Back to the couch to brood?

Exercise Doesn't Improve Mood
By Crystal Phend, Staff Writer, MedPage Today
Published: August 04, 2008
Reviewed by Zalman S. Agus, MD; Emeritus Professor
University of Pennsylvania School of Medicine. Earn CME/CE credit
for reading medical news



AMSTERDAM, Aug. 4 -- Regular exercise does not reduce anxiety and depression in the general population, researchers here found.

Genetically identical twins showed no difference in anxious and depressive symptoms when one in the pair exercised more than the other, reported Marleen H. M. De Moor, M.Sc., of VU University Amsterdam, and colleagues in the August issue of the Archives of General Psychiatry.

Nor did symptoms decline over time as individuals started exercising more. Although the large, population-based study of twin families confirmed that getting less regular exercise was associated with worse mental health symptoms in the general population as seen in prior studies, the findings show that the effect is not causal, the researchers said.

But the lack of direct benefits for mood shouldn't change physicians' recommendations for the general population, the investigators cautioned.


"These findings do not detract from the beneficial effects of regular exercise on numerous aspects of physical health such as cardiovascular disease and type 2 diabetes mellitus," De Moor and colleagues wrote.


For patients diagnosed with an anxiety or depressive disorder, studies have suggested that prescribed exercise may alleviate symptoms. But that effect may be more environmentally driven, the researchers said.


"The antidepressant effects of exercise may only occur if the exercise is monitored and part of a therapeutic program," they wrote.


To test the causal link suggested by prior studies, De Moor's group analyzed data from 5,952 twins in the Netherlands Twin Register, 1,357 of their other siblings, and 1,249 parents. The analysis included only adults ages 18 to 50.


Overall, there were small but significant correlations between a higher level of leisure-time physical activity and fewer anxious and depressive symptoms (correlation coefficient −0.12, 95% confidence interval −0.14 to −0.09).


Notably, genetics appeared to account for the associations with statistically significant correlation coefficients ranging from −0.16 to −0.24.


Exercise level predicted mental health symptoms in genetically identical monozygotic twin pairs but not in genetically distinct dizygotic twins and sibling pairs.


When genetically identical twins differed from each other in exercise level, the twin who exercised more was not significantly less depressed than the one who exercised less (intrapair difference score range −0.04 to 0.03).


This finding did "not support the hypothesis that exercise causes relief in anxious and depressive symptoms," the researchers said.


Individual environmental correlations -- reflecting level of exercise and other nongenetic factors -- were not significant (coefficients −0.07 to 0.05), "suggesting that the association between exercise behavior and symptoms of anxiety and depression is not explained by a causal effect," the researchers again noted.


In the longitudinal analysis, individual symptom and exercise levels reported periodically from 1991 through 2002 were significantly correlated over time with coefficients comparable to the cross-sectional analysis (−0.07 to −0.14).


Again, genetics were significantly correlated with the longitudinal link between symptoms of anxiety and depression and exercise level whereas environmental correlations were not significant.


Changes over time in the number of metabolic equivalent task (MET) hours of exercise reported were not associated with either significant decrease or increase in anxious and depressive symptoms in the subsequent two-, four-, seven-, nine-, or 11-year intervals.


Which genes might "simultaneously affect the regulation of exercise drive and mood" and thereby account for the association between exercise behavior and risk of anxiety and depression is unknown, the researchers said.


Likely candidates might be those in the dopaminergic, opioidergic, norepinephrenergic, or serotonergic pathways of the brain, they speculated.


They cautioned that the study may have been affected by selection bias, as individuals from families with more participants in the study were less depressed and anxious.


They also noted that the study's power was limited to detect small correlations in men in particular, that the study could not examine more complex causality, and that generalizability could be limited to different age groups and countries.