We performed a combined manual and computer search of the FMS literature to identify controlled clinical trials in FMS from 1980 to June 1994 inclusive. Our specific objectives were: 1) to determine which outcome measures have been used in clinical trials for FMS, and the methods utilized to measure these outcomes; 2) to identify which outcome measures were most and least sensitive in distinguishing between treatment groups, and 3) to identify weakness in trial design. Our analysis of 24 clinical trials demonstrates the large diversity of outcome measures and measurement instruments that have been used to detect differences between treatment and placebo in the management of FMS. Whereas certain outcomes, such as self-reported pain and sleep quality, were frequently measured, other clinically important outcomes, such as functional and psychological status, were infrequently included in data collection. Finally, we identified several significant potential sources of bias, including potential flaws in subject selection and group allocation, inadequate randomization, incomplete blinding, errors in outcome measurement, and inappropriate analysis of data.