Leisure time physical activity, risk of depressive symptoms, and inflammatory mediators: the English Longitudinal Study of Ageing.

OBJECTIVES: To examine if inflammatory markers (CRP, fibrinogen) might partly explain the association between physical activity (PA) and risk of depression.

DESIGN/SETTING: The English Longitudinal Study of Ageing, a prospective study of community dwelling older adults.

PARTICIPANTS: 4323 men and women (aged 63.4+/-9.7 yrs) free from depression at baseline.

MEASURES: Self reported leisure time PA levels and depressive symptoms (a score of > or = 4 using the 8-item CES-D scale) were assessed at baseline and 4 yrs follow up. The inflammatory markers, CRP and fibrinogen, were assessed at a 2 yrs intermediate time point between baseline and follow up.

RESULTS: At follow up 8% of the sample reported depressive symptomatology. In comparison with participants reporting none or light PA, the odds of depressive symptomatology for those reporting moderate or vigorous PA were 0.71 (95% CI, 0.54-0.95) and 0.58 (0.41-0.81), respectively, after adjustments for baseline CES-D score, age, gender, social-occupational class, smoking, alcohol, and chronic illness. Each standard unit increase in log CRP was associated with higher odds of depressive symptomatology at follow up (1.32, 1.13-1.55) and CRP was inversely associated with physical activity. The association between PA and depressive symptomatology was not, however, substantially modified by further adjustment for CRP (odds for none vs. vigorous PA=0.60, 0.43-0.84).

CONCLUSIONS: These data suggest that low grade systemic inflammation, as indexed by CRP, is a risk marker for depressive symptomatology, although this mechanism explains only a modest (approximately 5%) amount of the association between PA and risk of depression.