In a recent study published in JAMA Network Openresearchers examined the relationship between physical activity and cognitive decline by examining domain-specific and global cognition and examining dose-response associations and potential moderators.
Record
Physical exercise is associated with a reduced risk of cognitive decline. However, most evidence comes from short-term studies, possibly due to reverse causality bias. Research shows that physical activity improves cognitive outcomes and reduces the risk of dementia, but the evidence is inconsistent.
Recent interventional studies caution against linking cognitive function to physical activity, and most existing evidence comes from observational studies with short follow-up and no prior cognitive levels. Modeling knowledge as a continuous variable would improve statistical power.
About the study
In the present meta-analysis, researchers investigated the association between physical activity and cognitive decline, assessing the impact of factors such as length of follow-up, baseline age, amount of exercise, and quality of studies included in the association.
The team searched PubMed, Scopus, PsycINFO, Web of Science, CINAHL, and SPORTDiscus databases up to November 2, 2022 for relevant full-text records accessible in English. The cohort included prospective observational cohort or case-control studies, including subjects aged ≥20 years, follow-up of one year or more, and estimates of associations between physical activity and cognition for the meta-analysis. The included studies reported validated measures of cognitive outcome in later life (mean and maximum age 55 and 65 years, respectively).
The group included studies that assessed physical activity using devices, questionnaires or interviews. The primary outcome of the study was the association between baseline physical activity and global cognition or specific cognitive domains (such as episodic memory, executive function, verbal ability, verbal naming and fluency, processing speed, visuospatial ability, and working memory) at follow-up.
The team excluded studies that retrospectively recorded levels of physical activity, levels of cardiorespiratory fitness, periods of physical activity, physical activity extending beyond follow-up, and statistical reallocations for physical activity. They also excluded studies that included subjective measures of cognition and those that assessed cognitive function using disability level registries. They also excluded cohorts with dementia, specific diseases, or cognitive impairment at baseline.
Two reviewers independently screened titles, abstracts and full texts of identified records, assessed quality of evidence and extracted data, and a third reviewer resolved disagreements in the event of a lack of consensus. The team performed random-effects modeling and used logistic regressions to determine risk ratios (RRs) and regression coefficients using moderators, scatterplots, and funnel plots for physical activity. They examined pooled estimates of the relationships of physical activity with global and specific cognitive domains. They analyzed data between January and August 2023, with final analyzes in December of that year.
Results
Initially, the team identified 18,669 records and excluded 17,861 while screening title abstracts and 703 while screening full text. As a result, 104 records were analyzed, including 341,471 individuals. The binary outcome analysis included 45 records and 102,452 subjects. the subsequent cognitive analysis included 14 records and 41,045 subjects. and the analysis of global cognitive changes included 25 records and 67,463 subjects.
Among 45 studies with binary outcomes, one study, 13 studies and 31 studies were of high, medium and low quality, respectively. Among 14 studies evaluating global knowledge tracking, no study, four studies and ten were of equal quality, respectively. Among the 25 studies that assessed global cognitive changes, none, five studies and 20 were of high, medium and low quality, respectively. The included studies for all outcome measures were highly heterogeneous, with I2 values of 70% for binary outcomes, 76% for subsequent cognition, and 67% for changes in global cognition.
Physical exercise was associated with reduced cognitive impairment or impairment after funnel plot corrections (pooled RR, 0.97). However, there were no statistically significant associations in follow-up analyzes over ten years. Physical exercise was associated with global cognition monitoring domains and their changes with standardized regression coefficients of 0.030 and 0.010, respectively, obtained by trim-and-fill assessment, without measure or dose-response associations by participant age, duration of follow-up, adjustments cognitive baseline or study quality.
Specifically, verbal fluency and episodic memory were related to physical activity with standardized regression coefficients of 0.05 and 0.03, respectively. Amount of physical exercise had a higher inverse association with cognitive impairment or decline up to 5,000 metabolic equivalent minutes of work each week (ie, moderate to vigorous physical activity for 16 hours each week).
conclusion
Overall, the study findings showed a weak but positive association between physical activity and cognitive decline that persisted regardless of cognitive level or age of the cohort and is crucial for population health to delay multifactorial diseases that cause dementia.
Follow-up duration, pace, type of exercise measurement, and quality influenced the association reported in studies with binary cognitive outcomes. However, funnel plots detected potential bias. The study also revealed weak correlations between verbal fluency and episodic memory, with mixed results for executive function between follow-up and change analyses.
