The frequency of overweight in men and women was studied in relation to the weight of their spouses in a 17-year prospective study. Survival and overall mortality risk were also assessed depending on the weight gradient of their spouse. A primary screening was made in 1988-1991, by examining a random residential sample (1546 men and women of 20-59 years of age); a rescreening was made in 2002-2005. A cohort family study included all married couples (n=427) from the examinees. The follow-up lasted 1 to 17 years (mean 15.47 years). Overweight was recorded if Quetelet's index was ≤25 kg/m2. A relationship was found in the weight of couples: if one of the couples was found to be overweight, the probability that his/her spouse would be obese increases. The relative risk of obesity for the men having overweight wives was 1.23 (95% CI, 1.08-1.41). The risk of obesity for the women having overweight husband was 1.36 times as high as that for the women whose husband had normal weight (95% CI, 1.10-1.68). In the spouses of overweight men and women, the mean Quetelet index was much higher than that in the spouses of men and women without this risk factor. In the men having obese wives, the 17-year risk of death turned out to be 2.53 times higher than that in those whose spouses had normal weight (95% CI, 1.27-5.06). The men whose wives were overweight were found to have lower survival rates (p=0.016). The mortality risk and survival rates in the women did not depend on the weight gradient of a spouse.