The Longitudinal Relation of Childhood Height to Subsequent Obesity in a Large Electronic Health Record Database

Objective: Several cross-sectional studies have shown that height in childhood is correlated with BMI and with body fatness, and two longitudinal studies have reported that childhood height is associated with adult BMI. This study explored this longitudinal association in an electronic health record database of 2.8 million children. Methods: Children were initially examined between the ages of 2 and 13.9 years and, on average, were reexamined 4 years later. Results: As expected, there was a cross-sectional correlation between height-for-age z score and BMI that increased from r = −0.06 (age of 2 years) to r = 0.37 (age of 9-10 years). In addition, height-for-age at the first visit was related to subsequent BMI and obesity, with the prevalence of subsequent obesity increasing about fourfold over six categories of height-for-age at the first visit. About 40% of this longitudinal association was independent of initial BMI, but its magnitude decreased with initial age. For example, the initial height-for-age of children who were 12 years of age or older was only weakly associated with subsequent BMI. Conclusions: Health professionals should recognize that greater childhood height-for-age before 12 years of age may be a marker for increased risk of subsequent obesity.