Predictors of weight change over two years among a population of working adults: the Healthy Worker Project.

The present study examined behavioural predictors of body weight cross-sectionally and longitudinally in a cohort of 1639 male and 1913 female employees in 32 companies participating in a worksite intervention study for smoking cessation and weight control. Dietary intake, current and previous dieting behaviours, and physical activity were examined for their association with body weight over the two-year period. Cross-sectionally in both men and women, history of previous dieting, previous participation in a formal weight loss programme, current dieting and meat consumption were positively related to body weight while high intensity activity was negatively related to body weight. Prospectively, history of participation in a formal weight loss programme and dieting to lose weight at baseline, and increased consumption over time of french fries, dairy products, sweets and meat, independently predicted increases in body weight in women. Women who were dieting to lose weight or who had previously participated in a formal weight loss programme at baseline gained 1.99 lb and 1.74 lb more, respectively, than those who were not dieting to lose weight or who had not previously participated in a formal weight loss programme. Increased exercise, either walking or high intensity activity, predicted decreases in body weight in women (1.76 lb and 1.39 lb, respectively, for each session increase per week). In men, previous participation in a formal weight loss programme predicted increases in body weight over the two-year period. Men who had previously participated in a formal weight loss programme at baseline gained 4.83 lb more than those who had never previously participated in a formal weight loss programme. Increases in consumption of sweets and egg were prospectively related to increases in body weight, while increased walking and high intensity activity were related to decreases in body weight (0.86 lb and 3.54 lb, respectively, for each session increase per week). These results suggest the role that specific diet and exercise behaviours may play in body weight changes over time.