Seminars

General Departmental Seminar Series

Individual and Population Effectiveness of Therapies in Observational Studies

Stephen J. Gange, PhD, Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health

Friday, Oct 18, 2002, 12:00 pm

G5/142 Clinical Sciences Center (600 Highland Ave.)

ABSTRACT

The evaluation of therapies in observational studies is challenged by selection-by-indication biases, whereby individuals who choose or are chosen to be on therapies are generally in poorer health. As such, analyses of therapies in these settings must be appropriately formulated, and can be framed in two ways. Individual effectiveness analyses supplement clinical trials by direct comparisons of individuals on and off therapy, adjusting for time-varying confounding effects using 'causal analysis' approaches (e.g. marginal structural models). Population effectiveness analyses complement clinical trials through the use of instrumental variables to evaluate the change in disease burden across groups with heterogeneous therapy use. Here, we demonstrate some examples of individual and population effectiveness analyses in the context of evaluating highly-active antiretroviral therapy in HIV/AIDS observational studies.

Back to General Departmental Seminar Series