Validation of Indicators Related to Emergency C-Sections in Urban Ghana

Background: Cesarean delivery is the only surgery for which we have nearly global population-based data, thanks primarily to large-scale surveys of reproductive-aged women in low income countries.  Few surveys provide any other cesarean-related data. Given weaknesses in many health information systems, health planners will likely need to rely on nationally representative surveys for years to come.  Surveys may be able to provide valid information on cesarean delivery. The objective of this study is to validate self-reported data on the classification of cesarean deliveries among women delivering in Korle Bu Teaching Hospital in Accra, Ghana.

Methods: This study compares women’s self-report of the timing of the decision for cesarean and other indications regarding cesarean delivery against hospital-based data, considered the gold standard. Data were abstracted from individual medical records and interviews with women having undergone cesarean within specified dates in the hospital were conducted just prior to their hospital discharge. The study assessed sensitivity and specificity of self-reported responses to questions regarding emergency versus non-emergency cesarean delivery, and estimates the percent of emergency cesarean deliveries that would be obtained from a survey, given the observed percent, sensitivity and specificity resulting from this study. 

Results: Hospital data were matched with exit interviews for 659 women delivered via cesarean. Sensitivity and specificity were 79% and 82%, respectively, for emergency cesarean delivery;  95% and 98% , respectively, for previous cesarean delivery;  84% and 74%, respectively, for spontaneous labor and 67% and 95% respectively for labor induction. The percent of emergency cesareans estimated from a survey would be only 7% higher than the true rate, would nearly replicate the rate for previous cesarean and would be 17% and 33% higher for spontaneous labor and labor induction, respectively.

Conclusions: Population-based surveys are promising sources for data on emergency cesarean and some other indicators related to cesarean delivery and intrapartum care.