By Ellen Feldman, MD

Altru Health System, Grand Forks, ND

Dr. Feldman reports no financial relationships relevant to this field of study.

SYNOPSIS: A systematic review and meta-analysis incorporating more than 500,000 subjects revealed that moderate to vigorous physical activity is associated with a significant reduction in development of Parkinson’s disease; this relationship is most pronounced in men.

SOURCE: Fang X, Han D, Cheng Q, et al. Association of levels of physical activity with risk of Parkinson disease: A systematic review and meta-analysis. JAMA Netw Open 2018;1:e182421.

And is not the bodily habit spoiled by rest and idleness, but preserved for a long time by motion and exercise? Plato from Theaetetus, 360 BCE1

In recent years, the importance of exercise to health, well-known for centuries, has become the focus of medical studies attempting to quantify and better understand the nuances of this relationship. Fang et al attempted to clarify an association between physical activity and the risk of Parkinson’s disease (PD). Undertaking a broad-based meta-analysis to uncover relationships obscured by the limits of individual trials, they identified eight eligible studies for this investigation.

PD is one of the more common neurodegenerative disorders (ranking second to dementia). The risk for developing the disorder increases with age. Although there is regional variation, the overall estimated prevalence of PD in North America in persons 45 years of age and older is 572 per 100,000. Gender also plays a role; PD is more common in men than in women, with a prevalence of 667 per 100,000 in men compared to 488 per 100,000 in women older than 45 years of age.2,3

Fang et al noted that the etiology of PD most likely involves a combination of genetic and environmental factors, and that physical activity is among the modifiable risk factors suspected to play a role in protecting against the development of this disorder. Studies of this relationship have been inconsistent in terms of methodology and outcome, making it difficult to prove an association. In fact, of eight studies identified for review and meta-analysis, only one revealed a statistically significant correlation between physical activity and a reduced risk of PD. Yet, when the pooled group was considered and a standardized approach to measuring physical activity was employed, an association between physical activity and reduced risk of PD became more apparent.

The gender balance within the 544,346 participants across the eight studies was close to equal. More than 2,100 cases of PD developed over the follow-up period. Fewer than one-third of these cases were in women.

According to Fang et al, each study included in the meta-analysis had different cutoffs for the physical activity category, making comparisons and conclusions difficult. To overcome this barrier, the group quantified physical activity from each study using a standardized measure of metabolic activity, termed MET hours. One MET hour represents the amount of energy expended when sitting quietly (approximately 1 kcal/kg/hour). MET hours required for moderate to strenuous activity are in the range of three to six times this baseline.4 By moving physical activity to a consistent measure of energy expenditure, the authors created categories of physical activity. These categories ranged from minimal physical activity to light physical activity to moderate/vigorous physical activity.

Fang et al reported positive or statistically significant results, as well as results that did not show statistical significance. The results in this study appear much more conclusive for men than for women. There are possibly several factors at work here. It is not unexpected that fewer women overall developed PD in the follow-up period, as this disorder occurs more frequently in men. However, the numbers may not be sufficient to uncover a true relationship between relative risks of PD and exercise or activity level in this gender. Fang et al noted that more studies looking harder at this relationship in women are needed to clarify if the relationship changes with more female participants.

The authors also considered the possibility of reverse causation overall and attempted to control for this. The mean length of the studies was 12 years. This seems sufficient to minimize the risk of mistaking low physical activity due to subclinical PD with low physical activity as a risk factor for PD. Nevertheless, they noted even with a time-lag meta-analysis (excluding the first several years of the study), the relationship between physical activity and reduced risk of PD in men remains consistent. With longer studies in the pipeline, this concern should be laid to rest eventually.

This is a convincing study with immediate clinical application, especially for patients with suspected genetic predisposition to PD. Smoking tobacco and caffeine consumption are among the modifiable risk factors linked to the onset of PD.5,6 Although both activities appeared to be inversely related to the development of PD, mitigating factors make prescribing or endorsing the use of each (especially cigarettes) less desirable. Exercise, with its known widespread benefits and central role in maintaining wellness, is a more natural fit for a prescription.

Future studies in this arena remain necessary. Robust longitudinal studies incorporating quantifiable and consistent definitions of physical activity verified by means other than self-report should help clarify the relationship between physical activity and reduced risk of PD. However, it is not necessary to wait for such studies to start recommending at least daily moderate to vigorous levels of physical activity to patients.


  1. Plato. Theaetetus. Available online at: Accessed April 4, 2019.
  2. Parkinson’s Foundation. Statistics. Available at: Accessed Feb. 23, 2019.
  3. Marras C, Beck JC, Bower JH, et al; Parkinson’s Foundation P4 Group. Prevalence of Parkinson’s disease across North America. NPJ Parkinsons Dis 2018;4:21.
  4. University of South Carolina. Arnold School of Public Health Prevention Research Center. The Compendium of Physical Activities. Available at: Accessed April 4, 2019.
  5. Rodriguez T. The Troubling Link Between Parkinson’s and Smoking: Can We Deny the Benefits? Neurology Advisor, Dec. 10, 2015. Available at: Accessed April 4, 2019.
  6. Munoz DG, Fujioka S. Caffeine and Parkinson disease: A possible diagnostic and pathogenic breakthrough. Neurology 2018;90:205-206.