University Student Mental Health Before and After COVID-19: Heterogeneous Symptom Profiles




Vincent, Katelin

Journal Title

Journal ISSN

Volume Title



Introduction: Research on the effects of the COVID-19 pandemic on mental health has produced conflicting findings. Several studies claim the effects of quarantines related to COVID-19 could be of substantial harm to mental health, while others claim that the public has shown substantial mental health resilience despite pandemic-related stress. The objective of this study is to identify psychiatric symptom patterns in University students as they relate to the implementation of COVID-19 quarantines. Method: The U.S. college-based Stimulant Norms and Prevalence (SNAP) online survey was distributed across seven universities preceding and succeeding university COVID-19 closure announcements (CCAs) to identify mental health profiles from before and after COVID-19. We evaluated these survey data with unsupervised machine learning, through the use of latent profile analysis. Domains considered included self-report measures of depression, anxiety, sleep, anger, marijuana use, marijuana use consequences, binge drinking, alcohol consumption, and alcohol consumption consequences. Results: We found eight mental health profiles from the pre-CCA data, and the same eight mental health profiles emerged from in post-CCA data. For both sets of profiles, the trends for the profiles mirrored each other, with only small differences in the percentage of the sample that was associated with each profile. Among the eight profiles identified, three profiles consisted of lower severity in psychopathology, three profiles consisted of moderate-level psychopathology, and the remaining two were categorized as higher-severity psychopathology. Alcohol and marijuana consumption varied across these profiles. Conclusion: The data preceding and succeeding the COVID-19 pandemic was remarkably stable, with the same eight profiles emerging pre- and post-CCA, and the proportion of participants in each profile differed by less than 5% when comparing pre- and post-CCA profiles. Results from this study can contribute to the development of efficient quarantine treatment plans aimed to treat mental health and substance abuse.



COVID-19, heterogeneous symptom profiles, unsupervised machine learning, latent profile analysis, Honors College


Vincent, K. A. (2022). University student mental health before and after COVID-19: Heterogeneous symptom profiles. Honors College, Texas State University.


Rights Holder

Rights License

Rights URI