HHS Public Access Author manuscript Author Manuscript
J Okla State Med Assoc. Author manuscript; available in PMC 2016 November 01. Published in final edited form as: J Okla State Med Assoc. 2015 November ; 108(11): 450–454.
Oklahoma “Tobacco Stops with Me” Media Campaign Effects on Attitudes toward Secondhand Smoke Ashley White, MPH, The University of Oklahoma Health Sciences Center, College of Public Health
Author Manuscript
Cati G Brown-Johnson, PhD, Stanford School of Medicine, Stanford Prevention Research Center Sydney Martinez, MPH, The University of Oklahoma Health Sciences Center, College of Public Health Sjonna Paulson, APR, and Oklahoma Tobacco Settlement Endowment Trust Laura A. Beebe, PhD The University of Oklahoma Health Sciences Center, College of Public Health
Abstract
Author Manuscript
Importance—Public education campaigns in tobacco control play an important role in changing tobacco-related knowledge, attitudes and behaviors. The Oklahoma Tobacco Stops with Me campaign has been effective in changing attitudes overall and across subpopulations towards secondhand smoke risks. Objective—Investigate campaign impact on secondhand smoke policy and risk attitudes. Design—Serial cross-sectional data analyzed with univariate and multivariable models. Setting—Random-digit dialing surveys conducted in 2007 and 2015 Participants—Oklahomans 18-65 years old Main Outcome(s) and Measure(s)—1) Support for smokefree bars; 2) risk assessment of secondhand smoke (very harmful, causes heart disease, causes sudden infant death); and 3) likelihood of protecting yourself from secondhand smoke
Author Manuscript
Results—With Tobacco Stops with Me exposure, from 2007 to 2015, Oklahomans demonstrated significant increases in: 1) supporting smokefree bars (23.7% to 55%); 2) reporting beliefs that SHS causes heart disease (58.5% to 72.6%), is very harmful (63.8% to 70.6%) and causes sudden infant death (24% to 34%); and 3) reporting they are very likely to ask someone not to smoke
Corresponding Author: Laura Beebe, PhD, The University of Oklahoma Health Sciences Center, 801 NE 13th St, Room 317, Oklahoma City, OK 73104, [email protected], Tel: (405) 271-2229 x48061, Fax: (405) 271-2068. Financial Disclosures: A. White and L. Beebe are funded by TSET to evaluate their marketing campaigns through a contract with the University of Oklahoma Health Sciences Center (OUHSC). S. Martinez is an evaluator in the Biostatistics, Epidemiology, and Research Design core at the OUHSC. C. Brown-Johnson consults with TSET. S. Paulson is currently TSET's Director of Health Communication. Conflicts of Interest: Authors have no other conflicts of interest to declare.
White et al.
Page 2
Author Manuscript
nearby (45% to 52%). Controlling for demographics, smokers and males showed reduced attitude change. In uncontrolled comparisons, high-school graduates faired better than non-diploma individuals, who lacked significant attitude changes. Conclusions and Relevance—Tobacco Stops with Me achieved its mission to more closely align public perception of SHS with well-documented secondhand smoke risks. Efforts to target women were particularly successful. Smokers may be resistant to messaging; closing taglines that reinstate individual choice may help to reduce resistance/reactance (e.g., adding Oklahoma Helpline contact information).
Introduction
Author Manuscript
Educational mass media campaigns have been shown to be an effective way to change attitudes and beliefs around smoking.1,2 However, research is limited on the effects of educational campaigns on attitudes about secondhand smoke (SHS) and support for smokefree policies in the United States. One study conducted in 2012 in Mobile County, Alabama evaluated the effects of a local antismoking mass media campaign on support of smokefree environments.3 Results of that study showed a positive association between exposure to the media message and increased support for smokefree protections.3 Our goal with the current study was to determine whether an Oklahoma-specific antismoking mass media campaign, Tobacco Stops With Me (TSWM), would impact knowledge and attitudes about SHS, and support for smokefree policies, particularly smokefree bars.
Author Manuscript
Oklahoma has one of the highest tobacco use rates in the nation.4 In addition to high tobacco use, Oklahoma's preemption clause in the state's Smoking in Public Places Act continues to limit communities' abilities to protect citizens from secondhand smoke despite longstanding calls for its repeal.5 Furthermore, Oklahoma is one of only 19 states that allow smoking in all bars and nightclubs.6 In 2008, as part of an ongoing commitment to reduce tobacco use and alleviate the harmful health effects of SHS, the Oklahoma Tobacco Settlement Endowment Trust (TSET) in partnership with the Oklahoma State Department of Health initiated a multi-phase health communications campaign to highlight how tobacco negatively impacts individuals whether they use tobacco or not. The campaign, developed around the tagline, “Tobacco Stops With Me,” sought to raise awareness about the dangers of tobacco and SHS.7
Author Manuscript
This study of TSWM was designed to assess the effect of exposure to the campaign on support for smokefree policy and knowledge about the dangers of SHS. Our secondary goal was to determine any differences in the effect of exposure to the campaign among subpopulations.
Methods The baseline survey, conducted in November and December of 2007 prior to the launch of any TSWM messaging, included 4,001 Oklahomans. Follow-up longitudinal surveys (data not analyzed or reported herein) were conducted in 2008 and 2009, and results of that longitudinal study highlighted the successful reach of the campaign.7 In preparation for the
J Okla State Med Assoc. Author manuscript; available in PMC 2016 November 01.
White et al.
Page 3
Author Manuscript
longitudinal follow-up, smokers were oversampled at baseline to ensure enough smokers over time. Conducted in collaboration with Westat, a survey research firm located in Rockville, Maryland, and the University of Oklahoma Health Sciences Center (OUHSC), the baseline survey was a dual-frame cross-sectional sample with both landline telephone and cellular telephone numbers. Eligibility criteria included Oklahoma residency, English speaking, age 18-54 years, and verbal consent. Questions regarding tobacco use, knowledge of SHS dangers, and support for policy were taken from previously validated surveys, including the Behavioral Risk Factor Surveillance System and the Adult Tobacco Survey.
Author Manuscript
For this study, the second cross-sectional survey (n=1,001 Oklahomans) largely mirrored the first and was conducted between May and June 2015 by the OUHSC Sooner Survey Center in collaboration with the OUHSC evaluation team. Eligibility criteria were the same as baseline except that the upper age of the respondents was increased to 65 due to changing media campaign targets. Tobacco Stops with Me Campaign
Author Manuscript
Tobacco Stops with Me is a best-practice based multi-media, multi-phase, anti-tobacco campaign developed by the Oklahoma Tobacco Settlement Endowment Trust and the Oklahoma State Department of Health. The campaign, launched in February 2008, was designed to help reduce tobacco use and the adverse health effects of secondhand smoke. Many of the advertisements, which appeared at intervals throughout subsequent years, featured family relationships and portrayed situations showing that the negative effects of tobacco may not be realized for many years. The ads sought to raise awareness of each person's responsibility for making “tobacco stop with me” and helping communities be tobacco-free. The campaign message was intended for all Oklahomans with special focus on young adults, families, and pregnant women. These groups were targeted with specific, culturally relevant messaging through a multiplatform approach. Media outlets included paid media (television, radio, print, online, social media, sponsorships and outdoor advertising, etc.) and earned media (news releases, on-air interviews, articles, personal testimonials, social media interactions, etc.). The campaign was also represented by a dynamic website at StopsWithMe.com and a social media presence on Facebook, Twitter and YouTube. Measures
Author Manuscript
Previous evaluation of TSWM showed that by 2011, the vast majority of Oklahomans (81%) were aware of the media campaign.7 With the exception of those with less than a high school education, message awareness had permeated to all subpopulations.7 The reach and frequency of the media messaging has remained steady since this point of saturation. Because of this high level of awareness and consistent media coverage, campaign exposure was assumed to be present in 2015 for the follow-up cross-sectional survey, but absent at baseline in 2007. Demographic characteristics were used to define subpopulations of interest in the descriptive analyses. These measures included age, gender, race/ethnicity, and level of education. Respondents were asked questions in both surveys to gage support for smokefree bar policies and knowledge and attitudes about SHS. We assessed support for smokefree bar
J Okla State Med Assoc. Author manuscript; available in PMC 2016 November 01.
White et al.
Page 4
Author Manuscript
policies by asking participants if they thought smoking should be allowed in bars (2007 response options: in both indoor and outdoor areas, in outdoor areas only, not at all; 2015 response options: strongly agree, agree, disagree, strongly disagree). Knowledge about the dangers of SHS was assessed using three questions based on the CDC's Adult Tobacco Survey. Participants were asked: 1) to rate the harmfulness of SHS (very harmful to one's health, somewhat harmful…, not very harmful…, not at all harmful…); 2) if they believed that breathing SHS causes Sudden Infant Death Syndrome (SIDS) (yes, no) and heart disease (yes, no); and 3) how likely participants would be to ask someone not to smoke around them (very likely, somewhat likely, not too likely, not at all likely). Dichotomous categorical variables were created for questions about support for smokefree policies in bars and harmfulness of SHS, with responses such as “don't know” combined with “no” responses, “agree” responses combined with “strongly agree” responses, and “disagree” responses combined with “strongly disagree” responses.
Author Manuscript
Statistical methods Post-stratification weights were calculated along dimensions of age, gender, race, income, and education using American Community Survey data to adjust for non-response and sources of undercoverage. Imputation was used for missing responses in raking dimension variables. We used SAS version 9.4 to conduct descriptive statistics and binary logistic regression. Survey procedures and sampling weights were used to obtain population-level estimates. We used Pearson chi square tests of significance to compare categorical survey responses between baseline and follow-up. In order to adjust for inflated error rates caused by multiple comparisons, comparisons with p≤0.01 are considered significant.
Author Manuscript
We calculated odds ratios and corresponding 95% confidence intervals using multivariable logistic regression to evaluate the effect of campaign exposure on outcome measures. Odds ratios were adjusted for age, gender, age, race/ethnicity, and level of education. Regression models were stratified by smoking status because of the previously reported differences that exist in knowledge and attitudes towards SHS and support for smokefree policies between smokers and non-smokers. We explored effect modification between gender and exposure to the campaign in each model. In the presence of significant interaction, we estimated genderspecific odds ratios. All methods were approved by the Institutional Review Board of the University of Oklahoma Health Sciences Center.
Results Author Manuscript
For the 2007 survey (n=4001), respondents self-identified as White (75.8%), American Indian (11.0%), African American (5.5%), Hispanic (4.8%), and other races (2.8%). Smokers were oversampled and represented 38.3% of the respondents. The sample was 58.7% female; 13.3% ages 18 to 24, 23.3% ages 25 to 34, 25.4% ages 35 to 44, and 38.0% ages 45 to 54. Most respondents had at least a high school education: 8.8% had less than high school, 29.6% a high school or equivalent degree, 31.5% some college, and 30.1% a college degree.
J Okla State Med Assoc. Author manuscript; available in PMC 2016 November 01.
White et al.
Page 5
Author Manuscript
The 2015 sample (n=1001) was made up of 73.2% of respondents who self-identified as White, 9.8% as American Indian, 8.0% as African American, 6.6% as Hispanic, and 2.5% as another race. Smokers made up 18.1% of the sample population. Most respondents were female (54.2%); 12.0% were ages 18 to 24, 19.4% ages 25 to 34, 17.9% ages 35 to 44, 21.6% ages 45 to 54, and 29.2% 55 to 65. Most respondents had at least a high school education: 6.1% had less than high school, 25.0% a high school or equivalent degree, 31.6% some college, and 37.3% a college degree.
Author Manuscript
At the baseline survey, prior to any exposure to the TSWM campaign, only 23.7% of Oklahomans supported smokefree policies in bars (Table 1). Support for smokefree bars increased significantly post-campaign to 55.0% overall (p
J Okla State Med Assoc. Author manuscript; available in PMC 2016 November 01. Published in final edited form as: J Okla State Med Assoc. 2015 November ; 108(11): 450–454.
Oklahoma “Tobacco Stops with Me” Media Campaign Effects on Attitudes toward Secondhand Smoke Ashley White, MPH, The University of Oklahoma Health Sciences Center, College of Public Health
Author Manuscript
Cati G Brown-Johnson, PhD, Stanford School of Medicine, Stanford Prevention Research Center Sydney Martinez, MPH, The University of Oklahoma Health Sciences Center, College of Public Health Sjonna Paulson, APR, and Oklahoma Tobacco Settlement Endowment Trust Laura A. Beebe, PhD The University of Oklahoma Health Sciences Center, College of Public Health
Abstract
Author Manuscript
Importance—Public education campaigns in tobacco control play an important role in changing tobacco-related knowledge, attitudes and behaviors. The Oklahoma Tobacco Stops with Me campaign has been effective in changing attitudes overall and across subpopulations towards secondhand smoke risks. Objective—Investigate campaign impact on secondhand smoke policy and risk attitudes. Design—Serial cross-sectional data analyzed with univariate and multivariable models. Setting—Random-digit dialing surveys conducted in 2007 and 2015 Participants—Oklahomans 18-65 years old Main Outcome(s) and Measure(s)—1) Support for smokefree bars; 2) risk assessment of secondhand smoke (very harmful, causes heart disease, causes sudden infant death); and 3) likelihood of protecting yourself from secondhand smoke
Author Manuscript
Results—With Tobacco Stops with Me exposure, from 2007 to 2015, Oklahomans demonstrated significant increases in: 1) supporting smokefree bars (23.7% to 55%); 2) reporting beliefs that SHS causes heart disease (58.5% to 72.6%), is very harmful (63.8% to 70.6%) and causes sudden infant death (24% to 34%); and 3) reporting they are very likely to ask someone not to smoke
Corresponding Author: Laura Beebe, PhD, The University of Oklahoma Health Sciences Center, 801 NE 13th St, Room 317, Oklahoma City, OK 73104, [email protected], Tel: (405) 271-2229 x48061, Fax: (405) 271-2068. Financial Disclosures: A. White and L. Beebe are funded by TSET to evaluate their marketing campaigns through a contract with the University of Oklahoma Health Sciences Center (OUHSC). S. Martinez is an evaluator in the Biostatistics, Epidemiology, and Research Design core at the OUHSC. C. Brown-Johnson consults with TSET. S. Paulson is currently TSET's Director of Health Communication. Conflicts of Interest: Authors have no other conflicts of interest to declare.
White et al.
Page 2
Author Manuscript
nearby (45% to 52%). Controlling for demographics, smokers and males showed reduced attitude change. In uncontrolled comparisons, high-school graduates faired better than non-diploma individuals, who lacked significant attitude changes. Conclusions and Relevance—Tobacco Stops with Me achieved its mission to more closely align public perception of SHS with well-documented secondhand smoke risks. Efforts to target women were particularly successful. Smokers may be resistant to messaging; closing taglines that reinstate individual choice may help to reduce resistance/reactance (e.g., adding Oklahoma Helpline contact information).
Introduction
Author Manuscript
Educational mass media campaigns have been shown to be an effective way to change attitudes and beliefs around smoking.1,2 However, research is limited on the effects of educational campaigns on attitudes about secondhand smoke (SHS) and support for smokefree policies in the United States. One study conducted in 2012 in Mobile County, Alabama evaluated the effects of a local antismoking mass media campaign on support of smokefree environments.3 Results of that study showed a positive association between exposure to the media message and increased support for smokefree protections.3 Our goal with the current study was to determine whether an Oklahoma-specific antismoking mass media campaign, Tobacco Stops With Me (TSWM), would impact knowledge and attitudes about SHS, and support for smokefree policies, particularly smokefree bars.
Author Manuscript
Oklahoma has one of the highest tobacco use rates in the nation.4 In addition to high tobacco use, Oklahoma's preemption clause in the state's Smoking in Public Places Act continues to limit communities' abilities to protect citizens from secondhand smoke despite longstanding calls for its repeal.5 Furthermore, Oklahoma is one of only 19 states that allow smoking in all bars and nightclubs.6 In 2008, as part of an ongoing commitment to reduce tobacco use and alleviate the harmful health effects of SHS, the Oklahoma Tobacco Settlement Endowment Trust (TSET) in partnership with the Oklahoma State Department of Health initiated a multi-phase health communications campaign to highlight how tobacco negatively impacts individuals whether they use tobacco or not. The campaign, developed around the tagline, “Tobacco Stops With Me,” sought to raise awareness about the dangers of tobacco and SHS.7
Author Manuscript
This study of TSWM was designed to assess the effect of exposure to the campaign on support for smokefree policy and knowledge about the dangers of SHS. Our secondary goal was to determine any differences in the effect of exposure to the campaign among subpopulations.
Methods The baseline survey, conducted in November and December of 2007 prior to the launch of any TSWM messaging, included 4,001 Oklahomans. Follow-up longitudinal surveys (data not analyzed or reported herein) were conducted in 2008 and 2009, and results of that longitudinal study highlighted the successful reach of the campaign.7 In preparation for the
J Okla State Med Assoc. Author manuscript; available in PMC 2016 November 01.
White et al.
Page 3
Author Manuscript
longitudinal follow-up, smokers were oversampled at baseline to ensure enough smokers over time. Conducted in collaboration with Westat, a survey research firm located in Rockville, Maryland, and the University of Oklahoma Health Sciences Center (OUHSC), the baseline survey was a dual-frame cross-sectional sample with both landline telephone and cellular telephone numbers. Eligibility criteria included Oklahoma residency, English speaking, age 18-54 years, and verbal consent. Questions regarding tobacco use, knowledge of SHS dangers, and support for policy were taken from previously validated surveys, including the Behavioral Risk Factor Surveillance System and the Adult Tobacco Survey.
Author Manuscript
For this study, the second cross-sectional survey (n=1,001 Oklahomans) largely mirrored the first and was conducted between May and June 2015 by the OUHSC Sooner Survey Center in collaboration with the OUHSC evaluation team. Eligibility criteria were the same as baseline except that the upper age of the respondents was increased to 65 due to changing media campaign targets. Tobacco Stops with Me Campaign
Author Manuscript
Tobacco Stops with Me is a best-practice based multi-media, multi-phase, anti-tobacco campaign developed by the Oklahoma Tobacco Settlement Endowment Trust and the Oklahoma State Department of Health. The campaign, launched in February 2008, was designed to help reduce tobacco use and the adverse health effects of secondhand smoke. Many of the advertisements, which appeared at intervals throughout subsequent years, featured family relationships and portrayed situations showing that the negative effects of tobacco may not be realized for many years. The ads sought to raise awareness of each person's responsibility for making “tobacco stop with me” and helping communities be tobacco-free. The campaign message was intended for all Oklahomans with special focus on young adults, families, and pregnant women. These groups were targeted with specific, culturally relevant messaging through a multiplatform approach. Media outlets included paid media (television, radio, print, online, social media, sponsorships and outdoor advertising, etc.) and earned media (news releases, on-air interviews, articles, personal testimonials, social media interactions, etc.). The campaign was also represented by a dynamic website at StopsWithMe.com and a social media presence on Facebook, Twitter and YouTube. Measures
Author Manuscript
Previous evaluation of TSWM showed that by 2011, the vast majority of Oklahomans (81%) were aware of the media campaign.7 With the exception of those with less than a high school education, message awareness had permeated to all subpopulations.7 The reach and frequency of the media messaging has remained steady since this point of saturation. Because of this high level of awareness and consistent media coverage, campaign exposure was assumed to be present in 2015 for the follow-up cross-sectional survey, but absent at baseline in 2007. Demographic characteristics were used to define subpopulations of interest in the descriptive analyses. These measures included age, gender, race/ethnicity, and level of education. Respondents were asked questions in both surveys to gage support for smokefree bar policies and knowledge and attitudes about SHS. We assessed support for smokefree bar
J Okla State Med Assoc. Author manuscript; available in PMC 2016 November 01.
White et al.
Page 4
Author Manuscript
policies by asking participants if they thought smoking should be allowed in bars (2007 response options: in both indoor and outdoor areas, in outdoor areas only, not at all; 2015 response options: strongly agree, agree, disagree, strongly disagree). Knowledge about the dangers of SHS was assessed using three questions based on the CDC's Adult Tobacco Survey. Participants were asked: 1) to rate the harmfulness of SHS (very harmful to one's health, somewhat harmful…, not very harmful…, not at all harmful…); 2) if they believed that breathing SHS causes Sudden Infant Death Syndrome (SIDS) (yes, no) and heart disease (yes, no); and 3) how likely participants would be to ask someone not to smoke around them (very likely, somewhat likely, not too likely, not at all likely). Dichotomous categorical variables were created for questions about support for smokefree policies in bars and harmfulness of SHS, with responses such as “don't know” combined with “no” responses, “agree” responses combined with “strongly agree” responses, and “disagree” responses combined with “strongly disagree” responses.
Author Manuscript
Statistical methods Post-stratification weights were calculated along dimensions of age, gender, race, income, and education using American Community Survey data to adjust for non-response and sources of undercoverage. Imputation was used for missing responses in raking dimension variables. We used SAS version 9.4 to conduct descriptive statistics and binary logistic regression. Survey procedures and sampling weights were used to obtain population-level estimates. We used Pearson chi square tests of significance to compare categorical survey responses between baseline and follow-up. In order to adjust for inflated error rates caused by multiple comparisons, comparisons with p≤0.01 are considered significant.
Author Manuscript
We calculated odds ratios and corresponding 95% confidence intervals using multivariable logistic regression to evaluate the effect of campaign exposure on outcome measures. Odds ratios were adjusted for age, gender, age, race/ethnicity, and level of education. Regression models were stratified by smoking status because of the previously reported differences that exist in knowledge and attitudes towards SHS and support for smokefree policies between smokers and non-smokers. We explored effect modification between gender and exposure to the campaign in each model. In the presence of significant interaction, we estimated genderspecific odds ratios. All methods were approved by the Institutional Review Board of the University of Oklahoma Health Sciences Center.
Results Author Manuscript
For the 2007 survey (n=4001), respondents self-identified as White (75.8%), American Indian (11.0%), African American (5.5%), Hispanic (4.8%), and other races (2.8%). Smokers were oversampled and represented 38.3% of the respondents. The sample was 58.7% female; 13.3% ages 18 to 24, 23.3% ages 25 to 34, 25.4% ages 35 to 44, and 38.0% ages 45 to 54. Most respondents had at least a high school education: 8.8% had less than high school, 29.6% a high school or equivalent degree, 31.5% some college, and 30.1% a college degree.
J Okla State Med Assoc. Author manuscript; available in PMC 2016 November 01.
White et al.
Page 5
Author Manuscript
The 2015 sample (n=1001) was made up of 73.2% of respondents who self-identified as White, 9.8% as American Indian, 8.0% as African American, 6.6% as Hispanic, and 2.5% as another race. Smokers made up 18.1% of the sample population. Most respondents were female (54.2%); 12.0% were ages 18 to 24, 19.4% ages 25 to 34, 17.9% ages 35 to 44, 21.6% ages 45 to 54, and 29.2% 55 to 65. Most respondents had at least a high school education: 6.1% had less than high school, 25.0% a high school or equivalent degree, 31.6% some college, and 37.3% a college degree.
Author Manuscript
At the baseline survey, prior to any exposure to the TSWM campaign, only 23.7% of Oklahomans supported smokefree policies in bars (Table 1). Support for smokefree bars increased significantly post-campaign to 55.0% overall (p
