Categorization of Potential Classes Using the University Counseling Center Service Application
10.4306/jknpa.2022.61.3.170
- Author:
Sangil LEE
1
;
Kwanglo LEE
;
Seonmi LEE
;
Yugyeung LEE
;
Myungsung KIM
;
Dooyoung JUNG
Author Information
1. Graduate School of Artificial Intelligence, Ulsan National Institute of Science and Technology, Ulsan, Korea
- Publication Type:ORIGINAL ARTICLE
- From:Journal of Korean Neuropsychiatric Association
2022;61(3):170-177
- CountryRepublic of Korea
- Language:English
-
Abstract:
Objectives:The purpose of this study was to classify the users of a university counseling center and define their characteristics using the information from the service application written by them as one of the approaches to improve the efficiency of the center.
Methods:The participants were a total of 300 users of the university counseling and affiliated psychiatry center in the local science and technology institute. The service application form used in this study contained questions on personal information, overall satisfaction with university life and interpersonal relationships, problem areas and symptoms, depression, anxiety, sleep, family history, suicidal thoughts, experiences of suicide/self-harm attempts, and a functional impairment questionnaire. First, a latent class analysis was performed to classify the layers and define them. Also, a tree-based model was used to confirm the characteristics of the service applications of the suicide/self-harm attempt experience group.
Results:A total of 5 layers were found to be the most suitable fit. They were named the ‘health concern type,’ ‘relation problem type,’ ‘depression-sleep problem combined type,’ ‘complex problem type,’ and ‘sleep problem type.’ The variable that best discriminates the suicide/selfharm attempt experience group was question 5 of the Patient Health Questionnaire (PHQ-9). The frequency of selection in the impulsivity domain among the problem types and question 9 of the PHQ-9 was followed.
Conclusion:The results of this study indicate that five types of discriminatory interventions are possible based on the characteristics of the service application. The application can thus be used as basic data for increasing service efficiency in university counseling.
