The best and most reliable form of research is the double-blind, placebo-controlled study. A treatment cannot really be said to be proven effective unless it has been examined in properly designed and sufficiently large studies of this type.
In these experiments, one group of subjects receives the "real thing"—the active substance being tested. The other half receives a placebo designed to appear, as much as possible, like the real thing. Individuals in both groups don't know whether they are getting the real treatment or placebo (they are "blind"). Furthermore, the researchers administering placebo and real treatment 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. It is true, although hard to believe, that people given placebo (fake) treatment frequently report dramatic and long-lasting improvements in their symptoms. However, if the people in the real treatment group fare significantly better than those in the placebo group, it is a strong indication that the treatment really works.
What you've just read is a highly simplified introduction to a crucial and non-intuitive topic. If you'd like a more in-depth analysis, see the article Why Does This Database Rely on Double-blind Studies?