Comparability of sociodemographic and pregnancy characteristics of pregnancy-related deaths identified via the sisterhood method versus the household/verbal autopsy method

Objective To compare sociodemographic and pregnancy characteristics of pregnancy-related deaths identified by the direct sisterhood and the verbal autopsy with household mortality (HHVA) methods. Methods Nationally representative data for 1997-2001 were obtained from the household, verbal autopsy, and women's questionnaires of the Bangladesh Maternal Health Services and Maternal Mortality Services Survey, 2001. Sociodemographic and pregnancy characteristics were compared for maternal deaths identified by the two methods. Characteristics of deceased women were reported directly with HHVA, but extrapolated in the direct sisterhood method using the reporting sister as proxy. Results Overall, 201 pregnancy-related deaths were identified via HHVA and 388 through DS reporting. There were no significant differences between reporting sister characteristics and deceased women's characteristics in educational attainment, working status, husband's educational attainment, and spouse educational parity. However, timing of death relative to pregnancy phase, number of previous live births, and years since death did differ (P < 0.05). Conclusion The sociodemographic characteristics of women with pregnancy-related deaths identified via the two methods were similar. However, some pregnancy characteristics differed significantly, suggesting that different policy interventions are required. Before considering using sister proxy characteristics to target services, issues responsible for these differences should be resolved, and generalizability of evaluated indicators must be considered.