Reducing regional inequality in community-based treatment of childhood pneumonia in Ethiopia: A sub-national distributional cost-effectiveness analysis


 Background

Scaling up coverage of community-based treatment of childhood pneumonia (CCM) is part of the strategy to promote equity and reduce under-five mortality rate (U5MR) in Ethiopia. However, urban children with symptoms of pneumonia are still more than twice as likely to receive treatment compared with rural children having similar symptoms. There are no sub-national cost-effectiveness analyses available to inform decision makers on the most equitable scale-up strategy.

Objectives

To model sub-national cost-effectiveness and inequality impacts of scaling up coverage of CCM in each of the 11 Ethiopian regions. We also explore three different scale-up strategies: reducing geographical inequalities, health maximization and universal scale-up.

Methods

For each region, we developed a Markov model and estimated the cost-effectiveness of scaling up coverage to 90 percent. Data inputs were collected through literature review. Effects were modeled as life years gained and under-five deaths averted. Inputs on unit costs were adjusted to the proportions of rural and urban population in each region. In scenario analysis, we estimated costs, health effects and, by the use of the Gini measure applied to health, the inequality impacts of three different scale-up strategies: 1) maximizing health by prioritizing the regions where the intervention was the most cost-effective, 2) reducing geographical inequality by prioritizing the regions with the highest baseline U5MR and 3) universally scaling up to 90% coverage in all the regions.

Results

Universal scale-up of CCM would cost about 1.3 billion USD and prevent about 90,000 under-five deaths. This is less than 15,000 USD per life saved and translates to an increase in life expectancy at birth of 1.6 years across Ethiopia. The regional incremental-cost effectiveness ratio (ICER) of scaling up the intervention coverage varied from 26 USD per life year gained in Addis to 199 USD per life year gained in the SNNP region. In scenario analysis, we found that prioritizing regions with high U5MR is effective in reducing geographical inequalities, although at the cost of some fewer lives saved.

Conclusions

Our model results illustrate a trade-off between maximizing health and reducing health inequalities, two common policy-aims in low-income settings.


Introduction
Universal health coverage (UHC) is achieved when the entire population has access to essential medical services and can access these services without being exposed to financial hardship [1]. To achieve universal health coverage is high on the agenda of the 2030 sustainable development goals and tightly interlinked with all the focus areas regarding poverty, inequality and health. However, health resources are limited and there is an inevitable need to set priorities when progressing towards universal health coverage. Policy makers allocating scarce resources need evidence on expected costs and benefits associated with different scale-up strategies for essential health interventions.
Ethiopia has achieved considerable improvements in population health and child survival over the past decade. However, the progress seems to be less apparent in some regions, and many preventable deaths are still occurring. In 2016, the under-five mortality rate ranged from 39 deaths per 1000 live birth in Addis Ababa to 125 deaths per 1000 live birth in Afar [2]. Pneumonia was among the three major causes of under-five mortality throughout Ethiopia. Effective treatment is available, but the coverage remains low. Scaling up coverage of community-based treatment for childhood pneumonia (CCM) through the Health Extension Program could be an effective strategy to further reduce U5MR and promote equity.  [2].
Despite efforts to reach rural and poor population with high impact interventions through the Health Extension Program, evidence suggests that inequalities in important child health indicators have increased [3]. Distance to nearest health facility and region of residence are two important geographical factors that influence both health service utilization and child health outcomes [4,5]. In general, poor and rural populations are often deprived of health compared to rich and urban populations. Reducing these inequalities should be an important equity concern. Previous studies have already shown that pneumonia treatment is pro-poor and provides financial risk protection [6,7]. However, no studies have investigated the intervention impact on regional inequalities.
There are currently substantial inequalities in health service coverage and child health outcomes between the major regions of Ethiopia [3]. These geographical health inequalities are avoidable and could be associated with geographical inequalities in the distribution of other resources, such as income. When progressing towards UHC, policy makers need to consider both equity-impacts and the cost-effectiveness of prioritizing different strategies. It has been shown previously that scaling up coverage of pneumonia treatment would be pro-poor and cost-effective at a national level [6,8].
However, the regions of Ethiopia are diverse in terms of influential factors such as available infrastructure, population wealth, local epidemiology, background mortality and baseline coverage of pneumonia treatment.
The regional factors mentioned above are likely to affect both the costs and the expected benefits of providing pneumonia treatment at community level, and we believe that data on reginal costeffectiveness would inform policy planning more effectively than national averages alone [9]. However, to our knowledge, no studies on sub-national cost-effectiveness of pneumonia treatment have yet been published, nor have previous studies investigated the impacts of scaling up treatment coverage on geographical inequalities. Our aim is to model the sub-national cost-effectiveness and inequality impacts of scaling up coverage of community-based treatment of childhood pneumonia (CCM) in each region separately and nationally. We also explore three different scale-up strategies: 5 decreasing geographical inequalities, health maximization and universal scale-up.

Methods
We used Markov modeling in the software program TreeAge to model health and economic impacts of scaling up coverage of CCM from baseline to a target coverage of 90 percent in each of the eleven major regions. Each regional model was populated with 2016 region-specific data on cost per treatment, incidence of childhood pneumonia, background mortality, and baseline coverage of the intervention. Untreated case fatality rate and treatment effect were assumed to be similar in all the regions. Region-specific inputs are displayed in Table 1 and fixed model inputs are displayed in Table   2. We assumed that all treatments were provided outpatient under universal public coverage.  Figure 2). We extrapolated results by running the model for a total of 120 cycles until everyone in the model cohort was dead. Since the model starts at age zero and runs for 120 cycles, output adds up to life expectancy. Model inputs were specific for each region, but the structure of the models did not change.

Model inputs on incidence of pneumonia
Data on prevalence of pneumonia, all-cause, under-five mortality rates and baseline coverage in each region were collected from the Demographic and Health Survey 2016 (DHS 2016). The DHS 2016 estimated the prevalence of pneumonia by asking mothers of children under five years of age whether their child had experienced clinical symptoms of pneumonia during the two weeks prior to the interview. Clinical symptoms are defined as cough accompanied by short, rapid breathing that was chest-related, and/or difficult breathing that was chest-related [2]. In this survey, baseline treatment coverage was estimated as the percentage of children with symptoms of pneumonia who received clinical examinations and oral antibiotics from a health professional.
We used the following formula to estimate the annual incidence of pneumonia among children below the age of five: Incidence = prevalence/duration of disease [10] The estimated incidence of pneumonia varied from 0.453 cases per year in the Amhara region to 0.040 cases per year in the Harari region ( Table 2). Our calculations were comparable to estimates of incidence of childhood pneumonia in comparable settings [11].

Model inputs on background mortality
Age-specific mortality rates among adults affect the incremental life years gained by reducing U5MRs, and since U5MR is an acknowledged predictor of general population health, the large regional inequalities in U5MRs are likely to be reflected in mortality among older age groups. However, only life tables representing national averages were available for age-specific mortality rates among adults and children older than five years of age. Therefore, we modeled adult morality rates based on the assumption that the U5MR is associated with mortality among older age groups of the same population. In practice, we selected Ethiopian abridged life tables from the 2015 UN world population prospect [12] and matched each region to a national life table in a time period with a similar U5MR as the one observed for the region in 2016 [13]. The life table from this time period was used as a proxy for adult mortality rates in that region.
There are no census data from 2016. Without taking into account the effect of still births, we used regional fertility rates and data on the total number of women in each region to yield a rough estimate of the number of births in 2016 [2,14]. We used these estimates as size variables in the calculations of weighted averages of the effects observed in each region, the budget impact and the geographical Gini coefficients.

Model inputs on the effectiveness and unit costs of CCM
Data on effectiveness of the treatment were collected from a previously published systematic review 7 of studies assessing CCM of pneumonia in developing countries [15]. The review concluded that CCM on average reduces the case fatality rate of pneumonia in children less than five years of age by 70% ( Table 2). We applied this as our input for treatment effects in all the regions.
Data on treatment costs were collected through literature review. We did not encounter any data on regional cost per treatment. However, previous studies indicate that there are significant differences in costs of providing community health services in different geographical contexts [9,16]. The inputs for costs per treatment in each region were therefore adjusted for rural and urban residency.
One study from Kenya showed that it was 7.2 times more expensive to provide community health services in rural areas compared to urban areas [9]. We assumed that rural Ethiopia would observe a similar increase in cost per treatment compared to urban Ethiopia. We applied a cost per treatment provided for patients with urban residency of 45 USD [8,17], and the cost per treatment provided for patients with rural residency was modelled to be 7.2 times more expensive (Table 1). However, most regions of Ethiopia have both rural and urban population [18]. The following formula was used to estimate average costs per treatment in each region: where x is the proportion with urban residency while y is the proportion with rural residency.
The studies we relied on to estimate the costs per treatment costed the intervention from a providers' perspective. Cost items classified as personnel costs, capital costs or supply costs were included.
These were further divided into patient care costs and overhead costs [17]. In our Markov modeling, we did not include initial training of health personnel or capital costs. Costs were discounted at a 3 percent rate.

Estimation of the intervention effects on health and health inequalities
We modeled effects of scaling up coverage of CCM as life expectancy gains for children less than five years of age. The children who recovered from pneumonia were assumed to continue to live with the same health risks as the overall population. We did not apply disability weights to the effect measure as we were primary interested in mortality reduction afforded by the intervention. Hence, the incremental effects of the intervention represent gains in life expectancy at birth. We half-cycle 8 corrected and discounted the effects at a 3% rate. The incremental cost-effectiveness ratios were calculated by dividing incremental costs by incremental life years gained.
The Gini coefficient is a measure of inequality (here applied to health), represented by a number between 0 and 1 where 0 represents absolute equality in the distribution of a chosen variable, and 1 represents a situation in which all of the chosen variable belong to one individual. Gini coefficients can be used to describe inequalities in life expectancy between individuals or population groups [19].
We used the DASP extension of the STATA software to calculate the Gini coefficients quantifying inequalities in life expectancy between the regions and between individuals within each region. In our calculations of geographical inequalities, we applied model results on regional life expectancies at birth as the health variable, and estimated numbers for children born in 2016 as the size variables.
Data from survival curves provided by the Markov models were applied as the size and health variables for calculation of interindividual health inequalities.

Scale-up scenarios
In the regional scenario analysis, we explored three possible objectives: health maximization, reducing geographic inequality, and universal scale-up. The first two scenarios have lower costs and could be seen as possible pathways to universal scale-up. We estimated incremental costs, reduction in national U5MR, incremental effects, and interindividual GINI impacts of 1) maximizing health by scaling up to 90% coverage in the six regions where the intervention is the most cost-effective, 2) reducing geographic inequality by scaling up to 90% coverage in the three regions with the highest under five mortality rates, and 3) universally scaling up to 90% coverage in all regions. For all scenarios, we estimated expected health impacts across Ethiopia by adding the weighted averages of the effects observed in each region.

Results
In total, scaling up coverage of CCM to 90 percent in all regions would decrease the Ethiopian U5MR from 67 to 52 deaths per 1000 live birth. Figure 3 shows the impact on U5MR in each region.
Reduced U5MR translates as increased life expectancy. Scaling-up treatment coverage to 90 percent in all regions would increase the average life expectancy at birth from 63.18 to 64.73 years, a 1.55 9 years gain. At a regional level, the incremental effects of scaling up the intervention coverage to 90% varied from 0.13 life years gained in Harari to 1.93 life years gained in Oromia ( Table 3). The highest incremental effects were observed in regions with a high incidence of pneumonia and low baseline coverage of the intervention. Table 4 Table 6 shows the incremental costs, the incremental life years gained, the decrease in U5MR and the interindividual and geographical Gini impacts of three possible scenarios: health maximization, decreasing geographical inequalities, and universal scale up. Addis Ababa, Gambela, Dire Dawa, Harari and Tigaray were the regions in which the intervention was the most cost-effective, and Afar, Beni-Shangul and Somali were the regions with the highest baseline under five mortality rates.

Scenario analysis
Universal scale-up of the intervention in all regions would substantially alter the survival curves and decrease interindividual inequalities in life expectancy ( Table 6). The health maximizing strategy and the strategy prioritizing the worse off both, with much lower incremental costs, yielded slight decreases in interindividual Gini ( Table 6). As shown in Table 6, only the targeted strategy to increase treatment coverage in the regions with the highest U5MR decreased regional inequalities in life expectancy. The other two strategies, health maximization and universal scale up, both increased regional inequalities in life expectancy.

Discussion
According to the final report of the WHO Consultative Group on Equity and Universal Health Coverage, cost-effectiveness, priority to the worse off, and financial risk protection are important concerns for evaluating which health care interventions should be ranked first in a universal health care package [20]. They also argue that efforts should be made to prevent underprivileged sub-populations from being left behind. Our results indicate that scaling up coverage of pneumonia treatment would be cost-effective in all the elven major regions of Ethiopia, and that the intervention could reduce both interindividual and geographical inequalities in life expectancy. Moreover, making targeted efforts to scale up first the coverage of high priority interventions in underprivileged regions would prevent groups disadvantaged by residence from being left behind.
Our results on incremental cost effectiveness ratio of scaling up pneumonia treatment coverage in the eleven major regions are comparable to those of published literature, in which we found incremental cost effectiveness ratios (ICERs) ranging from 26,6 to 208 USD per DALY averted [21,22]. Geographical factors varying between the different study settings may account for the large dissimilarities found in previously published cost-effectiveness analysis. However, few previous studies have investigated subnational cost-effectiveness of community case management of pneumonia, nor the impacts of rural residence on costs per treatment.
There are several data limitations and weaknesses in this study. The first limitation is our estimation of costs per treatment in rural compared to urban areas. The study we relied on for these calculations was small and from Kenya, where they have a different strategy for delivery of community-based services [9]. However, we assumed that geographically varying factors identified in the Kenyan study, such as population density and attrition rates among local health workers, would comparably influence the costs of providing quality community health services in Ethiopia. In 2016, Berman et al.
reported the costs of providing primary health care services as varying by a factor of more than five between the regions of Ethiopia [16]. Their results support our estimates of region-specific costs per 11 treatment.
The second uncertain assumption we made was that untreated case fatality rate of pneumonia is similar in all the regions of Ethiopia. This may not be the case. Malnutrition, coinfection with HIV and low birth weight are some of the factors that influence the severity of pneumonia infections in children [23]. The prevalence of these risk factors varies between regions. Essentially, stronger evidence on sub-national cost-effectiveness would require more primary studies on how much geographical and regional factors influence case fatality rates and other key inputs.
We based our estimations of yearly incidence of pneumonia on data form the Demographic and Health survey 2016 [2]. However, some our estimations of incidence appear to be excessively high The final weakness to be discusses is that we did not include costs of demand generation, although evidence suggests that demand side barriers are major causes of sustained low intervention coverage [25,26]. To focus on demand generation would be especially appropriate in Ethiopia where community health workers have already been deployed throughout the country but the utilization of key health services remains low [26]. Strategies such as the health development army have been initiated, but to our knowledge there is yet no evidence available on its associated costs and effects.
The cost-effectiveness threshold is the upper ICER to be considered cost-effective within a given health budget. Investing in health interventions that are above this threshold would draw resources away from more cost-effective interventions and would lead total population health being foregone[27]. If we adopt a cost-effectiveness threshold of 50% of the GDP per capita, as suggested by Woods et al, scaling up coverage of CCM to 90% would be cost effective in all the regions of Ethiopia. However, even the threshold of 50% of the GDP per capita could be too high. Woods et al.
indicate that a cost-effectiveness threshold between 4 and 51% of the GDP per capita could be more appropriate for low to middle income countries, such as Ethiopia [27].

Concerns about reducing inequalities may justify giving priority to health investments in regions
where the health interventions are not proven cost-effective compared other regions where the investments would be more cost-effective [20]. This trade-off depends on the decision makers' level of aversion to inequality (ref: Wagstaff 2002). All the scale-up strategies we explored reduced interindividual inequality at a regional and national level. However, only the strategy prioritizing the worst-off regions reduced geographical inequalities. Since there are considerable inequalities in wealth among the Ethiopian regions, reducing the geographical inequalities in U5MR could also indirectly reduce socioeconomic inequalities [3].

Consent for publication
Not applicable.

Availability of data and materials
All data used as model inputs for analysis are available from sources with open access.
Data on prevalence of pneumonia, on under five mortality rate and on baseline coverage of the intervention were collected from the Demographic and Health survey 2016:

Funding
Funding was obtained though the student research program at the University of Bergen. The funding body had no role in the design of the study, analysis, interpretation of data or in writing the manuscript.

Authors' contributions
MO conducted the analysis in TreeAge, collected data from online sources, and contributed to writing major parts of the manuscript.
OFN and STM functioned as supervisors and contributed to study design, data collection, data analysis, interpretation of results and writing of the manuscript.
All authors read and approved the final manuscript 24.
McAllister, D.A., et al., Global, regional, and national estimates of pneumonia morbidity and mortality in children younger than 5        Incremental cost effectiveness ratio (ICER) in each region