BMJ  2003;327:E262 (20 December), doi:10.1136/bmj.327.7429.E262

BMJ USA: Editor's choice

EtOH and the numbers

I was taught in medical school to refer to ethanol as "EtOH," both in writing and in conversations that patients might overhear ("Nurse, is there a possibility of EtOH here...?"). I guess the reason for this code was to conceal my diagnosis or concern from patients. Spare them embarrassment. Or spare me the embarrassment of talking about a touchy subject.

Codes aside, we don't do a very good job of recognizing problem drinking in primary care. Alcohol dependence is only one end of the spectrum. More common is chronic heavy drinking or frequent bingeing without dependence, both of which put the drinker at risk for current or future problems.

And alcohol is a big problem. Cirrhosis kills more than 25 000 Americans a year. When you include alcohol-related traffic crash fatalities and other alcohol-related mortality, the number climbs to over 110 000. The deaths skew young, as anyone working in an ER on a Saturday night will attest. As a result, what epidemiologists call "years of potential life lost" are high for alcohol-related traffic crash deaths: Over 300 000 years of life before age 75 are wasted each year because of drinking. Direct and indirect costs related to alcohol abuse approach $200 million annually.

Shouldn't we be screening our patients for problem drinking? Short screening tests are available. Brief interventions are effective for those not alcohol-dependent, more intensive ones work for those who are. The very conservative US Preventive Services Task Force (USPSTF) recommended alcohol screening in primary care in 1996, and their new guideline coming out next year will likely renew that recommendation. Yet, when Anders Beich and colleagues systematically reviewed the primary care research literature on screening for excessive drinking (p 590), they concluded that it's probably not worth it. Why?

It has to do with the numbers and how you interpret them. In evaluating screening tests, a positive recommendation hinges on several factors: the seriousness of the problem, whether it can be discovered in an early stage, the adequacy of the treatment, and whether early detection and treatment actually improves outcome. Beich et al calculated that 1000 primary care patients would have to be screened to end up with 2 or 3 who drank less a year later. Not a very good deal.

We asked Evelyn Whitlock, a physician epidemiologist who works on alcohol screening with the USPSTF, to comment on the study (p 579). She has concerns about the way it was performed. We've also published edited excerpts from some of the dozens of Rapid Responses that the BMJ received in response to the article (p 596). It is an important issue, worthy of careful scrutiny and reasoned debate.

CiteULike    Complore    Connotea    Del.icio.us    Digg    Reddit    Technorati    What's this?