|
Abstract:
|
A key issue in planning a clinical trial is the population to target. To carry out trials most efficiently one wishes to have a study population that 1) has the condition or disease under study; 2) is likely to benefit from an effective treatment; and 3) can show benefit quickly enough to establish effects in a feasible time frame. Some choices are obvious: some disease/conditions affect only one gender (e.g., prostate cancer), relate to age (e.g, menopausal symptoms), or are specific to geographical areas (e.g., tropical diseases). Other criteria have been controversial. Studies of prevention of cardiovascular events routinely limited study entry to men on the grounds that because men were at higher risk of such outcomes a smaller sample size would have adequate power to detect treatment effects. After concerns arose in the 1990s that effects might be different in women and that prevention of cardiovascular disease was an important issue for women as well as men, Congress mandated inclusion of both genders in studies that affected both genders. Other examples of study selection criteria that raised questions about the generalizability of study results will be discussed.
|
