Summary: Contrary to some earlier reports, a long-term study led by Johns Hopkins researchers found no association between physical fitness in midlife and cognitive health in later life. The study did find that greater physical activity in later years correlated with modest short-term improvements in cognitive performance.
Source: Johns Hopkins University.
Johns Hopkins Bloomberg School of Public Health researchers followed 646 adults for three decades and report that exercise in midlife was not linked to cognitive fitness or dementia risk in later life.
The study challenges the commonly held expectation that being physically active in midlife will protect against cognitive decline such as dementia or Alzheimer’s disease many years later. Alzheimer’s and other dementias are progressive syndromes characterized by memory loss, disorientation, and loss of daily functioning; with no cure currently available, researchers continue to search for reliable preventive strategies.
Published in the Journal of Alzheimer’s Disease, the research did observe that participants reporting higher levels of physical activity in the years immediately preceding cognitive testing showed better cognitive performance two years later. This finding supports prior short-term studies that link recent exercise to improved cognitive function.
“Physical activity offers a range of health benefits — cardiovascular health, weight management, bone and muscle maintenance — but it is premature to conclude it will prevent Alzheimer’s,” says Alden L. Gross, assistant professor in the Bloomberg School’s Department of Epidemiology. “Long-term prevention remains an open question.”
The investigators designed the study to address limitations of much previous research, which often relies on cross-sectional data or brief follow-up periods. Cross-sectional designs compare different individuals at a single time point and cannot reliably establish causation because of reverse causation: people who are developing dementia may reduce activity as the disease progresses. Longitudinal studies that follow the same individuals over many years are better suited to assess whether midlife behaviors influence late-life outcomes.
Researchers analyzed data from the Johns Hopkins Precursors study, which enrolled medical students between 1948 and 1964 and collected annual health questionnaires over decades. This relatively homogeneous cohort reduced the influence of some confounding factors that can complicate observational research.
In 1978 the median age of participants was 46; by 2008 the median age was 77. Questionnaires administered periodically from 1978 through 2008 asked about exercise, routine physical activity, and physical limitations. From responses the team calculated metabolic equivalents (MET hours per day) to quantify activity levels and recorded whether participants reported exercising to the point of sweating.
Cognitive testing was conducted in 2008, and dementia diagnoses were adjudicated from medical records through 2011. Of 646 participants included in the analysis (598 men, 48 women), 28 individuals — about 4.5 percent — were identified with Alzheimer’s disease during follow-up.
The core result: no measure of midlife physical activity (from 1978 measures) predicted late-life cognitive performance or future dementia risk. In contrast, activity measured near the time of cognitive testing was associated with better cognitive scores two years later. The researchers also examined trajectories of activity across the lifespan and found no consistent links between activity patterns and cognitive outcomes.
Biologically, the idea that exercise could influence dementia risk is plausible. In animal models, physical activity has been associated with reduced accumulation of beta-amyloid plaques, a proteinopathy implicated in Alzheimer’s disease. Exercise also improves cerebral blood flow, which supports cognitive processing and may explain the short-term associations seen in many studies.
Gross notes the implications for future intervention research: “We need to focus on the causes and mechanisms of Alzheimer’s disease because we still do not know which preventive measures, if any, will be effective long-term. This study suggests that cross-sectional associations between activity and cognition can be misleading. Continued long-term follow-up studies are essential to clarify whether and how lifestyle factors influence dementia risk.”
Funding: The research received support from the National Institutes of Health and the National Institute on Aging (grants AG01760, DK02856, DK07732, K01-AG050699, and K01-AG048765).
Source: Barbara Benham, Johns Hopkins University.
Image source: Public domain image used by NeuroscienceNews.
Original research: The study, titled “Physical Activity in Midlife is not Associated with Cognitive Health in Later Life Among Cognitively Normal Older Adults,” was authored by Alden L. Gross, Haidong Lu, Lucy Meoni, Joseph J. Gallo, Jennifer A. Schrack, and A. Richey Sharrett and published in the Journal of Alzheimer’s Disease (online August 14, 2017).
Abstract
Physical Activity in Midlife is not Associated with Cognitive Health in Later Life Among Cognitively Normal Older Adults
Background:
Existing links between physical activity and dementia are often based on cross-sectional studies or studies with limited follow-up, which leaves open concerns about reverse causation.
Objective:
Using three decades of follow-up from a well-established cohort, the study tested whether midlife physical activity predicts late-life cognitive performance and dementia incidence.
Methods:
The Johns Hopkins Precursors study (n = 646) collected physical activity data from 1978 onward, from which metabolic equivalents (MET h/day) and exercise habits were derived. Cognitive assessments were performed in 2008 and dementia adjudication continued through 2011. Analyses used linear regression for cognitive outcomes and Cox proportional hazards models for dementia, adjusting for age, sex, smoking, diabetes, and hypertension.
Results:
No measure of physical activity from 1978 predicted late-life cognition or dementia onset. Activity measures taken near the time of cognitive testing (MET h/day and reports of regular exercise in 2006) were associated with better cognition in 2008, consistent with cross-sectional findings and short-term trial results.
Conclusion:
This 30-year cohort study found that recent physical activity, but not midlife activity, was linked to late-life cognitive performance. The results align with null findings from randomized trials and long-term observational studies, and highlight that cross-sectional associations may reflect reverse causation rather than a true long-term protective effect of midlife exercise on dementia risk.