Catquest-9SF questionnaire and eCAPS: Validation in a Canadian population

Abstract

Background: Visual acuity alone has limitations in assessing a patient's appropriateness and prioritization for cataract surgery. Several tools, including the Catquest-9SF questionnaire and the electronic cataract appropriateness and priority system (eCAPS) have been developed to evaluate patients-reported visual function as related to day-to-day tasks. The aim of this study was to validate Catquest-9SF and eCAPS in a Canadian population and propose a shorter version of each, in an attempt to extend their applicability in clinical practice.

Methods: The English translation of the Swedish Catquest-9SF and eCAPS were self-administered separately in pre-operative patients in tertiary care in Peel region, Ontario. Rasch analysis was used to validate both scales and assess their psychometric properties, such as category threshold order, item fit, unidimensionality, precision, targeting, and differential item functioning.

Results: A total of 313 cataract patients (mean age = 69.1, 56.5% female) completed the Catquest-9SF and eCAPS. Catquest-9SF had ordered response thresholds, adequate precision (person separation index = 2.09, person reliability = 0.81), unidimensionality and no misfits (infit range 0.75-1.35, outfit range 0.83-1.36). There mean for patients was equal to -1.43 (lower than the mean for items which is set automatically at zero), meaning that tasks were relatively easy for respondent ability. eCAPS had 3 items that misfit the Rasch model and were excluded (infit range 0.82-1.30, outfit range 0.75-1.36). Precision was inadequate (person separation index = 0.19, person reliability = 0.04). 78.8% of subjects scored≤9 (answered that they had no issues for most questions).

Conclusions: Catquest-9SF demonstrated good psychometric properties and is suitable for assessing visual function of care-seeking patients referred for cataract surgery in Ontario, Canada. There was some mistargeting, suggesting that the tasks were relatively easy to perform, which is consistent with previous research. On the contrary, eCAPS is not sensitive in differentiating patients who had impaired visual functioning.

Fig 1. Category probability curves for Catquest-9SF.

A) all 9-items, B) global assessment items Ca and Cb only, and C) daily-life activities items C1-C7 only.

Fig 2. Person-item map for Catquest-9SF.

Respondents are shown on the left side of the dashed line and items are shown on the right side. Respondents who are less disabled (have higher visual function) and items that are more difficult (respondents are more likely to rate having greater difficulties with tasks) are located at the bottom. Each ‘#’ represents 2 respondents and each ‘.’ represents 1 respondent. M = mean, S = 1 standard deviation, T = 2 standard deviations. The scale is in logits.

Fig 3. Category probability curve for eCAPS questionnaire.

Fig 4. Person-item map for eCAPS.

Respondents are shown on the left side of the dashed line and items are shown on the right side. Respondents who are less disabled (have higher visual function) and items that are more difficult (respondents are more likely to rate having greater difficulties with tasks) are located at the bottom. Each ‘#’ represents 12 respondents and each ‘.’ represents 1–11 respondents. M = mean, S = 1 standard deviation, T = 2 standard deviations. The scale is in logits.