OBJECTIVES: Prospective observational studies, which inform on the effectiveness of interventions in natural settings, may complement results from randomised clinical trials in the evaluation of health technologies. However, observational studies are subject to a number of potential methodological weaknesses, mainly selection and observer bias. This analysis applies various methods to control selection bias in the estimation of treatment effects. We also address the issues of estimation and inference in a multilevel setting. METHODS: We describe and compare the use of regression methods, propensity score matching, fixed-effects models incorporating investigator characteristics, and a multilevel, hierarchical model using Bayesian estimation techniques in the control of selection bias. To illustrate these methods, we have used data from the Schizophrenia Outpatient Health Outcomes (SOHO) study, a large, prospective, observational study of health outcomes associated with the treatment of schizophrenia. RESULTS: The methods used to adjust for differences between treatment groups that could cause selection bias yielded comparable results and reinforces the validity of the findings. CONCLUSIONS: Observational studies, when properly conducted and when using adequate statistical methods, can provide valid information on the evaluation of health technologies.