One of the primary reasons for low healthcare utilization rates in low-income countries is lack of affordable health insurance coverage. In recent years, Community-Based Health Insurance programs are widely implemented across developing countries aimed at increasing healthcare utilization and providing financial protection. This study investigates the causal effects of the program on utilization of healthcare services in Rwanda using nonrandomized household survey data. In a Bayesian potential outcomes framework with Markov Chain Monte Carlo simulation techniques, we address issues of selection bias on observable and unobservable dimensions. In addition, we address heterogeneity by estimating treatment effects at the individual-level. We find that Community-Based Health Insurance schemes significantly increase the likelihood of utilizing medical consultation and screening services but not utilization of drugs. We also find considerable heterogeneity in treatment effects with married women and under-five children benefiting the most.