Objective: The purpose of this study was to assess the construct validity of two global assessment questions, the Patient Global Impression of Severity and of Improvement, in female patients with stress urinary incontinence.
Study design: This was a secondary analysis of data from two double-blind, placebo-controlled studies that evaluated duloxetine for the treatment of predominant stress urinary incontinence in the United States (n = 1133 patients). Assessment variables included incontinence episode frequency, the Incontinence Quality of Life Questionnaire results, fixed volume (400 mL) stress pad test results, and the Patient Global Impression of Improvement and of Severity question results.
Results: Spearman correlation coefficients were 0.36, 0.20, and -0.50 among the Patient Global Impression of Severity question and incontinence episode frequency, stress pad test, and Incontinence Quality of Life Questionnaire results, respectively (all P <.0001). Mean incontinence episode frequency and median stress pad test results increased and mean Incontinence Quality of Life Questionnaire results decreased with increasing Patient Global Impression of Severity question severity levels. Similarly, significant (P <.0001) correlations were observed between the Patient Global Impression of Improvement question response categories and the three independent measures of improvement in stress urinary incontinence (0.49, 0.33, and -0.43 with incontinence episode frequency, stress pad test, and Incontinence Quality of Life Questionnaire results, respectively). As with the Patient Global Impression of Severity question, differences in mean changes for Incontinence Quality of Life Questionnaire and median percent changes for incontinence episode frequency and stress pad test among the Patient Global Impression of Improvement question response categories were highly significant (P <.0001). These relationships indicate appropriate and significant associations between the Patient Global Impression of Severity and of Improvement questions and the three independent measures of stress urinary incontinence severity and improvement, respectively.
Conclusion: The Patient Global Impression of Severity and of Improvement question responses were correlated significantly with incontinence episode frequency, stress pad test, and Incontinence Quality of Life Questionnaire measures, which established the construct validity of these two global assessment questions for baseline severity and treatment response, respectively.