Childhood diet and insulin-like growth factors in adulthood: 65 year follow-up of the Boyd Orr Cohort

Abstract

Objective: High levels of insulin-like growth factor-I (IGF-I) are associated with an increased cancer risk and reduce risk of diabetes and coronary heart disease. We investigated associations of diet in childhood, in particular energy intake, with the IGF system in adulthood to determine if IGF-I – disease associations could be linked to early nutrition. Design: Retrospective cohort study. Setting: Sixteen survey centres in England and Scotland that originally participated in the Carnegie (Boyd Orr) Survey of Diet and Health in Pre-War Britain, 1937–1939. Subjects: Seven hundred and twenty-eight participants (679 with complete data) in the Boyd Orr cohort. Methods: Participants were originally surveyed between 1937 and 1939 (at median age 5.8 years; inter-quartile range: 2.9–9.6) and were followed up for 65 years. Dietary exposure in childhood was assessed from 7-day household food inventories. Outcomes are expressed as regression coefficients for the change in IGF per standard deviation increased childhood nutrient or food intake, as derived from levels of household consumption. Results: In fully-adjusted models, energy-rich family diets in childhood were not associated with IGF-I (regression coefficient: 0.9 ng/ml; 95% confidence interval (CI): 1.8, 3.7), IGF-II, IGF binding proteins (IGFBP)-2 or IGFBP-3 in adulthood. IGF-I was associated inversely with childhood family-diets high in milk ( 2.5 ng/ml; 5.1, 0.1; P¼0.05) and positively with vegetable-rich diets (3.5 ng/ml; 0.9, 6.1; P¼0.009). IGF-I was not associated with family diets rich in protein, carbohydrates, fats, calcium, meat or fruit. IGF-II, IGFBP-2 and IGFBP-3 were not related to childhood family diet. Conclusions: This study suggests that energy-rich family diets in childhood do not program the IGF system in adulthood. As childhood diet was based on household consumption, however, measurement error may obscure individual-level diet-IGF associations. The associations of milk- and vegetable-rich family diets in childhood with IGF-I could be chance findings, but nevertheless are consistent with recent publications and warrant further investigation. Sponsorship: Funded by The World Cancer Research Fund (Grant No. 2001/31) with additional funds provided by UK Survivors. RMM was funded by a Wellcome Trust research training fellowship in clinical epidemiology (GR063779FR).