Counts describe this library, not the strength of evidence.
Cómo interpretar la evidencia
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.
In a pilot randomised trial of 99 cognitively normal adults aged 60-80, 6 months of cycling raised fitness more in the high-intensity group than in the moderate-intensity and control groups, but there was no direct effect of exercise on cognition. As a pilot, it was not powered to detect small cognitive effects or explain individual differences.
Among 3281 UK adults aged 50 and over, adjusted depression scores (PHQ-9) in 2020 were 3.23 (95% CI 3.01-3.44) in lonely people versus 1.22 (95% CI 1.12-1.32) in those not lonely, whose score did not change. This is observational and self-reported, and the lonely and less active groups may differ in ways that were not measured.
In a web survey of 1,396 Chinese college students during COVID-19 quarantine, 52.3% had inadequate physical activity, and higher activity went with lower anxiety and depression scores (high activity and anxiety β = -0.121, P < 0.001). A single-time self-reported survey cannot show whether activity lowered symptoms or low mood reduced activity.
In a meta-analysis of 19 papers covering 841 older adults, mean anxiety scores fell from 38.7 ± 5.6 before exercise programs to 33.7 ± 3.4 afterward. The comparison is before versus after in the intervention group, not against a control group, so it cannot isolate the effect of exercise.
Of 8 meta-analyses of randomized trials covering 134 studies, 6 found that exercise significantly reduced depressive symptoms in the general population and 2 found non-significant effects (effect sizes -0.10 to -0.81). None actually measured onset of depression, and heterogeneity was high, so the evidence is about symptoms, not prevention.
Among 3,971 adult twins surveyed in late March and early April 2020, 42% of pairs reported less physical activity, and a perceived decrease was associated with higher stress and anxiety. The stress link was confounded by genetics and shared environment, and the anxiety link was no longer significant after adjusting for age and sex.
In a pilot trial of 33 older adults with cognitive frailty, 91% (30/33) were retained and the smartphone programme raised walking time, step count and brisk walking time, with significant gains only in the intervention group. With 16 and 17 people and within-group tests only, these results are preliminary and need a larger, assessor-blinded trial.
In 61 adults aged 65 and over, those doing open-skilled activity (like tennis) scored better on inhibition and cognitive flexibility, while those doing closed-skilled activity (like swimming) scored better on selective attention and visuospatial function, compared with a no-activity group. With only 61 volunteers tested once, this shows association, not that either sport improves thinking.
Among adults aged 60 and over in a European ageing survey, being socially active at baseline was related to better cognitive function 4 years later, partly through better mental health and more physical activity 2 years later. The abstract gives no effect sizes or sample size, and this observational pathway model cannot prove that social activity causes the benefit.
Pooling 31 articles describing 30 school-related interventions, activity programmes improved resilience (Hedges' g = 0.748), well-being (0.877), positive mental health (0.405) and anxiety (0.347). Heterogeneity was moderate to high (I² = 59-98%) and publication bias was likely, so the true effect for any one school is uncertain.
Among 50,054 Norwegian students aged 18-35, less exercise went with more mental health problems in a dose-response pattern, strongest for exercise frequency. Women with low activity had near three-fold odds of high distress or depression, but the survey is cross-sectional, so direction of cause is unknown.
In 91 older people with dementia (69 analyzed), a 24-week walking and leg-strength program raised gait speed by about 0.05 m/s after the high-intensity phase (Cohen's d = 0.41), but the gain faded at follow-up and cognition did not improve. Many participants dropped out and attendance averaged about 60%.
Four meta-analyses covering 30 studies and 2,110 participants aged 5-20 found a medium overall effect of exercise on depression (d = -0.50) and a small to medium effect in clinically depressed samples (d = -0.48). The evidence is thin for young people, and the abstract gives no confidence intervals or quality ratings.
In 223 Chinese colorectal cancer survivors, a 12-month dietary intervention improved general quality of life (SF-6D mean difference 0.042) and lowered depression (P = 0.015) at 12 and 24 months. Exercise alone helped only at 6 months, and the abstract gives no results for the combined group.
Journal of cancer survivorship : research and practice
In 937 children followed from age 10-11 for 6 years, much of the link between more years of physical activity and later mental health was statistically accounted for by feelings of autonomy (71%), relatedness (51%) and competence (27%), and by self-reported moderate-to-vigorous activity (27-31%). Mediation estimates from observational data do not prove these feelings are the cause, and activity was self-reported.
The international journal of behavioral nutrition and physical activity
In 928 English 12-13 year olds followed for three years, accelerometer-measured physical activity showed no evidence of an association with mental wellbeing or overall symptoms of mental health disorder at age 15-16. Higher activity volume was linked to lower scores on the emotional problems subscale, but only about 72-73% had complete follow-up data and activity was measured only once.
The international journal of behavioral nutrition and physical activity
Pooling 13 randomised trials, mind-body exercise (tai chi, health qigong, yoga) lowered anxiety (SMD -0.76) and depression (SMD -0.86) scores in people with chronic obstructive pulmonary disease. Heterogeneity was moderate for anxiety (I2 = 47.4%) and high for depression (I2 = 71.4%), and the authors themselves say more robust trials are needed.
International journal of environmental research and public health
In a pilot trial of 59 older adults with mild cognitive impairment, 8 weeks of exercise using simulated daily tasks improved memory (p = 0.009), functional status (p = 0.005) and caregiver burden (p = 0.037) more than some of the comparison groups. With only 59 people split across four groups and no follow-up beyond the end of the programme, the results are preliminary.