Abstract P294: Association between Anxiety Levels and Weight Change in the Multi-Ethnic Study of Atherosclerosis
Background: Psychological factors may influence weight gain. This study assessed the association between anxiety and weight change in a multi-ethnic cohort followed for approximately 5 years.
Hypothesis: Anxiety level, measured by the Spielberger State-Trait Personality Inventory (STPI), is associated with weight change.
Methods: The study population consisted of participants of the Multi-Ethnic Study of Atherosclerosis (n=5,798), age range 45-84 at baseline. Weight was measured at baseline and at subsequent follow-up exams with an average of 1.6 years between exams. A locally weighted scatter plot revealed a non-linear association between anxiety and weight change. STPI scores were then divided into sex-specific quartiles (QANX) based on the STPI scoring manual. The relationship between anxiety level and weight change was examined using a mixed-effect model with weight as the dependent variable, QANX and time as the independent variables, and adjusted for age, race, gender, marital status, education level, income level, and MESA site. Time entered the model as a random effect, whereas all other variables were entered as fixed-effects.
Results: Overall, 52.5% were female and mean age at baseline was 61.7 years. Mean baseline weight was 78.9 kg (range 32.5 to 146.1 kg). The total average weight change was -0.55 kg for men and -0.56 kg for women over the 5 year period. Average annual change in weight (range) was -0.15 kg (-6.90 to 7.13 kg) for QANX 1, -0.11 kg (-5.81 to 7.27 kg) for QANX 2, -0.08 kg (-8.69 to 8.38 kg) for QANX 3, and -0.16 kg (-7.12 to 6.01 kg) for QANX 4. Mean baseline weight for those in Quartile 4 (most anxious) was 78.1 kg (range 38.4 to 146.1 kg) and mean weight change was -0.16 kg. Mean baseline weight and weight change for the bottom 3 QANX combined was 79. 1 kg (32.5 to 142.7 kg) and -0.12 kg, respectively. Individuals in the highest QANX on average weighed 1.63 kg less compared to individuals in the lower three QANX after adjustment for all covariates (95% CI -2.64, -0.61). Interactions terms between anxiety and race, gender, education, and income level on average weight did not reach statistical significance (P >0.1). No significant association was noted between anxiety quartiles and time (0.007 kg/yr; 95%CI -0.09, 0.11) after confounder adjustment.
Conclusions: Among adults age 45-84, high anxiety defined by being in the upper STPI quartile was associated with a significantly lower average weight in the MESA cohort but it was not associated with weight change. It is possible that that anxiety may influence weight change at a younger age.
- © 2013 by American Heart Association, Inc.