Abstract
OBJECTIVE: To develop and validate a pain scale that measures the extent of body pain. METHODS: A total of 12,799 patients with rheumatoid arthritis (RA), osteoarthritis (OA), and fibromyalgia (FM) completed a mailed survey regarding the location and intensity of their pain in 38 articular and nonarticular regions. The data were analyzed using item response theory (IRT) by nonparametric Mokken analysis followed by Rasch analysis. The resultant scale was examined for its association with clinical severity variables and its ability to distinguish patients diagnosed with and without FM. RESULTS: The resultant 19 item regional pain scale (RPS) was composed primarily of nonarticular regions. The scale had strong scalability as measured by the Mokken H statistic (H = 0.52), and satisfied the Mokken monotonicity and double monotonicity criteria. The RPS also fit the Rasch model and had satisfactory reliability and separation statistics. Of all clinical variables assessed by survey, the RPS best identified patients diagnosed with FM. In addition, the scale correlated with measures of clinical severity, regardless of diagnosis, and predicted measures of utilization. CONCLUSION: The RPS is a valid scale of pain extent. It can be useful to identify patients with FM or can be used to develop a new definition of FM, even among patients with concomitant illnesses such as RA and OA. In addition, it is a measure of pain extent that is disease independent, and works as well in RA and OA as in FM to identify patients with increased severity and resource utilization.
