Differential Item Functioning With the National Eye Institute Visual Function Questionnaire-25 in Patients With Vitreomacular Traction

Author(s)

Benedicte Lescrauwaet, MSc1, Konrad Pesudovs, PhD2;
1Xintera bv, Director, Outcomes Research, Ghent, Belgium, 2University of New South Wales, Professor, Faculty of Medicine and Health, Sydney, Australia
OBJECTIVES: The National Eye Institute Visual Function Questionnaire-25 (NEI VFQ-25) is a patient-reported outcome measure (PROM) frequently used in ophthalmic studies. This PROM was administered in Phase 3 ocriplasmin randomized clinical trials to evaluate changes in vision-related function in patients with vitreomacular traction (VMT) with or without epiretinal membrane (ERM) or macular hole (MH). Differential item functioning (DIF) occurs when subgroups of respondents with comparable levels of ability/function respond differently to an item, implying the estimate of item difficulty is influenced by sample characteristic. We investigate if items of the visual functioning scale (VFQ-VFS) and socioemotional scale (VFQ-SES) function differently by types of respondents using DIF analysis.
METHODS: We combined the untreated cohorts from 3 trials. For DIF testing, respondents were stratified by clinical subgroup (VMT; VMT+ERM; VMT+MH), gender (male; female), age (<70; ≥70 years), visual acuity (VA) (<70; ≥70ETDRS letters), lens (phakic; pseudophakic), vitreomacular adhesion (VMA) width (broad; focal). Rasch analysis(Winsteps 5.8.4) explored possible interaction of respondent-groups with each item, one at a time. We defined DIF based on the magnitude as follows: small/absent difference: ≤0.50 logit; minimal difference: >0.50-<1.0 logit;notable difference: ≥ 1.0 logit.
RESULTS: The dataset included 1,131 responses. There was no DIF for VA and VMA width. We found minimally different responses between respondent-groups: for VFQ-SES items “Frustrated” (0.64 harder for ≥70 years; 0.56 harder for VMT+MH vs. VMT+ERM), “See how people react” (0.54 harder for VMT+MH vs VMT) and “Visiting with people” (0.57 harder for VMT+ERM vs VMT+MH); for VFQ-VFS items “Going down steps in dim light” (0.61 harder for female), “Eyesight” (0.51 harder for VMT+MH vs. VMT+ERM).
CONCLUSIONS: Notable DIF was not observed with VFQ-VFS or VFQ-SES items. Minimal DIF or item bias (interactions between individual items and types of respondents defined by sample characteristics) were observed with 3 VFQ-SES items and 2 VFQ-VFS items.

Conference/Value in Health Info

2025-05, ISPOR 2025, Montréal, Quebec, CA

Value in Health, Volume 28, Issue S1

Code

PCR107

Topic

Patient-Centered Research

Topic Subcategory

Instrument Development, Validation, & Translation

Disease

No Additional Disease & Conditions/Specialized Treatment Areas, SDC: Sensory System Disorders (Ear, Eye, Dental, Skin)

Your browser is out-of-date

ISPOR recommends that you update your browser for more security, speed and the best experience on ispor.org. Update my browser now

×