Field test results of the motherhood method to measure maternal mortality

BACKGROUND & OBJECTIVES Measuring maternal mortality in developing countries poses a major challenge. In Nepal, vital registration is extremely deficient. Currently available methods to measure maternal mortality, such as the sisterhood method, pose problems with respect to validity, precision, cost and time. We conducted this field study to test a community-based method (the motherhood method), to measure maternal and child mortality in a developing country setting. METHODS Motherhood method was field tested to derive measures of maternal and child mortality at the district and sub-regional levels in Bara district, Nepal. Information on birth, death, risk factors and health outcomes was collected within a geographic area as in an unbiased census, but without visiting every household. The sources of information were a vaccination registry, focus group discussions with local health workers, and most importantly, interview in group setting with women who share social bonds formed by motherhood and aided by their peer memory. Such groups included all women who have given birth, including those whose babies died during the measurement period. RESULTS A total of 15,161 births were elicited in the study period of two years. In the same period 49 maternal deaths, 713 infant deaths, 493 neonatal deaths and 679 perinatal deaths were also recorded. The maternal mortality ratio was 329 (95%CI: 243-434)/100,000 live birth, infant mortality rate was 48 (44-51)/1000LB, neonatal mortality rate was 33 (30-36)/1000LB, and perinatal mortality rate was 45 (42-48)/1000 total birth. INTERPRETATION & CONCLUSIONS The motherhood method estimated maternal, perinatal, neonatal and infant mortality rates and ratios. It has been field tested and validated against census data, and found to be efficient in terms of time and cost. Motherhood method can be applied in a time and cost-efficient manner to measure and monitor the progress in the reduction of maternal and child deaths. It can give current estimates of mortalities as well as averages over the past few years. It appears to be particularly well-suited to measuring and monitoring programmes in community and districts levels.