Effects of the WeCareAdvisor Web App on Behavioral Symptoms and Caregiver Outcomes: A Randomized Trial.

BACKGROUND: Nonpharmacological strategies are first-line treatments for behavioral symptoms, the most distressful aspect of dementia care. Few caregivers have knowledge of or access to effective strategies. We tested the effects of WeCareAdvisor, a web app to help caregivers problem-solve and identify nonpharmacological strategies to manage behaviors. PARTICIPANTS/SETTING: Two hundred and sixty-two community-living caregivers of people with dementia. METHODS: Prospective two-group randomized trial comparing immediate app users (N = 131) to 3-month delayed app users (controls, N = 131). Interviewers masked to participant allocation conducted baseline, 1-(check-in), and 3-month (main endpoint) interviews. Using the DICE approach, immediate users answered brief questions to describe targeted behaviors and investigate contributors from which the app creates "Prescriptions" detailing strategies that caregivers evaluate if effective. Educational materials were also available. Three primary outcomes (caregiver distress, confidence managing behaviors, behavioral frequency by severity) and seven pre-specified secondary (number, frequency, severity of behaviors, functional dependence and caregiver-associated upset, wellbeing, and communications) outcomes were examined. RESULTS: Caregivers were 63.8 (SD = 10.6) years old, mostly female (82%), spouses (65%), with some college (93%); 83% were White, 10% Black, 3% Asian, or other/multi-racial group (3.9%). Most (87%) lived with people with dementia who averaged 75.1 (SD = 10.5) years, were primarily male (57%) and White (82%), with moderate-severe dementia (62%) and averaged 6.4 (SD = 2.8) behaviors. There were no treatment differences between groups in behavioral symptoms, or caregiver distress and confidence at 3 months. However, the difference in caregiver distress between immediate versus delayed user groups was apparent over time by 3 months (Bonferroni corrected p for interaction between time and intervention = 0.012). In secondary analyses, by 3 months, immediate users versus controls, reported fewer behaviors (p = 0.02), less functional dependence (p = 0.05) and associated caregiver upset (p = 0.005), and improved caregiver wellbeing (p < 0.001). CONCLUSIONS: WeCareAdvisor use was associated with significant and clinically meaningful improvements in caregiver distress and wellbeing, although differences in frequency/severity of behaviors were not significant. TRIAL REGISTRATION: NCT05012410; https://clinicaltrials.gov/study/NCT05012410. Initial registration date: 08/12/2021.