Incident hypercholesterolemia in relation to changes in vigorous physical activity.

PURPOSE: Test whether changes in vigorous exercise affect the risk for hypercholesterolemia.
METHODS: : Incident physician-diagnosed hypercholesterolemia was compared in recreational runners whose weekly distances run increased >0.5 km x d (4034 men, 1897 women), remained constant (+/-0.5 km x d; 4685 men, 1904 women), or decreased >0.5 km x d (15,678 men, 6224 women) during 7.8 yr of prospective follow-up.
RESULTS: Relative to those whose distance was maintained or reduced, the odds ratio (95% confidence interval) for hypercholesterolemia for those whose running increased was significantly less than 1 for men {0.82 (0.72-0.93), P = 0.0006}, women {0.71 (0.54-0.92), P = 0.01}, and men and women combined adjusted for sex {0.80 (0.72-0.89), P < 0.0001}. The significance of the odds ratios was somewhat diminished when adjusted for baseline and follow-up BMI in men {0.89 (0.79-1.01), P = 0.06}, women {0.71 (0.54-0.92), P = 0.08}, and the adjusted sexes combined {0.88 (0.79-0.98), P = 0.02}. When average distance run {i.e., (baseline + follow-up) / 2} was compared with incident hypercholesterolemia in the 6589 runners whose distance remained constant, each kilometer-per-day increment in average running distance was associated with an odds ratio significantly less than 1 in men {0.90 (0.86-0.93)}, women {0.91 (0.84-0.98)}, and the adjusted sexes combined {0.90 (0.87-0.93)}, which was also attributable in part to BMI in men {0.94 (0.90-0.97), P = 0.0005}, women {0.96 (0.89-1.04), P = 0.35}, and the adjusted sexes combined {0.94 (0.91-0.97), P = 0.0003}.
CONCLUSIONS: The odds for hypercholesterolemia 1) decrease in runners who increase their running mileage and 2) decline in association with the higher dose of vigorous activity even in the absence of any change in exercise. These effects are associated in part to the runners' BMI.