Menu
Menu
Menu
Menu
Background: Maternal and newborn mortality rates in Nigeria are among the highest globally, and large socioeconomic inequalities exist in access to maternal, newborn, and child health (MNCH) services in the country. Inequalities also exist in catastrophic health expenditure among households in Nigeria. We aimed to estimate the health and financial risk protection benefits across different wealth groups in Nigeria if a policy of public financing of MNCH interventions were to be introduced. Methods: We did an extended cost-effectiveness analysis to estimate the health and financial risk protection benefits, across different household wealth quintiles, of a public-financing policy that assumes zero out-of-pocket costs to patients at the point of care for 18 essential MNCH services. We projected health outcomes (deaths in children aged <5 years [under-5 deaths] and maternal deaths) and private expenditure averted using the Lives Saved Tool with data extracted from national surveys. We modelled three scenarios: 1) coverage expansion at a rate equal to the trend observed between 2013 and 2018 (status quo); 2) annual coverage expansion by 5% compared with the status quo (uniform scale-up scenario); and 3) annual coverage expansion by 10%, 8%, 6%, 4%, and 2% compared with the status quo from the poorest to the wealthiest quintiles, respectively (pro-poor scale-up scenario). Findings: Our analysis shows that, if an additional 5% increase in coverage was provided for all wealth quintiles between 2019 and 2030, this uniform scale-up policy would prevent more than 0·11 million maternal deaths and 1·05 million under-5 deaths, avert US$1·8 billion in private expenditure, and avert 3266 cases of catastrophic health expenditure. The incremental cost effectiveness ratio would be $44 per life-year gained, which is highly cost-effective when compared with the gross domestic product per capita of Nigeria for 2018 ($2028). The policy would prevent a higher number of under-5 deaths and catastrophic health expenditure cases in poorer quintiles, but would prevent more maternal deaths and private expenditure in wealthier quintiles. If poorer populations experienced a greater increase in service coverage (ie, the pro-poor scale-up scenario), more maternal and under-5 deaths would be prevented in the poorer quintiles and more private expenditure would be averted than would be under previous scenarios. Interpretation: Public financing of essential MNCH interventions in Nigeria would provide substantial health and financial risk protection benefits to Nigerian households. These benefits would accrue preferentially to the poorest quintiles and would contribute towards reduction of health and socioeconomic inequalities in Nigeria. The distribution would be more pro-poor if public financing of MNCH interventions could target poor households. Funding: WHO Partnership for Maternal, Newborn, and Child Health.
We did an extended cost-effectiveness analysis of a policy to publicly finance essential MNCH interventions in Nigeria.10 Compared with traditional cost-effectiveness analyses that estimate the levels of cost and effectiveness of certain interventions, an extended cost-effectiveness analysis further estimates: 1) the distribution of the health benefits across a spectrum; 2) the private health-care expenditure averted by the policy; and 3) the financial risk protection benefits that the policy provides. Under this policy, MNCH interventions are provided to the patient without any cost at the point of care (ie, no out-of-pocket payments). The projection period was between 2019 and 2030 and costs were reported in US$ (whereby $1 was equal to 305·79 Nigerian Naira; 2018 Central Bank of Nigeria rates). Wealth quintiles were defined at the beginning of the projection period and we assumed that an individual's wealth quintile remained stable throughout the projection period. All analysis and projections were conducted using the Lives Saved Tool (version 4.761) and results were reported according to the Consolidated Health Economic Evaluation Reporting Standards checklist. The Lives Saved Tool is a linear, deterministic mathematical model; further details are in the appendix (p 28). The research protocol was reviewed and approved by Duke Campus Institutional Review Board (2020-0122). We selected interventions that met the following inclusion criteria: 1) recommended as priority MNCH interventions by the Partnership for Maternal, Newborn and Child Health (PMNCH),11 an international multi-stakeholder alliance, including governments, UN agencies, health-care professional associations, youth-led organisations, and non-governmental organisations; 2) data on coverage for these MNCH interventions available by wealth quintiles; 3) interventions not currently provided for free in Nigeria; and 4) interventions included in the list of Lives Saved Tool's default interventions. We included 18 interventions (panel; appendix p 3). Excluded interventions and reasons for exclusion are in the appendix (p 3). We divided the total 2018 population of Nigeria (191 million)12 into five quintiles based on household wealth using 2018 Nigeria Demographic and Household Survey data.13 For each wealth quintile, we obtained the quintile-specific baseline disease prevalence and service coverage from 2018 Nigeria Demographic and Health Surveys and Nigeria Multiple Indicator Cluster Surveys,4, 13 and estimated the average annual rates of change (AARC) for each quintile using the quintile-specific data from the previous two surveys (appendix p 5). Proxies for service coverage were used when no reliable data were available (appendix p 6). We adjusted the default settings of the Lives Saved Tool for target population and population in need by using quintile-specific estimates for total fertility rates, proportion of women with low BMI (<18·5 kg/m2), neonatal mortality rates, infant mortality rates, and under-5 mortality rates (appendix p 7). We modelled three scenarios: 1) coverage for interventions will expand at an AARC equal to the trend observed between 2013 and 2018 (referred to hereafter as status quo; appendix p 9); 2) coverage will increase by 5% compared with the status quo scenario every year14 (referred to hereafter as the uniform scale-up scenario); and 3) coverage will increase by 10%, 8%, 6%, 4%, and 2% compared with the status quo from the poorest to the wealthiest household quintiles, respectively, every year (referred to hereafter as the pro-poor targeted scale-up scenario). For all three scenarios, if coverage reached 95%, it would remain stable until the end of the projection (appendix pp 9–21). Health outcomes were the number of under-5 deaths and maternal deaths averted and life-years gained. We modelled deaths averted using the Lives Saved Tool. To estimate life-years saved, for under-5 deaths, we multiplied the number of deaths averted at age 5 by the remaining life expectancy at age 5, and for maternal deaths (ie, among women of reproductive age), we multiplied the number of deaths averted at 30 years (midpoint between 15 and 45 years) by the remaining life expectancy at 30 years.10 The Lives Saved Tool estimates costs on the basis of a target population, proportion of population in need, coverage, treatment inputs, and cost per service. The Lives Saved Tool methodology has been published previously (appendix p 29).15, 16 We reported the intervention cost from the Lives Saved Tool, which included health-care provider fees, diagnostic costs, and medication costs, but excluded indirect costs such as transportation costs or the opportunity cost of lost employment or wages (appendix p 8). We applied a scale-up cost of 20% to estimate the total health system cost of the policy. Private expenditure refers to pooled resources that are not controlled by the government, such as voluntary health insurance, and direct payments or out-of-pocket payments from households. In Nigeria, voluntary health insurance accounted for only 0·55% of current health expenditure in 2018.9 Therefore, to estimate the private expenditure averted, we used the total cost of interventions from the Lives Saved Tool, and multiplied the cost by the current out of-pocket payment ratio in Nigeria (77%). We defined catastrophic health expenditure as private expenditure for health expenditures that exceeded 10% of household income.17 We used the private expenditure for each intervention to estimate whether this intervention could lead to catastrophic health expenditure and accumulated the number of catastrophic health expenditure cases for each intervention. We used the income information from the Nigeria Living Standards Survey, disaggregated by wealth quintiles18 (appendix p 8). Incremental cost-effectiveness ratios were reported from a modified societal perspective that includes the perspective of the payer and the household. We conducted a sensitivity analysis of the effect of the total fertility rate on estimates, to assess the effect of quintile population size and population growth rate. We repeated our analysis with different total fertility rate assumptions (using fertility rates 10% lower and higher than estimated fertility rate and using the national mean total fertility rates for each quintile). We also conducted sensitivity analyses on discount rates (at 3%, 5% and 10%), scale-up costs (at 10%, 30%, and 40%), and service coverage levels (at 4% and 6%). The study funder was involved in study design, but had no role in data collection, data analysis, data interpretation, or writing of the report.
