JMIR Publications

ABSTRACT

Background:

The increasing number of older people presents challenges for healthcare systems, particularly in nursing homes, where care at night is ensured by limited staff. In this context, artificial intelligence (AI)-based remote monitoring technologies could support care staff by improving responsiveness and resident safety. However, their acceptability and impact on workload remain underexplored.

Objective:

This pilot study aimed to evaluate (1) the acceptability of an AI-based remote monitoring system among night care staff in nursing homes, and (2) its effect on the perceived workload of healthcare professionals.

Methods:

We conducted a multimethod study in four French nursing homes over an 8-month period. The technology called the “augmented ear”, detects abnormal sounds and sends real-time alerts to staff smartphones. Acceptability was assessed using the extended Unified Theory of Acceptance and Use of Technology (UTAUT) at three time points (pre-use, mid-use, post-use), and workload was measured via the NASA Task Load Index (NASA-TLX). Semi-structured group interviews were conducted to enrich quantitative findings.

Results:

Twenty-three night care staff participated. UTAUT scores showed correct initial acceptability, which remained stable, particularly regarding performance expectancy and intention to use. However, ethical concerns (e.g., resident privacy, professionals’ surveillance) and technical limitations (e.g., connectivity, alert reliability) emerged. Workload trends suggested a decrease in physical and mental demand over time, but these changes were not statistically significant. Qualitative data revealed both perceived benefits, like improved fall detection and faster response, as well as burdens, like alert fatigue, decision overload and fear of reduced professional autonomy.

Conclusions:

The findings suggest a generally favorable level of acceptability for the “augmented ear” device by the night care staff. Nevertheless, the study highlights the need for better alert calibration, ethical consideration, and device integration to prevent cognitive overload and ensure alignment with care staff values.