The prevalence of fibromyalgia in the general population: a comparison of the American College of Rheumatology 1990, 2010, and modified 2010 classification criteria.

OBJECTIVE

The American College of Rheumatology (ACR) 1990 fibromyalgia classification criteria are based on the presence of widespread pain and tenderness. In 2010, new criteria were proposed that focused more on multiple symptoms, and these criteria were later modified to require only self report of symptoms. The current study aimed to determine the population prevalence of fibromyalgia and to compare differences in prevalence using the alternative criteria.

METHODS

A cross-sectional survey was conducted. Questionnaires, including items on pain, symptoms, and rheumatologic diagnoses, were mailed to 4,600 adults in northeast Scotland. Participants who had chronic widespread pain or those who met the modified 2010 criteria, plus a subsample of other participants, were invited to attend a research clinic. Attendees completed an additional questionnaire and underwent a rheumatologic examination, and their signs and symptoms were classified according to the ACR 1990, 2010, and modified 2010 criteria. The prevalence of fibromyalgia according to each set of criteria was calculated, weighting back to the target population by age, sex, and area of residence.

RESULTS

Of 1,604 questionnaire participants, 269 were invited to attend the research clinic, and 104 (39%) attended; 32 of these subjects (31%) met ≥1 set of fibromyalgia criteria. The prevalence of fibromyalgia according to the 1990, 2010, and modified 2010 criteria was 1.7% (95% confidence interval [95% CI] 0.7-2.8), 1.2% (95% CI 0.3-2.1), and 5.4% (95% CI 4.7-6.1), respectively. The ratio of females to males was 13.7:1, 4.8:1, and 2.3:1 of those meeting the respective criteria sets.

CONCLUSION

Fibromyalgia prevalence varies with the different sets of classification criteria applied. In particular, prevalence is higher and a greater proportion of men are identified with the modified 2010 criteria as compared to the criteria sets requiring clinician input. This has important implications for the use of the new criteria, both in research and in clinical practice.