Background: An enhanced surveillance system that aimed to capture all births, perinatal and maternal deaths in a rural district of Pakistan was established for 12 months in 2015, capturing 7,580 pregnancies and 7,273 live births.
Aim: To assess the completeness of this enhanced surveillance system and to calculate corrected rates by adjusting for missing births and deaths, using capture-recapture methodology.
Methods: A stratified random cross-sectional survey was conducted six months after the start of the enhanced surveillance. Births and deaths collected from this were matched with data from the enhanced surveillance system. The Petersen–Lincoln equation was used to estimate the total number of births and deaths. Mortality rates were adjusted for the estimated missed births and deaths, and mortality rate ratios and 95%CIs calculated.
Results: Of the 175 births enumerated in the survey, 173 (99%) were captured by the enhanced surveillance system. Nine of ten neonatal deaths and six of seven (86%) early neonatal deaths were matched with data from the enhanced surveillance system, as were all five stillbirths and the one maternal death. Rates calculated from the enhanced surveillance system underestimated neonatal mortality by 5% (40 vs 42 per 1,000 livebirths) and perinatal mortality by 7% (60 vs 63 per 1,000 live births), but these differences were not statistically significant.
Conclusion: The completeness of birth and death recording in the enhanced surveillance system was high. Capture-recapture methodology is useful for assessing completeness of existing surveillance systems and for correcting any under-counts in births and deaths in the population.