The best studies are done with randomized control trials (RCT). This means patients are randomly put into one of two groups. The first group is the experimental group. The second group is the placebo (control) group. Subjects are placed in groups by flipping a coin. The surgeon may open a card marked "experimental" or "control" for each patient in the operating room just before the surgery.

RCTs allow researchers to say, "This caused that." A blinded RCT means that no one in the study knows who is in each group. Neither the subjects nor the researchers are aware of the group differences. This helps reduce and sometimes prevent bias toward one result over another.

RCTS still have some problems. If the subject groups contain volunteers the study can't be blinded and there may be some bias just in the fact that they volunteered. If withholding treatment for the control group is unethical then an RCT can't be done for that study.