An augmented reality approach for ADL support in Alzheimer’s disease: a crossover trial

Background Dementia of the Alzheimer’s type can impair the performance of activities of daily living and therefore severely impact independent living. Assistive technologies can support such patients when carrying out daily tasks. Methods In this crossover study, we used an augmented reality approach using a Microsoft HoloLens to support patients in a tea making task. During task execution, subjects received three-dimensional dynamic holograms of the sub-steps necessary to complete the task. Ten patients suffering from Alzheimer’s disease were tested and post-hoc semi-structured interviews were conducted to assess usability. Results The patients committed errors when executing the task with and without holographic assistance. No differences in success rates or error frequencies were observed (p success = .250, p errors = .887). Patients revealed prolonged trial durations (Glass’ Δ = 1.475) when wearing the augmented reality headset. A model of multiple linear regression (R 2 adjusted = .958) revealed an influence of the errors in the control condition and a moderation by the errors in the experimental condition. Patients with more severe problems in the natural performance of the task showed lower increases in trial durations when wearing the HoloLens. Conclusions We assume that the application was a secondary task requesting its own resources and impairing performance on its own. The regression suggests however that the given assistance was compensating these additional costs in patients with stronger needs of support. Interview data on usability revealed an overall positive feedback towards the application although the hardware was considered uncomfortable and too large. We conclude that the approach proved feasible and the acceptability was overall high, although advances in hardware and the patient-interface are necessary to assist patients suffering from Alzheimer’s disease in daily activities. Trial registration DRKS, DRKS00014870. Registered 11 June 2018 - Retrospectively registered, TrialID = DRKS00014870 .