Reading the research literature. Putting randomized clinical trials into practice. 2

Randomized clinical trials (RCT) are our best available method for determining whether a treatment works or has what we technically call efficacy. RCTs set out to demonstrate that a particular treatment regimen works better than a conventional standard treatment or placebo, for a particular type of patient, with a particular condition or indication. The results of an RCT are, generally, used by the Food and Drug Administration (FDA) as the basis for approving treatments--especially drug treatments. The approval of a drug generally indicates that at least two independently conducted, randomized clinical trials have demonstrated that the treatment has efficacy for at least one indication. Once a drug is FDA approved, a physician has the authority to prescribe it for the approved, as well as other, indications. However, if used for another indication, a physician may risk a greater liability, should the outcome not be as expected. Unfortunately, RCT conducted for one indication cannot be used to support either the efficacy or safety of the treatment for another indication. Even when used for an approved indication, it is difficult to know what to expect when a new treatment is put into practice. The most difficult part of reading medical literature is to apply or put it into practice--what we call extrapolation. Extrapolation differs from interpretation, which refers to results of patients included in the investigations, while extrapolation asks questions about those who receive the treatment in practice. This important distinction is called effectiveness--when a treatment works in practice, as opposed to efficacy, which indicates it works under the conditions of an RCT. Authors of a randomized clinical trial draw conclusions about effectiveness as well as efficacy, because they are eager to have their successful treatment used and, despite being tempered by the peer review process, are likely to encourage its widespread adoption. Even if the author is not biased toward adopting the treatment, it is important to recognize that you, as the practitioner, know your patients best. Your patients may be quite different from those included in the RCT. Therefore, we need a systematic approach in order to draw conclusions from an RCT and apply them in practice. This paper will take a look at two basic questions in order to apply the results of an RCT. It will ask how large an impact can we expect on average for patients who are different from those in the RCT, and what conclusions can we draw about safety from an RCT?