Tobacco use: prevention, cessation, and control.

Objectives: The RTI International-University of North Carolina at Chapel Hill Evidence-based Practice Center (RTI-UNC EPC) systematically reviewed the evidence on (a) the effectiveness of community- and population-based interventions to prevent tobacco use and to increase consumer demand for and implementation of effective cessation interventions; (b) the impacts of smokeless tobacco marketing on smoking, use of those products, and population harm; and (c) the directions for future research.

Data sources: We searched MEDLINE, Cumulative Index to Nursing and Applied Health (CINAHL), Cochrane libraries, Cochrane Clinical Trials Register, Psychological Abstracts, and Sociological Abstracts from January 1980 through June 10, 2005. We included English-language randomized controlled trials, other trials, and observational studies, with sample size and follow-up restrictions. We used 13 Cochrane Collaboration systematic reviews, 5 prior systematic reviews, and 2 meta-analyses as the foundation for this report.

Review methods: Trained reviewers abstracted detailed data from included articles into evidence tables and completed quality assessments; other senior reviewers confirmed accuracy and resolved disagreements.

Results: We identified 1,288 unique abstracts; 642 did not meet inclusion criteria, 156 overlapped with prior reviews, and 2 were not published articles. Of 488 full-text articles retrieved and reviewed, we excluded 298 for several reasons, marked 88 as background, and retained 102. Evidence (consistent with previous reviews) showed that (a) school-based prevention interventions have short-term (but not long-term) effects on adolescents; (b) multicomponent approaches, including telephone counseling, increase the number of users who attempt to quit; (c) self-help strategies alone are ineffective, but counseling and pharmacotherapy used either alone or in combination can improve success rates of quit attempts; and (d) provider training and academic detailing improve provider delivery of cessation treatments, but evidence is insufficient to show that these approaches yield higher quit rates. New evidence was insufficient to address the following: (a) effectiveness of population-based prevention interventions; (b) effectiveness of provider-based interventions to reduce tobacco initiation; (c) effectiveness of community- and provider-based interventions to increase use of proven cessation strategies; (d) effectiveness of marketing campaigns to switch tobacco users from smoking to smokeless tobacco products; and (e) effectiveness of interventions in populations with comorbidities and risk behaviors (e.g., depression, substance and alcohol abuse). No evidence was available on the way in which smokeless tobacco product marketing affects population harm.

Conclusions: The evidence base has notable gaps and numerous study deficiencies. We found little information to address some of the issues that previous authoritative reviews had not covered, some information to substantiate earlier conclusions and recommendations from those reviews, and no evidence that would overturn any previous recommendations.