“Health claims -- whether mainstream or "alternative" -- should be evaluated by asking, "what is the evidence?" Reliable information comes mainly through use of the scientific method, a procedure for exposing hypotheses (assumptions) to critical examination and testing. The scientific method does not rely on testimonials as evidence of fact. Rather, it provides an objective way to collect and evaluate data.

The scientific method has at least three noteworthy characteristics.

First, it is self-correcting. Scientists do not assume that this method discovers absolute truth but rather that it produces conclusions that subsequent studies may modify. In this sense, science is cumulative.

Second, the scientific method requires objectivity. Findings must not be contaminated by the personal beliefs, perceptions, biases, values, or emotions of the researcher. Only when the results speak for themselves should the conclusions be considered valid. Research results often lead to new questions that need exploration.

Third, experiments must be reproducible. One study, taken alone, seldom proves anything. To be valid, one researcher's findings must be repeatable by others.

Competent researchers may use anecdotal reports for suggesting new hypotheses, but never as supporting evidence. The fact that a person recovers after doing something is rarely sufficient to demonstrate that the recovery was caused by the action taken and is not simply coincidental (due to the natural course of the ailment) or due to a placebo effect (a favorable treatment response that does not result from pharmacologic effect or other direct physical action). Moreover, reports of personal experiences can be biased, inaccurate, or even fraudulent. Well-designed experiments involving many people are needed to establish that a treatment method is effective. Without them, even honest, competent practitioners can be misled by their clinical experiences.

Controlled clinical trials compare an experimental group of people who receive the treatment being tested and a control group of people who receive a different treatment or no treatment. For example, members of the experimental group may receive a pill with active ingredients, whereas those in the control group receive another treatment, an inert substance (placebo), or no treatment. Studies may be conducted "blind" or "double-blind" to minimize or eliminate the effect of bias on data collection and interpretation. In blind studies the participants do not know which treatment they receive. In double-blind studies neither the people administering the treatment nor the experimental subjects know who gets what.

Medical "facts" are determined through a process in which hundreds of thousands of scientists share their observations and beliefs. Editors and editorial boards of scientific journals play an important role by screening out invalid findings and enabling significant ones to be published. Expert panels convened by government agencies, professional groups, voluntary health agencies, and other organizations also contribute to this effort. When controversies arise, further research can be devised to settle them. Gradually, a shared set of beliefs is developed that is considered scientifically accurate. Reference: Scientific Method