RT Journal Article
SR Electronic
T1 Childhood dairy and calcium intake and cardiovascular mortality in adulthood: 65-year follow-up of the Boyd Orr cohort
JF Heart
JO Heart
FD BMJ Publishing Group Ltd and British Cardiovascular Society
SP 1600
OP 1606
DO 10.1136/hrt.2009.168716
VO 95
IS 19
A1 J C van der Pols
A1 D Gunnell
A1 G M Williams
A1 J M P Holly
A1 C Bain
A1 R M Martin
YR 2009
UL http://heart.bmj.com/content/95/19/1600.abstract
AB Background: Dairy consumption in childhood may have long-term effects on cardiovascular mortality through influencing the development of risk factors or programming effects.Objective: To investigate whether dairy and calcium consumption in childhood is associated with adult mortality due to coronary heart disease (CHD), stroke and all causes.Methods: In 1937–9, 4999 children in England and Scotland participated in a study of family food consumption, assessed from 7-day household food inventories. Cause of death was ascertained between 1948 and 2005 in 4374 traced cohort members with complete data. Per capita household intake estimates for dairy products and calcium were used as proxies for individual intake.Results: No strong evidence that a family diet in childhood high in dairy products was associated with CHD or stroke mortality was found. However, childhood calcium intake was inversely associated with stroke mortality (multivariable adjusted hazard ratio (HR) for highest versus lowest calcium group: 0.41; 95% confidence interval (CI) 0.16 to 1.05; p for trend = 0.04), but not CHD mortality. All-cause mortality was lowest in those with the highest family dairy (HR = 0.77; 95% CI 0.61 to 0.98; p for trend = 0.04) and calcium intake (HR = 0.77, 95% CI 0.60 to 0.98; p for trend = 0.05).Conclusions: Children whose family diet in the 1930s was high in calcium were at reduced risk of death from stroke. Furthermore, childhood diets rich in dairy or calcium were associated with lower all-cause mortality in adulthood. Replication in other study populations is needed to determine whether residual confounding explains part of these findings.