Background/objective: The purpose of this study was to examine the longitudinal effects of body size over- and underestimation and other psychosocial factors on the risk for onset of overweight (OW) or obesity (OB) 1 year later among US adolescents.
Subjects/methods: Participants with non-missing height and weight were drawn from the first two waves of the National Longitudinal Study of Adolescent to Adult Health (n=13,568). Multinomial logistic regressions were conducted to assess longitudinal risk factors for OW and OB onset, controlling for baseline weight status, age, race/ethnicity, parent education and family structure. Analyses were stratified by sex.
Results: Compared with accurate body size perception, body size overestimation increased the relative risk (RR) of OW onset among women and men (RR=3.34, confidence interval (CI)=2.39-4.68; RR=6.01, CI=4.09-8.83, respectively, P<0.001) in fully adjusted models including body mass index z-scores. Body size underestimation decreased the RR of OW onset among women and men (RR= 0.08, CI=0.03-0.20; RR=0.13, CI=0.06-0.27, respectively) and OB onset (RR=0.05, CI=0.02-0.14; RR=0.19, CI=0.08-0.47, respectively, P<0.001 for all) in fully adjusted models. Dieting, extreme weight loss behaviors and skipping breakfast at Wave 1 increased the risk of OB onset by Wave 2.
Conclusions: Contrary to a common assumption, body size underestimation did not increase, but in fact decreased, the RR for the onset of OW and of OB among adolescents using a prospective longitudinal design. However, body size overestimation predicted onset of OW 1 year later. Body size self-concept and other psychosocial factors have an important role in risk for OW and for OB among both males and females during adolescence.