Tailored Online Physical Activity Coaching for Middle-Aged and Older Adults With Cognitive and Mental Health Concerns: Single-Arm Pre-Post Intervention Study.

Background: Low levels of physical activity (PA) increase dementia risk, and for middle-aged and older adults with co-occurring cognitive concerns and mental health symptoms, dementia risk increases further. Despite clinical trials showing lower adherence to PA interventions in high-risk groups, there is a sparsity of interventions tailored to support unique behavior change needs. In EXCEL (Exercise for Cognitive Health) phase 1, we developed a model to understand the needs of this population and identified tailoring requirements to enhance engagement. Here we report the findings of a pilot online intervention designed to support middle-aged and older adults with subjective cognitive decline or mild cognitive impairment and mild to moderate symptoms of depression or anxiety to meet PA guidelines. Objective: We aimed to measure (1) efficacy of the intervention in promoting adoption of PA in line with national guidelines, (2) acceptability, feasibility, and safety of the intervention, (3) changes to dementia risk, PA levels, mental health symptoms, and stages of change, and (4) changes in potential cognitive mechanisms of change. Methods: A pilot individual 12-week online home-based PA intervention. Participants aged 45-80 years, experiencing both cognitive and mild-to-moderate anxiety and depression symptoms, were prescribed individually tailored PA programs combining aerobic and strength PA, plus balance training as indicated, with fortnightly online coaching. Results: A total of 55 participants were enrolled (46 females/9 males; mean age 62.2, SD 7.6 y). The intervention was effective; at baseline, only 3 of 55 (6%) participants met all applicable PA guidelines, compared to 24 of 55 (44%) participants post intervention (P<.001). Participants were at least 8 times more likely to meet the age-appropriate guidelines for aerobic, strength, and balance activities after the intervention. Retention rates were high (95% completion rate), and feedback indicated 98% found the program useful. Safety was successfully monitored by a clinical panel via email, using a system of alerts. Dementia risk was reduced (d=-0.32, P=.008), and reductions in depression, anxiety, and stress scores were large and clinically meaningful (d=-1.31, d=-0.89, and d=-1.18, respectively; all P<.001). Over half of the cohort (27 participants) transitioned to a higher stage of change at postintervention. Improvements in action planning (d=0.66, P<.001) and positive outcome expectancies (d=0.33, P=.01) indicate potential cognitive mechanisms of change. Conclusions: Our findings showed that this tailored, home-based online intervention successfully supports an at-risk cohort of aging adults to adopt PA, in line with national guidelines. The intervention was acceptable, feasible, and safe and led to significant reductions in dementia risk and mental health symptoms and was associated with progression through stage of change. Reflecting the Capability, Opportunity, and Motivation Behavior model underlying the intervention design, improvements in action planning and outcome expectancy appeared to play a role in bridging the PA intention-behavior gap.