Researchers propose health coaching guided by wearable forecasts
Lifestyle coaches preferred forecast-informed plans in 71.4% of paired reviews over plans based on wearable history alone.
Paper by Lee SA, Esmaeilpour Z, Zhou M et al.
Our read
AI & data- Studied:
- Preprint: study in people
- Why it’s aging research:
- It builds or tests a tool for aging research.
Editor of record: Boris Djordjevic · AgingResearch.News
Participants in three cohorts—WEAR-ME (2,780), All of Us (22,748) and a longitudinal Fitbit cohort (1,464)—provided data for a modeling study of wearable-based health coaching. The researchers developed Coachable Health Age, a score built from modifiable body measurements and behaviors recorded by consumer wearables. It improved prediction of new cardiovascular and metabolic disease beyond age, sex and body mass index.
They combined the score with a model that forecast several weeks of physiological responses to candidate lifestyle changes. The system generated personalized 12-week coaching plans. In blinded paired reviews, certified lifestyle coaches preferred forecast-informed plans for overall quality and anticipated adherence. Board-certified clinicians judged 99.2% of these plans to meet clinical safety criteria, compared with 97.6% of plans based on wearable history alone.
Why it matters
The work explores how an age-based health score could guide behavior changes aimed at improving cardiovascular and metabolic health.
Caveats
Forecast accuracy and reviewers’ ratings do not show whether following the plans improved health or whether people actually followed them. The work is a preprint and has not been peer-reviewed.
From the paper
- Limitation
First, although SensorWM captures longitudinal temporal dynamics across behavioral and physiological signals, it is trained on observational wearable data rather than randomized interventional trials; consequently …
Quoted word for word from the full text on arXiv.
The paper
Personalized health coaching via coachable health age and counterfactual wearable trajectories
Lee SA, Esmaeilpour Z, Zhou M et al.
arXiv, Nov 5, 2026, Open access, CC BY, Preprint, not yet peer-reviewed
Comments
Be the first to comment.