PURPOSE:Generic instruments to assess health utilities can be used to express the burden of health problems in widely used indexes. That is in contrast with what can be obtained with condition-specific instruments, outcomes are very specific and difficult to compare across conditions. The purpose of this study was to assess health and visual outcomes and its determinants in patients with visual impairment (VI) using the EQ-5D-3L and the Activity Inventory (AI). METHODS:Participants were recruited in different hospitals during the PCVIP-study. A total of 134 patients with acuity 0.30 logMAR or less in the better eye were interviewed. The AI includes 46 goals split between three objectives: social functioning, recreation and daily living, and was used to measure visual ability. The EQ-5D consists of five questions covering one domain each and was used to provide a measure of health states. Responses to each domain were combined to produce a single individual index. RESULTS:The AI and the EQ-5D-3L showed enough discriminatory power between VI levels (p < 0.001), and their results were strongly correlated r(134) = 0.825, (p < 0.001). Explanatory factors for visual ability were level of VI in better eye, age and gender, R2  = 0.43, (p < 0.001). Explanatory factors for the EQ-5D-3L were level of VI in the better eye, comorbidities and gender, R2  = 0.36, (p < 0.001). CONCLUSION:Our results showed that the EQ-5D-3L is useful when characterizing the burden of VI and to compute, when necessary, quality-adjusted-life-years (QALY) changes due to VI. However, it is important to consider that the EQ-5D-3L uses a coarse response scale, assesses a limited spectrum of domains and is influenced by comorbidities. This might limit its responsiveness to small changes in visual ability.