Measurement of maternal mortality, United Republic of Tanzania.

Maternal mortality, defined as the death of a woman during pregnancy or shortly after childbirth, is a key target of the sustainable development goals. Yet, in many low- and middle-income countries, accurate measurement of maternal mortality remains a major challenge. Reliable death registration systems are often incomplete or absent, making it difficult to calculate the maternal mortality ratio. However, opportunities exist to improve measurement. In about two thirds of low- and middle-income countries, more than 80% of births now occur in health facilities. This shift means that health facilities can serve as a valuable source of data not only for local quality improvement but also for informing population mortality estimates. However, this requires high-quality data. To improve the quality of facility-based data, we propose four simple quality metrics: (i) plausibility of reported national and subnational levels and patterns; (ii) the ratio of reported stillbirths to maternal deaths; (iii) the distribution of causes of death; and (iv) the share of deaths among women aged 35 years and older. Using data from the United Republic of Tanzania as an example, we show how these metrics can be applied to assess the completeness and internal consistency of routine health data systems. Our example aims to encourage quality assessment of facility-based data in other countries, alongside: (i) individual death reporting using standardized classification of causes of death; (ii) centralized data quality assessment based on the proposed metrics; and (iii) regular national on-site assessment of the completeness and accuracy of maternal and perinatal death reporting.