Abstract
Objective: We examined psychopathological conditions accounting for the highest risk of role impairment among university students.
Method: Cross-sectional online survey of first-year students (60 universities, 10 countries) assessing role impairment due to emotional problems in previous 30 days with the 3-item emotional subscale of the VR-36. Cross-tabulations and Poisson regression examined associations between 12 psychopathological conditions (8 probable mental disorders and 4 SITB) and socio-demographics with "significant role impairment" ("most"/"all of the time" in 2+ VR-36 items). We used machine learning methods to predict probabilities (risk) of significant role impairment based on the 12 psychopathological conditions. Observed prevalence of significant role impairment was examined within and across the 20 population ventiles to assess risk concentration using sensitivity (SN) and positive predictive value (PPV).
Results: 43,990 students responded to the survey (median age = 19, IQR = 18-23). 27.1 % had significant role impairment due to emotional problems and 65.8 % had at least one psychopathological condition. Students with one or more of these conditions were more likely to have significant impairment than those without (RR = 3.9; 95 %CI: 3.6-4.2). In multivariable analyses, probable Depression and Bipolar disorders were the strongest correlates of significant impairment. Most (69.6 %) of the observed significant role impairment occurred among the roughly 35 % of respondents with highest predicted risk (PPV = 53.2 %).
Conclusion: Significant role impairment due to emotional problems is highly prevalent among university students. High risk of significant role impairment concentrates in one third of the students, those with several psychopathological conditions. Assessing these conditions should help identifying highest role impairment risk university students.