In this type of study, two groups of subjects receive a treatment. One group receives an untested or new drug, and the other group receives an older drug about which more is typically known. Individuals in both groups don't know what type of treatment they are getting. Furthermore, the researchers administering the treatments are also kept in the dark about which group is receiving which treatment (making it a "double-blind" experiment). This last part is important, because it prevents the researchers from unintentionally tipping off the study participants, or unconsciously biasing their evaluation of the results. The purpose of this kind of study is to eliminate the power of suggestion.
Such studies are especially useful in determining whether a new treatment offers any advantages over an older one. For statistical reasons, they are not quite as good at proving whether a treatment is effective as a
double-blind placebo-controlled study, the best and most reliable form of research.
A good double-blind study should enroll at least 100 people, preferably as many as 300. Dramatically effective treatments can prove themselves in somewhat smaller trials; however, research involving 30 or fewer people generally doesn't prove anything at all.