Abstract

---

"Objectives: This paper asks whether measles vaccination can reduce socioeconomic differentials in under five mortality rates (U5MR) in a setting characterized by extreme poverty and high levels of childhood mortality. Design: Longitudinal cohort study based on quasi experimental design. Setting: Data come from the phased introduction of a measles vaccine intervention in Matlab, Bangladesh in 1982. Subjects: There were 16 270 Bangladeshi children aged 9–60 months. Intervention: The intervention cohort received measles vaccine. Main outcome measures: Socioeconomic differentials in U5MR between the lowest and highest socioeconomic status (SES) quintiles in a cohort of 8135 vaccinated children and a cohort of unvaccinated age matched controls. Mantel–Haenszel rate ratios for the lowest to highest SES quintile were computed. SES was measured by factor analysis of maternal schooling, land holdings, dwelling size, and number of rooms. Results: The U5MR ratio of lowest SES to highest was 2.27 (95% CI=1.62–3.19) in the unvaccinated population and 1.42 (95%CI=0.94–2.15) in the vaccinated population. The difference between unvaccinated and vaccinated U5MR ratios was statistically significant (p50:10) and robust across alternative measures of SES. Conclusion: Children from the poorest quintile were more than twice as likely to die as those from the least quintile in the absence of measles vaccination. Universal distribution of measles vaccination largely nullified SES related mortality differentials within a high mortality population of children."