Menu
Menu
Menu
Menu
Background: High coverage of care is essential to improving newborn survival; however, gaps exist in access to timely and appropriate newborn care between and within countries. In high mortality burden settings, health inequities due to social and economic factors may also impact on newborn outcomes. This study aimed to examine equity in co-coverage of newborn care interventions in low- and low middle-income countries in sub-Saharan Africa and South Asia. Methods: We analysed secondary data from recent Demographic and Health Surveys in 16 countries. We created a co-coverage index of five newborn care interventions. We examined differences in coverage and co-coverage of newborn care interventions by country, place of birth, and wealth quintile. Using multilevel logistic regression, we examined the association between high co-coverage of newborn care (4 or 5 interventions) and social determinants of health. Results: Coverage and co-coverage of newborn care showed large between- and within-country gaps for home and facility births, with important inequities based on individual, family, contextual, and structural factors. Wealth-based inequities were smaller amongst facility births compared to non-facility births. Conclusion: This analysis underlines the importance of facility birth for improved and more equitable newborn care. Shifting births to facilities, improving facility-based care, and community-based or pro-poor interventions are important to mitigate wealth-based inequities in newborn care, particularly in countries with large differences between the poorest and richest families and in countries with very low coverage of care.
The Demographic and Health Survey Program (DHS) collects health data in high burden mortality settings including newborn care. Surveys are completed at household- and individual-levels, focusing on report from women of reproductive age (15–49 years). Complex, multi-stage sampling and stratification produce nationally-representative results for each country [18]. We analysed secondary DHS survey data since 2015 from low- and low middle-income countries in sub-Saharan Africa and south Asia which included questions on newborn care interventions in the first 2 days of life. We included last (most recent) live births in the 2 years before the survey. Exclusion criteria were births in the 2 days before the survey and neonates who died before the second day. Table 1 shows the included countries, survey years, and number of women interviewed. List of included countries from DHS, survey year, and sample sizes aWeighted, from ICF International [19] b includes only most recent live-born children surviving the first 2 days of life We created a co-coverage index of provider-initiated early newborn care interventions, using a method similar to Victora et al. [20] and Carvajal-Aguirre et al. [9]. We included five provider-initiated interventions included in the WHO standards for maternal and newborn care [21]: 1) examining the umbilical cord, 2) taking the newborn’s temperature, 3) counselling on danger signs in the newborn, 4) counselling on breastfeeding, and 5) observing breastfeeding (Table 2). The primary outcome measure was receipt of 4–5 of these interventions provided in the first two days of life and hereafter called “appropriate newborn care”. Newborn care intervention survey questions. Newborn care interventions and question wordings from the phase seven DHS model questionnaire [22] Key predictor variables focused on social determinants of health using an adapted person-centred conceptual framework (Fig. (Fig.1)1) where individual women and their newborns sit at the centre, encircled by their families and the wider community and structural contexts [17]. Conceptual framework for social determinants of health, adapted from the United Nations Development Programme [17] Simple weighted descriptive statistics on individual intervention coverage and co-coverage by birth location, wealth quintile, and country were calculated. We visually examined patterns of wealth-based inequities (Victora et al. [20]) assigning “top inequity,” “bottom inequity,” and “linear inequity”. Top inequity, also described as ‘mass deprivation’ [26], is when the majority of the population are deprived and only a minority have access to care. Bottom inequity, also described as ‘marginalisation’ [26], is when the majority of the population have access to care but a minority are excluded. Linear inequity, also described as queuing, lies somewhere between top and bottom inequity, with a linear relationship between wealth and access [26]. In descriptive analyses, individual-level weights were applied to account for sampling probability and non-response to ensure each sample was nationally representative. Descriptive results are presented for facility birth and home birth separately. Multilevel, multivariable logistic regression models were fitted, by country for facility birth only to assess the association between the factors in the conceptual framework and appropriate newborn care. For the multilevel models, individual-level weights were denormalised and cluster-level weights were approximated with equal allocation between individual and cluster levels (α = 0.5) using a method described by Elkasabi et al. [27]. The models were adjusted for the independent variables listed above. All statistical analyses for this study were conducted in R [28] and Stata, using the survey package [29] in R and applying the svy commands in Stata where appropriate to adjust for the complex sampling design.
