The cornerstone of successful medical treatment is some assurance that the proposed treatment can alleviate symptoms. This assurance is based on the results of properly designed and performed scientific research studies. Unfortunately psychiatry, especially child and adolescent psychiatry, lacks the sheer number of studies performed in other areas of medicine. This imbalance should begin to change as the Federal and Drug Administration (FDA) requires that studies for approval of medications should include children.
Most FDA medications are approved for use in adults for a particular purpose. Physicians, however, may use them in other manners. When this is done, as when medications are prescribed for children, they are referred to as being used "off label". This practice is not limited to psychiatry. In fact, drugs commonly prescribed for children with asthma are often "off label". Additionally, certain medical advances have been achieved by the "off label" use of medications. For example, benefits were realized in new areas of health when medication originally approved for depression proved beneficial for some forms of enuresis, and when the same drug given to adults with schizophrenia was found to help reduce nausea and vomiting in chemotherapy patients.

Although much of the initial information regarding psychiatric medication use comes from adult studies, child and adolescent psychiatrists may consider trying them in younger age groups and encourage future approval for children. This process can begin when professionals describe their successes in single cases. Then studies are designed in which several patients are exposed to the treatment in a standardized procedure. If the outcome is positive, further exploration will proceed. The next step is the "gold standard" - the double blind, placebo controlled study. This research design requires the treating physicians and patients to be "blind" as to whether the patient is on active medication or a placebo. Thus bias is reduced and a comparison of medication effect to that of a placebo can be conducted. The "ultimate" test is to repeat the study in other settings in order to replicate the findings.