A Randomized Trial of a Multicomponent Cessation Strategy for Emergency Department Smokers


  • Presented at the 2008 Society for Academic Emergency Medicine annual meeting, Washington, DC, May 29–June 1, 2008.

  • Supported by Grant R21DA017812 from the National Institute on Drug Abuse. ClinicalTrials.gov number NCT00297466.

  • Dr. Steven L. Bernstein has served as an expert witness for the plaintiff in litigation against the tobacco companies.

  • Supervising Editor: Rebecca M. Cunningham, MD.

  • A related commentary appears on page 635.

Address for correspondence: Steven L. Bernstein, MD; e-mail: Steven.bernstein@yale.edu. Reprints will not available.


ACADEMIC EMERGENCY MEDICINE 2011; 18:575–583 © 2011 by the Society for Academic Emergency Medicine


Objectives:  The objective was to determine the efficacy of an emergency department (ED)-based smoking cessation intervention.

Methods:  This study was a randomized trial conducted from January 2006 to September 2007 at an urban ED that treats 90,000 adults per year. Discharged adults who smoked at least 10 cigarettes per day were randomized to 1) usual care, receiving a smoking cessation brochure; or 2) enhanced care, receiving the brochure, a motivational interview (MI), nicotine patches, and a phone call at 3 days. Interventions were performed by a peer educator trained in tobacco treatment. Blinded follow-up was performed at 3 months.

Results:  A total of 338 subjects were enrolled, mean (±SD) age was 40.2 (±12.0) years, 51.8% were female, and 56.5% were either self-pay or Medicaid. Demographic and clinical variables were comparable between groups. Enhanced and usual care arms showed similar cessation rates at 3 months (14.7% vs. 13.2%, respectively). The proportion of subjects making a quit attempt (69.2% vs. 66.5%) and decrease in daily cigarette use (five vs. one; all p > 0.05) were also similar. In logistic modeling, factors associated with quitting included any tobacco-related International Classification of Diseases, ninth revision (ICD-9), code for the ED visit (odds ratio [OR] = 3.42, 95% confidence interval [CI] = 1.61 to 7.26) or subject belief that the ED visit was tobacco-related (OR = 2.47, 95% CI = 1.17 to 5.21). Conversely, subjects who reported having a preexisting tobacco-related illness were less likely to quit (OR = 0.22, 95% CI = 0.10 to 0.50).

Conclusions:  The primary endpoint was negative, reflecting a higher-than-expected quit rate in the control group. Subjects whose ED visit was tobacco-related, based either on physician diagnosis or subject perception, were more than twice as likely to quit. These data suggest that even low-intensity screening and referral may prompt substantial numbers of ED smokers to quit or attempt to quit.