Relationship between changes in physical activity and plasma insulin during a 2.5-year follow-up study

The association between changes in physical activity, body weight, and diet and fasting plasma insulin was analyzed in a 2.5-year follow-up study of 146 men aged 50 to 60 years. Physical activity was assessed by a 7-day physical activity recall interview, diet by a 4-day food record, and plasma insulin radioimmunologically. Total physical activity decreased from (mean +/- SD) 45.1 +/- 10.1 to 39.0 +/- 6.1 metabolic equivalent (MET) hours (METh).d-1 and conditioning physical activity (> 5.0 METs) from 8.0 +/- 11.2 to 2.7 +/- 5.0 METh.d-1, whereas plasma insulin increased from 8.2 +/- 5.8 to 9.2 +/- 6.7 mU.L-1 and body weight from 80.5 +/- 12.0 to 81.6 +/- 11.6 kg during the follow-up period (P < or = .001 for all). The change in conditioning physical activity correlated inversely (r = -.34, P < .001) and change in body weight positively (r = .42,P < .001) with the change in plasma insulin level. With data adjusted for the baseline insulin level, cardiovascular health status, alcohol intake, change in body weight, smoking, age, and follow-up time, the odds ratio for an increase in fasting plasma insulin was 8.9 (95% CI, 2.1 to 37.1; P = .003) for men with the greatest decrease in conditioning physical activity (< -7 METh) compared with men who reported an increase in conditioning physical activity. The same logistic regression model showed an odds ratio of 9.9 (95% CI, 2.1 to 45.4; P = .003) for the increase in plasma insulin for subjects who gained more than 3.3 kg body weight compared with subjects who lost at least 0.6 kg. Men who consumed at least 12 g.d-1 alcohol at both examinations had an odds ratio of 12.8 (95% CI, 1.7 to 94.5; P = .012) compared with nondrinkers. These data suggest that in middle-aged men, a reduction in physical activity increases the risk for increased plasma insulin independently of alcohol intake and changes in body weight.