Abstract MP23: Association of Trajectories of Adulthood Weight Gain with Cardiovascular Disease Risk Factors in Participants 50-74 Years in the Hispanic Community Health Study/Study of Latinos (HCHS/SOL)

Abstract

The CVD health of US Hispanic/Latino individuals—the second largest ethnic group in the US—is a top public health priority. Although Hispanic/Latino individuals have higher rates of obesity than non-Hispanic Whites, the pattern of adulthood weight gain in this diverse group and its relationship with CVD risk factors are unknown. We examined trajectories of weight gain and the associations of these trajectory classes with dyslipidemia, diabetes, physical inactivity, poor diet quality and hypertension among HCHS/SOL participants baseline ages 50-74 years. HCHS/SOL is a population-based cohort study of self-identified Hispanic/Latino adults from four urban US communities. At baseline HCHS/SOL participants were asked to recall/report their body weight at 21, 45, 65 years old, last year and currently (kg or lb, 2008-2011). We used a finite mixture model (selected for ≤10 classes, ≥3% of sample per class and favored by change in the Bayesian Information Criterion) to assign individuals with two or more self-reported weights to their most likely trajectory class. We then described the distribution of class membership across categories of CVD risk factors (Wald Chi-square test). All analyses accounted for the complex sampling design of HCHS/SOL and were adjusted for baseline age, height, and gender. There were 6,779 individuals who reported weights at 2-5 time points (25,687 observations). In the final four-group quadratic trajectory mixture model the two most common classes comprised individuals who on average maintained a normal weight in adulthood (18.5-24.9kg/m 2 ) but became overweight (≥25kg/m 2 ) in their 50s (class 1: 38% of sample), or who gained enough weight in early adulthood to become overweight by their 30s and obese (≥30kg/m 2 ) by their 50s (class 2: 43%). The other two classes were characterized by more severe weight gain across adulthood, but these classes were less common (class 3: 16%; class 4: 3%). Individuals grouped in class 3 on average had a normal weight in their 20s, gained weight steadily, and became obese by their 40s. Class 4 represented the most dynamic trajectory, wherein individuals began overweight in their 20s and gained >40kg by their late 50s when they began to lose weight. At baseline dyslipidemia, untreated diabetes (p<0.007), and physical inactivity (p=0.8) were most common in class 3. Class 4 was characterized at baseline by the poorest diet quality and highest prevalence of hypertension and treated diabetes (p<0.001). In summary our results indicate that although weight gain was ubiquitous, more than a third of participants aged 50-74 years belonged to a trajectory class that did not develop obesity during adulthood. Trajectories of extreme weight gain were associated with a higher burden of CVD risk factors. Future studies should investigate modifiable factors that influence trajectories and could serve as targets for public health interventions in CVD.