Digital mindfulness training linked to lower stress in dementia caregivers over 12 weeks

At baseline, participants reported a median stress rating of 7 out of 10 and mindfulness of 4 out of 10. Using generalized linear mixed models, researchers tracked how weekly app use and self-reported mindfulness and stress evolved throughout the trial. The largest stress reduction occurred between baseline and week 3 (a decrease of 2 points), while mindfulness peaked between baseline and week 4 (an increase of 2.5 points). For every 10 minutes of HMP-C use in the prior week, mindfulness ratings were 0.14 points higher and stress ratings were 0.14 points lower—effects that persisted across all 12 weeks, though the interaction effect was small. The relationship between mindfulness gains and stress reduction was strongest among caregivers who started with the highest baseline stress. A key limitation is that this was a secondary analysis of a feasibility trial with no control arm reported here, so we cannot isolate the app's effect from time or placebo effects. The findings suggest that brief, consistent engagement with digital mindfulness tools may support stress management during the critical early weeks of skill-building.

The data reveal that mindfulness and stress changes were not linear: most improvement happened in weeks 1–4, then plateaued. This is consistent with skill-acquisition curves—early practice drives the steepest gains. Notably, the relationship between mindfulness and stress was strongest among caregivers with high baseline stress (ρ01=0.86), meaning those under the most pressure benefited most from mindfulness gains. The effect size per 10 minutes of use (0.14 points on a 0-10 scale) was small in absolute terms but was consistent across all 12 weeks, suggesting sustained benefit if caregivers keep using the tool. The study lacks a control condition and relies entirely on self-report, so we cannot rule out regression to the mean (high-stress participants naturally improving over time) or placebo. However, the specificity of the dose–response relationship (more use = more benefit) and the differential benefit for high-baseline-stress participants strengthens the plausibility that the app drove real change, not just natural recovery.