Namibia - Selected development impact of HIV/AIDS

Report No. 22046-NAM Namibia Selected Development Impact of HIV/AIDS April 10, 2001 Macroeconomic Technical Group Africa Region Document of the World Bank  CURRENCY EQUIVALENT (April 10, 2001) Currency Unit = Namibia Dollar Namibia (N$) 1.00 = US$ 0.1242 US$ 1.00 = N$ 8.05 ABBREVIATIONS AND ACRONYMS AIDS Acquired Immune Deficiency Syndrome ANC Antenatal Clinics CBO Community Based Organizations CGE Computable General Equilibrium DHS Demographic Health Survey GDP Gross Domestic Product HIV Human Immunodeficiency Syndrome IEC Information, Education and Communication MOHSS Ministry of Health and Social Services MTC Mother to Child Transmission NACP National AIDS Control Program NACOP National AIDS Coordinating Program NER Net Enrollment Rate NGO Non-Government Organizations PLWHA Persons Living With HIV/AIDS PROST Pension Reform Options Simulation Tool-Kit RON Republic of Namibia SSS Sentinel Surveillance System STD Sexually Transmitted Disease TB Tuberculosis UN United Nations UNAIDS United Nations AIDS Agency USAID United States Agency for International Development USBOC United States Bureau of Census WHO World Health Organization Vice President Callisto E. Madavo Country Director Fayez Omar Sector Manager Philippe Le Houerou Task Team Leader James Sackey  Namibia: Selected Development Impact of HIV/AIDS Table of Contents Page No. Executive Summary ......................... ................................... v Chapter I Introduction: .............................................. I A. Nature of the Epidemic ............................................. 2 B. Likely Impact . .............................................. 3 C. Organization . .............................................. 5 Chapter II Status of HIV/AIDS in Namibia ...................... . 7 A. HIV Prevalence .............................. ................ 7 B. AIDS Cases and Deaths ............................................. 11 C. Sexual Practices and Knowledge .................................... 13 D. Need to Improve Monitoring ....................................... 15 Chapter III Demographic Impact of AIDS .......................................... 17 A. Impact on General Population .. .................................... 18 B. Implications for the Labor Force ................... ............... 21 C. Conclusions ...................... ........................ 22 Chapter IV HIV/AIDS and the Economy............................................ 23 A. Macroeconomic Impact .23 B. Impact on Selected Sectors .26 C. Conclusions .30 Chapter V Response to HIV/AIDS.............................................. 32 A. Control Strategies in Nanibia .33 B. Proposals to Strengthen Ongoing Initiatives .34 C. Some Lessons to Consider .42 Bibliography . .............................................................................. 44 Annex I: Modeling the Impact of HIV/AIDS in Namibia .46 Annex II: Namibia: Assumptions Used in Costing Selected HIV/AIDS Programs .56 List of Tables 2.1: HIV Prevalence Among Pregnant Women Attending ANC 1992- 1998 (In Percentage). 8 2.2: HIV Prevalence Among SDT Patients in 1998. 9 2.3: Regional Distribution of Reported HIV cases, HIV/AIDS Hospitalization and HIV/AIDS Deaths (Rate Per 1,000 Population) in 1998 .12 3.1: Summary of Selected Demographic Impact of AIDS .18 4.1: Indicators of Macroeconomic Impact of HIV/AIDS (in Percentage) .25 4.2: Summary of Long-term Demographic Indicators .27 .i.i  5.1: Strategies and Options .............................................. 35 5.2: Summary of Estimated Costs of Selected Policy Interventions .... 41 List of Figures 2.1: Age Specific HIV Prevalence in STD Patients .......... ............ 10 2.2: New and Cumulative AIDS Cases in Namibia .......... ............. 12 2.3: Adolescent Pregnancy .............................................. 14 2.4: Namibia - Percentage of Female Who Ever Used a Condom ....... 15 3.1: Population Growth With and Without AIDS ........... .............. 19 3.2: Population Profile in 2015, With and Without AIDS ........ ........ 21 3.3: Labor Force, With and Without AIDS ................. ................ 22 4.1: Namibia - Gross Domestic Product (With and Without AIDS)-- 25 4.2: A Young Through Rapidly Aging Population and the Uncertainty of AIDS .............................................. 28 5.1: Hospital Bed Days Needed for AIDS Patients ........................ 39 This report was prepared by James Sackey (Task Manager), Tejaswi Raparla, and Shashi Kolavalli based on a number of missions to Namibia during FYOO-0 1. The report team was supported by the staff of NEPRU, especially Hopolang Phororo. Anirudha Bonnerjee prepared the pension system projections. Valuable comments were received from colleagues at AFTMI, members of the Namibia Country Team, the peer-reviewers, Mead Over (DECRG) and Rene Bonnel (AFRHV). Secretarial assistance was provided by Felicidad Santos. This report was discussed extensively at the National Planning Commission in Namibia and by the UNAIDS Theme Group in Windhoek during March, 2001. Coniments from the participants are duly acknowledged. iv  EXECUTIVE SUMMARY Since independence in 1990, Namibia's economy has grown moderately with real GDP growth averaging 4 percent annually in the early nineties. Nevertheless, it is estimated that close to one half of the population in 1993/94 could be considered to be poor, if they are defined as households with expenditure on food exceeding 60 percent of income. Poverty is also concentrated, with two-thirds of the poor living in northern parts of the country. The government has responded by developing a poverty reduction strategy in 1994 aimed at reducing the percentage of the poor and extremely poor from 47 and 13 in 1994 to 40 and 7 respectively by 2000. Population policy complements the poverty strategy with the following objectives: reduce the growth rate of population from 3.2 percent during 1991-96 to 3 percent by 2006 and to 2 percent by 2015; infant mortality from nearly 57 per 1,000 live births to 40 per 1,000 live births by 2006 and to 30 per 1,000 live births by 2015; and total fertility rate to 5 by year 2006 and to 3.5 by 2015. While modest improvements have taken place in Namibia's human development, as its social spending is one of the highest in the world, the onset of HIV/AID is likely to substantially weaken the government's ability to meet its development objectives. HIV/AIDS has emerged as the number one cause of death in 1996. It is estimated by the UNAIDS that 19.94 percent of adults (15 - 49 years) was living with HIV (excluding those with AIDS) in Namibia in 1999. The estimated prevalence rate among 15 to 24 year-olds was 18.8 to 20.8 percent for females and 7.9 to 10.4 percent for males. The estimated total number of individuals living with HIV/AIDS in 1999 was nearlyl60,000. Of them, about 150,000 are adults (15-49). Among adults, the estimated number of women was 85,000 (UNAIDS, 2000). These estimates, for obvious reasons, are nearly three times more than the number reported by the health directorate on the basis of testing in the health system which indicated the cumulative of infection to be 53,330 in 1999. Nevertheless the number of HIV positive cases reported annually by the system has exceeded 10,000 since 1996. All indications are that the epidemic is widespread in the country. The objective of this report is to provide an overview of selected likely development impact of HIV/AIDS in the light of existing data. It was initiated as an exercise to assist policy makers in Namibia in their effort to incorporate HIV/AIDS into the planning process on a regular basis. As such it is directed at officials at the agencies for planning and finance. Because of data limitations, the review at this stage does not provide detailed costing of the impact of HIV/AIDS on various sectors of the economy. Detailed costing and policy design alternatives should be the focus of subsequent analysis. Impact of HIV/AIDS Demographic: Examining the demographic consequences of the epidemic is the first step in making a comprehensive assessment of the likely impact on various sectors or the economy. Using the Spectrum Models (developed by the Futures Group International), with comparative reference to the results generated by other studies, it is concluded that the primary impact of the epidemic could be in the reduction of life expectancy at birth (LEB). The 1981 and 1991 censuses indicate that LEB has already fallen in Namibia from nearly 58 to 55 years. The decline is confirmed by current projections, which indicate that in the v  absence of AIDS Namibians could have expected to achieve an LEB of nearly 70 years by 2015. As a result of AIDS related mortality of young adults, they might expect to live for 40 years. Comparable estimates by the US Census Bureau suggest that LEB would be reduced to nearly 40 by 2010, while the Ministry of Health and Social Services (MOHSS) estimates point to 40 years by 2005 before rising to about 45 by 2021. Although AIDS reduces the population growth, as sought in population and human development strategies of the country, it does so in perverse ways. In the desired form of demographic transition, the demand for children is brought down through improved education particularly of women and reduced infant and child mortality rates. The fewer children that are born receive greater investments in their development. In the case of AIDS-induced transition, on the other hand, child and infant mortality rates do not fall as much, crude birth rate does not fall significantly, but population growth rate is reduced by higher mortality of young adults. The demographic transition brought about by HIV/AIDS may have a number of consequences. A smaller economically active population may have to take care of the very young and the very old. Family and community structure may be affected significantly as more than 10,000 children are estimated would be orphaned annually and cumulative 115,000 by 2015. The labor force is also likely to be affected. Preliminary analysis suggests that AIDS deaths are likely to exacerbate the skill shortage in Namibia. While the existing data do not permit a detailed analysis of skill gap, the projected gap in the labor force as a result of AIDS related mortality is estimated to be nearly 20 percent by 2015. A substantial share of this gap could be in the skill category, made up of people with the means for greater mobility. Failure to deal with the skills shortage could exacerbate Namibia's growth constraints. HIV/AIDS and the Economy: From the macroeconomic perspective, HIV/AIDS is likely to affect the savings/investment relation. To the extent that AIDS leads to the reduction of savings, availability of skilled labor and profitability in the private sector, its impact on economic growth could be significant. Using a growth model, extended to incorporate the increase in morbidity and mortality resulting from HIV/AIDS, it is estimated that in the absence of AIDS, GDP could grow at about 3.5 percent during 2000- 15, approximately at the rate of population growth. With decreased investments, the rate of growth would be lower, projected at 2 percent by 2015, instead of the 3.5 percent that could have been expected in the absence of AIDS. Furthermore, the rate of per capita growth in GDP would have declined even in the absence of AIDS, as economic growth has not been keeping pace with population growth. With AIDS, it would decline much more. Depending on the nature of labor force employed in the sector, some sectors may be more affected by the epidemic than others. The key factors to consider are the skills level of the labor force in the particular sector, which highlights the difficulty associated with replacing those affected by AIDS, and the vulnerability of the sectoral labor force to HIV infection, which may be dependent on its socio-economic characteristics. As a case study a number of features of the civil service, for example, make it vulnerable to disruption through loss of manpower: vi  * Based on a selected sample, it is estimated that about 80 percent of the civil service has less than five years in service experience, with less than one percent above 11 years. There is thus a missing middle level staff in the hierarchy. * Close to half of the civil service fall in the 25-40 years age category, which may account for the large number of staff with limited experience but fall into the high-risk age category of contracting HIV. Given the young age of the civil servants and their predisposition to contracting HIV, the service is likely to lose a significant number of its members in the near future, assuming that they have same risks as the rest of the population. As the skills gap would have led to a staffing crisis in any case in a few years, AIDS losses would worsen the situation. Namibia continues to have problems staffing its schools with trained teachers. It is noted that many of teaches in schools, especially in the north, do not have academic qualifications beyond secondary school and very few have undergone teacher training. As some estimates suggest that nearly 35 percent of teachers are likely to be HIV-positive, there would be dramatic losses in the education sector due to AIDS. Even taking into account a decrease in the overall demand for teachers, the loss of teachers would imply that teacher training institutions would have to double their yearly output over the next ten years. This implies that almost one in every six secondary school graduates would need to complete teacher-training college and enter teaching to keep up with the demand. The impact on the private sector is understudied and therefore little understood. AIDS may increase costs for local businesses. Cost increases may be both internal and external. Internal costs may include increased absenteeism, higher pension payouts, and breakdown in worker discipline and morale. There is no sufficient information available for firms in Namibia. Studies in other countries suggest that the costs may be as low as 1 percent of profit and as high as 9 percent, much of the cost deriving from increased absenteeism. External cost increases could be associated with wage increases resulting from likely skilled labor shortage. Response to HIV/AIDS Developing a response to the epidemic in Namibia requires cognizance of ongoing public, donor and private activities to mitigate its impact. In this context, the Government launched the National AIDS Control Program (NACP) in 1990 after independence. A short-term plan was implemented between 1990 and 1992 followed by a medium term plan between 1992-1997, which focused on prevention messages aimed at the general population, condom promotion, training of counselors and home based care providers, STD/HIV/AIDS case management issues for health workers, and epidemiological surveillance. An external review of NACP conducted in 1996 recommended that to improve program management NACP should include a broader range of actors from other sectors and should explore options for greater collaboration with NGOs. Non-health sectors had played limited role in responding to the epidemic. The review also found that high level support, commitment to action and sense of urgency which are critical to AIDS control programs were lacking as a result of which control was not effective despite the implementation of two plans. vii  The strategic medium term plan II (MTP2) and the national AIDS Coordination Program (NACOP), which were launched in March 1999, focus on (i) social mobilization, (ii) prevention, (iii) access to services, (iv) reduction of discrimination, (v) policy and strategy development, (vi) program management, and (vii) capacity building through research and human resource development (HRD). The bulk of the resources mobilized would be allocated to information dissemination and condom promotion. In addition, greater attention would be paid in this plan to improve access to services for those affected. Available evidence suggests that efforts by the National AIDS Control Program (NACP), with the help of several NGOs and international agencies, has increased awareness of HIV/AIDS, its mode of transmission and consequences. However, there is little evidence to suggest that the average citizen considers herself/himself to be at risk of acquiring AIDS, has a good understanding of the asymptomatic nature of STD/HIV infection, or has altered her/his behavior to lower the risk of STD/HIV acquisition. Educational programs have turned out to be ineffective in controlling the spread of the epidemic. By the middle of the last decade it is estimated that more than 90 percent of the population had an adequate knowledge of HIV, but the information had not resulted in behavioral change. Building upon the Government's initiatives, this study provides an illustrative set of policy actions and costs a wide range of them with focus on reducing the transmission of HIV; prolonging life and reducing AIDS morbidity and mitigating the negative impact of HIV/AIDS on the economy. While the policies are not exhaustive, the preliminary estimates suggest that the cost of various individual elements of prevention could range between 0.1 to 1.3 percent of GDP per year, adding up to an average of 2.7 percent per year. In terms of mitigation, orphan care is estimated at about an average of 2.8 percent of GDP per year. The largest single cost element is hospital care, which is estimated at about 3.1 percent of GDP per annum. The latter points to the need for alternative programs for handling AIDS and terminal care. Although the estimates are preliminary, and merely for illustrative purposes, they point to the likely magnitude of the cost of response and suggest that the programs can be accommodated by existing resources. Compared to the estimated loss of GDP associated with HIV/AIDS (about 18 percent per year on average), the benefits of mitigation far outweigh the cost. Furthermore, since the costs of mitigation are substantially higher than those associated with prevention, it is prudent for Government to intensify preventive measures. Finally, international experience suggests that some of the features of effective national responses are: (a) political will and leadership, (b) societal openness and determination to fight against stigma, (c) strategic responses, (d) multi-sectoral and multilevel action, (e) community based responses, (f) social policy reform to reduce vulnerability, (g) sustained response and (h) learning from experience. To be strategic and to learn from past lessons, it is necessary to strengthen the information base. One of the difficulties in targeting behavioral modification strategies is the absence of behavior studies. Experience suggests that AIDS control programs that are not well focused on groups that are at risk are not effective. Standard prevention messages alone are not likely to effectively change behavior. Focused behavioral research is required to develop effective communication strategies. Therefore, an effective monitoring system is essential as part of a response program. viii  CHAPTER I Introduction Since independence in 1990, Namibia's economy has grown moderately (largely on account of political stability), and there has been greater commitment to social development. Real GDP grew at an average of 4 percent annually in the early nineties reversing a decline in per capita income between 1981 and 1991. Nevertheless, it is estimated that close to one half of the population inI993/94 could be considered to be poor, if poverty is defined as household expenditure on food exceeding 60 percent of income (RON, 1998). Poverty is also concentrated, with two-thirds of the poor living in northern parts of the country. The government has responded to poverty by developing a poverty reduction strategy in 1994 aimed at reducing the percent of poor and extremely poor from 47 and 13 in 1994 to 40 and 7 respectively by 2000 (RON, 1997). Complementing the poverty strategy is a population policy whose objective is to reduce the growth rate from 3.2 during 1991-96 to 3 percent by 2006 and 2 percent by 2025; infant mortality from nearly 57 per 1,000 live births to 40 per 1,000 live births by 2006 and 30 per 1,000 live births by 2015; and total fertility rate to 5 by year 2006 and 3.5 by 2015. Some improvements have taken place in Namibia's human development, as its social spending is one of the highest in the world. Spending on social services is almost 50 percent more than South Africa, which has similar poverty and inequity profile. Health and education continue to receive increasing share of the budget up from nearly 38 percent in 1996/97 to 42 percent in 1997/98. There is a significant improvement in human development indicators, mainly on account of increases in enrolments at all levels of the education system. Improvements in education, in particular, have been significant. The number of schools increased from 1,171 in 1990 to 1,457 in 1997. The share of children aged 7 to 13 years attending school has increased from 87.5 percent to 95.2 percent between 1991 and 1995 (RON, 1998). Primary school enrolments increased from 60 percent in 1990 to 95 percent in 1997. By 1996, more than 90 percent of all the children of school-going age were enrolled in school and enrollment in grade 12 had increased from 3,000 in 1990 to 12,000 in 1996. Gender equality also may have improved. Female enrollment has increased and female literacy rates exceed that of males. Improvements in health indicators have not been as significant as in education largely because of the likely effect of HIV/AIDS on the population during the last decade. Under five mortality fell from 110 per 1,000 live births in 1977 to 87 in 1992, but life expectancy at birth may have declined from 58.8 in 1995 to 55.8 in 1998 (UNDP 1998).' Despite significant improvements in the last decade, HIV/AIDS epidemic is likely to substantially weaken the government's ability to meet its development objectives. l The decline in life expectancy, as will be discussed in subsequent chapters, could be attributed to he impact of HIV/AIDS. 1  HIV/AIDS has emerged as the number one cause of death in 1996 (UNDP, 1998). Ministry of Health (MOH) statistics indicate that nearly 7,000 people have died from AIDS by 1998. An estimated 20 percent of the adult population may be living with HIV infection. The nature and magnitude of HIV/AIDS makes it in some ways unique among epidemics or other events that are sources of shocks to economic systems and societies. The objective of this chapter is to review the nature of the epidemic and its likely effect on the development prospects of affected countries. A. NATURE OF THE EPIDEMIC AIDS is caused by Human Immunodeficiency Virus (HIV), which is transmitted through body fluids.2 In Namibia, it is mainly a Pattern Two epidemic in which transmission is primarily through heterosexual contacts. The virus kills individuals by weakening the immune systems, and making them vulnerable to other infections and cancers. MOH statistics indicate that tuberculosis is one of the major opportunistic infections and cause of death of HIV-infected people. AIDS is also fatal, but it can be contained with a cocktail of antiviral drugs to extend productive life for a few years. Recent evidence suggests that these treatments may have significant side effects.3 However, such treatment is expensive and out of reach of the poor in low and medium income developing countries. The significant characteristic of HIV/AIDS is that it affects the most economically and socially active population that would otherwise be least likely to be seriously ill. As the epidemic spreads primarily through sexual contacts, it affects the population that is sexually active, which is in the age group 15 - 49 years. The period of illness could last for many years in the initial stages of which the effect may only be seen in terms of increased absenteeism. Ultimately the affected will withdraw from the workforce. The uniqueness of the epidemic comes from the unusual rate of severe illness and premature death of economically productive population. Some aspects of the epidemic limit individual action to prevent infection and also make public action to control its spread challenging. First, the epidemic is characterized by a long gap between infection and the onset of any signal of infection. Individuals infected with HIV may not experience any symptoms for as long as 5 to 7 years. During this period, infected individuals may continue their sexual activity, spreading the virus to their sexual partners. As infected individuals do not show any outward symptoms during this period, other individuals may not be able to take precaution against infection by avoiding unsafe sexual contact with infected individuals. 2 Like other sexually transmitted diseases (STDs), HIV is difficult to transmit except by sex or other direct contact with the bodily fluid of an infected person. The major modes of transmission are sexual intercourse, reuse of contaminated syringes by injecting drug users, infection via birth or nursing from mother to child, reuse of needles in medical settings, and transfusion of contaminated blood or blood products. 3 Toxic effects of drug therapy over extended periods are becoming apparent now. An important side effect is the redistribution of fat in the body resulting in humped back and distended bellies. The need to maintain a strict regimen of drugs also substantially reduces quality of life. 2  Second, there is stigma attached to those that are infected with HIV as it is considered to be acquired through 'immoral' behavior. Families may reject infected individuals and communities may ostracize affected families. In addition to the morality- related stigma, fear of being infected through association, and economic losses from debilitation and death of the infected individuals, also deny HIV/AIDS patients and their family members access to services and livelihood opportunities. They may not be able to get the usual services at health centers. Their children may be harassed at schools. They may be denied employment or terminated from employment under a pretext.4 Since public knowledge of one's infection brings emotional, social and economic consequences, individuals would not have the incentive to find out their status or to reveal the information to casual sex partners or spouses as it suggests that they have been unfaithful. Those that feel that they can do little about the disease except to wait for debilitation and death, would be particularly hesitant even to find out for themselves whether they are infected. HIV/AIDS affects, depending on the country and prevailing sexual practices, people at both ends of the socio-economic spectrum. At the upper end, they could include the rich, the educated and migrant workers (separated from family for long periods) with more than usual opportunities to engage in risky behavior. At the other end, the poor and the unemployed (especially the young women) who could engage in risky behavior for commercial purposes. B. Likely Impact The most likely impact of HIV/AIDS is a "shock to a existing system of livelihood at the national, regional, communal or household level which reduces the level and expectations of life and welfare" (UNAIDS, 2000). AIDS impacts livelihood through premature morbidity and mortality of labor. Some individuals, households, communities and economies may be able to cope with the virus better than others, but the epidemic can have significant effect at all levels of society. Impact at Household Level: The most significant effect of AIDS may be at the household level. Households may experience dramatic decrease in income, consumption and savings because of loss of income earners. Loss of income and substantial medical expenses could deplete family assets and could accentuate poverty. It may lead to food insecurity especially for children who may be taken out of school to save resources and to take care of ailing parents. The breakdown of family structure may have serious effect on how the children are taken care of. Elderly population living on meager retirement income may be burdened with bringing up children orphaned by the epidemic. 4 Reid (1993) characterized the HLIVAIDS epidemic as going through for phases: The first phase is felt as opportunistic infections occur, psychological trauma is experienced and there is increasing illness and death; the second phase is characterized by the change in demographic and societal structure, with many children and elderly without support; at the third phase, the depletion of labor is felt, especially in the productive and social service sectors; and the fourth phase is characterized by widespread destitution, social and economic disintegration and the possibility of social and political unrest. 3  The impact of the epidemic is often disproportionately higher on women as they tend to be more vulnerable to infection than male counterparts (World Bank, 1996). Women are physiologically more proned to sexual transmission of AIDS/AIDS. In addition, because of their perceived inferior status they are less able to take precaution as they also have less control over decisions relating to sex than their male counterparts. Women with limited economic opportunities, many of whom may also be raising children out of wedlock, are pressured into sexual relations that offer them economic security. They are likely to have sexual relations with older men who have had more exposure to infection in the past. Even when they are not affected, the burden of taking care of the sick often falls on women. Grandmothers also usually raise orphaned children. Demographic Impact: HIV/AIDS is likely to have substantial impact on the demographics of the country, which has implications for both households and the economy. The most important effect will be decreased life expectancy at birth as mortality rate increases. The population will grow at a slower rate than otherwise and there will be smaller labor force. The country may be left with a different population structure, with possibly increased burden. In addition to increasing numbers of orphans, family and community structures may be destroyed. Human Development: The impact of AIDS will be more significant on human development - life expectancy, education and economic opportunities - than on economic growth. Life expectancy has already fallen in many of the African countries. A worsening in the human development status of the country may have consequences for economic growth as well in the medium to long term. Economic Impact: Although AIDS causes considerable suffering to households and demographic changes could be dramatic, the estimates of the macroeconomic impact are sensitive to assumptions about how AIDS affects saving and investment rates, and whether AIDS affects skilled vis-a-vis unskilled/unemployed labor force. A likely economic impact is a reduction in savings as AIDS may require diversion of resources to uses that would not have been necessary in its absence. The costs of AIDS may be both direct (medical care) and indirect (labor productivity losses and training costs). Productivity loss measured in terms of weighted healthy years saved per case is higher for AIDS compared to common diseases such as TB and malaria, but less than that of sickle cell anemia, neonatal tetanus, and severe malnutrition (Becker, 1990). Macroeconomic effects are manifested through their impact on productivity, from changes in population growth, size of labor force and its skill composition. In addition to reduction in domestic savings, external capital flow may be affected because of reduced profitability, as well as reluctance of foreign labor to move into AIDS affected regions. The epidemic may also have substantial impact on government expenditures. It will be most apparent in expenditures at the health sector, where demand for services is likely to grow. Unless governments can expand services to meet increased demand, the access to and quality of services will decline. A study of the effect of the epidemic on health care services in a Nairobi hospital indicates that as the number of HIV infected patients grew, because of crowding out, the mortality of non-HIV infected people increased by 4  more than two-thirds (Over, 1998). The government is also likely to bear an increasing share of the health costs as private firns pass the burden to households and to government (Simon, 2000). All sectors of the economy may be affected, depending on the extent to which the relevant labor force is affected. The impact on the public sector itself may be substantial, affecting the ability of government to maintain critical services in health, education and defense, for example. Labor-intensive sectors, such as mining in Africa with disproportionate number of both domestic and foreign migrants in the labor force, are particularly vulnerable. However, the evidence on whether the impacts are significant appears to be mixed (Simon, 2000). The impact of the epidemic on social and economic institutions is particularly difficult to anticipate. For example, high mortality, absence of public information, and uncertainty as to whether someone will be around and for how long may affect trust and willingness to enter into contractual relationships, which are essential for economic growth. Similarly, it is not clear what would be the social impact of a missing generation of adults on children that are neglected or even brought up in orphanages (home-based or otherwise). Despite harsh features of the epidemic - absence of sympathy for the affected, high costs of treatment and certain death - its impact in some countries may be no more significant than that of other catastrophic events. AIDS can balloon into something that can significantly erode the progress in human development made in recent years by many developing countries (Reid, 1993). Public action to control the spread of the disease and to monitor how the disease is affecting the society at various levels is therefore necessary. C. Organization The objective of this report is to provide an overview of selected likely development impact of HIV/AIDS for Namibia. The purpose is to use to the outcome of the analysis to engage in dialogue with Government, relevant stakeholders and the donor community on the appropriate actions to pursue in support of the Government's development strategy on the epidemic. This report examines the status and the likely impact of HIV/AIDS in Namibia, focusing on the likely impact on the demographics of the country and on the economy. It employs conventional demographic and economic models to analyze selected development impacts of the epidemic on the economy, thereby providing an illustration of how these impacts can be incorporated in the regular planning processes. The study was also initiated as an exercise to assist policy makers in Namibia in their effort to incorporate HIV/AIDS into the planning process on a regular basis. As such, it is directed at officials in the agencies of finance and development planning. The nature of public action required and challenges posed in responding to the epidemic are also examined. The review at this stage does not provide a detailed costing of the impact of the epidemic on various sectors of the economy because of data limitations. Detailed costing and policy design alternatives should be the focus of subsequent analysis. 5  The report is organized into five chapters. The second chapter examines status of the epidemic in Namibia, the sources of epidemiological information and the limitations of available information for examining the full range of impacts on the economy and for human resource planning. The demographic implications of the epidemic are presented in the third chapter. The impacts on the economy, focusing on likely sectoral implications are discussed in chapter four. The final chapter deals with the likely policy responses, providing some quantification of the costs associated with implementation over the next fifteen years. 6  Chapter II Status of HIV/AIDS in Namibia Monitoring the epidemic in order to understand how it is progressing and the factors that are contributing to its development is essential for the fornulation of effective strategies to control it. A key tool for doing this is the Sentinel Surveillance System (SSS), which was initiated in 1992 as the principal source of information for monitoring the epidemic in Namibia. The surveillance system entails survey of prevalence among pregnant women who visit antenatal clinics, and STD and TB patients who seek medical assistance. Antenatal clinics are usually chosen as they provide access to a cross section of healthy, sexually active women in the general population. Samples of all blood donors are also tested in Namibia, and this could also serves as sources of information.' The government also offers confidential testing; the results of which are also analyzed centrally by the MOHSS. In addition, many private firms including insurance companies test employees, job applicants or clients for HIV prevalence, but this information is not available to the public. Sentinel surveys have been conducted in Namibia at selected sites every two years since 1992. The number of sites has been increased from 8 in 1992 to 13 in 1998. A sample of 200 is taken at each selected ANC and STD clinic; the subjects enter the sample as they come until the desired number is reached. Unlike the ANC survey, the sample of STD patients includes both sexes and the sample can be stratified by the nature of sexual disease the participants suffer from. In this chapter, a review of prevalence rates observed in antenatal and STD clinics, and estimates for the general population is undertaken. This is followed by information on recorded cases of AIDS, hospitalization and deaths in the country. Information on sexual practices and knowledge of preventive measures is presented next. Finally, the last section explores the need for monitoring and discusses key features of the next- generation monitoring systems. A. HIV Prevalence Antenatal Clinics: HIV prevalence rates observed among pregnant women sampled at antenatal clinics (presented in Table 2.1) indicate that the simple mean has increased from 4.2 percent in 1992 to 17.4 percent in 1998. The median increased from 3 percent in 1992 to 15 percent in 1998. Prevalence among 15-19 year olds increased from 6 percent in 1994 to 12 percent in 1998. Among 20 to 24 year olds, it increased from llpercent in 1994 to 20 percent in 1998 (UNAIDS, 2000a). In 1998, prevalence rate ranged from 6 percent to 34 percent among various sites. ' Blood donor are representative of the general population. In 1998, 1.6 percent of blood donors was diagnosed to be HIV positive. Prevalence was 0.6 percent among regular donors and 5.5 percent among first time donors. 7  Table 2. 1: HIV Prevalence Among Pregnant Women Attending ANC 1992-1998 (In Percentage) Area Site 1992 1994 1996 1998 Rural Andara 2 11 16 Engela 7 18 17 Nankudu 13 Nyangana 6 5 10 Onandjokwe 8 17 21 Opuwo 3 1 4 6 Urban Katima Mulilo 14 25 24 29 Oshakati 4 14 22 34 Otjiwarongo 2 9 16 Rundu 8 8 14 Windhoek 4 7 16 23 District Gobabis 1 9 Keetmanshoop 3 8 7 Swakopmund 3 7 17 15 Walvis Bay 29 National 4.2 8.4 15.4 17.4 Source: Republic of Namibia: Report of the 1998 HIV Sentinel Sero Survey, Ministry of Health and Social Services, April 1999. There is wide variation among regions. In some of the urban sites such as Katima Mulilo and Oshakati fairly high levels of infection were observed in the early nineties itself. The wide variation in prevalence among regions may reflect regional differences in the timing of the first HIV incidence. At most locations, the observed prevalence among pregnant women showed an upward trend consistent with the hypothesis of an S-shaped progression in the epidemic.2 As has been observed in other parts of Sub-Saharan Africa, prevalence observed at sites in major urban areas and those outside major urban areas (rural)3 have converged, although it is difficult to distinguish between rural and urban areas as some of the survey sites serve both rural and urban residents. Extrapolation of HIV prevalence observed among pregnant women attending antenatal clinics to the general population is fairly robust particularly in populations in which virus transmission is primarily through heterosexual contacts, but attention needs to be paid to see if those who attend antenatal clinics are representative of the general population of pregnant women as well as the overall general population (UNAIDS, 2000b). The SSS excludes those who do not use public antenatal clinics. It also excludes those who use private facilities, typically those with higher incomes, and those who do not use any, typically poor rural residents without access to health facilities. Both these groups could potentially run higher risks of infection, the former because of mobility 2 HIV prevalence increases slowly during the first phase, sharply rises during the second phase, and levels off in the third phase as the number of newly infected persons equals the number of persons dying of HIV/AIDS. 3As urban usually refers to only metropolitan areas, rural areas may include some non-metropolitan urban areas. 8  afforded by their socio-economic status, and the latter because of limited access to information and lower ability to exercise control over their sexual partners or relations. Additionally, as the epidemic matures, fertility declines among those affected by the virus but this is not captured by surveys at antenatal clinics. Therefore, the observations from ANC may potentially distort the true extent of HIV/AIDS in the general population. Sampling methods adopted in Namibia also introduce biases in the ANC data. First, because of the dispersion of the population, several clinics are included in some sites (Table 2.1), thereby making it difficult to ascertain the catchments area for the site;4 second, surveys have not been done at all the sites consistently (notably, Gobabis, Nankudu, and Walvis Bay) thus making it impossible to identify trends and patterns in different communities; and finally, the sentinel sites do not necessarily correspond to the sampling frame used by the national population census, thereby making sub-national projections difficult. Table 2.2: HIV Prevalence Among STD Patients in 1998 le HIV ~~~~95% Area _S_t _ Samp 1 e Negative Positive Prvalencnc __________ ______Size_Prevalence Interval Rural Engela 96 50 46 48 38.58 Nankudu 65 47 18 28 17.39 Onandjokwe 193 125 68 35 28.42 Opuwo 58 52 6 10 3.18 Urban Katima Mulilo 178 70 108 61 53.68 Oshakati 173 108 65 38 30.45 Otjiwarongo 104 76 28 27 18.35 Rundu 199 107 92 46 39.53 Windhoek 222 123 99 45 38.51 District Gobabis 133 109 24 18 12.25 Keetmanshoop 84 61 23 27 18.37 Swakopmund 137 91 46 34 26.41 Walvis Bay 203 122 81 40 33.47 National 1,845 1,141 704 38A __ Source: Republic of Namibia: Report of the 1998 HIV Sentinel Sero Survey, Ministry of Health and Social Services, April 1999. STD Clinics: The observed prevalence rates among STD patients presented in Table 2.2 are substantially higher than observed among pregnant women. In 1998, a median of 42 percent of STD patients tested in Windhoek and Walvis Bay were HIV positive. Median prevalence from the 11 sites outside the major urban areas was 34 percent (UNAIDS, 2000a). As observed in antenatal clinics, differences in the rates among sites may be due to differences in the onset of the epidemic at different locations. However, they reflect largely the relative homogeneity of HIV prevalence rates at the 4 This is necessitated by Namibia's small population and limited coverage at each clinic, especially outside the main metropolitan areas. 9  different sites. In general, at all major urban and sites near major transport corridors, HIV prevalence was higher than 30 percent. As STD patients carry higher risk of infection, prevalence among STD population is likely to be higher than prevalence among the general population for obvious reasons. Though it is not clear whether there was adequate sampling of sexes in different age groups5, the following tentative observations can be made: (i) there are no significant differences in HIV prevalence among men and women, the rates were close to 38 percent for both; (ii) and women are infected at an earlier age than males. For example, 24 percent of women under 20 years were infected compared to only 5 percent of men; 42 percent of women between 20-24 years were infected compared to 24 percent of men in the same age group (Figure 2.1). In general, at all sites, a higher percentage of younger women (under 24 years) was infected relative to their male counterparts.6 Figure 2.1: Age Specific HIV Prevalence in STD Patients 60t -'1 iZl~ ' " 40 WWWgg-g-W 30 El EFemale- - - - 20 > > *l | _ | _ _Ma 10 <20 20-24 25-29 30-34 35-39 40-44 45+ Source: Republic of Namibia: Report of the 1998 HIV Sentinel Sero Survey, Ministry of Health and Social Services, April 1999. The difference in HIV infection rates between men and women could be attributed to the difference in the age of the first onset of sexual activity and the age difference between partners. Since the data suggest that female STD patients, on average, were younger (median age of 26 years) than male patients (median age of 28 years), this may suggest (subject to improved data) the possibility of sexual contacts by younger women with substantially older men. National Estimates: The UNAIDS/WHO working group on Global HIV/AIDS Surveillance, in collaboration with regional and national experts, derives national HIV/AIDS prevalence rates by adopting the following approach: * Available data from SSS in Namibia are reviewed and median rates calculated for "major urban areas" and "outside major urban areas". The figures are then applied to the official urban-rural population distribution in Namibia to get national prevalence data for the year in which the survey is conducted. 5In the total sample, 51 percent were males and 49 percent were females. 6Higher prevalence among STD patients is not surprising as both HIV and STD are transmitted through unprotected sexual contacts. Moreover, those affected by STD are at higher risk of contracting HIV. 10  * The WHO Epimodel 27 is utilized to derive estimates of prevalence and incidence of AIDS and AIDS deaths, as well as the number of children infected through mother-to-child transmission of HIV, taking into account age-specific fertility rates, for other years. On the basis of the above, UNAIDS estimated that 19.94 percent of adults (15 - 49 years) was living with HIV (excluding those with AIDS) in Namibia in 1999. The estimated prevalence rate among 15 to 24 year olds was 18.78 percent to 20.82 percent for females and 7.89 percent to 10.38 percent for males. The estimated total number of individuals living with HIlV/AIDS in 1999 was nearlyl60,000.8 Of them, about 150,000 were adults (15-49) and 6,600 were children. Among adults, the estimated number of women was 85,000 (UNAIDS, 2000). These estimates, for obvious reasons, are nearly three times more than the number reported by the health directorate on the basis of testing in the health system which indicated the cumulative of infection to be 53,330 in 1999. The number of HIV positive cases reported annually by the health system has exceeded 10,000 since 1996. All indications are that the epidemic is widespread in the country, as it is established in the general population. One or more sub-populations may contribute disproportionately to the spread of the virus, but it appears that sexual networking in the general population is adequate to sustain the epidemic independent of high-risk sub populations (WHO&UNAIDS, 2000). B. AIDS Cases and Deaths According to the National AIDS Coordinating Program (NACOP), which maintains information on reported full-blown AIDS cases in Namibia, HIV/AIDS has become the leading cause of death in hospitals in 1996. In 1998, a reported 2,179 deaths due to HIV/AIDS accounted for 23 percent of all reported deaths and for 41 percent of deaths in the 15-49 years age group9. The total number of cumulative reported AIDS cases by end-December 1997 was 6,784 (Figure 2.2). UNAIDS estimated that the AIDS related deaths in 1999 alone were 19,000.10 The full extent of deaths due to AIDS may not be usually captured for the following reasons: (a) many deaths due to opportunistic diseases such as TB are not likely to attributed to AIDS; (b) AIDS patients in remote rural areas with limited access to health facilities may not seek hospital services; and (c) especially in early stages of the epidemic, there may have been poor reporting of AIDS cases by health units. ' Epimodel is a microcomputer program originally developed by the WHO Global Program on AIDS to make medium (under 5 years) terms estimates of the number of persons with HIV/AIDS. 8 The estimated total adult population in 1999 was 790,000 out of a total population of 1,689,000. 9 HIV related diseases are also an increasing cause of hospitalization, estimated to have increased from 355 hospitalizations in 1993 to 5155 in 1998 (Republic of Namibia, 1999). '