“…They are usually necessary if the CT is to generate valid results. Nor is this meant to suggest that these compromises with the ''best'' treatment are done casually or that clinical trials researchers are not concerned to care for their subjects as best as possible given the requirements of the designs of the trials (see, e.g., Zelen, 1979;Kopelman, 1983;Gifford, 1986;Freedman, 1987;Makuch and Johnson, 1989;Volavka, Cooper, Laska, & Meisner, 1996). Indeed, the opposite is often true.…”