Method: We enrolled adults with active RA (6 or more tender joints and physician judgment of activity) who were either starting a new anti-rheumatic medication or who were having escalation of their current medication regimen. Patients had physical examinations and completed a pain visual analog scale (0 - 100), global visual analog scale (0 - 100) and the Health Assessment Questionnaire Disability Index (HAQ; 0 - 3). They were reassessed after either 1 month (for those started on prednisone) or 4 months (all others). At the followup visit, patients were asked to report if they had improvement or worsening in each symptom. These judgments were related to measured changes in pain, patient global assessment, and HAQ.
Result: We studied 215 patients (mean age 52; 75% women; median duration of RA 7 years; 74% seropositive). At entry, they had a mean pain score of 59 ± 25, mean global assessment of 55 ± 25, and a mean HAQ score of 1.3 ± 0.7. Fifty-one patients were treated with prednisone, 64 patients were treated with a new disease-modifying medication, and 100 patients had dose escalation of current medications.
At follow-up, 61% reported improvement in pain, 63% reported improvement in global arthritis status, and 56% reported improvement in function. Changes in measured pain scores, global scores, and HAQ differed among patients who reported improvement, no change, or worsening in each aspect of health status.
|Mean (SD) Change in Pain||Mean (SD) Change in Global Assessment||Mean (SD) Change in HAQ|
|Improved||-26.6 (31.1)||-22.3 (25.4)||-.53 (.59)|
|No Change||-8.0 (20.6)||-7.0 (20.6)||-.08 (.42)|
|Worsened||3.5 (22.7)||0.8 (22.7)||.05 (.35)|
Of those with improvement, 89% rated the improvement as at least “moderately important” for each measure. Areas under receiver operating characteristic curves distinguishing improvement from no change/worsening were .74 for pain, .72 for global assessment, and .75 for HAQ. If the criterion for improvement were required to have a specificity of at least 0.75, an absolute decrease in pain score of 16, an absolute decrease in patient global assessment of 13, and an absolute decrease in HAQ of 0.375 were the cutpoints with maximum sensitivity (0.62, 0.62, and 0.59, respectively).
Conclusion: Patient ratings can accurately distinguish improvement in pain, global assessment, and functioning from no change or worsening. In this sample, decreases in pain score of 16 or more, decrease in global assessment of 13 or more, and decreases in HAQ of 0.375 or more, were specifically associated with clinically important improvement.
Disclosure: M. M. Ward, None; L. C. Guthrie, None; M. I. Alba, None.