An update report on the Cantonese version of Comprehensive Aphasia Test

Abstract

<p><strong>Introduction</strong></p><p>The Comprehensive Aphasia Test (CAT) contains 34 subtests under 3 sections: Cognitive Screen, Language Battery, and Disability Questionnaire. With the absence of a comprehensive standardized assessment tool for people with aphasia (PWA) in Hong Kong, CAT was adapted to a Cant-CAT with control on Cantonese-specific psycholinguistic variables (Kong & Ng, 2022); the adaptation involved modification of content and scoring criteria to ensure its applicability in Hong Kong’s linguistic and cultural context (Kong, Wong, & Hui, 2022). Due to the relatively small size of Cant-CAT’s normative sample, and the uneven distribution of participants in terms of sociodemographic variables (Kong, Hui, & Wong, 2022), there is a need to further expand on the sample. In the recent second edition of CAT (Swinburn, Porter, & Howard, 2023), a new Aphasia Impact Questionnaire (AIQ) forms the final section, replacing the Disability Questionnaire.</p><p><br></p><p>This study aimed to expand on the normative samples of Cant-CAT, with establishment of updated means and cut-off scores for each subtest. We also evaluated Cant-CAT’s concurrent validity, intra- and inter-rater reliability, particularly for the newly translated Cantonese AIQ. Furthermore, an electronic version of the AIQ (e-AIQ) was developed and its agreement between the paper-based version was compared. Finally, the effects of age, gender, and education on participants’ performance on Cant-CAT was investigated.</p><p><br></p><p><strong>Methods</strong></p><p>A forward-backward translation of the questions in AIQ was carried out, followed by modification of wordings in Chinese with careful considerations of cultural appropriateness. The translated AIQ was then included in the new edition of Cant-CAT (i.e., Cant-CAT-2).</p><p><br></p><p>Fifty-five PWA and 146 unimpaired controls were administered the Cant-CAT. Twenty-three PWA received testing of AIQ and e-AIQ in two separate sessions two weeks apart. Cut-off scores for each subtest in Cant-CAT were determined at the lowest 5th percentile based on the performance of unimpaired participants. All PWA also completed assessments of the Hong Kong version of Oxford Cognitive Screen (HK-OCS; Kong et al., 2016), Cantonese version of Western Aphasia Battery (CAB; Yiu, 1992), and Cantonese FACS (Lau, 2001).</p><p><br></p><p>To determine concurrent validity, PWA participants’ scores from most subtests in Cant-CAT were compared against those assessing similar domains in HK-OCS, CAB, and Cantonese FACS using Spearman’s rank correlations. A MANOVA with post-hoc Bonferroni corrections was adopted to investigate the effects of age, gender, and education on healthy participants’ performance.</p><p><br></p><p><strong>Results</strong></p><p>The cut-off scores obtained for Cant-CAT-2 subtests were all close to ceiling. The updated cut-off scores, based on a more balanced normative sample compared to Kong Hui, and Wong (2022), were more representative. Our results revealed good psychometric properties for Cant-CAT-2, with high concurrent validity, intra- and inter-rater reliability. A strong agreement was also found between the AIQ and e-AIQ. Significant multivariate effects were found for education and age on subtest scores of healthy participants; a similar gender effect was absent.</p><p><br></p><p><strong>Conclusions </strong></p><p>The current findings suggested the Cant-CAT-2 to be a promising clinical assessment tool for Cantonese-speaking PWA. Further research on virtual administration of this battery is underway to explore the feasibility of its application in telemedicine.</p>