Responsiveness of the SF-36 and the Health Assessment Questionnaire Disability Index in a systemic sclerosis clinical trial

Abstract

Objective: This study compares the responsiveness to change of the Medical Outcomes Study Short Form Health Survey (SF-36), a measure of health related quality of life (HRQOL), and the Health Assessment Questionnaire Disability Index (HAQ-DI), a function instrument, in a randomized clinical trial for treatment of systemic sclerosis (SSc).

Methods: A phase 2/3, multicenter, prospective, placebo controlled trial was conducted to evaluate human recombinant relaxin treatment in patients with diffuse SSc over 24 weeks. At baseline, subjects had stable, moderately severe, diffuse SSc of disease duration < or = 5 years, modified Rodnan skin score > or = 20, serum creatinine < 2.0 mg/dl, percentage forced vital capacity (% FVC) predicted > or = 50%, and % DLCO predicted > or = 40% and were not receiving concomitant disease modifying therapies. Internal consistency reliability of multi-item scales was estimated using Cronbach's alpha. Responsiveness to change of the SF-36 and HAQ-DI was computed between Weeks 0 and 24. Subjects were classified as unchanged or having a meaningful change in 4 different external measures: Change in (1) skin score > or = 30%; (2) % FVC predicted of > or = 15%; (3) self-reported patient global assessment by visual analog scale (VAS) > or = 20%; and (4) physician global assessment by VAS of > or = 20%. Responsiveness indices were computed and Cohen's effect size criteria were used to assess the magnitude of change.

Results: A total of 239 patients participated in this trial, with 196 completing the 24 week trial. Cronbach's alpha for the SF-36 scales ranged from 0.76 to 0.93 and for the HAQ-DI ranged from 0.69 to 0.91 (good to excellent). The SF-36 had a larger magnitude of responsiveness in overall disease (patient and physician global assessment) compared to the HAQ-DI, while the HAQ-DI had a larger magnitude of responsiveness in clinical measures (i.e., change in skin score and % FVC predicted) than the SF-36.

Conclusion: These data support inclusion of both the SF-36 and HAQ-DI as outcome measures in future clinical trials of diffuse SSc.