Abstract P106: Stable versus Changing BMI Trajectories in Relation to Cardiometabolic Risk Factor Trajectories in Adolescent Girls: The NHLBI Growth and Health Study

Abstract

Introduction: Adolescence is a period of increased independence in lifestyle choices, as well as shifts in cardiometabolic risk factors. Understanding how stability or change in BMI status may be superimposed on risk factor changes will enable focused attention on girls whose risk factors are likely to deteriorate during adolescence. Methods: Girls were enrolled at age 10 into the NHLBI Growth and Health Study (NGHS) at 3 centers (n=2379) and assessed annually to age 19. Ideal cardiovascular risk factors (smoking, dietary sodium, dietary saturated fat, BMI, blood pressure (BP), total cholesterol (TC) and glucose), and related risk factors (HDL and LDL cholesterol, and triglycerides (TG)) were assessed. Trajectories of adherence to ideal risk factor definitions were evaluated using group-based modeling, with risk factor trajectories layered over BMI trajectories in change-over-change analysis. Results: Four trajectories of BMI across adolescence were: ideal (51%), deteriorating (14%), improving (11%) and poor (25%). Two to 4 trajectories were identified for other risk factors, with similar trajectory shapes (e.g., ideal, improving, deteriorating and poor). Ideal smoking prevalence only deteriorated, with trajectories differing by timing of smoking uptake (early, middle or late adolescence). Most adolescent girls had poor adherence to diet metrics throughout adolescence, with diet intake trajectories improving somewhat at different ages (early or later adolescence) in a small percent of participants. Stable poor or ideal BMI trajectories were associated primarily with stable risk factor trajectories (poor or ideal, respectively), for LDL, HDL, TG, BP (all p=0.0001). The poor BMI trajectory was also associated with late adolescent deteriorating glucose (p=0.0001) and TC trajectories (p=0.02). Changing BMI trajectories, either improving or deteriorating, can unmask the sensitivity of risk factors to change in BMI status. The improving BMI trajectory was more likely to be in the later-improving saturated fat (p=0.003), improving HDL (p=0.0002) and moderate-improving TG trajectories (p=0.01) and less likely to be in the low-deteriorating BP trajectory (p=0.04) compared to the deteriorating BMI trajectory. Smoking trajectories did not differ by BMI trajectory. Conclusions: Stable poor BMI is associated with poor, but not with deteriorating cardiometabolic risk profiles during adolescence, suggesting these changes must have occurred before age 10. Decreasing ideal TC and glucose appear to be late adolescent responses to persistently poor BMI. However, improving BMI status during adolescence is associated with improving dietary saturated fat intake and blood lipid profiles, while deteriorating BMI is associated with deteriorating BP and lipids. Thus, specific ideal health factors may be differentially sensitive to BMI changes in adolescent girls.