Genetic variability of adult body mass index: A longitudinal assessment in Framingham families

Objective: To explore the contribution of genetics to the mean, SD, maximum value, maximum less the mean, and change over time in body mass index (BMI) and the residual of body weight after adjustment for height. BMI is frequently used as a general indicator of obesity because of its ease and reliability in ascertainment. Cross-sectional twin and family studies have shown a moderate-to-substantial genetic component for BMI. However, the contribution of genetics to the long-term average, variability, or change over time in BMI is less clear. Research Methods and Procedures: Longitudinal data from the Framingham heart study were used to create pedigrees of age-matched individuals. Heritability estimates were derived using variance-decomposition methods on a total of 105 1 individuals from 380 extended pedigrees followed for a period of 20 years. All subjects were followed from approximately age 35 to 55 years. Results: Moderate heritability estimates were found for the mean BMI (h(2) = 0.37), maximum BMI (h(2) = 0.40), and the mean residual of body weight (h(2) = 0.36). Low heritability estimates (h(2) congruent to 0.20) were found for the maximum less the mean in BMI and the SDs of BMI and residual of body weight. No additive genetic contribution was found for the average change over time in BMI or the residual of body weight. Discussion: These findings suggest that there is a significant genetic component for the magnitude of BMI throughout an individual's middle-adult years; however, little evidence was found for a genetic contribution to the variability or rate of change in an individual's BMI.