ABSTRACT
Background
Standardised aphasia assessment tools may not always be available in a variety of languages, posing challenges for speech and language therapists to adequately assess and diagnose aphasia in speakers of those languages. In 2013, Working Group 2 (WG2) Aphasia Assessment & Outcomes, part of the Collaboration of Aphasia Trialists network, was formed with the purpose of developing reliable and valid aphasia assessment tools and their cross-linguistic adaptations. Over the past decade, WG2 has undertaken important adaptation projects, including the cross-linguistic adaptation of the Comprehensive Aphasia Test (CAT; Swinburn et al., 2004).
Aims
This review aims to achieve three objectives: (a) describe the adaptation procedure of the CAT within WG2, (b) summarise common guidelines and recommendations for future adaptations, and (c) provide concrete solutions for specific cross-linguistic and cross-cultural challenges encountered during the adaptation and validation procedures of the CAT.
Methods
Between 2013 and 2023, WG2 employed a committee approach and fully adapted the CAT into Catalan, Croatian, Dutch, French, Hungarian, Norwegian, Spanish, and Turkish. Further adaptations are in progress for Arabic (Moroccan), Basque, Cantonese Chinese, German, Greek, Icelandic, Lithuanian, Serbian, Slovenian and Swedish. The review comprehensively addresses the linguistic/cultural adaptation and validation procedure for the three components of the battery: the Cognitive screening, the Language battery and the Aphasia Impact Questionnaire. Critical outcomes and some best practice recommendations from psychometric norming and piloting are also discussed.
Outcomes and results
This review builds upon prior work (Fyndanis et al., 2017) and serves as a practical guide for researchers and clinicians undertaking cross-linguistic adaptations of the CAT, with specific conclusions and recommendations drawn from WG2’s adaptations in 19 languages with diverse typological properties. Building on the work exemplified in this paper, future initiatives can direct their efforts towards adapting the CAT for PWA from different linguistic backgrounds for whom validated assessment instruments may be unavailable. This can be achieved through rigorous systematic adaptation procedures for the establishment of comparable language versions of this tool, valuable for various clinical applications. Such endeavours have the potential to provide access to valuable shared datasets for their use across international aphasia trials, and for comparable clinical work within the aphasiology community.
Acknowledgment
We are grateful to the members of Working Group 2 Aphasia Assessment and Outcomes of the Collaboration of Aphasia Trialists (CATs) for their support and contributions towards aphasia test adaptations to several languages in the last 10 years. We also thank the Croatian Science Foundation for allowing the members of Working Group 2: Aphasia Assessment and Outcomes to use some of the pictures from the Croatian database created by the professional illustrator Marko Belić (grant no. HRZZ-2421-UIP-11-2013), as well as Eva Soroli who also created drawings for different language versions of the CAT.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Notes
1. The Danish adaptation of the CAT (Swinburn et al., Citation2014) has been independently completed by two SLTs and is not necessarily comparable to the adaptations reported here. This paper version (with manual) includes new illustrations for all items, however, the imageability data used were taken from English and the test was normed only with 20 healthy adults. There is also a version in Japanese by Watamori and colleagues close to completion. As in the case of the Danish CAT, this adaptation does not adhere strictly to the methodology described here.
2. Construct validity reflecting lexical processing refers only to the language part of the CAT.
3. A gap between the high and low imageability ratings is necessary. For instructions on how to conduct an imageability study, see Paivio et al. (Citation1968) and Toglia and Battig (Citation1978). See Rofes et al. (Citation2018) for further details and data on imageability ratings in published and ongoing CAT adaptations.
4. A gap between low and high frequency ratings is necessary. For reference, the range in the original English version was set at 2–15 occurrences per million for low-frequency items and 50–150 per million for high-frequency items in naming subtests, and above 100 per million for high frequency items in repetition subtests (Swinburn et al., Citation2005).
5. Predicting Language Outcome and Recovery After Stroke. See https://www.ucl.ac.uk/ploras/project-information-versions/project-info-standard
6. For the solution implemented in Spanish, see Martinez-Ferreiro et al. (Citation2024).
7. More about the main concepts produced by PWA and HS from the Spoken picture description can be found in Kuvač Kraljević et al. (Citation2023).
8. Please note that in reliability studies normative sample sizes are often recommended to be larger, such as 400 individuals (Karakaya & Alparslan, Citation2022). Here we follow recommendations for adapting tests; see Ivanova and Hallowell (Citation2013).
9. See also Python et al. (Citation2023) for a comparison between first versus second language users’ performance on the French version of the CAT and the importance to consider second language issues during the standardisation process to better distinguish between stroke-induced language deficits and second language knowledge limitations, and thus to avoid misdiagnosis.