Compared with clinical impressions, systematic assessment methods substantially improve identification of adolescents likely to have AUDs.10
Screening for AUDs in adults traditionally relies on interview questions about alcohol-related problems organized to yield memorable mnemonics.1
Such tools may be readily incorporated into clinical practice. Two such tools have been examined in several studies: CAGE (Cut down, Annoyed, Guilty, Eye-opener) and CRAFFT (Care, Relax, Alone, Family, Friends, Trouble). CAGE, the most widely known of these screening methods, has been found to have relatively poor sensitivity and specificity in underage drinkers.1
begins with 3 questions about the presence or absence of alcohol, marijuana, or other drug use, then inquires about problems related to “alcohol or drugs” in 6 questions (). At a threshold score of 2 of 6, some studies have found that CRAFFT has acceptable sensitivity and specificity for identifying adolescents likely to have AUDs or other substance use disorders (eg, sensitivity, 0.80; specificity, 0.86; positive predictive value, 0.53; and negative predictive value, 0.96).10
However, other studies have found that CRAFFT has poor specificity (0.33 in the study by Cook et al12
and 0.44 in the study by Kelly et al13
). The advantages of CRAFFT include a broadened screening focus to the inclusion of “drugs,” the interview format, and simple response options (yes or no). Nonetheless, problem-focused screens do not eliminate the need for determining substance use patterns. Adolescents whose screening results are negative for problems may actually be engaging in high-risk drinking. A positive problem-focused screen needs to be followed by inquiries about substance consumption patterns and a diagnostic evaluation.
Alcohol and drug use screening questions from CRAFFT (Care, Relax, Alone, Family, Friends, Trouble). From the Center for Adolescent Substance Abuse Research, Children's Hospital Boston, with permission.
The 10-item Alcohol Use Disorders Identification Test (AUDIT)14
includes consumption (AUDIT-C; 3 items) and problem (7 items) subscales. AUDIT has consistently shown acceptable sensitivity and specificity in adults.15
AUDIT items may be presented by interview or questionnaire.16
The AUDIT takes approximately 2 minutes to administer and approximately 1 minute to score. This instrument is included in the National Institute on Alcohol Abuse and Alcoholism (NIAAA) clinician's guide.17
AUDIT has been tested in several adolescent samples. In a study of 415 adolescents aged 13 through 19 years who were seen in an emergency department,18
an AUDIT total score threshold of 3 yielded acceptable sensitivity (0.95) and specificity (0.75). Among 538 adolescents aged 14 through 18 years who were seen in a primary care clinic,19
an AUDIT total score threshold of 3 yielded acceptable sensitivity (0.88) and specificity (0.77). For adults, a score of 8 has typically been recommended, although some authors have suggested that a lower score should be adopted.15
Studies of older adolescents (ie, >18 years) have found that a threshold score of 8 has yielded acceptable results on both sensitivity and specificity.12,13
We recommend a threshold of 3 to maximize sensitivity of the scale for use in adolescents younger than 18 years. For those aged 18 through 20 years, a threshold of 8 is recommended.
Abbreviated methods assessing alcohol consumption levels may be effective for identifying adolescents likely to have AUDs. These methods are based on a minimum quantity of one standard drink (ie, 12 oz of beer, 8 oz of malt liquor, 5 oz of wine, and 1.5 oz of spirits). The 3 AUDIT-C items () and the 10-item AUDIT have shown similar sensitivities and specificities in adult samples.15
A single question asking the frequency of drinking episodes in the prior month with a threshold of 3 episodes was found to be 90% sensitive and 84% specific as a screen for adolescent AUDs.20
Among adults aged 18 through 20 years seen in an emergency department, AUDIT-C, CRAFFT, and the Diagnostic and Statistical Manual of Mental Disorders
(Fourth Edition, Text Revision) (DSM-IV-TR
) 2-item scale were effective screens.21
The 2-item screen required a positive response to one of 2 DSM-IV-TR
symptoms, those involving hazardous use (A2) or drinking in larger amounts or for a longer period than intended (D3; ). In this sample, the 2-item DSM-IV-TR
screen showed 88% sensitivity and 90% specificity. The AUDIT-C, drinking frequency, and DSM-IV-TR
2-item screens need to be more thoroughly studied in adolescent samples. In the absence of a clearly superior approach among these methods, acceptable screening methods for possible underage drinkers include the 10-item AUDIT, AUDIT-C, drinking 3 or more times in the past month, endorsement of DSM-IV-TR
items A2 or D3, and CRAFFT.
Alcohol Use Disorders Identification Test–Consumption (AUDIT-C) Questions, Responses, and Scoring