Gender Differences in Self-Rated Health among University Students in England, Wales and Northern Ireland: Do Confounding Variables Matter?

INTRODUCTION- We assessed gender differences in self-rated health (SRH) while considering physical health, health complaints, health service use, wider wellbeing, and health behaviours.METHODS- 3706 undergraduates at 7 Universities in the United Kingdom completed a self-administered questionnaire (2009–2008). Logistic regressions with excellent/very good SRH as dependent variable assessed the variables that explained the SRH sex difference. RESULTS- Females had more health complaints, illness periods, lower quality of life, more burdens, and took medication/s more often. The crude (unadjusted) odds ratio (OR) proposed that females were less likely to report excellent/very good SRH than males [OR 0.79, 95% CI 0.68-0.94]. Adjusting only for physical health and health service use, females’ OR increased considerably, and the association between female sex and SRH was no longer significant. Also, when adjusting only for wider well-being or when adjusting only for health behaviour, the negative association between females and SRH was no longer significant. Adjusting for all the variables simultaneously (physical health, health service use, wider well-being, health behaviours) resulted in considerable increase of females’ OR indicating now a positive association between female sex and SRH [OR 1.33, 95% CI 1.04-1.74]. CONCLUSION- Females’ lower SRH found in the crude analyses was confounded by their higher stress level, lower quality of life, lower physical activity and by more illnesses or health complaints when compared with males. Gender-related SRH research should control for many potential confounders to prevent overestimation of the gender effect. Health promotion programs should consider these factors when tackling gender health disparities.