[Skip to Content]
[Skip to Content Landing]
June 9, 1993

Evaluating Behavioral Interventions: Need for Randomized Controlled Trials

JAMA. 1993;269(22):2845. doi:10.1001/jama.1993.03500220031011

To the Editor.  —The impact of the acquired immunodeficiency syndrome (AIDS) epidemic may be influenced by the development of effective treatments, vaccines, or methods to assist behavior change. Prevention strategies may be guided by an assessment of the quality of evidence supporting different possible interventions. We wanted to assess the strength of evidence produced by studies of these interventions. To do this, we reviewed computerized abstracts (AIDSLINE) from the international AIDS conferences of 1989 through 1992 using the US Preventive Services Task Force hierarchy for strength of evidence.1 Abstracts were first selected if they contained the words trial: (presence of a colon after the word means any ending to the word will be selected) with either random: or control:. Second, we selected abstracts containing the word cohort: with either drug therapy as a Medical Subject Heading or behavior: and change: or vaccine:. These abstracts were then reviewed independently by