Skip to main content

A useful test for monitoring alcohol use

June 01, 2009

Carbohydrate-deficient transferrin, known as “%CDT,” is a blood test to detect heavy alcohol use (five or more drinks a day) over the past two weeks.1 This alcohol biomarker can provide clinicians with an objective way to screen and monitor clients in treatment for alcohol use disorders. %CDT monitoring is used by counselors to evaluate whether heavy alcohol use is complicating other diagnoses (e.g., cocaine use, psychiatric conditions); to evaluate a treatment plan's effectiveness; to detect “slips” to prevent relapse; and to provide an objective measure of drinking that avoids the counselor/client debate about whether the client is drinking.

Transferrin is a glycoprotein (a molecule made of amino acids and sugar groups) produced in liver cells. Its purpose is to “transfer” (hence, the word “trans”) iron molecules (hence the word “ferrin”) from the intestine, through the blood stream, to cells and organs that need iron to function; it is present in everyone under normal conditions. Normal transferrin has sugar groups (carbohydrates) attached to it as “side-chains” (the protein is like a motorcycle and the sugar groups like a side car). When someone drinks alcohol at the rate of about five drinks a day, the liver cell does not manufacture transferrin in the normal way (it eliminates some of the side cars). The molecule becomes deficient in sugar side-chains. This is where the name “carbohydrate-deficient transferrin” comes from. Heavy alcohol use causes the transferrin molecule to be abnormal or “carbohydrate-deficient.”

Under normal conditions, only about 1 to 2% of transferrin is carbohydrate-deficient. As a result of heavy drinking, it can go as high as 10%. Studies have found that a %CDT of 2.6% and higher is indicative of heavy drinking. About 50 to 70% of heavy drinkers will have a %CDT greater than 2.6%.

Evaluating the numbers

A client who has a %CDT of 2.6% or higher probably has been drinking at least five drinks a day for the past two weeks or longer. This is the case for 19 out of 20 people with this level of %CDT, but the test may be wrong in one out of 20 individuals who is not a heavy consumer.

There are not many other diseases or any drugs that cause an elevation of %CDT. For the small percentage of non-heavy drinkers who have a %CDT above 2.6%, this typically results from end-stage liver disease or a rare genetic defect. In most alcoholics who have moderate liver disease, the liver is not compromised sufficiently to cause an elevation of %CDT without the presence of heavy drinking.

Once a baseline %CDT has been established (e.g., at the beginning of treatment), the monitoring of future increases or decreases can prove very informative. A 30% decrease from a baseline measure strongly indicates that a client has reduced drinking substantially or has been abstinent since the last test. A 30% increase strongly indicates that a client has increased drinking substantially since the last test.

In a typical scenario, a recently heavy-drinking person entering treatment will have a %CDT over 2.6% and during early abstinence (two to four weeks) the %CDT level will decrease (such as to 1.8%) and stay there while abstinence is maintained. If relapse drinking occurs, perhaps six weeks into treatment, the level could increase by 30%, to greater than 2.3% (note that this is not above 2.6% but still 30% above the lowest level achieved during abstinence).

Case examples

Two case studies illustrate the clinical use of %CDT. “S.K.” is a 56-year-old male being seen in a mental health clinic for alcohol dependence and bipolar disorder. At the beginning of treatment, his %CDT was 5.2%, indicating chronic heavy drinking. During the first four months of treatment, monthly %CDT values were all below 2.6%, indicating abstinence or moderate drinking.

At that point, S.K. began missing appointments and, after a month of this pattern, was observed to be depressed and suicidal. He had not been taking his bipolar disorder medication. He denied alcohol use. He was admitted to the psychiatric unit for stabilization of his suicidal ideation and depression. Initial admission labs included a negative urine drug screen, a zero blood alcohol level, and negative liver function tests. His %CDT was 4.2%.

When the %CDT lab value was discussed with S.K., he admitted to drinking up to 18 beers a day for the past month. Since he had experienced severe alcohol withdrawal in the past, he was placed on withdrawal precautions. After hospitalization, he was discharged to the substance abuse treatment clinic for outpatient treatment and routine %CDT monitoring.

“R.L.” is a 47-year-old male with a history of chronic pain resulting from a neck injury requiring the use of opioid pain medication. His pain was being treated with methadone (10mg three times daily), but there was concern about his misuse of the medication. He had a history of alcohol abuse that was in reported remission. His most recent urine drug screen was negative for opiates, positive for methadone, and positive for marijuana. The positive marijuana finding prompted a referral to the substance abuse clinic.

R.L. admitted using marijuana and denied alcohol use. He agreed to submit to another urine drug screen and %CDT testing. While his urine drug screen was negative, his %CDT was 4.2%. When the %CDT results were discussed with him, he reluctantly admitted that he had been drinking to excess over the past year. As a result, %CDT monitoring was instituted as part of his pain medication contract so that medication would be prescribed only if %CDT values were negative.

%CDT can be a useful tool for the clinician in providing ongoing information about a client's heavy alcohol use. It can provide concrete evidence that the treatment plan is having the desired effect. The addition of %CDT testing to the practice of addiction treatment constitutes a positive step that enhances patient care, provides non-discriminatory feedback, and allows for a more objective dialogue between counselor and client.

Since %CDT is not 100% accurate in identifying all heavy drinkers (more like 50 to 70%), a low value (negative test) cannot be used as proof that a given person is not drinking heavily. However, even if a person has a level below 2.6%, he/she still can drop further (30% or so) when abstinence is achieved and maintained. Knowledge of this test and its interpretation can be of great benefit in addiction treatment.2

Peter M. Miller, PhD, is a Professor of Psychiatry and Behavioral Sciences at the Center for Drug and Alcohol Problems, Medical University of South Carolina. His e-mail address is Jeffrey Cluver, MD, is an Assistant Professor of Psychiatry and Behavioral Sciences at the university, and Raymond F. Anton, MD, is a Professor of Psychiatry and Behavioral Sciences and directs the Center for Drug and Alcohol Problems.


  1. Miller PM, Dominick C, Anton RF. Carbohydrate-deficient transferrin test: a tool for detecting alcohol abuse. Current Psychiatry 2005; 4:80-7.
  2. Cluver J, Miller PM, Anton RF. Case studies of the utility of serum carbohydrate-deficient transferrin (%CDT) in the clinical management of alcoholics. J Addict Med 2007; 1:44-7.

Behavioral Healthcare 2009 June;29(6):38-40

Back to Top