MNH and poverty - World Bank
February 2005
Maternal-Neonatal Health (MNH) and Poverty: Factors Beyond Care that Affect MNH Outcomes[1]
Thomas W. Merrick
World Bank Institute
Washington, DC
Table of Contents
Introduction 64
The Pathways Framework 64
Data on MNH and Poverty 67
Data and Measurement Issues 67
Regional and Cross-National Differences 67
Table 1: Comparison of 1995 and 2000 Regional and Global Totals 68
Table 2: 2001 Global and Regional Estimates of Neonatal Mortality 68
Poverty Linkages 69
Table 3: Attendance at Delivery by a Medically Trained Person 69
by Wealth Quintile 69
Table 4: Averages of Pathways Indicators for Countries Grouped by MMR Level 70
Linkages with Other Factors 70
Table 5: Pathways Indicators for Large Countries 71
Table 6: Pathways Indicators for Countries with Highest MMRs 72
Cross-National Analyses. 72
Table 7: Correlations between Pathways Variables and MMRs 73
Table 8: Averages of Pathways Indicators for Countries Grouped by Neonatal Mortality Rate 73
Review of Evidence on Pathways Variables 76
Other Reproductive Health Risk Factors 76
Table 9: Total Fertility Rates by Wealth Quintile and Region 77
Table 10: Contraceptive Prevalence by Wealth Quintile and Region 77
Table 11: Adolescent Fertility Rates by Wealth Quintile and Region 78
Table 12: Selected Findings on Other Reproductive Health Risks 82
Household and Community Factors 82
Health System Failures 86
Table 13: Selected Findings on Health System Issues 90
Other Sectors 90
Table 14: Selected Fndings on Transport Interventions 92
Table 15: Low Body-Mass in Reproductive-Age Women by Wealth Quintile and Region 94
Table 16: Child Malnutrition by Wealth Quintile and Region 94
Public Policy and Governance: 95
Table 17: Governance Indicators for Countries Grouped by MMR Level 95
Addressing Obstacles and Information Gaps 95
Policy and Program Actions 95
Strengthening the Evidence Base 98
Table 19: Estimates for Burden of Disease for sub-Saharan Africa, 2002 98
Table 20: Longitudinal Survey Programs 100
Conclusion 101
Annex 1: Maternal Mortality in India 102
Annex Table 1: MMRs and Other Indicators for Indian States 103
Annex 2: Country-Level Data Table 104
References 108
Introduction
Every year more than half a million maternal deaths and around four million perinatal deaths occur in low and middle-income countries, mostly among the poorest groups within these countries. There is an even larger toll of morbidities (more than 8 million each year according to Koblinsky et al., 1993) resulting from non-fatal complications of delivery. Most of these deaths and disabilities are preventable, and the interventions required to prevent them are known. The sad reality is that in many instances these interventions are either not available to poor women or so poor in quality that they are ineffective.
Other reports in this series address the “what” and “how to” of health care interventions to prevent maternal and neonatal mortality and morbidity. This paper focuses on the obstacles that prevent poor women from benefiting from the knowledge and technical expertise that is available, and on the factors beyond care that shape maternal and neonatal health (MNH) outcomes. It employs an adapted version of the “Pathways” framework from the World Bank’s guidelines for Poverty Reduction Strategy Papers to link factors at various levels—from individuals, households and communities to government policies in health and other sectors—that directly or indirectly affect MNH outcomes.
The paper begins with an explanation of what the Pathways framework is and how it will be used to guide this discussion. That is followed by a brief discussion of measurement issues and by a review of cross-national evidence about linkages between factors identified in the Pathways framework and MNH outcomes. More detailed country-level evidence on factors at each of the levels (households and communities, health and other sectors, policy) will then be reviewed, leading to recommendations for actions that could be taken to address obstacles at each of these levels as well as research needed to strengthen the evidence base to assess the impact of these actions.
In adapting the Pathways framework to MNH outcomes, the paper draws on other frameworks that have focused specifically on contextual factors affecting maternal health outcomes, for example the one developed by the Prevention of Maternal Mortality collaborators at Columbia University and in West Africa (McCarthy and Maine, 1993; Thaddeus and Maine, 1994; McCarthy, 1997) as well as broader frameworks that address a range of health outcomes (for example, Hanson et al., 2003). These frameworks address the role of household and community-level variables on outcomes.[2]
The Pathways Framework
The chart below depicts a simple framework for assessing the impact of factors inside and outside the health system that influence health, including MNH, as shown in the left-hand column of the chart. These factors operate at three levels, shown in the other three columns of the chart: (1) households and communities, (2) the health system (including health care, health finance, health promotion) and sectors other than health such as education, infrastructure (water and sanitation, transport and communications) that indirectly influence health outcomes, and (3) public policies and actions that affect health systems and outcomes directly (health reforms, for example) or indirectly (macro-economic policies).
[pic]
Households and Communities: Good health is dependent not only on the provision of good medical services when required but also on healthy behavior. Healthy behavior means avoiding or minimizing risks (e.g. practicing family planning and safe sex) and requires knowledge about how to prevent disease and promote health and the ability to act on this information. Many health-promoting behaviors (for example dietary habits, sanitary practices, fertility regulation, childcare, and utilization of health services) take place within families. Such behaviors are often related to ‘household assets’ such as income, education, access to health services, roads and communication, membership in formal and informal support networks, as well as general knowledge and information. Health too is itself a household asset. Economists view these household-level determinants as demand-side factors, as opposed to the supply of health care (Ensor and Cooper, 2004).
Economic research has tended to treat all members of a family, or household, as a single unit, assuming that whatever benefits one member will benefit the entire household. This is clearly not the case in reality, as it is widely acknowledged that intra-household differences in gender and age may significantly affect how decisions are made and whether a decision is beneficial for all members (Case and Deaton, 2002). Thus an understanding of household decision-making is critical to an understanding of how policy decisions affect the welfare of families as a unit or of individual members within them.
Gender disparities in access to education, credit and political influence have considerable impact on how individual family members are valued and on the degree to which women as well as men have a voice in household decision-making. Recognition of the importance of individuals and households in producing good (or poor) reproductive health outcomes should lead policy makers to focus on the constraints faced by vulnerable households and vulnerable members within those households.
Household-level behaviors and risk factors are influenced and reinforced by conditions in the community (Tinker, 2000). Community factors include both the values and norms that shape household attitudes and behaviors and the physical and environmental conditions of the community, for example terrain and weather conditions that affect households’ capacity to produce better outcomes. Community factors that typically influence health outcomes are:
• Gender norms, which are influenced by social and cultural values that shape the roles of and relationships between men and women;
• Existence of effective community groups, social cohesion (sometimes called social capital) that support positive behaviors individuals and families or organize actions to improve health outcomes directly (community health insurance or pooling of resources to transport emergency cases) or indirectly (micro-credit programs);
• Community access to public services (inside and outside the health sector); and
• Environmental conditions (safe water, location—distance from a health facility, terrain, weather conditions).
Health System and Other Sectors: The supply of health care and health information are key determinants of maternal and neonatal health outcomes for the poor. Since these interventions are covered in detail in other papers in the series, this paper will address two potential supply-side obstacles that may prevent poor women from benefiting from these services. These are: (1) financial obstacles including fees and/or coverage of critical services in insurance benefit packages; and (2) the organizational and institutional obstacles to scaling up effective interventions so that poor women can access them.
Actions in other sectors also affect health outcomes, for example:
• Education, either formal education or training that enhance earnings capacity of household members as well as their capacity for effective health-seeking behaviors;
• Transport and infrastructure, for example the availability of services and the quality of roads that can affect travel time when a mother requires transport for management of an obstetric emergency;
• Energy and communications, for example coverage in a cell-phone network so that help can be sought it an emergency;
• Water and sanitation, important for avoiding infections; and
• Nutrition.
Government Policies and Actions can affect both the health system and related factors in other sectors. Most countries have undertaken reform as a way to profoundly change the fundamentals of the health sector including changes in organization and accountability, revenue generation allocation and purchasing, as well as regulation. Government policies and actions in other sectors also affect health because of their influence on attitudes and behavior and the supply of related services such as education, transport, water and food security. Fees for schools and health services are examples of public policies that may have an indirect effect on MNH outcomes. Taxation may also be a strong influence on health, for example “sin taxes” on tobacco or alcohol.
Data on MNH and Poverty
Data and Measurement Issues
The paper utilizes cross-national estimates of maternal mortality from the WHO/UNICEF/UNFPA database prepared by AbouZahr and Wardlaw (2003). Estimates are provided for 172 countries and are derived from a variety of sources, including vital registration systems, direct and indirect estimating methods based on survey and census data, and statistical modeling for the 62 countries for which no national data on maternal mortality were available. This paper focuses on 142 countries in the World Bank's low and middle-income categories, and most of the 62 countries with estimated maternal mortality ratios (MMRs) are in those categories. The fact that statistical modeling was used to estimate MMRs creates some major limitations for the analysis of cross-country differences, because many of potential explanatory variables (fertility rates, GDP per capita, percentage of births assisted by a skilled attendant, and regional dummy variables) have been used to estimate the proportion of deaths that are considered “maternal” in country-level model life tables. The authors of the estimates emphasize that they should not be used for trend analysis and urge caution in cross-national comparisons for the reasons just stated.
Cross-national data on neonatal mortality rates (NMRs : deaths of liveborns during the first 28 days of life per 1000 live births) are from a compilation published in the State of the World’s Newborns 2001 (Save the Children, 2002). Data for other cross-national indicators (per-capita income, poverty, education, transport, governance, etc) are taken from the World Bank’s World Development Indicators 2003, UNDP’s 2003 Human Development Report, and other sources (see Annex 2).
In addition to cross-national comparisons and analysis, the paper will examine country-level tabulations of key indicators from the Demographic and Health Surveys that have been tabulated using a composite ‘household asset’ measure to show rich-poor differences in those indicators by wealth quintiles (Gwatkin et al, 2004). Additional evidence from country and topic-related studies will also be employed to fill out the picture of how factors beyond care may directly or indirectly impact on MNH outcomes for the poor.
Regional and Cross-National Differences
Global, regional and country-level estimates of maternal mortality (Table 1) show a clear connection between high maternal mortality ratios (MMRs) and poverty.
More than 99 percent of maternal deaths occur in developing regions, and more than 85 percent occur in the poorest countries of Sub-Saharan Africa and South Central Asia. Country-level estimates show that more than a quarter of those deaths occurred in India, and that several other poor countries in these two regions (Bangladesh, Ethiopia, the Democratic Republic of Congo, Nigeria, Pakistan, Tanzania) account for another quarter of them. The highest maternal mortality ratios are found in poor countries in Sub-Saharan Africa. With the exception of Afghanistan, all of the countries having maternal mortality ratios of 1000 or higher are found in Africa.
Table 1: Comparison of 1995 and 2000 Regional and Global Totals
|Region |2000 |1995 |
| |Maternal Mortality |Maternal deaths |Maternal Mortality |Maternal deaths |
| |Ratio |(000s) |Ratio |(000s) |
|WORLD TOTAL |400 |529,000 |400 |515,000 |
|DEVELOPED REGIONS* |20 |2,500 |21 |2,800 |
| Europe |28 |2.2 |36 |3.2 |
|DEVELOPING REGIONS |440 |527,000 |440 |512,000 |
| Africa |830 |251,000 |1,000 |273,000 |
| Northern Africa |130 |4,600 |200 |7,200 |
| Sub-Saharan Africa |920 |247,000 |1,100 |265,000 |
|Asia |330 |253,000 |280 |217,000 |
| Eastern Asia |55 |11,000 |60 |13,000 |
| South-central Asia |520 |207,000 |410 |158,000 |
| South-eastern Asia |210 |25,000 |300 |35,000 |
| Western Asia |190 |9,800 |230 |11,000 |
|Latin America & the Caribbean |190 |22,000 |190 |22,000 |
|Oceania |240 |530 |260 |560 |
Includes Canada, United States of America, Japan, Australia and New Zealand, which are excluded from the regional averages.
Data on neonatal mortality appear to be even scarcer than those for maternal mortality. Regional patterns for neonatal mortality (Table 2) are very similar to those for maternal mortality. Africa and South Asia account for over 93 percent of global deaths.
Table 2: 2001 Global and Regional Estimates of Neonatal Mortality
|Region |Number of live births |Neonatal deaths |Neonatal death rate |
| |(1000s) |(1000s) |(per 1000 live births) |
|Africa |28,865 |1,205 |42 |
|Asia* |76,090 |2,561 |34 |
| South-Central Asia |38,442 |1,757 |46 |
| Other Asia |37,648 |804 |21 |
|Latin America and the |11,553 |196 |17 |
|Caribbean | | | |
|Pacific Islands* |225 |8 |34 |
|Europe |7,374 |44 |6 |
|North America |4,098 |18 |4 |
|More Developed Countries |13,045 |65 |5 |
|Less Developed Countries |116,550 |3,970 |34 |
|Global |129,596 |4.035 |31 |
* Japan, Australia and New Zealand are included with the More Developed Countries but not in the regional sub-estimates.
Source: Save the Children, 2002
Death rates are highest in the South Central Asia region, at 46 per thousand live births, followed by Africa, with 42. The rate is lower for the Other Asia group because China is included there and has a substantially lower neonatal death rate (23), compared to much higher rates for the large countries in South Asia—Bangladesh (48), India (43), and Pakistan (49). While the Pacific Islands account for a small proportion of neonatal deaths, their NNM rate is comparatively high.
Poverty Linkages
Measuring maternal and neonatal mortality for sub-groups of the population within countries is even more challenging than country-level estimates. Graham and colleagues have developed a technique for estimating rich-poor differentials in maternal mortality using Demographic and Health Survey (DHS) data for 10 countries (Burkina Faso, Chad, Ethiopia, Indonesia, Kenya, Mali, Nepal, Peru, Philippines and Tanzania) with large sample sizes using wealth-quintile methodology developed at the World Bank (Graham et al, 2004; Gwatkin et al, 2004). In the country with the largest sample size and also with two surveys, Indonesia, they found that the poorest quintile accounted for one-third of all maternal deaths in both surveys, compared to fewer than 13 percent of deaths in the richest quintile. They also found a high level of association between the survival status of women and poverty status in all of the countries, and a highly significant correlation between education and survival status.
Table 3: Attendance at Delivery by a Medically Trained Person by Wealth Quintile
|Region |No. of |Regional average |Poorest quintile |Richest quintile|Poor/rich |
| |countries | | | |difference |
|East Asia |4 |53.6 |26.6 |90.4 |63.8 |
|Europe/Central Asia |6 |94.9 |88.4 |99.2 |10.8 |
|L. America, Caribbean |9 |66.0 |43.2 |93.3 |50.1 |
|Middle East, N. Africa |4 |52.5 |33.6 |80.3 |46.7 |
|South Asia |4 |21.5 |7.0 |56.7 |49.7 |
|Sub-Saharan Africa |29 |43.5 |24.2 |77.1 |53.4 |
|All country average |56 |51.6 |32.7 |81.7 |49.1 |
Source: Gwatkin et al, 2004
It is also possible to get a sense of rich-poor differences for other countries (56 in all, including the 10 for which MMRs by wealth quintiles have been calculated) by using the same DHS data for countries on deliveries attended by medically trained persons. This indicator is known to be highly correlated with both maternal and neonatal mortality (and has, in fact, been used to estimate the proportion of maternal deaths in countries lacking other maternal mortality data). Table 3 shows a 49 percentage point difference in the proportion of skilled attendance between the richest and poorest quintiles for all 56 countries for which the tabulations have been made—with the poorest quintile averaging 32.7 percent compared to 81.7 percent for the richest quintile.
Rich poor differences are greatest (64 percentage points) in East Asia, though only four countries are included in the tabulations (Cambodia, Indonesia, the Philippines, and Vietnam) and least (11 percentage points) for Europe/Central Asia (six countries: Armenia, Kazakhstan, the Kyrgyz Republic, Turkey, Turkmenistan, and Uzbekistan). That region also has the highest average level of attendance, 88 percent. South Asia (four countries: Bangladesh, India , Nepal and Pakistan) has the lowest overall level of skilled attendance, 21.5 percent and a 49.7 percentage point poor-rich differential. Latin America has the second highest overall level of skilled attendance (43 percent), but also a comparatively high (50 percentage point) differential between the rich and the poor. Sub-Saharan Africa has a higher overall average for attended deliveries (43.5 percent) than South Asia, which is puzzling given the MMR estimates for Africa. DHS data are available for 29 countries in Africa, suggesting that the issue may be poor quality of delivery care rather than under-representation of regional experience in the tabulations. Rich-poor differences are in the middle range (53 percentage points) in Africa, though attendance for the poorest quintile (25 percent) is second lowest in the tabulations after South Asia. The Middle East/North Africa (MENA) group (four countries: Egypt, Morocco, Jordan, Yemen) is in the middle of the range in terms of the regional average (52.5 percent) and rich-poor differential (46.7 percentage points).
Another way of looking at the poverty link is to group countries by MMR level and then compare them to average levels of per-capita GNP and other indicators in the Pathways framework (female school enrollment, gender equality, paved roads, and governance).[3] In Table 4, 142 low and middle-income countries are divided into five broad MMR groupings (0-30, 31-100, 101-300, 301-1000, over 1000). GDP per capita for the group with the lowest MMRs exceeds $10,000 (dollars in purchasing power parity terms—PPP), compared to just over $1,000 for the countries with the highest MMR levels, a ratio of greater than ten to one. Public health expenditures per capita (also in PPP dollars) are more than ten times greater ($315) in the low MMR group compared to the countries with the highest MMRs ($23).
Table 4: Averages of Pathways Indicators for Countries Grouped by MMR Level
|MMR Level |Number of |GDP per capita |Public health |Female school |UNDP gender |Paved |Effective |
| |countries |(PPP$) |spending per |enroll-ment % |Index |Roads |gover-nance |
| | | |capita (PPP$) | | |% | |
|0-30 |17 |10,629 |315 |72 |.82 |79 |0.41 |
|31-100 |39 |6,952 |168 |74 |.79 |59 |-0.07 |
|100-300 |32 |4,760 |118 |70 |.70 |37 |-0.35 |
|301-1000 |41 |1,610 |30 |43 |.48 |43 |-0.37 |
|Over 1000 |13 |1,025 |23 |31 |.37 |31 |-1.08 |
Source: See Annex 2
Linkages with Other Factors
Large differences between the lowest and highest MMR groups are also seen for other indicators. Female enrollment rates[4] for the lowest group are more than twice as high as for the highest group. The same is true for UNDP’s index of gender equality that combines a range of gender-related indicators (where 1 is the top score). Countries with the lowest MMRs score more than twice as high as those with high MMRs. A difference in the patterns for female school enrollment and gender equality is that the first three MMR groups (those lower than 300) show little variation in the three indicators, while countries with MMRs greater than 300 have significantly lower levels.
There are also large differences between the low and high MMR groups for the percentage of paved roads and effective governance scores. For the low MMR group, 79 percent of roads are paved, compared to 31 percent for the high group. There is not the clustering among the first three groups that was seen for school enrollment and gender equality. In the case of governance scores (which can range from 2.5 as the best score to –2.5,the worst) the low MMR group has the only positive average score (.41), while the high MMR group average –1.08.[5] The table shows simple averages. When population weighted averages were calculated, large countries like China and India significantly outweighed the other countries in their groups. Further, some interesting patterns emerge if we examine the indicators for these two countries, along with several other large countries. Table 5 shows indicators for the eight most populous countries.
Table 5: Pathways Indicators for Large Countries
| | | |Public health exp. per capita|
| | |GDP per capita |(PPP$) |
|Country |MMR |PPP$ | |
|Per capita public health |142 |4.12 |-.733 |
|expenditures (log) | | | |
|Female school enrollment |140 |60.4 |-.661 |
|(percent of school age pop) | | | |
|Gender equality index |116 |.64 |-.878 |
|(1=maximum) | | | |
|Percentage of paved roads |140 |41.4 |-.701 |
|Government effectiveness |139 |-.35 |-.574 |
|Safe water (percent of |140 |75.8 |-.615 |
|households) | | | |
Source: See Annex 2; all correlations are significant at the .01 level.
A similar pattern appears when neonatal mortality data are grouped by level (Table 8) for the indicators. These data are available for only 100 countries, with fewer richer countries as is clear when looking at the group with the lowest NNM rates (0-10), for which the average GDP per capita is $7,734 compared to an average of $10,629 for the lowest MMR group.
Table 8: Averages of Pathways Indicators for Countries Grouped by Neonatal Mortality Rate
| | | |Public health |Female school | | |Effective |
|NNM level |Number of |GDP per capita |spending. per cap. |enroll-ment |UNDP gender |Paved |gover-nance |
| |countries |(PPP$) |(PPP$) | |index |Roads | |
|0-10 |22 |7,534 |256 |74 |.88 |62 |+.15 |
|11-20 |19 |5,461 |124 |71 |.73 |43 |-.22 |
|21-30 |22 |5,116 |77 |66 |.59 |48 |-.27 |
|31-40 |16 |1,458 |35 |47 |.46 |25 |-.74 |
|Over 40 |21 |1,365 |17 |36 |.43 |18 |-.76 |
Source: See Annex 2
For the 21 countries with the highest NNM rates (over 40), average per capita GDP ($1,365) is less than one-fifth that of the low NNM group. Public health expenditure per capita falls off more steadily than GDP per capita. The education/NNM pattern is similar to the one for MMRs, with the first three groups having higher rates than the two lower groups, while the remaining three indicators decline steadily as the NNM levels rise. Overall, the relationships between the NNM groups and the Pathways indicators are quite similar to those for the MMR groups.
Analytical work by World Bank researchers on what would be required to attain the Millennium Development Goals in health has focused on these relationships using multivariate regression techniques to study cross-national variation in the WHO’s 1995 maternal mortality estimates (Wagstaff and Claeson, 2004). One of their main findings is that if interventions that are known to be effective (skilled attendance at delivery, effective referral and management of obstetric emergencies) could be increased from current levels to 99 percent, nearly four-fifths of the 529,000 maternal deaths that occur each year might be averted.
The authors and also address the question of how to ensure that these interventions are implemented, particularly in ways that benefit poor women whom they are not now reaching. A key question is whether additional government spending would help. One school of thought holds that added government spending is likely to have little effect because of corruption and the poor management of public services, but the authors of this study take the view that added government spending could make a difference provided it is combined with improvements in governance and is effectively targeted on the poor. They also call attention to the key role of households, both as consumers of health services and as producers of health outcomes, in ensuring that increases in delivery care will actually reduce maternal mortality and morbidity.
The study employed a simulation modeling process to project possible impacts of increased government spending and improvements in government effectiveness as well as changes beyond the health sector (added economic growth, increased education, improved water supply) on achievement of the maternal mortality reduction target of the Millennium Development Goals. The results are summarized in the chart below.
[pic]
For each of the World Bank’s regions (the same regional breakdown as shown in the DHS tabulations), the chart shows the rate of decline in MMRs from 1990 to 2000 (labeled “current” with darker shading at top of bars) and the added decline that could occur (lighter shading at bottom of each bar) if an additional 2.5 percentage points were added to the annual growth of government health spending as a percentage of GDP, provided that countries achieve a level of effectiveness in governance that is one standard deviation about the mean of cross-national scores in the World Bank’s Country Policy and Institutional Assessment (CPIA) data.[7] The chart also shows (in the cross-hatched area in each bar) the potential contribution of “extra-sectoral contributions” - added economic growth, better roads, quicker growth in female secondary schooling, and improved access to drinking water.
Finally, the chart shows the rates of decline between 1990-2015 required in order to meet the regional MDG goals for reduced maternal mortality (the black line across the graph–5.75 percent annually) and rates that would be required between 2000 and 2015 for regions whose declines were below the required level during 1990-2000 in order to catch up and reach the 2015 goal (the small black squares in the bar).
Comparing the simulated results by region, the chart shows that only one region (the Middle East and North Africa) experienced declines in the MMR during the 1990s at a rate that would enable them to reach the 2015 MDG goal. All of the others need higher rates of increase during 2000-2015 in order to catch up. In three regions (Europe/Central Asia, South and East Asia) a combination of extra-sectoral contributions and additional government health expenditures would bring the rate of decline up to the require level, but in two regions (Sub-Saharan Africa and Latin America/Caribbean) the combination would not bring the rate up to the required level. In the case of SSA, the projected impacts are just too weak to have an impact, and in LAC, the levels of extra-sectoral contributions are currently high enough that added change does not appear likely (though compared to other regions, MMR levels in LAC are already low).
This pessimism about the prospects for SSA has been countered by Jeffrey Sachs and his colleagues at the Millennium Project, who recognize that governance issues have inhibited efforts to reduce poverty in Africa but argue that the region suffers from a "poverty trap" stemming from geopolitical and environmental problems. To escape from this trap, countries need substantially larger investments in physical and human capital (Sachs et al, 2004). The Millennium Project calls for massive increases in donor financial support to poor countries in Africa. Critics of this recommendation caution that the effort is likely to run into substantial bottlenecks in absorbing and effectively utilizing large increases in funding. The problems arising from shortages of health personnel to scale up HIV/AIDS treatment is a case in point (see Marchal et al, 2004).
In making the case for increased health expenditures, the World Bank authors emphasize that they are talking not about across-the-board increases in expenditures but about targeted expenditures to increase the quality and accessibility of the key interventions needed to reduce the high maternal mortality ratios experienced by poor women. The report notes two promising approaches to such targeting. One of these is Marginal Budgeting for Bottlenecks (MBB), which targets added health expenditures on bottlenecks in health system performance. The second is financing through Social Funds, which are targeted on improving the demand and capacities of poor localities and households where these interventions need to occur.
MBB was developed by UNICEF, the World Bank, and WHO in conjunction with efforts to ensure that funds created by debt relief actually benefit the poor (Soucat et al, 2004). It involves the formulation of national or provincial-level medium expenditure plans that allocate newly available resources to achieve an MDG like reduced MMR. MBB starts by using proxy indicators from Demographic and Health Surveys and other data sources to identify potential bottlenecks in five broad categories: (i) gaps in physical accessibility, (ii) human resource bottlenecks, (iii) constraints related to supplies and logistics, (iv) demand and utilization constraints, and (v) bottlenecks due to technical and organizational quality. In Mali, for example, the proportion of deliveries attended by trained staff was used as a performance measure and a costing and budgeting program was developed to reach an attended delivery “performance frontier.” An epidemiological model was calculated to measure and then monitor the impact of increased expenditures on reaching this frontier.
Social Funds (SFs) are “agencies that finance small projects to benefit a country’s poor and vulnerable groups.” Projects are subject to specific eligibility criteria and are generated and managed by communities (Wagstaff and Claeson, 10). Evaluations of SFs in several countries have shown them to be effective mechanisms for channeling funding to poorer communities and improving both the demand for and quality of health services. In Bolivia, for example, it was reported that medicines and essential drugs were more available in SF facilities and, by going beyond traditional approaches to investment in infrastructure, the SF raised the utilization of services and contributed to a reduction in under-five mortality rates in SF communities (Newman et. al, 2002). An evaluation of the impact of social funds on health in four countries (Bolivia, Honduras, Nicaragua and Zambia) found that social fund health interventions had a positive impact on infrastructure quality and on the availability of medical equipment, furniture, and essential drugs. This, in turn, increased utilization of these facilities for critical services, including maternal and child health (Rawlings et al, 2004).
Review of Evidence on Pathways Variables
Other Reproductive Health Risk Factors
Fertility and Contraceptive Use: High fertility rates and low levels of contraceptive use are associated with poor maternal and neonatal health outcomes. Studies show that a woman’s chances of dying as a result of pregnancy and delivery are affected by her age and parity (Maine et al, 1994). Short birth intervals are associated with higher neonatal mortality (Population Reports, 2002). While data on these associations are widely available, the causal relationships underlying these linkages, both biological and social, have proved to be more difficult to untangle.
Poor women generally have higher fertility and lower rates of contraceptive use than non-poor women. Tabulations by wealth quintiles of Demographic and Health Surveys data carried out during the late 1990s for 56 countries provide supporting evidence. Table 9 shows regional averages in poor/rich differentials in the total fertility rate (rates for the poorest and richest quintiles and the difference between the two) from these tabulations. Overall, women in the poorest quintiles average nearly 3 more children than women in the richest quintile, with the largest difference occurring in Latin America/the Caribbean and the lowest in the Europe/Central Asian region.
Table 9: Total Fertility Rates by Wealth Quintile and Region
|Region |No. of |Regional average |Poorest quintile |Richest quintile|Poor/rich |
| |countries | | | |difference |
|East Asia |4 |3.2 |4.4 |2.0 |2.4 |
|Europe/Central Asia |6 |2.7 |3.7 |1.8 |1.9 |
|L. America, Caribbean |9 |3.7 |6.1 |2.2 |3.9 |
|Middle East, N. Africa |4 |4.6 |6.7 |3.9 |2.8 |
|South Asia |4 |3.8 |4.6 |2.6 |2.0 |
|Sub-Saharan Africa |29 |5.4 |6.7 |3.9 |2.8 |
|All country average |56 |4.6 |6.0 |3.2 |2.8 |
Source: Gwatkin et al, 2004
Poor/rich differences in fertility reflect similar patterns of differential use of modern contraception. Table 10 shows average rates of contraceptive use for women in the poorest and richest quintiles for the same regional groupings, along with the percentage point difference between those quintiles. The average difference between the contraceptive use rates for rich and poor women is 18.6 percentage points for the 56 countries, with the largest differentials occurring in Latin America/Caribbean, South Asia, and the Middle East, while the smallest are in Europe/Central Asia and East Asia. Africa has the lowest overall prevalence and lower rates for most quintiles than the lowest quintile in other regions.
Table 10: Contraceptive Prevalence by Wealth Quintile and Region
|Region |No. of |Regional average |Poorest quintile |Richest quintile|Rich/poor |
| |countries | | | |difference |
|East Asia |4 |39.4 |31.3 |41.8 |10.5 |
|Europe/Central Asia |4 |44.3 |38.2 |48.1 |9.9 |
|L. America, Caribbean |9 |47.1 |33.1 |56.8 |23.7 |
|Middle East, N. Africa |4 |34.2 |22.4 |45.0 |22.7 |
|South Asia |4 |32.8 |22.9 |45.8 |22.9 |
|Sub-Saharan Africa |29 |13.0 |6.5 |25.1 |18.6 |
|All country average |56 |26.7 |18.3 |36.9 |18.6 |
Source: Gwatkin et al, 2004
The two most commonly recognized factors linking fertility/contraceptive use and maternal/neonatal mortality rates are that risks are higher (1) when births occur at a very young age, and (2) when birth intervals are short. A third line of argument is that older, higher-parity women are more at risk.
Rani and Lule (2004) analyzed DHS data for 12 poor countries and found young women from the poorest households were more likely than those from the richest ones to be married by age 18 and to have had at least one child by that age. These women were also more likely to have had a mistimed birth and were less likely to practice family planning, use maternal health services and have knowledge about prevention of sexually transmitted infections. Additional DHS tabulations of age-specific fertility rate for women aged 15-19 (Table 11) show that the rate for women in the poorest quintile is more than twice that of women in the richest group for 55 countries as a whole (Egypt was not included in this tabulation), and nearly five times greater for poor women in Latin America and the Caribbean. The poor/rich differential is lowest in the three MENA countries and in Europe/Central Asia and East Asia, which also have lower regional average rates.
Table 11: Adolescent Fertility Rates by Wealth Quintile and Region
|Region |No. of |Regional average |Poorest quintile |Richest quintile|Poor/rich |
| |countries | | | |difference |
|East Asia |4 |46.0 |76.5 |15.8 |60.8 |
|Europe/Central Asia |4 |52.7 |73.0 |31.3 |52.7 |
|L. America, Caribbean |9 |94.7 |172.6 |36.9 |135.7 |
|Middle East, N. Africa |3 |62.7 |111.7 |99.0 |12.7 |
|South Asia |4 |108.8 |146.3 |56.0 |90.3 |
|Sub-Saharan Africa |29 |131.9 |169.6 |79.5 |90.0 |
|All country average |55 |106.5 |148.6 |62.6 |86.1 |
Source: Gwatkin et al, 2004
Research on the links between early childbearing and poor maternal and neonatal health outcomes have concluded that most of the adverse health consequences (delivery complications and maternal mortality, prematurity and higher perinatal death rates) of teen pregnancy are associated more with socioeconomic factors than with the biological effects of age (Makinson, 1985; Miller, 1991). These studies are for populations in richer countries. For developing countries, women aged 15-19 are twice as likely to die from childbearing as women in their 20s, and women under age 17 face especially higher risk. Young women who become pregnant are at risk of obstructed labor if they have not grown to their full height or pelvic size, and are also more likely to suffer from eclampsia, which threatens them and their babies (Upadhyay and Robey, 1999). In her review of research on the consequences of adolescent childbearing in India, Jejeebhoy (1996) reports that infants of adolescent mothers are more likely to suffer higher perinatal and neonatal mortality, that levels of anemia and complications of pregnancy are higher for adolescent mothers, but that they are less likely to obtain antenatal care and trained attendance at delivery than older mothers.
Births that are too closely spaced are also associated with higher perinatal mortality, and may be a risk factor for maternal mortality (Miller, 1991). Infants whose births were spaced more than two years have been less likely to be premature or suffer from low birth weights. Analysis of DHS data for 18 countries (and encompassing more than 430,000 pregnancies) showed that children born three to five years after a previous birth are more likely to survive (Venugopal and Upadhyay, 2002). A cross-national study of 18 Latin American countries found that women with interpregnancy intervals of less than six months had a higher risk of maternal death and complications of delivery than those conceiving at 18 to 23 months, and also that intervals greater than 59 months were associated with higher risks of eclampsia (Conde-Agudelo and Belizan, 2000).
The association between short birth intervals and poor maternal and neonatal health has been ascribed to the so-called “maternal depletion syndrome,” in which maternal nutrition plays a critical role (Winkvist et al, 1992; King, 2003). As the biological competition for nutrients increases during pregnancy, an inadequate supply contributes to poor fetal development and may also be factor in higher maternal mortality. King found that “maternal depletion of energy and protein resulting from short inter-pregnancy intervals or early pregnancies leads to a reduction in maternal nutritional status at conception and altered pregnancy outcomes” (King, 2003: 1735s).
While there is a biological basis for the close associations between early or too closely spaced pregnancies and poor maternal and neonatal health, causal explanations need to address the possibility that other factors may influence both at the population level. In a study for rural Bangladesh on the relationship between childbearing and maternal survival, Menken and colleagues found no significant effects of early or closely spaced pregnancies on these outcomes once other factors (education, height) were controlled. But while there were no significant differences in the risk of dying during delivery between births that were early, closely spaced, or numerous, their findings do suggest that lifetime childbearing does affect a woman’s survival chances. “Each time a women has a child, she faces an increased risk of dying in the relatively short period (two to three years) after that birth. Thus a woman who bears seven children has seven chances of succumbing to this risk, whereas a woman who bears two children has only two chances” (Menken et al, 2003). The authors refer to this as “extended maternal risk”. Their findings complement earlier work by Ronsmans and Campbell (1998) who found no evidence that short birth-to-conception intervals affected the risk of maternal mortality.
The interplay of biological and contextual factors is also emphasized by Zabin and Kiragu (1998) in their review of the health consequences of adolescent sexual and fertility behavior in Sub-Saharan Africa, in which they document the “immense reservoir of suffering cause by childhood marriage and immediate post pubertal childbearing among girls given in marriage at ages as young as ten or 12.” They emphasize that analysis of the role of age in these adverse outcomes should also take into account other proximate causes and cite two examples: radical forms of circumcision and cephalopelvic disproportion, which is a factor in obstructed deliveries and which, if not adequately managed, can cause obstetric fistulae. Circumcision often leaves so much scar tissue that that the first child is lost - this health effect is related to age because teenagers are more likely to be primiparous than older women but could still occur if childbearing is delayed to a later age. Cephalopelvic disproportion occurs more frequently among malnourished young girls whose bone development is incomplete, leading the authors to conclude that “the root causes of this sort of tragic delivery are malnutrition and a lack of access to or use of professional care”.
Unintended Pregnancies and Abortion: In addition to the risks associated with the number and spacing of births, there are those associated with unintended pregnancies, including those that are unwanted and those that are mistimed. While interpretation of data on pregnancy intentions continues to be debated, they have proved to be valuable in addressing pregnancy-related health risks. Women with unintended pregnancies are less likely to seek prenatal care, more likely to use alcohol and tobacco during pregnancy, and more likely to suffer physical abuse and violence (Santelli at al, 2003). Many unintended pregnancies end in abortion. Where abortion is safe, abortion-related mortality and morbidity are lower than birth-related mortality and morbidity. However, WHO estimates that 20 million unsafe abortions occur annually, mostly in developing countries. This means that one out of ten pregnancies is terminated by an unsafe abortion, with a ratio of one unsafe abortion per seven live births. South Central Asia accounts for the largest share of unsafe abortions, followed by Africa and Latin America (Ahman et al, 2003).
Unsafe abortion is a significant cause of maternal mortality and morbidity. Nineteen million unsafe abortions are estimated to have taken place during the year 2000, 98 percent of them in developing countries. Over the period 1995-2000, an estimated 78,000 maternal deaths, approximately 13 percent of all maternal deaths, were attributable to unsafe abortion. AbouZahr and Ahman (1998) estimated that the abortion-related mortality risk was at least 15 times higher in developing areas and that in some regions it may be 40-50 times higher than in more developed regions. While unsafe abortion and abortion-related mortality risk are much greater in poor countries, comprehensive data on rich-poor differences in risk are not available. However, the fact that poor women have higher fertility and lower contraceptive use would imply that they are at much greater risk. There is also evidence, still incomplete, that the incidence of unsafe abortion and resulting mortality may be rising among unmarried adolescent women in urban areas of developing countries.
Violence, Conflict and Refugee Settings: Mothers and children in conflict settings, including those who are refugees, are subject to greater risk of poor reproductive and neonatal health outcomes. UNFPA estimates that women and children account for 75-80 per cent of the world’s 37 million refugees and displaced persons at risk from war, famine, persecution and natural disaster; that 25 per cent of this population at risk are women of reproductive age, and that one in five is likely to be pregnant (UNFPA, 2001; Save the Children 2002). Maternal mortality for women in conflict zones of the Darfur region of the Sudan is higher than the already high national average of 1500 for the country as a whole (Collymore, 2004).
Women and children living in such circumstances are exposed to a range of risk factors, including gender-related violence, rape, poor nutrition, psychological trauma, abuses in camp settings, and a lack of access to basic health care. One of the “collateral” effects of conflict is the destruction or breakdown of basic services. Conflict also creates obstacles for relief agencies, further exacerbating health problems and making it difficult to measure outcomes. Reviews of available information on the reproductive health in war-affected populations identified a range of risk factors that depend on what was going on in specific settings at specific times (McGinn, 2003). Numerous case studies have documented deterioration in maternal and neonatal health indicators for refugees in conflict settings, including Congo, Guatemala, Sierra Leone, and Afghanistan (Ward, 2002). A 1999-2000 study of maternal mortality among Afghan refugees in Pakistan, revealed a rate was 50 percent higher than the rate for Pakistan (291 vs. 200), though lower than the estimate for Afghanistan itself (820). Most of the maternal deaths in camps were preventable and resulted from lack of access to emergency services (Bartlett et al, 2002). However, research on reproductive health in refugee camps in a number of countries (Sudanese and Somalis in Ethiopia, Sudanese and Rwandans in Uganda, Burundians and Congolese in Tanzania) found better outcome indicators (including maternal and neonatal mortality) than for refugees’ home countries or their host countries, suggesting that once reproductive health services are established in camps they can have a positive impact (Hynes et al, 2002).
Research on the links between gender violence and reproductive outcomes shows that women who experience violence have poorer outcomes, including more unintended pregnancies, low birth weight, fetal wastage and infant deaths (Nasir and Hyder, 2003; Heise et al, 1999). While none of these reviews distinguished between poor and non-poor women in these conditions, it is safe to assume that poor women represent a significant, if not substantial proportion of the total. Violence in pregnancy is another risk factor and is a cause of poor delivery outcomes. Heise and colleagues (1999) report that violence in pregnancy accounted for 16 percent of low birth weight deliveries in Nicaragua, and that violence may be responsible for a sizeable but under-recognized proportion of pregnancy-related deaths on the Indian subcontinent. The risk of pregnancy-related deaths associated with violence was substantially higher for pregnant teenagers than for other age groups.
Infections and Other Risk Factors: In addition to malnutrition, other risk factors may also trigger or exacerbate obstetric complications (Koblinsky, 1995; Tinker, 2000). In addition to poor nutrition and violence (discussed above) these include infections (sexually transmitted infections and HIV/AIDS, malaria, tuberculosis, hepatitis), substance abuse, and harmful practices such as female genital mutilation (FGM). A major gap in many programs aimed at prevention of the transmission of HIV infection from mothers to children is treatment for mothers after they have delivered, since most programs seek to save the life of the child but not the mother - ignoring evidence that survival chances for children are lower for those whose mothers have died, not to mention the basic ethical issues involved in such choices.
Some of the world’s highest MMRs are found in countries where the practice of FGM is widespread. Among the long-term consequences of FGM are increased risk of obstructed labor, delayed delivery, and increased risk of stillbirths (PATH, 1997). Another consequence is increase vulnerability to sexually transmitted infections (STIs) and HIV/AIDS. STIs are known to affect delivery and pregnancy outcomes, including premature birth and intrauterine growth retardation (Haberland et al, 1999). The link between HIV/AIDS and maternal mortality has been recognized in the latest estimates of MMRs, which include HIV prevalence in the estimating equation for the proportion of deaths in reproductive ages that are considered “maternal”. In South Africa, where the HIV prevalence rate is one of the highest in the world, AIDS-related respiratory infections (pulmonary TB and pneumonia) are important factors in pregnancy-related mortality (Kruger, 2003).
Malarial infection is another cofactor in pregnancy-related deaths of mothers and newborns. Meremikwu (2003) notes that malaria is typically a more severe disease in pregnant women and is a major contributing factor to maternal mortality in malarial areas of Africa. Malaria is associated with maternal anemia, which is another cofactor in maternal death. Santosi (1997) adds that malaria is associated with chronic health problems that frequently become acute during pregnancy and are associated with higher rates of maternal mortality and morbidity in malarial areas. Etard and colleagues (2003) report that malaria is a probably cofactor in seasonal swings in maternal deaths in Senegal. One of the reasons why Malawi’s MMR remains so high despite a comparatively high rating on education is its high prevalence of malaria (55-80 percent during the rainy season) along with severe anemia in pregnant women.
Table 12: Selected Findings on Other Reproductive Health Risks
|Country, location |Type of study |Citation |Findings |
|Latin America |Cross-national |Conde-Agudelo and |Women with interpregnancy intervals of 5 months or less had higher |
| |analysis |Belizan (2002) |risks for maternal deaths and complications; women with intervals |
| | | |greater than 59 months had higher risk of eclampsia. |
|Bangladesh |Matlab surveillance |Menken et al. (2003) |No significant effects of early or closely spaced pregnancies once |
| |data | |other factors (education, height) controlled; however, the overall |
| | | |number of deliveries affects survival chances |
| | | | |
| | |Ronsmans and Campbell |No evidence that short birth-to-conception intervals increase |
| | |(1998) |maternal mortality |
|Global |Reviews |Santelli et al (2003) |Unwanted pregnancy and unsafe abortion contribute to high MMRs, even |
| | |Ahman et al (2003) |in countries where abortion is legal. |
|Africa |Review |Zabin and Kiragu (1998)|Biological and contextual forces interact in determining the adverse |
| | | |effects of adolescent sexual and fertility behavior; malnutrition and|
| | | |lack of access to care exacerbate these problems, which affect both |
| | | |married and unmarried adolescents |
|Refugee populations|Reviews |McGinn (2003), |Refugees and women who experience violence experience worse MNH |
| | |Ward (2002), Heise et |outcomes; RH services for refugee populations can mitigate these |
| | |al (1999) |risks. |
|Africa |Reviews |PATH (1997), Kruger |Harmful practices (FGM) and infectious diseases (TB, malaria and |
| | |(2003), Meremikwu |HIV/AIDS) increase the risks of obstetric complications and higher |
| | |(2003) |MNM rates. |
Household and Community Factors
As noted in the introduction to the Pathways framework, households and communities are important but often-neglected actors in the effort to improve maternal and neonatal health outcomes. Building on the Pathways framework, the World Bank analytical report on the achievement of MDGs described in the previous section looked at households both as consumers of health services and as producers of health outcomes. The report cited two main obstacles that accounted for underutilization of key health interventions by poor households: (1) the prices that households pay for those interventions; and (2) lack of knowledge about those interventions and their importance to the health of the household members.
In addition, the Pathways framework recognizes that intra-household relationships, particularly gender relations, affect utilization patterns.
Gender deserves special attention because of the inequity in social relationships that restricts the rights of women to make decisions for themselves or to have fair access to household assets; and the greater the inequity, the greater the obstacle to poor women’s access to life-saving interventions. In Indonesia, women’s control over assets, which were found to be highly correlated with education, positively affected their chances to get prenatal and delivery care (Beegle et al, 2001). The authors note that while the government has made progress in reducing price and distance barriers to obtaining care, the inequitable distribution of power in social relationships remains a significant obstacle. Another Indonesian study based on interviews about the circumstances surrounding more than 100 maternal deaths found “many families whose personal poverty excludes them from even considering attempts to gain access to emergency obstetrical care” and characterized these women as “too poor to live” because families failed to take steps to seek care for complications because they feared the cost could be greater than they could bear (Islander et al, 1996: p. 80).
The case study in Box 1 illustrates household and community level obstacles that prevent poor women from obtaining life-saving interventions when they suffer an obstetric complication. The case is taken from an oral autopsy that was conducted in Bangladesh during the early 1990s and which played an important role in shaping the development of a maternal health strategy for Bangladesh later in that decade (Blanchet, 1991). It describes the plight of a poor woman who died as a result of complications during her tenth delivery. While the midwife diagnosed her condition as “extreme anemia”, and the symptoms also suggest septicemia (once a leading cause of maternal deaths, but now rare in richer countries), the case reveals much more about the complex weave of causes of Safar Banu’s death.
What we observe is a complex interplay of forces that undermined Safar Banu’s survival chances, as well as those of poor women around the world, in the face of life-threatening obstetric emergencies. These include:
• Her subordinate status in the household, and her willingness to endure this;
• The fact that she was not using contraception, and that she had no voice in this despite her concerns about having a tenth child;
• Her poor nutritional status (being the last and least fed in her household);
• The lack of prenatal care despite her experiencing swelling and dizziness;
• The fact that the household could not afford to pay for medicines;
• The husband’s reliance on the advice of the traditional healer, who failed to provide adequate care, and his refusal to listen to the midwife;
• The misinformation about her condition among those around her, which delayed treatment at the clinic; and
• Transport costs, for which the family had to sell land (creating a legacy of resentment) and which caused further delay in her treatment.
No single factor was the “cause” of her death, but the combination of circumstances described in the case capture many of the household/community-level factors mentions earlier: gender relationships in the household, household behaviors relating to fertility regulation and nutrition, lack of information, and household poverty and inability to pay for care. Further discussion of the cost issue is found in the next section, which addresses health system failures.
Gender relationships played a key role in that the traditional birth attendant and Safar’s mother knew that something was seriously wrong but would not go against the husband’s preference to rely on the local quack. They delayed consulting a midwife until it was too late, and even then the husband discounted the midwife’s advice.
The case highlights the importance of community factors (the gender system, misinformation, lack of community support) and the potential of community mobilization to contribute to the reduction of maternal and neonatal mortality. A number of community mobilization initiatives are already proving themselves to be effective, including more involvement of community health workers to create awareness about complications of delivery and demand for effective management of emergencies (discussed in more detail in the section on the health workforce), community support for transport (see the transport section), and support for mothers during delivery. An example of the latter is the promotion of home-based life-saving skill training. This approach involves both families and their communities to support birth preparedness and involvement of key decision makers to take action when it is needed and reduce delays in reaching referral centers (Silbey et al, 2001). Community-level safe motherhood committees and local health volunteers can provide effective support for such initiatives, particularly when they are linked effectively to the health system. Community groups were mobilized to conduct maternal death audits in Malaysia, enabling them to identify and address the conditions that contributed to the death of a mother in their community (Koblinsky et al, 1999).
A frequently cited strategy for reducing distance-related delays for women who live in remote areas is to bring all of them to delivery centers or maternity waiting homes prior to delivery. In her review of experience with this strategy, Figa-Talamanca (1996) cites the experience of Cuba, which reduced its maternal mortality rate from 118 to 29 per 100,000 between 1962 and 1989. To facilitate accessibility to facilities with capacity to manage obstetric emergences, Cuba located waiting homes near these hospitals for women who lived in remote areas. Community organizations were mobilized both to build and run the waiting homes and to encourage their utilization, for example by providing childcare for pregnant mothers during the time of delivery. Waiting homes were also a key factor in Honduras, where the MMR dropped 40 percent in seven years (from 182 to 108) after the introduction of a multi-pronged strategy that included upgrading of facilities and staff capacity as well as community organization and infrastructure development (Danel, 2003). Figa-Talamanca's review also identified a number of factors that need to be addressed in introducing waiting homes, including the acceptability of the practice in societies where home delivery is the norm, the question of selecting women at greater risk of complications (since any delivery has the potential for complications), and the range of services that such centers provide and their costs to families.
Broader community efforts such as micro-enterprise and credit programs targeted at poor women can reduce the gender disparity by ensuring that women have control over the money they earn and ensuring that information and education networks reach beyond the traditional boundaries and restrictions faced by poor women. In Africa, social action programs targeted to poor households have simultaneously provided funding for family planning and reproductive health services and created paying jobs for women. An evaluation of one such program in Malawi found that it had directly improved women’s reproductive health outcomes and had indirectly improved their status in the family, through its woman-focused educational, credit and employment initiatives (Marc et al, 1995). In Bangladesh, women’s participation in rural credit programs impacted positively on their demand for health care (Nanda, 1999). Women’s empowerment, including higher control over resources, was addressed in the design of these programs.
Health System Failures
It is widely recognized that improved maternal and neonatal health outcomes require a continuum of care, from the household and community through the referral process to an effectively functioning health system. Poor health system performance is one of the reasons why poor women do not get life-saving interventions when they experience an obstetric complication. In a cross-national review of health system failures that contribute to high maternal mortality, Sundari (1992) identified several critical problems: shortage of trained personnel, lack of equipment and facilities (including consumables such as blood products and antibiotics), and poor patient management. These problems are also pinpointed in the World Bank assessment of obstacles to the achievement of health-related MDGS. In addition to the cost issue mentioned above, the report also mentions the quality of care as reflected in health workforce performance, availability of medicines, as well as inadequate and ineffective public spending. On the last point, a review of the benefit-incidence of public spending on health care in Africa by Castro-Leal and colleagues found that subsidies are poorly targeted on the poor and typically favor the better off, a phenomenon that is not limited to Africa.
Costs to consumers include payments for care and transport as well as opportunity cost, for example lost wages or time that would have otherwise been spent in household production. Even when publicly provided care is nominally free, under-the-table or side payments may be required to obtain services or medicines. Evidence suggests that the poor already pay a lot out of pocket, particularly for medicines, and could get better care for their money if health system performance were improved (Nahar and Costello, 1998).
Health reformers have argued that cost recovery could improve the sustainability, quality and equity of health services by bringing payments into the open, rationalizing the use of services (for example, through incentives to use the level of care appropriate for the treatment required), and by giving providers control over resources and the incentive to use resources to improve quality. If combined with effective exemption schemes for the poor, cost recovery could also re-direct public subsidies that typically benefit the rich more than the poor.
Evaluating the impact of cost recovery schemes on MNH outcomes for the poor is not an easy task because of the many contextual and institutional influences that shape their implementation. Gilson’s (1997) review of experiences with user fees in Africa, where many countries introduced them during the 1980s, reports that:
• By themselves, fees tend to dissuade the poor more than the rich from using services;
• Fees, especially for community managed schemes like the Bamako Initiative that focus on medicines, may be associated with quality improvements that offset some of that negative impact on utilization;
• The equity impact depends a lot on the nature of the payment mechanism—direct payments have a more negative effect;
• Fees did not appear to generate sufficient revenue to enable the hoped-for reallocation of public subsidies to the poor;
• Exemption schemes did not really protect the poor, and often helped other groups (e.g. public sector employees) more than the poor;
• Differential geographic implementation of fee schedules may only exacerbate geographic inequalities in access to care; and
• There is very limited evidence on the impact of fees on the budgets of poor households and their demand for health care.
Available research on the response of poor households to fees shows that there is a substantial drop in utilization immediately after their introduction, followed by a partial return to pre-fee levels, and that poor households are somewhat more sensitive to price changes. Kutzin (1995) reports on research findings from Zimbabwe showing that intensified enforcement of user charges contributed to a 30 percent decrease in maternal health services compared to the year before enforcement, and that the numbers of babies born before reaching hospital increased by 4 percent as a possible result of mothers seeking to avoid per diem hospital charges. Mwabu and colleagues (1995) found that attendance at public clinics in Kenya dropped by 50 percent during the period when cost recovery was enforced (which lead to a suspension of fees), while Newbrander and colleagues (2000) found that poor consumers delayed seeking care more than rich ones in order to avoid costs. This response is particularly troubling in the case of maternal and neonatal care since such delays contribute to higher mortality. That review also noted that poor consumers tended to be unaware of exemption schemes or about how to take advantage of them even when such schemes existed. Nanda (2002) reports that exemption schemes do not include all reproductive health services and that their implementation is vulnerable to subjectivity and distortion by providers.
Removal of financial obstacles to care has been a key policy change in countries that have successfully reduced MMRs. Both Malaysia and Sri Lanka provided maternity care free to clients who could not pay for services (Pathmanathan et al, 2003). Care was provided in public facilities, whose quality was improved in terms of accessibility, health worker performance, and availability of medicines. Countries that still have higher MMRs are attempting to reduce financial barriers using alternative financing options, though their impact on maternal mortality has yet to be measured. Bolivia's National Maternal and Child Health Program has increased coverage of maternal and child health care, though in its initial phase the poorest quintile appeared not to benefit because location and social exclusion proved to be as serious an obstacle as cost, so that special outreach to them was added in a later phase (Seoane et al, 2003). A study of the comparative impact of user fees versus community-based financing on utilization of services in Niger found that the community schemes had a more positive effect (Diop et al, 2000), though attention to the coverage of maternity care in such schemes needs to be watched. In Rwanda, the evaluation of an experimental community insurance program found that women who are members of a prepayment scheme were three times more likely to deliver with professional assistance than nonmember women, who were more likely to delivery at home and alone (Schneider et al, 2001). Mexico's PROGRESA program targets cash payments to poor households for education, nutrition and preventive health care and has had a positive impact on the nutrition of pregnant women and reduced low birth weight in newborns (Gertler, 2000). PROGRESA is also reported to have given poor women more say in household decision making and control over household resources, as well as increased schooling for poor children (Skoufias and McClaferty, 2001).
Poor health workforce performance is another critical obstacle to poor women’s access to life-saving interventions, particularly in view of the key role that skilled attendance at delivery is known to play in the reduction of maternal and neonatal mortality. The large rich-poor differentials in skilled attendance documented in Table 3 suggest that even when countries have skilled attendants in their health workforce, these attendants are not serving the poor. Sundari’s (1992) review noted both the scarcity of such workers, particularly in rural areas, as well as lack of training and incentives to motivate workers. Health workforce problems are further exacerbated by losses of staff who die from AIDS or emigrate. An effort to upgrade delivery care in rural areas of Bangladesh during the 1990s was hindered by the fact that attendants were not available in facilities even when they were posted to them. Professional care was provided in the case of scheduled normal deliveries, but not available for emergencies that could occur at any time. In their study of contextual determinants on maternal mortality, Midhet and colleagues (1998) found that lack of trained staff at peripheral health facilities and access to those facilities accounted for most of the variation in maternal mortality in sixteen rural districts of rural Pakistan. Kwast (1996) reviewed performance issues in several countries (Bolivia, Guatemala, Indonesia and Nigeria), and also found that appropriate training of front-line staff combined community outreach and empowerment of women to recognize the importance of their own reproductive, maternal and neonatal health problems was critical to the reduction of physical and financial obstacles to care.
In its discussion of links between health systems problems and efforts to improve maternal and neonatal health outcomes, the interim report of the Millennium Project Task Force on Maternal and Child Health reports that deficiencies in health workforce capacity are a major bottleneck. It calls for development of health workforce strategies that expand the supply of critical skills, including skilled birth attendants (Freedman et al, 2004). The recommended approach includes more effective involvement of traditional healthcare workers (including traditional birth attendants—TBAs), on whom many poor women still rely for reasons of cost, convenience, services offered and trust. Up to now, TBA-based maternal care programs alone have failed to reduce MNM, in large part because TBAs were not linked to a functioning health system. Current thinking about building health workforce capacity calls for more effective involvement of TBAs in skill-attendant strategies, particularly strengthening their role as advocates for skilled care and linking them more effectively to a functioning referral system (WHO, 2004). Experience in Malaysia and Brazil has demonstrated that effective involvement of TBAs in community mobilization, awareness and demand creation, and referral of emergencies can be effective, particularly as countries move from situations in which delivery by a skilled attendant is rare toward fully functioning systems with deliveries by professional attendants in comprehensive obstetric care facilities (Koblinsky et al, 1999).
Decentralization of health system management is another tool that health reformers have employed in an effort to improve quality and accountability of front-line services. The rationale for such reforms is that if local authorities, and the communities that they serve, have greater control over human and financial resources, the system should be more responsive to local needs. Evaluation of experience of the impact of decentralization on poor women’s access to maternal health care is not easy because many forces are at work. In many cases, local managers are responsible for managing care but do not have real control of people and money. Resource allocation algorithms typically rely on formulas based on such measures as the number of hospital beds in an area rather than health needs, and control over personnel decisions remains centralized, so that local managers end up with the worst of both worlds. Further, decentralization often weakens specialized technical units (in this case, maternal health units that might have been providing critical technical leadership prior to decentralization) at the central ministry, and there is not enough expertise at the local level to pick up the slack.
On the positive side, targeted strategies such as the Marginal Budgeting for Bottlenecks described earlier have focused on training and incentives to improve workforce performance. A few African countries have experimented with the use of funding freed up through debt relief to create special incentive funds to support the redeployment of health workers to where they are needed. For example, in Mauritania a MBB bottleneck-identification exercise found that a lack of nurses and midwives in rural areas was a key obstacles to achievement of better maternal and child health outcomes. In response funds that were freed up through debt relief were used to the create incentive mechanisms to get staff to work in rural areas (Soucat et al, 2002). A similar effort in Mali focused on geographic availability of care as well as staffing. Safe motherhood strategies in a number of countries are focusing resources on the training and deployment of skilled attendants in facilities that serve poor women.
The lack of consumables and equipment in facilities is another bottleneck. Even when an emergency is recognized and the woman suffering it has been referred to a treatment facility, she may still die if, in the case of a hemorrhage, blood products are not available, or antibiotics in the case of infection. Facilities where poor women might go are chronically short of consumables in many countries, though in some cases targeted cost sharing arrangements such as the Bamako Initiative have been able to overcome this obstacle (Gilson, 1997). That model relies on community management to ensure that revenues are used to address quality constraints and ensure local accountability.
Performance problems seldom occur in isolation from each other. McPake and colleagues (1999) have documented the behavioral responses of public health workers in Uganda to poor incentives, scarcity of drugs and lack of management capacity. The situation encouraged drug leakage, side payments to obtain drugs, and understaffing of public health centers. Quality of service might have improved for consumers who could afford to pay, but probably at the expense of poorer ones who had to rely on “free” publicly provided treatment. A summary of findings on financial barriers, costs and cost recovery is presented in Table 13.
Table 13: Selected Findings on Health System Issues
|Country, |Type of |Citation |Finding |
|location |study | | |
|Global |Review |Kutzin (1995) |Studies in many countries have shown that poor people are |
| | | |more likely to be put off by price increases; travel costs|
| | | |have a similar deterrent effect. |
|Africa |Review |Newbrander et al (2000)|There were significant inequalities in access to health |
| | | |care under the user fee systems studied. The poor delay |
| | | |and wait longer for care, and often pay the same fees as |
| | | |the non-poor. |
|Africa |Review |Gilson (1997) |Poor people reduce utilization more than the rich when |
| | | |fees introduced, but quality improvements can improve |
| | | |utilization; administrative costs often high relative to |
| | | |fees, and exemptions schemes are hard to manage. |
|Zimbabwe |Impact Assessment |Reported in Kutzin |User charges brought 30 percent reduction in use of |
| | |(1995) |maternal health services and increased births outside of |
| | | |hospitals by 4 percent. |
|Malaysia, |Case studies |Pathmanathan et al |Removal of financial barriers a key policy change in |
|Sri Lanka | |(2003) |countries that have reduced maternal mortality |
|Global |Review |Ensor and Witter (2000)|Unofficial, under-the-table fees are charged even when |
| | | |services are nominally free |
|Global |Review |Freedman et al (2004) |Health workforce limitations are a critical bottleneck in |
| | | |efforts to improve MNH. |
|Global |Reviews |Koblinsky et al (1999);|TBAs can be effectively involved in skilled-attendant |
| | |WHO (2004) |strategies for community mobilization, demand creation, |
| | | |and as referral agents. |
|Africa |Case study |Soucat et al (2002) |MBB is an effective mechanism for targeting debt-relief |
| | | |funds to overcome health system obstacles. |
Other Sectors
Transportation: Delays in reaching a treatment facility pose key life-threatening obstacles for women who experience an obstetric emergency. Such delays can be the result of physical accessibility factors such as distance to a facility, the availability and cost of transport, and the condition of roads, all of which affect the time required to get a mother to a facility once the decision to seek care has been made (Thaddeus and Maine, 1994). A number of countries with high MMRs (Afghanistan, Pakistan, Nepal) also have large segments of their population living in remote areas that have poor road links to facilities that can provide life-saving interventions. In Zimbabwe, unavailability of transport is reported to have been a factor in 28 percent of deaths in a rural area that was studied (Fawcus et al, 1996). In the case of hemorrhage, 50 percent of deaths were attributable to transport-related delays.
Measurement of road networks is complicated by differences in geographic settings and population distribution. The index of road quality in Table 8 above and used in the World Bank's analytical work on attainment of MDGs standardizes the proportion of the country's roads that are paved by dividing that proportion by the area of the country. While the index provides an approximation of variability in road access across countries, caution is needed in interpreting it for countries with a large land area and whose populations (and roads) are concentrated along coasts or in a smaller segment of the total land area (China, India, Brazil, and Nigeria, for example).
Though there are few studies that focus specifically on these factors, and hardly any that link poverty to adverse MNH outcomes for poor women, the available evidence suggests that transport is a factor in the delays that threaten the lives of poor women. In their review of health-seeking behavioral responses to cost recovery, Newbrander and colleagues (2000) found that poor people in Tanzania traveled an average of over 60 kilometers for care, whereas the non-poor traveled only 15 kilometers. There were similar findings in Kenya. The authors mention several possible reasons, including the likelihood that the non-poor have their own transport and that the poor travel farther in order to attend a facility where fees would be waived. In their review of obstacles to health care, Ensor and Cooper (2004) found studies that reported transport accounting for 28 percent of all total patient costs in Burkina Faso, 25 percent in northeast Brazil, and 27 percent in the United Kingdom. In Bangladesh, transport was reported to be the second most expensive item for patients after medicines. To quote from one of the focus groups in that study: “The hospital is far away and it costs a lot to travel there. We can easily buy medicines from the village doctors with this money. We spend money to go to the hospital but we don't even get medicines there, so why should we go to the hospital?” (CIET-Canada, 2001, p. 38).
Country-level poverty analyses conducted by the World Bank have shown that the quality rural road networks is a factor in the social and economic isolation of the rural poor. Research on the impact of improved rural road networks has focused mainly on travel time. For example, a poverty assessment for Guatemala found that road closures were a major constraint on access to schools, work and markets, and that households in the poorest income quintiles were much more affected (45 vs. 12 percent) than the richest (World Bank, 2003). Villagers identified “giving birth” as a risk because mothers could not reach health centers due to inadequate road access, particularly during the rainy season. Improved roads cut farm-to-market travel time from more than 10 hours to 1-2 hours. While the impact on access to emergency obstetric care was not assessed in the research, these reductions in travel time have clear implications for reducing distance related obstacles to timely management of emergencies. An earlier study of the impact of roads on access to services in Morocco found similar patterns. The road project contributed to clear gains in women’s utilization of health services (World Bank, 1996). Similar results were shown for girls’ school attendance, which increased much more (40 % vs. 10%) than for boys, for whom lack of roads posed a lesser obstacle than for girls.
The quality and availability of roads is only one facet of the transport obstacles. The availability, type and cost of transport is clearly another. Research conducted by African partners in the Prevention of Maternal Mortality Network also found that poor roads, lack of vehicles and high transport costs were major causes of delay in deciding to seek and in reaching emergency obstetric care (Samai and Sengeh, 1997). That study reports the positive impact of efforts to improve transport and communication, along with community support and education activities, on the numbers of women getting treatment for obstetric emergencies, with consequent reductions in maternal deaths in the project area. Similar results have been reported for Nigeria (Essien et al, 1997), Uganda (Lalonde et al, 2003), and northwestern Tanzania (Ahluwalia et al, 2003). These studies emphasize the role of community organization in planning for emergencies when they occur, including preparation of delivery plans, mobilizing resources through community funds, reducing transport costs, and strengthening the referral chain.
Mention of the referral chain reminds us that improved roads and transport may in some cases be necessary to reduce delays in management of obstetric emergencies but their overall impact on outcomes depends on many other factors. This is illustrated by research in India over a ten-year period when improved roads and transport led to increases in the number of women reaching hospital but little reduction in case mortality rates: the improved roads made it possible for women living farther away to get to the hospital but they were already in a moribund condition. Poverty, social inequality and gender conditions (including very early marriage) in their villages eroded the positive impact of improvements in infrastructure and treatment facilities (Pendse, 1999). A summary of findings on obstacles relating to distance and transportation is presented in Table 14.
Table 14: Selected Findings on Transport Interventions
|Country, |Type of |Citation |Finding |
|Location |study | | |
|Global |Review |Thaddeus and Maine, 1994 |Distance and road conditions an obstacle to emergency |
| | | |care; better roads can help, but financial obstacles |
| | | |also need to be addressed |
|Global |Review |Ensor and Cooper, 2004 |Community health insurance schemes that include |
| | | |transport costs in benefit package have contributed to |
| | | |better access to care |
|Rural Guatemala |Poverty study |World Bank, 2003 |Long travel time to health facilities when roads poor, |
| | | |contributing factor to why “giving birth” considered a |
| | | |health risk |
|Rural Morocco |Road impact study |World Bank 1998 |Upgraded road network contributed to increased use of |
| | | |health services by rural women |
Education: Education of women influences reproductive health through a variety of channels, including childbearing attitudes, health-seeking behaviors, and earning opportunities. Early gains in female literacy played an important role in MMR declines in Malaysia and Sri Lanka (Pathmanathan, 2003). In her review of linkages between women’s education, autonomy and reproductive behavior, Jejeebhoy (1995) notes that education enhances women’s knowledge about the outside world and makes them more aware than uneducated women of the importance of family health and hygiene as well as the treatment and prevention of illness. Another consequence that she notes is greater decision-making autonomy within the home. At the same time, she cautions that contextual factors influence the impact of education on women’s participation in household decision making, so that this participation is likely to be weaker in a society characterized by a high degree of gender stratification.
Focusing specifically on maternal mortality, McCarthy (1997) has noted several possible channels though which women's education might impact maternal mortality:
• By reducing the number of pregnancies (and lifetime risk of complications) through later marriage and increased use of contraceptives;
• By enabling women to be better informed about symptoms of complications and more likely to make more timely decisions to seek treatment;
• By being healthier and less likely to suffer a complication;
• By having better physical access to treatment facilities (for example, because a higher proportion of educated women live in urban centers); and
• By being better off and more able to pay for care, or be well treated by care providers because of their status.
None of these potential linkages guarantees that education will have the hypothesized impacts. As both McCarthy (1997) and Thaddeus and Maine (1994) note, the empirical evidence on linkages between maternal education and utilization of health services is not at all clear-cut. We are reminded that educated women are more likely to rely on self-care and self-medication and to postpone visits to a facility until after such methods fail. They also note that if education is associated with desire for fewer births and later marriage, there may be more unintended pregnancies and higher abortion rates, which would pose greater risk when access to safe abortion is limited.
Education is closely linked to gender status and the ways in which gender stratification affects access to household resources and utilization of health care services (Kunst and Houweling, 2001). In their work on intra-household bargaining power in Indonesia, Beegle and colleagues (2001) found that women who were more educated than their husbands were more likely to obtain prenatal care and, generally, that education enables a woman to make decisions regarding her reproductive health care. Education is also linked to several of the other factors that may enhance or limit access to life-saving interventions. Research on the impact of cost recovery on utilization of services has shown that educated women are more likely to understand and use exemption schemes (Newbrander et al, 2000), and the transport literature also highlights the links between education and access to and utilization of transport to get to health facilities.
Water and sanitation: Provision of safe water has been cited has a factor in the declines of mortality in developed countries (Van Poppel and Van der Heijden, 1997), and lack of sanitation and safe water, along with poor personal hygiene, are known to be major factors in the wide prevalence of parasitic diseases in poor countries. Studies of the impact of safe water on infant and childhood mortality typically do not focus separately on neonatal mortality, but recognize that waterborne diseases can undermine the health of pregnant women because they cause anemia, a risk factor for mothers as well as their newborns (Santiso, 1997). Paul (1993) cites unsafe water supply as well as pollutants from fuels used in cooking as risk factors in the high MMRs of the African countries he studied. When comparing countries by the level of MMR, there is a sharp difference in the percent of households with safe water in the countries with MMRs under 30 (92 percent) compared to those with MMRs over 1000 (51 percent). The link between water supply and MMRs/NMRs involves both household and community factors. A household’s consumption of water may be constrained by prices, income and other household variables even if water is supplied at the community level. Jalan and Ravallion (2001) observed that health gains largely bypassed poor children when piped water was available in their community, particularly when the mother was poorly educated.
Nutrition: Poor nutrition is another key co-factor in maternal and neonatal mortality. The section on birth spacing identified the close link between poor nutrition and closely spaced births and maternal and neonatal mortality. The discussion in the previous section on malaria during pregnancy also highlighted the role of anemia. Poor nutrition among pregnant women in a number of the very high MMR countries is a factor contributing to those high rates. In India, anemia is reported as an indirect factor in 64.4 percent of maternal deaths (Buckshee, 1997). As the Safar Banu case demonstrated, gender stratification and attitudes also contribute through household behaviors that deprive poor women of adequate nutrition, not only during pregnancy but also during their childhood and adolescence, which leads to small stature and higher risk of delivery complications.
The DHS tabulations cited earlier also document major rich-poor differentials in maternal and child nutrition within countries, as shown in Tables 15 and 16. For the 36 countries that collected data on body mass for reproductive-age women, those in the poorest wealth quintile had an average of 14.5 percent of women with low body mass (LBM), compared to 7.7 percent in the highest wealth quintile.
Table 15: Low Body-Mass in Reproductive-Age Women by Wealth Quintile and Region
|Region |No. of |Regional average |Poorest quintile |Richest quintile|Poor/rich |
| |countries | | | |difference |
|East Asia |0 |* |* |* |* |
|Central Asia + Turkey |4 |5.1 |6.9 |4.1 |2.8 |
|L. America, Caribbean |8 |5.5 |7.3 |3.5 |3.8 |
|Middle East, N. Africa |3 |10.2 |16.0 |5.1 |10.9 |
|South Asia |2 |40.2 |45.1 |27.0 |18.1 |
|Sub-Saharan Africa |19 |12.9 |15.8 |8.5 |7.3 |
|All country average |36 |11.7 |14.5 |7.7 |6.8 |
Source: Gwatkin et al, 2004 *not all countries collected these data
South Asia had both the highest overall average of LBM and the largest differential between rich and poor quintiles. Latin America and the Caribbean along with Turkey and the two Central-Asian countries had the lowest overall average and the lowest rich-poor differentials. Sub-Saharan Africa and the Middle East/North Africa fell in between.
Table 16: Child Malnutrition by Wealth Quintile and Region
|Region |No. of |Regional average |Poorest quintile |Richest quintile|Poor/rich |
| |countries | | | |difference |
|East Asia |0 |* |* |* |* |
|Central Asia + Turkey |4 |23.1 |35.1 |13.1 |22.0 |
|L. America, Caribbean |9 |23.0 |36.0 |6.5 |29.5 |
|Middle East, N. Africa |3 |35.2 |45.1 |21.1 |24.0 |
|South Asia |4 |46.6 |56.6 |29.8 |26.8 |
|Sub-Saharan Africa |21 |34.2 |40.7 |22.6 |18.1 |
|All country average |41 |31.9 |41.0 |18.7 |22.7 |
Source: Gwatkin et al, 2004 *not all countries collected these data
Somewhat different patterns are found in the data for childhood malnutrition. Overall, the poorest quintiles have more than twice the percentage of malnourished children than the richest quintile. South Asia has the highest overall average, but Latin America and the Caribbean have the largest rich-poor differential, echoing the high level of income inequality for that region (for example, in Table 5 Brazil’s Gini coefficient was .59).
Public Policy and Governance:
Earlier in the paper, the table with countries grouped by the level of MMR showed that those with lower MMRs had better governance than those with higher rates when they were compared using a “governance effectiveness” indicator. That indicator was one of six compiled by World Bank experts (Kaufmann et al, 2003) covering 199 countries using several hundred variables drown from 25 separate data sources. The six indicators include government effectiveness, voice and accountability, political stability, regulatory capacity, rule of law, and control of corruption. The indicators have a mean of zero and a standard deviation of one, so that virtually all scores fall between –2.5 and +2.5, with higher scores indicating better performance. In the countries in the MMR sample, few countries (Bahamas, Singapore) had scores greater than one, and most were below the mean of zero. In Table 17, tabulation of average scores by MMR group for each of the indicators in the table below shows consistent falloff in performance as MMRs rise, and that the group with MMRs greater than 1000 has an average score near or below one standard deviation below the mean for all of the indicators. Political instability stands out as the indicator with the lowest average for this group, and the highest average for the countries with low MMRs.
Table 17: Governance Indicators for Countries Grouped by MMR Level
|MMR Level |Number of |Effective |Voice & |Political |Regula-tory |Rule of Law |Control of |
| |countries |gover-nance |account-ability|stability |quality | |corrup-tion |
|0-30 |17 |0.41 |.29 |.47 |.51 |.35 |.41 |
|31-100 |39 |-0.07 |-.24 |.04 |-.07 |-.12 |-.16 |
|100-300 |32 |-0.35 |-.23 |-.20 |-.27 |-.36 |-.40 |
|301-1000 |41 |-0.37 |-.35 |-.37 |-.34 |-.38 |-.36 |
|Over 1000 |13 |-1.08 |-.86 |-1.13 |-1.07 |-1.06 |-.94 |
Source: Kaufmann et al, 2003
Addressing Obstacles and Information Gaps
Policy and Program Actions
The strong negative correlation between MMRs and per-capita health expenditures suggests that more health spending is needed, but as the World Bank study on reaching the Millennium Development Goals in health reminds us, spending alone is not enough. The added health funding needs to be targeted on key obstacles that reduce poor women’s chances of accessing life-saving interventions when they are needed. It also has to be combined with improvements in government effectiveness and motivation of individuals and communities to make use of these interventions. Success stories like Sri Lanka and Malaysia demonstrate that sustained efforts to improve health system performance, increase women’s education, and improve infrastructure, along with targeted investments and policy change to establish a cadre of trained midwives to ensure safe delivery and effective management of emergencies do work (Pathmanathan et al, 2003). Other countries that have reduced maternal and neonatal mortality (Honduras, an example cited earlier, and the Indian state of Kerala[8]) have proved that a combination of targeted investments in life-saving interventions as well as social infrastructure can make a difference.
The Marginal Budgeting for Bottlenecks approach described earlier suggests that increased public expenditures on health will have an even greater impact if they are targeted on key obstacles that prevent poor women from accessing life-saving interventions. The first step in the MBB approach is the identification of such obstacles. Using the Pathways framework outlined in the introduction, this paper has shown that obstacles are both inside and outside the health system, beginning with individuals, households and communities and including both health care and health financing as well as other sectors—education, transportation, water/sanitation, and nutrition. The relative importance of obstacles will vary by country and region, but a number of them appear to affect most of the countries that currently have high maternal and neonatal mortality. MBB also requires costing of interventions to address these obstacles and tracking of expenditures and performance of the interventions that are funded to ensure that the targeted spending is being used effectively and that the poor are benefiting. Table 18 summarizes obstacles described in the paper and identifies possible program and policy changes to address them.
Table 18: Actions to Address Obstacles
| | | |
|Obstacle/Pathways Level |Action |Indicator |
|Other reproductive risks | | |
|Unintended pregnancy & unsafe abortion |Increase access to FP and ensure |Unmet need for family planning |
|Violence and refugee status |abortion safety | |
| |Improve RH services in refugee settings |MMRs in refugee camps |
|Infectious diseases (TB, malaria, |Link RH and infectious disease programs | |
|HIV/AIDS) | |HIV, TB and malaria treatment rates for pregnant|
| | |women |
|Household level | | |
|Poor hygiene |Health education |Hand washing |
|Poor nutrition |Micronutrients during pregnancy |Maternal anemia |
| |Ensure that credit schemes & social | |
|Women's lack of autonomy in decisions |funds give decision power to poor women |Women having control over money |
|Community level | | |
|Gender attitudes |Leadership support of improved women's |Religious and community leaderships groups |
| |status |engaged in gender issues |
| | | |
|Community support |Community mobilization to ensure safe |Percent of communities that participate in |
| |delivery and referral of emergencies |life-saving skills training and implementation |
|Health system level | | |
|Inadequate staffing; unsafe procedures,|Targeted expenditures on training and |Percent of outreach or first referral points |
|including unsafe abortion |deployment of midwives and key staff |with trained midwives |
| | | |
| |Involve TBAs and others in community in | |
| |creating awareness and referrals |Percent of TBAs who are engaged as community |
| | |agents |
| |Targeted expenditures to reduce gaps | |
|Lack of consumables and medicines |Targeted investments to upgrade | |
|Inadequate facilities |facilities |Percent of facilities with standard list |
| | |available |
| | |Percent of communities with delivery homes and |
| | |access to referral facilities meeting service |
| | |delivery norms |
|Other sectors | | |
|Education |Increase female enrollment |Enrollment rates |
|Transport |Increase paved roads |Percent of paved roads |
|Water & sanitation |Improve access to safe water |Percent of households with safe water |
|Nutrition |Nutrition supplementation |Anemia rates |
|Public policy | | |
|Public health expenditures |Targeted increases in health spending |Percent of public expenditures on health |
| | |(identifying, if possible, those targeted on MMR|
| |Reforms in financial management and |and NNM) |
|Good governance |resource allocation |Implementation of effective budget tracking |
| | |procedures |
If the performance of targeted interventions is to going to be tracked effectively, then better information is needed on rich-poor differentials in mortality, and on the morbidities that follow on poorly managed obstetric emergencies. Expenditure tracking also needs to be fine-tuned so that pro-poor expenditures on these investments can be monitored.
Achievement of lower maternal and neonatal mortality for poor women is possible if the obstacles to their utilization of interventions that are known to be effective can be overcome. As noted above, the countries that have reduced mortality rates have done this through a combination of investments inside and outside the health system including improving the skills and deployment of midwives and other key staff, reducing financial obstacles to care, improving the quality of care and access to referral facilities, and mobilizing households and communities to ensure safe delivery and effective management of emergencies when they occur. Parallel investments in nutrition, malaria control, education, roads, and water/sanitation have also contributed in situations where low performance in those areas has undermined the reproductive health of poor women.
Strengthening the Evidence Base
While the tabulations of maternal and neonatal health-related indicators in Demographic and Health Surveys (DHS) for wealth quintiles have helped to advance our understanding of linkages between MNH and poverty, there are still significant gaps in the evidence base. Data on maternal and neonatal mortality and morbidities are generally scarce, and particularly so for studying rich-poor differences. Sample sizes in most DHS surveys are typically too small for calculating rich-poor differences in maternal mortality, though Graham and colleagues (2004) have developed methodology for calculating rich-poor differentials in maternal mortality from a subset of larger DHS surveys. Special DHS supplements like the recent Bangladesh Maternal Health Survey also offer an opportunity for gaining further insights into links between maternal mortality and poverty.
If there are problems in addressing rich-poor differentials in maternal and neonatal mortality and their consequences, there is an even greater evidence gap concerning the life-long morbidities suffered by women who survive a poorly managed obstetric emergency. A comparison of estimates of the burden of disease for maternal conditions with BOD for TB, malaria and HIV/AIDS suggests that the impact of these morbidities is substantial. Table 19 shows estimates of years of life lost (YLL) and years lived with disability (YLD) for the population aged 15 and over for these four conditions for sub-Saharan Africa in 2002:
Table 19: Estimates for Burden of Disease for sub-Saharan Africa, 2002
|Condition |YLL (1000) |YLD (1000) |Ratio YLD/YLL |
|TB |6,994 |795 |.11 |
|Malaria |1,748 |528 |.30 |
|HIV/AIDS |43,849 |5,514 |.13 |
|Maternal conditions |6,865 |4,884 |.71 |
Source: WHO, GBD Estimates, www3.WHO.int/whosis, accessed 12/04/04
While the number of years of life lost due to maternal causes is substantially lower than for HIV/AIDS (though about the same as for TB, a cause of death associated with HIV/AIDS, and larger than for malaria, where deaths are concentrated among children), the years lived with disability associated with maternal conditions is nearly as large as for HIV/AIDS. One of the most poignant examples of this latter burden is that suffered by mostly young and poor African women who survive obstructed labor but live on with an obstetric fistula, which has terrible social and economic consequences. While fistulae and other obstetric morbidities can be treated, resources are often not allocated to them, and women who suffer them continue to endure their consequences as part of what may often be accepted as their normal lot. In an era in which allocation of health resources is increasingly being driven by such evidence-based algorithms as disability-adjusted life years, the combination of poorly measured and comparatively rare maternal mortality (compared to other causes, including major communicable diseases and non-communicable diseases) plus poorly measured morbidities highlights the importance of improving the measurement of maternal morbidities, particularly for the poor in order to support the call for allocation of adequate resources to address these problems.
Expenditure tracking is an important policy and advocacy tool in the health field and needs to be employed more effectively by the champions of maternal and neonatal health. Countries that are participating in debt-relief and/or poverty-reduction programs are required by the IMF and World Bank to prepare Poverty Reduction Strategies. These strategies include analyses of the causes and consequences of poverty as well as expenditure plans to address specific poverty issues. The Medium Term Expenditure Framework (MTEF) is both a prospective mechanism for allocating funding across key sectors that affect poverty and a retrospective tracking mechanism to ensure that planned allocations are actually made. While the level of detail in the typical MTEF is typically limited to the broad sectoral level (education, health, water and sanitation, etc), expenditure tracking through the MTEF process offers an important opportunity for champions of MNH to track key expenditures required to improve MNH (training and deployment of trained birth attendants, sustained funding of facilities and referral networks for the management of obstetric emergencies, etc.; see Reinikka and Svensson, 2002).
A stronger evidence base is also needed on linkages between poverty and MNH outcomes and on the effectiveness of interventions inside and outside the health sector to address them. When research funding is scarce, one way to achieve this is to add special inquiries of maternal and neonatal health outcomes to on-going survey programs, particularly longitudinal surveys that track groups of respondents over time. These permit analyses that show both the consequences of poor MNH outcomes on the well-being, consumption and economic productivity of households and different income/wealth levels and the impact of interventions inside (e.g., increase in skilled attendance) and outside (credit programs, community mobilization) the health system on those outcomes. Survey initiatives in several countries (Indonesia, Kenya, Malawi, Mexico, and the Philippines; see Table 20) might be engaged in this task.
Table 20: Longitudinal Survey Programs
| | | | |
|Country |Data set |Population |Type of Survey |
|Mexico |PROGRESA panel data |PROGRESA program focuses on population in extreme poverty in rural |Longitudinal |
| | |areas. Program addresses poverty through monetary and in-kind | |
| | |benefits, and encouraging investments in education, health and | |
| | |nutrition. | |
|Philippines |Cebu Longitudinal Health and|The Cebu Longitudinal Health and Nutrition Survey is part of an |Longitudinal |
| |Nutrition Survey |ongoing study of a cohort of Filipino women who gave birth between | |
| |cpc.unc.edu/projects/ceb|May 1, 1983 and April 30, 1984. The cohort of children born during | |
| |u/ |that period, their mothers, other caretakers, and selected siblings | |
| | |have been followed through subsequent surveys conducted in 1991-2, | |
| | |1994, and 1999. | |
|Indonesia |Indonesian Family Life |The Indonesian Family Life Survey (IFLS) is an on-going longitudinal |Longitudinal |
| |Survey |survey in Indonesia. The first wave of the IFLS (IFLS1) was conducted| |
| |FLS/IFLS |in 1993/94. IFLS2 and IFLS2+ were conducted in 1997 and 1998, | |
| | |respectively. IFLS2+ covered a 25% sub-sample of the IFLS households.| |
| | |IFLS3 was conducted in 2000 and covered the full sample. | |
|Kenya |ssc.upenn.edu/Social_Net|The first survey of the Kenya Diffusion and Ideational Change Project|Cross-sectional, |
| |works/Level%203/Kenya/level3|(KDICP-1) was carried out in 1994-5, and interviewed 925 ever-married|individual-level data |
| |_kenya_data.htm |women of childbearing age and 859 men (of which 672 were husbands of | |
| | |the currently married women ). In 1996-7 and in 2000, respectively, |Longitudinal, |
| | |the second and third round of the survey (KDICP-2 and KDICP-3) |individual-level data |
| | |followed-up the same respondents (if eligible), and also interviewed | |
| | |any new spouse. All rounds of the KDICP were carried out in four | |
| | |sites in Nyanza Province, in south-west Kenya. Within each village, | |
| | |currently married women and their husbands were eligible respondents | |
|Malawi | MDICP is a sister project of the Kenya project above also focuses|Cross-sectional, |
| |ial_Networks/Level%203/Malaw|on the role of social networks in changing attitudes and behavior |individual-level data |
| |i/level3_malawi_data.htm |regarding family size, family planning, and HIV/AIDS in Malawi. |for men and women; |
| | | | |
Conclusion
This paper has reviewed the evidence base on obstacles that poor women and newborns face when they suffer life-threatening obstetric complications. It has drawn on available cross-national comparisons and analyses of data, broken down where possible to show rich-poor differences in MNH outcomes, as well as on the extensive literature on factors within and beyond the healthcare system that affect MNH outcomes. The Pathways framework was used to organize the review of evidence. While it is not the only conceptual framework available, it has the advantage of being the one recommended for the planning of health expenditures at country-level in Poverty Reduction Strategies currently being implemented in poor countries where maternal and neonatal mortality rates are the highest. The evidence suggests that these obstacles can seriously undermine poor women’s chances of getting life-saving care when they need it, but that a combination of multi-sector efforts can overcome such obstacles. It has also shown that there are important knowledge gaps to be addressed by building a stronger evidence base on maternal and neonatal mortality and morbidities experienced by poor women, how well expenditure is targeted to address obstacles, and the effectiveness of interventions.
Annex 1: Maternal Mortality in India
Of the estimated 529,000 maternal deaths estimated by WHO to have occurred globally during the year 2000, 136,000 were estimated to have occurred in India. This reflects both the high MMR for India (the MMR for year 2000 is estimated to have been 540) and India’s large population (over 1 billion). Only China has a larger population, though with its MMR of 56 it had fewer than one-tenth the number of maternal deaths (11,000 deaths).
Data on MMR differences within India are limited. In its National Human Development Report 2001, India’s Planning Commission (2002) published MMRs for 15 large states for 1997-1998 based on the Sample Registration System (SRS). Bhat (2001) has applied indirect estimation techniques to those data to produce adjusted estimates. Bhat’s indirect method produced a national level MMR estimate of 479 for the period 1987-1996, compared to a figure of 407 from the SRS. Both estimates are lower than the WHO estimate for 2000, though Bhat’s figure is very close to the 1995 estimate of 470 produced by WHO/UNICEF/UNFPA. Both direct and indirect SRS estimates suggest substantial variation by state within India. The SRS data vary widely, and give rates that are implausibly low for Gujarat and Tamil Nadu and high for Kerala, suggesting that it would be better to use Bhat’s adjusted estimates. These are produced in column 2 of the table on the next page, along with other indicators from the Planning Commission that relate to the Pathways framework.
In the table, states are ranked from the highest to lowest MMRs, with estimates ranging from over 700 in Assam, Uttar Pradesh and Madhya Pradesh to under 200 in Punjab, Tamil Nadu and Kerala. SRS estimates are used for Punjab and Kerala because Bhat reported that the rates in those two states were too low to estimate using indirect methods. Their SRS estimates may be high. The populations in many Indian states are larger than those of many countries for which WHO reports its MMR estimates. For example, Uttar Pradesh is as large as Brazil, the world’s fifth largest country.
State-wise MMR differences are most closely correlated with the indicator for attended deliveries (r = .89), with a general pattern of low rates of attended delivery (20-30 percent) in the poorer northern tier states and higher rates (60-90 percent) in the more prosperous southern ones. The Planning Commission’s gender disparity index (female attainment relative to male in literacy, life expectancy and economic participation) ranges from 0.5 to 0.6 in the high MMR states to around 0.8 in the low MMR states.
Data for other indicators used in cross-country comparisons (road density, public health expenditures) do not reveal clear relationships with MMR differentials. The high MMR states are generally the poorer ones, although Kerala has a much lower GDP per capita than Gujarat, but Gujarat’s MMR is three times that of Kerala. Prevalence of anemia is generally higher in the high MMR states than in the lower ones. The three low MMR states have the highest per capita public health expenditures, but Assam with the highest MMR spends nearly as much as most of the lower MMR states. The low level of attended deliveries in Assam reinforces the point that it is not just the amount of expenditure, but how it is spent. Clearly Assam, UP and other high MMR states are not spending enough to ensure that all deliveries have skilled attendants.
Annex Table 1: MMRs and Other Indicators for Indian States
| |MMR |Popu-lation |Att. |Gen-der |Road Density |Female Anemia |Publ. Health|Net State or |
| | |(Mil) |Del. |Index |(Km/ |(%) |Exp/P in $ |National |
| | | |(%) | |100KM2) | | |Prod/P |
|Assam |984 |26.6 |21.5 |.58 |87.2 |69.7 |3.1 |1675 |
|Uttar Pradesh |737 |174.5 |23.0 |.52 |86.8 |48.7 |2.2 |1725 |
|Madhya Pradesh |700 |81.1 |30.1 |.66 |45.1 |54.3 |2.0 |1922 |
|Orissa |597 |36.7 |33.7 |.64 |168.7 |63.0 |2.3 |1666 |
|Gujarat |596 |50.6 |53.5 |.71 |46.7 |46.3 |3.0 |3918 |
|Rajasthan |580 |56.5 |36.2 |.69 |37.9 |48.5 |3.1 |2226 |
|Bihar |513 |109.8 |23.5 |.47 |50.8 |63.4 |1.7 |1126 |
|Karnataka |480 |52.7 |59.2 |.75 |75.1 |42.4 |3.0 |2866 |
|Haryana |472 |21.1 |42.0 |.71 |63.7 |47.0 |2.8 |4025 |
|West Bengal |458 |80.2 |44.5 |.63 |85.0 |62.7 |2.7 |2977 |
|Maharashtra |380 |96.8 |59.7 |.79 |117.6 |48.5 |3.0 |5032 |
|Andra Pradesh |283 |75.7 |65.1 |.80 |64.7 |49.8 |2.5 |2550 |
|Punjab |199* |24.3 |62.7 |.71 |127.7 |41.4 |3.7 |4389 |
|Tamil Nadu |195 |62.1 |84.1 |.81 |158.6 |56.5 |3.6 |3141 |
|Kerala |195* |31.8 |94.1 |.83 |374.9 |22.7 |4.0 |2490 |
| | | | | | | | | |
|INDIA |479 |1027.2 |42.3 |.68 |74.9 |51.8 |2.3 |2840 |
Sources: Bhat (2001) and GOI Planning Commission (2002); MMRs for Kerala and Punjab from SRS. Public health expenditures per capita and net state domestic product per capita in constant 1980-81 rupees. Health expenditure data from Peters et al. (2002).
Annex 2: Country-Level Data Table
|Country |MMR Deaths/ |UNDP status of |Govern |Combined female|Gini |Health |Safe Water |Paved roads as |
| |100,000 |women index |-ance index |enrollment data|Coef- |Expend. |(% of pop. with|% of total road|
| |births | | | |ficient |Per |access) |KM |
| |(2000) | | | | |capita | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
|MMU ................
................
In order to avoid copyright disputes, this page is only a partial summary.
To fulfill the demand for quickly locating and searching documents.
It is intelligent file search solution for home and business.
Related searches
- world bank unesco education
- world bank financial inclusion
- world bank poverty line 2019
- world bank financial inclusion 2020
- world bank business environment index
- world bank doing business 2020
- world bank ease of doing business 2019
- world bank doing business rankings
- symbolic interactionism and poverty examples
- world bank retiree portal
- world bank pension log in
- world bank retiree website