HIV/AIDS in Zambia Lisa Garbus, MPP AIDS Policy Research Center, University of California San Francisco Published March 2003 Updated March 2003 (c) Regents of the University of California Table of Contents (click on page number to go directly to that section) PREFACE 4 NOTE ON DATA SOURCES 5 ACKNOWLEDGMENTS 5 CONTACT INFORMATION 5 EXECUTIVE SUMMARY 6 EPIDEMIOLOGY 6 POLITICAL ECONOMY AND SOCIOBEHAVIORAL CONTEXT 6 IMPACT 8 RESPONSE 9 EPIDEMIOLOGY 12 AT A GLANCE 12 HIV SENTINEL SURVEILLANCE 13 UNAIDS ESTIMATES 20 AIDS CASES 21 SEXUALLY TRANSMITTED INFECTIONS (STIS) 21 TRANSMISSION PATTERNS 21 AIDS MORTALITY 22 DATA QUALITY ISSUES 23 POLITICAL ECONOMY AND SOCIOBEHAVIORAL CONTEXT 24 AT A GLANCE 24 POPULATION MOBILITY 39 FOOD CRISIS 43 HEALTH SECTOR 44 GENDER 46 KNOWLEDGE OF HIV/AIDS 51 STIGMA 53 SEXUAL BEHAVIOR 55 MALE CIRCUMCISION 58 ALCOHOL USE 58 IMPACT 60 AT A GLANCE 60 MACROECONOMIC IMPACT 65 LABOR FORCE 65 DEPENDENCY RATIO 66 HEALTH SECTOR 66 EDUCATION 67 HOUSEHOLDS 69 ORPHANS AND OTHER VULNERABLE CHILDREN 71 AGRICULTURE 72 INDUSTRY 72 MILITARY 73 PRISONS 73 BURIAL 74 RESPONSE 75 AT A GLANCE 75 GOVERNMENT 82 DONORS 87 MINISTRIES OUTSIDE HEALTH 89 HUMAN RIGHTS 90 TRADITIONAL LEADERS 91 TRADITIONAL HEALERS 91 NGOS AND CBOS 91 FAITH-BASED ORGANIZATIONS 92 PMTCT 92 TREATMENT OF OPPORTUNISTIC INFECTIONS AND ART 93 MALE CONDOMS 94 FEMALE-CONTROLLED PREVENTION TECHNOLOGIES 95 ORPHANS AND OTHER VULNERABLE CHILDREN 95 YOUNG PEOPLE 95 VCT 96 HOME-BASED CARE 97 INDUSTRY 98 LINKS 100 Preface This research was undertaken as part of the Country AIDS Policy Analysis Project, which is managed by the AIDS Policy Research Center at the University of California San Francisco. The project is funded by the U.S. Agency for International Development, Cooperative Agreement PHN-A-00-01-00001-00. Stephen F. Morin, PhD, is the project's Principal Investigator. The views expressed in this paper do not necessarily reflect those of USAID. The overarching objective of the Country AIDS Policy Analysis Project is to inform planning and prioritizing of effective and equitable HIV/AIDS prevention and treatment interventions through multidisciplinary research on HIV/AIDS. The project evolved from the acute need for analysis of the epidemiology of HIV/AIDS in tandem with analysis of countries' political economy and sociobehavioral context ¾ at household, sectoral, and macro levels. This multidisciplinary analysis aims to: * help inform national HIV/AIDS policies * strengthen ability to plan, prioritize, and implement effective interventions * highlight the range of sectoral interventions that may affect or be affected by HIV/AIDS * facilitate multisectoral/interministerial coordination * facilitate intercountry information sharing * increase national and subregional capacity for effective partnerships The project develops and disseminates online, easy-to-download, continually updated analyses of HIV/AIDS in 12 USAID Rapid Scale-Up/Intensive Focus/Basic Program countries: Ethiopia, Kenya, Malawi, Senegal, South Africa, Uganda, Tanzania, Zambia, Zimbabwe, Brazil, Cambodia, and India. The primary audience for the country analyses is in-country HIV/AIDS planners, including those from government ministries and agencies, multi- and bilateral donors, international and local NGOs, health care institutions, prevention programs, academia, faith-based organizations, affected communities, and the private sector. International investigators and policymakers also report using the analyses in their work. All country analyses undergo peer review at the AIDS Research Institute of the University of California San Francisco. In addition, two in-country experts from each profiled country serve as peer reviewers. A scientific advisory board also reviews all analyses. Each analysis is linked with national strategic plans for HIV/AIDS prevention, care, and support. Analyses also include extensive links to related resources. An online database comprising 73 HIV/AIDS and socioeconomic indicators for 168 countries and 13 regions is under development and will allow users to conduct a variety of comparative analyses. Project staff are in regular contact with national HIV/AIDS professionals who provide and verify data as needed. Staff continually assess and incorporate new data to maintain the timeliness of the analyses. Note on Data Sources All racial categorizations and nomenclature used in the data sources cited throughout this paper have been maintained; they do not constitute an endorsement of any particular terminology. Acknowledgments The following individuals served as peer reviewers and provided valuable inputs to this paper: Dr. William McFarland and Dr. George Rutherford, University of California San Francisco; and Dr. E.M. Chomba, School of Medicine, University Teaching Hospital, Lusaka, and Chair, Research Ethics, Alternative Remedies & Documentation Committee, Zambia National HIV/AIDS/STD/TB Council. They are not responsible for any errors of fact or judgment. Contact Information Because this analysis is continually updated, comments and suggested sources of new data are welcome and may be sent to the coinvestigator/project director at UCSF's AIDS Policy Research Center: Lgarbus@psg.ucsf.edu Executive Summary Epidemiology In 1986, the Zambian government established the National AIDS Surveillance Committee. The Zambian ANC-based National HIV/STD Sentinel Surveillance System (NSS) was launched in 1990. During the latter half of the 1990s, the dominant trend in Zambia's HIV epidemic has been a decline in HIV prevalence among women ages 15 to 19, largely limited to four sites in Lusaka. Changes in reported sexual behavior in Lusaka appear to be consistent with declining HIV prevalence among 15- 19-year-olds in Lusaka. At the end of 2001, UNAIDS estimated that 1.2 million Zambians were living with HIV/AIDS (estimate range: 930,000 to 1.4 million). Of them, 1 million were adults (ages 15 to 49), with adult prevalence at 21.50 percent. Of adults infected with HIV, 590,000 (59 percent) are women. Zambia's first AIDS case was identified in 1984. AIDS cases among women peak between ages 20 and 29 and among men, between ages 30 and 39, suggesting significant transmission from older males to younger females. The epidemic is driven largely by heterosexual transmission. However, MTCT is significant, accounting for about 30,000 new infections each year. Safe blood product needs are met in Lusaka, but it is unclear whether this is occurring nationwide. Though ANC data are widely used, they are imperfect. For example, comparative studies have shown that HIV prevalence among pregnant women in sub-Saharan Africa underestimates prevalence in women of reproductive age because fertility among HIV-positive women is substantially lower than among uninfected women. A mid-1990s population-based survey in Chelston (Lusaka) and Kapiri-Mposhi found that ANC data tended to overestimate HIV seroprevalence among women ages 15 to 19, with a reverse relationship in those ages 30 to 39. Another factor is that the population attending ANCs varies during the different stages of the epidemic. Poverty, the current food crisis, and other factors may also affect ANC attendance in Zambia. Political Economy and Sociobehavioral Context Zambia achieved independence in October 1964. Until two decades ago, Zambia was one of the most prosperous nations in sub-Saharan Africa. It is now one of the world's poorest countries. Zambia's 2000 per capita income of US$300 is half of what it was at the time of independence in 1964. A World Bank- and IMF-guided structural adjustment program did not generate economic growth, nor did it adequately address impacts on the poor and the need for social safety nets. Other factors contributing to the deteriorating economic situation of the 1990s included the falling price for copper, extremely high debt level, drought, and persistently high population growth rate. According to Zambia's Ministry of Finance and Economic Planning, HIV/AIDS also adversely affected economic growth and exacerbated poverty during the 1990s. Zambia has a long history of men migrating to work in large agricultural estates as well as to the mines in Copperbelt Province. Generally, Zambian men and women have a fairly high level of mobility. Key mobile groups in Zambia include truck drivers, sex workers, fishermen/women and fish traders, migrant and seasonal workers, cross-border traders (especially young girls), miners, military personnel, prisoners, and refugees. The food crisis is also spurring population dislocation. In May 2002, the Government of Zambia declared a national disaster due to actual and anticipated food shortages. About 2.9 million Zambians (29 percent of the population) require food aid. Reports from aid agencies who have recently conducted missions in Zambia concur that although erratic weather has contributed to the current food crisis, one of the key underlying factors is the depletion of human resources as a result of HIV/AIDS. Despite the government's stated commitment to fighting corruption, Zambians do not believe that corruption has diminished. Hospitals and clinics, among others, are perceived to have a high level of corruption. The general health status of Zambia's population worsened substantially during the 1990s. The country bears an enormous burden of malaria. Other major health problems include TB, leishmaniasis, Guinea Worm disease, measles, malnutrition, respiratory diseases, polio, and diarrheal disease. Almost all health facilities lack adequate personnel, drugs, and/or equipment. Physical infrastructure and equipment are deteriorating. There is overlap between the Ministry of Health and the Central Board of Health, leading to duplication and other inefficiencies. As part of structural adjustment, the government and its donor partners began a process of health sector reform in 1994. Serious concerns have been raised about underlying elements of the reform process, particularly the application of user fees. Despite constitutional and legislative provisions, women still experience disadvantages in enforcement of laws regarding property ownership, inheritance, and marriage. Widows may be particularly vulnerable to HIV because of sexual cleansing and wife inheritance. "Grabbing," wherein close relatives take possession of the deceased's household assets, exacerbates the already precarious economic (and social) situation of widows and their children. Despite legislation to address this issue, widows continue to suffer extreme harassment from and impoverishment at the hand of their husbands' relatives. AIDS morbidity and mortality are leading to many girls' (versus boys') being taken out of school to care for family members who are ill and to assist with household tasks and/or generate income as breadwinner die. Girls' lack of education is associated with lower formal sector employment opportunities and thus lessened economic autonomy, which may render them more vulnerable to acquiring HIV. AIDS is also leading to increasing numbers of female orphans and household heads, who are vulnerable to exploitation, including sexual abuse. Women in Zambia have little power in sexual negotiation with their husbands. In many rural areas, HIV/AIDS continues to be viewed as disease of women and, more specifically, of sex workers. Despite Zambians' high knowledge of HIV/AIDS, major gaps in fuller understanding of HIV transmission persist. These are particularly of concern with regard to gender and urban-rural differentials. Misconceptions about HIV/AIDS also persist and perpetuate stigma. Condom use within marriage or other consensual unions is very low and has not risen since 1998. Condom use in nonregular relationships is much higher than for other relationships, but still low. Urban-rural disparities in condom use are often wide and may be related to limited awareness and availability of condoms in rural areas, as well as lower knowledge and higher degree of misperceptions in rural areas. Girls' vulnerability to acquiring HIV infection heightens in relationships with older partners because of a lack of power to negotiate safe sex and the threat of violence. Alcohol use at last sex is reported by about one-fifth of men and of women. Impact During 2000-05, Zambia will have the world's lowest life expectancy at birth: 32.4. Although Zambia's life expectancy is projected to increase, AIDS will reduce life expectancy by 26 to 39 percent through 2050. By 2000, 749,000 Zambians had died because of AIDS, with AIDS having increased the number of deaths in the country by 32 percent. By 2015, AIDS will have increased the number of deaths by 83 percent, bringing the cumulative total of AIDS deaths to 2.8 million. By 2050, 6.2 million Zambians will have died because of the epidemic. In the medium term, Zambia will experience a 5.8 percent reduction in GDP per capita because of HIV/AIDS. The ILO projects that Zambia will lose 19.9 percent of its labor force by 2020 (compared with the labor force size without HIV/AIDS). HIV/AIDS is consuming an increasingly larger share of the health budget. AIDS patients are projected to occupy 45 percent of all hospital beds by 2014. At least 68 percent of adult patients with TB at University Teaching Hospital, the nation's largest hospital, are HIV-positive. High death rates among Zambian teachers are dramatically affecting the supply of education. On the demand side, the number of Zambian children of primary school age will be 20 percent lower by 2010 than pre-AIDS projections. In most cases, households and communities have responded to HIV/AIDS with no external financial or material support. Because of HIV/AIDS, poverty, and food shortages, traditional coping mechanisms in Zambia have become largely irrelevant. To survive, some engage in activities such as sex work or border trading, increasing their risk of exposure to HIV. Households affected by AIDS report annual income 30 to 35 percent lower than unaffected households. Within households, the burden of caring for PWHA usually falls on women. During the 1990s, there was a steady increase in the prevalence of orphanhood in Zambia. At the end of 2001, there were an estimated 570,000 AIDS orphans (ages 0 to 14) living in Zambia. The percent of Zambia's orphans due to AIDS rose from 11.5 percent in 1990 to 65.4 percent in 2001. In most cases in Zambia, extended families take in orphans who have lost both parents. These families are themselves likely to be poor and must therefore stretch already inadequate resources to provide for both orphans and their own children. As the traditional safety net of fostering fails, orphans may be become heads of households and responsible for caring for younger siblings. Orphans are particularly vulnerable to malnutrition, illness, abuse, child labor, and sexual exploitation. Concurrently, they suffer the stigma and discrimination associated with HIV/AIDS. A study conducted in the Konkola Copper Mines found that HIV prevalence ranged from 14.4 to 20.1 percent among employees. Firms are reporting increasing costs of productivity losses, sick leave, retraining, and funerals. A major study found that HIV prevalence among prisoners is 27 percent. Response The first AIDS case was reported in 1984. In 1986, Zambia created the National AIDS Surveillance Committee and National AIDS Prevention and Control Program. In 1987, a short-term emergency plan was established to deal with the blood supply. Two medium-term plans were subsequently created. Constraints to the government's response through the late 1990s included lack of high-level political commitment, strategic management of the HIV/AIDS program, analysis of HIV/AIDS in the context of macroeconomic or gender policy, programs tailored to different populations, implementation evaluation, and intragovernmental collaboration. In 2000, Zambia established the National HIV/AIDS/STD/TB Council to serve as the single, high-level institution responsible for national and technical leadership, strategic management, and effective coordination of all government and civil interventions. The national council is guided by a strategic framework (2001-03), which states that HIV/AIDS must be addressed in the country's overall development ¾ not just health ¾ program. Human and financial resources remain highly inadequate. Zambia relies heavily on donor funding. It is slated to receive US$42 million from the second phase of the World Bank's Multicountry HIV/AIDS Program for Africa (MAP). In April 2002, the Global Fund to Fight AIDS, TB & Malaria approved Zambia's US$92.9 million proposal to address HIV/AIDS. Several Zambian ministries have adopted workplace programs to raise awareness of HIV/AIDS among their staff, train peer educators, and distribute condoms. However, budget constraints are impeding full implementation of these workplans. Another concern is whether sufficient mechanisms for employee confidentiality have been established. The Ministry of Education has less than two full-time equivalent staff addressing the sector's response to HIV/AIDS. In 1993, the Ministry of Defense created the Zambia Defense Forces HIV/AIDS Prevention, Care and Support Program. In March 2003, ZDF announced that it will implement mandatory HIV testing of potential recruits and serving personnel. Testing positive for HIV will automatically disqualify one from joining the military. Serving members with HIV will not be discharged but will be placed in lower categories and offered available medical care. The Zambia AIDS Law Research and Advocacy Network has raised concerns that mandatory HIV testing is discriminatory and a violation of the right to autonomy. There are no HIV/AIDS-specific laws in the country. Stigma and unfair discrimination against PWHA are common in Zambia, but persons discriminated against have no system for redress. Zambian NGOs and CBOs have played a critical role in responding to HIV/AIDS, shouldering much of the country's response to HIV/AIDS. Among their constraints are high dependence on foreign grants and little financial assistance from government, limited coverage and scale, and poor distribution across country. There are national and district interfaith HIV/AIDS working groups to sensitize religious leaders and train clergy and lay religious leaders in counseling and supporting communities. Many church leaders appear to be recognizing that they need to play a greater role in HIV/AIDS prevention and care. However, some still object to HIV prevention messages that include mention of condoms. Some traditional leaders have played a critical role in HIV/AIDS prevention. A prime example is how some chiefs, especially in Southern Province, have been instrumental in modifying sexual cleansing practices. Traditional healers are represented on the National HIV/AIDS/STD/TB Council. Guidelines for conducting research on herbal remedies have been developed, and such research is awaiting funding. Only 2 percent of the population estimated to need PMTCT services received them during 2001. Zambia launched its PMTCT Initiative in 1999, which is now being implemented at six pilot sites. As of January 2002, less 1 percent of Zambians living with HIV/AIDS were receiving ART. The Zambian government has developed national guidelines for ART and is establishing nine provincial treatment centers to provide ART to 10,000 people. In response to the growing number of orphans and OVC, the Zambian government launched the Social Welfare Scheme; however, to date, it reaches only an estimated 10 percent of the target population. Zambia is seeking to establish effective legislation with regard to children, youth, and HIV/AIDS. There are numerous NGO and CBO projects targeted to orphans and OVC. Zambia developed the Code of Ethics and Practice for Counseling to establish standards of competence and conduct for counselors, trainers, and supervisors. The code is reinforced through the more detailed Guidelines on HIV/AIDS Counseling in Zambia, produced by the Ministry of Health in 2000. By 2000, public health centers in 23 districts (out of national total of 72) were providing VCT. Outside government, NGOs and CBOs have been providing VCT for many years. During 2001, 43 percent of the estimated population in need of VCT services in Zambia was receiving them. Zambia was one of the first countries in Africa to implement home-based care. Zambian NGOs, particularly those related to religious organizations, took the lead on HBC and developed a variety of approaches, many of which serve as best practice for other countries. Government has played a very limited role in HBC provision. There are currently over 50 HBC programs in Zambia, primarily found in urban areas and covering at most 20 percent of PWHA. Demand for HBC is enormous, and programs are overwhelmed. A recent study found that HBC is unlikely to increase significantly without greater government involvement. In December 2002, Copperbelt Electricity Corporation announced that it would provide ART to its workers and their spouses. The Zambia Federation of Employers has encouraged its members to assist their workers in accessing ART. However, most Zambians either work in the informal sector or hold low-level, nonunionized positions that do not offer medical assistance. Although some firms have implemented HIV/AIDS prevention and care programs, the majority of Zambian employers have no such policies nor programs. Epidemiology At a Glance The At a Glance section summarizes the more detailed data found below it. HIV Sentinel Surveillance * In 1986, the Zambian government established the National AIDS Surveillance Committee. The Zambian ANC-based National HIV/STD Sentinel Surveillance System (NSS) was launched in 1990. * The same 22 antenatal clinics were involved in Zambia's NSS in 1994, 1998, and 2001, whereas the 1993 NSS involved just 10 sites. There was a major gap in collection of NSS data between 1994 and 1998. As of February 2003, the results from the 2001 NSS had not yet been approved by the Central Board of Health (CBOH). * During the latter half of the 1990s, the dominant trend in Zambia's HIV epidemic has been a decline in HIV prevalence among women ages 15 to 19, largely limited to four sites in Lusaka. These sites showed an average decline in HIV prevalence from 28 percent in 1993 to 15 percent in 1998. * Changes in reported sexual behavior in Lusaka appear to be consistent with declining HIV prevalence among 15- 19-year-olds in Lusaka. * HIV prevalence may be declining among 15-19 in some urban areas outside Lusaka (and in one rural area), though these trends have yet to be validated. * As the CBOH stressed in 1999, the declines mentioned above have not yet resulted in a decline in the country's overall HIV prevalence. However, declines in the 15-19 age cohort are the most encouraging because prevalence in this age group is mostly like to reflect recent trends in HIV transmission. The CBOH does believe that national prevalence is stabilizing, albeit at a very high level. * The urban-rural differential remains wide. In several rural areas, overall HIV prevalence and HIV prevalence among the youngest age cohort appear to be increasing. * Verification of trends is required, including examination of the 2001 NSS and 2001-02 Zambia Demographic and Health Survey findings. Moreover, persistent poverty, economic uncertainty, food crises, and inadequate funding for HIV/AIDS projects are also shaping the epidemic's dynamic, rendering it difficult to pinpoint trends and make projections. UNAIDS Estimates * At the end of 2001, UNAIDS estimated that 1.2 million Zambians were living with HIV/AIDS (estimate range: 930,000 to 1.4 million). Of them, 1 million were adults (ages 15 to 49), with adult prevalence at 21.50 percent (the sixth-highest adult prevalence in the world). * Of adults infected with HIV, 590,000 (59 percent) are women. HIV prevalence among women ages 15 to 24 ranges from 16.78 to 25.18 percent; the comparable range for men in the same age cohort is 6.45 to 9.68 percent. * There were 150,000 Zambian children (ages 0 to 14) living with HIV/AIDS at the end of 2001. AIDS Cases * Zambia's first AIDS case was identified in 1984. * The Ministry of Health estimates that 100,000 Zambians develop AIDS each year. * AIDS cases among women peak between ages 20 and 29 and among men, between ages 30 and 39, suggesting significant transmission from older males to younger females. Transmission Patterns * The epidemic is driven largely by heterosexual transmission. However, MTCT is significant, accounting for about 30,000 new infections each year. * Safe blood product needs are met in Lusaka, but it is unclear whether this is occurring nationwide. AIDS Mortality * By 2000, 749,000 Zambians had died because of AIDS, with AIDS having increased the number of deaths in the country by 32 percent. By 2015, AIDS will have increased the number of deaths by 83 percent, bringing the cumulative total of AIDS deaths to 2.8 million. By 2050, 6.2 million Zambians will have died because of the epidemic. Data Quality Issues * ANC data currently serve as Zambia's primary sentinel surveillance of HIV/AIDS. Though ANC data are widely used, they are imperfect. For example, comparative studies have shown that HIV prevalence among pregnant women in sub-Saharan Africa underestimates prevalence in women of reproductive age because fertility among HIV-positive women is substantially lower than among uninfected women. * A mid-1990s population-based survey in Chelston (Lusaka) and Kapiri-Mposhi found that ANC data tended to overestimate HIV seroprevalence among women ages 15 to 19, with a reverse relationship in those ages 30 to 39. * Another factor is that the population attending ANCs varies during the different stages of the epidemic. Poverty, the current food crisis, and other factors may also affect ANC attendance in Zambia. HIV Sentinel Surveillance With technical assistance from the (then) Global Program on AIDS at WHO, the Zambian government established the National AIDS Surveillance Committee in 1986, part of its National AIDS Control Program. Currently, the Central Board of Health (CBOH) within the Ministry of Health manages the core HIV epidemiological surveillance and research system (Zam Core HIV-EPI), which comprises national sentinel surveillance in antenatal clinics (ANCs), population-based surveys (using saliva-based HIV testing), and research studies.1 The Zambian ANC-based National HIV/STD Sentinel Surveillance System (NSS) began in 1990 and is the main source of data on the epidemic and its dynamics. CBOH validates NSS data using population-based survey data from selected sentinel surveillance sites (these data are discussed in the Data Quality section below).2 Prior to 1993, NSS involved few sites and no information on age.3 The 1993 NSS survey was carried out in 10 sites and did include information on age. 4 However, in analyzing the 1993 data, the CDC reports that the overall prevalence found appears to be correct but that age-specific prevalences are likely to be incorrect. 5 Where available, data for 1993 are included in the discussion below, but should be viewed with this caveat in mind. In 1994, NSS underwent a major expansion to increase geographic representation (27 sites with at least one urban and one rural site in each of Zambia's nine provinces; of the 27 sites, seven overlapped with the 1993 sites). The 1994 NSS also expanded collection of sociodemographic characteristics of participants (in addition to age, marital status, residence, and educational attainment). The NSS indicated overall HIV prevalence of 19.8 percent in 1994.6 The next NSS survey did not occur until 1998. This was primarily because of difficulties in procuring HIV test kits. 7 The 1998 survey involved 22 sites, with Chilonga, Mporokosso, Samfya, Kamuchanga, Chitokoloki omitted; the government believed that the omission of these sites did not affect the geographical representation of the NSS.8 As of February 2003, the results from the 2001 NSS had not yet been approved by the Central Board of Health.9 The 2001 NSS was conducted at the same 22 sites as in 1998; the target of enrolling 400 to 500 women at each site was met at all but two or three sites.10 Zambia is highly dependent on donors to fund HIV sentinel surveillance NSS, although the government covers a substantial portion of costs as well. In addition to the WHO/GPA, Sweden and Norway have been major funders.11 Zambia is a priority country of the Global AIDS Program of the U.S. Centers for Disease Control and Prevention (CDC), and the CBOH is working closely with the CDC on NSS data collection and analysis. The CBOH is considering conducting the NSS at least every two years, if not annually.12 Methods For the 1994 and 1998 NSSs, women attending ANCs for the first time were enrolled consecutively over a four-month period. In 1994, the required sample size of 500 was not achieved at clinics with low population coverage.13 In 1998, a sample size of 500 was sought at all sites except Livingstone, Chelston, Kapiri- Mposhi, and Ndola, where the sample size target was higher to permit more refined age-specific analysis and comparison with population-based studies. 14 In 1994, a total of 11,517 women had blood collected and were interviewed; 15 in 1998, this figure was 12,001. 16 In 1994 and 1998, HIV testing was conducted in an unlinked anonymous fashion using blood samples collected for syphilis screening, part of the routine standard of antenatal care. A venous sample of blood was collected from each ANC attendee. Trained laboratory technicians tested all sera for HIV antibody using Capillus Rapid Test HIV-1, HIV-2. Quality control and confirmatory testing was conducted at two national reference laboratories: Virology Laboratory, University Teaching Hospital, Lusaka, and Immunology Laboratory, Tropical Diseases Research Center, Ndola. 17, 18 In 1994, 50 percent of all HIV antibody reactive sera and 5 percent of nonreactive sera (randomly selected) were retested using RTD. Discordant results were retested using Western Blot or an enzyme-linked, immunosorbent assay (HIV-1, ELISA). No further testing of sera was performed in geographic sites with no discrepancies between the first and second test. False positives were the most common form of discrepancy detected. In all sites where false positives appeared, the remaining 50 percent of reactive samples were retested. In sites where false negatives were detected, the remaining 95 percent of nonreactive samples were retested. In 1998, all positive samples and 4 percent of nonreactive sera (randomly selected) were retested using Wellcozyme HIV Recombinant HIV-1. Samples with discordant results were confirmed using Bionor HIV-1 and 2. For sites with false negatives, over 50 percent of negative samples were retested on Wellcozyme ELISA. 19 Findings In 1998, the sentinel sites with the highest HIV prevalences were: * Livingstone: 31.0 percent (1994: 31.9 percent). A tourist town, as well as major route for truckers and railway. Many women engage in cross-border trading with Victoria Falls in Zimbabwe. Located in Southern Province, which has country's highest rates of polygamous relationships and sexual cleansing. Widow inheritance, wherein close relatives take over possessions of widows, is common in Southern Province. 20 (More detail on sexual cleansing and widow inheritance is found in the Political Economy and Sociobehavioral Context section). * Ndola: 28.4 percent (1994: 27.5 percent). Located in Copperbelt Province. Has drawn men from around the country to work in mines. The privatization of Zambia Consolidated Copper Mines (ZCCM) entailed job losses, as have declining world prices for copper. (More detail found in Political Economy and Sociobehavioral Context section). * Chipata: 27.3 (1994: 30.3 percent) In Eastern Province on border with Malawi; major trucking and trading route.21 * Mongu: 27.3 percent (1994: 28.4 percent) In Western Province on Zambezi River. Starting point for tours of Liuwa Plains National Park. * The lowest prevalence in 1998 was found in Kasaba in Luapula Province (5.2 percent; 1994: 12.0 percent).22 Kasaba is a rural area, sparsely populated, not on a major transport route, and with no major economic activities. Ethnic groups in the area are predominantly matrilineal, and less likely to be polygamous than in other areas of Zambia.23 In 1998, in all sites except Kalabo, Kapiri- Mposhi (discussed below), and Kashikishi, prevalence among those ages 15-19 was lower than that for the age cohorts 20-24, 25-29, and 30-39. (NB: Data for women over age 39 not provided.) 24 The NSS data from the 1990s indicate that the major patterns in the HIV epidemic in Zambia are: HIV prevalence among pregnant women ages 15 to 19 is declining in Lusaka During the latter half of the 1990s, the dominant trend in the HIV epidemic has been a decline in HIV prevalence among women ages 15 to 19, largely limited to four sites in Lusaka. These sites showed an average decline in HIV prevalence from 28 percent in 1993 to 15 percent in 1998. (Though, as mentioned above, the 1993 data may not be a useful baseline; in addition, sample sizes in 1993 were far lower compared with those from 1994 and 1998.) Data for the four Lusaka sites covering 1994 and 1998 are found in figure 1 and table 1. 25 Figure 1. HIV Prevalence Among Pregnant Women Ages 15 to 19, Lusaka, 1993-98 Source: POLICY Project. HIV/AIDS in Southern Africa: Background, Projections, Impacts, and Interventions. Washington, DC: The Futures Group International, October 2001 Table 1. HIV Prevalence among ANC Sentinel Sites in Lusaka, 1993-1998 Site Overall Ages 15 to 19 Year N HIV Prevalence (%) CIs Year N HIV Prevalence (%) Chelston 1998 812 25.9 23.0 - 29.0 1998 199 15.1 1994 462 24.7 20.7 - 28.6 1994 106 21.7 1993 299 26.8 21.7 - 31.8 1993 56 28.6 Chilenje 1998 510 27.3 2.5 - 31.4 1998 114 16.7 1994 456 35.3 30.9 - 28.3 1994 102 29.4 1993 287 22.0 17.1 - 26.8 1993 53 30.2 Kalingalinga 1998 499 26.7 22.9 - 30.8 1998 120 16.7 1994 512 21.7 18.1 - 25.3 1994 113 14.2 1993 442 23.5 19.6 - 27.5 1993 100 26.0 Matero 1998 502 29.1 25.2 - 33.4 1998 123 10.6 1994 394 28.4 24.0 - 32.9 1994 101 24.8 1993 288 27.1 21.9 - 32.2 1993 63 25.4 Source: Zambian Ministry of Health, Central Board of Health. Zambian HIV Sentinel Surveillance: Time Trends in the HIV Epidemic in the 1990s. Lusaka: 1999. Changes in reported sexual behavior in Lusaka appear to be consistent with declining HIV prevalence among 15- 19-year-olds in Lusaka The CBOH does not believe that selection bias is a factor in this trend. Data from representative surveys of urban Lusaka (1990-98) and the country as a whole (1992-98) that examined trends in knowledge and sexual risk behaviors appear to validate this trend. For example: * A decline in premarital sexual activity has been observed in urban Lusaka. In 1990, 50 percent of never-married women reported no sexual experience, compared with 60 percent in 1998 (p = .003); among men, the figures were 38 and 53, respectively (p < .001). Fewer married women (1990: 8 percent; 1998: 2 percent; p < .001) and men (1990: 31 percent; 1998: 19 percent; p = .07) reported extramarital partners. Change in urban Lusaka was observed primarily from 1990 to 1996; the changes in men's behavior observed between 1996 and 1998 were also observed in the national estimates for those years. National figures for other indicators from 1992 to 1998 were less encouraging. Apart from an increase in having ever used condoms, no change in women's sexual behavior was observed. Fewer men had premarital sex from 1996 to 1998 (1996: 64 percent; 1998: 46 percent; p < .001), but condom use with nonregular partners decreased among men (1996: 38 percent; 1998: 29 percent; p = .02). 26 * A comparison of findings from two cross-sectional surveys implemented in Lusaka between 1996 and 1999 found statistically significant reductions in casual partnerships among men and women. Respondents with higher socioeconomic status were more likely to reduce casual sex. Condom use increased but the change was not statistically significant.27 * The 2000 Zambian Sexual Behavior Survey found that (NB: these data are not specific to Lusaka): ? The percentage of all unmarried women and men ages 15-19 who had ever had sex declined between 1998 and 2000. ? In 2000, among all unmarried adolescent women, 41 percent reported condom use with a nonregular partner, whereas this figure was 33 percent for all women. ? In 2000, unmarried urban women ages 15-24 were less likely their rural counterparts to have had sex in the last year and were more likely to have used a condom when they did have sex. They were also less likely than rural women to have had multiple partners in the last year.28 (The 2000 ZSBS was undertaken by the Zambian Central Statistical Office and Zambian Ministry of Health, with support from USAID and the MEASURE Evaluation Project at the University of North Carolina Chapel Hill. It was the second sexual behavior survey undertaken in Zambia, following that of 1998. [Findings from Zambia's 2001 Demographic and Health Survey have not yet been released; it was previously conducted in 1992 and 1996.] For the 2000 ZSBS, a representative national sample of 1,851 households, 2,034 women, and 1,798 men was achieved in the same clusters from which the sample for the 1998 survey was selected. The 2000 ZSBS findings were published in April 2002.) HIV prevalence may be declining among 15-19 in some urban areas outside Lusaka (and in one rural area), though these trends have yet to be validated CBOH also notes that this trend may also be occurring in Ndola, Livingstone, Solwezi, and Kabwe, although this requires further verification, using the findings of the 2001 NSS, among others. A similar dynamic of declining prevalence among 15- to 19-year-olds was found in only one rural site, Macha, where prevalence declined from 10 percent in 1993 to 6.2 percent in 1994 to 5.2 percent in 1998 (overall HIV prevalence was 10.0, 9.1, and 8.0, respectively). Any prevalence declines among 15-19 have not yet resulted in a decline in the country's overall HIV prevalence As the CBOH stressed in 1999, the declines mentioned above have not yet resulted in a decline in the country's overall HIV prevalence. However, declines in the 15-19 age cohort are the most encouraging because prevalence in this age group is mostly like to reflect recent trends in HIV transmission. The CBOH does believe that national prevalence is stabilizing, albeit at a very high level. 29 HIV incidence appears to be decreasing HIV prevalence and trend analysis is impeded by the lack of age-specific data prior to 1993. Prevalence changes within the age group 15-19 might be a better indicator of HIV incidence. CBOH posits that the decline in HIV prevalence in the 15-19 age cohort, seen primarily in Lusaka, is most likely attributable to reduced HIV incidence, which it believes has been declining since the early 1990s as a result of favorable behavior change.30 Further verification is needed, including examination of the 2001 NSS and 2001 ZDHS findings. Moreover, persistent poverty, economic uncertainty, food crises, and inadequate funding for HIV/AIDS projects are also shaping the epidemic's dynamic, rendering it difficult to pinpoint trends and make projections. (All discussed in detail in the Political Economy and Sociobehavioral Context section.) The urban-rural differential remains wide In 1994 and in 1998, the urban-rural HIV prevalence ratio was 2- 2.5/1.31 In 1994, overall HIV prevalence in areas of Lusaka and provincial capitals ranged from 22 to 35 percent. Prevalence among predominantly rural areas averaged 13 percent, though prevalence ranged widely, from 5 to 23 percent. Among those 15- to 19-years old, 11 to 28 percent in urban areas and 4 to 11 percent in rural areas were infected with HIV.32 As discussed above, in some urban settings, preliminary indications are that HIV prevalence is stabilizing, albeit at high levels. In some rural areas, HIV prevalence is still rising. 33 In several rural areas, overall HIV prevalence and HIV prevalence among the youngest age cohort appear to be increasing In two of 10 rural areas surveyed (Kalabo and possibly Kabompo), the 1998 NSS found increasing overall HIV prevalence and increasing HIV prevalence among the youngest age cohort. In Kalabo, overall HIV prevalence increased from 4.9 percent in 1993 to 10.2 percent in 1994 to 12.8 in 1998; among 15- to 19-year-olds, these percentages were 3.7, 7.1, 8.0, respectively. In Kabompo, the small sample size renders data interpretations difficult. In 1994, n=318, the third-lowest sample size among all HIV sentinel sites nationwide; in 1998, n=259, the lowest sample size among all sites. Overall prevalence in Kabompo was 5.0 percent in 1994 and 9.3 percent in 1998. The 1994 survey found a prevalence of 0.0 among 15- to 19-year-olds (n=63); this figure was 4.1 percent in 1998 (n=74). 34 Other In Kapiri-Mposhi (on the highway to Dar es Salaam and a stop on the Tanzania-Zambia Railway Authority), there was marked decline in health outreach services from 1994 to 1998, resulting in a significant decline in rural women attending ANCs. The increase in overall HIV prevalence ¾ 13.0 to 16.5 percent ¾ and among 15- to 19-year-olds ¾ 14.0 to 16.3 ¾ during this period may be related to this selection bias. Data from a follow-up population-based survey in Kapiri Mposhi found a stabilization in HIV prevalence from 1995 to 1998. 35 The government also underscores that in 1998, changes in provision of ANC meant that data collection in some sites was conducted in district health units (versus main hospitals in 1994) and that this may have affected the population covered (e.g., women attending district health units may be more likely to have lower HIV prevalence than those attending main hospitals). For example, it believes that the large decrease in HIV prevalence found in Kasama is likely the result of this dynamic and is viewing results from this site with caution. (In Kasama, overall HIV prevalence fell from 23.8 percent in 1994 to 14.7 percent in 1998; among 15- to 19-year-olds, these figures were 12.2 and 4.0, respectively.) 2001-02 Zambia Demographic and Health Survey Preliminary findings from Zambia's 2001-02 DHS were released in October 2002. This latest DHS round included voluntary HIV and syphilis testing. The findings are not comparable to the NSS and thus cannot be used for trend analysis. Moreover, they reflect prevalence among those who were willing to be tested for HIV. The 2001-02 DHS found that 16 percent of those ages 15 to 49 who agreed to an HIV test were HIV-positive. Prevalence among women (17.8 percent) was higher than among men (12.9 percent). Lusaka, Copperbelt, and Southern provinces had the highest HIV prevalences. Among women, HIV prevalence peaked in the 30-34 age group (29.4 percent), followed by ages 25-29 (25.1 percent), 35-39 (22.6 percent), 40-44 (17.3 percent), 20-24 (16.3 percent), 45-49 (13.6 percent), and 15-19 (6.6 percent). The urban-rural prevalences among women were 26.3 and 12.4 percent, respectively. 36 Among men, HIV prevalence peaked in the 35-39 age group (22.4 percent), followed by ages 30-34 and 40-44 (both 20.5 percent), 45-49 (20.2), 25-29 (15.0 percent), 20-24 (4.4 percent), and 15-19 (1.9 percent). The urban-rural prevalences among men were 19.2 and 8.9 percent, respectively. 37 The 2001-02 DHS reports that almost all individuals who agreed to an HIV test also agreed to be tested for syphilis. Preliminary results indicate that among those ages 15 to 49 who agreed to a syphilis test, 6.4 percent of women and 7.6 percent of men tested positive on RPR and TPHA/Determine. Among both men and women, prevalence was highest among those ages 25-29 (women: 9.3 percent; men: 10.4 percent) and in Copperbelt, Lusaka, and Eastern provinces.38 Once it is released, the final 2001-02 DHS will be integrated into this paper. Detailed background on methodology and sample characteristics will also be included. UNAIDS Estimates At the end of 2001, UNAIDS estimated that 1.2 million Zambians were living with HIV/AIDS (estimate range: 930,000 to 1.4 million). Of them, 1 million were adults (ages 15 to 49), with adult prevalence at 21.50 percent.39 (At the end of 1999, UNAIDS estimated adult prevalence at 19.95 percent.40) Zambia's adult HIV prevalence is the sixth-highest in the world, following that of Botswana (38.8 percent), Zimbabwe (33.7), Swaziland (33.4 percent), Lesotho (31.0 percent), and Namibia (22.5 percent).41 UNAIDS estimates that of adults infected with HIV, 590,000 (59 percent) are women. HIV prevalence among women ages 15 to 24 ranges from 16.78 to 25.18 percent; the comparable range for men in the same age cohort is 6.45 to 9.68 percent. 42 According to UNAIDS, there were 150,000 Zambian children (ages 0 to 14) living with HIV/AIDS at the end of 2001. 43 AIDS Cases Zambia's first AIDS case was identified in 1984.44 AIDS has been a notifiable disease since 1984. Returns for all cases of AIDS diagnosed in health facilities are supposed to be sent to the Ministry of Health. However, due to diagnostic, logistic, and personnel constraints, only a small proportion of AIDS patients are reported.45 The Ministry of Health estimates that 100,000 Zambians develop AIDS each year.46 According to Zambia's National HIV/AIDS/STD/TB Council, AIDS cases among women peak between ages 20 and 29 and among men, between ages 30 and 39, suggesting significant transmission from older males to younger females.47 (See Age Mixing section.) Sexually Transmitted Infections (STIs) Zambia's National STD Control Program was launched in 1980. The main source of epidemiological information from 1980 to 1996 was: 1. a passive reporting system for syphilis, gonorrhea, and pelvic inflammatory disease from outpatient clinics to the Ministry of Health 2. 64 specialized centers through the country 3. 24 sentinel sites that perform syphilis screening of ANC attendees 4. selected studies Since 1996, national STI data have not been readily available. 48 Transmission Patterns The epidemic is driven largely by heterosexual transmission.49 However, MTCT is significant, accounting for about 30,000 new infections each year. 50 Data on transmission via men who have sex with men are unavailable. In 1987, an emergency plan on safe blood supply was launched. All district, provincial, and central referral hospitals have blood transfusion facilities. All blood products used in these health institutions are screened for HIV and syphilis and, to a lesser extent, hepatitis B.51 Safe blood product needs are met in Lusaka, but it is unclear whether this is occurring nationwide.52 A residual risk of transfusion-associated HIV transmission may persist due to incomplete screening is some areas, donations made during the window period, laboratory false negatives, and human error. AIDS Mortality UNAIDS estimated that were 120,000 adult and child AIDS deaths in Zambia during 2001.53 (The comparable figure for 1999 was 99,000.54) It estimates that at least 700,000 Zambian adults and children died because of AIDS since the epidemic's beginning through the end of 2001.55 According to the U.N. Population Division, by 2000, 749,000 Zambians had died because of AIDS, with AIDS having increased the number of deaths in the country by 32 percent. The U.N. Population Division projects that by 2015, AIDS will have increased the number of deaths by 83 percent, bringing the cumulative total of AIDS deaths to 2.8 million. By 2050, 6.2 million Zambians will have died because of the epidemic (tables 2 and 3).56 Table 2. Projected Number of Deaths with and without AIDS, 1980-2000, 2000-2015, and 2015-2050 (Thousands) Period 1980-2000 2000-2015 2015-2050 With AIDS Without AIDS With AIDS Without AIDS With AIDS Without AIDS 3,110 2,361 4,463 2,445 9,677 6,284 Source: Population Division of the Department of Economic and Social Affairs of the United Nations Secretariat. World Population Prospects: The 2002 Revision. Highlights. New York: February 2003 Table 3. Excess Deaths Because of AIDS, 1980-2000, 2000-2015, and 2015-2050 Period 1980-2000 2000-2015 2015-2050 Excess Deaths (Thousands) Percentage Increase Excess Deaths (Thousands) Percentage Increase Excess Deaths (Thousands) Percentage Increase 749 32 2,018 83 3,393 54 Source: Population Division of the Department of Economic and Social Affairs of the United Nations Secretariat. World Population Prospects: The 2002 Revision. Highlights. New York: February 2003 The 2000 ZSBS found that causes of death most commonly reported for all persons in households were tuberculosis (17 percent), malaria (15 percent), other infectious diseases (16 percent), and diarrhea (13 percent). Among adults, tuberculosis was the main cause (32 percent), followed by malaria (14 percent) and other infectious diseases (25 percent). Although over half of deaths were reported as having been associated with prolonged illness, AIDS was rarely mentioned as a cause of death, Given high levels of stigma, it is reasonable to assume that some percent of AIDS-related deaths were reported under other causes such as TB.57 Research undertaken by UNAIDS and WHO found that for Zambia, the HIV-attributable under-5 mortality rate (per 1,000 and corrected for competing causes of mortality) was 33.6 during the 1990s. (Rates among the 39 countries studies ranged from Madagascar [0.2] to Botswana [57.7].) The HIV-related population proportional attributable risk of dying before age 5 (i.e., the proportion of all-cause under-5 mortality attributable to HIV) was 20.7 percent; the average for the 39 sub-Saharan African countries studies was 7.7 percent, ranging from 0.1 percent in Madagascar to 42.4 percent in Botswana. 58 (More information on AIDS mortality is found in the Demographic Impact section.) Data Quality Issues As mentioned, the same 22 antenatal clinics were involved in Zambia's NSS in 1994, 1998, and 2001, whereas the 1993 NSS involved just 10 sites. There was a major gap in NSS data between 1994 and 1998. As Zambia enters its second decade with HIV/AIDS, dynamics continue to change. For example, the population attending ANCs will vary during the different stages of the epidemic. 59 Poverty, the current food crisis, and other factors may also affect ANC attendance. ANC data currently serve as Zambia's primary sentinel surveillance of HIV/AIDS. Though ANC data are widely used, they are imperfect (see box 1). Comparative studies have shown that the HIV prevalence among pregnant women in sub-Saharan Africa underestimates prevalence in women of reproductive age because fertility among HIV-positive women is substantially lower than among uninfected women. 60 For example, Gregson et al. have found 25 to 40 percent lower fertility in women with HIV in high-prevalence African countries; they attribute about half of this "subfertility" directly to HIV infection.61 Researchers from Uganda and Europe adjusted HIV prevalence among ANC attendees in Chelston (Lusaka) (1994, 1996, 1998), and Ndola (1998). They found that the nonadjusted HIV prevalence among ANC attendees underestimated HIV prevalence among the general female population by 8.0 percent in Chelston in 1998 and by about 22 percent in all other cases.62 In 1995-96, researchers from Zambia's University Teaching Hospital, Tropical Diseases Research Center, and National AIDS Program undertook a population-based survey (PBS) to assess the representativeness of ANC data. The PBS was carried out in Chelston (Lusaka) and Kapiri-Mposhi (rural area). Adults over age 15 were selected through stratified random cluster sampling (n=4,195). Saliva-based HIV tests were used. The researchers found that in both sites, ANC data tended to overestimate HIV seroprevalence among women ages 15 to 19, with a reverse relationship in those ages 30 to 39.63 Box 1. HIV Sentinel Surveillance: Evaluating Data from Antenatal Clinics In many developing countries, estimates on the magnitude of and trends in the HIV epidemic are obtained through HIV seroprevalence surveys. These surveys are primarily conducted using sentinel populations. The most frequently used sentinel populations are women attending antenatal clinics and persons attending clinics for diagnosis and treatment of sexually transmitted infections. The objectives of sentinel seroprevalence surveys include: 1. obtaining information on the prevalence of HIV infection in the sentinel population 2. monitoring trends in HIV prevalence in the sentinel population 3. providing information for estimating future number of AIDS cases 4. providing information for program planning and evaluation of interventions Seroprevalence surveys are usually conducted annually at preselected clinics or hospitals. Surveys of women attending antenatal clinics can provide a reasonable estimate of HIV prevalence within the general population. The surveys are conducted among women ages 15 to 49 years attending the antenatal clinic for the first time during a current pregnancy. Surveys are usually conducted in an unlinked manner, in which serum remaining from routine syphilis screening is tested for HIV infection after all personal identifying information is removed from the specimen. Sampling is usually conducted during an 8- to 12-week period, and all eligible women are sampled consecutively until the desired sample size is achieved. In general, samples of 250 and 400 women are usually sufficiently large as to provide reasonable estimates of HIV prevalence over time. Although these surveys are extremely useful, there are several limitations to consider when interpreting the survey results. The surveys are not based upon a probability sample and therefore may not be representative of the population as a whole. True population-based surveys have found antenatal clinic data may overestimate or underestimate HIV prevalence. Moreover, the ANC studies do not provide information on mortality or HIV-associated morbidity. In addition, although monitoring trends in HIV prevalence provide information on the magnitude of the HIV epidemic, trends in prevalence cannot be relied upon to indicate trends in HIV incidence. However, examining trends in HIV prevalence in younger populations, particularly 15- to 19-year-olds, may provide some indication of trends in recently acquired HIV infection , as this group is unlikely to have been infected for a long period of time. Prepared by Sandy Schwarcz, MD, MPH Director, HIV/AIDS Statistics and Epidemiology Section, San Francisco Department of Public Health Adjunct Assistant Professor, Department of Epidemiology and Biostatistics, University of California San Francisco Political Economy and Sociobehavioral Context At a Glance The At a Glance section summarizes the more detailed data found below it. * Many of the factors discussed in this section exist in countries that, unlike Zambia, have low HIV prevalence; these include poverty, gender inequality, and history of colonialism and political and economic disenfranchisement. The relationship between HIV prevalence and socioeconomic factors is highly complex. Increasingly, risk of HIV infection is recognized as related to, inter alia, one's socioeconomic status as well as the socioeconomic profile of the community in which one is situated. Additionally, the country's current food crisis affects and is affected by HIV/AIDS. Economy * Zambia achieved independence in October 1964. It had considerable financial resources at its disposal, given its mineral wealth. However, it also faced major challenges. The legacy of colonialism resulted in few trained and educated Zambians available to run the government, with the economy largely dependent on foreign expertise. * Zambia embarked on a major program of developing its social, physical, and economic infrastructure. Education was made compulsory, and health services were provided free of charge. * Until two decades ago, Zambia was one of the most prosperous nations in sub-Saharan Africa. It is now one of the world's poorest countries. In 1994, per capita income was estimated at one-third its value in 1978. Zambia's 2000 per capita income of US$300 is half of what it was at the time of independence in 1964. * A World Bank- and IMF-guided structural adjustment program did not generate economic growth, nor did it adequately address impacts on the poor and the need for social safety nets. * Other factors contributing to the deteriorating economic situation of the 1990s included the falling price for copper, extremely high debt level, drought, and persistently high population growth rate. * According to Zambia's Ministry of Finance and Economic Planning, HIV/AIDS also adversely affected economic growth and exacerbated poverty during the 1990s. Trends in Public Expenditures * During the 1990s, Zambia's military expenditures fell dramatically, from 3.7 percent of GDP to 0.6 percent in 2000. * Zambia's public expenditure on health increased from 2.6 percent of GDP in 1990 to 3.6 percent in 1998. * Public expenditure on education fell from 3.1 percent of GDP during 1985-87 to 2.2 percent during 1995-97. However, the percent of the government education budget allocated to tertiary education rose during this period, whereas allocations to primary and secondary education fell. Zambia's net primary enrollment ratio declined from the mid-1980s to the end of the 1990s; secondary school enrollment ratios have also been falling. In addition to reduced government expenditures, these declines are likely due to AIDS mortality and poverty. Debt * In 1986, Zambia's total debt servicing accounted for 28.5 percent of its gross national income, rendering it one of the world's most indebted nations. * In December 2000, Zambia qualified for debt relief under the Heavily Indebted Poor Countries Initiative (HIPC). HIPC is not debt cancellation; rather it is a restructuring of debt repayment through provision of grants. Zambia was granted US$3.8 billion as total debt relief from all its creditors. * To date, HIPC has disbursed US$135 million to Zambia, of which US$18 million has been channeled into health and education. * Many international NGOs and social justice groups express concern that HIPC has reshuffled Zambia's enormous debt rather than cancel it. After receiving debt relief, Zambia will need to continue borrowing to maintain debt servicing obligations and to purchase key imports. They also express concerns that the assumptions underlying HIPC are unrealistic. Governance * Despite the government's stated commitment to fighting corruption, Zambians do not believe that corruption has diminished. Hospitals and clinics as well as the Lusaka City Council, among others, are perceived to have a high level of corruption. * There were numerous allegations of fraud during the 2001 elections. Currently, a case regarding the legitimacy of President Mwanawasa's election victory is before the Zambian Supreme Court. * In February 2003, former president Frederick Chiluba was arrested and charged with over 50 counts of theft and abuse of public office. Declining Human Development * Zambia's Human Development Index value, which declined during the 1990s, is 0.433, placing Zambia among "low human development" countries. * This decline reflects the fall in national income per capita and decreased spending on education. The government's spending on health rose during the 1990s, but clearly the enormous impact of AIDS mortality drastically reduced the life expectancy component of the HDI value. * A critical indicator of the well-being of children is the under-five mortality rate. For 2000, UNICEF reports that Zambia had the world's 10th-highest under-five mortality rate: 202 deaths per 1,000 live births. Moreover, Zambia's under-five mortality rate for 2000 (202) exceeded that of all the least-developed countries (161) and of sub-Saharan Africa (175). Population Mobility * Zambia has a long history of men migrating to work in large agricultural estates as well as to the mines in Copperbelt Province. * More generally, Zambian men and women have a fairly high level of mobility. Poverty may lead to increased migration, both within Zambia and to other countries, as people move from rural to urban areas in search of work ¾ or return to families if they lose their jobs or fall ill and cannot afford care. * The food crisis is increasing migration to towns in Central and Eastern provinces. * Key mobile groups in Zambia include: ? truck drivers ? sex workers ? fishermen/women and fish traders ? migrant and seasonal workers ? cross-border traders, especially young girls ? miners ? military personnel ? prisoners (in the sense that they often return to their families/communities upon release) ? refugees Food Crisis * On May 28, 2002, the Government of Zambia declared a national disaster due to actual and anticipated food shortages. About 2.9 million Zambians (29 percent of the population) require food aid. * The most vulnerable groups are located in Southern and Western provinces and include the elderly and households headed by children and women; those with persons who are disabled or ill; and those with widows not supported by other households. * Reports from aid agencies who have recently conducted missions in Zambia concur that although erratic weather has contributed to the current food crisis, one of the key underlying factors is the depletion of human resources as a result of HIV/AIDS. Unsustainable debt and deteriorating public services have also played a role. * Oxfam highlights that after years of World Bank- and IMF-supported agricultural sector reforms, Zambia still faces chronic food insecurity. It argues that these reforms were imposed too rigidly and too quickly, often leaving poor farmers without support from or access to either state or market institutions. Health Sector * The general health status of Zambia's population worsened substantially during the 1990s. * The country bears an enormous burden of malaria. Other major health problems include TB, leishmaniasis, Guinea Worm disease, measles, malnutrition, respiratory diseases, polio, and diarrheal disease. * Almost all health facilities lack adequate personnel, drugs, and/or equipment. Physical infrastructure and equipment are deteriorating. * Erratic distribution methods lead to frequent shortages of essential drugs and medical supplies in hospitals and health centers. * Conditions (low pay, inadequate equipment, poor supplies) in Zambia's public hospitals have deteriorated and have led to long-running strikes by health care providers. * There is overlap between the Ministry of Health and the Central Board of Health, leading to duplication and other inefficiencies. Health Sector Reform * As part of structural adjustment, the government and its donor partners began a process of health sector reform in 1994. Serious concerns have been raised about underlying elements of the reform process, particularly the application of user fees. * Zambia's National HIV/AIDS/STD/TB Council raises concerns about how decentralization has rapidly shifted the burden of HIV/AIDS to all districts across the country in a "one size fits all" approach. * Poverty, debt burden, falling copper prices, and the enormous impact of HIV/AIDS have also affected the health sector. Sexual and Reproductive Health * UNFPA ranks Zambia a category "A" country, meaning that it is furthest from achieving the sexual and reproductive health goals of the International Conference on Population and Development (ICPD), held in Cairo in 1994. * Zambia's maternal mortality ratio is 870 deaths per 100,000 live births. * Young women bear a high burden of fertility. STIs * Herpes simplex virus-2 is highly prevalent in young women shortly after they begin sexual activity. * Between 1998 and 2000, the proportion of Zambians who used a condom during sexual acts when they knew they were infected with an STI declined substantially, from 36 to 23 percent for women and 46 to 24 percent for men. Gender * HIV prevalence among Zambian women ages 15 to 24 ranges from 16.78 to 25.18 percent, whereas the comparable range for men in the same age cohort is 6.45 to 9.68 percent. * Despite constitutional and legislative provisions, women still experience disadvantages in enforcement of laws regarding property ownership, inheritance, and marriage. * A key component of women's poverty is their inability to obtain loans from banking institutions, which renders them less able to enter into safe, profitable economic activities. * Widows may be particularly vulnerable to HIV because of sexual cleansing (kusalazya) and wife inheritance (kunjilila mung'anda). * "Grabbing", wherein close relatives take possession of the deceased's household goods, land, livestock, clothes, and other assets, exacerbates the already precarious economic (and social) situation of widows and their children. Despite legislation to address this issue, widows continue to suffer extreme harassment from and impoverishment at the hand of their husbands' relatives. Burden of AIDS * AIDS morbidity and mortality are leading to many girls' (versus boys') being taken out of school to care for family members who are ill and to assist with household tasks and/or generate income as breadwinner die. * Girls' lack of education is associated with lower formal sector employment opportunities and thus lessened economic autonomy, which may render them more vulnerable to acquiring HIV. * AIDS is also leading to increasing numbers of female orphans and household heads, who are vulnerable to exploitation, including sexual abuse. * Studies have found that men are more likely to share their HIV status with their wives (in the expectation of a supportive response), whereas women are much less likely to disclose their HIV-positive status to husbands for fear that this might precipitate divorce or violence. Sexual Negotiation * Women in Zambia have little power in sexual negotiation with their husbands. * Polygamy, more common in rural than urban areas, also influences sexual negotiation. In 1996, 17 percent of married women in Zambia were in polygamous unions. Perception of Women * In many rural areas, HIV/AIDS continues to be viewed as disease of women and, more specifically, of sex workers. * Women's own self-perception and definition of what constitutes acceptable female behavior are also factors. Sexual Violence * Sexual assault of girls in Zambia is widespread, including abuse of female orphans by male guardians. There are numerous institutional and sociocultural barriers to reporting and prosecuting such abuse. Knowledge of HIV/AIDS * Despite Zambians' high knowledge of HIV/AIDS, major gaps in fuller understanding of HIV transmission persist. These are particularly of concern with regard to gender and urban-rural differentials. Misconceptions about HIV/AIDS also persist and perpetuate stigma. Stigma * HIV/AIDS-related stigma in Zambia is a massive problem. The most extreme forms of stigma have been found in homes and in health clinics, where the most intensive AIDS care is provided. * Adolescents exhibit even less acceptance of people with HIV/AIDS than do adults. Sexual Behavior * Condom use within marriage or other consensual unions is very low and has not risen since 1998. * Condom use in nonregular relationships is much higher than for other relationships, but still low. * Urban-rural disparities in condom use are often wide and may be related to limited awareness and availability of condoms in rural areas, as well as lower knowledge and higher degree of misperceptions in rural areas. * Girls' vulnerability to acquiring HIV infection heightens in relationships with older partners because of a lack of power to negotiate safe sex and the threat of violence. Zambian men are a median of 4.8 years older than women in all nonmarital/noncohabiting sexual partnerships. The median age difference between married men and their extramarital partners is 5.6 years. Transactional Sex * Among unmarried Zambian women ages 15 to19, 38 percent report having recently received money or gifts in exchange for sex. Male Circumcision * Male circumcision is generally low (except in Northwestern Province). Some observational studies from sub-Saharan Africa have indicated that male circumcision may reduce the risk of HIV acquisition, though circumcision does not appear to affect transmission from HIV-positive men to their partners. Alcohol Use * Alcohol use at last sex is reported by about one-fifth of men and of women. Among young people, ever having used alcohol and drugs is a risk factor for ever having had sex, having more sexual partners over lifetime, and having more than one partner during the last three months. Many of the factors discussed in this section exist in countries that, unlike Zambia, have low HIV prevalence; these include poverty, gender inequality, and history of colonialism and political and economic disenfranchisement. The relationship between HIV prevalence and socioeconomic factors is highly complex. Increasingly, risk of HIV infection is recognized as related to, inter alia, one's socioeconomic status as well as the socioeconomic profile of the community in which one is situated.64, 65 Additionally, the country's current food crisis affects and is affected by HIV/AIDS. This section does not seek to demonstrate causality; rather, it aims to analyze key political economy and sociobehavioral contextual elements to highlight the range of sectoral policies and interventions that may affect or be affected by HIV/AIDS. Postcolonial Context Zambia achieved independence in October 1964. The major figure in Zambian politics from 1964 to 1991 was Kenneth Kaunda, who had led the fight for independence and traditionally bridged the rivalries among the country's various regions and ethnic groups. A new constitution introduced in August 1973 provided for a strong president and a unicameral National Assembly. In 1973, Kaunda, head of the United National Independence Party (UNIP), was the sole candidate in presidential elections. Upon his election, all other political parties were banned. President Kaunda's mandate was renewed in December 1978 and October 1983 in a "yes" or "no" vote on his candidacy.66 After attaining independence, Zambia had considerable financial resources at its disposal, given its mineral wealth. However, it also faced major challenges. The legacy of colonialism resulted in few trained and educated Zambians available to run the government, with the economy largely dependent on foreign expertise. Three of Zambia's neighbors ¾ Southern Rhodesia (now Zimbabwe) and the Portuguese colonies of Mozambique and Angola ¾ remained under white-minority rule. In addition, Zambia shared a border with South African-controlled South-West Africa (now Namibia). Zambia's sympathies lay with forces opposing colonial or white-minority rule, particularly in Southern Rhodesia. During the next decade, Zambia actively supported movements such as the Union for the Total Liberation of Angola (UNITA), the Zimbabwe African People's Union (ZAPU), the African National Congress of South Africa (ANC), and the South-West Africa People's Organization (SWAPO). Civil war in postindependence Angola and Mozambique spurred a flow of refugees into Zambia and caused transport problems. Zambia's strong support for the ANC, which had its external headquarters in Lusaka, created security problems as South Africa raided ANC targets in Zambia. 67 Despite this subregional instability, Zambia embarked on a major program of developing its social, physical, and economic infrastructure. Education was made compulsory, and health services were provided free of charge.68 In December 1990, at the end of a tumultuous year that included riots in Lusaka and a coup attempt, President Kaunda signed legislation ending UNIP's monopoly on power. In response to growing popular demand for multiparty democracy, and after lengthy, difficult negotiations between the Kaunda government and opposition groups, Zambia enacted a new constitution in August 1991. Growing opposition to UNIP's monopoly on power led to the rise in 1990 of the Movement for Multiparty Democracy (MMD). The MMD included prominent UNIP defectors and labor leaders. Zambia's first multiparty elections for parliament and the presidency since the 1960s were held in October 1991. MMD candidate Frederick Chiluba carried 81 percent of the vote to win over Kaunda. The MMD won 125 of the 150 parliamentary seats. The newly liberalized political environment heralded comprehensive economic reforms. 69 By the end of Chiluba's first term as president (1996), the MMD's commitment to political reform had faded in the face of reelection demands. Several prominent MMD supporters founded opposing parties. Relying on the MMD's overwhelming majority in Parliament, President Chiluba pushed through constitutional amendments in May 1996 that barred former President Kaunda and other prominent opposition leaders from the 1996 presidential elections (for example, a retroactive two-term limit and a requirement that both parents of a candidate be Zambian-born). In the presidential and parliamentary elections held in November 1996, Chiluba was reelected, and the MMD won 131 of the 150 seats in the National Assembly. Kaunda's UNIP party boycotted the parliamentary polls to protest the exclusion of its leader from the presidential race, alleging in addition that the outcome of the election had been predetermined due to faulty voter registration. Other opposition parties as well as NGOs also declared the elections neither free nor fair. As President Chiluba began his second term in 1997, the opposition continued to reject the results of the election amid international efforts to encourage the MMD and the opposition to resolve their differences through dialogue. 70 Early in 2001, supporters of President Chiluba mounted a campaign to amend the constitution to enable Chiluba to seek a third term. Civil society, opposition parties, and many members of the ruling party exerted sufficient pressure on Chiluba to abandon this attempt. 71 Zambia held its next presidential, parliamentary, and local government elections in December 2001, with both international and local monitoring. Levy Mwanawasa of the MMD won the presidential election by a narrow margin. The opposition, which charged the MMD with voting irregularities, now holds just over 50 percent of seats in Parliament.72 Transparency International reports that EU observers substantiated allegations of electoral fraud. Currently, a case regarding the legitimacy of President Mwanawasa's election victory is before the Supreme Court.73 In February 2003, former president Chiluba was arrested and charged with over 50 counts of theft and abuse of public office.74 Economy Landlocked Zambia's economy is heavily dependent on copper, cobalt, and zinc mining. Copper is the country's primary export commodity; copper and other metal exports account for about 75 percent of the country's foreign exchange earnings.75 Most Zambians are subsistence farmers. 76 In 2000, agriculture accounted for 85 percent of total employment (formal and informal). Maize (corn) is the main cash crop as well as staple food.77 Other important agricultural products include sorghum, rice, groundnuts, sunflower seeds, vegetables, horticultural products, tobacco, cotton, sugarcane, livestock, coffee, and soybeans.78 As mentioned, immediately following independence, there were improvements in Zambia's economy (and human development, see below). However, in the mid-1970s, the price of copper severely declined, reducing Zambia's export earnings; concurrently, oil prices rose drastically, increasing the country's import bill. This resulted in a severe balance-of-payment crisis and economic growth began to stagnate.79 From the late 1980s to the late 1990s, economic growth declined. In 1994, per capita income was estimated at one-third its value in 1978.80 Per capita income fell almost 5 percent annually between 1974 and 1990.81 During the 1990s, per capita income fell from US$450 to US$300. (During the 1990s, per capita income for sub-Saharan Africa (from US$550 to US$470) and low-income economics (from US$430 to US$410) also declined.) Zambia's 2000 per capita income of US$300 is half of what it was at the time of independence in 1964.82 It is also is well below the average for all low-income economies (US$410) as well as for the sub-Saharan Africa region (US$470). 83 These declines can be explained by numerous factors, including the falling price for copper, extremely high debt level, drought, persistently high population growth rate, and macroeconomic policies. Although its past effect on income per capita has not been quantified ¾ and certainly encompasses a complex interplay with the factors just mentioned ¾ HIV/AIDS was likely a major factor in income decline during the 1990s (for analyses that have been undertaken on the impact of HIV/AIDS on the Zambian economy, see the Impact section). Moreover, the enormous impact of HIV/AIDS can be inferred from the fact that Zambia's income continues to trial even the rest of the world's poorest countries. In 1992, President Chiluba's government initiated an IMF-guided structural adjustment program (SAP), which included privatization of most parastatals, whose losses represented an enormous drain on public revenues and thus a diversion of public resources for investment in public and propoor expenditures 84 To date, about 300 state-owned enterprises have been commercialized or privatized, including the Zambia Consolidated Copper Mines (ZCCM), which was a condition that had to be met before donors would resume their balance-of-payment support. 85 The public sector still represents 44 percent of total formal employment, and telecommunications and electricity parastatals still await privatization.86 Recently, the government has stated that it does not intend to privatize the power utility ZESCO, the telephone company ZAMTEL, and the Zambia National Commercial Bank.87 (Privatization of this last is a condition of debt relief, see Debt section below.) Following ZCCM's privatization in 2000, the Zambian economy began to show some signs of recovery. In 2001, Zambia recorded its first year of increased productivity since 1973.88 In December 2001, the World Bank believed ¾ though cautiously ¾ that Zambia's economic outlook had improved with the sale of ZCCM, fiscal reforms, improved copper prices and production, HIPC interim debt relief (see below), and favorable weather conditions. It did caution that a possible slowdown in world demand for and the price of copper could threaten growth and income. 89 Indeed in early 2002, the mining conglomerate Anglo American, citing low copper prices, announced that it was withdrawing from Zambia's Konkola copper mines, a decision expected to shrink the economy as job losses increase and export earnings fall.90 In Zambia, as in many African countries, the loss of a job can have devastating secondary effects, as numerous family and household members may be supported by the salary of one formal sector employee. (About 105,000 formal sector jobs have been lost through privatization over the last 10 years.91) Moreover, favorable weather did not ensue, as the 2000-2001crop failure resulted in a massive maize shortage. The Zambia National Farmers Union anticipates more grain shortages in 2002 following erratic rains in the 2001-02 growing season.92 Officials from the World Food Program, among others, have been stressing that HIV/AIDS has exacerbated (and been exacerbated by) the food crisis, rendering it more acute and complex than past crises (discussed in detail below). From a purely macroeconomic standpoint, the food shortage diverts the government's scarce resources to maize importation, rather than investment in job creation and social services ¾ investments that are intertwined with HIV/AIDS prevention and treatment. Structural Adjustment In its 2002 Poverty Reduction Strategy Paper (which some claim is simply a new name for SAPs), Zambia's Ministry of Finance and National Planning states that SAP implementation was: ...often piecemeal and failed to fundamentally alter the economic structure. In particular, the design and implementation of SAP often failed to sufficiently address the poverty challenges that increasingly became evident as the structural changes took hold.93 In an analysis of structural adjustment loans between 1980 and 1999, Dr. William Easterly of the Center for Global Development states that: This intensive adjustment lending group includes some notable disasters. Zambia received 18 adjustment loans but had sharply negative growth, large current account and budget deficits, high inflation, a high black market premium, massive real overvaluation, and a negative real interest rate....These results do not prove that adjustment lending was ineffective in promoting good macroeconomic policies and good growth outcomes. It may be that performance would have been even worse without intensive adjustment lending. However, these results place bounds on our intuition on the counterfactual outcomes. It is necessary to believe that a worst case scenario like Zambia would have had even more negative growth, even higher inflation, even more extreme overvaluation and black market premiums, and even more financial repression without repeated adjustment lending than it did with repeated adjustment lending....The adverse selection of repeated failures is a plausible description of what happened in many countries, but this raises questions about why the Fund and Bank make new loans to countries that have failed to deliver reform in response to old loans. Putting external conditions on governments' behavior through structural adjustment loans has not proven to be very effective in achieving widespread policy improvements or in raising growth potential. If the original objective was "adjustment with growth", there is not much evidence that structural adjustment lending generated either adjustment or growth.94 Easterly goes on to state that "The IMF and World Bank declaring a country eligible for debt relief is an admission that past loans, including adjustment loans, did not bring enough current account adjustment and export and GDP growth in that country to keep debt ratios within reasonable bounds."95 (See Debt section below.) In a paper commissioned by the World Bank, researchers from the Chr. Michelsen Institute in Norway and Michigan State University argue that both the Zambian government and donors: have failed to express a coherent strategy of economic growth. Instead, both the government and the donors have made fiscal austerity an end in itself and a measure of reform commitment....As a result, a partial reform syndrome characterized by uneven implementation and limited commitment to policy reform has been supported by the inability of donors to apply the conditionality instruments in a coherent manner.96 Oxfam notes that SAPs exacerbated the exclusion of the poorest from the market while further undermining human development and food security. 97 Poverty Until two decades ago, Zambia was one of the most prosperous nations in sub-Saharan Africa. 98 It is now one of the world's poorest countries. Despite early human development accomplishments and general freedom from conflict since independence (although Human Rights Watch and Amnesty International voiced some concerns over government actions during elections during the 1990s), Zambia contends with worsening poverty. Currently, 73 percent of Zambians are living below the poverty line. Poverty is more prevalent in rural areas compared to urban areas (83 percent and 56 percent respectively) but has risen faster in urban areas due to failing industries and rising unemployment. Most of the rural poor are small-scale farmers.99 According to Zambia's Ministry of Finance and National Planning: In the 1990s, the HIV/AIDS pandemic and other diseases have worsened the poverty situation. At the time when resources were already low, HIV/AIDS has increased the disease burden beyond the individual level to adversely impact on the economics of the family, the health system, the working environment as well as human capital and many others. Principally, AIDS threatens the country's capacity building efforts because it strikes the educated and skilled as well as the uneducated. Consequently, it reverses and impedes the country's capacity by shortening human productivity and life expectancy. The long periods of illness of the skilled personnel in employment has translated into severe loss in economic productivity, which leads to considerable loss to the employer in lost person-hours. The complex relationship between economic growth and HIV/AIDS is increasingly being recognized: the epidemic is as much likely to affect economic growth as it is affected by it.100 Shortly after taking office in 2002, President Mwanawasa's government took some concrete steps to reduce poverty. Among other things, it has eliminated cost-sharing fees in primary schools and set aside resources for free inputs for the poorer peasant farmers.101 It remains unclear, however, whether the funding needed to underwrite these initiatives will be found. Governance As mentioned above, there were numerous allegations of fraud during the 2001 elections. Zambia's Auditor General reported that funds from ZESCO were diverted to fund the ruling party's campaign during the run up to the 2001 elections.102 In February 2003, former president Frederick Chiluba was arrested and charged with over 50 counts of theft and abuse of public office. Several watchdog organizations, such as Freedom House, have highlighted that Zambia's privatization process has been marked by corruption.103 In a case study of the privatization of the Luanshya/Baluba Mine, Transparency International found that: * "The President and his Cabinet contravened the law in conducting the privatization outside the provisions of the Privatization Act. * The Donor Community, including the World Bank, exerted undue influence on the Zambia Government to privatize quickly, resulting in costly mistakes being made. * Multinational Corporations, such as Anglo-American Corporation, took advantage of the government vulnerability to drive a hard bargain. * The Government has been unable to account for the sales proceeds of ZCCM assets. * Inadequate attention was given to the social impact of the sale of the mines resulting in untold human suffering. * The sale of the mines was compromised to some extent by the self-interest of those charged with the disposal of the mines. * The privatization exercise as a whole has lacked transparency and has been characterized by open disregarded of the law, particularly the Privatization Act."104 In a recent report, Transparency International examined Zambians' perception of the prevalence of corruption in institutions with which they have daily dealings, e.g., police, schools, and local courts. TI found that despite the Zambian government's stated commitment to fighting corruption, hospitals and clinics, the Lusaka City Council, the revenue authority, and the customs office were all perceived to have a high level of corruption. Financial institutions such as commercial banks also ranked very low in the TI survey. The public did not believe that corruption levels had diminished. 105 Trends in Public Expenditures In a 2001 analysis of public expenditures in Zambia, the World Bank expressed concern that Zambia's overall public sector deficit ¾ which includes the deficits of the central government, local governments, extrabudgetary accounts, state-owned enterprises, and the Central Bank ¾ remains very high and continues to threaten macroeconomic stability, growth, and poverty reduction. Moreover, the beneficiaries of public expenditure have generally not been the poor. 106 The accompanying indicator table includes some information on trends in public expenditures. During the 1990s, Zambia's military expenditures fell dramatically, from 3.7 percent of GDP to 0.6 percent in 2000. However, public expenditure on education (as a percent of GNP) fell from 3.1 percent during 1985-87 to 2.2 percent during 1995-97. However, the percent of the government education budget allocated to tertiary education rose during this period, from 18.3 to 22.3 percent. The increased allocation to tertiary education ¾ while allocations to primary (43.9 to 41.5) and secondary education (26.9 to 18.4) fell ¾ is of concern. In Zambia, the poor do not benefit from the substantial subsidies to university education. Children attending primary school are more likely to be from the poorest groups than from better-off groups; those who are not yet or able to be in school are also disproportionately poor. Secondary education shares this pattern, though the effect is not as pronounced as in primary education. UNDP notes that Zambia's net primary enrollment ratio declined from the mid-1980s to the end of the 1990s (88 to 73 percent).107 Secondary school enrollment ratios have also been falling.108 These declines are likely due to AIDS mortality as well as reduced government expenditures and poverty. Declining education spending and outcomes are particularly worrying, as several studies have found that education can play a protective role vis-à-vis HIV. For example, Family Health International interviewed 2,328 youth ages 10 to 24 years residing in greater Lusaka. Using these data on self-reported behaviors, FHI found that attaining higher levels of education and being currently enrolled in school were associated with a lower likelihood of ever having had sex among females and, for both genders, with a lower likelihood of having had multiple recent sexual partners, as well as a higher likelihood of consistent condom use. 109 Zambia's public expenditure on health increased from 2.6 percent of GDP in 1990 to 3.6 percent in 1998. 110 The beneficiaries of health expenditures are generally more complex and less clear than those of education spending. For example, the largest group using public primary health facilities tends to be in the middle of the income distribution. Despite increased public health expenditures ¾ at least through 1998, the last year for which data were available from UNDP¾ health outcomes have been deteriorating. This is likely the result of a combination of factors, including the burden of HIV/AIDS on the health sector as well as inefficiencies within the sector. For example, there are two national health agencies: the Ministry of Health and the Central Board of Health. In principle, MOH formulates policies, whereas CBOH implements them. Health workers are expected to be CBOH employees, subject to uniform conditions of service and disciplinary code. However, some health workers are subject to MOH's conditions of service. This scenario leads to work disruptions through strikes, as health workers compare conditions of service. 111 There is also a lack of transparency and expenditure wastage with regard to drug procurement.112 Although a purchasing department exists within CBOH and although CBOH's director general is a member of the tender board, MOH continues to purchase drugs on behalf of CBOH, an inefficient system that does not respond to CBOH's priorities. Efforts are under way to change this situation.113 Debt In the mid-1970s, as the price of copper plummeted and oil prices rose, Zambia turned to foreign and international lenders. It external debt more than doubled in 1976.114 However, copper prices remained depressed, and it became increasingly difficult for Zambia to service its growing debt. In 1986, Zambia's total debt servicing accounted for 28.5 percent of its gross national income, making it one of the world's most indebted nations.115 In December 2000, Zambia qualified for debt relief under the Heavily Indebted Poor Countries Initiative (HIPC). HIPC is not debt cancellation; rather it is a restructuring of debt repayment through provision of grants. Zambia was granted US$3.8 billion as total debt relief from all its creditors. To date, HIPC has disbursed US$135 million to Zambia, of which US$18 million has been channeled into health and education. 116 Programs to be supported by HIPC resources are focused on poverty reduction and social activities. In the health sector, the focus is on HIV/AIDS, malaria, essential drugs, and underfunded public hospitals.117 According to IMF estimates, HIPC will result in continually increasing social services expenditures in Zambia through 2005, in absolute terms and as a percent of government revenues.118 (Zambia has also received additional debt relief from several bilaterals.) The World Bank notes that Zambia's debt will not be sustainable until at least 2004,119 a projection that concerns, among others, Zambia's National HIV/AIDS/STD/TB Council. 120 This projection is primarily based on assumptions about Zambia's export earnings; however, sole reliance on this criterion is problematic for countries such as Zambia, dependent on one export commodity (copper).121 Indeed, in September 2002, the IMF and World Bank noted that Zambia's debt-to-export ratio had deteriorated, as export earnings fell because of lower world copper prices and Anglo American's decision to discontinue its mining operations in Zambia (mentioned above).122 The Bank's projections regarding debt sustainability are also based on what many see as overly optimistic assumptions, for example, that real GDP growth will be 5 percent from 2001 onward and that foreign direct investment will rise from US$250m in 2002 to US$281m in 2010. 123 Zambia is currently in its interim HIPC period, meaning that to qualify for the full amount of debt relief available via HIPC, it must successfully meet its creditors' requirements. 124 Among these is that it sell the state-owned Zambia National Commercial Bank.125 Regardless, after receiving debt relief, Zambia will need to continue borrowing to maintain debt servicing obligations and to purchase key imports.126 According to Jubilee Plus, by 2004, new debt will account for 63 percent of Zambia's overall debt. By 2019, 75 percent of Zambia's average annual debt service will be for new debt. 127 Many international NGOs and propoor advocacy groups ¾ including Bretton Woods Project, Catholic Fund for Overseas Development (CAFOD), Christian Aid, European Network on Debt and Development (EURODAD), Jubilee Plus, Oxfam U.K., and World Development Movement ¾ also express concern that HIPC has reshuffled Zambia's enormous debt rather than cancel it.128 Declining Human Development One method of tracking human development in Zambia is to analyze trends in its Human Development Index. The HDI was created by UNDP to measures average achievements in life expectancy at birth; adult literacy and combined primary, secondary, and tertiary gross enrollment ratios; and GDP per capita (most UN agencies are now calling this gross national income [GNI]; details on its calculation can be obtained from the World Bank). An HDI of 0.800 or above = high human development; 0.500 - 0.799 = medium human development; less than 0.500 = low human development. In 2000, Zambia's HDI value was 0.433, placing it among "low-human development" countries and ranking it 153 out of the 173 countries for which UNDP calculated an HDI. Zambia's HDI value is lower than that of the median for the world's least-developed countries (0.445) as well as for sub-Saharan Africa (0.471).129 What is particularly worrying is that although Zambia's HDI value is already very low, it declined further during the 1990s. Between 1975 and 1985, the HDI value rose from 0.449 to 0.480, a reflection of, inter alia, the government's concrete efforts to increase educational attainment and health outcomes. From 1985 onward, however, the HDI value has been falling, mirroring the country's economic decline. In 1990, the HDI value was 0.468, falling to 0.432 in 1995 and to 0.433 in 2000.130 This decline reflects the fall in national income per capita, discussed above, and as the public expenditure trends section above highlights, Zambia's decreased spending on education during the 1990s. The government's spending on health rose in that decade, but clearly the enormous impact of AIDS mortality (see Impact section) drastically reduced the life expectancy component of the HDI value. A critical indicator of the well-being of children is the under-five mortality rate. In 2000, UNICEF reports that Zambia had the world's 10th-highest under-five mortality rate: 202 deaths per 1,000 live births. The preindependence (1960) figure was 213. Moreover, Zambia's under-five mortality rate for 2000 exceeds that of all the least-developed countries (161) and of sub-Saharan Africa (175). 131 Infant mortality, another key human development indicator, fell somewhat though not dramatically between 1960 (126) and 2000 (112). And, as with under-five mortality, Zambia's infant mortality rate for 2000 exceeds that of all the least-developed countries (102) and of sub-Saharan Africa (108). 132 Population Mobility Throughout southern Africa, high levels of movement between urban, rural, and mining areas facilitate HIV transmission.133 Zambia has a long history of men migrating to work in large agricultural estates in Chinwag in rural Lusaka Province and the Namable Sugar Estates in Southern Province, as well as to the mines in Copperbelt Province. On sugar estates, for example, men leave their families to work as cane cutters from March to November.134 Migrant labor separates men from their families, places them in close proximity to "high-risk" sexual networks, and often results in their having an increased number of sexual contacts. Concurrently, it may also lead to women's reliance on sex to supplement their incomes while their male partners are away for long periods. Thirty-two percent of Zambia's population is urban.135 Zambia's transport infrastructure links major urban and industrial centers, and rail lines tie it into an extensive central African railroad network.136 The 2000 ZSBS found that Zambians have a fairly high level of mobility. Of men, 29 percent had lived less than five years in their current location; for women, this figure was 31 percent. In the last month, 18 percent of women and 23 percent of men had spent at least one night away from home.137 Poverty may lead to increased migration, both within Zambia and to other countries, as people move from rural to urban areas in search of work ¾ or return to families if they lose their jobs or fall ill and cannot afford care. This scenario can place an additional burden on receiving households, which concurrently lose any remittance income from the person who has fallen ill.138 The U.N. reports that the food crisis is increasing migration to towns in Central and Eastern provinces.139 The political and economic turmoil in Zimbabwe is also likely to be spurring population dislocations within the subregion. Key mobile groups in Zambia include: * truck drivers * sex workers * fishermen/women and fish traders * migrant and seasonal workers * cross-border traders, especially young girls * miners * military personnel * prisoners (in the sense that they often return to their families/communities upon release) * refugees In November 1999, Family Health International undertook research in four towns along the Durban-Lusaka highway, including Chirundu, Zambia. The findings illustrate the intersection of population mobility and HIV vulnerability (see box 2). Box 2. Chirundu: An Examination of HIV/STI Vulnerability in a Zambian Border Town Chirundu is situated on the Zambian side of the Zambezi Valley in Zambia's Southern Province, 142 kilometers south of Lusaka and 366 kilometers north of Harare. The population is estimated at 7,000. There are no recreational facilities, apart from taverns. The largest sources of formal income in urban Chirundu -- freight, construction, retail, customs, domestic service, teaching, immigration, and police -- employ over 400 people. The largest sources of urban informal income are sex work, money changing, and trading. Most women in Chirundu rely primarily on trading and sex work for income. Many female traders also practice sex work. Chirundu has about 300 permanent and 200 transient sex workers. Sex workers seek clients primarily at hotels and taverns. There is no street-based sex work, except on the main highway. Sex workers' major clients are truckers, especially those who pay in foreign currency. Sex workers value South African clients, reporting that they pay up to US$20, over 10 times more than local men can pay. Moneychangers are also clients. Competition for foreign clients is severe, and most sex workers remain poor, though perhaps less so than other women. Sex workers reported that they preferred to use condoms, but -- until recently -- condoms were difficult to obtain. Many sex workers reported that they entered sex work when they became pregnant and were abandoned by their boyfriends and families. Girls as young as 12 are reported to engage in sex work. Older sex workers complained that younger rural girls would accept clients for food or soap. Thirty-six trucking companies use the Chirundu, Zambia, route. Truckers' major southbound destination is South Africa, and their major northbound destinations are Lusaka, the Copperbelt, and the DRC. The war in the DRC increased its reliance on imports, and trucking traffic at Chirundu has increased as a result. The average age of truckers is 35, and most have been driving on the Chirundu route for five years or more. Many have steady girlfriends as well as sex worker contacts. The farms outside Chirundu are large, employing over 400 permanent workers and 2,000 seasonal workers. These farming communities also influence the social and sexual character of Chirundu, with rural poverty spurring movement to and from the town. Sources: Family Health International. Corridors of Hope in Southern Africa: HIV Prevention Needs and Opportunities in Four Border Towns. Arlington, Va.: 2000 Truck Drivers Long-distance trucking plays a crucial role in Zambian economy., Zambia has six major trucking routes. The Chirundu-Lusaka and Lusaka-Copperbelt routes each have 100 trucks daily. Livingstone-Lusaka has over 40 trucks a day. Lusaka-Chipata, Lusaka-Mumbwa and Mongu, and Kapiri Mposhi-Nakonde each host 20 trucks a day. On average, about 300 trucks use Zambia's main highways each day. There are 1,500 registered commercial trucks. Approximately 3,500 drivers and assistants are away from their families for extended periods. When they travel, their sex partners are mainly mobile sex workers from low-income communities.140 Sex Workers Tasintha, an NGO that works with sex workers, estimates that there are at least 6,000 full-time sex workers in Zambia. Other studies put the figure at 24,000 (7,000 in Lusaka and 17,000 in major tourist locations, major highways, and border and trading towns). Sex workers are based in, inter alia, brothels, nightclubs, and the street. Male clients control use of male condoms by offering to pay more for condom-free sex. 141 For most Zambian women, marriage remains the key to economic survival. Inability to form a long-term partnership or the collapse of a marriage can have disastrous financial consequences for women; some may be forced into sex work, which is illegal and highly stigmatized. 142, 143, 144 The penalty for prostitution is one month's imprisonment, a fine of about 80,000 Kwacha (about US$17, an enormous amount in a country where average monthly income is US$25), or both.145 In a study of 300 SWs in Ndola, researchers from the London School of Hygiene and Tropical Medicine and Institute of Tropical Medicine in Belgium found that 69 percent were infected with HIV. Longer duration of sex work in the city and genital ulceration and herpes simplex virus-2 infection were significantly associated with HIV infection. 146 With funding from USAID, World Vision has been working with Zambia's Tropical Diseases Research Center, National AIDS Council, FHI, and Belgium's Institute of Tropical Medicine to reach SWs and truck drivers in five major border posts and truck stops: Livingstone, Chirundu, Chipata, Kapiri-Mposhi, and Kasumbalesa. The project seeks to change behavior through outreach and peer education, social marketing of condoms, and improved STI care. 147 In 2000, the project recruited 636 SWs for a study: 267 at Livingstone, 145 at Chirundu, and 224 at Chipata. The study population was young, with a mean age of 23 years and a high proportion (37 percent) of adolescents. Most (81 percent) had completed at least primary school; almost half (48 percent) had ever been married. The distribution by ethnic group and religion was similar to that of the general Zambian population. The population was mobile, with 46 percent originally from a province other than the one in which they were currently residing and a median period of current residence of 5.9 years. One third (33 percent) reported an occupation other than sex work. Over half (58 percent) reported that they were supporting others. 148 The median age of sexual début was 15 years and of starting sex work, 17 years. Almost all women (99.5 percent) reported having previously engaged in sex work elsewhere. The median time period of sex work in the current residence was 1.7 years. The median number of sexual partners in the last seven days was 3. The median price per client was 20,000 Kwacha (about US$4.40). Slightly over half of women (54 percent) reported having used a condom at last sexual contact with a paying client. The most frequent reasons mentioned for not using a condom were that the client objected (36 percent), that they themselves did not like condoms (21 percent), or that they did not think of using one (21 percent). When asked how regularly they used condoms with clients over the last 30 days, 25 percent reported using condoms every time or almost every time, 59 percent sometimes, and 17 percent never. 149 Fifty percent of the women reported a nonpaying sex partner in the last seven days. The median frequency of sexual intercourse with this partner was four times in the last 30 days. Less than half (44 percent) had used a condom at the last sexual intercourse with this partner, and the majority reported having used condoms with nonpaying partners during the past 12 months only sometimes (66 percent) or never (18 percent). 150 Researchers from Population Services International interviewed 14 nightclub-based and six street-based SWs in Lusaka. PSI found that SWs are subject to frequent police raids and the risk of violence from clients or members of the public. The lack of a supportive social environment may reduce their ability to adopt effective safer behaviors. 151 Fishermen/women and Fish Traders Fishermen/women and fish traders converge in populous lake basins. Fishermen often leave their wives and families in rural areas for several months and, in most instances, enter into "marriages of convenience." Their steady cash earnings can also support a number of temporary sexual liaisons. Married female fish traders may exchange sexual favors for preferential road and water transport to and from fishing camps. Fishermen are also reported to frequently demand sexual favors as a condition for selling fish. These communities have limited knowledge of HIV/AIDS; for example, those who die because of AIDS are presumed to have returned to their home, thus masking the effects of the disease. Access to health information and care is low, as is reported condom use.152 Border Traders In Zambia, some female traders exchange sex with truck drivers for transport. Police and customs officials may demand sex from women caught crossing borders illegally or who want to avoid paying customs duties. Children are increasingly entering this trade, and young girls are especially vulnerable to sexual exploitation. 153 Military Zambia's military has had a high level of mobility, participating in U.N. peacekeeping initiatives in Mozambique, Rwanda, Angola, and Sierra Leone.154 Refugees Zambians have shown tremendous leadership and generosity in welcoming large groups of refugees escaping conflict in other African countries. 155 Zambia is host to about 285,000 refugees, mainly from Angola and the Democratic Republic of Congo (DRC), as well as Rwanda, Burundi, and Somalia. Many refugees have been in Zambia for several years.156 According to UNHCR, despite progress on peace accords in the DRC, around 100 Congolese cross into Zambia each month fleeing sporadic skirmishes. (The U.N. also notes that there have been reports of gunmen from the DRC harassing Zambian villagers.) 157, 158 Influxes of refugees put increased pressure on the budgets of Zambian districts that border areas of conflict. 159 The Zambian government imposes restrictions on refugees who seek to use government health facilities. 160 (There are some health centers in refugee camps, many run by NGOs.) According to a February 2002 report on refugees in Zambia produced by the Women's Commission for Refugee Women and Children: Perception of HIV risk differs within the refugee population. There are few diagnosed HIV/AIDS patients, and a persistent skepticism about the existence of the disease persists. Gender-based violence is a topic most people are reluctant to discuss. Domestic violence, exacerbated by alcohol and drug use/abuse, is reported to be the most common form of violence. Most refugee health facilities lacked protocols to manage the consequences of rape. Victim Support Units are in place at some camp police stations but it is not clear that the units' staff are adequately trained to care for victims of violence. Reproductive health services for adolescents are limited and ad hoc at best. There are nascent efforts by NGOs to establish youth-friendly centers, youth anti-AIDS clubs and use of peer educators to target the adolescent population. However, adolescents are clearly a sexually active population and are particularly vulnerable, given the lack of comprehensive services targeting their needs.161 In October 2002, several Western governments announced that they would provide support to Zambia and UNHCR to integrate Angolan refugees in Zambia's Western Province. Part of this process includes establishment of an STI/HIV/AIDS drop-in center to promote HIV/AIDS awareness among refugees and their host community. Zambia's approach to refugee integration through linking of relief and development is being used as a model in other countries.162 Food Crisis Over the past year, the situation in Zambia has become critical. In 2001, excessive rains in parts of the country resulted in floods that destroyed large cultivated areas. Districts in Eastern and Southern provinces along the Zambezi and Laneway rivers were the most affected and overall production of maize, the staple crop, fell by an estimated 24 percent compared to the previous year. In 2002, erratic rains and drought hit large parts of the same provinces during the growing season, exacerbating an already precarious food situation. 163 Moreover, underutilization of land is a ongoing issue (currently, only 20 percent of arable land in Zambia is cultivated164). On May 28, 2002, the Government of Zambia declared a national disaster due to actual and anticipated food shortages. 165 The government's Disaster Management and Mitigation Unit, under the office of the Vice President, is working with U.N. agencies, bi- and multilateral donors, civil society, and the private sector. In July 2002, the U.N. launched an appeal for US$71 million to combat the humanitarian crisis in Zambia. In September 2002, USAID reported that 2.9 million Zambians (29 percent of the population) would require food aid totaling 224,000 metric tons between September 1, 2002, and March 31, 2003. Concurrently, Zambia has made the least progress in filling its cereal gap, meeting less than 9 percent of its requirements, compared to the average 25 percent imported to date by the rest of the countries in the region.166 The most vulnerable groups are located in Southern and Western provinces and include the elderly and households headed by children and women; those with persons who are disabled or ill; and those with widows not supported by other households. Significant numbers of these households include orphans and other vulnerable children. Reports from aid agencies who have recently conducted missions in Zambia concur that although erratic weather has contributed to the current food crisis, one of the key underlying factors is the depletion of human resources as a result of HIV/AIDS.167 Unsustainable debt and deteriorating public services have also played a role. Oxfam highlights that the failure of agriculture liberalization policies are also a factor. After years of World Bank- and IMF-supported agricultural sector reforms, Zambia still faces chronic food insecurity. Oxfam argues that these reforms were imposed too rigidly and too quickly, often leaving poor farmers without support from or access to either state or market institutions. 168 In August 2002, the Government of Zambia announced that it would not accept biotech-derived food aid due to health and environmental concerns. In September 2002, President Mwanawasa indicated that the he remains open to accepting to accepting biotech-derived food contingent upon evidence demonstrating its safety.169 (Some have suggested that Zambia's refusal of biotech-derived food is related to its desire not to offend Europe ¾ its largest export market ¾ which bans genetically modified food.170) Health Sector The health and human development indicators in the accompanying table highlight the magnitude of the poor health status of Zambians. As discussed in the Human Development section, the general health status of Zambia's population worsened substantially during the 1990s.171 In 1998, public health expenditures in Zambia accounted for about 51 percent of total health care spending.172 Almost all public health expenditure is financed by taxation; public health insurance does not play a significant role.173 Private insurance accounts for about 30 percent of private expenditures on health. 174 The country bears an enormous burden of malaria (see indicator table). Other major health problems include TB (see Impact section), leishmaniasis, Guinea Worm disease, measles, malnutrition, respiratory diseases, polio, and diarrheal disease. Sexual and Reproductive Health UNFPA ranks Zambia a category "A" country, meaning that it is furthest from achieving the sexual and reproductive health goals of the International Conference on Population and Development (ICPD), held in Cairo in 1994. Group A countries have the greatest need for external assistance and the lowest capabilities for mobilizing domestic resources to close this gap.175 The accompanying table provides selected indicators of sexual & reproductive health. The burden of fertility on young women is high. Although little documentation exists, UNFPA ¾ using hospital and clinic admission data ¾ surmises that unsafe abortions are common.176 STIs As mentioned in the Epidemiology section, Zambia's National STD Control Program was launched in 1980. Since 1996, no national data have been readily available.177 In 2000, the National HIV/AIDS/STD/TB Council reported that STIs account for 10 percent of all documented outpatient attendance in public facilities. A 1997 study by UNICEF and Zambia's National AIDS Program found that adolescents made up 40 percent of STI clientele in public outpatient facilities.178 The Study Group on the Heterogeneity of HIV Epidemics in African Cities ¾ which includes Ndola, as well as Kisumu, Kenya; Cotonou, Benin; and Yaoundé, Cameroon ¾ found that infection with herpes simplex virus 2 is extremely prevalent in young women shortly after they begin sexual activity. 179 What is also of concern is that between 1998 and 2000, the proportion Zambians who used a condom during sexual acts when they knew they were infected with an STI declined substantially, from 36 to 23 percent for women and 46 to 24 percent for men. 180 Health Sector Reform As part of its structural adjustment measures, the government and its donor partners began a process of health sector reform in 1994. This approach emphasizes devolution of administrative, financial, and technical responsibility for essential health services to district health and hospital management boards. (Policymaking, resource mobilization, legislation, and other facilitation and coordination functions are retained at national level.) The aim of health sector reform was to improve health status by increasing access to and the quality of a national package of essential and cost-effective health services in a decentralized health care delivery system. This process also included a shift away from (expensive) tertiary-level institutions towards primary-level healthcare. The program was estimated to cost $537 million, and received support from several donors. The World Bank's Zambia Health Sector Support Project provided US$56 million. 181 USAID developed the Zambia Integrated Health Program (ZIHP) in partnership with government to address key health problems and continue the process of the government's health reform. ZIHP is implemented by a group of American and international agencies and organizations in partnership with Zambian public and private institutions and other international cooperating agencies. With a budget of US$41 million from 1999-2002, ZIHP focuses on demonstration districts. 182 As part of the health sector reform process, user fees at public health facilities were introduced. A system of exemptions from payment of user fees was introduced in 1993 to provide a social safety net and covers: * children under 5 and those ages 65 and above * all ante- and postnatal visits * all visits for chronic illnesses such as TB, STI, and HIV/AIDS * those affected by disaster or involved in accidents Those who are unable to pay can seek exemption under the Health Care Cost Scheme.183 The World Bank characterizes the status of the health reform process as "disappointing" and slow. 184 As is usually the case when the Bank evaluates projects and policies it has championed, it lays most of the blame on poor government implementation. Certainly, this is the case to some degree, but serious concerns have been raised about underlying elements of the reform process itself. For example, there have been inequities in the exemption policy mentioned above. Research has shown that that there have been very high errors of exclusion and inclusion; those who can afford to pay or are ineligible under the criteria have been granted exemption, whereas many who were eligible have been denied exemption.185 Moreover, exemption mechanisms, even if they worked as intended, would not necessarily address inequalities in the use of services related to income or distance to health facility. The poorest sections of the population are found in remote areas that are not easily accessible. For example, households in several districts in Central, Northern, and Western provinces have an average of over 60 km to the nearest health facility and an average of over 50 km to the nearest transport facility. Exemption schemes are of almost no benefit to them.186 Researchers from Kanyama Health Center in Lusaka, Michigan State University, and UCLA undertook a cross-sectional analysis of healthcare utilization in a large Zambian hospital for children ages 3 to 6 between August and September 2000. They found that female children may be less likely to present for care when user fees are imposed.187 The National HIV/AIDS/STD/TB Council raises concerns about how decentralization has rapidly shifted the burden of HIV/AIDS to all districts across the country in a "one size fits all" approach. 188 There was no analysis of individual districts' capacity to assume the responsibilities of decentralization, a particularly critical issue for those districts that were already understaffed, underfunded, and/or located in the poorest parts of the country. Nor was there analysis of the characteristics of clients served by different health centers (e.g., by occupation, mobility pattern) and how increased demands on health care staff resulting from decentralization would affect them. Currently, the health sector struggles with many problems, including: * Almost all health facilities lack adequate personnel, drugs, and/or equipment. * Physical infrastructure and equipment are deteriorating. * Erratic distribution methods lead to frequent shortages of essential drugs and medical supplies in hospitals and health centers. There have been instances of government personnel making ineligible drug expenditures and questionable contractual arrangements made by government for storage and distribution of drugs and medical supplies. * Conditions (low pay, inadequate equipment, poor supplies) in Zambia's public hospitals have deteriorated and have led to long-running strikes by health care providers. * The Ministry of Health gave the Central Board of Health executive responsibility for health service delivery. 189 However, as mentioned above, the functions and staff of the Ministry of Health and the Central Board of Health overlap, leading to duplication and other inefficiencies.190 Certainly, one cannot hold health reform solely responsible for the state of the Zambian health sector. Poverty, debt, falling copper prices, and the enormous impact of HIV/AIDS have played a role as well. In 2001, the government adopted a new five-year National Health Strategic Plan (2001-2005) to continue the reforms and improve service delivery. However, financing support from donors has been low, given the "disappointing" results of health reform thus far. 191 All this is particularly worrying with regard to the sector's capacity to manage HIV/AIDS. Gender HIV Prevalence among Women As discussed in the Epidemiology section, UNAIDS published data in July 2002 indicating that HIV prevalence among Zambian women ages 15 to 24 ranges from 16.78 to 25.18 percent, whereas the comparable range for men in the same age cohort is 6.45 to 9.68 percent. 192 Researchers from the Study Group on the Heterogeneity of HIV Epidemics in African Cities found that among sexually active 15- to 19-year-olds in Ndola (and Kisumu, Kenya), HIV was six times more prevalent in women than in men. Among 20-24-year-olds, HIV infection was three times more prevalent in women. HIV was equally widespread among women and men in the 25-49 age group. Behavioral factors did not seem to explain this difference in HIV susceptibility. The presence of other STIs, especially those that result in ulcerated lesions, may account for some increased transmission of HIV. In both cities, for example, the rate of infection with herpes simplex virus type-2 was roughly four times greater among young women than among young men. Even when no other STI was present, however, young women were still at higher risk for HIV infection. There is evidence of high rates of HIV positivity following only a few episodes of sexual intercourse, suggesting that young women have a higher susceptibility to infection. 193 Women's Status The 2000 ZSBS found that 85 percent of women and 92 percent of men have had at least primary education A much larger proportion of rural (21 percent) than urban (6 percent) women have no education. Forty-four percent of men and 33 percent of women have some secondary education. The proportions of men and women with secondary education were more than twice as high in the urban compared to the rural areas.194 Few women hold high-level decisionmaking positions in any sector of the Zambian economy. Despite constitutional and legislative provisions, women still experience disadvantages in enforcement of laws regarding property ownership, inheritance, and marriage.195 UNFPA's ranking of Zambia as a category "A" country connotes a high level of gender inequality, women's low socioeconomic status, and their poor sexual and reproductive health and rights.196 Many of the indicators found in the accompanying table are useful in analyzing women's socioeconomic status. For example, the maternal mortality ratio measures the number of deaths to women per 100,000 live births that result from conditions related to pregnancy, delivery, and related complications. It is a standard indicator not only of access to safe delivery services, but of overall human development. UNFPA estimates that Zambia's 2001 MMR was 870. This is the second-highest MMR in southern Africa (MMR for Mozambique was highest, at 980). It is over twice the global MMR; however, it is lower than the MMR for Africa (includes North Africa) and for all least-developed countries.197 UNDP measures gender inequality by using the unweighted average of three component indices: life expectancy, education index, and income index. Its Gender-related Development Index (GDI) value ranges from 0 (lowest gender equality) to 1 (highest gender equality). In 2000, UNDP calculated Zambia's GDI value at 0.424, ranking it 129 out of 146 countries on this index. (For comparison, GDI values range from 0.263 [Niger] to 0.956 [Australia].) Women and Poverty A key component of women's poverty is their inability to obtain loans from banking institutions, which renders them less able to enter into safe, profitable economic activities.198 Widows Widows may be particularly vulnerable to HIV because of sexual cleansing (kusalazya) and wife inheritance (kunjilila mung'anda). The traditional practice of sexual cleansing is still highly prevalent, especially in Southern Province. To be purged of evil forces assumed to have caused the death of the spouse, the widow or widower is cleansed through sexual intercourse with a relative of the deceased. 199 In response to analyses of sexual cleansing in the context of HIV transmission, many Zambian local leaders (especially in Southern Province) have encouraged alternative rituals. These include: * sliding over a half-naked person or jumping over an animal ("cow-jumping") * using herbs and roots * cutting of hair * applying powder * having a (proxy) married couple perform the cleansing act of sexual intercourse Malungo notes that many of the alternatives had been used for some time, but have only become popularized since the advent of AIDS.200 Another phenomenon is called "grabbing", wherein close relatives take possession of a the deceased's household goods, land, livestock, clothes, and other assets. This scenario exacerbates the already precarious economic (and social) situation of widows and their children.201 Despite legislation to address this issue (Intestate Law of Succession and Administration of Estates acts), widows continue to suffer extreme harassment from and impoverishment at the hand of their husbands' relatives.202 Sexual Negotiation Data from the 2000 ZSBS suggest that women in Zambia have little power in sexual negotiation with their husbands. Less than half of men (49 percent) and women (45 percent) who have heard of STIs believe that if a woman's husband has an STI, she can either refuse to have sex with him or ask him to use a condom. These figures were, however, an improvement over those of 1998, when only 25 percent of men and 30 percent of women believed that a woman could protect herself from an STI if her husband had one. A strong urban-rural differential was found for this indicator in 2000: percentages for urban women (55%) and urban men (58%) were much higher than in rural areas (women=37%, men=43%).203 Among respondents who reported that a woman can protect herself from an STI, 65 percent of respondents mentioned refusing sex; 66 percent of men and 56 percent of women mentioned condom use. A woman insisting upon condom use was a more common response among urban residents than rural residents. As the authors note, using condoms and refusing sex are largely "male-driven."204 Polygamy, more common in rural than urban areas, is also a factor. In 1996, 17 percent of married women in Zambia were in polygamous unions. The highest rate of reported polygamous relationships was in Southern Province (32 percent). 205 Perception of Women As in many countries, the image of women as the vectors of disease has been present