Maternal Cash Transfers Led To Increases In Facility Deliveries And Improved Quality Of Delivery Care In Nigeria

Abstract

Ninety-nine percent of global maternal deaths occur in low- and middle-income countries. The high mortality rates are often attributed to a large portion of births occurring outside of formal health care facilities. This has prompted the creation of programs to promote the use of formal delivery care. However, poor-quality care in health facilities in low- and middle-income countries is well documented. It is not clear that shifting births into health facilities in these settings necessarily leads to better-quality care. We present results from a randomized controlled trial in Nigeria that evaluated a conditional cash transfer intervention that paid pregnant women to deliver in a health facility. We found that the intervention led to a 41 percent increase in facility deliveries. We also found improvements in the quality of delivery care (as a result of more births taking place in formal health care settings) and in overall satisfaction with care. We found no evidence of a reduction in preventable complications that led to maternal deaths, though we found some improvements in self-reported health. Our results indicate that promoting facility deliveries can improve the quality of care received, even in settings where formal care quality is poor. However, modest quality improvements might not be sufficient to substantially improve health outcomes.

Keywords: Access and use; Children’s health; Conditional incentives; Education; Health savings accounts; Mortality; Organization of care; Quality Of Care; global health; health policy; maternal care; quality improvement.

Figures

Exhibit 2. Percentage-point differences in facility birth and skilled attendance between the treatment and control arms of the study of a conditional cash transfer for delivery care in Nigeria

SOURCE Authors’ analysis of data from household surveys carried out by the authors during 2017 and 2018. NOTES Each point represents a coefficient from a separate linear probability model that regressed the relevant outcome on the conditional cash transfer indicator, controlling for mother’s age, ethnicity, education, number of prior births, prior birth history (history of stillbirths, miscarriages, or abortions), dummies for each household wealth quintile (constructed using principal component analysis), time between the interview and the birth, birth year × month fixed effects, and health facility service area fixed effects. The error bars indicate 95% confidence intervals that were estimated using standard errors clustered by the health facility service area.

Exhibit 3. Percentage-point differences in the quality of delivery care between the treatment and control arms of the study of a conditional cash transfer for delivery care in Nigeria

SOURCE Authors’ analysis of data from household surveys carried out by the authors in 2017 and 2018. NOTES Each point represents a coefficient from a separate linear probability model that regressed the outcome on the conditional cash transfer indicator, controlling for the variables listed in the notes to exhibit 2. The error bars indicate 95% confidence intervals, as explained in the notes to exhibit 2. Receipt of oxytocin, controlled cord traction, uterine stimulation, and pain medication during labor represent technical quality. Having no physical or verbal mistreatment by delivery care providers represents interpersonal quality. “Cord traction” is explained in the text.

Exhibit 4. Percentage-point differences in maternal health between the treatment and control arms of the study of a conditional cash transfer for delivery care in Nigeria

SOURCE Authors’ analysis of data from household surveys carried out by the authors during 2017 and 2018. NOTES Each point represents a coefficient from a separate linear probability model that regressed the outcome on the CCT indicator, controlling for the variables listed in the notes to exhibit 2. The error bars indicate 95% confidence intervals, as explained in the notes to exhibit 2.