Health and the family = La famille et la santé [full issue]

Vol. 46, No. 4, 1993 World Health STATISTICS Quarterly Rapport trimestriel de STATISTIQUES sanitaires mondiales Health and the family La famille et la sante World Health Organization Organisation mondiale de la Sante Geneve The World health statistics quarterly replaces (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 Val 41 (1988), the Ouarterlycontains articles in either French or English with a summary in both languages. Annual subscription Sw. fr. 100.- Price per copy Sw. fr. 28.- 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 Organ12at1on welcomes requests for permission to reproduce or translate its publ1cat1ons. 1n part or in 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 editions. and reprints and translations already available. © World Health Organization 1994 Publ1cat1ons of the World Health Organizat1on enioy copyright protection 1n accor- dance with the provisions ol Protocol 2 of the Universal Copyright Convention All rights reserved The des1gnat1ons employed and the presentation of the material 1n this publ1cat1on do not imply the expression of any opinion whatsoever on the part of the Secretariat of the World Health Organization concerning the legal status of any country. territory, city or area or of its autont1es, or concerning the delimitation of its frontiers or boundaries. The mention of spec1f1c companies or of certain manufacturers' products does not imply that they are endorsed or recommended by the World Health Organization in preference to others of a similar nature that are not mentioned. Errors and omissions excepted, the names of proprietary products are d1st1ngu1shed by initial capital letters. Le Rapport trimestriel de statistiques sanitaires mondiales remplace (depuis 1978) le Rapport de statistiques sanitaires mondiales (publ1e depu1s 1967) et son precurseur le Rapportepidemiologique etdemogra- phique (publie depuis 1947). II presente des analyses detail lees sur des sujets specif1ques d'interet courant. A compter du Val 41 ( 1988). le Trimestriel presente des articles orig1naux en frarn;:ais ou en anglais, accompagnes d'un resume dans les deux langues. Prix de l'abonnement annuel Fr. s. 100.- Le numero Fr. s. 28.- La reproduction d'extraits du Trimestriel est autorisee, sous reserve d'indication de la source. Explication des signes Donnee preliminaire, approximative au estimative Donnee non d1sponible Zero au quantile negligeable Categorie non applicable T Total M Mascul1n F Feminin N Nombres absolus L'Organisat1ori mondiale de la Sante est toujours heureuse de recevoir des deman- des d'autorisation de reprodu1re ou de traduire ses publications, en part1e ou 1ntegralement. Les demandes a cet effet et les demandes de rense1gnements do1vent E!tre adressees au Bureau des Publications, Organisation mondtale de la Sante, Geneve, Suisse, qui se fera un pla1s1r de fournir les rense1gnements Jes plus recents sur les changements apportes au texte, les nouvelles editions prevues et les reimpress1ons et traduct1ons deJli disponibles © Organisation mondiale de la Sante 1994 Les publications de !'Organisation mondiale de la Sante benef1cient de la protection prevue par les dispositions du Protocole No. 2 de la Convention universelle pour la Protection du Dro1t d'Auteur. Taus droits reserves Les appellations employees dans cette publication et la presentation des donnees qu1 y figurent n'1mpl1quent de la part du Secretariat de !'Organisation mondiale de la Sante aucune prise de position quant au statut 1urid1que des pays, terntoires, v1lles ou zones, ou de leurs autorites. 11 quant au trace de leurs frontieres ou lim1tes. La mention de firmes et de produ1ts commerciaux n'impl1que pas que ces firmes et produits commerciaux sont agrees ou recommandes par !'Organisation mondiale de la Santa de preference a d'autres. Sauf erreur ou omission. une ma1uscule initiale 1nd1que qu'il s'agit d'un nom depose IX ISSN 0043-8510 Printed in Switzerland 93/9881 -Atar SA, Geneva - 5100 World Health Statistics Quarterly Rapport trimestriel de statistiques sanitaires mondiales Vol. 46, N°4, 1993 Health and the family La famille et la sante Contents Somma ire Introduction 212 Introduction [anglais seulement] 212 Child-care programmes for health and family Programmes de garde d'enfants en faveur de support la sante et de la famille [resume] Judith Evans & P.M. Shah 214 Judith Evans & P.M. Shah 220 Demographic and health surveys (OHS): Enquetes demographiques et de sante : contributions and limitations source d'information sur la sante des families J. Ties Boerma & A. Elisabeth Sommerfelt 222 dans les pays en developpement [resume] J. Ties Boerma & A. Elisabeth Sommerfelt 226 Natural and man-made disasters: the Secours et reconstruction en cas de vulnerability of women-headed households and catastrophe: problemes relatifs aux femmes children without families et aux enfants [resume] Debarati G. Sapir 227 Debarati G. Sapir 233 Psychosocial and mental health aspects of Determinants psychosociaux de la sante women's health mentale des femmes [resume] Lorraine Dennerstein 234 Lorraine Dennerstein 236 Family health and the use of psychoactive Sante de la famille et usage de substances substances psychoactives [resume] Lee-Nah Hsu 237 Lee-Nah Hsu 241 Surveillance for equity in maternal care in Egalite dans les soins de sante : le cas Zimbabwe du Zimbabwe [resume] Carl Taylor et al 242 Carl Taylor et al 247 Vol. 46, 1993 Vol. 46, 1993 Annual table of contents 248 Table annuelle des matieres 248 Wld hlth statist quart., 46 (1993) 211 Health and the family: Introduction The United Nations General Assembly has pro- claimed 1994 as the International Year of the Fam- ily, with the aim of creating among governments, policy-makers and the public a greater awareness of the family as the fundamental unit of society, and of promoting activities in support of families. As part of its contribution to this event, and in recognition of the central role of families in health promotion, protection and care, WHO has dedi- cated this issue of the Statistics quarterly to the family. The family, while still the fundamental unit of all societies, has undergone dramatic changes. The traditional concept of the nuclear family, consist- ing of two parents and their children, is not always relevant in today's society. The concept has ex- panded to include single-parent families, extended families, and numerous other variations. Whatever its form or composition, however, the family re- tains its primary protective, reproductive, produc- tive and educational functions. Indeed it is through their families that children first experi- ence the multiple aspects of their own culture and environment. Given its crucial functions, the fami- ly must be a major focus of all health strategies. Family health has been radically affected by recent socioeconomic, environmental and popula- tion changes. Some of these changes have been positive, to wit, modem medicines combined with public health interventions such as sanitation, clean water and immunizations have reduced the toll of infectious diseases. On the other hand, changes such as industrialization, urbanization, en- vironmental degradation, migration and war have placed great strain on the family's ability to pro- tect its members. Poverty, which affects more than half of the world's population, is the most dam- aging, since it marginalizes even more those who are most vulnerable - the mother and the child. The acceptance of equal rights for women, at least in principle in many countries of the world, has irrevocably changed the status of women. This has had the effect of loosening traditional family ties and enabling more women to work outside the home. While these changes may be beneficial, many women in both developing and developed countries are subjected to the "double burden" of work - at home and outside. Not only is their physi- cal health jeopardized, but also their mental health. Stress and stress-related illnesses are no longer the sole domain of men. For women, partic- 212 ularly for the growing number who head house- holds, mental health care and child care have be- come critical needs. This edition of World health statistics quarterly examines some of these key issues for women. In their article on child-care programmes for health and family support, Judith Evans and P.M. Shah assess the strategy of integrating health care into the rapidly expanding systems of child care. While there are some constraints to this strategy, the au- thors believe that its advantages largely outweigh these constraints. Lorraine Dennerstein studies the psychosocial and mental health aspects of women's health and concludes that women are disproportionately affected by mental health prob- lems. These problems must be considered within the context of social, political and economic issues. In this regard, more attention should be given to the rights of women and steps taken to acknowl- edge the conflicting demands which are often placed on them as providers, carers and wage earn- ers. At the same time, the responsibilities of men as partners, fathers and family members need to be underlined. Health services can assist individuals, families and communities to assume a shared re- sponsibility for health and well-being. One major challenge for public health plan- ners is to identify and implement health interven- tions that will allow families to thrive, and to obtain help when they need it. All too often, families are asked to assume all the consequences of unequal development and to shoulder all the burdens of social and economic change. Families can adapt to changing circumstances, but to do so effectively, they need support, particularly when it comes to health. Families also need support when dealing with the implications of changing socioeconomic cir- cumstances. The erosion of traditional values has resulted in changing mores, especially in the area of sexual relations. The use of harmful substances such as alcohol, tobacco and psychoactive drugs has become more prevalent, especially among young people. In their article on family health and the use of psychoactive substances, Lee-Nah Hsu and Oussama Tawil illustrate the potential impact of drug abuse on the family, while affirming the role of the family in the prevention, treatment and rehabilitation of drug abusers. They recommend strengthening the social network and families' Rapp. trimest. statist. sanit. mond., 46 (1993) coping mechanisms to reduce the demand for and the consequences of substance abuse by family members. It is clear that, with the increasing forms and intensity of socioeconomic change, women and children have become more vulnerable. This is particularly true in the case of natural and man- made disasters. Debarati G. Sapir examines the situation of women and children in the current world emergency context. She identifies four broad policy areas that affect women and chil- dren in disaster situations and illustrates these with examples and field evidence. The author hyppothesizes that the ability of any country to respond effectively to disasters depends on the strength of its health and social infrastructure, and on its overall development status. The article concludes by identifying seven areas in which concrete measures could be taken to improve the current situation. In any context, effective family health strategies cannot be developed without adequate informa- tion and data. Two articles in this edition examine information and data collection from different perspectives. Demographic and Health Surveys (DHS) illustrates how these surveys can be an important Wld hlth statist quart., 46 (1993) source of data on family health in developing countries. The other article, a case study of Zimba- bwe, proposes a system of periodic surveys to iden- tify groups among whom maternal care problems are concentrated. The system of surveillance for equity in maternal care would establish a process of finding appropriate local adaptations of solutions that are cost-effective and sustainable. In taking up the challenge of promoting and ensuring health in the face of change, family health is a key element. The stability of the family, regard- less of its form or composition, needs to be main- tained through all social policies, especially those relating to health. It is clear that there are contra- dictions within all family structures. The family can be a shelter, a system of mutual solidarity and sup- port; it can also be restrictive, hindering individual and social development, even providing the setting for child abuse, sexual abuderives from the inabili- ty of the family, in the absence of social support mechanisms, to cope with the stresses of changed circumstances. Thus the great challenge of public health is to seek ways to empower families not just to cope with these changes but to thrive. To meet this challenge, a positive relationship between fam- ilies and the health sector is essential. 213 Child-care programmes for health and family supporta Judith Evansb & P.M. Shahc Introduction Over the past few decades there has been a ongo- ing dialogue about child care. In recent years the debate has been stilled somewhat, not by research suggesting the efficacy or harmful attributes of child care, but by the reality of people's lives. Demographic, social, and economic factors, such as population growth, changes in the composition of the labour force, the shift from agricultural to industrial employment, migration from rural to urban areas, and changes in patterns of family life have all affected the way people live and the types of health and social service supports they require. For many, child care is a necessity, not an option (Tabk 1). The response by families, communities, non- governmental organizations and governmental a This article is excerpted from: Child-care programmes as an entry point for maternal and child health romponents of primary health care (WHO document in preparation). b Co-director, Consultative Group on Early Childhood Care and Development, Haydenvilla, USA. c Responsible Officer, Child Health and Development, Division of Family Health, World Health Organization, Geneva. Table 1 bodies to the need for child care has been to devel- op a wide variety of early-childhood programmes available to children from soon after birth until the time they enter the formal school system. These systems exist throughout the world, serving people in a wide variety of settings. Concurrently with these developments, the health sector has been seeking ways to expand the delivery of maternal and child health services. While there are a number of reasons for the lack of utilization of health-care centres by certain segments of the population, a contributing factor is that preventive health care is not often seen as a priority by traditional and low-education groups. Where maternal and child health services are avail- able they are often seen as sources of therapeutic or remedial rather than preventive care. Moreover, planning by the health and social sectors has typi- cally failed to incorporate many of the perceived needs and expectations of the communities for whom the planning has been undertaken. As a result, the services that are available have neither attracted the attention, nor inspired the confi- dence, that would have ensured their appropriate use. Therefore, the challenge is to find ways of Labour force participation rates of married and single women with children under 18, and child-care rates for children of different ages, selected countries Tableau 1 Taux d'appartenance a la population active des meres (mariees au celibataires) d'enfants de mains de 18 ans, et proportions d'enfants de differents ages pris en charge, dans certains pays Percentage of mothers in the Percentage of children in child labour force - pourcentage de care - pourcentage d'enfants meres dans la population active pris en charge Country- Pays Year- Married - Single 0-3 3-5 An nee Mariees Celibataires years -ans years-ans Canada 1988 67 64 12 31 Denmark- Danemark 1988 873 873 44 85 Germany - Allemagne 1986 48 70 2 76 France 1988 66 85 24 95 Italy - ltalie 1986 44 45 5 90 Sweden - Suede 1988 893 893 73 United Kingdom - Royaume-Uni 1988 59 52 2 43 United States of America - Etats-Unis d'Amerique 1990 663 663 20 70 •According to data from official sources, there are no significant differences in the labour-force participation rates of women by marital status in Denmark, Sweden and the United States of America. - D'apres des donnees de sources officielles, le taux de participation des femmes a la population active ne differe pas sensiblement en fonction de leur etat matrimonial au Danemark, en Suede et aux Etats-Unis d'Amerique. Source: Ref. - Ref. (8). 214 Rapp. trimest. statist. sanit. mond., 46 (1993) delivering primary health-care services, particular- ly to low-income families in both urban and rural settings, that will bring those services to popula- tions not currently being served. One possible strategy is to integrate health care into the rapidly expanding systems of child care that are develop- ing out of family and community needs. There are a range of terms that refer to group care of young children (from birth until the age at which the child enters the formal school system, generally between the ages of 5 and 7) outside the home. For example, the word "creche" generally refers to services for children under the age of 2. "Child-minding" or "family day care" is care pro- vided in a home, usually in the neighbourhood. "Daycare" generally describes full-time care pro- vided for children, most often in a centre. In addi- tion, there are preschool programmes designed to prepare young children for entry into the formal primary school system. To encompass all these situ- ations, and for ease of usage, the terms "child care" and "early-childhood programmes" will be used to describe group care for young children, regardless of the age group or the setting. The rapid pace of change is having an impact on families worldwide. Dramatic changes resulting from technological advances in the industrial and agricultural fields are leading to the breakdown of traditional social institutions. In both urban and rural environments there is an increase in the number of woman-headed households. This neces- sarily impacts on women's work load. Women have always played multiple roles that compete for their time and physical and emotional energy. Re- gardless of the context within which children are raised, care of children, particularly young chil- dren, is still the woman's responsibility. In addi- tion, the woman is responsible for household man- agement and operations, and economic/produc- tive activity, some of which takes her outside the home. Women's work outside the home may pro- duce direct benefits for children, particularly if the woman has control of the money she earns. In these instances, children are likely to have im- proved health and participate more fully in education.d However, there are also costs. De- mands on time and energy make it increasingly difficult for women to provide the kind of care their children need. The rapid growth and early acquisition of mo- tor and cognitive skills typical of infancy and early childhood make children vulnerable to a variety of health, nutritional, psychosocial and other envi- ronmental conditions. Those living in poverty are particularly at risk of poor physical and psychoso- d Engle, P. The intersecting needs of working women and their young chUdren: a repm to the ford Foundation, 1980. Wld hHh statist. quart., 46 (1993) cial development. Thus it can be argued that a critical time to assure survival and sustained devel- opment is during the early years. The various developmental transitions that the young child goes through in the first years require constant adjustment and present potential threats to health and survival. These transitions are com- pounded by stresses of poor socioeconomic, sani- tary and environmental conditions that are them- selves changing rapidly. The younger the child, the greater the susceptibility to morbidity and mortali- ty and the more difficult it is to disassociate the physiological and psychological factors within the environment that govern health. The role of care in the child's development The conditions within the physical environment which are important in supporting children's growth and development are combined with the conditions within the family context. An important dimension of the family environment is care. The quality of care the young child receives determines his/her development. During the postpartum and early infancy stages the mother is generally the primary caregiver, sometimes with considerable support from others and sometime alone. She is responsible for the safety, care and feeding of the child, and, to varying degrees, for providing stimu- lation for the child's cognitive and psychosocial development. As the child grows older, others in the family and community play an increasingly im- portant role in the care of the child. Care includes much more than keeping the child safe and free from harm. Caregiving behav- iours include: breastfeeding; providing emotional security and reducing the child's stress; provid- ing shelter, clothing, feeding, bathing, and super- vision of the child's toilet; preventing and at- tending to illness; nurturing and showing affec- tion, interaction and stimulation; playing and so- cializing; protecting from exposure to pathogens; and providing a relatively safe environment for exploration (l).e A second set of caregiving behav- iours includes the use of resources outside the family, among them curative and preventive health clinics, prenatal care, the use of traditional healers, or members of the extended family network. f All of these behaviours are a part of supporting the devel- opment of young children. The type of caregiving behaviour which is re- quired depends to a large degree on the child's developmental age and the health and nutritional risks the child is facing. For example, during the e Zeitlin, M. Chil,d ro.re and nutrition: the findings from positive deviance research. Final rep<m to UNICEF, the Italian government and Tufts University (Positive deviance in nutrition research project, 1987-1992). 1993. f Engle, P. Care and chil,d nutrition: theme paper for the International Nutrition Conference (ICN), March 1992. 215 first year of life the child is at the greatest risk of mortality. During late infancy ( or when comple- mentary foods are introduced) and during the tod- dler period, the child is at greatest risk of growth faltering. While faltering growth may be the result of inadequate nutrition, there is clear evidence to suggest that the feeding of young children is more that providing them with food. Feeding includes the interactive process which accompanies the in- take of food. This process is part of caregiving, and the way that care is provided during the feeding process can have a profound effect on the child's later development. Studies comparing children who are well- and poorly-nourished suggest that there are positive caring practices associated with children with better nutritional status, even though the children come from the same type of environment as their malnourished peer~.e In the context of feeding, caring practices in- clude frequent physical contact, being consistently responsive to the child's needs and showing af- fection to the child. Care also includes active feeding which means being aware of how much the child is eating, offering the child a second helping, assist- ing the child in the use of a utensil instead of expecting complete self-feeding, and offering praise for eating.£ Another example comes from longitudinal studies of children growing up in poverty. In Ha- waii, Werner & Smith (2) were able to identify children who were able to thrive even though the conditions under which they were raised would suggest they would be malnourished and not devel- op well. They called those children that did well "resilient". Those characterized as resilient showed greater autonomy and competence and clearly did better developmentally, over time, than their non-resilient peers. They did better because they were able to "elicit predominantly positive responses from their environment." These chil- dren were found to be "stress-resistant" compared to children who elicited negative responses from their environment. Werner & Smith sought to identify what makes some children more resilient. They concluded, as have others, that the differences between children who thrive and those who falter are determined by the type of interaction that occurs between the child and the environment. Important elements of this interaction that help determine the child's ability to thrive include the amount of affective stimulation and the kind of care the child receives. It is important to note that the linkage between the child and the primary caregiver, usually the moth- er, is not the only important variable. Even when this linkage is positive, the level of support the caregiver receives from others in the family and from society plays an important part in the kind of care that person is able to provide. 216 It is also important to note that what the child brings to the situation is as important as what the environment provides. Early on, a cycle can be established that leads to and reinforces poor devel- opmental outcomes. The cycle can begin with an unresponsive infant who does not engage with adults. As a result, caregivers find the child unap- pealing and fail to provide the stimulation that is needed to promote development. The results seen in young children are apathy, reduced responsive- ness to people and the environment, and the in- ability to attend to tasks. The cycle gets reinforced. The child is even less appealing and adults do not choose to interact with the child. The child be- comes more invisible and, without significant inter- action with adults, does not achieve optimal devel- opment. In order to create a programme for young chil- dren, it is critical to have an indication of develop- mental needs across different ages. Such a scheme is provided by Donohue-Colletta (3) as follows: Infants (birth to 1 year) need: • protection from physical danger • adequate nutrition and health care • an adult with whom to form an attachment • an adult who can understand and respond to their signals • things to look at, touch, hear, smell, and taste • opportunities to explore the world • appropriate language stimulation. Toddlers (1-3 years) need these things as well as: • support in acquiring new motor, language and thinking skills • a chance to develop some independence • help in learning how to control their own be- haviour • opportunities to begin to learn to care for themselves • daily opportunities to play with a variety of ob- jects. Children (3-6 years), in addition to the above, need: • opportunities to develop fine motor skills • encouragement of language through talking, reading, singing • activities which will develop a positive sense of mastery • opportunities to learn cooperation, helping, sharing • experimentation with pre-writing and pre-read- ing skills. In any child-care setting, these elements should be present to support children's overall develop- ment. As the health sector begins to examine ways in which appropriate linkages can be made be- tween child-care settings and health services, a sim- ilar differentiation can be made between the Rapp. trimest. statist. sanit. mond., 46 (1993) health needs of children depending on their devel- opmental ages. As a reference point, Tabl,e 2 presents a visual summary of the types of promotional, preventive and curative health actions that can be successfully delivered via child-care programmes, at each stage of child development. As can be seen, a wide vari- ety of critical health activities can be undertaken within child-care settings. In addition to providing the minimum package sketched in Tabl,e 1, the more formal child-care programmes are able to do more feeding and to carry out growth monitoring. Depending on the number of children being served, it could also be possible to provide direct health care through the presence (full- or part-time) of a nurse. It must be noted, however, that it is only with adequately trained staff that these health services can be developed and delivered. Thus another step in deciding which MCH (maternal and child health) components can be delivered through child-care programmes is to differentiate between those that can be delivered by child-care person- nel, with appropriate back-up and referral mecha- nisms, and those that require the skills and knowl- edge of health-care personnel. While caregivers can be trained to provide a range of health-care activities, particularly those associated with promo- tion and prevention, a professional from the health sector needs to be available to deliver cer- tain services, such as immunizations. In sum, there are a variety of MCH components that could logically be delivered through early- childhood programmes. The question is, how feasi- ble is it to deliver these components through child- care systems? In the next section an argument is made that while there are pros and cons of using early-childhood programmes for the delivery of health care, clearly the scale is tilted toward the positive aspects of such a collaboration. Table 2 There are some compelling reasons for using child-care programmes for the delivery of health- care components. But, in order to determine the feasibility of doing so, it is important to define the supports for integrating child care and health sys- tems and the constraints to such a process. Positive/supportive factors Early-childhood care and development pro- grammes exist in some form in all parts of the world ( Tabl,e 3). They have been begun under the trees in rural Africa, and organized in the streets of Chile. They have also been developed through na- tional initiatives devoted to promoting the healthy growth and development of all children, as in In- dia, Mauritius and parts of South east Asia. While in many instances an infrastructure exists to support these programmes, they are frequently under-uti- lized in terms of the range of services that could be provided. In general, those programmes not spon- sored by the health sector have not taken up the health-care and health-monitoring needs of chil- dren and mothers. While early-childhood programmes generally reach only a small percentage of the eligible popu- lation, they have the potential to reach many more. In many developing countries approximately 50% of the population is under the age of 15. More than a third of these are under the age of 5 (4). If young children's families are taken into account, it is possible for child-care programmes to touch the lives of more than half the population ( Tabl,e 3). Early-childhood programmes are expanding rapid- ly due to increasing demand. The demand has come from urban and rural communities, from those wanting better health care, from those con- cerned with supporting the child's cognitive de- velopment, and from those seeking a safe place for their child. The demand is illustrated by the in- Possible integration of health and development activities by child-care setting Tableau 2 Integration possible d'activites de sante et de developpement, par milieu dans lequel l'enfant est pris en charge Hygiene - Hygiene Creative play - Jeu createur Feeding - Alimentation Nutrition supplement - Complement nutritionnel Affective stimulation - Stimulation affective Monitor psychosocial development - Observation du developpement psychosocial Monitor growth - Observation de la croissance Immunization - Vaccination Parental discussion - Discussion avec les parents Wld hlth statist. quart., 46 (1993) Parents Child x x x x x x x x minding Prise en charge x x x ? x x Community Workplace Collectivite Lieu de travail x x x x x x x x x x x x x x ? x x Institution Institution x x x x x x x 217 Table 3 Early-childhood care and education (ECCE): total facilities and enrollment Tableau 3 Prise en charge et education dans la petite enfance: nombre total d'installations et d'enfants inscrits Total facilities- Country - Pays Nb. total d'installations Austria- Autriche 5 075 Denmark - Danemark 24 760 Poland - Pologne 26 289 Czech Rep. - Rep. tcheque 13 003 Spain - Espagne 16 884 India- lnde 194 718 Indonesia - lndonesie 32 030 Thailand - Tha"ilande 27 804 China - Chine 299 061 Malaysia - Malaisie 5188 Rep. of Korea - Rep. de Coree 10 501 Benin - Benin 588 Ghana 4 048 Mauritius - lie Maurice 1 391 Zimbabwe 5 220 Bolivia - Bolivie 2 148 Chile-Chili 8 563 Costa Rica 713 Jamaica - Jama"ique 1 983 Mexico - Mexique 47 386 Peru-Perou 7 982 Egypt - Egypte 3 871 Iran (lslamicRep. of- Rep. islamique d') 2162 Jordan - Jordanie 755 United Arab Emirates - Emirats arabes unis 94 crease in the number of child-care centres avail- able over the years. For example, in 1969 in Tanza- nia there were only 280 centres. By 1980 there were 3 500. The increase is much more dramatic in Kenya where in 1968, early-childhood programmes served 175 OOO children; by 1980 that number was 300 OOO, and it had increased to 800 OOO children in 1992 (1). Even so, the programmes which exist serve only a small percentage of the world's eligi- ble population. But the situation is changing rapid- ly: increasing demand is leading to the develop- ment of a wide variety of child-care programmes. Early-childhood care and development pro- grammes have a social attractiveness and accept- ability that invites participation from families that might not avail themselves of primary health-care services. For people in many parts of the world, early-childhood programmes are a priority. For ex- ample, an informal education programme for pre- school children in Puno (Peru) has been used as the base for the implementation of a number of health programmes. It provides a setting for a regional immunization programme, touching the lives of families in the community as well as the 218 Total % of population enrollment- 0-6 in ECCE- Nb. total % de la population d'enfants inscrits O a 6 ans pris en charge 180 337 253 724 65.6 673 669 14.8 642173 40.1 1127 348 32 1161 500 1540504 5.1 1 267112 14.9 18 545 300 293 369 9.5 607 296 11.5 14 497 1.2 283 323 8.1 36 200 2 .4 205 600 9.5 86114 5.3 465 323 23.6 43 375 8.2 141 180 33.1 2 960 278 18.8 852 960 19.8 152 371 1.4 146 409 1.2 47 654 4.6 160 159 6.7 children served in the programme. Given a choice, a community may well begin by creating an early- childhood programme. An example comes from Kuala Lumpur (Malaysia) where the health sector was interested in establishing health-care services, but the community was interested in early-child- hood care and income-generating activities. Once these were established, the community was open to the development of health-care services. Thus, sup- port for and the utilization of early-childhood care and development programmes in many settings may be more regular than is often the case with health-care and family-planning services. Primary health care is a logical extension of the current services being provided within early-child- hood care and development programmes. Health care is a part of the child's total development, and as such should be seen as a part of and provid- ed within the context of children's services. Health services are currently being provided as part of a variety of child-care programmes. Child care offers parents and the community the possibility of participating in the health care of their own children through the monitoring of Rapp. trimest. statist. sanit. mond., 46 (1993) physical growth and development, the early detec- tion of physical and psychosocial handicaps, nutri- tion surveillance, immunization against major dis- eases, prevention and treatment of common dis- eases and health education. In addition, child-care programmes can provide a channel of communica- tion between families and health services, which in tum can assist in making parents more aware of better hygienic and nutritional practices. Many of the health and nutritional problems associated with conditions of poverty, if identified early, can be managed using appropriate and simple technol- ogies. Technology in this instance is meant to in- clude "not only scientifically sound equipment and information but also certain organizational elements" (5). These technologies are currently available and can be managed by persons with relatively low levels of training, and delivered through other than health-care centres. There is every reason to believe that the health and nutri- tion technologies are suited to most early-child- hood care and development programmes which are in place, or could be developed with a mini- mum of cost and back-up support (6). The many positive factors listed would appear to provide a solid rationale for using child-care settings to deliver MCH components. However, there are some constraints that are likely to be encountered. The multiplicity of mandates which have lead to the development of early-childhood programmes make it difficult to identify an entry point for pri- mary health care. Early-childhood care and devel- opment, because it is closely intertwined with child rearing, is a phenomenon in which almost every- one feels he or she should have a say, from those in health, education, public welfare, social services, community development, and labour. What this means is that there is no one sector that can be worked with to better integrate health care into child-care service delivery. By looking at Thailand, for example, it is possible to illustrate the issue. It is estimated that about 15% of the 0-6 age group are being served through some type of early-childhood effort ( Tabk 3) _g Yet, even with so few children being served, the following programmes have been developed: child nutrition centres are run by the Department of Health; child development centres are operated by the Department of Community Development; kindergarten or "head start" cen- tres are run by the Department of Education; day- care centres are operated by metropolitan city mu- nicipalities; and private enterprises provide servic- es to yet another group of children. To make things even more difficult, those oper- ating early-childhood programmes may not be g UNESCO. Regional training workshop for supervisors and administrators of early childhood care and education. Bangkok, Thailand, 2-7 May, 1983. Wld hlth statist. quart., 46 (1993) aware of other child-care programmes. Private-sec- tor providers frequently do not coordinate their efforts with public-sector programmes offered by health, education, or social welfare ministries. And, since the public and private sectors may have different goals for offering services to young chil- dren, it is not uncommon for parallel programmes to be developed. In some instances this leads to more spaces being available within child-care pro- grammes than there are children in the communi- ty, while in other communities there are no ser- vices. In the developed countries the primary empha- sis has been on developing child-care programmes that include educational and/ or stimulation com- ponents, with little thought being given to the es- sential health elements. Models of these child-care systems, particularly in the form of preschool pro- grammes, have been transposed to developing countries. The pressing health needs and prob- lems of infants and young children remain either unattended or are aggravated in such centres. Where access to the health-care system is inade- quate there is now a clear need to recognize the full dimensions of child-care systems in terms of their potential to integrate basic health care and developmental stimulation, complementing exist- ing primary health-care structures and activities. The multiple needs of children require a multi- sectoral approach to the integration of the MCH components and early-childhood programming. Such an approach is consuming of both time and energy and can only be developed with the cooper- ation of all sectors involved. A successful approach to multi-sectoral programming is to begin by pro- viding a core service. Once that becomes well estab- lished, it is then possible to add services. An exam- ple comes from Indonesia where a parent-educa- tion programme, which includes weekly meetings, a toy lending library and extensive follow-up with families, was added to an already successful family- planning and nutrition programme.h The infra- structure developed in the existing programme allowed the parent education programme to be added successfully. A primary constraint on using child-care pro- grammes to implement primary health-care com- ponents is that many are developed on a pilot basis, but they neither continue beyond the project-de- velopment phase, nor are they disseminated to other communities. Even though a project shows promise, it may not have staying power, and thus it may not be appropriate to expend limited resourc- es to integrate MCH components into the project. Reasons for the failure of projects fall within four categories: resource constraints; lack of political h Stein, H.D. Review of programme on the rol.e of women in r.omprehensive chil.d development. UNICEF, 1985. 219 commitment at various levels; weak demand for services; and organizational, implementation and management problems (7).i There are also issues related to how to take a successful pilot project and disseminate it to a wider audience. Many organiza- tional characteristics of community-based projects necessarily change if the project is to be imple- mented nationally (7). Thus, before investing in projects it is necessary to question whether or not they could successfully undergo these organiza- tional changes to be implemented on a large scale. Because of the scarcity of resources, low-cost alternative approaches are being sought for the delivery of health care. Early-childhood care and development programmes are low-cost services. While this might at first appear to be an advantage, it is in fact a constraint. Early-childhood pro- grammes are frequently developed by the commu- nity and rely on trained volunteer or paraprofes- sional staffing for day-to-day programme opera- tion. What this means is that those who work in these programmes generally have low levels of edu- cation, and once they receive training they are able to find higher-paid employment in related fields. This has three implications. Firstly, pro- grammes are constantly in the position of training new staff; it is difficult to build on experience and develop a cadre that can provide role models, peer support and supervision. Secondly, health-care planners need to be realistic about what health- care components can be introduced into early- childhood programmes. Although health-care technology makes it possible for people with low levels of education to deliver a wide variety of ser- vices, paraprofessional staff may not have the ca- pacity to develop the skills necessary to deliver sophisticated health-care services. Thirdly, training materials and methodologies have to be geared to a very practical level, with ongoing support and supervision built into the training system. On balance, experience indicates that the posi- tive factors supporting the introduction of health- care elements into child-care programmes largely outweigh the constraints. Further, if given due con- sideration in the planning and implementation phases, the constraints can be overcome. Summary Demographic, social and economic factors have affect- ed the types of health and social service support that the people require. For many, child care is a necessity, not an option. Child care refers to group care of children in creche, child minding, family day care, kindergarten and day care in different settings. Maternal and child i Myers, R. Going to scal.e. Paper prepared for the second inter- agency meeting on community-based child development. New York, October 29-31, 1984. 220 health services are often seen as sources of curative rather than preventive care. Planning by the health and social sectors has failed to incorporate many of the perceived needs of the communities. As a result, the services that are available have neither attracted the attention, nor inspired the confidence of the people. One possible strategy is to integrate health care into the rapidly expanding systems of child care that are devel- oping out of family and community needs. These health services can be developed and delivered by adequately trained child-care as well as health-care personnel. With appropriate and simple back-up tech- nologies and skills and referral mechanism, the child- care personnel can undertake some of the MCH care activities in child-care centres. When needed, either a primary health worker or nurse-midwife from health sector should deliver services. Early-childhood care programmes are in increasing demand. Such programmes have potential to reach more than half of the population consisting of the chil- dren under-five years of age and their parents. Child- care and development programmes have a social at- tractiveness and acceptability that invites participation from families that might not avail themselves of primary health-care services. Thus, support for and the utiliza- tion of early-childhood care and development pro- grammes in many settings may be more regular than is often the case with health-care and family-planning services. Primary health care is a logical extension of the current services being provided within early-childhood care and development programmes. Moreover, child care offers parents and the community the possibility of participating in the health care of their own children through the monitoring of physical growth and develop- ment, the early detection of physical and psychosocial handicaps, nutrition surveillance, immunization, pre- vention and treatment of common diseases and health education. Some constraints are likely to be encoun- tered. However, the experience indicates that positive factors supporting the introduction of health-care ele- ments into child-care programmes largely outweigh the constraints. Where access to the health-care system is inadequate there is now a clear need to recognize the full dimension of child-care systems to integrate basic health care and developmental stimulation, comple- menting existing primary health-care structures and activities. Resume Programmes de garde d'enfants en faveur de la sante et de la famille Les conditions demographiques, sociales et economi- ques influent sur la nature des services de sante et des services sociaux dont la communaute a besoin. Pour beaucoup de families, le recours a des services de garde d'enfants est une necessite. Ces services com- prennent les creches, les nourrices, les garderies fami- liales ou publiques, les jardins d'enfants et d'autres formes de garde. On considere souvent que les servi- ces de sante maternelle et infantile ont plus un r61e Rapp. trimest. statist. sanit. mond., 46 (1993) curatif qu'un role preventif. Le secteur sanitaire et social n'a pas tenu compte, dans sa planification, de bon nombre des besoins ressentis par les communautes, si bien que celles-ci ne s'interessent pas aux services disponibles ou s'en mefient. Pour y remedier, on pourrait integrer des soins de sante aux systemes de garde d'enfants qui se developpent rapidement pour repon- dre aux besoins des families et des communautes. Ces services de sante peuvent etre developpes et dispenses par des puericultrices et des agents de sante qualifies. Avec des techniques et des connaissances appropriees et simples et un systeme d'orientation ade- quat, le personnel des garderies peut accomplir certai- nes t~ches relevant des soins de sante maternelle et infantile. En cas de besoin, les services seront fournis par des agents de sante primaires ou des infirmieres- sages-femmes du secteur de la sante. Les services de puericulture sont de plus en plus de- mandes. lls ont la capacite de toucher plus de la moitie de la population composee des enfants de mains de 5 ans et de leurs parents. Ces services sont socialement attractifs et acceptables, ce qui favorise la participation de families qui, sans cela, n'auraient pas acces aux soins de sante primaires. En consequence, dans bien des cas, leur clientele peut etre plus reguliere que celle des services de sante et de planification familiale. Les soins de sante primaires sont un prolongement logique des services dispenses actuellement dans le cadre des programmes de puericulture. Ces programmes don- nent en outre aux parents et a la collectivite la possibilite de participer aux soins de sante donnes aux enfants, notamment par la surveillance de la croissance et du developpement physique, par la detection rapide des handicaps physiques et psychosociaux, par la sur- veillance nutritionnelle et la vaccination, par la preven- Wld hHh sratist quart., 46 (1993) lion et le traitement des maladies courantes et par !'education sanitaire. Quels que soient les obstacles, !'experience montre que les elements positifs militent en faveur de !'integration des soins de sante dans les programmes de gardes d'enfants l'emportent large- ment sur les contraintes. La ou l'acces aux soins de sante est limite, ii est indispensable aujourd'hui de tirer pleinement parti de la possibilite d'integrer aux syste- mes de garde d'enfants des soins de sante de base et des activites de stimulation du developpement, pour completer les structures de soins de sante primaires existantes. References/References 1. Myers, R. The twelve who survive: strengthening programmes of early chil.dhood deuel.opment in the Third World. London, Routledge, 1992. 2. Werner, E.E. &: Smith, R.S. Vulnerable but invincible: A longitudinal study of resilient chil.dren and youth. New York, McGraw Hill Book Company, 1982. p. 158. 3. Donohue-Colletta, N. Cross-cultural chil.d deuelopment: a training course fur program staff. Richmond, VA, Christian Children's Fund, 1992, p. 65. 4. World Bank. Wurl.d deuel.opment repurt 1984. New York, Oxford University Press, 1984. 5. Rohde, J.E. &: Hendrata, L. Development from below: transformation of village-based nutrition projects to a national family nutrition program in Indonesia In: Scrimshaw, N.S. & Wallerstein, M.B.(eds) Nutrition implementation: issues and experience. New York, Plenum Press, 1982, pp. 209-230. 6. Werner, D. "7im there is no doctur. Palo Alto, CA, Hesperian Foundation, 1977. 7. Pyle, D. Life after project: a multi-dimensional analysis of implementing social deuel.opment programs at the community leveL Boston,John Snow, Inc., 1984. 8. Himes,J.R.etal. (eds.) Women, wmiandchil.dcare(lnnocenti global seminar-summary report), Florence, UNICEF, 1993. 221 Demographic and health surveys (OHS): contributions and limitations J. Ties Boermaa & A. Elisabeth Sommerfelfb Introduction A variety of data sources can be employed to gener- ate data relevant to health planning, and the im- plementation, monitoring and evaluation of health programmes in developing countries. The routine health information system, based on health facility data, and population-based surveys are the most important sources of data. The focus is on a selected set of health indicators, such as those proposed in the context of primary health care (1,2) or child survival initiatives (3). In general, surveys can provide information on health indicators for the whole population. Nation- al surveys conducted in the context of the Demo- graphic and Health Surveys (DHS) programme constitute an important source of information on family health indicators. In this article, we first provide some back- ground information on the DHS programme, in- cluding a brief overview of the health information collected in DHS surveys. Subsequently, the contri- bution of DHS surveys to our knowledge of family health issues in developing countries is discussed, followed by an assessment of the limitations of DHS surveys. The article concludes with a summary of the role of DHS-type surveys in family health measurement during the remainder of the 20th century. OHS The DHS programme, funded by the United States Agency for International Development, started in 1984 as a successor to the World Fertility Survey (WFS) (4). The WFS had mainly focused on collect- ing information on fertility, child mortality and family planning, but DHS expanded this to include information on health and nutrition. DHS surveys are based on nationally-representative samples. The respondents are women of childbearing age. An increasing number of surveys include a men's questionnaire as well, focusing on family planning and AIDS-related issues. DHS surveys are implemented by national orga- nizations: usually the Bureau of Statistics, Ministry a Senior Technical Adviser, TANESA project (Tanzania- Netherlands project to Support AIDS/HIV Control), Mwanza, United Republic of Tanzania. b Senior Health Expert, DHS, Columbia, MD, United States of America. 222 of Planning or Ministry of Health, often in collabo- ration. DHS staff provide technical assistance to the implementing organization in the areas of gen- eral survey design, questionnaire development, sampling, data processing and editing, data analy- sis, report writing, dissemination of results and fur- ther analysis. The DHS model or core question- naire is adapted in each country, as are other sur- vey documents such as interviewers' manuals and tabulation plans. An integrated software pack- age (ISSA: integrated system for survey analysis) has been developed specifically for DHS surveys. Tab/,e I lists countries with completed DHS sur- veys as of September 1993 with year of survey and sample size. Several countries have now conducted two DHS surveys. Health indicators in OHS The model DHS questionnaire includes questions to collect information on the following health indi- cators: Fertility and mortality Estimates of fertility and child mortality, as well as information on ages of children, are based upon a full birth history from women 15-49 at the time of the survey. The respondent is asked to provide information on all her children born alive, starting with the most recent birth. This information in- cludes the child's sex, birth date, and survival status. If a child died, the age at death is recorded. The birth history data can be used to obtain estimates of current fertility levels and trends, and levels and trends in neonatal, infant, early child- hood and under-5 mortality for the 15 years prior to the survey. Mortality estimates are mostly made for 5-year periods. Several countries have included maternal mortality modules with questions on adult mortality for both sexes which can also be used to estimate male and female adult mor- tality (5). Anthropometry Initially, data on weight and height were only col- lected for children under 3 or under 5 years, but more recently maternal anthropometry has also been included in DHS surveys. Levels of wasting, stunting and underweight at the time of the survey are estimated. Rapp. trimest statist. sanit. mond., 46 (1993) Table 1 Countries with OHS surveys, year of survey and number of respondents (mostly women 15-49 years). Tableau 1 Pays participant au programme "Enquete demographique et de sante", et nombre de repondants (surtout des femmes, a.gees de 15 a 49 ans) Region/Country Year of survey Number of respondents Region/Pays Annee de l'enquMe Nombre de repondants Sub-Saharan Africa -Afrique subsaharienne Botswana 1988 4 368 Burkina Faso 1992/93 6 OOO Burundi 1987 3 970 Cameroon - Cameroun 1991 3 871 Ghana 1988 4 488 Kenya 1988/89 7150 Kenya 1993 8 OOO Liberia - Liberia 1986 5 239 Madagascar 1992 6 260 Malawi 1992 4 850 Mali 1987 3 200 Namibia - Namibie 1992 5 421 Niger 1992 6 503 Nigeria - Nigeria 1990 8 781 Nigeria - Nigeria (Ondo State - etat d'Ondo) 1986/87 4 213 Rwanda 1992 6 551 Senegal - Senegal 1986 4 415 Senegal - Senegal 1992/93 6 500 Sudan - Soudan 1989/90 5 860 Togo 1988 3 360 United Republic of Tanzania - Rep.-Unie de Tanzanie 1991/92 7 650 Uganda - Ouganda 1988/89 4 730 Zambia - Zambie 1992 7 060 Zimbabwe 1988/89 4 201 North Africa and Asia -Afrique du Nord et Asie Egypt - Egypte 1988/89 8 911 Egypt - Egypte 1992 9 864 Indonesia - lndonesie 1987 11 884 Indonesia - lndonesie 1991 22 909 Jordan - Jordanie 1990 6 462 Morocco - Maroc 1987 5 982 Morocco - Maroc 1992 9 256 Nepal - Nepal (KAP- CAP/gap)a 1987 1 623 Pakistan 1990/91 6 611 Philippines 1993 15 OOO Sri Lanka 1987 5 865 Thailand - Thailande 1987 6 775 Tunisia - Tunisie 1988 4184 Turkey- Turquie 1993 7 500 Yemen 1991/92 5 687 Latin America and Caribbean - Amerique latine et Caraibes Bolivia - Bolivie 1989 7 923 Brazil - Bresil 1986 5 892 Northeast Brazil - Nord-Est Bresil 1991 6 222 Colombia - Colombie 1986 5 329 Colombia - Colombie 1990 8644 Dominican Republic - Republique dominicaine 1986 7 649 Dominican Republic (experimental)b - Republique dominicaine (experimental)b 1986 3 885 Dominican Republic - Republique dominicaine 1991 7 320 Ecuador- Equateur 1987 4 713 El Salvador 1985 5 207 Wld hlth statist. quart., 46 (1993) 223 Table 1 (continued) Countries with OHS surveys, year of survey and number of respondents (mostly women 15-49 years). Tableau 1 (suite) Pays participant au programme «Enquete demographique et de sante», et nombre de repondants (surtout des femmes, a.gees de 15 a 49 ans) Region/Country Region/Pays Guatemala Mexico - Mexique Paraguay Peru-Perou Peru (experimental)b - Perou (experimental)b Peru-Perou Trinidad and Tobago - Trinite-et-Tobago Year of survey Annee de l'enquete 1987 1987 1990 1986 1986 1991/92 1987 Number of respondents Nombre de repondants 5160 9 310 5 827 4 999 2 534 15 882 3 806 •This survey explored the discrepancies ("gap") between fertility preferences, and knowledge, attitudes and practices (KAP) of family planning.· Cette enquete examinait les ecarts ("gap") entre les preferences en matiere de fecondite et les connaissances, attitudes et pratiques (CAP) dans le domaine de la planification familiale. bMethodological issues. - Aspects methodologiques. Family planning Detailed information is collected on current and past use of family planning methods, and knowl- edge of contraceptive methods. Several questions are asked to assess (met and unmet) needs for family planning and availability of family planning services. Maternity care Questions are asked on antenatal care (attendant, number of visits, timing of first visit), tetanus tox- oid vaccination (number of doses received during each pregnancy), and delivery care (place and at- tendant). In addition, data are collected on birth weight, both recalled birth weight and subjective size of the baby as reported by the mother. Recent- ly, a question on complications during delivery has been added. Child feeding DHS surveys provide data on breastfeeding and supplementary feeding patterns. Vaccination Coverage estimates are made using information from child health cards and mother's recall of vaccinations for all children under 5 years. Child morbidity Questions on the prevalence of diarrhoea, fever and cough, with or without rapid breathing, are used to assess curative health services utilization and treatment patterns. AIDS Questions about knowledge of HIV I AIDS have re- cently been added to the model questionnaire, and 224 an AIDS module includes questions about sexual behaviour and condom use. Contributions made by OHS DHS surveys have made a significant contribution to family health programmes. They provide a glo- bal, comparable, high-quality data base on a wide range of health indicators. Firstly, DHS surveys are alleviating the paucity of data on health status in developing countries. Very few sources of data on fertility and child mor- tality are available, and child (and maternal) an- thropometry data from DHS surveys also make an increasingly important contribution. Extensive use is made of DHS surveys in many health studies by national and international organizations and re- searchers. Long-term trends in fertility and mortal- ity and MCH (maternal and child health) indica- tors can be assessed and used to challenge or sup- port current opinion about priority health inter- ventions. Comparative studies using DHS data are powerful tools to ascertain commonalities or ex- ceptional patterns, and to compare countries or regions (6-10). Secondly, DHS surveys provide population- based coverage data of key health services indica- tors. These indicators can be disaggregated and differentials can be assessed by geographic, biode- mographic (sex of child, birth interval, etc.) and socioeconomic characteristics (mother's educa- tion, rural-urban residence, attributes of the house- hold, etc.). Previously, rather limited data were available on health status and health care utiliza- tion differentials. Continuous evaluation of data quality (11,12) and inputs from international experts are used to create the best possible health interview instru- ments. Health interview surveys are a complicated and relatively under-developed field ( 13), and im- Rapp. trimest. statist. sanit. mond., 46 (1993) provements in the health section of the DHS ques- tionnaire are made continuously. Extensive train- ing and supervision during field work, almost con- current data entry and editing, and use of data- quality tables with feedback during field work are methods which ensure data quality. A particular strength ofDHS surveys is the quick- ness of publication of results after the end of field work. Through standard approaches in question- naire design and contents, data processing and tabulation programmes, most tabulations are avail- able within 2 or 3 months after the end of field work. limitations Virtually all information collected in DHS surveys is subject to reporting and recall biases ( except weight and height measurements and vaccination data copied from the child health card). Current- status data are used to estimate duration of breast- feeding, postpartum amenorrhoea, etc., since that methodology is least vulnerable to biases. Certain information such as the age of the respondent, birth dates of children, and age at marriage refers to events in the past. This inevitably causes biases, although detailed evaluation of DHS data has shown that these data are reasonably well reported. Omission of births or deaths, the most serious problem of cross-sectional surveys, is generally within limits. Most health information is based on women's reports: for example, concerning child diarrhoea and respiratory symptoms in the past two weeks, or use of maternity care. Misclassification biases are known to occur. The magnitude of the bias is often unknown and correcting for the bias is difficult. As long as the biases are fairly random, the aggregate estimates of indicators will be fairly adequate, but individual-level data will have to be interpreted more carefully, especially when mak- ing causal interpretations. DHS surveys are also limited to health indica- tors which can be measured with relatively few questions. For instance, malaria, tuberculosis, and AIDS are illnesses for which no satisfactory ques- tions are available. Recall of diets for children is cumbersome and requires multiple questions and considerable expertise from the interviewer. Deter- mination of economic status is limited to a short list of durable goods. Determination of cause of death through verbal autopsy gives a rough idea of the importance of selected causes of death, but is not precise enough for evaluation of the impact of health interventions or assessment of trends in cause-specific mortality. In addition, DHS data on health services are limited to assessing availability and utilization, and no data on quality of care are collected. There are better methods than cross-sectional surveys to as- sess the quality of care ( 14). The disadvantage of using a standardized ques- tionnaire is that there are limited opportunities to Wld hlth statist quart., 46 (1993) adapt the questionnaire to be locally relevant. Ad- ditions, deletions and changes are made in every DHS survey, but the number of modifications is limited in order to maintain comparability, limit complexity of the survey, and keep the length of the questionnaire within limits. Disaggregation to district level is desirable, since a district is often the major unit of implemen- tation of health programmes. DHS surveys are gen- erally not designed to yield estimates of health indicators at the district level, since this is too cost- ly. A sample of at least 1 OOO to 1 500 women is required in order to obtain valid estimates of fertil- ity and child mortality. Other maternal and child health indicators do not need such a large sample size, but sampling errors are usually large at district level. Hitherto, the DHS questionnaire has focused mainly on child health indicators, and much less on adolescent and adult health. Although it may be possible to ask questions about adult health, the current length of the questionnaire prohibits con- siderable expansion. Special modules are, howev- er, being considered. Nationally representative surveys are costly and require considerable expertise. For most develop- ing countries external assistance is required to cov- er the local costs of 3-6 months of field work (50-80 staff in the field), and data processing and editing, while the costs of technical assistance may amount to half of the survey expenses. Discussion DHS surveys are an important source of data on health of families in developing countries. Both at the national and international level DHS surveys provide much needed data on fertility and family planning, on mortality and nutrition, and on health services utilization. Adult health and health examination data ( e.g., physical examination and laboratory investiga- tions) are likely to receive more attention in DHS surveys later during the decade. For instance, DHS surveys could be used to provide more detailed data on the spread and determinants of the cur- rent HIV I AIDS pandemic, if simple diagnostic techniques are used, such as a saliva test or blood- spotted filter paper analysis. DHS surveys could also be the main source of data for monitoring the health goals of the World Summit for Children. The summit set specific goals for improvements in maternal and child health for the decade 1990-2000. A review of the indicators shows that, although several goals can be monitored with a well-functioning national health information system, more than two-thirds of the health goals require nationally representative sample surveys. Therefore, each country committed to the goals should have a DHS-type survey at the beginning and at the end of the decade. A survey conducted be- 225 tween 1989 and 1994 could provide baseline data and the data necessary for programme planning. A survey ten years later should be conducted to evalu- ate achievements during the decade. Summary Surveys conducted in the context of the Demographic and Health Surveys (DHS) programme are an important source of data on health of families in developing coun- tries. Both at the national and international level, DHS surveys provide much-needed data on fertility and fam- ily planning, on mortality and nutrition, and on health services utilization. The use of uniform survey instru- ments allows detailed international and subnational comparisons of health status and health care. Limita- tions of the DHS surveys are also discussed. Resume Enquetes demagraphiques et de sante: source d'infarmatian sur la sante des families dans /es pays en develappement Les enquetes conduites dans le cadre du programme «Enquetes demographiques et de sante» constituent une importante source d'informations sur la sante des families dans les pays en developpement. Sur le plan national comme sur le plan international, elles fournis- sent des renseignements particulierement precieux sur la fecondite et la planification familiale, la mortalite et la nutrition ainsi que sur !'utilisation des services de sante. Le recours a une instrumentation uniforme d'enquete permet des comparaisons detaillees de la situation sanitaire et des soins de sante a l'echelle internationale et infranationale. Les limites des enquetes conduites dans le cadre du programme sont egalement passees en revue. 226 References/References 1. World Health Organization. Development of indicators in the context of primary health care and health f <Yr all IYy the year 2000. Health for All Series, Geneva, WHO, 1981. 2. Hansluwka, H.E. Measuring the health of populations: indicators and interpretations. Social science & medicine 20: 1207-1224 (1985). 3. Cuh, R. et al. (eds), The Unicef GOBIFFF programtM. Beckenharn, Croom Helm, 1987. 4. l.apharn, R.J. &: Westoff, C.F. Demographic and health surveys: population and health information for the late 1980s. Population index52(1): 28-34 (1986). 5. Rutenberg, N. &: Sullivan,J.M. Direct and indirect estimates of maternal mortality from the sisterhood method. In: DHS world conference proceedings, IRD/Macro International Inc., Columbia, Maryland, USA, Vol. III: 1669-1696. 1991. 6. Boerma, J.T. et al. Immunization: levels, trends, and differentials. DHS comparative studies No. 1, Macro International Inc., Columbia, Maryland, USA, 1990. 7. Boenna,J.T. et al. Child morbidity and treatment patterns. DHS comparative studies No. 2, Macro International Inc., Columbia, Maryland, USA, 1991. 8. Govindasamy, P. et al. High-risk births and maternity care. DHS comparative studies No. 8, Macro International Inc., Columbia, Maryland, USA, 1993. 9. Bicego, G.T. &: Boerma, J.T. Maternal education and child survival: a comparative study of survey data from 17 countries. Social science &medicine36(9): 1207-1227 (1993). 10. Hobcraft,J. Child spacing and child mortality. In: DHSworld conference proceedings, IRD/Macro International Inc., Columbia, Maryland, USA, Vol. II: 1157-1181, 1991. 11. Institute for Resource Development. Assessment of DHS-1 data quality. DHS methodol.ogical reports No. 1, IRD/Macro Systems, Col~bia, Maryland, USA, 1991. 12. Demographic&: Health Surveys. Assessment ofDHS-1 health data quality. DHS methodol.ogical reports No. 2, Macro International Inc., Columbia, Maryland, USA, 1993. 13. Ross, D. &: Vaughan, J.P. Health interview surveys in developing countries. Studies in family planning 17 ( 1): 78-94 (1986). 14. Newman, J.S. Assessing the quality of health services. In: Boerma J. T. (ed.), Measurement of maternal and child murtality, morbidity and health care: interdisciplinary approaches. Liege, Editions Ordina-Derouaux for IUSSP. Rapp. trimest. statist. sanit. mond., 46 (1993) Natural and man-made disasters: the vulnerability of women- headed households and children without familiesa Debarati G. Sapirb Introduction The impact of natural or man-made disasters does not fall equally or at random. Certain characteris- tics and factors may be used to identify communi- ties at higher risk. Since 1980, over 2 million peo- ple have died as an immediate result of natural and man-made disasters. The refugee population has grown 500% since 1970 compared to a 20% growth in the world population, registering nearly 16 mil- lion refugees in 1992. This estimation does not include the internally displaced, of which there are 1.2 million in the Philippines alone. More than half of these are women and children. In 1992 alone, more than 300 million people had their homes or livelihoods destroyed directly by disas- ters. The human impact of natural and man-made disasters has evolved over the last three decades. Since the recent unfolding of the disasters in So- malia, Sudan, former Yugoslavia, Cambodia and Afghanistan, the world is recognizing that econom- ic dislocation, natural disasters, collapsing political structures, famines and mass displacements have all woven together to affect millions in ways both profound and prolonged. Analysis of disaster im- pact and relief effectiveness has been seriously hampered by the lack of consistent and accurate data or standard management information. Data collection for any an·alytical purpose has been a task in itself and therefore policy-making has re- mained ad hoe. Since the Sahelian famines of the mid-l 970s, followed by their recurrence in the mid-l 980s, world interest in disasters has increased and consequently, reporting has improved. Natural disasters In terms of frequency of occurrence, floods and wind-related phenomena are by far the most com- mon. They represent more than 60% of all disas- ters requiring external assistance. (Fig. 1) Famines and droughts, while fewer in number, have a great- er and more profound impact on populations, gen- erally affecting extensive areas and very large popu- lations. Increasingly, since the 1970s, famines and droughts have been linked to civil strife and armed a This article was adapted from a document prepared for the WHO Expert Committee Meeting on Maternal and Child Health and Family Planning, December 1993. b Universite catholique de Louvain, 30.34, Clos Chapelle aux Champs, 21200 Brussels, Belgium. Wld hlth statist. quart., 46 (1993) Fig. 1 Percentage distribution of disasters by type, 1960-1989 Repartition en pourcentage des catastrophes, par type, 1960-1989 Earthquakes Tremblements de terre 15.97% WH094025 Source: EM-DAT database, WHO Collaborating centre for research on the epidemiology of disasters (CRED), Brussels. I Base de donnees EM-DAT. Centre collaborateur de l'OMS pour la recherche sur l'epidemiologie des catastrophes (CRED), Bruxelles. conflicts. Pure famines, such as the Great Bengal Famine in 1942, have become rare occurrences. Armed conflicts (generating famine) have started to claim larger and larger shares of total disaster mortality. Fig. 2 displays a combined chart of per- centage distribution by type of disaster and per- centage distribution of mortality due to these events. Civil strife and famines, although relatively infrequent, have a disproportionate effect on pop- ulations. Representing a little over 20% of all disas- ters, they account for nearly 70% of the direct mortality. The human impact of disasters consists of two elements, the catastrophic event and the vulnera- bility of people. While countries like Bangladesh and the Philippines are in geographically vulnera- ble situations, there is no doubt that their main susceptibility comes from their weak social and economic structures. Housing quality, pre-existing health and nutritional status, social welfare infra- structure, and economic resilience determine the magnitude of the disaster effect and its long-term 227 Fig. 2 Distribution of disaster mortality and type as proportion of all categories, 1960-1989 Repartition des deces par catastrophes et types de catastrophe en proportions de toutes les categories, 1960--1989 % 70ltlJ5WjJ5Wj7]~~~~~~~~~---i 60 50 40 30 20 10 0 Famines Civil unrest/ Famines Situation de troubles Earthquakes/ T remblements de terre Floods/ lnondations Wind related/ Phenomenes eoliens Category of disasters/Categorie de catastrophe I@% killed/% tues - % of events/% d'evenements catastrophiques Source: EM-DAT database, WHO Collaborating centre for research on the epidemiology of disasters (CRED), Brussels, 1993. I Base de donnees EM-DAT, Centre collaborateur de !'OMS pour la recherche sur l'epidemiologie des catastrophes (CRED), Bruxelles, 1993. sequelae. Furthermore, broader ecological factors, such as population pressures on land, increasing urbanization, unplanned land-use, and marginal- ization of populations are aggravating the poten- tial for increased losses when disaster does occur. This is well illustrated by comparing the cases of earthquakes in Managua (1972) and San Fernando Valley, California (1971) (]) . As shown in Tabl,e 1, despite a lower Richter scalec reading, a smaller range of destruction on the Mercalli scaled and a much smaller population in affected area, Man- agua suffered 80 times more casualties than Cali- fornia. Similarly, in 1974, Hurricane Fifi left an estimated 8000 dead in Honduras, crashing through at a wind speed of 250 km/h and causing 80% destruction in the impact area. In the same year, Cyclone Tracy, with comparable wind speed and impact zone destruction, killed 49 persons in Darwin, Australia (2) . Finally, the differential vul- c The Richter scale measures the intensity of the seismic activity at the epicentre on a logarithmic scale. This means that a one- unit increase represents an important proportion. d The Mercalli scale measures the extent of physical damage over surface area on a scale of I-XII. 228 nerability of persons within an affected zone is illustrated by the 1976 earthquake in Guatemala where, of the 1 200 persons killed and 90 OOO left homeless, the large majority were from the poorer slum sections of the city (3) . Floods, although less fatal than earthquakes, affect much larger numbers of people and in long- lasting ways. Harvests are lost, land is salinated, and cattle are drowned, thus destroying people's means of livelihood. Similarly, fewer people die as a direct result of famines or droughts but the scope of destitution is higher than in other disasters such as earthquakes and cyclones (4) . Armed conflict By far the largest proportion of total victims of disasters in recent years is that caused by civil armed conflict (Fig. 2) . In 1990, war-related famine affected about 20 million people only in southern and north-eastern Africa. In addition, armed con- flicts have killed almost three times as many civil- ians as soldiers (5) . Of these, the great majority are women and children (6-8). Protracted civil unrest leaves in its wake dislocated people, disrupted economies, poverty and famine that are long-term phenomena. Furthermore, direct actions to affect civilian populations, such as burning of harvests (e.g., western sub-SahelianAfrican countries, 1983- 1987), contamination of wells or other drinking- water sources (e.g., southern Sudan, 1990) or di- version of food aid to the military ( e .g., Ethiopia, 1988) are standard practices. Maternal and child health conditions, precarious in normal circum- stances, are further aggravated by these actions. Food as an instrument of war. An especially reprehensible practice of primary sig- nificance for children is the use of food as an instrument of war. Scorched-earth policies, restric- tion of the passage of humanitarian food aid, and diversion of food to the military are all common occurrences in most of the conflicts experienced in recent times. Diversion to the military of emergen- cy food aid intended for vulnerable groups is so common as to be, in some cases, counted into the calculation for needed supplies as the percentage reserved ( or lost) to diversion. In Asmara (Eritrea), for example, the militia was paid in food-aid grain (9). In Somalia, Askin estimated that only 12% of the food aid reached the civilian victims for whom it was destined ( 10). Besides diversion, feeding cen- tres for children and vulnerable groups are fre- quently bombed or attacked. McRae & Zwi (5) report that feeding centres were attacked in all of the study countries which included Sudan, Mozam- bique, Ethiopia, Angola and Liberia. Very few fac- tual reports exist on these issues, partly because systematic reporting has been neglected and partly because the publication of such information could Rapp. trimest. statist. sanit. mond., 46 (1993) Table 1 Comparison of characteristics of earthquakes in Managua, Nicaragua (1972) and San Fernando Valley, California (1971) Tableau 1 Comparaison des caracteristiques des tremblements de terre survenus a Managua au Nicaragua (1972) et dans la vallee de San Fernando en Californie (1971) Disaster characteristics - Caracteristiques de la catastrophe Managua San Fernando Valley Richter scale reading - Degre de magnitude sur l'echelle de Richter Extent of destruction (Mercalli range VI-VII) - 5.6 100 km2 6.6 1 500 km2 Etendue des destructions (I-II sur l'echelle de Mercalli) Population in affected area - Population des zones touchees Dead - Nombre de marts Injured - Nombre de blesses Source: Seaman, 1984 (1). jeopardize the implementation of emergency food aid programmes. Mines and disability. The use of mines, like direct atrocities, serve to remind the community that the rebel groups exist and command a certain power. They have a devas- tating effect on rural communities, particularly be- cause of their continued power to disable and de- stroy for years following the war. They limit the community's ability to migrate which is, in many cases, tantamount to survival. The number of per- sons disabled from mine injuries is growing global- ly with the increasing use of this method of destabi- lization. Save the Children Fund-UK reported that more than one million mines have been planted in Somalia (11). Certain countries such as Angola, Table 2 Selected disasters with large numbers of unaccompanied children Tableau 2 420 OOO 5 OOO 20 OOO 7 OOO OOO 60 2 540 Mozambique and Cambodia are now home to the largest numbers of mine-disabled people in the world (5). Most of the victims of mine-related inju- ries are civilians and a significant proportion are children. Less aware of mined areas or unseen dangers, children wander across the countryside at will, exposing themselves to greater risk. Unaccompanied children, violence and disruption Abandoned or unaccompanied children are a mis- erable but inevitable corollary to many emergen- cies, especially of the type we experience today. In most catastrophic situations - war, famine, refugee movements, natural disasters - children have been separated from their families ( Tabl.e 2). These chil- dren may be abandoned, orphaned, lost, abducted Selection de catastrophes dans lesquelles ant ete denombres de nombreux enfants non accompagnes Year An nee 1915 1919 1936 1939 1948 1950 1954 1954 1956 1960 1970 1970 1970 1972 1975 1975 1979 Disaster event Evenement catastrophique Armenian massacre - Massacre en Armenie Russian famine and revolution - Famine en Russie et revolution Spanish Civil War -Guerre civile espagnole World War II -Deuxieme Guerre mondiale Greek Civil War -Guerre civile en Grece Korean War- Guerre de Goree Tibetan refugees - Refugies tibetains Viet Nam War -Guerre du Viet Nam Hungarian Revolt - Soulevement en Hongrie Cuban Revolution -Revolution cubaine Nigerian Civil War - Guerre civile au Nigeria Bangladesh cyclone and tidal wave - Cyclone et raz de maree au Bangladesh Bangladesh War of Independence - Guerre d'independance du Bangladesh Famine in Ethiopia - Famine en Ethiopie Viet Nam refugee exodus - Exode des refugies du Viet Nam Laotian refugees - Refugies laotiens Cambodian crises - Crises cambodgiennes Wld hlth statist. quart., 46 (1993) Estimated number of unaccompanied children Estimation du nombre d'entants non accompagnes 132 OOO 800 OOO 90 OOO 13 OOO OOO 37 500 100 OOO 2 OOO 880 OOO 6 OOO 17 OOO 100 OOO 7 OOO 400 OOO 2 OOO 22 OOO 2 OOO 11 OOO 229 or recruited into the war effort. Ressler et al. esti- mated that unaccompanied children number in the millions and cited 13 million orphaned or abandoned children during the Second World War and about 100 OOO each during the civil wars in Spain, Korea and Nigeria ( 12). While most often the unaccompanied children are older, in some cases, such as the Korean War or Nigerian Civil War, large proportions have been infants. Child abandonment figures from the Korean War show that two-thirds of the children abandoned in the first two years of the war (61.3 and 66.5%) were less than one year of age. Of these, the majority were girls. Studies of children's response to extreme vio- lence, death, abuse, and hunger indicate that they are able to resist emotional stress and physical hardship as long as they remain with their families and parents ( 13). Emergencies become significant as soon as separations occur and the child's pri- mary attachments are disrupted. In general, chil- dren are most often separated from families in which a death has occurred, the parents are sepa- rated, or where there is a continuing threat to safety, abject poverty or displacement. These con- ditions mostly occur in armed conflicts, refugee situations or famines. Acute natural disasters such as earthquakes or cyclones are less likely to present such conditions. Sexual violence against women (and, apparent- ly, men and children) ( 14), consequent pregnan- cies and their care have been thrown into the lime- light in the recent conflict in Bosnia, although it is far from a rare occurrence in mass conflicts. Wom- en are often forced to provide sex in exchange for food and shelter for themselves or their children and the implications for sexually transmitted dis- eases, unwanted pregnancies and their termina- tion are significant. While data are extremely limit- ed on all issues related to sexual violence against women, high rates of pregnancy, sexually transmit- ted diseases and HIV are recognized as common in these situations and are considered indicative of the levels of desired or undesired sexual practice. Apart from a brief period of media attention in early 1993 along with or in the wake of the UN Expert Team Report on allegations of rape in the former territory of Yugoslavia (E/CN.4/1993/50), there have been no discernible signs of follow up within the large humanitarian programmes operat- ing in the conflict zone. The exemplary effort of family planning offered to Khmer refugees in Thai- land (Editorial, Lancet, April 10, 1993) does not seem to have been replicated elsewhere. Differential risks for mortality and morbidity among women and children in emergencies It seems obvious that children and women are particularly vulnerable in disaster situations. How- ever, until recently, natural disasters were thought 230 to affect people in an non-discriminatory fashion. This notion has not been borne out, at least with regard to sex discrimination, as reported by Rivers and Sen in relation to famines (8,15) and the work of Beinin ( 16) on earthquakes. Rivers has made a convincing case regarding differential mortality among girls and women in disaster situations. Al- though most of his case is based on famines, the conclusions are generalizable across other situa- tions of extreme social and environmental stress. Tab/,e 3 displays sex differences in the prevalence of malnutrition in two African famines and that in Bangladesh, which, Rivers argues, contradicts exist- ing evidence pointing to greater physiological re- sistance of females compared to males, all other things being equal. Among the health effects of man-made disas- ters on children, mental illness is a much neglected but important aspect, particularly in violent social disruptions, refugee situations or displacement. The impact on children is especially serious as they are less able to resist the psychological and physical aggression that surrounds them. For example, in the Philippines, there are over one million inter- nally displaced persons, 60% of whom are chil- dren. Although figures are not available, reports from the agencies working in the evacuation and drop-in centres of the "militarized zones" indicate that the prevalence of children who arrive at these centres in states of apathy and mental derange- ment is alarmingly high. The local health system, including the non-governmental agencies, are bad- ly prepared to handle these children and in most cases, mental diseases in children are left to degen- erate until death. With regard to children in situations of war, a Harvard study team visited paediatric wards of sev- eral hospitals in Iraq following the Gulf crisis in 1990 ( 17). They reported high mortality and high prevalence rates of preventable diseases which they attributed to the war and trade sanctions. Gastro- enteritis and severe malnutrition featured high among the causes of children's admission to hos- pital. Reduction in breastfeeding, based on a UNICEF survey in 1990, was suggested as a contrib- utory factor by the Harvard authors for both mal- nutrition and gastroenteritis. The authors also cit- ed reports of mothers who substituted other, inad- equate foods (e.g., rice water or sugar solutions) for infant formula which was prohibitively priced. Local studies in Somalia with small samples have reported astonishingly high crude death rates of children under 5 years of age. A study done in the Baidoa camps (18) reported that 75% of the children under 5 years of age had died in an 8- month period. Risk factors reported included dis- placement and mortality from preventable diseases as the main cause of death. In another study of the displaced population in Merca and Qorioly ( 19), south of Mogadishu, reported a mortality of 25% Rapp. trimest. statist. sanit. mond., 46 (1993) r:~l:i~erences in the prevalence of malnutrition in children in famines in Africa and Bangladesh Tableau 3 · · B I d h Differences selon le sexe dans la prevalence de la malnutrition chez les enfants lors de famines en Afnque et au ang a es Males Females F:M ratio Sexe Sexe Rapport masculin feminin F:M Bangladesh Mean energy intake -Apport energetique moyen 809 694 86 Mean protein intake - Ration moyenne en proteines 23.0 20.2 88 Nunmber of visits for diarrhoea treatment/1 OOO population of each sex - 136 82 60 No. de consultations pour soigner les diarrhees/1 OOO personnes de chaque sexe <60% W/A- P/A 5.1 14.4 282 <755 W/A- P/A <85% H/A- T/A <90% H/A- T/A Burkina Faso <80% W/H - Pff Ethiopia - Ethiopie <80% W/H Pff Adapted from - D'apres: Ref. - Ref.(B). W/A: Weight for age. - P/A: poids selon l'age. H/A: Height for age. - T/A: taille selon l'age. W/H: Weight for height. PIT: poids selon la taille. among children less than 5 years of age in the year preceding the study. They cited malnutrition as the main cause followed by war casualties. A study in famine-suuck areas of Chad in 1985, compared households which had moved out of their villages in search of food to those which had remained in place (20). They found children from displace~ families to have lower vaccination rates and nutri- tional status than those who remained in their villages ( Tabl.e 4). Malnutrition was associated not only with displacement from villages, but also with woman-headed families. The men had typically left earlier in search of food and revenue. Following a long absence of the main income-earner and final- ly, total destitution of the village, the family moved to an urban centre with the mother as head of household. Studies in the Thai-Kampuchea camps (21) re- port highest risks of death in children from acute respiratory infections, diarrhoeal diseases, malaria and malnutrition, but diarrhoea remains the most significant killer in most instances. The extre~e vulnerability of under-fives is reflected by mortality rates that ranged from 5 to 8 times the normal rates ( Tabl.e 5). In terms of direct trauma and vulnerability of children to earthquakes, evidence from a study done in Italy (22) following the 1980 Campania earthquake shows that mortality among older chil- dren (5-9 years) was disproportionately higher. W/d hlth statist quart., 46 (1993) 54.8 59.6 109 16.7 34.8 208 26.2 32.7 125 12.1 18.9 163 3.8 6.1 138 Similarly, the mortality data from the Sumpango earthquake in Guatemala and the Managua earth- quake in Nicaragua, both in 1976, show excess mortality in the older age-groups of children. In conclusion, most of these studies provide much-needed insights into the patterns and trends in mortality and morbidity in these special situa- tions. But in general, knowledge of the diseases or health risks to which children are especially ex- posed in these conditions remains limited. It is hard to draw convincing conclusions from these studies and generalize for policy change. Further confirmation and support for better studies are required. It is true that the chaotic nature of civil wars and disasters makes systematic evaluation of health effects on civilian populations very difficult. But a better understanding of these aspects of the impact of disasters on children would greatly im- prove response planning and public health servic- es. Summary Since 1980, over 2 million people have died as an immediate result of natural and man-made disasters and by 1992, the refugee population registered nearly 16 million people. This article reviews the human impact of disasters as a composite of two elements: the cata- strophic event itself and the vulnerability of people. It 231 Table 4 Comparative mortality rates of infants and children from studies in situations of civil strife and famine Tableau 4 Comparaison des taux de mortalite chez les nourrissons et les enfants a partir d'enquetes en situations de guerres civiles et de famines Sample size <1 1-4 <5 Source Taille de l'echantillon Mozambique (6)3 Gaza 329 40.0 90.0 In ham bane 358 233.0 125.0 Manica 210 83.0 80.0 Tete 972 172.0 95.0 Thailand - Tha"ilande (21)b Nov. 1979 NA/SO 10.7 7.6 Sudan - Soudan (21)b Camp D NA/SO 5.6 23.8 Somalia - Somalie (21)b NW Camps, Sept. 1980 NA/SO 27.0 14.0 Baidoa, Apr-Nov 1992. 62 32.0 (18)b Afgoi (urban) Apr-Nov 1992 211 10.4 Afgoi (rural) 1989 NA/SO 117.1 (23) 3 Merca/Qorioly April 1992 (19)3 Resident 442 115.4 Displaced in camps 586 240.6 Displaced in towns 232 86.2 • Rates expressed as deaths per 1 OOO per year. - Taux exprimes en deces pour 1 OOO par an. b Rates expressed as deaths per 10 OOO per day. - Taux exprimes en deces pour 10 OOO par jour. Table 5 Mortality-related risk factors among children of famine-affected households by displacement status (Batha province, Chad 1985) Tableau 5 Facteurs de risque lies a la mortalite chez les enfants de foyers affectes par la famine selon la categorie de deplacement (province de Batha, Tchad, 1985) Risk factor Facteur de risques Immunization coverage - Couverture vaccinale: Measles - Rougeole BCG Wt/Ht <80% of reference population Pff <80% de la population de reference Non- displaced Non deplaces 22.4 20.1 12.7 also examines the specific case of women and children in the current world emergency context. It identifies four broad policy areas that affect women and children in disaster situations and discusses them with examples and field evidence. The first policy area addresses humanitarian assistance and armed conflicts, and armed conflict and international humanitarian law, the use of food as instrument of war, mines and civilian disability, and rape and sexual violence are discussed within this context. The second problem discussed is the issue of unaccompanied and abandoned children in terms of its magnitude and implications for relief re- sponse. Thirdly, the article examines the differential 232 Displaced - Deplaces Permanent Temporary Permanents Temporaires 4.4 2.2 20.0 9.6 10.5 11.9 8.7 9.1 12.2 Total risks in emergencies for mortality and morbidity, specif- ically for women and children. Finally, it addresses certain policies and approaches to disaster rehabilita- tion which effectively mirror and reinforce inherent ineq- uities in the affected society. The article notes that: (i) the largest proportion of disas- ter victims today arise from civil strife and food crises and that the majority of those killed, wounded and permanently disabled are women and children; and (ii) the ability of any country to respond effectively to di- sasters depends on the strength of its health and social infrastructure, and its overall developmental status. It Rapp. trimest. statist. sanit. mond., 46 (1993) concludes by identifying seven areas where concrete measures could be taken to improve the current situa- tion. Resume Secours et reconstruction en cas de catastrophe: problemes re/atifs aux femmes et auxenfants Depuis 1980, plus de deux millions de personnes sont mortes des suites immediates de catastrophes naturel- les ou dues a l'homme et le nombre total des refugies avoisinait les 16 millions en 1992. Le present document decrit l'impact sur l'homme des catastrophes, sous deux angles: d'une part l'evenement catastrophique et d'autre part la vulnerabilite de la population; ii examine aussi en particulier le cas des femmes et des enfants dans le contexte mondial actuel des situations d'urgen- ce. II recense quatre grands domaines qui affectent les femmes et les enfants en cas de catastrophe et les examine a l'aide d'exemples et de temoignages re- cueillis sur le terrain. Le premier domaine concerne l'aide humanitaire et les conflits armes et l'auteur exami- ne sous cette rubrique les conflits armes et le droit humanitaire international, !'utilisation des denrees ali- mentaires en tant qu'instruments de guerre, les mines et les incapacites, le viol et les violences sexuelles dans la population civile. La deuxieme partie traite de la ques- tion des enfants sans famille ou abandonnes sous l'angle de son ampleur et de ses incidences pour la fourniture des secours. En troisieme lieu, le document examine l'ecart, en situation d'urgence, entre les risques de mortalite et de morbidite, en particulier pour les femmes et les enfants et, enfin, ii aborde la question des politiques et des approches relatives a la readaptation a la suite d'une catastrophe qui renforcent en fait et refletent les inegalites inherentes a la societe affectee. L'article note que i) ce sont surtout les troubles civils et les situations de penurie alimentaire qui font aujourd'hui le plus grand nombre de victimes en cas de catastrophe et que la majorite des personnes tuees, blessees ou handicapees a vie sont des femmes et des enfants et ii) que !'aptitude d'un pays a faire face efficacement aux catastrophes depend de l'etat de son infrastructure sanitaire et sociale, ou de son niveau de developpement general. II distingue en conclusion sept domaines ou des mesures concretes aideraient a ameliorer la situa- tion. References - References 1. Seaman,J. Epidemiol.ogy of natural disasters. Basel, Karger, 1984. Wld hlth statist. quart, 46 (1993) 2. Sapir, D.G. & Lechat, M.F. Reducing the impact of natural disasters : why aren't we better prepared? Health policy and planning, 1(2): 118-126 (1986). 3. Glass, R & Urrutia,JJ. Earthquake injuries related to housing in a Guatemalan village. Science, 197: 638-643 (1977). 4. Sapir, D.G. & Lechat, M.F. The impact of natural disasters: a brief analysis of characteristics and trends. Journal of the world association of emergency and disaster medicine. 2 ( 1-4) : 221- 223 (1986). 5. McRae,J. & Zwi, A.B. Food as an instrument of war in contemporary African famines: a review of the evidence. Disasters, 16(4): 299-321 (1992). 6. Cliff, J. & Razak Noormahomed, A Health as a target: South Africa's destabilization of Mozambique. Social science and medicine, 27(7): 717-722 (1988). 7. Ascherio, A. et al. Effect of the Gulf War on infant and child mortality in Iraq. NewEnglandjournalofmedicine, 327: 931-936 (1992). 8. Rivers,J.P.W. Women and child last: an essay on sex discrimination in disasters. Disasters, 6( 4): 256-267 ( 1982). 9. Keen, D. War, famine and flight in Sudan: III. A disaster for whom? Local interests and international donors during famine among Dinka of Sudan. Disasters, 15(2): 150-165 (1991). 10. Askin, S. Food aid diversion. Midd/,e East report, March-April 1987: 38-39. 11. Save the Children Fund. Emergency updates -Sudan, Ethiopia, Somalia. Overseas Department, London, 1991. 12. Ressler, E.M. et al Unacrompanied children: care and protection in wars, natural disasters, and refugee movements. New York, Oxford, Oxford University Press, 1988. 13. Black, D. Children and disaster. British medical journa~ 285(6347): 989-990 (1982). 14. Black, M.E. et alReproductive freedom for refugees. Lancet, 341: 1285 (1993). 15. Sen, A.K P(ll)erty and famines. An essay on entitkment and deprivation. Oxford, Oxford University Press, 1981. 16. Beinin, L. An examination of health data following two major earthquakes in Russia. Disasters, 5(2): 142-146 (1981). 17. Harvard Study Team . The effect of the Gulf crisis on the children of Iraq. New England journal of medicine, 325(13): 977-980 (1991). 18. Moore, P.S. et al. Mortality rates in displaced and resident populations of central Somalia during 1992 famine. Lani:et, 341: 935-938 (1993). 19. Manoncourt, S. et al.Public health consequences of the civil war in Somalia, April 1992. Lancet, 340: 176-177 ( 1992). 20. Sapir, D.G. et al.Risk factors and vulnerability of children to drought-related famine, the case of Chad. Proceedings of the World Congress of the International Epidemiological Association, Helsinki, 1987. 21. Toole, MJ. & Waldman, RJ. An analysis of mortality trends among refugee populations in Somalia, Sudan, and Thailand. Bulletin of the World Health Organization, 66(2): 237- 247 (1988). 22. De Bruycker, M. et al. The 1980 earthquake in southern Italy: rescue of trapped victims and mortality. Bulletin of the World Health Organization, 61 (6): 1021-1025 ( 1983). 23. Persson, L.A. et al. Famine in Somalia (letter). Lani:et, 341, 5June 1993. 233 Psychosocial and mental health aspects of women's health a Lorraine Dennersteinb Recent decades have seen increasing recognition of the stresses that differentially affect women by virtue of their unequal social status especially in their family roles. Circumstances and conditions that society ac- cepts as normal or ordinary often lead to emotion- al problems in women. Women face dilemmas and conflicts in the contexts of marriage, family rela- tionships, reproduction, childrearing, divorce, ag- ing, education and employment. Stresses that have more impact on women and may contribute to a higher risk for depression in- clude: physical and sexual abuse; sexual harass- ment; gender discrimination; unwanted pregnan- cy; divorce; poverty and powerlessness. Russo (1) reports that epidemiological data link mental dis- order with alienation, powerlessness, and poverty, conditions more commonly experienced by wom- en. A range of studies indicates that women are disproportionately affected by mental health prob- lems and that their vulnerability is closely associat- ed with marital status, employment and roles in society. For example, an official Australian Health Survey in the 1980s found that 357 OOO females and 197 OOO males reported mental health problems. Russo (2) reports that community-based, epide- miological studies have found that for the 15 diag- nostic groups studied there were substantial gen- der differences in lifetime prevalence rates. Wom- en predominated in major depression, agorapho- bia, and simple phobia whereas men predominat- ed in antisocial personality and alcohol abuse/ dependency. Women were more likely than men to have received a diagnosis of dysthymia, obsessive- compulsive disorder, somatization disorder, and panic disorder. No gender differences in manic episodes or cognitive impairment were apparent. It should be noted that there is controversy about the inclusion of alcohol disorders (as well as drug and personality disorders) in the definition of mental disorder as these disorders may not cause personal distress or mental disorganization. There were also marked gender differences evident in the utiliza- a Based on: Dennerstein L. et al. Psychosocial and mental health aspects of women's health. Unpublished WHO document WHO/FHE/MCH/93.1. b Associate Professor, Key Centre for Women's Health in Society, Victoria, Australia. 234 tion of mental health services in the United States of America. Gender differences varied by marital status and race/ethnicity and cannot be explained by biomedical models. Never married and separat- ed/ divorced men have higher overall admission rates to mental health facilities than women in the same marital-status categories. In contrast, married women have higher admission rates than married men. However this does not hold for all diagnoses. These findings emphasize the importance of un- derstanding complex relationships among gender, ethnicity, sex roles and mental health. Research has rarely considered such joint effects. Gove & Tudor (3) had earlier argued that be- cause of the roles women occupy in western soci- ety, women are more likely than men to have emo- tional problems. Major reasons proposed were that: (a) men have two sources of gratification - employment and family, whereas women have only one - family; (b) raising children and keeping house is frustrating; (c) the role of housewife is relatively unstructured and invisible; (d) when a married woman works outside the home, she is in a less satisfactory position than the married man; and (e) expectations confronting women are un- clear and diffuse. These authors noted that both community-based and treatment-based studies clearly showed that more women than men had mental-health problems. This applied to married women and married men. Among single persons, men were more likely to be mentally ill. Prior to World War II, more studies showed a higher rate of mental illness for men than for women, indicating that social change has adversely affected women. Also in communities undergoing economic de- pression, there was a higher incidence of mental illness and the rates were higher in men than in women. In contrast in an integrated French Aca- dian village - traditional, family-oriented and cul- turally isolated - there were lower overall rates of mental illness and women had lower rates than men. Ibrahim (4), in an earlier review of the relation- ship between marital status and mental health, found that married women experience higher rates of mental disorders than married men, al- though single women exhibit rates of mental disor- ders similar to or even lower than the rates shown for single men. Married individuals of both sexes experience better physical health than the unmar- ried. It would appear that being married as corn- Rapp. trimest. statist. sanit. mond., 46 (1993) pared with being single is generally associated with better physical health for both men and women, but is not associated with better mental health for women unless they are gainfully employed. Such employment under certain circumstances may have detrimental consequences. Ross et al (5) tested the hypothesis that in societ- ies which value the family and the woman's role in the home, psychological distress levels of mar- ried men and women will be more similar. These authors utilized a questionnaire to compare an Anglo-American community with a Mexican com- munity. The gap between the psychological distress levels of married men and women was less in Mexi- can culture than in Anglo culture. However, educa- tion and the wife's participation in the labour force affected marital satisfaction which in tum affected the level of psychological distress. Educa- tion increased marital satisfaction whereas paid employment decreased it. Being employed de- creased women's psychological distress directly in both cultures. Parry & Shapiro (6) found that in the case of working-class women, working outside the home was associated with less depression where there was good social support, but more depres- sion where good support was not available. Russo (2) in her review noted that parenthood, particularly when children are young, increases the symptoms of psychological distress for women whether or not they work outside the home and these symptoms appear to increase with the num- ber of children living in the home. Whether or not employment brings mental health costs or benefits to women depends substantially on their satisfac- tion with and their husband's attitudes toward child care. For employed mothers, if child care was accessible and husbands shared in it, depression rates were low. In contrast, employed mothers without accessible child care and with sole respon- sibility for childcare had extremely high depres- sion levels. For non-employed wives, children in- creased depression levels. Rosenfield (7) proposed that role overload causes greater symptoms for the same reasons as low power, i.e., through lowering an individual's sense of personal control. Thus employment may trade one source of low control for another. Personal control may also explain dif- ferences in symptoms among social classes. Depression, the most prevalent psychiatric con- dition, is the most frequently encountered wom- en's health problem in many western countries. Depression may vary imperceptibly through sub- clinical distress to a normal mood which is part of universal human experience. Defining a threshold for the disorder and separating this from normal experience has been important. Community preva- lence studies indicate that about 5% of the popula- tion satisfy the criteria for psychiatric depression in a &month period. Weissman & Klerman (8) report- ed that sex ratios in treated cases of depression in Wld hlth statist. quart., 46 (1993) western cultures show a 2: 1 predominance of wom- en to men.Jorm (9), using a qualitative synthesis of published prevalence data, shows that the gender differences in depression are age-specific, and also a function of social situations. In children and the very old there is little gender difference, but in at ages 15-50, depression is more prevalent in wom- en. Reasons for gender differences in depression advanced by Paykel ( 10) include the following: (i) Although both sexes suffer equally from de- pression, women may more often seek help for the condition. However, in community sur- veys, in terms of specific depressive symptoms and anxiety, females predominate. (ii) Given that there is a difference, does it reflect biological factors such as chromosomal factors or female sex hormones? There is no good evidence for X-linkage in mood disorders and hormonal factors have so far not been identi- fied to explain disorders linked to the repro- ductive cycle such as premenstrual tension and postpartum depression. (iii) Another explanation is based on the social effects of life-event stress, of social vulnerabili- ty factors and absence of support, and of wom- en's role in society. Community studies sug- gest that women react to the same stress with higher intensities of symptoms. (iv) A final explanation for differences in depres- sion rates may reflect a difference in acknowl- edgment and direction of distress, e.g. "'Depressive-spectrum' disease" in which women tend to have depression, while male family members have alcoholism and antiso- cial personality. The Old Order Amish of Pennsylvania, a religious group with strong prohibitions on drink and antisocial behav- iour have high rates of affective disorder and show approximately equal gender incidence. Recent western community-based studies re- peated 10 to 20 years apart show a trend for the sex ratios to equalize (11). This may reflect societal changes with less gender role differentiation, which might change the acknowledgment and the social stress. No major sex differences were found in certain occupational groups. These trends were evident in studies from the United States, Sweden, Germany, Canada and New Zealand, but not in those from Korea and Puerto Rico (12). Summary Both community-based studies and studies of treatment seekers indicate that women are disproportionately af- fected by mental health problems and that their vulner- ability is closely associated with marital status, employ- ment and roles in society. 235 Women's mental health cannot be considered in isola- tion from social, political and economic issues. When women's position in society is examined, it is clear that there are sufficient causes in current social arrange- ments to account for the surfeit of depression and anxiety experienced by women. Resume Determinants psychosociaux de la sante mentale des femmes Les etudes conduites au sein d'une collectivite ou parmi les patientes en quete de traitement montrent qu'un nombre exceptionnellement eleve de femmes sont de- mesurement touchees par les problemes de sante men- tale et que leur vulnerabilite est etroitement liee a leur statut marital, a leur travail et au role qu'elles assument dans la societe. L'etude de la sante mentale des femmes ne saurait etre dissociee de celle des problemes sociaux, politiques et economiques auxquels elles sont confrontees. Lorsque l'on examine la situation des femmes dans la societe, on constate que !'infrastructure sociale actuelle fournit des raisons en quantite amplement suffisante pour expli- quer la depression et l'anxiete qui affectent trap souvent les femmes. 236 References/References 1. Russo, N.F. A Woman's Mental Health Agenda. Washington DC. American Psychological Association, 1986. 2. Russo, N.F. Overview: Forging research priorities for women's mental health. American psychologi.st, 45: 368-373 (1990). 3. Gove, W.R. &:Tudor,J.F. Adult sex roles and mental illness. American journal of sociology, 78: 812-835 (1973). 4. Ibrahim, M.A. The changing health state of women. Americanjoumalofpuhlichealth, 70: 120-121 (1980). 5. Ross, C.E. et al. Distress and the traditional female role: a comparison of Mexicans and Anglos. American journal of sociology, 89: 670-682 ( 1983). 6. Pany, G. &: Shapiro, D.A. Social support and life events in working class women. Archives of general psychiatry, 43: 315-323 (1986). 7. Rosenfield, S. The effects of women's employmnent: personal control and sex differences in mental health. Journal of health & social behaviour, 30: 77-91 (1989). 8. Weissman, M.M. &: Klerman, G.L Sex differences and the epidemiology of depression. Archives of general psychiatry, 34: 98-111 (1977). 9. Jonn, A.F. Sex and age differences in depression: a quantitative synthesis of published research. Australia and New 7.eal.and journal of psychiatry, 21: 46-53 ( 1987). 10. Paykel, E.S. Depression in Women. British journal of psychiatry, 158 (Supplement 10): 22-29 (1991). 11. Murphy,J.M. Trends in depression and anxiety: men and women. Acta psychiatrica scandinavica, 73:113-27 (1986). 12. Klennan, G.L. &: Weissman, M.W. Increasing rates of depression.JAMA, 261: 2229-35 (1989). Rapp. trimest. statist. sanit. mond., 46 (1993) Family health and the use of psychoactive substances Lee-Nah Hsua Introduction An individual's inappropriate or non-medical use of psychoactive substances affects his or her primary relationships, particularly the family, as well as society in general. Substance misuse is often a result of underlying family-related problems, therefore, effective preventive interventions of sub- stance abuse can start with the family. This article analyzes: (i) drug and alcohol use by parents, spouses and children and (ii) the role of the family in preventing psychoactive substance abuse. The data used in this article come from countries in different regions of the world, in order to reflect diverse sociocultural and economic envi- ronments. As psychoactive substances, we include illicit, licit or controlled drugs, alcohol, and other substances such as glue and organic solvents. Children with substance-abusing parents Fetal defects have been linked to parental use of alcohol, cocaine, heroin, cannabis, methadone and combinations of multiple drugs, and to the related socioeconomic environment (1,2,3). Some effects of parental drug use on the fetus are not apparent until the child reaches four to six years of age (4). One European study reported that 87% of the Human Immunodeficiency Virus (HIV)-sero- positive children were born to HIV-infected moth- ers who used injecting drugs (5). In addition, chil- dren born to substance-abusing parents have high- er rates of morbidity, mortality, accidents and inju- ries when compared to children whose parents do not abuse substances (6). Substance-abusing parents may divert their fi- nancial resources from household food needs and children's educational expenses to support their own drug use. These parents are often not able to provide proper care and nurturing for their chil- dren, resulting in childhood malnutrition, health problems and child neglect (7). Many children with substance-abusing parents are found to manifest antisocial personalities, ag- gressive behaviour, difficulties in adjusting and achieving in school, and involvement in delin- quent behaviours, including early initiation into psychoactive substance use (8). Psychosomatic symptoms such as sleep disturbances, headaches or a Short-term consultant, Programme on Substance Abuse, World Health Organization, Geneva, Switzerland. Wld hlth statist. quart., 46 (1993) stomach aches were also reported among these children. Some of these children may experience frequent changes of residence, and may be cared for by a single parent, separated from their own parents, or placed in institutions. Studies have re- ported that up to 48% of infants born to heroin- using parents were living with substitute parents and by preschool age 91 % were living with some- one other than their own parents (7, 9). Spouses with substance-abusing partners The use of psychoactive substances affects the equi- librium in the household relationship. This dis- equilibrium includes the inability of substance abusers to function in their expected roles within a household; to participate in family activities; to perform at work; or to keep up with personal hy- giene and appearance. The substance-abusing spouse may also contribute to aggressive interac- tions within the family. The likelihood of separa- tion or divorce increases when one partner has a substance-abuse problem (10). Tabk 1 gives some examples of data from coun- tries in different regions of the world. These data show that among adults, there are more male than female users of alcohol and illicit drugs. However, more females than males misuse pain medications, sedatives or tranquillizers. Many women whose spouses use psychoactive substances are responsible for the financial sup- port of the entire household while the male spouse is unable to contribute to the family either finan- cially or psychologically (11). Many women with substance-abusing partners either get into sub- stance abuse themselves or resort to using pain medications, tranquillizers or sedatives to cope with the tension in the relationship, real or poten- tial domestic violence, or the stress of seeking an income while caring for the household. A North American study reported that 48-78% of domestic violence against women involved a male perpetrator with concurrent alcohol use ( 12). Although data from developing countries are limit- ed, the societal perception of male and female roles in a family relationship, increasing rural-to- urban migration, and the accompanying changes in family structure may contribute to links between substance abuse and domestic violence. For exam- ple, a report on Central America prepared for the World Health Organization found that 97% of do- mestic violence involved a male perpetrator who 237 Table 1 Examples of country statistics on psychoactive substance use among adulta males and females Tableau 1 Exemples de statistiques par pays sur la toxicomanie chez les adultesa hommes et femmes Country Pays Zimbabwe Japan - Japon Poland - Pologne Estonia - Estonie Brazil - Bresil Sri Lanka Type of substance Type de substance sleeping pills - somniferes alcohol - alcool opiates - opiaces polydrug use - usage de plusieurs drogues inhalants - produits par inhalation sedatives - sedatifs psychotropics - psychotropes heroin - heroine Year An nee 1992 1984 1990 1993 1993 1993 Male:female Other substances increasing in use ratio Types de toxicomanie en augmentation Rapport homme:femme 1 :7 beer- biere 1.5:1 analgesics - analgesiques 1:1 opiates - opiaces 24:1 cannabis sedatives - sedatifs 9:1 2.5:1 1 :2 crack mixed with coca paste - crack melange avec de la pate de coca psychotropics - psychotropes 32:1 heroin - hero'ine Source: Hsu, L·N. Compilation of data from country reports of the women and drug abuse project at WHO Programme on Substance Abuse, WHO/PSA/93.13, Geneva, Switzerland, 1993 Hsu, L ·N: Compilation de donnees a partir des rapports par pays du projet sur les femmes et la toxicomanie, elabore par le programme de l'OMS de lutte contre les toxicomanies, OMS/PSA/93.13, Geneve, Suisse, 1993. •Adults include those over the age of 20 - Entrent dans la categorie adulte les personnes agees de 20 ans et plus. abused substances. Recent studies found that drug use, or combined use of drugs and alcohol, is asso- ciated with the more extreme incidents of domes- tic violence ( 13). Regular or periodic use of psychoactive sub- stances does not necessarily lead to marital prob- lems. For example, small amounts of alcohol may assist in reducing tension in an interaction and thus facilitate communication ( 14). However, chronic or excessive drinking is associated with sexual dysfunction in males (15) and with repro- ductive system dysfunction in females (16). These associated dysfunctions have potential implications for the family. In addition, persons at greatest risk of HIV in- fection through heterosexual contacts are the part- ners or spouses of seropositive drug users. One North American study reported that HIV seroprev- alence is up to 38% among women who are not drug users but who are sexual partners of male injecting-drug users (IDUs) (17). Adolescent use of psychoactive substances Often, a child's initial exposure to psychoactive substance use is at home. The adult example of drug use, the availability, the curiosity for experi- mentation and peer pressure can contribute to subsequent use by adolescents. Tabl.e 2 gives exam- ples of statistics from many countries reporting youth experimenting with a wide range of both licit and illicit substances. The countries which previ- 238 ously were transit routes for drug traffic are now experiencing an increase in drug consumption, especially among young people. Certain societies have continued to use the psy- choactive substances that were traditionally used in their culture, for example, the use of opium, home-brewed alcoholic beverages, hashish, coca- leaf, khat and betel nuts. However, the pattern of use and mode of administration may be changing. In developed countries, psychoactive substance use among youth became popular in the 1960s and gradually extended to adults and now includes fashionable drugs such as Ecstasy. Psychoactive substance use has spread across socioeconomic, cultural and gender lines. Today young women are becoming as active as their male peers in this drug use trend (18). For example, studies found no gen- der difference in the likelihood of cocaine use among adolescents (19). Twenty percent of fatal automobile accidents in one North American city involved drivers who test- ed positive for cocaine metabolites. Among these cocaine-using drivers, many were adolescent males who were also under the influence of alcohol (20). Psychoactive substance use among adolescents is also related to a greater frequency of accidents and suicides and homicides, both attempted and completed. These problems experienced by youths in developed countries are becoming increasingly common among youth in developing countries where there is rapid social change and increased migration into cities (21,22). Rapp. trimest. statist. sanit. mond., 46 (1993) Table 2 Examples of country statistics on psychoactive substance use among youtha, both male and female Tableau 2 Exemples de statistiques par pays sur la toxicomanie chez les jeunesa, hommes et femmes Country Pays Egypt - Egypte Japan - Japan Honduras Greece- Grace United Rep. of Tanzania Rep.-Unie de Tanzanie Type of substance Year Type de substance An nee hashish - haschisch 1990 tranquillizers - tranquillisants amphetamines - amphetamine 1985 alcohol 1992 licit drugs - drogues licites 1988 illicit drugs - drogues illicites stimulants 1992 Male:female ratio Rapport homme:femme 9:0 5:3 1:2 3:1 1 :2 4:1 3:1 Other substances increasing in use Types de toxicomanie en augmentation heroin (injecting) - injection d'hero"ine amphetamine (injecting) - injection d'amphetamine psychotropic drugs - psychotropes inhalants - produits par inhalation amphetamines - amphetamine cannabis cocaine - coca"ine plastic-cement smoke inhaling - inhalation d'emanations de colle forte a prise rapide cocaine - coca"ine cannabis heroin - hero"ine psychotropics - psychotropes cannabis heroin- hero"ine Source: Hsu, L-N. Compiled from country report data for the women and drug abuse project. WHO Programme on Substance Abuse, WHO/PSA/93.13, Geneva, Switzerland, 1993 Hsu, L-N. Compilation de donnees a partir des rapports par pays du projet sur les femmes et la toxicomanie elabores par le programme de l'OMS de lutte contre les toxicomanies, OMS/PSA/93.13, Geneve, Suisse, 1993. a Youth includes those 13 to 24 years of age - Entrent dans cette categorie les jeunes ages de 13 a 24 ans. Furthermore, teenage substance abusers are as- sociated with unprotected sexual practices which can result in unplanned pregnancies and exposure to sexually transmitted diseases, including HIV. It is estimated that one-half of all teenage pregnan- cies occur when one or both partners are using drugs or alcohol (23). In one North American city, the number of 13-21-year-old females infected with HIV has risen by 77% between 1991 and 1993 (24). Common factors Some common factors associated with the effects of substance abuse on family health are summa- rized in Tabk 3. Discussion The movement from traditional to modified or new patterns of psychoactive substance use implies the ineffectiveness of regulatory control and a weakening of social norms. Rural-to-urban migra- tion and the subsequent changes in family struc- tures, along with conditions of poverty in both the developed and the developing countries also con- tribute to substance use. Today, many adolescents and adults are increasingly exposed to lifestyles, occupations and activities related to substance use. This exposure is enhanced with the increasing Wld hlth statist quart., 46 (1993) availability of psychoactive substances and the re- laxing of traditional forms of community and fami- ly bonds. Although the number of new users is greater among youth, use of psychoactive substanc- es may continue at later ages, with possible changes in the pattern and types of substances used. Resources and social support networks are nec- essary for families to effectively adapt to and cope with traumatic experiences or hardships, so as to avoid any member of the family resorting to sub- stance abuse. These resources may, to a certain extent, depend on the strength oflocal community networks and family ties. If social tolerance exists for the particular form of substance abuse, the family may still be able to benefit from external support. If a particular abused substance is not approved by the society, the abuser- especially if he or she becomes HIV-infected - together with the family, may face difficulty in continued acceptance by the community. Nutritional status often deteriorates among drug users because of their lack of attention to proper food intake, and among children with drug-abusing parents, because of the diversion of family income normally used for food purchase to drug purchase. The association between drug use and a family's nutritional status is an area that needs further investigation. 239 Table 3 Factors associated with substance abuse in the family Tableau 3 Facteurs de risque associes a la toxicomanie au sein de la famille Prenatal-neonatal period Childhood-adolescence period - Periode prenatale-neonatale - Periode de l'enfance et de !'adolescence • difficulty in pregnancy and fetal development • malnutrition, illnesses and intoxication from substances - difficulte au cours de la grossesse et repercussion sur - malnutrition, maladies et intoxication imputables aux le developpement du foetus. droques • maternal to child transmission of HIV and other • risks of accidents and injuries infectious diseases - risques d'accidents et de blessures - transmission par la mere a l'enfant du VIH et • weakened parent-child bonding d'autres maladies infectieuses - affaiblissement du lien parents-enfants • inadequate prenatal care • child developmental problems - insuffisance de soins pendant la grossesse - problemes de developpement chez l'enfant • premature birth and low birth weight • parent-child separation - naissance prematuree et insuffisance de poids a la - separation parents-enfants naissance • physical and psychological abuse • mental retardation and physical malformations - abus physiques et psychiques - arrieration mentale et malformations physiques • educational problems • neonate drug-withdrawal syndrome - problemes d'education - syndrome du sevrage de drogue chez le nouveau-ne • social maladjustment - inadaptation sociale • risk of becoming substance users - risque de devenir toxicomane Partner-spouse relationship Family unit - Relation partenaire/epoux - Cellule familiale • stress to mental and physical health • employment or financial difficulties - stress pour la sante mentale et physique - difficultes d'emploi ou difficultes financieres • marital instability • cost of substance consumption and cost for health care - instabilite du couple - coot engendre par la consommation de drogues et • separation, divorce coot des soins de sante - separation, divorce • social marginalization, frequent residential moves • interpersonal violence - marginalisation sociale, changements de residence - violence interpersonnelle frequents • risk of sexually transmitted diseases • family break-down - risque de maladies sexuellement transmissibles - dislocation de la famille • difficulties in coping - difficultes a faire face aux problemes • risk of becoming substance users - risque de devenir toxicomane Source:Tawil, 0. Psychosocial implications of drug abuse on family health, WHO/PSN93.6, Geneva, Switzerland, 1993 Tawil, 0. Implications psychosociales de la toxicomanie sur la sante de la famille, OMS/PSN93.6, Geneve, Suisse, 1993 Summary Substance abuse is spreading to countries previously unaffected, to all age groups, including the elderly, and to women: the potential impact of drug abuse on the family is therefore increasing while at the same time, the family can play a crucial role in prevention, treatment and rehabilitation. The family is the arena for learning healthy behaviour, including ways of handling problems associated with 240 psychoactive substance use. The ability of family mem- bers to cope with life events is influenced by the avail- able behavioural role models, the quality of communica- tions among family members and the options a family has within the community. Strengthening the social network and families' coping mechanisms may reduce the demand for and the consequences of substance abuse by family members. Rapp. trimest. statist. sanit. mond., 46 (1993) Resume et conclusion Sante de la tamille et usage de substances psychaactives Les toxicomanies s'etendent actuellement a des pays jusqu'ici epargnes; elles atteignent tous les groupes d'age, y compris les personnes ages et les femmes: la famille risque done de plus en plus d'etre touchee alors meme qu'elle peut jouer un r61e capital en matiere de prevention, de traitement et de readaptation. La famille est le cadre ou s'acquierent les comporte- ments propices a la sante et notamment les fac;:ons de gerer les problemes lies a l'abus des drogues. L'aptitu- de des membres de la famille a faire face aux differentes circonstances de la vie depend des modeles de r61es comportementaux disponibles, de la qualite des com- munications existant entre eux ainsi que des options qui leur sont offertes au sein de la collectivite. En renforc;:ant le maillage social et les mecanismes de reaction des families, on pourra vraisemblablement reduire la de- mande mais aussi les consequences de l'usage des drogues pour les membres de la famille. References/References 1. Gingras,J. etal. Cocaine and development: mechanisms of fetal toxicity and neonatal consequences of prenatal cocaine exposure, Early human devtlbpfllfflt, 31: 1-24 (1992) .. 2. Bongain, A. et al. Drug addiction and pregnancy, Rernu du practicien, 42(8): 1004-9 (1992). 3. Johnson, H. 8c Rosen, T. Mother-infant interaction in a multirisk population, American j01J.mal of orthopsychiatry, 60(2): 281-288 (1990). 4. Feng, T. Substance abuse in pregnancy, Current opinion in obstetrics andgynecowgy, 5(1): 16-23 (1993). 5. Sanchez.Ruiz, E. et al. Vertical transmission of HIV: descriptive epidemiology, risk factors and survival, Anales espanolesd8pediatria, 37(5): 367-71 (1992). 6. MacDonald, D. Cocaine leads ED drug visits, JAMA, 258: 2029 (1987). 7. Casado-Flores, J. et al. Social and medical problems in children of heroin-addicted parents: a study of 75 patients, Wld hlth statist. quart, 46 (1993) American journal of diseases of chudren, 144: 977-9 ( 1990). 8. Drake, R. 8c Vaiilant, G. Predicting alcoholism and personality disorder in a 33-year longitudinal study of children of alcoholics, British j01J.mal of addiction, 83: 799-807 (1988). 9. Wilson, G. Clinical studies of infants and children exposed prenatally to heroin, in: Hutchings, E. (ed.), Prenatal abuse of licit and illicit drugs, New York, New York Academy of Sciences, 1989. 10. Edwards, G. &Arif, A. Drugproblmns in the sociocultural context: a basis f<W policies and programme planning, World Health Organization, Geneva, 1980. 11. Holmila, M. Young families and alcohol use in Finland and the Soviet Union, Contempqrary drug problmns, 14: 649-72 (1987). 12. Appleton, W. The battered woman syndrome, Annalr of emergmcy medicine, 9: 84-91 (1980). 13. Browne A. When battered women It.ill, New York, The Free Press, 1987. 14. Dunn, N. et al. Marital stability in alcoholic-spouse relationships as a function of drinking pattern and location, J01J.mal of abn<Wmal psychowgy 96 (2): 99-107 ( 1987). 15. Fahrner, E-M. Sexual dysfunction in male alcohol addicts: prevalence and treatment, Archives of sexual behaviour, 16(3): 247-57 (1987). 16. Gavaler,J. 8c Van Theil, D. Reproductive consequences of alcohol abuse: males and females compared and contrasted, Mutation research, 186: 269-277 ( 1987). 17. Friedman, S.R. et al. Drug injectors and heterosexual AIDS, in Sherr, Lorraine (ed.), AIDS and the heterosexual population, Chur, Switzerland, Harwood Academic Publishers, 1993, pp 41-65. 18. Henderson, S. Fun, fashion and frisson: incompatible with existing perspectives on women and drug use? International journal of drug policy, 4(3) (1993). 19. Marzuk1 P.M. etal. Prevalence ofrecentcocaine use among motor vehicle fatalities in New York City,JAMA 263: 250-256 (1990). 20. Crumley, F.E. Substance abuse and adolescent suicide behaviour,JAMA, 263: 3051-3056 (1990). 21. Blum, R.W. Contemporary threats to adolescent health in the United States,JAMA, 257: 3390-3395 (1987). 22. Blum, R. W. Global trends in adolescent health, JAMA, 265: 2711-1729 (1991). 23. Burke, T. The economic impact of alcohol abuse and alcoholism, Public health report, 193(6): 564-8 (1988). 24. Hein, K. Invisible epidemic now becomes visible as HIV/ AIDS pandemic reaches adolescents.JAMA, 270:16 (1993). 241 Surveillance for equity in maternal care in Zimbabwe Carl Taylor.a David Sanders,b Mary Bassetfb & Stella Goingsa Introduction After independence, Zimbabwe declared equity to be the central principle of the first health plan which was titled Planning for equity in health care (1). Zimbabwe's experience in trying to imple- ment this goal is instructive for other countries, especially because of the recent impact of financial constraints imposed by structural adjustment poli- cies in the past decade. The lessons learned help in defining a new approach to surveillance for equity in maternal care which only recently has been rec- ognized as one of the important challenges for primary health care and family health. The concept of surveillance for equity in maternal care as a practical management tool The standard approach in efforts to reduce mortal- ity and morbidity from pregnancy-related prob- lems has been high-risk monitoring. All women should be encouraged to start a series of visits to health facilities early in pregnancy to identify com- plications or indicators of potential risk. Early de- tection of potential problems should initiate pre- ventive measures, including making sure that the woman reaches a health facility which can provide whatever level of care her condition requires at the appropriate time. This system works well where health care cover- age is relatively complete and health personnel and facilities are functioning well. It can reduce expenses because it provides for those with identifi- able problems to get care appropriate to their needs while the vast majority of women who have uncomplicated deliveries can be cared for with less expensive arrangements. This has proved difficult: around the world, 42% of deliveries each year, or 56 million pregnancies, are still without a trained attendantc. The coverage gap is greatest among the most needy. The proposed surveillance for equity has sever- al distinguishing characteristics: a Johns Hopkins Institute of International Programs, The Johns Hopkins University, Baltimore, MD, United States of America. b Department of Community Medicine, UniversityofZimbabwe School of Medicine, Harare, Zimbabwe. c World Health Organization. Coverage of maternity care (Docu- ment WHO/FHE/89.2) 1989. (English only - Anglais seulement) 242 ( a) A critical difference between surveillance for equity and high-risk pregnancy monitoring (2) is that surveillance is population-based rather than focused on individuals. High-risk monitor- ing requires infrastructure to reach pregnant women. The two approaches are complementa- ry with surveillance for equity being most need- ed where infrastructure is limited. (b) Population-based surveillance can sample pop- ulations in order to identify subgroups with the greatest concentration of problems. These sub- groups can then be the focus of intensive high- risk monitoring. ( c) The equity principle contributes directly to health improvement (3). Equality of access is not enough. Inequity can be measured by de- finable clusters of health and social factors which tend to be synergistic. ( d) Surveillance is also distinguished from monitor- ing because it should trigger community action, using standardized interventions to correct problems. (e) A current dilemma is that many health systems are turning to self-financing rather than public- ly supported primary health care. This aggra- vates the unfortunate tendency of health sys- tems to try to maximize efficiency by focusing care on families who are readily accessible, edu- cated and affluent enough to pay. (j) Surveillance for equity provides information to help define local priorities, corrective actions and indicators. It goes beyond the biomedical indicators of high-risk pregnancy monitoring (2) to include a concern for underlying socio- economic variables. Such combinations of equi- ty indicators tend to make sense to community members. (g) The health system should empower the com- munity to solve its own problems especially in applying interventions that depend on behav- ioural change and community action. Surveil- lance should combine bottom-up and top-down approaches in order to facilitate dialogue and action in a balanced relationship between health personnel and community members. (h) Public acceptance of the need for equity re- quires fundamental changes in attitudes and values. The most extreme disparities between groups will be difficult to change because of deeply ingrained cultural practices and tradi- tions. Rapp. trimest. statist. sanit. mond., 46 (1993) (i) When problems arise for which there is no known solution, a surveillance system can help identify questions and approaches for field re- search. In situations where resources are limit- ed, the appropriateness, acceptability and cost/ effectiveness of alternative interventions need to be tested. (j) A good way of promoting local problem-solving is through a continuing relationship between health services and a local academic centre. A Longitudinal Area-Based Study Site (LABSS) can serve as a laboratory for successive testing of alternative options. Practical steps for implementation The process of establishing surveillance for equity adapted to local conditions should start with situa- tion analysis and priority setting. Policies can be set centrally about what needs to be done and adapta- tions can be made locally about how those policies will be implemented. This analysis should balance the knowledge, understanding and potential of local people with the objectivity and facts provided by the health system. Where data are not available simple rapid assessment surveys (RAP) (4) and rap- id evaluation methods (REM)d can fill specific gaps. Agreement on priorities by both health sys- tem and community representatives should lead to continuing dialogue that makes primary health care responsive and effective. Only a few carefully selected indicators should be used in surveillance. They should relate directly to control measures so that understanding of caus- al linkages does not get lost in intervening vari- ables. Each indicator should be measurable at min- imal cost in money and time. They should be dis- crete, readily analyzed and easily understood by the people. A menu of possible indicators can help district and community decision-makers think through their alternatives and concerns. A spec- trum is desirable which should include: input indi- cators to measure equity in access; process indica- tors to quantify coverage and utilization so as to facilitate rapid feedback for change; and outcome indicators directly measuring the impact of servic- es. If only input and output indicators are used they tend to produce a conceptual conflict be- tween equity and efficiency. Individuals and fami- lies that are most efficiently reached and con- vinced are seldom those with the greatest needs. Utilization may achieve high levels statistically but, even with two-thirds to three-fourths coverage, the population segments missed may include most of the priority problems. d World Health Organization. Rapid evaluation method guidelines fur maternal and chi/,d health, family pl.anning and other health servir.es (Document WHO/MCH-FPP/MEP/93.1) 1993 (English only, French in preparation - Anglais seulement, fram;ais en preparation.) Wld hlth statist. quart., 46 (1993) Simplicity and speed in analysis and feedback reporting are more important for management purposes than the detail and precision of data needed for research. Only the most necessary items of data should be gathered through routine reporting because it is easy to overload the infor- mation system. For the highest-priority problems in an area, surveillance can help standardize responses. Devia- tions from expected trends in indicators or local- ized high prevalence should trigger action. The number of interventions implemented at one time should be limited and introduced in a phased se- quence. Specific interventions can be integrated into primary health care with much more attention to building infrastructure than has been the practice in targeted, high-intensity campaigns. For long- term sustainability a balance must be maintained in the roles of the health system and community. Technical expertise must come from health work- ers, but the methods and reporting should be kept simple enough to give community members a sense of ownership and involvement in the process especially in activities that require behavioural change. A particular strength of community participa- tion is that local people understand how to com- municate with and convince their neighbours, es- pecially the difficult-to-reach groups who must be included to achieve equity. Choice among surveillance indicators for maternal care A tentative list of potential surveillance indicators for maternal care is provided, including measur- able items from which local health units and com- munities can choose those that are most suitable for local conditions ( Box 1). The items are grouped under three headings: biomedical, socioeconomic and services. Choices can be made about which sources of data will be most feasible and outlined in a matrix with two columns: data gathered in peri- odic surveys and data gathered by routine, continu- ous reporting. Background information on Zimbabwe's problems and priorities in maternal and child health care In pre-independence Zimbabwe the burden of death and disease fell with particular inequity on infants, young children and women. The inequity of the health care system is evident in great dispar- ities in mortality by race, socioeconomic status and geographical area. At the time of independence in 1980 the ratio of infant mortality rates among whites, urban blacks and rural blacks was 1:3.5:10. The relationship in per capita income for the same three groups was the reverse: 39:5: 1. The ratio for 243 Box 1 Potential indicators for surveillance for equity in maternal care in Zimbabwe Biomedical indicators • maternal deaths and causes of death • parity and number of living children • previous pregnancy history • antenatal risk factors • weight and height of mothers • haemoglobin and nutritional status • infection rates for HIV, STDs, malaria, etc. Socioeconomic indicators • income • housing • education of women • nutrition Availability and use of services • coverage with antenatal care • coverage with trained midwifery care • time and convenience of referral care • family planning the mean cost of all health care received was 36:8: 1 (5). The infant mortality rate (IMR) for whites was 17 per 1 OOO live births, but for rural and urban blacks combined, it was 120 per 1 OOO live births with much underreporting. Mortality risk for women was especially great in relation to child-bearing. Maternal mortality was estimated to be 145 per 100 OOO live births at the time ofindependence.e In 1988 it was estimated to have dropped to 76 for facility-based deliveries with a range from 142 in Manicaland province to 49 in Masvingo province for hospital deliveries. Howev- c Minisay of Health, MCH annual report for 1987 (mirneo), Harare, Ministry of Health, 1987. 244 Routine reporting Surveys er, a community-based study in Masvingo in 1989 (personal communication from S. Fawcus and Mbizvo) indicated that overall maternal mortality was probably about 120 per 100 OOO live births. HIV prevalence is 7% in urban blood donors and is an increasingly important cause of maternal death. Child health problems continue to be severe although improvement is suggested by a reported decline in IMR (presumably mainly in urban facili- ties) from 83 per 1 OOO live births in the 1982 census, to 73 in 1987 (6) and 53 in 1988. However, using Brass techniques several studies have shown results that range between 81-88 per 1 OOO live births up to the 1990s. Disparities continue with urban IMR in 1984 estimated to be half that in rural areas. HIV has become the first cause of Rapp. trimest statist. sanit. mond., 46 (1993) death among children in urban hospitals. Al- though severe malnutrition seems to have de- clined, stunting is still extremely common especial- ly among the rural poor (7-8). Efforts to reduce disparities in maternal care in Zimbabwe Implementation of the policy of equity has been mixed because of severe constraints imposed by the pattern of health care left by the colonial ad- ministration. The sophisticated hospital- and spe- cialty-oriented health system was designed to pro- vide colonials with health care equivalent to what they would have had in their home country. Hospi- tals have continued to absorb most of the available resources, particularly because of a general policy of not disrupting prior economic arrangements, government-financed urban hospitals and a medi- cal school that provides sophisticated clinical train- ing. With independence, Zimbabwe undertook vig- orous promotion of primary health care. However, after a brief post-independence boom, severe eco- nomic problems were precipitated by the interna- tional recession, a prolonged drought, and exter- nal military destabilization which hampered all so- cial development. The recent imposition of IMF (International Monetary Fund) economic-adjust- ment policies has resulted in further severe cuts to all social services in the already depressed econo- my. The negative impact on health care has been aggravated by the loss of up to one-third of Zim- babwe's health professionals to South Africa. In the first few years after independence, the small proportion of resources devoted to starting new patterns of care for the 90% of the population in rural areas was sufficient to begin the process of serving those in greatest need. Hospitals in all dis- tricts were designated as district hospitals and a beginning was made in providing coverage through health centres and mass public health programmes. By 1987 the government had built 274 health centres in rural areas, bringing such centres within reach of the majority of the rural population. ( Tabk 1) Little reallocation of resources has occurred in the proportion of government funding for curative as compared with preventive services. In 1980 the figures were 89.5% for curative care as compared with 6. 7% for prevention, in 1982 the ratios had shifted to 81.5% and 13.1 %, and after that they stabilized at 82% and 14% (9). The minimal shift in patterns of funding is also shown by the proportion of grants going to mission hospitals which provide 64% of rural hospital beds, compared with the main teaching hospital in Harare. The proportion of the teaching-hospital grant compared to the total mission grant was 97.1 % in 1980, 48.1 % in 1982 and 76.1 % in 1989 (10). Wld hlth statist quart., 46 (1993) The large urban hospitals are supposed to serve as referral centres for the rural areas. However, the lack of outreach from tertiary-care hospitals is shown by a 1986 study in one district ( 11) (Tabk 2). The percentages of support for primary care as compared with secondary care were: urban black hospital 11 % vs. 67.8%, and urban elite hospital 50.9% vs. 42.4%. The pattern and severity of illness seen at the three levels was remarkably similar. The outcome in terms of survival did not vary but hospi- tal stay was longer at central hospitals. The daily unit costs were: district hospital 2$13.69, provincial hospital 2$33.66, urban black hospital Z$38.67 and urban elite hospital Z$92.26. Disparities were most marked in term of whether mothers had ac- cess to antenatal care and had been "booked" for delivery. Booked deliveries had perinatal mortality of 25.3 per 10 OOO as compared with 142.5 for the unhooked (12). A major factor in producing ineq- uitable use of health services is rapidly rising costs of care. One example is the sharp decline in num- Table 1 Population with access to a hospital or a health clinic Tableau 1 Population beneficiant de soins dans un hopital ou un dispensaire Within 8 km Over 30 km Area and facility Zones et installations Situees dans un Situees dans un rayon rayon de 8 km superieur a 30 km (%) (%) Rural - Milieu rural Hospital - Hopital Health clinic - Dispensaire Urban - Milieu urbain Hospital - Hopital Health clinic - Dispensaire Source: Ref. Ref. (17) Table 2 12.4 69.8 49.6 91.5 66.8 14.5 12.7 7.2 Hospitalized patients: distance from residence to hospital Tableau 2 Patients hospitalises : distance entre leur domicile et l'hopital Hospital - Hapital Within 1 O km - Dans un rayon de 10 km (%) District - District 32.6 Provincial - Province 36.7 Urban Black - Milieu urbain, 59.3 population noire Urban elite - Milieu urbain, 54.6 elite Source: Ref. Ref. (13) Over 31 km - Oans un rayon superieur a 31 km (%) 35.0 38.5 30.5 25.8 245 hers of normal deliveries in Harare health centres after the maternity fee was sharply increased in 1988. The number of health centre deliveries had increased from 12 439 in 1980 to about 17 500 between 1981 and 1987, but after a fee increase in 1988 the number fell to 14 225. The influence of cost on care is especially marked in the shift to private medical care for the urban population. Although nominally pro- vided by medical aid societies, private care contin- ues to receive hidden subsidies from the public sector. A conservative estimate, calculated from medical aid society payments to health care pro- viders, is that private patients have at least ten times the amount spent per capita as patients in the public sector ( 10). Manpower distribution also reflects efforts to correct earlier inequities but with only partial success. A great increase oc- curred in all categories of health workers. Most important was the training of over 7 OOO village health workersf ( 13). Curricula were reoriented toward health promotion and preventive care, emphasizing oral rehydration, growth monitoring and antenatal care. The numbers of doctors grad- uated increased by 40% but the number of doc- tors in public service increased by only 10% be- tween 1981 and 1989. About 60% of doctors are in private practice. Less than 15% of local gradu- ate physicians were in government service 5 years after graduation.g Most of the district and provin- cial posts are still filled by expatriate doctors. In 1983 doctors were found at the central (67%), provincial (15%), and district/mission levels (15%). By 1988 they were at: central level, 72%; provincial, 12%; and district/mission, 16%. Start- ing in 1980 new medical graduates were bonded to serve for several years in rural areas but in 1988 junior doctors had a strike which successful- ly changed this policy. The differentials found in physician distribution are reflected in all other categories of health manpower. Improving the proportion of deliveries by trained personnel has been stressed, achieving over 90% coverage in 1988 (10). The ratios are: urban coverage 96%, rural communal areas 90% and large-scale-farming areas 94%. About 15% of mothers seek antenatal care during the third tri- mester of pregnancy. Deliveries in health facilities are: urban 92%, communal areas 64% and large- scale farming areas 56%. Low-birth-weight babies were only 5.1 % of hospital deliveries but communi- ty surveys showed figures of 10-11 % suggesting that those who delivered at home were poor and mal- nourished women. f Ministry of Health. Repurt on PHC/MCH/EPI surue:ys (mimeo), Harare, Ministry of Health, October 1988. g University of Zimbabwe Medical Faculty. Data on graduates 1979-1989 (mimeo), Harare, Medical Faculty, 1989. 246 Some primary health care programs have been successful. For instance, the expanded program on immunization (EPI) increased coverage from 25% in 1982 to 42% in 1984 and by 1986 to between 50 and 80% in various parts of the country (14) and eventually to over 80% generally in the country. The poorer areas had the lowest coverage,e for instance in a peri-urban squatter settlement near Harare only 52.2% of children 12-23 months of age were immunized. It is among these children that epidemics are most common and sequelae most severe. Summary and conclusion The great hope and promise of post-independence efforts to promote equitable health care in Zimbabwe started with three years of dramatic improvement. Com- mitment to correcting inequities which were as discrimi- natory as any country in the world produced rapid extension of health centre infrastructure and the impro- vement of district hospitals. The major constraint was the entrenched pattern of sophisticated, high-technology health care left by colonial administrators which conti- nued to monopolize resources. In spite of the excellent beginning, development of services for the poor was thwarted by recession, prolon- ged drought and external military destabilization. The cutbacks in funding for health care have been particu- larly severe as a result of economic adjustment policies imposed by IMF. Political pressure moved the health system toward private entrepreneurship returning to earlier patterns of discrimination in favour of whites and urban residents. Efforts to promote high-risk monitoring have had little impact among the poor and those living in remote areas. Equity has become symbolic rather than real. The government of Zimbabwe maintains a continuing commitment to the original goals of equity through primary health care. International agencies also would like to find a way to help reallocate services. There seems to be recognition that little will be accomplished in improving health conditions unless services are pro- vided to those in greatest need. Disparities in maternal care are especially severe and can be improved only by building infrastructure to provide antenatal and peri- natal services. The timing seems right to try surveillan- ce for equity as a means of using limited resources to reach the most needy mothers with targeted services. A system is proposed involving periodic surveys to identify groups among whom maternal care problems are concentrated and to tailor actions to the major causes of maternal mortality and morbidity. It would also establish a process of finding locally appropriate adaptations of cost-effective and sustainable solutions. The highest-priority interventions would be defined and implemented locally by strengthening the primary health care infrastructure and community participation. Health systems research in demonstration areas could lead to national extension. Rapp. trimest. statist. sanit mond., 46 (1993) Resume Egalite dans les soins de sante : le cas du Zimbabwe Apres !'accession a l'independance du Zimbabwe, le grand espoir et la promesse d'amelioration qu'impli- quaient les efforts deployes pour promouvoir des soins de sante equitables ont commence a se concretiser au cours des trois premieres annees marquees par des progres spectaculaires. La volonte de corriger des inegalites non mains discriminatoires que dans n'impor- te quel autre pays s'est traduite par le developpement rapide de !'infrastructure des centres de sante et par !'amelioration des h6pitaux de district. L'un des princi- paux obstacles residait dans la structure heritee de !'administration coloniale d'un systeme profondement enracine de soins de sante sophistiques reclamant des technologies de pointe, qui continuait a monopoliser les ressources disponibles. En depit de cet excellent debut, la mise au point de services con<;:us pour les pauvres a ete contrecarree par la recession et la secheresse persistante, ainsi que par des manoeuvres de destabilisation militaire d'origine exterieure. Les coupures imposees aux budgets des soins de sante ont ete particulierement severes du fait de la politique d'ajustement economique voulues par le FMI. Des pressions politiques ont amene le systeme de sante a se tourner vers !'initiative privee, ce qui a marque le retour des anciennes formes de discrimination favo- rables aux blancs et aux citadins. Les efforts deployes pour promouvoir la surveillance des cas a haut risque n'ont eu que peu d'effets sur les pauvres et les habitants des zones reculees. Plus qu'une realite, l'egalite est devenue symbole. Le Gouvernement du Zimbabwe reste attache a son engagement d'atteindre ses objectifs premiers d'egalite dans le cadre du systeme de soins de sante primaires. Les institutions internationales souhaiteraient egale- ment trouver le moyen d'aider a reorienter les services de sante. Autant reconnaitre que la situation sanitaire ne sera guere amelioree tant que des services ne seront pas offerts a ceux qui en ont le plus grand besoin. Les disparites qui caracterisent les soins dispenses aux meres sont particulierement graves et ne peuvent etre ameliorees que par l'etablissement d'une infrastructure susceptible d'assurer la prestation de services prena- tals et obstetricaux. Le moment semble venu d'experi- menter la surveillance de l'egalite pour mieux tirer parti des ressources limitees et dispenser aux meres qui en ont le plus besoin des services con<;:us a leur intention. 11 est propose de mettre en reuvre un systeme d'enquetes periodiques en vue d'identifier les categories de fem- mes les plus touchees par les lacunes des soins prodi- gues aux meres et de prendre les mesures en rapport avec les principales causes de mortalite et de morbidite maternelle. Un processus serait alors instaure qui per- Wld hlth statist quart., 46 (1993) mettrait d'adapter les solutions aux c0nditions locales de fa<;:on a la fois rentable et durable. On pourrait ainsi identifier les interventions qui revetent une priorite abso- lue et les mettre en reuvre a l'echelle locale grace a une infrastructure de soins de sante renforcee et a la parti- cipation de la collectivite. Les recherches sur les syste- mes de sante dans des zones d'experimentation pour- raient contribuer a leur extension nationale. Acknowledgements We are pleased to acknowledge the assistance rende- red by the staff of the Ministry of Health, including the Permanent Secretary; the WHO Representative and members of his staff; the UNICEF Representative; offi- cials of USAID and other international agencies and others. We express sincere thanks to all for their time, insights and forthright sharing of their wisdom. References/References 1. Ministry of Health.Plannineforequity in health: aser:tmalreview and policy statement. Harare, Government Printers, 1984. 2. Omran, A.R. et al. High risk mothers and newhorns-aetection, management and prevention, Thun, Switzerland, Ott Publishers, 1987. 3. Taylor, C.E. Surveillance for equity in primary health care: policy implications from international experience, lntemationaljoumalofepidemiol.ogy, 21: 1043-1049 (1992). 4. Scrimshaw, N.S. & Gleason, G. Rapid assessment prowl.ures: qualitative methodol.ogies for planning and evaluation of health related programs, INFDC, Boston, 1992. 5. Sanders, D. Equity in health: Zimbabwe nine years on, Journal of social devel.opment in Afri.w, 5: 5-22 (1990). 6. Central Statistical Office. Statistical yearbook, Harare, Government Printers, 1987. 7. Loewenson, R. & Sanders, D. The political economy of health and nutrition, In: Stoneman, C.(ed.). Zimbabwe's prospects. London, Macmillan, 1988. 8. Government of Zimbabwe-UNICEF. Situation a~sis of women and chiW.ren in Zimbabwe: 1980-1990, Harare,Jongwe Printers, 1991. 9. Government of Zimbabwe. Estimates of expenditure, Harare, Government Printers, 1980-89. 10. Lowenson, R. et al.Challenges to equity in health and health care: a Zimbabwean case study, Social science and medicine, 32: 1079-1088 (1991). 11. Sanders, D. et al. Utilisation of different kvels of health r.are and the operation of the referral system in Zimbabwe, The IEA Africa regional conference report, August 1989. 12. Fawcus, S. et al. Booked vs. unhooked deliveries at Harare Central Hospital: maternal characteristics and fetal outcome. The IEA Africa regional conference report, August 1989. 13. Ministry of Health. Primary health r.are review 1987, Harare, Government Printers, 1987. 14. Ministry of Health. z.EPI plan of operations 1986-1990, Harare, Government Printers, February 1986. 15. Zimbabwe Central Statistical Office and Institute for Resources Development. Zimbabwe service availability survey 1989/1990, Ministry of Finance, Economic Planning, and Development, Harare, Zimbabwe and Macro International, Inc. Columbia, Maryland, USA, May 1991. 247 World Health Statistics Quarterly Vol. 46, 1993 Annual table of contents Alcohol Alcohol consumption and related problems. J. Lehto cancer Thyroid cancer in children in Belarus after Chernobyl. K. F. Baverstock Cardiovascular disease mortality Cardiovascular disease mortality in the developing countries (Introduction). Andreas T. Wielgosz Assessing the burden of mortality from cardiovascular diseases. Alan D. Lopez Cardiovascular diseases in the Eastern Mediterranean Region. Ala'din A. S. Alwan Cardiovascular diseases in India. K.S.Reddy The cardiovascular disease situation in Seychelles. P. Hungerbuhler et al. The changing pattern of cardiovascular diseases in China. Yao C. et al. The pattern of cardiovascular diseases in Indonesia. R. Boedhi-Darmojo Cardiovascular diseases in Africa. W. F. T. Muna Cardiovascular disease mortality in the Americas. E. Nicholls et al. CCEE/NIS EUROHEAL TH Programme. M. Danzon & S. K. Litvinov Health in the central and eastern countries of the WHO European Region: an overview. A. Nanda et al. Health care reforms on the European scene: evolution, revolution or seesaw? M.A. Vienonen & W. C. Wlodarczyk Raising the nursing profile, the case of t he invisible nurse. J. Savage 248 195 204 90 91 97 101 108 113 119 125 134 153 158 166 170 Rapport trimestriel de statistiques sanitaires mondiales Vol. 46, 1993 Table annuelle des matieres Aide humanitaire Aide humanitaire: evaluation technique et appui en matiere de sante publique aux operations de secours coordonnees en ex-Yougoslavie. [resume] X. Leus Al cool La consommation d'alcool et problemes a pparentes dans les PECO/NEI. (resume] J. Lehto Cancer Cancer de la thyro'ide chez Jes enfants apres Tchernobyl. [resume] K. F. Baverstock Femmes et violence La violence contre Jes femmes: un probleme de sante cache [resume]. L. Heise lnfirmieres Ameliorer le profil des infirmieres: le cas de l'infirmiere invisible. [resume] J. Savage Maladies cardiovasculaires Mortalite par maladies cardio-vasculaires dans Jes pays en developpement (Introduction) [anglais seulement]. Andreas T. Wielgosz Evaluation du poids de la mortalite due aux maladies cardio-vasculaires [resume]. Alan D. Lopez La prevalence des maladies cardio-vasculaires dans la Region de la Mediterranee orientale [resume]. Ala'din A. S. Alwan Les maladies cardio-vasculaires en lnde [resume]. K. s. Reddy Les maladies cardio-vasculaires aux Seychelles (resume]. P. Hungerbuhler et al. Evolution des maladies cardio-vasculaires en Chine [resume]. Yao C. et al. Caracteristiques des maladies cardio- vasculaires en lndonesie [resume] R. Boedhi-Darmojo Les maladies cardiovasculaires en Afrique [resume]. W. F. T. Muna Mortalite par maladies cardio-vasculaires dans les Ameriques [resume]. E. Nicholls et al. 203 198 208 78 176 90 5 99 107 111 118 124 131 150 Rapp. trimest. statist. sanit. mond., 46 (1993) Communicable diseases in the CCEE/NIS. C. Roure & G. Oblapenko Tobacco or health. T. Piha, E. Besselink & A. D. Lopez Alcohol consumption and related problems. J. Lehto Humanitarian assistance: technical assessment and public health support for coordinated relief in the former Yugoslavia. X. Leus Thyroid cancer in children in Belarus after Chernobyl. K. F. Baverstock Child abuse Child abuse: measuring a global problem. M.A. Belsey Communicable diseases Communicable diseases in the CCEE/NIS. C. Roure & G. Oblapenko1 Corrigenda Vol. 46, 1993 Health transition Health in the central and eastern countries of the WHO European Region: an overview. A. Nanda et al. Health care reforms on the European scene: evolution, revolution or seesaw? M.A. Vienonen & W. C. Wlodarczyk EUROHEAL TH Programme. M. Danzon & S. K. Litvinov Humanitarian assistance Humanitarian assistance: technical assessment and public health support for coordinated relief in the former Yugoslavia. X. Leus Mortality See Cardiovascular disease mortality Nursing Raising the nursing profile, the case of the invisible nurse. J. Savage Suicide The epidemiology of suicidal behaviour: a review of three continents. 177 188 195 99 204 69 77 209 158 166 153 199 170 R. F. W. Diekstra & W. Gulbinat 52 Tobacco Tobacco or health. T. Piha, E. Besselink & A. D. Lopez 188 Violence and health Violence and health (Introduction) [French only] 2 Comparative analysis of violent deaths in the developed countries and in some developing countries, 1985-1989 [summary]. R. Bourbeau 31 Wld hlth statist. quart., 46 (1993) Maladies transmissibles Les maladies transmissibles dans les PECO/NE! [resume]. C. Roure & G. Oblapenko 186 Mortalite Vair Maladies cardlo-vasculaires PECO/NE I Le programme EUROHEALTH. M. Danzon & S. K. Litvinov [anglaisseulement] 156 La sante dans les pays d'Europe centrale et orientale de la Region europeenne de l'OMS; aperc;u general [resume]. A. Nanda et al. 165 Reforme des soins de sante sur le continent europeen; evolution, revolution o u mouvement de bascule? [resume] M.A. Vienonen & W. C. Wlodarczyk 169 Ameliorer le profil des infirmieres: le cas de l'infirmiere invisible [resume]. J. Savage 176 Les maladies transmissibles dans les PECO/NEI [resume]. C. Roure & G. Oblapenko 186 Tabac ou sante [resume]. T. Piha, E. Besselink & A. D. Lopez 194 La consommation d'alcool et problemes apparentes dans les PECO/NEI [resume]. J. Lehto 198 Aide humanitaire: evaluation technique et appui en matiere de sante publique aux operations de secours coordonnees en ex-Yougoslavie [resume]. X. Leus 203 Cancer de la thyro"ide chez les enfants apres Tchernobyl [resume]. K. F. Baverstock 208 Rectiflcatifs Vol. 46, No. 1, 1993 Suicide Epidemiologie des comportements suicidaires: analyse sur trois continents [resume]. R. F. W. 209 Diekstra & W. Gulbinat 52 Tabac ou sante Tabac ou sante [resume]. T. Piha, E. Besselink & A. D. Lopez 194 Transition sanitaire Le programme EUROHEALTH. M. Danzon & S. K. Litvinov [anglais seulement] 156 La sante dans les pays d'Europe centrale et orientale de la Region europeenne de l'OMS; aperc;u general [resume]. A. Nanda et al. 165 Reforme des soins de sante sur le continent europeen; evolution, revolution ou mouvement de bascule? [resume] M.A. Vienonen & W. C. Wlodarczyk 169 Violence et sante Violence et sante (Introduction) 2 249 Intentional violence among adolescents Analyse comparative de la mortalite violente and young adults: an epidemiological dans les pays developpes et dans quelques perspective. pays en developpement durant la periode 0. Jeanneret & E. A. Sand 34 1985-1989. R. Bourbeau 4 The epidemiology of suicidal behaviour: Violence intentionnelle chez les adolescents a review of three continents. et les jeunes adultes: perspective R. F. W. Diekstra & W. Gulbinat 52 epidemiologique [resume]. 0. Jeanneret & E. A. Sand 4 Child abuse: measuring a global problem. M.A. Belsey 69 Epidemiologie des comportements suicidaires: analyse sur trois continents [resume]. Violence against women: the hidden health R. F. W. Diekstra & W. Gulbinat 52 burden. L. Heise 78 La maltraitance des enfants: mesurer un probleme mondial [resume]. M.A. Belsey 69 Women Violence against women: the hidden health La violence centre les femmes: un probleme burden. L. Heise 78 de sante cache [resume]. L. Heise 78 250 Rapp. trimest statist. sanit. mond., 46 (1993) Publications of the World Health Organization 1994 WORLD HEALTH FORUM An international journal of health development (Separate editions in English, French, Spanish, Arabic, Chinese and Russian) World htalth 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 public 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 of WHO'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. 68.- Sw. fr. 20.- BULLETIN OF THE WORLD HEALTH ORGANIZATION The Bu/ktin is the principal scientific organ of WHO: its role is to 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. 160.- Sw. fr. 28.- Prepared for the guidance of health administrations and health authorities, the Weekly ,pidemiol.ogi.cal record contains notifications made under the International Health, Regulations and information concerning their application. The &cord also contains epidemiological information on communicable diseases of international importance. Annual subscription ......................................................... . Sw. fr. 190.- WORLD HEALTH STATISTICS QUARTERLY The World htalth statistics quarterly replaces (since 1978) the World htalth statistics report (published since 1967) and its forerunner the Epukmiol.ogical and vital statistics report (published since 1947). It deals with the detailed analysis of selected health topics of current interest. The Quarterly contains articles in eithtr French or English with a summary in both languages. Annual subscription ......................................................... . Price per copy ............................................................... . WORLD HEALTH STATISTICS ANNUAL (Bilingual: English and French) Sw. fr. 100.- Sw. fr. 28.- The forerunner of this series was the Annual epidtmiol.ogi.cal report of the Leagill of Nations. It was followed by the Annual epidemiol.ogi.cal and vital statistics issued by the World Health Organization. Latest publications: 1989. Vital statistics and life tables, the WHO MONICA Project, causes of death (single volume), 443 pages .............................................................. . 1990. Vital statistics and life tables, international statistics on causes of disability, causes of death (single volume), 447 pages (out of stock) ............................ : ....................................... . 1991. 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 Healh 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 ................................. . Sw. fr. 90.- Sw. fr. 90.- Sw. fr. 100.- Sw. fr. 100.- Sw. fr. 100.- Publications de /'Organisation mondiale de la Sante 1994 FORUM MONDIAL DE LA SANTE Revue internationale de developpement sanitaire (Editions separees en fran~ais, anglais, espagnol, arabe, chinois et russe) Forum mondial,ulasante est une revue trimestrielle destinee aux responsables 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 sante. Forum se consacre a l'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'elles soient ou non soutenues par l'OMS. TI est le principal organe a la disposition des Etats Membresde l'OMS pour l'echange international d'informationssanitairesen meme tempsqu'un instrument de cooperation technique entre pays en developpement. Abonnement (4 numeros) ............................................... . l..e numero ····································································· Fr. s. 68.- Fr. s. 20.- BULLETIN DE L'ORGANISATION MONDIALE DE LA SANTE Le Bulktin est le principal organe scientifique de !'OMS; ii a pour role de passer en revue les progres des sciences medicales et apparentees (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. Le Bulktin contient des articles originaux en fran~s ou en anglais, accompagnes d'un resume dans l'autre langue. Bimensuel. Abonnement (6 numeros) ............................................... . Lenumero ····································································· RELEVE EPIDEMIOLOGIQUE HEBDOMADAIRE (Bilingue: fran~s et anglais) Fr. s. 160.- Sw. fr. 28.- Destine aux administrations sanitaires et aux services de sante, le &/.eve ipidhnwl.ogi.que hebdomadaire contient les notifications exigees par le Reglement sanitaire international, ainsi que d'autres renseignements concernant !'application de ce reglement. Le &/.eve contient egalement des informations epidemiologiques concernant les maladies transmissibles d 'une importance internationale. Prix de l'abonnement annuel .......................................... . Fr. s. 190.- RAPPORT TRIMESTRIEL DE STATISTIQUES SANITAIRES MONDIALES Le Rapport trimestri.el ,u statistiques mondiaks, remplace ( depuis 1978) le Rapport ,u statistiques sanitaires mondiaks (publie depuis 1967) et son precurseur le Rapport epidemwl.ogi.que et demographique (publie depuis 194 7) . II presente des analyses detaillees sur des sujets specifiques d 'interet courant. Le Trimestriel presente des articles originaux en fran~ais ou en anglais, accompagnes d'un resume dans les deux langues. Prix de l'abonnement annuel .......................................... . Le numero .................................................................... . Fr. s. 100.- Fr. s. 28.- ANNUAIRE DE STATISTIQUES SANITAIRES MONDIALES (Bilingue: fran~ais et anglais) Cet annuaire remplace les Statistiques epidemwl.ogiques et demographiques annuell.es publiees par ('Organisation mondiale de la Sante et qui avaient, elles-memes, rem place le Rapport epidemiol.ogi.que annuelpublie par la Societe des Nations. Publications recentes: 1989. Mouvementde la population et tables de survie, le Projet MONICA de !'OMS, causes de deces (1 seul volume), 443 pages .................................. . 1990. Mouvement de la population et tables de survie, statistiques intemationales sur les causes d'incapacite, causes de deces (1 seul volume), 447 pages (epuise) ............................................... . 1991. Mouvement de 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 reuvre de la Strategie mondiale de la sante pour tous d'ici !'an 2000 (1 seul volume), 480 pages .............................................................. . 1993. Mouvement de la population et tables de survie, donnees sanitaires etdemographiques, methodes permettant de recueillir des donnees sur les causes de deces et les faits de l'etat civil en ('absence de systeme universe( d'enregistrement (1 seul volume), 662 pages .................................. . Fr. s. 90.- Fr. s. 90.- Fr. s. 100.- Fr. s. 100.- Fr. s. 100.-