Abstract :
Outcomes research (OR) proposes to study the appropriateness of health care services in order to identify the manner by which diseases can be most effectively prevented, diagnosed and treated. Proponents of OR claim that this methodology will do this better than the randomized clinical trial (RCT), which we have traditionally used as the gold standard to obtain answers to many of the same questions. OR, like RCTs, uses the tools of epidemiology and biostatistics but OR studies the factors involved in the process of care in existing large data bases, not in specifically recruited and highly selected cohorts. Contrary to what some critics of RCTs have said, both methods try to understand what is responsible for better clinical outcomes but OR concentrates more on patient satisfaction and the cost-effective use of health care resources. RCTs, which have been called “efficacy” studies, have actually provided us with considerable data on outcomes and much valuable guidance for therapy. These efficacy studies, however, have been widely criticized for not reflecting the “real world” of medical care and for creating artificial outcomes which can not replicated in the usual clinical setting. This presentation will review clinical trial methodology, discuss RCTs which are designed to be “effectiveness” studies (large simple trials) and compare to RCTs to OR. I will suggest that unless OR finds a way to control bias and identify the innumerable confounders that can effect the therapeutic process, it will not be able to accomplish its goals.
