Profiles for Family Planning and Reproductive Health Programs 116 Countries
Publication date: 2005
2nd EDITION2nd EDITION Profiles for Family Planning and Reproductive Health Programs 116 COUNTRIES Profiles forFam ily Planning and Reproductive H ealth Program s 116 COUNTRIES R o ss ❖ Sto ver ❖ A d ela ja John Ross ❖ John Stover ❖ Demi Adelaja Profiles for Family Planning and Reproductive Health Programs 116 COUNTRIES John Ross John Stover Demi Adelaja Futures Group 2005 2nd EDITION Futures Group 80 Glastonbury Boulevard Glastonbury, Connecticut 06033 USA Cover design: Kim Farcot Printing: Paladin Commercial Printers Newington, Connecticut ISBN 1-59560-002-7 Copyright © 2005 by the Futures Group Any part of this volume may be copied or adapted to meet local needs without permission from the authors or the Futures Group, provided that the parts copied are distributed free or at cost (not for profit). Any commercial reproduc- tion requires prior permission from the Futures Group. The authors would appreciate receiving a copy of any materials in which the text or tables in the volume are used. Contents iii ContentsContentsContentsContentsContents Foreword . vii I. Geographic Patterns of Reproductive Health Problems .1 II. Past Trends in Contraceptive Use . 5 Total Contraceptive Use .5 Use by Method .5 Plateaus in Contraceptive Prevalence and Total Fertility Rates .6 Use by Source by Method .8 III. Future Trends in Contraceptive Use . 17 Projections for the Percentage Using Contraception . 18 Projections for Total Numbers Using Contraception . 19 Projections for Commodities Needed, by Method . 21 Projections for Commodity Costs . 21 IV. Demands on Services . 23 Growing Numbers of Women, Married Women, and Deliveries . 23 Youth: Current Needs and Services . 25 Unmet Need for Youth . 26 HIV/AIDS and Youth . 27 V. Maternal Health . 29 Maternal Mortality and Morbidity . 29 Antenatal, Delivery, and Tetanus Care . 31 Induced Abortion and Postabortion Contraception . 33 Program Efforts to Improve Maternal Health . 35 VI. Child Health . 39 Rates and Numbers of Child Deaths . 39 Risks of Death by Birth Categories . 39 Immunizations, ARI, and ORS . 41 VII. HIV/AIDS Programs and Shortfalls . 43 HIV/AIDS Incidence and Prevalence . 43 Goals and Strategies . 45 VIII. Five Program Objectives . 47 Goal No. 1: To Provide Full Access to a Variety of Contraceptive Methods . 47 Goal No. 2: To Satisfy Unmet Need and Intention to Use a Method . 50 Goal No. 3: To Reach the Desired Fertility Level . 56 Goal No. 4: To Attain the Replacement Fertility Level . 60 Goal No. 5: To Satisfy Millennium Development Goals and the Cairo Programme of Action . 66 Appendices: A: Supporting Tables (list follows) . A.1 B: Technical Appendix for Projection Methods . B.1 iv Contents Appendix A: Supporting Tables Sources for Supporting Tables . A.1 A.1. Contraceptive Use by Method Among Currently Married Women: All Surveys 1980 to Present, Developing Countries . A.3 A.2. Source of Supply for Modern Contraception Methods . A.17 A.3. Population, Number of All Women (15-49), and Married Women (15-49), for 2005, and Percent Using Contraception (latest survey), and Number of Users . A.22 A.4. Number of Married Women of Reproductive Age (15-49) for Four Dates, and Percent Currently Married . A.24 A.5. Projected Contraceptive Prevalence by Method Among Married Women of Reproductive Age . A.26 to A.33 a. Year 2005 . A.26 b. Year 2010 . A.28 c. Year 2015 . A.30 d. Year 2020 . A.32 A.6. Projected Number of Contraceptive Users by Method Among All Women (Aged 15-49) . A.34 to A.41 a. Year 2005 . A.34 b. Year 2010 . A.36 c. Year 2015 . A.38 d. Year 2020 . A.40 A.7. Projected Contraceptive Commodities by Method Among All Women (Aged 15-49) . A.42 to A. 49 a. Year 2005 . A.42 b. Year 2010 . A.44 c. Year 2015 . . A.46 d. Year 2020 . A.48 A.8. Projected Contraceptive Costs by Method Among All Women (Aged 15-49) . A.50 to A. 57 a. Year 2005 . A.50 b. Year 2010 . A.52 c. Year 2015 . A.54 d. Year 2020 . A.56 A.9. Unmet Need, Percent Using, and Percent of Demand Satisfied . A.58 A.10. Intention to Use Contraception . A.60 A.11. Relationship of Unmet Need and Intention to Use Contraception . A.62 A.12. Ideal Number of Children, Total Fertility Rate and Wanted Fertility Rates, and Fertility Planning Status . A.66 A.13. Percent Distribution of the Gap to 75% Contraceptive Prevalence, by Region, Year 2005 Estimates . A.68 A.14. 1999 Program Effort Scores: Total and Four Dimension Scores as Percent of Maximum . A.70 A.15. Maternal Mortality Ratio (MMR), Number of Deaths Annually, Lifetime Risk, and Percent of Female Deaths (ages 15-49) That Are Pregnancy Related . A.72 A.16. Number of Abortions, Abortion Rate, and Abortion Ratio (1999 Estimates) . A.74 A.17. Percentage Receiving Antenatal Care, Tetanus Injections, and Delivery Care . A.76 A.18. Maternal and Neonatal Program Effort Index (MNPI), 1999 and 2002 Surveys: Country Scores as Percent of Maximum . A.78 A.19. Number of Births, Infant and Child Mortality Rates, and Number of Deaths, 2003 Estimates . A.82 Contents v A.20. Births According to Risk Category . A.84 A.21. Immunizations, ARI, and Re-hydration . A.86 A.22. Estimated Number of People Living with HIV/AIDS, Estimated Number of Orphans (AIDS and non AIDS), and Estimated AIDS Deaths, at the End of 2003 . A.88 A.23. Condom Needs . A.90 to A. 93 a. 2005 Projections . A.90 b. 2010 Projections . A.92 A.24. Comparative Information on Youth . A.94 A.25. Demographic Dividend: Percent of the Population Aged 15-59, from 1950 through 2000 and from 2005 through 2050 . A.96 to A.99 vii FFFFForororororeeeeewwwwworororororddddd This volume, an updated and enlarged edition of the first edition, was conceived as a way to assist ac- tion programs by bringing together much of the comparative data that bear upon family planning and reproductive health. A matrix for 116 countries was constructed to embrace time trends for each of nu- merous data sets. The object was to provide both reference information through supporting tables, and basic analyses through textual presentation. The body of the text comments on the chief patterns and trends of each feature, usually by region. The topics chosen embrace a continuum from the demographic context to past and future contraceptive use, to service burdens, maternal and child health, HIV/AIDS, and, finally, to a selection of alternative action objectives. Large countries are given special attention in most sections. Chapter 1 provides an overview of the disparate geographic pattern of reproductive health problems as a backdrop to the rest of the volume. Chapter 2 uses over 300 national surveys to describe contracep- tive use, including trends by method and source. Chapter 3 introduces a special projection method to anticipate future contraceptive use, again by method, with estimates of commodity needs and costs. Chapter 4 summarizes demands on services due to growing population numbers, with particular atten- tion to youth. Chapter 5 concerns maternal morbidity and mortality along with antenatal and delivery care, abortion frequency, and program effort to improve maternal health. Chapter 6 is a parallel chapter for child mortality, risks by type of birth, and program efforts for immunizations and other measures. Chapter 7 is devoted to the HIV/AIDS pandemic. Finally, Chapter 8 presents five action goals, includ- ing full access to contraception, satisfaction of unmet need and intention to use a method, achievement of the desired fertility level, attainment of replacement fertility, and movement toward the Millennium Goals. A set of 25 appendix tables supports these various topics. The intended audience encompasses the many international agencies active in family planning and re- productive health programs. It also includes officials and researchers in individual countries, who can find here a convenient source of information on their own situation, as well as comparative data for their region. We hope also that the text discussions will lead to a deeper understanding of some of the dy- namics that bear on each topic. Any compilation of information that covers numerous topics and many countries incurs a corresponding degree of indebtedness. We wish to express our gratitude to the David and Lucile Packard Foundation, the World Population Society, and The Centre for Population and Development Activities (CEDPA) for underwriting the work that produced this volume. We are grateful to a number of institutions that freely shared data, including Macro International, The United Nations Population Division, The United Na- tions Children Fund (UNICEF), the World Health Organization (WHO), and the World Bank. Our appre- ciation also goes to Cathy Johnson for manuscript preparation, layout, and production. The grant from the World Population Society that inspired this Profiles effort was in fact the Society’s last activity. For many years the Society was active in promoting greater understanding of the issues of population growth and the need for family planning, and in building political commitment to address these issues through effective action. The driving force behind the World Population Society was Phi- lander Claxton. Through his volunteer work at the World Population Society and his official positions with the US State Department and later at the Futures Group, Phil made enormous contributions to the field of international population assistance. He contributed to the post-World War II efforts to establish the United Nations, the World Bank, and country level institutions to help carry out his vision. Those who worked with Phil over the years learned much from his dedication to international development and his tireless efforts to promote appropriate policies and programs to ensure that all couples have the right and the means to achieve their desired family size. We hope that this book will make a contribu- tion to achieving that goal. The distribution of people and events is cast heavily into a relatively few coun- tries, with the remainder spread thinly over about a hundred others. The “size of the problem” is a complex topic: no matter what one considers, whether peo- ple, pregnancies, or deaths, a few coun- tries dominate the globe; a few countries also dominate within each region. We have chosen 116 countries as the subject of this report. These are restrict- ed to those having over one million pop- ulation, covering 98% of the developing world. Included are countries in Latin America, Asia, sub-Saharan Africa, and Middle East/North Africa, together with the five Central Asian Republics, the three Caucasus countries, and the set of Russia, Ukraine, and Moldova. The 116 countries contain over 5 billion people in the UN projections for 2005. A convenient breakdown is as follows (in thousands): China 1,322,273 India 1,096,917 Rest of Asia (except Japan) 1,123,870 Latin America 558,281 Sub-Saharan Africa 693,901 Middle East/North Africa 374,092 The five Central Asian Republics 58,881 The three Caucasus Countries 16,596 The developing world 5,244,811 The set of Russia, Ukraine, and Moldova 193,594 Grand total 5,438,405 As another overview still using the UN 2005 figures, ➤ China has 25% and India has 21% of the developing world, for nearly half of the total (see Figure 1.2). ➤ The top 9 countries (including those) contain two-thirds of the total. ➤ The top 16 countries contain three- fourths of the total. While China and India dominate the whole developing world, a similar im- balance exists within each region (see Figure 1.1). ➤ In the rest of Asia the next largest country, Indonesia, has only 6% of the region’s total; however it has one-sixth of the rest of Asia after the two giants are removed. It is also the world’s fourth largest country, since the breakup of the USSR. ➤ In Latin America, Brazil contains one- third of the total (33%) and Mexico has one-fifth (19%), for over one-half to- gether. The next two, Colombia and Ar- gentina, have only 8% and 7%. Peru and Venezuela have 5% each. Eight of the 24 countries each contains less than 1% of the region’s total. ➤ In sub-Saharan Africa, Nigeria has 19% of the total. Next is Ethiopia with 11% and D.R. Congo with only 8%. The top five, including South Africa and Tan- zania, dominate the 40 members of the region. With 49% of the total population they contain about one-half of all births, infant deaths, and maternal deaths in the region, with the other half spread over the other 37 countries. Fifteen countries each have less than 1% of the region’s population. ➤ In the Middle East/North Africa re- gion, Egypt and Turkey together have 40% of the total, with 20% each. The next two, Algeria and Sudan, have only 9% each. Morocco has 9%; all together these five (of 16) countries contain two- thirds (67%) of the region’s total. The number of women of childbearing age (15-49) is distributed very much as the total population. Concentration is heavy in China and India, and then with- in 2-5 countries within every region. The picture is similar for married/cohabiting women. Outside of China, the number of women aged 15-49 is growing by 9% from 2005 to 2010, and will grow by 17% from 2005 to 2015, totaling a one- sixth increase over 10 years. Growth for married/cohabiting women is essentially similar. Overall, 69.5% are married/co- habiting and while this may decline somewhat the percentage is not expected to change substantially over the planning period to 2005. See Appendix Tables A.6 and A.7. There is also a vast range among devel- oping countries in the pattern of deliver- ies, infant and child mortality, and ma- ternal mortality. Figure 1.2 depicts the uneven geographic distribution of some of these features, as follows. Deliveries follow much the same geo- graphic pattern as populations, except that China’s share is much smaller and sub-Saharan Africa’s share is much larg- er, reflecting their especially low and es- pecially high fertility rates in relation to the rest of the world. However for attended deliveries, the pat- tern changes sharply. China has 25% of all attended deliveries, whereas it has only 15% of all deliveries. Sub-Saharan Africa’s share drops from 20% of deliv- eries to only 15% of attended deliveries. Not surprisingly, infant and child deaths reverse that pattern. Maternal deaths reverse it even more: sub-Saharan Africa’s 41% exceeds the total for all other regions together out- side of India. GEOGRAPHIC PATTERNS OF REPRODUCTIVE HEALTH PROBLEMS Chapter 1Chapter 1Chapter 1Chapter 1Chapter 1 2 Chapter 1 Figure 1.1. Population in Developing Countries China 35% India 29% Other Asian Countries 20% Indonesia 6% Pakistan 4% Bangladesh 4% Viet Nam 2% Brazil 33% Other Latin American Countries 23% Mexico 19% Colombia 8% Argentina 7% Peru 5% Venezuela 5% OtherMiddleEast / NorthAfricanCountries 28% Egypt 20%Turkey 20% Algeria 9% Sudan 9% Morocco 9% Iraq 6% Populations of Asia Populations of Latin America Populations of the Middle East/North Africa Populations of the 15 Largest Developing Countries Populations of Sub-Saharan Africa China 34% India 28% Indonesia 6% Brazil 5% Pakistan 4% Bangladesh 4% Nigeria 3% Mexico 3% 2% Iran 2% Philippines 2% Egypt 2% Turkey 2% Thailand 2% Ethiopia 2% Viet Nam Other Sub-Saharan African Countries 37% Nigeria 19% Ethiopia 11% D.R. Congo 8% South Africa 7% Tanzania 6% Kenya 5% Uganda 4% Ghana 3% Chapter 1 3 Figure 1.2. Populations and Patterns of Deliveries, Infant and Child Mortality, and Maternal Mortality Populations by Region Deliveries Maternal Deaths Infant Deaths Child Deaths Deliveries Attended China 25% India 21% Rest of Asia 23% Sub-Saharan Africa 13% Latin America 11% Middle East/North Africa 7% China 15% India 21% Rest of Asia 24% Sub-Saharan Africa 20% Latin America 12% Middle East/North Africa 8% China 25% India 16% Rest of Asia 21% Sub-Saharan Africa 15% Latin America 14% Middle East/North Africa 10% China 2% India 25% Rest of Asia 19% Sub-Saharan Africa 41% Latin America 9% Middle East/North Africa 4% China 8% India 22% Rest of Asia 21% Sub-Saharan Africa 39% Latin America 4% Middle East/North Africa 6% China 8% India 21% Rest of Asia 20% Sub-Saharan Africa 42% Latin America 4% Middle East/North Africa 6% 4 Chapter 2 5 Chapter 2 presents past trends in contra- ceptive use for (1) total use, (2) use by method, (3) use by source by method, and (4) plateaus in contraceptive preva- lence and total fertility rates. Total Contraceptive Use The rich body of national surveys now available, encompassing some 301 sur- veys in 109 countries taken since 1980 (Appendix Table A.1), and many others prior to 1980, documents the revolution in family planning that has swept much of the developing world since the 1960s. The 1965 average was about 10% of couples using a method; now it is at 60% (UN 2003). The upward trends in most individual countries that dominate Fig- ure 2.1 testify to this revolution. The patterns by region follow. Sub-Saharan Africa. Change has lagged in sub-Saharan Africa but even there certain countries are impressive exceptions, enough so to undermine ear- ly fears that African cultures were near- ly immune to contraceptive adoption. Moreover, clear evidence has appeared of fertility declines in numerous African countries (Cohen, 1998; Kirk and Pillet, 1998; UN 2003). However, prospects are fundamentally different between Anglophone and Fran- cophone Africa (Figures 2.1a and 2.1b). Contraceptive use is rising in Anglo- phone countries and has reached signifi- cant levels in some, as in Kenya, Zimba- bwe, Botswana, South Africa, Namibia, and Zambia. Other trends are also up, though at lower levels, and Nigeria, the largest of all, is flat at a very low level. Francophone countries give a far differ- ent picture, one of very low use levels and only modest suggestions of change. All but Togo fall below 20% of couples using a method, even including tradi- tional methods, and any upward slopes are quite gentle. Remarkably, modern method use is below 13% everywhere. Latin America. Both South America and the Central America and Caribbean regions show patterns of steady rises in use, to substantial levels for many coun- tries (Figures 2.1c and 2.1d). High val- ues occur for Brazil, with its population of partly European extraction; Mexico, with a strong government program; and Colombia, with a strong private sector, along with Costa Rica, Cuba, Domini- can Republic, Jamaica, Peru, and Puerto Rico. Lower values appear for Haiti and Guatemala and Peru’s high values come heavily from traditional methods, though all three have risen somewhat. The Middle East/North Africa. Contra- ceptive use has risen steadily over the years in most countries surveyed (Figure 2.1e). Six of the 16 countries are at or above 60% of couples using a method: Algeria, Egypt, Lebanon, Morocco, Tuni- sia, and Turkey. However, Iraq, Sudan, Oman, and Yemen are at very low levels, although the latter two have risen recently. Asia. The immense continent of Asia is divided here into three sub-regions (Fig- ure 2.1). East Asia has the fewest mem- bers and the highest use levels, in China, Taiwan, Hong Kong, and South Korea. Considerable diversity is present else- where. In Southeastern Asia, the wide spread in use is accompanied by differ- ences in slope: Myanmar and Viet Nam are sharply up, while Indonesia’s rise is relatively steady though at a much high- er level. Southern Asia also presents a diverse pic- ture, from Pakistan at a remarkably low level to Bangladesh’s impressive rise over the years, to Sri Lanka with the highest level in the area. India is at 40-45% but is exceedingly diverse internally. In summary, contraceptive use has risen historically in much of the developing world. It is already at ceiling levels in some countries, and it continues to rise in many others. However, the pattern is uneven: a few of the largest countries, such as India, Pakistan, Nigeria and oth- ers, have far to go, and much of sub-Sa- haran Africa still registers low levels of use. Appendix Table A.1 provides the full results in surveys from 1980 on, and Appendix Tables A.5a-d give the pro- jected level for 116 countries at four dates from 2005 to 2020 (see Chapter 3). Use by Method Current use of each contraceptive meth- od reflects the history of its past adop- tions together with its continuation pat- tern. This is far different for resupply methods such as the pill or condom, where use can cease at any moment, than it is for sterilization, where protec- tion continues automatically for many years. This is one reason why steriliza- tion use has risen to substantial levels in some countries even though rather few couples adopt it in each year. The time trend for each method in each of 22 large countries appears in Figure 2.2. Note that the vertical scales differ, to better clarify the method patterns. The outstanding feature in most countries is the dependence upon only two or at most three methods (and only one method in India and Algeria). However in some countries the sum of all other methods, in the aggregate, protects an appreciable proportion of couples. Seven modern methods of contraception have been available for enough time to reveal the emergent international pat- terns. These are immediately evident in Appendix Table A.1. Overall, the pill and female sterilization are the front runners. For family planning, the con- dom is not dominant in any country, but HIV/AIDS campaigns in some countries have substantially increased its distribu- tion. Except for a few countries (includ- ing China) vasectomy is unimportant. The IUD is important in some countries but not in most. Vaginal applications have won only trivial use, and the new implant methods are of significant use thus far mainly in Indonesia. Each of the methods is now discussed in more detail. Sterilization stands out in Asia, with high figures in the group of China, Tai- wan, South Korea, Hong Kong, and Singapore, and also in Thailand, Sri PAST TRENDS IN CONTRACEPTIVE USE Chapter 2Chapter 2Chapter 2Chapter 2Chapter 2 6 Chapter 2 Lanka, Nepal, Bangladesh, and India. Major exceptions, with little steriliza- tion use, are Indonesia, Viet Nam, and Myanmar. Most Asian sterilization is for females, but male sterilization is sub- stantial in China, South Korea, and Ne- pal (and historically in India and Bang- ladesh, although less so now). Latin America has also seen extensive use of female sterilization, in the two largest countries of Brazil and Mexico, and in Colombia, Costa Rica, Cuba, Do- minican Republic, Ecuador, El Salva- dor, Guatemala, Honduras, Jamaica, Nicaragua, Panama, and Puerto Rico. Little use is made of male sterilization. An exception to the extensive use of fe- male sterilization is the group of Mus- lim countries in the Middle East/North Africa region. In Appendix Table A.1, all countries but one are below 5% of couples using sterilization. Tunisia is the exception; there women with at least a few children have been able to obtain sterilization, and 15% were using it by 1994. On the other hand, Egypt has very little sterilization activity and follows what amounts to an informal policy against it. As a result, the IUD is espe- cially prominent in most Middle East/ North African countries. Sub-Saharan Africa has registered only small figures for sterilization, except for 18% of couples using it in South Africa (as of the 1998 national survey). The trend is up however, in Mauritius at 7%, and Namibia at 8%. The other countries with surveys show nearly negligible lev- els of use. The pill accounts for more use than any other method except sterilization; it is prominent in countries in all regions. Among the 22 large countries in Figure 2.2, it plays an important role (10% or more couples using it) in Brazil and Co- lombia in Latin America (only 8% in Mexico); Bangladesh, Indonesia, Philip- pines, and Thailand in Asia; Algeria, Egypt and Iran in Middle East/North Af- rica; and South Africa in sub-Saharan Africa. Among smaller countries, at above 10% are Hong Kong; Libya, Mo- rocco, United Arab Emirates; Botswana, Mauritius, Zimbabwe; Costa Rica, Do- minican Republic, Ecuador, Honduras, Jamaica, Nicaragua, Panama, Paraguay, and Trinidad and Tobago. In a number of other countries pill use is below 10% of couples but still serves an appreciable share of users. All in all, the pill plays a considerably wider role across many countries than does the IUD or inject- able. The IUD’s pattern is one of minor use in most countries but with major excep- tions – most notably in China, where over one-third of all couples use it. It is the number one method in Egypt, where 37% of couples rely on it; 24% do so in Jordan, 22% in Tunisia, 11% in Libya, 16% in Syria, and 20% in Turkey. Viet Nam has always stressed use of IUD, es- pecially in the North, and nearly 40% of couples use it nationwide. In Taiwan 22%, and in South Korea 13% do so. In Cuba a full 44% of couples use the IUD. The IUD is also prominent within the Central Asian Republics: 38% of cou- ples use it in Kyrgyzstan, 42% in Kazak- hstan, and 52% in Uzbekistan, for the highest figure recorded. A remarkable instance of regional con- trasts is the popularity of the IUD in the Middle East/North Africa versus its near absence in sub-Saharan Africa. By far most surveys, throughout the entire re- gion, show barely 3% of couples using it. The absence of this long-continuation method, together with the neglect of the sterilization method, help explain the low levels of contraceptive prevalence and the elevated fertility rates in the continent. The injectable method has won steadi- ly increasing popularity in many coun- tries. The outstanding cases are Indone- sia (a rapid rise to 28% of couples using it in the 2000/03 survey, or nearly half of all contraceptive users) and South Africa (23% in 1998). In five Latin American countries use has risen rapidly and 10% to 15% of couples now use it. Increases are also notable in some sub-Saharan Af- rican countries, especially Botswana (10%), Kenya (16%), Malawi (16%), and Swaziland (12%). The condom sees relatively little use by married couples nearly everywhere. The only recent surveys of married couples reporting more than 10% using it are in the East Asia cases of Hong Kong, Sin- gapore, South Korea, and Taiwan. Oth- ers include Costa Rica (16%), Jamaica (17%), Trinidad and Tobago (12%), Mauritius (13%), Botswana (11%), and Ukraine (11%). Other countries are around the 5% level of use, or close to zero. (One qualification is that informa- tion on methods comes chiefly from fe- male respondents, who may underreport condom use.) Also, special programs for HIV/AIDS have raised condom use in some countries. Traditional methods of withdrawal and rhythm are still very important. They account for a substantial share of all use in many countries. Appendix Table A.1 provides figures for the percentage of couples relying on traditional methods. In summary, most countries are quite se- lective in their use, or non-use, of the seven principal modern methods. Most use only two or three to any appreciable degree. Sterilization and the pill have emerged as the favorites, but with some irregularity, since most Muslim coun- tries shy away from sterilization (in fa- vor of the IUD) and certain large coun- tries make little use of the pill. The IUD is prominent in selected countries both in the Middle East and elsewhere; the injectable in fewer. Condoms are used least (although some HIV/AIDS cam- paigns are increasing its use). Plateaus in Contraceptive Prevalence and Total Fertility Rates Contraceptive prevalence, as shown in Figure 2.1, has shown impressive in- creases in many countries of the devel- oping world, so much so that pauses in the upward paths have seemed strange. They have also caused considerable concern by the agencies that implement population programs and the donors that provide assistance. Therefore the ques- tion has repeatedly been raised as to how common plateaus have been. A search for plateaus was conducted for the 52 developing countries that have had three or more surveys each, since a Chapter 2 7 plateau can be evident only when a rise appears between two surveys that fails to continue to the next survey. The flat period that ensues, if flat “enough,” con- stitutes a plateau. Clearly, most intervals between surveys show prevalence on the rise (Figure 2.3). In a few cases, at the lower end of the figure, the change has been negative or near zero. Some of those cases how- ever occurred in countries where preva- lence has always been very low and a clear upward trend has not yet taken form. In the same way, prevalence at high levels naturally levels off, since it cannot rise forever. We therefore focus upon those countries where prevalence has reached at least 25% and is still be- low 60%. The rule employed is that the prevalence rise has to be at or below 0.3 points per year (e.g., from 30 to 30.3) to qualify as a plateau. Alternative rules were also applied in the original analy- sis (Ross, Abel, and Abel, 2004). The results appear in Table 2.1. Only ten plateaus were found out of a possi- ble 159, or about 6%. Thus they are few, and they are not concentrated in any one region. The general conclusion, even if the rule were made more inclusive, is that plateaus have been quite uncom- mon in the middle range of prevalence, between 25% and 60%. Table 2.1. Number of Plateaus by Region Possible No. Actual Percent East and SE Asia 32 0 - South Asia 21 2 9.5 Latin America 49 1 2.0 Middle East/No. Africa 24 3 12.5 Sub-Saharan Africa 33 4 12.1 All Regions 159 10 6.3 Other parts of the analysis produced im- portant features: Brevity: nearly all plateaus terminate after only one survey interval. Only five countries had plateaus that lasted for two consecutive intervals. Non-repetition: plateaus rarely repeat. Only five countries experienced two plateaus at different times. Balance between modern and tradition- al methods: in some plateaus, use of modern methods still rose but use of tra- ditional methods fell, leaving overall use flat. In other cases traditional use rose but modern use fell. Thus substitu- tion of one type of use for the other can occur within an overall failure to rise. Causes of plateaus: multiple factors can be at work when prevalence levels off: ➤ Program weakening: the public pro- gram suffers losses of funding or leader- ship or serious interruption of supply lines. ➤ Narrow method mix: the sub-group of users who want (or can tolerate) the few methods available becomes saturated. ➤ Lack of long-term methods: if preva- lence is based just on resupply methods, with their poor continuation rates, it sel- dom rises to a high level. Most countries with high prevalence have considerable use of either sterilization or the IUD or both. ➤ Competition from HIV programs in some African countries may be a factor, if they attract away personnel and re- sources from family planning activities. Also, high levels of HIV can decimate the capacity of health infrastructures. ➤ A very strong program, as in Indone- sia historically, can saturate demand, leaving little unmet need. ➤ Measurement error can be at work, es- pecially when two surveys come close together and the real prevalence change between them is relatively small. Why Don’t Plateaus Repeat? One like- ly reason for the non-repetition of pla- teaus is the dismay they create among managers and donors. Corrective ac- tions have been taken in the historical experience of Bangladesh and Egypt for example, when a national survey showed an interruption of the upward course of contraceptive use. This is not surprising given the substantial commit- ments made in policy support and re- sources. A further possibility for the persistent rise in prevalence in so many countries over so many surveys is the combined influence of program improvements, private sector motivation, and general modernization including better educa- tion, more females in the labor force, ur- banization, and ideational change. Interesting Cases: considerable interest has been devoted to the stalling of prev- alence found in the 2003 Kenyan survey. This is under an intensive analysis that testifies to the concern occasioned by a failure to improve in every survey. Indo- nesia and Jordan are unusual in having a series of annual surveys, and both have 1 5 2 9 21 27 31 29 17 8 4 1 4 0 5 10 15 20 25 30 35 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 No. in each interval (inclusive up to each mark) Fr eq ue nc y Figure 2.3. Annual Pace of Prevalence Increase (points per year for 159 survey intervals) 8 Chapter 2 showed flat trends in recent years. How- ever the latest (2003) DHS survey in In- donesia showed a rise from the 1997 DHS survey (Figure 2.4). Program Responses: these take numer- ous forms and are usually selective. A first step is analytic, to determine whether the plateau occurred in every province, in both rural and urban areas, in each education group, etc. If the lack of change is common to all sub-groups that is one thing, but if for example it occurs only in the rural sector or the least favored provinces, that suggests other responses. Program strengthening can include a change in provincial lead- ership, an improvement in the method mix, stimulation of the private sector through more lenient regulations, or other measures. In summary, plateaus are uncommon; they seldom repeat; there is seldom more than one in a country; and they of- ten stimulate corrective actions. The overriding tendency is that once preva- lence reaches the range of about 25% it continues upward. At very low levels and very high levels the patterns are flat, but for quite different reasons. Use by Source by Method The available cross-national informa- tion on the sources of contraceptive sup- ply/services appears in Table 2.2 and Appendix Table A.2. Cautions are in order regarding data on both levels and trends for sources. The four categories used in the tables are the only workable ones across multiple sur- veys and time periods. Definitions of public and private have varied, some- times even in successive surveys in the same country. Also, the boundaries be- tween government and private are some- times unclear to the respondent, often justifiably so where they are truly mixed, as in some social marketing out- lets. By far most contraceptive users in the developing world rely upon the govern- ment for their supplies or services. Con- sidering only large countries that con- tain a substantial body of users, public (largely government) supply dominates in China, India, Bangladesh, Pakistan, Thailand, Viet Nam, Mexico, Kenya, Tanzania, and the Philippines. By merg- ing all modern methods, Table 2.2 shows the overall reliance upon govern- ment sources in many countries. Appen- dix Table A.2 gives the full detail for source by method, for countries with past surveys. Since the most popular method is steril- ization or the pill, it follows that govern- ment is heavily involved. Most steriliza- tion users reside in China and India, as well as Mexico and a few other countries of substantial size where government generally dominates services. Unlike sterilization sources, pill sources are mixed. When the pill became inex- pensive due to mass production in the late 1960s, governments began to add it to their method mix, and soon afterward private companies extended its use in countries where ministries of health per- mitted its sale under prescription or where informal practices flourished. Government is responsible for most pill supply in the Philippines, Thailand, and Tanzania. However, the private sector is the major source in Brazil, Colombia, Mexico, Egypt, Turkey, and Indonesia. In Kenya, public and private sources have equal shares of pill supply. The IUD is next in total use, after steril- ization and the pill. China has by far the largest body of IUD users, over 70 mil- lion or over three-fifths of all IUD users in the developing world. Government also provides most IUDs in Indonesia, Pakistan, Philippines, Thailand, Viet Nam, Kenya, Tanzania, Morocco, Tur- key, Mexico, and a number of smaller countries. (Appendix Table A.6 projects users by method for each country.) The various education and residence groups obtain their methods from some- what different sources, especially since commercial outlets and private medical personnel exist chiefly in the urban sec- tor; also, the education level is higher there. Age also matters, since older and higher parity women tend toward the longer, automatic-continuation methods of the IUD and sterilization more than younger women do. A summary of these differentials appears in Curtis and Neit- zel (1996). Trends in the Source Mix. A leading question is whether a shift in contracep- tive supply is occurring in favor of the private sector. That would relieve some of the burden on government and inter- national donors, especially as popula- tions grow and prevalence of use rises. However, in surveys to date there is only the most limited evidence that such a shift is occurring. Among 28 countries with time trend in- formation, the percentage of use due to the government (public) rose in 13 and Figure 2.4. Indonesia: Percent of Couples Using Contraception, by Type of Survey 26.8 38.5 47.8 49.7 54.7 57.4 60.3 54.2 55.5 55.4 55.4 54.8 52.5 54.2 0 10 20 30 40 50 60 70 19 80 19 81 19 82 19 83 19 84 19 85 19 86 19 87 19 88 19 89 19 90 19 91 19 92 19 93 19 94 19 95 19 96 19 97 19 98 19 99 20 00 20 01 20 02 20 03 Pe rc e n t U si ng DHS SERIES SUSENAS SERIES Chapter 2 9 fell in 5, with 10 changing very little. Definite changes occurred in Indonesia, where the public percentage fell over time from 76% to 43% to 30%. In Ken- ya the public share fell from 68% to 58% to 46%. In Peru however, the per- centage rose from 48% to 70% to 79%. Cost burdens on governments and do- nors are relieved when the private sector grows, with greater roles for pharma- cies, shops, and private medical person- nel and facilities. This may occur when a method such as the injectable gains popularity and is available through pharmacies or private midwives, as in Indonesia. It can occur when a vigorous private agency establishes a full service network, as in Colombia. It can also oc- cur when a government deliberately en- livens the private sector, though that is rare, occurring mainly in Indonesia his- torically. References Cohen, Barney. “The Emerging Fertility Transition in Sub-Saharan Africa.” World Development 26(8):1431-61. Au- gust 1998. Curtis, Sian L. and Katherine Neitzel. Contraceptive Knowledge, Use, and Sources. DHS Comparative Studies No. 19. Calverton, Maryland: Macro Inter- national Inc. 1996. Kirk, Dudley, and Bernard Pillet. “Fer- tility Levels, Trends, and Differentials in Sub-Saharan Africa in the 1980s and 1990s.” Studies in Family Planning 29(1):1-22. 1998. Ross, John, Edward Abel, and Katherine Abel. “Plateaus During the Rise of Con- traceptive Prevalence.” International Fam- ily Planning Perspectives 30(1):39-44. 2004. United Nations, World Population Pros- pects: the 2002 Revision, Volume 1. New York: United Nations Population Division. 2003. United Nations, “World Contraceptive Use 2003: Trends in Contraceptive Prevalence, 1990-2000.” Wall Chart. New York: United Nations Population Division. December 2003. 10 Chapter 2 Table 2.2. Sources of Supply for Modern Contraception Methods* Distribution of Modern Use By Source Private Other Country Year Public Medical Private Other Sum Asia Bangladesh 1999/2000 64.9 21.9 6.8 5.4 100 Cambodia 2000 45.5 24.0 24.6 2.5 100 India 1998/99 76.0 17.3 5.3 0.3 100 Indonesia 2003 29.6 61.0 7.2 2.2 100 Nepal 2001 79.4 7.7 7.3 4.8 100 Pakistan 2001 54.1 9.5 32.1 4.3 100 Philippines 1998 72.0 26.4 1.3 0.1 100 Sri Lanka 1987 85.3 6.7 4.0 2.0 100 Thailand 1987 83.6 7.9 7.0 1.0 100 Viet Nam 2002 74.8 23.9 0.2 1.2 100 Latin America Bolivia 1998 41.5 55.8 0.7 0.7 100 Brazil 1996 43.1 54.1 1.1 0.7 100 Colombia 2000 27.4 69.4 - 2.3 100 Dominican Republic 2002 34.8 26.3 33.1 5.7 199 Ecuador 1987 46.5 44.7 1.4 6.6 100 El Salvador 1985 88.8 3.3 6.4 1.4 100 Guatemala 1998/99 34.5 62.2 0.5 1.4 100 Haiti 2000 24.1 30.4 28.2 16.9 100 Mexico 1987 61.7 14.3 21.8 1.9 100 Nicaragua 2001 58.0 18.4 19.5 4.1 100 Paraguay 1990 18.7 18.5 58.1 4.2 100 Peru 2000 79.3 16.6 2.2 1.6 100 Trinidad and Tobago 1987 38.4 23.3 36.9 0.6 100 Middle East/North Africa Egypt 2003 41.2 56.5 0.1 2.2 100 Jordan 2002 43.3 56.5 - 0.2 100 Morocco 1995 62.6 37.1 0.2 0.2 100 Sudan 1990 58.1 13.1 26.2 2.0 100 Tunisia 1988 76.5 8.8 14.1 0.2 100 Turkey 1998 55.8 42.2 0.7 0.7 100 Yemen 1987 49.4 47.5 - 0.2 100 Sub-Saharan Africa Benin 2001 45.5 23.4 26.7 3.2 100 Botswana 1988 94.2 3.5 1.3 0.3 100 Burkina Faso 1998/99 75.3 6.3 15.7 2.6 100 Burundi 1987 86.7 1.2 9.3 2.8 100 Cameroon 1998 31.9 39.5 27.8 0.4 100 Central African Rep. 1994/95 49.3 31.7 13.2 2.5 100 Chad 1996/97 59.3 11.5 4.2 24.4 100 Côte d’Ivoire 1998/99 30.8 35.8 28.6 3.6 100 Eritrea 2002 58.6 19.6 0.0 21.8 100 Ethiopia 2000 77.5 15.5 5.8 0.3 100 Gabon 2000 26.5 50.7 17.2 4.2 100 Ghana 1998 46.7 46.1 4.7 1.4 100 Guinea 1999 49.9 21.0 21.1 4.6 100 Kenya 2003 45.8 41.0 11.7 1.6 100 Liberia 1986 31.1 53.9 13.1 1.6 100 Madagascar 1997 52.1 39.2 7.6 0.6 100 Malawi 2000 68.0 15.3 4.0 12.4 100 Mali 2001 51.8 33.6 11.1 1.4 100 Mauritania 2000/01 69.2 22.4 0.1 6.5 100 Mozambique 1997 82.7 8.5 4.6 2.0 100 Namibia 2000 77.4 19.6 2.8 0.2 100 Niger 1998 83.6 9.1 6.9 - 100 Nigeria 2003 34.2 53.4 1.7 10.8 100 Rwanda 2000 69.0 22.6 7.2 0.9 100 Senegal 1997 68.3 21.1 9.1 0.7 100 South Africa 1998 83.6 14.4 0.3 0.6 100 Tanzania 1999 67.2 21.8 10.4 0.1 100 Togo 1998 48.0 14.8 35.8 0.2 100 Uganda 2000/01 36.0 46.1 15.7 1.4 100 Zambia 2001/02 60.9 20.4 17.1 0.4 100 Zimbabwe 1999 76.7 16.5 2.6 3.8 100 Central Asia Republics Kazakhstan 1999 89.5 6.9 - 2.5 100 Kyrgyzstan 1997 96.9 0.6 1.1 0.3 100 Turkmenistan 2000 98.5 1.0 0 0.2 100 Uzbekistan 1996 98.3 0.3 0.2 0.2 100 Caucasus Armenia 2000 88.2 3.0 - 1.8 100 *Male sterilization and vaginal methods are omitted due to small numbers of users in some survey samples. Column figures may not add to 100% since “missing” and “don’t know” replies are omitted. Source: Demographic and Health Surveys. Chapter 2 11 0 10 20 30 40 50 60 70 80 90 1980 1985 1990 1995 2000 Botswana Eritrea Ghana Kenya Lesotho Malawi Namibia Nigeria South Africa Swaziland Tanzania Uganda Zambia Zimbabwe 0 10 20 30 40 50 60 70 80 90 1980 1985 1990 1995 2000 Benin Burkina Faso Burundi Cameroon CenAfrRep Chad Congo, DR Cote d'Ivoire Guinea Madagascar Mali Mauritania Mauritius Niger Rwanda Senegal Togo 0 10 20 30 40 50 60 70 80 90 1980 1985 1990 1995 2000 Costa Rica Cuba Domin. Rep. El Salvador Guatemala Haiti Honduras Jamaica Mexico Nicaragua Panama Puerto Rico Trin & Tob 0 10 20 30 40 50 60 70 80 90 1980 1985 1990 1995 2000 Brazil Colombia Ecuador Paraguay Peru 0 10 20 30 40 50 60 70 80 90 1980 1985 1990 1995 2000 Algeria Egypt Iran Iraq Jordan Kuwait Lebanon Libya Morocco Oman Saudi Arabia Sudan Syria Tunisia Turkey U.A.E. Yemen 0 10 20 30 40 50 60 70 80 90 1980 1985 1990 1995 2000 China Hong Kong Korea, Rep. Taiwan 2005 2005 2005 2005 2005 2005 Figure 2.1. Percentage Using Contraception Figure 2.1a Figure 2.1b Figure 2.1c Figure 2.1d Figure 2.1e Figure 2.1f Anglophone Africa Francophone Africa Central America and Caribbean South America East AsiaMiddle East/North Africa 12 Chapter 2 Algeria 0 5 10 15 20 25 30 35 40 45 50 1980 1985 1990 1995 2000 2005 F ster M ster Pill Inj/Implant IUD Condom Vaginals Traditional Bangladesh 0 5 10 15 20 25 1980 1985 1990 1995 2000 2005 F ster M ster Pill Inj/Implant IUD Condom Vaginals Traditional Brazil 0 5 10 15 20 25 30 35 40 45 1980 1985 1990 1995 2000 2005 F ster M ster Pill Inj/Implant IUD Condom Vaginals Traditional China 0 5 10 15 20 25 30 35 40 1980 1985 1990 1995 2000 2005 F ster M ster Pill Inj/Implant IUD Condom Vaginals Traditional 0 10 20 30 40 50 60 70 80 90 1980 1985 1990 1995 2000 Cambodia Indonesia Laos Malaysia Myanmar Philippines Singapore Thailand Viet Nam 0 10 20 30 40 50 60 70 80 90 1980 1985 1990 1995 2000 Bangladesh India Nepal Pakistan Sri Lanka 2005 2005 Figure 2.1g Figure 2.1h Figure 2.1. Percentage Using Contraception (Cont.) Figure 2.2. Time Trends for Percent of Married Women using Each Contraceptive Method Figure 2.2a Figure 2.2b Figure 2.2c Figure 2.2d Southeastern Asia Southern Asia Chapter 2 13 Colombia 0 5 10 15 20 25 30 1980 1985 1990 1995 2000 2005 F ster M ster Pill Inj/Implant IUD Condom Vaginals Traditional Egypt 0 5 10 15 20 25 30 35 40 1980 1985 1990 1995 2000 2005 F ster M ster Pill Inj/Implant IUD Condom Vaginals Traditional Ethiopia 0 1 1 2 2 3 3 4 1980 1985 1990 1995 2000 2005 F ster M ster Pill Inj/Implant IUD Condom Vaginals Traditional India 0 5 10 15 20 25 30 35 40 1980 1985 1990 1995 2000 2005 F ster M ster Pill Inj/Implant IUD Condom Vaginals Traditional Indonesia 0 5 10 15 20 25 30 35 1980 1985 1990 1995 2000 2005 F ster M ster Pill Inj/Implant IUD Condom Vaginals Traditional Jordan 0 5 10 15 20 25 1980 1985 1990 1995 2000 2005 F ster M ster Pill Inj/Implant IUD Condom Vaginals Traditional Figure 2.2e Figure 2.2f Figure 2.2g Figure 2.2h Figure 2.2i Figure 2.2j Figure 2.2. Time Trends for Percent of Married Women Using Each Contraceptive Method (Cont.) 14 Chapter 2 Kenya 0 2 4 6 8 10 12 14 16 18 1980 1985 1990 1995 2000 2005 F ster M ster Pill Inj/Implant IUD Condom Vaginals Traditional Mexico 0 5 10 15 20 25 30 35 1980 1985 1990 1995 2000 2005 F ster M ster Pill Inj/Implant IUD Condom Vaginals Traditional Myanmar 0 2 4 6 8 10 12 14 1980 1985 1990 1995 2000 2005 F ster M ster Pill Inj/Implant IUD Condom Vaginals Traditional Nigeria 0 1 2 3 4 5 6 7 8 1980 1985 1990 1995 2000 2005 F ster M ster Pill Inj/Implant IUD Condom Vaginals Traditional Pakistan 0 1 2 3 4 5 6 7 8 1980 1985 1990 1995 2000 2005 F ster M ster Pill Inj/Implant IUD Condom Vaginals Traditional Philippines 0 5 10 15 20 25 1980 1985 1990 1995 2000 2005 F ster M ster Pill Inj/Implant IUD Condom Vaginals Traditional Figure 2.2k Figure 2.2l Figure 2.2m Figure 2.2n Figure 2.2o Figure 2.2p Figure 2.2. Time Trend for Percent of Married Women Using Each Contraceptive Method (Cont.) Chapter 2 15 South Africa 0 5 10 15 20 25 30 1980 1985 1990 1995 2000 2005 F ster M ster Pill Inj/Implant IUD Condom Vaginals Traditional Tanzania 0 1 2 3 4 5 6 7 8 9 10 1980 1985 1990 1995 2000 2005 F ster M ster Pill Inj/Implant IUD Condom Vaginals Traditional Thailand 0 5 10 15 20 25 30 1980 1985 1990 1995 2000 2005 F ster M ster Pill Inj/Implant IUD Condom Vaginals Traditional 0 5 10 15 20 25 30 35 1980 1985 1990 1995 2000 2005 F ster M ster Pill Inj/Implant IUD Condom Vaginals Traditional Turkey Viet Nam 0 5 10 15 20 25 30 35 40 45 1980 1985 1990 1995 2000 2005 F ster M ster Pill Inj/Implant IUD Condom Vaginals Traditional Zimbabwe 0 5 10 15 20 25 30 35 40 1980 1985 1990 1995 2000 2005 F ster M ster Pill Inj/Implant IUD Condom Vaginals Traditional Figure 2.2q Figure 2.2r Figure 2.2s Figure 2.2u Figure 2.2t Figure 2.2. Time Trend for Percent of Married Women Using Each Contraceptive Method (Cont.) Figure 2.2v Chapter 3 17 This chapter presents projections from 2005 through 2020 for numbers of wom- en, contraceptive users, needed commod- ities, and commodity costs. The chapter is based upon the UN 2004 projections for numbers of women aged 15-49 and their total fertility rates, as well as upon a large body of national surveys that provide prevalence of use by method. The starting point is a projection of fu- ture prevalence for 116 countries, based upon the relationship of prevalence to the UN total fertility rates as shown in Figure 3.1. The projections are adjusted at the beginning to agree with the latest survey estimates or, for those lacking a survey, the regional average. Given the prevalence projections, other results fol- low as described below. Percentage using each method. To pro- duce projections for individual contra- ceptive methods, the body of past na- tional surveys was used to establish the relationship between total use and each method’s use. On average each method’s share changes through time as total use rises (Figures 3.2a and 3.2b), so the projected method mix in each country depends upon its path for total use. Two different sets of equations were used (given in the Technical Appendix) for Muslim and non-Muslim countries, since the mix in Muslim countries tends on average to contain less sterilization and more IUD use than elsewhere. The starting method mix was adjusted to match the actual mix in the most recent survey, just as total use was adjusted. Appendix Tables A.5.a-d provide the es- timates for total use and use by method for the 116 countries from 2005-2020. Also of interest is Appendix Table A.1, which contains all national surveys from 1980 onward, with sources provid- ed at the end. Numbers of users by method. To project the number of users of each method, the UN projections for numbers of women aged 15-49 were employed in combina- tion with the percent of women relying on each method. This was done for all women and also for married/cohabiting Figure 3.1. Contraceptive Prevalence and Total Fertility Rate Chapter 3Chapter 3Chapter 3Chapter 3Chapter 3 FUTURE TRENDS IN CONTRACEPTIVE USE women, using the UN proportions of married women as updated from recent surveys (proportions married are kept constant through time). Appendix Ta- bles A.6.a-d give the number of users by method for 2005-2020. (The propor- tions married appear in Appendix Table A.4.) Adjustments were made to these proce- dures to allow for a few special country cases, as when the original method mix was very unusual, and especially if total prevalence was already high and quite stable; for example in China most use for many years has been of male and fe- male sterilization and the IUD, at high levels, and in Brazil most use has been stable for female sterilization and the pill. Wherever total prevalence was al- ready above 65% the mix in the latest survey was kept constant, and this rule was applied also to India and to Indone- sia, where the mix was expected to change little, at least in the next five years. Commodities needed by method. The quantities of commodities needed were calculated for each method and each fu- ture year by reference to couple years of protection (CYP). One year of protection requires 15 pill cycles (rather than 13, to allow for wastage), 120 condoms or vaginal tablets, or 4 injectables. An IUD lasts 3.5 years on average and a male or female sterilization lasts about 9 years. (For CYP data see Stover et al., 1997, and Stover, Bertrand, and Shelton, 2000.) Quantities of commodities need- ed were obtained by these CYP values, applied to the numbers of users in each year. For the IUD and sterilization the commodities required depend upon new adoptions, which on average follow the time path for increasing users. (Method- ological details appear in the Technical Appendix.) Numbers of commodities by method for 2005-2020 are in Appendix Tables A.7.a-d. Costs for each method. Commodity costs were calculated by simply multi- plying the cost per method times its quantity in each year. The costs were in- creased by 10% to allow for internation- al transportation, but they do not in- clude costs for personnel, facilities, or any of the associated services and they do not adjust for inflation. Costs were calculated at US 24 cents per pill cycle, 96.5 cents per injectable dose, 3.5 cents per condom, 7.2 cents per vaginal ap- plication, 57.6 cents per IUD adoption, US$9.09 for female sterilization adop- tion and US$4.95 for male sterilization adoptions. (If modifications of these rules are preferred they can be easily 0 10 20 30 40 50 60 70 80 90 0 1 2 3 4 5 6 7 8 Total Fertility Rate C o n tr ac ep ti ve P re va le n ce 18 Chapter 3 applied to the numbers of users in the Appendix to generate alternative cost estimates.) Appendix Tables A.8.a-d give costs for each commodity by meth- od for 2005-2020. Summary. The various approaches above cover the essential features and yield estimates for prevalence, method mix, users, commodities, and costs for the 116 countries through time, taking into account numbers of women and proportions married. The following sec- tions discuss the results. Projections for the Percentage Using Contraception Because our projections for increases in contraceptive prevalence are governed by the declines in total fertility rates (TFRs), there is a U-shaped pattern to the prevalence projections. Countries with very high or very low TFRs are pro- jected to change least, while countries in the middle range are projected to change more rapidly. When the TFRs are translated to prevalence values the re- sults appear as shown in Table 3.1. Countries that in 2005 have prevalence below 10% improve only by 4.4 points by 2020, while countries in the middle, at 30-39%, improve by a full 15.9 points. At the highest level, 70% or above, the av- erage change is zero, with some coun- tries slightly above and some slightly below, depending upon the small chang- es in the TFR when it is near replace- ment. Next, changes in the method mix follow the average patterns in Figures 3.2a and 3.2b as noted above. In Muslim coun- tries, as total prevalence rises to about 25% (bottom axis), the share due to tra- ditional methods lessens while the shares due to the IUD and the pill gain sharply. At higher prevalence levels the pill loses ground to the IUD. The picture is quite different in the non-Muslim countries, where traditional method use is replaced by a gradual increase in fe- male sterilization, and secondarily by an increase for the IUD. The injectable plays a larger role in non-Muslim than in Muslim countries. The share of the pill is not much different between the Figure 3.2b. Method Mix – Non-Muslim Countries This model shows, on average, how method mix changes as total prevalence of use rises, as registered in past national surveys in many developing coun- tries. Traditional method use, at the top, declines as a proportion of all use, while female sterilization increases considerably. Pill use declines, while IUD use increases somewhat, as does condom use. Minor roles on average are played by male sterilization, injectables, and such vaginals as foaming tablets. Note that all these changes are relative ones, adding to 100 percent of use. Be- cause total use is increasing, up to about 80% of all couples, the number of pill users for example will be larger than suggested by the relative decline. This model is used for the projections of use by method in this report, with ad- justments for Muslim countries, whose method mix shows less sterilization and more IUD use, etc., and for certain other countries. See Appendix B, Technical Appendix for Projection Methods, for details. Figure 3.2a. Method Mix – Muslim Countries Pill Injection IUD Condom Female Sterilization Traditional Method 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 5 10 15 20 25 30 35 40 45 50 55 60 65 70 Contraceptive Prevalence Male Sterilization Vaginal Application Pill Injection IUD Condom Female Sterilization Traditional Method 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 5 10 15 20 25 30 35 40 45 50 55 60 65 70 Contraceptive Prevalence Male Sterilization Vaginal Application Pill Injection IUD Condom Female Sterilization Traditional Method 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 5 10 15 20 25 30 35 40 45 50 55 60 65 70 Contraceptive Prevalence Male Sterilization Vaginal Application Pill Injection IUD Condom Female Sterilization Traditional Method 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 5 10 15 20 25 30 35 40 45 50 55 60 65 70 Contraceptive Prevalence Male Sterilization Vaginal Application Pill Injection IUD Condom Female Sterilization Traditional Method 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 5 10 15 20 25 30 35 40 45 50 55 60 65 70 Contraceptive Prevalence Male Sterilization Vaginal Application Pill Injection IUD Condom Female Sterilization Traditional Method 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 5 10 15 20 25 30 35 40 45 50 55 60 65 70 Contraceptive Prevalence Male Sterilization Vaginal Application Chapter 3 19 two; the chief difference is in the strong preference for the IUD over sterilization in the Muslim group at the higher TFR levels. Projections for Total Numbers Using Contraception The numbers of contraceptive users will increase very substantially in the future due to population growth. As seen in Table 3.2, sizeable growth will occur in the numbers of married women and all women. (Because the percents married are kept constant the rates are nearly the same; minor differences occur be- cause country weights change as some countries grow more slowly than oth- ers.) Within five years 100 women will be replaced by 107 women, and by 115 women within 15 years. This is highly variable by country; China’s very slow growth affects the averages, which are higher elsewhere and are especially high in sub-Saharan Africa. The same overall growth will occur for married users, and it will be substantial- ly more for all users, to 119 users for ev- ery 100 now. Growth for users is much higher in many countries due to the dou- ble force of population growth and in- creases in the proportions using a meth- od. The figures also vary by region: they are much higher for sub-Saharan Africa and are higher in general when China is omitted. All this creates similar increases in resources needed, including supplies, facilities, training, and service arrange- ments. Even in the next five years, an increase of 9.1% must be absorbed. The sharp re- gional differences are apparent in Figure 3.3a, from China’s 2% rise to sub-Sahar- an Africa’s 30%. Note however that the predicted rise in contraceptive use in these projections is approximately tied to the UN projections for declines in the total fertility rates, and those are proba- bly optimistic in the case of sub-Saharan Africa. If fertility falls less than expect- ed, the associated rise in contraceptive use will be less. However, it is best for donors and governments to plan for more users rather than fewer, as a hedge against the usual shortages and interrup- tions in supply lines and services. Paralleling Figure 3.3a is Figure 3.3b, to show the regional pattern for the 15-year increases. It is very similar in appearance to Figure 3.4 except that sub-Saharan Af- rica’s increase is now more than double the next highest region, reflecting the long-range effects of the very young age structure in most countries there. In Chi- Table 3.2. Percentage Increases in Numbers of Married Women, All Women, Married Users, and All Users No. of No. of All No of No. of Married Women Women Married All Aged 15-49 Aged 15-49 Users Users 2005 - 2010 6.6 6.7 7.1 9.1 2005 - 2015 11.6 11.8 13.2 16.3 2005 - 2020 15.3 15.6 17.8 22.0 Figure 3.3a. Percent Increase Over the Next Five Years in Number of Contraceptive Users, 2005-2010, by Region Figure 3.3b. Percent Increase over the Next Fifteen Years in Number of Contraceptive Users, 2005-2020, by Region Table 3.1. Projected Increases in Prevalence According to Starting Level Starting Level Gain from (in 2005) 2005 2010 2015 2020 2005 to 2020 0-9% 7.3 8.0 9.5 11.7 4.4 10-19% 14.9 16.7 19.6 23.5 8.6 20-29% 25.4 28.5 32.8 37.6 12.3 30-39% 33.9 38.1 43.8 49.7 15.9 40-49% 44.8 48.5 52.6 56.3 11.5 50-59% 56.6 59.3 62.4 65.2 8.6 60-69% 64.8 66.3 67.6 68.4 3.6 70+ 76.5 76.7 76.7 76.5 0.0 1.7 15.6 12.1 9.2 16.6 30.1 0.7 9.1 0 5 10 15 20 25 30 35 China India Rest of Asia Latin America Middle East/North Africa Sub- Saharan Africa Central Asia Republics Caucasus Grand Total Pe rc en t I nc re as e 10.9 (6.8) 42.4 29.7 21.6 49.2 119.8 25.5 (8.2) 22.0 -20 0 20 40 60 80 100 120 140 China India Rest of Asia Latin America Middle East/North Africa Sub- Saharan Africa Central Asia Republics Caucasus Grand Total Pe rc en t I nc re as e 20 Chapter 3 na and in the Caucasus small declines occur, again due to the changing age structure. Turning to absolute numbers to be served, Figure 3.4 shows the changes. China again, although having the larg- est numbers, stabilizes at about 200 mil- lion users. India is slated to experience a drastic rise, nearly reaching China’s lev- el, due to the combination of population growth and a rise in the proportion using contraception. The rest of Asia will also face rapid growth, due not only to rising prevalence of use but also to the young age structures in Pakistan, Bangladesh, Indonesia, and elsewhere. Latin America and the Middle East/North Africa can expect milder increases, unlike sub-Sa- haran Africa where rapid growth is pro- jected. By method, users in 2005 are distribut- ed for each region in Table 3.3. Some 618 million users are estimated, 72% of them in Asia (see also the top three lines in Figure 3.5). Male and female steril- ization dominate in Asia, where they ac- count for 50% of users, and sterilization accounts for 45% of users in Latin Amer- ica. The IUD is next at over one-fifth of all users due largely to its extensive use in China. In percentage terms its share of use is exceptionally high in the Middle East/North Africa at one-third and in the former Soviet Union areas at above one- fifth. After the IUD comes the pill at 12% of all users. In general, only two methods account for most use within each region, al- though the two methods are not always the same: sterilization and the IUD in Asia; sterilization and the pill in Latin America; the IUD and pill in the Middle East/North Africa (and the IUD alone in the Central Asia Republics). In sub-Sa- haran Africa over 40% of use is due to the pill and injectable; 45% is due to the pill and IUD in the Caucasus. The overall pattern is that two modern methods, out of the seven candidates, emerge in each region as dominant. The eighth choice, traditional methods, con- tinues to be very important in both sub- Saharan Africa and the Middle East/North Africa, at 25%-30% of use respectively, and also in the former Soviet Union. User increases overall are projected at 9% by 2010 and by 16% by 2015, but higher for the pill (21%). Lesser increases occur for other methods, except for vaginals (small base in 2005) and traditionals. For absolute numbers the main annual in- creases are for female sterilization, 50 million more in 2020 than in 2005. Sterilization Any Region Total Female Male Pill Injectable IUD Condom Vaginals Traditional Number of Users (000), 2005 Asia 444,695 191,181 33,084 39,358 19,856 107,028 21,395 220 32,572 Latin America 72,639 30,624 1,735 14,600 3,612 7,946 4,844 265 9,013 Middle East/North Africa 30,406 2,266 45 7,287 1,223 10,000 1,882 272 7,430 Sub-Saharan Africa 35,498 3,953 260 7,724 7,984 2,347 2,431 250 10,550 Central Asia Republics 6,723 363 10 656 221 4,042 373 30 1,027 Caucasus 1,524 279 15 357 116 328 85 12 331 Moldova, Russia, Ukraine 26,761 6,488 826 3,105 690 5,871 3,920 210 5,651 Grand Total 618,246 235,154 35,976 73,087 33,703 137,562 34,930 1,259 66,575 Percent Distribution Within Each Region, 2005 Asia 100 43.0 7.4 8.9 4.5 24.1 4.8 0.0 7.3 Latin America 100 42.2 2.4 20.1 5.0 10.9 6.7 0.4 12.4 Middle East/North Africa 100 7.5 0.1 24.0 4.0 32.9 6.2 0.9 24.4 Sub-Saharan Africa 100 11.1 0.7 21.8 22.5 6.6 6.8 0.7 29.7 Central Asia Republics 100 5.4 0.2 9.8 3.3 60.1 5.5 0.4 15.3 Caucasus 100 18.3 1.0 23.4 7.6 21.5 5.6 0.8 21.7 Moldova, Russia, Ukraine 100 24.2 3.1 11.6 2.6 21.9 14.6 0.8 21.1 Grand Total 100 38.0 5.8 11.8 5.5 22.3 5.6 0.2 10.8 Total Numbers by Date (000), 2005-2020 2005 618,246 235,154 35,976 73,087 33,703 137,562 34,930 1,259 66,575 2010 674,595 256,401 37,456 81,408 36,788 146,061 38,274 1,537 76,672 2015 719,142 273,040 37,904 88,758 39,679 151,018 41,143 1,804 85,798 2020 754,284 285,353 37,246 95,575 42,107 153,319 44,013 2,078 94,593 Percent Growth in Users by Period, 2005-2020 2005 to 2010 9.1 9.0 4.1 11.4 9.2 6.2 9.6 22.0 15.2 2005 to 2015 16.3 16.1 5.4 21.4 17.7 9.8 17.8 43.3 28.9 2005 to 2020 22.0 21.3 3.5 30.8 24.9 11.5 26.0 65.1 42.1 Table 3.3. Projected Numbers (000s) and Percent of Contraceptive Users Among All Women, by Method, by Region for 2005, and for Four Dates 2005-2020 Chapter 3 21 Projections for Commodities Needed, by Method The user numbers above translate direct- ly into commodity needs by the rules explained above, based on couple years of protection. For example, the number of pill cycles needed in any country in 2005 is simply 15 times the number of users in that year. Note that the steriliza- tion figures are simply estimates for the number of procedures done annually, as a basis for country calculations (not in- cluded) of the kits and other supplies needed. Appendix Table A.5 projects the needs for each method by country and year, with regional totals. Here we sim- ply show the overall rise in total com- modities needed for pills, condoms, IUDs, and injectables for the developing world as a whole (Figure 3.4). Note that the condom figures are understated since they omit the need for disease pre- vention. Chapter 7 covers additional condom requirements. Percent Growth Region 2005 2010 2015 2020 2005 - 2010 2005 - 2015 2005-2020 China 190,332 193,657 189,661 177,471 1.7 (0.4) (6.8) India 145,278 167,896 188,581 206,841 15.6 29.8 42.4 Rest of Asia 257,291 282,694 302,364 317,716 9.9 17.5 23.5 Latin America 134,559 146,785 155,665 163,140 9.1 15.7 21.2 Middle East/North Africa 49,685 56,832 63,212 69,328 14.4 27.2 39.5 Sub-Saharan Africa 83,065 103,543 130,669 163,725 24.7 57.3 97.1 Central Asia Republics 6,688 9,004 10,869 12,060 34.6 62.5 80.3 Caucasus 2,793 2,807 2,671 2,556 0.5 (4.3) (8.5) Moldova, Russia, Ukraine 44,103 40,753 37,109 34,835 (7.6) (15.9) (21.0) Total 913,793 1,003,971 1,080,801 1,147,672 9.9 18.3 25.6 Table 3.4. Total Commodity Costs by Region and Year (thousands of U.S. dollars) Figure 3.4. Users of All Contraceptive Methods, 2005-2020, by Region Finding ways to cope with these large increases in users and commodities will occupy donors and governments for the foreseeable future. The role of the pri- vate sector will be quite important as a way of relieving these burdens. Howev- er, more creative ways are needed than those used heretofore if private sectors in many countries are to significantly enlarge their contribution. Projections for Commodity Costs As explained above, costs are calculated only for the purchase of commodities, with an addition of 10% to allow for in- ternational transportation costs. Table 3.4 presents total costs (for all methods) by region and year. China and India mainly cover their own commodities, and the largest costs occur in the Rest of Asia. Latin American comes next with its relatively high prevalence of use, and then sub-Saharan Africa, where preva- lence and the number of users is smaller. However sub-Saharan Africa will experi- ence an exceptionally fast growth of costs, rising by one-fourth by 2010, by over one-half by 2015, and almost dou- bling by 2020. Growth is rapid in India and in the Middle East/North Africa. It is also rapid in the Central Asian Repub- lics, reflecting the expected shift to modern contraceptives there. Costs should decline in China since total prev- alence is already high and flat, and the age structure will change toward fewer women aged 15-49. This is also expect- ed in Moldova, Russia, and Ukraine. For the developing world as a whole costs are projected to increase by 10% by 2010, 18% by 2015, and 26% by 2020. References Abou-Zahr, Carla and Tessa Wardlaw, “Maternal Mortality in 2000: Esti- mates Developed by WHO, UNICEF, and UNFPA.” http://www.childinfo.org/ maternal mortality in 2000.pdf. Henshaw, Stanley K., Susheela Singh, and Taylor Haas. “The Incidence of Abortion Worldwide.” International Family Planning Perspectives. Vol. 25, Supplement. Pages S30-S37. 1999. Stover, John et al. Empirically Based Conversion Factors for Calculating Couple-Years of Protection. The EVAL- UATION Project. Carolina Population Center, Tulane University, and The Fu- tures Group International. 1997. Pub- lished also in the Evaluation Review, Vol. 24, No. 1, pp. 3-46, Feb. 2000. Sage Foundation. - 50,000 100,000 150,000 200,000 250,000 2005 2010 2015 2020 China India Rest of Asia Latin America Middle East/North Africa Sub-Saharan Africa Central Asia Republics N um be r o f U se rs (0 00 ) 22 Chapter 3 UNAIDS. Report On the Global AIDS Epidemic: 4th Global Report. Geneva: Joint United Nations Programme on HIV/AIDS. 2004. United Nations Population Division. World Population Prospects: The 2002 Revision. Volume I: Comprehensive Ta- bles. New York: United Nations. 2003. WHO Division of Reproductive Health. Unsafe Abortion: Global and Regional Estimates of the Incidence of Unsafe Abortion and Associated Mortality in 2000. Fourth Edition. Geneva: WHO. 2004. Figure 3.5. Projection of Annual Commodity Needs, Four Methods, 2005-2020, Developing World Total Number of Pill Cycles Needed (000) Total Number of Condoms Needed (000) for Family Planning (see Chapter 7 for information on Condom needs for HIV/AIDS) Total Number of IUDs Needed (000) Total Number of Injectables Needed (000) 1,096,306 1,221,123 1,331,368 1,433,625 0 200,000 400,000 600,000 800,000 1,000,000 1,200,000 1,400,000 1,600,000 2005 2010 2015 2020 4,191,594 4,592,916 4,937,106 5,281,562 2005 2010 2015 2020 39,303 41,732 43,148 43,806 2005 2010 2015 2020 134,811 147,151 158,716 168,429 0 20,000 40,000 60,000 80,000 100,000 120,000 140,000 160,000 180,000 2005 2010 2015 2020 0 1,000,000 2,000,000 3,000,000 4,000,000 5,000,000 6,000,000 37,000 38,000 39,000 40,000 41,000 42,000 43,000 44,000 45,000 Chapter 4 23 This chapter concerns demands on ser- vices, in both public and private sectors, that are implicit in increases in the sheer numbers of women, married women, and deliveries – depending however upon the region. In a few instances the population numbers will actually decline, but growth overall will be substantial in ev- ery five-year period. Births are projected to grow mainly in sub-Saharan Africa, not overall. Special attention goes to the growing numbers of youth, and contra- ceptive use among both married and sin- gle women. Unmet needs among youth are presented both for attendance at birth and for contraception services. Finally, HIV/AIDS among youth is briefly pre- sented. Growing Numbers of Women, Married Women, and Deliveries While the number of women, and mar- ried women, will certainly rise substan- tially in the developing world, the num- ber of births will not, according to the UN’s projections (Appendix Tables A.3- A.4). However this differs by region: Figure 4.1 (on the next page) shows the trends for China, India, and the major regions. Note that the numbers of women to ex- pect in the next 15 years are already born so the projections are fairly reli- able; also the proportions married through time are held constant, so the numbers married mirror those for all women but at lower levels. China’s age structure is such that a decline is expect- ed in numbers of women; the number of births also falls. India is different: large increases are projected for numbers of women, but the UN anticipates enough of a fall in the fertility rate to cause an actual decline in the number of births. In the rest of Asia, large increases are ex- pected for the population of women but births decline slightly. The other regions vary: in sub-Saharan Africa both women and births increase very sharply. In Latin America and the Middle East/North Africa, women be- come more numerous but births do not. The other three regions, all parts of the former USSR, are variable: more women are projected only in the Central Asia Republics, and the least declines in births in the Caucasus. The percentage changes for the regions are given in Table 4.1. Apart from Chi- na and the former USSR areas, all devel- oping areas experience substantial growth in each five-year period, on an ever-growing base. The picture for births is quite different, as explained above. Table 4.1. Percent Increases for Women, Married Women, and Births, by Region, 2005-2020 Percent Increases in Numbers of Women Aged 15-49 2005 to 2010 2010 to 2015 2015 to 2020 2005 to 2020 China 1.7 (2.0) (5.7) (6.0) India 8.9 7.2 5.0 22.5 Rest of Asia 8.9 6.8 5.7 22.9 Latin America 6.3 4.5 3.5 15.0 Middle East/North Africa 10.7 8.8 8.3 30.5 Sub-Saharan Africa 12.4 12.6 12.8 42.7 Central Asia Republics 6.2 3.1 3.3 13.1 Caucasus 0.6 (4.5) (3.2) (6.9) Moldova, Russia, Ukraine (6.7) (8.1) (6.2) (19.6) ALL REGIONS 6.7 4.8 3.5 15.6 Percent Increases in Numbers of Married Women Aged 15-49 China 1.7 (2.0) (5.7) (6.0) India 8.9 7.2 5.0 22.5 Rest of Asia 9.1 7.1 5.9 23.7 Latin America 6.3 4.5 3.5 15.0 Middle East/North Africa 10.7 8.9 8.3 30.5 Sub-Saharan Africa 12.9 13.1 13.3 44.7 Central Asia Republics 6.4 3.3 3.4 13.6 Caucasus 0.6 (4.5) (3.2) (7.0) Moldova, Russia, Ukraine (6.7) (8.1) (6.2) (19.6) ALL REGIONS 6.6 4.7 3.3 15.3 Percent Increases in Numbers of Births China 0.1 (1.5) (5.2) (6.5) India (1.2) (1.5) (3.1) (5.7) Rest of Asia 1.4 (0.1) (1.9) (0.6) Latin America (1.8) (2.3) (2.8) (6.8) Middle East/North Africa 2.5 (0.1) (0.7) 1.8 Sub-Saharan Africa 5.7 4.0 2.8 13.0 Central Asia Republics 0.0 (3.3) (6.1) (9.2) Caucasus 2.1 0.4 (5.3) (2.8) Moldova, Russia, Ukraine (0.2) (4.8) (9.1) (13.6) ALL REGIONS 1.4 0.1 (1.6) (0.1) Births for 22 large countries are high- lighted in Table 4.2. This list, in order by number of births in 2005, echoes the large role being played by India in all demographic matters. It has more births than China, and more than the next five countries together, or alternately, more than the bottom 14. Ten of these countries are projected by the UN to experience birth declines over the next 5 years. Posed against those are the large percentage of increases coming in Afghanistan (13%), D.R. Congo (9%), DEMANDS ON SERVICES Chapter 4Chapter 4Chapter 4Chapter 4Chapter 4 24 Chapter 4 Table 4.2. Number of Births Annually and Percentage Change for 22 Large Countries, 2005-2020 Number of Births (000s) Percentage Change 2005 2010 2015 2020 2005-2010 2010-2015 2015-2020 2005-2020 India 25,027 24,733 24,355 23,607 (1.2) (1.5) (3.1) (6) China 18,795 18,824 18,534 17,565 0.1 (1.5) (5.2) (6.5) Pakistan 5,672 6,022 6,189 6,164 6.2 2.8 (0.4) 8.7 Nigeria 4,909 5,088 5,136 5,093 3.6 0.9 (0.8) 3.7 Indonesia 4,479 4,380 4,242 4,074 (2.2) (3.1) (4.0) (9.1) Bangladesh 4,187 4,151 4,094 4,020 (0.8) (1.4) (1.8) (4.0) Brazil 3,438 3,324 3,188 3,048 (3.3) (4.1) (4.4) (11.4) Ethiopia 3,089 3,324 3,531 3,657 7.6 6.2 3.6 18.4 Congo DR 2,776 3,038 3,253 3,447 9.4 7.1 6.0 24.2 Mexico 2,265 2,189 2,100 2,000 (3.4) (4.1) (4.8) (11.7) Philippines 1,995 1,979 1,965 1,934 (0.8) (0.7) (1.6) (3.1) Egypt 1,967 2,020 1,960 1,885 2.7 (3.0) (3.8) (4.2) Viet Nam 1,646 1,682 1,684 1,611 2.1 0.2 (4.3) (2.1) Iran 1,465 1,569 1,553 1,408 7.1 (1.0) (9.4) (3.9) Turkey 1,459 1,404 1,325 1,274 (3.8) (5.6) (3.9) (12.7) Tanzania 1,456 1,481 1,476 1,441 1.7 (0.4) (2.3) (1.0) Uganda 1,393 1,589 1,794 2,002 14.1 12.9 11.6 43.8 Russia 1,237 1,241 1,182 1,073 0.3 (4.8) (9.2) (13.3) Afghanistan 1,217 1,378 1,486 1,559 13.2 7.8 4.9 28.1 Myanmar 1,147 1,101 1,065 1,025 (4.0) (3.3) (3.8) (10.7) Sudan 1,095 1,081 1,066 1,062 (1.2) (1.4) (0.4) (3.0) Thailand 1,061 1,023 993 970 (3.6) (2.9) (2.4) (8.6) Figure 4.1a. Numbers of Women Aged 15-49, by Region, 2005-2020 (000) Figure 4.1c. Numbers of Births Annually, by Region, 2005-2020 (000) Figure 4.1b. Numbers of Married Women Aged 15-49, by Region, 2005-2020 (000) 0 50,000 100,000 150,000 200,000 250,000 300,000 350,000 400,000 2005 2010 2015 2020 3 1 2 6 4 5 9 7 8 (1) China (2) India (3) Rest of Asia (4) Latin America (5) Middle East/North Africa (6) Sub-Saharan Africa (7) Central Asia Republics (8) Caucasus (9) Moldova, Russia, Ukraine Key: 50,000 100,000 150,000 200,000 250,000 300,000 350,000 400,000 2005 2010 2015 2020 0 3 1 2 6 4 5 9 7 8 0 5,000 10,000 15,000 20,000 25,000 30,000 35,000 2005 2010 2015 2020 3 1 2 6 4 5 9 7 8 Chapter 4 25 Ethiopia (8%), Iran (7%), Pakistan (6%), and Uganda (14%). This points to the need for plans that are specific to each country, notwithstanding the overall pic- ture of stability in birth numbers. Moreover, nearly every country needs substantial improvement in coverage and quality of services, so any relief from rising numbers of births does not allow for any relaxation of effort. Quite the contrary, especially since the picture given here depends heavily upon future fertility trends that may differ from those assumed. Youth: Current Needs and Services About one billion youth, aged 15-24, in- habit the developing world. They make up a fifth (19%) of the total population and will become a considerably larger share over the next decade in many countries. Selected highlights follow from UN projections and Appendix Ta- ble A.24, with information also on un- met need and HIV/AIDS. Distributions: China and India each has one-fifth of all youth (Table 4.3, column 1) and the rest of Asia has over a fifth, for 70% of the grand total. Africa con- tains 19% and Latin America 11%. There is very little variation across re- gions in the balance between youth 15- 19 and youth 20-24; it is essentially half and half (52%/48%). The share of population due to youth does not vary much around the overall 19% figure, but it is as low as 14%-16% in Cuba, Uruguay, the Republic of Ko- rea, Taiwan, and Ukraine. On the other hand, the percents run higher in sub-Sa- haran Africa, to 24%-26% in Botswana, Lesotho, Rwanda, and Swaziland, where fertility rates are higher. Time Trends: Overall, the developing world will see a 5% increase of youth by 2010, with rather little additional in- crease to 2020 (Table 4.3). However the picture is very different for the UN list of “least developed” countries (second row): growth is a full 12% by 2010, 24% by 2015, and 35% by 2020. That means that in the next 15 years every 100 mem- bers of this age group will be replaced by 135. This is reflected also in the row for Africa, which contains many of the least developed countries, with 30% growth by 2020. At the other extreme, China’s youth pop- ulation will decline, becoming 18% smaller by 2020 than it is now, a total re- duction of over 40 million persons be- tween 2005 and 2020 (bottom panel of table). Over the long run that accelerates the trend toward an aging population: it shrinks the numbers at the bottom of the Table 4.3. Time Trends from 2005 to 2020 for the Population Aged 15-24 Numbers Aged 15-24 (000) Percent Growth from 2005 Areas 2005 2010 2015 2020 2005-2010 2005-2015 2005-2020 Less developed 993,515 1,041,923 1,045,330 1,049,228 4.9 5.2 5.6 Least developed 154,844 174,118 191,802 209,319 12.4 23.9 35.2 Other less developed 838,671 867,806 853,528 839,910 3.5 1.8 0.1 Less developed ex. China 775,172 822,401 848,509 870,683 6.1 9.5 12.3 Africa 188,597 207,688 224,987 245,412 10.1 19.3 30.1 Asia 711,633 737,388 721,711 704,005 3.6 1.4 (1.1) China 217,349 218,593 195,952 177,735 0.6 (9.8) (18.2) India 211,254 224,657 231,221 232,353 6.3 9.5 10.0 Rest of Asia 283,030 294,137 294,539 293,916 3.9 4.1 3.8 Latin America 105,665 107,543 108,637 109,625 1.8 2.8 3.7 Added Numbers (000) Areas 2005-2010 2010-2015 2015-2020 Less developed na 48,408 3,407 3,898 Least developed na 19,274 17,684 17,517 Other less developed na 29,135 (14,278) (13,618) Less developed ex. China na 47,229 26,108 22,174 Africa na 19,091 17,299 20,425 Asia na 25,755 (15,677) (17,706) China na 1,244 (22,641) (18,217) India na 13,403 6,564 1,132 Rest of Asia na 11,107 402 (623) Latin America na 1,878 1,094 988 Source: UN 2004 Estimates and Projections. 26 Chapter 4 population pyramid and later provides few in the working ages to support the heavy load of older parents and grand- parents. India’s youth will increase by 10% in 2020 but most of that growth will occur before 2015, ten years from now. Some 13 million youth will be added by 2010 and another 6 million by 2015. Latin America will be the slowest grow- ing of the regions shown, up by only 2% by 2010 and 4% by 2020. All of these patterns reflect the current and future age distributions of childbearing-age women and the projected age-specific fertility rates. Marriage: The percent already married or cohabiting in the age group 15-19 var- ies a great deal. In Asia the variation is from 47% in Bangladesh or 34% in India down to 1% in the Republic of Korea, 8% in Malaysia and the Philippines, and 7% in Sri Lanka. Similar extremes, both high and low, occur in Latin America and in sub-Saharan Africa. These are remark- able differences in the age of first formal union and, to a large extent therefore, in the age at first birth. Single women are not included in most surveys in Asia or the Middle East/North Africa, but in Latin American surveys, up to 11% of single women aged 15-19 admit to sexual activity. The sub-Saharan pattern is quite different: the high range is about 40% admitting sexual activity, and numerous countries are in the teens and 20s. Contraception by single women: In the same two regions there is a substantial percent of sexually active single women who say they are using contraception. In Latin America, for the young age group 15-19, four countries report percents above 60%; other countries are in the 30s to 50s. In sub-Saharan Africa the range is greater because it starts lower, reflecting the low presence of contraception in gen- eral in some countries. Contraception by married women: The percents reported above are usually much lower for married women. Sexual- ly active adolescents are more likely to use some method, even if erratically, than the average married woman, since many of the latter are currently preg- nant, amenorrheic, or not sexually ac- tive. Childbearing is already common among women in the next age group, at 20-24. However in China, only 8% of all women 20-24 have given birth and only 16% have done so in Viet Nam and Sri Lanka. The other extremes are found in Bangladesh (61%), India (47%), and Nepal (51%). Figures in Latin America are restricted to a narrow range in the 30s and 40s. Many sub-Saharan African countries are well above that; several are in the 50s, 60s, and even the 70s. The lowest range is in the Middle East/North Africa (Appendix Table A.24). All births: Finally, the two age groups, 15-19 and 20-24, contribute a large share of all births, from a third to one- half. Very few of the 116 countries in Appendix Table A.24 are above the 50% figure. Most countries in Latin America and sub-Saharan Africa are clustered in the 40s, whereas somewhat more fall be- low that in Asia and especially in the Middle East/North Africa. Attended births: Large proportions of births are still unattended by trained per- sonnel (last column of Appendix Table A.24). Less than half attended is not un- common in Asia and sub-Saharan Africa. All births should be attended, but only 16 countries outside of the former USSR exceed 80% of births attended by trained personnel. Clearly, young adults constitute a major share of all pregnancy experience and childbearing throughout the developing world. The corresponding needs for edu- cation and services are substantial, par- ticularly since so much of this early ex- perience is for first pregnancies and first births. Unmet Need for Youth Youth account for much of the unmet need, just as they do for numbers of births. Most unmet need information is for married women, so they are dis- cussed first. Based on an international analysis of unmet need in 55 national surveys (Ross and Winfrey, 2002), the two age groups 15-19 and 20-24 account for a full one-third (32.8%) of unmet need in the whole married group, or 34.9 million women. In the Middle East/ North Africa region the percent falls well below the average, at only 23.3%, and in the Central Asia Republics the percent is lower too at 27.9%. Within sub-Saharan Africa the figure is 30.3%, Numbers (000) Ages 15-19 Ages 20-24 Ages 15-24 Total 11,445 23,469 34,914 Latin America 1,064 2,405 3,469 Sub-Saharan Africa 2,465 4,663 7,128 Asia (except China) 7,390 14,679 22,069 Middle East/North Africa 479 1,457 1,936 Central Asia Republic 47 265 312 Percents Ages 15-19 Ages 20-24 Ages 15-24 Total 24.5 22.5 23.1 Latin America 27.2 20.2 21.9 Sub-Saharan Africa 25.6 26.0 25.9 Asia (except China) 24.4 22.7 23.3 Middle East/North Africa 18.0 17.4 17.5 Central Asia Republic 14.0 15.8 15.5 Source: Ross and Winfrey, 2002. Table 4.4. Numbers and Percentages of Currently Married/In Union Women with Unmet Need, by Region, for Ages 15-19 and 20-24. Chapter 4 27 within Latin America 31.3%, and within Asia 35.4%. In sheer numbers young married women weigh heavily in the balance. They tend to be at low parities but many are interested in limiting child birth as well as spacing. More unmet need exists at ages 20-24 than at ages 15-19: 23.5 million vs. 11.4 million in Table 4.4. The age group 15- 19 is larger than the 20-24 group but fewer members of it are married or in union, therefore in terms of sheer num- bers the 20-24 group contains twice the number in need. The differential is even more extreme in the Middle East/North Africa and in the Central Asia Repub- lics. It is slightly less marked in sub-Sa- haran Africa because cohabitation be- gins earlier, and the entire 15-19 age group is considerably larger in relation to the 20-24 age group than it is else- where. The two young age groups are quite sim- ilar in the percent of married women having an unmet need, except in Latin America, as shown in Table 4.4. Overall, regional differences are small, although the Middle East/North Africa and the Central Asia Republics have 15%-17% in need compared to 22%-26% else- where. Sub-Saharan Africa has the highest per- cent with unmet need for both young married women and all married women, at about 25% for both groups (not shown here; see section on unmet need). How- ever the other regions show differences; young women have more unmet need by a considerable margin in Latin America (21.9% vs. 13.7%), in Asia (23.3% vs. 16.4%), and in the Central Asian Repub- lics (15.5% vs. 11.3%), though by rath- er little in the Middle East/North Africa (17.5% vs. 15.6%). Unmarried Women: Information is se- verely limited for unmarried women at young ages, except in sub-Saharan Afri- ca. There, among never-married women, the percentage with unmet need is 7.3% at ages 15-19 and 10.7% at ages 20-24. Among previously married women it is 15.4% and 15.7% respectively. HIV/AIDS and Youth Unmet need in the larger sense is made worse by the AIDS epidemic. Youth aged 15-24 account for half of all new HIV in- fections worldwide, and some 6,000 contract the virus each day (UNAIDS, 2004). Of all youth already living with HIV, two-thirds are in sub-Saharan Afri- ca, and of them most (75%) are female. At the end of 2003 the regional distribu- tion for youth living with HIV was as follows: Sub-Saharan Africa 62% Asia 22% Latin America 7% Middle East/North Africa 1% Eastern Europe & Central Asia 6% High-income countries 2% The future of the AIDS epidemic turns heavily upon the success of programs directed to youth. While many are al- ready infected, a majority are not, and if behavioral change can be established it will persist to later ages and help reduce the base of HIV carriers that generates new cases of AIDS. HIV prevalence, especially in sub-Sahar- an Africa, is much higher among teenage girls than boys. Ratios there range from 2 to 1 to as high as 4.5 to 1. An aggravat- ing factor is that large age differences exist between girls aged 15-19 and their male sexual partners, many of whom are already infected. (See the main section on HIV/AIDS in Chapter 4.) References UNAIDS, 2004 Report on the Global AIDS Epidemic: 4th Global Report. See p. 22. Ross, John, and William Winfrey. “Un- met Need for Contraception in the De- veloping World and the Former Soviet Union: An Updated Estimate.” Interna- tional Family Planning Perspectives 28(3):138-143. 2002. 28 Chapter 4 Chapter 4 29 MATERNAL HEALTH Chapter 5Chapter 5Chapter 5Chapter 5Chapter 5 Maternal health is a complex subject, too large to consider in great detail. Here we discuss four aspects, supported by Ap- pendix Tables A.15 through A.21. ➤ Maternal Mortality and Morbidity ➤ Antenatal, Delivery, and Tetanus Care ➤ Induced Abortion and Postabortion Contraception ➤ Program Efforts to Improve Maternal Health Maternal Mortality and Morbidity Maternal mortality has received continu- ing attention as a major problem of the developing world, but clear evidence of progress against it is lacking. The three international series of estimates of mater- nal mortality ratios (MMRs) for 1990, 1995, and 2000 show erratic trends for many countries, due to defective data and also to some changes in methodology. In addition the small number of deaths in some data sets produces large sampling errors. Still, the average MMR for all de- veloping countries declined from 528 in 1990 to 502 in 1995 to 444 in 2000 (the medians were 300, 255, and 230). De- clines also appear in long-term trend data for selected countries, including China (UN, 2002). Essentially no developing country has an MMR anywhere near the ratios in the West, where the average ratio in 2000 was 20. On average, in developing coun- tries about 1 woman in 61 can expect to die from pregnancy-related causes some- time during her reproductive career (Ta- ble 5.1); at worst, in some countries this is 1 in 10. The basic facts are not in dispute; here are examples from a World Bank (1999) review: ➤ Nearly 99% of the more than 500,000 maternal deaths each year occur in the developing world. ➤ Of all the human development indica- tors, the greatest discrepancy between developed and developing countries is in the risk of maternal death. ➤ Complications of pregnancy and child- birth are the leading cause of death and disability among women of reproductive age in developing countries. ➤ One in four women in these countries suffers from acute or chronic conditions related to pregnancy. ➤ At least 20% of the burden of disease among children below age five is attribut- able to conditions directly associated with poor maternal health, nutrition, and the quality of obstetric and newborn care. ➤ Most of this loss and suffering is pre- ventable. Within the developing world the regions differ significantly in average risk, but all are far above the rates for Europe and North America. In the following section we discuss three features, drawn from Appendix Table A.15: numbers, ratios, and lifetime risks. Due to defects in the original data, all fig- ures are approximations and the patterns presented must be viewed in general terms. Numbers of Deaths. The total numbers of deaths reflect population sizes and birth rates as well as the mortality risks, so the numbers are very uneven by re- gion. Asia has nearly one-half of the total (Table 5.2), due largely to India, and sub- Saharan Africa has about as many. India’s maternal deaths far exceed those of any other country. Its 136,000 deaths (Figure 5.1) compose over one-fourth (27%) of the developing world total; this vastly exceeds China’s total of 11,000, reflecting China’s fewer births and its much lower maternal mortality ratio of 56 vs. India’s 540. Other countries follow after India. Nige- ria has about 37,000 annual maternal deaths; Pakistan, D.R. Congo, and Ethio- pia have about 25,000 each; Afghanistan and Bangladesh have between 16,000 to 20,000; and numerous others have about 10,000 each. While all these estimates are subject to considerable error, Figure 5.1 identifies a rough ordering for large countries. The ratios (MMRs) are much higher in sub-Saharan Africa than in other regions. Within each region the MMRs vary sharply, as the distributions in Figure 5.2 demonstrate. However the central ten- dencies follow a clear ordering: in sub- Saharan Africa the range is from about 1800 to one-tenth of that. In Asia, apart from the single high figure (for Nepal), the range is from about 650 to quite low figures, and in Latin America the range starts at about 400 and also falls to low figures. The lowest ratios are in areas of the former Soviet Union, where the ex- tensive health systems have covered most women. Table 5.1. Women’s Lifetime Risk of Dying from Pregnancy and Childbirth Region Risk of Dying Developed Countries 1 in 2800 All Developing Countries 1 in 61 Sub-Saharan Africa 1 in 16 South Asia 1 in 43 East Asia and Pacific 1 in 360 Middle East and North Africa 1 in 100 Latin America and Caribbean 1 in 160 Source: AbouZahr and Wardlaw, 2004. Table 4.2. Table 5.2. Number and Percent of Maternal Deaths Annually (2000), Developing Countries Number of Maternal Deaths Country Annually Percent Asia 237,665 47.3 Latin America 21,570 4.3 Middle East/North Africa 20,362 4.1 Sub-Saharan Africa 220,925 44.0 Central Asia Republics 1,000 0.2 Caucasus 140 0.0 Moldova, Russia, Ukraine 990 0.2 Total 571,652 100.0 30 Chapter 5 These ratios partly overlap with the pic- ture for numbers of deaths. High ratios occur in Nepal and Pakistan, in Sudan and Yemen, and most especially in 17 sub-Saharan countries that have ratios of 1,000 or higher. The lifetime risks vary across a vast range, from 1 woman in 6 or 7 dying in Afghanistan, Angola, Malawi, Niger, and Sierra Leone to only 1 in 1,100 or more 136,000 37,000 26,000 24,000 24,000 20,000 16,000 11,000 11,000 11,000 10,000 10,000 9,700 9,300 8,700 7,900 6,800 6,400 6,000 5,400 - 20,000 40,000 60,000 80,000 100,000 120,000 140,000 160,000 Ind ia Nig er ia Pa kis tan Co ng o, D.R . Eth iop ia Afg ha nis tan Ba ng lad es h Ch ina An go la Ke ny a Ind on es ia Ug an da Nig er Ma law i Bra zil Mo za m biq ue Ma li Su da n Ne pa l Bu rkin a Fa so Figure 5.1. Number of Maternal Deaths, 2000 (top 20 countries) 0 200 400 600 800 1000 1200 1400 1600 1800 2000 Asia SubSaharan Africa Latin America Middle East/North Africa in South Korea, Chile, Cuba, and the former USSR. (These can be converted by their reciprocals to the percents im- plied: for example, 14% for a 1 in 7 risk.) The risk in each country reflects both the average number of births per woman (the TFR) and the risk per birth (the MMR), so women in countries with high fertility rates and high ratios will have the highest lifetime risks. (See Appendix Table A.15.) The MMR is normally stated as maternal deaths per 100,000 births. The same number of deaths can be compared to pregnancies rather than births. If about 120 million births occur in the develop- ing world, they represent perhaps two- thirds to three-fourths of all pregnancies. If there are at least 500,000 maternal deaths in these countries the overall ratio is about 415. However, with the 160 to 180 million pregnancies as the denomi- nator the ratio is less, at 277 to 313. That however still translates to one death per minute year around, in addition to many times more for serious disability. Strategies. A strong argument has been urged that highly specific measures are essential to reduce the numbers of mater- nal deaths. General improvements in so- cioeconomic status will not significantly lower maternal mortality rates, since most deaths occur for lack of well- equipped medical facilities close at hand and ready at short notice to assist the woman experiencing difficulty. Screen- ing in advance to identify high-risk sub- groups is not efficient, since “.the vast majority of high-risk women will deliver without incident. Furthermore, most women who develop life-threatening complications belong to low-risk groups.” (Maine and Rosenfield, 1999), Figure 5.2. MMR Values by Region, 2000 (3 groups to right are 5 CARs, 3 Caucasus, and Moldova, Russia, Ukraine) 136,000 Chapter 5 31 apparently because of their sheer num- bers in the population. The implications of such analyses are that sheer numbers of deaths will not fall greatly until there is close access to appropriate medical ser- vices to treat emergency cases. The exception is broad-scale family plan- ning since that reduces the overall num- ber of unplanned and unwanted pregnan- cies in the first place. Moreover, enlarged contraceptive use offsets abortions that would otherwise occur, many of which produce maternal deaths from septic pro- cedures. (See abortion section.) A full strategy to reduce the total number of maternal deaths in the developing world must take into account their high- ly concentrated geographic distribution (Figure 5.1). Within any country, deaths are a function of the number of women, the birth rate, and the risk per birth (or pregnancy). The latter is to some extent a function of unsafe abortions. Wider con- traceptive use addresses both the abortion rate and the birth rate, but the numbers of deaths will remain far too high until med- ical facilities improve in close proximity to most women. References AbouZahr, Carla, and Erica Royston. Maternal Mortality: A Global Factbook. Geneva: WHO. 1991. AbouZahr, Carla, and Tessa Wardlaw. “Maternal Mortality in 2000: Estimates Developed by WHO, UNICEF, and UNFPA.” Geneva: WHO. 2004. Adamson, P. “Women: Maternal Mortal- ity.” In: Adamson, P., ed. Progress of Na- tions. New York: UNICEF. Pp. 2-7. 1996. Maine, Deborah. Safe Motherhood Pro- grams: Options and Issues. New York: Columbia University, Center for Popula- tion and Family Health. 1991. Maine, Deborah, and Allan Rosenfield. “The Safe Motherhood Initiative: Why Has It Stalled?” American Journal of Public Health, Vol. 89, No. 4. Pages 480- 482. April 1999. Tsui, Amy, Judith N. Wasserheit, and John G. Haaga, eds. Reproductive Health in Developing Countries. Expanding Di- mensions, Building Solutions. Panel on Reproductive Health, Committee on Pop- ulation, Commission on Behavioral and Social Sciences and Education. National Research Council. Washington, DC: Na- tional Academy Press. 1997. UNICEF. The Progress of Nations. New York: UNICEF. 1996. United Nations. World Population Moni- toring 2002: Reproductive Rights and Reproductive Health: Selected Aspects. UN Commission on Population and De- velopment, Thirty-fifth Session, April 1- 5, 2002. Draft, page 107, citing data from WHO and UNICEF databases. WHO. Coverage of Maternal Care: A Listing of Available Information, Fourth Edition. Geneva: WHO. 1997. WHO. Mother-Baby Package: A Road Map for Implementation in Countries. Geneva: WHO, Division of Family Health. 1993. WHO and UNICEF. “Revised 1990 Esti- mates of Maternal Mortality: A New Ap- proach by WHO and UNICEF.” Geneva: WHO and UNICEF. April 1996. WHO / UNICEF / UNFPA. “Maternal Mortality in 1995: Estimates Developed by WHO, UNICEF, UNFPA.” Geneva: World Health Organization. 2001. World Bank. Safe Motherhood and the World Bank: Lessons from Ten Years of Experience. Washington, DC: The World Bank. June 1999. Antenatal, Delivery, and Tetanus Care Burdens upon the health system reflect large numbers of births and constraints that cause major shortfalls in the propor- tions of women currently served. Three functions are discussed below: antenatal care, tetanus immunizations, and delivery attendance. Antenatal care. Only about 70% of births are preceded by even a single ante- natal visit in the developing world as a whole. Across regions (Table 5.3) the range is from 65% to 69% in Asia, the Middle East/North Africa, and sub-Sa- haran Africa, and up to 87% in Latin America. It is 93% in the Central Asia Republics as an inheritance from the former USSR system. In terms of numbers of women unserved, a full 36 million women receive no ante- natal care annually (Table 5.3 and Ap- pendix Table A.17). By region, the esti- mates are 21.6 million in Asia, 1.5 mil- lion in Latin America, 2.9 million in the Middle East/North Africa, and 9.5 mil- lion in sub-Saharan Africa. Although the United Nations estimates that the total number of births per year in the develop- ing world as a whole has leveled off, the number will still increase in such large countries as Ethiopia and Pakistan, so even if the proportions of women assist- ed improve, the absolute numbers un- served may still rise. The distribution of countries according to the percent of women receiving antenatal care appears in Table 5.4, for 80 coun- tries. One-third (19) of countries fall be- low the 70% mark (column 2). That is, in these countries less than 70% of women receive antenatal care. In 15% of coun- tries, or about one in six, less than 50% of women receive care. However the decade of the 1990s saw notable progress in antenatal care accord- ing to a careful report issued recently (AbouZahr and Wardlaw, 2004). Defin- ing care as one or more antenatal visits and using data from 49 countries with multiple surveys, the overall trend was up from about 53% of pregnant women in 1990 to about 64% in 2000. The rise was sharpest in Asia and least in sub-Saharan Africa and the Middle East/North Africa. An encouraging finding was that over half of women receiving any care re- ceived at least four visits. Exceptions, however, include Bangladesh, Ethiopia, Morocco, Nepal, and Yemen, all of which have substantial percentages of women who have only one visit. South Asia, overall, had the lowest levels of antenatal care with only 50% of women getting even one visit. Delivery care. Professionally trained birth attendants, whether paramedics or doctors, and whether serving at home or 32 Chapter 5 in facilities, are the focus here. Coverage of births by professional attendants rests at about 59% of births for the developing world as a whole (Table 5.3). The range of variation across regions is consider- ably greater than it is for either tetanus or antenatal care since only 41% of deliver- ies are attended in sub-Saharan Africa. The Asia figure is a low 60%; a high fig- ure for China is offset by lower ones for Bangladesh, India, and Pakistan (Appen- dix A.17). (All regional figures above weight countries by number of deliver- ies.) Converted to numbers, these percentages mean the neglect of 28.4 million women in Asia, 2.1 million in Latin America, 2.7 million in the Middle East/North Africa, and 16.3 million in sub-Saharan Africa. Improvements in the proportions of births attended will tend to offset the increasing absolute numbers of births in countries like those mentioned above, but will still leave vast numbers unattended. The dis- tribution of countries by the percent of births attended is shown in Table 5.4. Again, a single fact is eloquent: one-half of countries attend less than 60% of births. One-fourth of countries attend less than 40% of births. Tetanus immunizations. A similar anal- ysis for tetanus shows about 70% of women receiving care, but compared with antenatal care more receive care in Asia and fewer in Latin America, Middle East/North Africa, and sub-Saharan Afri- ca (Table 5.3). It must be said that the original country data are quite rough and approximate; nevertheless the picture of serious shortfalls cannot be doubted. Translated to numbers about 16.1 million women are omitted from tetanus protec- tion in Asia, 3.7 million in both Latin America and Middle East/North Africa, and 10.8 million in sub-Saharan Africa. Again, the coming five-year increases in both women and births in certain large countries will elevate these numbers un- less the proportions served rise enough to counteract them. The distribution of countries by the per- cent of women receiving tetanus immuni- zations is shown in Table 5.4. It is consid- erably worse than the one in that table for antenatal care. Over half of countries fall below the 70% mark for women treated. Three-way comparison. Most countries in the developing world have large defi- ciencies in maternal care. Figure 5.3 gives the cumulative distribution of all countries by the percent of women re- ceiving care for the three services of an- tenatal visits, tetanus protection, and de- livery attendance. The ideal curve would stay very low along the bottom axis, indi- cating that few countries have low per- centages of care, and it would then rise very sharply at the right, placing most countries at the favorable high percentag- es. The space beneath each line reflects the failure to provide care. Thus the best curve is for antenatal visits and the worst is for attended deliveries. In between is tetanus; it crosses the 50% point for countries at the unfavorable point of only 60% of women with tetanus protection. Figure 5.3 is based upon 80 developing countries that have data for all three types of care, to create a fair comparison. The omitted countries include those that lack data on one or more types, and several of the omitted countries happen to be those with stronger health systems. When they are included the antenatal and tetanus curves are essentially unchanged, but the delivery curve is more favorable, lying close to the tetanus curve. Therefore the figure represents the situation for the less favored countries, which is probably more relevant to action planning. The burdens of care and the needs for ser- vices will rise inexorably in those devel- oping countries with increasing numbers of births. A race is under way between those increases and the effort to improve coverage of services, made more chal- lenging by the drive to also improve quality. Progress on coverage and quality may be largely cancelled in these coun- Table 5.3. Mean Regional Values for Care Received, with Numbers Unserved Percent of Women Receiving Care Numbers Unserved No. of Deliveries Antenatal Deliveries Tetanus (000s) Antenatal Deliveries Tetanus Asia 69.4 59.6 77.0 70,430 21.580 28,445 16,183 Latin America 86.9 82.1 68.2 11,511 1,504 2,061 3,660 Middle East/ North Africa 70.5 71.8 61.9 9,806 2,894 2,761 3,738 Sub-Saharan Africa 65.4 40.7 60.7 27,508 9,525 16,311 10,813 Central Asia Republics 92.5 93.7 u 1,187 89 74 u Caucasus 75.8 88.3 u 232 56 27 u Developing World 70.4 58.7 70.3* 120,675 35,748 49,796 35,812 u - unavailable *Figure would be somewhat higher with data for the two missing regions. Table 5.4. Distribution of Countries by the Percent of Women Receiving Care Percent Antenatal Care Attended Deliveries Tetanus Immunizations of Women No. of Cumulative No. of Cumulative No. of Cumulative Receiving Care Countries Percent Countries Percent Countries Percent 0-9 - - 1 1 2 3 10-19 - - 6 9 - 3 20-29 3 4 5 15 4 8 30-39 3 8 8 25 7 16 40-49 6 15 13 41 11 30 50-59 1 16 7 50 9 41 60-69 12 31 12 65 10 54 70-79 17 53 6 73 7 63 80-89 17 74 9 84 15 81 90-100 21 100 13 100 15 100 No. of countries 80 80 80 Source: Appendix Table A.10. Chapter 5 33 tries by the rising numbers of births un- less both efforts and resources are great- ly augmented. References AbouZahr, Carla and Tessa Wardlaw, An- tenatal Care in Developing Countries; Promises, Achievements, and Missed Op- portunities. An Analysis of Trends, Lev- els, and Differentials, 1990-2001. WHO and UNICEF 2004. UNICEF. The State of the World’s Children 2005. New York: UNICEF. Dec. 2004. United Nations. World Population Pros- pects: The 2002 Revision. Volume I, Comprehensive Tables. New York: Unit- ed Nations Population Division. See also the 2004 Revision. WHO. The World Health Report 1998. Geneva: World Health Organization. 1998. See also the 2005 edition. Induced Abortion and Postabortion Contraception Planners need to know the level of abor- tion activity to sense the burdens that weigh upon both maternal health and ser- vice networks. They also need to gauge the requirements for preventive care, in- cluding contraception and postabortion contraception. Here we provide three measures of abortion activity (Appendix Table A.16), using data drawn primarily from the World Health Organization and the Alan Guttmacher Institute. These data are only rough estimates for many coun- tries and should be used with caution. Recent regional data on abortion are shown in Table 5.5 (see also Henshaw, Singh, and Haas, 1999). Asia dominates the number of abortions done annually, with 55% of the table total and 57% of the developing world total. China is re- sponsible for much of this, but the rest of Asia still contains over half of abortions in the developing world with China re- moved. This reflects the large numbers in India, Indonesia, Pakistan, and the Phil- ippines. Latin America is next, due to the presence of Brazil, Mexico, and Venezu- ela. Next is sub-Saharan Africa, where Nigeria ranks first; in the Middle East/ North Africa, Turkey and Egypt have the largest numbers. In the five Central Asian Republics, Uzbekistan has an estimated two-thirds of the total. The number of abortions reflects two determinants: the number of pregnancies and the proportion aborted (the abortion ratio is however abortions per births). From another perspective, the numbers reflect the abortion rate (the proportion of all women having an abortion annual- ly). Thus, the size of the population, the number of pregnancies, and the proclivi- ty to terminate them, all enter in. Further, each of these three has its own prior de- terminants; most notably, increased con- traceptive use reduces the number of pregnancies and therefore the total num- ber of abortions. Rates vary across a wide range, from lows of only 6 to 10 (annual abortions per 1,000 women aged 15-49), to highs in the 60s and 70s. The rates and ranks in Appendix Table A.16 reflect these ex- tremes. An overview of regional differ- ences appears in Table 5.5; the rates are especially low in Western Asia and Northern Africa (corresponding largely to our designation of the Middle East/ North African region) and in sub-Saharan Africa. Next highest is Asia, and then Latin America. Much higher rates have been registered in the former Soviet Union regions. Within every region, there is rather wide country variation. The ratios of abortions per 100 births present a different picture from the rates for many individual countries, but the re- gional patterns are largely unchanged. The ratios are quite low in the Middle East/North Africa, higher in Asia and even higher in Latin America. In Appen- dix Table A.16 the highest averages are for the Central Asian Republics and for the group of Moldova, Russia, and Ukraine. Five of the ten highest country ratios occur in the former Soviet Union. Widespread contraceptive practice great- ly reduces the number of pregnancies and therefore the rate, but the ratio may either rise or fall. It may fall if most of the re- maining pregnancies are wanted ones, but it may rise if there are many contra- ceptive failures and a high proportion is aborted. In that case, most abortions serve as backup for defective contracep- tive methods (especially traditional meth- ods) or defective use of the methods. In other situations, where abortion is a pri- mary instrument of birth control, the abortion ratio can be very high, and it matters considerably whether the ratio is based on pregnancies or births. If in the Russian Federation two-thirds of preg- nancies are aborted that leaves one-third for births, so while the ratio is 66 per 100 pregnancies it is 66 per 33 births, or a standard ratio of 150. 10 20 30 40 50 60 70 80 90 100 0-9 10-19 20-29 30-39 40-49 50-59 60-69 70-79 80-89 90-100 Antenatal Deliveries Tetanus 0 C um ul at iv e P er ce nt o f C ou nt rie s Figure 5.3. Cumulative Distribution of Countries for Maternal Care: Antenatal Care, Tetanus Immunization, and Delivery Attendance Cumulative Percent of Women Receiving Care 34 Chapter 5 Regardless of whether abortion is legal or illegal, increased contraceptive use will cut into the base of unwanted pregnan- cies. In particular, postabortion contra- ception addresses the population most concerned. A strategy to offer advice and methods during the abortion episode is vital, since many women will not be seen again and many will go on to repeat abor- tions. Deaths: Unsafe abortions are thought to account for some 68,000 maternal deaths each year, or about 13% of all maternal deaths. Thus they constitute about one- eighth of the maternal deaths that occur in the developing world. Safer medical procedures, including new non-surgical ones, will save lives and reduce maternal morbidity. Abortion is made safer and less traumatic by the use of vacuum pro- cedures; these are now in common use in many developing countries. There are three general strategies to re- duce the numbers and rates of abortion, whether safe or unsafe: 1. Gross numbers will continue to occur mainly in a relatively small number of countries with large populations and high rates; therefore, regional and internation- al strategies should take account of the geographic pattern. 2. The abortion rates within countries will fall as the pregnancy rate falls, the key to which is the spread of reliable con- traceptive use. 3. The focused strategy of contraceptive offerings at the time of abortion will re- duce repeat abortions and orient action to the subgroup most prone both to have re- peat unwanted pregnancies and to termi- nate them. The scourge of unsafe abortions in much of the developing world can be reduced by better contraceptive offerings, to pro- vide more methods and easier access to more people. Abortion can be made saf- er through the spread of vacuum aspira- tion equipment. Contraception offered at the time of each abortion is especially important to reduce future unwanted pregnancies and repeat abortions. Con- traceptive failure can be reduced by the provision of better counseling and a wid- er choice of reliable methods. Table 5.5. Global and Regional Estimates of Annual Incidence of Unsafe Abortion and Mortality Due to Unsafe Abortion, by United Nations Regions, Around the Year 2000 Unsafe Abortion Incidence Mortality Due to Unsafe Abortion NUMBERS RATIOS RATES Unsafe Unsafe Abortions Unsafe Abortions Number of Percent of all Unsafe Abortion Abortions to 100 per 1000 Women Maternal Deaths Due Maternal Deaths to (thousands) Live Births Aged 15-44 to Unsafe Abortion Deaths 100,000 Live Births World 19,000 14 14 67,900 13 50 Developed countriesa 500 4 2 300 14 3 Developing countries 18,400 15 16 67,500 13 60 Africa 4,200 14 24 29,800 12 100 Eastern Africa 1,700 16 31 15,300 14 140 Middle Africa 400 9 22 4,900 10 110 Northern Africa 700 15 17 600 6 10 Southern Africa 200 16 17 400 11 30 Western Africa 1,200 13 25 8,700 10 90 Asiab 10,500 14 13 34,000 13 40 Eastern Asia - - - - - - South-central Asia 7,200 18 22 28,700 14 70 South-eastern Asia 2,700 23 21 4,700 19 40 Western Asia 500 10 12 600 6 10 Latin America & Caribbean 3,700 32 29 3,700 17 30 Caribbean 100 15 12 300 13 40 Central America 700 20 21 400 11 10 South America 2,900 39 34 3,000 19 40 Europe 500 7 3 300 20 5 Eastern Europe 400 14 6 300 26 10 Northern Europe 10 1 1 - 4 - Southern Europe 100 7 3 <100 13 1 Western Europe - - - - - - Northern America - - - - - - Oceaniab 30 12 17 <100 7 20 a Figures may not add exactly to totals because of rounding. b Japan, Australia and New Zealand have been excluded from the regional estimates, but are included in the total for developed countries. – No estimates are shown for regions where the incidence is negligible. Source: WHO, 2004. Chapter 5 35 *For details and additional information see Ross and Begala (2005) and other references below. The 13 components were based upon 81 items in a stan- dard questionnaire completed by 10-25 expert ob- servers in each country. References Henshaw, Stanley K., Susheela Singh, and Taylor Haas. “The Incidence of Abortion Worldwide.” International Family Plan- ning Perspectives, Vol. 25, Pages S30- S37. Supplement, 1999. Ross, John A. and Elizabeth Frankenberg. “Induced Abortion,” Ch. 9 in Findings from Two Decades of Family Planning Research. New York: The Population Council. 1993. Rossier, Clementine. “Estimating Induced Abortion Rates: A Review.” Studies in Family Planning 34(2):87-102, June 2003. WHO. “Global and Regional Estimates of Incidence of and Mortality due to Un- safe Abortion with a Listing of Available Country Data,” Table 2. Maternal and Newborn Health: Unsafe Abortion. Third Edition. Geneva: WHO. 1998. See also the Fourth Edition, Table 3. 2004. Singh, S., J. E. Darroch, M. Vlassoff, and J. Nadeau. Adding It Up: the Benefits of Investing in Sexual and Reproductive Health Care. New York: Alan Guttmach- er Institute and UNFPA. 2003. Program Efforts to Improve Maternal Health (See Appendix Table A.18) National programs to reduce maternal mortality and morbidity exist in every country. Surveys were conducted in both 1999 and 2002, in some 50 countries, to assess both the types and levels of effort devoted to these programs. A maternal and neonatal health programs index (MNPI) was used to measure 13 compo- nents of the programs.* Results in 1999 and 2002 matched close- ly by component, for the 43 countries that participated in both surveys (Figure 5.4). (Each score is the percent of maxi- mum, and the total score is the average of the 13 component scores.) The close match in pattern is reassuring as to the re- liability of the methodology since the surveys were done independently; it also indicates that, overall, little change oc- curred in the three years between sur- veys. The improvement was only from 54.9% in 1999 to 57.1% in 2002 for the total score, and it was also small for each of the components. That is not surprising since major shifts in effort large enough to affect the international average could hardly be expected in a short time. Among the 13 components the highest rating goes to newborn services (70% in both years), due in part to the successful immunization campaigns in many coun- tries. The bottom place is held by three components at about 50% of maximum effort: health center capacities, resources, and public education. Another low score is for access to services and it is especial- ly low for rural access. In the middle are six components at about 60% of maxi- mum: district hospital capacities, antena- tal and delivery services, family planning at health centers and at district hospitals, and policies bearing on safe pregnancy. The final two components, for training and monitoring/evaluation, are near 55%. The range across the 13 component scores is from 45% to 70%, demonstrat- ing that a strong selectivity exists in what government programs emphasize. Among regions there are also substantial differences. Table 5.6 for 2002 (all 55 countries) shows the 13 component scores for each region. The chief features are: ➤ The South Asia region (India, Paki- stan, Nepal, and Bangladesh) has the lowest total score, and it is lowest or tied for lowest on most of the components. ➤ Francophone sub-Saharan Africa is next lowest both on the total score and on the number of components for which it holds the bottom rank. ➤ East and Southeast Asia shows the highest total score, and it is highest on most components. ➤ The other regions fall at intermediate levels for the total score and most com- ponent scores. ➤ The greatest difference among re- gions is in access to services by most of the population. Thus public access to ma- ternal services varies greatly, both in the rural sector and in the urban sector. Access to services in the rural sector, where most of the population lives, is far worse than urban access. This appears in every region regardless of the overall lev- els (Figure 5.5). The disparity is close to 30% in every case. Figure 5.4. Comparison of 2002 and 1999 Surveys, by Component (43 Countries Included in Both Years) -10 0 10 20 30 40 50 60 70 80 90 100 He alt h C en ter Ca pa citi es Dis tric t H os pit al Ca pa citi es Pe rce nta ge wit h A cce ss to Ca re An ten ata l S erv ice s De live ry Se rv ice s Ne wb orn Se rvic es FP at He alt h C en ter s FP at Dis tric t H os pit als Po lici es tow ard Sa fe Pre gn an cy Re so ur ce s & Pr iva te Se cto r Inf or m ati on , Ed uc ati on Tra inin g A rra ng em en ts Mo nit or ing , E val ua tio n TO TA L Pe rc en t o f M ax im um S co re 2002 1999 Difference 36 Chapter 5 Figure 5.5. Rural and Urban Access Scores, by Region, 2002Relation to Maternal Mortality: The association between program effort and the maternal mortality ratio appears in Figure 5.6. All countries in the study are divided into three groups according to their MMR levels. The figure shows the score for each group on each of eight program features taken selectively from the various components. The heavy line is for countries with the lowest MMRs; this line lies to the outside of the figure, showing the highest effort scores on ev- ery feature. The sharpest relationships between pro- gram effort and maternal mortality ap- pear for access to postpartum family planning and to safe abortion, where the lines are farthest apart. They are also far apart for access to emergency treatment. Each of these has a plausible relationship to lower maternal mortality. Postpartum family planning programs encourage contraceptive use, which in turn avoids unwanted pregnancies with their elevated risks. Safe abortion services help avoid septic complications that cause deaths, and emergency treatment is vital in cases of hemorrhaging and other life-threaten- ing complications. Smaller differences appear where all scores are relatively high, as with antena- tal care access, and especially with im- munizations, which have received special attention in recent years. East and Latin America Middle East Anglophone Francophone Southeast South and the and Sub-Saharan Sub-Saharan All Asia Asia Caribbean North Africa Africa Africa Regions 1. Capacities of health centers 52.6 45.4 49.3 50.8 51.5 54.7 51.3 2. Capacities of district hospitals 65.9 58.8 61.8 75.6 61.4 58.5 62.8 3. Total access 70.2 37.2 56.1 68.8 54.4 39.6 53.3 3A. Rural access 62.2 30.7 40.6 53.8 45.5 30.5 42.4 3B. Urban access 85.1 59.2 67.6 80.9 71.0 58.4 69.0 4. Care at antenatal visits 60.5 50.0 65.9 58.7 67.9 67.4 64.2 5. Care at delivery 68.2 42.7 62.6 62.8 61.2 57.7 60.2 6. Care for newborns 75.1 57.0 75.7 76.8 72.0 70.0 72.1 7. Family planning at health centers 62.5 47.4 56.2 65.5 65.9 51.0 58.6 8. Family planning at district hospitals 68.2 59.8 59.3 57.3 61.4 51.1 58.5 9. Policies toward safe pregnancy 69.0 57.6 58.4 60.8 65.6 67.0 63.5 10. Resources 50.9 50.4 48.8 55.6 52.4 37.6 48.2 11. Information, education 61.8 51.5 41.4 51.8 53.6 50.0 50.5 12. Training arrangements 63.5 50.1 56.3 56.1 52.8 47.6 53.5 13. Monitoring, evaluation 65.8 48.6 57.4 55.3 59.7 54.2 57.1 Total Score 64.2 50.5 57.6 61.2 60.0 54.3 58.0 Note: The total score for all regions is 58.0%, for the 55 countries in 2002, while it is 57.1% for the 43 countries common to both surveys. 0 20 40 60 80 Emergency treatment access Attended deliveries Antenatal care access Safe abortion access Postpartum FP access Immunizations Training doctors Training nurses, midwives ____ MMR <250 __ __ MMR 250-749 - - - MMR 750+ 30.9 40.9 51.8 41.0 30.7 57.9 56.8 67.8 82.3 66.7 59.2 86.1 0 10 20 30 40 50 60 70 80 90 100 Francophone Sub-Saharan Africa Anglophone Sub-Saharan Africa Middle East and North Africa Latin America and the Caribbean South Asia East and Southeast Asia Percent of Maximum Urban Rural Table 5.6. Average Effort Scores by Component and Region, 2002 Figure 5.6. Program Efforts for Three Groups of Countries with Different MMR Levels Chapter 5 37 Summary: National programs to im- prove maternal health vary across a great range, from very low scores to relatively high ones. Differences are less at the re- gional level but are still substantial, de- pending upon the component of effort. However, sheer access to services is seri- ously compromised especially in the ru- ral sector, which puts pregnant women with complications in jeopardy. Separate surveys in 1999 and 2002 show very lit- tle improvement for countries taken as a whole. References Bulatao, R.A. and J.A. Ross. “Rating Maternal and Neonatal Health Services in Developing Countries.” Bulletin of the World Health Organization. 2002; 80:721-727. Hill, K., C. AbouZahr, and T. Wardlaw. “Estimates of Maternal Mortality for 1995.” Bulletin of the World Health Or- ganization. 2001; 79(3):182-198. See also WHO and UNICEF. Revised 1990 Estimates of Maternal Mortality: A New Approach by WHO and UNICEF. April 1996. See also AbouZahr, Carla and Tes- sa Wardlaw, “Maternal Mortality in 2000: Estimates Developed by WHO, UNICEF, and UNFPA.” Geneva: WHO. 2004. Ross, J.A. and J. Begala. “Measures of Strength for Maternal Health Programs in 55 Developing Countries: The MNPI Study. Maternal and Child Health Jour- nal, Vol. 9, No. 1, March 2005. Ross J.A., O.M.R. Campbell, and R.A. Bulatao. “The Maternal and Neonatal Programme Effort Index (MNPI).” Trop- ical Medicine and International Health. 2001; 6(10):787-798. Chapter 4 39 CHILD HEALTH Chapter 6Chapter 6Chapter 6Chapter 6Chapter 6 To provide an overview of child health in developing countries, this chapter takes up three features: ➤ The rates and numbers of child deaths ➤ The risks of death according to birth categories (birth intervals, birth orders, mother’s age) ➤ The status of programs for immuniza- tions, ARI, and ORS. Rates and Numbers of Child Deaths (see Appendix Table A.19) One determinant of child deaths is the number of births, so as birth rates fall the numbers of infant and child deaths either fall or grow more slowly than they other- wise would have. Birth rates have fallen by about half in the last four decades, and the large increases in contraceptive use deserve considerable credit for sav- ings in the numbers of infant and child deaths. Death rates are another matter: the risk to each child born can remain high even while births decline. However there is a selectivity that also works to reduce death rates. When birth rates fall there are fewer high-risk births to older wom- en and fewer short-interval births, and that lowers death rates even more. Latin America has the lowest regional infant mortality rate; at 27 per 1,000 births, it is well below that for the Mid- dle East/North Africa (45) or Asia (50). A doubling of those rates exists in sub- Saharan Africa (104). In the Central Asian Republics the rate is 62, but it is only 16 in the group of Moldova, Ukraine, and Russia (Table 6.1). Within regions countries vary consider- ably, as the Appendix table shows. The lowest IMR figures in Asia are a mere 3 to 7, while the highs are in the 80s and 90s (and 165 in Afghanistan). The range is narrower in Latin America; it starts as low as Asia does but rises only into the 30s, except for Bolivia (53) and Haiti (76). In the Middle East/North Africa the range is also from about 7 to the 30s, plus Iraq (102 est.). Sub-Saharan Afri- ca’s lowest figures are in the 40s, and the highs exceed 150. Of special interest is the ratio of child mortality to infant mortality. Where death rates are high, as in sub-Saharan Africa, the ratio is also high (1.69). When rates fall, the risk at ages 1-4 falls faster than the IMR does, partly because a portion of the IMR is due to birth de- fects, congenital malformations, etc. that do not respond to general environment improvements. Consequently the ratio is lowest in Latin America (1.18), where the two rates are lowest. In between are Asia (1.36) and the Middle East/North Africa (1.26). Widespread improve- ments, over time, lower both child and infant mortality rates, but more so for the child rates. Numbers of deaths are shown in Table 6.1. They total an estimated 11 million per year, equal to the entire population of some countries. The number would have been far more had birth rates not fallen by about half in the last four de- cades, but they remain far too high. Most deaths occur in Asia due to its popula- tion size, but sub-Saharan Africa’s high rates place it second, with four-fifths as many deaths as in Asia. China and India illustrate the interplay of population size, the birth rate, and the mortality rates. China’s larger popula- tion is offset by its low birth rate, result- ing in only 19 million births, well below India’s 25 million. China also has a child mortality rate of only 37 compared to India’s 87. The result is that China expe- riences 692,000 child deaths annually but India experiences 2.2 million. Risks of Death by Birth Categories (see Appendix Table A.20) Infants die at higher rates if they arrive too close to a preceding birth, if the mother is too old or too young, or if the birth is of a high order. DHS surveys have documented these elevated risks for many countries, as listed in the Appen- dix. Illustrative countries appear in Ta- ble 6.2. Figures in the first set of col- umns distribute all births according to a risk calculation: first births are bound to Table 6.1. Number of Births, Infant and Child Mortality Rates, and Number of Deaths, 2003 Estimates Under Age 5 Infant Annual Mortality Annual Annual No. Mortality Infant Rate Child of Births Rate (IMR) Deaths (U5MR) Deaths Asia 70,052,000 50 3,511,012 68 4,735,000 Latin America 11,538,000 27 313,896 32 372,000 Middle East/North Africa 9,657,000 45 431,910 57 547,000 Sub-Saharan Africa 26,813,000 104 2,790,496 176 4,705,000 Central Asia Rep. 1,178,000 65 76,474 80 94,000 Caucasus 229,000 62 14,102 70 16,000 Moldova, Russia, Ukraine 1,684,000 16 27,025 21 36,000 Total/Mean 121,151,000 59 7,164,915 87 10,505,000 40 Chapter 6 occur and are counted separately, and some births do not fall into any of the risk categories just mentioned. However other births fall into one or more catego- ries and are termed “any risk” births. The second set of columns show how many births fall into each type of risk: a short birth interval between the birth and the preceding one (if any), a birth order of 4 or above, and a birth to a mother at either age 35+ or below age 18. (Be- cause many births fall into two or more categories the four numbers total more than 100%.) Finally, the ratio of mortality among “any risk” births to “no risk” births is shown in the last column. (The mortali- ty risk is calculated as deaths among all births occurring in the last five years, so the exposure times vary depending upon how long ago the birth occurred.) The ratios shown vary greatly and are not available for all countries, so no region- al values are given. However the ranges are of interest: Asia about 1.3 to 3.3 Latin America about 1.2 to 3.1 Middle East/N. Africa about 1.5 to 2.0 Sub-Saharan Africa only 1.1 to 1.6 Remarkably, across 31 surveys in sub- Saharan Africa, not a single country has a ratio above 1.6. The apparent reason is that because the base mortality rates are high, even the births that fall into the safe categories carry high risks, not far below the others. That makes it clear that the ratios can decline over time either because the death rates in the risky categories fall, or because the rates in the safe categories rise. If the latter rates fall faster than the former ones do, the ratios will rise. In any case, the trend in a ratio can only be understood by reference to the underly- ing mortality rates in each category, to- gether with the distribution of births across the categories. Trends: When contraceptive use rises the percent of births at high orders falls and that reduces the risk of infant and child deaths. Each line in Figure 6.1 is for one country, showing the decline in the IMR as the percent of high order births fell. In most countries the two fell together. Moreover, the main pattern, in the overall diagonal from upper right to lower left, is consistent: high values go with high values, and low with low. Roughly, a 10% fall in the percent of high order births is accompanied by a 27 point fall in the IMR. Table 6.2. Births According to Risk Category Distribution of All Births Separate Risks by Birth Type* Birth Mortality Ratio, First No-Risk Any-Risk Total Interval Birth Age Below Any-Risk Birth Births Births Births Births <24 no. Order 4+ 35+ Age 18 to No-Risk Birth Asia Bangladesh 1999/2000 13.9 33.1 53.1 100 11.4 28.7 5.7 17.2 1.80 Philippines 1998 22.4 20.7 56.9 100 26.3 37.0 15.0 2.3 1.70 India 1998/99 20.1 29.3 50.7 100 20.3 28.7 4.1 9.1 2.00 Indonesia 2003 30.4 35.6 34.0 100 8.4 20.8 13.5 4.3 1.58 Pakistan 1990/91 14.8 19.1 66.1 100 27.3 50.1 12.9 3.6 2.00 Latin America 100 Brazil 1996 26.2 29.3 44.5 100 18.7 21.5 9.6 8.3 2.00 Colombia 2000 28.1 29.5 42.4 100 16.0 17.8 9.6 8.7 2.00 Haiti 2000 17.7 20.9 61.4 100 20.5 45.3 18.0 5.2 1.30 Mexico 1987 17.7 22.6 59.7 100 25.6 39.4 11.3 6.1 2.40 Peru 2000 24.9 26.8 48.3 100 13.8 30.9 15.2 5.5 1.30 Middle East/North Africa 100 Egypt 2003 26.5 32.5 41.0 100 14.4 26.0 10.8 3.1 1.57 Jordan 2002 18.7 20.0 61.3 100 33.4 43.0 13.9 1.5 1.65 Morocco 1992 16.8 18.9 64.3 100 20.4 51.5 19.7 2.3 1.60 Turkey 1998 29.9 29.9 40.2 100 16.8 22.3 7.1 4.4 2.00 Yemen 1997 11.3 14.1 74.5 100 30.6 59.0 16.2 4.7 1.60 Sub-Saharan Africa 100 Nigeria 2003 13.7 21.2 65.1 100 19.1 46.5 14.2 8.9 1.43 Senegal 1997 12.4 23.3 64.3 100 14.6 52.8 18.4 5.6 1.20 South Africa 1998 26.1 32.1 41.8 100 9.1 26.7 15.0 6.8 1.50 Tanzania 1999 17.2 26.2 56.6 100 12.7 43.0 12.8 6.5 1.10 Uganda 2000/01 11.2 21.8 67.0 100 22.6 49.2 11.5 7.6 1.20 * Because many births fall into two or more categories the four numbers total more than 100%. Chapter 6 41 Immunizations, ARI, and ORS (Appendix Table A.21) UNICEF’S annual reports provide com- prehensive immunization data, and the latest report (UNICEF 2005) contains regional estimates for 2003. The pattern in Figure 6.2 and Table 6.3 is fairly clear. Two regions, South Asia and sub- Saharan Africa, have the farthest to go, while the other four regions perform about equally well. They have attained a level of about 90% for four of the five immunization types shown. However a level of only about 70% has been at- tained for hepatitis (no data for South Asia). Figures for individual countries are in Appendix Table A.21. Regional results for acute respiratory in- fections (ARI) also appear in Table 6.3 for three regions: the Middle East/North Africa, South Asia, and sub-Saharan Af- rica. Only 12%-19% of children are re- ported to have suffered from ARI in the last two weeks, but of those only 39- 69% saw a health provider. Treatment is even more disappointing for children with diarrhea with oral rehydration solu- tions (ORS): in the four regions shown only 25% to 36% receive ORS. (Country figures are in Appendix Table A.21). 0 20 40 60 80 100 120 140 10 20 30 40 50 60 70 Percent of Births at High Orders IM R Figure 6.1. Relation of the Infant Mortality Rate to the Percent of Births at Orders 4+ (Trends Across Multiple Surveys in 36 Countries) Figure 6.2. Percent Coverage for Immunizations, ARI, and ORS, by Region Source: UNICEF. State of the World’s Children 2005, p. 117. Indicator Notes: a Percentage of infants that received three doses of diphtheria, pertussis (whooping cough) and tetanus vaccine. b Percentage of infants that received three doses of hepatitis B vaccine. c Percentage of children (0-4 years) with acute respiratory infection (ARI) in the last two weeks. d Source: DHS, MICS, and other national household surveys. e Percentage of children (0-4 years) with ARI in the last two weeks taken to an appropriate health provider. f Source: DHS, MICS. g Percentage of children (0-4 years) with diarrhea in the last two weeks preceding the survey who received either oral rehydration therapy (oral rehydration solutions or recommended homemade fluids) or increased fluids and continued feeding. 0 10 20 30 40 50 60 70 80 90 100 TB DPT3 Polio3 Measles HepB3 Percent Immunized Central and Eastern Europe East Asia and Pacific Latin America and Caribbean Middle East and North Africa South Asia Sub-Saharan Africa Table 6.3. Percent Coverage for Immunizations, ARI, and ORS, by Region Under Age Five % with Diarrhea % with ARI Receiving Taken to ORS and 1-Year-Olds, % Immunized to Health Continued TB DPT3 Polio3 Measles HepB3 % with ARI Provider Feeding Notes - a - - b c d,e f,g East Asia & Pacific 91 86 87 82 66 - - - Latin America 96 89 91 93 73 - - 36 Middle East/N. Africa 88 87 87 88 71 12 69 - South Asia 82 71 72 67 1 19 57 26 Sub-Saharan Africa 74 60 63 62 30 14 39 32 Central & Eastern Europe 95 88 89 90 81 - - 25 Chapter 4 43 HIV/AIDS PROGRAMS AND SHORTFALLS Chapter 7Chapter 7Chapter 7Chapter 7Chapter 7 Table 7.1. Global Estimates of the HIV/AIDS Epidemic as of the End of 2003 People newly infected with HIV in 2003 Total 4.8 million Adults 4.1 million Children <15 years 630,000 Number of people living with HIV/AIDS Total 37.8 million Adults 35.7 million Children <15 years 2.1 million AIDS deaths in 2003 Total 2.9 million Adults 2.4 million Children <15 years 490,000 Total number of AIDS orphans in sub-Saharan Africa 12 million (AIDS orphans are defined as children under the age of 18 who have lost one or both parents to AIDS.) HIV/AIDS Incidence and Prevalence The HIV/AIDS epidemic has surprised the world in its rapid growth and in its severity. It strikes the young population of working age, mainly in the cities, but in some African countries it pervades the whole society and affects more than one-fourth of all adults. Global estimates. A summary of global estimates is provided by UNAIDS, which publishes an annual status report for HIV/AIDS in all regions. Most of this section is drawn from the 2004 re- port, as is Appendix Table A.13, which provides estimates by country for sever- al items of information. Highlights include the following (Table 7.1): ➤ In some regions AIDS is already the leading cause of death among adults (aged 15-49). ➤ Globally it is among the top ten caus- es of death. At current levels of new HIV infections, it may move into the top five. ➤ About 40 million people are infected with HIV; most will die within a decade although treatment programs to extend life are scaling up rapidly. Almost 3 mil- lion deaths in 2003 were due to AIDS. ➤ New infections are continuously be- ing added: about 4.8 million in 2003 alone. ➤ Over 600,000 children were infected with HIV in 2003, mostly through their mothers before or during birth or through breastfeeding. ➤ The high number of AIDS deaths to adults has created a large number of orphans. Figure 7.1 shows that there are almost a million orphans due to AIDS in five countries (Nigeria, South Africa, Zimbabwe, Tanzania and Uganda) and over 500,000 in four other countries. ➤ Only a tiny fraction of those with HIV know they have it. This disguises the ex- tent of the epidemic, invites denial, and hampers efforts to expand treatment. Growth rates. Countries differ in the patterns by which the HIV virus spreads. The prevalence of HIV infection among adults has stabilized in many countries in sub-Saharan Africa and Latin Ameri- ca but is still growing rapidly in parts of Eastern Europe and Asia. There are exceptions such as Thailand, Uganda, and Kenya that have experi- enced declines in HIV prevalence. The highest growth rates can be high in- deed, even where prevalence already falls within a high range. Figure 7.2 shows the 14-year HIV trend among pregnant women in parts of South Africa. The epidemic is most severe in southern Africa (Figure 7.3), where Namibia, South Africa, and Zimbabwe have infec- tion levels of 20-25% of all adults and Lesotho, Botswana, and Swaziland are even higher at about one-third or more of all adults aged 15-49. Nigeria South Africa Zimbabwe Tanzania Uganda Ethiopia Kenya Zambia Malawi Mozambique Côte d’Ivoire Burkina Faso Cameroon Burundi Ghana Rwanda Botswana Central African Republic Angola Lesotho Figure 7.1. Twenty African Countries with the Highest Number of Orphans Due to AIDS, 2003 0 200,000 400,000 600,000 800,000 1,000,000 1,200,000 1,400,000 1,600,000 1,800,000 2,000,000 Number of AIDS Orphans 44 Chapter 7 In Asia the pattern is different: levels are low in general, but population sizes are greater and small changes in rates can produce very large numbers of new cas- es (Table 7.2). Information is sparse for much of Asia but China’s prevalence probably doubled in recent years to over 800,000 cases. India has more infected people than any other country except South Africa, but this is still less than 1% of all adults. Figure 7.4 shows South Africa and India as first among the 20 countries with the largest absolute num- bers of HIV/AIDS cases. These reflect a balance between the country’s size and the percent affected. Ten countries ap- pear in both Figures 7.3 and 7.4 with the unfortunate distinction of having both very large numbers of cases and very high rates. These are Cameroon, Côte d’Ivoire, Kenya, Malawi, Mozambique, Nigeria, South Africa, Tanzania, Zam- bia, and Zimbabwe. Table 7.2. Adults and Children Living with HIV/AIDS by Region North Africa and Middle East 480,000 Sub-Saharan Africa 25,000,000 Caribbean 430,000 Latin America 1,600,000 East Asia and Pacific 900,000 South and Southeast Asia 6,500,000 Australia and New Zealand 32,000 North America 1,000,000 Western Europe 580,000 Eastern Europe and Central Asia 1,300,000 Total 37,822,000 Prevalence remains low (0.1% or less) in some large Asian countries: China, Indo- nesia, Pakistan, Philippines, and Sri Lanka. However, prevalence is notably higher in Myanmar, Thailand, and espe- cially Cambodia. In Latin America the picture is mixed. Adult prevalence is estimated at less than 1% in most countries, but overall some 1.6 million cases exist. Infection levels are quite high in certain popula- tions, especially injecting drug users, men who have sex with men, and com- mercial sex workers. In 1986 in Brazil women constituted one in 17 AIDS cas- es; now it is one in three. For HIV prev- alence in Latin America as a whole, about one-third of cases are women. Figure 7.2. HIV Prevalence Among Pregnant Women, Selected Provinces, South Africa, 1990-2003 Source: Department of Health, South Africa, various surveillance reports. Figure 7.3. Twenty Countries with the Highest Adult HIV/ AIDS Prevalence Levels, 2003 Figure 7.4. Twenty Countries with the Highest Number of People Living with HIV/AIDS (Adults and Children), 2003 South Africa India* Nigeria Zimbabwe Tanzania Ethiopia Mozambique Congo, D.R.* Kenya Zambia Malawi Russian Federation China Brazil Côte d’Ivoire Thailand Cameroon Uganda Sudan Ukraine Swaziland Botswana Lesotho Zimbabwe South Africa Namibia Zambia Malawi Central African Rep. Mozambique Tanzania Gabon Côte d’Ivoire Cameroon Kenya Burundi Liberia Haiti Nigeria Rwanda 0 1,000,000 2,000,000 3,000,000 4,000,000 5,000,000 6,000,000 Number of Adults and Children* 2002 Estimates 0 5 10 15 20 25 30 35 40 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 H IV P re va le nc e (% ) KwaZulu Natal Gauteng Free State Eastern Cape 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0 Prevalence of Adults (15-49) Chapter 7 45 Because contraceptive use is widespread in Latin America fewer women become pregnant, which reduces the absolute number of mother-child transmissions; consequently there are fewer infected in- fants and fewer orphans when women die. In North Africa and the Middle East, the smallest region, data are very thin, but Sudan is the only country that estimates its adult HIV prevalence at above 1%. Only about 443,000 cases are estimated to exist (Table 7.2), less than 1% of the world’s total. Eastern Europe has some of the fastest- growing epidemics. Russia ranks twelfth in terms of countries with the most peo- ple living with HIV/AIDS (Figure 7.4). The epidemic there and elsewhere in the region is fueled by transmission among injecting drug users who share injection equipment. The HIV/AIDS epidemic has also contributed to an exploding tuber- culosis epidemic that is producing new strains of TB resistent to many of the drugs used to treat TB. Methods. Figures on prevalence and mortality are subject to much error. Prevalence estimates in sub-Saharan Af- rica come chiefly from women attending antenatal clinics and from some nation- al surveys. In the other regions estimates are based on surveillance among popula- tions at high risk. It is not enough just to measure current HIV prevalence, since it reflects the three different components of the recent flow of new cases, the in- herited bulk of cases from the past, and AIDS deaths. This can confuse any prognosis of growth. In a mature epi- demic prevalence may be stable, but this stability simply means that the number of new infections every year equals the number of people dying from AIDS each year. When prevalence stabilizes at a high level, such as 20-25% in Zimbabwe and South Africa, it means that about 2% of adults are dying each year from AIDS and another 2% are newly infected each year. Goals and Strategies The international community has adopt- ed several goals to guide the fight against HIV/AIDS. In 2001 the United Nations General Assembly Special Ses- sion on AIDS (UNGASS) developed a consensus on a strategy for addressing the epidemic. Other United Nations or- ganizations and meetings have elaborat- ed on particular parts of the strategy and many donors have developed their own targets. Among the key goals are: ➤ Reduce HIV prevalence among young people by 25% by 2005 in the most af- fected countries and by 2010 everywhere (UNGASS). ➤ Reduce infections due to mother-to- child transmission by 20% by 2005 and 50% by 2010 (UNGASS). ➤ Expand anti-retroviral treatment pro- grams to reach 3 million people by the end of 2005 (WHO). ➤ Halt and begin to reverse the HIV/ AIDS epidemic by 2015 (Millennium Development Goal). ➤ Provide anti-retroviral treatment to 2 million people by 2008, avert 7 million new HIV infections by 2010, and pro- vide care and support to 10 million peo- ple living with HIV/AIDS and orphans made vulnerable by HIV/AIDS by 2008 (U.S. President’s Emergency Plan for AIDS Relief). Comprehensive prevention programs have focused on mass media, condom promotion, control of sexually transmit- ted infections, voluntary counseling and testing, blood safety, school-based AIDS education, and outreach programs for sex workers, men who have sex with men, and injecting drug users. Condom distribution has increased dramatically in many countries, expanding to 50-100 million condoms per year in countries such as Ethiopia, Kenya, Nigeria, and Zimbabwe; however, even more are needed to cover all risky sex acts (Ap- pendix Table A.23). Recently more attention has been fo- cused on programs to promote absti- nence and reduce the number of people with multiple partners. Programs to pre- vent mother-to-child transmission are expanding rapidly in many countries. There have been some successes. Ugan- da and Thailand were able to dramatical- ly reduce prevalence during the 1990s. Senegal took early action to keep preva- lence at low levels. Recently prevalence has begun to decline in Kenya. Latin America has led the way in ex- panding access to anti-retroviral treat- ment with many countries there provid- ing universal access. In the last several years a new emphasis on treatment has led to a rapid expansion of the numbers of people receiving anti-retroviral thera- py in other regions as well, to over 700,000 by the end of 2004. This figure is expected to grow to 3 million in the near future. The expansion of treatment will mean better survival for millions of HIV-positive people. Efforts are also being made to address the consequences of the epidemic. Countries in sub-Saharan Africa are attempting to expand programs to support the millions of orphans and other children made vul- nerable by HIV/AIDS. Efforts are being made to address stigma and discrimina- tion and to improve the legal, cultural, and economic practices that increase vul- nerability for women, although progress has been slow. Strategies to control the epidemic differ according to the type of epidemic. In countries with low-level epidemics (prevalence less than 5% in all popula- tion groups) and concentrated epidemics (prevalence above 5% in some popula- tion groups but less than 1% in pregnant women), prevention efforts need to focus on those populations with the highest risks where most new infections are oc- curring. This includes injecting drug us- ers, men who have sex with men, and sex workers and their clients. Interven- tion now to stop transmission in these populations can prevent much larger ep- idemics in the future. In generalized ep- idemics (prevalence above 1% in preg- nant women) efforts need to be directed to all segments of the population. Ex- panding treatment access to meet the 46 Chapter 7 large numbers needing treatment re- quires not only more funding but also increased capacity to deliver drugs and manage patients on long-term care. Ex- panded access to treatment will not only save lives but may also help to improve the environment for prevention. Some prevention programs have expand- ed at rapid rates in the past few years. A survey of the coverage of prevention and care services in 2003 produced by the 0 500,000 1,000,000 1,500,000 2,000,000 2,500,000 3,000,000 3,500,000 2001 2003 0 1,000,000 2,000,000 3,000,000 4,000,000 5,000,000 6,000,000 2001 2003 0 50,000 100,000 150,000 200,000 250,000 300,000 350,000 2001 2003 0 10,000,000 20,000,000 30,000,000 40,000,000 50,000,000 2001 2003 Figure 7.6. The AIDS Program Effort Index by Region, 2003 Number of people receiving voluntary counseling and testing (VCT) Number of pregnant women offered services to prevent mother to-child transmission (PMTCT) Number of PLHA receiving anti-retroviral therapy (ART) Number of secondary school students receiving AIDS education Figure 7.5. Number of People Receiving Services in 2001 and 2003 for Countries Reporting in Both Surveys for VCT, PMTCT, ART, and AIDS Education - 10 20 30 40 50 60 70 80 90 100 Po litic al su pp or t Po licy an d pla nn ing Or ga niz ati on al str uc tur e Pro gra m re so ur ce s Ev al, m on , re se ar ch Le ga l an d re gu lat ory Hu m an rig hts Pre ve nti on pro gra m s Ca re pro gra m s Mi tiga tio n Eastern and Southern Africa Western and Central Africa Asia Latin America and Caribbean Eastern Europe POLICY Project (2003a) shows large in- creases in the past two years in the num- ber of people receiving voluntary coun- seling and testing, treatment to prevent mother-to-child transmission of HIV, anti-retroviral therapy and school-based AIDS education (Figure 7.5). Coverage of the population in need is still low for these services, however, and in other ar- eas the need is even greater, such as ef- fective programs for injecting drug users and men who have sex with men. National governments, bilateral and in- ternational donors, and private founda- tions have significantly increased the resources devoted to HIV/AIDS pro- grams to US$6 billion in 2004, but an estimated US$10–20 billion will be needed annually in the coming years to successfully address all the prevention, care and treatment, and mitigation needs. Although efforts to control the spread of AIDS and address its consequences have improved markedly in the past 5 years, much remains to be done. Figure 7.6 presents a profile of effort showing where efforts are strongest and where they are weakest. The AIDS Program Ef- fort Index (API) measures overall effort in 10 categories through interviews with national experts (POLICY Project 2003b). The 2003 round of the API, con- ducted in 54 countries, shows that most countries have done well in providing political support and the policy and planning required to address the epidem- ic. However, effort lags far behind in mobilizing and using financial resources and in protecting the human rights of people living with HIV/AIDS. References UNAIDS. Report on the Global HIV/ AIDS Epidemic, June 2004. Issued by UNAIDS/WHO, Joint United Nations Programme on HIV/AIDS (UNAIDS). World Health Organization. 2004. UNAIDS. Children on the Brink 2004. A Joint Report of New Orphan Estimates and a Framework for Action. UNAIDS, UNICEF, USAID. July 2004. POLICY Project 2003a. “Coverage of Selected Services for HIV/AIDS Preven- tion, Care and Support in Low- and Mid- dle-Income Countries in 2003.” USAID, UNAIDS, WHO, UNICEF, POLICY Project. 2003. POLICY Project 2003b. “The Level of Effort in the National Response to HIV/ AIDS: The AIDS Program Effort Index (API), 2003 Round.” USAID, UNAIDS, WHO, and the POLICY Project. Dec. 2003. Chapter 8 47 Introduction This chapter discusses five program ob- jectives: 1. To provide full access to a variety of contraceptive methods 2. To satisfy unmet need and intention to use a method 3. To reach the desired fertility level 4. To attain the replacement fertility level 5. To satisfy the Millennium Develop- ment Goals and the Cairo Programme of Action. These objectives have been stressed in various national plans or in the interna- tional discourse concerning the proper goals of action programs. The first objec- tive is the aim of providing good access to a full range of contraceptive methods to the population; this may be regarded as a necessary condition to attaining other objectives, and so it logically comes first. The second is meant to ad- dress unmet need and the level of ex- pressed intention to use contraception. The third overlaps with that; it is to assist couples in reaching the desired level of fertility. The fourth is to move toward the replacement fertility level, an objective that is present in numerous national plans. The fifth concerns both the Mil- lennium Development Goals and the Cairo Programme of Action: two sets of objectives recognized by the interna- tional community, that match to a great extent. Goal No. 1: To Provide Full Access to a Variety of Contraceptive Methods The Cairo ICPD meeting stressed the goal of providing full availability to family planning methods. The Pro- gramme of Action declared: “All countries should take steps to meet the family planning needs of their popu- lations as soon as possible and should, in all cases by the year 2015, seek to pro- vide universal access to a full range of safe and reliable family planning meth- ods.” And they should: “Recognize that appropriate methods for couples and individuals vary according to their age, parity, family-size preference and other factors, and ensure that women and men have information and access to the widest possible range of safe and ef- fective family planning methods.” However there is a long way to go before most couples are given a true choice of alternative methods. In this section we use selected scores from past cycles of the International Family Planning Effort Study (Ross and Stover, 2001). (See Ap- pendix Table A.14.) Of the 30 scores in the study, five concern the availability of contraceptive methods to the population. Country experts estimate what propor- tion of the population has ready access to each method – pill, IUD, male steril- ization, female sterilization, and con- dom. This is explicitly not the propor- tion currently using the method, but rather the proportion who have reason- able access to it. Table 8.1 and Figure 8.1 summarize the combinations of methods and the avail- ability of individual methods. The rule employed is that at least half of the pop- ulation must have access to a method for it to be considered available. For exam- ple, the pill is considered available by this rule in 83% of 88 countries (Table 8.1, top panel) and the IUD in 63%. Male sterilization is available in only 25% of countries while condoms are available in 85%. Our focus however is not upon single- method access but upon something clos- er to “full availability of contraception,” Table 8.1. Percent of Countries Making Contraceptive Methods and Combinations of Methods Available as of 1999 Sub- Latin Middle East/ Saharan Eastern All Asia America North Africa Africa Europe Countries Pill 94.4 100.0 90.9 61.3 80.0 83.0 IUD 77.8 82.6 90.9 22.6 100.0 62.5 Female sterilization 72.2 60.9 45.5 19.4 20.0 44.3 Male sterilization 55.6 30.4 18.2 6.5 20.0 25.0 Condom 88.9 91.3 90.9 77.4 80.0 85.2 Pill and IUD 77.8 82.6 90.9 22.6 80.0 61.4 Pill and female sterilization 72.2 60.9 45.5 19.4 20.0 44.3 IUD and female sterilization 72.2 60.9 45.5 16.1 20.0 43.2 Pill, IUD and female sterilization 72.2 60.9 45.5 16.1 20.0 43.2 Pill, IUD, female sterilization, condom 72.2 60.9 45.5 16.1 20.0 43.2 At least one long-term method 77.8 82.6 90.9 25.8 100.0 63.6 At least one short-term method 94.4 100.0 90.9 77.4 100.0 89.8 At least one long-term and at least one short-term method 77.8 82.6 90.9 25.8 100.0 63.6 No. of countries 18 23 11 31 5 88 Chapter 8Chapter 8Chapter 8Chapter 8Chapter 8 FIVE PROGRAM OBJECTIVES 48 Chapter 8 the topic of this section. That leads to the question of how many countries provide multiple methods. Therefore the second panel of Table 8.1 gives the percent of countries where both pill and IUD availability meet the 50% rule – similarly for each other combina- tion shown. Finally, the bottom panel al- lows for some flexibility in the provision of methods: in the first row a country qualifies if any long-term method ex- ceeds 50% availability – either male or female sterilization or the IUD. Numer- ous countries offer either the IUD or fe- male sterilization, which produces the relatively high 64% in the last column. In the second row a country qualifies if any short-term (resupply) method meets the 50% rule, either the pill or condom. Again, many countries offer at least one, so 90% qualify in the last column. The last line combines the two previous lines: a country qualifies only if it meets both criteria, which reduces the figure to 64%. The numbers necessarily decline as more stringent conditions are applied. Eighty-three percent of countries qualify for the pill alone, but only 61% qualify for the pill and IUD together, and only 43% for those plus female sterilization. Thus less than one-half of all countries provide those three methods to at least one-half of the population – a far cry from the ICPD goal. Even those may not do so uniformly throughout the country: Figure 8.1. Availability of Multiple Contraceptive Methods certain areas may have most access to just the pill and condom, and other areas only to the pill and sterilization. Regions differ in the combinations of methods they provide. The Middle East/ North Africa does poorly on combina- tions that include sterilization, whereas Asia and Latin America do considerably better. However the Middle East/North Africa improves in the bottom panel, where flexibility is allowed as to which methods qualify. Overall, Francophone sub-Saharan Afri- ca has the least method availability. By the Cairo ICPD mandate, it is farthest from the goal of “full availability.” It does best for condoms (top panel) but poorly for all other methods, which hurts its ratings in the combinations. The eas- iest route to an enlarged choice of meth- ods in the short term would be to add the pill to numerous outlets in both public and private sectors. However long-term methods are also needed both for auto- matic continuation and reliability. Distance to go to “full availability.” Observers have long noted the adjust- ments needed in certain country pro- grams. India and Nepal have a near-ex- clusive stress on sterilization, which calls for a better balance with temporary methods. Conversely, Indonesia’s cau- tion regarding sterilization has left many couples with unsatisfactory alternatives, and this occurs in Egypt as well. Viet Nam’s preoccupation with the IUD alone has driven many couples to high- failure temporary methods and to exces- sive numbers of abortions. The tendency in numerous countries toward only one or two methods appears in detail in the method mix data in Chapter 2. A different kind of problem prevails in some Francophone countries where no method at all is widely available to the mass of the population, as in Chad, Mali, Mauritania, and both Congos. This is equally true in some Anglophone coun- tries, such as Ethiopia and Nigeria. All these and others face the elementary need to deploy services to most of the population. Moreover, physical avail- ability of several contraceptive methods needs to be accompanied by services that are convenient and congenial to po- tential clients. Thus the overall picture is quite bleak for public access to a variety of contra- ceptive choices. The rule used here, that a method is “available” if it is readily accessible to at least half of the popula- tion, is a lenient one. Yet less than two- thirds of countries provide at least one long-term method and one short-term method to half of the population. Nevertheless there has been improve- ment over time. Figure 8.2 and Table 8.2 display the average availability scores from the international family planning effort studies, for male and fe- male sterilization, IUD, pill, and con- dom (each score is the percent of maxi- mum*). The figure is based on the 69 countries included in all years, which clarifies the time trend, but data are also shown in Table 8.2 for the 109 countries that were in at least one study. The general regional pattern is one of in- creasing availability, except for Latin America and Asia after 1989 when they were already at relatively high levels. The Middle East/North African region *The original scores ranged from zero to four, where four was any percentage over 80. Therefore the “maximum” here is best regarded as about 85%, so that a score of 60% in the table or figure is in fact about 51%. As a result, the figures shown are somewhat elevated. 0 10 20 30 40 50 60 70 80 90 100 Pill and IUD Pill and F ster IUD and F ster Pill, IUD, F ster Pill, IUD, F ster, condom At least one long-term method At least one short-term method At least one long-term and at least one short- term method Pe re nt o f C o u n tri e s R ea ch in g 50 % o f P o pu la tio n Chapter 8 49 Table 8.2. Time Trend of the Average Availability Score, 1982-1999, from Family Planning Effort Studies 69 Countries Included in All Studies No. of 1982 1989 1994 1999 Countries Latin America 54.6 69.4 69.6 71.2 19 Asia 59.8 66.7 64.4 70.3 16 Anglophone SSA 12.7 33.2 44.9 48.0 12 Francophone SSA 6.7 19.9 21.2 24.6 12 Middle East/N. Africa 26.7 50.0 61.9 64.6 10 All regions 35.8 50.8 54.6 57.9 69 109 Countries in One or More Studies No. of 1982 1989 1994 1999 Countries Latin America 54.8 68.6 66.0 69.9 24 Asia 56.9 59.2 63.0 71.3 24 Anglophone SSA 13.6 33.5 50.1 50.3 19 Francophone SSA 6.6 20.4 23.6 29.7 21 Middle East/N. Africa 18.2 36.4 52.5 61.2 16 Central Asia Rep. - - 17.7 52.7 5 All regions 32.5 44.6 51.9 57.5 109 has improved noticeably over the years. Anglophone sub-Saharan Africa has also improved, though in 1999 it was still below the 50% mark, and the bottom re- gion, Francophone sub-Saharan Africa, improved very little after 1989, persist- ing at a low level. The five Central Asian Republics (see table) improved marked- ly from 1994 to 1999 in making contra- ceptives available. The leveling off of the curves at about 70%, and the slowing of improvement by the Middle East/North Africa as it ap- proaches that level, suggest a kind of practical ceiling to average availability. Average availability is depressed partly by the inclusion of male sterilization, which still suffers from poor access in many countries. Access improvements have been selective, and have been most impressive for the pill and condom. References Ross, John. “The Question of Access.” Studies in Family Planning 26(4):241- 242. 1995. Ross, John, and W. Parker Mauldin. “Family Planning Programs: Efforts and Results, 1972-94.” Studies in Family Planning 27(3):137-147. 1996. Ross, John and John Stover. “The Fami- ly Planning Program Effort Index: 1999 Cycle.” International Family Planning Perspectives 27(3):119-129. Sept. 2001. United Nations Population Fund. Pro- gramme of Action: Adopted at the Inter- national Conference on Population and Development. Cairo, September 5-13, 1994. Pages 51, 53. Booklet published 1996. Wilkinson, Marilyn I., Wamucii Njogu, and Noureddine Abderrahim. The Avail- ability of Family Planning and Maternal and Child Health Services. DHS Com- parative Studies No. 7. Calverton, Mary- land: Macro International. 1993. Figure 8.2. Mean Availability Score for Five Contraceptive Methods, 69 Countries in All 4 Studies (Percent of Maximum Score) 0 10 20 30 40 50 60 70 80 1982 1989 1994 1999 M ea n Sc o re Latin America Asia Anglophone SSA Francophone SSA Middle East/N. Africa All regions 50 Chapter 8 Goal No. 2: To Satisfy Unmet Need and Intention to Use a Method Besides serving as one indicator of the public’s need for contraceptive assis- tance, unmet need can be supplemented by information on a woman’s own ex- pressed intention to use a method. Here we provide first the unmet need perspec- tive, and then additional information on intention to use. The unmet need concept has been useful through the years as a humane rationale for action programs and as evidence of a large subgroup in nearly every popula- tion whose needs have not yet been ad- dressed. As a counterpoint to target-driv- en approaches it has helped ease interna- tional opposition to family planning, partly by demonstrating that satisfying unmet need in many populations would raise contraceptive prevalence as much as meeting the targets would. That helps justify the discontinuance of worker tar- gets (Sinding et al., 1994) and in princi- ple can release workers from general re- cruitment efforts and let them focus on simply helping the truly interested cou- ples. Unmet need therefore has served as one of the considerations for program plan- ning, as it was in the 1994 Cairo ICPD meeting. At both international and re- gional levels it has been an important rationale for justifying donor funding and for winning the support of a broad spectrum of interest groups. This is true also within some individual countries, where it serves to help gauge the inter- ested market for family planning. For planning purposes however, where sur- vey data permit, the unmet need esti- mates should be reduced for women who do not intend to use, but increased to recognize omitted couples who intend to use a method. Unmet need reflects the puzzling gap between the desire to avoid pregnancy and the failure to use contraception. This gap changes in size during the transition from very low prevalence of contracep- tive use, as in Ethiopia for example, to very high prevalence, as in Thailand or Colombia over time. Unmet need starts small, since the desired family size is large, and ends small, since nearly ev- eryone is using a method. In between, unmet need tends to be rather large, since usually there is a serious lag in supplies and services to address the pub- lic’s growing desire to avoid unwanted pregnancies. Therefore the time trend in unmet need can disguise program im- provements if they are outrun by a rapid decline in desired family size. The data used here are based primarily upon the DHS definition of unmet need* since that is available for numer- ous countries. However the figures would be higher if the definition were expanded to sexually active single men and women, dissatisfied users, tradition- al method users, and amenorrheic wom- en close to the return of menses. In Viet Nam for example unmet need rises from 14% to 36% if traditional method users (who have a high abortion rate for fail- ures) are included (Phai et al., 1996). How rapidly can unmet need be erased? Programs work best by satisfying the in- terest that already exists, as good in it- self and as the best way to enlarge that interest. International experience (Ta- ble 8.3) indicates that an annual rise of about two points in prevalence is as much as can be expected, unless the pro- gram is exceptionally strong and the public is especially ready. A 2% rise per year in Ethiopia would require ten years for unmet need to fall from 35% to 15%, which is the current level in Bangladesh. Historically however, unmet need and prevalence have often not moved togeth- er. For example, in Kenya prevalence rose by 1.5 points annually from 1989 to 1993 and by 1.3 points annually from 1993 to 1998. However unmet need fell only slightly from 1989 to 1993, from 38.0 to 35.5, since the desire to avoid pregnancy changed nearly as fast as the prevalence level did. The ICPD directive to reduce unmet need translates in practice to a rise in contraceptive prevalence, in programs that focus on women or couples who are genuinely interested in postponing preg- nancy and in using contraception. Un- met need may rise during an intermedi- ate stage but it finally diminishes as con- traceptive prevalence increases to a high level. Meanwhile there is movement in and out of the pool of users and the var- ious unmet need categories. Trends in unmet need. In some coun- tries, most of them with medium to high prevalence, unmet need has declined over time, as illustrated by 15 of the 18 countries in Figure 8.3. It shows unmet need, contraceptive prevalence, and total demand (the sum of the other two) at two dates. In nearly all cases prevalence has risen and need has decreased. De- mand varies from 33% in Nigeria to over 80% in Colombia. Data covering most surveys over time confirms that unmet need has in general been declining as contraceptive use has risen. The proportion of total demand that is accounted for by contraceptive use is a useful measure; it is simply use divided by the sum of use plus need. Table 8.4 shows this measure by wealth quintiles; it documents clearly that persons in the bottom quintiles have the least part of their demand met by contraceptive use. The lowest figures are only 12.9% and 23.7% in the lowest quintile in Africa. Across the quintiles, each next wealthier group has more demand satisfied. That holds true in every region and in both time periods. A favorable feature in the table is that all quintiles and regions show improvement from the earlier to the later period. This general pattern is echoed in Figure 8.3. When the desired family size has fallen quickly, before contraceptive use has caught up, unmet need is large as a proportion of total demand. This pattern is visible in the demand range between 40% and 55%, as in Benin, Tanzania, *Women with unmet need are those who are mar- ried/cohabiting, fecund, not using a method, and wish to postpone birth at least two years. Women who are pregnant or amenorrheic have unmet need if they did not want the current pregnancy or recent birth either at that time or at all, but if a contracep- tive failure was responsible the woman is treated as having no unmet need. Chapter 8 51 Prevalence Circa 1990 Less than 1.0 1.0-1.9 2.0 or More Less than Rwanda (0.55) Niger 1.09 15 percent Angola (0.38) Gambia (0.22) Eritrea - Iraq - Afghanistan 0.12 Mali 0.24 Sudan 0.37 Ethiopia 0.38 Mauritania 0.47 Burkina Faso 0.67 Guinea 0.70 Chad 0.79 Senegal 0.79 Mean 0.24 15-34 Madagascar 0.32 Nigeria 1.03 Oman 2.16 percent Benin 0.44 Togo 1.13 Cambodia 2.26 Cameroon 0.52 Comoros 1.18 Yemen 2.27 Burundi 0.54 Uganda 1.49 D.R. Congo 2.37 Ghana 0.64 Pakistan 1.56 Central African Rep. 2.62 Swaziland 0.65 Haiti 1.59 Myanmar 3.18 Côte d'Ivoire 0.69 Zambia 1.62 Lesotho 0.96 Tanzania 1.89 Malawi 1.91 Laos 1.94 Mean 0.59 Mean 1.53 Mean 2.48 35-49 Trinidad and Tobago (1.12) Syria 1.09 percent Dominica 0.13 Philippines 1.14 Guyana 0.24 Kenya 1.31 Botswana 0.62 Guatemala 1.36 Saint Lucia 0.66 Nepal 1.62 India 0.75 Qatar 0.99 Mean 0.32 Mean 1.30 50-64 Turkey 0.14 Malaysia 1.03 Paraguay 2.00 percent Azerbaijan 0.30 Zimbabwe 1.05 Bolivia 2.01 Lebanon 0.32 Indonesia 1.11 Algeria 2.18 South Africa 0.66 Barbados 1.22 Grenada 2.27 Dominican Rep. 0.96 El Salvador 1.27 Morocco 2.57 Bahrain 1.40 Egypt 1.50 Honduras 1.51 Tunisia 1.70 Kuwait 1.73 Bangladesh 1.84 Jordan 1.95 Mean 0.48 Mean 1.44 Mean 2.21 65+ Mongolia (0.19) Colombia 1.08 Nicaragua 2.10 percent Mauritius (0.10) Brazil 1.09 Iran 2.37 Thailand 0.02 Jamaica 1.23 Uzbekistan 2.90 Cuba 0.25 Mexico 1.34 Rep. of Korea 0.49 Ecuador 1.39 Puerto Rico 0.52 Peru 1.43 China 0.62 Viet Nam 1.71 Sri Lanka 0.68 Kazakhstan 1.90 Costa Rica 0.81 Mean 0.35 Mean 1.40 Mean 2.46 Source: United Nations Population Division, "World Contraceptive Use 2003" Wall Chart. Table 8.3. Annual Percentage-Point Increase in Contraceptive Prevalence Around 1990-2000, by Prevalence at the Start of the Period 52 Chapter 8 Uganda, Malawi, and Guatemala. Note that while unmet need is large, its share of total demand declines over time as the prevalence bars grow more than the un- met need ones do. Another way to approach the interaction between prevalence and unmet need is by calculating the annual increase in prevalence as a percent of need at the time of the first survey. That is, how much of the initial unmet need is trans- lated into prevalence increases during the ensuing years? In Figure 8.4 the range is from 1% to 11% around an av- erage of about 5%. That suggests that in the middle range of prevalence, half of the initial unmet need could be satisfied over a decade. There is of course consid- erable circulation of individuals in and out of using statuses and unmet need cat- egories over time, but the net changes can be positive. Global estimates of need. Surveys pro- vide estimates of unmet need for many countries (Appendix Table A.9). By as- signing to unknown countries the region- al averages of the known countries we can obtain a crude picture of unmet need for most of the developing world. Table 8.5 presents the results: because unmet need has declined somewhat as contra- ceptive prevalence has risen in the devel- oping world, the overall level of unmet need for married women is about 17%, and about 13% for all women. The num- bers in need remain large due to popula- tion growth: 105 million for married women and another 8 million for unmar- ried women.* Most live in Asia; large numbers are also in sub-Saharan Africa, with smaller numbers in Latin America and Middle East/North Africa. Needs for spacing and limiting are about evenly balanced at 8%-9% each except in sub- Saharan Africa, where the ratio is about two to one. *Including Russia, Eastern Europe, the Caucasus, and the Baltic Republics another 9.1 million wom- en (both married and unmarried) are added, for a total of 122 million. 0 10 20 30 40 50 60 70 80 90 Nigeria 1990 Cameroon 1991 Benin 1996 Tanzania 1992 Uganda 1995 Malawi 1992 Guatemala 1995 India 1992.5 Zimbabwe 1994 Indonesia 1991 Egypt 1992 Jordan 1990 Bangladesh 1993.5 Turkey 1993 Bolivia 1994 Peru 1992 Dominican Rep. 1991 Colombia 1990 Percent Unmet Need Prevalence Figure 8.3. Trends in Unmet Need, Prevalence, and Demand Q1 Q2 Q3 Q4 Q5 All Quintiles Africa 1990-1995 12.9 16.0 17.7 24.7 36.9 22.9 1996-2000 23.7 25.2 29.1 37.2 48.0 33.9 Latin America 1990-1995 29.4 40.7 49.0 57.4 65.3 50.0 1996-2000 38.1 48.8 57.3 61.7 68.0 56.4 Asia 1990-1995 48.2 51.0 57.5 57.5 67.6 57.1 1996-2000 50.0 57.1 59.8 63.7 70.7 60.9 N. Africa/W. Asia 1990-1995 36.8 45.7 53.7 58.2 65.4 53.8 1996-2000 49.7 59.0 62.0 67.5 70.5 62.7 Global Average 1990-1995 29.2 35.8 40.4 46.2 55.6 42.7 1996-2000 37.9 44.4 49.3 54.7 62.1 50.8 Source: Bernstein, 2004. Table 8.4. Proportions of Demand Satisfied by Modern Contraceptive Use, by Wealth Quintile (Q1: poorest) (unweighted averages for selected countries within each region) Chapter 8 53 India has by far the most women with unmet need (Figure 8.5): some 32 mil- lion, accounting for 32% of all need in the developing world (China is assumed to have none by the DHS definition). The next country, Pakistan, has 7 million in need, for 8% of the total. Half of all couples in need live in the top five coun- tries, including Bangladesh, Indonesia, and Nigeria. Two-thirds of all need is in only 11 countries, showing the sharp geographic concentration of unmet need. Youth in need. The two age groups, 15- 19 and 20-24, account for 33% of all un- met need among married women, or 34.9 million women. The 20-24 age group contains twice the number in need as in the 15-19 age group (23.5 vs. 11.4 million). Sub-Saharan Africa has the highest proportion with unmet need – about one in four for both youth and for women of all ages. However, the other regions show differences: young women have more unmet need than do all wom- en by a considerable margin in Latin America (22% vs. 14%), in Asia (23% vs. 16%), and in the Central Asian Re- publics (16% vs. 11%), but by rather lit- tle in the Middle East and North Africa (18% vs. 16%). Details are in Chapter 4; see the “Youth” section. Intention to use a method. The unmet need perspective can be adjusted by in- formation on women’s own statements as to their intention to use contraception. Table 8.6 cross-classifies intention and need for 14 countries selected from Ap- pendix Table A.11 (it contains data from 78 surveys in 48 countries). The figure shows both sides of the adjustment: some of those with unmet need do not intend to use a method, but others with- out apparent need do plan to use. In Ken- ya 28% of nonusers (right-hand column) intend to use a method even though they are classified as having no unmet need – the same women represent 17% of all married women (left-hand column). They more than balance out the smaller group that has unmet need but plans not to use (10% on the right and 6% on the left). Among non-users in Table 8.6, from 15% to 47% of women lacking unmet need by the DHS definition still intend Figure 8.4. Percent of Unmet Need Converted to Annual Prevalence Increases Table 8.5. Number (000s) and Percent of Women with an Unmet Need for Contraception, by Region and Marital Status, 2000 Married Women Unmarried All Women All Spacing Limiting Women Numbers Developing world 113,647 105,205 55,402 49,803 8,442 Asia (ex. China) 63,650 61,142 31,658 29,484 2,508 Sub-Saharan Africa 27,997 23,550 15,269 8,281 4,447 Latin America 11,837 11,088 4,615 6,473 749 Middle East/North Africa 8,925 8,306 3,345 4,961 619 Central Asia 1,238 1,119 515 604 119 Percents Developing world 13.0 17.1 9.0 8.1 3.2 Asia (ex. China) 12.9 16.4 8.5 7.9 2.0 Sub-Saharan Africa 19.4 24.2 15.7 8.5 9.5 Latin America 8.5 13.7 5.7 8.0 1.3 Middle East/North Africa 10.6 15.6 6.3 9.3 2.0 Central Asia 8.5 11.4 5.2 6.2 2.6 Figure 8.5. Number of Married Women with Unmet Need: Top 20 Countries 5,000,000 10,000,000 15,000,000 20,000,000 25,000,000 30,000,000 35,000,000 Ind ia Pa kis tan Ba ng lad es h Nig er ia Ind on es ia Eth iop ia Ira n Ph ilip pin es Th ail an d Co ng o D .R. My an m ar Ne pa l Eg yp t Tu rke y Ug an da Ta nz an ia Ke ny a Ar ge nti na Ye m en Vie t N am 0 0 2 4 6 8 10 12 Gh an a Ma li Ma da ga sca r Be nin Ca me roo n Tu rke y Cô te d'Iv oir e Bo livi a Nig er Ke ny a Se ne ga l Ha iti Ug an da Nig eri a Eg ypt Ph ilip pin es Ma law i Ne pa l Zim ba bw e Ind ia Ta nz an ia Gu ate ma la Jo rda n Za mb ia Pe ru Ba ng lad esh Ind on es ia Co lom bia Do mi nic an Re pu blic Ka za kh sta n AV ER AG E Pe rc en t 54 Chapter 8 Table 8.6. Relation of Unmet Need and Intention to Use* to use. Among all married women the average across the 14 countries is 15% who plan to use even though they are classified without need. Among non-us- ers the average is 31%, a remarkable finding. This occurs due to an oddity in the DHS definition, which says that women who want a birth within two years have no need. Actually substantial numbers of those women do not want to become pregnant just yet; they wish to insure a delay of their next conception within the two-year period in question. These “in- tenders without need” need to be taken into account in program planning. One methodological note is in order: in- tention to use is divided in some surveys into intention to use within the next year vs. intention to use later. However many surveys do not use this distinction, so the data presented here refer to intention to use at anytime, whether in the next year or later. If data were available every- where on use within the next year that would be preferable, since that signals a firmer resolve in the short term and probably contains fewer courtesy re- plies. All the intention figures here would be lower if restricted to the next year. However the figures would be larger if amenorrheic women were included in the unmet need count. Nearly all women near their last birth say they do not want to become pregnant again soon, and many are close to resuming ovulation. For program purposes they are eminent- ly in need of information and access for contraception. Including all amenorrhe- ic women raises the proportion with un- met need by half, from an average of 22% to 33% in a study of 27 countries (Ross and Winfrey, 2002). (All amenor- rheic women were included regardless of future intention statements, but as noted, few such women wish to get preg- nant soon and many would intend to use.) Some analysts prefer a less inclu- sive rule, to retain only women who are at least four months from their last birth. That focuses upon women who are near- er to the resumption of menses, and avoids what may appear as an excessive estimate of need. Among All married women Among Married Non-Users Unmet Need Unmet need Yes No Using Yes No Totals Asia Bangladesh 2000 Intend to use Yes 13 20 Yes 28 44 71 No 3 11 No 6 23 28 15 31 54 33 67 100 India 1999 Intend to use Yes 12 19 Yes 22 38 60 No 4 16 No 8 31 40 16 35 48 31 68 100 Indonesia 2002 Intend to use Yes 4 13 Yes 11 32 43 No 4 18 No 11 46 56 9 31 60 22 78 100 Philippines 1998 Intend to use Yes 10 12 Yes 19 23 42 No 9 21 No 17 41 58 19 33 48 36 64 100 Latin America Colombia 2000 Intend to use Yes 5 11 Yes 23 47 70 No 1 6 No 4 27 30 6 17 77 27 73 100 Dominican Rep. 2002 Intend to use Yes 9 11 Yes 30 36 66 No 2 8 No 6 28 34 11 19 70 36 64 100 Peru 2000 Intend to use Yes 7 10 Yes 23 33 56 No 3 11 No 9 34 43 10 21 69 33 67 100 Middle East/North Africa Egypt 2000 Intend to use Yes 7 19 Yes 17 42 59 No 3 15 No 8 33 41 11 33 56 24 76 100 Turkey 1998 Intend to use Yes 7 12 Yes 19 33 52 No 3 14 No 9 38 47 10 26 64 28 72 100 Sub-Saharan Africa Ethiopia 2000 Intend to use Yes 24 18 Yes 26 20 46 No 11 39 No 12 42 54 35 57 8 38 62 100 Kenya 2003 Intend to use Yes 18 17 Yes 30 28 58 No 6 19 No 10 32 42 25 36 39 40 60 100 Nigeria 1999 Intend to use Yes 6 13 Yes 8 15 23 No 11 53 No 12 62 76 17 66 15 21 78 100 Tanzania 1999 Intend to use Yes 13 16 Yes 18 21 39 No 8 37 No 11 50 61 22 53 25 29 71 100 Uganda 2000 Intend to use Yes 26 22 Yes 34 28 62 No 8 21 No 11 27 38 35 43 23 45 55 100 *Intention to use at any time in the future. Source: Special tabulations provided courtesy of ORCMacro, Demographic and Health Surveys. Chapter 8 55 An examination of five DHS countries for women with sexual experience found that the inclusion of those amenorrheic women who either wanted no more chil- dren or wanted to wait increased the per- cent with unmet need by roughly half (Bernstein, 2005), a result that matches the one-half increase found just above. For surveys that do not ask about amen- orrheic status, reliance upon when the next child is wanted is especially rele- vant. In any case, explicit clarification of the rule is essential. Intention to use is a marker for the extent of serious interest in contraceptive use. In a Morocco panel study (Curtis and Westoff, 1996) stated intention was the best predictor of which women in fact adopted contraception after the initial survey. Intention to use can be measured in various groups of programmatic inter- est: one group is all married women; others are women classified with unmet need and women who are not now using a method (all three measures are includ- ed in Appendix Table A.11). The percent intending to use varies for several rea- sons: in Brazil few married women in- tend to use because most already do so, whereas in Nigeria the desired family size is still large, and access to methods is poor. In Pakistan religious objections and husband opposition may help ex- plain why 33% of married women have unmet need but the percent intending to use is still low. The Philippines presents yet another combination of factors, in- cluding some religious ambivalence. All three countries lack vigorous program action in the rural sector. In Kenya the high proportions planning to use perhaps reflect a better supply system and great- er personal freedom by women to adopt a method. It must be remembered that many wom- en who say they intend to use will in fact not do so, at least not in the near future. A wide range of deterrents exists such as personal ambivalence, family opposi- tion, and weak programs that provide neither information nor physical access to a choice of methods. Nevertheless the intention to use suggests the presence of a market for contraception that supple- ments unmet need information. In sum- mary, close attention should be paid by managers and planners to levels and trends for both intention to use and un- met need. They are the best gauges of public interest in contraceptive use, whether supplied by the public or private sector. References Bernstein, Stan. Personal communica- tion, January 17, 2005. Bernstein, Stan. “A Proposal for Includ- ing a Measure of Unmet Need for Con- traception and Adolescent Fertility or Early Marriage Levels as Indicators of the Reproductive Health Component of Gender Equality.” Report of the Inter- agency and Expert Group Meeting on MDG Indicators, 27 September 2004, link (labelled “Proposal for a Reproduc- tive Health-related Gender Equality In- dicator,” at www.unstats.un.org/unsd/mi/ techgroup/subgroups/Gender.htm), ac- cessed January 17, 2005. Curtis, S.L., and C.F. Westoff, “Intention to Use Contraceptives and Subsequent Contraceptive Behavior in Morocco.” Studies in Family Planning 27(5:239- 250). 1996. Phai, Nguyen Van, John Knodel, Mai Van Cam, and Hoang Xuyen. “Fertility and Family Planning in Vietnam: Evi- dence from the 1994 Intercensal Demo- graphic Survey.” Studies in Family Plan- ning 27(1):1-17. 1996. Robey, Bryant, John Ross, and Indu Bhushan. “Meeting Unmet Need: New Strategies.” Population Reports. Series J, No. 43, September 1996. Ross, John, and Laura Heaton. “Intend- ed Contraceptive Use Among Women Without an Unmet Need.” International Family Planning Perspectives 23(4): 149-154. December 1997. Ross, John and William Winfrey. “Un- met Need for Contraception in the De- veloping World and the Former Soviet Union: An Updated Estimate.” Interna- tional Family Planning Perspectives 28(3):138-143. 2002. Sinding, Steven W., John A. Ross, and Allan G. Rosenfield. “Seeking common Ground: Unmet need and demographic goals.” International Family Planning Perspectives 20(1):23-27. March 1994. United Nations. Levels and Trends of Contraceptive Use as Assessed in 2002. New York: United Nations Population Division. Westoff, C.F., and A. Bankole. “Unmet Need: 1990-1994.” DHS Comparative Studies No. 16, Calverton, Maryland: Macro International. June, 1995. 56 Chapter 8 Goal No. 3: To Reach the Desired Fertility Level (See Appendix Table A.12) The desired family size has fallen over time in most countries, and has consis- tently stayed below actual fertility as it too has fallen. A reasonable goal for a national program is to hasten movement to the desired level, and this section trac- es the mutual changes in both actual and desired fertility. A useful measure of desired fertility is the total wanted fertility rate (TWFR). In each survey it is the same as the total fertility rate (TFR) except that any birth that exceeds the respondent’s ideal fam- ily size is considered unwanted. Remov- ing such births leaves the TWFR. Thus for each woman her wanted fertility is al- ways below or equal to her actual fertili- ty as of the survey. The comparison of the two measures ap- pears in Figure 8.6. The TFR is every- where above the TWFR. On average the difference is about one child between the number wanted and the number in fact, and in some countries it is a two-child difference. Over the developing world this amounts to a large body of unwant- ed childbearing as expressed by the women themselves, quite apart from any public policy regarding fertility. Trends: The goal of bringing actual fer- tility in line with wanted (“desired”) fer- tility can be quite elusive. Desired fertil- ity is a receding target, since it has been declining in most developing countries. It is common for the current TFR to have fallen below earlier desired levels, but for the current desired level to have fallen well below the current TFR. A special analysis was conducted for time trends in 40 countries with multi- ple DHS surveys. Each line in Figures 8.7a – 8.7e shows the joint change in the TFR and the TWFR between the earliest and latest DHS surveys. The five figures are scaled identically, to highlight the regional differences in general levels. (Legends show the dates of each initial survey.) When the wanted fertility rate falls fast- er than the total fertility rate the gap be- tween the two increases with time rath- er than diminishes. One way of captur- ing this is by the slopes in Figure 8.7a, since each line shows how rapidly the TFR fell in relation to the fall in the TWFR. If they fell exactly together the slope would be 1.0; if the TFR fell fast- er the slope would exceed one, and if the TWFR fell faster the slope would be less than one. For example, in Nigeria, from 1990 to 2003, the TFR fell from 6.0 to 5.7, but in the meantime the TWFR fell from 5.8 to 5.3. The 0.3 fall in fertility was exceed- ed by the 0.5 fall in wanted fertility, en- larging the gap between the two. The re- sulting slope was 0.60: the ratio between the two changes. Here are the average slopes by region: (no. of countries in parentheses) Sub-Saharan Africa 0.84 (20) Other Regions 1.98 (20) Asia 1.39 (6) Latin America 1.86 (8) Middle East/North Africa 2.76 (6) All countries 1.38 (40) In sub-Saharan Africa, desired fertility has been declining faster than actual fer- tility, outpacing it and increasing the gap between the two. The averages for it and “other regions” are 0.84 and 1.98 respec- tively, a very large difference that re- flects the failure of fertility to fall as rap- idly as desired in sub-Saharan Africa. This is also reflected in the high levels of unmet need in that region and the pro- portions of births that are ill-timed or not wanted at all. In the other regions the TFR has fallen at nearly double the pace of the TWFR (1.98 average slope above), so behavior there has been catching up with desires. Gaps still remain however: all lines in Figure 8.7a (for 40 countries) fall above the line of equality between the TWFR and the TFR. That is one demonstration of the way yet to go to achieve the goal of matching desired fertility to actual fertility. The disparity is greatest in sub- Saharan Africa, where the trends show the failure of fertility to fall as rapidly as desired fertility has (Figure 8.7c). Births not wanted: Another way to state the goal in this section is by the reduc- tion in births that are ill-timed or not wanted. Survey respondents are asked about their most recent birth (or current pregnancy) and the results are tabulated as in Appendix Table A.12. They are dis- played by region in Figures 8.8a and 8.8b. Figure 8.6. Relation of the Total Fertility Rate (TFR) to the Total Wanted Fertility Rate (TWFR), 140 National DHS Surveys - 1 2 3 4 5 6 7 8 - 1 2 3 4 5 6 7 8 Total Wanted Fertility Rate (TWFR) To ta l F er tili ty R at e (T FR ) Chapter 8 57 Figure 8.7a Trends in the Relation of the Total Fertility Rate (TFR) to the Total Wanted Fertility Rate (TWFR) (40 countries with multiple DHS Surveys) Figure 8.7b Asia Figure 8.7d Latin America Figure 8.7c Sub-Saharan Africa Figure 8.7e Middle East/North Africa 1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 8 TF R TWFR 1 2 3 4 5 6 7 8 TF R Bangladesh 1993/94 India 1992/93 Indonesia 1987 Nepal 1996 Philippines 1993 Viet Nam 1997 1 2 3 4 5 6 7 8 TWFR Benin 1996 Burkina Faso 1998/99 Cameroon 1991 Côte d'Ivoire 1994 Eritrea 1995 Ghana 1988 Kenya 1989 Madagascar 1992 Malawi 1992 Mali 1987 Namibia 1992 Niger 1992 Nigeria 1990 Rwanda 1992 Senegal 1986 Tanzania 1992 Togo 1988 Uganda 1988 Zambia 1992 Zimbabwe 1988 1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 8 TWFR TF R Bolivia 1989 Brazil 1986 Colombia 1986 Dominican Republic 1986 Guatemala 1987 Haiti 1994/95 Nicaragua 1997/98 Peru 1986 1 2 3 4 5 6 7 8 TF R 1 2 3 4 5 6 7 8 TWFR Egypt 1988 Jordan 1990 Morocco 1987 Turkey 1993 Yemen 1991/92 1 2 3 4 5 6 7 8 TWFR 2 3 4 5 6 7 8 TF R 1 58 Chapter 8 0 10 20 30 40 50 60 70 Ph ilip pin es Ne pa l 2 00 1 Ba ng lad es h 1 99 9/2 00 0 Ca m bo dia 20 00 Vie t N am 20 02 Pa kis tan 19 90 /91 Ind ia 19 98 /99 Ind on es ia 20 02 /20 03 Bo livi a 2 00 3 Pe ru 20 00 Ha iti 2 00 0 Co lom bia 20 00 Bra zil 19 96 Nic ar ag ua 20 01 Do m inic an Re pu blic 20 02 Gu ate m ala 19 98 /99 Pa ra gu ay 19 90 Ye m en 19 97 Mo ro cc o 19 92 Jo rda n 20 02 Tu rke y 1 99 8 Eg yp t 2 00 0 Arm en ia 20 00 Ka za kh sta n 19 99 Ky rgy z R ep ub lic 19 97 Uz be kis tan 19 96 Tu rkm en ista n 20 00 Pe rc en t o f B irt hs Figure 8.8a Percent of Births Ill-Timed or Not Wanted: Recent DHS Surveys in Four Regions Figure 8.8b Percent of Births Ill-Timed or Not Wanted: Recent DHS Surveys in Sub-Saharan Africa 0 10 20 30 40 50 60 70 So uth Af ric a 19 98 Na m ibia 20 00 Ke ny a 20 03 Ga bo n 20 00 To go 19 98 Za mb ia 20 01 /02 Ma law i 2 00 0 Gh an a 20 03 Ug an da 20 00 /01 Zim ba bw e 19 99 Eth iop ia 20 00 Rw an da 20 00 Se ne ga l 1 99 7 Cô te d'Iv oir e 19 98 /99 Ma ur ita nia 20 00 /01 Ca m er oo n 19 98 Eri tre a 20 02 Mo za m biq ue 19 97 CA R 1 99 4/9 5 Bu rki na Fa so 20 03 Be nin 20 01 Ta nz an ia 19 99 Ma li 2 00 1 Gu ine a 19 99 Ma da ga sca r 2 00 3/2 00 4 Nig er ia 20 03 Nig er 19 98 Ch ad 19 96 /97 Pe rc en t o f B irt hs Countries vary considerably within each region, and the regions differ in their central tendencies. Latin America has the highest percentages of ill-timed or unwanted births, over 40% in seven of the nine countries shown. In sub-Sahar- an Africa only 8 of 28 countries are above the 40% level, partly because de- sired fertility levels are higher there. Using a cutoff of 20%, altogether, 42 of the 55 countries fall above that level, in- dicating that one birth in five is ill-timed or not wanted. Percent wanting no more children. A further measure of time trends in desired fertility is increases in the percent of women who say that they want no more children. Nineteen countries have infor- mation on this at two or three points in time, from World Fertility Surveys of the late 1970s through multiple DHS sur- veys to the early 1990s (data pertain only to fecund women in union) (Bankole and Westoff, 1995). The de- gree and consistency of upward trends in Figure 8.9 are remarkable, and they oc- cur in both periods shown. Table 8.7 adds the mean values: the average in- crease over the 10-15 year period hap- pens to be in the range of 10-15 points, a substantial shift, especially considering that the increase probably included more younger women, at lower parities. The percent wanting no more children increases sharply by family size (not shown), but regions differ sharply in the gradient. In much of sub-Saharan Africa few women want to stop unless they have Chapter 8 59 three children, but in Latin America many wish to stop at one child, and one- half to two-thirds of those with two chil- dren wish to stop. Middle East/North Africa is between these extremes, and Asian countries vary the range. More recent information, for more coun- tries, is given in Appendix Table A.12. Information is not available for all coun- tries in all regions, but the averages are as follows: Mean Percent Wanting No No. of More Children Countries Asia 45.0 10 Latin America 47.9 13 Middle East/North Africa 48.3 7 Sub-Saharan Africa 26.4 31 Central Asia Republics 51.2 4 All countries 37.4 65 The results are quite uniform in four re- gions, with 45% to 51% of married women not wanting more children. Sub-Saharan Africa falls well below that at only 26%, since desired family sizes are lower there and a higher percent of the population is made up of younger women, who still want another child. Figure 8.9. Trends in Desire for No More Children Table 8.7. Trends in Desire for No More Children WFS DHS-I DHSII,III Increase Sub-Saharan Africa Cameroon 3.0 15.0 12.0 Ghana 12.0 22.0 34.0 22.0 Kenya 17.0 49.0 52.0 35.0 Nigeria 5.0 14.0 9.0 Rwanda 19.0 36.0 17.0 Senegal 7.0 17.0 19.0 12.0 Sudan 17.0 23.0 6.0 Zimbabwe 32.0 31.0 (1.0) Means 11.4 28.6 28.7 17.3 Middle East/North Africa Egypt 53.0 64.0 68.0 15.0 Jordan 42.0 54.0 12.0 Morocco 42.0 48.0 53.0 11.0 Turkey 57.0 72.0 15.0 Means 48.5 56.0 61.8 13.3 Asia Indonesia 39.0 50.0 52.0 13.0 Pakistan 43.0 39.0 (4.0) Philippines 54.0 64.0 10.0 Means 45.3 50.0 51.7 6.3 Latin America Colombia 61.0 70.0 66.0 5.0 Dominican Republic 52.0 64.0 66.0 14.0 Paraguay 32.0 45.0 13.0 Peru 61.0 73.0 75.0 14.0 Means 51.5 69.0 63.0 11.5 Overall Means 34.2 46.5 47.5 12.1 Source: Table 4.4 in Bankole and Westoff, 1995. 0 10 20 30 40 50 60 70 80 Late 1970s Mid-Period Early 1990s Pe rc en t W an tin g No M o re Cameroon Ghana Kenya Nigeria Rwanda Senegal Sudan Zimbabwe Egypt Jordan Morocco Turkey Indonesia Pakistan Philippines Colombia Dominican Republic Paraguay Peru 60 Chapter 8 Goal No. 4: To Attain the Replacement Fertility Level Replacement fertility is normally set at a total fertility rate of 2.1, slightly above 2.0 to allow for some mortality. By this standard a number of developing coun- tries have approached or surpassed the goal of replacement. In Asia these in- clude most prominently China as well as South Korea (and perhaps North Korea), Taiwan, Hong Kong, Singapore, Thai- land, Viet Nam, Sri Lanka, and at least Kerala State in India. In Latin America there are Cuba, Puerto Rico, Trinidad and Tobago, probably Brazil, and nu- merous small Caribbean populations. Others are Mauritius, Kazakhstan, and Tunisia. Many other developing countries have moved far along the path toward low fer- tility and smaller family sizes, enough to produce, for the developing world as a whole, a 79% decline toward replace- ment over the past 35 years. Table 8.8 shows the United Nations TFR estimates for 1960-1965 and 2000-2005 (regions according to UN definitions). All regions began at traditionally high fertility lev- els, and all fell to levels that reflect tru- ly historic changes in marriage and re- productive behavior. East Asia, with China, has fallen below replacement, and Southeastern Asia and Latin Ameri- ca have fallen 89% of the way. The North Africa and Western Asia regions are the closest to the “Middle East/North Africa” region used elsewhere in this re- port; they fell by 78% and 68% respec- tively of the distance to replacement. Sub-Saharan Africa has moved only 30% of the way. Finally, a group of 48 countries (33 in Africa) identified by the UN as “least developed” (second row of table) have fallen only a third of the way. It is important to bear in mind that the total fertility rate is only one measure of fertility behavior. Unlike the crude rate or general fertility rate, it gives equal weight to every age group, and it is sometimes sensitive to short-term fluctuations, for example in age at first birth. It does not reflect population momentum: popula- tions will continue to grow for some de- cades after replacement is reached.