79 Clemens RA, Hernell O, Michaelsen KF (eds): Milk and Milk Products in Human Nutrition. Nestlé Nutr Inst Workshop Ser Pediatr Program, vol 67, pp 79–97, Nestec Ltd., Vevey/S. Karger AG, Basel, © 2011. Milk and Linear Growth: Programming of the IGF-I Axis and Implication for Health in Adulthood Richard M. Martin a,b , Jeff M.P. Holly c , David Gunnell a a Department of Social Medicine, b MRC Centre for Causal Analysis in Translational Epidemiology, Department of Social Medicine, and c Clinical Sciences North Bristol, University of Bristol, Bristol, UK Abstract There is increasing awareness that childhood circumstances influence disease risk in adulthood. As well as being strongly influenced by genes/genetic factors, stature acts as a marker for early-life exposures, such as diet, and is associated with risk of several chronic diseases in adulthood. Stature is also a marker for levels of insulin-like growth factor (IGF)-I in childhood. Levels of IGF-I are nutritionally regulated and are therefore modifiable. Milk intake in childhood and in adulthood is positively associ- ated with higher levels of circulating IGF-I and, in children, higher circulating IGF-I promotes linear growth. Studies conducted by our team and others, however, indicate that the effect of milk is complicated because consumption in childhood appears to have long-term, programming effects which are opposite to the immediate effects of consuming milk. Specifically, studies suggest that the long-term effect of higher levels of milk intake in early childhood is opposite to the expected short-term effect, because milk intake in early-life is inversely associated with IGF-I levels throughout adult life. We hypothesize that this long-term programming effect is via a resetting of pituitary control in response to raised levels of IGF-I in childhood. Such a programming effect of milk intake in early life could potentially have implications for cancer and ischemic heart disease risk many years later. Copyright © 2011 Nestec Ltd., Vevey/S. Karger AG, Basel Introduction The opportunities to directly examine the relationship between early nutri- tion and diseases in adulthood are limited because few cohort studies have information from birth until old age. To date, therefore, most epidemiological