Supplements

A hybrid factorial SMART-MRT intervention combining human and app-based coaching reduces caregiver stress in family caregivers of people with dementia.

This protocol describes a trial testing a BPSD management intervention with 550 family caregivers randomized to five conditions including human and app-based coaching. The primary outcome is caregiver stress, with secondary outcomes including sleep quality and psychological well-being. The study uses adaptive trial designs and machine learning to personalize just-in-time support.

Last updated: Jul 16, 2026•1 RCTs•📖 Read as article →

Evidence Score

Evidence Score42/100
Human RCT★★☆☆☆
Meta-analysis☆☆☆☆☆
Mechanism★★★★★
Safety★★★★☆
Confidencelow

Study Evidence

Study 1. Machine learning-enabled behavioural and psychological symptoms of dementia management intervention for dementia caregivers: protocol for a hybrid factorial SMART-MRT trial.

observational

Cheung DSK, Kor PPK, Chu AMY, Xie H, Qian M, Chu LW, Zarit S, Chou KL · BMC geriatrics (2026)

Participants: N/A
Duration: 10 weeks (6 weeks initial + 4 weeks re-randomization)
Intervention: BPSD management intervention with combined human- and app-based coaching, human coaching only, app-based coaching only, no coaching, or psychoeducation control; includes telephone-based psychoeducation and app-based support with potential re-randomization to intensified coaching or booster session.
Outcome: Caregiver stress (primary), caregiver burden, psychological well-being, anxiety, depression, sleep quality, health-related quality of life, caregiving self-efficacy, severity of BPSD in care recipients.
Effect Size: N/A
Population: 550 family caregivers of community-dwelling people with dementia.

Result:

PubMed: 42443790DOI: 10.1186/s12877-026-07894-w

Mechanism Graph

Caregivers receive psychoeducation and BPSD management training via telephone
↓
App-based support provides personalized motivational messages and symptom-specific guidance
↓
Machine learning models predict caregiver receptivity and optimize notification timing and content
↓
Result: Reduced caregiver stress and improved secondary outcomes

Limitations

  • ⚠This is a protocol paper; no results are yet available to confirm effectiveness.
  • ⚠The complex adaptive design may introduce implementation challenges and potential confounding in real-world settings.

Frequently Asked Questions

What is the primary outcome of this trial?▼

The primary outcome is caregiver stress.

How many caregivers will be recruited?▼

A total of 550 family caregivers of community-dwelling people with dementia will be recruited.

What adaptive trial designs are used?▼

The trial integrates a five-arm RCT, a two-stage SMART, and three embedded micro-randomized trials (MRTs).

How does machine learning contribute to the intervention?▼

Supervised machine learning models and reinforcement learning (e.g., contextual multi-armed bandits) are used to predict caregiver receptivity and optimize notification timing, frequency, and content.

Products

Affiliate links coming soon. We only recommend products that match the doses and forms used in the cited research.

References

  1. 1.Cheung DSK, Kor PPK, Chu AMY, Xie H, Qian M, Chu LW, Zarit S, Chou KL. "Machine learning-enabled behavioural and psychological symptoms of dementia management intervention for dementia caregivers: protocol for a hybrid factorial SMART-MRT trial.." BMC geriatrics, 2026. PMID: 42443790 DOI: 10.1186/s12877-026-07894-w
Disclaimer: This content is for educational purposes only and is not medical advice. Evidence scores reflect the quality and quantity of available research, not clinical recommendations. Always consult a healthcare professional before starting any supplement or intervention.