Abstract
Purpose: :
The purpose was to assess and re-engineer a conventional visual disability questionnaire, the Catquest questionnaire, using Rasch analysis in order to optimize item fit to the construct, minimize test length and create a linear measure of visual disability.
Methods: :
Catquest data from 21,364 consecutive cataract patients were extracted from the Swedish National Cataract Register. This included both first and second eye cataract patients both with and without comorbid eye disease. The Catquest contains 7 visual disability items, 6 items assessing the frequency of performing activities, 2 symptoms items and 2 global assessment items. The fit of these 17 items to a single construct was assessed with Rasch analysis using Winsteps v3.64.2 applied with an Andrich rating scale for each question type. Due to the large sample size, tests of significance were not performed.
Results: :
The existing response scale thresholds were ordered for all items. All items could not be included in a single construct. The only subgroup of items which formed a valid measure were the 7 visual disability items (person separation, 2.08, infit range 0.78 to 1.28, outfit range 0.78 to 1.26). The frequency items did not form a valid measure and corrupted the measurement of visual disability if combined with the visual disability items. The symptoms items grossly misfit if included with the visual disability items (infit range 2.30 to 2.58, outfit range 3.05 to 3.70). The global assessment items contributed to the measurement so were included in the final short-form Catquest (person separation, 2.50; reliability, 0.86; infit range 0.81 - 1.31 and outfit range 0.77 - 1.26). The items were well targeted to the subjects (mean difference -0.31 logits).
Conclusions: :
The Catquest Questionnaire has been modified to create a 9-item short-form which validly quantifies the visual disability of cataract surgery patients on a linear scale. This revision of the Catquest will simply visual disability measurement for major national registries. Rasch scaling coefficients which convert categorical data into a continuous variable are supplied to simplify clinical application.
Keywords: quality of life • cataract • clinical research methodology