Electronic supplementary material
The online version of this article (https://doi.org/10.1186/s12884-018-1734-0) contains supplementary material, which is available to authorized users.
Having high-quality data available by 2020, disaggregated by income, is one of the Sustainable Development Goals (SGD). We explored how well coverage with skilled birth attendance (SBA) is predicted by asset-based wealth quintiles and by absolute income.
We used data from 293 national surveys conducted in 100 low and middle-income countries (LMICs) from 1991 to 2014. Data on household income were computed using national income levels and income inequality data available from the World Bank and the Standardized World Income Inequality Database. Multivariate regression was used to explore the predictive capacity of absolute income compared to the traditional measure of quintiles of wealth index.
The mean SBA coverage was 68.9% (SD: 24.2), compared to 64.7% (SD: 26.6) for institutional delivery coverage. Median daily family income in the same period was US$ 6.4 (IQR: 3.5–14.0). In cross-country analyses, log absolute income predicts 51.5% of the variability in SBA coverage compared to 22.0% predicted by the wealth index. For within-country analysis, use of absolute income improved the understanding of the gap in SBA coverage among the richest and poorest families. Information on income allowed identification of countries – such as Burkina Faso, Cambodia, Egypt, Nepal and Rwanda – which were well above what would be expected solely from changes in income.
Absolute income is a better predictor of SBA and institutional delivery coverage than the relative measure of quintiles of wealth index and may help identify countries where increased coverage is likely due to interventions other than increased income.