Psychosocial and Behavioral Predictors of Energy Intake Plausibility and Weight Loss in Overweight Perimenopausal Women

Abstract

The analyses in this dissertation were designed to 1) extend the knowledge of characteristics associated with and predictive of energy intake plausibility (under or overreported energy intake), and 2) extend previous research in a sub-sample of this study population of baseline short-term weight loss predictors to evaluate within the full sample whether baseline psychosocial, behavioral and dietary predictors of weight loss varied by energy intake plausibility. Subjects were 155 overweight or obese perimenopausal women participating in a 4mo lifestyle weight loss program. Based on self-reported intake from 3-d dietary records, women were categorized as energy underreporters (n=71), accurate energy reporters (n=27), or energy overreporters (n=57), using the cut-off values for energy plausibility defined by Goldberg. All subjects completed a comprehensive behavioral and psychosocial battery assessing diet and weight history, life status, weight loss readiness, psychology, eating behavior, physical activity, and self-image. Results from logistic regression models showed that y of education, weight loss aspirations, exercise perceived competence, social support to exercise, and measures of body image were the best predictors of energy underreporting. Dietary carbohydrate and fat intake, health related quality of life, and profile of mood states (anger) were the best predictors of energy overreporting. Baseline predictors of successful weight loss did vary by energy plausibility group, with unique predictors for energy underreporters including fewer previous dieting attempts and exercise perceived obstacles, and energy overreporters including higher TEE, more negative mood status and higher perceived hunger. Overall, more successful weight loss was also associated with higher baseline fruit and vegetable intake. Validation of these findings will help lead to establish factors to account or adjust for bias from energy misreporting, reduce health or disease risk underestimation and improve understanding of nutrition, health and disease relationships. Further, identification of successful weight loss predictors unique to energy under- and overreporters will enhance weight loss profiling and tailoring of interventions to optimize success.