Abstract

OBJECTIVE: To investigate the relation between childhood height, its components--leg length and trunk length--and mortality in adulthood. DESIGN: Cohort study based on the Carnegie (Boyd Orr) Survey of diet and health in pre-war Britain, 1937-9. SETTING: The 14 centres in England and Scotland that participated in the Carnegie Survey and where children were examined. Scottish centres: Aberdeen, Dundee, West Wemyss, Coaltown of Wemyss, Hopeman, Methlick, Tarves, Barthol Chapel. English Centres: Liverpool, York-shire, Barrow in Furness, Wisbech, Fulham, and Bethnal Green. SUBJECTS: 2990 boys and girls aged between 2 years and 14 years 9 months when they were examined in 1937-9. These children were drawn from 1134 families who underwent a one week assessment of family diet and home circumstances. Of these, 2547 (85%) have been traced and flagged using the NHS Central Register. MAIN OUTCOME MEASURES: Age adjusted overall, coronary heart disease, and cancer mortality in men and women in relation to age and sex specific z scores for height, leg length, and trunk length. All analyses were adjusted for the possible confounding effects of childhood and adult socioeconomic circumstances and childhood diet. RESULTS: Leg length was the component of childhood height most strongly associated with socioeconomic and dietary exposures. There was no significant relation between childhood height and overall mortality. Height-mortality relations were observed in relation to both coronary heart disease (CHD) and cancer. Leg length was the component of height most strongly related to cause specific mortality. In men and women CHD mortality increased with decreasing childhood leg length. Men in the lowest leg length quintile had a relative risk (RR) of 2.5 (95% CI 1.0 to 6.2) compared to those with the longest legs (linear trend p = 0.14). Similarly, women in the lowest leg length quintile had a RR of 3.9 (95% CI 0.8 to 19.0; linear trend p < 0.01). Adjustment for childhood and adult socioeconomic circumstances had little effect on these trends. In men, but not women, those who as children had long legs experienced increased cancer mortality. The significant relations between anthropometry and both CHD and cancer mortality were restricted to those aged < 8 years when measured. CONCLUSIONS: These findings suggest that adverse diet and living conditions in childhood, for which leg length seems to be a particularly sensitive indicator, are associated with increased risk of CHD in adulthood and possibly reduced cancer risk. It is likely that these influences operate after birth, during the first few years of life.