Related Experiment Video
Updated: Aug 6, 2026
12:18
A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
Published on: January 11, 2020
An Alternative Approach for Detecting Problematic Alcohol Use: Developing the Student Alcohol Risk Assessment
Şükrü Alperen Korkmaz1,2, Pınar Mutlu1,3, Sibel Oymak1,4
1Çanakkale Onsekiz Mart University Center for Combating Addiction Application and Research, Çanakkale, Türkiye.
International Journal of Methods in Psychiatric Research
|July 20, 2026
Summary
A new AI-supported tool (SARAS-15) effectively assesses risky alcohol use in university students, identifying low, medium, and high-risk levels. This brief, 15-item scale shows high accuracy and reliability, complementing existing screening methods.
Area of Science:
- Psychology
- Artificial Intelligence
- Public Health
Background:
- Risky alcohol use is prevalent among university students, negatively impacting health.
- Existing screening tools overlook crucial psychosocial factors.
- There's a need for a brief, student-focused assessment tool incorporating AI.
Purpose of the Study:
- To develop an AI-supported, brief tool for assessing alcohol-related risk in university students.
- To evaluate the preliminary reliability and validity of this new tool.
- To create a student-focused instrument addressing behavioral and psychosocial aspects.
Main Methods:
- A 15-item risk score model (SARAS-15) was developed using AI from a 59-item pool.
- Machine learning models (Logistic Regression, Random Forest) evaluated model performance.
- Internal consistency and correlations with AUDIT, RAPS4-QF, and CAGE were analyzed.
Main Results:
- The logistic regression model achieved 93.5% accuracy, with excellent ROC discrimination (AUCs 0.88-0.96).
- The SARAS-15 demonstrated strong internal consistency (Cronbach's alpha = 0.811).
- High correlations were found with established screening instruments (AUDIT r=0.861, RAPS4-QF r=0.793, CAGE r=0.631).
Conclusions:
- The AI-supported SARAS-15 is a valid and reliable tool for assessing risky alcohol use in students.
- The 15-item scale accurately identifies three risk levels and aligns with traditional measures.
- Its multi-domain structure makes it valuable for early risk identification and intervention planning.