Comparing the Usability and Acceptability of Wearable Sensors Among Older Irish Adults in a Real-World Context: Observational Study

Background: Wearable devices are valuable assessment tools for patient outcomes in contexts such as clinical trials. To besuccessfully deployed, however, participants must be willing to wear them. Another concern is that usability studies are rarelypublished, often fail to test devices beyond 24 hours, and need to be repeated frequently to ensure that contemporary devices areassessed.Objective: This study aimed to compare multiple wearable sensors in a real-world context to establish their usability within anolder adult (>50 years) population.Methods: Eight older adults wore seven devices for a minimum of 1 week each: Actigraph GT9x, Actibelt, Actiwatch, Biovotion,Hexoskin, Mc10 Biostamp_RC, and Wavelet. Usability was established through mixed methods using semistructured interviewsand three questionnaires, namely, the Intrinsic Motivation Inventory (IMI), the System Usability Scale (SUS), and an acceptabilityquestionnaire. Quantitative data were reported descriptively and qualitative data were analyzed using deductive content analysis.Data were then integrated using triangulation.Results: Results demonstrated that no device was considered optimal as all scored below average in the SUS (median, IQR;min-max=57.5, 12.5; 47.5-63.8). Hexoskin was the lowest scored device based on the IMI (3.6; 3.4-4.5), while Biovotion, Actibelt,and Mc10 Biostamp_RC achieved the highest median results on the acceptability questionnaire (3.6 on a 6-point Likert scale).Qualitatively, participants were willing to accept less comfort, less device discretion, and high charging burdens if the deviceswere perceived as useful, namely through the provision of feedback for the user. Participants agreed that the purpose of use is akey enabler for long-term compliance. These views were particularly noted by those not currently wearing an activity-trackingdevice. Participants believed that wrist-worn sensors were the most versatile and easy to use, and therefore, the most suitable forlong-term use. In particular, Actiwatch and Wavelet stood out for their comfort. The convergence of quantitative and qualitativedata was demonstrated in the study.Conclusions: Based on the results, the following context-specific recommendations can be made: (1) researchers should considertheir device selection in relation to both individual and environmental factors, and not simply the primary outcome of the researchstudy; (2) if researchers do not wish their participants to have access to feedback from the devices, then a simple, wrist-worndevice that acts as a watch is preferable; (3) if feedback is allowed, then it should be made available to help participants remainengaged; this is likely to apply only to people without cognitive impairments; (4) battery life of 1 week should be considered as a necessary featire to enhance data capture; (5) researchers should consider providing additional information about the purpose of devices to participants to support their continued use.