Completeness of a Maternal and Perinatal Mortality Enhanced Surveillance System in Pakistan: Evidence from Capture-Recapture Methods.

Objectives An enhanced surveillance system that integrated health information systems and extended surveillance to previously uncovered areas to capture all births, perinatal and maternal deaths in a rural district of Pakistan was established in 2015, and this study uses capture-recapture methodology to assess completeness. Methods Births and deaths collected by the survey were matched with the data captured by the enhanced surveillance system. Capture-recapture methodology was used to estimate the total number of births and deaths, measure the degree of underestimation, and adjust mortality rates. Results Of all births, 99% were captured by the enhanced surveillance system. Ninety percent of neonatal deaths and 86% of early neonatal deaths were recorded. The recorded neonatal mortality rate was 40 per 1000 live births (95% CI 35-44), and after adjustment for under-enumeration was 42 per 1000 live births (95% CI 37-46). Recorded rates underestimated neonatal mortality by 5% and perinatal mortality by 7%. Five stillbirths were recorded by the survey and all were matched to recorded stillbirths. The one maternal death recorded by the survey was matched with the maternal death captured by the enhanced surveillance system. The maternal mortality ratio prior to adjustment for under-enumeration was 247 per 100,000 live births (95% CI 147-391), whereas after adjustment it was 246 per 100,000 live births (95% CI 146-389). Conclusion Application of capture-recapture methods to the enhanced surveillance system indicated a high completeness of birth and death recording by the surveillance system.