Validation of the visual disability questionnaire (VDQ) in India.

PURPOSE To validate visual disability questionnaire (VDQ) in patients with low vision in India, and explore whether the two latent traits "importance" and "difficulty" associated with performance of daily activities are valid and independent constructs. METHODS The VDQ consisting of 25 items was administered verbally to 137 subjects with low vision aged 16 to 89 years. Responses for each item were rated for importance and difficulty using a 5-category Likert scale. Rasch analysis was used to estimate interval measures of response ratings. RESULTS Subjects could discriminate only three response categories for importance and difficulty. Content validity was demonstrated by good separation indices for importance (4.24 and 2.59 for the item and person parameters, respectively) and difficulty ratings (7.64 and 3.33, respectively). High reliability scores were recorded for importance (0.95 and 0.87) and difficulty ratings (0.98 and 0.92). The most important items were "grooming" (1.15 logits) and "reading newspaper" (0.97 logits). Although "threading a needle" was the least important item (-2.79 logits), it was the most difficult task (3.13 logits). The least difficult item was moving around in familiar places (-2.51 logits). A poor correlation was observed between the item measures (r = -0.19, p = 0.34) and also between person measures (r = 0.18, p = 0.03) of importance and difficulty. CONCLUSIONS The VDQ is a valid questionnaire with importance and difficulty ratings found to be independent constructs. This questionnaire could be used to prioritize the goals for rehabilitation intervention in patients with low vision.