J Wardle1, S Williamson, F Johnson, C Edwards. 1. Department of Epidemiology and Public Health, Cancer Research UK Health Behaviour Unit, University College London, London, UK. j.wardle@ucl.ac.uk
Abstract
OBJECTIVE: Community studies relating depression to obesity in adolescents have generated inconsistent results. It has been argued that the variability in findings is due to effect modification by demographic characteristics that vary across samples. The aim of this study was to test the hypothesis that the strength of the obesity-depression association is moderated by gender, ethnicity and socioeconomic status (SES). RESEARCH METHODS AND PROCEDURES: Data were from two large, school-based, studies of adolescent health and well-being (n=4320; n=1824). Students completed one of two measures of depressive symptoms (SDQ; CES-D) in school and were weighed and measured. Gender and ethnicity were self-reported and SES was indexed by residential neighbourhood characteristics or individual family deprivation. RESULTS: There was barely any association between obesity and depressive symptoms in either sample. There was also no evidence that obese participants who were female, white or from higher SES backgrounds were especially vulnerable to depressive symptoms. DISCUSSION: The results indicate that in community samples of adolescents, regardless of gender, SES or ethnicity, reports of depressive symptoms are not significantly higher in obese than normal-weight groups. The results are discussed in terms of obese adolescents' resilience in the face of multiple social adversities.
OBJECTIVE: Community studies relating depression to obesity in adolescents have generated inconsistent results. It has been argued that the variability in findings is due to effect modification by demographic characteristics that vary across samples. The aim of this study was to test the hypothesis that the strength of the obesity-depression association is moderated by gender, ethnicity and socioeconomic status (SES). RESEARCH METHODS AND PROCEDURES: Data were from two large, school-based, studies of adolescent health and well-being (n=4320; n=1824). Students completed one of two measures of depressive symptoms (SDQ; CES-D) in school and were weighed and measured. Gender and ethnicity were self-reported and SES was indexed by residential neighbourhood characteristics or individual family deprivation. RESULTS: There was barely any association between obesity and depressive symptoms in either sample. There was also no evidence that obeseparticipants who were female, white or from higher SES backgrounds were especially vulnerable to depressive symptoms. DISCUSSION: The results indicate that in community samples of adolescents, regardless of gender, SES or ethnicity, reports of depressive symptoms are not significantly higher in obese than normal-weight groups. The results are discussed in terms of obese adolescents' resilience in the face of multiple social adversities.