Abstract
BACKGROUND: Estimation of health-related quality of life (HRQoL) in diabetic retinopathy (DR) patients is important to assess the impact of disease, to monitor the treatment outcomes, and conduct health technology assessments. The study aimed to measure the HRQoL in DR patients using the generic as well as disease specific instruments, identify the determinants of HRQoL, empirically investigate the concurrent validity between the QoL instruments, and to develop statistical approaches to map NEI-VFQ-25 (National Eye Institute- Visual Function Questionnaire- 25) composite score based on EQ-5D-5 L (EuroQol 5-dimensions 5-levels) utility values. METHODS: A facility based cross-sectional study was carried out to measure the HRQoL of 300 DR patients using EQ-5D-5 L, EuroQol visual analogue scale (EQ-VAS), and NEI-VFQ-25 instruments. Mean HRQoL scores, along with DR state specific and visual acuity specific utility values were analysed. Pearson correlation coefficient was used to ascertain concurrent validity between NEI-VFQ-25 composite score and its sub-scale scores, EQ-5D-5L index value, and EQ-VAS score. Lastly, we evaluated statistical models using predictor sets based on the EQ-5D-5 L utility scores to estimate NEI-VFQ-25 scores. RESULTS: The mean EQ-5D-5 L utility value for DR patients was 0.69 (95% CI: 0.65-0.73). The mean NEI-VFQ-25 composite score and EQ-VAS score were 61.1 (95% CI: 57.5-64.5), and 67.6 (95% CI: 65.6-69.5), respectively. Both EQ-5D-5 L and EQ-VAS had a strong concurrent validity with NEI-VFQ-25 composite score. 'Tobit regression with log of composite score' was assessed to the preferred model to predict the NEI-VFQ-25 composite score using EQ-5D-5L utility values. CONCLUSION: Diabetic retinopathy has a decremental impact on quality of life, which increases with severity in vision loss. Both generic as well as disease-specific outcome measures are equally reliable to determine quality of life of patients with DR. The EQ-5D-5 L can be utilized for measurement of HRQoL in DR in clinical settings to optimize time of clinicians, with further derivation of NEI-VFQ-25 values through application of the crosswalk algorithm for predicting vision- related QoL.