Abstract
Objectives: The purpose of this study is to demonstrate how traditional risky drinking indicators, such as binge drinking, are inefficient when identifying college students who are in need of alcohol-related services. Methods: Using data collected from the 2010 National College Health Assessment II (NCHA-II), the sensitivity, specificity, positive predictive value and negative predictive value of binge drinking was analyzed relative to experiencing negative alcohol-related consequences in the previous 12 months. Results: Despite a considerable mathematical advantage, binge drinking demonstrated a sensitivity of 0.80 and a specificity of 0.69. Additionally, more than 56% of the sample who were identified as binge drinkers did not experience a major negative consequence in the previous 12 months. Conclusions: Risky drinking indicators constructed solely of frequency, quantity and duration items have a limited capacity to discriminate at-risk college drinkers from the general student body. New indicators comprised of novel drinking behaviors are necessary to properly identify students at high risk.