OBJECTIVES: Previous Rasch analysis on the National Eye Institute Visual Function Questionnaire (NEI VFQ-25) has suggested that a modified instrument may provide improved coverage and targeting of patient-reported visual function for patients with eye diseases. The aims of this study were to propose a Rasch-based scoring structure, the NEI VFQ-25-R, and to assess its psychometric performance. METHODS: The NEI VFQ-25-R was developed by combining response levels for disordered thresholds for 15 NEI VFQ-25 items, excluding four poorly fitting items and including six optional near and distance subscale items. NEI VFQ-25-R items were regrouped into two subscales, Activity Limitation and Socio-Emotional Impact, based on literature, item face validity and clinician input. Performance was evaluated using Rasch analysis by assessing item fit validity, threshold targeting, item dependency, and reliability using pooled baseline data (2487 person measurements) from six clinical trials across four retinal diseases. RESULTS: No disordered thresholds were observed for any NEI VFQ-25-R items, and only three misfitting items were reported. The estimated Person Separation Index was 0.94, suggesting good reliability. Only two items had a residual correlation >0.30, demonstrating minimal local dependency between items. Distribution of item thresholds revealed a reduced ceiling effect compared with the NEI VFQ-25. Threshold targeting was better for patients with neovascular age-related macular degeneration, lower visual acuity, and low reported visual function. CONCLUSIONS: Collapsing disordered response categories and excluding misfitting items improved psychometric performance for the NEI VFQ-25-R compared with the NEI VFQ-25. The reduced ceiling effect may reflect more effective population targeting and improved instrument coverage. Additional research will determine whether the two subscale structure of the NEI VFQ-25-R provides better interpretability of potential differences in patient-reported visual functioning than the NEI VFQ-25.