Measures of activity and damage in rheumatoid arthritis: Depiction of changes and prediction of mortality over five years

Abstract
Objective. To analyze various quantitative measures of inflammatory activity and joint damage, including articular, radiographic, laboratory, questionnaire, and physical function measures, in regard to changes in status in surviving patients and prediction of mortality in non‐survivors over 5 years in a cohort of patients with rheumatoid arthritis (RA) monitored in the mid‐1980s and early 1990s. Methods. A comprehensive evaluation, which included a complete joint count, radiograph, laboratory tests, physical measures of function, and self‐report questionnaire scales, was performed at baseline and 5 years later in 210 consecutive patients with RA. Results. Five years after baseline, 206 of the 210 patients were accounted for: 37 had died, 130 had a comprehensive repeat assessment, and 39 had a more limited repeat assessment. In surviving patients, most measures of activity were generally unchanged or somewhat better, including joint tenderness, pain on motion, and swelling; erythrocyte sedimentation rate and rheumatoid factor; as well as questionnaire scores for pain, global status, helplessness, and difficulty in performing 8 activities of daily living (ADL) according to a modified Health Assessment Questionnaire (MHAQ). By contrast, measures of damage, including joint deformity, grip strength, walk time, and radiographic scores, indicated worse status. Mortality over 5 years was predicted significantly in univariable analyses by American College of Rheumatology (formerly the American Rheumatism Association) Functional Class, limited joint motion, scores for MHAQ, global status, helplessness, grip strength, walk time, button time, and number of comorbidities and duration of diseases, as well as the sociodemographic measures of age and formal education. In multivariable Cox regressions, age, comorbidities, MHAQ, and other measures of functional status were the most effective predictors of 5‐year mortality. Conclusion. In patients with RA, most measures of inflammatory activity were unchanged and sometimes better, while measures of damage indicated worse status in the same patients over 5 years. Measures indicating functional disability, as well as age and comorbidities, predict 5‐year mortality more effectively than radiographic and laboratory data. Measures of inflammatory activity may underestimate long‐term outcomes in RA, and long‐term studies should include measures of damage.