Abstract
OBJECTIVE: To develop a short, 4-item arthritis severity questionnaire that is simple to score, clinically useful and meaningful, and suitable for use in primary care, where osteoarthritis (OA) is the primary prevalent arthritis illness. METHODS: Data and items from the Medical Outcomes Study Short Form 36 (SF-36), the Western Ontario McMaster Osteoarthritis Index (WOMAC(c)), and visual analog scales (VAS) for pain and patient global severity were studied in 16,519 patients with arthritis. The Short Arthritis Assessment Scale (SAS) was developed by performing multivariable analyses that involved individually adding/subtracting items in differing regression models. The candidate items and scales were then studied by Rasch analysis, and tested for effect size, sensitivity to change, and reliability. The resultant scale was validated using data from a recent OA clinical trial. RESULTS: The VAS pain and VAS global severity scales and 2 items from the WOMAC in the VAS format, difficulty going down stairs and difficulty shopping, were found to be the best predictors of change in health status. The 4-item SAS was reliable (Cronbach's alpha = 0.87), demonstrated good test-retest reliability (Lin's concordance coefficient = 0.85), was unidimensional, and was strongly correlated with other important clinical measures, indicating good construct validity. Using data from a recent randomized clinical trial in OA, the SAS performed better than the WOMAC pain scale and the SF-36 physical component score in detecting change, and at least as well as the clinical trial VAS pain scale. CONCLUSION: The SAS is a 4-item arthritis severity questionnaire that can be easily administered in primary care for patients with OA, but is suitable for use across all arthritis illnesses. Scoring is simple, requiring only the addition of four 10-point scales, and interpretation is straightforward. The SAS may have a role in rapid assessment of the arthritis patient in primary care practice.
In this issue
