Using relative and absolute measures for monitoring health inequalities: experiences from cross-national analyses on maternal and child health

Abstract

Background: As reducing socio-economic inequalities in health is an important public health objective, monitoring of these inequalities is an important public health task. The specific inequality measure used can influence the conclusions drawn, and there is no consensus on which measure is most meaningful. The key issue raising most debate is whether to use relative or absolute inequality measures. Our paper aims to inform this debate and develop recommendations for monitoring health inequalities on the basis of empirical analyses for a broad range of developing countries.

Methods: Wealth-group specific data on under-5 mortality, immunisation coverage, antenatal and delivery care for 43 countries were obtained from the Demographic and Health Surveys. These data were used to describe the association between the overall level of these outcomes on the one hand, and relative and absolute poor-rich inequalities in these outcomes on the other.

Results: We demonstrate that the values that the absolute and relative inequality measures can take are bound by mathematical ceilings. Yet, even where these ceilings do not play a role, the magnitude of inequality is correlated with the overall level of the outcome. The observed tendencies are, however, not necessities. There are countries with low mortality levels and low relative inequalities. Also absolute inequalities showed variation at most overall levels.

Conclusion: Our study shows that both absolute and relative inequality measures can be meaningful for monitoring inequalities, provided that the overall level of the outcome is taken into account. Suggestions are given on how to do this. In addition, our paper presents data that can be used for benchmarking of inequalities in the field of maternal and child health in low and middle-income countries.

Figure 1

a-d Rate Ratio (comparing the poorest 40% and richest 40% population group) by overall level of the outcome: under-5 mortality, full childhood immunisation coverage, skilled delivery attendance, and skilled antenatal care, for 43 low and middle-income countries. Exponential curves were fitted through the data.

Figure 2

a-d Rate Difference (comparing the poorest 40% and richest 40% population group) by overall level of the outcome: under-5 mortality, full childhood immunisation coverage, skilled delivery attendance, and skilled antenatal care, for 43 low and middle-income countries. Parabolic curves were fitted through the data.

Figure 3

Comparing poor-rich Rate Ratios (richest 20% – poorest 20% population group) in skilled delivery attendance with poor-rich Rate Ratios in prevalence of no skilled delivery attendance.

Figure 4

Rate Ratio (comparing poorest 40% and richest 40% population group) by overall level of the outcome, and mathematically defined ceiling to value of RR. The curves presented for the health-related outcomes correspond to those shown in Fig. 1a-d. The upper x-axis gives the overall-level for under-5 mortality. The lower x-axis gives the overall level for immunisation coverage, antenatal care and skilled delivery attendance.

Figure 5

Rate Difference (comparing poorest 40% and richest 40% population group) by overall level of the outcome, and mathematically defined ceiling to value of RD. The curves presented for the health-related outcomes correspond to those shown in Fig. 2a-d. The upper x-axis gives the overall-level for under-5 mortality. The lower x-axis gives the overall level for immunisation coverage, antenatal care and skilled delivery attendance.