Abstract
Objectives. We assessed the impact of a tobacco control initiative over 10 years on cessation and prevention.
Methods. We examined 2000–2009 Behavioral Risk Factor Surveillance System cases of a metropolitan statistical area (MSA) with systematic tobacco control efforts throughout the decade (El Paso, TX) and 2 comparison MSAs similar in size and population with less coordinated tobacco control efforts (Austin-Round Rock, TX and San Antonio, TX).
Results. Yearly, El Paso exhibited a 6% increase in the prevalence of former smokers, a 6% decrease in prevalence of daily smokers, and a 7% decrease in the prevalence of established smoking (≥ 100 cigarettes per lifetime and currently smoking); we did not observe similar trends in the comparison MSAs. There was no change in the prevalence of nondaily smokers in any of the MSAs.
Conclusions. The coordinated tobacco control activities in El Paso are related to cessation among daily smokers and prevention of established smoking at the population level but have not stimulated cessation among nondaily smokers. Comprehensive tobacco control should focus more on not only daily smokers but also nondaily smokers.
A decrease of 1% in the prevalence of tobacco users represents a meaningful improvement in population health.1 However, achieving population reductions in smoking is challenging. Although the adequate financing of tobacco control efforts is associated with effective control activities and policies,2–4 coordinated and comprehensive evidence-based ecological efforts are necessary to realize smoking reductions.5
A Smoke-Free Paso del Norte, funded by the Paso del Norte Health Foundation and launched in 1999, is a comprehensive tobacco control initiative founded on Centers for Disease Control and Prevention (CDC) best practices.6 The initiative's elements include the promotion of strong clean indoor air ordinances (CIAO); tobacco use prevention and cessation media; the development, promotion, and availability of youth prevention and adult cessation services; the establishment and maintenance of a tobacco control network; and the implementation of an ongoing evaluation plan driven by regular surveillance (Table 1).7
TABLE 1— A Smoke-Free Paso del Norte Initiative Activities by Year: El Paso, Texas, 2000–2009
| Year (No. of Funded Programs) | Focus | Activities |
| 2000 (8) | Cessation | Cessation media campaign (billboards, radio, television, and community events) |
| Implementing clinical practice guidelines for screening tobacco use | ||
| Dejate De Ese Vicio—smoking cessation and relapse prevention education | ||
| Fresh Start—group-based tobacco cessation support program | ||
| Freedom from Smoking—personal support to help employees quit smoking | ||
| Not On Tobacco—American Lung Association's Best Practice Curriculum helping high-risk youths cease tobacco use | ||
| Prevention | Prevention media campaign (billboards radio, television, and community events) | |
| Tobacco prevention education to colonia residents | ||
| Policy | A Smoke-Free Paso del Norte Coalition and Las Cruces Smoke-Free Coalition—community coalitions to promote clean indoor air ordinances | |
| Other | Program evaluation and surveillance | |
| 2001 (9) | Cessation | Continuing activities: media, screening, Dejate de Ese Vicio, Fresh Start, Freedom from Smoking, and Not on Tobacco |
| Prevention | Continuing activities: media and tobacco prevention education to colonia residents | |
| Teens Against Tobacco Use—peer-to-peer education on dangers of smoking | ||
| Policy | Continuing activities: A Smoke-Free Paso del Norte Coalition and Las Cruces Smoke-Free Coalition | |
| Other | Continuing activities: program evaluation and surveillance | |
| 2002 (10) | Cessation | Continuing activities: media, screening, Dejate de Ese Vicio, Freedom from Smoking, Fresh Start, and Not On Tobacco |
| Prevention | Continuing activities: media, tobacco prevention education to colonia residents, and Teens Against Tobacco Use | |
| Tobacco prevention targeting college females | ||
| Policy | Passage of El Paso, TX clean indoor air ordinance—environmental tobacco smoke protection in workplaces and public places, including restaurants, bars, bingo facilities, and bowling alleys; not outdoor | |
| Train El Paso and Hudspeth County law enforcement officers and judicial officials on tobacco enforcement | ||
| Doña Ana County South Central New Mexico Prevention Coalition—enforcement of tobacco laws | ||
| Other | Continuing activities: program evaluation and surveillance | |
| 2003 (8) | Cessation | Continuing activities: media, screening Freedom from Smoking, Fresh Start, and Not On Tobacco |
| Prevention | Continuing activities: media and tobacco prevention targeting college women | |
| Healthy Beginnings—tobacco prevention and cessation program implemented in prenatal and parenting classes | ||
| Policy | Continuing activities: A Smoke-Free Paso del Norte Coalition and Doña Ana County South Central New Mexico Prevention Coalition | |
| Otero County Tobacco Educational Coalition—clean indoor air ordinance promotion in Alamogordo, NM | ||
| Other | Continuing activities: program evaluation and surveillance, published “Settling the Smoke 2003: A Status Report on Smoking in the Paso del Norte Region” | |
| 2004 (2) | Cessation | Continuing activities: media and Not On Tobacco |
| Prevention | Continuing activities: media and Healthy Beginnings | |
| Policy | Continuing activities: A Smoke-Free Paso del Norte Coalition | |
| Other | Continuing activities: program evaluation and surveillance | |
| 2005 (6) | Cessation | Continuing activities: media, Fresh Start, and Not On Tobacco |
| Dejar de Fumar—self-help smoking cessation program | ||
| Student Health Center Pilot—a brief cessation intervention for college students | ||
| Prevention | Continuing activities: media, Healthy Beginnings, and Dejate De Ese Vicio | |
| Mi Familia No Fuma—environmental tobacco smoke prevention using a family cohesion model in the colonias and rural communities | ||
| Policy | Continuing activities: A Smoke-Free Paso del Norte Coalition | |
| Coalition for a Smoke-Free Socorro—support for a clean indoor air ordinance in Socorro, TX | ||
| Junior Educator Corps—provide merchant education about tobacco sales to minors | ||
| Other | Continuing activities: program evaluation and surveillance | |
| 2006 (9) | Cessation | Continuing activities: media and Not on Tobacco |
| Prescription for a Smoke-Free Community—smoking cessation services and nicotine replacement therapy education at participating pharmacies | ||
| StopLite—a brief tobacco cessation intervention for light and intermittent smokers | ||
| Prevention | Continuing activities: media, Healthy Beginnings, and Mi Familia No Fuma | |
| Policy | Continuing activities: A Smoke-Free Paso del Norte Coalition, Coalition for a Smoke-Free Socorro, Otero County Tobacco Educational Coalition, and Junior Educator Corps | |
| Other | Continuing activities: program evaluation and surveillance | |
| 2007 (10) | Cessation | Continuing activities: media, Dejate De Ese Vicio, Fresh Start, Not on Tobacco, Prescription for a Smoke-Free Community, and StopLite |
| Last Drag—tobacco cessation intervention for HIV positive, and lesbian, gay, bisexual, and transgender individuals | ||
| Prevention | Continuing activities: media, Healthy Beginnings, and Mi Familia No Fuma | |
| Smoke-Free Journey—social influences model of prevention, focusing on social and psychological factors to promote tobacco prevention | ||
| Smoke-Free Socials—tobacco prevention activities in an underserved community designated the empowerment zone | ||
| Policy | Continuing activities: Otero County Tobacco Educational Coalition and Junior Educator Corps | |
| Passage of Socorro, TX clean indoor air ordinance—environmental tobacco smoke protection in all enclosed areas within a place of employment to include auditoriums, classrooms, conference and meeting rooms, private offices, company-owned vehicles, employee lounges, stairs, restrooms, and all other enclosed facilities; not outdoor. | ||
| Horizon City Smoke-Free Coalition—organized community organizations and representatives to promote a clean indoor air ordinance in Horizon City, TX | ||
| Other | The University of Texas at El Paso began coordination of initiative efforts | |
| El Paso Tobacco Control Network—Smoke-free community coalition converted to a network | ||
| La Red de Control de Tabaco—established tobacco control network in Ciudad Juarez, Chihuahua, Mexico | ||
| Program evaluation and surveillance—published “Settling the Smoke 2006: A Status Report on Smoking in the Paso del Norte Region” | ||
| 2008 (9) | Cessation | Continuing activities: media, Dejate De Ese Vicio, Fresh Start, Last Drag, Not on Tobacco, Prescription for a Smoke-Free Community, and StopLite |
| Freedom from Smoking—personal support to assist employees to quit smoking | ||
| Texas Youth Tobacco Awareness Program—mandated tobacco awareness classes for minors in possession of tobacco | ||
| Prevention | Continuing activities: media, Mi Familia No Fuma, and Smoke-Free Socials | |
| Policy | Continuing activities: Otero County Tobacco Educational Coalition, Junior Educator Corps, Horizon City Smoke-Free Coalition, and Coalition for a Smoke-Free Socorro (promoting awareness after ordinance passage) | |
| Other | Continuing activities: El Paso Tobacco Control Network and La Red de Control de Tabaco, program evaluation, and surveillance | |
| 2009 (8) | Cessation | Continuing activities: media, Not on Tobacco, StopLite, and Texas Youth Tobacco Awareness Program |
| A Clinical Toolkit for Treating Tobacco Dependence—tobacco cessation promotion with health care providers | ||
| Prevention | Continuing activities: media and Mi Familia No Fuma | |
| Breathe Smart from the Start—tobacco prevention and cessation intervention implemented in prenatal and parenting classes | ||
| 10 Minute Talk—tobacco prevention in high-risk populations | ||
| Get Real About Tobacco—K–12th grade multimedia tobacco prevention program | ||
| Policy | Continuing activities: Junior Educator Corps and Horizon City Smoke-Free Coalition | |
| Tobacco sales checks in Alamogordo, NM | ||
| Coalition for a Smoke-Free Clint—organized community organizations and representatives to promote a clean indoor air ordinance in Clint, TX | ||
| Other | Continuing activities: El Paso and Ciudad Juarez Tobacco Control Network (integrated El Paso and Ciudad Juarez tobacco networks), program evaluation, and surveillance |
Such an initiative enables the coordination of multiple resources to promote comprehensive community tobacco control. Businesses, schools, medical care organizations, universities, foundations, and community agencies all play a role in reducing tobacco use through policy advocacy, communication campaigns, client services, and community programs.5,8,9 A network among these and other partner organizations facilitates the development of goals, the stability of efforts, the integration of activities, and connectivity among public health providers. A network also promotes inclusiveness, the sharing of resources, and the development of new skills and knowledge among members.5
Successful tobacco control involves efforts to reduce demand, decrease supply, improve surveillance, and promote the exchange of scientific and technical information.6,10 Activities that reduce demand include public education and awareness about the effects of tobacco, counteradvertising, increasing tobacco excise taxes, and prevention and cessation programs. Activities that reduce supply include advocating policy to reduce illegal trade and the smuggling of cigarettes, and surveillance encompasses reducing sales to minors and supporting smoke-free indoor air ordinances. Tobacco control networks offer the opportunity to exchange scientific and technical information to emphasize to multiple partners the importance of protecting the public health environment and to promote alternatives to smoking.
Evaluation is imperative for understanding and documenting results and program planning and improvement.11 Discrete program interventions for individuals are more easily evaluated than are multicomponent community-wide initiatives.12 It is rarely feasible to create well-controlled laboratory environments13 and repeat population studies within a true experimental design. Further complicating community-wide evaluation, early change resulting from population-level tobacco control efforts is not easily detectable,14 thereby necessitating the evaluation of longitudinal population trends.
The Behavioral Risk Factor Surveillance System (BRFSS) is the largest continuously active telephone health surveillance system in the world.15 Although limited city-specific inferential ability was reported as a barrier to greater BRFSS use in the past,16,17 steady expansion and larger sample sizes have made it possible to use data for the evaluation of initiatives implemented at the regional, county, or metropolitan level. Given the importance of surveillance and evaluation in tobacco control,6,10 the extensive coverage of large urban areas in Texas makes BRFSS a useful tool for community tobacco control stakeholders in the state. The assessment of smoking status and established smoking (defined as having smoked ≥ 100 cigarettes in a lifetime and currently smoking) enables the observation of both reductions in smoking prevalence and the transition to regular smoking over time.
We assessed the longitudinal impact of a regional tobacco control initiative across 10 years on population-level smoking. We compared cessation- and prevention-related indicators for El Paso, Texas, with the same measures in similar Texas communities from 2000 to 2009. We hypothesized that significant reductions in current smoking and established smoking would be observed in El Paso relative to the 2 other metropolitan statistical areas (MSAs).
We used 10 years of data (2000–2009) from BRFSS,18 a yearly national random-digit dial landline telephone survey of health behaviors the CDC conducts in partnership with state data collection agencies. To account for the population diversity across the nation, the CDC implements disproportionate stratified sampling to provide population estimates of health behaviors. We obtained yearly BRFSS survey data19 and compiled the data for 3 MSAs across the 10 years for which all 3 MSAs have reliable data sets available. We selected Austin-Round Rock, Texas (MSA1) and San Antonio, Texas (MSA2) as comparison MSAs for 4 reasons. First, they are all located in Texas. We did not select cities outside Texas because state tobacco control efforts can be substantially different, particularly over time, and tobacco excise taxes vary by state yet are consistent in Texas. Second, there were no known systematic tobacco control initiatives in the 2 comparison MSAs that spanned the entire 10 years of the study period. For example, it appears that Texas settlement-funded tobacco control efforts in the first half of the decade focused on other communities (e.g., Port Arthur, Beaumont, and Harris County),20 whereas according to most recent evaluation reports, more coordinated efforts appear to have begun in the 2 comparison MSAs much later in the decade.21 However, coordinated efforts were ongoing throughout the decade in El Paso. In addition, Dr. Huang, medical director and health authority of the Austin and Travis County Health and Human Services and formerly chief of the Bureau of Chronic Disease and Tobacco Prevention at the Texas Department of Health, indicated that although documenting the lack of coordinated tobacco control efforts is challenging, it seems clear that San Antonio and Austin had some isolated ongoing tobacco control activities in comparison with El Paso's coordinated, comprehensive approach to tobacco control throughout the decade (written communication, May 2011). Third, to analyze the impact of the initiative over time, statistical power is improved if the maximum number of years is used for trend analysis. All 3 MSAs were surveyed since 2000—the start of the tobacco control initiative in El Paso. Finally, we selected Austin because it is very similar in population size and growth rate to El Paso, and we selected San Antonio because it is very similar to El Paso in demographic makeup.22–24 A more detailed description of BRFSS design and survey methodology is available.18,19
We used BRFSS survey items as outcomes. We defined established smoking25,26 as reporting ever having smoked more than 100 cigarettes in an individual's lifetime and reporting being a current smoker. Given the variability of smoking patterns within the region (e.g., light and intermittent smoking),27 this item was of particular interest because it can be used as a proxy for prevention effects achieved as a potential result of the initiative. Although BRFSS samples only adults, an observed trend toward reduction in the prevalence of established smokers would apply not only to adults but also to individuals who were aged 8 to 9 years or older in 2000 because these individuals would be reflected in BRFSS estimates when they were aged 18 years or older by 2009.
The second outcome was current smoking status. This survey item contains 4 categories: nonsmoker, former smoker, nondaily smoker, and daily smoker. For 1 year (2002), Texas BRFSS data did not contain a 4-category smoking status variable. Rather, the 2002 BRFSS categorized individuals as nonsmokers, nondaily smokers, or daily smokers; former smokers were considered nonsmokers. We did not exclude this year from analyses for 3 important reasons. First, the exclusion of 2002 in smoking status would prevent reasonable yearly trend analysis in nondaily and daily smokers. Second, the exclusion of 2002 likely decreases statistical power to detect trends across MSAs in other smoking status categories. Third, the estimate of the population distribution of the 3 categories of smoking status in 2002 did not statistically vary across the 3 MSAs (P > .14), suggesting that the survey was not biased toward differential smoking status identification among MSAs in 2002. This suggests that trends for each MSA would be similarly influenced by this change in coding in 2002 and less likely to be biased toward estimate changes in one MSA but not another.
In addition to smoking status outcomes, we used demographic and socioeconomic indicators to adjust inferential models. Demographic indicators included gender, age, and ethnicity. We also adjusted education level, income level, and reports of some form of health insurance in inferential models. We included health insurance status because a large number of individuals report no access to health insurance in El Paso,15 and it may be a particularly useful indicator of socioeconomic status along the United States–Mexico border, where El Paso is situated. Individuals may lack health insurance not only because they lack the ability to pay but also because they opt to seek health care across the border in Ciudad Juarez, Chihuahua, Mexico, where care is more affordable and may be more familiar.28 To that end, health insurance serves as a useful proxy for potential acculturation and immigration differences among the 3 MSAs not well nor consistently tracked by BRFSS yet.
We used Stata version 11.0 (StataCorp LP, College Station, TX) for analyses. We employed complex survey design analysis to provide inferences of the population; we used variance linearization and CDC-calculated weights that account for selection and poststratification in the BRFSS disproportionate stratified sampling design. Reports of established smoking and of being a former smoker were not rare outcomes. For that reason, we estimated multivariate relative risk models. The use of relative risk calculation makes the prevalence ratios (PRs) calculated here more conservative and less likely to identify trend effects not actually present in the population relative to the more familiar odds ratio (OR).29,30 We calculated trends in the same manner as previous reports using BRFSS data.31–33 The covariates of interest in these models were the interaction terms produced between linear year (2000–2009) and MSA. We assessed quadratic interaction terms, but they did not offer additional explanation of the data, and we excluded them from final models. We first calculated interaction terms without adjusting for gender, age, ethnicity, education level, income level, and health insurance status (unadjusted models) and then adjusting for these covariates (adjusted models).
The full sample across the 10 years comprised 5132 individuals surveyed from the El Paso MSA, 6930 from MSA1, and 6817 from MSA2. Established smoking and smoking status prevalence point estimates and confidence intervals (CIs) appear in Table 2.
TABLE 2— Yearly Smoking Estimates for each Metropolitan Statistical Area: A Smoke-Free Paso del Norte, El Paso, Austin-Round Rock, and San Antonio, Texas, 2000–2009
| El Paso MSA, % (95% CI) | Comparison MSA1, % (95% CI) | Comparison MSA2, % (95% CI) | |
| BRFSS 2000 | |||
| Smoking status | |||
| Nonsmoker | 0.60 (0.51, 0.69) | 0.51 (0.45, 0.57) | 0.53 (0.47, 0.59) |
| Former smoker | 0.21 (0.14, 0.29) | 0.24 (0.19, 0.30) | 0.25 (0.20, 0.30) |
| Nondaily smoker | 0.07 (0.03, 0.12) | 0.08 (0.04, 0.11) | 0.06 (0.04, 0.09) |
| Daily smoker | 0.12 (0.05, 0.18) | 0.17 (0.12, 0.22) | 0.15 (0.11, 0.19) |
| Smoked > 100 cigarettes in lifetime and currently smoking | |||
| No | 0.76 (0.67, 0.85) | 0.67 (0.61, 0.74) | 0.71 (0.65, 0.78) |
| Yes | 0.24 (0.15, 0.33) | 0.33 (0.26, 0.39) | 0.29 (0.22, 0.35) |
| BRFSS 2001 | |||
| Smoking status | |||
| Nonsmoker | 0.60 (0.52, 0.68) | 0.55 (0.49, 0.61) | 0.61 (0.56, 0.66) |
| Former smoker | 0.19 (0.13, 0.25) | 0.25 (0.20, 0.29) | 0.23 (0.19, 0.28) |
| Nondaily smoker | 0.06 (0.02, 0.11) | 0.08 (0.05, 0.12) | 0.05 (0.03, 0.07) |
| Daily smoker | 0.15 (0.09, 0.21) | 0.12 (0.08, 0.16) | 0.10 (0.07, 0.13) |
| Smoked > 100 cigarettes in lifetime and currently smoking | |||
| No | 0.74 (0.65, 0.82) | 0.73 (0.67, 0.79) | 0.80 (0.75, 0.85) |
| Yes | 0.26 (0.18, 0.35) | 0.27 (0.21, 0.33) | 0.20 (0.15, 0.25) |
| BRFSS 2002 | |||
| Smoking status | |||
| Nonsmoker | 0.81 (0.74, 0.87) | 0.83 (0.78, 0.88) | 0.74 (0.66, 0.81) |
| Former smoker | … | … | … |
| Nondaily smoker | 0.04 (0.02, 0.07) | 0.07 (0.04, 0.11) | 0.07 (0.04, 0.10) |
| Daily smoker | 0.15 (0.09, 0.22) | 0.10 (0.06, 0.13) | 0.20 (0.12, 0.27) |
| Smoked > 100 cigarettes in lifetime and currently smoking | |||
| No | 0.58 (0.50, 0.66) | 0.57 (0.50, 0.64) | 0.56 (0.49, 0.63) |
| Yes | 0.42 (0.34, 0.50) | 0.43 (0.36, 0.50) | 0.44 (0.37, 0.51) |
| BRFSS 2003 | |||
| Smoking status | |||
| Nonsmoker | 0.62 (0.53, 0.70) | 0.55 (0.49, 0.60) | 0.56 (0.50, 0.61) |
| Former smoker | 0.19 (0.13, 0.25) | 0.24 (0.20, 0.29) | 0.22 (0.18, 0.27) |
| Nondaily smoker | 0.09 (0.03, 0.15) | 0.07 (0.04, 0.10) | 0.08 (0.04, 0.12) |
| Daily smoker | 0.10 (0.05, 0.16) | 0.14 (0.10, 0.18) | 0.14 (0.10, 0.19) |
| Smoked > 100 cigarettes in lifetime and currently smoking | |||
| No | 0.76 (0.67, 0.85) | 0.72 (0.66, 0.78) | 0.71 (0.65, 0.78) |
| Yes | 0.24 (0.15, 0.32) | 0.28 (0.22, 0.33) | 0.29 (0.22, 0.35) |
| BRFSS 2004 | |||
| Smoking status | |||
| Nonsmoker | 0.61 (0.56, 0.66) | 0.59 (0.54, 0.63) | 0.57 (0.53, 0.62) |
| Former smoker | 0.21 (0.17, 0.25) | 0.23 (0.19, 0.27) | 0.21 (0.17, 0.24) |
| Nondaily smoker | 0.09 (0.06, 0.12) | 0.08 (0.05, 0.10) | 0.08 (0.06, 0.11) |
| Daily smoker | 0.09 (0.06, 0.12) | 0.11 (0.08, 0.14) | 0.14 (0.10, 0.17) |
| Smoked > 100 cigarettes in lifetime and currently smoking | |||
| No | 0.77 (0.72, 0.82) | 0.76 (0.71, 0.81) | 0.72 (0.68, 0.77) |
| Yes | 0.23 (0.18, 0.28) | 0.24 (0.19, 0.29) | 0.28 (0.23, 0.32) |
| BRFSS 2005 | |||
| Smoking status | |||
| Nonsmoker | 0.62 (0.58, 0.67) | 0.60 (0.54, 0.65) | 0.58 (0.53, 0.63) |
| Former smoker | 0.19 (0.16, 0.23) | 0.22 (0.18, 0.26) | 0.25 (0.21, 0.29) |
| Nondaily smoker | 0.08 (0.05, 0.11) | 0.08 (0.04, 0.12) | 0.06 (0.03, 0.09) |
| Daily smoker | 0.10 (0.07, 0.13) | 0.11 (0.07, 0.14) | 0.11 (0.08, 0.14) |
| Smoked > 100 cigarettes in lifetime and currently smoking | |||
| No | 0.77 (0.73, 0.82) | 0.76 (0.70, 0.82) | 0.77 (0.69, 0.83) |
| Yes | 0.23 (0.18, 0.27) | 0.24 (0.18, 0.30) | 0.23 (0.17, 0.31) |
| BRFSS 2006 | |||
| Smoking status | |||
| Nonsmoker | 0.66 (0.60, 0.72) | 0.58 (0.53, 0.64) | 0.59 (0.53, 0.65) |
| Former smoker | 0.19 (0.15, 0.23) | 0.23 (0.19, 0.27) | 0.23 (0.18, 0.27) |
| Nondaily smoker | 0.10 (0.05, 0.16) | 0.06 (0.04, 0.09) | 0.03 (0.01, 0.05) |
| Daily smoker | 0.05 (0.02, 0.07) | 0.12 (0.09, 0.16) | 0.15 (0.10, 0.21) |
| Smoked > 100 cigarettes in lifetime and currently smoking | |||
| No | 0.79 (0.73, 0.84) | 0.76 (0.70, 0.81) | 0.77 (0.73, 0.80) |
| Yes | 0.21 (0.16, 0.27) | 0.24 (0.19, 0.30) | 0.23 (0.20, 0.27) |
| BRFSS 2007 | |||
| Smoking status | |||
| Nonsmoker | 0.64 (0.60, 0.67) | 0.59 (0.55, 0.62) | 0.60 (0.57, 0.64) |
| Former smoker | 0.18 (0.16, 0.20) | 0.24 (0.21, 0.27) | 0.21 (0.19, 0.24) |
| Nondaily smoker | 0.08 (0.06, 0.11) | 0.07 (0.05, 0.09) | 0.06 (0.04, 0.08) |
| Daily smoker | 0.10 (0.08, 0.12) | 0.10 (0.08, 0.13) | 0.12 (0.09, 0.14) |
| Smoked > 100 cigarettes in lifetime and currently smoking | |||
| No | 0.77 (0.74, 0.81) | 0.77 (0.74, 0.81) | 0.76 (0.72, 0.80) |
| Yes | 0.23 (0.19, 0.26) | 0.23 (0.19, 0.26) | 0.24 (0.20, 0.28) |
| BRFSS 2008 | |||
| Smoking status | |||
| Nonsmoker | 0.61 (0.55, 0.67) | 0.56 (0.52, 0.61) | 0.59 (0.55, 0.63) |
| Former smoker | 0.23 (0.18, 0.28) | 0.24 (0.21, 0.28) | 0.23 (0.20, 0.25) |
| Nondaily smoker | 0.08 (0.05, 0.11) | 0.10 (0.06, 0.13) | 0.07 (0.05, 0.10) |
| Daily smoker | 0.08 (0.05, 0.11) | 0.10 (0.06, 0.13) | 0.11 (0.09, 0.14) |
| Smoked > 100 cigarettes in lifetime and currently smoking | |||
| No | 0.79 (0.73, 0.84) | 0.74 (0.68, 0.80) | 0.76 (0.72, 0.80) |
| Yes | 0.21 (0.16, 0.27) | 0.26 (0.20, 0.32) | 0.24 (0.20, 0.28) |
| BRFSS 2009 | |||
| Smoking status | |||
| Nonsmoker | 0.65 (0.61, 0.69) | 0.58 (0.53, 0.63) | 0.59 (0.54, 0.64) |
| Former smoker | 0.20 (0.17, 0.23) | 0.29 (0.24, 0.33) | 0.26 (0.22, 0.31) |
| Nondaily smoker | 0.06 (0.04, 0.09) | 0.04 (0.02, 0.07) | 0.06 (0.03, 0.08) |
| Daily smoker | 0.09 (0.06, 0.11) | 0.09 (0.06, 0.12) | 0.09 (0.06, 0.12) |
| Smoked > 100 cigarettes in lifetime and currently smoking | |||
| No | 0.81 (0.77, 0.85) | 0.81 (0.77, 0.86) | 0.80 (0.75, 0.85) |
| Yes | 0.19 (0.15, 0.23) | 0.19 (0.14, 0.23) | 0.20 (0.15, 0.25) |
Note. BRFSS = Behavioral Risk Factor Surveillance System; CI = confidence interval; MSA = metropolitan statistical area.
El Paso exhibited a statistically significant adjusted (PR = 0.93; 95% CI = 0.89, 0.98) yearly trend for prevalence reduction in established smoking. However, trend coefficients were not statistically significant for MSA1 or MSA2 in the adjusted binomial models (both linear trend Ps > .36). The covariate adjusted predicted prevalence change of established smokers each year is presented in Figure 1 for all 3 MSAs. Unadjusted models demonstrated the same pattern of results.
FIGURE 1— Change in each metropolitan statistical area (MSA) in predicted prevalence in established smoking: A Smoke-Free Paso del Norte, El Paso, Austin-Round Rock, and San Antonio, TX, 2000–2009.
Note. The y-axis denotes the predicted prevalence of established smoking (≥ 100 cigarettes smoked in lifetime and currently smoking). The x-axis represents each year of Behavioral Risk Factor Surveillance data.*P < .01.
We observed no significant reduction in nondaily smokers in any MSA in the adjusted model (all Ps > .27). Whereas MSA1 and MSA2 exhibited no statistically significant growth trends in former smokers in the adjusted multinomial model (all Ps > .43), El Paso exhibited an adjusted increase (PR = 1.06; 95% CI = 1.01, 1.11) in the PR of former smokers. Further, El Paso evidenced a trend of reduced prevalence of daily smokers in adjusted models (PR = 0.94; 95% CI = 0.89, 0.99)—an approximate 6% reduction in the prevalence of daily smokers per year. Whereas statistical trends were observed in El Paso, the comparison MSAs exhibited no statistical change over this 10-year period. The covariate adjusted predicted prevalence change of both former and daily smokers each year is presented in Figure 2 for all 3 MSAs. Unadjusted models demonstrated the same pattern of results.
FIGURE 2— Covariate adjusted predicted change in prevalence of former and daily smokers across 3 metropolitan statistical areas (MSAs): A Smoke-Free Paso del Norte, El Paso, Austin-Round Rock, and San Antonio, TX, 2000–2009.
Note. The y-axis represents the predicted prevalence of former and daily smokers. The x-axis represents each year of Behavioral Risk Factor Surveillance data. The 3 lines in each of the 2 plots represent linear trends.*P < .05.
Using external information available from the CDC,34 an economic extrapolation of the daily smoker point estimate in El Paso is possible. Given the daily smoker prevalence change estimate per year and the year 2000 population estimate of adults in El Paso (smallest population estimate) and assuming the lowest observed point estimate of daily smokers (5% in 2005) across the 10 years, the number of people changing their daily smoking status can be estimated. This extrapolation translates to approximately 0.3% fewer adult daily smokers each year relative to the year before in El Paso. This estimate translates to approximately 1167 fewer daily smokers per year. The CDC34 estimated that the economic cost to the nation is $3391.00 per smoker per year, or $4020.65 when adjusting for inflation. Using this adjustment for yearly cost of inflation to the US economy and assuming these 1167 residents will not resume smoking, the inflation-adjusted economic savings associated with El Paso tobacco control efforts is $4.69 million per year. Economic savings would likely be even greater if significant reductions in the prevalence of established smoking could be taken into account.
Although these repeated cross-sectional population-level trends are correlational, they suggest that a community initiative can promote tobacco prevalence reductions within a population. Rates of established smoking and daily smoking both decreased in El Paso but not in the comparison areas, where tobacco control efforts were less coordinated and comprehensive. Nondaily smoking prevalence did not demonstrate similar declines in any of the studied MSAs. As many low-level smokers do not self-identify as smokers,35 population-based efforts may be particularly poorly suited to nondaily smokers; thus, future efforts may need to take into account emerging characteristics associated with nondaily smoking.36
The isolation of the individual impact of any given component of a tobacco control initiative is challenging. Nevertheless, highlighting how the most prominent components may have contributed to reduce established and daily smoking despite not helping nondaily smokers can meaningfully contribute to future tobacco control efforts by noting past and present successes as well as future directions for enhanced population-based assessment and intervention.
Advocating for CIAOs within regional municipalities was and continues to be a centerpiece of the initiative. The impact of these CIAOs on smoking prevalence, enacted for El Paso in 2002 (vs 2005 and 2008 in Austin and 2010 in San Antonio), cannot be underestimated. Further, both smokers and nonsmokers in the region have viewed CIAOs favorably.7 CIAOs are critical in comprehensive tobacco control,6,37 and relative to the rest of Texas and the nation, El Paso's is strong.38,39 Not only do CIAOs strengthen tobacco control efforts at the policy level, but clean air policy changes also appear to increase media and community attention toward smoke-free environments,40–42 promote attention and activity in neighboring communities,43 provide a public health learning experience to those interested in stronger tobacco control,43,44 and heighten the awareness of and demand for prevention and cessation services. Nondaily smokers may be less affected by CIAO efforts, particularly if they do not self-identify as smokers. Future efforts may wish to assess nondaily smokers’ attitudes, beliefs, and behaviors associated with CIAOs with particular attention to how policy promotion and change may better affect tobacco prevalence reductions in low-level smokers.
Beyond national campaigns that all comparison areas may have been exposed to, the El Paso community has had consistent smoking cessation– and tobacco use prevention–based media messaging with particular aims to reduce regional tobacco prevalence. Mass media are effective in promoting cessation among adults, demonstrate some effectiveness in preventing tobacco use in youths, and appear to educate the public about smoking consequences, tobacco industry efforts, and the benefits of cessation.45–47 Frequency, duration, and intensity (e.g., emotional content) of messaging have all been shown to increase mass media effectiveness.45,46,48 Regional campaigns may have targeted regular smokers well in terms of cessation and youths in terms of prevention; however, recent messaging efforts may be missing the nondaily smoking group. Recognizing the high prevalence of low-level smoking in the region,27 the adult cessation campaign in almost all messaging has attempted to reach nondaily smokers by beginning calls to action with the phrase “whether you smoke a little or a lot.” However, current findings and recent studies suggest the need for more targeted nondaily smoker messaging with attention to frequency, duration, and intensity of content to promote cessation among this group.45,46,48
Although prevention programming, particularly in schools, has met with mixed results in terms of reductions in tobacco initiation,49 multiple agencies in the region have been funded by the initiative to focus on tobacco prevention, particularly through education and policy enforcement. Current results suggest that comprehensive efforts have resulted in reduced transition to established smoking within the region.
The population impact of cessation services likely has both explicit (cessation) and implicit (promoting a smoke-free community) outcomes. The Paso del Norte Health Foundation funded regional nonprofit organizations, school districts and universities, pharmacies, and the state and city health departments to offer cessation services, including QuitLine and QuitNet access. Having access to multiple screening and cessation services increases the likelihood that an individual will quit.50,51 Despite the consistent presence of cessation programs, only recently has an initiative-funded program targeted light and intermittent smokers. More or more intense programming may be warranted to promote nondaily smoking reductions, or these programs’ impact may take more time to accrue.
Furthermore, active encouragement of cessation promotes a smoke-free culture52 because even when individuals are not necessarily ready to quit, cessation programs serve as a source of support and education.53 Notably, the majority of smokers quit unaided by intervention54; however, the personal impetus to quit smoking may not be unaided. The degree to which the presence of cessation resources and other tobacco control efforts in the community motivate smokers to quit may be significant at the population level. Less is understood of those who opt to quit unaided,54 and future community success in promoting cessation may benefit from increased focus on what aspects of comprehensive tobacco control motivate individuals to quit smoking not only with professional help but also without it, particularly as unassisted cessation may be related to prevalence changes in nondaily smoking.
When coalitions advocate reasonably, meaningful change can be achieved in tobacco control policies.55,56 Furthermore, coalitions are in a strong position to better involve and collaborate with stakeholders in public and private sectors57 and can educate the public about the importance of collective health.58 Similarly, networks can build and maintain partnerships among tobacco control stakeholders to share information and resources and to improve the inclusiveness of health expertise.5,59 El Paso has had a network of regional tobacco control experts and stakeholders for the past 10 years. The steady increase in systems thinking about tobacco control during the initiative has likely helped to contribute to a collective effort5 demonstrating a population-based impact in reductions of established and daily smoking but not nondaily smoking. Future network efforts should maintain a strong focus on policy promotion, media messaging, and prevention and cessation programming; but an emphasis on ways to promote nondaily smoking cessation in a comprehensive fashion is warranted.
Four study limitations are noteworthy. First, the isolation of individual effects of any given component of the initiative on prevalence assessments is difficult to achieve using aggregate population data60 and quasi-experimental methods. As such, this aggregate assessment remains correlational, although we made every effort to select communities suitable to examine prevalence trends relative to El Paso. Future researchers may wish to enhance measures and subsequent study design and analyses to enable a more nuanced understanding of individual tobacco control components and their synergistic impact on tobacco reductions. Second, data are cross-sectional, which may be limited by selection bias not captured in the complex survey weights and design; however, a meaningful benefit is that the data are representative of the population over the period.61 Third, evidence suggests some response bias with the BRFSS survey methodology and design, leading to some groups being underrepresented (e.g., younger individuals).62 Finally, as El Paso is situated on the border of the United States and Mexico, issues such as cigarette smuggling63 may impede initiative efforts and dampen observed effects, whereas border crossing to access more affordable health care in Mexico (which is difficult to document or measure) may artificially inflate observed effects.
A strength of this study is that it assessed the relationship of a tobacco control initiative to smoking prevalence at the population level. This assessment used population-level data supported by the CDC over 10 years of sustained activities, thereby facilitating an assessment of where population trends are present and where they are not.
We observed a clear trend toward reduction in the prevalence of established and daily smokers for El Paso over the course of a 10-year tobacco control initiative. This trend was not present for cities of similar size and demographic makeup within the state. Meaningful reductions in the prevalence of smoking at the community level can occur through the utilization of multiple tobacco control activities housed within a comprehensive tobacco control initiative. Nevertheless, a more extensive focus on the growing subgroup of nondaily smokers seems especially warranted.
Acknowledgments
A Smoke-Free Paso del Norte, an initiative of the Paso del Norte Health Foundation funded this work.
The authors would like to acknowledge Enrique Mata, Ida Ortegon, Dennis Smith, members of the Prevention and Treatment in Clinical Health lab, and members of the El Paso–Ciudad Juarez Tobacco Control Network.
Human Participant Protection
Institutional review board approval was not needed as data were accessed through the Texas Department of State Health Services Center for Health Statistics and no primary data collection occurred at any affiliated institution.
