OBJECTIVES: The 25-item National Eye Institute Visual Function Questionnaire (NEI VFQ-25) is widely used to assess patient-reported visual functioning. Previously, the reliability and validity of the NEI VFQ-25 has been assessed in patients with neovascular age-related macular degeneration (AMD). Its measurement performance in other ocular indications, and in large pooled datasets, remains unclear. This study aimed to evaluate the psychometric properties of the NEI VFQ-25 in a large sample of patients across multiple retinal diseases. METHODS: Dataset included pooled baseline NEI VFQ-25 data from six clinical trials in diabetic macular edema (DME), macular edema from branch and central retinal vein occlusion (RVO), neovascular AMD, and myopic choroidal neovascularisation. Rasch analysis was conducted by assessing item fit validity, threshold targeting, item dependency, reliability and stability. RESULTS: Measurement performance was evaluated for 2487 person measurements (mean age: 64±9 (SD) years; gender: 53% male). Mixed psychometric properties of the NEI VFQ-25 were identified. Key strengths identified included a high Person Separation Index of 0.93 (suggesting good reliability), predominantly low residual correlations (suggesting minimal local dependency between items), and no statistically significant Differential Item Functioning (suggesting stability of scoring function across country, study and gender). Some potential psychometric limitations of the NEI VFQ-25 included disordered thresholds for 15 of the 25 items, poor fit according to the Rasch model for several items, and mismatched distributions of person and item threshold locations. These limitations suggest sub-optimal item targeting, potential problems with conceptual clarity, lack of discrimination across some response options and a consistent ceiling effect across all indications. CONCLUSIONS: Rasch analysis of NEI VFQ-25 identified many positive characteristics of this widely used instrument. Several suboptimal characteristics, however, were also identified, suggesting that changes to the NEI VFQ-25 could be considered to improve its psychometric performance in clinical trials of retinal diseases.