Marital status and mortality in middle-aged Swedish men

In a large primary prevention trial among middle-aged men in Gothenburg, Sweden, register data were used to establish marital status, alcohol abuse, and economic problems for nearly all of the study population in 1970-1973. Married men had a higher participation rate in the examinations for the trial than non-married men, with non-married alcoholic men having the lowest participation rates. Among the participants, 26% of divorced men, but only 5% of married men were registered with the social authorities for alcohol problems. Serum cholesterol, body mass index, and diabetes were not associated with marital status, but smoking was more common among widowers and divorced men. Nonfatal myocardial infarction was not related to marital status among participants, after a mean follow-up of 11.8 years. Death from coronary heart disease was more common in non-married men in univariate analysis, but not when other risk factors were taken into consideration. In participants, married men had a mortality rate of 9%, compared with 20% for divorced men. After adjustment for other risk factors, including registration for alcohol problems, smoking, and occupational class, the association between marital status and total mortality was still highly significant. Among nonparticipants in the trial, 13% of married men were registered for alcohol problems, compared with 41% of divorced men. Nonparticipants had higher all-cause mortality, 18% for married men and 33% for divorced men.