The impact of directly observed therapy on the efficacy of Tuberculosis treatment: a Bayesian multilevel approach

We propose and discuss a Bayesian procedure to estimate causal effects for multilevel observations in the presence of confounding. This work is motivated by an interest in determining the causal impact of directly observed therapy on the successful treatment of Tuberculosis. We focus on propensity score regression and covariate adjustment to balance the treatment allocation. We discuss the need to include latent local-level random effects in the propensity score model to reduce bias in the estimation of causal effects. A simulation study suggests that accounting for the multilevel nature of the data with latent structures in both the outcome and propensity score models has the potential to reduce bias in the estimation of causal effects.

Automated summary

We propose a Bayesian procedure to estimate causal effects for multilevel observations in the presence of confounding. The motivation is to determine the causal impact of directly observed therapy on the successful treatment of Tuberculosis. We discuss the inclusion of latent local-level random effects in the propensity score model to reduce bias in the estimation of causal effects. A simulation study suggests that accounting for the multilevel nature of the data with latent structures in both the outcome and propensity score models has the potential to reduce bias in the estimation of causal effects.