Body mass index, coronary heart disease and stroke in Swedish women. A prospective 19-year follow-up in the BEDA study

BACKGROUND: Although obesity is an important determinant of an unfavourable risk factor pattern reported associations between cardiovascular disease and obesity in women have been remarkably inconsistent. DESIGN: Longitudinal observational population study. METHODS: 1408 Göteborg women without prior cardiovascular disease aged 39 to 65 years at baseline were examined with respect to cardiovascular risk factors, including body mass index (BMI), in 1979 to 1981. Quartiles of BMI were formed of <22 (reference), 22 to 24, 24 to 27, and >27 kg/m(-2). Follow-up was conducted by use of the Swedish patient and cause-specific death registers. RESULTS: All trends with respect to incident coronary heart disease (CHD--myocardial infarction or revascularization), stroke and all cardiovascular disease were positive and significant (P<0.05). No significant increase in risk was noted in women with BMI 22-24, compared with women below 22. After adjustment for smoking, women with BMI 24 to 27 had a doubled risk of CHD [hazard ratio(HR) 2.41 (1.06-5.50)] and of any cardiovascular disease [HR 1.89 (1.05-3.37)] whereas the increase in stroke risk was non-significant [HR 1.80 (0.81-4.01)]. Hazard ratios in the heaviest women, with BMI >27, were 3.75 (1.68-8.37) for CHD, 2.84 (1.32-6.12) for stroke, and 2.98 (1.70-5.21) for any cardiovascular disease, after adjustment for smoking. After further adjustment for other cardiovascular risk factors, all trends became non-significant. However, women with BMI >27 still displayed a statistically independent association with respect to coronary disease [adjusted HR 2.67 (1.10-6.47)] and all cardiovascular disease [HR 2.23 (1.23-4.04)], but not stroke [HR 2.08 (0.94-4.61)]. CONCLUSION: The influence of BMI on cardiovascular disease in women may be greater than previously thought and, although to a great extent explained by the influence of obesity on other risk factors, associated with adverse outcomes already at moderately increased body weight.