Improved and Equitable Access

Well-targeted investments along the continuum of care can respond to a fundamental human right: the right to health. Increasing equitable access to RMNCH services is a key strategy for moving closer to universal health coverage, defined by the WHO as when all people obtain the health services they need without suffering financial hardship when paying for them (WHO 2010, ix).

Health System Benefits

Investments in women’s and children’s health strengthen the entire health system. For example, the capacity to provide 24-hour emergency obstetric care requires that health system components, such as qualified health workers, medications, facilities, and a functioning referral system, be in place across geographic areas.

Extended Lifecycle Benefits

Investments in RMNCH bring benefits across age groups. For example, investments in nutrition have long-lasting effects beyond the immediate improvement in nutritional status, such as improvements in cognitive development, school performance, and future earnings (Ruger and others 2012).

Cost-Effective Interventions

A considerable body of research, including Disease Control Priorities in Developing Countries, second edition, has established that RMNCH interventions are among the most effective and cost-effective available (Jamison and others 2006). Recent evidence confirms these findings. A study of diarrhea and pneumonia interventions finds that 15 highly cost-effective interventions exist that, if implemented at scale, would prevent 95 percent of deaths from diarrhea and 67 percent of deaths from pneumonia in children under age five years by 2025 (Bhutta and others 2013). Evidence from Afghanistan suggests that an approach combining improved family planning with incremental improvements in skilled birth attendance, transport, referral, and appropriate intrapartum care in high-quality facilities could prevent 75 percent of maternal deaths at a cost of less than US$200 per year of life saved (Carvalho, Salehi, and Goldie 2012).

Improved Integration of Services

Opportunities exist to deliver packages of interventions when women and children present at health facilities, for example, to prevent sexually transmitted infections in conjunction with family planning programs (Church and Mayhew 2009). Findings of the Multi-Country Evaluation of the Integrated Management of Childhood Illness (IMCI) suggest that integrated care can lead to cost savings: the annual cost of providing health care to children was considerably lower in districts with IMCI compared with districts without it (Adam and others 2005).

Selecting the Interventions

National policy makers must choose which services to provide, taking into account budgetary constraints and financial ceilings allocated by the ministry of finance and other financing partners. Evidence on cost-effectiveness, current health system capacity, feasibility, and acceptability will inform investment strategies. The framework outlined in this chapter includes interventions that were identified in a 2011 review as essential and cost-effective RMNCH interventions (PMNCH, WHO, and Aga Khan University 2011). Table 16.1 lists the 50 selected interventions grouped into six broad packages that follow program structures in many national health systems: family planning, maternal and newborn health, malaria, HIV/AIDS, immunization, and child health, with nutrition included in several packages.

Table 16.1

Preventive and Treatment Interventions Modeled.

The effective delivery of high-quality interventions depends on key enablers, including national policies, functional health systems, community engagement, and innovation. Strategies modeled include those supporting both the supply side (for example, expanding health system access by constructing new hospitals and facilities) and the demand side (for example, mass media campaigns to encourage breastfeeding and care seeking for childhood illnesses).

Estimating the Costs

The second stage of the analysis is to use modeling tools to estimate the health and fertility impacts of the interventions and the investment costs required.

With respect to costs, attempts have been made to estimate the resources required to scale up the provision of essential RMNCH services in LMICs. Most of these are disease- or program-specific cost studies that determine costs more or less specific to the disease or age group (Bhutta and others 2013; Singh and Darroch 2012; Stenberg and others 2007). Such studies tend to include patient-level costs for intervention-specific commodities, such as vaccines, bednets, and nutritional supplements, as well as some estimates of the time and related cost of health workers involved in providing the health services. Ideally, these studies would also include program support costs, for example, for training in disease-specific management, epidemiological surveillance, and provision of vehicles specific to the program activities. However, the studies may not always do so, or the specific methods used to estimate such costs are not always well described. Finally, program- or disease-specific estimates may miss resources needed for broader health-system-strengthening activities, thereby underestimating the true resource needs for the provision of services. Health-system-strengthening activities to consider include preservice training and deployment of clinical staff, development of a functioning referral system, strengthening the health information system, and upgrading facility infrastructure. Figure 16.1 shows the 12 components of the full cost of scaling up the interventions that are estimated in the analysis presented in this chapter.

Estimating Health and Fertility Impacts

The scale up of a comprehensive package of care will have interactions across diseases and age groups. The interlinkages built into the OneHealth Tool and accompanying impact modules (box 16.1) eliminate double counting of lives saved, and they take into account the reduction in need for treatment as preventive care is scaled up. Increases in coverage are translated into reductions in maternal, newborn, and child mortality, along with declines in some aspects of morbidity such as prevalence of wasting and stunting. Fertility rates are modeled to decrease with increasing targets for contraceptive prevalence rates, in turn affecting population growth projections over time.

Box 16.1

Translating Coverage Increases into Cost and Health Impacts: The OneHealth Tool. The OneHealth Tool (OHT) is a software program that aims to support integrated planning processes in low- and middle-income countries by bringing together disease-specific (more...)

Assessing the Economic and Social Benefits of Achieved Outcomes

Once the improved health outcomes arising from the interventions—lives saved, morbidity averted, and unwanted pregnancies avoided—are determined, the task is to measure the benefits arising from these better outcomes. Some of these benefits will be strictly economic, reflected in higher GDP from increased workforce participation and from higher productivity. However, other benefits, although equally real and certainly economic in a broader welfare sense, will not be reflected in conventional GDP measures. A mother’s life saved so that she is able to look after her children and support her community has great social value even if she does not enter the paid workforce. Equally, the value of a child’s life saved does not depend only on his or her participation in the labor force when an adult. We refer to the benefits not captured in existing GDP measures as social benefits.

A strong consensus exists among economists that measurements of economic and social change need to move beyond production or conventional GDP to sustainable well-being (Stiglitz and others 2009) and that these more inclusive measures are especially important in relation to health (Arrow and others 2013; Suhrcke and others 2012). These more inclusive methods can be referred to as full income methods. They include additional benefits from improved health outcomes and more inclusive gauges of income than those included in GDP as it is currently measured.

The Lancet Commission on Investing in Health (Jamison and others 2013) argues strongly for a full income approach to measuring the benefits of investment in health, defined as measured increases in conventional GDP plus the value of additional life years gained. The approach presented in this chapter goes further, because it does not limit the analysis to the benefits arising from lives saved. We attempt to include in an explicit manner estimates of economic and social benefits from morbidity averted, and to estimate the economic benefits derived from control of fertility and hence from the reduction in unwanted pregnancies—the demographic dividend. Accordingly, the approach in this chapter allows for a more comprehensive approximation of the estimated benefits than that used in Jamison and others (2013) and other studies.

Modeling an Increase in Coverage Level

An investment case may take into account different scenarios of specific packages of services (content), levels of investment (level of ambition), and strategies (for example, community-based versus facility-based delivery) to achieve the set goals. At the country level, the various different scenarios should be assessed to inform national policy discussions regarding the most-effective resource allocations. Here, for illustrative purposes and in the interest of assessing the benefits of investing in a set of high-mortality countries, we scale up the same package of interventions across all countries. A scenario that maintains current baseline (2012) coverage is defined as Low, while an ambitious scenario with coverage increasing for all 50 interventions until 2035 is defined as High (table 16.2).²

Table 16.2

Parameters of the Investment Analysis.

The relative level of coverage across the scenarios drives the differences in intervention costs and impact so that the incremental effect of an investment strategy (that is, the High scenario) compared with maintaining current coverage without strengthening the health system (the Low scenario) can be assessed and valued. The analysis is centered on the comparison between scenarios; it is important to note that the main counter-factual in our example is the Low scenario with constant coverage levels and a growing population. Accordingly, the results should not be interpreted as additional spending above current levels of health expenditure, but rather as the cost and impact of bending the curve and accelerating progress compared with a Low scenario in which coverage remains at the 2012 level while population increases.

We applied tools that have been developed by the international community, including the OneHealth Tool (box 16.1) to assess intervention-specific costs and health and fertility impacts. Intervention-specific costs are driven by increases in coverage, with costs distributed to different levels of care (community, outreach, facility, and hospital). Program- and systems-related costs draw upon estimates made by the Taskforce on Innovative International Financing for Health Systems (WHO 2009) and are described in detail in Stenberg and others (2014). Costs are generally estimated using an ingredients approach (quantity times price), with the exception of supply chain costs, for which a mark-up ratio is applied.

Key Methodological Assumptions

The costs and the benefits are defined as the incremental costs and benefits between two scenarios. However, when fertility management tools are an important part of the suite of interventions, the populations in the two scenarios diverge substantially. The approach we adopt is to assess only those benefits that apply to those alive in the High scenario, and we compare their situation to what it would have been in the Low scenario.

Three broad types of benefit are identified:

• Some have the benefit of life because their lives were saved through the interventions.
• Others are in much better health because of the morbidity averted.
• The whole community has the benefit of higher per capita incomes arising from the reduction in unintended pregnancies and from the processes that the fall in fertility rates sets in motion.

The difference in deaths and in morbidity for children between the Low and High scenarios will reflect two different factors: the impact of the health interventions for a given level of births, and the reduction in the number of births (due to expanding family planning) for a given level of health. We partition the reduction in child deaths and in morbidity between the Low and High scenarios into these two components. We use only the former, which we refer to as lives saved, in calculating benefits.³ The reduction in child mortality from scaling up contraceptives is thus counted in the health impact results but not in the cost-benefit analysis. All maternal deaths prevented are considered to be lives saved; that is, the full reduction in maternal mortality is translated into economic benefits.

GDP per capita paths were derived from World Bank data and were combined with population estimates from the OneHealth Tool projections; these were extended to 2070 on the basis of convergence to zero population growth in each country by that year. Per capita GDP estimates and assumptions about productivity are combined with data on labor force participation of those affected by the intervention⁴ (ILO 2013).

Economic and Social Benefits of Years of Life Saved

A vast literature discusses the value of a statistical life⁵ and, by implication, the value of life years saved (VLY). Most studies use a willingness-to-pay approach, either in the form of analyses of revealed preferences evident in wage and risk data or analyses of stated preferences. Viscusi and Aldy (2003) review the revealed preference literature and suggest, albeit with a wide uncertainty margin, an implied value for a life year of about 4.0 times GDP per capita, with an income elasticity of about 0.6. These two facts, in turn, imply a value of a life year for LICs of 1.5 to 2.0 times GDP per capita.

Jamison and others (2012) estimate the VLY as 2.3 times GDP per capita in LMICs at a 3 percent discount rate, with estimates by World Bank region ranging from 1.4 for Latin America and the Caribbean and for the Middle East and North Africa to 4.2 for Sub-Saharan Africa. Cropper, Hammitt, and Robinson (2011) note the recent expansion of the stated preference literature, in which individuals are asked about how they would act in hypothetical situations, and that the value of a statistical life emerging from these studies is much lower than for revealed preference studies.

The revealed preference studies refer to both the economic and social value of a life year; by economic value we mean the value that would be captured in conventional GDP measures, primarily through labor force effects; the social value refers to all other benefits of an additional year of life to an individual or a community not captured in GDP. We regard it as useful to distinguish between the social and economic components of the VLY because they may have different roles in some investment analyses.

We have constrained the total value of a life year across these two components to 1.5 times GDP per capita for the sample as a whole, which we regard as being at the lower end of the range used in the literature. The calculated economic benefits of increased labor force participation amount to about 1.0 times GDP per capita, calculated as the sum of GDP for all 74 countries divided by the population for all 74 countries. A social VLY equal to 0.5 times the GDP per capita of the full set of sample countries is then applied as a common value across countries. Although the strictly economic value of an additional year of life will vary with local economic parameters, there is no reason to think that the social value is lower in poorer countries than in richer ones. Although we do not use any age adjustment for the social value of a life, our procedure results in some discounting of the overall value of a life year for age, and the economic benefits of children’s lives saved only begin to accrue when they enter the labor force.

Benefits of Morbidity Averted

Many women and children who survive adverse RMNCH events suffer serious and sustained disabilities (Ashford 2002; Blencowe and others 2013; Mwaniki and others 2012; Souza and others 2013) that undoubtedly have substantial human, social, and economic costs. The interventions studied here should be expected to generate important benefits through lower morbidity. In spite of its acknowledged importance, few attempts have been made to quantify the burden of maternal and child morbidity or to estimate its economic and social cost; we attempt to begin the process in this study.

Although the OneHealth Tool estimates the lives saved as a result of scaling up the interventions, it does not measure the morbidity averted (other than for wasting and stunting) or the impact on mortality in subsequent years from averting morbidity in the initial year. We estimate morbidity averted for four causes for children (preterm birth complications, birth injury, congenital abnormalities, and malnutrition) and two for mothers (obstructed labor and other maternal disorders), and calculate economic and social benefits.⁶ Moreover, we derive parameters relating improved nutritional outcomes—prevention of low stature and low birth weight—to lifetime earnings and apply these to estimates of reduced wasting and stunting by country.

Benefits of Reduced Fertility Rates

The third benefit is the economic impact of the reduction in fertility rates, which is well documented in the literature.⁷ Ashraf, Weil, and Wilde (2013) identify a range of channels through which a reduction in the total fertility rate (TFR), that is, the number of children born to the average woman during her lifetime, affects growth in GDP per capita; these channels can be grouped into three types of effect, each affecting GDP:

• A dependency effect because a reduction in births reduces the dependent population. Given that the nondependent population produces the GDP, the fall in the dependent population for a given level of GDP increases the level of GDP per capita.
• A labor supply effect because adults are able to devote more time to working. (With fewer births, women and other caregivers will have an increased propensity to enter the labor force, leading to increased labor supply per capita and hence to increased GDP per head.)
• A productivity effect covering a range of factors influencing long-term productivity, such as higher saving by households and higher investment in schooling. More generally, with lower birth rates, more of a society’s resources can be devoted to capital deepening, thereby increasing productivity, rather than to capital widening to meet the needs of the expanding population.

The estimates of the demographic dividend draw on and adjust the methods of Ashraf, Weil, and Wilde (2013), who developed estimates of key parameters based on a review of relevant literature. We derive from their model an aggregate relationship between the reduction in the TFR and the change in GDP per capita over time, out to 2070, and apply this to the change in the TFR in each country to estimate the impact on per capita GDP and hence on overall GDP.⁸

In summary, we present economic benefits, valued in GDP terms, derived from the following:

• Lives saved
• Morbidity averted
• Demographic dividend.

Social benefits, also valued in GDP terms, are derived from the following:

• Lives saved
• Morbidity averted.

In the Low scenario, GDP per capita paths converge to an annual growth rate of 2 percent by 2070. The economic benefits here refer to the difference in GDP growth between the High and Low scenarios.

Two Country Case Studies

To illustrate how the investment framework could be applied at the country level, we present two case studies of LICs. One is a country in Asia that has seen increased coverage of RMNCH interventions and reductions in the fertility rate to about 2.5. The other is a country in Sub-Saharan Africa with low coverage of many RMNCH interventions and continued high fertility rates (table 16.3).

Table 16.3

Parameters of Country Case Studies.

Benefit-Cost Ratios for Investments

Applying a discount rate enables benefits and costs to be expressed as a net present value (NPV). The benefit-cost ratio for a given discount rate is the ratio of the NPV of benefits and costs at that discount rate.

Table 16.5 reports for all countries considered as a whole, and for groups of countries, the benefit-cost ratios calculated using rising discount rates over the period: 3 percent for 2013–35, 5 percent for 2013–50, and 7 percent for 2013–70. We present results individually for China and India given the significant size of these countries. China is also a particular case in that there is limited additional demographic dividend to gain (table 16.6). Although the 3 percent rate is commonly used in this type of analysis (appendix 3 in Jamison and others 2013), the use of rising discount rates for longer periods is one way of taking account of higher uncertainty over the longer term as well as myopic time preferences and likely increases in consumption over time.

Table 16.5

Benefit-Cost Ratios for High Compared with Low Scenarios, Selected Periods and Discount Rates.

Table 16.6

Analysis of Contribution to Benefit-Cost Ratio, High versus Low Scenarios, 5 percent Discount Rate for Net Present Value, 2013–50.

The benefit-cost ratios shown in table 16.5 indicate high returns on increased investment in RMNCH in most countries, especially when benefits beyond the intervention period are included. For all countries considered as a group, the benefit-cost ratio is 8.7 for the intervention period to 2035 at a 3 percent discount rate, 27.6 at 5 percent for the period to 2050, and 34.2 at 7 percent for the period to 2070. The benefit-cost ratio is generally higher for lower-middle-income countries and UMICs than for LICs, especially post 2035, as well as for those 43 countries in Sub-Saharan Africa and 5 in South Asia where maternal and child mortality are highest.

Analysis of Benefits and Benefit-Cost Ratios by Type of Benefit

Tables 16.6 and 16.7 show the contribution from the three sources of benefits to 2050 comparing the High and Low scenarios (using a 5 percent discount rate) expressed in two ways: as a contribution to the overall benefit-cost ratio and as a percentage share of all benefits in NPV terms. These tables illustrate four points about the distribution of benefits.

Table 16.7

Analysis of Contribution to Benefits, High versus Low Scenarios, by percentage Shares, 5 percent Discount Rate for Net Present Value, 2013–50.

Results from Two Case Studies

The Asian country is considerably larger in both population and GDP than the Sub-Saharan African country and has a somewhat higher level of GDP per capita. The Sub-Saharan African country has higher mortality rates for both children and mothers in addition to a higher fertility rate and a higher level of labor force participation by women (table 16.3).

Although the total cost of the intervention for the Asian country is larger than that for the Sub-Saharan African country, reflecting the disparity in population size, the additional cost per person for the High versus the Low scenario is considerably lower at US$2.65 (versus US$6.88). This result is due to the higher fertility and maternal and child death rates in the Sub-Saharan African country, which require a higher level of intervention and a greater cost per capita. Despite the differences in population size, the numbers of maternal, child, and stillbirth deaths prevented by the interventions are similar in the two countries, with a proportionally greater impact in the Sub-Saharan African country.

Table 16.5 shows a high benefit-cost ratio for the Sub-Saharan African country, with results similar to those for the average of all 74 countries and for the group of LICs. Although positive, the benefit-cost ratio for the Asian country is more modest, again reflecting the differences in initial fertility and death rates.

A more detailed description of the sources of the benefits that arise from the intervention for the two country case studies is provided in tables 16.6 and 16.7. For the Asian country, the biggest contributors to the benefit-cost ratio are those benefits arising from the social value of lives saved and morbidity averted (65.7 percent). The contributions from the increase in GDP from workforce-related benefits (15.2 percent) and from the demographic dividend (20.0 percent) are more modest but still significant. Considered solely as a function of either the increase in GDP from workforce-related benefits or from the demographic dividend, the benefit-cost ratio still shows benefits outweighing costs (ratios of 1.4 and 1.8, respectively).

For the Sub-Saharan African country, in contrast, the contributions from the economic and social benefits are virtually equal (48.1 percent and 51.9 percent, respectively). The demographic dividend is about twice as important as for the Asian country (39.0 percent), while the contribution from additional GDP is lower (9.1 percent). Again as a function of either the increase in GDP from workforce-related benefits or from the demographic dividend, the benefit-cost ratio shows benefits outweighing costs (ratios of 2.3 and 9.6, respectively) and the ratios are higher than for the Asian country.

Large Economic and Social Returns

First, investments in high-impact interventions across the continuum of care in RMNCH have large economic and social returns in addition to the impact on health outcomes. The benefit-cost ratio of investments in the High scenario for the full country sample is 8.7 in 2035. Findings are robust to variations in the methods of analysis, such as discount rates.

Affordable Investments

Second, the required investments are affordable for most countries. On average for the 74 countries, an additional US$4.48 per capita would be needed in 2035 to finance the High scenario. However, affordability needs to be examined in the context of fiscal sustainability as issues related to universal coverage, financial protection, quality, responsiveness, and efficiency will affect the policy dialogue around public investment in health, and macroeconomic conditions will set the overall boundaries for what can be achieved. The Global Financing Facility to Advance Women’s and Children’s Health, created in 2014, will support countries in overcoming fiscal constraints in the short term and in setting up mechanisms to achieve long-term sustainable domestic financing.¹⁰

Variable Returns on Investment

Third, the magnitude of returns on investment varies across country groupings. By income, the highest returns are realized in lower-middle-income countries. This finding might be explained by two factors: First, economies of lower-middle-income countries with higher GDP have higher returns operating through workforce benefits and the demographic dividend compared with LICs. Second, returns in UMICs might be lower than in LMICs, given their already lower mortality rates and more strongly diminishing returns.

The findings vary by individual countries, reflecting the epidemiological and demographic situation in each, current health systems performance, and country-specific economic factors. The substantially different findings of the two country case studies confirms that individual countries will find considerable value in undertaking their own investment analyses, to give results specific to their circumstances. For example, the returns on investment in the Sub-Saharan African case study country, with low coverage of most RMNCH interventions, and therefore still facing high child and maternal mortality rates and high fertility rates, are more than twice as large as the case study country in Asia, which has managed to increase coverage of RMNCH interventions and reduce fertility rates to less than 2.5. The country case studies confirm the importance of investing in family planning; the effect of the demographic dividend is substantial even when the investment reduces the TFR by a small amount.

Similarly, we present results for China and India separately, given the size of their populations and economies. Given current low birth rates in China, no significant economic benefits are to be derived from increasing the availability of family planning. This is not to argue that significant benefits could not be bought by increasing the quality of current programs and ensuring their responsiveness to population needs (Kane and Choi 1999). In India, our model estimates high economic benefits from increasing the contraceptive prevalence rate to respond to the unmet need.

High Rates of Return for a Comprehensive Approach, Including Family Planning

Fourth, investment in each of the elements in the continuum of care matters. The analysis finds that family planning programs generate particularly high returns, especially in countries with current high fertility rates, primarily through its effect on the demographic dividend. We have not separated out the rate of return on investment in maternal versus child health because the analysis deals with investing across the full spectrum of RMNCH; however, we note that there may be specifically high returns on investment in maternal care for adolescents, given that adolescent pregnancies pose a much higher risk for both mother and newborn compared with pregnancies among women of older age groups (Patton and others 2009; WHO 2008, 2011).

Returns on Investment Vary over Time

Fifth, the different types of interventions often generate benefits in different time frames, so that the rate of return varies over time. Returns increase substantially over time, particularly beyond the investment period of 2013–35. For example, at a discount rate of 3 percent, the benefit-cost ratio for the full sample of 74 countries is about four times larger in 2070 (34.2) as in 2035 (8.7). Although policy makers often make decisions in much shorter time horizons, it is nevertheless important to note that returns are realized well beyond the investment period.

An Extended Modeling Approach

Finally, on a methodological note, the overall economic and social benefits are driven by the demographic dividend generated by the investment. For example, in 2050 the demographic dividend accounts for 48.3 percent of the benefit-cost ratio (for the 74 countries). Workforce-related benefits and social benefits account for about 25 percent each. The relative share of morbidity-averted benefits compared with lives saved benefits is low because only a few sources of morbidity are included in the model, and the gains in morbidity are adjusted for the degree of disability averted. For LICs, the social benefits predominate because these are valued using the average GDP per capita of all countries; the workforce-related benefits are valued using country GDP per person in the workforce.