Written Advice Given by African American Smokers to Their Peers: Qualitative Study of Motivational Messages

Introduction

Smoking is the leading cause of preventable death in the United States [1]. Smoking cessation programs have reduced overall smoking rates, but their impacts have not been uniformly beneficial [2,3]. Smoking-related disparities have noticeably expanded, resulting in calls for targeted cessation programs to meet the needs of minority and disadvantaged populations [3]. African Americans are disproportionately affected by tobacco-related diseases in terms of mortality and morbidity. African Americans are 20% more likely to die from heart disease than non-Hispanic Whites and have a higher incidence of lung cancer compared with the general population [4-7].

Although African Americans have the lowest rates of smoking onset, have slower progression to daily smoking [8], and report attempting to quit (63%) more often than non-Hispanic White smokers (53%) [9], they are less likely to achieve long-term cessation [10]. Tailored cessation interventions are needed to support African American smokers [11].

We conducted formative research to adapt a technology-assisted peer message intervention for African American smokers. The effectiveness of a peer-written message intervention has been demonstrated in a large randomized trial conducted among mostly non-Hispanic White smokers, who were recruited from medical practices across the United States [12,13]. In this study, participants were enrolled into a web-assisted tobacco intervention. The intervention group was emailed one peer and one expert (written by a team of behavioral scientists and clinicians) message weekly for 6 months. Compared with the control group, receiving a combination of peer- and expert-written messages significantly increased the odds of 7-day point prevalence cessation at 6 months (odds ratio [OR] 1.69, 95% CI 1.03-2.8) [12,13]. We further compared the effect of peer versus expert-written messages on intervention engagement, which was defined by return visits to the intervention website. Peer-written messages were more likely to generate a return visit to the intervention website (OR 2.03, 95% CI 1.74-2.35) than expert-written messages [14]. There were significant interactions of the effect of the peer-written message with time from registration (P<.001). As the time interval from registration increased, the benefits of peer-written messages were accentuated over time [14]. Preserving engagement in digital interventions is challenging yet critical to intervention fidelity and subsequent cessation. These results highlighted the benefit of peer message interventions [15]. Further details of our prior evaluation demonstrating the feasibility and effectiveness of a peer message intervention in a general population of smokers has been published [12-14].

Research shows that African American communities may be influenced by unique sociocultural dimensions, such as interconnectedness, level of health socialization, and ecological and health care system factors [16]. Messages written by health experts may not incorporate these dimensions. Thus, we hypothesize that African American smokers may particularly benefit from peer messages (messages written by African American smokers) as these messages could reflect shared experiences and exemplify difficulties of quitting, skills, and strategies needed for smoking cessation [17]. African American participants have reported greater engagement and improved outcomes when they felt a strong sense of community or support from social networks [18,19]. Prior studies that have tested a peer-driven digital intervention among African American participants have reported improved satisfaction and outcomes [17,18,20]. The peer messages developed in our prior work were written by a population of smokers (mainly non-Hispanic White smokers) [12,13]. To adapt the prior intervention, we asked African American adults, who were current or former smokers, to write messages to their peers guided by various hypothetical scenarios. In this paper, we focused on the qualitative assessment of the peer-written messages and report on the cessation strategies that emerged from the messages written by African American current and former smokers. In the future, these peer messages will be tested in an adapted and tailored technology-assisted behavioral intervention to promote cessation in African American smokers.

Methods

Study Design

We conducted a formative qualitative study of motivational messages written by African American current or former smokers. Writers (n=41) responded to various hypothetical scenarios (Multimedia Appendix 1) via an online semistructured questionnaire. Details on participant recruitment, data collection procedures, and analysis of peer-written messages are provided below. This protocol was approved by the University of Massachusetts Medical School Institutional Review Board.

Participant Recruitment

Participants were recruited online via ResearchMatch between April 2017 and November 2017. ResearchMatch is an online web-based recruitment database freely available to individuals interested in volunteering for research studies [19]. Researchers can search the database and filter the list of volunteers by demographic information, such as age, race, ethnicity, and tobacco use. Participants were eligible if they self-identified as African American, were current or former smokers, were 18 years or older, and were willing to write messages for peers (n=41). Participants received a US $50 Amazon gift card for their contribution in the study.

Data Collection

We collected data on sociodemographic factors and smoking characteristics. Sociodemographic factors included age, gender, education, and ability to pay for medical care. The ability to pay for medical care measure was used as a proxy for socioeconomic status, providing insights into household resources and demands placed on participants by medical expenditures [21,22]. Smoking characteristics included readiness to quit, number of cigarettes smoked per day, past year quit attempts, past 30-day use of menthol cigarettes, level of nicotine dependence, and electronic cigarette (or e-cigarette) use. The readiness to quit measure was based on stages of change as described in the transtheoretical model of behavior change [23], and was captured by asking current and former smokers to indicate whether they (1) were not thinking about quitting, (2) were thinking about quitting, (3) had already quit, (4) had set a quit date, or (5) had quit today.

Current and former smokers were matched to hypothetical scenarios by age group (45 years or younger and older than 45 years) and gender (male and female). We requested that participants write motivational messages about quitting based on each scenario. We suggested they describe experiences that might be useful to other smokers before, on, and after a set quit date. We also asked them to include suggestions for maintaining motivation and addressing barriers to quitting (eg, how to handle urges and cravings). Each participant responded to the following two hypothetical scenarios: for a smoker ready to quit and one not ready to quit. All responses to the scenarios were captured through a semistructured questionnaire that was distributed through Research Electronic Data Capture (REDCap) (Multimedia Appendix 1).

Data Analysis

The percentages and mean distributions of smoker characteristics were calculated. Informed by our previously published methods [14], we reviewed all messages to identify those with specific characteristics. First, the messages had to be directly quotable, include sufficient details, and be self-explanatory. Directly quotable messages with sufficient details were those that were written in sentence form and contained clear and detailed information targeted to the scenario. Second, the messages had to have behavioral content that reflected personal quitting experiences. The underlying assumption for this criterion is that personal experiences are considered credible and trustworthy and therefore can potentially motivate behavior change [24]. Messages were selected by research team members with relevant expertise in smoking cessation (CSN and RSS), health behavior change (CSN, AJM, JMF, RSS, and JHW), and qualitative analysis (JHW and AJM). Out of the 1200 messages written, 318 messages (27% of the original messages) met the criteria.

Examples of the messages selected are as follows:

She can speak with doctor and discuss methods of quitting, and what's available to help her. Try to reach out and see if there's a quit smoking program that either she can go to in person or she can call when she has urges.

Figure out what triggers her to smoke throughout the day have a plan. If you typically smoke after meals, you might want to have something planned after you eat to keep mind off it. Sometimes certain rooms of the house cause cravings if that's where you smoked. You might want to change the decor, try to reduce the smell of smoke, etc., to make the environment less triggering.

Examples of the messages excluded are as follows:

Secondhand smoking kills.

The cause and the effect.

Two coders (CSN and DM) reviewed selected messages in an open coding process until concepts became apparent, at which point a code was assigned. Labels and definitions of themes were based on words in the message text. This approach to developing sets of codes is a grounded theory approach to qualitative data analysis that limits researchers from erroneously “forcing” a preconceived result [25]. Data analysis was conducted using MAXQDA qualitative data analysis software developed and distributed by VERBI Software based in Berlin, Germany.

Results

Demographic and Smoking Characteristics

Among the participants, 60% (25/41) were female. Most of the participants had a college-level education (27% [11/41] were college graduates and 59% [34/41] had attended some college or technical school). The daily average number of cigarettes smoked was 8.5 (SD 5.7). More than half (23/41, 56%) of smokers were thinking about quitting, and 29% (12/41) had set a quit date. About half (21/41, 51%) of the participants reported a past year quit attempt. Most participants smoked within 30 minutes of waking up (22% [9/41] within 5 minutes and 51% [21/41] within 6-30 minutes). Moreover, 27% (11/41) of smokers had used e-cigarettes some days or every day in the past 30 days. Among smokers who had ever tried e-cigarettes, 59% (17/29) had used e-cigarettes to quit or cut down on smoking (Table 1).

Frequencies of Subthemes in Peer-Written Messages

Results are organized via the following themes: behavioral strategies, seeking help, improvements in quality of life, attitudes and expectations, and mindfulness/religious or spiritual practices. Under the behavioral strategies theme, distraction strategies were the most frequently suggested behavioral strategies (referenced 84 times in the 318 messages). Other behavioral subthemes included evidence-based treatments/cessation strategies (referenced 41 times), avoidance strategies (referenced 37 times), removal of smoking triggers (referenced 28 times), use of rewards (referenced 34 times), setting restrictions on one’s smoking and goal setting (each referenced 23 times), and use of reminders (referenced 17 times). In the seeking help theme, subthemes included seeking help or support from family/friends or close social network (referenced 56 times), former smokers (referenced 24 times), and health care professionals (referenced 22 times). Subthemes that emerged from the improvements in quality of life theme included concerns about the negative health effects of smoking (referenced 13 times), improving one’s health (referenced 22 times), and quality of life (referenced 21 times). Subthemes that emerged from the attitude and expectations theme included practicing positive self-talk (referenced 27 times), autonomy/independence from the smoking habit and financial cost of smoking (each referenced six times), and reducing second-hand exposure to others (referenced five times). Within the mindfulness/religious or spiritual practices theme, the subthemes included use of self-awareness techniques (referenced 36 times) and religious or spiritual practices to cope (referenced 13 times). Example messages for each subtheme are provided in Textbox 1 and Textbox 2.

Discussion

Principal Findings

Our approach to adapting a peer messaging intervention generated a rich set of messages that can be incorporated into text message cessation interventions tailored to support African American smokers. The peer-written messages aligned with evidence-based practices for supporting cessation and corresponded to the following five themes: behavioral strategies; seeking help; improvements in quality of life; attitudes and expectations; and mindfulness/religious or spiritual practices. The themes reflected in the messages described in this paper can be applied to smokers at all stages of behavioral change.

Within the distraction behavioral theme, use of evidence-based treatments/cessation treatment, and avoidance were the most frequently referenced subthemes. Distraction is often suggested as a way to prevent relapse for smokers [26]. Distraction techniques recommended by participants included physical activity and listening to music. Engaging in short bouts of physical activity is associated with a reduction in smoking cravings [27,28]. Listening to music helps participants manage emotional states and cravings that may act as cues to smoke [28]. Participants also suggested use of evidence-based treatment or cessation strategies such as nicotine replacement therapy. African American smokers reported less positive attitudes toward using cessation treatments and are less likely to use a Food and Drug Administration–approved cessation aid compared to White smokers [29]. Increasing the use of evidence-based treatment/cessation strategies among African American smokers is an important research priority in tobacco cessation [30]. Having a peer suggest use of cessation medications may increase the uptake of evidence-based strategies among African American smokers, and this needs further research. Avoiding situations that trigger smoking or peers who smoke is also an often-suggested strategy for smokers interested in quitting [31].

Participants frequently suggested seeking help from family and friends, and health professionals during cessation. Past studies indicated that social network support significantly contributes to cessation in African American individuals who smoke [32,33], and ways to increase this support for cessation need to be explored further [34]. Peers also suggested requesting help from health professionals. Brief interventions from health providers are shown to increase cessation among smokers [35]. However, African American smokers are less likely to receive these interventions [29,36]. Taking the initiative to request for help from a health professional may alleviate differences in receiving cessation advice from providers among African American smokers.

Improvements in the quality of life theme encompassed benefits of quitting and motivations to quit smoking. The most frequent subthemes were (1) improving one’s health and (2) quality of life. Given that the health effects of smoking and benefits of quitting are well-known, peers highlighting the health benefits of smoking may serve as reinforcement to African American smokers. Participants also highlighted several benefits of smoking to improve one’s quality of life, including eliminating the smell of cigarettes, breathing more easily, and having better taste, which are linked to one’s intentions to quit [34]. Smokers are more receptive to information that emphasizes the benefits of quitting [37]. The added benefit of having these messages written by peers is likely to enhance their impact on cessation.

Subthemes under the attitudes and expectations theme included practicing positive self-talk, autonomy/independence from the smoking habit, and the financial cost of smoking. Practicing positive self-talk has been associated with an increase in self-efficacy [38], and in turn, an increase in self-efficacy is associated with smoking cessation [39]. African American smokers are likely to benefit from practicing general self-regulatory activities, including self-talk, as these have been shown to result in smoking fewer cigarettes for up to 15 months [40]. Participants also urged peers to consider the financial costs of smoking as a motivation to quit. Smoking-related cost is an important motivator for cessation among smokers in lower socioeconomic groups [41,42].

Within the mindfulness/religion or spiritual practices theme, smokers were advised to use self-awareness techniques and religious or spiritual practices as coping mechanisms during cessation. In our past work in which we collected messages written mostly by non-Hispanic White smokers (32 out of 39 participants) [14], mindfulness, religious, and spiritual practices did not emerge from the peer-written messages. Although not exclusive to African American communities, use of spiritual and religious practices, as a coping response to health issues, is often adopted in this group [40], with positive and promising results in health outcomes [43]. Other health behavior change studies in African American populations have shown more positive outcomes if participants reported having strong ties within a religious congregation [41]. Exploring the use of religious practices for cessation is particularly important for African American smokers, since they are disproportionately impacted by tobacco-related health conditions. Participants also advised use of self-awareness techniques, which are shown to be effective for reducing the number of cigarettes smoked and for relapse maintenance [44,45]. The inclusion of self-awareness messages in cessation messaging interventions will be valuable.

To our knowledge, this is the first study to explore the content (and frequency of the content) of cessation messages developed by African American smokers. Peer-written messages developed in this study can be used to support cessation efforts among smokers motivated to quit, as well as among those less motivated. Smokers who are motivated to quit may experience withdrawal symptoms and smoking cravings during a quit attempt [39]. Messages that highlight distraction and avoidance strategies are therefore relevant to this group. Smokers who initiate quit attempts are vulnerable to declines in motivation, which can lead to resumption of smoking. Messages that highlight healthier lifestyle modifications (eg, engaging in physical activities), along with positive reinforcements, may be particularly helpful for maintaining motivation. For smokers currently not motivated to initiate a quit attempt, challenges, such as low self-efficacy and lack of support, reduce the chances of early quitting success [39]. Messages from peers that highlight ways to increase self-efficacy or ways to tap into sources of social support may be of particular benefit to this group.

This study had some limitations. Recruitment through ResearchMatch resulted in enrollment of highly motivated smokers. Therefore, the subthemes obtained from the peer-written messages may reflect behavioral strategies implemented by highly motivated smokers and could differ from strategies that might be implemented by smokers less motivated to quit. Our sample lacked diversity, as we enrolled a high proportion of female, older, and highly educated current and former smokers. Given that smokers with low socioeconomic status are the most likely population to have difficulty quitting [32] and that smoking is more prevalent in men than women [33], health behaviors observed in our sample could differ from those present in other populations. The meaning of messages may not have corresponded with the coders’ interpretations of the messages. We limited the effects of the researchers’ perceptions on the study findings by implementing coding protocols that preserved and used original words written by smokers.

Conclusion

For populations experiencing tobacco-related disparities, a reduction in smoking rates may require development of tailored cessation interventions. In this formative work, we were successful in developing a rich set of peer messages written by African American current and former smokers that incorporated many evidence-based strategies and can be used to support both low- and highly-motivated African American smokers. In future research, we plan to assess the impact of these peer-written messages within a technology-assisted intervention for smoking cessation among African American smokers.