|
Alcohol Use Disorders Identification Test consumption questions (AUDIT-C)
Please note that this section is an archive (last updated in June 2006). [disclaimer]
Sections:
Overview |
Instrument Reviews |
Construct Overviews |
Book Compendium Reviews |
Internet Site Reviews
|
|
|
|
Practical Information
|
Instrument Name:
|
Alcohol Use Disorders Identification Test consumption questions (AUDIT-C)
|
Instrument Description:
|
The AUDIT, developed by the World Health Organization, is a 10-item screener used to identify hazardous and harmful alcohol use in adults. The intention was to develop an instrument that could not only be used to detect alcoholism, but also problem drinking at an earlier stage, and thus in a presumably less severe form. The AUDIT-C, the subject of this review, contains the first three items from the first AUDIT subscale. The AUDIT-C measures the quantity and frequency of alcohol consumption and may be used to screen for active alcohol abuse and dependence. (Ref: 2) Another name for the AUDIT-C is QF-AUDIT. (Ref: 4)
|
Price:
|
Free (Ref: 6). Available: Ref 1 Appendix and Ref 2 “Measures” section.
|
Administration Time:
|
No information was found for AUDIT-C. However, the full AUDIT takes only 2 mins. (Ref: 5)
|
Publication Year:
|
1998
|
Item Readability:
|
The measure has a Flesch-Kincaid Grade Level of 6.1, which means that an individual reading at a 6th grade level should understand the items in the scale. There are 12.3 words per item and language is simple.
|
Scale Format:
|
Items are answered using a 5-point likert-type scale. (Ref: 6)Selections such as ‘never’ receive a value of zero and selections such as ‘daily,’ or the equivalent, receive a value of four. (Ref: 2)
|
Administration Technique:
|
AUDIT-C is a self-administered scale, but has also been administered over the telephone and on computer. (Ref: 2,5-6)
|
Scoring and Interpretation:
|
Scores range between the values of 0 and 12. Lower scores indicate an individual who does not consume large amounts of alcohol. Males have demonstrated higher AUDIT-C scores than females (t=22.70; df=3321.82, p<0.001). (Ref: 6)
Researchers used an AUDIT-C cutoff score of three to determine the sensitivity and specificity of screening patients with alcohol abuse, dependence or heavy drinking. AUDIT-C had a sensitivity of 90% when screening alcohol abuse or dependence, and 98% when screening patients for heavy drinking. (Ref: 2) Specificity was less, 60%. (Ref: 2) When increasing the specificity to 72%, the sensitivity decreases to 86% for the aforementioned groups. (Ref: 2)
|
Forms:
|
The AUDIT was translated into German and used in a variety of countries. (Ref: 1,6) However, no information was found regarding the psychometric characteristics of the other language versions of the AUDIT-C. AUDIT 3 is a shorter version of the AUDIT and AUDIT-C. It contains one item, adopted from the original form, that gauges the amount of drinks an individual consumes in a single occasion. This item may also be used as a brief screening test. (Ref: 2)
|
|
|
Research Contacts
|
Instrument Developers:
|
A collaborative group of researchers developed the AUDIT/AUDIT-C for theWorld Health Organization (WHO): JB Saunders, OG Aasland, TF Babor, JR De la Fuente, M Grant.
|
Instrument Development Location:
|
WHO, Avenue Appia 20, 1211 Geneva 27, Switzerland. Collaborating investigators came from Geneva, Australia, Bulgaria, Kenya, Mexico, Norway, and USA.
|
Instrument Developer Email:
|
publications@who.int
|
Instrument Developer Website:
|
www.who.int
|
|
|
Annotated Bibliography
|
1. Saunders JB, Aasland OG, Babor TF, de la Fuente JR, Grant M. Development of the Alcohol Use Disorders Identification Test (AUDIT): WHO Collaborative Project on Early Detection of Persons with Harmful Alcohol Consuption --II. Addiction 1993 Jun;88(6):791-804. [PMID:8329970]
|
Purpose:
To develop a measure that would screen for a broad range of drinking concerns and for excessive alcohol consumption, the AUDIT.
|
|
|
Sample:
Of the 1888 individuals that participated in the study, 36% were classified as non-drinkers, 48% were labeled as drinking patients and the remainder was assessed as alcoholics. No information was found regarding mean age and ethnicity.
|
|
|
Methods:
Patients were recruited from primary care healthcare settings located in one of six countries: Australia, Bulgaria, Kenya, Mexico, Norway and USA. Recruitment was based on a structured interview that labeled patients as nondrinkers, drinkers or alcoholics. A trained interviewer interviewed each eligible participant. A 150-item schedule was used to extract needed information for the study.
|
|
|
Implications:
Researchers mention that the first alcohol screening tests had the objective of identifying alcoholics. These screening tests were successful in the identification process, with sensitivities exceeding 95%. However, the screening instruments were not as successful in identifying individuals who consume hazardous or harmful amounts of alcohol. The AUDIT was developed to fulfill this task. In this study, the AUDIT was administered to a culturally diverse population, and the developers encourage incorporating diverse cultures in studies. Lastly, the developers suggest that administering the AUDIT as it exists or as a part of a large questionnaire requires exploration.
|
2. Bush K, Kivlahan DR, McDonell MB, Fihn SD, Bradley KA, for the Ambulatory Care Quality Improvement Project (ACQUIP). Arch Intern Med 1998 Sep 14;158:1789-95. [PMID:9738608]
|
|
Purpose:
To evaluate the performance of the AUDIT-C in assessing the typical quantity and frequency of drinking.
|
|
|
Sample:
Participants (n=243) were all males and 89% of those with available data were white; 39% were 60-69 years and 44% were greater than 70 years; most had multiple medical problems.
|
|
|
Methods:
General medical patients were recruited from three VA medical centers (Washington, Vermont, Massachusetts). Individuals were excluded if they resided in a nursing home, participated in a conflicting study, did not have a telephone, did not answer calls for two weeks, were too ill or deaf or were women. Measures were administered to participants by a telephone interview or mail. These measures included the Health History Questionnaire (HHQ) (mailed) and the Drinking Practices Questionnaire (DPQ) (telephone administered). The DPQ consisted of the AUDIT, the AUDIT-C, a retrospective drinking diary, questions addressing previous provider advice, and questions concerning the individual’s want to change.
|
|
|
Implications:
Researchers reported that the AUDIT-C had higher sensitivity and specificity for heavy drinking than for alcohol abuse and dependence. Researchers concluded that the AUDIT-C better assesses heavy drinking than the AUDIT. However, the AUDIT performed better in assessing alcohol abuse and dependence when compared to the AUDIT-C. The AUDIT-C performed better that the CAGE in assessing alcohol abuse and dependence.
|
3. Reid MC, Voynick IM, Peduzzi P, Diellin DA, Tinetti ME, Concato J. Alcohol Exposure and Health Services Utilization in Older Veterans. J Clin Epidem 2000;53:87-93. [PMID:10693908]
|
|
Purpose:
To investigate if the increased use of alcohol results in the greater utilization of healthcare services among older veterans.
|
|
|
Sample:
Participants (n=129) were divided into three distinct groups, with 43 individuals per group. The abstainers group had a mean age of 70 years, 98% were male and 83% were Caucasian. The social drinking group had a mean age of 71 years, 98% were male and 93% were Caucasian. The problem drinking group had a mean age of 70 years, 100% were male and 91% were Caucasian. All individuals in the study were veterans.
|
|
|
Methods:
Participants were selected based on results of a 1993 study that screened for alcohol use. The 1993 study recruited civilians from the General Internal Medicine Center and veterans from the West Haven VA Medical Center. Patients were given the CAGE and the AUDIT-C. Forty-five veterans were labeled as problem drinkers, based on the CAGE results in 1993, and medical charts were located for 43 individuals. The current study selected the 43 veterans with charts and selected 86 other individuals above 65 years of age to analyze. Forty-three were placed in the social drinkers category and 43 were placed in the abstainers group (those that were labeled as such by the CAGE). Medical records were reviewed and baseline characteristics were assessed with the Charlson Index.
|
|
|
Implications:
Researchers did not find an association between alcohol exposure and the utilization of outpatient services by veterans.
|
4. Gordon AJ, Maisto SA, McNeil M, Kraemer KL, Conigliaro RL, Kelley ME, Conigliaro J. Three Questions Can Detect Hazardous Drinkers. Jour of Family Practice 2001 Apr;50(4).[PMID:11300982]
|
|
Purpose:
To determine the performance of the AUDIT, AUDIT-C and AUDIT-3 in assessing heavy drinkers in the primary care setting.
|
|
|
Sample:
Participants (n=12,867) were 53% male. Ages in years varied with a minimum of 21 years. The larger percentage of patients fell between the ranges of 41-50 years (22.6%) and 61-75 years (25.2%). The demographic variables associated with the respondents were as follows: 58.6% had training beyond high school, 80% were white, 59.6% were married and 52.7% were employed.
|
|
|
Methods:
Individuals were recruited from the waiting rooms of 12 primary care settings. These included a Veterans Affairs Medical Center, university-affiliated clinics, HMO clinics and private practice family medicine clinics. Individuals received an eight page self-administered questionnaire, which included questions concerning lifestyle habits. The questionnaire included the AUDIT instrument to assess drinking behaviors.
|
|
|
Implications:
Researchers concluded that the AUDIT and the AUDIT-C were better screens for hazardous drinkers than the AUDIT-3. For this particular study, the AUDIT, AUDIT-C and AUDIT-3 were sensitive in screening for hazardous drinking with cut-off scores of greater than of equal to eight, three and one respectively.
|
5. Reinert DF, Allen JP. The Alcohol Use Disorders Identification Test (AUDIT): a review of recent research. Alcohol Clin Exp Res 2002 Feb;26(2):272-9. [PMID:11964568]
|
|
Purpose:
To compare the properties of AUDIT to other measures that measure similar dimensions.
|
|
|
Sample:
N/A
|
|
|
Methods:
A literature review was conducted to retrieve publications associated with the AUDIT, its short forms and other comparable measures.
|
|
|
Implications:
The authors mentioned that the AUDIT performed at a level similar to or better than instruments measuring the same concept. The unique aspect of the AUDIT is that it not only focuses on alcohol dependence, but on risky drinking and the current behavior of the individual as well. The shorter form, the AUDIT-C is believed to be a brief and accurate screening tool, but more psychometric information is needed to make a full assessment.
|
6. Rumpf H-J, Hapke U, Meyer C, John U. Screening for Alcohol Use Disorders and At-Risk Drinking in the General Population: Psychometric Performance of Three Questionnaires. Alcohol & Alcoholism 2002; 37(3):261-68.[PMID:12003915]
|
|
Purpose:
To compare the performances of the AUDIT, the AUDIT-C and the LAST and determine appropriate cut-off points for each measure. Researchers also assessed the age and gender effects and the psychometric properties of each instrument and to test if sensitivity and internal consistency varied among hospital and general practice subsamples.
|
|
|
Sample:
German participants (n=3551) were 50.8% male and had a mean age of 41.2 years. The majority of the sample was married (59.6%) and 45% had at least 9 years of education.
|
|
|
Methods:
Individuals were recruited from a north German city. Eligibility criteria included individuals with a German nationality who had consumed alcohol within the last 12 months. Questionnaires were administered to participants by 56 trained interviewers.
|
|
|
Implications:
Researchers found that the validity of the scales differ when used in the general population. It was reported that the scales are less reliable in the general population compared to clinical settings. The authors suggest ways to improve the accuracy of screening measures. First, the use of more than one similar measure may enhance accuracy. Second, the screening measures may need to use simple criteria. Lastly, modified screeners may be useful when evaluating specific settings such as emergency room, general population or general hospital.
|
|
|
top
|
Factors and Norms
|
Factor Analysis Work:
|
No information found for specific factor analysis of the AUDIT-C. However, a review of the literature shows that six studies have examined the structure of the full AUDIT and all suggest that items one through three (AUDIT-C) load on one factor. (Ref: 5)
|
Normative Information Availability:
|
No information found
|
|
|
Reliability Evidence
|
Test-retest:
|
Using a three-month interval and a cut-off score of 3, the AUDIT-C had test-retest coefficients ranging from 0.65 to 0.85. (Ref: 5)
|
Inter-rater:
|
No information found
|
Internal Consistency:
|
The AUDIT-C reported a Cronbach’s alpha coefficient of 0.56. Corrected item-total correlations ranged from 0.30 (item 1) to 0.52 (item 3). (Ref: 6) Alpha for a subsample of patients with a hospital admission in the previous 12 months (vs. a general practice visit) was 0.66. (Ref: 6)
|
Alternate Forms:
|
No information found
|
|
|
Validity Evidence
|
Construct/ Convergent/ Discriminant:
|
No information found
|
Criterion-related/ Concurrent/ Predictive:
|
In one study, the AUDIT-C (AUROC=0.891) performed better than the AUDIT (AUROC=0.881) when measuring heavy drinking (p=0.03). However, the AUDIT had higher AUROC (0.811 VS. 0.786) when measuring active alcohol abuse and/or dependence. (Ref: 2) When determining if an individual is a heavy drinker or abuses alcohol, the AUDIT-C and AUDIT have similar AUROC results (p= 0.83). (Ref: 2) In a separate study, the AUDIT-C was compared to the AUDIT (the criterion measure of hazardous drinking) and AUDIT-3 by using the area under the receiver operating characteristic curve (AUROC) analysis. Researchers reported that the three scales were significantly more discriminating than the line of identity (AUROC= 0.5) when comparing heavy drinkers versus non-heavy drinkers. (Ref: 4) Furthermore, AUDIT-C AUROC was significantly different from the AUDIT AUROC (z=2.69, p=0.004) and the AUDIT-3 AUROC (z=12.69, p<0.001). (Ref: 4)
In another study, the authors assessed concurrent validity by comparing the AUDIT-C with other similar measures and determining the differences in sensitivity. (Ref: 6) A non-parametric McNemar test and ROC comparison were used. Cut-off scores of 3 and 4 showed low specificity for the AUDIT-C. Results were specificities of 0.40 and 0.62 and sensitivities of 0.95 and 0.83 respectively. (Ref: 6) The AUDIT and AUDIT-C yielded similar results for at-risk drinking (X2=0.61, df=1, p=0.44), alcohol dependence (X2=3.15, df=1, p=0.08) and alcohol misuse (X2=0.37, df=1, p=0.54). (Ref: 6)
|
Content:
|
No information found
|
Responsiveness Evidence:
|
No information found
|
Scale Application in VA Populations:
|
Yes. (Ref: 2-3)
|
Scale Application in non-VA Populations:
|
Yes. (Ref: 1-2,4-6)
|
|
|
Comments
|
The AUDIT-C is a briefer version of the AUDIT, containing only three items that measure quantity and frequency of alcohol consumption. It may be used to screen for active alcohol abuse and dependence. The AUDIT-C is a self-report scale that may also be administered via telephone or computer.
Overall Usefulness for a Certain Population: The AUDIT-C has been successfully used to identify alcohol use among veterans and nonveterans, females and males, in both clinical and general settings, among culturally diverse populations.
Advantages: The AUDIT-C is extremely brief, comes from a rich developmental history, and is easily scored.
Disadvantages: Alpha reliability is relatively low, probably due to the low number of items relative to the magnitude of the item-total correlations, and more validity evidence is required. Potentially more problematic, is the relatively low specificity when using the recommended cut-score of three. Researchers interested in using the AUDIT-C as a screening tool should be aware that they may have a large false positive rate. Further, this may get worse with time. Recent, widely publicized, research on alcohol consumption has suggested that high frequency (several times a week), low-level (1-2 drinks at a time) consumption of alcohol may have noticeable health benefits. Someone who consumed alcohol at this rate would very likely screen positive for alcohol abuse or dependence using the currently recommended cut-score.
Recommendation: While the AUDIT has been well-studied and its psychometric properties evaluated, more psychometric work is needed for the AUDIT-C. Available internal consistency evidence is somewhat weak, and indicates that reliability should be further examined in different populations, both at the individual and group comparison level. Further work is also needed to determine the best cut-scores in a variety of different populations.
|
|
|
|
