Observational Studies

University Of Pennsylvania, Philadelphia PA

Investigators

Abstract

An observational study is an empirical investigation of the effects of a treatment, policy, intervention or exposure that was not randomly assigned to subjects, as it would be in a randomized experiment. The central difficulty in an observational study is that, because treatments were not randomly assigned, the subjects receiving different treatments may not be comparable, so differing outcomes after treatment may not be effects caused by the treatment. If the treatment groups differ before treatment in ways that have been measured, there is an overt bias that often can be removed by adjustments, such as matching. Often there is concern that treatment groups differed before treatment in ways that were not measured, that is, concern about hidden biases. Hidden biases cannot generally be removed by adjustments and must be addressed by other means. Statistical methodology for observational studies concerns adjustments for overt biases and methods of design and analysis to address hidden biases. Building upon earlier work funded by NSF, the current project will extend statistical methods for overt and hidden biases. Observational, or nonexperimental, studies of treatment effects are common in most fields that study people, because harmful or unwanted treatments cannot be imposed on human subjects for experimental purposes. They are common in studies of public health, public policy, medicine and economics. New statistical methods will be developed for use in observational studies, and they will be illustrated with examples from labor economics, genetic epidemiology, injury prevention, and cancer treatment for children.