Investigating the Equivalence and Test-Retest Reliability of WAIS-5 Test Results Between Telehealth and In-Person Assessment Modalities

Administering cognitive assessment via telehealth could expand service access, but home-based telehealth assessment is under-explored. We examined the feasibility, acceptability, equivalence and test-retest reliability of the Wechsler Adult Intelligence Scale – 5th Edition (WAIS-5 A&NZ ) in this format. Excluding unusable cases, 98 English-speaking participants (47 male, 51 female, M age = 47.1 years) across all age bands completed each subtest via videoconference. A matched in-person comparison group was drawn from the WAIS-5 A&NZ normative sample. Attrition and material losses were low, and post-assessment surveys indicated telehealth acceptance. Differences in group performance were within equivalence margins for full-scale IQ and all core indices except Verbal Comprehension, and all but two of the subtests for which we could not infer equivalence were related to this index score. However, since they were slower in these untimed subtests, telehealth participants may have simply given more considered responses due to feeling less rushed in our extended sessions. Test-retest reliability was similar at the full-scale IQ (ICC telehealth = .81, ICC in-person = .84), index (Range telehealth = [0.62, 0.93], Range in-person = [0.64, 0.89]), and scaled subtest levels (Range telehealth = [0.40, 0.93], Range in-person = [0.40, 0.86]). Our sample included only cognitively healthy volunteers in Australia and New Zealand, which may limit generalisability somewhat. Nevertheless, these results offer promising support for home-based telehealth assessment, as well as guidance for future research.