Weight change, weight cycling and mortality in the ERFORT male cohort study

Objective To investigate the effect of weight change and weight fluctuations on all-cause-mortality in men. Methods Within a prospective population-based cohort of 1160 men aged 40-59 years at recruitment, complete weight change patterns from baseline and three follow-up examinations during a period of 15 years of follow-up was used to categorize the 505 men into stable obese, stable non-obese, weight gain, weight loss and weight fluctuation groups. For these men (age range: 55-74 years at start time of survival analysis) further survival was analyzed during the subsequent 15 years. Results Overall, 183 deaths were observed among the 505 men. Only weight fluctuations had a clear significant impact on all-cause mortality. Adjusted hazard rate ratio (HRR (95%-CI)) was 1.86 (1.31-2.66) after adjustment for age group, pre-existing cardiovascular disease or diabetes mellitus, smoking and socio-economic status. The risk rate due to weight loss was borderline significant (HRR = 1.81 (0.99-3.31)). Risk of death due to weight gain (HRR = 1.15 (0.70-1.88)) or stable obesity (HRR = 1.16 (0.69-1.94)), however, were not significantly increased compared to men staying non-obese for the first 15 years after cohort recruitment. Conclusion Weight fluctuations are a major risk factor for all-cause mortality in middle aged men. Moreover, stable obesity does not increase further mortality in men aged 55-74 years in long-term follow-up.