National Institute on Alcohol Abuse and Alcoholism No. 8 PH 285 April 1990
Screening for Alcoholism
It is estimated that at least 20 percent of adults who visit a physician have had an alcohol problem at one time (1). In a survey of patients admitted to an inpatient service, 12 to 30 percent screened positively for alcoholism (2). Yet, several recent studies indicate that physicians in various health care settings often do not recognize and treat alcoholism (3,2,4). These findings underscore the need for effective and accurate procedures that will enable clinicians to screen for alcoholism.
Alcohol screening identifies individuals in a patient population who have begun to develop or who are at risk for developing alcoholism. Although physicians customarily take a patient's medical history, routine use of a standard alcoholism-detection instrument is valuable because these instruments provide a structured, disciplined, and consistent means to detect individuals at risk
Two types of alcoholism-screening instruments are available. The first type includes self-report questionnaires and structured interviews; the second type includes clinical laboratory tests which can detect pathophysiology associated with excessive alcohol consumption . Both types of screening instruments should be valid (i.e., measure what the clinician or researcher is attempting to measure) and should yield reliable results (i.e., consistent across raters and time).
Sensitivity and specificity, key properties of every screening test, are related to validity. Sensitivity refers to the test's accuracy in identifying individuals with an alcohol problem (i.e., persons with the disease test as positive). Specificity refers to the test's effectiveness in identifying people who do not have an alcohol problem (i.e., persons without the disease test as negative).
Two issues should be addressed in the discussion of sensitivity and specificity. First, it is not possible to optimize both properties in a screening instrument. The likelihood of overidentifying alcohol abuse occurs with increased sensitivity; conversely, the possibility of missing people who have an alcohol problem heightens with increased specificity (5). Second, in determining the utility of a screening test, knowledge of the test's sensitivity and specificity is not sufficient. The prevalence of the particular condition in the screened population also must be taken into account (6). Sensitivity and specificity are functions of a screening test's cutoff score (a value that clinicians use to define a positive result from a screening instrument). If the base rate of a condition in a population is low, most of the cases identified by a sensitive test will be false positives (5).
In addition, it is important to note that no "gold standard" exists for evaluating the accuracy of screening instruments. Clinicians compare evidence for alcoholism from medical records and clinical examination with screening results to confirm the accuracy of a particular alcoholism-screening instrument.
The CAGE questionnaire, a mnemonic for attempts to cut down on drinking, annoyance with criticisms about drinking, guilt about drinking, and using alcohol as an eye-opener (7,8), is a self-report screening instrument that appears to be suited to a busy medical setting when there is limited time for patient interviews (9). The CAGE, which can be self-administered or conducted by a clinician, poses four overt yes-no questions and requires approximately 1 minute to complete. Bush and colleagues (10) used the CAGE to screen 518 patients in a community teaching hospital At a cutoff score of "2" (in this case, meaning two "yes" answers), the investigators found that the test correctly identified 75 percent of alcoholics (sensitivi ty) and 96 percent of nonalcoholics (specificity). For routine health screening, the test may identify individuals with alcohol problems that might have been missed otherwise (11).
The Michigan Alcoholism Screening Test (MAST) (12) is a formal 25-item questionnaire that requires approximately 25 minutes to complete. The MAST focuses on the consequences of problem drinking and on the subjects' own perceptions of their alcohol problems. Recent studies have reported that a cutoff score of "12" or "13" achieves balanced rates of false positives and false negatives (13,14). Two shortened forms of the MAST, a 13-item Short MAST (SMAST) (15) and a 10-item brief MAST (b-MAST) (16), have been constructed using items from the original test that are highly discriminating for alcoholism. A cutoff score of "3" is suggested for the SMAST; a cutoff score of "6" is suggested for the b-Mast.
The Self-Administered Alcoholism Screening Test (SMST) (17) is a 35-item questionnaire or interview with a yes-no format. A score of "10" or greater denotes probable alcoholism. Hurt and colleagues (18) evaluated the use of the test with patients undergoing general examinations and recommended the instrument as an adjunct to a physician interview and an examination. A more recent study reported that the original SAAST and an abbreviated nine-item version are useful for screening medical patients for alcoholism (19).
The Alcohol Dependence Scale (ADS) (20) is a self-report questionnaire designed to measure elements of the alcohol dependence syndrome described by Edwards and Gross (21). The test, which yields an index of severity of alcohol dependence, addresses core features of dependence, including an individual's compulsion to drink excessively, repetitive experiences of withdrawal symptoms, and loss of control overdrinking. Ross and colleagues (14) compared the merits of the ADS and the MAST as screening tools and determined that the specificity and sensitivity for the two tests were approximately equivalent.
Alcohol is the most widely used drug by young persons between the ages of 12 and 17 years (22). Routine screening, however, is relatively rare in pediatric practices (23). Because life problems for adolescents and adults differ, many screening instruments are inappropriate for younger individuals. The Adolescent Drinking Inventory (24) is a self-report instrument developed specifically to screen adolescents. The inventory's 25 questions focus on drinking-related loss of control as well as social, psychological, and physical symptoms of alcohol problems. Allen and colleagues (9) reported that the inventory correctly identified 88 percent of adolescents with alcohol problems and 82 percent of those without alcohol problems.
Heavy alcohol intake during pregnancy can have severe adverse effects on a developing fetus; yet, maternal drinking can be difficult to detect. Sokol and colleagues (25) developed the T-ACE questionnaire to identify pregnant women who consume quantities of alcohol that potentially can damage the fetus (defined in this study as daily intake of 1 ounce of absolute alcohol or greater). The questionnaire takes approximately 1 minute to complete and incorporates the C, A, and E items of the CAGE. The G item is replaced with a question that addresses alcohol tolerance (T). The investigators considered a woman tolerant if she needs more than two drinks to feel the effects of alcohol. Perhaps because many individuals do not understand the implications of tolerance, the tolerance question did not appear to trigger psychological denial. In study of 971 pregnant women, the T-ACE correctly identified 69 percent of those women who consumed more than an ounce of alcohol per day.
Clinical laboratory procedures, the second type of sreening test, frequently are used to corroborate results of physicians' interviews; and of self-administered questionnaires. Biochemical markers of heavy alcohol consumption can provide objective evidence of problem drinking, especially in patients who deny any drinking problem. However, the sensitivities and specificities of these biological laboratory markers can be modified by nonalcoholic liver iniury, by drug use, and by metabolic disorders or individual metabolic differences.
Several clinical tests may be useful in detecting harmful alcohol use. Increased activity of serum gamma-glutamyltransferase (GGT) is a relatively sensitive index of liver damage in alcoholics and heavy drinkers (26,27). However, this test lacks diagnostic specificity because all types of liver damage and a variety of diseases may cause elevated serum activity of this enzyme. Results may be more discriminating when interpreted in conjunction with the measure of mean corpuscular volume (MCV) (28). MCV, an index of red blood cell size, increases with excessive alcohol intake. Although MCV has a high correlation with alcohol consumption, the index, alone, is not a useful screening marker (29). The liver enzyme aspartate aminotransferase (AST) can be a useful marker for alcohol abuse: The ratio of levels of mitochondrial AST to total AST has been found effective in differentiating alcoholics from other patients and in detecting chronic excessive drinking (30,31,32).
It is important to note that self-report interviews and questionnaires have greater sensitivity and specificity than routine blood tests for biochemical markers (29). Laboratory tests may be used most successfully in conjunction with self-report instruments to enhance objectivity (32,33,34).
Screening for Alcoholism A Commentary by
NIAAA Director Enoch Gordis, M.D.
Untreated alcoholism often results in severe or fatal outcomes and can be the underlying cause of other illnesses. Therefore, screening all patients for alcohol problems--particularly in primary health care settings--is a medical necessity.
Structured interviews and self-report instruments are useful for screening. Both are rapid, inexpensive, noninvasive, and relatively accurate tools. A screening instrument should be selected on the basis of staff experience and training, available testing time, and characteristics of the patient population and should be used consistently.
Although laboratory tests such as the GGT can provide useful information to supplement knowledge gained through an interview or self-report, Laboratory tests are not adequate when used alone to screen for alcohol problems.
Developing objective markers of alcohol use currently is an important goal of alcohol research. An objective marker should testify to the subject's blood alcohol level over a period of several weeks, even if the subject is abstinent at the time of sampling.
(1) CLEARY, P.D.; Miller, M.; Bush, B.T.; Warburg, M.W.; Delbanco, T.L.; and Aronson, M.D. Prevalence and recognition of alcohol abuse in a primary care population American Journal of Medicine 85:466-471. 1988. (2) MOORE, R.D.; Bone, L.R.: Geller, G.; Mamon, J.A.: Stokes, E.J.; and Levine, D.M. Prevalence, detection, and treatment of alcoholism in hospitalized patients. Journal of the American Medical Association 261(3):403-407. 1989. (3) CLEMENT, S. The identification of alcohol-related problems by general practitioners. British Journal of Addiction 81:257-264, 1986 (4) MOORE, R.D. & Malitz, F.E. Underdiagnosis of alcoholism by residents in an ambulatory medical practice. Journal of Medical Education 61( 1 ):46-52, 1986. (5) RICE, J.P. Statistical issues in the interpretation of tests. In: Screening for Alcoholism in Primary Care Settings. Rockville, MD: National Institute on Alcohol Abuse and Alcoholism, 1987. pp. 3-5. (6) GRANT, B.F.; Hasin, D.S.; and Hartford, T.C. Screening for major depression among alcoholics: An application of receiver operating characteristic analysis. Drug and Alcoh ol Dependence 23:123-131.1989 (7) MAYFIELD, D.G.; McLeod. G. and Hall, P. The CAGE questionnaire: Validation of a new alcoholism screening instrument. American Journal of Psychiatry 131:1121-1123, 1974. (8) EWING, J.A. Detecting alcoholism: The CAGE questionnaire. Journal of the American Medical Association 252(14):1905-1907,1984. (9) ALLEN, J.P.; Eckardt, W.; and Wallen, J. Screening for alcoholism: Techniques and issues. Public Health Reports 103(6):586-592,1988. (10) BUSH, B.;Shaw. S.; Cleary, P.; Delbanco. T.L.; and Aronson, M.D. Screening for alcohol abuse using the CAGE questionnaire. American Journal of Medicine 82:231-23S, 1987. (11) CLARK, W.D. Alcoholism: Blocks to diagnosis and treatment. American Journal of Medicine 71 :275-286, 1981. (12) SELZER, M.L. The Michigan Alcoholism Screening Test: The quest for a new diagnostic instrument. American Journal of Psychiatry 127(12):89-94,1971. (13) JACOBSON, G R. & Lindsay, D. Screening for alcohol problems among the unemployed. In: Galanter, M., ed. Currents in Alcoholism: Recent Advances in Research and Treatment. Vol. VlI. New York: Grune and Stratton, 1980. pp. 357-371. (14) ROSS, H.E; Gavin, D.R.; and Skinner, H.A. Diagnostic validity of the MAST and the Alcohol Dependence Scale in the assessment of DSM-III alcohol disorders. Journal of Studies on Alcohol, in press. (15) SELZER, M.L.; Vinokur A.; and van Rooijen, L. A self-administered Short Michigan Alcoholism Screening Test (SMAST). Journal of Studies on Alcohol 36(1):117-126, 1975. (16) POKORNY, A.D.; Miller, B.A.; and Kaplan, H.B. The brief MAST: A shortened version of the Michigan Alcoholism Screening Test. American Journal of Psychiatry 129(3):118-121, 1972. (17) SWENSON, W.M. & Morse, R.U. The use of a self-administered alcoholism screening test (SAAST) in a medical center. Mayo Clinic Proceedings 50(4):204-208, 1975. (18) HURT, R.D.; Morse, R.M.; and Swenson, W.M. Diagnosis of alcoholism with a self-administered alcoholism screening test. Mayo Clinic Proceedings 55:365-370, 1980. (19) DAVIS, L.J., JR.; Hurt, R.D.; Morse, R.M.; and O'Brien, P.C. Discriminant analysis of the self-administered alcoholism screening test. Alcoholism: Clinical and Experimental Research 11(3):269-273, 1987. (20) SKINNER, H.A. & Horn, J.L. Alcohol Dependence Scale (ADS) Users Guide. Toronto: Addition Research Foundation,1984. (21) EDWARDS, G. & Gross, M.M. Alcohol dependence: provisions description of a clinical syndrome. British Medical Journal 1 (6017): 1058-1061, 1976. (22) FORNEY, P.D.; Forney, M.A.; and Ripley, W.K. Profile of an adolescent problem drinker. Journal of Family Practice 27(1):65-70, 1988. (23) KLIZNER, U.; Schwartz, R.H.; Gruenewald, P.; and Blasinsky, M. Screening for risk factors for adolescent alcohol and drug use. American Journal of Diseases of Children 141(Jan.):45-49, 1987. (24) HARRELL, A V. & Wirtz P.W. Screening for adolescent problem drinking: Validation of a multidimensional instrument for case identification. Psychological Assessment 1:61 -63,1989. (25) SOKOL, R.J.; Martier, S.S.; and Ager, J.W. The T-ACE questions: Practical prenatal detection of risk-drinking. American Journal of Obstetrics and Gynecology 160(4) :863-870,1989. (26) SCHUCKIT, M.A. & Griffiths, J.C. Gamma-glutamyltransferase values in nonalcoholic drinking men. American Journal of Psychiatry 139(2):227-228. 1982. (27) GJERDE, H.; Amundsen, A.; Skog, O.-J.; Morland, J.; and Aasland, O.G. Serum gamma-glutamyltransferase: An epidemiological indicator of alcohol consumption? British Journal of Addiction 82:1027-1031,1987. (28) CHICK, J.; Kreitman, N.; and Plant, M. Mean cell volume and gammaglutamyl-transferase as markers of drinking in working men. Lancet 1:12491251, 1981. (29) BERNADT, M.W.; Mumford, J.; Taylor, C.; Smith, B.; and Murray, R.M. Comparison of questionnaire and laboratory tests in the detection of excessive drinking and alcoholism. Lancet 1:325-328, 1982. (30) NALPAS, B.; Vassault, A.; LeGuillou, A.; Lesgourgues, B.; Ferry, N.; Lacour, B.; and Berthelot, P. Serum activity of mitochondrial aspartate aminotransferase: A sensitive marker of alcoholism with or without alcoholic hepatitis. Hepatology 4(5):893-896, 1984. (31) NALPAS, B.; Vassault, A.; Charpin, S.; Lacour, B.; and Berthelot, P. Serum mitochondria aspartate aminotransferase as a marker of chronic alcoholism: Diagnostic value and interpretation in a liver unit. Hepatology 6(4):608-614, 1986. (32) LUMENG, L. New diagnostic markers of alcohol abuse. Hepatology 6(4):742-745, 1986. (33) WATSON, R.R.; Mohs, M.E.; Eskelson, C.; Sampliner, R.E.; and Hartmann, B. Identification of alcohol abuse and alcoholism with biological parameters. Alcoholism: Clinical and Experimental Research 10(4):364-385, 1986. (34) YATES, D.W.; Hadfield, J.M.; and Peters, K. The detection of problem drinkers in the Accident & Emergency department. British Journal of Addiction 82:163-167, 1987.
All material contained in the Alcohol Alert is in the public domain and may be used or reproduced without permission from NIAAA. Citation of the source is appreciated.
Copies of the Alcohol Alert are available free of charge from the Scientific Communications Branch, Office of Scientific Affairs, NIAAA, Willco Building, Suite 409, 6000 Executive Boulevard, Bethesda, MD 20892-7003. Telephone: 301-443-3860.
U.S. DEPARTMENT OF HEALTH AND HUMAN SERVICES
Public Health Service * National Institutes of Health
Updated: October 2000