Exercise 'protects brain from ageing'

Friday December 24 2010

“Regular exercise may be the best way to keep ageing brains sharp,” says The Daily Telegraph. The newspaper has reported that a new review of research has shown that even low to moderate exercise prevents the milder forms of cognitive decline in older age.

The research pooled the results of 15 studies and found that low to moderate exercise, such as playing a round of golf once a week or tennis twice a week, was linked to a 35% reduction in the risk of cognitive decline. The researchers think that this could be due to physical activity increasing blood flow to the brain.

This review of observational studies was well conducted and reported. It includes an analysis of over 30,000 people and it seems likely that the researchers have sourced the most important studies on this topic. Although these were not randomised studies, the consistency and strength of the evidence appears to provide the best current estimate of activity’s ability to prevent normal, age-related brain changes such as the decline of memory.

Where did the story come from?

The study was carried out by researchers from the University of Florence and other institutions in Italy. No external sources of funding are noted. The study was published in the Journal of Internal Medicine.

The Daily Telegraph summarised this meta-analysis accurately, placing the research in context and reporting some of the researchers’ estimates of the size of exercise’s effect.

What kind of research was this?

This was a meta-analysis that aimed to pool the results of prospective cohort studies on physical activity and cognitive decline. The researchers pooled the data from 15 studies that between them included over 30,000 non-demented subjects who had been followed for a period of one to 12 years. Among this population, over 3,000 new cases of cognitive decline occurred.

The meta-analysis showed that individuals who were physically active at the start of the study (baseline) had a significantly reduced risk of developing cognitive decline during follow-up.

The researchers explain that it is already known that physical activity has positive effects on a wide range of health measures, reducing the risk of heart disease and stroke, diabetes, obesity, hypertension and some cancers.

Attention, memory and concentration (known as cognitive functions) typically decline with age,  becoming slower and less efficient, much as physical functions such as walking and balance do. The authors argue that these cognitive changes can become noticable and can cause mild disability, even if a state of dementia is not reached.

The authors have described using a comprehensive procedure for finding relevant research and, importantly, only included studies if patients recruited had received a clinical evaluation at the start of the study and did not suffer from dementia. As participants were followed up over time, the new evidence provided by this meta-analysis supports the role of exercise programmes in preventative medicine, as maintaining activity levels into later life seems to slow the onset of memory loss associated with normal ageing.

What did the research involve?

In this study the researchers searched a number of computer databases including Medline, Embase, Google Scholar, Web of Science and the Cochrane Library. They retrieved and assessed articles published up until January 2010, plus studies cited within these articles. Studies were included only if the association between physical activity and cognitive decline in subjects without dementia was analysed prospectively (ie were prospective cohort studies).

The researchers used and reported best practice systematic review methods, including assessment of studies by two separate people and appraisal and statistical analysis for any publication bias among the studies they found. They excluded studies of other design, such as case control or cross-sectional studies, plus any that included people with dementia at the start.

They adjusted for a range of other factors that could have influenced the result such as age, education, smoking, alcohol, use of NSAID medication, self-rated health and some chronic conditions. They also appropriately used a random effects model, a type of statistical analysis that in part takes into account the statistical differences in studies included.

What were the basic results?

Fifteen publications of 12 prospective cohorts were included in the final analysis, from a total of 58 papers identified by the researchers. These studies included 33, 816 people without dementia who were followed for up to12 years. A total of 3,210 patients (about 9.5%) showed cognitive decline during the follow-up.

The analysis of all the studies showed that subjects who performed a high level of physical activity were significantly more protected (by 38%) against cognitive decline (memory loss etc) during the follow-up, compared to people who reported being sedentary (hazard ratio [HR]) 0.62, 95% confidence interval [CI] 0.54 to 0.70).

The researchers also looked at the effect of exercise of a low to moderate level, and found that this too protected against cognitive impairment compared to being sedentary. It provided a significant protection of 35% (HR 0.65, 95% CI 0.57 to 0.75).

They tested to see if the studies were similar enough to allow them to pool the results in a valid way and found that they could. Technically there was no significant heterogeneity (variation) among the studies (I2 = 17%; P = 0.26) and no publication bias.

How did the researchers interpret the results?

The researchers claim that this is the first meta-analysis to evaluate the role of physical activity on cognitive decline among people without dementia. The results, they say, suggest a “significant and consistent protection for all levels of physical activity against the occurrence of cognitive decline”.


These results have highlighted the important role that even low levels of physical activity can play in protecting people from the decline in mental function that can routinely occur in healthy people as they age.

The importance of the study lies in its application to an ageing population and the study has both strengths and some weaknesses:

  • One clear strength is the size of the study, with a large number of people for whom the researchers had data. This increases confidence in the result.
  • Publication bias was not evident in the studies analysed, which supports the validity of this meta-analysis. Publication bias is the  tendency for those involved in studies to handle the reporting of positive results (those that show a significant finding) differently from results that are negative or inconclusive ones.
  • A limitation to the study was that the methods used to measure cognitive decline and physical activity varied across the included studies. The MMSE test (a recognised cognition test) was the most frequently used tool for diagnosing cognitive decline, but other tests were used in some studies. Though an unavoidable weakness of this study, the researchers tested for the effect and concluded it was not a significant problem.
  • This study did not find a clear ‘dose dependent’ effect, ie an association in which increasing levels of activity resulted in increasing levels of protection.
  • The protective effect appears stronger for women than for men, and it is not clear why.
  • The studies included in the analysis may have had different definitions as to what are moderate and what are high levels of physical activity. Further clarification may be needed to see how much physical activity elderly people should aim to do.

Randomised trials, though possible in the area of physical activity, would need to be large and follow people for a long time to find these sorts of results. The practical constraints of performing such a study suggest that for the time being this well-conducted meta-analysis provides probably the best evidence that this important link exists.

The authors now call for further studies to determine the best “type, frequency and intensity of exercise” or physical activity that maintains memory into old age.

The study is reliable, well conducted and reported. Though the findings may be unsurprising, as some individual studies had already shown significant results, the summary of a large body of evidence adds weight to the science behind the established link between low levels of physical activity and cognitive decline.

Analysis by Bazian
Edited by NHS Choices