Definition

Smoking bans and restrictions are private, nongovernment, and government policies, regulations, and laws that limit smoking in workplaces and public areas. Smoking bans entirely prohibit smoking in geographically defined areas; smoking restrictions limit smoking to designated areas. Smoking bans and restrictions can be implemented with additional interventions, such as education and tobacco use treatment programs.

Background

Businesses establish smoking policies to protect employees and customers from exposure to ETS in the workplace. Accrediting agencies set regulations to protect employees and patrons within their organizations (e.g., Joint Commission on Accreditation of Healthcare Organizations). Federal, state, or local laws are implemented to protect persons from ETS exposure in public areas, and to establish minimum standards for both public and private workplaces. For regulations and laws establishing smoking restrictions, standards often include the size, location, and ventilation requirements for designated smoking areas.

Review of evidence: effectiveness

Our search identified a total of 54 studies regarding the effectiveness of smoking bans or restrictions. ^{23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76} Of these, 17 studies evaluated the impact of the intervention on exposure to environmental tobacco smoke. ^{23, 27, 28, 30, 31, 35, 36, 38, 40, 42, 47, 48, 53, 55, 57, 61, 72} Thirty-nine studies evaluated only the effect of smoking bans and restrictions on tobacco use behaviors, and these studies are considered below. Five papers provided additional information regarding an already-included study. ^{77, 78, 79, 80, 81} Seven of the 17 studies were not included in the review of effectiveness because of least suitable study designs and fair or limited execution quality. ^{23, 27, 30, 38, 40, 48, 53} Details of the 10 qualifying studies ^{28, 31, 35, 36, 42, 47, 55, 57, 61, 72} are provided in Appendix B-1, and at the website: www.thecommunityguide.org.

The 10 qualifying studies provided a total of 12 measurements of the impact of smoking bans and restrictions on exposure to environmental tobacco smoke (Figure 2). Overall, 9 of the 10 studies ^{28, 31, 35, 36, 47, 55, 57, 61, 72} observed reductions or differences in ETS exposure in workplaces that had smoking bans or restrictions. In four studies, environmental measurements of components of ETS (such as nicotine vapor) were collected before and after implementation of the smoking ban or restriction. ^{28, 35, 55, 72,} Environmental measurements of ETS components decreased by a median relative percentage difference of -72% (range -44% to -97%) in assessments conducted between 6 months and 12 months after implementation of the ban or restriction. Six studies provided a total of 8 measurements of differences in self-reported exposure to ETS. ^{31, 36, 42, 47, 57, 61} In assessments conducted between 4 and 18 months post-implementation, the median relative percentage difference in self-reported ETS exposure was -60% (range of +4% to -94%).

Figure

Figure 2. Relative percentage changes in exposure to environmental tobacco smoke attributable to workplace smoking bans and restrictions from studies that qualified for inclusion in this review.

Four studies evaluated the impact of smoking restrictions, ^{36, 42, 47, 55} four studies measured the impact of smoking bans, ^{28, 35, 57, 72} and two studies measured differences in workplace ETS exposure for both. ^{31, 61} In general, reductions in ETS exposure were greater in workplaces that had smoking bans than in those with only smoking restrictions (Figure 2).

Review of evidence: applicability

The same body of evidence used to assess effectiveness was used to assess the applicability of smoking bans and restrictions to different settings and populations. Smoking bans and restrictions were evaluated in a variety of settings including hospitals and medical centers, ^{28, 57, 72} health care provider offices, ⁵⁷ government or public sector workplaces, ^{35, 47, 55} and a university. ⁴² Studies on representative samples of employed persons in California, ^{31, 61, 79, 80} and in Missouri ³⁶ demonstrated that smoking bans and restrictions reduced self-reported ETS exposure in workplaces community-wide. Studies included representative samples of indoor workers in the states of California ^{31, 61, 79, 80} and Missouri, ³⁶ and large, diverse samples of government employees in Texas ⁴⁷ and HMO employees in Oregon. ⁵⁷ The evidence of effectiveness in these studies should extend to most indoor workers in the United States.

In 4 studies, smoking bans or restrictions were implemented and evaluated in response to a government law. ^{35, 36, 47, 55} In 4 studies the bans or restrictions were the result of private-sector policies. ^{28, 42, 57, 72} Two studies measured the effect of smoking bans or restrictions created through workplace policies or local ordinances. ^{31, 61}

No studies were found that evaluated the effect of smoking bans or restrictions in public settings outside of the workplace, such as public transportation systems or sports and entertainment venues.

Review of evidence: other positive or negative effects

Fifty of the 54 studies in the body of evidence measured the effectiveness of smoking bans or restrictions on one or more tobacco use behaviors, including consumption (cigarettes per shift or per day), cessation attempts, cessation, or smoking prevalence. ^{24, 25, 26, 27, 28, 29, 30, 32, 33, 34, 35, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 49, 50, 51, 52, 54, 55, 56, 57, 58, 59, 60, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 80} However, 41 of these studies ^{24, 25, 26, 27, 28, 30, 32, 33, 35, 37, 38, 39, 40, 41, 43, 44, 45, 46, 49, 51, 54, 55, 56, 58, 59, 60, 63, 64, 65, 66, 67, 68, 69, 71, 72, 73, 74, 75, 76, 77, 80} were not included in the review of effectiveness because of least suitable study designs. Details of the 9 qualifying studies ^{29, 34, 42, 47, 50, 52, 57, 62, 70} are provided in Appendix B-2, and at the website: www.thecommunityguide.org.

All 9 qualifying studies measured self-reported cigarette consumption, and 8 of the 9 studies ^{29, 34, 42, 47, 50, 52, 62, 70} observed either reductions or lower levels of reported cigarette consumption in the group or population exposed to smoking bans or restrictions. Overall, the median absolute change was -1.2 cigarettes per day (range: no change in consumption to -4.3 cigarettes per day) with follow-up periods of up to 2 years.

Four studies of smoking bans ^{34, 52, 62, 70} measured self-reported smoking cessation in sampled smokers over follow-up periods of 4 weeks to 18 months. In one small study, ³⁴ none of the smokers quit in the 4 weeks of follow-up. In each of 3 studies with study periods of 12-18 months, ^{52, 62, 70} a larger proportion of smokers subject to a workplace smoking ban quit compared with smokers subject to lesser or no workplace smoking restrictions. In 2 of these studies, ^{62, 70} the absolute percentage changes were +7.9 and +9.6 percentage points, respectively. The third study ⁵² reported an adjusted relative risk of 1.7 (95% CI 1.2, 2.4).

Six studies ^{29, 42, 47, 50, 57, 62} measured changes or differences in tobacco use prevalence. Three studies ^{29, 47, 62} observed reductions in prevalence following implementation or exposure to smoking bans or restrictions (absolute percentage changes of -1.4, -3.4, and -11.4 percentage points). Three studies ^{42, 50, 57} observed small increases in tobacco use prevalence (absolute percentage changes of +0.4, +0.8, +1.0 percentage points).

This body of evidence led the Task Force to the conclusion that smoking restrictions and bans appear to have an effect on tobacco consumption and cessation, but the evidence of an effect on tobacco use prevalence is less consistent.

Additional benefits of smoking bans or restrictions include reduced workplace cleaning costs and reduced risk for fires. ⁸² Seven papers describing six studies provided evidence that smoking ordinances do not have an adverse economic effect on businesses (including bars and restaurants) or on tourism. ^{83, 84, 85, 86, 87, 88, 89}

Review of evidence: economic

One study conducted in the United States modeled the costs and benefits of a proposed national smoke-free environment act to restrict or ban smoking inside all non-residential buildings regularly entered by 10 or more persons per week. ⁸² Costs included implementation of the restriction or ban by the establishment, construction, and maintenance of smoking lounges, and enforcement. Benefits included savings on medical costs by averting heart disease, value of lives saved obtained by willingness-to-pay methods, costs averted by reduced smoking-related fires, and productivity improvements. The net present benefit to society adjusted to the Community Guide reference case was in the range of $42 to $78 billion. This range was based on high and low estimates of benefits and costs. This study was classified as very good, based on the criteria for quality assessment of economic data used in the Community Guide. ⁹⁰ (See also Appendix C, Interpreting the Economic Data.)

Barriers to intervention implementation

A major barrier to efforts by local governments to adopt smoking bans is preemption, which is the passage or presence of a state law with weaker smoking restrictions that prevents implementation and enforcement of stronger local laws. ^{91 92} Eliminating preemption statutes is one of the tobacco objectives of Healthy People 2010. ¹⁹ Another major barrier to the adoption of local, state, and national smoking bans is political opposition by smokers, businesses concerned about potential changes in revenue, and tobacco industry-sponsored groups. ⁹³

Conclusion

According to the Community Guide's rules of evidence, there is strong scientific evidence that smoking bans and restrictions reduce exposure to environmental tobacco smoke in the workplace.

Definition

Community education includes all efforts to increase knowledge and to change attitudes about the health effects of exposure to environmental tobacco smoke. Techniques include mass media messages, small media messages (including educational materials), and counseling provided outside of health care settings.

Background

Community education provides information to parents, other occupants, and visitors to the home about the health risks of ETS for nonsmoking adults and for children. For infants and children, most ETS exposure occurs in the home. ⁹⁴ Information could change the knowledge and attitudes of smokers, prompting them to reduce or eliminate smoking indoors, reduce consumption, or quit entirely. Non-smokers might increase their support and encouragement to smoking household members to quit, or create and enforce home smoking bans or restrictions. The combination of reduced indoor smoking and increased cessation would result in a reduction in indoor ETS exposure, with a consequent reduction in morbidity and mortality.

Review of evidence: effectiveness

Our search identified a total of 3 studies that evaluated effectiveness of education in the adoption of home smoking policies, in reducing ETS exposure in the home, or in changing tobacco use behaviors. ^{44, 95, 96} Two studies were excluded because of limitations in execution, ⁹⁶ or a least suitable study design. ⁴⁴ Details of the qualifying study ⁹⁵ are available at the website: www.thecommunityguide.org.

The single qualifying study evaluated a randomized trial of home nurse visits to assist families in reducing infant exposure to ETS. The nurse visits provided a variety of home education aids and an opportunity to discuss options for home smoking policies. At a 12-month follow-up the absolute percentage difference in reported infant exposure to ETS was -4 percentage points (relative decrease of 12%). However, households in the intervention group reported an increase in infant exposure to ETS over the period of the study, from 33% at baseline to 49% at 12-month follow-up.

Review of evidence: applicability

The only qualifying study evaluated home nurse visits to families in central North Carolina. ⁹⁵

Review of evidence: other positive or negative effects

Potential benefits of education to reduce ETS exposure in the home include changes in tobacco use behaviors such as an increase in cessation attempts and successful cessation. No harms of community education to reduce ETS exposure in the home were identified in the literature or by the chapter development team.

Review of evidence: economic

Economic evaluation information was not reviewed because there was insufficient evidence of effectiveness of the intervention.

Barriers to intervention implementation

No barriers were identified to the implementation of community education strategies focused on reducing ETS exposure in the home.

Conclusion

According to the Community Guide's rules of evidence, the evidence of effectiveness of education strategies in reducing exposure to ETS in the home environment is insufficient because of the small number of available studies and limitations in the design and execution of available studies.

Effectiveness

The effectiveness of smoking bans and restrictions in reducing exposure to environmental tobacco smoke in the workplace is established. Research issues, which have been identified by others, ²⁰ overlap with questions generated as a result of this review.

• What are the relative effects of smoking bans and restrictions on tobacco use behaviors? What is the full range of effects on tobacco use behaviors that occurs in response to smoking bans and to smoking restrictions? In addition to reducing exposure to ETS, what are the effects on tobacco use cessation and tobacco use prevalence?
• What are the extended effects (beyond 1-2 years post implementation) of smoking bans and restrictions on the tobacco use behaviors of workers? Do these effects increase over time?
• What is the full range of effects that occurs in communities in response to different types of local ordinances?
• How does the effectiveness of smoking policies vary by the specific requirements of the legislation and vigorousness of enforcement?
• How is the impact of workplace smoking policies affected by the size and composition of the workforce?
• What is the full range of health benefits that results from reducing or eliminating exposure to ETS in workers and customers currently exposed to ETS on a regular basis (e.g., in restaurants, bars, and casinos)?
• How do cultural characteristics of businesses and workers contribute to increased or decreased effectiveness of smoking bans and restrictions?

The effectiveness of community education interventions in reducing exposure to ETS in the home has not been established. Basic research questions proposed by others ^{94, 97} overlap the questions generated from this review.

• How effective are educational methods in reducing exposure to ETS in the home?
• What are the relative contributions to reducing home ETS exposures of (1) adherence to policies that ban or restrict smoking in the home? and (2) smoking cessation?
• Do policies in the home that ban or restrict smoking reduce exposure to ETS? In adults? In children? Are households with children more likely to adopt policies that ban or restrict smoking in the home?
• Are home smoking bans more effective than smoking restrictions?
• What information or message is effective in prompting and maintaining practices in the home?
• What channels are effective for dissemination of information to reduce home ETS?

Applicability

Workplace smoking bans and restrictions should be applicable in most workplace settings and populations. However, possible differences in the effectiveness of each intervention for specific subgroups of the population could not be determined. Several questions regarding applicability of these interventions in settings and populations other than those studied remain.

• Are smoking bans effective in high schools in reducing exposure to ETS and/or tobacco use?
• Are smoking bans and restrictions effective in universities in reducing exposure to ETS and/or tobacco use?
• Are smoking bans effective in child care settings in reducing exposure to ETS?
• Do meaningful differences exist in effectiveness of smoking bans and restrictions based on the level or scale of implementation (private, local, state, national)?

Other positive or negative effects

Smoking bans and restrictions may have important effects on such tobacco use behaviors as consumption, cessation attempts, and cessation. Assessing the full range of effects of smoking bans and restrictions is important and was addressed above under "Effectiveness." In addition, research on the following issues would be useful:

• How effective are workplace smoking bans and restrictions in reducing relapses?
• Do smoking bans and restrictions divert tobacco consumption from cigarettes to smokeless tobacco?
• To what extent, if any, do workplace smoking bans and restrictions increase consumption and ETS exposures in the home?
• What effects do workplace smoking bans and restrictions have on productivity?

Economic evaluations

The available economic information consisted of a single evaluation. Considerable research is therefore warranted regarding the following questions:

• What are the costs of these interventions?
• What is the cost-effectiveness, net cost, or net benefit of smoking bans and restrictions when the cost-effectiveness analysis includes cost of illness averted?

Barriers

Research issues important to communities and local governments identified in this evaluation include:

• What aspects of efforts to prevent or to overturn state preemption laws are effective?
• What aspects of efforts to pass local smoking bans are effective in addressing local concerns and industry-organized opposition? What arguments for smoking bans are most persuasive to voters? To local legislative bodies?

Definition

Interventions to increase the unit price for tobacco products include legislation at the state or national level to raise the product excise tax. Although other factors affect tobacco product pricing, excise tax increases have historically resulted in an equivalent or larger increase in tobacco product price. ⁹⁹

Background

Excise taxes on tobacco products make the use of tobacco products less attractive to adolescents and young adults who have limited resources and a variety of options for spending available money. Excise tax increases have primarily occurred in individual states, as most attempts at the federal level have been unsuccessful. State excise tax increases have occurred as a result of legislative action, and in some states as a result of statewide referendum. Referenda passed in California, Massachusetts, Oregon, and Arizona provided various proportions of excise tax funds by mandate or recommendation to support statewide education programs and mass media campaigns. ¹⁰⁰

Review of evidence: effectiveness

Our search identified 8 studies regarding the effectiveness of increasing the price of tobacco products on changing the tobacco use behaviors of adolescents and/or young adults. ^{101 102 103 104 105 106 107 108} Three additional papers provided more information on an already-included study. ^{109 110 111} All of the studies were conducted in the United States. Five studies examined the effect of product price on tobacco use in adolescents (13-18 years), ^{103 104 105 106 108} and three studies examined the effect of product price on tobacco use in young adults (18-24 years, or < 25 years). ^{101 102 107} All of the identified studies were of moderate or greatest suitability of study design and fair or good quality of execution. Details of these 8 qualifying studies are provided at the website: www.thecommunityguide.org.

All eight studies employed econometric methods in analysis of single or sequential cross-sectional surveys of populations of students and/or young adults. Local tobacco product price and price changes or differences over the period of study were combined with the survey responses on tobacco use and consumption to calculate price elasticity of demand estimates (the percentage change in quantity demanded resulting from a 1% change in price). Price elasticity of demand estimates provided in these studies included participation (i.e., tobacco use prevalence), tobacco product consumption (such as cigarettes smoked per day), and an overall estimate (participation and consumption). A negative price elasticity of demand estimate reflects a decrease in tobacco use in response to an increase in tobacco product price. All of the studies attempted to control for concurrent tobacco prevention and control efforts including differences in smoking restrictions, youth tobacco access laws, school tobacco education programs, and exposure to anti-tobacco media.

The study periods differed for each qualifying study, with some overlap. Five studies evaluated the effect of price on tobacco use for study periods that included the 1990s, ^{101 102 103 104 106} while three studies reported the effect of price on tobacco use for periods before 1990. ^{105 107 108}

Price elasticity of demand estimates from seven studies ^{101 102 103 104 105 106 107} demonstrated that higher tobacco product prices are associated with lower levels of tobacco use by adolescents and young adults. One study ¹⁰⁸ did not find a statistically significant effect of price on adolescent tobacco use, after controlling for such tobacco use regulations as smoking restrictions. For tobacco use prevalence, the price elasticity estimates ranged from "no statistically significant effect" to -1.19 with a median of -0.37 (suggesting that a 10% increase in product price would result in a 3.7% decrease in the prevalence of tobacco use among adolescents). For tobacco consumption (6 studies provided measurements), the price elasticity estimates ranged from 0 to -0.68 with a median of -0.23 (suggesting that a 10% increase in product price would result in 2.3% decrease in the quantity of product consumed by adolescent users).

The subset of five studies conducted on surveys from adolescents (13-18 years) ^{103 104 105 106 108} determined price elasticity of demand estimates for participation (prevalence) that ranged from "no statistically significant effect" to -1.19 with a median of -0.38. Four of these studies also reported estimates for tobacco consumption, ranging from 0 to -0.47 with a median of -0.27.

In the subset of three studies evaluating populations of young adults (18-24 years), ^{101 102 107} the price elasticity measurements were similar to those observed in studies on adolescents. Price elasticity of demand for tobacco use participation ranged from -0.07 to -0.52 with a median of -0.37. For the effect on consumption, two studies reported estimates of of -0.21 and -0.68.

Overall, the price elasticity of demand estimates in 7 of 8 studies demonstrate that increases in tobacco product price result in decreases in both the overall prevalence of tobacco product use and the quantity consumed. Increases in product price resulted in reductions in tobacco use in both adolescents and in young adults.

Review of evidence: applicability

The same body of evidence used to assess effectiveness was used to assess the applicability of these interventions to different tobacco products, settings, and populations. All of the qualifying studies measured differences in tobacco product price across jurisdictions such as states. Differences in tobacco product prices included, but were not limited to, differences in state excise taxes.

Studies have evaluated the effect of product price on use and consumption of cigarettes ^{101 102 103 104 105 106 107 108} and of smokeless tobacco products. ^{107 110}

All of the studies were conducted in the United States and most of the studies used national datasets. The study samples are representative of populations of adolescents and young adults. In addition, some studies reported stratified analyses, demonstrating evidence of effectiveness of price on tobacco use and consumption among whites, ^{101 103 104} blacks, ^{101 103} and Hispanics. ¹⁰¹ Two studies observed that both black adolescents and young adults were more responsive to differences in product price than were white adolescents and young adults, respectively. ^{101 103} Studies providing analysis by gender found that tobacco product price increases had a greater effect among males than among females. ^{101 103 104 106}

Studies conducted on nationally representative population samples suggest that the evidence of effectiveness should apply to most adolescents and young adults in the United States.

Review of evidence: other positive or negative effects

Increases in the price of tobacco products also reduce tobacco use in adults. A review of this body of evidence and additional positive and negative effects are presented in "Part III: Strategies to Increase Tobacco Use Cessation." No information about other positive or negative effects, relevant to reducing initiation in adolescents, was identified.

Review of evidence: economic

Econometric analyses were used to evaluate the effectiveness of this economic intervention. The results are reported above in the section, "Review of evidence: effectiveness."

Barriers to intervention implementation

Excise tax increases require passage of legislation or statewide referendum. Political opposition is well organized and funded at both the federal and state levels. Published reports provide information on the components and experiences of both successful and unsuccessful state initiatives that proposed an excise tax increase on tobacco products. ^{100 112 113}

Conclusion

According to the Community Guide's rules of evidence, strong scientific evidence demonstrates the effectiveness of increasing the price of tobacco products on reducing tobacco use prevalence and consumption among both adolescents and young adults.

Definition

Campaigns are mass media interventions of an extended duration, using brief, recurring messages to inform and motivate individuals to remain tobacco-free. Message content is developed through formative research, and message dissemination includes the use of paid broadcast time and print space, donated time and space (as public service announcements), or a combination of paid and donated time and print space. Mass media campaigns can be combined with other interventions, such as tobacco product excise tax increases, school-based education, or other community programs.

Background

Mass media techniques primarily include broadcast messages on television and radio, although other formats such as billboards, print media, and movies have been used. Campaigns can focus on messages targeting children and adolescents or can include such messages as part of an overall anti-tobacco effort (e.g., including messages targeting tobacco users to increase cessation and messages about reducing exposure to ETS). The content of mass media campaigns designed to educate and motivate children and adolescents to remain tobacco-free can vary, but a recent review identified two primary strategies: agenda setting and demand reduction education. ¹¹⁴ Agenda-setting messages increase awareness of strategies used by the tobacco industry to promote tobacco use, and attempt to facilitate changes in both tobacco use behaviors and public policies concerning tobacco. Demand reduction education messages provide information and support to young people to help individuals decide to remain tobacco-free.

Review of evidence: effectiveness

Our search identified 14 studies regarding the effectiveness of mass media campaigns in reducing tobacco use in adolescents. ^{115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128} Two studies were excluded from the analysis of effectiveness because of least suitable study designs. ^{123 126} Details regarding the 12 qualifying studies are provided at the website: www.thecommunityguide.org. Twenty-eight papers provided additional information on an already-included study. ^{129, 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 153, 154, 155, 156}

The qualifying studies evaluated a variety of mass media campaigns. Nine studies evaluated interventions in which the mass media component focused on youth. ^{115, 116, 117, 118, 119, 120, 121, 122, 124} In three studies, the mass media effort included youth-targeted messages within a larger anti-tobacco campaign. ^{125, 127, 128} Only one study employed mass media alone (through a variety of outlets). ¹²⁰ In 11 studies, the mass media campaign occurred in coordination or concurrently with other interventions including contests, ^{115, 122} school-based education programs, ^{116, 117, 118, 119, 121, 122, 124, 125, 128} and community education programs. ^{116, 121, 124, 125, 127, 128} Two studies were conducted in settings with recognized excise tax increases on tobacco products. ^{124, 127}

The duration of the intervention differed significantly among the studies. Two of the mass media efforts were less than 3 weeks in duration, ^{117, 118} three interventions were less than two years in duration, ^{115, 121, 122} and seven campaigns were two or more years in duration. ^{119, 120, 124, 125, 127, 128, 133}

The 12 qualifying studies provided 12 measurements of tobacco use. Five studies reported differences in rates of self-reported tobacco use between intervention and comparison groups. ^{115, 116, 122, 124, 125,} In follow-up periods that ranged from 2 to 5 years, the reported absolute percentage differences ranged from +0.02 to -9.5 percentage points with a median of -2.4 percentage points. Six studies reported intervention and comparison group outcomes expressed as an odds ratio for tobacco use at follow-up. ^{117, 118, 119, 120, 121, 127,} Two studies did not find any effect on tobacco use behaviors from exposure to the intervention. ^{117, 118,} In four studies, the follow-up periods ranged from 2 to 4 years, and the adjusted odds ratios were similar in magnitude and direction (range 0.49 to 0.74: median 0.60). ^{119, 120, 121, 127,} Finally, one study observed an absolute percentage difference of -11 percentage points in group mean tobacco use prevalence at 15-year follow-up for students who received a school education program and were exposed to community and mass media education interventions to reduce cardiovascular disease risk factors. ^128,

All seven of the studies evaluating mass media campaigns of two or more years' duration observed a reduction in tobacco use prevalence in the intervention group when compared to the control group. ^{119, 120, 124, 125, 127, 128, 133,} The observed differences were more consistent and slightly greater in magnitude of effect for both tobacco use prevalence (range -2.4 to -11, median -8.0 percentage points) and in the odds ratios (range of outcomes 0.49 to 0.74, median result 0.74).

The contributions of individual components to the overall effectiveness of the interventions cannot be attributed.

Review of evidence: applicability

The same body of evidence used to assess effectiveness was used to assess the applicability of these interventions to different settings and populations. Interventions were performed in the United States, ^{115, 116, 117, 118, 119, 121, 122, 124, 125, 127,} Norway, ^120, and Finland. ^128, Interventions included statewide campaigns in Florida, ^116, Massachusetts, ^127, and Minnesota. ^{124, 125,} Community and regional interventions were conducted in the Southeast, ^115, Northeast, ^119, Midwest, ^121, and ^122, in Montana ^119, and Southern California. ^{117, 118,} Interventions were initiated in the 1990s, ^{116, 120, 127,} in the 1980s, ^{115, 117, 118, 119, 121, 122, 124, 125,} and in the 1970s. ^128, Outcomes were evaluated in representative samples of adolescents identified in the general population ^{115, 127,} and in representative or selected samples of schools. ^{116, 117, 118, 119, 120, 121, 122, 124, 125, 128} Outcomes were evaluated in student populations recruited or surveyed before grade 6, ¹¹⁹ grade 6-7, ^{116, 117, 118, 121, 122, 125, 128} or after grade 7. ^{116, 120, 124, 125} Outcomes from the two most recent studies demonstrate a greater benefit among younger adolescents. ^{116, 127} Evaluation of the Florida campaign provides evidence of effectiveness of mass media campaigns among girls, boys, whites, blacks, and Hispanics. ^{116, 133}

Review of evidence: other positive or negative effects

Mass media campaigns are effective in reducing population consumption of tobacco products and in increasing cessation in adult tobacco users (see Part III: Strategies to Increase Tobacco Use Cessation). Messages and broadcast times that are effective for children and adolescents, however, may have less or no effect on adult tobacco use. No other positive or negative effects were identified in this review.

Review of evidence: economic

One four-year study ¹⁴⁸ conducted in Montana, New York, and Vermont compared the effect of a mass media campaign combined with a school smoking prevention program to a school smoking prevention program alone, on students from grades 5-7 through grades 8-10. Students were followed for two years after conclusion of the intervention. Costs included personnel, travel, data entry, message research and development, and television and radio advertising. The absolute percentage point change was -5.5 percentage points in smoking initiation (tobacco use prevalence was 25% in the comparison groups at follow-up). Adjusted program cost per smoker averted was $6,069, and the adjusted program cost per quality-adjusted life year (QALY) was $333. Based on the explicit quality assessment criteria used in developing the Community Guide, this study was classified as very good. ⁹⁰ Study details, adjusted results, and quality scoring are listed online in the economic summary table provided at the website: www.thecommunityguide.org. (See also Appendix C, Interpreting the Economic Data.)

Barriers to intervention implementation

The main barrier to implementation of mass media campaigns is the cost of purchasing broadcast time. The costs of developing and test marketing messages can be offset by cooperation between tobacco prevention and control programs. Programs can lower message development costs, for example, by using existing television, radio, print, and outdoor ads from CDC's Media Campaign Resource Center, a clearinghouse of high-quality materials produced by states and other organizations. ⁵

Conclusion

According to the Community Guide's rules of evidence, strong scientific evidence exists that mass media campaigns are effective in reducing tobacco use prevalence in adolescents, when combined with other interventions. The contribution of individual components to the overall effectiveness of these interventions cannot be attributed.

Effectiveness

The effectiveness of increasing the unit price of tobacco products and mass media campaigns in reducing tobacco use by adolescents is established. Important questions remain regarding the composition and content of effective campaigns and the effectiveness in different settings and populations. Some issues raised by others ^{4 157} overlap with questions generated as a result of this review.

• What interventions are most effective in combination with mass media campaigns? What interventions are least effective?
• What are the relative effects of these interventions on adolescent initiation, consumption, access to tobacco products, and cessation?
• What is the required intensity (frequency of spots and the broadcast exposure) of media messages for an effective campaign?
• What are the independent contributions of particular intervention features (e.g., components, content, intensity, and duration) to overall intervention effectiveness?
• What are the most effective ways to maintain reductions in youth tobacco use into young adulthood?
• Does tobacco use in adults respond to mass media campaigns that are youth-focused?

Applicability

The effectiveness of these interventions should be applicable in most settings and populations. However, there could be differences in the effectiveness of these interventions for specific subgroups of the population. Remaining questions regarding the applicability of these interventions in various settings and populations are:

• Are there differences in the responses of adolescents to tobacco product price increases by age, race, and ethnicity?
• Are the effects of mass media campaigns on adolescents by gender, race, and ethnicity similar to or different from those observed in Florida?

Other positive or negative effects

The studies in these reviews did not provide information on other positive or negative effects. Research questions pertinent to interventions to increase the price of tobacco products are presented in the section on Strategies to Increase Tobacco Use Cessation. Some issues generated by the review of mass media campaigns are:

• Do mass media campaigns that target children and adolescents result in increases in tobacco initiation among young adults, by delaying the age of initiation?
• What are the most effective ways to maintain reductions in youth tobacco use into young adulthood?

Economic evaluations

Available economic information was limited to a single study of mass media campaigns. Therefore, considerable research is warranted regarding the following questions:

• Are the costs and cost-effectiveness, net cost, or net benefit of mass media campaigns similar to or substantially different from those that have been previously reported?
• How do the costs per tobacco user averted compare with other tobacco prevention strategies?
• How do specific characteristics of mass media campaigns contribute to economic efficiency?
• What combinations of components in multicomponent interventions are most cost-effective?

Barriers

The strategies evaluated in this section require political action and support. Research issues generated in this review include:

• What characteristics are effective in successful legislative and referendum campaigns?
• How can adequate funding levels be maintained for mass media campaigns?

Methods of intervention research

Evaluations of mass media campaigns should provide information on the costs, scale, duration, and content of the campaign. Researchers should identify concurrent tobacco control efforts, especially excise taxes or changes in tobacco product price, and attempt to control for these changes in their analyses.

Definition

Interventions to increase the unit price of tobacco products include municipal, state, and federal legislation that raises the excise tax on these products. Although other factors also affect tobacco product pricing, excise tax increases historically have resulted in an equivalent or larger increase in tobacco product price. ⁹⁹

Background

Excise taxes on tobacco products make the continued use of tobacco products less attractive to users. In the United States, excise tax increases have primarily occurred in individual states, as most legislative attempts at the federal level have been unsuccessful. State excise tax increases have occurred as a result of legislative action, and in some states as a result of statewide referendums. Referenda passed in California, Massachusetts, Oregon, and Arizona provided various proportions of excise tax funds by mandate or recommendation to support statewide education programs and mass media campaigns. ¹⁰⁰

Review of evidence: effectiveness

Our search identified 56 papers regarding the effectiveness of increasing the price of tobacco products on tobacco use behaviors among individual smokers and general populations. ^{101, 107, 108, 111, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 187, 188, 189, 190, 191, 192, 193, 194, 195, 196, 197, 198, 199, 200, 201, 202, 203, 204, 205, 206, 207, 208, 209} A number of papers, however, conducted analyses on the same datasets for similar, identical, or overlapping periods of time. For the purposes of this evaluation, the Task Force consolidated papers into aggregate studies based on similarities in location, the period of study, and the dataset employed (details available at www.thecommunityguide.org). For example, we considered all of the identified papers evaluating activities in California in the period 1989-1994 a single aggregate study. ^{158, 159, 160, 161, 162, 163, 164} After consolidating overlapping papers, the body of evidence consisted of 22 aggregate studies. Five studies were excluded because of limitations in execution quality. ^{204, 205, 206, 207, 208} Details of the 17 qualifying studies, including the selection of representative outcome measurements, are available at www.thecommunityguide.org.

The 17 qualifying studies include consolidated evaluations of the effect of tobacco product price on tobacco use in the states of California, ^{158, 159, 160, 161, 162, 163, 164} Massachusetts, ¹⁶⁵ Oregon, ¹⁶⁶ 11 Western states, ¹⁶⁷ national evaluations conducted in the 1990s, ^{101, 168, 169} national evaluations conducted in the 1980s, ^{107, 108, 111, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184} national evaluations conducted in the 1970s, ^{185, 186, 187, 188, 189, 190, 191, 192, 193} three studies conducted in Canada, ^{194, 195, 196} three studies conducted in the United Kingdom, ^{197, 198, 199} and single studies conducted in Austria, ²⁰⁰ Finland, ²⁰¹ Switzerland, ^{202, 203} and New Zealand. ²⁰⁹

The most common measurement described in the body of evidence to estimate the effect of tobacco product price increases was the price elasticity of demand (the percentage change in consumption that results from a 1% change in price). In econometric analysis of tobacco product price increases (resulting from an excise tax increase, a tobacco industry price increase, or both), a negative price elasticity of demand estimate indicates a decrease in consumption. In studies that measure changes in tobacco product consumption (typically estimated from national or regional tobacco sales figures), observed changes capture the combined effect of increases in cessation, decreases in initiation, and reduced consumption by continuing users. In studies that use data from individuals to measure tobacco use prevalence and quantity consumed, the price elasticity of demand calculations include distinct estimates for participation (i.e., tobacco use prevalence) and tobacco product consumption (e.g., cigarettes smoked per day), and overall estimates (participation and consumption).

Thirteen of the seventeen consolidated studies included measurements of price elasticity of demand. ^{101, 107, 108, 111, 111, 158, 159, 160, 161, 162, 163, 164, 167, 168, 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 187, 188, 189, 190, 191, 192, 193, 194, 197, 198, 199, 200, 201, 202, 203, 209} Ten aggregate studies provided measurements based on cigarette sales data, with a range of price elasticity estimates of -0.27 to -0.76 and a median estimate of -0.41 (suggesting that a 10% increase in product price would result in a 4.1% decrease in population consumption). ^{101, 107, 108, 111, 158, 159, 160, 161, 162, 163, 164, 167, 168, 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 187, 188, 189, 190, 191, 192, 193, 194, 200, 201, 202, 203, 209} Seven studies ^{101, 107, 108, 111, 158, 159, 160, 161, 162, 163, 164, 168, 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 187, 188, 189, 190, 191, 192, 193, 197, 198, 199} provided measurements based on individual responses, although one study provided only an overall price elasticity of demand estimate. ¹⁹⁹ For the overall price elasticity estimate the values ranged from +0.5 to -0.84 with a median of -0.42. For the price elasticity of participation, the range was -0.08 to -0.26 with a median of -0.15. For the estimate of quantity consumed, the values ranged from -0.09 to -0.61 with a median of -0.19.

Overall, this body of evidence documented consistent effects of increases in tobacco product price on reducing tobacco use, regardless of the measurements reported or calculated, the setting or period of time evaluated, or differences in the control of potential confounders. Three aggregate studies, which used information collected in the 1990s, provide more recent price elasticity of demand estimates of the effect of tobacco product price increases on reducing tobacco use. ^{101, 158, 159, 160, 161, 162, 163, 164, 168, 169, 194} Price elasticity estimates from tobacco sales figures provided in these studies range from -0.30 (state tax elasticity in California) to -0.42 (US national estimate) with a median of -0.40. Results from two papers within the US national set ^{101 169} based on individual responses in two national surveys were very similar, with estimates of -0.29 and -0.25 overall, -0.14 and -0.15 for participation, and -0.15 and -0.10 for quantity consumed.

Four studies provided measurements other than price elasticity of demand in assessing the impact of tobacco product price changes on tobacco use. ^{165, 166, 195, 196} One study reported the effect of federal and provincial excise tax increases on tax paid sales of cigarettes in Alberta, Canada. ¹⁹⁶ In the period 1985-1995, the price index for cigarettes increased from 1.00 (1985-1986 baseline) to 1.78 while the consumption index decreased from 1.00 to 0.58. A second study from Canada in 1994-1995, following a federal excise tax reduction, compared tobacco use in survey samples in provinces that reduced excise taxes on cigarettes with provinces that did not. ¹⁹⁵ Although tobacco use declined in all of the provinces over the study period, the rate of decline slowed significantly in provinces that reduced the excise tax on cigarettes. Finally, the states of Massachusetts and Oregon reported reductions in population consumption of tobacco following increases in the excise tax for tobacco products and implementation of multicomponent tobacco control programs that included a mass media campaign. ^{165 166} Compared to the rest of the United States, the relative percent decreases in tobacco consumption were -12.8% and -9.8% over periods of 4 years and 2 years, respectively. The independent effect of the excise tax change on tobacco consumption could not be determined in these studies.

Review of evidence: applicability

The individual papers included in the qualifying studies were used to provide information on the applicability of this intervention to different products, settings, and populations. All of the qualifying studies evaluated the effect of price on the consumption of cigarettes. In addition, studies evaluated the effect of price increases on consumption of smokeless tobacco, ^{107, 111} cigars, ²⁰¹ and pipe tobacco. ²⁰¹ Studies demonstrated the effectiveness of state excise tax increases ^{159, 161, 162, 165, 168, 177} and federal excise tax increases. ^{160 168} Most studies using individual data employed representative samples of state or national populations. Studies have demonstrated effectiveness in stratified analyses for whites, ¹⁰¹ blacks, ^{101, 162} Hispanics, ^{101, 162} men, ^{101, 162, 187, 191, 198} and women. ^{101, 187, 191, 198, 199} Studies also demonstrated effectiveness in persons with incomes below median, ^{101, 169} less than a high school education, ¹⁸⁸ and across most social classes. ^{198, 199}

Review of evidence: other positive or negative effects

Increasing the unit price for tobacco products will decrease tobacco use in adolescents and young adults (see "Part II: Strategies to Reduce Tobacco Use Initiation"). Increases in illegal cross-border transport and sale of untaxed tobacco products (smuggling) are potential negative effects of increases in tobacco product excise taxes. No recent studies of the effects of organized smuggling in the United States were identified in this review, and an analysis in 1985 reported a significant reduction in activity following federal legislation in 1978. ²¹⁰

This review identified several additional effects of tobacco product price increases that might reduce (but would not eliminate) the potential health benefits caused by increases in cessation and reductions in consumption. These effects include: (1) legal, individual cross-border purchases of tobacco products; (2) substitution of tobacco products (e.g., smokeless tobacco for cigarettes) created by unequal taxation on different kinds of tobacco products; ¹¹¹ and (3) modification of individual tobacco use behaviors, such as smoking cigarettes longer or changing to a higher-tar, higher-nicotine brand. ¹⁷⁴

Review of evidence: economic

Econometric analyses were used to evaluate the effectiveness of this economic intervention. The results are reported above in the section, "Review of evidence: effectiveness."

Barriers to intervention implementation

Excise tax increases require passage of legislation or statewide referendum. Efforts to increase the federal excise tax have largely been unsuccessful. Political opposition is well organized and funded at both the federal and state levels. Reports of state tobacco tax initiatives, successful and unsuccessful, have been published. ^{17, 100, 112, 113}

Conclusion

According to the Community Guide's rules of evidence, strong scientific evidence demonstrates that increasing the unit price for tobacco products is effective in increasing tobacco use cessation and in reducing consumption.

Definition

Campaigns are mass media interventions of an extended duration, using brief, recurring messages to inform and to motivate tobacco product users to quit. Message content is developed through formative research, and message dissemination includes the use of either paid air time and print space (advertisements), donated time and space (public service announcements), or a combination of the two. Campaigns can be combined with other interventions, such as an excise tax increase, or additional community education efforts.

Background

The historical foundation for mass media education to reduce tobacco use is the national experience from 1966 through 1970 of a long-duration, high-intensity anti-smoking broadcast campaign that resulted from a complaint to the Federal Communications Commission (FCC). The complaint requested application of the Fairness Doctrine to mandate reply time to counter cigarette advertising on television and radio. In June 1967, the FCC initiated a mandate (upheld on appeal) that required broadcasters to provide broadcast time free of charge for anti-smoking messages. At the peak, anti-smoking messages were broadcast (often in prime time) at a rate of one message for every three cigarette commercials. This policy lasted until 1971, when a ban on cigarette broadcast advertising went into effect. Cigarette consumption declined each year during the campaign and rose again after the cigarette advertising ban ended free access to broadcast time for anti-smoking messages. ^{211 212 213}

Review of evidence: effectiveness

Our search identified 24 studies regarding the effectiveness of mass media campaigns in reducing tobacco use. ^{164, 165, 166, 214, 215, 216, 217, 218, 219, 220, 221, 222, 223, 224, 225, 226, 227, 228, 229, 230, 231, 232, 233, 234} An additional 14 papers provided more information on an already-included study. ^{152, 161, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246} Four papers with limited quality of execution, ^{214, 219, 227, 233} and two papers with least suitable study designs ^{216, 232} were not included in the body of evidence. Three studies measured the effect of a mass media campaign in increasing use of a telephone cessation information service, ^{215, 229, 231} and these results are described later in this section, in the review of multicomponent interventions including patient telephone support. Details of the remaining 15 qualifying studies are provided at the website: www.thecommunityguide.org.

All of the qualifying studies evaluated the effectiveness of a mass media campaign either coordinated with or concurrent with other interventions. These other interventions included an excise tax increase (6 studies), ^{164, 165, 166, 220, 223, 225} community education programs such as the distribution of self-help cessation information (12 studies), ^{164, 165, 166, 217, 218, 220, 223, 224, 225, 228, 230, 234} individual or group cardiovascular disease risk factor reduction or smoking cessation counseling (7 studies), ^{218, 221, 222, 223, 224, 226, 228} and other mass media efforts (2 studies). ^{218, 234}

The 15 qualifying studies provided 15 measurements of changes in tobacco use, consisting of 5 measurements of changes in individual tobacco use cessation, 3 measurements of changes in population consumption (measured by statewide sales of cigarettes), and 7 measurements of changes in the prevalence of tobacco use in the study population. Two studies ^{164 165} provided both measurements of population consumption and tobacco use prevalence. For these studies, the measurements of population consumption are reported.

Five studies evaluated the effectiveness of mass media campaigns combined with additional interventions in increasing tobacco use cessation in groups of recruited tobacco users. ^{217, 221, 224, 226, 228} The duration of the interventions ranged from less than 1 year to 5 years. Cessation rates in the intervention group ranged from 3.9% (confirmed) to 50% (self-reported), with a median of 7% in follow-up periods of 6 months to 5 years (median, 14 months). The absolute percentage differences in cessation between intervention group participants and comparison group participants (who were potentially exposed to the media component in 3 studies) ranged from -2 percentage points to +35 percentage points, with a median difference of +2.2 percentage points.

Three studies evaluated the effectiveness of mass media campaigns in reducing tobacco consumption (as measured by state-wide sales of cigarettes) in a state population. ^{164, 165, 166} These studies evaluated the effect of ongoing state-funded mass media campaigns coordinated with excise tax increases and funding for community and school-based education programs. All three studies, with follow-up periods of 2-8 years, observed decreases in state per capita consumption of cigarettes compared to per capita consumption in the rest of the United States. The observed differences ranged from -9 cigarette packs per capita per year (-9.8% relative decrease) to -20.4 packs per capita per year (-17.5% relative decrease), with a median of -15 packs per capita per year (-12.8% relative decrease).

Seven studies evaluated the effectiveness of mass media campaigns in reducing the prevalence of tobacco use in study populations. ^{218, 220, 222, 223, 225, 230, 234} These include investigations conducted in California (including targeted campaigns for Vietnamese and Hispanic populations as part of the larger California media campaign previously described), ^{218, 220, 223, 225} Minnesota, ²²² Australia, ²³⁰ and Finland. ²³⁴ Five studies reported differences in tobacco use prevalence compared with a concurrent population, ^{218, 220, 222, 225, 234} and two studies reported before-after changes in a single study population. ^{223, 230}

Six of the seven studies observed decreases in tobacco use prevalence, ^{220, 222, 223, 225, 230, 234} over study periods that ranged from 6 months to 20 years (median 6 years). In the five studies with concurrent comparison populations, absolute percentage differences in tobacco use prevalence reported ranged from +0.2 to -7 percentage points (median -3.4 percentage points). ^{218, 220, 222, 225, 234} In the two studies without a concurrent comparison population, the absolute percentage changes observed over time were -2.5 ²³⁰ and -4.7 percentage points. ²²³

Review of evidence: applicability

The same body of evidence used to assess effectiveness was used to assess the applicability of these interventions to different settings and populations. All of the qualifying studies focused on differences or changes in the consumption and use of cigarettes. Studies included campaigns conducted nationwide in Scotland, ²²¹ statewide in the United States, ^{164, 165, 166} region-wide in the United States ²¹⁷ and Finland, ²³⁴ in large cities, ²³⁰ and in smaller communities and localized populations. ^{218, 220, 222, 223, 224, 225, 226, 228}

Evidence of effectiveness from the state campaigns should be generally applicable to the U.S. population. Studies have been performed among specific U.S. populations including Hispanics ^{223, 224} and Vietnamese men. ^{220, 225}

Review of evidence: other positive or negative effects

Several studies ^{215, 217, 229, 231} observed significant increases in use of telephone cessation information or support services when the mass media messages directed viewers to call for further information or support. The mass media campaigns in California, Oregon, and Massachusetts included messages targeting children and adolescents, and messages about the health effects of exposure to environmental tobacco smoke.

Review of evidence: economic

Two studies were included for review, one conducted in the Netherlands ²⁴⁷ and the other in Scotland. ²⁴⁸ Both studies evaluated community-wide interventions to increase cessation among adult tobacco users. Both studies reported program costs per quitter. In addition, one study ²⁴⁸ reported program cost per life-years saved, which was converted to dollars per QALY based on preference weights provided by Fiscella and Franks. ²⁴⁹

The intervention conducted in the Netherlands ²⁴⁷ evaluated the effect of a mass media campaign, self-help manual, hotline, and a 9-session group cessation program. Self-reported 7-day abstinence was determined at 6-month follow-up. Costs included were wages, overhead, calls to the hotline, participant time, transportation, and charges for the group session. The effect size was estimated as the difference in smoking prevalence before and after the intervention. The self-reported cessation rate was 11% for participants using the self-help manual alone and 22% for participants using the self-help manual combined with the cessation program. Based on this effect size, adjusted program cost per quitter ranged from $796 to $1593. Based on the explicit quality assessment criteria used in developing the Community Guide, this study was classified as satisfactory. ⁹⁰

The study conducted in Scotland ²⁴⁸ consisted of a mass media campaign, telephone helpline, and information booklet. At the 12-month follow-up, 9.8% of program participants reported having quit smoking for at least 6 months. Costs included research, production, design fees, printing, dissemination, staff salaries, and overhead. The adjusted program costs per quitter in this study ranged from $298 to $655. Adjusted program costs per QALY ranged from $151 to $328. The range was based on lower and upper bound estimates of adult participants. Based on the explicit quality assessment criteria used in developing the Community Guide, this study was classified as satisfactory. ⁹⁰ Study details, adjusted results, and quality scoring are provided at the website: www.thecommunityguide.org. (See also Appendix C, Interpreting the Economic Data.)

Barriers to intervention implementation

The primary barriers to the implementation of mass media campaigns are obtaining and preserving the funding needed to develop and to maintain an extended duration, high-intensity campaign using paid and targeted broadcast times for messages that resonate with target audiences. The barriers encountered by the tobacco control program in the state of California provide important lessons on the need for both public and political support and vigilance in maintaining an effective campaign. ^{250 251}

Cooperation between tobacco control programs can reduce program development costs. Programs can lower message development costs, for example, by using existing television, radio, print, and outdoor ads from CDC's Media Campaign Resource Center, a clearinghouse of high-quality materials produced by states and other organizations. ⁵

Conclusion

According to the Community Guide's rules of evidence, strong scientific evidence exists that mass media campaigns combined with other interventions are effective in increasing tobacco use cessation and in reducing consumption of tobacco products.

Definition

Cessation series are mass media interventions using recurring instructional segments to recruit, inform, and motivate tobacco product users to initiate and maintain cessation efforts. Cessation series can be coordinated with pre-series broadcast or print promotion, community education such as distribution of self-help cessation materials, and organization of cessation groups in the community. The series can extend for a period of several weeks to several months, and techniques include nightly or weekly segments on news or informational broadcasts providing expert advice or peer group experiences on a variety of cessation issues (e.g., dealing with the symptoms of withdrawal).

Background

The goals of smoking cessation series are to increase the number of tobacco users who attempt to quit and to increase the number of quitters who succeed. The series can provide motivation to tobacco product users to join in a community-wide targeted quit effort. Over the course of the broadcasts, viewers can receive ongoing support and assistance from cessation experts and recruited peers.

Review of evidence: effectiveness

Our search identified 20 studies regarding the effectiveness of cessation series in increasing tobacco use cessation. ^{244, 252, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270} Ten additional papers provide information on an already-included study. ^{271, 272, 273, 274, 275, 276, 277, 278, 279, 280} Ten studies with limited execution, ^{244, 252, 254, 257, 258, 261, 263, 265, 266, 268} and one study with a least-suitable study design, ²⁵⁵ were not included in the body of evidence. Details of the nine qualifying studies ^{253, 256, 259, 260, 262, 264, 267, 269, 270} are provided at the website: www.thecommunityguide.org.

All of the qualifying studies evaluated the effectiveness of cessation series combined with other interventions, such as community education (typically access to or distribution of self-help cessation manuals), ^{253, 256, 259, 260, 262, 267, 269, 270} organized cessation groups or programs, ^{256, 260, 262, 264, 267, 269, 270} or telephone cessation support. ^{256, 260} Eight studies evaluated televised cessation series broadcast over periods extending from 20 days to 3 months, ^{256, 259, 260, 262, 264, 267, 269, 270} and one study evaluated a week-long newspaper cessation series. ²⁵³

Five of the nine qualifying studies present evaluations of broadcast cessation series conducted in the Chicago metropolitan area (in one of three waves) between 1985 and 1987. ^{256, 262, 267, 269, 270} These studies were evaluated separately because of differences in the study populations or settings.

Of the nine qualifying studies, only one investigation compared exposed and unexposed groups of recruited participants. ²⁵⁹ That study observed no significant difference in self-reported cessation rates at 6-month follow-up. One study compared self-reported cessation based on recalled exposure to various campaign interventions. ²⁶⁰ Persons who recalled watching the cessation series were more likely to report sustained cessation at interview (odds ratio 1.36; 95% CI 1.13, 1.65). In five studies, smokers in both the intervention and comparison groups were exposed to the cessation series, with participants in the intervention group receiving additional interventions, such as small group sessions. ^{256, 262, 264, 267, 270} These studies reported absolute percent differences in cessation that ranged from +4 to +8 percentage points (median, +5 percentage points) with follow-up periods of 4 to 24 months (median, 12 months). However, these studies share a potential limitation in comparing smokers motivated to quit (enrolled or offered cessation group support) with smokers potentially less motivated to quit. Of the two remaining studies, one reported differences in cessation by intervention setting, with all of the participants receiving a similar intervention. ²⁶⁹ The other study reported an increase in cessation attempts in persons who recalled exposure to a week-long newspaper cessation series. ²⁵³

The qualifying studies provided insufficient evidence of effectiveness of the broadcast series in increasing tobacco use cessation. The differences in cessation observed in these studies might be the result of (1) baseline differences in motivation to quit between intervention and comparison smokers, and (2) the small group cessation sessions provided to the intervention group participants.

Review of evidence: other positive or negative effects

No additional information regarding other positive or negative effects was identified in this review.

Conclusion

According to the Community Guide's rules of evidence, the available studies provide insufficient evidence to assess the effectiveness of cessation series in reducing tobacco use. Evidence is insufficient because of (1) inconsistent results, and (2) inadequate comparison populations/groups.

Definition

Cessation contests are short-duration, community-wide events using mass media for the promotion, recruitment, and motivation of tobacco product users to participate in a targeted cessation date or period. This evaluation included contests that offered additional incentives for participation and successful cessation, as well as targeted quit events conducted without additional incentives.

Background

Cessation contests are included in this section because this intervention uses mass media (as well as small media such as posters and flyers) as the primary tool for promotion and recruitment of tobacco product users in the community. Contests can work to increase cessation in the community by changing attitudes of tobacco product users about cessation, recruiting tobacco product users to initiate a quit attempt, and motivating those who attempt to quit to remain abstinent (through incentives or by mobilizing support from family, friends, and other participants).

Review of evidence: effectiveness

Our search identified 17 studies regarding the effectiveness of tobacco cessation contests. ^{257, 281, 282, 283, 284, 285, 286, 287, 288, 289, 290, 291, 292, 293, 294, 295, 296} Three additional papers provided information on an already-included study. ^{297, 298, 299} Eleven studies measuring tobacco use cessation in contest participants in the absence of a comparison group were excluded from further evaluation. ^{281, 282, 283, 284, 286, 287, 288, 290, 293, 294, 296} Five studies with limitations in quality of execution were not included in the evaluation of effectiveness. ^{257, 285, 289, 291, 292} Details of the one qualifying study ²⁹⁵ are provided at the website: www.thecommunityguide.org.

The single qualifying study evaluated a multicomponent smoking cessation program in New York City. Interventions included a cessation manual and video, telephone cessation support, and the opportunity to participate in smoking cessation contests. The contest participation rate was 13% and entering the contest was significantly associated with self-reported cessation at 6-month followup (odds ratio 3.0; 95% CI 1.7, 5.4). Overall, the absolute percentage difference in self-reported cessation at 6-month follow-up was +3.3 percentage points compared with smokers who received only general health education materials.

Review of evidence: other positive or negative effects

No information on other positive or negative effects was identified in this review.

Conclusion

According to the Community Guide's rules of evidence, the available studies provide insufficient evidence to assess the effectiveness of cessation contests in increasing tobacco use cessation. Evidence is insufficient because of the small number of qualifying studies.

Definition

Provider reminder systems include efforts to identify patients who use tobacco products and efforts to prompt providers to discuss and/or advise patients about cessation. Techniques by which reminders are delivered include chart stickers, vital sign stamps, medical record flow sheets, and checklists. The content of provider reminders can vary, and provider reminder systems are often combined with other interventions, such as provider education and patient education. Multicomponent interventions including provider reminders are considered below, in the section on Provider Reminder + Provider Education with or without Patient Education.

Background

Reminder systems prompt providers to interact with patients about tobacco use at every encounter. Reminders aim to increase recognition of patient tobacco use status, resulting in an increase in provider delivery of advice to quit. Because even brief provider advice to quit has a demonstrated effect on patient tobacco use cessation, ^{14 300} increasing the delivery of advice by providers will increase the number of patients who quit.

Review of evidence: effectiveness

Our search identified eight studies evaluating the effectiveness of provider reminder systems when used alone. ^{301, 302, 303, 304, 305, 306, 307, 308} One paper with a least-suitable study design was not included in the body of evidence. ³⁰³ Details of the seven qualifying studies are provided at the website: www.thecommunityguide.org. The evaluated techniques for prompting providers were chart prompts or stickers, ^{301, 306, 307, 308} "expanded vital signs" that include tobacco use status, ³⁰⁵ and flowsheets. ^{304, 307} In one qualifying study, the intervention sites received an office consultation that encouraged implementation of a provider reminder system. Analysis was based on receipt of the consultation, not on actual implementation of the reminder system. ³⁰²

The 7 qualifying studies reported 4 measurements of differences in documentation of patient smoking status, ^{301, 304, 306, 308} 5 measurements of chart-documented or patient-reported provider delivery of advice to quit to tobacco-using patients, ^{301, 302, 305, 307, 308} and 1 measurement of patient smoking cessation. ³⁰⁷ For determination of patient smoking status, 4 studies ^{301, 304, 306, 308} measured absolute percentage differences of +26 to +57.6 percentage points (median, +32.5 percentage points) in periods extending from 8 to 24 months post implementation (median, 15 months). For provider delivery of advice to quit (Figure 3), 5 studies ^{301, 302, 305, 307, 308} observed differences that ranged from +7 to +31 percentage points (median, +13 percentage points) in assessments extending 2 to 24 months post implementation (median, 8 months). One study reported an absolute percentage difference in biochemically confirmed smoking cessation of +4 percentage points, 6 months post clinic visit. ³⁰⁷

Figure

Figure 3. Percentage point change in patient receipt of advice to quit tobacco use from providers attributable to provider reminders from studies that qualified for inclusion in this review.

Review of evidence: applicability

The same body of evidence used to assess effectiveness was used to assess the applicability of these interventions to different tobacco products, settings, and populations. Studies were conducted in the United States ^{301, 302, 304, 305, 307, 308} and Canada. ³⁰⁶ Studies were conducted in primary care clinics, ^{302, 307} family practice clinics, ^{304, 305, 306, 308} and in pulmonary clinics. ³⁰¹ Most studies did not provide demographic information on the patient populations.

Review of evidence: other positive or negative effects

Two studies, in which provider reminders included other preventive services, observed increases in some or all of the prompted activities. ^{304 306} Other positive or negative effects were not identified in this review.

Review of evidence: economic

We did not identify any economic evaluations reporting on cost, cost-effectiveness, cost-benefit, or cost utility in this review.

Barriers to intervention implementation

One potential barrier to the implementation of a provider reminder system is the administrative burden. Administrative burden was not identified as a problem in any of the qualifying studies, and most of the reminder systems (e.g., "expanded vital signs") were easily implemented.

Conclusion

According to the Community Guide's rules of evidence, sufficient scientific evidence documents that provider reminder systems when implemented alone are effective in increasing provider delivery of advice to quit to tobacco-using patients.

Definition

Provider education involves giving information about tobacco and tobacco use cessation to providers, to increase their knowledge and change their attitudes and practices. Techniques by which information is delivered include lectures, written materials, videos, and continuing medical education seminars. Provider populations include physicians, nurses, physician assistants, health care students, and other office staff.

Provider education efforts are frequently combined with other interventions, such as provider reminders and patient education efforts. Multicomponent interventions including provider reminders are considered below, in the section on Provider Reminder + Provider Education with or without Patient Education.

Background

Provider education attempts to increase or improve providers' interactions with tobacco-using patients. Changes in provider performance could include increases in the identification of tobacco-using patients, increases in the delivery of advice to quit, improvement in the quality of providers' advice to quit, and both an increase and an improvement in providers' efforts to assist tobacco users' attempts to quit and to maintain abstinence.

Review of evidence: effectiveness

Our search identified 25 studies regarding the effectiveness of provider education interventions when implemented alone. ^{302, 307, 309, 310, 311, 312, 313, 314, 315, 316, 317, 318, 319, 320, 321, 322, 323, 324, 325, 326, 327, 328, 329, 330, 331} Two additional papers provided information on an already-included study. ^{332, 333} Five studies with limited quality of execution, ^{310, 312, 315, 319, 328} and four studies with least suitable study designs ^{317, 321, 323, 331} were not included in the body of evidence. Details on the 16 qualifying studies are provided at the website: www.thecommunityguide.org.

Provider education techniques evaluated in the qualifying studies include day-long seminars, ^{302, 311} lectures for practitioners, ^{311, 314, 327} lectures for resident physicians, ^{307, 316, 318, 320, 326} lectures and office visits or contacts, ^{322, 329} small group tutorial sessions, ^{309, 324} mock interviews with feedback, ^{325, 330} and education materials. ³¹³ The total duration of the provider education sessions ranged from 2 hours to 3 days (median, 2.5 hours) in the 11 studies that provided this information.

The 16 qualifying studies reported a total of 19 measurements of changes in provider counseling skills or behaviors, and two measurements of patient tobacco use behaviors. Four studies measured changes or differences in cessation counseling skills or use of patient education materials, ^{309, 313, 324, 325} which were not considered in the evidence of effectiveness. Five studies measured differences in provider determination of patient smoking status, ^{311, 318, 322, 326, 330} with absolute percentage differences ranged from +0.1 to +35 percentage points (median, +8 percentage points). Ten studies reported differences in provider delivery of advice to quit (see Figure 4), ^{302, 307, 311, 314, 316, 318, 320, 322, 329, 330} with absolute percentage differences that ranged from -5 to +73 percentage points (median, +2.2 percentage points). Four of these 10 studies observed either no effect or a negative effect on provider delivery of advice to quit. ^{302, 316, 329, 330} Two studies reported differences in patient tobacco use cessation. ^{307 316} Absolute percentage differences in cessation were +5.2 and +1.7 percentage points in follow-up periods of 3.5 months and 6 months, respectively.

Figure

Figure 4. Percentage point change in patient receipt of advice to quit tobacco use from providers attributable to provider education from studies that qualified for inclusion in this review.

Review of evidence: applicability

The same body of evidence used to assess effectiveness was used to assess the applicability of these interventions to different settings, provider populations, and patient populations. Settings have included HMOs, ³¹⁴ academic centers, ^{307, 316, 318, 320} and private practices. ^{302, 311, 329} Studies have included such providers as medical and nursing students, ^{309, 325, 327} resident physicians, ^{307, 316, 318, 320, 324, 326, 330} physician practicioners, ^{302, 311, 313, 314, 326, 329} and non-physician practioners. ³²⁹ Physician specialties included internal medicine, ^{307, 311, 324} family practice, ^{307, 311, 316, 320, 324} and pediatrics. ^{307, 318} Four studies evaluated interventions targeting providers community-wide. ^{311, 313, 322, 329} Few studies provided information on the patient populations.

Review of evidence: other positive or negative effects

No information regarding other positive or negative effects was identified in this review.

Conclusion

According to the Community Guide's rules of evidence, available studies provide insufficient evidence to assess the effectiveness of provider education interventions when implemented alone. Evidence is insufficient because (1) few studies evaluated the effect on patient tobacco use cessation, and (2) studies that evaluated provider delivery of advice to quit demonstrated inconsistent results.

Definition

Multicomponent strategies to increase tobacco use cessation include efforts to educate and to prompt providers to identify and to intervene with tobacco-using patients, as well as to provide supplementary educational materials when indicated. The components of this intervention are a provider reminder system and a provider education program, with or without patient education materials such as self-help cessation manuals.

Background

A multicomponent intervention can provide an integrated strategy to increase and improve tobacco use cessation by patients. These strategies can educate, motivate, and prompt providers to increase and improve their interaction with tobacco-using patients. These strategies can also improve patient cessation by increasing knowledge and motivation to quit and to remain abstinent. The multicomponent interventions evaluated in this section include at least one component directed at providers.

Review of evidence: effectiveness

Our search identified a total of 37 studies regarding the effectiveness of multicomponent health care system interventions in reducing tobacco use among patients. ^{302, 307, 314, 320, 326, 328, 329, 334, 335, 336, 337, 338, 339, 340, 341, 342, 343, 344, 345, 346, 347, 348, 349, 350, 351, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363} Fifteen additional papers provided information on an already-included study. ^{364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378} Two studies with limited quality of execution ^{328, 334} and four studies with least suitable study designs ^{345, 350, 355, 360} were not included in the body of evidence. Details of the 31 qualifying studies are available at the website: www.thecommunityguide.org.

Twenty-one studies included at minimum the combination of a provider reminder system and a provider education program. ^{302, 307, 314, 320, 326, 329, 335, 336, 337, 338, 339, 343, 344, 347, 348, 349, 352, 356, 357, 358, 361} An additional component, patient education materials, was provided in 14 of these studies. ^{314, 337, 338, 339, 343, 344, 347, 348, 349, 352, 356, 357, 358, 361} One study measured changes in adolescent tobacco use initiation and is not considered further in this section. ³⁴⁴ Three studies evaluated a provider reminder system combined with other interventions without a provider education program. ^{354, 359, 363} Seven studies evaluated a provider education component combined with a patient education program without a provider reminder system. ^{340, 341, 342, 346, 351, 353, 362}

The 20 qualifying studies, evaluating the effectiveness of a multicomponent intervention containing a minimum of a provider reminder system and a provider education program, reported 15 measurements regarding provider advice to quit and 14 measurements regarding patient tobacco use cessation (Figure 5). Overall, the absolute percentage differences in patient receipt of provider advice to quit ranged from +5.2 to +60 percentage points (median, +20 percentage points). In follow-up periods that ranged from 5 weeks to 12 months (median, 10 months) the absolute percentage differences in patient tobacco use cessation ranged from -1.0 to +25.9 percentage points (median, +4.7 percentage points).

Figure

Figure 5. Percentage point change in tobacco use cessation attributable to multicomponent interventions (minimum of a provider reminder and provider education components) from studies that qualified for inclusion in this review.

A subset of 7 studies evaluated the minimum combination of a provider reminder system and a provider education program. ^{302, 307, 320, 326, 329, 335, 336} These studies provided 6 measurements of differences in provider advice to quit and 4 measurements of patient tobacco use cessation. The absolute percentage differences reported ranged from +6 to +39 percentage points (median, +12.5 percentage points) for provider advice to quit and -0.3 to +6.4 percentage points (median, +4.5 percentage points) for patient tobacco use cessation.

A subset of 13 studies evaluated interventions that included patient education materials combined with a provider reminder system and a provider education program. ^{314, 337, 338, 339, 343, 347, 348, 349, 352, 356, 357, 358, 361} These studies provided 9 measurements of provider advice to quit and 10 measurements of patient tobacco use cessation. The absolute percentage differences reported ranged from +5.2 to +60 percentage points (median, +22 percentage points) for provider advice to quit, and from -1 to +25.9 percentage points (median, +5.7 percentage points) for patient cessation.

Of the remaining studies, 3 included provider reminders and patient education without a provider education program. ^{354, 359, 363} These studies reported 3 measurements of differences in patient tobacco use cessation. The absolute percentage differences ranged from -0.6 to +5.2 percentage points (median, +1.2 percentage points) in follow-up periods of 6-12 months.

Seven studies evaluated combinations that included a provider education program without a provider reminder system. ^{340, 341, 342, 346, 351, 353, 362} These studies reported 5 measurements of differences in patient tobacco use cessation. Absolute percentage differences in cessation ranged from -1.5 to +4.6 percentage points (median, +0.5 percentage points) in follow-up periods of 6-12 months. ^{340, 341, 342, 346, 362} Only three studies measured changes in receipt of provider advice to quit (absolute percentage changes of +14.6, ³⁴⁶ -9, ³⁴² and +22.7 ³⁶² percentage points.

Review of evidence: applicability

The same body of evidence used in the assessment of effectiveness was used to assess the applicability of these interventions to different settings, populations, and forms of tobacco. The 21 studies of multicomponent interventions that included at least a provider education program and a provider reminder system were conducted in a variety of health care settings including HMOs, ^{314, 339, 352, 356} private practices, ^{302, 337, 338, 347, 348, 349, 357, 358} academic health care centers, ^{307, 329, 348, 351} physician training programs, ^{307, 320, 326, 335, 343} and public health clinics. ³⁶¹ Studies were implemented in a variety of provider populations including primary care, ^{307, 337, 347, 348, 349} internal medicine, ^{302, 314, 338, 339} family medicine, ^{302, 320, 326, 358} obstetric care providers, ^{329, 343, 352} pediatricians, ³⁵⁷ and dental care providers. ^{336, 344, 356} One study focused on reducing use of smokeless tobacco. ³⁵⁶

Review of evidence: other positive or negative effects

Provider reminder systems have included prompts for other important preventive services. No other positive or negative effects were identified in this review.

Review of evidence: economic

No economic evaluations reporting on cost, cost-effectiveness, cost-benefit, or cost utility were identified in this review.

Barriers to intervention implementation

Administrative burden is a potential barrier to the implementation of provider reminder systems.

Conclusion

According to the Community Guide's rules of evidence, strong scientific evidence demonstrates that multicomponent health care system interventions that include a minimum of a provider reminder system and a provider education program are effective in increasing both provider delivery of advice to quit and patient tobacco use cessation. Additional effectiveness was demonstrated by studies that also included patient education, such as self-help cessation materials.

Definition

Feedback interventions use retrospective assessment of provider performance in the identification of patient tobacco use status and/or the delivery of advice to quit, to inform and motivate providers. Techniques of assessment include chart reviews or the use of computerized records. Assessment and feedback interventions can be combined with other activities, such as provider reminders and provider education, and these combinations are considered in this section.

Background

Provider assessment and feedback can motivate providers to increase and improve their delivery of effective provider interactions with patients, such as advice to quit. Evaluation of provider assessment and feedback is timely because (1) clinical information systems are improving and are increasingly common; (2) effective cessation therapies are available and increasing provider interactions with tobacco-using patients could increase the use of these therapies; and (3) quality assurance approaches such as the Health-plan Employer Data and Information Set (HEDIS) are being used more often.

Review of evidence: effectiveness

Our search identified five studies evaluating the effectiveness of provider assessment and feedback interventions on provider behaviors with tobacco-using patients. ^{379, 380, 381, 382, 383} One study with limited quality of execution ³⁷⁹ and one study with a least-suitable study design ³⁸⁰ were not included in the evaluation of effectiveness. Details of the three qualifying studies are provided at the website: www.thecommunityguide.org.

A provider assessment and feedback program was evaluated alone in one study, ³⁸¹ and in combination with other components in two studies including a provider education program, ^{382, 383} and a provider reminder flowsheet. ³⁸² In all three studies, provider documentation or recognition of a patient's tobacco use status was only one of several preventive care practices for which assessment and feedback was provided (range, 3-26 items). Only one study used a computer system to collect information and enable providers to obtain feedback information. ³⁸³

None of the qualifying studies attempted to measure changes in provider delivery of advice to quit or patient tobacco use behaviors. The three qualifying studies provided three measurements of effectiveness in increasing provider recognition of patient tobacco use status. In study periods that ranged from 3 months to 6 years, the absolute percentage improvements in provider recognition of patient tobacco use status ranged from +13 to +39 percentage points (median, +21 percentage points).

Review of evidence: other positive or negative effects

Implementation of a provider assessment and feedback program increased provider delivery of other preventive care practices in these studies. ^{381, 382} No other positive or negative effects were identified in this review.

Conclusion

According to the Community Guide's rules of evidence, the available studies provided insufficient evidence to assess the effectiveness of provider assessment and feedback interventions in increasing either provider delivery of advice to quit or patient tobacco use cessation. Evidence is insufficient because the small number of available studies did not provide measurements on the outcomes required for an evaluation of effectiveness in this review (such as increasing provider delivery of advice to quit, or patient tobacco use cessation).

Definition

This intervention includes efforts to reduce the financial barriers to patient use of effective cessation therapies such as nicotine replacement, ¹⁴ other pharmacologic therapy, ¹⁴ and/or behavioral therapies such as cessation groups. ¹⁴

Background

Reducing out-of-pocket costs for effective cessation therapies has as its objectives: (1) to increase use of effective therapies; (2) to increase the number of persons who attempt to quit; and (3) to increase the number of persons who make successful cessation efforts.

Review of evidence: effectiveness

Our search identified five studies regarding the effectiveness of reducing patient out-of-pocket costs for effective tobacco cessation therapies. ^{384, 385, 386, 387, 388} All five studies were of fair quality of execution and greatest or moderate suitability of study design and were included in the evaluation of effectiveness. Details of the qualifying studies are available at the website: www.thecommunityguide.org.

All five studies evaluated interventions that reduced or eliminated patient costs for nicotine gum, ^{384, 386, 387} or nicotine replacement. ^{385, 388} In two studies, nicotine gum or replacement were provided as part of, ³⁸⁴ or in addition to, ³⁸⁵ a behavioral program. Access to a behavioral program was provided but rarely used in a third study. ³⁸⁸ In three studies, nicotine gum or nicotine replacement was provided free of charge to participants in the intervention group. ^{384, 386, 388} In one study, the out-of-pocket costs of the combination of behavioral program and nicotine replacement were $52.50 for comparison group patients and $10 for intervention group patients. ³⁸⁵

One study conducted in an HMO setting reported significant differences in the use of nicotine gum (measured in pieces of gum per user) by the level of drug co-payment, but the results could not be meaningfully expressed as a percentage point difference. ³⁸⁷ The remaining four studies ^{384, 385, 386, 388} provided four measurements of differences in use of cessation therapies and four measurements of differences in tobacco use cessation among study populations of recruited smokers ^{384, 386, 388} or among a general patient population. ³⁸⁵

All four studies observed increases in the use of cessation therapies. In three studies, the absolute percentage difference reported ranged from +6.5 to +28 percentage points (median +7 percentage points). ^{384, 385, 386} In one study, the difference in use was reported as an adjusted odds ratio of 2.26 (95% CI 1.60, 3.19). ³⁸⁸

All four studies also observed increases in tobacco use cessation, either measured as differences in the observed cessation rates ^{384, 386, 388} or as differences in the overall rates of cessation calculated for the study population. ³⁸⁵ In follow-up periods that ranged from 6 to 12 months (median, 9 months), absolute percentage differences in cessation ranged from +2.1 to +11 percentage points (median, +7.8 percentage points). In addition, one study reported an adjusted odds ratio for cessation of 1.63 (95% CI 1.14, 2.35). ³⁸⁸

Review of evidence: applicability

The same body of evidence used to assess effectiveness was used to assess the applicability of these interventions to different settings and populations. All of these studies were conducted in the United States. Studies were conducted in several settings, including HMOs, ^{385, 387, 388} private practices, ³⁸⁶ and a Department of Defense hospital. ³⁸⁴ Studies were conducted in rural ³⁸⁶ and mixed rural-urban settings. ^{385, 388} One study was conducted among a low-income population. ³⁸⁶

Review of evidence: other positive or negative effects

One study observed an increase in the extended use of nicotine gum beyond the recommended duration (4 months). ³⁸⁶ No other potential benefits or harms were identified in this review

Review of evidence: economic

Of two studies reviewed, ^{385, 386} one was conducted in Washington state and reported program cost per quitter. ³⁸⁵ This study was conducted in a health care setting for employees enrolled in a health plan. The intervention consisted of insurance coverage for patients in a behavioral program that included nicotine replacement, with a 12-month follow-up. There were four types of insurance coverage, which differed according to the user's out-of-pocket-costs: (1) a 50 % copayment for the behavioral program and the usual $5 copayment per prescription for nicotine replacement therapy (standard coverage); (2) a 50% copayment for both the behavioral program and nicotine-replacement therapy (reduced coverage); (3) no copayment for the behavioral program but a 50% copayment for nicotine-replacement therapy (flipped coverage), and; (4) no copayment for the behavioral program and the usual $5 copayment per prescription for nicotine replacement therapy (full coverage). Costs measured included drugs, personnel, and cost of the behavioral program. Development, marketing, and implementation of the coverage plan were not included in the analysis. The adjusted program costs per quitter were $135, $141, $149, and $195 for standard, reduced, flipped, and full coverage, respectively. This study was classified as good, based on the quality assessment criteria used in the Community Guide. ⁹⁰

The second study was a cost-benefit analysis conducted in Vermont, ³⁸⁶ reporting net benefit. This study was conducted at a rural family practice clinic with low-income patients. The intervention consisted of brief physician advice and a prescription for free nicotine gum, with a 6-month follow-up. Costs measured included physician time, nicotine gum, smoking cessation booklets, and patient time. Development, promotion, and evaluation costs were not included. The adjusted quit rate for the intervention group was 9.4%. When costs and benefits from averted illness were compared, the intervention was shown to be cost-saving. This study was classified as satisfactory, based on the quality assessment criteria used in the Community Guide. ⁹⁰ (See also Appendix C, Interpreting the Economic Data.)

Barriers to intervention implementation

Potential barriers to increased use of effective cessation therapies are coverage requirements that tie pharmacotherapy to behavioral therapy interventions. Recent reviews have identified each strategy as effective alone. ¹⁴ These combinations demonstrate a higher cost-effectiveness, ³⁸⁹ but also limit use of effective therapies for smokers who are unwilling to participate in the behavioral program. These barriers might be reduced by including proactive telephone counseling as a behavioral therapy option. ³⁹⁰

Conclusion

According to the Community Guide's rules of evidence, sufficient scientific evidence demonstrates that reducing out-of-pocket costs for effective cessation therapies increases both use of the effective therapy and patient tobacco use cessation.

Definition

Telephone support interventions provide tobacco product users with cessation counseling or assistance in initiating and/or maintaining abstinence. Telephone support can be reactive (tobacco user initiates contact) or proactive (provider initiates contact or user initiates contact with provider follow-up). Techniques for delivery of telephone support include the use of trained counselors, health care providers, or taped messages in single or multiple sessions. Telephone support sessions usually follow a standardized protocol for providing advice and counseling. The telephone support component is usually combined with other interventions, such as patient education materials, individual or group cessation counseling, or nicotine replacement therapies.

Background

Telephone contact can increase motivation to tobacco users to attempt to quit, and can provide support and assistance to recent quitters to reduce relapses. When implemented in a community setting, the telephone component typically provides access to cessation information such as self-help materials and available local resources (such as group sessions), and may provide counseling and motivation sessions. When implemented in a clinical setting, telephone follow-up calls usually support other clinical cessation interventions such as provider counseling, group cessation sessions, or nicotine replacement or other therapies.

Review of evidence: effectiveness

Our search identified 39 studies regarding the effectiveness of telephone cessation support. ^{24, 232, 277, 314, 343, 351, 354, 355, 356, 361, 363, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418} Sixteen additional papers provided information on an already-included study. ^{365, 366, 367, 370, 371, 372, 376, 377, 378, 419, 420, 421, 422, 423, 424, 425} Four studies with limited quality of execution, ^{391, 394, 396 411} and three studies with least suitable study designs, ^{24, 232, 355} were not included in the body of evidence. Details of the 32 qualifying studies are available at the website: www.thecommunityguide.org.

In all of the qualifying studies, telephone support was coordinated with additional interventions including patient education (29 studies), ^{277, 343, 354, 356, 361, 363, 392, 393, 395, 397, 398, 399, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 412, 413, 414, 415, 416, 417, 418} provider-delivered counseling (17 studies), ^{314, 351, 354, 356, 361, 363, 392, 395, 397, 398, 402, 405, 410, 412, 413, 414, 416} nicotine replacement (4 studies), ^{401, 405, 413, 417} a smoking cessation clinic (1 study), ⁴⁰⁰ and a televised cessation series (1 study). ²⁷⁷ The qualifying studies included evaluations of telephone support interventions that were proactive (27 studies), ^{314, 343, 351, 354, 356, 361, 363, 392, 395, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 409, 410, 412, 413, 414, 416, 417, 418} or reactive (5 studies). ^{277, 393, 407, 408, 415}

Thirty of the 32 qualifying studies reported a total of 31 measurements of differences in patient tobacco use cessation (Figure 6). ^{277, 343, 351, 354, 356, 361, 363, 392, 393, 395, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 412, 413, 414, 415, 416, 418} In follow-up periods of 5 weeks to 34 months (median, 12 months), these studies reported absolute percentage differences in cessation ranging from -3.4 to +23 percentage points (median, +2.6 percentage points). Seven measurements from six studies that compared telephone support and patient education to patient education alone ^{399, 403, 404, 406, 408, 418} provided similar results with absolute percentage differences in continuous tobacco use cessation ranging from +0.9 to +6.3 percentage points (median, +2.4 percentage points). Five of these six studies evaluated proactive telephone support systems. The median relative percent difference in tobacco use cessation in these studies was +41%.

Figure

Figure 6. Percentage point change in tobacco use cessation attributable to multicomponent interventions that included telephone cessation support from studies that qualified for inclusion in this review.

Review of evidence: applicability

The same body of evidence used to assess effectiveness was used to assess the applicability of these interventions to different settings and populations. Telephone cessation support interventions have been implemented nationwide, ⁴⁰⁷ statewide, ⁴¹⁸ and in regions and cities. ^{277, 399, 408, 409, 415} Studies were conducted in a variety of health care settings including HMOs, ^{314, 356, 363, 392, 401, 403, 404, 405, 406, 414, 415} private practices, ⁴¹² public health clinics, ³⁶¹ medical centers and hospitals, ^{351, 354, 398, 402, 410, 412} and resident training programs. ^{343 351} Provider specialties included dentistry, ³⁵⁶ obstetrics, gynecology, and family planning, ^{343, 361, 395, 397, 403, 404, 412} primary care, ^{314, 363, 392, 406} family practice, ^{351, 363} and internal medicine. ^{351, 363} Patient populations included hospitalized smokers, ^{354, 398, 402, 405, 413, 414} veterans, ^{413, 416} pregnant women, ^{343, 361, 395, 403, 412} African-Americans, ⁴⁰⁷ and older (60+years) smokers. ⁴⁰⁹ One study focused on changing smokeless tobacco use. ³⁵⁶

Review of evidence: other positive or negative effects

No positive or negative effects of telephone cessation support interventions were identified in this review.

Review of evidence: economic

Five studies were reviewed. ^{423, 426, 427, 428, 429} Two studies reported program costs per quitter, ^{427, 429} and three studies ^{423, 426, 428} reported program costs per life-year saved ratios, which were converted to cost per QALY using preference weights reported by Fiscella and Franks. ²⁴⁹

Both studies reporting program costs per quitter evaluated interventions to increase tobacco use cessation among pregnant women. ^{427, 429} The first study was conducted in Southern California in an HMO setting. ⁴²⁷ The intervention consisted of a combined prenatal nutrition counseling and smoking cessation program aimed at reducing the incidence of low birthweight infants. The smoking cessation program consisted of an eight-week home correspondence program that included weekly telephone calls to an automated answering service. The absolute percentage change in smoking cessation was +12 percentage points, with a comparison group cessation rate of 38%. Costs included salaries, overhead, supplies, printing, phone, and postage. Adjusted program cost per quitter was $677. This study was classified as good, based on the quality assessment criteria used in the Community Guide. ⁹⁰

The second study was a nationwide, modeled intervention in the United States. ⁴²⁹ It consisted of a single 15-minute counseling session conducted by a nonmedical counselor, instructional material, and two follow-up telephone calls. The absolute percentage change in cessation was +15 percentage points, modeled from earlier randomized trials of smoking cessation among pregnant women. Costs included instructional materials, staff time, overhead, and training. Adjusted program cost per quitter was $292. This study was classified as good, based on the quality assessment criteria used in the Community Guide. ⁹⁰ The difference in program cost between the two studies can be explained by the fact that the first study looked at a comprehensive intervention using more resources. This study was classified as good, based on the quality assessment criteria used in the Community Guide. ⁹⁰

Of the three studies reporting program costs in terms of life-years saved, two studies ^{423 426} looked at interventions conducted in hospital settings with adult patients who smoked. One study was conducted at the Mayo Clinic in Minnesota, ⁴²⁶ and the second was conducted at HMO hospitals in Oregon and Washington. ⁴²³ The comparison group quit rates for the studies were 10.7% and 9.2 %, respectively. The absolute percentage changes were +12 percentage points and +4 percentage points, respectively. The Mayo clinic intervention consisted of two programs: an individual nicotine-dependency treatment program, and a relapse prevention program. The relapse prevention program included telephone follow-up calls, letters, and a mailed survey. The comparison was a group of patients with no program. Costs included personnel, supplies, telephone, drugs, and capital equipment. Adjusted program cost per QALY was $2,532. The HMO hospital intervention consisted of a 20-minute bedside counseling session, a video, self-help materials, and follow-up calls. Costs included program development, personnel, communications, and overhead. Adjusted program cost per QALY of this intervention was $1,248. In comparing the two studies, the Mayo clinic intervention had a higher program cost per QALY in spite of showing a higher net effect size. This was because programs costs of the Mayo clinic intervention included letters, surveys, and treatment for nicotine dependency in addition to counseling and telephone calls. These studies were classified as good, based on the quality assessment criteria used in the Community Guide. ⁹⁰

The third study ⁴²⁸ was conducted at a Boston hospital with patients who had had an acute myocardial infarction. The modeled intervention consisted of nurse-managed smoking-cessation counseling including telephone support after discharge. The comparison group consisted of patients exposed to standard smoking cessation counseling designed for survivors of acute myocardial infarction. The comparison group had a quit rate of 45%. The absolute percentage change was +26 percentage points. Costs included were personnel and instructional materials. Time spent on the phone, follow-up time, program development, and training costs were not included. The adjusted program costs per QALY was $73. This study was classified as good, based on the quality assessment criteria used in the Community Guide. ⁹⁰ (See also Appendix C, Interpreting the Economic Data.)

Barriers to intervention implementation

One paper identified in this review ⁴²⁰ reported extremely low utilization of a reactive telephone support line implemented in an HMO setting. Several studies identified in the evaluation of mass media campaigns ^{215, 229, 231, 253} observed significant increases in use of telephone cessation services when the mass media messages directed viewers to call for further information or support.

Conclusion

According to the Community Guide's rules of evidence, strong scientific evidence exists that telephone cessation support is effective in increasing tobacco use cessation when implemented with other interventions (e.g., other educational approaches and/or clinical therapies) in both clinical and community settings. The minimum intervention with sufficient evidence of effectiveness identified in this evaluation was proactive telephone support combined with patient cessation materials.

Effectiveness

The effectiveness of increasing the unit price for tobacco products and mass media campaigns (when implemented with other interventions) is established. However, research issues regarding the effectiveness of these interventions remain.

• What intervention components contribute most to effectiveness of multicomponent interventions? What components contribute the least?
• What are the minimum and optimal requirements for the duration and intensity of mass media campaigns?
• What are the most effective combinations of messages for mass media campaigns?
• Do tobacco users respond differently to changes in product price that result from excise tax increases than to industry-induced increases?
• How long do the effects of a single excise tax increase last?

Because the effectiveness of mass media cessation series and smoking cessation contests has not been established, basic research questions remain.

• Are these interventions effective in increasing tobacco use cessation in the population?
• Do recruited tobacco users exposed to these interventions quit at a greater rate than recruited tobacco users not exposed to these interventions?
• What are the rates of participation in these interventions?

Applicability

The effectiveness of increasing the unit price and of mass media campaigns in reducing tobacco use in the population is established. However, identifying differences in the effectiveness of each intervention for specific subgroups of the population remains important.

• Do significant differences exist regarding the effectiveness of these interventions based on the level of scale (i.e., national, state, local) at which they are delivered?
• What are the effects of mass media campaigns among populations that differ by race and ethnicity?

Other positive or negative effects

Several potential negative effects of tobacco product price increases were reviewed in this evaluation. Although further research on the potential negative effects is warranted, evaluating the impact of potential positive effects of reductions in tobacco use should also be investigated, to provide a complete picture of the effects of state and federal excise tax increases.

• What are the effects of these interventions on reducing smoking-related fires? What are the effects on ETS exposure?
• What proportion of smokers substitute tobacco products and/or modify their smoking habits in response to an increase in tobacco product price? How much of the potential health benefit of a price increase is reduced by these behaviors? How can these potential problems be reduced?
• Do mass media campaigns focusing on tobacco have additional effects on other drug use?

Economic evaluations

The available economic information on mass media campaigns was limited. Considerable research is therefore warranted regarding the following questions:

• What are the costs of mass media campaigns, especially campaigns that achieve an effective intensity over an extended duration?
• How do the costs per additional quitter compare with other interventions intended to reduce tobacco use?
• What is the cost-benefit, cost-utility, or cost per illness averted of these interventions?

Barriers

Implementation of these interventions requires political action and support. Research issues generated in this review include:

• What components of successful legislative and referendum campaigns are most effective? What components are least effective?
• What information is most important in gaining public support for these interventions? In gaining legislative support?
• What are the most effective ways to maintain adequate funding levels for mass media campaigns?

Effectiveness

The effectiveness of recommended and strongly recommended interventions in this section (i.e., multicomponent provider reminder + provider education with or without patient education materials; provider reminder systems alone; multicomponent interventions including telephone cessation support; and reducing patient out-of-pocket costs for cessation) is established. However research issues regarding the effectiveness of these interventions remain.

• Which characteristics of provider-based interventions contribute to increased or decreased effectiveness?
• What are the least and most effective combinations of services in multicomponent interventions?
• What is the effect of provider reminder systems on patient tobacco use cessation when implemented alone?
• What is the relative effectiveness of provider reminders that focus on determination of patient tobacco use status versus reminders that prompt for delivery of advice to quit?
• How do content and method of delivery of provider reminders relate to effectiveness?
• Can reducing patient costs for effective cessation services increase the effectiveness of provider-based interventions?
• What is the most effective level of implementation for telephone cessation support services?
• Is the use and effectiveness of telephone cessation support increased when community and clinical cessation support programs are coordinated?

Because the effectiveness of two interventions (provider education when used alone, and provider feedback systems) has not been established, basic research questions remain. This is especially true for provider assessment and feedback systems for which the number of available studies was small.

• What are the effects of provider assessment and feedback interventions on provider delivery of advice to quit to tobacco-using patients? On patient tobacco use cessation?
• What is the effectiveness of HEDIS, as a form of assessment, feedback, and benchmarking, in improving patient receipt of advice to quit and patient tobacco use cessation? Does effectiveness vary by practice setting?
• What frequency, duration, and format of provider education efforts are required to obtain consistent improvements in provider performance and patient response?

Applicability

Each recommended and strongly recommended provider-based intervention should be applicable in most relevant target populations and settings. However, possible differences in the effectiveness of each intervention for specific subgroups of patient and provider populations could not be determined. Several questions regarding the applicability of these interventions in settings and populations other than those studied remain.

• Do provider-based interventions differ in effectiveness in different patient populations?
• Are provider-based interventions effective in increasing cessation or in reducing initiation in adolescent populations?
• Do significant differences exist regarding the effectiveness of these interventions based on the level of scale at which they are delivered?

Other positive or negative effects

With the exception of the use of provider reminder systems to prompt action on other preventive services, studies in this review did not report on other positive and negative effects of these interventions. Research on the following questions would be useful:

• Do provider-based interventions for tobacco use cessation interfere with office flow or efficiency? If so, how can this effect be minimized?
• Do provider-based interventions increase or decrease the delivery of other preventive services?

Economic evaluation

Available economic information was limited in this section. Considerable research is warranted regarding the following questions:

• What are the costs for provider-based interventions?
• What are the costs for patient-based interventions?
• How do the costs per additional quitter compare with other interventions intended to reduce tobacco use?
• What is the cost-benefit, cost-utility, or the cost per illness averted of these interventions?
• What is the cost-effectiveness for provider interventions targeting tobacco alone compared with provider interventions targeting multiple preventive services?

Barriers

Research questions regarding the potential barriers identified for the interventions evaluated in this section include:

• How can provider-based interventions that place minimal administrative burden on providers or systems be implemented?
• What information is needed to overcome potential barriers to the implementation of provider assessment and feedback interventions?
• What information is needed to overcome potential barriers to reducing patient out-of-pocket costs for effective cessation therapies?
• What is the effect on use of combining effective pharmacologic therapies and behavioral programs as a criterion for reimbursement? What is the impact on use and effectiveness if these cessation options are provided independently?