To Name :
To Email :
From Name :
From Email :
Comments :

Evidence-Based Reviews

Getting to the bottom of problem drinking: The case for routine screening

More-accurate screening tests and biochemical measures make it easier to recognize alcohol problems, motivate change in drinking, and reinforce abstinence.

Vol. 2, No. 6 / June 2003

Do you know which of your patients have alcohol problems? Though alcohol use disorders may be difficult to detect, self-report and biochemical measures followed by a thorough face-to-face assessment improve diagnostic accuracy. New tools—such as the serum carbohydrate-deficient transferrin (CDT) test—are changing how psychiatrists screen for alcohol problems, provide motivational feedback, and monitor patients for relapse.


Screening for excessive alcohol consumption is important in psychiatric practice because:

  • Alcohol use disorders coexist with many psychiatric problems, most notably affective and anxiety disorders and—not surprisingly—other substance abuse disorders (Table 1).1,2
  • Patients with psychiatric comorbidity who abuse alcohol have poorer prognoses, are less adherent to treatment, and are more likely to drop out of treatment than are psychiatric patients who do not have alcohol problems.3
  • Alcohol interacts with many psychotropics, and chronic heavy drinking can cause pharmacokinetic changes that affect a patient’s response to medications.
  • Alcohol-dependent patients are more likely than nondrinkers to become dependent on anxiolytics and sedative-hypnotics.

Table 1

Overlap of alcohol problems with common psychiatric disorders


Risk of alcohol use disorder (odds ratio)

Source of data (population survey)

Drug use disorder






Major depression



Obsessive-compulsive disorder



Generalized anxiety disorder






Posttraumatic stress disorder



Panic disorder



NLAES: National Longitudinal Alcohol Epidemiological Survey

NCS: National Comorbidity Survey

ECA: Epidemiologic Catchment Area

Because alcohol problems are common in psychiatric patients, routine screening for alcohol abuse and dependence at the onset of any treatment can be very useful. Thereafter, screening can be done periodically—perhaps annually or more often if the patient’s functioning declines.


Many self-report alcohol screening scales are available,4 the most popular being the CAGE 5 and the Michigan Alcoholism Screening Test (MAST).6 Though both instruments can help identify alcohol problems, each has shortcomings:

  • The CAGE performs less reliably in women and adolescents than in men, and its validity depends on the patient’s sensitivity to the emotional impacts of alcohol dependence.
  • The MAST is long (25 items), concentrates on late-stage alcoholism symptoms, and uses differential weighting—not validated in subsequent studies—of particular items in deriving the score.

Neither addresses drinking behavior or when symptoms occurred and thus may misclassify recovered alcoholics or former problem drinkers.

AUDIT. A more reliable choice is the Alcohol Use Disorders Identification Test (AUDIT).7 It was designed by the World Health Organization (WHO) to be valid across gender and culture and to identify even early stage problem drinking. The AUDIT’s 10 items deal with drinking behavior, dependence on alcohol, and adverse consequences of drinking during the past year (Box). The survey takes less than 5 minutes; can be administered orally, in writing, or online; and it retains its validity when given as part of a comprehensive health risk appraisal.8

The WHO offers an excellent manual detailing how to administer and interpret the AUDIT (see Related resources). A patient’s score is computed by summing the values associated with his or her responses to each item. A score of 8 or greater indicates excessive alcohol consumption, although some researchers have argued that for women a more accurate threshold might be 6 or 7 points.

Standardized for adults, the AUDIT also appears to accurately gauge drinking behavior in adolescents9 and in psychiatric patients, although only three studies have explored its use in the latter population.10-12 Abbreviated AUDIT versions have been found to be psychometrically sound8 and may be useful in an emergency room or busy primary care clinic. In comparisons with other screening tools, the AUDIT almost always has been found to be more valid. 9,13


Self-report screens for alcohol problems, especially the AUDIT, are highly sensitive and specific, though their accuracy depends on the patient's memory, understanding of the questions, and candor. In chronic heavy drinkers, biochemical measures (Table 2) can augment self-reports.14

Self-report and biochemical screens have different strengths and weaknesses (Table 3). It is important to see them as complementary because each contributes to accurate screening.

CDT. Most biomarkers screen indirectly for alcohol problems by measuring damage to an end organ-typically the liver-caused by chronic excessive alcohol consumption. False positive results are common because of nonalcohol-related organ damage, medications, smoking, obesity, and other confounding factors. An exception appears to be the serum test for carbohydrate-deficient transferrin (CDT), a biomarker for heavy drinking approved in kit form 3 years ago by the Food and Drug Administration.

The value of measuring CDT levels is that few conditions other than excessive alcohol consumption elevate them. For unclear reasons,15 average daily consumption of >60 grams of alcohol (about five standard drinks) during the previous 2 weeks causes a higher percent of transferrin—a glycoenzyme that transports iron in the body—to lack its usual carbohydrate content.

Bio-Rad Laboratories ( offers a reagent kit (%CDT Turbidimetric Immunoassay). It quantifies CDT as a percent of total serum transferrin, rather than total CDT, thus correcting for individual variations in transferrin levels. CDT values are obtained from a 100-microliter serum sample. The blood is clotted and the serum separated. The sample may be stored at 2 to 8 °C if the test is to be run within 1 week. Samples must be tested at a reference lab (Bio-Rad offers a list of labs). Results are available in a few days.

Patients who deny problem drinking may need convincing to submit to a blood draw. It may help to explain that alcohol use can exacerbate emotional problems and that the test can provide information on possible risky alcohol use.

GGT. Using a second biochemical marker may improve the sensitivity of CDT to detect heavy drinking.16-18

The most-researched choice for a second marker is gamma glutamyltransferase (GGT). Patients are considered to have tested positive for an alcohol problem if either CDT or GGT levels are elevated. Combining these tests may be especially useful in alcohol-dependent women, in whom the reliability of CDT testing alone has been questioned.

Recommendation. Start with a self-report screening measure. If the patient scores slightly below the threshold for an alcohol problem, follow up with the more costly CDT and GGT tests.

For example, biomarkers might be useful for follow-up in men with AUDIT screening scores of 6 or 7 or women with scores of 5 to 7. Biomarkers also are recommended when you suspect an alcohol problem for another reason or question whether the patient responded accurately to the self-report measure.


Screening for problem drinking: the Alcohol Use Disorders Identification Test (AUDIT)

  1. How often do you have a drink containing alcohol?
  2. How many drinks containing alcohol do you have on a typical day when you are drinking?
  3. How often do you have 6 or more drinks on one occasion?
  4. How often during the last year have you found that you were not able to stop drinking once you had started?
  5. How often during the last year have you failed to do what was normally expected from you because of drinking?
  6. How often during the last year have you needed a first drink in the morning to get yourself going after a heavy drinking session?
  7. How often during the last year have you had a feeling of guilt or remorse after drinking?
  8. How often during the last year have you been unable to remember what happened the night before because you had been drinking?
  9. Have you or someone else been injured as a result of your drinking?
  10. Has a relative, friend, doctor, or other health worker been concerned about your drinking or suggested that you cut down?

The World Health Organization offers a manual on how to administer and interpret AUDIT (

Source: World Health Organization

Table 2

Biochemical markers of heavy drinking


Time needed for return to normal limits

Level of drinking characterized


Gamma glutamyltransferase (GGT)

2 to 6 weeks of abstinence

~70 drinks/wk for several weeks

Most common and reliable of the traditional markers of heavy drinking; many sources of false positives

Aspartate aminotransferase (AST) (formerly SGOT)

7 days, but much variability in declines with abstinence

Unknown, but heavy

Present in many organs; many sources of false positives; moderate correlations with GGT

Alanine aminotransferase (ALT) (formerly SGPT)


Unknown, but heavy

Many sources of false positives and less sensitive than AST; ratio of AST to ALT may be more accurate

Macrocytic volume (MCV)

Unknown; half-life ~40 days

Unknown, but regular and heavy

Poor sensitivity and specificity; even with abstinence, very slow return to normal limits and may increase at first; little, if any, gender effect

Carbohydrate-deficient transferrin (CDT)

2 to 4 weeks of abstinence

>60 grams/day for approximately 2 weeks

Few sources of false positives; excellent indicator of relapse


MET. Motivational enhancement therapy (MET) has gained popularity as a means of changing problematic drinking.19 Project MATCH—a multi-site trial on alcohol abuse treatment—studied MET and two other interventions. MET required fewer sessions but equaled the other interventions in reducing drinking days and aver-age amount of alcohol consumed.20

A key component of MET—and other brief interventions—is to provide patients with empathetic, nonjudgmental feedback.19-21 Responses to the first three AUDIT items can provide such feedback to patients with drinking problems. Amazingly, heavy drinkers and alcoholics often do not realize how much more they drink than other people. To help them develop this insight, show them their self-report responses in contrast with national normative data.

Biomarker results can be used similarly, in this case comparing the patient’s score with the test’s reference range values. Kristenson et al22 showed that giving men with elevated scores recurrent biomarker information and advice significantly reduced morbidity and mortality and improved their work performance.

Using visual aids can deepen patients’ under-standing of motivational feedback. For example, displaying sequential test results on a timeline can reinforce motivation by showing how their drinking behavior has improved with continuing treatment and sustained effort.

Table 3

Self-report and biochemical measures of drinking: Pros and cons





High validity
Flexible window of assessment
Immediate results

Easily feigned
Accuracy depends on patient’s verbal skills and memory


Results may be more compelling to patients than self-reports
May reflect organ damage
Useful in tracking treatment progress

Window of assessment is limited to recent past
Results often not immediately available
May be more costly than self-report measures


Although treatment for alcohol problems is often successful,23 relapse to some level of drinking is not uncommon, especially during the first 3 or 4 months after patients complete treatment.20 Recognizing relapse quickly can help you:

  • decrease risk of harm from resumed alcohol use
  • reduce the potential for drinking to again become habitual
  • identify circumstances and cues that may have triggered the drinking episode, for use in tailoring future interventions.

In clinical practice, relapse is most often revealed via comments from the family, direct observation by the clinician, or voluntary acknowledgment by the patient. Interestingly, CDT has demonstrated a relapse “heralding effect,”24 meaning that it tends to rise well before a patient will admit he or she resumed drinking.24-26 Although other markers may also rise following relapse, their elevation tends to be delayed and less dramatic.

The sensitivity and specificity of CDT alone and with GGT in identifying relapse have been evaluated. Across male-only studies, CDT’s median sensitivity (percent of relapsed patients with elevated scores) was 0.73, with a specificity (percent of those not relapsed who had low scores) of 0.91. In the two female-only studies, median sensitivity and specificity for CDT were 0.32 and 0.86, respectively. For women, using CDT and GGT in combination substantially raised median sensitivity to 0.62, although specificity fell slightly to 0.80.27

Recommendation. When using biomarkers to identify relapse, examine the temporal pattern of test results to date. Assume that an increase of 30% or more above the lowest observed lab value indicates a relapse.28

Frequent testing—probably biweekly—is recommended during the first 3 or 4 months after patients complete treatment. If there is no indication of relapse, testing frequency could be tapered down.

Did you miss this content?
Psychosis and catatonia after dancing with a dangerous partner