Physical activity and inflammation in a multiethnic cohort of women

Purpose: Many cross-sectional studies using data from a single time point have reported that higher levels of physical activity or fitness are associated with lower levels of inflammatory markers, but data examining change are limited, as are race/ethnicity-specific data. The purpose of this study was to examine the associations between physical activity and inflammation assessed at two time points among women of different race/ethnicities. Methods: A total of 1355 postmenopausal women (301 whites, 300 blacks, 300 Hispanics, 300 Asians/Pacific Islanders, and 154 American Indians) age 50-79 yr were studied. Participants were from 40 US cities and were free of cardiovascular disease and cancer. At baseline and year 3, women reported their recreational physical activities and provided blood samples, which were analyzed for several inflammatory markers. Results: In cross-sectional analyses, after adjusting for several potential confounders including body mass index, higher physical activity levels were generally related to lower inflammatory marker concentrations. For example, P values for a linear trend of lower C-reactive protein levels across physical activity tertiles at baseline were <0.0001 in all women and 0.94, 0.09, 0.002, 0.20, and 0.10, respectively, for the five race/ethnic groups listed above. For interleukin 6, the corresponding P values were <0.0001, 0.0007, 0.01, 0.03, 0.37, and 0.004, respectively, at baseline. Relationships at year 3 were similar to baseline. However, there was no relation between changes in physical activity and changes in inflammatory markers during the 3-yr period. Conclusions: Among middle-age and older women overall, there were strong, inverse, cross-sectional associations between physical activity level and inflammatory markers. However, changes in inflammatory markers were unrelated to changes in physical activity. These data suggest a noncausal association between physical activity and inflammatory markers.