Hazardous alcohol use is a common and important public health problem. Men and women who drink more than 2 to 3 drinks a day or more than 4 to 5 drinks per occasion are at significant risk for accidents, injuries, family violence, medical complications, tobacco and drug addiction, and a variety of comorbidities.1,2 The prevalence of hazardous drinking varies from 10% to 25% depending on sex, age, and clinical setting.3,4 Primary care settings and emergency departments offer some of our best opportunities to screen the US population on a regular basis.
However, time, cost, provider skill level, and competing agendas are all barriers to incorporating alcohol screening into primary care. Providers are being asked to screen patients for an increasing number of medical, mental health, social, and family problems. Many health care systems require physicians to screen all patients for a variety of cancers, heart disease risk factors, family violence, tobacco use, seatbelt use, immunizations, occupational exposures, and other health risks.
Examples of Rapid, Simple Screening Tests
The 2 articles presented in this issue of JFP go a long way in our search to develop rapid simple screening tests for identifying hazardous alcohol use. Williams and Vinson5 conducted a study in the emergency department setting to assess the sensitivity and specificity of a single alcohol screening question. The authors administered the question, “When was the last time you had more than (4 drinks [women] or 5 drinks [men])?” in a sample of 2335 patients seen for an acute injury. Responses to this single question were compared with 2 criteria standards. The first criteria standard used a 29-day timeline follow-back calendar procedure to assess the frequency of hazardous drinking.6 The second standard used the Diagnostic Interview Schedule (DIS) to diagnose alcohol abuse or dependence.7
This single question detected 86% of all hazardous drinkers or persons meeting the Diagnostic and Statistical Manual of Mental Disorders, 4th edition, criteria for alcohol abuse or dependence. Interestingly, this question was positive for 22 of 27 patients with blood alcohol levels over 0.10 who were misclassified as nonhazardous drinkers by the 29-day calendar procedure and the DIS. The single question compares favorably to other screening tests, such as the Alcohol Use Disorder Inventory Test (AUDIT) and the CAGE questionnaire. If the findings of the study are replicated by other scientists and in primary care settings, this single question may be the answer to our quest to develop a rapid simple screening test to detect one of our patients’ most common and important health risks.
The second study reported in this issue compared the sensitivity and specificity of: (1) the AUDIT-3 (the first 3 quantity frequency questions of the AUDIT), (2) the AUDIT-C (the binge-drinking question contained in the AUDIT), and (3) the full AUDIT. Gordon and colleagues8 administered the 10-question AUDIT to 13,438 patients being seen for routine care at 12 primary care sites. The criteria standard was 3 quantity frequency questions similar to the AUDIT questions. The study did not use the timeline follow-back procedure or the DIS. The authors found that the AUDIT-C had similar psychometric properties to the full 10-question AUDIT. The single question, “How often do you have 6 or more drinks on one occasion?” had significantly lower sensitivity and specificity. This study suggests that asking patients about frequency, quantity, and binge drinking provides similar information to the full 10-question AUDIT.
Screening in Practice
These studies confirm an increasing amount of evidence that supports the use of quantity frequency and binge-drinking questions as the first line of inquiry into a patient’s current alcohol use. The alternative CAGE screening instrument is recommended as a second-line set of screening questions for patients who test positive on the basis of the quantity-frequency questions. The CAGE may be more useful as a primary screening instrument in patients who are ambivalent about revealing their alcohol use because of legal or other issues and in patients who are alcohol dependent. Other screening instruments such as the AUDIT or the Michigan Alcohol Screening Test (MAST) are more appropriate as assessment instruments rather than screening tools.
Routine systematic implementation of the single question developed by Williams and Vinson or the first 3 questions of the AUDIT require less than 60 seconds of a clinician’s time. As with tobacco use, alcohol use questions can be administered by a nurse assistant or other clinic staff support person. Incorporation of these questions into routine health questionnaires and annual health surveys may be another strategy for systematically screening all patients for hazardous drinking. Screening and conducting a brief intervention in persons who screen positive will have a significant impact on reducing alcohol-related harm. 9