The Visual Function Questionnaire: Utility Index: Does It Measure Glaucoma-related Preference-based Status?

Abstract

As health budgets tighten globally, evaluating the cost-effectiveness of glaucoma services is vital; however, there is a lack of validated instruments that measure utility specific to glaucoma patients. We thus evaluated the validity of the Visual Function Questionnaire-Utility Index (VFQ-UI) as a measure of vision-related function and preference-based status in glaucoma. Cross-sectional study of 141 volunteer patients over 40 years of age with moderate (n=64) or severe (n=36) glaucoma, and 41 controls. Sociodemographic information, visual acuity, and visual field test parameters were obtained. The VFQ-UI and Glaucoma Activity Limitation-9 patient-reported outcome tools were administered, and their psychometric properties explored using Rasch analysis. Criterion, convergent, and divergent validity of the VFQ-UI were also assessed. Mean age was 69.7 (±10.8) years, with a mean better eye visual acuity (LogMAR) of 0.074 (±0.356) (approximate Snellen acuity of 20/24). Overall, the VFQ-UI had satisfactory fit to the Rasch model, however targeting and precision were suboptimal with a person separation index of 1.72 and person reliability coefficient of 0.75. VFQ-UI scores significantly increased as glaucoma severity worsened between controls (-4.54, ±1.15) and patients with moderate (-3.77, ±1.57) and severe glaucoma (-1.98, ±2.98), indicating good criterion validity (P<0.001). Strong correlation between VFQ-UI and Glaucoma Activity Limitation-9 scores (r=0.764, P<0.001) demonstrated good convergent validity. There was no significant correlation between VFQ-UI scores and age and sex (r=0.150 and 0.026, respectively), demonstrating good divergent validity. The VFQ-UI shows potential to measure preference-based status in a cohort of glaucoma patients, with the means to calculate utility. Although the VFQ-UI displayed reasonable fit to the Rasch model overall and had good criterion, convergent and divergent validity, its lack of precision is a limitation. The difficulty of the VFQ-UI in discriminating among patients with varying glaucoma severity may be improved by including items specific to glaucomatous visual dysfunction.