Situation analysis for cervical cancer diagnosis and treatment in east, central and southern African countries.

Situation analysis for cervical cancer diagnosis and treatment in East, Central and Southern African countries Zvavahera M. Chirenje,1 Simbarashe Rusakaniko,2 Leah Kirumbi,3 Edward W. Ngwalle,4 Pulani Makuta-Tlebere,5 Sam Kaggwa,6 Winnie Mpanju-Shumbusho,7 & Lucy Makoae8 Objective To determine the factors influencing cervical cancer diagnosis and treatment in countries of East, Central and Southern Africa (ECSA). Methods Data were collected from randomly selected primary health care centres, district and provincial hospitals, and tertiary hospitals in each participating country. Health care workers were interviewed, using a questionnaire; the facilities for screening, diagnosing, and treating cervical cancer in each institution were recorded, using a previously designed checklist. Findings Although 95% of institutions at all health care levels in ECSA countries had the basic infrastructure to carry out cervical cytology screening, only a small percentage of women were actually screened. Lack of policy guidelines, infrequent supply of basic materials, and a lack of suitable qualified staff were the most common reasons reported. Conclusions This study demonstrates that there is an urgent need for more investment in the diagnosis and treatment of cervical cancer in ECSA countries. In these, and other countries with low resources, suitable screening programmes should be established. Keywords: cervix neoplasms, diagnosis, therapy, surgery; cervix dysplasia, diagnosis; ctodiagnosis; vginal smears; community health centers; hospitals, district; hospitals, general; Zimbabwe; Lesotho; Kenya; United Republic of Tanzania; Uganda. Mots cle´s: tumeur du col de l’ute´rus, diagnostic, the´rapeutique, chirurgie; dysplasie du col de l’ute´rus, diagnostic; cytodiagnostic; frottis vaginal; centre de sante´ communautaire; hoˆpital re´gional; hoˆpital ge´ne´ral; Zimbabwe; Lesotho; Kenya; Re´publique-Unie de Tanzanie; Ouganda. Palabras clave: neoplasmas del cuello uterino, diagno´stico, terapia, cirugı´a; displasia del cuello uterino, diagno´stico; citodiagno´stico; frotis vaginal; centros comunitarios de salud; hospitales de distrito; hospitales generales; Zimbabwe; Lesotho; Kenya; Repu´blica Unida de Tanzanı´a; Uganda. Bulletin of the World Health Organization, 2001, 79: 127–132. Voir page 131 le re´sume´ en franc¸ais. En la pa´gina 132 figura un resumen en espan˜ol. Introduction Cancer of the cervix is having a devastating impact on women’s health around the world, especially in developing countries, where it is the most common cancer and the leading cause of death from cancer in women. It is estimated that 500 000 new cases occur every year worldwide, themajority (80%) being in the developing world (1). Although cervical cancer is a preventable disease, it still remains a major burden on public health resources in sub-Saharan Africa. Countries in this region have some of the world’s highest age- standardized death rates from cervical invasive cancer, e.g. 67 per 100 000 people in Harare, Zimbabwe (2) and 40.8 per 100 000 in Kampala, Uganda in 1997 (3). Data from hospital-based registers inNairobi, Kenya have indicated that cancer of the cervix accounted for 70–80% of all cancers of 1 Senior Lecturer, Department of Obstetrics & Gynaecology, University of Zimbabwe, PO Box A178, Avondale, Harare, Zimbabwe. Correspondence should be addressed to this author. 2 Lecturer, Department of Obstetrics & Gynaecology, University of Zimbabwe, Harare, Zimbabwe. 3 Lecturer, Department of Obstetrics & Gynaecology, University of Nairobi, Nairobi, Kenya. 4 Senior Lecturer, Department of Obstetrics & Gynaecology, University of Dar es Salaam, United Republic of Tanzania. 5 Consultant, Ministry of Health, Maseru, Lesotho. 6 Regional Director, Commonwealth Regional Health Community Secretariat, Arusha, United Republic of Tanzania. 7 Lecturer, Makerere University, Kampala, Uganda. 8 Clinical Nursing Officer, Ministry of Health, Maseru, Lesotho. Ref. No. 99-0444 127Bulletin of the World Health Organization, 2001, 79 (2) # World Health Organization 2001 the genital tract and 8–20% of all cancer cases for the period 1981–90. It has been reported that there are 10–15 new cases of cervical cancer in Nairobi, Kenya each week (4). These findings are similar to data from Zimbabwe where cancer of the cervix accounted for 30% of all registered cancers and 80% of gynaeco- logical cancers in the entire country in 1995 (2). The effectiveness of cervical cancer screening by exfoliative cytology was demonstrated in the Scandi- navian countries, where a substantial fall in incidence of cervical cancer occurred 5–10 years after the introduc- tion of screening programmes (5). The Canadian Cancer Institute has also shown that there was a threefold reduction in the age-standardized death rate from invasive cervical cancer between 1969 and 1990, thanks to the national screening programme (6). Accurate data on the magnitude of morbidity and mortality from cervical cancer in the countries of East, Central and Southern Africa (ECSA) are scanty and usually hospital based. Cancer registration in most of the low-resource countries is difficult because of inadequate sources of information within the health delivery system, and lack of reliable population data for estimating accurate incidence rates. To facilitate the proper planning of services for the prevention, early diagnosis and treatment of cervical cancer in the ECSA countries, a situation analysis was carried out. The objective of the study was to establish which factors influence cervical cancer diagnosis and treatment in the ECSA region. Materials and methods Amulticentre cross-sectional study was carried out in Kenya, Lesotho, Uganda, United Republic of Tanzania, and Zimbabwe between 1 May and 31 December 1997. For each participating country, data were collected from tertiary hospitals, provincial hospitals, district hospitals, and primary health care (PHC) centres selected at random by computer. Definitions of health care facilities in ECSA countries Primary health care (PHC) centres are the basic operational facilities in the health care delivery system, focusing mainly on disease prevention and health promotion. Each centre usually serves a catchment area of 10 000 people living within a radius of 8–10 km from the centre. In this study, dispensaries, health centres, maternity homes and clinics were classified as PHC centres. District hospitals usually have surgical facilities, are headed by a medical officer and are the first referral unit for the PHC centres in the district. Provincial hospitals have specialists such as gynaecologists and general surgeons, and are referral centres for patients from the district hospitals in the province. Tertiary hospitals provide highly specialized care and usually have subspecialities in various disciplines (e.g. renal medicine, cardiology, and gynaecological oncology). Data collection Interviews were conducted at each level of health care facility, and were carried out by pre-trained health workers, using a questionnaire. At PHC centres, the nurse in charge and other relevant health workers were interviewed. At district and provincial hospitals, the gynaecologist, medical officer in charge, director and nurse in charge of the maternal and child health services were questioned. At tertiary centres, the consultant gynaecologist, nurse in charge of the gynaecology outpatient clinic, nurse in charge of thematernal and child health services, and hospital administrator were interviewed. Each questionnaire addressed issues concerning the screening, diagnos- tic, and treatment facilities; human resources; and the communication and referral structures available to cervical cancer patients. The facilities for screening, diagnosing and treating cervical cancer were recorded during visits to all the selected health institutions, using a previously designed checklist. The data collected in each country were coded by the principal investigator before entry into the computer, using the EpiInfo Version 6 programme, and cleared by each in-country coordi- nator before being sent to the coordinating centre in Harare, Zimbabwe. The data from the five countries were merged into one file and then analysed using EpiInfo. Results The situation analysis was carried out in 159 PHC centres, 48 district hospitals, 24 provincial hospitals, and 12 tertiary hospitals in the five ECSA countries (Table 1). The tertiary and provincial hospitals in the study were all public institutions to whichmost of the referral cases from both public and private PHC centres and district hospitals were sent for treatment. Primary health care (PHC) centres A total of 159 PHC centres participated in the study (42 in Kenya, 20 in Lesotho, 28 in Uganda, 40 in the United Republic of Tanzania, and 29 in Zimbabwe); this sample included 66 dispensaries (14 in Kenya, 20 in Uganda, and 32 in the United Republic of Tanzania). As seen in Table 2, most of the PHC centres surveyed had an examination room (92%), or an examination couch (87%) and screen (80%), where a patient could have a pelvic examination and a cervical smear taken. Basic prerequisites for taking a cervical smear (i.e. a speculum, spatula, glass slides, and a working light source) were available in 70% of the PHC centres. Although 80% of the health care workers in the five countries were aware of what a cervical Pap (Papanicolaou) smear was, only 17% of them had taken a smear in the preceding six months. Lesotho and Zimbabwe were the only countries in the ECSA region performing cervical cancer screen- ing at a PHC centre. The number of Pap smears performed per centre across the five countries averaged four per month (nine per month in 128 Bulletin of the World Health Organization, 2001, 79 (2) Research Zimbabwe, while the other countries had one or none per month). Some of the reasons given by health workers why very few women were screened in the ECSA regionwere the absence of policy guidelines, frequent shortages of materials needed for taking Pap smears, and the long distance and cost of sending the smear to the processing centre. The PHC centres surveyed in the five countries showed an average of six trained staff members per centre (two nurse clinicians, two state registered nurses, and two state certified nurses). It was reported that these numbers were adequate for staffing a cervical cancer screening service. The records and follow-up of the 17% of patients who were screened in Lesotho and Zimbabwe showed an absence of precise manage- ment protocols, and resulted in poor outcomes. Some 31% of the PHC centres surveyed had seen at least two cases of invasive cervical cancer in the preceding year. Among the common problems and difficulties faced by women with cervical cancer, as reported by PHC staff, were the high cost of travel to the referral centre, the advanced stage of the disease due to late presentation in the PHC centre, and inadequate palliative care for terminally ill cases. There were no differences in the availability of facilities, staffing levels, and the number of women screened per month between the public-sector and privately-owned PHC centres in the mining and commercial farming areas. District and provincial hospitals Of the 72 district and provincial hospitals surveyed, 37 (52%) carried out screening of patients for cervical cancer (Table 3). There were fewer institutions for screening in Uganda (15%) and the United Republic of Tanzania (5%) than in the other countries. The infrastructure required for taking a Pap smear (i.e. an examination room, or an examination couch and screen) was available in 99% of the institutions, while a speculum, spatula and glass slides were available in 80% of the hospitals. Shortages of test materials (especially fixatives) and of staff (including cytology technicians), and lack of policy guidelines were the most common reasons given for not being able to carry out screening. The institutions performing screening were mostly family planning clinics (45%), postnatal clinics (29%), and gynaecology outpatient clinics (20%). A high propor- tion (57%) of the smears collected were taken by doctors. The number of Pap smears taken averaged only 24 permonth per hospital, and the processingwas carried out in the provincial hospitals (14%), tertiary hospitals (65%), and private laboratories (21%). The absence of a cytology technician or histopathologist in all the district and most provincial Table 1. Distribution of health care facilities (number existing and number surveyed) in the five ECSA countries Health care Kenya Lesotho Uganda United Zimbabwe Total level Republic of Tanzania Primary Existing 200 157 117 103 1114 1691 Surveyed 42 20 28 40 29 159 District Existing 60 NA 39 63 52 214 Surveyed 12 NA 12 16 8 48 Provincial Existing 7 16 4 17 8 52 Surveyed 3 10 4 4 3 24 Tertiary Existing 1 2 1 4 4 12 Surveyed 1 2 1 4 4 12 NA = not applicable in this country. ECSA = East, Central and Southern African countries. Table 2. The number of diagnostic facilities (in the checklist) available in primary health care centres surveyed in the five ECSA countries Diagnostic Kenya Lesotho Uganda United Zimbabwe Total facilities Republic of Tanzania Examination room 37 20 20 40 29 146 Examination couch 37 19 22 31 29 138 Light source 29 14 17 22 27 109 Screen 33 14 16 36 28 127 Gloves 34 18 19 35 29 135 Sterilizers 29 16 21 37 28 131 Kidney dishes 35 20 21 37 29 143 Vulsellum forceps 14 16 9 27 27 93 Speculum 31 20 0 32 29 112 Spatula 30 15 14 31 28 118 Glass slides 16 13 12 38 27 106 Fixatives 3 11 0 20 15 49 Slide holder 8 9 0 27 8 52 Stationery 20 12 11 38 23 104 Marker pencil 10 11 11 37 15 84 Detergents 29 11 16 40 26 122 ECSA = East, Central and Southern African countries. 129Bulletin of the World Health Organization, 2001, 79 (2) Cervical cancer diagnosis and treatment hospitals surveyed made it necessary for the speci- mens taken for screening to be dispatched to the tertiary hospitals and private laboratories. It took, on average, 4 weeks to obtain the results of the Pap smear and sometimes 6–8 weeks. The survey also showed that only 31% of district and provincial hospitals had facilities to perform a biopsy on a woman with an abnormal Pap smear. Colposcopy was not available in any of the district hospitals, but 6% of provincial hospitals had this facility. The only mode of treatment offered for cervical precancer was cone biopsy (31%) at provincial hospitals. Equipment for cryotherapy and loop electrosurgical excision procedure (LEEP) was, however, not available in all the district and provincial hospitals surveyed. Only 21% of the provincial hospitals had a gynaecologist in place during the survey; their absence was a common reason given for failure to offer any cervical cancer treatment at these institutions throughout the ECSA region. Tertiary hospitals All 12 tertiary hospitals in the participating countries were surveyed (1 inKenya, 2 in Lesotho, 1 inUganda, 4 in the United Republic of Tanzania, and 4 in Zimbabwe). All the hospitals reported that screening for cervical cancer was being carried out at the time of the study, except in the United Republic of Tanzania, where only one out of the four tertiary hospitals carried out screening. Screening facilities were available in all the hospitals, but the three Tanzanian hospitals that did not carry out screening reported infrequent supplies of materials such as fixatives and laboratory reagents as the most common constraint. An average of 86 smears per month were carried out at each hospital, with the Tanzanian hospitals having the lowest number (40 per month). All the tertiary hospitals had facilities to process cervical smears although they reported having an inadequate number of cytology technicians. All had at least two histopathologists in place at the time of the survey. Punch biopsy, cone biopsy, and colposcopy services were available in all. Cryotherapy and LEEP were available in two hospitals: one in Zimbabwe and the other in Lesotho. The remaining hospitals used cone biopsy as the primary mode of treatment for cervical precancer. Surgical treatment services were available in all hospitals, but only those in Zimbabwe had two functional radiotherapy units with adequate staffing at the time of the survey. Hospitals in Lesotho had no radiotherapy equipment in place, while those in Kenya, Uganda and the United Republic of Tanzania, which had equipment, re- ported shortages of staff (a radiation oncologist and physicists) and isotopes. Patients with advanced cervical cancer in most of the tertiary hospitals were offered palliative radiotherapy, analgesia, psychosocial support (through Island Hospice), blood transfusion and haematinics (agents to increase the haemoglobin content of the blood), and antibiotics for secondary pelvic infections. Discussion As screening for cervical cancer in most developing countries is minimal or non-existent, there are few opportunities to diagnose precancerous lesions. Consequently the majority of patients are diagnosed with advanced-stage disease, which carries a high morbidity and mortality rate (1). In several ECSA countries, cervical cancer accounts for 80% of all gynaecological cancer admissions among young women, with a peak incidence in women aged 45 years; more than 50% of the patients present cancers beyond stage 3 (4, 7). This study showed that 95% of the institutions surveyed at the primary, district, provincial and tertiary levels in ECSA countries had the basic infrastructure to perform exfoliative cervical cytology screening. However, only very few women were actually being screened for cervical cancer at the time of the survey, with an average of only four Pap smears per month at the primary level. On average, 20 smears per month were performed at the district Table 3. Percentages of screening and treatment facilities, including the percentages of diagnostic facilities (in the checklist), in the district and provincial hospitals surveyed in the five ECSA countries Facilities Kenya Lesotho Uganda United Zimbabwe Total Republic of Tanzania % % % % % % Screening 56 94 15 5 88 52 Types of clinic carrying out screening Gynaecology 7 36 5 5 47 20 outpatient Antenatal 1 19 0 0 6 5 Postnatal 9 58 0 10 66 29 Family planning 64 74 5 12 69 45 Average percentage 5 6 10 0 7 6 of women screened per week Punch biopsy 38 24 45 45 22 35 Cone biopsy 31 24 25 5 69 31 Colposcopy 13 0 5 5 5 6 Hospitals with 19 0 5 5 5 6 a colposcope Diagnostic facilities Examination room 100 100 95 100 100 99 Examination couch 100 100 100 95 100 99 Light source 100 930 80 65 100 88 Speculum 100 100 95 85 100 96 Spatula 88 100 40 65 97 78 Detergents 81 87 65 68 97 80 ECSA = East, Central and Southern African countries. 130 Bulletin of the World Health Organization, 2001, 79 (2) Research and provincial levels, and 86 smears per month at the tertiary level. Uganda and the United Republic of Tanzania showed the lowest numbers of women screened per month. This study thus confirms the findings from WHO, which reported that <5% of women in developing countries were screened for cervical cancer per year, compared with 45–50% of women in industrialized countries (8). The consider- able differences in cervical cancer incidence rates are a result of the very low screening coverage in developing countries. This survey showed that there were no clear policy guidelines on cervical cancer screening throughout the entire ECSA region. The few women who were screened were screened more by chance: they were often women under 25 years of age (at which age precancer lesions are not common), or symptomatic women who were harbouring invasive disease. The lack of policy guidelines could partly explain the very low number of women being screened in institutions that have the basic infra- structure and facilities for cervical cancer screening. However, the infrequent supply of materials (such as fixatives and laboratory reagents for staining the smears) and lack of cytology technicians in most of the surveyed institutions were often the reasons for missed screening opportunities. Facilities to perform cone biopsy were available in 42% of the district and provincial hospitals, and yet it was carried out in only 31% of the provincial hospitals. Cone biopsy still remains an important procedure for both diagnostic and therapeutic management of cervical precancer, which ought to be available in all district and provincial hospitals. Only 4% of institutions surveyed in the region had the equipment to perform cryotherapy, although this mode of cervical precancer treatment is relatively inexpensive, does not require general anaesthesia, and has an 84% cure rate for treating cervical intra- epithelial neoplasia grade III (carcinoma in situ) lesions (9). This study demonstrated that there is an urgent need for more investment in the diagnosis and treatment of cervical cancer in ECSA countries. Several studies have consistently shown that visual inspection of the cervix (VIA) or cervicoscopy, after staining the cervix with 3–4% acetic acid, has a sensitivity of between 60% and 70% (10–12). Thus, countries with low resources (including those in ECSA) where organized cytology screening pro- grammes are not feasible need to implement VIA programmes as an alternative to seeking new investments for the management of invasive cervical cancer. Currently, VIA increases treatment costs due to false-positive referrals. These high rates of false positives can be reduced by rigorous training programmes, especially if incorporated into a combined testing with VIA and treatment options with cryotherapy and LEEP (10). Surgical facilities to operate on women with cervical cancer were available in 46%of the provincial hospitals, of which 21% had a gynaecologist to perform the operation. Thus, the majority of women with invasive cervical cancer are referred to tertiary hospitals, resulting in huge transport costs to the patient. There is also a need for investment so that a cytology technician or histopathologist is available in all provincial hospitals of the ECSA region. For any cervical cancer screening programme to be effective, each provincial hospital must have a resident pathologist and gynaecologist to avoid costly referrals to tertiary hospitals. The numbers of oncosurgeons, radiotherapists and physicists in the tertiary hospitals were inadequate. Radiotherapy centres were also few for the number of cervical cancer cases diagnosed each year. Kenya and the United Republic of Tanzania, each with a population of 30 million, have one radiotherapy centre for the entire country, in sharp contrast to Zimbabwe, which has a population of 11 million and two radiotherapy centres. A direct association between increased inci- dence of cervical cancer and HIV infection has not been demonstrated (13), although increased rates of abnormal cytology among HIV-infected females has been observed (14). It would be cost-effective for policy-makers in HIV endemic areas in the ECSA region to incorporate urgently cervical cancer screen- ing programmes with newly tested screening tech- nologies such as VIA or cervicoscopy, supported by cryotherapy and LEEP as treatment modes. n Acknowledgements We thank the Commonwealth Regional Health Community Secretariat (CRHCS) and the govern- ments of Kenya, Lesotho, Uganda, United Republic of Tanzania, and Zimbabwe for granting permission to carry out this study, and for scientific and technical support. Funds for this project were provided by CRHCS and the Rockefeller Foundation (for data analysis and report writing). Re´sume´ Diagnostic et traitement du cancer du col de l’ute´rus en Afrique orientale, centrale et australe : analyse de la situation Objectif De´terminer quels sont les facteurs qui in- fluencent le diagnostic et le traitement du cancer du col de l’ute´rus dans les pays d’Afrique orientale, centrale et australe. Me´thodes Dans chacun des pays participants, les donne´es ont e´te´ recueillies aupre`s d’e´tablissements de diffe´rentes cate´gories – centre de soins de sante´ primaires, hoˆpitaux de district et de province et hoˆpitaux tertiaires – choisis par tirage au sort. Le personnel soignant a e´te´ interroge´ au moyen d’un questionnaire, et on a utilise´ une liste de controˆle pre´e´tablie pour noter les moyens de de´pistage, de diagnostic et de traitement du 131Bulletin of the World Health Organization, 2001, 79 (2) Cervical cancer diagnosis and treatment cancer du col de l’ute´rus en usage dans chaque e´tablissement. Re´sultats Bien que 95 % des e´tablissements de sante´ des pays d’Afrique orientale, centrale et australe disposent, quel que soit leur niveau, de l’infrastructure de base pour le de´pistage cytologique du cancer du col de l’ute´rus, seul un faible pourcentage de femmes en be´ne´ficient re´ellement. L’absence de lignes directrices, les lacunes de l’approvisionnement en fournitures de base et le manque de personnel qualifie´ en sont les raisons les plus fre´quemment avance´es. Conclusion Cette e´tude montre qu’il est urgent que les pays d’Afrique orientale, centrale et australe investissent davantage dans le diagnostic et le traitement du cancer du col de l’ute´rus. Dans ces pays, ainsi que dans d’autres pays dont les ressources sont limite´es, des programmes de de´pistage approprie´s devront eˆtre e´tablis. Resumen Ana´lisis del estado del diagno´stico y el tratamiento del ca´ncer cervicouterino en paı´ses de A´frica oriental, central y meridional Objetivo Determinar los factores que influyen en el diagno´stico y el tratamiento del ca´ncer cervicouterino en paı´ses de A´frica oriental, central y meridional (AOCM). Me´todos Los datos proceden de una muestra seleccio- nada al azar de centros de atencio´n primaria, hospitales distritales y provinciales y hospitales terciarios de cada paı´s participante. Se entrevisto´ a los agentes de salud mediante un cuestionario, y se utilizo´ una lista de comprobacio´n previamente establecida para registrar los servicios de cribado, diagno´stico y tratamiento del ca´ncer cervicouterino de que disponı´a cada institucio´n. Resultados Aunque el 95% de las instituciones de todos los niveles asistenciales de los paı´ses de AOCM disponı´an de la infraestructura ba´sica para llevar a cabo ana´lisis citolo´gicos del cuello uterino, en la pra´ctica so´lo se cribaba a un pequen˜o porcentaje de mujeres. Las razones ma´s citadas para explicar ese hecho fueron la falta de directrices de polı´tica, el insuficiente suministro de material ba´sico y la falta de personal cualificado. Conclusio´n Este estudio demuestra que es urgente invertir ma´s medios en el diagno´stico y el tratamiento del ca´ncer cervicouterino en los paı´ses de AOCM. Es necesario emprender programas de cribado adecuados tanto en esos paı´ses como en otros con recursos escasos. References 1. Parkin DM, Pisani P, Ferlay J. Estimates of the worldwide incidence of eighteen major cancers in 1985. International Journal of Cancer, 1993, 54: 594–606. 2. 1995 Annual Report. Harare, Zimbabwe National Cancer Registry, 1997. 3. Wabinga H. Kampala Cancer Registry Data. In: Parkin DM et al., eds. Cancer incidence in five continents. Vol. VII. Lyon, International Agency for Research in Cancer, 1997: 82–85 (IARC Scientific Publication No. 143). 4. Rogo KO et al. Carcinoma of the cervix in an African setting. International Journal of Obstetrics and Gynaecology, 1990, 33: 249–255. 5. Hakama M et al. Evaluation of screening programmes for gynaecological cancer. British Journal of Cancer, 1985, 52: 669. 6. Canadian cancer statistics, 1996. Toronto, National Cancer Institute of Canada, 1996: 15. 7. Kasule J, Coulson R, Akino V. Cancer of the cervix in Zimbabwe: a prospective clinico-epidemiological study of 211 patients. Obstetrics and Gynaecology of East and Central Africa, 1989, 8: 61–64. 8. Control of cancer of the cervix uteri (Memorandum from a WHO meeting). Bulletin of the World Health Organization, 1986, 64: 607–618. 9. Anderson ES, Husth M. Cryosurgery for cervical intra-epithelial neoplasia: 10 year follow-up. Gynaecologic Oncology, 1992, 45: 240–242. 10. Megevand E et al. Acetic acid visualisation of the cervix: an alternative to cytologic screening. Obstetrics and Gynaecology, 1996, 88: 383–386. 11. Wesley R et al. Evaluation of visual inspection as a screening test for cervical cancer. British Journal of Cancer, 1997, 75: 436–440. 12. Chirenje ZM et al. Visual inspection with acetic acid for cervical cancer screening: test qualities in a primary care setting. Lancet, 1999, 353: 869–873. 13. Beral V, Newton R. Overview of the epidemiology of immuno- deficiency-associated cancers. Journal of the National Cancer Institute, 1998, 23: 1–6. 14. Johnstone FD et al. A population-based controlled study of the relationship between HIV infection and cervical neoplasia. British Journal of Obstetrics and Gynaecology, 1994, 101: 986–991. 132 Bulletin of the World Health Organization, 2001, 79 (2) Research