chapter two obstacles to progress: context or policy? This chapter seeks to explain why progress in maternal and child health has apparently stumbled so badly in many countries. It shows in detail how stagnations, reversals and slow progress in some countries are clearly to poverty, HIV/AIDS, and humanitarian crises, leading to exclusion from to health care. In many countries, the strategies put in place to provide health services have not produced the hoped for results. While many countries have based their health care systems on health districts, with a backbone of health centres and a referral district hospital, there has often been a failure to implement this model successfully in an exceedingly resource-constrained context. The chapter argues that the health district model still stands as a rational way for governments to organize decentralized health care delivery. Although there has been, for decades now, a global consensus that than for children. Whatever the context, lack of progress the health of mothers and children is a public priority, much still needs is also due to failures of health systems to provide good- to be done. Most progress is being made by countries that were al- quality care and services to all mothers and children ready in a relatively good position in the early 1990s, whereas those Moving towards universal access to health care must less favourably placed, particularly in sub-Saharan Africa, have been take account of the contextual barriers to progress, the left behind. Much of this large and growing gap can be explained by reasons for exclusion from care, and the various pat the context in which health systems have developed. The stagnations, terns of exclusion. Many countries, and particularly reversals and slow progress seen in some countries are clearly related those that face the biggest challenges, have based their to contexts of poverty, humanitarian crisis and the direct and indirect health care systems on the health district model, with a effects of HIV/AIDS (see Table 2. 1). These lead to an increasingly vis- backbone of health centres and a referral district hospi- ible gap between people who have access to health care and others tal. This chapter argues that the disappointing situation ho are excluded from such benefits. Exclusion from health benefits in many countries often has more to do with the condi- leads to even greater inequalities in survival for mothers and newborns tions under which this strategy has been implemented
21 chapter two obstacles to progress: context or policy? Although there has been, for decades now, a global consensus that the health of mothers and children is a public priority, much still needs to be done. Most progress is being made by countries that were already in a relatively good position in the early 1990s, whereas those less favourably placed, particularly in sub-Saharan Africa, have been left behind. Much of this large and growing gap can be explained by the context in which health systems have developed. The stagnations, reversals and slow progress seen in some countries are clearly related to contexts of poverty, humanitarian crisis and the direct and indirect effects of HIV/AIDS (see Table 2.1). These lead to an increasingly visible gap between people who have access to health care and others who are excluded from such benefits. Exclusion from health benefits leads to even greater inequalities in survival for mothers and newborns than for children. Whatever the context, lack of progress is also due to failures of health systems to provide goodquality care and services to all mothers and children. Moving towards universal access to health care must take account of the contextual barriers to progress, the reasons for exclusion from care, and the various patterns of exclusion. Many countries, and particularly those that face the biggest challenges, have based their health care systems on the health district model, with a backbone of health centres and a referral district hospital. This chapter argues that the disappointing situation in many countries often has more to do with the conditions under which this strategy has been implemented This chapter seeks to explain why progress in maternal and child health has apparently stumbled so badly in many countries. It shows in detail how stagnations, reversals and slow progress in some countries are clearly related to poverty, HIV/AIDS, and humanitarian crises, leading to exclusion from access to health care. In many countries, the strategies put in place to provide health services have not produced the hoped for results. While many countries have based their health care systems on health districts, with a backbone of health centres and a referral district hospital, there has often been a failure to implement this model successfully in an exceedingly resource-constrained context. The chapter argues that the health district model still stands as a rational way for governments to organize decentralized health care delivery
22 The World Health Report 2005 Table 2.1 Factors hindering progress than with the failure of the strategy itself a new commitment is needed to create Decline of More than two years Adult HIV the conditions for moving towards effec- child mortality of humanitarian crisis since 1992 (weighted average) 1990-2002 in 1995 verage tive implementation international dollars CONTEXT MATTERS 93 countries are 3/93 countries 20049(0ECD) on track 4179(non-0ECD) Poverty undermines progress 51 countries are making 10/51 countries 07 Many of the countries whose child mor- slower progress tality rates are stagnating or reversing are oor in terms of gross domestic product 14 countries are 8/14 countries 10.2 others are facing economic downturn. (excluding South Africa) Conventional wisdom has it that income 29 countries have 11/29 countries poverty is on its way out because the stagnating mortality proportion and the total number of peo Towards Millennium Development Goal 4 ple around the world living on less than S1 per day is decreasing (1). However almost all of this progress has been made in Asia. Sub-Saharan Africa, where most of the countries whose child mortality rates have stagnated or reversed are to be found, has emerged as the region with the highest incidence of extreme poverty, and the greatest depth of poverty(2). Furthermore, the average income of poor people in Africa has been falling over time, in contrast with that of poor people in the rest of the developing world ( 3) But poverty also influences maternal health. When women die in childbirth it is usu- ally the result of a cascade of breakdowns in their interactions with the health system delays in seeking care, inability to act on medical advice, and failure of the health system to provide adequate or timely care. These breakdowns are more likely to occur and to come together into a fateful combination when the macroeconomic and social contexts deteriorate In Mongolia, for example, widespread social chaos and economic collapse followed the introduction of economic"shock therapy"in the early 1990s(4, 5), with a rapid increase in unemployment and widespread poverty. Government ex- Box 2.1 Economic crisis and health system meltdown: a fatal cascade of events shnyam, a 41-year-old housewife, was a very she did not follow their advice and of these serious symptoms, the doctor urged or migrant from the countryside to a provin- pregnant again. She did not seek prenatal care, her to go to the provincial hospital's maternity cial capital of Mongolia. She and her husband but the family doctor discovered her pregnancy waiting home. However, her admission was were unemployed and often homeless, with six during an antenatal examination of her 18- delayed for over a week to solve bureaucratic children. During her last pregnancy Dashnyam old daughter. Because of Dashnyam's history issues, initially because she had no proof of had oedema and pre-eclampsia and required and age, and because she said that she did not having health insurance, and then because manual extraction of the placenta. Afterwards, want the child, the family doctor urged her to go there were no beds available. Eventuall she said she wanted no more children and was to the provincial hospital for an abortion. How- Dashnyam delivered via caesarean section given an intrauterine device(IUD). She had ever, by the time she had collected sufficient but suffered severe haemorrhage. After delay problems with the IUD and finally, in 2002 after funds, her pregnancy was too far advanced and in finding the anaesthetist, the bleeding years of use, she asked to have it removed abortion was no longer an option. She returned eventually stopped by emergency surgery, but ecause of pelvic inflammatory disease and home and received antenatal care from the the hospital had no blood for transfusion. She associated pain. The obstetrician who removed family doctor. As she came closer to term, she died from haemorrhagic shock. Names and the IUd urged her to use another form of birth manifested symptoms of pre-eclampsia -high places have been changed control, and her primary care physician gave her blood pressure and oedema. Because of her the same advice. For reasons that are unclear, age, history of complications, and the presence Source: (7)
22 The World Health Report 2005 than with the failure of the strategy itself. A new commitment is needed to create the conditions for moving towards effective implementation. CONTEXT MATTERS Poverty undermines progress Many of the countries whose child mortality rates are stagnating or reversing are poor in terms of gross domestic product; others are facing economic downturn. Conventional wisdom has it that income poverty is on its way out because the proportion and the total number of people around the world living on less than US$ 1 per day is decreasing (1). However, almost all of this progress has been made in Asia. Sub-Saharan Africa, where most of the countries whose child mortality rates have stagnated or reversed are to be found, has emerged as the region with the highest incidence of extreme poverty, and the greatest depth of poverty (2). Furthermore, the average income of poor people in Africa has been falling over time, in contrast with that of poor people in the rest of the developing world (3). But poverty also influences maternal health. When women die in childbirth it is usually the result of a cascade of breakdowns in their interactions with the health system: delays in seeking care, inability to act on medical advice, and failure of the health system to provide adequate or timely care. These breakdowns are more likely to occur and to come together into a fateful combination when the macroeconomic and social contexts deteriorate. In Mongolia, for example, widespread social chaos and economic collapse followed the introduction of economic “shock therapy” in the early 1990s (4, 5), with a rapid increase in unemployment and widespread poverty. Government exshe did not follow their advice and was soon pregnant again. She did not seek prenatal care, but the family doctor discovered her pregnancy during an antenatal examination of her 18-yearold daughter. Because of Dashnyam’s history and age, and because she said that she did not want the child, the family doctor urged her to go to the provincial hospital for an abortion. However, by the time she had collected sufficient funds, her pregnancy was too far advanced and abortion was no longer an option. She returned home and received antenatal care from the family doctor. As she came closer to term, she manifested symptoms of pre-eclampsia – high blood pressure and oedema. Because of her age, history of complications, and the presence Dashnyam, a 41-year-old housewife, was a very poor migrant from the countryside to a provincial capital of Mongolia. She and her husband were unemployed and often homeless, with six children. During her last pregnancy Dashnyam had oedema and pre-eclampsia and required manual extraction of the placenta. Afterwards, she said she wanted no more children and was given an intrauterine device (IUD). She had problems with the IUD and finally, in 2002 after six years of use, she asked to have it removed because of pelvic inflammatory disease and associated pain. The obstetrician who removed the IUD urged her to use another form of birth control, and her primary care physician gave her the same advice. For reasons that are unclear, of these serious symptoms, the doctor urged her to go to the provincial hospital’s maternity waiting home. However, her admission was delayed for over a week to solve bureaucratic issues, initially because she had no proof of having health insurance, and then because there were no beds available. Eventually, Dashnyam delivered via caesarean section, but suffered severe haemorrhage. After delay in finding the anaesthetist, the bleeding was eventually stopped by emergency surgery, but the hospital had no blood for transfusion. She died from haemorrhagic shock. (Names and places have been changed.) Source: (7). Table 2.1 Factors hindering progress Decline of More than two years Adult HIV GDP per capita child mortality of humanitarian prevalence rate (weighted average crisis since 1992 (weighted average) 1990–2002 in 1995 international dollars) 93 countries are 3/93 countries 0.3 20 049 (OECD) on tracka 4179 (non-OECD) 51 countries are making 10/51 countries 0.7 2657 slower progressa 14 countries are 8/14 countries 10.2 1627 in reversal (excluding South Africa) 29 countries have 11/29 countries 4.1 896 stagnating mortality a Towards Millennium Development Goal 4. Box 2.1 Economic crisis and health system meltdown: a fatal cascade of events
obstacles to progress: context or policy? 23 penditure halved, reflecting a widespread Figure 2. 1 a temporary reversal in maternal mortality drop in investment in social services, health care and education. Hospitals Mongolia in the early 1990s clinics and maternity homes closed or curtailed operations(6). The health sec tor recovered eventually with the support before the meltdown of services had led to a temporary reversal in matemal mor- the dynamics of increasing poverty can create a fatal series of events are illus- 0.05 e direct and indirect effects管 In a number of countries, particularly in sub-Saharan Africa, the effects of pov erty and economic downturns on the 199119921993199419951996199719981999 environment in which people live, on their health and on the functioning of health systems are compounded by HIV/AIDS HIV/AIDS has direct and indirect effects. It directly affects the health of infected women and children(see Box 2.2). Globally, the direct contribution of HIv/AIDS to the number of children dying is limited, but it has been increasing steadily in sub-Saharan Africa. In 1990 HIV/AIDS accounted for around 2% of under-five mortality in that part of the world; 10 years later this had risen to 6.5%, although there are significant Box 2.2 How HIV/AIDS directly affects the health of women and children The HIVAIDS pandemic takes an increas(miscarriage, anaemia, postpartum haemor- Children of an HIv-positive mother have a g toll of women and children, especially in rhage, puerperal sepsis and post-surgical com- higher mortality risk than children of HIv-nega sub-Saharan Africa. Some 39 million people plications). AIDS is also a major indirect cause tive mothers(13 ) As parents die of AlDS, the are now living with HIv, of whom 2.2 mil- of maternal mortality through increased rates number of orphans increases: 9% of children lion are children under 15 years of age and of malaria and opportunistic infections such as under 15 years of age in 40 countries in sub 18 million are women. In 2004, there were tuberculosis(12). The combined effect of these Saharan Africa have lost one parent, and 1% 4.9 million new cases of infection, including different mechanisms may overshadow prog- have lost both (15). Orphans are especially vul- 640 000 children under 15 (8). Almost 90% ress made in reducing maternal mortality from nerable to social and health risks: they are less of paediatric infections occur in sub-Saharan other causes. In Rakai, Uganda, for example, likely to attend school and may live in house Africa, where there are both high fertility rates maternal mortality was 1687 per 100 000 live holds where conditions are less favourable for and high HIV prevalence rates among women births among HIV-infected women and 310 health and development than the average HIV (9). In 2004, 3.1 million people died of AlDS, among non-infected women (13). The maternal infection in children, almost always acquired 510 000 of whom were children (8). HIV/AIDs mortality ratio in the University Teaching Hospi- through mother-to-child transmission, causes has thus led to significant increases in mortal- tal in Lusaka, Zambia, has increased eightfold high mortality rates and some 60% die before ity in many countries: it is a leading cause of over the past two decades, mainly because of their fifth birthday (16). In Malawi, HIV/AIDS death among women and children in the most the increase in non-obstetric causes of death. accounts for up to 10% of child deaths, and in everely affected countries in sub-Saharan While such causes were almost negligible in one of the most affected countries, botswana Africa (10 1975, HIV-related tuberculosis and unspeci- child mortality doubled in the 1990s, and Hiv Across the world, around 2.2 million women fied chronic respiratory illnesses accounted AlDS was responsible for more than 60% of with HIV infection give birth each year (11). for 27% of all causes of maternal deaths in child mortality in 2000 (16) HIV infection in pregnancy increases the risk 1997(14) of complications of pregnancy and childbirth
obstacles to progress: context or policy? 23 Change in gross domestic product per capita Maternal mortality per 100 000 live births 0.05 0 -0.05 -0.1 250 200 150 1991 1992 1993 1994 1995 1996 1997 1998 1999 Figure 2.1 A temporary reversal in maternal mortality: Mongolia in the early 1990s (miscarriage, anaemia, postpartum haemorrhage, puerperal sepsis and post-surgical complications). AIDS is also a major indirect cause of maternal mortality through increased rates of malaria and opportunistic infections such as tuberculosis (12). The combined effect of these different mechanisms may overshadow progress made in reducing maternal mortality from other causes. In Rakai, Uganda, for example, maternal mortality was 1687 per 100 000 live births among HIV-infected women and 310 among non-infected women (13). The maternal mortality ratio in the University Teaching Hospital in Lusaka, Zambia, has increased eightfold over the past two decades, mainly because of the increase in non-obstetric causes of death. While such causes were almost negligible in 1975, HIV-related tuberculosis and unspeci- fied chronic respiratory illnesses accounted for 27% of all causes of maternal deaths in 1997 (14). The HIV/AIDS pandemic takes an increasing toll of women and children, especially in sub-Saharan Africa. Some 39 million people are now living with HIV, of whom 2.2 million are children under 15 years of age and 18 million are women. In 2004, there were 4.9 million new cases of infection, including 640 000 children under 15 (8). Almost 90% of paediatric infections occur in sub-Saharan Africa, where there are both high fertility rates and high HIV prevalence rates among women (9). In 2004, 3.1 million people died of AIDS, 510 000 of whom were children (8). HIV/AIDS has thus led to significant increases in mortality in many countries: it is a leading cause of death among women and children in the most severely affected countries in sub-Saharan Africa (10). Across the world, around 2.2 million women with HIV infection give birth each year (11). HIV infection in pregnancy increases the risk of complications of pregnancy and childbirth Children of an HIV-positive mother have a higher mortality risk than children of HIV-negative mothers (13). As parents die of AIDS, the number of orphans increases: 9% of children under 15 years of age in 40 countries in subSaharan Africa have lost one parent, and 1% have lost both (15). Orphans are especially vulnerable to social and health risks: they are less likely to attend school and may live in households where conditions are less favourable for health and development than the average. HIV infection in children, almost always acquired through mother-to-child transmission, causes high mortality rates and some 60% die before their fifth birthday (16). In Malawi, HIV/AIDS accounts for up to 10% of child deaths, and in one of the most affected countries, Botswana, child mortality doubled in the 1990s, and HIV/ AIDS was responsible for more than 60% of child mortality in 2000 (16). Box 2.2 How HIV/AIDS directly affects the health of women and children penditure halved, reflecting a widespread drop in investment in social services, health care and education. Hospitals, clinics and maternity homes closed or curtailed operations (6). The health sector recovered eventually with the support of sizeable development loans, but not before the meltdown of services had led to a temporary reversal in maternal mortality (see Figure 2.1). The ways in which the dynamics of increasing poverty can create a fatal series of events are illustrated in Box 2.1. The direct and indirect effects of HIV/AIDS In a number of countries, particularly in sub-Saharan Africa, the effects of poverty and economic downturns on the environment in which people live, on their health and on the functioning of health systems are compounded by HIV/AIDS epidemics. HIV/AIDS has direct and indirect effects. It directly affects the health of infected women and children (see Box 2.2). Globally, the direct contribution of HIV/AIDS to the number of children dying is limited, but it has been increasing steadily in sub-Saharan Africa. In 1990 HIV/AIDS accounted for around 2% of under-five mortality in that part of the world; 10 years later this had risen to 6.5%, although there are significant
24 The World Health Report 2005 In humanitarian crises, basic maternal and child health services are often disrupted (New Jalozai refugee camp, Peshawar, Pakistan) differences among countries. HIV prevalence rates are much higher in the countries in stagnation or reversal than in the others(17/); in 9 of the 10 African countries in reversal, HIV/AIDS was responsible for more than 10% of child deaths in 1999, a much higher proportion than in 1990 (18). But HIV/AIDS as a direct cause of death cannot explain all of the slowing or reversal of trends in child mortality HIV/AIDS also affects the health of mothers and children in a more indirect way Appropriate diagnosis and treatment of HIV/AIDS in women and children are rarely provided and pose particular challenges in resource-limited settings. HIV/AIDS puts an additional strain on fragile health systems. It generates demand for new services such as prevention of HIv transmission from mothers to infants, HIV testing and counselling and complex diagnostic and investigative procedures(19, 20). This calls for increased spending on infrastructure, equipment, drugs and human resources. Where increases in funding do not follow, maternal and child health services have to share their scarce resources and personnel. As health workers themselves fall ill, the workforce becomes overstretched(21). Work performance is further reduced by fear, lack of knowledge about HIV/AIDS and protective practices, and the stress of caring for patients whose condition appears hopeless. As a result it is increasingly difficult to recruit young people into medical and nursing professions, particularly obstetrics Conflicts and emergencies set systems back Of the 43 countries showing stagnation or reversal in child mortality, 19 have been the subject of a Consolidated Appeal Process for a humanitarian crisis with a duration of
24 The World Health Report 2005 differences among countries. HIV prevalence rates are much higher in the countries in stagnation or reversal than in the others (17); in 9 of the 10 African countries in reversal, HIV/AIDS was responsible for more than 10% of child deaths in 1999, a much higher proportion than in 1990 (18). But HIV/AIDS as a direct cause of death cannot explain all of the slowing or reversal of trends in child mortality. HIV/AIDS also affects the health of mothers and children in a more indirect way. Appropriate diagnosis and treatment of HIV/AIDS in women and children are rarely provided and pose particular challenges in resource-limited settings. HIV/AIDS puts an additional strain on fragile health systems. It generates demand for new services such as prevention of HIV transmission from mothers to infants, HIV testing and counselling, and complex diagnostic and investigative procedures (19, 20). This calls for increased spending on infrastructure, equipment, drugs and human resources. Where increases in funding do not follow, maternal and child health services have to share their scarce resources and personnel. As health workers themselves fall ill, the workforce becomes overstretched (21). Work performance is further reduced by fear, lack of knowledge about HIV/AIDS and protective practices, and the stress of caring for patients whose condition appears hopeless. As a result it is increasingly difficult to recruit young people into medical and nursing professions, particularly obstetrics. Conflicts and emergencies set systems back Of the 43 countries showing stagnation or reversal in child mortality, 19 have been the subject of a Consolidated Appeal Process for a humanitarian crisis with a duration of US$ 1.5 and US$ 3 per inhabitant per year), especially when compared with those available in refugee camps in the same area through relief agencies. Utilization of curative services and preventive coverage rates has actually increased: vaccination coverage has tripled. Maternal health activities have been intensified both quantitatively and qualitatively, with 52% of deliveries taking place in health centres and the hospital, and a population-based caesarean section rate of 1.4%; case-fatality of caesarean sections dropped from 7.2% to 2.9%. The Since the 1980s, in North Kivu Province of the Democratic Republic of the Congo (formerly Zaire), the socioeconomic environment has been deteriorating. The province also faced an influx of Rwandan refugees in July 1994. In these difficult circumstances the Rutshuru Health District was nevertheless able to adjust and maintain its medical activities. For 11 years the health care network remained accessible and functional, although human and financial resources were extremely limited (external assistance fluctuated between district was able to cope with a workload of 65 000 cases of various pathological conditions in Rwandan refugees settled outside the camps, a 400% increase in the curative workload. The district was under severe pressure but its services managed to respond effi- ciently to the repeated crisis situations, mainly by maintaining a solid district management structure rooted in ongoing communication and participation of the population (23). Box 2.3 Health districts can make progress, even in adverse circumstances In humanitarian crises, basic maternal and child health services are often disrupted (New Jalozai refugee camp, Peshawar, Pakistan). J.M. Giboux/WHO