Scaling up effective maternal and child health interventions in low- and middle-income countries

Abstract

In low- and middle-income countries, significant health gains could be made if health interventions that are known to be effective were delivered with greater coverage, quality, and efficiency to those that need them. In this dissertation, I examine three issues related to the scale-up of effective maternal and child health interventions in low-resource settings: the effectiveness of an intervention to improve immunization coverage and equity in India, the gaps in adherence to clinical guidelines for postpartum hemorrhage care in referral hospitals in Kenya, and the appropriate methods for estimating the costs of services delivered in health facilities. In my first study, I evaluate the effectiveness of a campaign-style intervention for increasing the coverage and equity of routine immunization delivery, in the context of India’s Intensified Mission Indradhanush (IMI) initiative. Using controlled interrupted time-series analysis, I compare vaccination trends in districts that participated in IMI with trends in districts that did not participate. I find that, during the four-month implementation period, IMI increased the delivery of thirteen infant vaccines, with a median 10.6% increase in delivery volume across these vaccines. I do not find evidence of a sustained effect after the implementation ended. These findings suggest that campaign-style interventions can successfully increase routine immunization coverage over short periods, particularly for early infant vaccines, but other approaches may be needed to achieve sustained coverage improvements. In my second study, I measure adherence to guideline-based quality indicators for postpartum hemorrhage prevention, identification, and management in referral hospitals in Kenya. Using direct clinical observation of deliveries, I measure the proportion of deliveries in which Kenyan and WHO clinical guidelines for PPH care were followed and the timeliness of care provided. I find significant gaps in adherence to clinical guidelines, particularly for guidelines related to monitoring patients for signs of complications: providers did not follow guidelines for postpartum patient monitoring in any of the observed deliveries. These findings suggest that efforts to reduce maternal morbidity and mortality in low- and middle-income country settings should emphasize improvements in PPH care generally, and in patient monitoring specifically. In my third study, I review methods used to estimate service delivery unit costs using data from multiple health care facilities in low- and middle-income countries and conducted a simulation study to evaluate the performance of different methods. I find that one common method, the simple mean of unit costs across sites, has a large upward bias when applied in multiple empirical datasets. Alternative methods, including the volume-weighted mean, regression, and calibration estimators, perform better in terms of root mean squared error and bias. These findings suggest that choice of summary estimator in multi-site costing studies can significantly influence study findings and, therefore, the economic analyses they inform.