Developing and Testing a Brief Mindfulness Just-in-time Adaptive Intervention for Reducing Stress of Dementia Caregivers

Objectives: To determine an effective JITAI decision rule to deliver a brief mindfulness intervention depending the current state of stress of caregivers and to examine if the adaptive model of machine learning algorithm generate higher receptivity of the brief intervention than the static and control model at the end of the four weeks. Hypothesis: The delivery of a prompt to engage the brief intervention will reduce the likelihood of being stressed in the subsequent two hours comparing with no prompt and the effect will be stronger if the prompt is delivered when the caregiver is stressed. The static and adaptive models of machine learning algorithm have significant higher receptivity than the control model and the receptivity to interventions delivered by the adaptive model will be higher than that by the static one at the end of four weeks. Design and subjects: Micro-randomization trial including 195 family dementia caregivers with mild to moderate stress level. Study instruments: Wireless sensor system, ecological momentary assessment (EMA) of stress and mood, seven validated scales used to measure perceived stress, depressive symptoms, caregiving burden, positive aspect of caregiving, sleeping quality, quality of life and mindfulness awareness. Main outcome measures: stress measured by sensor, EMA and self-reported scale. Data analysis: Linear mixed models will be adopted to compare the outcomes across different time points and groups. Expected results: This study has the potential to identify a just-in-time adapative intervention to reduce stress level of family dementia caregivers.

Funding Source: HKSAR Govt Related Organizations - Health and Medical Research Fund (HMRF)