child (12, 13). The male:female ratio of new tuberculosis noti- fications shows wide variation both between and within countries. In the United States of America in 1990, the male:female ratio of tuberculosis noti- fications was 2:1 (14). In many low-income coun- tries, the ratio of male to female cases notified was also 2:1 (]), however, in Zambia, the male:female notification ratio ranged from 0.7:1 to 1.7:1 in different provincesc. In Tanzania, the ratio for new smear-positive cases has consistently been 1.9:1 (15), while in a similar programme in Nicaragua, the ratio is 1.2:I.<l It is unclear whether the higher rates in men are due to biological mechanisms c Sichone, M. Tuberculosis and gender in Zambia: the burden on women. Unpublished Master\ thesis. Amsterdam: Royal Tropical [nstitutc, 1993. <L Cruz Gonzalez, J.R. lnforme Anual del Programa d" Control de Tuberculosis 1993. Programa de Control de Tuberculosis y Lepra, Ministerio de Salud, Gobierno de Nicaragua, 1993. 115 placing them at higher risk of developing tubercu- losis or whether the difference is due to under- notification in women, or both. An active case·· finding study conducted over two years in India showed the incidence of tuberculosis in women to be 30% of that in men (16) suggesting that the higher notification rates in men may be explained by biological mechanisms. However, studies in Nepal comparing active case-finding with passive case-finding (self-referral) of patients to the exist- ing services showed that a higher proportion of women were undiagnosed in the community. The overall male:female ratio in active case-finding was 1.2: 1 while in the passive case-finding group, the ratio was 2.6: 1 ( J'ij. This suggests that, in pro- grammes which rely on passive case-finding, as most programmes do, under-diagnosis of tubercu- losis in women may occur. Socioeconomic and cul- tural factors may well be the cause of this under- notification and these factors are discussed below. Women of reproductive age have higher mor- tality and case-fatality rates from tuberculosis than men of the same age. A prevalence survey in China found that women had higher tuberculosis mortal- ity rates than men from birth to 30 years of age.c A study in India found that females aged 5-24 had a case-fatality rate over 35% greater than males of the same age (18). The higher case-fatality rate noted in young women may in part be due to increased biological susceptibility to tuberculosis related to child-bearing. It also may in part be due to a lower level of awareness of tuberculosis among women leading to delays in diagnosis and treat- ment, or greater constraints in accessing health care than men. Social and economic consequences al tuberculosis in women Tuberculosis can run an insidious course, gradual- ly disabling a woman. Without treatment, the dis- ease is fatal in 50 to 60% of cases, 5 years after onset of symptoms ( 19). A number of studies have shown that women tend to wait longer to report their illness than men and only do so when the illness severely interferes with their daily activities (20, 21). In the case of tuberculosis, this leads to late presentation and a worse prognosis. Women have also been found to consult with traditional healers more frequently than men, further delaying the diagnosis and treatment of tuberculosis (22). The fear and stigma associated with tuberculo- sis have a greater impact on women than on men, often leaving them in a more precarious social and economic position (23). In Pakistan, women with tuberculosis were more likely to be divorced and c Ministry of Public Health of the People's Republic of China. National random survey for the epidcmioloh'T of tuberculosis in 1990, 1992. 116 husbands often took a second wife. Unmarried women with tuberculosis were more likely to have difficulty finding a marriage partner than those without tuberculosis (24). In India, a study found that married men and single women with tubercu- losis perceived a greater level of family support to initiate and complete treatment than married women (Nair, D. Tuberculosis in Bombay: new reasons for concern, unpublished document). An- other study in India found that male patients ex- pected to receive care from their wives while the reverse was seldom true (25). '.\farried women often tried to hide their illness for fear of desertion, rejection or blame for developing the disease. Tuberculosis in women affects child survival and family welfare. Studies on tropical diseases have shown that women's illness has a profound impact on the social and economic well-being of families and households (26, 27), although the im- pact of disease varies widely depending on the socioeconomic and cultural context. Tuberculosis, when untreated in mothers, can have serious re- percussions for their children. In Bangladesh, a study of children under age 10 found that a mother's death sharply increased the chances that her children would die within 2 years, especially for her daughters (28). Children whose mothers die are 3 to 10 times more likely to die within two years than those with both parents alive. A mother's death has twice the impact of a father's death on child survival (29). Tuberculosis in women puts their children at risk for tuberculosis infection, disease and death. This triple threat makes detect- ing and treating tuberculosis in women absolutely vital in order to ensure the survival and well-being of their children. Tuberculosis affect~ the economic productivity of women. The disease has the greatest impact on adults aged 15 to 49, who represent the most eco- nomically productive part of society. In the devel- oping world, women are responsible for up to three-quarters of the food produced annually. In parts of Africa, women produce 80% of the food consumed domestically and at least 50% of export crops. Women earn 40 to 60% of household in- come, if home production is included (30). Tuber- culosis in women reduces their ability to work and leads to significant loss of income. In this way, tuberculosis perpetuates poverty in women and this has a long term impact on the economic devel- opment of their community. Tuberculosis control in women It is increasingly recognized that socioeconomic and cultural factors play an important role in tu- berculosis control. These factors are particularly significant for women. Over 95% of tuberculosis cases occur in low-income countries. In these coun- tries, women's disproportionate poverty, low social status and reproductive role may lead to greater Rapp. trimest. statist. sanit. mond., 49 (1996) barriers in accessing health care. In the WHO tu- berculosis control strategy, DOTS (Directly Ob- served Treatment, Short course), passive case-find- ing is the recommended method of case detection. Suspected cases must therefore present themselves to the health services in order to be diagnosed. Once diagnosed, tuberculosis requires a minimum of six months supervised treatment. In many cul- tures, this places major demands on women which often conflict with their other duties. Knowledge of the disease and recognition and interpretation of its signs and symptoms require a certain level of education. In many low-income countries, women have lower standards of educa- tion and literacy. A lower level of education among women can contribute to delays in diagnosis. In Nepal, a study of new cases found that the mean reported duration of cough before diagnosis was 27 days for men and 49 days for women.f The association between tuberculosis and poverty has long been noted. It is estimated that women make up 70% of the world's poor (30). These women face the greatest obstacles to seeking health care for tuberculosis and achieving a com- plete cure. Poverty may curtail women's access to health services as poor women have less disposable income to spend on health. Transportation costs involved in using the health care services present greater barriers to women. Studies in Kenya and Peru confirm that distance from health institutions and user fees are a larger obstacle to women than to men in seeking medical care (31, 32). Work patterns and responsibilities such as child minding and crop growing may preclude women from hav- ing the time to visit a health facility to collect med- ications. Women face high opportunity costs for time spent on health care. The strongest evidence of gender differentials in health status and use of health services has been documented in South Asia. Community-based studies in India found that women had a higher rate of illness and disease than men in the same household, but used health services less often (30). Studies in other countries also found that even where there is no apparent gender difference in prevalence of disease, women were less likely than men to seek care for infectious diseases. In Thai- land and Columbia, about 6 times as many adult men as women attend malaria clinics for treatment (20, 33). Cultural factors such as restrictions on women travelling alone or being treated by male health care providers may restrict women's use of health services. Other factors such as stigma, household decision-making patterns, and the cul- tural value given to women's health may lead to delays in the timing of care-seeking behaviour and therefore diagnosis. f British Nepal Medical Trust. 1992 Annual critical report: TB Leprosy Programme. Kathmandu: unpublished, 1993. Wld hlth statist. quart., 49 (1996) Once in treatment, women may face greater difficulties maintaining compliance. In the Philip- pines, a high default rate among pregnant or lac- tating women was reported to be due to fear that the drngs could cause miscarriages, reduce the ability to breast feed or harm the baby (34). A study in India showed that parents of girls of marriage- able age were reluctant to continue sending their daughters to the tuberculosis clinic once symptoms subsided. It was felt that knowledge of the disease among the community reduced their daughters' chances of marriage (25). Women face major obstacles in obtaining treat- ment, and once it is begun, further obstacles in completing the course of treatment. Tuberculosis control programmes should be sensitive to the con- straints faced by women in seeking health care and maintaining compliance. Tuberculosis and HIV/AIDS in women HIV infection is the most potent risk factor for the development of disease in tuberculosis-infected in- dividuals. An HIV-negative individual infected with Mycobacterium tuberc,dosis has a I 0% lifetime risk of developing tuberculosis. However, for individuals co-infected with HIV, the annual risk of progression to disease has been estimated as between 3% and 13% (35, 36). Due to this interaction, tuberculosis has emerged as the most common life-threatening opportunistic disease associated with HIV in many developing countries. In one study in Africa, up to 54% of AIDS patients had clinical tuberculosis dur- ing the course of HIV infection (37). In Brazil and Haiti, almost one-quarter of AIDS patients had tu- berculosis( 38). In studies in lndia(39) and Thailand (38), up to 60% of AIDS patients had tuberculosis confirmed by clinical or autopsy findings. The HIV epidemic is increasing the burden of tuberculosis in women. During their reproductive years, when women are most at risk of progressing from tuberculosis infection to disease, they are also at greatest risk of HIV infection. A study in Tanza- nia on the association of tuberculosis and HIV showed that there has been a marked increase in notifications of smear-positive tuberculosis in 15-34 year-old women and men between 1985 and 1991 (15). The rates of HIV infection among smear- positive cases were higher in women than in men in the 15-34 year age group. A study in Zambia in 1988 showed that there were twice as many tuber- culosis notifications among women as men in the 14-24 age group. Seventy-four per cent of women were HIV positive compared to 48% of men (40). In Kenya, a study of hospital admissions showed that the average age of all admitted patients was the same for both sexes, yet the average age of HIV positive patients with tuberculosis was 24 years in women and 37 years in men. This suggests that HIV positive women of childbearing age have a marked- ly higher risk of developing tuberculosis (41). 117 The HIV epidemic is increasing the incidence of tuberculosis in women, particularly in young women in low-income countries. The full impact of HIV has not yet occurred in A~ia, where two-thirds of the world's tuberculosis occurs at present. The spread of HIV into Asian communities is likely to result in large increases in HIV-associated tubercu- losis in women in the future. Tuberculosis in women: the response Tuberculosis is one of the world's most neglected health problems. Population growth, the HIV epi- demic, increasing poverty and growing levels of drug resistance will lead to increasing numbers of women developing tuberculosis over the next few decades. However, an effective system of control- ling tuberculosis exist~ in v\THO's recommended DOTS strategy. The fundamental components are the detection of infectious cases and the supervi- sion and documentation of their treatment to the point of proven cure. It is important that DOTS programmes are gender-sensitive in their imple- mentation to ensure that women have access to tuberculosis treatment. It is also important that women are more aware of the risk of tuberculosis. Health education material for women should in- clude information on tuberculosis, particularly in maternal and child health and HIV I AIDS services. Tuberculosis treatment is one of the most cost- effective health interventions available in develop- ing countries (4). The greatest obstacle to reducing the burden of tuberculosis in women remains the lack of political will and inadequate financial re- sources available for tuberculosis control. Health policy makers, donor countries and non-govern- mental organizations need to be more aware of the impact of tuberculosis on women. It is imperative that women with tuberculosis have access to treat- ment through effective control programmes. Tu- berculosis control in women is an investment in the future health of women themselves, their chil- dren and society at large. Summary Tuberculosis is the leading infectious cause of death in women worldwide. The disease poses a major threat to women's health security. Population growth, the HIV epidemic, increasing poverty and rising levels of drug resistance will inevitably increase the burden of this disease in women. Women are at increased risk of progression to disease during their reproductive years However, in most low- income countries, twice as many men are notified with tuberculosis as women. Biological mechanisms may account for most of this difference but socioeconomic and cultural factors leading to barriers in accessing health care may cause under-notification in women. Tuberculosis control programmes should be sensitive 118 to the constraints faced by women in accessing health care, in order to empower women to commence and complete treatment. The fear and stigma associated with tuberculosis have a greater impact on women than on men. often leaving them in a more precarious social and economic posi- tion. Tuberculosis in women creates orphans, impover- ishes families and reduces the economic development of society. Tuberculosis is a major cause of preventable suffering and death in women. WHO's recommended tuberculo- sis control strategy, DOTS. represents a cost-effective response to the problem of tuberculosis in women. Tuberculosis is a major women's health issue. It is a global health priority that tuberculosis treatment be made available to women, particularly to those in low- income countries who are bearing the brunt of this epidemic. Resume Les femmes et la tuberculose La tuberculose est la principale cause infectieuse de deces chez les femmes dans le monde entier. Cette maladie constitue une grave menace pour la securite sanitaire des femmes. La croissance demographique, l'epidemie d'infection a VIH, l'accroissement de la pau- vrete et !'augmentation des niveaux de la pharmaco- resistance alourdiront inevitablement le fardeau de cet- te maladie chez les femmes. Les femmes courent de plus en plus le risque de contracter la maladie durant leurs annees reproductives. Toutefois, dans la plupart des pays a faible revenu, la tuberculose est notifiee deux fois plus chez les hommes que chez les femmes. Les mecanismes biologiques expliquent la plupart de ces differences, mais les facteurs socio-economiques et culturels qui entravent l'action de soins de sante sont peut-etre la cause de la sous-notification chez les fem- mes. Les programmes de lutte contre la tuberculose devraient tenir compte des difficultes rencontrees par les femmes dans l'acces aux soins de sante afin de donner les moyens aux femmes de commencer et de completer le traitement. La crainte et l'opprobre associes a la tuberculose ant davantage d'impact sur les femmes que sur les hom- mes, et !es laissent souvent dans une position sociale et economique plus precaire. La tuberculose chez les femmes cree des orphelins, appauvrit les families et reduit le developpement economique de la societe. La tuberculose est une cause principale de souffrances et de deces evitables chez les femmes. La strategie de lutte contre la tuberculose recommandee par l'OMS, DOTS, represente une reponse de ban rapport cout/ efficacite aux problemes de la tuberculose chez les femmes. La tuberculose est un probleme de sante majeur des femmes. La mise a disposition des femmes d'un traitement antituberculeux est une priorite sanitaire mondiale, et notamment dans les pays a faible revenu qui souffrent le plus de cette epidemie. Rapp. trimest. statist. sanit. mond., 49 (1996) References/References I. Kumaresan, J.A. et al. Tuberculosis. In: :\1urray, C.J.L. & Lopez, A.O., eds. The g/JJbal /Jurden ofdisea1eand riskftu:ton in 1990, Geneva, \\'orld Health Organization, 1996. 2. Dolin, P.J. etal. Global tuberculosis incidence and mortality during 1990-2000. Bulletin of the Wr,rltf Health Organization, 72: 213-220 (1994). Dolin, P.J. et al. Incidence mondiale de la tuberculose et mortal ire de 1990 a 2000. (Resume] Bulletin de l'Oq;ranisation rnondia/,e de la Sante, 72: 220 (1994). 3. Koehl, A. Tuberculosis: distribution, risk factors, mortality. lmmunobiology, 191: 32.'i-336 (1994). 4. Murray, C.J.L. et al. Tuberculosis in developing coumries: burden, intervention and cost. Bul&tin of the lntmwtional Union against Tubenulosis and Lung Disert5e, 65: 6-24 (1990). 3. The World Bank. W01ld development report 1993: investing in health. New York, Nt', Oxford l'nivcrsity Press. 1993. 6. Sutherland, I. & Styblo, K. The risk of tuberculosis infection in the Netherlands from 1967 to 1979. Tubercle, 64: 241-233 (1983). 7. Groth-Petersen, E. et al. Epidemiological basis of tuberculosis eradication in an advanced country. Bulletin of the World Health O,ganiw.tion, 21: 3-49 (1939). Groth-Petersen, E. et al. [resume sans titre] Bul&tin de !'Organisation mondiale de la SanlR, 72: 42-44 ( 1939). 8. Gedde-Dahl, T. Tuberculous infection in the light of tuberculous matriculation.Americaljoumal,fhyginie,56: 139- 214 (1932). 9. Fine, P.E.M. Immunities in and to tuberculosis: implications for pathogenesis and vaccination. In: Porter, J.O.H. & McAdam, KP.WJ., eels. Tubl'rcuwsi1: bark to the future. John ·wiley & Sons, Chichester, New York, Brisbane, Toronto & Singapore, 1991. 10. Schwabe, K.H. & Dobstadt, H.P. l.ungcntuberkulose und Schwangerschaft [Pulmonary tuherculosis and pregnancy]. Bl'itrage zur Klinik unrl Erforsrhung der Tuberkulose und der lungenkrankheiten, 134: 73-96 ( I 'l66). 11. Crombie, J.B. Pregnancy an<l pulmonai:· tuberculosis. British journal of tubercuw,is, 48: 97-101 ( 1954). 12. Snider, D. Pregnancy and tuberculosis. Cht'sl, 86: IOS-13S (1984). 13. Schaeffer, G. et al. Pregnancy and tuberculosis. Obstehics andgyner.olngy, 46: 706-713 (1973). 14. Centers for Disease Control. Summary of notifiable diseases, United States. Morbidity and m01tality weekly report, 39: 36-37 (1990). 13. Chum, H.J. et al. An epidemiological studv of tuberculosis and HIV infection in Tanzania, 1991-1993.All)S, l 0: 2~19-309 (1996). 16. Ray, D. and Abel, R. Incidence of smear-positive tuberculosis from 1981-83 in a rural area under an acti1·e health care programme in south India. Tubertkand lung disease, 76: 190- 193 (1995). 17. Cassels, A. et al. Tuberculosis case finding in eastern l\epal. Tuberde, 63: 173-185 ( 1982). 18. National Tuberculosis Institute. Tuberculosis in a rural population of South India: a live year epidemiological study. Bulletin of the World !-frtLlth Organization, 51: 473-488 (1974). National Tuberculosis Institute. La Lubcrculose clans une population rurale de l'lndc mcriclionale: etudc epidemi<>- logique de cinq ans [resume]. Bulletin de l'Organisation mondiale de la Santi, 51: 487-488 (1974). 19. Drolet, GJ. Present trend of case fatality rates in tuberculosis. Ameriwn review of tubertulosis, 37: 12.'i-131 (1938). 20. Ettling, M.B. et al. Evaluation of Malaria Clinics in l\faesot, Thailand: Use of Serology Lo Assess Coverage. Transaction.I of Wld hlth statist. quart., 49 (1996) the Royal Society of Tropical Medicine and Hygiene, 83: 323-330 (1989). 21. Cominsk"}', S. Women and health care on a Guatemalan plantation. Social s,ienre and medirinP, 24: 1 Hi3-l l 7'.l ( 1987). 22. Lieban, RW. Gender aspects of illness and practitioner use among Filipinos. Solial science and medicine, 17: 853-839 (1983). 23. Hudleson, P. Gender differentials in tuberculosis: a review of the role of sociocconomic and cultural factors. Tubertuu• and lung disease (in press). 24. Licfooghe, R. et al. Perceptions, social consequence.sand sttLges of illntess btehrwiour: a focus grou/1 study rf tuberculosis f){l/ients in SialJwt, Pahistrm. Antwerp, Instituut voor Tropishe Geneeskuncle, 1993. 23. Barnhoom, F. & Adriaanse, H. In search of factors responsible for non compliance among tuberculosis patients in Ward ha district, India. Social science and medicine, 34: 291-306 (1992). 26. Leslic,J. et al. Weathering economic crises: the crucial role of women in health. In: Bell, O.E. & Reich, \LR. eds. Health, nut1ition and aonomir crises: aJ,j1roarhes to policy in tlw third !L'orlrl. Dover, England, Auburn Publishing Company, 1988. 27. Watts, S.J. et al. Guinea worm: an indepth study of what happens to mothers, farniliesan<l Communities.Su,:ialscienr.e and medirirte, 29: 1043-1049 (1989). 28. Strong, M.A. The Health of Adult~ in the Developing World: The View from Bangladesh. Health Transition Review, 2: 213- 224 (1992). 29. Over, M. et al. The consequences of adult ill-health. In: Feachem, R.G.A. etal. c<ls. Thehmlthofarlults in thed.eveloping world. New York, Oxford L'niversity Press, 1992. 30. Tinker, A.G. et al. Hf!mm 's lwalth and nutrition: making a diffemnrt'. \,\'ashington, D.C., The World Bank, 1994. 31. Mwabu, G. et al. Quality of medical care an<l choice of medical treatment in Kenya: An Empirical Analysis. Technical Working Paper >lo. 9. The World Bank, Africa Technical Department, Human Resources and Poverty Di\ision, Washington, D.C., 1993. 32. Gertler, P. & Van der Gaag,J. The willingness to pay for nwdirnl wre. Baltimore: Johns Hopkins University Press for the World Bank, 1990. 33. Vlassof, C. & Bonilla, E. Gender-related differences in the impact of tropical diseases on women: what do we know? Journal of biosorial scinw,, 26: '.l7-33 ( 1994). 34. Nichter, M. Illness semantics and international health: the weak lungs/TB complex in the Phillipines. Social scienu and medirine, 38: 649-663 ( 1994). 33. Selwyn, P.A. et al. A prospective study of the risk of tuberculosis among intravenous drug abusers with human immunodeficiency \irus infection. Nm Engumd journal of medicine, 320: 343-339 (1989). 36. Antonucci, G. et al. Risk factors for tuberculosis in HIV- infected persons: a prospective cohort study JAMA, 274: 143- 148 (1993). 37. Lucas, S.B. et al. The mortality and pathology of HIV infection in a West African city. MDS, 7: 1369-1379 (1993). '.18. Raviglione, M.R. et al. Global epidemiology of tuberculosis: morbidity and monalitv of a worldwide epidemic. JAMA, 273: 220-226 (1993). 39. Kaur, A. et al. Clinical and laboratory profile of AIDS in I ndia . .foumal of tlw atquired immune rleflcimty .1yndrorne, 5: 883- 889 (1992). ·10. Elliott, A.M. et al. Impact of HIV on tuberculosis in Zamhia: a cross sectional study. British medical journal, 301: 412-413 (1990). 4i. Gilks, C.F. et al. Extrapulmonary and disseminated tuberculosis in HIV-1-seropositive patients presenting to the acute medical services in Nairobi. A.IDS, 4: 981-983 (1990). 119 Double jeopardy: women and leprosy in India Carol Vlassott,a Seemantinee Khotb & Shoba Rao c Introduction d Leprosy is a disease that has been characterized for centuries by immense social stigma. Given the fact that stigmatization occurs for both males and fe- males affected by leprosy, a number of interesting questions arise. Does stigmatization tend to equal- ize status differentials between men and women, as both may become outcasts from "normal" society? Or does the lower social status of females, typical of many cultures, result in greater suffering for women? And to what extent do status differences between men and women contribute to differen- tials in the detection and treatment of the disease? This article presents evidence from two states of India, Bihar and Maharashtra, on the process of "dehabilitation" (leaving or being sent away from home because ofleprosy) among male and female leprosy patients, and suggests gender-sensitive in- terventions to address existing problems in leprosy control. While the study investigated a wide range of gender differences in the impact of leprosy, this article focuses on only two - marriage and family reactions. Gender differences in the impact of leprosy: evidence from previous research Leprosy is often reported to be more prevalent among men than women, at least in its clinical forms, at a ratio of about 2:1 (1,2). However, similar prevalence rates have been found in some areas of Africa (2, 3), and in a survey of two slum areas in Bangalore, India (4), Marshall et al. found similar numbers of males and females with leprosy. Women appear to develop stronger immunologi- cal responses to M. !,eprae which are observed dur- ing puberty, in association with increasing estro- gen and other hormone levels (1). During preg- nancy, however, these responses may be weakened, with severe consequences for the progression of " Programme '.\fanager, Special Programme for Research and Training in Tropical Diseases, World Health Organization, 1211 Geneva 27, Switzerland. h Project Coordinator, Women and leprosy project, Biometrv Department, Agharkar Research institute, Pune, India. c Head, Biometry Department, Agharkar Research Institute, Punc, India. d This is a shortened version of an article by the same title in: Khan, M.E. (ed.), \l11rk, health and contmcej1tiun from womrn'.s pl'rsputive. Centre for Operations Research and Training, Baroda, lndia (in press). 120 the disease (5). While the incidence of leprosy in infants is rare, congenital transmission of the infec- tion to the fetus appears to be possible (1). It has been observed that leprosy services fail to take account of the special needs of women, per- haps because such services are often vertically ori- entated. For example, leprosy personnel may not counsel pregnant women or those of child-bearing ages about their risks. As Duncan (5) points out: ... physicians, who - being leprocentric and not obstetrically orientated - do not notice the bump in the abdomen, the neonate in a shawl on its mother's back, or that a young child is carrying the baby that cannot be held by the mother because of her "rheumatism" and (si- lent) neuritis; nor do they think to ask the date of the last menstrual period, or last delivery as a matter of course, let alone to take an obstetric history. It is worth pointing out that it is for lack of these very observations that retrospective re- views of patients' records regarding the effect of pregnancy on leprosy are so totally mis- leading ... Similarly, women are frequently not told about the side effects of the drugs they are given for leprosy treatment. In Karachi, Pakistan, for exam- ple, Mull et al. (6) noted that women were not forewarned that one of the drugs in a multi-drug therapy regimen (MDT), clofazimine, causes no- ticeable dark stains, especially in the area ofleprosy lesions. Empirical evidence of gender differences in the social and personal impact of leprosy is scarce and inconclusive. Naik et al. (7), in a study of female leprosy patients at leprosy centres in Goa and Bom- bay, found that around one-fifth acknowledged that they had experienced domestic problems when the disease was first diagnosed. However, they said that these "initial disturbances" had been resolved and that they were satisfactorily set- tled in their families. It is not clear, however, whether these patients were resettled in their origi- nal families or had formed new unions. Mull et al. (6) highlighted several gender differ- ences with respect to treatment among leprosy pa- tients in Karachi. They noted a much higher pro- portion of men (62%) than women (38%) among patients at the leprosy centre at which the survey was based. They observed that this imbalance may have been due partly to the practice of purdah in the area and the lack of female health workers, Rapp. trimest. statist. sanit. mond., 49 (1996) which may have prevented some women from visit- ing the centre. Also, a significantly higher propor- tion of females were suffering from tuberculoid leprosy, the milder form of the disease resulting from a powerful cell-mediated immune response to the infection, whereas more males suffered from the more severe, lepromatous type. Once diag- nosed, females were more compliant with treat- ment than males, possibly because women were socialized to conform to prescribed behaviour. Although the Karachi study does not examine differences in stigma by sex, the cases of severe suffering cited are mainly female, indicating, per- haps, that leprosy has more dramatic personal conse- quences for women. For example, several women who were abandoned by their husbands said that their children were taken away from them; another had been sent to live in a cave; and others had problems marrying because a male relative was known to have leprosy. Interestingly, leprosy was widely considered to be caused by having intercourse with a "bad"woman or with a menstruating woman. Data and methods A questionnaire survey of 2 495 respondents from Bihar ( 46% of the sample) and Maharashtra (54% of the sample) was conducted. The sample consisted of 934 non-dehabilitated patients, i.e., those who were taking treatment, staying in their own families and continuing to live relatively normal lives; 1 071 deha- bilitated patients, i.e., those who had to leave their family, community or job ( or, in most instances, all of these) due to leprosy; 100 rehabilitated patient<;, i.e., those who were once dehabilitated but had been integrated into their previous social milieu; 300 members of the families of non-dehabilitated pa- tients; and 90 health workers who were treating these patients. Of the non-dehabilitated sample 59% were male and 41 % female; while in the dehabilitated sample 63% were male and 37% female. Among the rehabilitated sample, 55% were male and 45% fe- male. The structured questionnaires were supple- mented by in-depth interviews and 25 case studies (13 males and 12 females). Table 1 Findings In Maharashtra and Bihar, the male:female ratio was similar to those reported elsewhere: 1.7:1. These samples were taken from registered leprosy cases rather than from a population survey. It is likely that some cases remained undetected due to the absence of regular and close surveillance, on the one hand, and the tendency to hide the disease on the other. This was also evident from the fact that as many as 10% of women in leprosy colonies were not taking any treatment although they were aware that they were suffering from leprosy. These women were not reflected in any register. One possible explanation for the lower number of wom- en on leprosy registers is that women may be more compliant and finish treatment more quickly than men, hence being removed from registers earlier (V. Pannikar, personal communication, 1993). Gender differences in family reaction In the non-dehabilitated group the majority of pa- tients, upon learning their diagnosis, had dis- cussed it with their families. However, the propor- tion of those who shared this information was higher among females (85%) than males (76%). Status within the household was an important de- terminant of whether or not a person was dehabili- tated. The large majority (72%) ofnon-dehabilitat- ed males were heads of household, compared to only 39% of the dehabilitated; among females the proportions were 15% and 11 %, respectively. Male household heads thus "Wielded sufficient power to maintain their position in the family despite their affliction. Women, few of whom were heads of the family, suffered an additional disadvantage by vir- tue of their inferior status. Table 1 indicates the reaction of the patient's spouse on learning of the leprosy diagnosis. N ega- tive reactions such as fear, shame or blame pre- vailed in both the dehabilitated and non-dehabili- tated groups, but these reactions were more fre- quent with respect to females than males. Positive responses such as sympathy and support were rela- tively infrequent, especially for women. Reaction of spouse and sex of dehabilitated and non-dehabilitated leprosy patients: percentage distributions Tableau 1 Reaction du conjoint et sexe des malades de la lepre desadaptes et non desadaptes: distributions des pourcentages Reaction of spouse - Reaction du conjoinU de la conjointe Positive Negative - negative Total x2 Wld hlth statist. quart., 49 (1996) Oehabilitated - Exclu de chez soi M 19 7 81 93 677 394 29.30, P<.00001 Non-dehabilitated - Non-exclu de chez soi M F 32 18 68 82 552 382 22.29, P<.00001 ' 121 Table 2 Isolation of patient in selected activities and sex of dehabilitated leprosy patients: percentage distributions Tableau 2 lsolement du malade pour certaines activites et sexe des malades de la lepre exclus de chez eux: distributions des pourcentages Activities - Activites Sitting - S'asseoir Eating - Manger Shares sleeping quarters - Partager les dortoirs Fetching water - Aller chercher de l'eau Physical contact (Touch) - Contact physique (toucher) Physical contact (Sex) - Contact physique (rapports sexuels) The degree to which leprosy patients in the dehabilitated sample were isolated from family ac- tivities is shown in Tab/,e 2. While isolation was not widely admitted by either males or females, women were isolated more often from all activities. Women said that when they were restrained from their roles of cooking and caring for children they felt redundant in the family and decided to leave. For women, the most difficult adjustment was refraining from touching others, especially their children. Women are accustomed to receiving re- inforcement for their caring role by the freedom to touch and be touched. When afflicted by leprosy, patients were frequently forbidden, or themselves declined, from touching others, for fear of contam- inating them. It was touch, more than anything, that women longed for, and the loss of this inti- mate female right symbolized isolation and rejec- tion. For leprosy patients, one of the most telling signs of rejection was indifference of others to- wards them. Patients who remained in their fami- lies frequently complained of this failure to ac- knowledge their presence, symbolizing their loss of significance to others. One woman said she had been served food in a dog's dish. Another said that she would have preferred her husband to beat her because such a strong reaction would have proved that he had some feelings towards her. Indiffer- ence, by contrast, was "worse than anything". Sev- eral women said that their husbands entered into extramarital relationships after learning of their wives' illness, but said nothing about it, their indif- ference sufficing to keep their partners silent and subdued. When they finally announced that they 122 Oehabilitated - Exclus M F x2 22 37 27.98 p<.00001 18 39 57.47 p.<.00001 20 30 13.83 P<.001 15 38 18.94 p<.00001 15 25 16.83 p<.0001 9 24 31.83 p<.00001 wanted a divorce, the wives had neither energy nor power to protest. In the non-dehabilitated sample, exclusion from family and community events was uncommon (less than 20% of the time for most activities) but again, women were more often ex- cluded than men. Tabk 3 gives a summary of the critical factors that influenced the decision of dehabilitated pa- tients to leave home. Women admitted to having experienced pressure to leave more frequently than males: 68% of the males said that they them- selves had made the decision to go, compared to only 49% of females. Also, 15% of the men said that they had been advised not to leave their homes, compared to only 10% of the women. Understandably, respondents were unwilling to discuss their innermost feelings, particularly those relating to hurtful events such as leaving home, with interviewers. In the in-depth interviews, how- ever, bittersweet memories revealed the depth of the hurt experienced. For example, one woman recalled the joy that everyone in the family once shared in the celebration of the festival of lights, Divali. But on this very day she was asked to leave the family. Thus she looked back on Divali with mixed feelings. Despite the rejection, she remem- bered how pleasant it had been before this hap- pened and commented, "How nicely we used to celebrate it!" Surprisingly few patients had developed severe deformities at the time of leaving home; many more ( 49% of males and 63% of females) said that it was the fear of deformity that compelled them to leave. Women's work tends to involve the hands, and if they lose sensation, things slip from their Rapp. trimest. statist. sanit. mond., 49 (1996) Table 3 Factors influencing decision to leave home among dehabilitated leprosy patients: percentage distributions Tableau 3 Facteurs influen~ant la decision de quitter le foyer parmi les malades de la lepre exclus de chez eux: distributions des pourcentages Factors - Facteurs Who took decision? - Oui a pris la decision? Self - La personne elle-meme Family - La famille Other - Autres Fear of deformity - Crainte de la difformite M 68 20 12 49 F 49 31 20 63 x2 37.82 P<.0001 27.70 p<.00001 Deformity at time of decision - Difformite au moment de la decision Patches - Taches 41 36 12 11 46 27 14 12 7.47 p>.05 Correctable - Difformite pouvant etre corrigee Boils/knots - Pustules/nreuds Loss of digits - Perte des doigts Experienced family pressure to leave - A subi la pression de la famille pour qu'il/elle parte hands, they cannot lift heavy articles, and their fingers are burnt easily while cooking. Women play a vital domestic role, and even the thought of being useless in housework and cooking was asso- ciated with the unbearable idea of becoming a parasite. Impact on marriage Marriage can be affected in many ways by stigmatiz- ing diseases such as leprosy. These include chances of marrying, problems during marriage and ability to stay married, and possibilities of remarriage if the first marriage fails. The qualitative data from the study indicate that girls with leprosy found it more difficult to marry than boys, and that they made more compromises in the choice of partners. Children ofleprosy patients also had difficulty mar- rying and again, girls seemed to suffer more. In fact, 35% of patients said that one of the reasons they left home was worry about the marriage pros- pects of their daughters. The interviews with children ofleprosy patients provided insights into an often overlooked dimen- sion of the disease, the impact on children in fami- lies affected by leprosy, and on their marriage pros- pects. One young boy explained: We don't mind that our parents are deformed and disfigured: we have accepted that fact. But do you have an answer to why I should suffer from a disfigured future? We have a literally hopeless future. It is branded not by our poten- tial, skills and dreams but by stigma that our parents own, and that is all that we inherit. Nobody wants to give us jobs when they take our address, but if we do not give our address we get Wld hlth statist. quart., 49 (1996) 49 63 27.70 P<.00001 no job. We will marry one day to one of the girls in this community itself. So there is no curiosity, no excitement. We know who are the probable matches, and they are few. But the situation of girls in our colony is worse. At least we boys go out and spend time on the main road, listening to songs and watching televisions that are displayed in the shops in the market. But these girls do not go out. You know we brew and sell liquor and all sorts of people come here. If any of these girls is found to be loose and outside the colony, the men won't leave them alone. Their world remains very small within the bounds of this colony. And then we boys find them very boring as partners. While most respondents (68% of males and 62% of females) reported that they were married, the term "married", in the context of this popula- tion, is misleading. Those who reported that they were married were not necessarily with their origi- nal spouses. On rechecking, many dehabilitated respondents admitted that they were previously married and had left first spouses when they be- came noticeably disfigured. In fact, most of the leprosy-affected couples living in colonies were in common-law unions. More wo~en than men reported that they were married. In-depth interviews revealed that many female patients continued to say they were "mar- ried" even though they were not living with their husbands, and sometimes had not seen their spouses for many years. They did so as a security mechanism, both because it was psychologically comforting and because it protected them from the possible advances of male patients in leprosy 123 homes (treatment centres) or colonies. For exam- ple, a 70 year-old woman who had claimed that she was married later disclosed that she had not even seen her husband for 50 years. She explained that she was married, "but presently he is not living with me". Generally, women were less likely to mention marital problems than men, but three factors stood out as distinguishing women who stayed with their families from those who left. These were, firstly, a strong marriage, in which the husband supported his wife in seeking early treatment. Secondly, prompt detection of the disease was important in preventing dehabilitation, as it could be arrested early with less possibility of disfiguration. In the non-dehabilitated group, 89% of married males and 95% of females had sought treatment at the early stages of the disease, whereas in the dehabili- tated sample, nearly one-third of males and females had severe damage before seeking care. Finally, whether a woman had children appeared to play a role in whether or not she was able to stay with her husband. In the dehabilitated group, of the women who said they had borne children, those currently married had an average of2.6 chil- dren as opposed to only one child for divorced females. More interesting, however, was the number who claimed that they had never had chil- dren - 119 of the 227 women - of whom only 24 were unmarried. This indicates considerable childlessness and sub-fertility for leprosy-affected women, and their dehabilitation status also implies a probable link between childlessness and depar- ture from their original marital union. Non-deha- bilitated married women had 2.5 children on aver- age, and only 6% of them were childless; their higher status due to motherhood may have helped them to remain in their families. Almost all women who left their homes before the 1970s were sent to, or dropped at, leprosy homes (treatment centres) by their families. At the time of leaving, half of them were unmarried and the rest said that they did not have children or that their husbands were living with other women. In the leprosy colonies which have grown up around leprosy homes, and which are growing in number as a result of the increasing cure rate and released, cured inmates, leprosy patients were making new alliances with other ex-patients. All the women who had joined a new union had settled down with severely handicapped men suffering from deformi- ties that were more advanced than their own. One remarried woman, wife of a leprosy colony leader explained, "Why would they take us if our hands were wasted? They have married our hands which cook and clean and sweep!" Among the 72 married couples in the four lep- rosy colonies, 45 households had both male and female leprosy patients. In the remaining house- holds 16 leprosy-affected men were married to 124 non-leprosy-affected, able-bodied women, four of whom were brought from an orphanage adjacent to the leprosy home. Only one female leprosy pa- tient was married to a non-leprosy-affected man who was physically handicapped due to a road acci- dent. Male leprosy patients in the colonies thus made fewer compromises than women in the choice of spouse or companion. Patients· understanding of treatment The study revealed insights concerning the way leprosy patienL~ experienced and interpreted the disease. For example, women were very concerned about the fact that their urine turned dark orange- yellow as a result of rifampicin, one of the drugs used in MDT. Women associated the change in urine colour with jaundice, which was considered as one of the deadliest diseases. Women and health workers alike were also un- aware of the possible side-effects of other drugs, such as prcdnisolone. Prednisolone is contraindi- cated for pregnant and lactating women because it can make their bones brittle and can cause severe bone pain. In an area where women already are deficient in calcium, iron and vital nutrients, the effects of the drug could be especially problematic. Yet when prednisolone or steroids are used, women report instant improvement and demand the same tablets repeatedly. Women's interpretation of the medicines re- ceived, in terms of perceived attributes such as colour, size and smell, was also enlightening. Dap- sone, for example, is very small in size and women tended to see it as less effective than the glossy, brightly coloured and attractively packaged multi- vitamin tablets. Further, the small tablets musr be taken in a rather rigid regimen that was difficult for women to follow: before going to bed every night and preferably not on an empty stomach. A "blis- ter pack" had been introduced to simplify the pro- cess, but this gave rise to the suspicion among women, especially mothers-in-law, that they were contraceptive pills being distributed surreptitiously by the family planning programme. Conclusions Several conclusions relating to improved leprosy control can be derived from these results: 1. One of the greatest obstacles to leprosy treatment is late reporting of signs of the disease. Because of social stigma, women may be afraid to disclose early patches when visited by leprosy workers. As women were very concerned about the condition of their skin, a possible entry point for detection of leprosy is that of skin disease. Asking women about the condition of their skin, rather than about "patches" as such, is more likely to be acceptable, and seeking treatment for skin ali- ments is much less subject to social stigma. Train- Rapp. trimest. statist. sanit. mond., 49 (1996) ing in skin disorders for health workers could also help them distinguish signs of leprosy from other, similar or overlapping problems. 2. Our study has pointed out the importance of people's perceptions of medicines. Smaller pills may be viewed as less effective, or be confused with contraceptive pills. It is important for leprosy work- ' ers to be aware of these concerns and alleviate them, as much as possible, with female patients. Family members should also be informed about the role of the medicines for curing the disease, and the importance of carefully following pre- scribed regimes. 3. Women should he informed about the risks of pregnancy when they have leprosy and of the need to carefully protect themselves by MDT dur- ing this period. Both women and health workers should be fully informed about possible negative side effects of drugs that should not be adminis- tered to pregnant women. More attention also needs to be given to the special problems of lepro- sy patieuts giving birth, when sensation in certain parts of the body may he lacking. 4. Following the MDT regimen is complicated, and the amount of drugs to he administered de- pends on type of disease, body weight and state of health of the patient. Simple calendars should be developed to help leprosy patients follow their reg- imens, particularly for women who are illiterate, non-numerate and hence at. considerable disad- vantage in treating their condition. The role of literate family members can be especially impor- tant in these situations. 5. Gender sensitization of leprosy workers may be an important step to more effective disease con- trol. For example, leprosy workers should be in- formed about the need for different approaches to women and men in their work. In India, most leprosy workers are male, and being touched or examined by men may be unacceptable to women. The involvement of local women can help to allay such fears or concerns. In some areas, especially where women are in purdah, female specialists may be required to perform these examinations. 6. The distinctive problems and needs of chil- dren of leprosy patients have received very little attention. Even though most of these children are not infected, they suffer severe deprivation in their daily experience and few chances for a better life. Special attention should be given to this group, in terms of assuring equal access to education, jobs and health care. Summary This article presents evidence from two states of India. Bihar and Maharashtra, on the process of "dehabilita- tion" among maie and female leprosy patients, and suggests gender-sensitive interventions to address ex- isting problems in leprosy control. While the study inves- Wld hllh statist. quart .• 49 (1996) ligated a wide range of gender differences in the impact of leprosy, this article focuses on only two - marriage ard family reactions. Important gender differences were apparent in the impact of the disease. While both men and women were negatively affected in terms of their family and marital lives. women suffered more isolation and rejection. Psychologically, women ap- peared more vulnerable because they were deprived of personal contact with others in the domestic environ- ment where they were accustomed to receiving their greatest emotional rewards. Women reported that indif- ference to them by other family members. or seeming negation of their presence. caused them the greatest suffering. This underscores the importance of providing information to both leprosy patients and their families about the disease and its treatment, including the pos- sibility of cure with MDT (multi-drug therapy) and of counselling family members about their crucial role in helping patients cope and recover. This support is even more critical for women. who often lack access to the variety of outside advice and assistance available to men. The evidence presented in the article demon- strates the importance of analysing leprosy from a gender perspective, not only because this approach helps to inform our understanding of the determinants and consequences of the disease, but also because it provides new insights for improved disease control. Resume Double peril: Jes femmes et la lepre en lnde Cet article presente des temoignages en provenance de deux Etats de l'lnde, le Bihar et le Maharashtra, sur le processus de «degradation» survenant chez les malades de la lepre. hommes et femmes, et suggere des interventions tenant compte du role specifique des deux sexes pour aborder les problemes de la lutte centre la lepre. Si l'etude a porte sur une vaste gamme de differences entre les sexes concernant l'impact de la lepre, le present document n'est axe que sur deux aspects: le mariage et les reactions de la famille. D'im- portantes differences entre les sexes sont apparentes pour ce qui est de l'impact de la maladie. Tant les hommes que les femmes sont touches de maniere negative dans leur vie familiale et leur vie maritale, mais les femmes souffrent davantage d'isolement et de rejet. Psychologiquement, les femmes semblent etre plus vulnerables car elles sont privees de contact personnel dans le milieu domestique ou elles avaient l'habitude de recevoir les plus grands stimulants affectifs. Les fem- mes ont fait etat de !'indifference qu1 leur est manifestee par les autres membres de la famille, qui pouvaient aussi faire semblant d'ignorer leur presence. et cela leur causait les plus grandes souffrances. Cela montre !'im- portance qu'il y a a fournir des informations tant aux lepreux qu·a leur famille sur la maladie et son traitement (y compris la possibilite d'une guerison grace a la polychimiotherapie) et a conseiller les membres de la famille sur le role special qu'ils peuvent jouer pour aider les malades a faire face a la rnaladie et a guerir. Cet appui est encore plus essentiel pour les femmes. qui. souvent n'ont pas acces a la gamme de conseils exte- rieurs et d'assistance dont disposent les hommes. Les 125 temoignages presentes dans ce document montrent qu'il est important d'analyser la lepre du point de vue de chaque sexe, non seulement parce que cette approche nous aide a comprendre Jes determinants et Jes conse- quences de la maladie, mais egalement parce qu'elle offre de nouvelles perspectives permettant de mieux !utter contre la maladie. References - References l. Ulrich, M. et al. Leprosy in women: characteristics and repercussions. Social science and medicine, 37 ( 4): 445-456 (1993). 126 2. Noordeen, S.K. The epidemiology of leprosy. In: Hastings, R.C. (ed.) Leprosy. Edinburgh, Churchill Livingstone, 1985 (pp. 15-30). 3. Fine, P .E.M. Leprosy: the epidemiology of a slow bacterium. Epidemiologi,cal review, 3: 161-188 (1982). 4. Marshall, J.T. et al. Prevalence of leprosy among slum dwellers. Leprosy in India, 53(1): 70-82 (1981). 5. Duncan, M.E. etal. The association of pregnancy and leprosy II: Pregnancy in dapsone-resistant leprosy. Leprosy revi.ew, 52: 263 (1981). 6. Mull,J. D. et al. Culture and "compliance" among leprosy patients in Pakistan.Socialscienceand medicine, 29(7): 799-811 (1989). 7. Naik, S.S. et al. Problems and needs of women leprosy patients in Bombay and Goa - a preliminary report. Indian journal of !,eprosy, 63(2): 213-222 (1991). Rapp. trimest. statist. sanit. mond., 49 (1996) Women and smoking: a global issue a Amanda Amosb Introduction Smoking kills over half a million women around the world each year and this number is increasing ( I). It is estimated that between 1950 and 2000, 10 million women will have died from their smok- ing habit. In several developed countries, such as the United States of America and the United King- dom, cigarette smoking is the single most impor- tant preventable cause of premature death in women, accounting for at least a third of all deaths in women aged 35 to 69 ( I). However smoking is still regarded in many countries as being a mainly male problem. This article explores the reasons behind this widespread view, and considers why more attention needs to be paid to issues around smoking among women. In particular it examines whether the factors influencing smoking uptake and cessation are the same for women and men. The experience of countries with the longest histo- ry of widespread female smoking is used to identify key issues for developed and developing countries, and to argue the case for gender-specific and gen- der-sensitive approaches to tobacco control. The focus is primarily on cigarette smoking, the most prevalent form of tobacco use among women. However, when relevant, reference will be made to other types of traditional tobacco use which are common in certain parts of the world. Patterns and trends - the growing epidemic A cursory look at the data on cigarette smoking in different countries might suggest that there are no consisten.t patterns or trends among women (Tabl.e I). Prevalence rates vary from 58% in Nepal and over one-third in European countries such as Denmark, to barely detectable levels in many Afri- can countries such as Cote d'Ivoire (2). More con- sistent is the finding that cigarette smoking rates are higher among men than women. However countries vary considerably, with nearly equal rates in the United States and United Kingdom, but a several-fold difference in China where less than 6% of women are daily smokers compared to 56% of men (Tabl.e I). However these variations are not random but reflect different stages of the smoking • A longer version of this article appeared in the British =diral bulletin, 52 ( 1), 1996. b Department of Public Health Sciences, Medical School, University of Edinburgh, Edinburgh, Scotland. Wld hlth statist. quart., 49 (1996) Table 1 Prevalence of cigarette smoking among women in selected countries Tableau 1 Prevalence de l'usage de la cigarette chez les femmes dans certains pays Region - Region Americas - Ameriques Bolivia - Bolivie Brazil - Brasil Honduras Trinidad and Tobago- Trinite-et-Tobago United States of America- Etats-Unis d'Amerique Europe Denmark - Danemark France Poland - Pologne Portugal United Kingdom- Royaume-Uni Africa Cote d'Ivoire Nigeria - Nigeria Swaziland Zambia - Zambie Eastern Mediterranean Bahrain - Bahre"in Egypt Sudan - Soudan Tunisia - Tunisie South East Asia India- lnde Indonesia - lndonesie Nepal - Nepal Thailand - Tha"ilande Western Pacific Australia - Australie China - Chine Japan - Japan Malaysia - Malaisie Prevalence - Prevalence(%) 38 33 11 5 26 45 30 35 12 28 1 10 7 4-7 20 2 19 6 0-673 10 58 4 27 8b 14 5 Date of survey- Date de l'enquete 1986 1990 1988 1986-89 1990 1988 1991 1989 1988 1992 1981 1990 1989 1984 1985 1981 1986 1984 1990 1991 1988 1986-89 1984 1990 1990 a Depends on area surveyed. -Depend de la region sur laquelle porte l'enquete. b See -Voir Ref·Ref. 38. Source: Ref·Ref. 2. 127 Fig.1 A model of the cigarette epidemic Un modele de l'epidemie de tabagisme 70 Stage - Stade I Stage - Stade II _ ....§.ta-2:- Stade Ill .-~~~~.--~~~~~~---.~~~~~~~-r-S-t-ag-e---St-ad_e_lV--,40 / ' 60 / ' I"' % males- I ' r.,,!!:l %d'hommes 1 ===i 50 \. ::J-o 'O"' I %females- "'"' %de femmes Cl<l> I c- 40 ON E~ I I <,l(.) "'"' 30 I I Q) :i I ..>< <l> I OE I E:::, / r.n- 20 -<» / / 0-0 ?f!.-;f?.. / / 10 / / .,., / / -O·~~::::::t::::::~~:::::::±::::::~~j___L___jO 0 10 20 30 40 50 60 70 80 90 100 Number of years - Nombre d'annees WH09ti716 Source: (Ref. - Ref. 1) epidemic in each country. A 4-stage model, based on evidence from countries with the longest histo- ry of cigarette smoking, describes the typical evolu- tion of cigarette smoking and subsequent mortality in a country (Fig. 1) (3). While the exact picture in each country will vary to some degree, countries can be categorized into one of four different evolu- tionary stages: Stage 1 (e.g., many developing countries mainly in sub-Saharan Africa): Male prevalence rates less than 15% but increasing rapidly, female preva- lence less than 5%. Health consequences not yet apparent. Stage 2 (e.g., China, Japan, several countries in Asia, Latin America and North Africa): Male preva- lence rates rising rapidly to 50-80% with few ex- smokers. Female rates lagging 10-20 years behind but increasing. Male smoking-related death rates starting to increase. Stage 3 (e.g., Eastern and Southern Europe): Male prevalence starting to decline, reaching around 40%. Female prevalence peaks and plateaus at a lower rate than among men and starts to decline. Stage 4 (e.g., United States, United Kingdom, Canada, Western Europe): Smoking declining slowly in both sexes. Male mortality from smoking peaks as female deaths begin to rise rapidly. This model has several important features relat- ing to women and smoking. I. Women generally take up cigarette smoking as a widespread habit after men. This is mainly due to sociocultural factors, such as it not being socially acceptable for women to smoke in public, religious attitudes, and women generally being less affluent than men and unable to afford cigarettes. However this model indicates that countries ·with low female smoking rates should not be complacent. If ciga- rette smoking is increasing among men then it is 128 likely to increase among women in the future. In- deed, while in the United Kingdom and the United States there was a 20-30-year lag between smoking becoming a widespread habit in men and women, evidence from Stage 2 and 3 countries suggests this lag may be diminishing, particularly where there is aggressive promotion of cigarettes to women and/ or rapid changes in women's socioeconomic posi- tion. The challenge for countries in Stages 1 and 2 is to take action now to counter the factors that may encourage smoking among women. 2. Even where female smoking is increasing (Stages 2 and 3) there may be little concern about this trend as female rates of smoking will be consid- erably lower than those in men and the health consequences for women will not yet have emerged. However the model predicts that unless effective action is taken, smoking is likely to contin- ue to increase among women and become a major cause of death. However this is not inevitable. For example, the approach adopted by Singapore in the 1970s, when it was at the beginning of Stage 2, showed that implementing a strong and compre- hensive tobacco control policy can reverse this trend and the progression from Stages 1 to 2. Female smoking declined from 9.5% in 1978 to 3% in 1992 (3). 3. Women who are most likely to start smoking in large numbers are affluent, well-educated and live in urban areas (Stages 1 and 2). For example, in Costa Rica a survey found that 24% of affluent, urban women were smokers compared with only 10% of poor, rural women (4). In Spain, in 1989, 52% of upper class compared to 12% of working class women smoked (5). More advantaged women are also more likely to show the first declines in smoking, reflecting both lower rates of uptake and higher cessation rates (Stages 3 and 4). Thus in Rapp. trimest. statist. sanit. mond., 49 (1996) countries where smoking is declining, smoking is increasingly associated with disadvantage. In these countries the typical female smoker has a limited education, a low status job or is unemployed, is on a low income and experiences high levels of depri- vation. For example, in the United Kingdom in 1992, 13% of women in socioeconomic group 1 smoked compared to 35% in socioeconomic group 6, and whereas 62% of professional women who had ever smoked had quit, only 36% of un- skilled manual women had succeeded in quitting (6). Also in many countries the highest smoking rates are among disadvantaged ethnic minorities. In Canada in 1989, 77% of Inuit women were smokers (7) and in New Zealand, Maori women have one of the highest smoking rates among women in the world (2). 4. Because of the time lag between smoking uptake and the health effects, even countries with the longest history of female smoking have not yet experienced the full impact on women's health. Because men in developed countries were the first to smoke cigarettes in large numbers, most of the epidemiological research has been carried out on these male populations. Thus our understanding of the true impact of smoking on women's health is limited. There is, however, strong evidence that where women have smoked cigarettes regularly for several decades, the percentage of female deaths attributable to tobacco is approaching the male figure (1). Among women in developed countries who have smoked regularly throughout their adults lives, tobacco will cause at least half and perhaps substantially more of all deaths in middle age. Whatever the stage of the smoking epidemic within a country, action is needed now to halt its Fig. 2. progression for there is no doubt that smoking causes the same diseases in women as in men (Fig. 2). The main tobacco-related killers in both sexes are cancers, especially lung cancer, heart dis- ease and chronic bronchitis. But smoking also af- fects women's health in ways which are specific to them. In developed countries (2) women who smoke have a 10-times higher risk of heart disease and an increased risk of stroke if they use oral contraceptives; have a 2-fold associated higher risk of cervical cancer; and experience detrimental ef- fects on their reproductive health, including dys- menorrhoea, reduced fertility and earlier meno- pause. Smoking during pregnancy increases the risk of miscarriage by one-quarter and the risk of perinatal death by a third, doubles the risk of pre- mature labour, and trebles the likelihood of a low- birth-weight baby (8). In developing countries where the health of mother and baby is already jeopardized through poverty and malnutrition, the effects of smoking are likely to be even greater. In Chile, for example, it is estimated that 10% ofnon- accidental perinatal deaths are attributable to smoking (9). In developing countries with high levels of tra- ditional tobacco use among women, such as chew- ing or smoking with the lit end of the chutta inside the mouth, there is an association with high levels of oral cancer. Smoking causes around one in five of all cancers in women in India ( 10), with women in Bangalore having the highest rate of mouth cancer in the world (2). Why women smoke To develop effective tobacco control strategies for women it is important to consider to what extent Estimated distribution of deaths from various smoking-related diseases among women in developed countries, 1985 Distribution estimative des deces dus a diverses maladies liees au tabac chez les femmes des pays developpes, 1985 Other respiratory diseases Autres maladies des voies respiratoires 2.7% Chronic obstructive pulmonary diseases (primarily chronic bronchitis and emphysema) Maladies pulmonaires chroniques obstructives (surtout bronchite chronique et emphyseme) 18.1% Other causes Autres causes 11 .0% Lung cancer Cancer du poumon 21.1% Cancer of the mouth, oesophagus, pharynx and larynx Cancer de la bouche, de l'oesophage, du pharynx et du larynx 2.4% Other cancers - Autres cancers 3.7% Cardiovascular disease Maladies cardio-vasculaires 41.0% Total 300 OOO deaths - Total 300 OOO dilC8S Source: (Ref. - Ref. 2) Wld hlth statist. quart. , 49 (1996) 129 factors which influence smoking uptake and cessa- tion differ between men and women. This section looks at some of the main findings from research which has explored these issues. Most studies have been carried out in developed countries which have the longest history of female cigarette use. However many of the issues raised are also likely to be relevant to developing countries. Starting the habit A recent WHO study of 10 European countries found that over one-third of girls have tried smok- ing by the age of 13, and this increases to around 60% among 15 year-olds ( 11). The initiation of smoking in girls, as in boys, is heavily influenced by social pressures and psychological needs including environmental influences, school and peer influ- ences, personal factors, and knowledge, attitudes and beliefs about smoking ( 12). Creating the market. The tobacco industry is depen- dent on a mass market. As smokers die or quit, the industry is keen to recruit new, young smokers to maintain profits, particularly in new markets such as developing countries and Eastern Europe. Mar- keting aims to tailor a product to appeal to target groups by altering its price, availability and image through packaging, advertising and promotions. Over the past few years the tobacco industry has targeted women by: (i) promoting brands through advertisements and sponsorship using images and messages which promote smoking as being glamor- ous, sophisticated, romantic, sexy, healthy, sporty, fun, relaxing, liberated, rebellious and slimming; (ii) producing "women only" brands and other types of cigarettes likely to appeal to women; and (iii) advertising in women's magazines to reach large numbers of women and discourage reporting about the health risks. Magazines dependent on tobacco revenues are less likely to cover the health hazards or take anti-smoking advertisements (13-17). Tobacco companies argue that cigarette adver- tisements do not encourage smoking but simply affect brand choice. Yet research in developed countries shows that children can identify the brands of edited cigarette advertisements and that awareness of cigarette brands is a strong predictor offuture smoking (18). Young teenage smokers are more appreciative of cigarette advertising than non-smokers and the most heavily advertised brands are more often bought by teenagers than adult smokers ( 19, 20). Many developing countries have few restric- tions on tobacco promotion and thus the influence on girls, who are likely to be less knowledgeable than adults about the harmful effects of smoking, may be even more powerful. Cigarette advertising in developing countries at present tends to be di- rected at the general public, but there is evidence 130 that women are becoming targets. In Hong Kong, where less than 1 % of young women smoke, one large tobacco company launched a major cam- paign to promote a brand targeted at women - a clear attempt to create a market among young women (21). In India an attempt has also been made to launch a "women only" brand. However, where comprehensive restrictions on tobacco pro- motion have been introduced there has been an immediate fall in smoking prevalence among young people, especially girls (22). How smoking is portrayed in the media more generally also affects the way young people view the habit. Glamorous models, female personalities, teenage pop idols and film stars feature in maga- zines, television soap operas, and films which de- pict smoking as being part of their success. These images reach many different audiences around the world and may create aspirational images of west- ern life which are both false and harmful. In South Asia, for example, concerts by Madonna and Paula Abdul have been sponsored by tobacco companies. The social setting. Adolescents are more likely to smoke if their parents smoke or have permissive attitudes about smoking. In developed countries, girls in particular appear to be influenced by their parents smoking habits and attitudes, although this decreases as they get older (12). The situation is much the same in developing countries and coun- tries where religious and cultural mores are ap- plied more strictly to girls than to boys (22). In countries where cigarette smoking is comparatively new among women, it is affluent professional wom- en who adopt the habit first (23). This may be due to the more liberated environment in which they live and work, their relative affluence and their urban environment which exposes them to tobac- co advertisements and makes cigarettes more ac- cessible. First cigarettes are usually smoked with friends and having a best friend who smokes is a strong predictor of becoming a smoker, though this may be more important for boys than girls (24). Adoles- cent smokers are more likely to be underachievers at school with low academic goals. In the United States in the 1980s women without a college educa- tion were over twice as likely to take up smoking than those who went to college (12). Personal factors. Many young people in developed countries experiment with drugs, such as tobacco, in an attempt to achieve the image of maturity, sophistication, attractiveness, sociability, masculini- ty or femininity to which they aspire. Of particular importance to girls are concerns about weight and self confidence. In Western countries, where the media promote an image of female attractiveness which equates being thin with desirability, weight control and dieting are major obsessions among adolescent girls (25). This concern has been picked Rapp. trimest. statist. sanit. mond., 49 (1996) up and used in advertisements for certain cigarette brands aimed at women which associate smoking with slimness and glamour (26). Research has repeatedly found that girls with low self-esteem are more likely to take up smoking. Adolescents who feel that they have a lot of control over their health and life are less likely to become smokers than those who feel they have little control (27). Using smoking to bolster self-confidence stems from the widespread belief that smoking helps calm nerves, control moods and alleviate stress -all important concerns during adolescence. By showing attractive young women with hand- some male partners or socializing with successful and confident people, tobacco advertisements ex- ploit young people's insecurities, and sell the idea that these desirable qualities are theirs if they smoke. There is also evidence that girls feel more dependent on cigarettes, compared to boys smok- ing similar amounts, and imagine that they would find giving up more difficult (28). Knowledge and attitudes. Whether a girl becomes a smoker also depends on her knowledge about the health risks, and whether these are personally rele- vant and outweigh the perceived benefits of smok- ing. In many developed countries school health education programmes have increased young people's knowledge about the health effects of smoking, their awareness of influences such as ad- vertising and social pressure, and helped develop their confidence, self-esteem and social skills to resist smoking. However it is not clear whether they ultimately reduce smoking or simply delay its onset (12). Since girls are more likely to believe that smoking helps them deal with stressful situations, developing their self-esteem and competency to solve problems is likely to be an important strategy. In developing countries, young girls' knowledge about smoking and health is likely to be limited due to the lack of health education programmes and structural barriers such as thinly spread rural populations and high levels of illiteracy. Why women stay hooked Putting up a smoke screen. As public awareness about the health effects of tobacco has grown, tobacco companies in many countries have increased the amount and variety of tobacco products and pro- motions targeted at women in an attempt to allay health fears (15,26). These include cigarettes which are lower in tar, lower in nicotine or men- tholated. Many women have changed to low-tar cigarettes believing that this reduces their health risks. There is, however, little medical evidence to support this view (29). The pressures on women. In countries with the long- est history of widespread female smoking, smoking is now most common among those on low in- Wld hlth statist. quart., 49 (1996) comes, who have low status jobs or are unem- ployed, are single parents or divorced, have low levels of academic achievement and are from un- derprivileged ethnic groups. One reason why these women continue this habit is that they believe that cigarettes help them cope (25, 30). A woman on a low income, tied to the home, bringing up small children, may smoke to deal with her feelings of stress. Although recognizing that it may ultimately damage her own health, she may feel that smoking is less damaging to her children than "letting off steam" some other way. Similarly women at home may structure their day with cigarettes, providing excuses for breaks and refuelling. Many women on low incomes see cigarettes as their only luxury- the only thing they do for themselves. Women in low status, repetitive and insecure jobs may also smoke to break the monotony or deal with their frustra- tion about the work. In reality smoking probably does little to relieve stress or reduce feelings of anger, but if women believe that it does, they are unlikely to feel that they can cope competently without cigarettes. Gaining weight- the price of giving up? Many women in developed countries believe that smoking helps control their weight and quitting leads to weight gain (31,32). Studies have shown a weight gain around 5-10 lbs (1.5-4.5 kg) after giving up smok- ing among some women. Even though the health costs of a small weight gain are more than offset by the health benefits of quitting smoking, for many women this is a high cost to pay in terms of their self-image. This needs to be recognized. An addictive habit. As a young woman starts to smoke regularly her body gets used to regular nicotine doses and she becomes physiologically addicted. She develops a pattern of daily smoking which together with having family and friends who smoke, and her beliefs about smoking, reinforce this dependency. To quit, women need help and support to overcome short-term physiological with- drawal, and to break behavioral patterns that may have developed over many years. Kicking the habit Cessation clinics in developed countries have tended to show that women are less successful than men in quitting (33) but this cannot be generalized to all women since the vast majority who quit do so independently. That women in the United States, United Kingdom and Australia are now giving up at about the same rate as men also refutes this notion (34). These figures however do not show whether women make more attempts at quitting than men before they are successful. Disadvantaged women are less likely to give up smoking than more affluent women (2,25,30). If these women are to be empow- ered to take control of their lives, it will be necessary 131 to adopt strategies that address not only their smok- ing but also the social and economic circumstances that reinforce their habit (35). In most countries where smoking is declining, the majority of women want to give up. Many have attempted to quit at least once and most make several attempts before they are r.uccessful. Confi- dence in one's ability to quit and the desire and resolve to succeed are of crucial importance (36). Most people go through a process that involves pre-contemplation, contemplation, action, mainte- nance and relapse. Cessation programmes that ad- dress all stages of the quitting process are likely to be the most effective, particularly if they are backed up by measures aimed at changing the social and environmental factors which make giv- ing up easier and support non-smokers. Issues for action Smoking is a complex issue and controlling its spread requires a comprehensive approach. While similar factors influence smoking among men and women there are important differences, often relating to girls' and women's own social worlds, which need to be taken into account. To achieve the overall aim of helping young women to resist pres- sures to start smoking and helping female smokers to quit, tobacco control strategies need to be both gender-sensitive and gender-specific and should encompass 3 areas of action: research, public policy and legislation; and education and support. Research To develop and evaluate effective programmes, countries need comprehensive and reliable data about women and smoking and smoking-related diseases. There needs to be a better understanding of women's reasons for smoking and quitting, and the factors which encourage smoking such as ciga- rette promotion aimed at girls and women. Fur- ther research is needed on the role that smoking plays in women's daily lives and how health promo- tion can help women develop alternative ways of dealing with the factors that keep them smoking. Public policy and legislation The aim is to create a social, economic and politi- cal climate which promotes non-smoking as the norm and reduces countries' economic depen- dence on the production, manufacture and sale of tobacco. This requires action at the international, national and community levels. It is essential that women become more visible and that the issue of women and smoking is placed high on the health agenda. This will be facilitated through exchang- ing information, expertise and research. Many countries have yet to acknowledge the potential seriousness of this issue. It is therefore important that health professionals and others seize the initia- 132 tive and raise awareness about the threat that smok- ing poses to women's health. Education and support Girls and women need knowledge, attitudes and skills to make informed decisions about smoking. Programmes should be culturally appropriate, rel- evant to women at different points in their lives, and related to the stage of the smoking epidemic within each country. There is a need for both gen- der-sensitive and gender-specific programmes. For example, women tend to have more contact with health services, creating considerable opportuni- ties for health education. Health education should also form an integral part of school education. However in many developing countries girls have limited access to schooling and, in countries in Stages 3 and 4, smoking is more common among those whose who are disenchanted with school. It is therefore important to involve organizations and networks which reach women in different ways such as youth organizations, community groups, women's organizations, workplaces and the media, including women's magazines. Conclusion Smoking is a major cause of ill-health and prema- ture death among women and this is increasing. Even in countries where smoking is still low among women, many women's lives are negatively affected by smoking, through their husbands spending scarce resources on cigarettes, their exposure to second-hand smoke, or having to cope with a spouse's death from smoking (37). While religious and cultural attitudes, often combined with poor economic status, have kept female smoking low in many countries, history shows that unless strong, comprehensive, tobacco control policies are imple- mented, female smoking will increase. The tobac- co industry has identified women as a key target group around the world. Countries with newly opened markets, such as China and Eastern Eu- rope, or those which permit tobacco promotion, are particularly vulnerable to targeting by the to- bacco industry. Also vulnerable are women in countries undergoing fast urbanization or industri- alization, where tobacco promoters attempt to as- sociate smoking with affluence, sophistication, mo- dernity and success. Firm action needs to be taken now to halt and ultimately reverse this epidemic. Summary Smoking kills over half a million women each year and is the single most important preventable cause of female premature death in several developed countries. How- ever in many countries smoking is still regarded as a mainly male problem. This paper explores the reasons why more attention needs to be paid to smoking and Rapp. trimest. statist. sanit. mond., 49 (1996) women, even in countries with low levels of female smoking. Included is an overview of the patterns and trends of smoking among women, the factors which influence smoking uptake and cessation in women. and some of the key challenges facing developed and developing countries.It is argued that to be successful in addressing the tobacco epidemic among women. tobacco control policies need to encompass both gen- der-specific and gender-sensitive approaches. Resume Les femmes et le tabac : un probleme mondial Le tabac tue plus d'un demi-million de femmes chaque annee et constitue l'unique et la plus importante cause evitable de deces premature des femmes dans plu- sieurs pays developpes. Cependant. dans de nom- breux pays, le tabac est toujours considere comme un probleme essentiellement masculin. Le present docu- ment explore les raisons pour lesquelles ii taut consa- crer davantage d'attention a la question du tabac et des femmes, meme dans les pays ou le nombre de fumeu- ses n'est pas aussi important. On y trouvera un aperi;:u des schemas et des tendances lies a l'usage du tabac chez les femmes. les facteurs qui influencent l'usage du tabac et sa cessation chez les femmes, ainsi que quelques-uns des problemes cles auxquels sont con- frontes les pays developpes et en developpement. On fait valoir que pour reussir a faire face a l'epidemie de tabac chez les femmes, les politiques de lutte antitabac doivent englober des approches a la fois sensibles aux disparites entre les sexes et des approches fondees sur le sexe. References/References 1. Peto, R. et al. Mortalityfromsnwking in d,,,1eluf1ed countries 1950- 2000. Oxford, Oxford l!niversily Press, 1994. 2. Chollat-Traquet, C. Womni and tobacro. Geneva, World Health Orgauizalion, 1992. ,\. Lopez, A.D. et al. A descriptive model of the cigarette epidemic in developed countries. Tobarr.o conlro~ 3: 242-247 (1994). 4. Rosero-Bixby, L. & Oberle, M.W. Tobacco use in Costa Rican women. Ciencias sociales, 35: 93- l 02 ( 1987). 3. F..1tudio de ws estiws de vu/a de la /1obl.acion adulta esf1a,inla. Madrid, Minislerio de Sanidad y Consumo, Direci6n General de Salud Publica, 1992. [in Spanish]. 6. Office of Population Census and Surveys. General Jwuseho/.d . ,uruey 1992. London, Her '.\fajesty's Stationery Office, 1994. 7. Miller, W. & Peterson,J. Tobauo u.>e l,y youth in the Canadian Arrtic, 1989. Ottawa, Health and Welfare Canada and Ministry of Health N\,\'T, I 989. 8. Smoking and the young. London, Royal College of Physicians, 1992. 9. United States Surgeon General. RPducing the Jualth mnsequences of smoking: 25 years of progmss. Bethesda, Maryland, United States Department of Health and Human Ser;ices, 1989. 10. Stanley, K. & Stjernsward, J. Lung cancer in developing countries. In: Hansen, H.H. (ed.) Ba.sicclinicalmnceptsoflw1g cancer. Boston, Kluver Academic Publishers, 1989. 11. Cunie, C. et al. Health behaviours of Scottish school children: &port 4: The cross-national penpative: Scotland comj1ared to other European amntri,is and Canada. Edinburgh, Health Education Board for Scotland, 1994. Wld hlth statist. quart., 49 (1996) 12. Preventing tobacco use amongyoungfieoph!: a refwrt of the Surgeon Ge1u>ml. Atlanta, United States Department of Hcahh and Human Services, 1994. 13. Ernster, V.L. 1'-Iixcd messages for women: a social hi~tory of cigarette smoking and advertising. J\'nv York State journal of m.edirirtP, 316: 72:i-732 (1983). 14. Amos, A. & Bostock, Y. Pulling women in the pie/um: rigarel/e advntising Jiolicy and tol'nage ofmwking and health in women's mogazin" in Eurr1J1:. London, RM:\, 1992. 1:i. Amos, A. I low women arc targeted by the tobacco indmtry. World health forum. 11: 41&-422 (1990). Hi. Warner, KE. et al. Cigarette advertising and magazine cm·erage oft he hazards of smoking: a statistical analysis. New En1;la111LJ01mwl 0/111.f'ditilw, 326: ]05-309 (1992). 17. Penna, R.M. The essence of censorship. Tobarm control, 3: I 04-1 O:i ( 1994). 18. Aitken, P.P. Children and cigarette advertising, In:Smoke-fi'P, Euro/1e 8. Copenhagen, WHO, 1988. 19. Difranza,J.R. et al. RJR ~ahisco's cartoon camel promotes Camel cigarettes to children. Joumal of the A. meriwn Mediwl Association, 266: 3!49-3Ei3 (1991). 20. Hastings, G.B. et al. Cigareue ad\·ercbing and children's smoking: why Reg was with,lrawn. British 11udical]ournal, 309: 933-9% (1'194). 2l. Chapman, S. Pushing smokc-wbacco advertising and pro- motion, In: Smokr:free Eumjie 8. Copenhagen, \\'HO, 1988. 22. Waldron, I. etal. Gender differences in tobacco use in Africa, Asia, the Pacific and Latin America.Social.sriencermd 11udici,I.I!. 27: 1269-127'> (1988). 23. Mackay, J. \Vo men and tobacco, the scale of the problem: de\·eloping countries. Pro,eedinr;s of the Finl lnt,mwtional Co11jr-rnia nn \fomen and Smoking, lklfast, Ulster Cancer Fo11ndation/Hc-alth Promotion Agency for Northern Ireland. p. 32-34, 199'.1. 24. Charlton, A. & Blair, V. Predicting the onset of smoking in boys and girls. Soda[ sriena rmd nU'dicine, 29: 8 l '.l-818 ( 1989). 2:i. Greaves, L. Winnen and tobarco: updat,•, Ottawa, Health and Welfare Canada. 1990. 2G. Davis, R.M. Current trends in cigan,ue advertising and marketing. .Vew England Joumfll of medirine, 361: 72:i-732 (1987). 27. Nutbeam, D. etal. Planning for a smoke-free generation. In: SmnkefrPe Ic'urof1i, 6, Copenhagen, WHO, 1988. 28. Goddard, E. Rhy childwn sta,t 1moking. London, Her :'vlajesty's Stationery Office, 1990. 29. Palmer, J.R. et al. 'Low yield' cigarettes and the risk of nonfatal myocardial infarction in women. N,w Erigland journal ufm.edicine, 320: 1'>69-1373 (1989). 30. Graham, H. vW1en life\ a drag - wmnnt, snwking and di.Vid11antagr,. London, Her l\fajesty's Statione,,Office, 1993. 31. Rigotti, N.A. Cigarette smoking and weight gain. New England Journal of ))l.f'11icinr., 320: 931-933 ( 1989). 32. Batten, L. Respecting sameness and diflerence: taking account of gender in research and smoking. Tobacco tontrol 2: 18:i-18n (1993). 33. United States Department of Health and Human Services . Tht hnzlth consrquenccs ofsmokinr;: nicotine addiction. A repmt of the Surgeon G,,neral. Maryland, Centers for Disease Control, 1988. 34. Jarvis, M.J. Gender difkrences in smoking cessalion: real or myth? Tobaffo wntrol, 3: ?,2-1-'.\28 ( 1994). 35. Crossan, E. & Amos, A. Under a dmul - womrm, low incom• and smoking. Edinburgh, Health Education Board for Scotland, 1994. 36. Marsh, A. & Matheson, J. Smoking attitude.> and b,,ha11iour. London, Her Maje~tv's Stationery Otlice, 1983. 37. Greaves, L. et al. (eds). Herstories. World smoking and health, 19:2 (1994). 38. Niv, S. et al. A national smoking prevalence and health consequence survey in China in 1991. Tobacco and Health: proce,,dings of the 9th worl.d conf ermce on tobacco and health (Paris, 1994), New York, Plenum Press, 1996, p 29. 133 Older women in an aging world: achieving health across the life cot1rsea Ruth Bonita b & Anna L. Howe c Introduction There are many reasons why the health of aging women should now be an issue of major public health concern: the numbers of aging women are increasing worldwide; women's life course now ex- tends for a substantial period beyond 50 years and is increasing; and there is very significant scope for improving their health. The great differences in the life courses of aging women in countries at different levels of development also deserve great- er attention. Increasing proportions of women in developing countries now reach adulthood. Al- though their remaining life expectancy is still shorter than that of adult women in the developed world, that stage of their life span is substantial and continuously increasing. Already two-thirds of the net annual increase in the number of older women in the world occurs in less developed countries. However, aging has yet to be recognized as an issue in most of these countries. Action to improve the health of aging women is a worldwide imperative if they are to achieve an acceptable quality of life in their old age and if society is to avoid the conse- quences that would otherwise ensue. Magnitude of the challenge worldwide In developed countries, population aging has brought about an extension of later stages of the life span - an experience now being extended to developing countries, as seen in Tabl,e 1. Female life expectancy at birth ranges from just over 50 years in the least developed countries to 60 years and even into the 70s in those undergoing rapid development. In many developed countries, the typical female life expectancy at birth is currently around 80 years and still increasing. These differ- ences in life expectancy at birth between devel- oped and developing countries result from both high infant mortality and high maternal mortality in the latter. a This paper is part of a longer report, \lnmen, agi,ng and health prepared by Dr Bonita for the Global Commission on Women's Health under the guidance of the \,VHO Aging and Health Programme. Copies of the full report are obtainable from Dr A. Kalache, Chief, Aging and Health Programme, World Health Organization, CH 1211 Geneva 27, Switzerland. b Associate ProfessorofEpidemiology, University Geriatric Unit, Department of Medicine, University of Auckland, Auckland, New Zealand. c Reader, Faculty of Health Sciences, School of Health Systems Sciences, La Trobe University, Melbourne, Australia. 134 Low life expectancies at birth, however, dis- guise the duration of the later life course in devel- oping countries. For women in developing coun- tries who survive the early life span stages to reach middle age, life expectancy approaches that of women in developed countries. This is largely due to the fact that once a woman reaches middle age, non-communicable diseases become by far the most frequent causes of death, wherever she lives. The risk factors for these diseases are by and large related to lifestyle (for example, smoking, over- eating and alcohol consumption) and are more common in the developed world. However, residu- al problems related to reproductive health, expo- sure to infectious diseases and limited access to health services all contribute to higher mortality rates for older women in developing countries, although the differential between developing and developed countries is much smaller than at earlier stages of life. Life expectancies at age 65 show much greater similarity between developing and developed countries - around 15 and 19 years respectively. At age 65, women in developing coun- tries now have about three-quarters of the remain- ing life expectancy of their counterparts in devel- oped countries, and the gap may narrow in the future if mortality continues to decline not only at younger but also older ages. It is the likelihood of reaching the later life course rather than its duration that differs most markedly between countries at different levels of development: that and the quality of life if later years continue to be lived in the context of poverty and deprivation. The smaller proportion of the population surviving into their 60s and beyond makes old age a less common individual and social experience in developing countries than in devel- oped ones. However, for those who do survive, the duration of this life stage is considerable. This par- adox explains the presence of some very old indi- viduals in countries with low average life expectan- cies. These demographic data reflect major restruc- turing of both family relationships and the social roles of aging women. In developed countries, these roles - while more established- are still undergoing change. In developing countries, the aging process is being compressed in time and in parallel with radical societal changes. Traditional roles of older women are being overtaken and new norms have yet to unfold. Economic, social, cultural and political Rapp. trimest. statist. sanit. mond., 49 (1996) Table 1 Life expectancy of men and women in 1991, and improvements over the period 1970-1991 Tableau 1 Esperance de vie feminine et masculine en 1991, et amelioration entre 1970 et 1991 Life expectancy Increase in Life expectancy at birth, 1991 - life expectancy, at age 65, 1991 - Esperance de vie 1970-1991- Esperance de vie Country/Pays a la naissance, Augmentation de a 65 ans, 1991 1991 l'esperance de vie, 1970-1991 Female Male Female Males Female Male Femmes Hom mes Femmes Hammes Femmes Hammes Sub-Saharan Africa - Afrique subsaharienne Mozambique 48 45 6 6 n.a. n.a. Nigeria - Nigeria 53 50 10 10 n.a. n.a. South Africa - Afrique du Sud 66 59 10 9 n.a. n.a. lndia-lnde 60 60 11 10 n.a. n.a. China -Chine 71 67 8 6 16 14 Other Asian countries and islands -Autres pays et iles d'Asie Bangladesh 52 53 8 7 n.a. n.a. Philippines 67 63 8 7 14 12 Malaysia - Malaisie 73 68 10 8 15 3 Rep. of Korea - Rep. de Coree 73 67 11 9 15 13 Middle Eastern Crescent-Croissant moyen-oriental Egypt - Egypte 62 60 10 10 13 12 Turkey - Turquie 70 64 11 9 n.a. n.a. Tunisia -Tunisie 68 67 13 13 14 13 Latin America -Amerique latine Brazil - Bresil 69 63 8 6 n.a. n.a. Mexico - Mexique 73 67 9 7 17 15 Argentina - Argentine 75 68 5 4 17 14 Former socialist economies of Europe -Anciennes economies socialistes d'Europe Romania - Roumanie 73 67 2 0 15 13 Poland - Pologne 75 67 1 0 16 12 Hungary - Hongrie 74 66 1 -1 16 12 Established market economies -Economies de marche bien etablies United States of America - Etats-Unis d'Amerique 79 72 4 5 19 15 Spain - Espagne 80 74 5 4 18 15 Australia - Australie 80 73 5 5 19 15 France 81 73 5 5 20 15 Japan - Japan 82 76 7 7 20 16 Sources: Bonita, R. Women, aging and health: achieving health across the life span• and World Bank, World development repon. 1993, New York, Oxford University Press. 1993. factors all affect the health of agingwomen. As these social transformations proceed, they have a major impact on the quality of life of women as they age and have to be taken into consideration by those responsible for health policy. The demographic transition: the "feminization of old age" The demographic and epidemiological transitions that have been widely used in analysing population Wld hlth statist. quart .• 49 (1996) aging bring with them a gender transition. Three phases of population aging can usefully be identi- fied on the basis of changing patterns of survival of women relative to men and the associated gender balance of older populations. In the early stages of the transition, life expectancy at birth for men and women is similarly low, for instance, in countries such as Mozambique, Nigeria and India. Female life expectancy is still limited by high maternal mortality and is close to that of males. Although by age 65 women's life expectancy has moved ahead 135 of men's, the gender balance is only marginally in women's favour in those countries where they out- number men only in the older population. In some developing countries - typically those in the Middle East, sub-Saharan Africa and South Asia - women's life expectancy at birth is lower than or only equal to that of men. Increases have been only marginally higher than those experi- enced by men over the last 20 years. The second stage of the gender transition re- flects two trends in mortality which are well estab- lished in the developed world and becoming in- creasingly evident in many developing countries. Mortality declines both early in life and as a result oflower maternal mortality so that a larger propor- tion of women reach middle age. Life expectancy increases, both at birth and in old age. The com- bined effects of improved female survival to middle age, followed by improved survival at older ages, result in the marked shift in gender balance in the older population. Populations in the second stage of the gender transition are characterized by increasing propor- tions of older women. This process, which has been described as the "feminization of aging", is the current experience of many developed countries and also of many rapidly industrializing countries such as Thailand, Korea and Brazil. There i11, however, great diversity between countries which are just entering this stage, such as Brazil and Argentina, and those where the process is more advanced, such as Korea and Malaysia. The very rapid transformation of age structure in Japan contrasts with the slower process in Western Eu- rope. The countries of Eastern Europe show yet another pattern; female life expectancies are ahead of those for men, but there have been virtu- Fig. 1 Geographical distribution of the world's aging women ally no improvements over the last 20 years and the gender imbalance is set to persist. The social and economic diversity among these countries is even greater than the variations in demographic struc- ture, and the health of aging women will be shaped by these conditions. A third stage of the gender transition is now emerging in advanced industrial societies. At this stage, very high life expectancies for women mean that there is limited scope for further major in- creases, while recent improvements in male life expectancy at older ages are starting to narrow the gap between males and females. Diversity within global aging Already, more than half of the world's women aged 60 years and over live in developing regions (148 million, compared to 121 million in devel- oped regions). It is only at age 70 years and over that the number of women in the developed re- gions exceeds the number in the developing re- gions, but even in this age group, the margin is small (60 million compared to 58 million). The future growth in the numbers and proportion of aging women in developing countries is foreshad- owed in the distribution of those now aged 45 to 59 years. Two-thirds of the women in this age group (213 million) live in developing countries and one- third live in developed countries (98 million). The distribution of the world's aging women in the 8 World Bank regions is shown in Fig. 1. Differences in life expectancy Population aging has drawn attention to differenc- es in life expectancy between men and women: women have a greater life expectancy at birth and Repartition geographique des femmes plus agees a l'echelle mondiale [21 45-59 years - ans [] 60+ years - ans Subsaharan Africa - Afrique subsaharienne Middle Eastern Crescent - Croissant moyen-oriental Latin America and Caribbean - Amerique latine et Carai"bes Other Asia and islands - Autres pays et iles d'Asie lndia- lnde China- Chine Former socialist economies of Europe - Anciennes economies socialistes d'Europe Established market economies - l=~=::;::~::::::::;:::=:::'::::::;:::::::::::::::::;::::::::::::::=;::::::=::::::;:::::::::::::::::;:::::::::::::::::;:::::=.. _ _J Economies de marche bien etablies 0 10 20 30 40 50 60 70 80 90 Millions Source: United Nations Population Division and World Bank. 136 Rapp. trimest. statist. sanit. mond., 49 (1996) Fig. 2A Population of Europe, by age and sex, 1995 Population europeenne selon l'age et le sexe, 1995 80+ 75-79 70-74 65-69 60-64 ~ 55-59 -~ 50-54 ~ 45-49 I 40-44 *l 35-39 .b 30-34 f 25-29 Males - Hom mes Females - Femmes 20-24 15-19 10-14 I 5-09 0 0-041--~~r-~..!::::¥======:;:==:======:;:==:=======i========~:i=::::::::::::::::;'f::==:==:::;:;f====::::#=====*-~~~~----li 30 25 20 15 10 5 Source: Medium-variant projections of the United Nations Population Division. Fig. 28 Population of Europe, by age and sex, 2025 Population europeenne selon l'age et le sexe, 2025 0 Millions 5 10 15 20 25 30 Females - Femmes ao+r-~~~~~~~~~~~~~~~~¥¥~~~~~~~~~~~~LI~tJ:.:.:.:.::::.::._:_:.:..:.:..:~ 75-79 70-74 65-69 6Q-64 Males - Hommes ~ 55-59 -~ 50-54 ~ 45-49 I 40-44 *l 35-39 .b 30-34 gi, 25-29 <t 20-24 15-19 10-14 ij 5-09 ~ 0 0-04 l-----.---f====¥===~==~~9f:b::~~~;;;;,e~~g~§!d.._,-----,-----l i 30 25 20 15 10 5 Source: Medium-variant projections of the United Nations Population Division. at older ages than men, although the gap is closing at older ages. Differences in life expectancy be- tween women in countries at different levels of development have received less attention; these differences are in fact as great as or greater than differences between men and women within coun- tries at a similar level of development. Life expectancy at birth for women in devel- oped countries exceeds that of women in middle- income developing countries by 15 years, and by as much as 30 years for women in the poorest coun- Wld hlth statist. quart., 49 (1996) 0 Millions 5 10 15 20 25 30 tries. The disparity in life expectancy at birth for women in rich and poor countries has improved only a little over the past 20 years: these differences in women's life expectancies represent major in- equities that require greater attention. While less pronounced, differences between rich and poor countries in life expectancy at age 65 are nonetheless evident. That life expectancy in later life in many developing countries is already approaching that in developed countries can be taken as an indicator of what can be achieved with 137 improved health over the life span. Further nar- rowing of this gap will be an indicator of improve- ments in life expectancy at birth and at older ages in developing countries. The extended life span already experienced by older women in both devel- oping and developed countries calls attention to issues of quality of life in these years. Differences over the age range The gender transition brings shifts not only in the proportions of older men and women in national populations, but also in the age structure of the population. Notwithstanding their dominance on a global scale, aging women in developing coun- tries remain a smaller proportion of the popula- tion than in developed countries. Whereas women aged 60 years and older account for around 20% of the female population in developed regions, they account for only some 7% of women in developing countries. The equivalent proportions for men 60 years and over are nearly 15% in developed regions and 6% in developing regions. The main trend in aging in developed coun- tries is the increase in the "oldest old", those 85 years and older. Women comprise the great majority of this age group, and will continue to do so in the foreseeable future, as shown in Figs. 2A & 2B. It is largely this group who will require access to a range of health and long-term-care services. However, it is important that the small absolute size of these very old cohorts be recognized so that the demographic aspects of the rapid increase project- ed for the "oldest-old" cohorts in developed coun- tries do not overwhelm other considerations. The "oldest old" constitute only a small fraction of the total population in developed countries; Sweden has the highest proportion of "oldest old": 4.4%. In the least developed countries, less than 1 % of the population is in the "oldest-old" cohort; in countries such as Bangladesh and Indonesia, the proportion is as low as 0.2%, although the absolute numbers are substantial. Cohort differences Future cohorts of older women will be very differ- ent from the current generations. A cohort ap- proach which follows each particular generation or age group over time is essential for developing social and health policies. Encouraging each co- hort to consider its own future health is an impor- tant means of alerting those who make health pro- motion policies to these different problems and potentials. For example, in Latin America it is projected that 60% of women who were aged 45-49 years in 1990 will survive to age 75-79 in the year 2020. The health that these survivors can ex- pect in their old age will depend on the social and physical environments in which they live over the next 30 years. 138 In both developed and developing countries, older women are the main care-givers for their partners, and many women who care for frail elder- ly parents are themselves aging. Changes in living arrangements associated with urbanization in de- veloping countries suggest that, in the future, even more of the care provided to older persons will be given by older women. Such changes, particularly in living arrange- ments, have other profound implications for the health of older women in the future. A substantial proportion of older women in the developed world already live alone; if current trends persist, these absolute and relative numbers will be very high in the future. The single most important reason for this outcome is the much higher rate of widow- hood among women compared to men in both developing and countries, as shown in Tab'le 2. In addition, in many societies, many women reach older ages unmarried, either because they have never married or because after divorce (which is becoming more and more common), they are less likely than men to remarry. Urbanization and the trend towards the nuclear family are likely to lead to an increase in the proportion of older women living alone in the developing world. They will be vulnerable on many counts: poor, without the fam- ily support that previous generations could count on, and not able to rely on public services as these are unlikely to be in place. Monitoring trends in health and well-being The marked difference in the health experiences of successive cohorts of aging women and the dif- ferences between countries at different levels of development means that aggregate data on total aged populations is inadequate for monitoring trends in health status. A simple but significant contribution could be made by the presentation of all relevant data in more detailed categories of age, sex and economic status. A second important basic development is the use of age-specific and sex- specific life expectancies as indicators of change in later life, rather than just life expectancies at birth. Improvements in demographic data are of direct importance to monitoring trends in the health status of aging women. They also provide the foundations for developing further indicators such as indices of healthy life expectancy, as rec- ommended by WHO ( 1). A number of technical problems are involved in developing these mea- sures which must also take into consideration the different social and cultural meanings that disabili- ty and handicap have for men and for women. Methodological issues of this kind have begun to be addressed in the work of the REVES group (Reseau Esperance de Vie et Sante), among others (2). The recognition of gender factors needs con- tinued attention. Measurement of quality of life must reflect older women's experience and their Rapp. trimest. statist. sanit. mond., 49 (1996) Table 2 Widowhood among aging women and men (60+ years) Tableau 2 Veuvage (hommes et femmes de 60 ans et+) Country-Pays Sub-Saharan Africa -Afrique subsaharienne Mozambique Burkina Faso South Africa - Afrique du Sud lndia-lnde China - Chine Other Asian countries and islands-Autres pays et lies d'Asie Bangladesh Philippines Thailand - Thai'lande Rep. of Korea - Rep. de Goree Middle Eastern crescent - Croissant moyen-oriental Egypt - Egypte Turkey - Turquie Latin America -Amerique latine Brazil - Bresil Mexico - Mexique Argentina - Argentine % widowed Women Men Veuves Veufs 55 19 50 8 48 10 64 19 58 27 66 8 41 15 53 17 64 13 60 12 50 15 47 12 38 12 44 13 Former socialist economies of Europe - Anciennes economies socialistes d'Europe Romania - Roumanie 50 16 Poland - Pologne 52 14 Hungary- Hongrie 52 16 Established market economies - Economies de marche bien etablies United States of America - Etats-Unis d'Amerique 26 6 Spain - Espagne 42 14 Australia - Australie 43 12 France 45 13 Japan - Japan 49 12 Source: US Department of Commerce, Economics and Statistics Bureau, Bureau of the Census 1991. Global aging: comparative indicators and future trends. Washington, DC. expectations about acceptable levels of function- ing in their daily lives. An important adjunct to these policies, and an area in which WHO has a key role, is the develop- ment of measures to monitor demographic trends and health outcomes. Strategies are needed to ex- pand the range of indicators in order to take into account functional status and well-being as these are experienced by aging women. Some such strat- egies are: inclusion of aging women's health when reviewing national health goals and targets; devel- opment by Member States of intersectoral policy to reduce the differences in life expectancy between aging women in developing and developed coun- tries, and between groups of aging women within countries at different levels of development; use of Wld hlth statist. quart., 49 (1996) age-specific and sex-specific life expectancies in conjunction with life expectancies at birth as basic indicators for monitoring changes in the later life course; and the development of appropriate and relevant indicators of the health of aging women. The latter would involve both the critical analysis of the current global measure of health status from the perspective of aging women as well as taking into account the way aging women perceive their quality of life and value their health. Adding life to years Changes in their life course and the diversity of their experience of later life have significant impli- cations for the health of aging women. Above all, 139 the higher life expectancy of women in developed countries sets a basic goal for global strategies for the health of aging women, namely to reduce in- equities in life expectancy between women in de- veloped and developing countries. This goal is not simply a matter of increasing the number of years oflife, but includes addressing the social, econom- ic and political factors that determine health and quality of life. Following on from the 1995 United Nations Fourth World Conference on Women in Beijing, the UN International Year of Older Persons in 1999 will provide an important opportunity to monitor progress. The importance of 1999 lies par- ticularly in raising awareness of the plight of older women and the potential for positive aging. The reactivation and reorientation of the WHO Aging and Health Programme also provides a major op- portunity for furthering global strategies through initiatives and interaction with the United Nations (Programme on Aging) and UN Specialized Agen- cies, other national and international agencies, non-governmental organizations (NGOs), \VHO Regional Offices and other technical programmes in WHO Headquarters. The Regional Offices have a particularly crucial part to play in identifying programmes that offer models of good practice for promoting the health of aging women. Further, the Regional Offices are well placed to promote the exchange of available information and sharing of practical experience, and to develop simple but effective guidelines to enhance the focus on aging women's health in future planning and pro- grammes. Because resources and expertise are es- pecially limited in developing countries, it is essen- tial that duplication of effort be avoided and the potential of an interdisciplinary approach be exploited to the greatest possible degree in this important task of improving the health of older women. Summary This article, based on a report (Women. aging and health: achieving health across the life span) prepared for the WHO Global Commission on Women's Health under the guidance of WHO's Aging and Health Pro- gramme, presents demographic data that clearly dem- onstrate the need for recognition of the health of aging women as a global issue of major public health concern. The authors show that, while female life expectancy at birth is significantly different in developed and develop- ing countries (because of high infant and maternal mortality in the latter), these differences tend to de- crease for women in developing countries who reach 140 middle age. The authors review the various facets of the "gender transition" brought about by demographic and epidemiological transitions, drawing attention to con- trasts between the situation in developing countries in Asia and Latin America and that in Eastern Europe, for example. The role of older women as care-givers is discussed, as is the likelihood of a future increase in the proportion of older women living alone in tre developing world (a factor which renders them particularly vulnerable in many socioeconomic and health respects). Sugges- tions are made on methodologies for monitoring health trends in aging women, and on the role of WHO in this respect. A basic goal for global strategies relating to the health of older women is formulated: reduction of the inequities in life expectancy between developed and developing countries. Resume Les femmes plus agees dans un monde vieillissant: conserve, la sante durant toute la vie Cet article, base sur un rapport (Women. aging and health: achieving health across the life span), prepare pour la Commission mondiale de l'OMS pour la Sante des Femmes, sous l'egide du programme Vieillissement et Sante de l'OMS, presente des donnees demographi- ques qui demontrent clairement qu'il est necessaire de reconnaTtre que la sante des femmes vieillissantes est un probleme mondial d'un interet majeur pour la sante publique. Les auteurs montrent que s'il y a des differen- ces considerables dans l'esperance de vie des femmes a la naissance entre les pays developpes et les pays en developpement (dues a une mortalite infantile et mater- ne!le elevee dans ce dernier cas), ces differences tendent a decro'ilre .chez les femmes des pays en developpement qui atteignent le milieu de leur vie. Les auteurs passent en revue les divers aspects de la «transition entre les sexes" entraines par les transitions demographiques et epidemiologiques, en montrant, par exemple, les contrastes qui existent entre la situa- tion des pays en developpement en Asie et en Amerique latine et la situation en Europe orientale. Cet article examine le role des femmes plus a.gees en tant que pourvoyeuses de soins. ainsi que la probabilite d'une augmentation future de la proportion de femmes plus a.gees vivant seules dans le monde en developpe- ment (facteur qui les rend vulnerables et par rapport a la sante et sur le plan socioeconomique). II comporte des suggestions concernant les methodes permettant de surveiller les tendances sanitaires chez les femmes vieillissantes et concernant le role de l'OMS ace sujet. II formule un but fondamental pour les strategies mon- diales relatives a la sante des femmes plus a.gees, a savoir, la reduction de l'inequite pour ce qui est de l'esperance de vie dans les pays en developpement et les pays developpes. Rapp. trimest. statist. sanit. mond., 49 (1996) References/References I. World Health Organization. Development of indicators for mnnitoring health for all by the year 2000 ("Health for all" Series, No. 4) Geneva, "WHO, 1981. Organisation mondiale de la Sante. Elaboration d'indicateurs pour ut surveillanu continue des progres realises dans la voie de la Santipourtousd'icil'an2000(Serie «Sante pour tous», No. 4) Geneve, OMS, 1981. Wld hlth statist. quart., 49 (1996) 2. Mathers, C.D. & Robine,.J.M. Health expectancy indicators: a review of the work of REVES to date. [n: Robine,J.M. et al. (eds.) Calculation of health expectancies, harmonisation, consensus achieved and future perspectives, Colloque INSERJ\1., Vol 226. France, John Libbey Eurotext and Les Editions lNSERM, 1993. (pp 1-21). 141 Combating female genital mutilation: an agenda for the next decade Etua Dorkenooa Hosken (1) and Toubia (2) estimate that there are at present over 120 million girls and women who have undergone some form offemale genital muti- lation - sometimes referred to as "female circum- cision" - and that 2 million girls per year are at risk of mutilation. Most of the girls and women who have undergone mutilation are reported to live in 28 African countries where it is practised by many ethnic groups, in northern, eastern and western Africa. Some female genital mutilation is practised in the southern parts of the Arabian peninsula and along the Persian Gulf and increasingly, among some immigrant populations in Europe, Australia, Canada and the United States of America. It has also been reported to be practised by a minority ethno-religious group - the Daudi Bohra Muslims, who live in India - and among Muslim populations in Malaysia and Indonesia (3). The arguments against this traditional practice are based upon recognized human rights stan- dards including the right to health. It is known that the physical and psychological effects of the prac- tice are very extensive and irreversible, affecting the health of girls and women, in particular sexual, reproductive and mental health and well-being. Furthermore, female genital mutilation reinforces the inequities suffered by women in the communi- ties where it is practised, and must be addressed if the health, social and economic development needs of women are to be met. Despite recognition of the importance of this sensitive issue, there are still major gaps in knowledge about the extent and nature of the problem and the kinds ofinterventions that can be successful in elimi- nating it. To begin the process of developing a sound technical basis for policy and action, WHO convened a Technical Working Group Meeting on Female Genital Mutilation inJuly 1995 (4). The recommen- dations which emanated from this meeting have drawn international attention to female genital muti- lation and its health consequences and have contrib- uted to setting the agenda for the next decade for accelerating the elimination of this practice. Definition Female genital mutilation entails the removal of part or all of the external female genitalia and/ or a Consultant in Women, Health and Development, in the Family and Reproductive Health Programme, WHO, Geneva, Switzerland. 142 injury to the female genital organs for cultural or other non-therapeutic reasons. This definition, adopted by the WHO Technical Working Group Meeting on Female Genital Mutilation, encom- passes the physical, psychological and human rights aspects of the practice. Classification of the types of female genital mutilation In order to strengthen policy formulation includ- ing legislation and to clear the path for research and training, the WHO Technical Working Group recommended for adoption the following classifi- cation for the different types of female genital mu- tilation. In Type I, the prepuce (clitoral hood) is removed, sometimes along with part or all of the clitoris. In Type II, both the prepuce and the clito- ris and part or all of the labia minora (inner vaginal lips) are removed. Type III (known as infibulation) involves the complete removal of the clitoris and labia minora, together with the inner surface of the labia majora. The raw edges of the labia majora are then stitched together with thorns or silk or catgut sutures, so that when the skin of the remaining labia majora heals, a bridge of scar tissue forms over the vagina. A small opening is preserved, by the insertion of a foreign body, to allow the passage of urine and menstrual blood. Since a physical barrier has been created for sexual intercourse, the infibulated woman has to undergo gradual dilata- tion by her husband over a period of days, weeks or months to allow for penetrative intercourse. This painful process does not always result in successful vaginal penetration and the opening may have to be re-cut. Type IV is a new category that encompasses other surgical procedures including manipulation of the genitalia. These include pricking, piercing or incision of the clitoris and/ or labia, stretching of the clitoris and/ or labia, cauterization by burn- ing of the clitoris and surrounding tissue, introci- sion, scraping of the vaginal orifice, cuts into the vagina and introduction of substances into the va- gina with the aim of tightening or narrowing the vagina. The commonest type of female genital mutila- tion is Type II. This constitutes up to 80% of all female genital mutilation practised. The most ex- treme form is infibulation. This is thought to con- stitute 15% of FGM and is widespread in Somalia, northern Sudan and Djibouti. It has been reported Rapp. trimest. statist. sanit. mond., 49 (1996) in parts of Ethiopia, Eritrea, northern Kenya and small parts of Mali and northern Nigeria. Practitioners Female genital mutilation is usually performed by a traditional practitioner with crude instruments and without anaesthetics. Although WHO has con- sistently issued statements opposing medicaliza- tion of any form of female genital mutilation, among the more aflluent and in urban centres, female genital mutilation is increasingly being per- formed in health care facilities by qualified medi- cal personnel. Age The age at which female genital mutilation is prac- tised varies from area to area. It is performed at a few days old (for example among the nomads of Sudan), at about 7 years old (as in Egypt and in countries in eastern Africa and the horn of Africa) or in adolescence. In Nigeria, for instance, FGM takes place shortly before marriage among the Ibo, but only before the first child among the Aboh in the midwest (5). Most experts agree, however, that the age at which FGM takes place is falling. Reasons It is not known when or where the tradition of female genital mutilation originated and a variety of reasons (sociocultural, religious, psychosexual, hygienic and aesthetic) are given for maintaining it. Female genital mutilation is practised by fol- lowers of a number of different religions including Muslims, Christians (Catholics, Protestants and Copts) and animists, and also by non-believers in the countries concerned. Although female genital mutilation is not mentioned in the Koran, it is frequently carried out in some Muslim communi- ties in the genuine belief that it forms part of Islamic tradition. The health complications of female genital mutilation The health effects ofFGM depend on the extent of cutting, the skill of the operator, the cleanliness of the tools and the environment, and the physical and psychological state of the girl or woman con- cerned. Immediate complications Immediate physical complications include haem- orrhage and severe pain which can lead to shock and in some cases death. Acute urinary retention and infections are common. Injury to adjacent tis- sue of the urethra, vagina, perineum and rectum can result from the use of crude instruments. Frac- tures of the clavicle, femur or humerus or disloca- tion of the hip joint can occur if heavy pressure is Wld hlth statist. quart., 49 (1996) applied to the struggling girl during the operation, as often occurs when several adults hold down the girl during the procedure. Group mutilations, in which the same unclean cutting instruments are used on each girl may give rise to a risk of transmis- sion of HIV and hepatitis B but this has not been confirmed. Medium-term problems include de- layed healing and the formation of abscesses due to primary infections resulting from faulty healing. Long-term complications The long-term complications include keloid scar formation, the formation of dermoid cysts and cli- toral neuroma, dyspareunia (painful intercourse), chronic pelvic infections and difficulties in men- struation as a result of partial or total occlusion of the vaginal opening. Problems in pregnancy and childbirth are common, particularly following type III mutilation, because the tough scar tissue that forms causes partial or total occlusion of the vaginal opening and prevents dilatation of the birth canal. Prolonged and obstructed labour can lead to tearing of the perineum, haemorrhage, fistula formation and uterine inertia, rupture or prolapse. These complications can lead to neona- tal harm (including stillbirth) and even maternal death. Psychosexual and psychological health Almost all the types of female genital mutilation involve the removal of or damage to part or the whole of the clitoris, which is the main female sexual organ, equivalent in its anatomy and physi- ology to the male organ, the penis. Infibulation removes larger parts of the genitals, and closes off the vagina, leaving areas of tough scar tissue in place of the sensitive genitals, thus creating perma- nent damage and dysfunction. Sexual dysfunction in both partners may be the result of painful inter- course and reduced sensitivity following clitoridec- tomy and narrowing of the vaginal wall. FGM may leave a lasting mark on the life and mind of the woman who has undergone it. The psychological complications may be submerged deeply in the child's subconscious mind, and may trigger the onset of behavioural disturbances. The possible loss of trust and confidence in care-givers has been reported as another serious effect of female genital mutilation. In the longer term, women may suffer anxiety, depression, chronic ir- ritability, frigidity and marital conflicts. Many girls and women, traumatized by their experience of FGM, may have no acceptable means of expressing their fears, and suffer in silence. Human rights agreements to guide action There are various international agreements in place that are legally binding on the parties (states) and which prohibit the practice of female genital 143 mutilation. The United Nations Convention on the Elimination of All Forms of Discrimination Against Women promotes the rights of women and specifi- cally addresses discriminatory traditional practices (6). The Convention on the Rights of the Child protects the right to gender equality, and Article 24.3 of the Convention explicitly requires States to take all effective and appropriate measures to abolish tra- ditional practices prejudicial to the health of chil- dren (7). Similarly, there are regional human rights agreements such as the African Charter on Human and Peop!,es 'Rights and the African Charter on the Rights and Welfare of the Child which protect women and children against harmful traditional practices (8, 9). Article 18 of the African Charter on Human and Peoples' Rights specifically re- quests states to "ensure the elimination of every discrimination against women and also ensure the protection of the rights of women and the child as stipulated in international declarations and con- ventions." Article XXI of the African Charter on the Rights and Welfare of the Child obliges state par- ties to eliminate harmful social and cultural prac- tices affecting the welfare, dignity, normal growth and development of the child. The Programme of Action of the International Con- ference on Population and Development (/CPD) held in Cairo in 1994 also included recommendations in regard to female genital mutilation, which commit governments and communities to "urgently take steps to stop the practice of female genital mutila- tion and to protect women and girls from all such similar unnecessary and dangerous practices." ( 10) The Platform for Action of the World Conference on Women, also included a special section on the girl child and urged governments, international orga- nizations and nongovernmental groups to develop policies and programmes to eliminate all forms of discrimination against the girl child including female genital mutilation ( 11). Gaps in knowledge There have been no comprehensive global surveys of the prevalence of female genital mutilation. Current information on types of mutilation and their prevalence is derived from inadequate, frag- mentary data. On the basis of government reports, anecdotal evidence and limited surveys with sam- ples that are not always representative, the preva- lence of female genital mutilation is estimated to range from 5% to 98% in African countries (1, 2). Sudan is the only country to have carried out nationwide surveys (12-14). They were based on a national sample which excluded the three south- ern provinces, where the practice is unknown (ex- cept by adoption through marriage to the domi- nant northern ethnic groups practising FGM), and indicated an initial prevalence of 89% which subse- quently declined by 8%. A study by the Nigerian 144 Association of Nurses and Nurse-Midwives (15) conducted in Nigeria in 1985-1986, using a sample of 400 women and men in each state, showed that 13 out of the 21 states had populations practising some form of female genital mutilation, with preva- lence ranging from 35% to 90%. However, the data could not be extrapolated to give a national pic- ture. Similar limited surveys exist for Chad, Ethio- pia, Gambia, Ghana and Kenya. Reliable and accurate data on the prevalence, incidence and recurrence rates of the different forms of female genital mutilation or its health consequences will provide baseline information for subsequent evaluations and to inform policy makers and national decision-making processes. At the local level, a rapid-intervention survey may be the most appropriate step. At the national level, more detailed incidence and prevalence rates can be obtained by incorporating modules on female genital mutilation into existing surveys. Existing government surveys (for example, national Demo- graphic and Health Surveys (OHS), household in- come and expenditure surveys, and fertility sur- veys) can, with the addition of some extra ques- tions, be used to provide data on female genital mutilation at a fraction of the cost of a specific survey. As has been noted, questions on female genital mutilation were incorporated in the 1989- 90 Demographic and Health Survey (OHS I) in the Sudan. The Central African Republic and Cote d'Ivoire have also incorporated a few questions on FGM into their national Demographic and Health Surveys (1994 and 1994-95). Egypt integrated 34 questions on female genital mutilation into its national Demographic and Health Survey in 1995. A full module on female genital mutilation with 20 questions (OHS III) was field-tested in Mali and Eritrea in 1995. These efforts will help to generate reliable incidence and prevalence data for coun- tries in future years. Where studies ofFGM are to be conducted, the magnitude of the practice should be reported for different socio-demographic groups and for each type of genital mutilation. The magnitude of the problem should be expressed in terms of preva- lence, incidence and recurrence rates. Repeat sur- veys of prevalence, incidence and recurrence rates over time will help to establish trends of genital mutilation in a given community or nation. Further descriptive research providing quanti- tative and qualitative information is needed to characterize the different forms of genital mutila- tion and the socio-demographic characteristics of those who practise FGM versus those who do not. Other sociological variables such as the age of the girl, the location, and the persons involved in per- forming or assisting in the practice are essential information for planning target interventions and health education programmes aimed at certain lo- cations (for example, schools or homes) or popula- Rapp. trimest. statist. sanit. mond., 49 (1996) tions (for example, nurses, midwives, doctors and traditional birth attendants) that could help elimi- nate the practice at the source. Investigations are also needed to gain a better understanding of the sociocultural factors that in- fluence female genital mutilation, including be- liefs, class differences, power structures within soci- ety, the social/festive character that has built up around mutilation rituals and the links with mar- riageability. Some beliefs recur in a number of population groups, but there are notable differ- ences and some themes are exclusive to certain areas. Efforts are needed to analyse these factors within countries so that information and commu- nication materials can be adapted to take account of local conditions. Similarly, an accurate analysis of the existing economic incentives that promote the continuation of the practice will suggest measures to counteract them and indicate appro- priate areas for intervention. With regard to the health complications, the physical complications are well known. What is unclear, however, is the actual prevalence of com- plications and their long term sequelae in relation to gynaecological and obstetric morbidity and their impact on maternal and childhood mortality. The nature and the degree of psychological and sexual damage in different groups are still largely unexplored. Given the scale of the practice of female genital mutilation in many communities, this information is most important for developing clinical support for girls and women who are suf- fering from the health complications of female genital mutilation. Lessons learned In the last decade, a wide range of organizations and individuals have attempted community-based activ- ities to eliminate FGM. Women's organizations from communities where the practice persists have been leading the campaign for the last decade. With very little resources, they embarked on awareness- raising campaigns and have managed to break the taboo surrounding FGM in their communities. They have also brought the problem to the attention of political, religious and community leaders. Some governments have made statements condemning female genital mutilation; a few have adopted a policy or passed laws banning the practice; but often they have taken little action on the issue. FGM is an issue that cuts across both health and human rights and a major lesson learned from past community actions is that efforts to stop the practice need to go beyond the medical model of disease eradication. A multidisciplinary approach must be developed. An agenda for action To achieve real change at the grassroots will re- quire more planning, and more sustained commit- Wld hlth statist. quart., 49 (1996) ment from governments and international agen- cies to the elimination of FGM. The gaps in knowl- edge that need to be addressed have been out- lined. While voluntary organizations can play a pivotal role in the elimination of FGM, it is impor- tant that governments act to initiate, support and coordinate actions against this practice. The broad actions to be taken at the national level, include the need to: • adopt a clear national policy for the abolition of FGM. This should focus on prevention and re- habilitation. It should also incorporate clear goals, targets and objectives, and schedules for their attainment. Legislation is important, but legislation alone is insufficient. It should be accompanied by appropriate community-based action. Laws and professional codes of ethics should prohibit the medicalization of all the different forms of FGM; • establish inter-agency coalitions with members from relevant government ministries, non- governmental organizations and professionals to follow up action on FGM; • promote research on FGM, including the inci- dence, prevalence, and health consequences, particularly the impact of FGM on mental and sexual health as well as on the sociocultural determinants ofFGM, in order to develop more effective approaches to its elimination; • organize strong community outreach and fam- ily life education programmes for all sectors of the public including village and religious leaders, men and young people; and • organize training for health workers - physi- cians, nurses, midwives and also traditional birth attendants and healers - to enable them to work for the abolition of female genital mutila- tion and to provide clinical and psychological care and support for girls and women who have undergone FGM. In order to sustain action for the elimination of FGM, activities on FGM must be integrated into existing health education programmes, reproduc- tive health services and population and develop- ment strategies at national, regional and at com- munity level. Given the United Nations commitment to hu- man rights, with emphasis on advancing and pro- tecting the health and the lives of women and children, including their mental and sexual health, it is the duty of WHO, UNICEF, UNFPA and other UN agencies, as well as bilateral, multi- lateral and international development agencies, to support policies and programmes that bring an end to this damaging practice in all its forms. Vari- ous approaches will need to be developed such as promotion, providing technical support, and mo- bilizing resources, so that national and local groups can initiate community-based activities aimed at eliminating all harmful practices that af- 145 feet the health of women and children, especially FGM. WHO, in particular, has a special responsibil- ity to increase knowledge of FGM and promote technically sound policies and approaches for the elimination ofFGM, including developing training guidelines to equip health care workers with the appropriate knowledge, skills and attitudes for pre- venting and eliminating FGM, providing clinical management of the health complications, and en- suring that FGM is incorporated into broader con- cerns of women's health, reproductive health and human rights. Conclusions Harmful practices such as female genital mutila- tion persist today in many communities for a vari- ety of reasons. However, the roots of the practice lie in the patriarchal family and in society at large. Although women who are the victims of FGM are the gatekeepers of the practice in their communi- ties, this should be understood within the context of their general powerlessness in male-dominated societies. Promoting gender equity and women's empowerment will invariably lead to a decrease in the incidence and prevalence of FGM within com- munities and to its total elimination. It has taken some time for women's and children's rights to be accepted as human rights. It is vital that efforts now be made to give the human rights declarations and conventions on women's and children's rights meaning at the national and local levels. A signifi- cant shift in societal attitudes towards women and girls is called for. Young people are the adults of tomorrow. Early introduction of gender-sensitive education in schools will help to foster respect for girls' and women's human rights. Finally, as in- creased education of women appears to be a major factor in decreasing the practice of FGM, efforts to promote female education would have to be cen- tral to the long-term strategy for the elimination of this harmful traditional practice. Summary Female genital mutilation (FGM) - sometimes locally referred to as "female circumcision" - is a deeply rooted traditional practice that adversely affects the health of girls and women. At present it is estimated that over 120 million girls and women have undergone some form of genital mutilation and that 2 million girls per year are at risk. Most of the girls and women affected live in 28 African countries where the prevalence of female genital mutilation is estimated to range from 5% to 98%. The elimination of female genital mutilation will not only improve women's and children's health; it will also promote gender equity and women's empowerment in the communities where the practice persists. To achieve change will require more planning, and more sustained 146 programmes for its elimination. The political will of gov- ernments is essential in order to eliminate this harmful traditional practice and concerted efforts from all con- cerned are required. Resume Lutte contre les mutilations genitales feminines: ordre du jour pour la prochaine decennie La mutilation genitale feminine - parfois connue locale- ment sous le nom de circoncision feminine - est une pratique traditionnelle profondement ancree qui nuit a la sante des fillettes et des femmes. On estime a present que plus de 120 millions de fillettes et de femmes ant subi la mutilation genitale feminine sous une forme ou une autre et que 2 millions de fillettes par an courent le risque d'etre mutilees de la sorte. La plupart des fillettes et des femmes touchees vivent dans 28 pays d'Afrique ou on estime que la prevalence de la mutilation genitale feminine va de 5 a 98%. L'elimination de cette mutilation n'ameliorera pas seulement la sante des femmes et des enfants; elle promouvra egalement l'egalite entre les sexes et donnera des moyens aux femmes dans les communautes ou cette pratique perdure. II faudra, pour realiser ces changements, davantage de planification et des programmes plus soutenus. II est essentiel que les gouvernements montrent la volonte politique d'elimi- ner cette pratique traditionnelle nefaste et on demande a tous des efforts concertes. References/References l. Hosken, F.P. Female genital mutilation, estimate: total number of girls and women mutilated in Africa, Lexington, Women's International Network News, 1995. 2. Toubia, N. Fema/,e genital mutilation, a call for global action, second edition, New York, RAINBO, 1995, 24-25. 3. Hosken, F.P. The Hosken report, genital and sexual mutilation of femaks, fourth revised edition, Women's International Network News, 1993. 4. Report of a WHO Technical Woming Group Meeting on Femal.e Genital Mutilation, 17-19 July, 1995. Geneva, 'WHO (forthcoming). 5. Dorkenoo, E. Cutting the rose: femak genital mutilation, the practice and its prevention, London, Minority Rights Group, 1994. 6. Convention on the elimination ofall forms of discrimination against women, in: Brownlie, I. (ed) Basic documents on human rights, 3rd edition, Oxford, Oxford University Press, 1992 (pp. 169-181). 7. Convention on the rights of the child, in: Brownlie, I. (ed) Basic documents on human rights, 3rd edition, Oxford, Oxford University Press, 1992 (pp. 182-202). 8. African charter on human and peoples' rights, in: Brownlie, I. (ed) Basicdocumentsonhumanrights, 3rdedition, Oxford, Oxford University Press, 1992 (pp. 555-566). 9. African charter on the rights and welfare of the child. Organization of African Unity, Doc. CAB/LEG/153/Rev. 2 (1960). l 0. International Conference on Population and Development (ICPD ), Report of the International Omference on Population and Development, UN Doc. A/CONF. 171.13 (1994). Rapp. trimest. statist. sanit. mond., 49 (1996) 11. Platform for Action of the World Conference on Women, in Report of the Fourth World Conference on Women, UN Doc. A/CONF. 177 /20. 12. El Dareer, A. Woman, why do you weep? London, Zed Books, 1982. 13. The Sudan fertility surol!)·, Department of Statistics, Ministry of Economic and National Planning, Khartoum, Sudan, I 979. Wld hlth statist. quart., 49 (1996) 14. Sudan demographic and health survey, Department of Statistics, Ministry of Economic and National Planning, Khartoum, Sudan, 1989/1990. 15. Adebajo, C.O. Female circumcision and other dangerous practices to women's health. In: Kisekka, M.N. Women's health issues in Nigeria, Zaria, Tamaza Publishing Company, 1992, 1-11. 147 Ethics and reproductive health: a principled approach Ruth Mackfina When we learn that an estimated 500 OOO women, at least, die annually from pregnancy-related causes ( 1) and that the vast majority of these deaths occur in the developing world (2), the ethical problem is self-evident. When we hear that in many developing countries, complications of pregnancy and child- birth are among the main causes of death in 15-19 year-old women and that in Jamaica and Nigeria, girls younger than 15 years of age are 4 times more likely to die during pregnancy and childbirth than 15-19 year-olds (3), the conclusion is inescapable. As a public health imperative, both govern- ments and health-related nongovernmental orga- nizations have an obligation to take all necessary steps to reduce this staggering rate of maternal mortality. The same ethical obligation exists for the individual physician caring for patients, the obligation to strive to prevent the most common causes of maternal mortality: deaths resulting from obstructed labour and ruptured uterus, postpar- tum haemorrhage, pregnancy-induced hyperten- sion, postpartum infection, or septic abortion (2). For the individual clinician, this obligation means both helping women to take proper preventive health measures and also administering appropri- ate therapy to patients. When we learn that be- tween 100 000 and 200 OOO of the annual deaths from pregnancy-related causes are estimated to be due to improperly performed and usually illegal abortions ( 1), the ethical problem is again self- evident, but more controversial. When all the figures are considered, not just the maternal deaths from preventable causes, but the staggering figures for maternal morbidity from un- safe abortions, a very grim picture of women's over- all health and reproductive health begins to appear. One way of addressing ethical issues in repro- ductive health is to look at the consequences of current laws, policies and practices and see whether the existing situation gives rise to a pre- ponderance of good or bad consequences. If the bad consequences outweigh the good ones, there is an ethical obligation to seek to change the laws, policies or practices. The data on maternal mortal- ity and morbidity in developing countries make it uncontroversial to state that the harmful conse- a ProfessorofBioethics, DeparunentofEpidemiologyand Social Medicine, Albert Einstein College of Medicine, Bronx, New York, United States of Amaerica. 148 quences for women of inadequate family planning programmes and of laws prohibiting abortion far outweigh the beneficial consequences. Furthermore, these harmful consequences not only affect women, but also the children they bear. For example, it has been shown that the death of a mother increases significantly both the morbidity and mortality rates of her surviving children, par- ticularly those under age 5 ( 1). And it is well known that prematurity and low birth weight in infants are frequent consequences of births too closely spaced, as well as of pregnancies occurring when women are too young or of advanced age for child- bearing. It is always easier to point to ethical problems than it is to devise solutions. Some solutions to poor reproductive health in developing countries require money those countries do not have, trained medical or paramedical personnel current- ly in short supply, and medical equipment of vari- ous kinds that is expensive or hard to maintain. There is no ethical principle that can dictate a solution to problems that stem from genuinely scarce or expensive resources. However, there are ethical principles that point to solutions to mater- nal mortality and morbidity from preventable causes. The barriers to these solutions are not lack of money but lack of political will, the indifference of men in power to death and disease among un- empowered women, and traditional customs that should be made to change for ethical reasons. Indi- viduals and groups can work towards a solution by using ethical principles to justify placing a high priority on reducing maternal mortality and mor- bidity from preventable causes. Universal agreement on ethical matters is a goal unlikely to be reached. This is not only because of the cultural and religious differences that exist in our world, but also because some issues are deeply problematic from an ethical point of view. This creates dilemmas for decision makers, be they cli- nicians or policy makers. Even people who share a common religious and cultural background often disagree about particular matters. But despite such disagreements about particulars, universal ethical principles do exist. Problems in applying these principles lie not with the principles themselves, but with the various ways in which they can be interpreted. Let us begin with the principle of be- neficence, which obligates people to try to produce more good than harm. Rapp. trimest. statist. sanit. mond., 49 (1996) Beneficence In the field of medicine, this principle obligates both individual clinicians and policy makers to strive to bring about more beneficial consequences than harmful ones. In clinical practice, the princi- ple takes the form of making risk-benefit assess- ments, on which physicians base treatment recom- mendations having the most favourable benefit- risk ratio for the individual patient. Risk-benefit assessments can also be used to evaluate family planning programmes and other reproductive health services. Applied to methods of family planning, the good and bad consequen- ces of each method must be examined, looking at those consequences both for the user and her sexual partner. Here, as elsewhere, good ethics will begin with good facts. In making risk-benefit assess- ments, accurate, up-to-date information must be employed. Care must be taken to use appropriate data for the locale in which the method is to be used. This is because data gathered about risks and benefits for women in developed countries or in some regions of the developing world may not be strictly applicable to women in other developing countries. For this reason, characteristics of the users of contraceptive methods must be factored into the risk-benefit assessment. For example, tak- ing a daily pill in a society in which pill-taking is otherwise rare is unlikely to gain compliance. The natural family planning method that relies on women accurately monitoring their cervical secre- tions is no doubt an alien concept to women in many cultures. And reliance on condoms in societies in which men steadfastly refuse to use them cannot protect women against unwanted pregnancies or sexually transmitted diseases. This last point demonstrates that there is an inevitable relativity in making risk-benefit assess- ments. Depending on the features of the service delivery system, the cultural aspects of a country or region, and the beliefs and attitudes of the people, some family planning methods will have a more favourable benefit-risk ratio than others. The risks include not only the medical and psychological risks to the users of a family planning method, but also the risk of not having effective family planning methods available and not having a backup for failed contraception. This underscores the point that risk-benefit assessments are likely to yield dif- ferent results at different times and in different places. This relativity in the application of ethical prin- ciples to different situations must be distinguished from the very different notion of ethical relativism, the view that ethics can be relative to time, place and circumstance. Although it is true that different cultures have different norms and customs, it does not follow that whatever cultural or national prac- tices exist are ethically justified. I will return to this point by way of conclusion. Wld hlth statist. quart., 49 (1996) The principle of beneficence, like other ethical principles, is objective. To say it is objective means that its proper application requires obtaining accu- rate data. With regard to reproductive health, the data must take into account the actual experiences of women. To do so necessarily relies on subjective reports given by women. But that does not under- mine the objectivity of the process. Information about subjective experiences can be gathered in an objective way, that is, a way that is scientifically valid and uses established methods of data-gathering. Social and behavioral research reveals a wealth of information about how women respond to differ- ent methods of family planning, which methods they find acceptable or unacceptable, the reasons for noncompliance with some methods, the role female literacy plays in reproductive health, and why women resort to unsafe abortions in countries where safe, legal abortions are not accessible to them. A failure to take into account women's ac- tual experiences and attitudes is likely to result in flaws in risk-benefit assessments made from the perspective of medical scientist~. For example, evidence can be cited to show that women's values have been either neglected or un- derrated in formulating risk-benefit assessment,;. Women's health advocates tend to define the "safety" of contraceptive methods in terms differ- ent from those typically employed by biomedical scientists. According to one report: Scientists' concern is to establish safety of meth- ods according to specific measurable parame- ters. They assess toxicity, first in animals and then in carefully controlled studies in human volunteers. Subsequent studies address efficacy and short- to medium-term safety .... ·women's health advocates ... give more priority to meth- ods that have fewer side effects and that protect against sexually transmitted diseases and their consequences such as infertility. \Alhile scien- tists have tended to give priority to methods which minimize users' control, women's health advocates prefer methods controlled by the user (4). The same mix of o~jective and sul~jective ele- ments enters into balancingrisks and benefits. That, too, is partly a scientific matter, based on available evidence regarding the probability and magnitude of anticipated risks and benefits; but it is also a subjective matter. Different people - he they medi- cal scientists, patients or healthy laypersons - may evaluate the risks and benefits differently. They may consider some risks worth taking in relation to expected benefits while other risks may be viewed as unacceptably high in relation to expected benefits. A point made by one participant in a meeting between women's health advocates and scientists is worth repeating, "On the question of side effects, there is always a tendency to over-emphasize the benefits and underplay the risks ... Most of the time 149 1t 1s we women who undergo the risks and the benefits are taken by the pharmaceutical companies or by population control experts or governments of Third World countries" (4). An example of the way women's assessments of the risks and benefits of contraceptives may diverge from that of clinicians, medical scientists or gov- ernmental agencies comes from Mexico. One par- ticipant at a conference on ethics and reproductive healthb suggested that if allowed a choice, women may choose a higher risk method of contracepti_on that has fewer side effects rather than a more effec- tive method with undesirable side effects. Women would rather take the risk of becoming pregnant than use a contraceptive method with unaccept- able side effects. It was noted that the population in Mexico does not want hormonal methods or IUDs, but these are the methods promoted by the family planning agency. A conference participant noted that the three contraceptive methods most pre- ferred by the population are sterilization, the rhythm method and the condom. However, as the governmental system in Mexico is currently man- aged, if couples accept the condom they do not officially count as "acceptors." There is a long-standing tradition in medical practice everywhere of paternalism, the ~ew that doctors and medical scientists know what 1s best for patients. Not only are they supposed t~ know wh~t is best, but their knowledge and authonty has tradi- tionally been used to justify decision-making by doctors on behalf of patients, and even coercion of patients "for their own good." As an illustration of this paternalistic attitude and disregard for women's own values, a marketer for one method was quoted at the conference in Mexico as saying, "This is a feminist method: what does it matter if it has side effects?" Risk-benefit assessments play a central role in the medical context, even though the risks and benefits have traditionally been defined and weighed more by physicians than by the patients who suffer the risks and enjoy the benefits of treatments. However, analysis that focuses on benefit and harm has too often been ignored among scholars in the field of bioethics, who have tended to focus more on the rights of patients and respect for their autonomy. Nevertheless, we cannot eliminate the language of rights when addressing ethical issues in repro- ductive health. That is because we cannot ignore the role of reproductive rights within a larger framework of human rights. International atten- tion to human rights has focused on violations of human rights in areas such as torture, the treat- ment of political prisoners, and denial of the fun- damental freedoms of citizens under authoritarian b This conference was part of a project conducted by the ~uthor from 1992 to 1994 on ethics and reproductive health, which was made possible by a grant from the Ford Foundation. 150 political regimes. It is now widely believed that at least some reproductive rights should be included among the fundamental human rights now ac- knowledged throughout the world. Respect for persons . The universal ethical principle from which the concept of human rights is derived is known as "respect for persons". As a universal ethical princi- ple, it presumes that all human beings have dignity and are worthy of respect. In today's world, virtual- ly every society pays lip service to human rights. Sadly, however, violations of human rights contin- ue in some nations and regions. But even in those societies that acknowledge respect for persons, it is often the case that equal respect is not shown to women as persons. Only when women are granted a status of respect as persons equal to that tradition- ally given to men can ethical issues in reproductive health be properly addressed. For example, countries that have laws or customs requiring a husband's authorization before a wom- an can be sterilized, receive contraception or under- go an abortion, fail to show equal respect for women as persons (5). Similarly, long-standing customs that sanction husbands' power to determine which means of birth control, if any, their wives may use quite clearly grant unequal respect to men's dec!- sion-making authority over that of women. Yet 1t cannot be denied that the effects of such decisions on women are more profound and potentially more devastating than on the men who make them. It is not men who are dying from complications of preg- nancy and childbirth or septic abortion. It is not male adolescents who suffer morbidity and mortal- ity from childbirth. And it is not men who become infertile as a result of these practices. The ethical principle that serves as a foundation to granting women reproductive rights is the prin- ciple of individual freedom or liberty. The man- date to promote reproductive freedom has been recognized, at least as a matter of principle, throughout the world. Since the first World Popu- lation Conference, held in Bucharest in 1974, more than 130 countries have signed the following statement regarding reproductive rights: "All cou- ples and individuals have the basic right to decide freely and responsibly the number and spacing of their children and to have the information, educa- tion and means to do so." (6) That statement was reiterated 10 years later at the Mexico City World Population Conference, and strengthened even further in 1994 at the International Conference on Population and Development in Cairo.c,d c &port of the International ConferenaonPopulation, 1984, CN Doc. E/CONF. 76.19 (1984). d International Conference on Population and Development (ICPD). &port of the International Confererue on Population and Development, UN Doc. A/CONF, 171.13 (1994). Rapp. trimest. statist. sanit. mond., 49 (1996) The ethical principle of "equal respect for per- sons" can be understood at an individual level and as an ingredient in social justice. At the individual level, showing equal respect for women as persons means recognizing their autonomy and treating them as capable decision-makers and full partici- pants in medical decisions. One example would be honouring women's feelings and preferences in modes of contraception and methods of abortion, even if it is more cost-effective to impose methods preferred by providers or funders. Another exam- ple is treating pregnant women with dignity and respect instead of merely as "fetal containers." The principle of "respect for persons" is viol- ated when women who seek abortions or ask for medical help following complications of an in- duced abortion are treated punitively by physi- cians or other health-care workers. Even worse than being treated punitively or in a degrading manner are situations in which physicians refuse to render medical assistance to women following a self-induced abortion. A member of the audience at the author's lecture in Santiago, Chile in 1993 reported the following case. A patient was brought the hospital after a self-induced abortion, dying and in need of medical attention. The physician refused to help the patient, saying that if a doctor tried to heal a patient who had committed a sin, the physician would also be complicit in that sinful behaviour. The doctor refused to assist the pa- tient. The physician in that case had unquestion- ably violated his obligation to care for sick and dying patients. A physician who is opposed to abortions on moral or religious grounds should not be forced to perform the procedure. However, this case did not involve a physician performing or assisting in an abortion but rather, treating a pa- tient who had made the attempt herself. So the physician could in no way be viewed as complicit in the patient's earlier act. Throughout history, women have always sought and will continue to seek an end to unwanted pregnancies. Women will continue to resort to abortion as a backup for failed contraception. A first step in seeking to reduce the number of abor- tions, be they safe or unsafe, legal or illegal, is the recognition of women's right to be adequately in- formed about available family planning methods. This includes information about what procedures are involved in proper use of the method, and about the risks and benefits of various available methods. The principle of respect for persons mandates that women's use of a particular method must be voluntary, that is, free coercion or undue inducements of any sort. In the abortion debate, the rights of the fetus are typically pitted against the rights of the preg- nant woman. In the political debate carried on in the United States of America and in other coun- tries, feminists have adopted the phrase "a Wld hlth statist. quart., 49 (1996) woman's right to control her own body", thus iden- tifying a right that could presumably override the right to life of the fetus. "The right to control one's own body" is another way of describing the right to self-determination. The underlying basis for this reproductive right is the right to liberty. The principle of liberty dictates that individuals have a right to freedom of decision and action, to the extent that their actions do not interfere with the rights of others. Opponents ofa woman's right to abortion do not disagree on the soundness of that fundamental ethical principle itself. Yet they disagree profoundly over its application: oppo- nents of abortion claim that the act of terminating a pregnancy does interfere with the rights of another (the fetus), while advocates ofa woman's right to procure an abortion deny that killing a fetus is a violation of rights. As is true of any conflict of rights, this one might be resolved in favour of either party - the woman or the fetus. If a right to life is ascribed to fetuses and a right to terminate a pregnancy is assigned to women, a higher priority could be giv- en to the rights of the woman. However, as impor- tant as the value of liberty is in philosophical and political thought, only rarely is it held to outweigh the value of human life when the two values con- flict. Therefore, the most reasonable way to resolve this apparent conflict of rights in favour of the woman is to deny that the fetus can properly be considered an entity having rights. The view that a fetus is a "person" from the moment of fertiliza- tion, and therefore possesses a right to life just like already-born individuals, is a precept of some, but by no means all religious faiths. One way of trying to think clearly about the right of pregnant women to control their own bod- ies is by changing the context of the abortion de- bate. A useful analogy to consider is that of organ or bone-marrow transplantation. Sometimes chil- dren are in need of a life-saving or possibly curative procedure and the best tissue match is a parent. No one has ever convincingly argued that an unwilling father should be compelled to undergo surgery for removal of a kidney or even a bone marrow extrac- tion for the benefit of his child. And no judge has ever ordered such a procedure. For the govern- ment or, perhaps, a husband to insist that a woman remain pregnant against her will, for the sake of the fetal life within her, is thus to grant a higher moral status to developing fetal life than to that of already-born children. Justice This article has explored the ethical principles of beneficence and respect for persons and their ap- plication to reproductive health. A third leading concern of bioethics is justice. Questions of justice arise with respect to the distribution of family plan- ning methods, including access to safe abortion in 151 cases of contraceptive failure. A just distribution of reproductive health services requires that methods be accessible to poor women as well as those who are better off, to the less educated as well as those who are better educated, to rural as well as urban residents. The principle of justice mandates that all indi- viduals who need family planning and health ser- vices should have equitable access to them. From an ethical perspective, "equitable access" means that use of these services should not be based on an ability to pay for them. Guaranteeing access by law has not proved sufficient for achieving social jus- tice in societies that do not recognize a right to health care. An additional precondition for access is information about the existence and nature of the services. A moral obligation exists to ensure that women have information, as well as the means to obtain family planning services. These conclusions apply with even more force to women in developing countries than they do in developed countries. Poor women everywhere dis- proportionately bear the burden of restrictive abortion laws and inadequate or nonexistent fam- ily planning services. Justice dictates not only that equal respect be shown to women, but also that the needs of the least advantaged members of society be addressed. Cultural and ethical relativism We now return to the topic of cultural and ethical relativism. Perhaps even more than in other areas of medicine and health care, the introduction of techniques and practices related to human repro- duction gives rise to ethical controversy stemming from social, cultural and religious differences. A culture may have a long-standing tradition prefer- ring or rejecting certain methods of family plan- ning. There may be a religious prohibition against contraception or abortion. In a culture character- ized by dominance of men over their women part- ners, or one in which the rights of women are not fully recognized and granted, some family plan- ning methods more desirable to women might be rejected by men, be they men who hold political power or men who are the husbands and sexual partners of women. By law or custom, the practice of "spousal authorization" for sterilization or con- traception remains in force in a number of devel- oping countries today (5). It requires enhanced sensitivity and a commit- ment to social progress to recognize that centuries- old cultural patterns of behaviour of men toward their wives, their sexual partners, and their daugh- ters are ethically unacceptable. Just because these patterns have existed for centuries does not make them right. Just because they have been wide- spread and largely tolerated in many cultures does not make them right. The fact that a majority of people in a society accept and subscribe to certain 152 social or cultural practices does not amount to an ethical justification of those practices. For example, consider the practice of genital mutilation of women, more politely called "fe- male circumcision." An estimated 120 million girls and women in the world are genitally mutilated (7). Should this be thought of simply as a "cultural rite" following the traditional custom where the ritual has been practised for centuries? Or should it more properly be thought of as mutilation of the body of female persons, more akin to torture than to a cultural ritual that causes temporary pain but is not overly harmful? Can any of the justifications offered for the practice of female genital mutila- tion stand up to ethical scrutiny? It would not be sufficient to show that the practice of genital muti- lation of females is merely approved of by the cul- tures that practice it. According to that method, slavery could be ethically justifiable if a majority of people in the society {that is, the slaveholders) approved of slavery. Using the ethical principle of beneficence, one would have to demonstrate that this cultural practice actually creates more good than harm, compared to the alternative of not engaging in the practice; or that it produces more beneficial consequences for the greatest number of people than it does harmful consequences. However, the morbidity and mortality of female infants, pubescent girls, and women later in their lives has been well-documented and is sufficient to condemn the practice on grounds of its health consequences for those who are subjected to it. The consequences include pain, trauma and se- vere physical complications in both the short and long term. The reasons given for this ritual surgery in- clude: (i) it is a good tradition needed for group identity; (ii) it is a religious requirement; (iii) it is necessary for cleanliness and health; and (iv) it is needed to preserve virginity and family honour, and to control immorality (8). It is beyond the scope of this paper to analyse and respond to each of these reasons. However, a careful analysis would have to address each of these reasons and question first and foremost, whether they rest on sound empirical evidence. For example, could this partic- ular cultural practice be abandoned while still pre- serving other traditional practices that promote group identity? Seconily, the claim that it is a religious requirement also must be subjected to ethical scrutiny. If some religion required the sacri- fice of the first-born child at age twelve, could that practice be ethically justified simply based on the fact that it is a feature of the traditional religion? Fundamental ethical principles can be used to eval- uate specific practices required or condoned by any religion, however prominent, dominant, or widespread the religion may be. Just because a religion dictates or condones a particular practice does not mean that the practice is ethically right. Rapp. trimest. statist. sanit. mond., 49 (1996) There could be no religious reform if it were not possible to stand outside the dictates of a particular religion and judge some of its features to be wrong. Unlike some ethical concerns, female genital mutilation is a practice in which all three leading bioethical principles are shown to be violated. The ethical principle of "respect for persons" is viol- ated in situations where women are controlled by men for the express purpose of satisfying men's sexual pleasure at the expense of their own, when women lose the capacity for sexual pleasure and expression of their sexual feelings, and when they are unmarriageable if they are not genitally muti- lated in accordance with the norms of that society. Furthermore, any society in which there are differ- ent standards of sexual morality for men and for women violates the principle of "equal respect for persons." Women are undeniably persons. If geni- tal mutilation is a cultural practice held to be justi- fied because it serves to control immorality, it is the alleged sexual immorality of women but not of men that it seeks to control. Failure to grant equal respect to women and to men, and perpetuation of practices that constitute oppression of women are violations of the principle of justice. To resist ethically mandated change because of long-held beliefs or practices is a philosophical error. The error lies in concluding that because a state of affairs has existed in the past, it ought to continue into the present and future. The flaw in that reasoning can easily be seen by reflecting on the fact that manifestly unjust social institutions, such as slavery and colonialism, would still be with us if history and tradition served as an infallible moral guide. Moral progress requires a critical eval- uation of past practices and institutions. Of course, many social practices and institutions will with- stand such critical evaluation, but others will not. Subjugation of women, denial of their right to self- determination in choosing an acceptable method of family planning, and preventing them from hav- ing access to safe, legal abortion are practices that cannot withstand critical ethical evaluation. Summary Universal ethical principles can be used to analyse problems in reproductive health. The principle of benef- icence obligates people to strive to bring about more beneficial consequences than harmful ones. The princi- Wld hlth statist. quart., 49 (1996) pie known as respect for persons presumes that all human beings have dignity and are worthy of respect. Showing equal respect for women as persons means recognizing their autonomy and treating them as capa- ble decision-makers and full participants in medical decisions. A third leading concern of bioethics isjustice. which requires a fair distribution of family planning methods. including access to sate abortion in cases of contraceptive failure. Resume L 'ethique et la sante reproductive sous /'angle des principes Les questions de sante reproductive peuvent s'analyser a !a lumiere de principes universelles d'ethique. Selan le principe de bienfaisance, le «primum non nocere» doit l'emporter sur toute autre consideration. Le principe de respect de la dignite humaine pose que toute personne humaine doit etre respectee et traitee dignement. En particulier, le respect du aux femmes en tant que per- sonnes implique que leur autonomie soit reconnue et qu'on admette leur capacite a prendre des decisions et notamment a etre partie prenante a la decision medi- cale. Entin, le principe de justice, preoccupation majeu- re de la bioethique, exige une diffusion equitable des methodes de planification familiale, y compris la possi- bilite de se faire avorter en toute securite en cas d'echec de la contraception. References/References 1. Rosenfeld, A. RIJ486, American journal of f1ublic hrnlth, 82(10): 1323 (1992). 2. Elkins, T.E. Maternal mortality and morbidity in the developing world: personal reflections and a profession's commitment, Wmnm s health issues, 2(3): 147 (1992). 3. Cook, R.J. International protection of women's repro- ductive right~. ,\'ew York University joumal of international lmowulpolitics, 24(2): 6H-717 (1992). 4. Special programme of research, development and research training in human reproduction & International Women's Health Coalition. Creating rommon ground: women '.s penpertives on the seledion and introdurtion of fertility regulation technologies, Geneva, World Health Organization, 1991 (p. 11). 5. Cook, R.J. & Maine, D. Spousal veto over family planning services, 1\merican journal of public health, 77: 339 ( 1987). 6. United Nations: World population plan of action. In: The popul.ation debate: dimensions nnd perspectives, Vol. 1, N cw York, UN Publications, 1973, pp. 155-167. 7. Toubia, N. Femak genital mutilation: fl call for global action. Second edition, New York, RAINBO, 1995. 8. Kopelman, L.M. Female circumcision/genital mutilation and ethical relativism, Second opinion, 20(2): 33-71 (1994). 153 Sexual abuse in the health professions - who's counting? Marilou McPhedrana Introduction: different perspectives This article will concentrate on surveys and other research which have begun to quantify the inci- dence of sexual exploitation of patients by health professionals. Itis intended to be an introduction to some of the work which has been done, and an invitation to the reader to explore this area further. However, the quantitative research will never tell us all we need to know about this phenomenon. There is much to learn about the qualitative aspects of the patient/health professional dynamic when the healing relationship becomes sexualized -from the perspectives of the patient and of the professional. As well, we are only beginning to appreciate the impact which sexual abuse in the childhood of a health professional and the experience of sexual harassment by, or sexual contact with, teachers in the course of professional training have on the ability of the professional to provide appropriate and effective care ( 1-2). In a survey of all Canadian psychiatric residents, of the 314 respondents, 6 women and 2 men reported sexual involvement with their educators.Although the majority of these 8 residents had positive or neutral feelings about this contact, 3 reported mixed feelings (3). The authors of this study concluded that resi- dency programmes should include education con- cerning student-educator sexual contact. One good model is the mandatory course for all psychi- atric residents given at the Jefferson Medical Col- lege in Philadelphia, United States of America. Another perspective was gained with the results of a national poll of 2 660 doctors in Canada, con- ducted by Decima Research and released in No- vember 1993. b Of the female respondents, 37% reported having been propositioned by a patient during a consultation, while 31 % of the male re- spondents reported that experience. Placing the available data in a sociopolitical con- text allows for gender analysis and the advantage of using both qualitative and quantitative research. Is sexual contact a breach of trust on the part of the health professional? The essence of the relationship between patient and health professional is trust. When this trust is a Corporate DirecLor, Women's Health ParLnerships, Women's College Hospital, Toronlo, Canada. b As reported inTheAledicall'osl, Ontario, Canada, November 1993. 154 abused, the results are devastating- devastating for the victim, usually affecting many aspects of her life: family, work, social relationships, mental and physical health (4-6) - and devastating for the faith society places in health professionals to govern themselves in the public interest (7). Patients seek help when they are in a vulnerable state - when they are sick, when they are needy, or when they are uncertain about what needs to be done. The unequal distribution of power in the health pro- fessional-patient relationship provides abundant opportunities for sexual exploitation (8). This vul- nerability gives professionals the power to exact compliance. Physical force or weapons are not nec- essary because the power comes from having the professional knowledge as well as being trusted by patients and society to act in the patient's best interest. Fiduciary duty is recognized in law as having the highest standard of conduct. The nature of the trust between lawyer and client has long been recognized as fiduciary, with severe consequences to a lawyer who breaches that trust; so too for the health professional (9). In Canada, this was clari- fied in a 1992 decision, when two female justices of the country's highest court, the Supreme Court of Canada, having reviewed the research (10-11), defined the doctor-patient relationship as follows: A fiduciary relationship is marked by the follow- ing characteristics: ( 1) the fiduciary has scope for the exercise of some discretion or power; (2) the fiduciary can unilaterally exercise that power or discretion so as to affect the beneficia- ry's legal or practical interests; and (3) the ben- eficiary is particularly vulnerable or at the mer- cy of the fiduciary holding the discretion or power. A physician owes his or her patient the classic duties associated with a fiduciary rela- tionship - loyalty, good faith, and avoidance of conflict of duty and self interest. ( 12) What constitutes sexual violation of patients? As yet, one particular term has not prevailed. Titles such as sexual abuse, sexual violation and sexual exploitation are virtually interchangeable in most jurisdictions. The process of defining sexual abuse has been based on actual incidents described by patients and from anonymous self-reporting by professionals. For example, Canada's first inquiry into the sexual abuse of patients by doctors, an Rapp. trimest. statist. sanit. mond., 49 (1996) independent task force commissioned by the Col- lege of Physicians and Surgeons of Ontario, (re- ferred to as the CPSO task force) heard over 303 detailed reports - 61 through public and private hearings, 203 through reports to the confidential 24-hour telephone line, 39 in letters. Sixteen of those reports pertained to male patients. Twenty- six were reports from third parties, including some from doctors reporting colleagues. As well, the task force members agreed to a request by an abusing physician to meet with him to hear his description of how his medical education and status "en- abled" multiple sexual contacts with several of his female patients ( 13). Testimony to the CPSO task force illustrated that sexual exploitation can take many forms. It can involve sexual intercourse (vaginal or anal), oral sex, touching breasts, genitals or any other sexual- ized body part for any purpose other than appropri- ate examination or treatment, encouraging the pa- tient to masturbate in the presence of the profes- sional or masturbation by the professional while the patient is present. Sexual exploitation can occur without physical contact. For example, sexual im- propriety can include gestures or expressions which are seductive or sexually demeaning, such as watch- ing as a patient undresses or dresses instead of providing privacy, making sexual comment or crit- icism about a patient's underclothing or body (ex- cept where clinically indicated), conversation with the patient about the professional's personal sexual preferences, problems or fantasies, kissing of a sex- ual nature, or requesting a date. Who's counting? Patients Canada has the most extensive consumer survey data on sexual abuse of patients in the world. The first such poll in Canada was conducted in 1991 in Ontario (which had over 22 OOO physicians li- censed to practice at the time) through telephone interviews in which 550 women over the age of 15 were asked two questions. The first question was designed to measure how many reported an expe- rience of sexual harassment and the second ques- tion was meant to gauge cases of sexual contact. Male interviewers conducted 383 of the interviews and in those, 6% of the respondents answered "yes" to at least one of the two questions, whereas in the 167 interviews conducted by women, 11 % answered "yes". Dr Earl Berger, whose company Canada Health Monitor conducted the survey, concluded that the blend of the two produced the conservative figure of 8% of Ontario women who said they had been either sexually harassed or abused by doctors, accurate within ±2. 7 percentage points, 19 times out of 20. b In 1992, in Canada, the Committee on Physician Sexual Misconduct of the College of Physicians and Wld hlth statist. quart., 49 (1996) Surgeons of British Columbia and the University of British Columbia departments of Health Care and Epidemiology and Psychiatry conducted "The Public Survey on Physician Sexual Misconduct", the first survey of its kind in the world. It was based on a random sample of the public, drawn from the voter registration list, utilising an anonymous survey sent by mail to 6 OOO women and 2 OOO men, 34. 7% of whom responded. The College of Physicians and Surgeons of British Columbia conducted a subse- quent "call-in" survey which found that of all the complaints concerning doctor-patient contact received by the College in 1993, 40% involved gen- eral practitioners, 17% gynaecologists, and 15% psychiatrists ( 14). Doctors Earlier in 1992, a survey of 600 doctors in Ontario, Canada ( 15) asked the question, "Has there ever been a time when you would have been required to report a colleague you had reason to believe was sexually abusing a patient?" to which 9% of the respondents said "yes". In this same survey, 57% agreed that "Doctors found guilty of sexual abuse should permanently lose their medical licences." In an American study, also conducted in 1992, of 10 OOO questionnaires sent to a national sampling of family practitioners, internists, obstetrician- gynaecologists and surgeons, 1 891 (about 19%) responded. Some of the answers to the survey prompted Dr Nanette Gartrell, the principal inves- tigator, to state in a media interview, "I am con- cerned about the potential for serious harm to patients. It's unfortunate that 9% of the physicians in this survey gratify their needs at the expense of patients and public trust." (16) The study defined sexual contact as "contact that arouses or satisfies sexual desires in the patient, physician or both." Other key findings of the study were that: • nearly 9 in 10 sexual contacts were between male doctors and female patients, 6% were be- tween female doctors and male patients, and 5% were homosexual or lesbian; • among those doctors who reported being sexu- ally involved, 42% had sex with more than one patient; and • close to two-thirds of all the doctors sur- veyed - including 38% of those who admitted having sex with their patients -said that doctor- patient sex was "always harmful" to the patient. Following this American study, the College of Physicians and Surgeons of British Columbia and the University of British Columbia departments of Health Care and Epidemiology and Psychiatry sur- veyed the 5 245 clinically active doctors in British Columbia, using the definitions developed by Gartrell, with the result that 3.5% of the respon- dents acknowledged sexual contact, 4. 7% acknowl- edged sexual touching, while 6% acknowledged making sexual remarks ( 14). 155 A 1994 national survey of Canada's obstetri- cians and gynaecologists yielded 487 men and 121 women who returned the questionnaires - a response rate of 78%. Overall, 10% of the respon- dents indicated that they knew of another obstetri- cian-gynaecologist who at some time had been sex- ually involved with a patient. In all, 3.2% of the men and 1 % of the women acknowledged that they had been sexually involved with a patient; the cor- responding proportions of those who reported having been accused of sexual abuse by a patient were 4% and 2%. Of the 17 who reported sexual involvement with a patient or former patient, 12 said the patient initiated the sexual contact. Most of the encounters were reported to have oc- curred in non-clinical settings, although sexual contact was repe;ted to have occurred on 5 occa- sions in the doctor's office. 74% supported loss of licence for proven sexual transgression. The au- thors concluded that obstetrician-gynaecologists have varied opinions about how sexual abuse of patients should be defined and sanctioned, result- ing in a discrepancy between proposed public poli- cy and the beliefs of physicians ( 17). Nurses In another Canada Health Monitor survey, con- ducted in 1993 for the College of Nurses of On- tario, a poll of more than 1 600 nurses and nursing assistants found that 46% had witnessed at least one example of physical or verbal abuse of a pa- tient by a nursing colleague, but witnessing sexual abuse was not reported ( 18). The intersection of health and law Released early in 1996, the most thorough review of North American data is found in the book Sexu- al abuse by professionals: a legal guide ( 19). The authors provide extensive citations and state the following general conclusions about the available research: • that sexual abuse is a breach of fiduciary trust; • that it is not restricted to any one profession, with perpetrators in education, health, law and religion; • that the power differential between profession- al and patient/client is not unique to North America; and • that there is no clear evidence to indicate either an increase or a decrease in the incidence of sexual abuse by those in positions of trust, but evidently, it is not "going away" as an issue. However, in Ontario, Canada, which enacted the Regulated Health Professions Act in 1994 re- quiring health professionals to report colleagues who allegedly committed abuse, an internal study conducted by the professional affairs department of the College of Physicians and Surgeons of On- tario (CPSO) noted that in 1992 and 1993, the 156 college received an average of 6 reports of abuse a month, then experienced an increase averaging over 20 a month, following passage of the new law. In a presentation to the International Confer- ence on Physician Health, held jointly by the Cana- dian and American Medical Associations in March 1996, Dr Laurel Dempsey of the CPSO was quoted as saying, "Does the data represent what is actually going on? Our data would suggest that there is lower activity than past physician self-reporting studies which found the incidence of sexual mis- conduct between 5 and 13%." (20) In her paper to the same conference, Dr Jennifer Schneider of Arizona, United States, reported on an assessment of 137 physicians found guilty of sexual misconduct which found that 54% of the offending physicians knew their behaviour was inappropriate but continued to do it. In their survey of the literature, Stotland and Harwood con- clude that "sexual and therapeutic relationships cannot coexist" (21). Emerging perspectives One of the most widely consulted experts in this field, American psychologist Gary Schoener, is in the process of preparing an update on the cases which focus on post-termination relationships and has expressed concern over the wide disparity among countries in the definition of sexual abuse (personal communication, April 1996). Dr Schoener points to the need for clear codes of conduct for health professionals, noting that self- reported data grossly underreport the incidence because many health professionals define only sexual intercourse as "sexual contact". Confer- ences, like the first Australian and New Zealand conference on sexual exploitation by health pro- fessionals, psychotherapists and clergy, held in April 1996, in Sydney, Australia, provide crucial opportunities to review and refine research in this evolving field. Conclusion As this generation addresses the health security needs and the human rights of women in our glo- bal society, it may be helpful to bear in mind that the sexual exploitation of patients is not a new issue, as the following quote from the Hippocratic oath demonstrates: \-\'hatever houses I may visit, I will come for the benefit of the sick, remaining free of all intentional injustice, of all mischief and in particular of sexual relations v:ith both male and female persons, be they free or slaves. Acknowledgment The author wishes to thank Dr Aleya El Bindari Hammad for her encouragement to prepare this article for an international audience. Rapp. trimest. statist. sanit. mond., 49 (1996) Summary In an examination of the health security of women, sexual exploitation of women patients when the perpe- trators hold the trusted position of care provider, merits some considerable attention. Availability of data varies, with the majority gathered in the United States and Canada, focusing on medicine, psychotherapy and nursing. However, even for those professions, this field of research is in its early stages. Within the past few years, research has been undertaken in Australia and New Zealand, Germany, the Netherlands, Norway and the United Kingdom. Central to any discussion of this aspect of women's access to safe and appropriate health care are 3 questions about data. Do we know whether women are the principal victims of this abuse? Do such numbers count? Do we in fact need more data collection as we learn to respond more effectively to this particular form of violence against women? The author. who chaired Canada's first inquiry into the sexual abuse of patients by doctors. answers all 3 questions in the affirmative, with an emphasis on combining data collec- tion with action. The sampling of research in this article demonstrates the need to combine qualitative and quantitative data to gain a more accurate understanding of the dynamics of abuse, within the social context of women's experience. Resume Abus sexuels dans /es professions de sante - quicompte? Lorsqu'on se penche sur la securite sanitaire des fem- mes. ii convient d'accorder une attention considerable a !'exploitation sexuelle des malades femmes lorsque ceux qui s'en rendent coupables detiennent la position digne de confiance de dispensateurs de soins. L'exis- tence de donnees varie selon les pays, la majorite etant regroupee aux Etats-Unis d'Amerique et au Canada, et etant axee sur la medecine, la psychotherapie et les soins infirmiers. Cependant, meme pour ces profes- sions, le present domaine de recherche n'en est qu'a ses debuts. Au cours des dernieres annees, des recher- ches ont ete entreprises en Australie et en Nouvelle- Zelande, en Allemagne, aux Pays-Bas, en Norvege et au Royaume-Uni. Au cceur de toute discussion sur cet aspect de l'acces des femmes a des soins de sante surs et appropries se trouvent trois questions sur les don- nees. Savons-nous si les femmes sont les principales victimes de ces abus? Ces nombres comptent-ils ? Avans-nous en realite besoin de recueillir davantage de donnees a mesure que nous apprenons a riposter de maniere plus efficace a cette forme particuliere de violence a l'encontre des femmes? L'auteur, qui a pre- side la premiere enquete canadienne sur les abus sexuels perpetres a l'encontre de leurs malades par des medecins, repond aces trois questions par l'affirmative, en soulignant qu'il taut combiner la collecte des don- nees avec l'action. L'echantillonnage des recherches dans cet article montre qu'il est necessaire de combiner des donnees qualitatives et quantitatives pour corn- Wld hlth statist. quart., 49 (1996) prendre plus precisement la dynamique des abus sexuels, dans le contexte social de !'experience des femmes. References/References 1. Gallop, R. et al. The impacL of child sexual abuse on the psychological well-being and practice of nursing. Archives of jisyrhiatric 1iu1,ing, 9: 137-143 (1993). 2. Gallop, R. et al. Inquiring about childhood sexual abuse as part of the nursing history opinions of abused and nonabused nurses. Archives of /1sychiatric nursing, 9: 146-131 (1993). 3. Carr, Melanie et al. A survey of Canadian psychiatric residenL~ regarding resident-educator sexual contact. American journal of /1sychiatr;, 148 (2): 216-220 (1991). 4. Robinson, G.E., TreaLing female patienLS. Canadian Medical Associalionjourna~ 150 (9): 1427-30 (1994). 3. Fahy, T. & Fisher, N. Sexual conLacL between doctors and paLienL~: almost always harmful. British medical journal, 304: 1319-20 (1992). 6. Talley, Nicholas J. et al. GastroinLestinal tract sympLoms and self-reported abuse: a population-based study. Gastrn- enlerowgy, 107 1040-1049 (1984). 7. Council on Ethical and Judicial Affairs, American Medical Association. Sexual misconduct in the practice of medicine: council report.journal of Lhe Amniwn ,WedicnlAssociation, 266 (19): 2741-3(1991). 8. Carr, M. & Robinson, G.E. Fatal atLraction: the eLhical and din ical dilemma of paLient- Lherapist sex. Canadian journal of psychiatry, 35: 122-27 (1990). 9. Feldman-Summers, S. Sexual coni.act in fiduciary rela- tionships. In: Gabbard, G.O. (ed.) Sexuale>.t1loitation in profes- sional rel.ationships. Washington, DC, American PsychiaLric Press Inc., 1968. 10. Coleman, Phyllis. Sex in power dependency relationships: taking unfair advantage of the "fair sex". Albany law review, 53: 93. l l. Jorgenson, Linda & Randles, Rebecca M. Time-out: the sLatute of limitations and fiduciary theory in psychothera- pist sexual misconduct cases. Oklahoma law review, 44: 181 (1991). 12. L'Heureux-Dube, Hon. Mme Justice & Mcl..achlin, Hon. Mme Justice. In: L-iuraNorbergv. Dr. Morris Wynrib and the Women's Legal Education and Action Fund as Intervener, Supreme Court of Canada, 2 SCR: 226 ( 1992) . 13. McPhedran, M. et al. The final report of the independent task force on sexual abuse of patients by physicians.Toronto, The College of Physicians and Surgeons of Ontario (1991). 14. Crossing Jhe boundaries, The College of Physicians and Surgeons of British Columbia, Canada, 1993. 13. Environics Research Group. Ontario medical review, February, 1992. 16. Gartrell, Nanette K. et al. Physician-patient sexual contact: prevalence and problems. H'iestern journal of medicine, 157: 139-43 ( 1992). 17. Lamont, John A. & Woodward, Christel. Patient-physician sexual involvement: a Canadian sun·ey of obstetrician- gynaecologim, Canadian MedicalAssociation journal, 150 (9): 1433-1439 (1994). 18. CollegeofNursesofOnlarioCommuniquti, 18: (4) I (1993). 19. Bisbing, Steven et al. Sexual abuse by professionals: a legal guide. Charlottesville, Virginia, The Michie Company (1993). 20. Quinn, M. Physician sexual misconduct higher Lhan rcporLS suggest, Family practice - the C.anadian newspaper of primary care, 8 (6): 1 (March 4, 1996). 21. Stotland, Nada L. & Harwood, Bryna. Sexual contact between patienL~ and physicians: romance or betrayal? Journal of women's health, 4 (2): 197-208 Mary Ann Liebert, Inc. (1993). 157 Women and migration: a public health issue Manuel Carballo,a Mandy Grocuttb & Asja Hadzihasanovicc Introduction Mass migration is the product of many factors. Sometimes they come together and act synergisti- cally; at other times they function more explicitly alone. Migration is usually the end result of pover- ty, limited opportunities, land pressures, ecological degradation, political conflict and the denial of human rights. In a world of rapid communication and growing international commerce and indus- try, people can move further and more quickly than at any other time in history. The potential impact of migration on their own health, that of the societies they leave, and that of the communi- ties they settle in has thus become an important component of national and international health development. The past 50 years have also seen uprooting and forced movement of people rise to a scale that has rarely been experienced before. Bet- ter transportation has made it possible for people fleeing natural and man-made disasters to move in larger numbers and further than before, making disaster-related mass migration a significant source of social, health and demographic change. Migration has historically been seen as process beginning in the search for new opportunities and ultimately culminating in the enjoyment of new and improved ways oflife. And indeed where there has been implicit or explicit respect for the funda- mental human rights of migrants, their social inte- gration and related health outcomes, as measured by such indicators as educational and occupational achievement on the one hand and pregnancy out- come on the other, have pointed to the positive results of migration (1-5). In the context of womens' personal development, for example, moving from traditional settings or from poor so- cial environments to more modern and economi- cally sound situations can often facilitate the edu- cational and professional mobility that would have otherwise been impossible. But no matter what the underlying causes of migration, or the form it takes, the process of migration almost always in- volves a degree of social and personal cost, and a Coordinator, International Centre for Migration and Health, Geneva, Switzerland h Senior Lecturer, Department of Emergency Medicine, Keele University Hospital, UK c Technical Officer, International Centre for Migration and Health, Geneva, Switzerland 158 even under the most ideal of circumstances, it is a stressful life event. Irrespective of how well orga- nized and prepared people are for it, migration often involves breaking well established ties with family, community and culture. It is usually fol- lowed by the insecurity of not knowing what will happen in the future and having to adapt to new value systems (6) and sometimes contradictory life- styles. Migration also tends to mean a temporary (at least) social marginalization which comes as a result of linguistic and social barriers and the often poor acceptance of newcomers by host communi- ties. The effect of migration on health can thus be complex. Its impact is likely to be determined by a wide range of circumstances, including the origi- nal "push factors" which precipitated the decision to move, the psychosocial and physical conditions which characterized the move itself, and the socio- economic, political and health environment in which people are resettled. In the context of the massive migrations which are now taking place in many parts of the world, the health outcomes asso- ciated with population movement have become a major challenge to international as well as national public health policy. Uprooting and movement nevertheless affect people in different ways and to different extents. For a variety of social and biological reasons, women are often more vulnerable than men to problems associated with migration. Some of those reasons have their origin in the status and role of women, especially in traditional and poor societies. Other reasons reflect the unique biological charac- ter and needs of women, particularly with respect to reproduction, sexual relations and physical power. Magnitude and nature of the problem Although there are few reliable global data on the size of contemporary uprooting and migration, the overall flow of people within and between coun- tries and regions appears to be increasing. In poor countries the flow of young people from rural ar- eas towards urban centres in search of work con- tinues to grow and involve millions of people per year. There is also an increase in the movement of people between developing countries, and the tra- ditional pattern of south-to-north migration may be gradually giving way to a more predominantly south-to-south migration. Even so, immigration to Rapp. trimest. statist. sanit. mond., 49 (1996) traditional "receiving" countries (United States of America, Canada, Australia) is estimated to still involve at least a million people per year. Mean- while, over 22 internal and inter-country conflicts around the world have produced around 50 mil- lion refugees and people internally displaced in their own countries. Migration has been typically characterized as being either voluntary or involuntary in origin, and much of the literature on the subject has revolved around this distinction. But while useful as a gross typology, the classification overlooks the fact that at some level or another most human migration is prompted by severe "push" as well as "pull" fac- tors. Ecological degradation, chronic poverty, pres- sures on scarce land and resources, as well as poor schooling and limited job opportunities are part of a complex interaction of factors leading to the difficult decision to uproot and move elsewhere. In recent years a number of changes have also taken place in the demography of migration. For while much of the early migration from Europe to the Americas and Australia was at least a nuclear- family affair, contemporary migration is far less so. Labour migration has become a highly age-selec- tive and sex-selective phenomenon favouring younger adults who can meet the stringent de- mands of the job market and immigration require- ments. On the one hand, primary industries such as mining, forestry and fishing, as well as construc- tion projects call for a predominantly male migra- tion, while on the other the flourishing "domestic help" and growing free-trade zone textile industries call for young women who are often expected to remain away from their homes for two to three years at a time. Meanwhile, the immigration and labour poli- cies that have emerged in receiving countries have not helped support or enhance family cohesion. Indeed in many instances, they have been de- signed to deny the possibility of families moving together. As a result, labour-driven population movements throughout much of the world are become highly skewed towards either men or wom- en. In southern Africa and the Caribbean, for example, it is often the men who are more likely to move and leave females, children and older people behind; in others regions such as South-East Asia, the growing industrial demand is for women migrants to move alone. In many parts of the world migration is also becoming a temporary or circular phenomenon in which people leave home for indefinite and often long periods of time to work abroad or in other regions of their own countries before returning for short visits at home. So-called circular migration has become an important feature of population movement in developing countries, and is a partic- ularly important source of family disruption, effec- tively depriving families of any real continuity or Wld hlth statist. quart., 49 (1996) "intactness" for months and years at a time even though it may involve relatively short distances. In the case of Mali, well over a half of all the circular migration involves women who leave during peri- ods of drought and then do not return until they have earned enough to help see the family through the next period of agriculture development (7). Elsewhere in Africa, Ghanaian women constitute a large percent.age of the people who travel to and fro along the coast of in search of employment. Involuntary or forced migration is typically the outcome of violence and conflict. Coming unex- pectedly, and providing little time for preparation, forced uprooting disperses families indiscriminate- ly and quickly. Because men are often conscripted or, as in the case of Bosnia, immediately rounded up and imprisoned, numerous reports indicate that women and children are by far the most nu- merous group in refugee and internally displaced populations (8-9). The physical and psychological violence associated with man-made disasters adds both immediate and long-term damage, and even in resettlement, refugee women often find them- selves politically and physically insecure for long periods of time. Because of the sex selectivity of contemporary migration, more and more women are either being left behind with responsibility for children and elderly relatives, but without any of the other sup- portive mechanisms which they may have been used to ( 10), or are being forced to flee alone or with children but without spouses or older relatives to share the psychological and physical burden. Social aspects Women move for much the same reason as men, and although migration may open up many new opportunities, the fact remains that many women face serious and resistant barriers to personal and professional development. For some women, mi- gration may mean an interruption in whatever lim- ited educational opportunities they had, and reset- tlement may offer few possibilities for continued education. When they come from traditional or socioeconomically poor backgrounds, they may face even more difficult problems of social integra- tion than men. They are often denied the possibil- ity of professional improvement in the communi- ties they settle in ( 11), because they move with few of the skills needed in modern industrial societies, and remain excluded from opportunities for per- sonal development (12-13). Official planning of economic migration schemes may also fail to acknowledge the implicit value of women, and as was the case in early Indo- nesian transmigration programmes, and base themselves on orthodox economic and statistical concepts which devalue the contribution of wom- en and anchor them more in domestic activities ( 14). Organized labour migration schemes, such as 159 those run by the private sector in Sri Lanka, often ' lead to abuses in fees, contract~. and fraudulent job offers, particularly in the case of female workers who are exploited in a variety of ways with little chance of legal recourse ( 15). As a result of barriers to economic mobility, many women migrants find themselves confined to recreating the lifestyles and conditions remi- niscent of their situation at home before they migrated (16). This is especially so when they lose the family-linked status which they had at home in traditional societies. In the case of Vietnamese ref- ugees in the USA, it meant that women had special difficulties in re-establishing themselves in a mod- em environment and felt at a loss to define a new identity for themselves ( 17). The loss of social sup- port networks among young women can also have serious implications for marriage and family life, and has been identified as a major source of psy- chological difficulty among Iraqi women in exile in the United Kingdom (18) and refugees from El Salvador in the United States ( 19). There is also a growing body of evidence that migrant women face considerable sexual harass- ment, violence and exploitation by employers, es- pecially but not only in countries where protec- tion of foreign workers is poor (15,20). Female migrants often find themselves in situations where potential unemployment and fear of deportation compel them to provide sex to employers and others (21). The extent to which transactional sex is imposed on women in return for legal and job rights is thought to be a major human and civil rights problem everywhere, but certainly poorly described. Displaced and refugee women are even more vulnerable, and even in the so-called safe-haven of refugee camps may have to provide sex in ex- change for food, shelter and "protection". Rape is not uncommon either, according to the United Nations High Commissioner for Refugees (UNHCR). Unaccompanied women in refugee camps are at considerable risk of sexual violence not only from external aggressors but also from other refugees, camp staff and even UN civilians and military forces there to protect the camps. As the UN High Commissioner for Refugees has high- lighted, "Rape, abduction, sexual harassment, prostitution, physical violence and the not infre- quent obligation to grant sexual favours in return for documentation and/ or relief goods remain a distressing reality for many refugee women" (22). Steps to provide women with security from sex- ual abuse are being given higher priority in the work of humanitarian organisations, but neither the UNHCR Statute nor the 1967 Protocol contains articles specifically protecting refugee women. The only relevant article places the family, which is where much of the violence against women occurs, outside the scope of refugee law (22). 160 Migration and sex work have also become more closely inter-linked in many parts of the world and primarily involve the exploitation of women from poor countries or bad<.grounds. Trafficking in women has become a global problem and takes a variety of forms (23). Many women are offered one- way tickets and are then forced to sell sex under the conditions laid down by their recruiters. In Amsterdam 90% of so-called "window prostitutes" are foreign to the Netherlands, and in Turkey there have been reports of organized and relatively large scale busing of young females from Rumania for purposes of prostitution. In Spain, France and Italy, women from sub-Saharan countries are high- ly represented among street prostitutes, and Rus- sian female sex workers have become prominent among prostitutes in Lithuania, the Czech Repub- lic, Slovakia and Hungary (23). The fact that in many of these situations they risk deportation makes them all the more vulnerable to further exploitation and the lack of legal recourse. The latter part of the 20th century has also seen sex tourism become a major international industry involving hundreds of thousands of clients per year. To maintain an increasingly lucrative indus- try, women and men (but especially the former) are being recruited from all over the world to vaca- tion centres specializing in organized sex. In Thai- land, one of the more open sex-vacation centres, the traditional pattern of recruiting young women from rural areas within the country is gradually expanding to include an even more expansive pro- cess in which women are now coming to work from as far away as Europe and Australia. While a num- ber of countries are considering the introduction of laws penalizing sex tourism involving minors, there has been little attempt to protect women of legal age or to control the industry in general. Health implications As a result of the physical and psychosocial condi- tions under which people move, both voluntary and forced migration can he associated with ad- verse health outcomes. In some cases these may reflect the nature of the migration itself, but in others they may reflect a more complex relation- ship between the health of the migrant prior to migration and appropriate and timely access to health services on resettlement (24). Adverse health outcomes may reflect administrative obsta- cles to care, including residence conditions which need to be fulfilled before services can be accessed, and may also reflect a lack of awareness about what services are available, or a culturally or psychologi- cally defined resistance to seeking help (25). In other instances adverse health outcomes may be the result of linguistic barriers which prevent mi- grants from making appropriate use of the health services that do exist and for which they would be eligible. In general, however, health care services Rapp. trimest. statist. sanit. mond., 49 (1996) are rarely defined according to the needs of mi- grants and refugees, nor are staff necessarily trained to look for what may be unique health problems and requirements among refugees (26). The relative absence of women in senior admin- istrative positions at refugee camps has also been highlighted as a contributing factor to the lack of attention to women's issues (27), but in general the fact remains that the health and well-being of refu- gees and migrants is itself often given low priority and few steps are taken to explicitly tailor services to their needs. As a result, traditional practices which also interfere with the promotion of health are often lost from sight in context of all the other crises surrounding refugee movements and their management. Thus, the practice of purdah among Afghan refugee women, which was found to be associated with generalized vitamin D deficiency as a consequence oflimited exposure to sunlight, and is also thought to have then contributed to higher- than-average rates of tuberculosis (28), was not identified until late in the process. The social status of women within their own migrant and refugee communities may also be a constraint to health seeking behaviour. A report on 28 OOO refugees in Sudan, for example, indicat- ed that while 75% were women and children, all in- patients and almost all out-patients were men. Whether this was because the health providers were men is not clear, but in many situations, tradi- tional attitudes and beliefs concerning the rela- tionship between men and women have proved to be a serious constraint to providing women refu- gees with much-needed health services while in camps (29). Human reproduction In general, pregnancy outcome and perinatal health indicators tend to be worse among migrants and refugees (3-5) and among displaced women. The situation in Bosnia demonstrated the need for careful and technically sound support to be pro- vided to women in disaster situations (30, 31). But where the fundamental right to timely and equal health care is respected and where steps are taken to make access to health services logistically and psychosocially feasible, there is evidence from sen- sitive indicators such as pregnancy outcome and perinatal mortality that refugees and migrants tend to "catch up" with the better standards of host countries. Studies of female migrants in sub-Saharan Afri- can countries also show that migration from rural to urban areas dramatically decreases the "risk" of conception and that fertility, once lowered, re- mains lower. Descriptive analysis suggests that this may be due to improvement'> in standards ofliving (32), but the fact that women tend to migrate alone in many parts of Africa may mean that the absence of the spouse is also a key factor. But the picture is Wld hlth statist. quart., 49 (1996) not consistent everywhere, and rural-to-urban fe- male migrants in Cameroon (33) showed no major change in fertility-related behaviour or outcomes. The impact of migration on fertility patterns has also been highlighted in the Eastern Caribbean where migration has replaced mortality as the lead- ing cause of child loss, and where high rates of reproduction may be linked to this (34). On the whole, however, cmde death rates of refugees are higher than baseline rates for coun- tries of origin, and within this pattern women may be the more vulnerable. In the case of Bangladesh the death rate among female refugees was several times higher than that of males (35). The health of women refugees is also threatened as a result of sexual violence, and this has become a major and possibly increasing public health problem. A study of women displaced by the war in Mozambique indicates that almost 9% of ali pregnant women had been raped (36), and more recently in Bosnia refugee women were reported to have been repeat- edly raped as part of the ethnic cleansing and genocide (37). One of the regrettable natural consequences of the rapid expansion in sex tourism, sexual exploi- tation, and violence against women has been the increased risk of sexually transmitted diseases, in- cluding HIV and AIDS. Among migrant<; and refu- gees, sexually transmitted diseases including HIV are often spread through heterosexual contact in which rape and prostitution are key factors (38). Sero-surveys of pregnant female refugees in Mozambique found a syphilis prevalence rate of 12%, and the fact that the prevalence of HIV was only 3% suggests that HIV may have been recently introduced into the population as a result of the sexual violence which women were exposed to in the war (37-39). HIV infection as a result of sexual violence against women in war situations in Africa has also been reported elsewhere (40). In part, the problem lies in the fact that repro- ductive health programmes and the protection of women in refugee situations has tended to be defi- cient. As a result, problems such as genital tract infections have often gone unidentified or poorly treated. The fact that victims of rape are unlikely to come forward, especially in traditional societies, has also added to the more strictly biological prob- lem that STDs in women may often be asympto- matic, and the perceived need to seek treatment may subsequently be low. Uprooting and migration may also interlere with traditional patterns of breast-feeding. In refu- gee situations breast-feeding practices may be ad- versely influenced by the indiscriminate introduc- tion and distribution of milk powder and other supplementary foods which can be and are given to women ·with young infants. Migration to urban centres and to economically more developed countries may also influence women to adopt local 161 customs and what they believe are more "mod- em" practices. This was the case with Indo-Chinese women moving to Australia (41) and has been an- ecdotally reported elsewhere. Its implications for the risk of new pregnancies in situations where family planning services may not be available, ac- cessible or culturally acceptable, is serious, espe- cially since the lactational amenorrhea associated with traditional breast-feeding practices may have been an important source of contraceptive protec- tion (42). Family health in general is also affected by mi- gration because in most parts of the world, wom- en are a vital source of family health care and care of children. Interruptions to the everyday status and life of women can immediately have an im- pact on their capacity to cope with the needs of other family members, and because all types of migration impose some changes on the role and status of women, their caring role may also be impaired. The loss of supportive social structures may equally mean that women are left to define new roles and seek employment, which in turn may create conflict within the household. Jewish wom- en refugees from World War II found it easier than men to adapt economically and fitted quickly into local job markets that sought cheap labour (43), and the more recent migration of Vietnam- ese refugees to the United States saw women often able to earn more than their male counterparts, but while this was a positive outcome from one perspective, it has sometimes disrupted traditional family values and roles, and increased intra-famil- ial stress ( 44). Psychological problems are probably the most pronounced of all health outcomes associated with migration, culture conflict and resettlement. In general women appear to be more vulnerable to the stress of uprooting and resettlement, or at least to be able express stress-related problems (17, 45, 46). Hmong immigrant women are re- ported to have a high incidence of unexplained nocturnal death which is thought to be associated with the confrontation between traditional beliefs (12) and modern society. Psychosomatic blind- ness has been reported among women refugees from Cambodia (6) and Cambodian women have also often presented with other complex somatic manifestations of stress (47) and have also report- ed more frequent use of alcohol for stress, insom- nia and pain (48). Among Soviet Jewish refugees in the United States, women appear to be more traumatized by t.he need to adopt new lifestyles, roles and values than men, and felt much more alienated in the process of trying to do so (43). Similar responses have been reflected in other studies of refugees (19), and have also been re- ported in populations displaced by natural disas- ters (49-51). 162 Summary The need to migrate is usually a function of the complex interaction of economic, social, familial and political factors. Among the most important. however, are the denial of access to education, employment. goods and services and the lack of respect for basic human rights. Because in many societies women are marginalized from these rights, migration to more economically and educationally open societies can often help improve their personal situation and their professional opportuni- ties. On the other hand, because the status of women is usually linked to their role and status within the family and is defined in relationship to their male partners, migration can place women in situations where they experience stress and anxiety due to the loss of their traditional social entourage and environment. Their so- cial integration in new settings may be equally limited by their initial lack of education and occupational experi- ence. The higher vulnerability of women to sexual abuse and violence also places them at risk of STDs, including HIV, and a range of post-traumatic stress disorders associ- ated with sexual violence. Their reproductive health needs often go unnoticed and unprotected even in well organized refugee and migrant situations, and the in- sensitivity of health staff to the needs of women is often more pronounced in refugee and migrant contexts than it is in general. Health monitoring of women in all migration-related situations has to be given greater priority. Similarly, much more attention at a health policy level is called for if the rights of women refugees and migrants are to be protected, and their contribution to health and social development is to be acknowledged and promoted. Resume Les femmes et les migrations: question de sante publique La necessite de migrer provient en general d'une inter- action cornplexe de facteurs economiques, sociaux, familiaux et politiques. Parmi les plus importants, toute- fois, figurent le refus de l'acces a l'education, a l'emploi, aux biens et services et !'absence de respect des droits fondamentaux. Du fait que les femmes sont dans de nombreuses societes ecartees de ces droits, la migra- tion vers des societes plus ouvertes sur les plans eco- nomique et educatif leur permet souvent d'ameliorer leur situation personnelle et leurs possibilites profes- sionnelles. Par ailleurs, du fait que le statut des femmes est habi- tuellement lie a leur r61e et a leur position au sein de la farnille et est defini par rapport a leurs partenaires masculins, la migration risque de placer les femmes dans des situations ou le stress et l'anxiete qu'elles connaissent sent dus a la perte de leur entourage et milieu social traditionnel. Leur integration sociale dans Rapp. trimest. statist. sanit. mond., 49 (1996) de nouveaux milieux risque d'etre egalement limitee par leur absence initiale d'education et d'experience pro- fessionnelle. La vulnerabilite plus grande des femmes vis-a-vis des abus et de la violence sexuelles leur fait egalement courir le risque de contracter des MST, et notamment le VIH, et les expose a des troubles dus au stress post- traumatique associe a la violence sexuelle. Leurs be- soins en matiere de sante reproductive passent souvent inapen;:us et non proteges, meme dans des situations de refugies et de migrants bien organisees, et l'insensi- bilite du personnel de sante vis-a-vis des besoins des femmes est encore plus prononce dans des contextes de refugies et de migrants qu'il ne l'est en general. II taut accorder une plus grande priorite a la surveillance sanitaire des femmes se trouvant dans toutes les situa- tions liees aux migrations. Parallelement, ii taut accorder davantage d'attention au niveau des politiques sanitai- res si l'on veut proteger les droits des femmes refugiees et migrantes et reconnaHre et promouvoir leur contribu- tion au developpement sanitaire et social. References/References I. Rojnik, B. et al. Women in difficult circumstances: war \~ctirns and refugees. international journal ofgynaemlogy and obstetrirs, 48(2): 311-315 (19Y5). 2. Mann, J. Health .i.nd human right~. B1istish mediral journa~ 7036(312): 924-925 (Aptil 1996). 3. Malamitsi, P.A. et al. Pretcrrn dcli\'ery and low birthweight among refugees in Greece. Paediatric and p,mnatal Pf1idemiol.ogy, 8 ( 4): 384-390 (1 Y94). 4. Wasse, H. et al. Pregnancy risk factors and birth outcomes in Washington State: a comparison of Ethiopian-born and US- born women. Ammimn journal of public health, 84(9): 1505- 1507 (IY94). 5. Doncet, H. et al. Risk of low birth weight and prematurity among foreign-born mothers. Ammam journal of public health, 83(3), 192-195 (1Y92). 6. Mattson, S. Mental health of Southeast Asian refugee women: an overview. Health mre womnz international, 14(2): 155-165 (1993). 7. Findley, S.E. Does drought increase migration? A study of migration from rural Mali during the 1982-1985 drought. International migration nroiew 28(3): 539-553 ( 1991). 8. Samary, C. Refugees from former Yugosl,l\ia facing the European fortress. Innovation, 7(2): 189-198 (1991). 9. Keely, C.R. The resettlement of women and children refugees. Migration world magazine, 20(4): 14-18 (1992). 10. Miller, L.C. Afghan refugee children and mothers. Archives of pediatrir. and adolesant medicine, 148(7): 704-708 (1991). 11. Adler, S.R. Ethnomcdical pathogenesis and Hmong immigrants' sudden deaths. Culture, medicine and psychiatry, 18(1): 23-59 ( 1994). 12. Knocke, W. Gender, ethnicity and technological change. Economic and indmtrial demorra0•, 15( I): 11-34 (l 9Y4). 13. Makinwa, A. & Paulina, K. Women migrants in Nigeria. International sociology, 9(2): 223-236 (1994). 14. Dawson, G. Development planning for women: the case of the Indonesian transmigration program. Women's studies international forum, 17(1): 69-81 (19Y4). 15. Soysa, G.D. Sri Lankan overseas women workers: policy and issues. A.sian migrant, 6(3-4): 89-95 (1993). Wld hlth statist. quart., 49 (1996) 16. Vicarelli, G. Emigration and the tabor market: gender differences. Sociologia del lavom, 47-48 (1992). l 7. Gold, S.J. Mental health and illness in Vietnamese refugees. Westemjournalofmedirine, 157(3): 290-294 (1992). 18. Al-Rasheed, M. The meaning of marriage and status in exile: the experience ofiraqi women.jm.trnalof refugeestudies,6(2): 89-101 (1993). 19. Jacob, A. Adaptation of Salvadoran refugees in Montreal. Migmtion worldmagazine, 20(1): 21-21 (1992). 20. Lipson,J.G. & Muler, S.Changing Roles of Afghan Refugee 'Women in the United States. Health care for women intemational, 15(3): lil-180 (1991). 21. Mardh, P.A. & Gen~, M. y[igratory prostitution with emphasis on Europe. Jounwl of travel mediline, 2( I): 28-32 (1995). 22. AJ-Omari, G. The legal protection of refugee women .Refugee f1artidpation network, 20: 14-16 (November 1995). 23. International Office on Migration (IOM). Trafficking and /1m.Hit11tio11: the growing exj,loitation of migra11t women from Cl'i1tral mul Ea{ltTn },'urof>P. Geneva, !OM, 1995. 21. King, P.A. Born before arri\'al. Australian and Xew 7.ealand journal of obstetrics and J;Ynaecowgy, 32 ( l): 6-9 ( 1992). 25. Carballo, M. & Siem H. Migration, migration policy and AIDS in migration. In: Ethnic minorities and AIDS. Social aspecL~ of AIDS series. London, Taylor & Francis, 1996. 26. Pickwell, S.M. et al. "lletclmania": betel quid chewing by Cambodian women in the United States and it~ potential health effects. m,sten1 journal of medicine, 160( 4): 326-330 (1991). 27. Walker, B. The question of gender: Refugee l'artirif){/tion ,\'dwork, 20: 8-13 (November 1995). 28. Field Studies Paper No. 2. Geneva, League of Red Cross and Red Crescent Societies, 1991. 29. Scott, V. et al. Repatriation of 150 OOO Sudanese refugees from Ethiopia: the manipulation of ci\'ilians in a situation of civil conflict. Disasters, 17(3): 202-217 (1993). 30. International Centre for Migration and Health. Reproductive health and pregnant} outcome among di.1puiced women: Report of the Techniwl Woming Gmup. Geneva, ICMH, 1995. 3 l. Carballo M., et. al. Health in countries torn by conflicts: lessons from Sarajern. Lancet, 348: 872-874 ( 1996). 32. Brockenhoff, M. & Yang, X. Impact of migration on fertility in Sub-Saharan Af'rica. Social biolr,gy, 41(1-2): 19-43 (1994). 33. Lee, B.S. The influence of rural-urban migration on migrant's fertility behaviour in Cameroon. international migration review, 24(4): 1416-1445 (1992). 34. Brittain, A.W. Anticipated child loss to migration and sustained high fertility in an Eastern Caribbean population. jrntmal of biology, 38( 1-2): 94-112 ( l 9Y l). 35. Bern. C., et al. Risk factors of mortality in the Bangladesh cyclone of 1991. Bulletin of the World Health Orgtmization 71(1): 73-78 (1993). 36. Cossa, H.A. et al. Syphilis and HIV infection among displaced pregnant women in rural Mozambique. Intemationaljournal of STD and Al/JS, 5(2): 117-123 (1994). 37. Kozaric-Kovacic, D. et al. Rape, torture and traumatization of Bosnian and Croatian women: psychological sequelae. A.111P1imnjoumal of orthop5ydiiatry, 65(3): 428-433 (1993). 38. Davidson, S. \,\,11at is reproductive health care? Refugee Participation Network, 20: 4-8 (November 1995). 39. Cade, K.J. Ethnicity, war and rape. Migracij'sketenll!, 8(2): 95- 104 ( 1992). 40. Seidel, G. Women at risk: gender and AIDS in Af'rica. DiS(L\tPYS, 17(2): !33-112 (1993). 41. Rossiter, J.C. et al. Indochincse women's breastfeeding practices following immigration to Sydney: a pilot study. A1tstralian journal of nursing, 10(3): 3-9 (1993). 42. Carballo, M. Contemporary patterns of breastfeeding. Geneva, World Health Organization, 1982. 43. Birman, D. & Tyler, F.B. Acculturation and alienation of Smietjewish refugees in the United States. Genetic, sodaland general psyrhology monographs, 120(1): 103-115 (1994). 163 44. Fox, P.G. Stress related to family change among Vietna· mesc refugees. Jourrwl of community nursing, 8(1): 45-56 (1991). 45. Chung, R.C. & Kagawa, S.M. Predictors of psychological distress among Southeast Asian refugees. Social science and medicine, 36(5): 631-639 (1993). 46. Summerfield, D. & Toser, L "Low intensitv" war and mental trauma in Nicaragua: a study in a rural community. Medicine and war, 7(2): 84-99 (1995). 47. Frye, B.A. & D'Avanzo C. Themes in managing culturally defined illness in the Cambodian refugee family.Joun.al of communit)' health nursing, 11(2): 89-98 (1994). 164 48. D'Avanzo, C.E. Stress in Cambodian refugee families. Jmag,,-thejournal of nursingscholarship,26(2): 101-105 ( 1994). 49. Viinamiki, H. et al. The Chernobyl accident and mental well- being: a population study. Acta psychiatrica scandinavica, 91 (6): 396-401 (1995). 50. Webster, R.A. et al. Effects of a natural disaster on immigrant~ and host population. journal of nervous and mental disease, 183(69): 390-397 (1995). 51. Green, B.L. et al. Children of disaster in the second decade: a 17 year follow-up of Buffalo Creek survivors. Journal of the American Academy of Chiul and Adolescent Psychiatry, 33( 1):71- 79 (1994). Rapp. trimest. statist. sanit. mond., 49 (1996) Publications of the World Health Organization 1996 WORLD HEALTH FORUM An international journal of health development (Separate editions in English, French, Spanish, Arabic, Chinese and Russian) World health forum is a quarterly journal for policy-makers, health planners, administrators, health educators, and public health workers of all kinds. It provides a medium for the presentation and discussion of new concepts in puhlic health and new approaches to health problems, and it is devoted to the improvement of health through the promotion of health services covering the entire population and the undertaking of a wide variety of public health measures, whether or not they are supported by WHO. The Forum is the main organ ofWHO's Member States for the international exchange of health information and an active instrument in technical cooperation among developing countries. Subscription ( 4 numbers) ............................................... . Price per copy ............................................................... . Sw. fr. 75.- Sw. fr. 20.- BULLETIN OF THE WORLD HEALTH ORGANIZATION The Bulktin is the principal scientific organ of WHO: its role is Lo review progress in medical and related sciences ("update" articles) and bring to light new knowledge by publishing original papers on scientific research in the laboratory and field. The Bulktin contains original articles in either English or French, with a summary in the other language. Bimonthly. Subscription (6 numbers) ............................................... . Price per copy ............................................................... . WEEKLY EPIDEMIOLOGICAL RECORD (Bilingual: English and French) Sw. fr. 176.- Sw. fr. 31.- Prepared for the guidance of health administrations and health authorities, the Weekly epidemiolllgical record contains notifications made under the International Health, Regulations and information concerning their application. The Record also contains epidemiological information on communicable diseases or international importance. Annual suhscription ......................................................... . Sw. fr. 209.- WORLD HEALTH STATISTICS QUARTERLY The World health statistics quarterly replaces (since 1978) the World health statistics report (published since 1967) and its forerunner theEpidemiolllgiwl and vital statistics report (published since 194 7). It deals with the detailed analysis of selected health topics of current interest. The Quarterly contains articles in either French or English with a summary in both languages. Annual subscription ......................................................... . Price per copy ............................................................... . WORLD HEALTH STATISTICS ANNUAL (Bilingual: English and French) Sw. fr. llO.- Sw. fr. 31.- The forerunner of this series was the Annual epidemiolllgiwl report of the League of Nations. It was followed by the Annual epidemiolllgiml and vital statistics issued by the World Health Organization. Latest publications: 1990. Vital statistics and life tables, international statistics on causes of disability, causes of death (single volume), 447 pages (out of stock) .................................................................... . 199 l. Vital statistics and life tables, health and human development, causes of death (single volume), 371 pages ..................................................... . 1992. Vital statistics and life tables, implementation of the Global Strategy for Heath for All by the Year 2000 (single volume), 480 pages ....................... . 1993. Vital statistics and life tables, health and demographic data, and alternative sources for the collection of cause-of-death and vital events data in the absence of universal vital registration (single volume), 662 pages ................................. . 1994. Vital statistics and life tables, and worldwide health and demographic data (single volume), 460 pages ........................................................... . Sw. fr. 90.- Sw. fr. 100.- Sw. fr. 100.- Sw. fr. 100.- Sw. fr. 100.- Publications de /'Organisation mondiale de la Sante 1996 FORUM MONDIAL DE LA SANTE Revue internationale de developpement sanitaire (Editions separees en fran(ais, anglais, espagnol, arabe, chinois et russe) Forum mondialdelasanti est une revue trimestrielle destince aux rcsponsables des politiques sanitaires, aux planificateurs, administrateurs et educateurs sanitaires, enfin aux travailleurs de la sante publique de toutes categories. Tribune pour la presentation et la discussion de nouveaux concepts en sante publique et de nouvelles approches des problemes de same. Forum se consacre it !'amelioration de la sante par la promotion de services de sante couvrant la population tout entiere et d 'une vaste gamme de mesures de sante publique, qu'ellessoientou non soutenues par !'OMS. Il est le principal organe it la disposition des Etats Membres de !'OMS pour l'echange international d'informationssanitaires en meme tempsqu'un instrument de cooperation technique entre pays en developpement. Abonnement ( 4 numeros) .............................................. . Lenumf:ro .................................................................... . Fr. s. 75.- Fr. s. 20.- BULLETIN DE L'ORGANISATION MONDIALE DE LA SANTE Le Bulktinest le principal organe scientifique de l'OMS; ii a pour role de passer en revue les progres des sciences medicales etapparentees (articles de la rubrique «Le point») et de mettre en lumiere les connaissances nouvelles en presentant dans des articles originaux les resultats de recherches scientifiques au laboratoire et sur le terrain. lc Bulktin contient des articles originaux en fran<;ais ou en anglais, accompagnes d'un resume dans l'autre langue. Bimensuel. Abonnement (6 numeros) .............................................. . Lenumero .................................................................... . RELEVE EPIDEMIOLOGIQUE HEBDOMADAIRE (Bilingue: fran<;ais et anglais) Fr. s. 176.- Sw. fr. 31.- Destine aux administrations sanitaires et aux services de sante, le Rekvi epidfmliolllgique hehdomadaire contient les notifications exigees par le Reglement sanitaire international, ainsi que d'autres renseignements concernant !'application de ce reglement. Le Rekvicontient egalement des informations epidemiologiques concernant les maladies transmissibles d'une importance intemationale. Prix de l'abonnement annuel ......................................... . Fr. s. 209.- RAPPORT TRIMESTRIEL DE STATISTIQUES SANITAIRES MONDIALES Le Rapporttrime,trieldestatistiques mondiales, remplace ( depuis 1978) le Rapport de statistiques sanitaires mondiales (publie depuis 1967) et son precurseur le Rapport epidimiologique et dimographique (publie depuis 1947). Il presente des analyses detaillees sur des sujets specifiques d'intcret courant. Le Trimestrul presente des articles originaux en fran<;ais ou en anglais, accompagnes d'un resume dans les deux langues. Prix de l'abonnement annuel ......................................... . Lenumf:ro .................................................................... . Fr. s. no.- Fr. s. 31.- ANNUAIRE DE STATISTIQUES SANITAIRES MONDIALES (Bilingue: fran<;ais et anglais) Cet annuaire rcmplace les Statistiques epidimiologiques et dbnographiques annuellts publiees par I 'Organisation mondiale de la Sante et qui avaient, elles-memes, remplace le Rapport epidbniolllgique annuelpublie par la Societe des Nations. Publications recentes: 1990. Mouvement de la population et tables de survie, statistiques intemationales sur les causes d'incapacite, causes de deces (l seul volume), 44 7 pages ( epuise) ............................................... . 1991. Mouvementde la population et tables de survie, la sante et le developpement humain, causes de deces (1 seul volume), 371 pages ...................... . 1992. Mouvement de la population et tables de survie, mise en a:uvre de la Strategie mondiale de la sante pour tous d'ici l'an 2000 (1 seul volume), 480 pages .............................................................. . 1993. Mouvement de la population et tables de survie, donnees sanitaires et demographiques, methodes permettant de recueillir des donnees sur les causes de deces et les faits de l' etat civil en !'absence de systeme universe( d'enregistremenl (l seul volume), 662 pages ................................. . 1994. Mouvement de la population et tables de survie, ainsi que donnees sanitaires et demographiques it l'echelle mondiale (1 seul volume), 460 pages Fr. s. 90.- Fr. s. 100.- Fr. s. 100.- Fr. s. 100.- Fr. s. 100.- Health security for women The articles appearing in this issue of World health statistics quarterly seek to demonstrate the need for women's health to be considered in a holistic rather than a fragmented manner. In particular, it is essential to take a lifespan perspec- tive, since health conditions in one phase of a woman's life not only affect subsequent stages of her own life but also have an impact on future generations. Underlying all the contributions is a concern for women's "health security", a concept that encompasses all aspects of the basic human right to the highest attainable standard of health. Some of the articles address reproductive health issues, others consider the greater vulnerability of women to infection (as well as the different impact of some diseases on women), while a number of articles examine the special health needs of specific groups of women. This issue seeks to provide health professionals and other interested persons throughout the world with new and critical data and information that may be useful in continuing endeavours to improve the health of women across age-groups and across borders. Many of the articles echo issues addressed in the historic Beijing Platform for Action, adopted at the conclusion of the September 1995 Fourth World Conference on Women. Securite sanitaire pour les femmes Les articles dans ce numero du Rapport trimestriel de statistiques sanitaires mondiales cherchent a montrer qu'il taut envisager la sante des femmes de maniere globalisee plut6t que fragmentee. II est notamment essentiel d'adopter une perspective portant sur la vie entiere, car l'etat de sante d'une femme au cours d'une des phases de sa vie n'affecte pas seulement les stades ulterieurs de sa propre vie, mais a egalement un impact sur les generations futures. A la base de toutes les contributions se trouve le souci de la securite sanitaire des femmes, notion qui englobe tous les aspects des droits fondamentaux jusqu'au niveau de sante le plus eleve possible. Certains articles traitent de la sante reproductive, d'autres examinent la vulnera- bilite plus grande des femmes vis-a-vis de !'infection (ainsi que l'impact different de certaines maladies sur les femmes), tandis qu'une serie d'autres articles etudient les besoins de sante specifiques de certains groupes de femmes. Le present numero cherche a fournir aux professionnels de la sante et a d'autres personnes interessees, au fil des travaux, des donnees essentielles et nouvelles et des informations pouvant etre utilisees a !'occasion des tentatives qui sont faites en permanence pour ameliorer la sante des femmes a travers les ages et au dela des frontieres. Un grand nombre de ces articles se font l'echo des questions qui ont ete abordees dans la Plate-forme d'action de Beijing, document historique adopte a l'issue de la Quatrieme Conference mondiale sur les femmes de septembre 1995.
Всемирная организация здравоохранения (ВОЗ / WHO) · Journal articles
Health security for women = Sécurité sanitaire pour les femmes [full issue]
Открыть оригинал документа
Полный текст размещён на сайте публикующей организации. lawenc.com индексирует метаданные и ведёт на официальный источник.
Вернуться к постатейному просмотруПолный текст