Influences of gender and social class on adolescents' perceptions of health

E. Goodman, B. C. Amick, M. O. Rezendes, A. R. Tarlov, W. H. Rogers and J. Kagan
Division of Adolescent/Young Adult Medicine, Children's Hospital, Harvard Medical School, Boston, MA 02115, USA.

OBJECTIVE: To explore how gender and social class affect perceptions of health status among 16-year-old adolescents. DESIGN: Cross-sectional survey. SETTING: University psychology laboratory. PARTICIPANS: Fifty upper-middle-class and 48 working-class adolescents stratified by gender. MAIN OUTCOME MEASURES: The general health perceptions (GHP) scale of the Medical Outcomes Survey 36-Item Short Form Health Survey (SF-36). The other self-reported health status domains and 3 measures of different aspects of psychological well-being were included as covariates in analysis of variance models. RESULTS: Upper-middle-class females reported the lowest and upper-middle-class males the highest GHP (76.7 vs 88.4, P=.003). A multivariate regression model (adjusted R2=0.08) revealed significant gender (P=.03) differences in GHP, but not a social class effect, and an interaction effect between gender and class (P=.01). With addition of psychological well-being covariates (P<.001), gender remained significant (P=.04) and a significant portion of the interaction effect (P=.13) was explained. When the self-reported physical health status scales (P<.001) were added to the model (adjusted R2=0.51), gender remained significant (P=.03) and the interaction effect was partially explained (P=.07). CONCLUSIONS: Gender is a crucial factor in understanding the complex relationships between sociostructural inequalities and health differentials. These data suggest that psychological well-being and self-reported physical health status mediate the effects of gender and the gender and social class interaction in explaining variation in GHP. Contrary to the social class gradient hypothesis, upper-middle-class females reported the lowest GHP. These results suggest that the paradigms applicable to early childhood and adulthood may not be appropriate to understand the complex dynamics of adolescence.