Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 53
Part II Papers

OCR for page 53

OCR for page 53
2 Aging in Sub-Saharan Africa: The Changing Demography of the Region Victoria A. Velkoff and Paul R. Kowal INTRODUCTION Population aging will become perhaps the most important demographic dynamic affecting families and societies throughout the world in the com- ing decades. Nearly 63 percent of the population age 60 and older currently resides in developing countries, and this percentage will increase to nearly 73 percent over the next 25 years. Yet the limited understanding of the demographics of aging in most developing countries stands in stark con- trast to the comparatively well-documented course and implications of ag- ing in developed countries. A combination of factors contributes to the limited understanding of the situation of older people in Africa: they constitute a smaller proportion of the population and their proportions are projected to grow fairly slowly relative to other areas in the world.1 In addition, other more pressing po- litical, demographic, and health issues have confronted the subcontinent over the past two decades, and the systems to collect data essential for 1While we recognize the limitations of using a chronological age to define older persons in sub-Saharan Africa, most comparisons in this paper will focus on the population age 60 and over. Largely derived from the creation of a state social welfare system for older workers in developed countries, the use of the age group 60 and over or 65 and over has evolved to become a relatively standard definition of old age worldwide. Age 60 and over was adopted by the United Nations as the standard definition. This standard is not able to fully account for the cultural and societal differences in the definition of “old” between and within countries; how- ever, using this chronological age to define “old” is practical and commonly used for official purposes. 55

OCR for page 53
56 AGING IN SUB-SAHARAN AFRICA accurate demographic estimates and projections are largely lacking. Re- sources available for addressing demographic changes and health prob- lems in Africa have focused on issues of more immediate concern to the great majority of people who are not yet old: infant, child, and maternal health; nutrition; and HIV/AIDS. However, the consequences of recent so- cial and political upheavals—HIV/AIDS, poverty, and violent conflicts— have shaken the core of societies and thrust older people into new roles in families and communities. Despite the fact that the older population makes up a small proportion of the population in most sub-Saharan African countries, the number of older people is growing. In 2005, there were 34 million people age 60 and over in sub-Saharan Africa, and this number is projected to increase to over 67 million by 2030. In fact, the number of older people is growing more rapidly in sub-Saharan Africa than in the developed world. This increase in the number of older people will occur despite the excess mortality due to AIDS that many countries are currently experiencing. This paper is divided into two sections. The first section focuses prima- rily on the demographic aspects of aging in sub-Saharan African countries with a special subsection examining the impact of AIDS on population ag- ing. The demographic data in this first section are from the U.S. Census Bureau’s International Programs Center’s International Data Base. The sec- ond section compares and contrasts the estimates and projections from the U.S. Census Bureau with those of the United Nations (UN) Population Di- vision. This section presents, compares, and contrasts these two sources of demographic estimates and projections, focusing on populations age 60 and older. The underlying models and assumptions, input data, and the result- ing output data are examined to describe the demographic aspects of aging in sub-Saharan African countries. The concluding section provides sugges- tions for future work in the area. DEMOGRAPHIC DIMENSIONS The world is aging at an unprecedented rate. The numbers of older persons and pace of aging vary widely between and within regions, and typically more developed regions have higher proportions of their popula- tions in older age groups than do developing regions (Figure 2-1). For ex- ample, nearly 21 percent of Europe’s population was age 60 and over in 2005. In contrast, less than 5 percent of sub-Saharan Africa’s population was age 60 and over. In other developing regions, those aged 60 and over make up between 7 and 9 percent of the population. In all regions of the world, the proportion age 60 and over is projected to increase in the future.

OCR for page 53
57 THE CHANGING DEMOGRAPHY OF THE REGION 20.7 Europe 30.6 16.9 Northern America 25.4 14.4 Oceania 21.6 9.4 Asia 17.5 2005 8.9 Latin America/Caribbean 16.8 2030 6.8 Near East/North Africa 12.4 4.7 Sub-Saharan Africa 5.5 FIGURE 2-1 Percentage age 60 and over by region of the world: 2005 and 2030. SOURCE: U.S. Census Bureau (2005). By 2030, over 30 percent of Europeans are projected to be age 60 and over. In Asia and Latin America and the Caribbean, the proportions age 60 and over are projected to nearly double in less than 25 years. Again, sub- Saharan Africa stands in contrast to the other regions of the world with the proportion age 60 and over projected to grow only slightly, from 4.7 per- cent in 2005 to 5.5 percent in 2030. The Misconception of “No Older People” in Africa The small increase in the proportion age 60 and over in sub-Saharan Africa masks a large increase in the number of people in this age group. The number of people age 60 and over in sub-Saharan Africa will nearly double from over 34 million in 2005 to over 67 million in 2030. The number of older people is growing more rapidly in sub-Saharan Africa than in the developed world and will continue to do so in the future (Figure 2-2). The average annual growth rate of the population age 60 and over in sub- Saharan Africa is over 2 percent and will increase over the next 50 years to nearly 4 percent. In contrast, the average annual growth rate of this popula- tion in developed countries is less than 2 percent and is projected to decline to less than 1 percent over the next several decades.

OCR for page 53
58 AGING IN SUB-SAHARAN AFRICA 4.0 3.5 Sub-Saharan Africa 3.0 2.5 Percentage 2.0 More developed countries 1.5 1.0 0.5 0.0 2000-2005 2005-2010 2010-2015 2015-2020 2020-2025 2025-2030 2030-2035 2035-2040 2040-2045 2045-2050 FIGURE 2-2 Average annual growth rates of the age 60 and over population in sub-Saharan Africa versus more developed countries: 2000 to 2050. SOURCE: U.S. Census Bureau (2005). Country Comparisons Population aging in sub-Saharan Africa is not uniform. Both the size of the 60 and over population and the proportion of the population they ac- count for varies among the countries of the region.2 Eight Countries Have at Least 1 Million People Age 60 and Over In 2005, Nigeria ranked among the top 30 countries in the world on the basis of the size of its population age 60 and over. Nigeria had the largest older population in sub-Saharan Africa, with over 6 million people age 60 and over; South Africa had just over 3.4 million (Figure 2-3). Six additional sub-Saharan African countries had over 1 million people age 60 and over in 2005. 2There are 50-53 countries in sub-Saharan Africa. The UN Population Division includes 50 countries and the U.S. Census Bureau, 51. This paper focuses on 42 countries that had total populations of at least 1 million in 2005. The countries not included in tables and figures are Cape Verde, Comoros, Djibouti, Equatorial Guinea, Mayotte, Reunion, Saint Helena, Sao Tome and Principe, and Seychelles.

OCR for page 53
59 THE CHANGING DEMOGRAPHY OF THE REGION Nigeria 6.4 South Africa 3.4 Ethiopia 3.2 Congo (Kinshasa) 2.4 Sudan 1.6 Tanzania 1.5 Kenya 1.2 Ghana 1.1 FIGURE 2-3 Sub-Saharan African countries with at least 1 million people age 60 and over: 2005 (number of people age 60 and over in millions). SOURCE: U.S. Census Bureau (2005). Nigeria 12.3 Ethiopia 6.6 Congo (Kinshasa) 5.1 South Africa 4.8 Sudan 4.5 Kenya 3.0 Ghana 2.8 Tanzania 2.7 Uganda 2.2 Madagascar 2.1 Cameroon 1.7 Cote d'Ivoire 1.5 Senegal 1.4 Mozambique 1.3 Burkina Faso 1.1 Niger 1.0 FIGURE 2-4 Sub-Saharan African countries with at least 1 million people age 60 and over: 2030 (number of people age 60 and over in millions). SOURCE: U.S. Census Bureau (2005). The size of older populations in many sub-Saharan African countries is roughly equivalent to certain developed countries. For instance, Nigeria’s older population is roughly the same size as those in South Korea and Canada. The list of countries with at least 1 million people age 60 and over is projected to increase to 16 by the year 2030 (Figure 2-4). Again, Nigeria will have the largest older population, with over 12 million people age 60 and over, and Ethiopia will rank second, with over 6 million people. Congo (Kinshasa) and South Africa are projected to have nearly 5 million older people in 2030. Burkina Faso, Cameroon, Cote d’Ivoire, Madagascar, Mozambique, Niger, Senegal, and Uganda are all projected to have their older populations grow to over 1 million people by 2030.

OCR for page 53
60 AGING IN SUB-SAHARAN AFRICA Mauritius Is the Oldest Country in Sub-Saharan Africa Although the proportion age 60 and over is just under 5 percent for sub-Saharan Africa as a whole, a number of countries have much higher proportions in this age group (Figure 2-5). Over 9 percent of Mauritius’s population was age 60 and over in 2005, making it the oldest country in sub-Saharan Africa. South Africa had 7.8 percent of its population age 60 and over in 2005 and nearly 7 percent of Lesotho’s population was in this age group. At the other end of the spectrum are such countries as Benin, Burundi, Kenya, Mauritania, Rwanda, Uganda, and Zambia, where the older population accounted for less than 4 percent of the total population. By 2030, nearly 22 percent of the population of Mauritius is projected to be age 60 and over. In South Africa over 12 percent of the population is projected to be 60 and over (Figure 2-6). While the proportion of this popu- lation group is projected to increase in some countries (for example, Congo [Brazzaville], Ghana, Mauritius, and South Africa), the proportion age 60 and over is projected to remain fairly stable in many sub-Saharan African countries. For instance, 4 percent of Burundi’s population in 2005 was age 60 and over, and this proportion is projected to stay the same in 2030. In other countries, the proportion is projected to decrease slightly. In Malawi, the percentage is projected to decrease from 4.2 percent in 2005 to 3.7 percent in 2030. Mauritius 9.5 South Africa 7.8 Lesotho 6.9 Central Africa Republic 6.1 Eritrea 5.5 Botswana 5.3 Ghana 5.2 Nigeria 5.0 Senegal 4.8 Mozambique 4.6 Cote d'Ivoire 4.6 Ethiopia 4.4 Malawi 4.2 Tanzania 4.1 Burkina Faso 4.0 Rwanda 3.8 Zambia 3.7 Kenya 3.6 Uganda 3.4 FIGURE 2-5 Percentage age 60 and over in selected sub-Saharan African countries: 2005. SOURCE: U.S. Census Bureau (2005).

OCR for page 53
61 THE CHANGING DEMOGRAPHY OF THE REGION Mauritius 21.9 South Africa 12.4 Ghana 8.7 Senegal 7.7 Lesotho 7.2 Eritrea 6.3 Cote d'Ivoire 5.9 Central African Republic 5.8 Kenya 5.7 Ethiopia 5.7 Mozambique 5.5 Rwanda 5.4 Nigeria 5.3 Botswana 5.2 Tanzania 4.8 Burkina Faso 4.0 Malawi 3.7 Zambia 3.5 Uganda 3.3 FIGURE 2-6 Percentage age 60 and over in selected sub-Saharan African countries: 2030. SOURCE: U.S. Census Bureau (2005). Although the proportion age 60 and over is on average projected to remain stable or decrease slightly in many countries, the absolute number of people in this age group is projected to grow in most countries. For example, the decrease in the proportion age 60 and over in Malawi between 2003 and 2030 masks an increase in the absolute number of people in this age group of around 280,000. Older Populations Projected to Grow in Sub-Saharan African Countries The change in the proportion of the population age 60 and over in most sub-Saharan African countries does not indicate the magnitude of change. The absolute number of people age 60 and over is projected to increase over the next three decades. However, there are exceptions, such as Botswana, Lesotho, and Swaziland. These three countries are severely affected by the AIDS epidemic, and their populations age 60 and over are projected to decrease between 2005 and 2030. Conversely, the number of older people in some countries is projected to more than double by 2030. In Sudan, for example, the number is expected to nearly triple (Figure 2-7). Composition of Older Age Groups In many countries in the world, the oldest old (those age 80 and over) is the fastest growing segment of the population. This is true for a majority of

OCR for page 53
62 AGING IN SUB-SAHARAN AFRICA Sudan 171 Somalia 167 Madagascar 152 Senegal 149 Ghana 147 Uganda 144 Kenya 144 Congo (Brazzaville) 114 Ethiopia 104 Cote d'Ivoire 96 Nigeria 92 Tanzania 80 Angola 71 Malawi 52 Zambia 50 Mali 46 South Africa 38 Zimbabwe 28 Botswana -7 Lesotho -10 Swaziland -17 FIGURE 2-7 Percentage increase in the population age 60 and over in selected sub- Saharan African countries: 2005 to 2030. SOURCE: U.S. Census Bureau (2005). sub-Saharan African countries as well. In sub-Saharan Africa there were around 2.4 million people age 80 and over in 2005, and this number is projected to nearly triple to 6.1 million by 2030. Despite the rapid growth in the number of people age 80 and over, the oldest-old population ac- counted for less than 1 percent of the total population of sub-Saharan Af- rica in the years 2005 and 2030. While the oldest old account for a very small proportion of the total population, they accounted for 7.1 percent of the 2005 population age 60 and over in sub-Saharan Africa. By 2030, the proportion will increase to 9.1 percent. In the more developed region, the oldest old will account for 22.6 percent of the population age 60 and over in 2030 and 12 percent in countries in the less developed regions. Factors Affecting Population Structure Impact of AIDS Seen in Population Pyramids The extensive spread of HIV started in sub-Saharan Africa in the late 1970s, but it was not until the late 1980s that the epidemic exploded in Southern Africa (Joint United Nations Programme on HIV/AIDS and World Health Organization, 2003). Whereas the HIV/AIDS pandemic has consisted of various distinct epidemics, with geographic and population differences, almost all countries in sub-Saharan Africa have generalized

OCR for page 53
63 THE CHANGING DEMOGRAPHY OF THE REGION epidemics. At the end of 2004, about 25.4 million of the estimated 39.4 million people worldwide living with HIV/AIDS were in this region, ac- counting for approximately two-thirds of the global burden (Joint United Nations Programme on HIV/AIDS, 2004). South Africa has the largest number of people living with HIV/AIDS in the world, 5.3 million. Botswana and Swaziland have the highest prevalence levels, both approach- ing 40 percent with no signs of leveling off. West Africa has been relatively less affected by HIV infection than other regions of sub-Saharan Africa, but the spread of HIV from forced migration in this subregion is a signifi- cant cause for concern. In those countries most affected by HIV/AIDS, the age-specific impact on mortality is reshaping population structures. The death of adults in their prime reproductive and economically productive years has changed age pyramids, through declining fertility and increasing mortality, resulting in very atypical age distributions both now and for the next few decades. Specific details of the impact are provided in the next section, but a good illustration of the impact is evident on examination of the population pyramids for Botswana and Zimbabwe. Comparisons of the age and sex structures over time for each country reveal the magnitude of the devasta- tion. Figures 2-8 through 2-11 show the age and sex structure of the popu- lations of Botswana and Zimbabwe for 2005 and 2030. These pyramids show the population estimates and projections with AIDS mortality incor- porated into the projections and what the population structures would have looked like without AIDS mortality. The 2005 population of Botswana is somewhat smaller than it would have been if there was no AIDS mortality (Figure 2-8). By 2030, Botswana’s population age and sex structure is projected to be dramatically different from what it would have been without AIDS mortality (Figure 2-9). Botswana’s total population in all age groups is projected to decrease slightly between 2005 and 2030, dropping from about 1.6 million in 2005 to 1.5 million in 2030. The population age 60 and over is also projected to de- crease slightly over the same time period. In 2005, there were 86,000 people age 60 and over, and by 2030 this population is projected to be 80,000. The age and sex structures for Zimbabwe also show the impact of AIDS mortality; however, the impact is slightly less severe than that on Botswana. Zimbabwe’s population in 2005 is somewhat smaller than it would have been without AIDS mortality (Figure 2-10). By 2030, the impact of AIDS can clearly be seen in the age and sex structure of the population (Figure 2- 11). Unlike Botswana, the total population in Zimbabwe will be larger in 2030 than it was in 2005, despite the impact of AIDS. The older population will also continue to grow in Zimbabwe. In 2005, there were 614,000 people age 60 and over, and this number is projected to grow to 783,000 by

OCR for page 53
81 THE CHANGING DEMOGRAPHY OF THE REGION before 1997 1997-1999 2000-2005 FIGURE 2-17 Most recent census dates. SOURCE: U.S. Census Bureau (2005). data collection for each country. While censuses provide invaluable data, they have the disadvantage of long time periods between rounds and time lags (sometimes significant) between data collection and data availability. Some sub-Saharan countries have data from a recent census (taken less than five years ago). However, others have postponed censuses from 2000 out to 2005 or later. Although certain countries took censuses near 2000, the data are not always available for use until much later (e.g., Senegal took their census in 2002 but the data have not yet been released). Other data sources, such as vital registration data, demographic sur- veillance sites, and national surveys are also used as input data. However, many of these other sources have limitations, particularly for the countries of sub-Saharan Africa. Vital registration systems with high coverage are uncommon in most countries in sub-Saharan Africa. Where they do exist, coverage is variable (the World Health Organization has data from only

OCR for page 53
82 AGING IN SUB-SAHARAN AFRICA nine countries in sub-Saharan Africa, with coverage rates ranging from 5 percent in Mozambique to 99 percent in the Seychelles) (Kowal, Rao, and Mathers, 2003). Thus, vital registration data that can be used as input data for projections are not available for use in most projections of sub-Saharan populations. Another source of input data is national demographic surveys, such as the Demographic and Health Surveys (DHS). These surveys are often con- ducted more frequently than censuses and produce high-quality data for estimations of fertility and infant and child mortality, but they do not pro- vide adult mortality estimates. Typically, infant and child mortality esti- mates derived from DHS data are matched to model life tables to produce estimates of adult mortality patterns. However, given that the model life tables available were developed before the onset of the HIV/AIDS epidemic, they cannot be used without major adjustments to take into account the impact of AIDS deaths. Demographic surveillance field sites, such as those in INDEPTH, po- tentially provide high-quality data; however, the data are not typically na- tionally representative.4 Data from demographic surveillance sites have not been used by the Census Bureau or the UN in the estimates and projections discussed in this paper. Migration data are derived from various sources, including the UN (the United Nations High Commission for Refugees, the United Nations Statis- tics Division, and the United Nations Economic Commission for Europe Statistics Division) and the International Organization for Migration. Mi- gration data are notoriously difficult to obtain and available data are gener- ally considered to be unreliable. The political will throughout Africa to address migration policies and to obtain these data is improving, yet the realities of current data collection systems suggest that improvements will take time (African Union Commission, 2004). Data on forced, internal, and international migration are fraught with problems. Cross-border migration and internal displacement continue to create migration flows that remain difficult to track as the frequency, timing, and duration of migration pat- terns are subject to rapidly changing factors, such as household disintegra- tion due to HIV/AIDS, economic forces driven by globalization, and natu- ral and manmade disasters, many of which disproportionately affect countries in this region. In addition, new migration patterns have developed as a result of AIDS, countering the urbanization trends: adult children “going home to die,” moving from urban areas to rural homes, to be cared for by their parents 4INDEPTH is an international network of field sites with continuous demographic evalua- tion of populations and their health in developing countries.

OCR for page 53
83 THE CHANGING DEMOGRAPHY OF THE REGION and families (Foster, 1995). AIDS deaths are contributing to the disintegra- tion of households, resulting in orphaned children being forced to relocate, and usually to poor areas (Richter, 2004). Both the Census Bureau and the UN use data on HIV/AIDS prevalence rates provided by the Joint United Nations Programme on HIV/AIDS (UNAIDS) and World Health Organization (WHO) Epidemiology Refer- ence Group (Joint United Nations Programme on HIV/AIDS and World Health Organization, 2004a). These prevalence rates are based on the best available data from different national sources, including antenatal clinic surveillance sites and national surveys. Both organizations used the UNAIDS 2004 release of HIV prevalence rates in their projections. The prevalence rates that underlie the mortality assumptions for both the Cen- sus Bureau and the UN projections are presented in Table 2-5. The charac- teristics and impact of HIV/AIDS vary throughout the subcontinent, which affects the magnitude and timing of their effects on demographic estimates and projections. In general, UNAIDS divides the magnitude of the infection into three states: (1) generalized, defined as HIV prevalence consistently over 1 per- cent in pregnant women; (2) concentrated, defined as HIV prevalence con- sistently over 5 percent in at least one subpopulation at highest risk and prevalence below 1 percent in the general adult population ages 15 to 49 in urban areas; and (3) low, defined as HIV prevalence has not consistently exceeded 5 percent in any defined subpopulation (Joint United Nations Programme on HIV/AIDS and World Health Organization, 2003). Fertility Assumptions Historically, declines in fertility have been the main determinant of population aging in developing countries. Countries that have experienced rapid declines in fertility have also experienced rapid increases in aging (for example, South Korea and Thailand). Fertility rates for most countries in sub-Saharan Africa are still high. In many, fertility is declining slowly, which contributes to the relatively slow rate of population aging in the region (United Nations, 2003a). Future trends in fertility will affect the way that countries in the region will age. According to Census Bureau estimates, sub-Saharan African countries accounted for 8 of the top 10 highest fertility rates in the world in 2005. Niger and Mali had the two highest total fertility rates in the world, at more than 7 births per woman, and Somalia’s estimated total fertility rate was 6.8. Only three countries in the subregion (Botswana, Mauritius, and South Africa) had total fertility rates below 3.0 children per women, and only Mauritius had a total fertility rate below the replacement level fertil- ity of 2.1.

OCR for page 53
84 AGING IN SUB-SAHARAN AFRICA TABLE 2-5 Prevalence Rates for Sub-Saharan Africa HIV Adults Ages 15 to 49 from the Joint United Nations Programme on HIV/AIDS: End of 2003 Adult (Low Estimate and Prevalence (%) High Estimate) Sub-Saharan Africa 7.5 [6.9 - 8.3] Angola 3.9 [1.6 - 9.4] Benin 1.9 [1.1 - 3.3] Botswana 37.3 [35.5 - 39.1] Burkina Faso 4.2 [2.7 - 6.5] Burundi 6.0 [4.1 - 8.8] Cameroon 6.9 [4.8 - 9.8] Central African Republic 13.5 [8.3 - 21.2] Chad 4.8 [3.1 - 7.2] Congo (Brazzaville) 4.9 [2.1 - 11.0] Congo (Kinshasa) 4.2 [1.7 - 9.9] Côte d’Ivoire 7.0 [4.9 - 10.0] Djibouti 2.9 [0.7 - 7.5] Eritrea 2.7 [0.9 - 7.3] Ethiopia 4.4 [2.8 - 6.7] Gabon 8.1 [4.1 - 15.3] Gambia 1.2 [0.3 - 4.2] Ghana 3.1 [1.9 - 5.0] Guinea 3.2 [1.2 - 8.2] Kenya 6.7 [4.7 - 9.6] Lesotho 28.9 [26.3 - 31.7] Liberia 5.9 [2.7 - 12.4] Madagascar 1.7 [0.8 - 2.7] Malawi 14.2 [11.3 - 17.7] Mali 1.9 [0.6 - 5.9] Mauritania 0.6 [0.3 - 1.1] Mozambique 12.2 [9.4 - 15.7] Namibia 21.3 [18.2 - 24.7] Niger 1.2 [0.7 - 2.3] Nigeria 5.4 [3.6 - 8.0] Rwanda 5.1 [3.4 - 7.6] Senegal 0.8 [0.4 - 1.7] South Africa 21.5 [18.5 - 24.9] Swaziland 38.8 [37.2 - 40.4] Togo 4.1 [2.7 - 6.4] Uganda 4.1 [2.8 - 6.6] Tanzania 8.8 [6.4 - 11.9] Zambia 16.5 [13.5 - 20.0] Zimbabwe 24.6 [21.7 - 27.8] SOURCES: Joint United Nations Programme on HIV/AIDS and World Health Organization (2004b).

OCR for page 53
85 THE CHANGING DEMOGRAPHY OF THE REGION Fertility rates used in the estimates are derived from census and na- tional household survey data. For its projections, the Census Bureau takes trends in observed fertility rates for a country and calculates the decline in the future based on a logistic function. The UN assumes that fertility de- cline follows a path derived from models of fertility decline that it has es- tablished on the basis of the past experiences of countries with declining fertility during the period 1950 to 2000 (United Nations, 2005). Projected fertility is compared with recent fertility trends in each country and ad- justed so that the projected fertility is consistent with the most recent fertil- ity trends. Projected fertility rates for a number of sub-Saharan countries are shown in Figures 2-18a (Census Bureau) and 2-18b (UN). Both the Census Bureau and the UN project fertility to decrease in all of the coun- tries of sub-Saharan Africa between 2005 and 2030; however, the size of the decrease differs. The Census Bureau projects that fertility will be at or below replacement level in only five countries in sub-Saharan Africa by 2030, and the UN projects that two countries will be at or below replace- ment by this date. The total fertility rate in 2030 is projected to remain above 4 children per woman in 15 of the 42 countries, according to the Census Bureau. These relatively high fertility rates ensure that the propor- tion in the older ages will remain fairly low in many sub-Saharan African countries well into the future. It is unclear how fertility rates will be affected by HIV/AIDS, but at the individual level, as the time infected increases, pregnancy rates drop. Over- all, the most likely result is that an HIV epidemic will slightly reduce fertil- ity, but at this point the data are not available to make reasonable assump- tions about the impact (Stover and Stanecki, 2003). Mortality Assumptions Although declines in fertility have historically been the driving force behind population aging in the countries of sub-Saharan Africa, mortality contributes to population aging, especially in countries highly affected by AIDS. The impact of AIDS has been so large in many of these sub-Saharan countries that it will significantly affect how their populations age. In countries with AIDS mortality, the impact is seen clearly in mortality rates for the adult age groups. These groups are projected to have high mortality rates when AIDS mortality is incorporated into the projections. The mortality rates are adjusted on the basis of HIV prevalence rates. The adult HIV prevalence rate for the countries in sub-Saharan Africa ranges from 0.6 in Mauritania to 38.8 in Swaziland (see Table 2-5) (Joint United Nations Programme on HIV/AIDS and World Health Organization,

OCR for page 53
86 AGING IN SUB-SAHARAN AFRICA 8 7 6 Births per woman 5 4 3 2 1 0 1990 1995 2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050 Botswana Kenya Malawi Senegal South Africa Uganda Zimbabwe FIGURE 2-18a Total fertility rates for selected sub-Saharan African countries: 1990 to 2050 (U.S. Census Bureau data). SOURCE: U.S. Census Bureau (2005). 8 7 6 Births per Woman 5 4 3 2 1 0 1990 1995 2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050 Botswana Kenya Malawi Senegal South Africa Uganda Zimbabwe FIGURE 2-18b Total fertility rates for selected sub-Saharan African countries: 1990 to 2050 (UN data). SOURCE: United Nations (2005).

OCR for page 53
87 THE CHANGING DEMOGRAPHY OF THE REGION 2004a).5 These figures are based on HIV prevalence in women attending antenatal clinics, from which assumptions about infection rates are applied to derive rates for the general population (Joint United Nations Programme on HIV/AIDS, 2004). Despite the limitations and without a functioning vital registration system or representative national-level surveys, prevalence rates derived from antenatal clinic data provide the best source of routinely collected information currently available.6 The Census Bureau incorporated AIDS mortality in 54 countries into their 2005 International Data Base. Of these 54 countries, 39 were in sub- Saharan Africa. The Census Bureau obtained estimates of AIDS-related mortality using a new application that incorporates estimates of HIV preva- lence from the Estimation and Projection Package (EPP)—an epidemiologi- cally realistic model developed and used by the WHO and UNAIDS. EPP produces a national “best fit” curve of adult HIV prevalence using sentinel surveillance data pertaining to pregnant women. The Census Bureau used country-specific adult HIV prevalence estimates from EPP for years from the beginning of the epidemic to 2010. The Census Bureau applied assump- tions from the WHO/UNAIDS Epidemiological Reference Group about the age and sex distribution of HIV incidence, sex ratios of new infections, mother-to-child transmission rate, and disease progression. The model al- lows for competing risk of death and projects HIV incidence implied by the EPP estimates of HIV prevalence through 2010, assuming a decline in HIV incidence of 50 percent by 2050. The model can include the impact of antiretroviral therapy, but the current projections assume no one will re- ceive treatment (U.S. Census Bureau, 2005). In its 2004 revision, the UN Population Division increased the total number of countries with substantial excess deaths caused by HIV/AIDS to 60; of these 60 countries, 40 are located in sub-Saharan Africa (United Nations, 2005). A slow pace of mortality decline in countries highly af- fected by the AIDS epidemic was used for mortality risk not related to HIV/ AIDS. For countries not considered “most affected” by HIV/AIDS, mortal- ity is projected based on models of changing life expectancy produced by the UN. Infection prevalence data from models created by UNAIDS were used 5The proportion of adults ages 15 to 49 living with HIV/AIDS at the end of 2003. 6There is recent evidence that using data from antenatal clinics to estimate prevalence rates for the entire population may not be appropriate. A recent national survey in Kenya, which tested respondents for HIV infection, found that only 7 percent of the adult population was HIV positive. This contrasts with the estimate of 15 percent prevalence estimated using ante- natal clinic data. In other words, the antenatal clinic overestimated the prevalence rate by 100 percent.

OCR for page 53
88 AGING IN SUB-SAHARAN AFRICA 80 70 60 50 In years 40 30 20 10 0 1985 1990 1995 2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050 Botswana Kenya Malawi Senegal South Africa Uganda Zimbabwe FIGURE 2-19a Life expectancy at birth in selected sub-Saharan African countries: 1990 to 2050 (U.S. Census Bureau data). SOURCE: U.S. Census Bureau (2005). to estimate past dynamics and create projections for annual incidences of HIV infection. The 2004 UN revision projects the impact of HIV/AIDS to be less severe than was previously forecast in the 2002 revision. This differ- ence is due to the revised and lower estimates of HIV prevalence in several countries (based on UNAIDS data for 2003) (United Nations, 2005; Joint United Nations Programme on HIV/AIDS, 2004). Also in the 2004 revi- sion, the UN has assumed that beginning in 2005 changes in behavior and treatment will reduce the chances of infection in the future. Both the Census Bureau and the UN project that life expectancy at birth will continue to decline for countries in which AIDS mortality is present (see Figures 2-19a and 2-19b). However, life expectancy at birth is projected to increase in most countries beginning some time after 2010. CONCLUSION Accurate statistics on basic demographic events are the foundation of rational health and public policy, yet many countries lack sound demo- graphic information. In particular, data on both the number and causes of death in sub-Saharan African countries are virtually nonexistent. Reliable adult mortality data on levels, let alone causes, simply do not exist for the majority of the countries in sub-Saharan Africa. Mortality estimates are modeled from limited sources of data, such as surveys, censuses, and demo- graphic surveillance sites (in the small number of countries where they ex- ist). Currently there is a paucity of high-quality country-level data on mor-

OCR for page 53
89 THE CHANGING DEMOGRAPHY OF THE REGION 80 70 60 Life Expectancy 50 (in years) 40 30 20 10 1985 1990 1995 2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050 Botswana Kenya Malawi Senegal South Africa Uganda Zimbabwe FIGURE 2-19b Life expectancy at birth in selected sub-Saharan African countries: 1990 to 2050 (UN data). SOURCE: United Nations (2005). tality for the sub-Saharan countries, and this has an impact on population estimates and projections. Efforts are currently under way to improve the collection of demographic data in many sub-Saharan countries, which will improve the future projections and assist in elucidating how these countries will age in the future. The impact of HIV/AIDS on adult mortality rates in sub-Saharan Af- rica has reshaped the population structure and age distribution in most countries. Approximately 2.3 million people died of AIDS in 2004 (Joint United Nations Programme on HIV/AIDS, 2004). By the year 2020, it is projected that a total of 75 million Africans will have lost their lives to AIDS since the beginning of the epidemic. The impact of AIDS dramatically affects how countries in sub-Saharan Africa will age over the next several decades. Despite the huge impact of AIDS, sub-Saharan Africa is aging and will continue to age. The number of older people is projected to nearly double in less than 30 years. These growing numbers of older people will age in countries that are ill equipped to deal with the challenges that aging populations pose. Explanation of Acronyms/Abbreviations AIDS Acquired Immunodeficiency Syndrome HIV Human Immunodeficiency Virus HMN Health Metrics Network

OCR for page 53
90 AGING IN SUB-SAHARAN AFRICA INDEPTH An international network of field sites with continuous demographic evaluation of populations and their health in developing countries IOM International Organization of Migration SAVVY Sample Vital Registration and Verbal Autopsy UN United Nations Population Division UNECE United Nations Economic Commission for Europe Statis- tics Division UNHCR United Nations High Commission for Refugees UNSD United Nations Statistics Division UNAIDS Joint United Nations Programme on HIV/AIDS USCB U.S. Census Bureau WHO World Health Organization REFERENCES African Union Commission. (2004, March). Experts Group Meeting on the Policy Framework on Migration in Africa, Addis Ababa, Ethiopia. Opening speech by Adv. Bience Gawanas, African Union Commissioner for Social Affairs. Foster, S.A. (1995). A study of adult diseases in Zambia: Final report. London, England: Overseas Development Agency. Joint United Nations Programme on HIV/AIDS. (2004). 2004 report on the global AIDS epi- demic. Executive Summary. Geneva, Switzerland: Author. Joint United Nations Programme on HIV/AIDS and World Health Organization. (2003, Sep- tember). A history of the HIV/AIDS epidemic with emphasis on Africa. Training Work- shop on HIV/AIDS and Adult Mortality in Developing Countries. New York: United Nations. Joint United Nations Programme on HIV/AIDS and World Health Organization. (2004a, De- cember). AIDS epidemic update. UNAIDS/04.45E. Joint United Nations Programme on HIV/AIDS and World Health Organization. (2004b, De- cember). Technical report on improving estimates and projections of HIV/AIDS. Based on a meeting of the UNAIDS/WHO Reference Group for Estimates, Modelling, and Pro- jections, Sintra, Portugal. Geneva, Switzerland: Author. Kowal, P.R., Rao, P.V.C., and Mathers, C. (2003). Report on a WHO workshop: Minimum data set on ageing and adult mortality data in Sub-Saharan Africa. Geneva, Switzerland: World Health Organization. Leitenberg, M. (2003). Death in wars and conflicts between 1945-2000. New York: Cornell University. Peace Studies Program. Richter, L.M. (2004, May). The impact of HIV/AIDS on the development of children. Pre- sented at the Seminar on HIV/AIDS, Vulnerability, and Children: Dynamics and Long- term Implications for Southern Africa’s Security, 4 April, 2003, Pretoria. Institute for Security Studies monograph no. 109. Smith D. (2003). The atlas of war and peace. New York: Penguin Putnam. Stover, J., and Stanecki, K.A. (2003). Estimating and projecting the size and impact of the HIV/AIDS epidemic in generalized epidemics: The UNAIDS Reference Group approach. Geneva, Switzerland: Joint United Nations Programme on HIV/AIDS. United Nations. (2001). World population prospects: The 2000 revision. New York: Author.

OCR for page 53
91 THE CHANGING DEMOGRAPHY OF THE REGION United Nations. (2003a). The impact of HIV/AIDS on mortality. (Workshop on HIV/AIDS and Adult Mortality in Developing Countries. UN/POP/MORT/2003/14). New York: Author. United Nations. (2003b). World population prospects: The 2002 revision. New York: Author. United Nations. (2005). World population prospects: The 2004 revision. (ESA/P/WP.193). New York: Author. U.S. Census Bureau. (2004). The AIDS pandemic in the 21st century. (International Popula- tion Reports No. WP/02-2, available from the U.S. Government Printing Office). Wash- ington, DC: U.S. Department of Commerce. U.S. Census Bureau. (2005). International programs center, international database. Available: http://www.cia.gov/cia/publications/factbook/ [accessed August 2006]. U.S. Central Intelligence Agency. (2004). The world fact book. Available: http://www. census.gov/ipc/wwf [accessed Sept. 2006]. Zaba, B., Marston, M., and Floyd S. (2003). The effect of HIV on child mortality trends in sub-Saharan Africa. Presented at the Training Workshop on HIV/AIDS and Adult Mor- tality in Developing Countries, Population Division, Department of Economic and Social Affairs, United Nations Secretariat, September 8-13, New York.