A discriminant-function analysis of screening tests for excessive drinking and alcoholism

Area of Science:

• Psychiatry
• Clinical Chemistry
• Medical Diagnostics

Background:

• Previous studies indicated rapid interviews outperform laboratory tests for screening excessive drinking.
• The efficacy of screening methods for alcoholism requires further investigation, particularly regarding information overlap and cut-off points.

Purpose of the Study:

• To re-examine the separation of patients with excessive drinking patterns from normal drinkers using discriminant analysis.
• To identify the most effective screening methods for excessive drinking and alcoholism.
• To evaluate the impact of laboratory test cut-off points on screening sensitivity.

Main Methods:

• Discriminant analysis was employed to re-evaluate screening performance.
• Comparison of three rapid interviews (including Brief Michigan Alcoholism Screening Test and Reich interview) against nine laboratory tests.
• Assessment of glutamate dehydrogenase (GDH), gamma-glutamyl transpeptidase, and mean corpuscular volume performance.

Main Results:

• Discriminant analysis confirmed the high performance of rapid interviews, specifically the Reich interview.
• Glutamate dehydrogenase (GDH) emerged as the best-performing laboratory test, comparable in efficacy to rapid interviews for excessive drinkers.
• The combination of the Reich interview and GDH demonstrated 100% sensitivity for excessive drinking and alcoholism without compromising specificity.

Conclusions:

• Rapid interviews and GDH testing are highly effective for screening excessive drinking and alcoholism.
• The optimal screening strategy involves combining the Reich interview with GDH testing for maximum sensitivity and specificity.
• Arbitrary cut-off points for laboratory tests do not improve screening performance compared to using the full range of results.