OBJECTIVES: The aim of the study was to estimate the association between self-esteem and subsequent self-rated health during college years, taking into account a wide range of potential confounders. DESIGN: Prospective longitudinal study. SETTING: The French i-Share cohort. PARTICIPANTS: The sample consisted of 1011 college students. PRIMARY AND SECONDARY OUTCOME MEASURES: The association between self-esteem and later self-rated health was evaluated using multivariate modelling.Data regarding self-rated health, global self-esteem and demographic, educational, social, behavioural, environmental and financial characteristics were collected through an internet-based questionnaire. RESULTS: The 1011 participants had a median age of 21.9 years and 79% (795/1011) were females. Self-rated health was assessed a median of 8 months after the self-esteem measurement. Twenty per cent of the students declared average to very poor health (203/1011). Students with higher levels of self-esteem were more likely to declare good or very good self-rated health (adjusted OR=1.40, 95% CI 1.15 to 1.72, p value=0.001). Other factors associated with good or very good self-rated health were low body mass index, a comfortable financial situation during childhood and three personality traits (low persistence and harm avoidance and high cooperativeness). CONCLUSIONS: This study offers novel findings on the impact of self-esteem on self-rated health among college students. Interventions targeting self-esteem should be experimented during university years in order to improve health outcomes.