Purpose: The present study examined the prospective association of neuroticism, extraversion and psychoticism with risk of hospital diagnosed mental disorder, examining intelligence as a potential confounder of this association. Methods: A total of 1118 Danish men and women completed the Eysenck personality questionnaire at the mean age of 27 years. Information on psychiatric diagnoses was obtained by linking the study population to the national Danish psychiatric registers, and risk of diagnoses associated with each personality trait was examined using multiple Cox regression in models including the three personality traits unadjusted and adjusted for intelligence. Participants with diagnosis from a psychiatric department prior to the personality assessment were excluded. Results: In total, 122 participants were diagnosed with a mental disorder during follow-up. Neuroticism significantly predicted risk of anxiety-, adjustment-, personality- and alcohol and substance abuse diagnoses. Extraversion did not significantly predict any diagnosis type, while psychoticism predicted a combined category of mood and anxiety diagnoses. Despite intelligence being a significant predictor of the majority of the included diagnoses, adjusting for intelligence did not substantially influence any trait-disorder associations. Conclusion: The results confirm high neuroticism as a prospective vulnerability factor for mental disorder and indicate high psychoticism to be a potential risk factor for mood and anxiety disorders. These associations are not confounded by intelligence.