Breast feeding and cardiovascular disease risk factors, incidence, and mortality: the Caerphilly study.

STUDY OBJECTIVE To investigate the association of having been breast fed with cardiovascular disease risk factors, incidence, and mortality. DESIGN Prospective cohort study. SETTING Caerphilly, South Wales. PARTICIPANTS All men aged 45-59 years living in and around the study area. Of 2818 eligible men, 2512 (89%) were seen. Altogether 1580 men (63%) obtained details of how they had been fed in infancy (ever breast fed or only bottle fed) from their mother or a close female relative. A subset of 1062 subjects reported on whether bottle fed or the duration of breast feeding if breast fed. MAIN RESULTS Breast feeding was not associated with stature, blood pressure, insulin resistance, total cholesterol, or fibrinogen. In fully adjusted models (controlling for age, birth order, and social position in childhood and adulthood), breast feeding was associated with greater body mass index than bottle feeding (difference: 0.41 kg/m(2) (95% CI: 0.01 to 0.81). There was a positive association between breast feeding and coronary heart disease mortality (hazard ratio: 1.73; 1.17 to 2.55) and incidence (1.54; 1.17 to 2.04) (fully adjusted models). There was no evidence of a duration-response effect, which might be expected if an adverse effect of breast feeding was causal. CONCLUSION These data provide little evidence of a protective influence of breast feeding on cardiovascular disease risk factors, incidence, or mortality. A possible adverse effect of breast feeding on coronary heart disease incidence was observed but may have a number of explanations, including selection and information bias. In view of these limitations, further long term studies with improved measures of infant feeding are required to confirm or refute these findings.