A Controlled Trial of Brief Intervention Versus Brief Advice for At-Risk Drinking Trauma Center Patients Carl A. Soderstrom, MD, Carlo C. DiClemente, PhD, Patricia C. Dischinger, PhD, J. Richard Hebel, PhD, David R. McDuff, MD, Kimberly Mitchell Auman, MS, and Joseph A. Kufera, MA Background: Numerous reports doc- ument that preinjury alcohol use is asso- ciated with all modes of injury requiring treatment in a trauma center, with 25% to 50% or more of patients testing positive for alcohol at the time of admission. There is evidence that in trauma patients unad- dressed alcohol use problems result in re- current injury requiring readmission to a trauma center and/or death. Methods: A randomized clinical trial was conducted to assess the effectiveness of two types of brief interventions to re- duce drinking and the consequences of drinking. Trauma patients defined as at- risk alcohol users (n  497) were ran- domized into two treatment options: a brief personalized motivational interven- tion (PMI), or brief information and ad- vice (BIA). After a brief assessment, PMI subjects received a motivational session, feedback letter, and two postdischarge telephone contacts, whereas the BIA group received a brochure and one post- discharge telephone contact. Both groups were reassessed at 6 and 12 months postinjury. Results: Both the PMI and BIA groups had statistically significant reduc- tions in drinking, binge episodes, and consequences related to drinking that per- sisted from the 6- to the 12-month follow- up. However, although not statistically significant, for those classified as lower- level drinkers (<1 drink per day), there was a consistent pattern of maintaining reductions for the PMI group at 12 months compared with the BIA group. Conclusion: Our results suggest that brief interventions (PMI and BIA) that link alcohol consumption with trauma in- jury and consequences of drinking can be effective in reducing drinking and conse- quences related to drinking in a signifi- cant portion of at-risk nondependent drinkers. J Trauma. 2007;62:1102–1112. N umerous reports document that preinjury alcohol use is associated with all modes of injury requiring treatment in a trauma center, with 25% to 50% or more of patients 1–5 testing positive for alcohol at the time of admis- sion. Further, depending on whether screening test results or diagnostic criteria were used, 25% to 50% of trauma center patients have been identified to have an alcohol use problem. 1,6,7 A prior study in the clinical site for the current study documented that 24% of admitted patients were alcohol dependent. 7 There is evidence that in trauma patients unaddressed alcohol use problems result in recurrent injury, requiring readmission to a trauma center and/or death. A 5-year longi- tudinal study 8 of urban trauma center patients indicated that two-thirds of patients sustaining a recurrent injury had an indication of substance abuse (alcohol and/or drugs) at the time of the initial injury episode. Another study found that trauma center patients who were positive for possible alco- holism were over three times more likely to be readmitted for treatment of a second trauma episode compared with other patients. 9 The most compelling reason to conduct a trial of brief intervention for injured patients with alcohol use problems comes from a postdischarge mortality study from the trauma center that was the site for this project. 10 In that study, over 27,000 patients were followed for a period of 1.5 to 14.5 years. In contrast to an overall 6.4% mortality rate from injury for the general population, 23% of trauma center pa- tients died from a subsequent injury. Over one-third (35%) of those who tested positive for alcohol or other drugs of abuse died as the result of a subsequent injury event after the initial trauma center admission (alcohol was the most common substance for which positive test results were obtained at the index injury admission). In contrast, 15% of trauma patients who tested negative died as the result of another injury episode. A trauma center admission per se does not appear to result in decreases in postinjury hazardous drinking patterns. Using several measures of past-month hazardous drinking, Dunn and colleagues 11 found that 41% of trauma center patients engaged in hazardous drinking before admission. Although there was an overall significant decrease in hazard- ous drinking among patients 1 month after injury, 55% of the Submitted for publication March 9, 2006. Accepted for publication February 2, 2007. Copyright © 2007 by Lippincott Williams & Wilkins, Inc. From the National Center for Trauma and EMS (C.A.S., P.C.D., K.A., J.A.K.), the Departments of Epidemiology (J.R.H) and Psychiatry (D.R.M.), University of Maryland School of Medicine, Baltimore, MD; and the De- partment of Psychology (C.D.C), University of Maryland Baltimore County, Baltimore, MD. Supported by the National Institute on Alcoholic Abuse and Alcohol- ism (grant 2 RO1 AA09050-04A2). Presented orally at the Research Society on Alcoholism Meeting, June 27, 2005, Santa Barbara, California. Address for reprints: Carl A. Soderstrom, MD, FACS, Maryland MVA, Medical Advisory Board, Glen Burnie, MD 21062; email: csoderstrom@umm. edu. DOI: 10.1097/TA.0b013e31804bdb26 The Journal of TRAUMA  Injury, Infection, and Critical Care 1102 May 2007