Identifying Distinct Risk Subgroups of Mental Health Problems Among College Students: A Latent Profile Analysis

Suicide, homicide and other violent events have become increasingly prevalent among college students in China. Universal mental health assessment among freshman provides an important opportunity to early diagnosis and treatment of high-risk individuals. In the past two decades the University Personality Inventory and the Symptom Checklist 90 (SCL-90; Hoffmann et.al., 1978; Huang, 2009) have commonly been used to screen for general mental health problems among Chinese college students (Wang , 2009).However, the traditional demarcation mark method is doubted by many researchers because of the false positive rate. The present study seeks to explore a new method to identify students at high risk based on the latent profile analysis (LPA).

The sample included 12,718 college students (Mage =19.1 years , 64% female) from all first-year students at a university in China, conducted in 2012 and 2013. Participants completed two mental health instruments, the UPI and SCL-90.UPI included 56 symptom items assess whether an individual usually experienced the described symptom (e.g., “Lack of interest in anything”) during the past year. Scores on the primary subscales of UPI were used as indicators in the latent profile analysis. The result of the latent profile analysis was compared with the traditional classification to test the reasonableness of UPI diagnostic screening standard. The mental health status of 644 students assessed for high risk, general and health by psychologist, counselor and peers. Using evaluation results and the 90 Symptom checklist (SCL90) positive detection rate as the "golden standard" diagnostic accuracy，sensitivity and specificity were compared between the latent profile analysis and the traditional method.

Results:Student’s mental health problems can be divided into three subgroups: high risk groups (9.4%), mental confusion groups (17.7%), health groups (72.9%). The positive symptoms of mental health risk in high risk groups is 61.21%, far above mental confusion groups(38.28%) and mental health groups(8.36%). High risk groups is characterized by prominent severe symptoms (Z-score≥2.6SD). Mental confusion groups shows high risks with emotional and cognitive symptoms.The latent profile analysis improved sensitivity by 8.93%-35.26% and showed better diagnostic accuracy compared to the traditional demarcation method.

This study demonstrated that the latent profile analysis provides a more sensitive approach comparing with the traditional method to identifying college student at risk for severe mental health problems in order to prevent related issues such as suicide. The present findings are useful for developing effective intervention and psychological counseling programs for different risky subgroups of mental health problems.