Digital ACT Psychological Support for Primary Caregivers of Older Adults With Severe Mental Illness

This is a single-center, parallel-group randomized controlled trial conducted in Tieli City, Yichun, Heilongjiang Province, China. A total of 64 eligible primary caregivers of older adults (≥60 years old) with severe mental disorders (SMD) will be enrolled and randomly assigned at a 1:1 ratio, with 32 participants in the intervention group and 32 participants in the control group. The sample-size calculation was performed using G*Power 3.1.9.7 for repeated-measures ANOVA (group-by-time interaction): effect size f = 0.25, α = 0.05, power (1-β) = 0.95, 2 groups, 3 measurement time-points. The calculated minimum analyzable sample size is 44 participants. A 30 % anticipated attrition rate was added, leading to the final recruitment target of 64 participants.

All participants in both groups will receive identical routine psychological support, including regular disease-related and care-related knowledge education and basic empathetic listening provided via telephone and WeChat three times per week.

Participants allocated to the intervention group will additionally receive an 8-week digital psychological support program based on Acceptance and Commitment Therapy (ACT), delivered through a lightweight WeChat mini-program. The ACT training module is the core component and consists of eight sequentially unlocked weekly sessions: an ACT introduction session using a choice-point tool to identify experiential avoidance, followed by acceptance, cognitive defusion, contacting the present moment, observing self, values clarification, and committed action, and a final session of review and consolidation. Each session combines short video or illustrated texts, scenario-based comprehension items and a guided self-reflection exercise. Scenario items are built on real caregiving situations and use neutral response options without negative feedback, in line with the accepting stance of ACT. The program delivers a pre-planned daily schedule consisting of six learning days and one weekly review day: each learning day requires approximately 7-8 minutes (a 2-minute optional daily emotion check-in plus 5-6 minutes of core learning content), and the weekly review day requires approximately 10 minutes of emotion-data review, knowledge consolidation and practice. Three supplementary modules can be used on demand without a fixed sequence: an emotion check-in module (daily emotion record, personal emotion calendar and structured emotional review), a caregiving skill workshop (management of acute relapse, medication refusal and hoarding, self-harm and suicide risk, aggressive behavior and emergency referral, routine home care skills, SMD knowledge education and scenario-based practice), and a resource navigation module (national, provincial, municipal and county-level mental health policies, local designated psychiatric facilities, welfare subsidies and medical insurance reimbursement procedures). Local grassroots staff will provide simple operational guidance and adherence reminders by phone or WeChat, without interpreting ACT professional content.

Participants in the control group will maintain the same daily contact frequency and overall time burden as the intervention group. General mental-health and care-related popular-science reading materials without any ACT-related or other psychological-regulation training content will be delivered via WeChat every day over the 8-week period, with reading duration (approximately 7-8 minutes per day) and total reading load matched to the daily learning volume of the intervention group. Participants only need to read the materials, with no homework, reflection tasks or interactive psychological exercises. Both groups receive no additional psychological treatment during the study period.

Outcome assessments will be performed at three time-points: baseline (T0, before randomization), immediately after completion of the 8-week intervention (T1), and one-month follow-up after intervention termination (T2). Questionnaires can be completed online via QR-code or using identical paper versions for participants with limited digital operation ability. The primary outcome is psychological flexibility measured by the Chinese version of the Comprehensive assessment of Acceptance and Commitment Therapy processes (CompACT), a 23-item instrument covering the dimensions of openness to experience, awareness and engaged action, scored on a 7-point Likert scale from 0 (never) to 6 (always), with higher total scores indicating greater psychological flexibility. Secondary outcomes include: family functioning measured by the Chinese revised Family Assessment Device (FAD), a 30-item scale covering affective involvement, positive communication, self-absorption, problem solving and family rules, scored on a 4-point Likert scale (total range 30-120, higher scores indicating better family functioning); psychological distress measured by the 12-item General Health Questionnaire (GHQ-12, Zhang Mingyuan Chinese version), a 12-item self-report screening instrument scored continuously; caregiver needs and resources measured by the Chinese version of the Caregiver Needs and Resources Assessment (CNRA), a 36-item instrument with two dimensions (needs and resources) across 12 domains, two reverse-scored items, scored on a 5-point Likert scale; disease- and care-related knowledge measured by an instrument adapted from the schizophrenia knowledge-attitude-practice awareness questionnaire, covering disease attribution, risk factors, treatment and social rehabilitation across 30 items and reworded for caregivers of older adults with SMD; and perceived social support measured by the Chinese version of the Perceived Social Support Scale (PSSS), a 12-item instrument covering family, friend and other support scored on a 7-point Likert scale. Objective system-usage logs (login frequency, module completion, learning-day adherence) and a self-reported satisfaction questionnaire will be collected to evaluate intervention feasibility and acceptability.

All adverse events and unanticipated problems reported by participants or identified through routine contact will be recorded and reported to the ethics committee. Participants are free to withdraw at any time without affecting their routine care.

After finishing all research assessments, participants in the control group will be granted full access to the digital ACT psychological support mini-program as a benefit for study participation.

Statistical analysis will follow modified intention-to-treat (mITT) and per-protocol (PP) principles. Generalized linear mixed models will be applied to examine group-by-time interaction effects for outcome indicators, with baseline values and covariates adjusted as appropriate. Missing data will be handled by multiple imputation, with sensitivity analysis conducted.