Am Fam Physician. 2005;71(2):347
Clinical Question: Does the addition of bupropion to nicotine replacement and counseling improve cessation rates?
Setting: Outpatient (any)
Study Design: Randomized controlled trial (double-blinded)
Synopsis: The investigators conducting this study enrolled 209 men and 34 women who were receiving care at a Veterans Affairs medical center in the United States. These primarily unmarried, middle-aged, white smokers were recruited by hospital-based advertising, as well as from lists of patients who previously had participated in smoking-cessation clinical studies. Presumably, the patients in the latter group had failed previous attempts at smoking cessation, although they may have taken up smoking to participate in the study (i.e., professional study participants).
The patients were randomized (allocation concealment uncertain) to receive placebo or bupropion in a dosage of 150 mg twice daily for seven weeks. Patients in both groups received cognitive-behavior counseling for 30 to 60 minutes with five follow-up counseling sessions by telephone. They also received nicotine-replacement therapy using transdermal nicotine patches. Initially, quit rates were high (i.e., 81 percent at one week). However, the clinically relevant outcome for smoking cessation is the quit rate at one year. In this study, biochemically confirmed quit rates at one year were not statistically different between the two groups: 19 percent in the bupropion-treated group compared with 24 percent in the placebo-treated group (P = not significant). These results were in the intention-to-treat analysis. Results were higher in the patients who were compliant, but not different between the two groups: 28 percent of compliant bupropion-treated patients and 22 percent of compliant placebo-treated patients. Although the study was large, as were the quit rates, the researchers did not provide a power calculation.
Bottom Line: The addition of bupropion does not further increase long-term quit rates in patients who receive nicotine replacement and cognitive-behavior counseling. Quit rates at one year were similarly high in both groups—22 and 28 percent (P = not significant). (Level of Evidence: 1b)