摘要：Using data from a experimental supportive intervention to India's malaria control program, we study the impact of local capacity on mosquito net usage and fever care seeking patterns. The intervention was conducted simultaneously by three NGOs in two endemic districts in the state of Orissa, and we find that program impact varies significantly by district. Examining three potential sources of this variation (differential population characteristics, differential health worker characteristics, and differential implementer characteristics), we provide evidence that the implementing agency can significantly affect both the success of a program and the external validity of RCT results.
