If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Address correspondence to: Rebekah R. Rhoades, MPH, Department of Biostatistics and Epidemiology, College of Public Health, University of Oklahoma Health Sciences Center, 801 NE 13th Street, Oklahoma City OK 73104
The Oklahoma Communities of Excellence in Tobacco Control (CX) program was established in 2004. Thirty-three CX grants have been funded to implement comprehensive tobacco control programs in 50 counties and one tribal nation.
Purpose
To describe local tobacco policy gains within CX counties and short-term and intermediate outcomes such as Oklahoma Tobacco Helpline registration and awareness, quit attempts, and home smoking bans among adults.
Methods
A before-and-after study design examined outcomes within CX counties and among CX urban and rural counties. Comparisons were made with non-funded counties when possible. Local policy tracking databases were reviewed for the number of policies implemented from 2004 to 2013 in CX counties. Population-level tobacco indicators, using Helpline registration and 2004–2010 Behavioral Risk Factor Surveillance System data, quantified changes over time. Data were collected in 2003–2013 and analyzed in 2013.
Results
Eight hundred thirty-one legislated and voluntary policies were implemented in CX counties and high levels of Helpline registration were maintained. Statistically significant increases were observed in CX counties for the proportion of smokers making a quit attempt, Helpline awareness, and home smoking bans among smokers. These observed increases were greater in rural CX counties than urban. Non-CX counties only experienced a statistically significant increase in Helpline awareness.
Conclusions
Using community-based best practices in tobacco control while focusing on social norm change, CX counties experienced positive changes in smoking-related attitudes and behaviors. This study expands the evidence base for statewide tobacco control programming and underscores the value of community-based tobacco control programs.
Introduction
To effectively counter pro-tobacco influences, local communities must be involved in strategies to change the “way tobacco is promoted, sold, and used while changing the knowledge, attitudes, and practices of tobacco users and nonusers.”
The Tobacco Settlement Endowment Trust (TSET) selected the Communities of Excellence Plus in Tobacco Control (CX) framework for its community-based program because of its grounding in social norm change theory and its success in other states.
California Department of Health Services, Tobacco Control Section. A model for change: the California experience in tobacco control. Sacramento CA, 1998.
The social norm change approach uses policy and systems strategies to alter the social and legal climate and establish environments where tobacco is not the norm. The community program launched in October 2004 to address four priority areas: (1) eliminate secondhand smoke exposure; (2) prevent youth initiation; (3) promote tobacco cessation services; and (4) reduce tobacco industry influences. A statewide request for proposals (RFP) solicited applications. Tribal nations and single counties, or a consortium of counties if a county’s population was <15,000, were encouraged to apply. Potential applicants were informed that only one application per county would be funded. Eligibility criteria included an existing coalition, a lead agency selected by the coalition, and, in later years, documentation that the lead agency had a tobacco-free workplace policy and did not accept tobacco industry funds. Proposals were reviewed by an external panel of national experts with final funding decisions made by TSET’s Board of Directors. Seventeen of the initially funded 20 grantees (CX1) continued after the planning year, which concluded in September 2005. In October 2007, the 5-year grant program expanded to seven new grantees (CX2), followed by a third cohort of ten grantees (CX3) in October 2010. Currently, 33 grantees have been funded to implement comprehensive programs in 50 counties and one tribal nation, and about 85% of Oklahomans live in a funded community (Figure 1, Table 1).
Figure 1Communities of Excellence in Tobacco Control grantees map.
Community grantees promote effective policies and programs to reduce tobacco use at the local level. Each year, grantees submit a strategic plan that includes detailed information on objectives designed to impact the grant’s outcomes, many of which measure passage of ordinances and policies. However, owing to preemption, communities are limited in their ability to implement key policies that are effective in reducing tobacco use and for which other states have seen successful, widespread grassroots movements. Oklahoma is one of few states that have preemptive provisions that limit local government authority to enact ordinances that restrict smoking in workplaces and public places, tobacco advertising, and youth access to tobacco products.
Policies that are not prohibited by preemption include ordinances that mirror state law, ordinances making city-owned/-operated properties (such as outdoor recreational areas) smoke-/tobacco-free, and voluntary policies by schools and private entities that establish tobacco-free properties. Pursuing local ordinances that mirror state law was prioritized to initiate relationships with city councils and other community leaders, create a public conversation to educate decision makers and the public, and increase support for tobacco control policies. This strategy also served to demonstrate the limitations preemption places on a community determined to improve opportunities for health and economic development. Case studies exploring grantees’ roles in promoting policy passage are reported elsewhere in this supplement.
CDC has highlighted their potential to help increase the number of tobacco users making quit attempts and improve their success rate, thus leading to higher overall cessation rates.
CX grantees support statewide promotion of the Helpline through relationships with local organizations, paid media, customizable press releases, community events, other forms of earned media, and widespread dissemination of Helpline materials.
The purpose of this paper is to describe local policy accomplishments within counties served by grantees and changes in key short-term and intermediate tobacco-related outcomes among adults, including quit attempts, home smoking bans among smokers, and Helpline awareness and utilization.
Methods
Available data on local policy gains and population-level tobacco indicators were systematically examined to describe changes in counties that were awarded TSET CX funding (CX counties). Data were collected from 2003 to 2013 and analyzed in 2013. A before-and-after study design was used to examine outcomes within funded counties over time. Comparisons were made with non-funded (non-CX) counties when possible. CX and non-CX counties are not entirely comparable because of self-selection into the CX program and differences in demographic characteristics, particularly race and education. Evaluating changes in the selected population-level tobacco indicators among non-CX counties provided information on secular trends that may be occurring in the absence of the CX program.
Local Policy
The Oklahoma State Department of Health (OSDH) tracks local policy among CX grantees. Grantees submit local policies to the OSDH for validation. Policies meeting inclusion criteria established by the grant’s outcome measures are then added to the database. This database was used to identify policies meeting grant criteria and passed from October 1, 2004, through June 30, 2013, in CX counties. Five types of policies were included in this study: (1) local ordinances mirroring the state Smoking in Public Places and Indoor Workplaces Act; (2) local ordinances mirroring the state Prevention of Youth Access to Tobacco Act; (3) 24/7 tobacco-free public school district policies (no tobacco use by staff, students, and visitors on school grounds, property, or school-related events); (4) tobacco-free worksite policies; and (5) local ordinances, resolutions, or voluntary policies establishing smoke-/tobacco-free outdoor recreational areas.
Data from all three waves of CX grants were included in the policy review. The strength of tobacco control policies (clean indoor air and youth access ordinances, schools, and worksites) was consistent across grantees because of the inclusion criteria established by the grant’s outcome measures and reinforced by OSDH-developed model policies. The level of community readiness influenced the extent of coverage for outdoor recreational policies, and grant outcome standards were not defined. Thus, outdoor recreational policies had greater variation. Policy adoption occurred along a continuum that included city ordinances, resolutions, and voluntary policy. Policies may have stipulated entirely tobacco-free properties, entirely smoke-free properties, or allowed for exempt areas. Policies passed in CX counties were monitored using the grant’s measured outcomes, work plans, and reports that provided details of meetings with key community leaders, presentations to city councils and school boards, and media. Policies passed in non-CX-funded counties are not systematically tracked or routinely included in the policy-tracking database.
Helpline Utilization
Helpline registrants are adult tobacco users who call the Oklahoma Tobacco Helpline and register for the one-call program, multiple-call program, or self-help materials. County of residence is collected at time of registration. To assess the reach of Helpline promotions in CX and non-CX counties, the average number of Helpline registrants per month per 10,000 smokers for each grant year was calculated for CX1, CX2, and non-CX counties. CX3 counties were treated as non-CX counties because they were not funded until the last quarter of 2010.
To identify differences in Helpline reach over time in CX and non-CX counties, the number of Helpline registrants per month per smoking population was also modeled using robust regression analysis, which accounted for outlying data such as the spike in registration due to the 2005 state tobacco excise tax and 2009 federal cigarette excise tax increases. Two models were created to account for the later date of implementation for CX2 counties. The first examined the time period from October 2003 through September 2010, and analyzed the differences in Helpline registrations in CX1 and non-CX counties. The second examined the time period from October 2007 through September 2010, and analyzed differences in CX1, CX2, and non-CX counties. Non-CX counties were defined as counties not included in either CX1 or CX2 per the specified grant year. Statistical significance was set at the α <0.05 level.
Quit Attempts, Home Smoking Bans, and Helpline Awareness
Behavioral Risk Factor Surveillance System (BRFSS) data were used to estimate key tobacco indicators, including quit attempts, home smoking bans among smokers, and Helpline awareness among adults in CX and non-CX counties. Additionally, because county population size or density may affect program outcomes, CX counties were designated as either urban or rural using the Oklahoma State University Center for Rural Health’s designation of Oklahoma counties as a guide.
The urban category includes the four most populous counties in the state and accounts for about 45% of the state’s population. The remaining CX counties were categorized as rural. CX3 counties were treated as non-CX counties because they were not funded until the last quarter of 2010. Baseline estimates were obtained from the 2004 Oklahoma BRFSS. Follow-up estimates were obtained from the 2010 Oklahoma BRFSS. Comparisons with 2011 or 2012 BRFSS estimates were not possible because of methodologic changes.
Estimates of Helpline awareness and the proportion of home smoking bans were obtained from state-added BRFSS questions. These questions were asked of all Oklahoma BRFSS participants in 2004 and half of the BRFSS participants in a split sample in 2010.
A quit attempt among smokers was defined as stopping smoking for ≥1 day within the preceding 12 months because the smoker was trying to quit smoking. Home smoking bans among smokers were defined as a rule that smoking is never allowed inside a participant’s home, not counting decks, porches, or garages. Awareness of Oklahoma’s Tobacco Helpline was established among all BRFSS respondents with a yes response to the question: Are you aware of any 1-800, or toll-free, smoking quitlines in Oklahoma that a smoker can call to receive help quitting smoking?
Weighted proportions for each measure were estimated to represent the Oklahoma population using Proc Surveyfreq in SAS, version 9.3. Ninety-five percent CIs for each proportion were calculated and reported. Statistical significance was assessed using a chi-square test with α <0.05.
Results
Local Policy
By 2013, a total of 92 ordinances mirroring the state Smoking in Public Places and Indoor Workplaces Act and 88 ordinances mirroring the state Prevention of Youth Access to Tobacco Act had been implemented in CX counties (Table 2).
Fifteen cities and towns passed resolutions calling for the Oklahoma State Legislature to repeal preemption and return local rights to the community (P. Warlick, American Cancer Society Cancer Advocacy Network, Inc., personal communication, 2013). The movement to prohibit smoking or tobacco use in outdoor recreational areas gained momentum with a total of 50 ordinances, resolutions, and policies; thirty-seven policies have been implemented since 2010. Thirty-four policies prohibited either smoking or all tobacco use in hundreds of city-owned/-operated outdoor recreational areas (Table 2). School districts in CX counties established tobacco-free environments for students, families, and the community, with 292 districts implementing tobacco-free school policies 24 hours a day 7 days a week (state law only requires Monday through Friday from 7:00am to 4:00pm). The number of school districts with policies prohibiting tobacco use on school property and events increased from 33 before the program launched to 292 (784% increase) at the close of Grant Year 2013. The number of tobacco-free worksite policies went from one known policy prior to the launch of the program to 309 documented policies at the close of Grant Year 2013 (Table 2).
Includes policies for five school districts that closed/merged.
33
51
78
108
129
157
185
202
270
292
Tobacco-free worksite policies
1
4
11
18
23
40
90
146
226
309
Smoke-/tobacco-free outdoor recreational area policies
0
0
0
1
4
9
13
21
43
50
CIA, Clean Indoor Air (Smoking in Public Places and Indoor Workplaces Act); CX, Communities of Excellence Plus in Tobacco Control; GY, Grant Year (October-September); YA, Youth Access (Prevention of Youth Access to Tobacco Act).
a CX2 funded.
b CX3 funded.
c Includes policies for five school districts that closed/merged.
CX1 counties had a higher average number of Helpline registrants per adult smoking population as compared to non-CX counties throughout all years (Table 3). CX2 counties had the highest average number of registrants per smoking population in the Grant Years 2009 and 2010.
Table 3Average number of Helpline registrants per month per 10,000 smokers in CX and non-CX counties
For the regression analysis of Helpline registration from October 2003 to September 2010 comparing CX1 to non-CX counties, there was a non-significant interaction between CX group and time (p=0.30), indicating that the change over time in registrants per 10,000 smokers was similar in CX1 and non-CX groups (i.e., there was no “time” effect). However, the average number of registrants per 10,000 smokers was significantly greater in CX1 than in non-CX counties (p<0.0001). Figure 2 shows the predicted number of registrants from the robust regression model with the actual number of registrants per 10,000 smokers.
Figure 2Registrants per 10,000 smokers, CX1 and non-CX counties, with predicted curve from robust regression (two images).
Results were similar for the regression analysis of Helpline registration from October 2007 to September 2010 comparing CX1, CX2, and non-CX counties. There was a non-significant interaction between CX group and time for CX2 versus non-CX counties (p=0.30) and for CX1 versus non-CX counties (p=0.60), indicating that the change over time in registrants per 10,000 smokers was similar among CX1, CX2, and non-CX groups. Again, the number of registrants per 10,000 smokers was significantly greater in CX1 (p<0.0001) and CX2 (p=0.008) counties when compared to non-CX counties (Figure 3).
Figure 3Registrants per 10,000 smokers, CX1, CX2, and non-CX counties, with predicted curve from robust regression.
Quit Attempts, Home Smoking Bans, and Helpline Awareness Among Adults
Cigarette smokers making at least one quit attempt in CX counties increased significantly, from 51.6% in 2004 to 63% in 2010 (p=0.0004). A small, non-significant increase was observed among smokers in non-CX counties. Although significant gains were observed in quit attempts among both urban and rural CX counties, the relative increase in quit attempts was twice as high in rural CX counties (34.5%) compared to urban CX counties (15.7%).
Significant increases in Helpline awareness were observed in both CX and non-CX counties. In CX counties, awareness increased from 25.0% in 2004 to 47.7% in 2010 (p<0.0001). In non-CX funded counties, awareness increased from 24.8% to 44.2% (p<0.0001). Among CX counties, significant increases in awareness were observed among both urban and rural CX counties (Table 4). Similar to quit attempts, Helpline awareness showed greater increases in rural CX counties (121.3%) than in urban CX counties (73.7%).
Table 4Short-term and intermediate tobacco-related outcomes in CX and non-CX counties,
Smokers reporting home smoking bans increased significantly in CX counties, from 36.1% in 2004 to 43.3% in 2010 (p=0.0286), whereas the proportion of smoking bans in smokers’ homes remained essentially unchanged in non-CX counties (38.2% to 39.5%, p=0.83) during the same time period. In addition, the proportion of smokers with home smoking bans significantly increased by 52.5% (28.2% to 43.0%, p=0.0030) in rural counties, but rose just 6.6% (40.9% to 43.6%, p=0.5293) among smokers living in urban CX counties.
Discussion
Funded communities implemented and maintained a strong local public health advocacy landscape, resulting in the passage of 831 legislated and voluntary policies. Pursuing voluntary policy and local ordinances that mirror state law called attention to tobacco use, and helped generate public debate that built support for other tobacco-related public policies, including repeal of preemption. Beginning in Grant Year 2009, some communities extended the reach of local policy by making outdoor recreational areas smoke-/tobacco-free. The social norm paradigm asserts that key systems and policy change can lead to positive trends in smoking-related attitudes and behaviors.
The Oklahoma Tobacco Helpline has been ranked in the top ten quitlines for reach and state investment since national benchmarking activities began in 2008.
This study demonstrated that CX counties maintained higher rates of Helpline registration than non-CX counties. Studies have shown that state quitline call volumes are driven by both statewide media promotion and local efforts.
In terms of statewide Helpline promotions, CX and non-CX counties experience similar levels of exposure. However, grant outcomes required Helpline promotion and most local earned and paid media within CX counties included the Helpline phone number. Thus, the “value added” for local promotion of the Helpline within CX counties is suggested by these analyses.
Tobacco-related disparities in rural communities may partially explain differences in rural and urban CX counties. The American Lung Association
found that rural residents are more likely to use tobacco, be heavy smokers, have higher rates of smokeless tobacco use, allow smoking in their homes and cars, and accept tobacco use as the norm. TSET was intentional in its efforts to fund CX grantees in all areas of the state, including highly rural counties. Implementation of these community-based programs offered the means and framework to approach tobacco control via a comprehensive effort in previously under-resourced rural communities. Comparison of population-level tobacco indicators in rural and urban CX counties demonstrated differences between rural and urban funded counties. Increases in quit attempts, Helpline awareness, and home smoking bans among smokers were observed for both urban and rural funded counties. However, the magnitude of change was larger in rural CX counties than in urban CX counties. The greater gains seen in rural CX counties compared to urban CX counties may be due to tobacco-related disparities, suggesting that the CX framework may have greater impact in less-populated areas. The urban–rural differences observed in this study warrant further investigation.
This study has several limitations. Causal relationships between provision of CX grant funding and the number of passed policies cannot be established with this study design, as there is no equivalent comparison group and systematic policy tracking was not performed for non-CX counties. Selection bias is a possibility for all studied indicators, as counties self-select into the intervention by responding to the RFP and accepting the grant award. CX counties also have different demographic profiles than non-CX counties, particularly in terms of race and education. Finally, the analyses did not include multivariate techniques designed to control for potential confounding by other variables such as age, gender, race, and SES.
Although the CX framework has existed for many years, the number of published studies from states implementing the framework in a systematic, statewide manner is limited. The results of this study expand the evidence base for community-based tobacco control programming, which is continually challenged when funding is limited. Within its limitations, this study demonstrates that Oklahoma’s funding of local communities to tailor and systematically implement the CX approach was associated with positive changes in key outcomes related to tobacco use. Communities have implemented evidence-based strategies using a comprehensive approach and worked with multiple sectors throughout the community. These efforts enhance Oklahoma’s strides in reducing tobacco consumption and prevalence.
Publication of this article was supported by the Oklahoma Tobacco Research Center (OTRC), with funding from the Oklahoma Tobacco Settlement Endowment Trust (TSET).
This study was funded by TSET.
No financial disclosures were reported by the authors of this paper.
References
CDC
Best Practices for Comprehensive Tobacco Control Programs—2007.
USDHHS, CDC, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health,
Atlanta GA2007
California Department of Health Services, Tobacco Control Section. A model for change: the California experience in tobacco control. Sacramento CA, 1998.
00577-7&id=gr1.jpg){kind=link}
00577-7&id=gr2.jpg){kind=link}
00577-7&id=gr3.jpg){kind=link}