Arthritis & Rheumatism, Volume 62,
November 2010 Abstract Supplement

Abstracts of the American College of
Rheumatology/Association of Rheumatology Health Professionals
Annual Scientific Meeting
Atlanta, Georgia November 6-11, 2010.

The Development of Fibromyalgia: Examination of Rates and Predictors in Patients with Osteoarthritis (OA).

Wolfe1,  Frederick, Hauser2,  Winfried

National Data Bank for Rheumatic Diseases, Wichita, KS
Technische Universität München, Munich, Germany


Using modified ACR 2010 fibromyalgia (FM) criteria, we have reported that in 9,739 rheumatoid arthritis (RA) patients without FM, a series of demographic, RA, FM, and psychological variables predict FM development. However, it is possible that RA damaged joints or systemic illness may interfere with FM measurements and predictors, and prediction and predictive strength might be different in RA compared with regional rheumatic disorders such as osteoarthritis (OA). To examine this possibility, to compare FM in RA and OA, and to confirm our prediction models, we analyzed predictors of FM in OA.


After excluding patients with FM and those with high levels of FM symptoms (fibromyalgianess score >10), we studied the FM development in 1,888 OA patients during 6,892 patient-years of follow-up over a mean follow-up time of 4.3 (SD 3.5) years (range, 5 to 11). We defined FM using a modification of the ACR 2010 FM diagnostic criteria. We used Cox regression to predict future FM, and examined the discriminatory power and accuracy of predictions using Harrell's C concordance coefficient.


At the last observation, 6.0% of OA compared with 7.4% of RA patients satisfied FM criteria, although 18.8% of OA and 19.8% with RA did so at some point during follow-up, for an OA FM incidence rate of 5.0 (95% CI 4.2, 5.3) and an RA rate of 5.3 (5.1, 5.6) per 100 pt-years. The rate in women was 5.5 (4.9, 6.2) and was 3.5, (4.7, 5.5) in men. Among those satisfying criteria, half of follow-up time after diagnosis was FM criteria (+), and was associated with markedly abnormal OA and FM variable scores. C-statistics, adjusted for age and sex, are shown in Figure 1.

The discriminatory power of variables was similar in OA and RA, except that FM variables were slightly more predictive in OA than RA. In multivariable analyses, demographics were weak predictors of FM (OA C = 0.623, RA C=0.604). Stronger predictors were demographics plus OA (C = 0.725) compared with demographics plus RA variables (C= 0.720), and demographic plus FM variables (OA C = 0.803, RA C = 0.765), and all predictors (OA C = 0.812) (RA C= 0.782) in the multivariable model shown below.

VariableH.R (95% CI)zP-value
Widespread pain index (0–19)1.28 (1.21, 1.36)8.41<0.001
HAQ (0–10)2.06 (1.67, 2.54)6.81<0.001
Symptom count (0–34)1.10 (1.07, 1.13)6.45<0.001
Fatigue (0–10)1.26 (1.17, 1.35)6.40<0.001
Mood (0–10)1.11 (1.02, 1.21)2.500.013
Opioid use1.29 (0.99, 1.67)1.870.062
  11.34 (0.97, 1.84)1.80.072
  21.72 (1.25, 2.37)3.350.001
  31.70 (1.15, 2.49)2.690.007
  4–91.35 (0.86, 2.12)1.290.196

Clinically important HRs were noted for cognition, depression, comorbidity and high levels of OA and FM continuous variables.


FM is predicted similarly in patients with OA and RA. Multiple factors contribute to FM development, including socio-demographic disadvantage, comorbidity, psychological distress, drug and service utilization, FM symptoms (particularly somatic symptom reporting), and functional status; but there is little evidence of the effect of underlying causes.

To cite this abstract, please use the following information:
Wolfe, Frederick, Hauser, Winfried; The Development of Fibromyalgia: Examination of Rates and Predictors in Patients with Osteoarthritis (OA). [abstract]. Arthritis Rheum 2010;62 Suppl 10 :111
DOI: 10.1002/art.27880

Abstract Supplement

Meeting Menu