Background and objective: According to the high and increasing prevalence of internet addiction, and the fact that little research has been done on the predictors of internet addiction in Iran, the purpose of this study was to examine the role of predictor variables of mental health and personality subscales in internet addiction of students in medical and non-medical universities of Sanandaj in 2014. Methods: The Method of this research was correlation and the statistical population were all of medical and non-medical students of Sanandaj Universities in 2014. In this study, 250 students (125 female and 125 male), were randomly selected, and completed the checklist of mental health symptoms, NEO personality questionnaire, and internet addiction questionnaire. Data were analyzed using Pearson Correlation, stepwise regression, and T test by SPSS software version 20. Results: The results indicated that obsession-compulsion, openness, consciousness, aggression and somatization subscales had predictor roles in internet addiction, and totally 51 percent of variances predicted the internet addiction (F=29.97; P=0.001). Conclusion: The internet addiction of university students is dependent upon their mental health and personality, and one can predict the internet addiction of students via subscales of mental health and personality.