AIMS: To explore the prevalence of overweight in men and women in a Swedish rural community and to examine its associations with all-cause mortality. METHODS: A community-based cohort study. A total of 1,109 men and women aged >or= 40 years participated in a survey of cardiovascular disease (CVD) risk factors in the city of Skara in Skaraborg, Sweden, in 1993-94. Overweight was defined as a BMI >or=25.0 kg m(-2). Vital status was ascertained to 31 December 1999 and sex-specific associations between overweight and mortality were explored. RESULTS: The prevalence of overweight according to the WHO's criteria exceeded 50% in men and 35% in women. In men there was an inverse association between BMI and mortality. Men in the lowest quartile of BMI experienced the highest mortality with 44.1 deaths per 1,000 person-years. The hazard ratio (HR) in the highest quartile was 0.6 (95% CI 0.4-0.9). In women there were no significant differences in mortality between quartiles of BMI. In both men and women with previous CVD the mortality rates decreased with quartiles of increasing BMI. The inverse association between BMI and mortality was confined to elderly men with a history of CVD. CONCLUSION: In both sexes the association between BMI and mortality differed across subgroups of age and of a history of previous CVD. No indication of overweight being negative for longevity was found in this population. Higher age and a history of previous CVD contribute to the excess mortality seen in subjects with low BMI.
AIMS: To explore the prevalence of overweight in men and women in a Swedish rural community and to examine its associations with all-cause mortality. METHODS: A community-based cohort study. A total of 1,109 men and women aged >or= 40 years participated in a survey of cardiovascular disease (CVD) risk factors in the city of Skara in Skaraborg, Sweden, in 1993-94. Overweight was defined as a BMI >or=25.0 kg m(-2). Vital status was ascertained to 31 December 1999 and sex-specific associations between overweight and mortality were explored. RESULTS: The prevalence of overweight according to the WHO's criteria exceeded 50% in men and 35% in women. In men there was an inverse association between BMI and mortality. Men in the lowest quartile of BMI experienced the highest mortality with 44.1 deaths per 1,000 person-years. The hazard ratio (HR) in the highest quartile was 0.6 (95% CI 0.4-0.9). In women there were no significant differences in mortality between quartiles of BMI. In both men and women with previous CVD the mortality rates decreased with quartiles of increasing BMI. The inverse association between BMI and mortality was confined to elderly men with a history of CVD. CONCLUSION: In both sexes the association between BMI and mortality differed across subgroups of age and of a history of previous CVD. No indication of overweight being negative for longevity was found in this population. Higher age and a history of previous CVD contribute to the excess mortality seen in subjects with low BMI.