Responsiveness of fibromyalgia clinical trial outcome measures

Objective. To assess the responsiveness of the Fibromyalgia Impact Question naire (FIQ), patient ratings of pain intensity, number of tender points, an d total tender point pain intensity score to perceived changes in clinical status in patients with fibromyalgia (FM). Methods. Using data from a randomized placebo controlled study evaluating e fficacy of magnetic therapy in patients with FM, the ability of primary out comes to detect clinically meaningful changes over a 6 month period was ass essed by: (1) degree of association between outcome change scores and patie nt global ratings of symptom change (Spearman rank-order correlations); (2) ability of these scores to discriminate among groups of patients whose per ceived health status had changed to varying degrees (ANOVA): (3) ability of these scores, individually and jointly, to discriminate between patients w ho had reported improvement and those who did not (logistic regression); (4 ) effect size, standardized response mean, and Guyatt's statistic were calc ulated to quantify responsiveness. Results. Correlations showed the outcome measures were moderately responsiv e to perceived symptomatic change. For FIQ, pain intensity ratings and numb er of tender points, differences in change scores between globally improved and unchanged groups and between globally improved and worsened groups wer e significant; for total tender point pain intensity, the globally improved differed from worsened group. FIQ outperformed the other measures in discr iminating between patients who reported improvement from those who did not. Summary statistics were consistent with discriminatory analyses, indicatin g the measures were sensitive to improvement, but relatively unresponsive t o decline. Conclusion. The FIQ was the most responsive measure to perceived clinical i mprovement and we recommend its inclusion as a primary endpoint in FM clini cal trials.