This Free Life Campaign: Increasing Intention to Quit Among LGBTQ+ Young Adult Nondaily Smokers in Minneapolis

Abstract

PURPOSE

LGBTQ+ young adults smoke at disproportionately higher rates than their non-LGBTQ+ counterparts, but prevention efforts are limited. Furthermore, prior to This Free Life (TFL), no known campaigns target LGBTQ+ nondaily smokers. In this study Blue Cross and Blue Shield of Minnesota evaluated a local partnership extension of the Food and Drug Administration (FDA) Center for Tobacco Products’ TFL campaign.

METHODS

The intervention featured a variety of LGBTQ+-tailored events, social/digital media, and out-of-home media placed in locations with a high density of LGBTQ+ young adults. Cross-sectional surveys (n = 1215) were collected from LGBTQ+ young adult (18-26) nondaily smokers at 4 time points between 2016 and 2019. The national TFL campaign was tracked in a separate evaluation conducted by the FDA.

RESULTS

43.0% of nondaily LGBTQ+ smokers reported awareness of the campaign (n = 522), and 63.4% of those also engaged with TFL (n = 330). Engagement was highest for gay, lesbian and transgender participants, and for Asian and Black/African American participants. Each additional instance of campaign engagement increased participants’ odds of intending to quit smoking by 20% (AOR = 1.20; 95% CI = 1.05, 1.36). The relationship between campaign engagement and intention to quit was fully mediated by the campaign’s impact on attitudes against smoking and perceived normative trends, but not by perceived behavioral control.

CONCLUSION

The local extension of TFL increased intentions to quit for LGBTQ+ nondaily smokers. Future research should further explore the relationship between perceived behavioral control and intentions to quit for nondaily smokers.

Introduction

Although rates of daily smoking have decreased in recent years, rates of nondaily smoking have significantly increased, particularly among young adults.^1-3 Additionally, research over the past 10 years highlights disparities in tobacco use among certain populations, and the prevalence of both daily and nondaily smoking is disproportionately high among LGBTQ+ young adults compared to non-LGBTQ+ young adults.^4-6

Researchers have proposed several factors that may contribute to elevated rates of tobacco use among LGBTQ+ young adults, including targeted marketing to the LGBTQ+ community by the tobacco industry and the use of smoking as a coping mechanism for feelings of stress associated with coming out, experiences of discrimination, and victimization.^7-10 While there are an increasing number of smoking interventions focused on adults within the LGBTQ+ community, recent reviews of smoking interventions have revealed a dearth of interventions focused on LGBTQ+ youth and young adults.^11-13 This is despite evidence that LGBTQ+ individuals report that tailored interventions would be more acceptable, salient and motivating than non-tailored interventions.^14,15

Prior to the TFL campaign, no existing interventions targeted nondaily smoking within the LGBTQ+ community, defined as smoking on some days in the past month but not every day. Nondaily smoking includes ‘social smoking,’ a term referencing the pairing of nondaily smoking with social situations.¹⁶ Long-term nondaily smokers have unique motivations that influence their smoking patterns and make them distinct from daily smokers.^17-19 Many uphold the belief that nondaily smoking is low risk, despite findings that there are substantial risks associated with this behavior.^20-24 Often nondaily smokers do not identify as ‘smokers’ or consider themselves addicted, and believe they can easily stop.^16,20,25 As such, there is an important opportunity to tailor interventions to address the distinct motivations and behavioral patterns of LGBTQ+ nondaily smokers.

Description of This Free Life

This Free Life (TFL) is a Social Branding® intervention developed for LGBTQ+ young adult, nondaily smokers with funding from the U.S. Food and Drug Administration (FDA) Center for Tobacco Products (CTP). At the time of campaign launch, Minneapolis had a number of established community supports for the LGBTQ+ community, including Rainbow Health, Outfront Minnesota, Gender Justice, and the Aliveness Project, among others. However, there were no existing targeted intervention services that sought to address nondaily smoking for the LGBTQ+ community. FDA launched TFL in 2016 with paid digital, print, and out-of-home media focused primarily in 12 designated market areas. With the goal of increasing local campaign engagement, Blue Cross and Blue Shield of Minnesota funded a local campaign extension that supplemented the nationwide campaign with local on-the-ground efforts in Minneapolis. This included events at LGBTQ+ bars, local influencers, brand ambassadors, and a locally-focused evaluation to assess these efforts (Table 1, Figure 1).

Table 1.

Key elements of a social branding intervention and relevant metrics for a local extension of “This Free Life” 2016-2019.

Social Branding Key Element	Minneapolis “This Free Life” Intervention
Social Brand	The TFL brand reflects the desire of LGBTQ+ young adults to be “free” in their lives and personal experiences, and connects this driver to the importance of being free from tobacco. The tagline for the campaign was “Freedom to Be, Tobacco-Free”.
Branded Events	Sixty-four TFL events took place at LGBTQ+ bars, clubs, and events in the Minneapolis designated market area. All venue spaces, inside and out, were tobacco-free for the event duration, and often required that all venue staff wear TFL apparel. All available screens at the venue were often leveraged to promote TFL messages. Events were attended by local influencers eager to promote tobacco-free living, including DJs and drag queens. The total estimated LGBTQ+ young adult attendance at bar/club-based events was 13 450, with an additional 38 250 LGBTQ+ young adults reached at the Twin Cities Pride Festival.
Brand Ambassadors	Brand Ambassadors were tobacco-free LGBTQ+ influencers in the Minneapolis community recruited to promote the brand while normalizing tobacco dialogue. Brand ambassadors attended monthly messaging trainings and endorse the campaign’s overarching message of living free from tobacco both at live events as well as through their personal social media channels. By September of 2019, there were over 55 brand ambassadors.
Relevant Messages	Messages included facts about tobacco’s impact on the LGBTQ+ community, elaborations on how tobacco’s addiction conflicts with personal freedom, and the message that the health consequences from tobacco make life more difficult.
Website and Social Media	Digital campaigns used social media and the TFL website to disseminate tobacco messaging, encourage online conversations and promote local events. From 2017 through September 2019, for Facebook and Instagram, there were a total of 14,149 877 impressions; 4 197 459 video views; 2141 post shares; 1595 post comments; and 103 403 post likes/reactions stemming from the Minneapolis DMA. The website received 148 390 visits; 8558 video views; and was shared 354 times. Digital campaigns began using Twitter in May 2018. From May 2018 through September 2019 this resulted in a total of 4 419 203 impressions; 1 904 548 video views; 108 post retweets; 116 post comments; and 2192 post likes/reactions.
Social Rewards	After completing a 4 question quiz about tobacco’s impact on the LGBTQ+ community, event attendees had the chance to win branded fashionable gear (eg t-shirts, sunglasses). Brand ambassadors gave individuals verbal and nonverbal approval for supporting a tobacco-free lifestyle. These rewards can accelerate the behavior change process.

Figure 1.

Sample This Free Life advertisement and event image.

Aligned with the Theory of Planned Behavior,²⁶ shifting attitudes about tobacco was a focal point of TFL, as LGBTQ+ young adults express permissive attitudes towards smoking that may facilitate use.⁹ The intervention leveraged LGBTQ+ young adults’ shared desire to be “free” in their lives and drew a connection between that value and the importance of being free from smoking, using the tagline, “Freedom to Be, Tobacco-Free.”

Shifting social norms, a second tenet of the Theory of Planned Behavior,²⁶ was also a central focus of the TFL campaign. It is widely recognized that normative perceptions of smoking are perpetuated through LGBTQ+ social venues where indoor and outdoor smoking is more common,^27,28 and perceived as an acceptable ‘social lubricant’ that increases connection within the LGBTQ+ community.^9,10 These locations can serve a particularly important role to young LGBTQ+ people given the difficulty finding other safe spaces to explore and establish the social aspect of their sexual and gender identity.^29,30 In these settings, TFL offered appealing tobacco-free social events and provided opportunities for attendees to interact with local brand ambassadors, which increased the visibility of LGBTQ+ peers who choose to live tobacco-free.

Attitudes and norms were the primary focus of the TFL strategy, while perceived behavioral control, a construct similar to self-efficacy in the Theory of Planned Behavior, was not a central focus. Given that the applicability of perceived behavioral control may depend on the population and behavior,³¹ the current study sought to better understand the role of this construct for LGBTQ+ nondaily smokers. Additionally, exploring the relationship between these constructs and intentions to quit is important because intention to quit smoking is considered the first step before quitting smoking behavior, and is clearly correlated with the rate of successful smoking cessation.^26,32

The purpose of this study was to evaluate awareness and engagement with a local extension of TFL, an intervention targeting LGBTQ+ young adult nondaily smokers, in the Minneapolis designated market area, along with evaluating how constructs of campaign engagement are associated with intentions to quit smoking. To our knowledge, TFL was the first intervention targeting LGBTQ+ young adult nondaily smokers. We also hypothesized (1) increased engagement with TFL would predict greater intention to quit smoking and (2) the relationship between campaign engagement and intention to quit smoking would be mediated by attitudes and norms. As an exploratory component, we sought to assess the relationship between perceived behavioral control and intention to quit smoking.

Methods

Participants and procedures

A convenience sample of 1215 LGBTQ+ nondaily smokers in the Minneapolis area completed in-person (n = 443, 36.5%) and online (n = 772, 63.5%) surveys. Wave 1 data collection was from February to May 2016 (22.3%), Wave 2 from June to November 2017 (26.5%), Wave 3 from June to November 2018 (29.0%) and Wave 4 from May to July 2019 (22.2%). Campaign events and digital presence were ongoing during this time. Participants were recruited in-person from LGBTQ+ bars/clubs, and online through targeted social media advertisements, effective recruitment methods for LGBTQ+ young adults.³³ Data collection occurred at the same place as recruitment. Those recruited through social media completed the survey online, while those recruited in-person completed a paper survey.

Venues for in-person data collection were randomly selected from LGBTQ+ bars/clubs, and bars/clubs with LGBTQ+ themed nights, in the Minneapolis area. Data collection did not coincide with any TFL events, and was never conducted at a venue where TFL even had occurred in the past month. Data collectors approached young adults in the venues and invited them to complete a screener survey in exchange for a $5 cash incentive. Online participants were recruited via Facebook (2016-2019) and Instagram (2017-2019) advertisements with ad-targeting features such as age-, geo-, and interest-based targeting applied. Respondents answered online screening questions to ensure eligibility. Eligible online participants were provided a $10 gift card for participation. Inclusion criteria for all participants included reported ages 18-26, self-identifying as LGBTQ+, residing in the Minneapolis designated market area, and providing informed consent. The analytic sample was restricted to nondaily smokers. The study protocol was classified as exempt by Chesapeake Institutional Review Board.

Measures

Sociodemographic variables

LGBTQ+ identity was determined based on the response to 3 separate survey items. Participants were asked “What is your current gender identity?” (female, male, transmale/transman, transfemale/transwoman, genderqueer/gender nonconforming, a different identity (please specify), prefer not to answer), “What sex were you assigned at birth, on your original birth certificate?” (female, male, prefer not to answer), and “Which of the following best represents how you think of yourself?” (lesbian, gay, straight, bisexual, queer, pansexual, trisexual, or omnisexual, asexual, something else (please specify), prefer not to answer). Answers were recoded into 5 discreet categories, “lesbian,” “gay”, “bisexual”, “transgender” or “gender nonconforming”. Date of birth was used to calculate a continuous measure of age. Racial/ethnic categories included non-Hispanic White, Asian, Black/African American, Hispanic/Latino, or other/multiracial. Due to small sample sizes, participants who identified as American Indian or Alaska Native, Native Hawaiian or other Pacific Islander, multiracial or other were combined into a single other/multiracial category.

Campaign engagement

As a specific goal of the Blue Cross and Blue Shield local extension effort in Minneapolis was to increase local engagement, participants were asked to report any type of engagement with TFL, including website, social media, and local events. These questions were, “How many times have you visited This Free Life’s website?”, “How many times have you visited This Free Life’s Facebook page?”, and “How many times have you been to a This Free Life event?”. Response options included: (0) “Never,” (1) “Once,” (2) “2 times,” (3) “3 times,” or (4) “4 or more times.” Responses from the 3 items were summed and all summations greater than 4 were classified as “4 or more times” resulting in a final 5-point item capturing the number of engagements across all 3 mediums, with categories of “None,” “1,” “2,” “3,” or “4 or more.”

Perceived normative trends

Perceived normative trends for LGBTQ+ young adult smoking was measured using a mean score of 2 questions, each on a 3-point scale: “Thinking about LGBTQ+ people your age, do you think tobacco use is…” (0) increasing, (1) staying the same, or (2) decreasing, and the question, “Compared to 1 year ago, people your age at LGBT bars, clubs, and events are smoking…” (0) more often, (1) about the same, (2) less often. Unlike traditional descriptive norms, which describe only the current prevalence of a behavior, trending norms have been shown to be an important category of norms that guide beliefs and behaviors, in some cases even more strongly than descriptive norms.³⁴

Attitudes against smoking

Attitudes against smoking were measured using a mean score of 3 items: “Tobacco’s addiction conflicts with the freedom to be yourself”, “Reducing tobacco use in the LGBTQ+ community is a priority”, and “No matter how difficult life is, tobacco’s health consequences make life more difficult”.^35,36 Responses were on a 5-point Likert scale from (1) strongly disagree to (5) strongly agree.

Perceived behavioral control

Perceived behavioral control was measured with a single item shown to reliably assess self-efficacy in the context of tobacco use,^37-39 “If you decided to give up smoking altogether in the next 12 months, how likely do you think you would be to succeed?” Response options were (1) very unlikely, (2) somewhat unlikely, (3) somewhat likely, (4) very likely.

Intention to quit

Intention to quit smoking was measured by the question, “I plan to stop smoking cigarettes for good within the next…”, and participants that selected the response options “7 days,” “30 days,” “6 months,” or “12 months” were classified as “intends to quit cigarette use in the next 12 months.”^40,41 Participants could also select “I do not plan to stop smoking cigarettes within the next year”.

Data analysis

All analyses were performed using SPSS Statistics for Mac.⁴² Campaign awareness and engagement rates were explored. Logistic regression explored the association between TFL campaign engagement and intention to quit smoking, controlling for demographics and survey year. A parallel multiple mediation model then examined the role of theoretical mediators as mechanisms to explain the potential impact of TFL campaign engagement on intention to quit.^43,44 Analyses were conducted using the PROCESS macro version 3.3 with a percentile bootstrap estimation approach generating 10 000 samples.⁴⁵ Diagnostic procedures were conducted to ensure that no covariates were highly correlated, and that the collinearity statistics were within acceptable limits (r < .28, Tolerance > .63, VIF < 1.6).^46,47

Results

Participant profile

Participants were on average 22.0 years old (SD = 2.4). Similar to demographics of Minneapolis, participants reported their race/ethnicity as non-Hispanic white (58.9%), Hispanic/Latino (10.7%), Black/African American (8.2%), Asian (6.6%), or other/multiracial (15.6%). Participants self-reported their sexual orientation and gender identity as lesbian (19.3%), gay (23.5%), bisexual (20.2%), transgender (13.0%), or gender nonconforming (24.0%). Additionally, the majority of participants reported alcohol use (86.7%) and marijuana use (69.6%) in the past 30 days. See Table 2.

Table 2.

Sample characteristics of survey participants (n = 1215).

Characteristics	Total Sample
Age, M (SD)	22.0 (2.4)
Average cigarettes smoked in past 30 days, M (SD)	10.0 (8.4)
LGBT Identity, %
Lesbian	19.3%
Gay	23.5%
Bisexual	20.2%
Transgender	13.0%
Gender nonconforming	24.0%
Sex Assigned at Birth, %
Female	61.1%
Male	34.7%
Prefer not to answer	4.2%
Race/Ethnicity, %
Non-Hispanic White	58.9%
Hispanic/Latinx	10.7%
Black/African American	8.2%
Asian	6.6%
Other/multiracial	15.6%
Survey Wave, %
Wave 1	22.3%
Wave 2	26.5%
Wave 3	29.0%
Wave 4	22.2%
Campaign Aware %	43.0%
Campaign Engagements among Aware, %
None	36.6%
One	12.5%
Two	12.2%
Three	8.1%
Four or more	30.6%

Campaign awareness and engagement

Among LGBTQ+ nondaily smokers, 43.0% reported awareness of the campaign, with no significant change in awareness across waves of study enrollment. There were no significant differences in awareness by LGBTQ+ identity. Of those who were aware, 63.3% of the sample had engaged with the campaign: 40.5% had attended a TFL event, 39.5% had visited the TFL website, and 53.1% had visited the TFL Facebook page. Of those aware of the campaign, 71.8% of lesbian, 74.7% of gay, 56.9% of bisexual, 60.0% of transgender, and 52.6% of gender nonconforming participants engaged with the campaign (P < .01). Similarly, of those aware of the campaign, 53.0% of non-Hispanic white, 64.7% of Hispanic/Latinx, 82.9% of Black/African American, 80.6% of Asian and 82.4% of other/multiracial participants engaged with the campaign (P < .001).

Association of campaign engagement with intention to quit smoking

Overall, 70.3% of participants intended to quit smoking in the next 12 months. The final logistic model fit the data well, as indicated by the Hosmer-Lemeshow test (χ²(8) = 8.00, P = .43). Each instance of engagement with the campaign increased participants’ odds of intending to quit smoking by 20% (AOR =1.20; 95% CI = 1.05, 1.36; P < .01; Table 3).

Table 3.

Regression analyses examining attitudes against tobacco, normative trends, and perceived behavioral control among LGBTQ+ nondaily smokers (n = 1031).

	Exp(B) Predicting Intention to Quit [95% CI]	B Predicting Attitudes Against Tobacco [95% CI]	B Predicting Normative Tobacco Trends [95% CI]	B Predicting Perceived Behavioral Control [95% CI]
Campaign Engagement	1.20 [1.05, 1.36]**	.14 [.10, .18]***	.06 [.03, .08]***	−.01 [−.05, .04]

Note. Reference group; *P < .05, **P < .01, ***P < .001.

Associations between campaign engagement and theoretical mediators

Mediation analysis found that increased engagement with the TFL campaign associated with intention to quit smoking through attitudes against tobacco and perceived normative trends. In Figure 2, the a-paths indicate the association between the independent variable (campaign engagement) and the mediators. Individuals who reported increased engagement with the TFL campaign endorsed stronger attitudes against tobacco (a1 = .14; 95% CI [.10, .18]; P < .001) (Figure 2, Table 3). The model including demographics, year, and campaign engagement significantly predicted attitudes against tobacco (R = .29, R2 = .08, F (11,1019) = 8.57, P < .001).

Figure 2.

Mediation model depicting path coefficients between TFL campaign engagement, theoretical mediators and intention to quit, with age, LGBT identity, year, and race/ethnicity entered as covariates (n = 1031). The indirect effect of TFL engagement on intention to quit through attitudes (a1b1 = .09) and norms (a2b2 = .02), but not through perceived behavioral control (a3b3 = .00) was significant. The total effect (c) from TFL engagement to intention to quit and the direct effect (c’) are described. *P < .05; **P < .01; ***P < .001

Greater campaign engagement was also associated with perceptions that tobacco use was decreasing within the LGBTQ+ community (a2 = .06; 95% CI [.02, .08]; P < .001, Table 3). The model including demographics, year, and campaign engagement significantly predicted perceived LGBTQ+ normative trends (R = .23, R2 = .05, F (11,1019) = 5.07, P < .001).

However, increased campaign engagement was not associated with increased perceived behavioral control (a3 = −.01; 95% CI [−.05, .04]; p = ns). While campaign engagement was not associated with participants’ beliefs that they would be successful if they decided to give up smoking, the model including demographics and year, significantly predicted perceived behavioral control (R = .27, R2 = .07, F (11,1019) = 7.43, P < .001).

Associations between theoretical mediators and intention to quit smoking

In Figure 2, the b-paths indicate the association between the mediators and intention to quit smoking. As hypothesized, stronger attitudes against tobacco predicted intention to quit smoking (b1 = .45; 95% CI [.27, .64]; P < .001), as did the perception that tobacco use was on the decline in the LGBTQ+ community (b2 = .39; 95% CI [.15, .63]; P < .01). We did not find evidence that increased perceived behavioral control was associated with intention to quit; rather, there was a non-significant trend in the opposite direction (b3 = −.15; 95% CI [−.31, .02]; P = .08). Unstandardized coefficients (log-odds) are reported in Table 4.

Table 4.

Logistic regression analysis predicting intention to quit using tobacco with theoretical mediators among LGBT nondaily smokers (n = 1031).

Independent Variables & Covariates	Log Odds for Intention to Quit [95% CI]
Attitudes against smoking	.46 [.27, .64]***
Perceived norms	.39 [.15, .64]**
Perceived behavioral control	−.15 [−.31, .02]
Age	.01 [−.05, .06]
LGBTQ+ identity
Gaya	—
Lesbian	.27 [−.20, .75]
Bisexual	−.37 [−.81, .06]
Transgender	−.16 [−.65, .33]
Gender nonconforming	−.53 [−.95, −.12]*
Race/Ethnicity
Non-Hispanic whitea	—
Black/African American	1.13 [.49, 1.77]***
Asian	.68 [.03, 1.32]*
Hispanic/Latino	.24 [−.23, .69]
Other/multiracial	.97 [.51, 1.42]***
Survey Year	.08 [−.05, .22]
Campaign Engagement	.09 [−.05, .22]

^aReference group; *P < .05, **P < .01, ***P < .001.

Indirect and direct paths in mediation analysis

The ab paths, or the specific indirect effects represent the portion of the effect of TFL campaign engagement on intention to quit smoking that is attributable to changes in attitudes, norms, and perceived behavioral control (Figure 2). The indirect effect of TFL engagement on intention to quit was attributable to changes in attitudes (ab1 = .07 [.04, .10]) and norms (ab2 = .02 [.01, .04]), but not to changes in perceived behavioral control (ab3 = .00 [−.01, .01]. Independent of its effect on attitudes and norms, awareness of the TFL campaign alone did not significantly impact intention to quit (c’ = .09, P = .19), supporting the full mediation hypothesis through attitudes and norms.

Discussion

The current study was designed to evaluate engagement with the intervention This Free Life among LGBTQ+ young adult nondaily smokers in Minneapolis. A substantial proportion of participants reported awareness of the campaign (43.0%) and the majority of those exposed also engaged with the campaign (63.4%). Additionally, increased campaign engagement was associated with increased intention to quit smoking, through attitudes and norms. This suggests that in Minneapolis, TFL is effectively reaching and engaging LGBTQ+ young adult nondaily smokers, a group that has been largely overlooked in smoking prevention and cessation efforts.^11,16,17

Campaign awareness was similar to other LGBTQ+ campaigns with no broadcast presence, however, this still leaves a majority of the target audience who was not reached by the campaign.⁴⁸ This is because TFL used highly targeted tactics, such as local events and digital messages, in order to drive engagement for LGTBQ+ individuals, rather than traditional television advertising, which is more efficient at driving awareness.^49,50 If it were a priority to increase awareness, that could be accomplished through traditional media, however this approach would reach a much broader group of people, including those outside of the narrow TFL campaign audience.⁵⁰ While the national TFL campaign focused on building awareness through an array of paid media, which it successfully did over time, the Minneapolis campaign extension focused on building community engagement without the financial spill of traditional media.

There is evidence that the local extension of TFL was also able to engage sexual, gender and racial minorities. Gay, lesbian and transgender participants were equally likely to engage with the campaign, all with high rates of engagement (74.7%, 71.8%, 60.6%). Given that LGBTQ+ tobacco campaigns have largely reached only gay males,¹² this suggests that in Minneapolis TFL successfully deployed a more inclusive campaign strategy. Bisexual and gender nonconforming individuals engaged with the campaign less frequently than gay males, suggesting these groups may further benefit from separate, targeted campaign components. Given limitations with availability of targeted media opportunities, this could include identifying additional local influencers that openly identify as bisexual and gender nonconforming. TFL also had increased engagement among Black/African American and Asian individuals in Minneapolis. This is encouraging given that compared to non-Hispanic whites, these racial/ethnic minorities are at increased risk for nondaily smoking.⁵¹

Although Theory of Planned Behavior would predict that perceived behavioral control would be associated with intention to quit, the current study did not find this relationship for nondaily smokers. This may be due to insufficient measurement of the construct within the instrument (a single item), or because the instrument measured self-efficacy, vs controllability.⁵² Additionally, increased perceived behavioral control may not increase quit intention among LGBTQ+ young adult nondaily smokers, particularly given the non-significant trend (P = .08) in the opposite direction. Nondaily smokers already report that they could quit smoking easily if they wanted to,^20,25 and may experience an “illusion of control” which could weaken the relationship between perceived behavioral control and intention.^53,54 It is possible that further increasing self-efficacy might make nondaily smokers feel even less vulnerable to the risks of addiction, reducing their need to stop and subsequently their intention to quit. Future research could further investigate the role that self-efficacy plays for nondaily smokers on intention to quit.

Overall, this study demonstrated an association between engagement with the TFL campaign, stronger attitudes against tobacco, and a shift in perceived norms, for an underserved and high-risk population in Minneapolis. These attitudes and perceived norms were then linked to intentions to quit smoking. According to the theoretical framework, we would expect these shifts in attitudes, norms and intention to ultimately result in behavioral shifts.²⁶ Given that smoking interventions are limited for LGBTQ+ nondaily smokers, this is a contribution to the tobacco-control landscape. TFL also represents a unique campaign that effectively sparked engagement across racial/ethnic groups within the LGBTQ+ community. The approaches used in this campaign offer important insights about how to reach marginalized communities that are frequently overlooked within the LGBTQ+ community.

Limitations

A limitation of the current study was the cross-sectional nature of the design. The current study also relied on a convenience sample obtained through social media and bar/club recruitment, which limits the generalizability of these findings. It is also worth nothing, that while gender identity and sexual orientation were collapsed into a single variable for analysis, these are not exclusive categories. It is possible that transgender and gender nonconforming participants could also identify as a sexual minority. Additionally, the study did not include data from a separate community where the campaign was not implemented as a control group, which would have strengthened the design. Finally, the current study measures intention to quit, a weaker measure than actual behavior.

Conclusion

This study evaluated a local extension of the campaign This Free Life, demonstrating that a targeted campaign approach may be associated with increased attitudes against tobacco and perceptions that tobacco use is less normative in LGBTQ+ young adult, nondaily smokers, an audience that has largely been overlooked in prevention and cessation interventions. The study further suggests certain sexual and gender minority groups within the larger LGBTQ+ community may benefit from separate, targeted campaign components. Additionally, this study contributes to the existing literature on LGBTQ+ young adult nondaily smokers, by using a theoretical foundation to highlight the unique ways that intention to quit smoking manifests in this audience, which could ultimately lead to behavior change.