The 36-Item Short Form Health Survey (SF-36) and Work Productivity and the Impairment (WPAI) Questionnaire in Panic Disorder

The SF-36 is a valid, reliable and responsive measure of QOL in panic disorder, and the WPAI did not perform as well, perhaps due to a combination of limited sample size and high variability in WPAi scores.

Abstract

Available evidence suggests that panic disorder causes severe degradation in quality of life and work productivity (WP). There is growing interest in evaluating pharmacotherapy in panic disorder, but its effects on quality of life or WP have not been assessed, perhaps because appropriate scales have not been validated specifically in panic disorder. This study examines the internal consistency, test/retest reliability, construct validity and responsiveness of the Medical Outcomes Study 36-Item Short-Form Health Survey (SF-36) and the Work Productivity and Activity Impairment (WPAI) questionnaire when used in panic disorder. Clinical evaluations were performed at baseline and at weeks 4 and 8. Patients self-administered the SF-36 and WPAI at weeks 4 and 8. The SF-36 performed well in terms of internal consistency and test/retest reliability, with the exception of the 'role limits due to emotional health' scale. For the other scales, Cronbach's α ≥ 0.83 and correlations between week 4 and 8 scores were significant (p < 0.01) for patients with a stable clinical condition. There were no significant differences in mean SF-36 scores obtained at weeks 4 and 8 for stable patients. Construct validity was also supported. As hypothesised, we detected significant correlations between the SF-36 Mental Health Component Summary (MCS) and clinical measures of disease severity. Evidence of responsiveness was also obtained: patients with fewer panic attacks during the second month of follow-up improved on the MCS, while those with a stable or increased frequency of panic attacks did not (8 point versus 1 point improvement, p < 0.01). The WPAI did not perform as well, perhaps due to a combination of limited sample size and high variability in WPAI scores. Only 44 of 77 individuals were employed at baseline and completed WPAI items. The standard deviation for the measure 'overall work productivity' was more than half the mean score. Neither test/retest reliability nor construct validity were demonstrated. Although WPAI scores improved for patients showing clinical improvement, this result is incongruous with the lack of evidence supporting its validity. The SF-36 is a valid, reliable and responsive measure of QOL in panic disorder. It is unclear whether the WPAI possesses these properties when used in panic disorder.