Randomized clinical trials are the gold standard for establishing the evidence that guides practice. They are the only way to be sure that unrecognized confounders are not influencing the differences in therapies. This is why guidelines and other authoritative documents place so much weight on randomized trials. There is confidence that if the same type of patients were enrolled with the same methodology in future trials, the average result will come out the same.

But does this answer the question of the patients who are not in the trial? Are there confounders that place them at one end of a spectrum rather than the average? Do they have features that cause them not to be enrolled in the trial? In multicenter revascularization trials such as BARI, COURAGE, and ISCHEMIA, many patients are screened at each site in order to enroll a few from each, resulting in significant selectivity. The same may be true for drug trials or device trials, but the impediment to enrollment is less since the differences in treatment are perceived by the patient to be minor and blinding is possible.