Assessing the Validity of Intervention Coverage Indicators for Maternal and Newborn Health Care in Kenya

Session: New Horizons in Measurement for Maternal and Newborn Health

Presentation: Ann Blanc, Population Council
Download presentation slides

Background: In the absence of high quality health system statistics, tracking the coverage of lifesaving maternal and newborn health interventions often relies on reports from surveys of women. Despite widespread use, most indicators proposed as measures of maternal and newborn intervention coverage have not been validated. This study assessed the ability of women delivering in two Kenyan hospitals to recall key interventions received during the intrapartum and early postpartum period at two time points: at discharge and 14 months following hospital delivery.

Methodology: Women’s self-reports of care received at both time points were compared against observations by trained third party observers at the time of delivery. Indicators selected for validation were either currently in use or have the potential to be included in population-based surveys. We calculated individual-reporting accuracy using the area under the receiver operating curve (AUC), estimated population-level accuracy using the inflation factor (IF), and compared the accuracy of women’s reporting at baseline and follow-up.

Results: Of 606 women who participated at baseline and agreed to follow-up, 515 were re-interviewed. Of 34 indicators, 27 had sufficient sample size for validity analysis at follow-up. Eleven indicators met one acceptability criteria: 4 for AUC (>0.6) and 7 for IF (0.75<IF<1.25). Five indicators met both criteria: main provider was a nurse-midwife, support person present during birth, injection received within first few minutes of delivery (i.e., uterotonic to prevent postpartum hemorrhage (PPH)), cesarean section, and low birthweight infant. Apart from uterotonic for PPH, these indicators also met both criteria at baseline. There was a significant decline in reporting accuracy for 5 indicators with time, none of which influenced the acceptability classification.

Conclusions: Women are able to report accurately on some aspects of labor and delivery care. Overall, results do not suggest that there is significant deterioration in reporting accuracy over time.