Impact of real-time stress monitoring in people with an intellectual disability

People with an intellectual disability are vulnerable to stress, which can result in challenging behaviour, such as apathy, self-harm, or aggression. By monitoring stress in real-time, professional caregivers can timely intervene to prevent escalations and improve the quality of life for both the client and themselves. The aim of this study was to investigate the impact of real-time stress monitoring using the stress-detection system HUME on the quality of life of people with a severe intellectual disability and their professional caregivers. The study comprised two parts. A case series study (n=12) was conducted with long-term care clients with intellectual disabilities to validate the HUME. HUME stress measurements, based on physiological data and trained artificial intelligence models, were collected, and compared with labelled video observations of professional caregivers. A second study was conducted to measure the impact of HUME and the induced interventions on quality of life. Physiology data and quality of life scores were collected. The HUME stress prediction was used 1) for early warning to deploy interventions based on what the professional caregiver deemed best, and 2) as an assessment tool to understand the effectiveness of care interventions. The quality of life for both the client (n=41) and professional caregiver (n=31) was evaluated via a questionnaire. Results showed that the HUME was able to detect stress in all cases, and stressful events detected by the HUME were consistent with the behavioural observations. The real-time stress monitoring using HUME, along with subsequent interventions, was effective. Clients with intellectual disabilities experienced reduced stress and an improvement in their perceived quality of life. Also, professional caregivers perceived an increase in the quality of life during the period the HUME was used. In most of the cases, HUME-based interventions led to a reduction in escalations, fixations, and self-harming behaviour. Further randomized controlled studies are needed to substantiate these results.