Randomized Controlled Trial Testing a Video-Text Tobacco Cessation Intervention among Economically Disadvantaged African American Adults

Abstract

Objective:

This pilot study tested the acceptability and short-term outcomes of a culturally specific mobile health (mHealth) intervention (Path2Quit) in a sample of economically disadvantaged African American adults. We hypothesized that Path2Quit would demonstrate greater acceptability, biochemically verified abstinence, and promote nicotine replacement therapy (NRT) use compared with a standard text-messaging program.

Method:

In a 2-arm pilot randomized trial, adults who sought to quit smoking (N=119) received either Path2Quit or the National Cancer Institute’s (NCI) SmokefreeTXT, both combined with a brief behavioral counseling session plus two weeks of NRT. Outcomes included acceptability (intervention evaluation and use), NRT utilization, 24-hour quit attempts, self-reported 7-day point prevalence abstinence (ppa), and biochemically-verified smoking abstinence at the 6-week follow-up.

Results:

Participants were 52% female/48% male, mostly single (60%), completed ≥ 12 years of education (83%), middle-aged, and 63% reported a household income < $10k/year. Participants smoked 11 (SD=8.2) cigarettes/day for 25 (SD=16) years, and reported low nicotine dependence. There were no differences in intervention evaluations or use (ps>.05), yet Path2Quit led to significantly greater NRT utilization at follow-up (p<.05). There was no difference in quit attempts between conditions or 7-day ppa (p>.05). However, Path2Quit resulted in significantly greater carbon monoxide confirmed ppa (AOR=3.55; CI: 1.32–9.54) at the 6-week follow-up.

Conclusions:

A culturally specific mHealth intervention demonstrated positive effects on NRT use and short-term abstinence. Additional research in a larger sample, and with longer-term follow-up is warranted.

Evidence of racial disparities in tobacco cessation (Kulak, Cornelius, Fong, & Giovino, 2016) and tobacco-associated diseases, including lung cancer, cerebrovascular disease, and cardiovascular disease, is robust and longstanding (National Cancer Institute, 2017; American Cancer Society, 2019; USDHHS, 1998). Multilevel factors contribute to greater difficulty achieving tobacco cessation among African American compared with White individuals, such as targeted tobacco marketing for menthol brands, less provider advice to quit, elevated stress, and less access to evidence-based interventions (Institute, 2017). However, African American adults report greater motivation to quit and more past-year quit attempts compared with White adults (Babb, Malarcher, Schauer, Asman, & Jamal, 2017; Kumar et al., 2016). Needed are theory-based interventions to address the distinct smoking patterns and cessation needs of African American individuals, with the potential for population-level reach. This study evaluated a mobile health (mhealth) tobacco cessation intervention developed specifically for African American adults.

Culturally specific tobacco cessation interventions are an evidence-based approach to increase engagement in efforts to become tobacco-free. Models of culturally specific intervention development discuss the importance of framing within an ethno-cultural context, incorporating theoretically and empirically-derived behavioral patterns, beliefs, norms and values, psychosocial stressors, and language(s) into the messaging and presentation (Bayer, 1994; Kreuter, Lukwago, Bucholtz, Clark, & Sanders-Thompson, 2003; Kreuter & McClure, 2004; Resnicow, Baranowski, Ahluwalia, & Braithwaite, 1999; Resnicow, Soler, Braithwaite, Ahluwalia, & Butler, 2000). African American adults who smoke report a preference for culturally specific interventions (Webb, 2009), and such approaches have demonstrated efficacy when paired with nicotine patch therapy (C. S. Ahluwalia, McNagny, & Clark, 1998), bupropion (J. S. Ahluwalia, Harris, Catley, Okuyemi, & Mayo, 2002), and in comparison with standard cognitive behavioral therapy (CBT) (Webb Hooper, Antoni, Okuyemi, Dietz, & Resnicow, 2016). There is a significant need to increase the reach of culturally specific tobacco interventions by delivery in acceptable and scalable formats.

The Promise of Mobile Health (mHealth) Tobacco Interventions among African American Adults

The majority (96%) of U.S. adults own a mobile phone (Center, 2019). Mobile technologies, specifically mobile phones, are used in healthcare settings, and for health information seeking (Free, Phillips, Galli, et al., 2013; Free, Phillips, Watson, et al., 2013). Compared with in-person interventions, mHealth offers wide reach potential and addresses key social determinants of health that create barriers to in-person care, such as numerous clinic visits, scheduling limitations, insurance, high cost, transportation, and employment and caregiving responsibilities (Robyn Whittaker et al., 2012). These concerns are even more prohibitive among underserved populations.

Text messaging interventions have been applied to the management of conditions such as diabetes, hypertension, asthma, and HIV treatment (Cole-Lewis & Kershaw, 210), and have demonstrated positive effects on smoking cessation (Free et al., 2011; Free et al., 2009; Hall, Cole-Lewis, & Bernhardt, 2015; Robyn Whittaker et al., 2012; R. Whittaker, McRobbie, Bullen, Rodgers, & Gu, 2016). Meta-analytic findings support the efficacy of text messaging interventions for smoking cessation relative to control conditions for both 7-day point prevalence abstinence and continuous abstinence (Scott-Sheldon et al., 2016; R. Whittaker et al., 2019). One trial found that a text program doubled the odds of quitting compared to usual care (Abroms, Boal, Simmens, Mendel, & Windsor, 2014).

Applying an equity lens is important for mHealth tobacco interventions, and there is a relative dearth of research focused on underserved groups. In a review focused on disadvantaged populations, Boland et al. (2018) found promising effects of text messaging support for smoking cessation among indigenous, psychiatric, and substance use disorder inpatients. Vidrine et al. (2019) compared abstinence rates among low-income individuals who smoke who received NRT, NRT plus text messaging, or combined NRT, text messaging, and a telephone counseling call. At the 6-month follow-up, they found biochemically confirmed abstinence rates of 12%, 12%, and 25.5%, respectively. In 2011, the NCI released SmokefreeTXT on smokefree.gov, which is a fully automated, two-way (push and pull) text messaging program. Messages sent from the program are drawn from an established library of behavior change strategies (Michie, Hyder, Walia, & West, 2011). However, African American adults enrolled in SmokefreeTXT reported less engagement and lower smoking abstinence compared with White enrollees (Robinson et al., 2019). This suggests opportunities for improvement in text messaging programs among African American adults, and that culturally specific approaches may add value in this domain.

The Current Study

The purpose of current study was to pilot test a culturally specific text messaging tobacco cessation intervention in a sample of economically disadvantaged African American adults. The intervention was a translation of a culturally specific intervention, Pathways to Freedom: Leading the Way to a Smoke Free Community© (PTF), into a video-text format. Applying frameworks guiding the development of culturally specific interventions (Bayer, 1994; Kreuter et al., 2003; Kreuter & McClure, 2004; Resnicow et al., 1999; Resnicow et al., 2000), PTF is a 60-minute documentary-style program that includes both surface (e.g., race-matched host, experts, and families; music, colors, images) and deep structure elements (e.g., content focused on menthol cigarettes and alternative tobacco products, the health and financial costs of smoking, tobacco cessation pharmacotherapy, psychosocial stressors, and culturally relevant coping strategies to manage smoking urges) (Bayer, 1994; Kreuter et al., 2003; Kreuter & McClure, 2004; Resnicow et al., 1999; Resnicow et al., 2000). Findings from an analogue randomized study supported the efficacy of PTF delivered on a Digital Video Disk (DVD) as a stand-alone intervention among non-treatment seeking, low-income individuals (Webb Hooper, Baker, & Robinson, 2014). Participants viewed the PTF video or a control video in the laboratory and were provided with a take-home copy. Compared to the control group, those in the PTF condition reported greater risk perceptions, motivation to quit, and were almost three times more likely to report smoking abstinence.

The rationale for developing a video-text intervention, Path2Quit, was two-fold. First, survey data from treatment-seeking participants in our tobacco research program indicated a strong preference for a text-messaging program versus computer-based or mobile phone apps. Second, an analysis of enrollees in five state quitlines demonstrated that African American adults were less likely to enroll in or utilize a web-only cessation program compared to White adults (Webb Hooper, Carpenter, & Salmon, 2019). Moreover, a video-text approach can be utilized on most mobile devices, reduces health literacy concerns, and provides visuals and graphic imagery. The aims of this pilot study were to examine the feasibility, acceptability, and initial efficacy of a culturally specific video-text intervention (Path2Quit). It was hypothesized that compared with SmokefreeTXT, Path2Quit would demonstrate greater acceptability and biochemically verified point prevalence abstinence (ppa). Secondary outcomes included intervention engagement, NRT use at the end-of-intervention, 24-hour quit attempts, and 7-day ppa.

Method

Study Design

This study was a 2-arm semi-pragmatic randomized controlled trial (ClinicalTrials.gov Identifier: NCT03524482). Low-income adults who reported current tobacco smoking were randomly assigned to receive (1) the newly translated Path2Quit video text-messaging program or (2) SmokefreeTXT (standard control). In both conditions, participants received one behavioral counseling session plus a starter supply (i.e., two weeks) of NRT. Sample size estimates for this pilot study were based on beta-testing of Path2Quit and the PTF development study (Webb Hooper et al., 2014). Assuming a medium effect size, d, of .55, alpha of .05, and two-tailed testing, a sample size of 106 (55 per condition) would yield a power of .80.

Participants

The study was conducted in an urban mid-sized Midwestern U.S. city. Participants were adults recruited from the community via flyers, radio commercials, and internet advertisements. Path2Quit is an intervention intended to reach and apply (in general) to African American audiences. It is not based on socioeconomics, but rather on race and culture. However, we partnered with the local public housing authority to promote the inclusion of economically disadvantaged individuals, and hired a community navigator (history of smoking and respected community resident; co-author M.B.) to assist with recruitment. Per the design of pragmatic randomized trials, we limited the number of inclusion criteria. Eligible participants selfi-dentified as African American, were age ≥ 18 years, reported low household income (based on federal standards that consider household size), smoked at least one cigarette/day or had a CO reading of at least 5 ppm, had a mobile phone with a data plan, and consistent with SmokefreeTXT, were willing to set a quit date within the next 14 days. Individuals with contraindications for NRT were included, but did not receive medication [only one participant reported a contraindication (pregnancy)]. Ineligible respondents were enrolled in the publicly available smokefreeTXT (if they had a mobile phone) or referred to the state quitline.

Figure 1 illustrates the flow of participants through the project. A total of 183 individuals were screened for eligibility, and 33 (18%) were ineligible. Of those who were eligible and attended the baseline visit, 119 individuals provided written informed consent and were randomly assigned to the Path2Quit (n = 61) or SmokefreeTXT (n = 58) condition. Of these, 104 (87%) completed the end-of intervention assessment, with no difference between conditions (p > .05).

Figure 1.

Participant Flow Chart

Measures

Sociodemographics.

Participants self-reported race and ethnicity (Hispanic [yes or no]), sex, household income, and completed education.

Smoking History.

Participants reported cigarettes per day and nicotine dependence using the Fagerström Test of Nicotine Dependence (Heatherton, Kozlowski, Frecker, & Fagerstrom, 1991) (current sample alpha = .76).

Intervention Acceptability and Use.

Using an 8-point Likert scale, ranging from “strongly agree” to “strongly disagree,” participants rated the degree to which the intervention was trustworthy, informative, and contained new information (17 items; current sample alpha = .95). Items included, “Overall, how satisfied are you with the text messaging program,” “The information in the text messages were trustworthy,” and “The messages in the text messaging program encouraged me to try to quit smoking.” Using a Likert scale ranging from “strongly agree” to “strongly disagree,” participants also indicated the degree to which they used the program (3 items; current sample alpha = .80).

NRT Utilization.

Participants received a starter supply (2-weeks) of nicotine patches or gum (based on preference), which is the equivalent of 14 days. Self-reported NRT utilization was defined as the number of days the starter supply of NRT was used, ranging from 0–14 days. Participants were asked to bring their box of NRT to their follow-up visit, and any remaining amounts were counted.

Smoking Behavior.

Quit attempts were defined as an intentional effort to abstain from smoking for at least 24-hours (yes or no). The FDA approved coVita, Micro+ ™ pro Smokerlyzer® was used to obtain exhaled breath carbon monoxide (CO) readings at the end-of-intervention (< 5 parts per million (ppm) = nonsmoker; 5ppm or greater = smoking) (Perkins, Karelitz, & Jao, 2013). Participants self-reported 7-day ppa (any smoking, even one puff during the previous seven days), which was coded as yes or no.

Intervention Conditions

Participants were randomly assigned (and blinded) to one of the following conditions.

SmokefreeTXT (standard text message intervention control).

SmokefreeTXT is a 6–8 week fully automated text-based cessation intervention (Augustson et al., 2017). The program is free to U.S. subscribers who have a mobile phone, and is available on smokefree.gov or by texting “QUIT” to short code 47848. At sign-up, participants report demographics, smoking frequency (“every day,” “most days,” “some days”, or “less than that”), and state of residence. The program continues for 6-weeks starting on the selected quit date, and includes motivational messages, quitting strategies (smoking facts, tips for quitting, managing urges, and ideas for smoke-free activities), and assessment questions (mood, cravings, and smoking status). The message library contains 166 messages and 41 behavior change techniques (Stoyneva, Pugatch, Sanders, Coa, & Cole-Lewis, 2016). Keywords (MOOD, CRAVE, or SLIP) are used to receive a relevant message from the system. With informed consent, we were able to access the data for study participants.

Path2Quit (culturally specific text message intervention).

The 75-minute PTF video program was translated into a short message service (SMS) intervention. The research team developed the programing requirements for the mobile intervention, including the order and timing of videos to send, the names of the keywords, and the assessment items. We beta-tested the usability of the system among 15 users, and applied their feedback on video message content, usability, timing, frequency, keywords, duration, and assessments to refine and finalize the Path2Quit program.

Path2Quit is a 6-week program with a comparable structure to SmokefreeTXT, yet delivers texts with links to PTF video segments. We developed an automated software system (mobile app) to send (i.e., “push”) video text messages and assessment items to all registered users at prescribed times, and to “pull” videos from a message library containing 85 (30–120 second) videos. The topic of the video (e.g., Never Quit Quitting) accompanied the text link. Messages were pushed to participants up to two times/day, with access to video messages pulled from keywords (HELP1, JONES, SLIP). The program also delivered “inquiry” questions, which pushed additional videos (e.g., the history of tobacco and African Americans, menthol smoking, the dangers of secondhand smoke) if participants indicated interest in the topic. The intervention includes cognitive behavioral strategies for tobacco cessation and relapse prevention, such as education about nicotine dependence and withdrawal, managing quit attempts, and coping skills training. Cultural adaptations were infused throughout the video segments and included, educational messages about tobacco use among African Americans, health consequences, benefits of quitting, managing withdrawal symptoms (e.g., urges to smoke), testimonials from African Americans who formerly smoked, pharmacotherapy concerns, menthol cigarettes and alternative tobacco products, environmental tobacco smoke, unique stressors such as racism and discrimination, race-specific weight issues and concerns, co-morbid addiction, neighborhood and environmental influences, and working as a community against the tobacco industry. Messages also incorporated empowerment themes, cultural assets, and the importance of social support.

Nicotine Replacement Therapy (NRT).

Participants in both conditions were offered the choice of a starter pack (i.e., two weeks supply) of nicotine patch therapy or nicotine gum. Participants received instructions on how to use the NRT, and information on the possible side effects. Participants were encouraged to start use upon a quit attempt.

Procedures

The study was approved by the Case Western Reserve University Institutional Review Board and the Case Comprehensive Cancer Center’s Protocol Review and Monitoring Committee. Respondents were screened via telephone or in person, and eligible participants selected a quit date and were scheduled for a baseline assessment and behavioral support session. The project locations were our community-accessible research center and a central public housing authority community center, and we provided support to offset transportation costs. Following informed consent and the baseline assessment, participants were randomly assigned to a condition (1:1 ratio and stratified in blocks of 50). Participants in both conditions received a single behavioral support session, delivered by trained masters or bachelors-level interventionists who were supervised by the Principal Investigator, and a starter supply (2-weeks) of nicotine patches or gum based on their preference (unless there were medical contraindications). The behavioral support session covered the benefits of cessation, the process of smoking cessation, understanding triggers, coping responses, garnering social support, the safety, efficacy, and importance of using FDA-approved cessation aids during quit attempts, home smoking restrictions, and maintaining motivation. Participants received verbal and written instructions on how to use the text-programs and research staff assisted them with enrollment in either SmokefreeTXT or Path2Quit. Follow-up assessments occurred at the end-of-intervention (i.e., at 6-weeks). Participants received $20 at baseline and $30 at 6-weeks for assessment completion.

Statistical Analyses

Analyses included descriptive statistics to characterize the sample’s sociodemographics, smoking history, and follow-up completion by intervention condition. The smoking-related outcomes were coded as binary (24-hour quit attempt, yes or no; 1 = abstinence, 0 = smoking). Smoking-related variables were analyzed using chi-square tests to compare quit attempts, CO-verified ppa, and 7-day ppa by condition. Next, hierarchical logistic regressions entering age as a covariate in step 1 (as it is a prognostic indicator and differed significantly between conditions at baseline) and intervention condition in step 2 were conducted to obtain odds ratios and 95% confidence intervals for each outcome. Analyses of covariance, adjusting for age, tested the effect of condition on intervention acceptability, engagement, and NRT utilization. There were no repeated measures. Approximately 13% of participants had missing data, which was managed for all outcome variables using multiple imputation. Alpha was set at 0.05 and analyses were conducted using SPSS version 26 (IBM).

Results

Sample Characteristics

Table 1 presents the characteristics of participants by condition. Participants were mostly female (52%), single (60%), and reported ≥ 12 years of education (83%). The sample was primarily low-income, and 63% reported an annual household income of < $10,000. Participants smoked an average of 11 (SD = 8) cigarettes per day (cpd) for 25 (SD = 16) years. The mean FTND score was 4.5 (SD = 1), indicating low nicotine dependence. Participants in the Path2Quit condition were significantly younger than those in the SmokefreeTXT condition [t = 4.05, df = 112, p < .001].

Table 1.

Sample Characteristics by Race/Ethnicity

	Study Condition
Characteristic	Path2Quit (n = 61)	SmokefreeTXT (n = 58)
Sex (Female)	57%	48%
Age (range = 25–73)	49 (12)	58 (10)
Marital Status (Single)	67%	51%
Completed ≥ 12 Years
Education	83%	78%
Income < $10K/Year	57%	69%
Cigarettes/Day (range = 1–40)	11 (9)	12 (8)
Years Smoking (range = 1-65)	25 (16)	25 (16)
FTND Total (range = 0–10)	4 (1)	4 (2)

Note: FTND = Fagerström Test of Nicotine Dependence

Intervention Acceptability and Use

As shown in Table 2, participants evaluated both Path2Quit and SmokefreeTXT as highly acceptable (e.g., strong agreement that both programs were trustworthy, informative, and contained new information), with no significant difference after adjustment for age. Controlling for age, there was also no significant difference in participants’ use of the two interventions.

Table 2.

Primary Outcomes by Study Condition (N = 119)

	Study Condition
	Path2Quit	SmokefreeTXT	p
Outcome Variable
Intervention Ratings (range = 0–119)	94.85 (27.36)	93.06 (35.65)	0.27
Use of Text-Messaging Program (range = 0–21)	16.05 (5.14)	14.20 (5.56)	0.13
Days of Nicotine Replacement Therapy Use (range = 0–14)	9.21 (5.50)	6.00 (5.57)	0.02

NRT Utilization

Sixty-four percent of participants chose nicotine patch therapy and 36% selected nicotine gum. There were no differences in NRT selection between conditions (p > .05). ANCOVAs controlling for age, tested days of NRT use at the end-of-intervention by condition. As shown in Table 2, there was a significant difference in NRT utilization. Specifically, participants in the Path2Quit condition reported a greater duration of NRT use compared with SmokefreeTXT, F(1, 97) = 5.55, p = .02.

Smoking Behavior

Most participants (88%) reported making a serious smoking quit attempt over the previous six weeks. The proportion of quit attempts was greater in the Path2Quit versus SmokefreeTXT condition, yet the difference in quit attempts between conditions was not statistically significant in bivariate (p = .14; Table 3) or adjusted analyses (p = .21; Table 4). At the end-of-intervention, 38% of the sample was biochemically confirmed as abstinent. As shown in Table 3, CO-confirmed ppa was significantly greater in the Path2Quit condition compared with SmokefreeTXT, Χ² (1, N = 119) = 5.32, p = .02. The effect of intervention remained significant after adjusting for age, such that the odds of biochemically-confirmed point prevalence abstinence were 3.55 times greater in the Path2Quit condition compared with SmokefreeTXT (p = .01; Table 4). Finally, 31% of the sample reported 7-day ppa at the end-ofintervention. Path2Quit resulted in a 30% increase in 7-day ppa compared with SmokefreeTXT, but the odds of quitting for a least one week were not significantly different between groups (p =.21; Table 4).

Table 3.

Intent-to-Treat Bivariate Analyses of Quit Attempts and Smoking Abstinence at the 6-Week Follow-Up (N = 119)

	Condition
	Path2Quit	SmokefreeTXT	p	Chi-Square Test
24-hour Quit Attempt (Yes)	91%	84%	0.12	X2(df= 1)= 2.40
Biochemically-Verified PPA	48%	26%	0.02	X2 (df= 1)= 5.32
7-day PPA	35%	27%	0.14	X2 (df= 1)= 2.20

Note: PPA = point prevalence abstinence

Table 4.

Logistic Regressions of Intervention Condition on Smoking Abstinence at Follow-Up (N = 119)

	Quit Attempt		7-Day PPA		CO-Verified PPA
	AOR	(95% CI)	AOR	(95% CI)	AOR	(95% CI)
Logistic Regression Modelsa
Age	1.04	(0.98–1.11)	1.03	(0.98–1.07)	1.02	(0.97–1.06)
Intervention
Condition
SmokefreeTXT	Reference		Reference		Reference
Path2Quit	3.02	(0.53–7.37)	1.97	(0.66–5.75)	3.55	(1.32–9.54)

Note: ppa=point prevalence abstinence, AOR= Adjusted Odds Ratio, CI= Confidence Interval; Models examined odds of quit attempts or abstinence; OR of less than 1.0 indicates decreased odds of abstinence

Discussion

There are few data available on the effects of mHealth for tobacco cessation among African American adults (and none focused on low-income individuals). Indeed, the NCI only began collecting race/ethnicity data among SmokefreeTXT enrollees in August 2017. This pilot research offers an important contribution to the body of knowledge, demonstrating the feasibility, acceptability, and initial efficacy of delivering a culturally specific video text messaging intervention and a standard text program among individuals who wish to quit smoking. Path2Quit and SmokefreeTXT were both rated as high quality and informative. Participants also indicated engagement with the interventions and their contents, with no difference between groups. As expected, Path2Quit demonstrated positive effects on behavior change compared with SmokefreeTXT. Specifically, Path2Quit resulted in a greater frequency of NRT utilization. The majority of participants attempted to quit smoking over the 6-week intervention, and biochemically confirmed ppa was significantly greater in the Path2Quit condition. Self-reported 7-day ppa was also greater in the Path2Quit condition, although this difference was not significant. Taken together, this study highlights the promise of a culturally specific text messaging tobacco intervention designed for African American adults.

The delivery of both Path2Quit and SmokefreeTXT was feasible and acceptable. This is comparable to previous research suggesting that text messaging tobacco cessation programs can be administered effectively (Abroms et al., 2014; Augustson et al., 2017). In the current study, both interventions were evaluated as high quality, credible, informative, and trustworthy. Participants also reported sharing the information with others, with no difference between groups. This is in contrast to Webb Hooper et al. (2014), who found greater satisfaction and content evaluations of the PTF video compared to a standard tobacco cessation video. This may be attributable to the difference in delivery format (i.e., full 60-minute video versus segmented messages delivered via mobile phones over a 6-week period). Nevertheless, the strong acceptability ratings in the current study serves to add to the evidence supporting the use of PTF in either format.

Participants in the Path2Quit condition also reported more days of NRT use compared with SmokefreeTXT. Most mHealth tobacco programs are stand-alone interventions, yet we provided a starter supply of NRT – consistent with quitline offerings. Addressing pharmacotherapy hesitancy was a culturally specific component of Path2Quit, given previous indications that African American adults are less likely to utilize tobacco cessation pharmacotherapy, and expressed concerns about becoming addicted (Hooper, Payne, & Parkinson, 2017). In contrast, SmokefreeTXT does not cover this topic. However, the content of the behavioral support session received by participants in both conditions covered the importance of NRT during cessation attempts, evidence for its safety and efficacy, and provided opportunities for participants to ask questions. Moreover, 100% of participants accepted the offer of NRT during the support session (excluding the one person with a known contraindication). The high motivation of the sample coupled with the provision of NRT may explain the greater abstinence rate compared to previous research [e.g., (Abroms et al., 2014)]. Overall, we observed differential benefits of Path2Quit for promoting NRT uptake, which is important for cessation success.

Path2Quit demonstrated significantly greater biochemically-verified ppa compared with SmokefreeTXT, and a clinically meaningful increase in 7-day ppa at the end-of-intervention. The current study extends findings from meta-analyses (R. Whittaker et al., 2019), providing evidence in support of text-messaging interventions for tobacco cessation – and in a sample of adults with multiple levels of social disadvantage. The overall end-of-intervention abstinence rate was greater than previous studies testing stand-alone text-messaging interventions. Possibilities for this difference include the reporting of short versus long-term abstinence (Vidrine et al., 2019; R. Whittaker et al., 2019), and the video versus text-only approach. Future studies are needed to examine long-term abstinence.

Understanding the feasibility intervention delivery and acceptability was a central goal of this initial study. National surveys indicate that almost everyone has internet access and a mobile phone, yet, our deep experience working with and serving this population suggested that the digital divide persists and that even among individuals who own mobile phones, digital inequities exist. With the proliferation of technology-based tobacco interventions, it is imperative that all populations have access and benefit equitably, or disparities in cessation and downstream health outcomes may increase. Given findings that African American adults who wish to quit smoking may be less interested in technology-based tobacco interventions (Webb Hooper et al., 2019) and are less likely to quit using SmokefreeTXT (Robinson et al., 2019), Path2Quit was designed as a targeted video-text approach that can reduce health literacy concerns and provide visuals and graphic images to increase impact. However, there were key feasibility challenges that point to digital inequity, which should be considered in future research. First, a notable minority of participants owned text-capable phones, but needed to call their phone carrier to activate texting (and were unaware of this beforehand). Second, many participants required significant assistance texting the short code needed to enroll in either text messaging program, and with texting keywords to pull messages. Third, 98% of participants had Android phones, and over 80% had no-contract (“pay as you go”) plans from smaller telephone carriers. Finally, of ineligible respondents (17% of screened), 41% did not have stable Internet access. These factors have important implications for engagement and outcomes, and may affect economically disadvantaged individuals broadly. It should be noted, however, that digital inequity was not a part of the study’s outcome measures. We collected this information, a priori, to contextualize the findings and to inform future digital intervention projects seeking to reach underserved groups. Thus, in-person enrollment in mHealth interventions may be an important consideration when possible, and even when the phones and data plans are provided to participants.

This study has strengths and limitations. This is the first study to develop and test in an RCT a video-text mHealth intervention that integrates ethno-cultural specificity, and accounts for tobacco-related differences between the general population and groups with multiple levels of social disadvantage (i.e., low-income and racial minority). We conducted a head-to-head test with SmokefreeTXT, a publicly available and evidence-based program that has been examined among African American enrollees. In addition, this pilot trial of Path2Quit is the first to address key gaps in the field, specifically health literacy and community competence (e.g., preferred methods of message/information delivery in the target population). The primary limitations include the small sample size, the single geographic location, limited assessment battery, and the short-term follow-up. The use of video versus written text messages could also be conceived as a limitation. However, this format was a key aspect of community competence, and based on our formative research and longstanding research on intervention preferences in this population (Gollop, 1997; Matthews, Sellergren, Manfredi, & Williams, 2002). Given the preliminary positive effects, a fully-powered RCT s warranted.

In conclusion, this study supports the feasibility and acceptability of text messaging tobacco cessation programs among economically disadvantaged African American adults. Findings from this preliminary investigation signal clinically meaningful and positive effects of Path2Quit relative to SmokefreeTXT on NRT utilization and short-term abstinence. There is a clear need to enhance the short-term intervention effects by providing additional treatment (behavioral support and pharmacotherapy) and to evaluate intervention effects over an extended follow-up duration. Findings from this initial study are suggestive of broader contextual considerations. First, culturally specific interventions may be an important health equity-promoting approach, by providing resources designed to fit the needs of populations that experience health disparities; and second, Path2Quit has the potential to eliminate social determinants of health that present challenges to tobacco cessation. Specifically, it is evidencebased, highly accessible (provides on-demand access), low cost, does not require transportation, attends to health literacy levels, and includes content that addresses specific, community-level influences such as targeted tobacco marketing. Moreover, Path2Quit is a population-based intervention that can fit within multiple tobacco control contexts. Future research should incorporate methods to address digital inequity and test Path2Quit as an adjunct to other interventions. Future work should also conduct mediational analyses to understand explanatory factors underlying intervention effects. This intervention has clear potential for scalability, dissemination, and reducing tobacco-related health disparities.

Public Health Significance:

This study provides preliminary evidence that a culturally specific tobacco intervention delivered via video text messages can increase utilization of nicotine replacement therapy and promote short-term tobacco cessation among economically disadvantaged African American adults.