Objectives The aim was to study physical function in individuals with rheumatoid arthritis (RA) as measured by SOFI (observed physical function) and HAQ (self-rated physical function) over 15 years and describe associated variables, with focus on gender differences.
Methods In all 2558 patients (847 men and 1711 women) age (SD) 58 (16) were recruited from the BARFOT inception cohort of patients with early RA. They had at inclusion a disease duration of one year or less and fulfilled the ACR 1987 criteria. At 15 years follow-up 663 out of 690 patients participated. Physical function was assessed by the SOFI (Signals of functional impairment) test, (scores 0-44, best to worst) which includes 12 performance tests measuring objective physical function, and the HAQ (Health Assessment Questionnaire) (scores 0-3, best to worst) measuring self-reported activity. A logistic regression model was performed to assess if being in the highest quartile of SOFI and HAQ, respectively, at the15 year follow up visit was associated to gender. Age, disease duration at inclusion, disease activity, smoking habits, RF positivity and pain were included in the model.
Results Women had lower mean SOFI than men at inclusion and during the first study year, p<0.001. HAQ showed a conversed pattern, where women reported worse physical function than men on all occasions, p<0.001. During the first year SOFI and HAQ decreased in both genders, p<0.001. Thereafter throughout the study period mean SOFI and HAQ increased in men and women, p<0.001, figure 1A and B.
At the 15 year follow up visit being in the highest quartile of SOFI (score ≥10) was not associated with gender while women had a higher risk to be in the highest quartile of HAQ (score ≥1.13) OR (95% CI) 2.86 (1.73-4.74), p<0.001. DAS 28 at inclusion showed a weak association with SOFI, OR (95% CI) 1.36 (1.11-1.68), p=0.003 while pain at inclusion was somewhat closer associated with HAQ, 1.02 (1.01-1.03), p<0.001. The correlation between the two functional test was r=0.54.
Conclusions Women had an almost three times higher risk of worse outcome of HAQ after 15 years while the outcome of SOFI was not associated with gender. These two measures provide information of different aspects of physical function and should be used concomitantly.
Disclosure of Interest None declared