Do Socioeconomic Inequalities in Neonatal Mortality Reflect Inequalities in Coverage of Maternal Health Services? Evidence from 48 Low- and Middle-Income Countries

Progress towards universal health coverage and inequalities in infant mortality: an analysis of 4·1 million births from 60 low-income and middle-income countries between 2000 and 2019

BACKGROUND

Expanding universal health coverage (UHC) might not be inherently beneficial to poorer populations without the explicit targeting and prioritising of low-income populations. This study examines whether the expansion of UHC between 2000 and 2019 is associated with reduced socioeconomic inequalities in infant mortality in low-income and middle-income countries (LMICs).

METHODS

We did a retrospective analysis of birth data compiled from Demographic and Health Surveys (DHSs). We analysed all births between 2000 and 2019 from all DHSs available for this period. The primary outcome was infant mortality, defined as death within 1 year of birth. Logistic regression models with country and year fixed effects assessed associations between country-level progress to UHC (using WHO's UHC service coverage index) and infant mortality (overall and by wealth quintile), adjusting for infant-level, mother-level, and country-level variables.

FINDINGS

A total of 4 065 868 births to 1 833 011 mothers were analysed from 177 DHSs covering 60 LMICs between 2000 and 2019. A one unit increase in the UHC index was associated with a 1·2% reduction in the risk of infant death (AOR 0·988, 95% CI 0·981-0·995; absolute measure of association, 0·57 deaths per 1000 livebirths). An estimated 15·5 million infant deaths were averted between 2000 and 2019 because of increases in UHC. However, richer wealth quintiles had larger associated reductions in infant mortality from UHC (quintile 5 AOR 0·983, 95% CI 0·973-0·993) than poorer quintiles (quintile 1 0·991, 0·985-0·998). In the early stages of UHC, UHC expansion was generally beneficial to poorer populations (ie, larger reductions in infant mortality for poorer households [infant deaths per 1000 per one unit increase in UHC coverage: quintile 1 0·84 vs quintile 5 0·59]), but became less so as overall coverage increased (quintile 1 0·64 vs quintile 5 0·57).

INTERPRETATION

Since UHC expansion in LMICs appears to become less beneficial to poorer populations as coverage increases, UHC policies should be explicitly designed to ensure lower income groups continue to benefit as coverage expands.

FUNDING

UK National Institute for Health and Care Research.

Impact evaluation of the TAMANI project to improve maternal and child health in Tanzania

BACKGROUND

The Tabora Maternal and Newborn Health Initiative project was a multicomponent intervention to improve maternal and newborn health in the Tabora region of Tanzania. Components included training healthcare providers and community health workers, infrastructure upgrades, and improvements to health management. This study aimed to examine the impact of trainings on four key outcomes: skilled birth attendance, antenatal care, respectful maternity care and patient-provider communication.

METHODS

Trainings were delivered sequentially at four time points between 2018 and 2019 in eight districts (two districts at a time). Cross-sectional surveys were administered to a random sample of households in all districts at baseline and after each training wave. Due to practical necessities, the original stepped wedge cluster randomised design of the evaluation was altered mid-programme. Therefore, a difference-in-differences for multiple groups in multiple periods was adopted to compare outcomes in treated districts to not yet treated districts. Risk differences were estimated for the overall average treatment effect on the treated and group/time dynamic effects.

RESULTS

Respondents reported 3895 deliveries over the course of the study. The intervention was associated with a 12.9 percentage point increase in skilled birth attendance (95% CI 0.4 to 25.4), which began to increase 4 months after the end of training in each district. There was little evidence of impact on antenatal care visits, respectful treatment during delivery and patient-provider communication.

CONCLUSION

Interventions to train local healthcare workers in basic and comprehensive emergency obstetric and newborn care increased skilled birth attendance but had limited impact on other pregnancy-related outcomes.

Epidemiology of neonatal mortality: a spatial and multilevel analysis of the 2019 mini-Ethiopian demographic and health survey data

BACKGROUND

Sub-Saharan African countries are a high-burden region of neonatal mortality and showed slow progress in its reduction. In developing countries, as long as the current trend of mortality persists, achieving a sustainable development target for neonatal mortality would be challenging. The aim of this study was to detect significant geographic areas and identify community and individual-level predictors of neonatal mortality in Ethiopia to draw attention to a policy.

METHODS

A weighted total sample of 24,136 mothers from the 2019 mini-Ethiopian demographic and health survey data were included in the analysis. Global Moran's I statistics was run to check the clustering of neonatal mortality and then kriging interpolation was done to predict the magnitude of neonatal mortality in Ethiopia. In addition, SaTScan analysis was also executed to identify hot spot clusters of neonatal mortality. Finally, a multilevel mixed-effect logistic regression model was used to identify community and individual-level predictors of early neonatal and neonatal mortality.

RESULTS

The lifetime early neonatal and neonatal mortality among mothers in Ethiopia was 5.08 (95% CI: 4.13-6.03) and 6.54 (5.55, 7.52) per 1000 births respectively. Neonatal mortality was spatially clustered in the country and the SaTScan analysis identified significant hotspot areas of neonatal mortality in the Amhara and Afar regions and some areas of the Somali and Oromia regions. Its predicted magnitude was > 8 per 1000 births in wide areas of the Amhara and Benishangul regions. A multilevel mixed-effect logistics regression analysis identified that a lower level of maternal education, being a twin neonate, and being a male neonate were predictors of both early neonatal and neonatal mortality. Whereas, the younger age of mothers predicted neonatal mortality.

CONCLUSIONS

Neonatal mortality in Ethiopia is geographically clustered and sociodemographic and obstetric factors played a significant role. Policy direction should focus on evidence-based practices like midwives-led community and facility-based continuum of care from preconception to postnatal periods to possibly reduce neonatal mortality.

Unacceptable persistence of territorial inequalities in avoidable under-five mortality in Colombia between 2000 and 2019: a multilevel approach

OBJECTIVES

This study aimed at evaluating territorial inequalities in avoidable mortality in children under 5 years of age in Colombia between 2000 and 2019.

STUDY DESIGN

This was an ecological study.

METHODS

An ecological, longitudinal, multigroup study was conducted using secondary sources. Because of the hierarchical structure of the data, the effect of territorial characteristics on the count of avoidable under-five deaths was estimated using a three-level negative binomial regression model with random intercepts for municipality and fixed intercepts for time and departments.

RESULTS

Between 2000 and 2019, there were 216,809 avoidable under-five deaths in Colombia (91.3% of all registered deaths of children under 5 years of age). A total of 1117 municipalities located in 33 departments were analyzed over five 4-year periods. Ecological relationships were found between avoidable under-five mortality and the percentage of adolescent births, female illiteracy, and multidimensional poverty at the municipal level (standardized mortality ratio: 1.43 95% confidence interval: 1.33-1.54 for the group with the highest level vs the group with the lowest level of poverty). Furthermore, multidimensional poverty was a confounding factor for the association between the percentage of the population living in rural areas and avoidable child mortality.

CONCLUSIONS

Systematic and avoidable gaps were observed in mortality in children aged under 5 years in Colombia, where the territory constitutes an axis of inequality. Implementing strategies and programs that contribute to improving the conditions of women and socio-economic conditions in the territories should be a priority.

Linking political exposures to child and maternal health outcomes: a realist review

BACKGROUND

Conceptual and theoretical links between politics and public health are longstanding. Internationally comparative systematic review evidence has shown links between four key political exposures - the welfare state, political tradition, democracy and globalisation - on population health outcomes. However, the pathways through which these influences may operate have not been systematically appraised. Therefore, focusing on child and maternal health outcomes, we present a realist re-analysis of the dataset from a recent systematic review.

METHODS

The database from a recent systematic review on the political determinants of health was used as the data source for this realist review. Included studies from the systematic review were re-evaluated and those relating to child and/or maternal health outcomes were included in the realist synthesis. Initial programme theories were generated through realist engagement with the prior systematic review. These programme theories were adjudicated and refined through detailed engagement with the evidence base using a realist re-synthesis involving two independent reviewers. The revised theories that best corresponded to the evidence base formed the final programme theories.

RESULTS

Out of the 176 included studies from the systematic review, a total of 67 included child and/or maternal health outcomes and were included in the realist re-analysis. Sixty-three of these studies were ecological and data were collected between 1950 and 2014. Six initial programme theories were generated. Following theory adjudication, three theories in revised form were supported and formed the final programme theories. These related to a more generous welfare state leading to better child and maternal health especially in developed countries through progressive social welfare policies, left-of-centre political tradition leading to lower child mortality and low birth weight especially in developed countries through greater focus on welfare measures, and increased globalisation leading to greater child and infant mortality and youth smoking rates in LMECs through greater influence of multinational corporations and neoliberal trade organisations.

CONCLUSION

We present a realist re-analysis of a large systematically identified body of evidence on how four key political exposures - the welfare state, democracy, political tradition and globalisation - relate to child and maternal health outcomes. Three final programme theories were supported.

Comparing socioeconomic inequalities between early neonatal mortality and facility delivery: Cross-sectional data from 72 low- and middle-income countries

Facility delivery should reduce early neonatal mortality. We used the Slope Index of Inequality and logistic regression to quantify absolute and relative socioeconomic inequalities in early neonatal mortality (0 to 6 days) and facility delivery among 679,818 live births from 72 countries with Demographic and Health Surveys. The inequalities in early neonatal mortality were compared with inequalities in postneonatal infant mortality (28 days to 1 year), which is not related to childbirth. Newborns of the richest mothers had a small survival advantage over the poorest in unadjusted analyses (-2.9 deaths/1,000; OR 0.86) and the most educated had a small survival advantage over the least educated (-3.9 deaths/1,000; OR 0.77), while inequalities in postneonatal infant mortality were more than double that in absolute terms. The proportion of births in health facilities was an absolute 43% higher among the richest and 37% higher among the most educated compared to the poorest and least educated mothers. A higher proportion of facility delivery in the sampling cluster (e.g. village) was only associated with a small  decrease in early neonatal mortality. In conclusion, while socioeconomically advantaged mothers had much higher use of a health facility at birth, this did not appear to convey a comparable survival advantage.

Does government expenditure reduce inequalities in infant mortality rates in low- and middle-income countries?: A time-series, ecological analysis of 48 countries from 1993 to 2013

Inequalities in infant mortality rates (IMRs) are rising in some low- and middle-income countries (LMICs) and decreasing in others, but the explanation for these divergent trends is unclear. We investigate whether government expenditures and redistribution are associated with reductions in inequalities in IMRs. We estimated country-level fixed-effects panel regressions for 48 LMICs (142 country observations). Slope and Relative Indices of Inequality in IMRs (SII and RII) were calculated from Demographic and Health Surveys between 1993 and 2013. RII and SII were regressed on government expenditure (total, health and non-health) and redistribution, controlling for gross domestic product (GDP), private health expenditures, a democracy indicator, country fixed effects and time. Mean SII and RII was 39.12 and 0.69, respectively. In multivariate models, a 1 percentage point increase in total government expenditure (% of GDP) was associated with a decrease in SII of −2.468 [95% confidence intervals (CIs): −4.190, −0.746] and RII of −0.026 (95% CIs: −0.048, −0.004). Lower inequalities were associated with higher non-health government expenditure, but not higher government health expenditure. Associations with inequalities were non-significant for GDP, government redistribution, and private health expenditure. Understanding how non-health government expenditure reduces inequalities in IMR, and why health expenditures may not, will accelerate progress towards the Sustainable Development Goals.

Trends and inequities in use of maternal health care services in Indonesia, 1986-2012

Purpose

Overall health status indicators have improved significantly over the past three decades in Indonesia. However, the country's maternal mortality ratio remains high with a stark inequality by region. Fewer studies have explored access inequity in maternal health care service over time using multiple inequality markers. In this study, we analyzed Indonesian Demographic and Health Survey (DHS) data to explore trends and inequities in use of any antenatal care (ANC), four or more ANC (ANC4+), institutional birth, and cesarean section (c-section) birth in Indonesia during 1986-2012 to inform policy for future strategies ending preventable maternal deaths.

Methods

Indonesian DHS data from 1991, 1994, 1997, 2002/3, 2007, and 2012 surveys were downloaded, merged, and analyzed. Inequity was measured in terms of variation in use by asset quintile, parental education, urban-rural location, religion, and region. Trends in use inequities were assessed plotting changes in rich:poor ratio, rich:poor difference, and concentration indices over period based on asset quintiles. Sociodemographic determinants for service use were explored using multivariable logistic regression analysis.

Findings

Between 1986 and 2012, institutional birth rate increased from 22% to 73% and c-section rate from 2% to 16%. Private sector was increasingly contributing in maternal health. There were significant access inequities by asset quintile, parental education, area of residence, and geographical region. The richest women were 5.45 times (95% CI: 4.75-6.25) more likely to give birth in a health facility and 2.83 times (95% CI: 2.23-3.60) more likely to give birth by c-section than their poorest counterparts. Urban women were 3 times more likely to use institutional birth and 1.45 times more likely to give birth by c-section than rural women. Use of all services was higher in Java and Bali than in other regions. Access inequity was narrowing over time for use of ANC and institutional birth but not for c-section birth.

Conclusion

Ongoing pro-poor health-financing strategies should be strengthened with introduction of innovative ways to monitor access, equity, and quality of care in maternal health.