Catquest questionnaire: re-validation in an Australian cataract population.

Background: The Catquest questionnaire was developed using traditional methodology to enable cataract surgery outcomes assessment in European countries. Recently, it has been validated using Rasch analysis in a Swedish population resulting in the Catquest-9SF. The aim of the present study was to assess the performance of the Catquest and the Catquest-9SF questionnaires using Rasch analysis in Australian cataract patients.

Methods: A total of 217 cataract patients awaiting surgery at Flinders Medical Centre, Adelaide, South Australia self-administered the Catquest questionnaire. This is a 19-item instrument containing frequency, difficulty, symptoms and global rating items. Rasch analysis was undertaken to assess the unidimensionality (whether all the items are measuring a single underlying construct using principal components analysis or PCA), person separation (ability of the questionnaire to distinguish between strata of patient ability) and targeting of item difficulty to person ability.

Results: Similar to the previous validation study, the original Catquest questionnaire required revision because of misfit and multidimensionality necessitating removal of the frequency items. The revised version was similar to the Catquest-9SF although an extra driving item was a valid optional inclusion. The Catquest-9SF performed well in the Australian cohort satisfying all criteria for valid measurement including unidimensionality. However, targeting of item difficulty to person ability was marginally worse compared with the Swedish cataract population indicating the Australian cataract patients present for surgery at lower levels of visual disability.

Conclusions: The Catquest-9SF is a reliable and valid measure of visual disability in the Australian cataract population.