Effects of marital transitions on changes in dietary and other health behaviours in US women.

BACKGROUND Previous studies have indicated that married people have lower mortality and are generally healthier. Most previous studies have been cross-sectional and few studies investigated the effect of marital transition on health. With a prospective design and repeated measures of variables, we sought to analyse the temporal relation between marital transition and change in health behaviours. METHODS We followed up 80 944 women aged 46-71 for 4 years (1992-1996). All information was self-reported. We used multivariate-adjusted linear and logistic regression models to examine the impact of changes in marital status on concomitant changes in health behaviours, controlling for potential confounders and baseline health behaviours. RESULTS Compared with women who remained married, women who divorced/widowed had body mass index (BMI) decreases of 0.65 kg/m(2) (P < 0.001) and 0.44 kg/m(2) (P < 0.001), respectively. Compared with women who remained unmarried, women who remarried had an increase in mean BMI of 0.41 kg/m(2) (P < 0.001). Women who divorced increased physical activity by 1.23 metabolic equivalent hours (MET)/week (P = 0.07) compared with women who stayed married. Among non-smokers and past smokers, women who divorced/widowed had more than a twofold increased risk of relapsing/starting smoking (OR = 2.47, 95% CI: 1.56, 3.89; OR = 2.08, 95% CI: 1.56, 2.76, respectively) than women who stayed married. Divorced and widowed women had decreased vegetable intake relative to women who stayed married (-2.93 servings/week [P < 0.001] and -1.67 servings/week [P < 0.001], respectively). CONCLUSIONS These patterns suggest both health-damaging and health-promoting changes accompanying divorce and widowhood, and generally health-promoting changes following remarriage.