Counts describe this library, not the strength of evidence.
How to read this evidence
Intervention trials and step-count associations should be read separately. Exercise can be part of care, but bias and evidence certainty affect estimates. Cognitive test improvements do not by themselves demonstrate dementia prevention; wearable stress is not a psychiatric assessment.
A scoping review of 9 studies (5 observational, 4 interventional) of healthy older adults in South and Southeast Asia found that exercise frequency, regularity and duration were positively linked to mental wellbeing, mainly less depression and better sleep. The review is small, mixes study types, and the authors note that physical activity was rarely measured objectively.
In 256 Koreans aged 65 or older living alone, flexibility exercise was significantly related to depression (p < 0.05), while sitting hours, aerobic activity, walking and muscle-strengthening exercise were not. This is one cross-sectional survey, so it cannot show that stretching protects against depression.
International journal of environmental research and public health
In Framingham Heart Study cohorts of 1861 middle-aged and 909 older adults, as little as 10-21.4 minutes a day of moderate-to-vigorous activity was linked to better executive function (P < .02). The analysis is cross-sectional, so it cannot show that activity improves thinking.
Among 9118 Chinese adults aged 45 and older, 150-299 min/week of moderate activity was linked to lower odds of depression in women (OR 0.49, 95% CI 0.28-0.87), while frequent or long vigorous activity was linked to higher odds in men. The data are from one 2015 survey, so we cannot tell whether activity changes depression or depressed people change how much they do.
Across 14 observational and 5 intervention studies, 12 of the 14 observational studies found cognitive benefits from open skill exercise (like ball or racquet sports), and 3 of the 5 intervention studies favoured it over closed skill exercise. Most evidence is cross-sectional, so it cannot show that open skill exercise causes the better cognition.
Among 370 hospitalized patients aged 75 or older in Spain, 5-7 days of multicomponent exercise improved executive function and MMSE scores more than usual care (for example MMSE 2.10 versus 0.27 points, p < 0.001). This is a secondary analysis of a trial in relatively fit patients, and many could not complete the tests, so benefits may be overstated for sicker patients.
In 124 healthy adults (mean age 61.51), executive functions, processing speed and language were linked to specific gait measures of variability, pace and asymmetry, especially while walking and doing a second task. The design is cross-sectional with no effect sizes in the abstract, so it cannot show that thinking ability determines walking or the reverse.
After 12 weeks of low-intensity aquatic exercise, older adults with depression had lower depression (53%) and anxiety (48%) scores, better balance and flexibility, and changes in oxidative stress markers. Every participant exercised and the abstract describes no non-exercise control group, so the changes cannot be credited to the exercise itself.
In 27 breast cancer survivors, 30 minutes of moderate-intensity treadmill walking shortened processing-speed reaction time compared with 30 minutes of seated rest (interaction P = .03), while spatial working memory effects only trended (P = 0.08 and 0.10). The sample is small and the effect was measured right after one session, so it says nothing about lasting benefit.
Among 800 women in a population-based sample followed from 1968 to 2012, midlife cognitive activity was associated with lower risk of dementia (HR 0.66; 95% CI 0.49-0.89) and Alzheimer disease (HR 0.54; 95% CI 0.36-0.82). Physical activity was linked to lower risk of mixed dementia and dementia with cerebrovascular disease, but this is observational, so it cannot show that activity prevents dementia.
In a randomized trial of 115 people with dementia, 12 weeks of exergame or aerobic training improved psychomotor speed compared with relaxation and flexibility exercises (mean difference 0.326 and 0.370 in domain score), with no effect on executive function, episodic memory or working memory. Only one of four cognitive domains improved, so the benefit is narrow.
Among 639 pregnant women in Brazil, a 16-week supervised exercise program did not significantly lower depressive symptom scores 3 months after birth (mean difference -0.6; 95% CI, -1.3 to 0.1). Only 40.4% attended at least 70% of sessions, so the trial may understate what regular exercise could do.
In a survey of Norwegian adolescents, sports club participation was linked to lower odds of depression symptoms after adjustment (OR 0.57, 95% CI 0.40-0.81), as were other organised activities such as dance and martial arts (OR 0.52, 95% CI 0.31-0.86). This is a cross-sectional self-report survey, so it cannot show that clubs prevent depression.
In a small trial of 50 women having chemotherapy for breast cancer, 12 weeks of home-based moderate walking improved perceived cognitive function versus usual care, but not attention, executive function, memory or visuospatial skills on objective tests. The abstract gives no numbers for the effect, and the sample is small.
Among 1298 sedentary adults aged 70-89 in the LIFE trial, a 24-month moderate-intensity physical activity programme was linked to 21% lower odds of worsening cognitive frailty than health education (OR 0.79, 95% CI 0.64-0.98), regardless of baseline IL-6. The analysis used data from an existing trial, and the upper confidence limit is close to no effect.
In 84 older adults randomised to 8 weeks of physical exercise (PE), virtual reality-based exercise (VRE), brain exercise (BE) or no training, PE gave the biggest gains in the Timed Up and Go and sit-to-stand tests, while BE and VRE gave larger gains on the Montreal Cognitive Assessment. The trial was small and short, with no longer follow-up.
European review of aging and physical activity : official journal of the European Group for Research into Elderly and Physical Activity
In 13 UK studies of exercise referral for people with cardiovascular, mental health and musculoskeletal disorders, schemes of 20+ weeks produced better health outcomes and adherence than schemes of 8-12 weeks, and all schemes raised physical activity. The authors call the review preliminary, and the comparison of scheme lengths is across different studies, not within one trial.
Across 15 randomised trials with 675 patients, aerobic exercise reduced raised anxiety compared with waiting-list controls (effect size -0.41, 95% CI -0.70 to -0.12), and high-intensity programmes worked better than low-intensity ones. The authors say conclusions are limited by the small number of studies and wide variation in how exercise was delivered.