rss
J Epidemiol Community Health 64:A16 doi:10.1136/jech.2010.120956.40
  • Society for Social Medicine abstracts
  • Cancer survival

040 Early and mid-adulthood BMI in relation to later cancer mortality: over 80 years of follow-up in the Harvard Alumni Health Study

  1. G D Batty1
  1. 1Medical Research Council Social and Public Health Sciences Unit, Glasgow, UK
  2. 2Harvard School of Public Health, Boston, MA, USA
  3. 3Brigham and Women's Hospital, and Harvard Medical School, Boston, MA, USA

Abstract

Objective The association between adiposity in early adulthood and subsequent development of specific malignancies is still unclear. Additionally, the potential mediating role of adiposity in middle age has not been well examined. We investigated the association of body mass index (BMI; weight(kg)/height(m2)) in early adulthood with later mortality from several cancers.

Design Cohort study of male Harvard University students who had a medical examination at university between 1914 and 1952 (mean age 18.4 years) when height and weight were measured. Data on smoking habits were recorded and physical activity details were ascertained from athletic records. Alumni were traced and mailed a health questionnaire in 1962 or 1966 (mean age 45.1 years) which included enquiries regarding height and weight. They were then followed for subsequent mortality experience – which is >99% complete—until the end of 1998 (mean age at follow-up 56.5 years). In Cox regression models, adjustment was made for university smoking and physical activity levels; joint models were used to explore mediation by BMI in middle age. Imputation was used to allow inclusion of 4040 men with missing data for at least one variable.

Setting US.

Participants 19 593 males in the Harvard Alumni Health Study cohort who had a medical examination at university and returned a mailed questionnaire in 1962 or 1966.

Main outcome measure Mortality from cancer.

Results There were 8445 deaths in total, 2395 of which were from cancer. A one SD increase in early adulthood BMI was associated with an increased risk of death from cancer from all sites combined (adjusted hazard ratio 1.11; 95% CI 1.05, 1.17), plus lung (1.24; 95% CI 1.10 to 1.40) and skin (1.29; 95% CI 0.96 to 1.75). Cancers of the pancreas, stomach, liver, brain, prostate, kidney, bladder, lymphatic and haematopoietic tissue were not associated with BMI in early adulthood. Results were equivalent when based on multiply imputed data and accounting for the role of middle age BMI.

Conclusion In this cohort, higher BMI in early adulthood appears to be a risk factor for some malignancies several decades later, and appeared to be neither confounded by lifestyle factors nor meditated via BMI in middle-age. These findings suggest that strategies to tackle obesity early in life may be important for the prevention of selected cancers.