Addressing the persistent problems of access to and delivery of health care in rural India requires a better understanding of the individual provider choice decision. The first essay investigates the determinants of outpatient health care provider choice in rural India in the mixed multinomial logit (MMNL) framework. The simulated maximum likelihood (SML) method is used to estimate the parameters. Using data from National Sample Survey Organization of India, we find the following: price and distance to a health facility play significant roles in health care provider choice decision; when health status is poor, distance plays a less significant role in an adult's provider choice decision; and outpatient care for children is more price-elastic than that for adults.

The SML method often becomes infeasible in estimating the MMNL model parameters when random coefficients are allowed to be correlated and when non-normal distributions are specified for the random parameters. Bayesian framework provides a natural way to deal with these complexities. The second essay provides a Bayesian analysis of the MMNL framework as applied to models of health care provider choice. Posterior distributions of the parameters are approximated using Markov Chain Monte Carlo methods. In particular, we use Metropolis-Hastings algorithm within a Gibbs sampler to make random draws from the resulting joint posterior distribution. The findings of the chapter show that models with normally distributed coefficients fit the data better. Price and distance were found to be significant determinants of health care provider choice in rural India.

The third essay analyzes ambulatory health care expenditures in rural India using a modified two-part model, where the second part models conditional quantiles of expenditure distribution. Evidence of differential impacts the determinants of health care expenditures was found. Specifically, the effect of household income, health status, illness duration and illness severity were found to vary across expenditure quantiles in rural India.