THE STANFORD-INDIA GVK EMERGENCY MANAGEMENT AND RESEARCH INSTITUTE (GVK EMRI) STUDY: EARLY FINDINGS

Purpose: Prior to 2005, India had no large-scale, centralized emergency medical system or ambulance service. Through a public-private partnership, the Emergency Management and Research Institute (GVK EMRI) has emerged as India’s largest ambulance service provider, currently serving about 730 million people. This study provides the first quantitative evidence of GVK EMRI’s early impact on population-level infant and maternal health outcomes. 

Method(s): We use data from GVK EMRI’s internal electronic records merged with birth level records from two population-based surveys conducted in Andhra Pradesh and Gujarat (the first two states served). Primary outcomes are indicators for neonatal death, infant death, and maternal health complications. Other outcomes include indicators for institutional delivery, skilled birth attendance, and delivery-related expenses. We exploit district-month variation in GVK EMRI service intensity to estimate changes in outcomes associated with the scale-up of services using ordinary least squares difference-in-difference regression models.

Result(s): In Andhra Pradesh, we find that GVK EMRI is associated with significant reductions in the probability of neonatal and infant mortality of about 7.7 and 11 per 1,000 live births (respectively).  In Gujarat, the relationship between GVK EMRI services and neonatal and infant mortality varies substantially by initial (pre-implementation) mortality conditions, with larger absolute reductions in districts with higher initial mortality rates.  For every standard deviation increase in the baseline mortality rate, GVK EMRI service intensity reduces the probability of neonatal and infant death by an additional 1.5 and 2.1 per 1000 (respectively), and reduces the probability of maternal health complications by 1 percentage point. We find little change in the probability of institutional delivery or skilled birth attendance. Consistent with qualitative reports, these findings together suggest that maternal and child health gains associated with GVK EMRI’s services may be linked to more timely initiation of active labor management, risk screening, and more effective treatment of women with obstetric emergencies.     

Conclusion(s): GVK EMRI’s services may have played an important role in improving maternal – and in particular, neonatal and infant – health outcomes in India in recent years.  More generally, this study provides important new evidence on the potential role of public-private partnerships for health service delivery in complex institutional environments.