Comparability and usability of Paired Associates Learning task (PAL) delivered on a smartphone. (20th December 2022)

• Record Type: Journal Article
• Authors: Baker, Elizabeth; Thorp, Emily; Anwyl‐Irvine, Alexander L; Cashdollar, Nathan; Zavitz, Kenton; Taptiklis, Nick; Cormack, Francesca K
• Is Part Of: Alzheimer's & dementia. Volume 18(2022)Supplement 7
• Page Start: n/a
• Page End: n/a
• Publication Date: 2022-12-20
• Subjects: Alzheimer's disease -- Periodicals; Alzheimer Disease -- Periodicals; Dementia -- Periodicals; Démence; Maladie d'Alzheimer; Périodique électronique (Descripteur de forme); Ressource Internet (Descripteur de forme); 616.83
• DOI: 10.1002/alz.065366
• Languages: English
• ISSNs: 1552-5260

Abstract

Abstract: Background: Delivering episodic memory tests on a mobile device can support increased accessibility across diverse health settings. This is particularly important for increasing access to memory screening tools for participants at risk of memory impairment including Alzheimer's and dementia. CANTAB PAL task is a short (8‐12 minutes) task of memory performance that is currently validated for delivery on a tablet, laptop or desktop device. Cambridge Cognition developed a prototype version of PAL that can be delivered on a smartphone (iPhone or Android). A key component of making the task available on smartphone is to understand the comparability to the original version of the task. For tasks delivered direct to patient, usability becomes even more important. This study was designed to evaluate the comparability and usability of PAL on smartphone device. Method: Participants (N = 98, mean age 55 ± 5.59 years, gender balanced (50% (49)) were recruited through the Prolific platform to remotely complete both smartphone and desktop versions of CANTAB PAL on two consecutive days followed by a usability questionnaire. A cross‐over design was used with 50:50 participants smartphone first: desktop first to counterbalance for exposure to the task. Participants used their own devices for assessment and the device operating system was captured. Result: Most participants used Android smartphones (74.1% (69)) and Windows desktops (90% (81)). Smartphone and desktop versions of PAL were shown to be highly correlated (r = 0.61, p < 0.05). Average performance was better on the desktop device; an upwards bias estimated as 3.52 (±1.96 SD 27 to ‐20) adjusted errors. The usability questionnaire found that 76.6% (59) participants strongly agreed they found the instructions clear and easy to follow and 42.9% (33) strongly agreed that they could see the symbols easily on their mobile phone. Most participants preferred using PAL on a desktop device in terms of ease of use (70.1% (54)) and enjoyment 66.2% (51). Conclusion: We have shown good correlation between iPhone and computer versions of the test, but with some important caveats. The usability data collected here will inform improvements to ensure suitability of PAL on a smartphone device for near patient assessments.