Estimation of Maternal Mortality Rate in Iran from 2010 to 2014 Using Capture-Recapture Method

Due to a strong correlation between maternal mortality rate (MMR) and developmental indicators, MMR is regarded as a development index in communities. One of the deficiencies in the MMR data registries is the under-reporting due to the rare incidence of maternal mortality in the target population and the diversity of information resources. This study used a three-source capture-recapture method to estimate the MMR. This method can be used to calculate the under-reporting of different data registries. The sample size included all maternal mortality cases registered between 2010 and 2014 in the National Maternal Survey System, National Organization for Civil Registration, and National Death Registration System. Population estimation models including log-linear models, sample coverage approach (SCA), and the Akaike’s information criterion (AIC) were employed by the capture-recapture method. The MMR was estimated to be 22.15, 21.49, 19.55, 19.77, and 18.97 per 100000 live births from 2010 to 2014, respectively. The coverage rate of the three sources of data was estimated at 73%, 83%, 82%, 80%, and 83% between 2010 and 2014, respectively. The results of the study indicate that the capture-recapture method using three sources of data can be utilized to estimate the MMR with higher validity than using only one data source.