Testing for Alcoholism
The single screening question recommended by the National Institute on Alcohol Abuse and Alcoholism (NIAAA) accurately identifies unhealthy alcohol intake.
About 15 percent of the U.S. population are problem drinkers. Approximately 5 percent to 10 percent of male drinkers and 3 percent to 5 percent of female drinkers could be diagnosed as alcohol dependent--this translates to about 12 million people in the U.S. alone. "But this situation is too often under-diagnosed," writes Peter C. Smith, MD, Boston University School of Medicine. "Practice guidelines recommend universal screening."
In the study, patients from primary care waiting rooms were asked, "How many times in the past year have you had X or more drinks in a day?" (X = 5 for men and 4 for women). A positive response to this single-question screen was defined as "more than 1." Patients were considered to have unhealthy alcohol use if a standardized diagnostic interview revealed the presence of an alcohol use disorder or if a validated 30-day calendar method showed risky consumption.
There are other questionnaires (e.g. AUDIT) that claim to have reasonable sensitivity and specificity for detecting alcohol abuse. The concern about these self-reporting surveys is that should the patient recognize the reason behind the survey, the appropriate answer to questions like the one above would be "None," or "Never."
There are a number of laboratory tests available in addition to blood alcohol, which is of value only within hours after drinking. These tests include carbohydrate deficient transferrin (CDT), long used in Europe and in some institutions in the U.S. CDT has been found to be more sensitive and specific than GGT (although the combination of the two is better than either alone).
More recently, phosphatidylethanol (PEth) and ethyl glucuronide (EtG) have been studied in serum and urine samples. For example, in one study patients were divided into two groups: AUDIT scores < 8 and AUDIT scores >or= 8. Twenty-five percent of the patients with AUDIT scores below the cut-off were tested positive for both PEth and UEtG. Of the patients who declared to be sober during the past 12 months, 38.5 percent were tested positive for PEth and UEtG. PEth discriminated well as %CDT for AUDIT scores >or= 8. The authors state that the "self-reports of alcohol consumption were unreliable."
It would seem that in addition to testing blood alcohol levels in suspected DUI cases, there are other situations (physician's suspicions, re-granting a driver's license, monitoring therapy, etc.) where the laboratory can play an important and objective
role in aiding the clinician.