It is unclear whether recent increases in mental health issues reported by students are comparable between young people in and out of higher education. Furthermore, university (non)attendance may combine with individuals’ other social identities to lead to mental health inequalities. Using quantitative longitudinal data from the national cohort study, Next Steps, we conducted multilevel analyses within an intersectional framework. We examined predictors of mental health during emerging adulthood with combinations of the following characteristics: adolescent mental health (measured at age 16), university attendance, sex, socioeconomic class, sexuality, and ethnicity. The findings indicated that respondents who had symptoms of mental ill health during adolescence were more likely to attend university. At age 19, respondents who had symptoms of adolescent mental ill health were more likely to report having no close friends and lower life satisfaction. Individuals attending university were more likely to report having one or more close friends, and greater life satisfaction, independently of their mental health during adolescence. Those who had symptoms of mental ill health during adolescence were more likely to have poorer mental health outcomes at age 25. The opposite pattern was found for those who had been to university, independently of their adolescent mental health. Mixed patterns of predictions were found for respondents’ other social identities. We found that intersectional effects were additive, suggesting that characteristics are layered and independent in predicting mental health outcomes (i.e. they do not amplify each other). This has potential implications for how interventions should be targeted towards particular intersectional subgroups.