As Mersy¹ points out in his article in this issue of American Family Physician, substance use problems are common and serious. They are also hidden.

Most patients' problems with hazardous drinking or substance use go unrecognized in most practices.² Some disorders can be identified by the symptoms and signs described in the article, especially if they occur in combination. Why should physicians be concerned? Because even when there are several indications that substance use may be playing a role in the patient's problems, we tend to miss the diagnosis.

Brief interventions for problem drinking work.³ Fleming and colleagues in Wisconsin⁴ replicated a study done in the United Kingdom.⁵ In both studies, patients in primary care practices were screened. If they were drinking more than safe limits (more than two drinks per day, on average), appointments were made for two 10- to 15-minute intervention visits with a family physician, plus two telephone contacts by an office nurse. One year later, approximately 40 percent of the patients in the intervention groups had moderated their drinking to safe levels, compared with 20 percent in the control groups. In the Wisconsin trial, the differences between intervention and control groups were still present four years later. Outcomes such as length of hospital stays were significantly reduced in the intervention group. For every $1 spent on brief interventions, $4.30 was saved.⁶

How should patients with more serious alcohol problems be managed? Here, too, the evidence shows that treatments work.⁷ In Miller and Wilbourne's article,⁷ brief interventions by primary care clinicians top the list of effective treatments. Many other treatments—social skills training, community reinforcement, behavior contracting, behavior marital therapy, case management—also have solidly documented effectiveness.

How should we screen? Mersy¹ is correct: Pick a screening test that works in your practice and use it. Written or oral tests are more sensitive than laboratory tests; carbohydrate-deficient transferrin does not become abnormal until the patient is drinking more than four drinks per day every day—considerably above the threshold of hazardous drinking. Furthermore, this test is available only through certain reference laboratories. The CAGE questions and the Alcohol Use Disorders Identification Test are well tested and effective.⁸ A single question also is effective: “When was the last time you had more than X drinks in one day?” where X = 4 for women and 5 for men.⁹ A positive screen would be within the preceding three months. Pick a screening approach, use it routinely, and develop a charting system so you do not have to screen patients more than once unless their situation changes.

What about drug abuse? Unfortunately, there are few validated screening instruments and few studies of brief interventions for substance use disorders other than alcohol. The CAGE questionnaire expanded to include drugs is one effective screening approach.¹⁰ Until we know more about which brief interventions are effective in patients with drug use problems, extrapolating findings from studies of brief interventions with problem drinkers is reasonable.

What should the content of a “brief intervention” be? Several approaches are effective, including a straightforward physicians' guide (available online atwww.niaaa.nih.gov/publications/physicn.htm), patient handout (www.niaaa.nih.gov/publications/handout.htm), and more involved, yet still readily learned motivational-enhancement techniques.¹¹ Problem drinking meets all the criteria for conditions that family physicians should screen for and address, and the U.S. Preventive Services Task Force agrees.¹² Despite this, we have not incorporated alcohol screening to the extent that is recommended.² Family physicians can be effective coaches in helping patients change their behaviors. Taking up that call will require that we change our own.