Tobacco smoking is a major cause of preventable mortality and morbidity [1] and health inequalities for people with severe mental illness (SMI), including extreme social, economic, and physical health disadvantages [2]. SMI diagnoses include, for example, schizophrenia and bipolar disorder. In Australia, the smoking rates for SMI populations have remained inequitably high [3] for more than a decade [4]. Smoking rates for SMI populations in South Australia are alarmingly high, at almost triple (43.6%), that of the general population (15%) [5]. The high prevalence of smoking in this population is indicative of mental illness and smoking being intrinsically linked [4,6]. This is evident as people with mental illness often identify as smokers [6] and use smoking as a form of self-medication to help them cope with and relieve their symptoms of mental illness [4,6,7]. Owing to the inseparable nature of mental illness and smoking, nicotine addiction in this population has been difficult to treat [4].

Smoking-related studies have revealed that most people with SMI want to quit [8] and often attempt to quit but can find it challenging to quit without support [4,9]. A review of smoking cessation interventions, such as motivational interviewing, found limited evidence to support the interventions’ effectiveness in assisting people with schizophrenia to quit [10]. Cutting-edge digital health technology, such as mobile health (mHealth) smoking cessation apps [11-13], may contribute to the solutions needed to address this significant public health problem [11,14].

There are currently hundreds of smoking cessation apps available for download; however, limited studies have been conducted to assess the quality of app design. Research assessing the quality of generic smoking cessation apps revealed that most do not adhere to best-practice guidelines for smoking cessation, such as recommending pharmacotherapy [15,16]. Many smoking cessation apps also rated low on technical quality [17]. A review of 112 smoking cessation apps found that only 6 of these apps rated high on technical quality, such as having aesthetic appeal [16]. Despite the lack of quality for the vast majority of apps, scientific studies on smartphone interventions are promising in increasing cessation. Smokers who received the smoking cessation interventions demonstrated as much as a 1.7 times higher quit rate than smokers who did not receive the interventions [18].

In contrast to the vast availability of generic smoking cessation apps, there are only 3 that have been tailored for SMI populations [19-21]. This highlights a substantial gap in the availability of smoking cessation apps for SMI populations. This is particularly important when considering the cognitive impairments many people with SMI have to endure [22], which limits their ability to use apps [15,23,24]. A study on the QuitPal app developed by the National Cancer Institute, found that people with SMI experienced problems navigating the app, such as entering data, which was particularly relevant among participants with cognitive impairments and tremors [25]. Vilardaga et al used co-design, a person-centered approach to technology design, to involve people with SMI in the tailoring of the Learn to Quit app, developed by the University of Washington. Findings indicated that adapting a user experience (UX) approach with simple functionality, including large buttons and simple screens, improved the utility, usability, and acceptability of the app among people with SMI [21]. UX is a human-centered approach to improving end users’ performance and their psychological experience of technology systems [26]. Furthermore, exploration of participants’ preferences for app features found that people with SMI were interested in gamification (application of game design components and game principles in nongame systems) [27], interactive strategies to develop quit skills, and tracking devices for monetary incentives [21,25].

Building on this existing knowledge, there is a need to gain a deeper understanding of the relationship between mental illness, smoking, and smoking cessation to inform the design of effective smoking cessation app approaches for this population. This can help to guide how an app can be tailored to meet the specific requirements of people with SMI to reduce their smoking and quit. Currently, there are no smoking cessation apps for SMI populations that have been investigated within an Australian context, regarding their acceptability, feasibility, and usefulness. This study aims to address these important issues in relation to tailoring the Kick.it app for SMI populations [14].

Kick.it is a generic Australian-based prototype app that was originally co-designed for use by the general population of smokers [28], using intervention mapping (IM), which is a rigorous multitheoretical intervention development framework [29]. This consisted of a comprehensive needs analysis of the literature and stakeholder input from health professionals and smokers to identify the problem behaviors and determinants for smoking cessation. A co-design principle has also been used in this study to tailor the app for SMI populations before releasing the app on the marketplace. The design of the app for SMI populations enables app users to create a profile (ie, input information about their psychiatric diagnoses and smoking) and receive a personalized quit program that offers smoking cessation approaches tailored to meet their unique needs. These smoking cessation approaches are based on multitheoretical perspectives [28], as follows.

The Theoretical Domains Framework, a valid multitheoretical approach [30], underpinned the determinants for smoking cessation (eg, knowledge and skills) and the change objectives required to assist app users to quit (eg, increased knowledge of and ability to implement quit strategies) [28]. The Behavior Change Technique Taxonomy (v1) was used to identify behavior change approaches [31] and behavior change outcomes. These were then translated into app features and practical applications for smoking cessation [28]. The Persuasive System Design, a framework for technology development that targets attitude and behavior change, was also applied to inform the choice of app features [32].

There are 4 core features contained in the Kick.it app. The smoke and crave profile feature is based on the principles of ecological momentary assessment [33], which tracks user’s smoking and quitting behaviors in real time and delivers in-time quit strategies at critical moments to prevent smoking relapse and support smoking cessation (see Multimedia Appendices 1 and 2 for screenshots of in-time interventions). When an app user logs a smoke or crave, it activates the tracking device and provides them with a progress report [28] (see Multimedia Appendix 3 for screenshot of the tracking device). The Kick.it stack feature contains education and strategies to assist app users during their quit attempt (see Multimedia Appendix 4 for screenshot of an educational video on nicotine replacement therapy). The social network platform is a unique app feature that leverages peer support and normative social influence for smoking cessation through chatrooms, community feeds, and links to other app users’ social networks [28]. To our knowledge, Kick.it is the first app to include a social network feature to enhance social support for SMI populations. The missions and treatment goals feature is based on an incentive and reward system that encourages app users to engage in daily health-enhancing activities and log their pharmacotherapy use. A comprehensive overview of the development and design of the Kick.it app has been published elsewhere [28]. Figure 1 presents the Kick.it app features and practical applications [28].

Assessing the feasibility, utility, and acceptability of tailoring the Kick.it app for SMI populations provides an ideal opportunity to address the limited availability of smoking cessation apps for this population.

This study aimed to gain a deeper understanding of the lived experiences of smoking and quitting among people with SMI and their perception of social support for smoking cessation in relation to the Kick.it app. These experiences informed the researchers understanding of their perceptions regarding the feasibility, utility, and acceptability of the Kick.it app features to guide the tailoring of the app to their specific needs and preferences. The following research questions were investigated: (1) what are the facilitators and barriers to smoking cessation experienced by people with SMI and their perceptions of social support for smoking cessation? and (2) what features of the generic Kick.it app are perceived as feasible, useful, and acceptable in supporting people with SMI to prevent smoking relapse and quit smoking? [14].

Co-design methodology was used for this qualitative inquiry because of its value in offering consumer involvement and collaboration [34] in the tailoring of the Kick.it app, which is well matched to meet the study aims [14]. Co-design methods included the triangulation of semistructured in-depth interviews, observation, and in-situ exploration [35] of the Kick.it prototype app with participants.

Research ethics approval was obtained from the Southern Adelaide Clinical Human Research Ethics Committee (reference no. 16.17). A sample of people with SMI was then drawn from the community mental health services (CMHS) and the South Australian Cancer Council’s Quitline, within the Adelaide metropolitan area, Australia, between February 2018 and June 2018. Quitline is a free telephone service that provides information and advice to support people to quit smoking [36]. A purposive sampling method was used to recruit participants who were reflective of the target population and who meet the selection criteria [37]. CMHS and Quitline staff identified and screened potential participants’ eligibility to be involved in the study using the selection criteria. The inclusion criteria were (1) self-reported diagnosis of an SMI, (2) adult smokers (aged 18 years or more) who had attempted to quit smoking in the past 12 months and ex-smokers (abstinence for 7 days or longer before the interview) [38], and (3) the ability to provide informed consent as confirmed by CMHS or their doctor. The exclusion criteria were (1) individuals with acute severe suicidality or current acute psychosis as confirmed by CMHS or their doctor, (2) a sensory or motor impairment affecting the individual’s ability to participate in the study, and (3) a severe cognitive impairment affecting the individual’s ability to provide informed consent as confirmed by CMHS or their doctor [15]. Self-reported smoking status was determined using the following smoking status question “Which of the following best describes your smoking status?” and prompted from responses, “I’m a smoker, I smoke daily” and “I’m an ex-smoker, I never smoke now” [39].

CMHS and Quitline staff invited eligible participants to participate in the study and provided the contact details to the research team. Team member (PK) followed up with participants to further describe the study and organize an interview time. At the time of interview, written informed consent was obtained.

Two consecutive semistructured in-depth interviews were conducted. This study included an iterative 2-staged interview approach to provide participants with an individualized experience [34]. This was particularly important during the stage 2 interviews, as it enabled people with SMI to receive the personalized assistance needed to navigate the app and the time to reflect and provide feedback on its features.

Stage 1 interviews (approximately 1 hour) consisted of open-ended questions to elicit rich data and a depth of understanding regarding participants’ smoking-related experiences. The interview guide was informed by the research questions, the relevant literature [40,41], and in consultation with the research team. Stage 2 interviews (approximately 1.5 hours) continued to explore participants’ smoking-related experiences in relation to the app. These sessions involved sitting with each participant, as they viewed the prototype app and asking questions in relation to their perception of its features in accordance with the stage 2 interview guide. Observations and field notes were also used to record any reflections on the interview process [37] and the participants’ ability to navigate the app [14]. The stage 2 interview guide was developed from review of the limited studies on tailoring smoking cessation apps for SMI populations [21,24,25], in consultation with the research team, and from preliminary analysis of the stage 1 interviews. The interview guides were reshaped somewhat in accordance with the iterative process of analyzing the data as it was being collected to enable a flexible approach that allows a review and refinement of the interview guide questions (eg, redundant questions were excluded) [37]. Table 1 gives examples of stage 1 and stage 2 interview guides. The interviews were audio recorded and transcribed verbatim by an accredited transcriber to preserve the meaning and authenticity of the participants’ responses. The transcribed interviews were then compared against the audio recordings to ensure their accuracy.

Thematic analysis was used as it provides a systematic approach to organizing, categorizing, and interpreting qualitative data [42,43]. The research team conducted open coding of the first 4 transcripts, independently of each other. In vivo coding was used to exemplify the meaning associated with participants’ responses [44]. A series of team meetings were then held to discuss and debate the initial codes and agree on a structure to guide the coding of the remaining interviews. Categories, selective codes, and emerging themes were captured in a spreadsheet and grouped to assist the researchers to gain a clearer sense of the themes emerging from the data [42-44]. Tables 2 and 3 give examples of stage 1 and stage 2 categories, selective codes, and participants’ frequency of responses.

An iterative process of reading and rereading the transcripts enabled the 3 researchers to reflect on and gain an in-depth understanding of participants’ stories [42,43]. Data were interpreted using a constant comparative approach within and between transcripts to help identify, review, and refine the codes and themes [45]. Mind maps were developed by ordering and linking the codes and categories to the themes. This process prompted robust debate among the researchers, which deepened the interpretation of meaning within the data and finalized the ordering of the themes [42,43]. A dualistic approach was used to utilize existing themes within the literature to build on the limited theory underpinning the design of smoking cessation apps for SMI populations (theory driven) and explore new emerging themes from this study’s findings (data driven) [37,46]. The triangulation of the different research team members’ interpretations and perspectives of the data added further methodological rigor [47]. Sample size was established according to evidence of data saturation being achieved [48]. The researchers concurred that data saturation was achieved by the sixth participant of the stage 2 interviews.

A total of 12 adults with SMI participated in the study, comprising 6 male and 1 female smokers, and 2 male and 3 female ex-smokers. All participants had been medically diagnosed with either an individual diagnosis of schizophrenia, borderline personality disorder or bipolar disorder, or psychiatric comorbidity. Most participants (75%, 9/12) were diagnosed with paranoid schizophrenia and psychiatric comorbidities, such as depression and anxiety. Some participants (58%, 7/12) had a socioeconomic disadvantaged status, as indicative of these participants’ receiving disability support pension as their primary source of income. All participants were in receipt of community-based support services. Participants’ characteristics are presented in Table 4.

Key findings highlighted several psychosocial factors as important in tailoring the Kick.it app for SMI populations. The key themes that emerged from the data in relation to participants’ lived experiences of smoking and quitting and their perceptions of the Kick.it app features are described below. Key findings aligned with broader psychosocial needs and experiences of perceived stigma and social isolation for this population, which indicated that smoking cessation efforts are inseparable from the environmental and personal context in which these smokers experience and cope with mental illness in their community. Examples of participants’ quotes that help to exemplify the meaning and interpretation of participants’ responses are also included. Multimedia Appendix 5 gives more examples of participants’ quotes.

Results suggest that tailoring of the Kick.it app is feasible, useful, and acceptable for people with SMI. Participants were interested in using mHealth interventions to support their quit smoking efforts. Similar to other studies [21,25], our findings indicated using a co-design approach to improve the utility, usability, and acceptability of smoking cessation apps for SMI populations is required.

Exploring participants’ smoking and quitting experiences highlighted the complexity associated with the inextricable link between mental illness and smoking [6,7], which has pervasive impacts on participants’ psychosocial well-being and affects every aspect of their daily lives. These findings suggest that specific tailoring of the app is needed to assist people with SMI to navigate the complex interaction between mental illness and smoking that impact on their capacity to quit. Consistent with previous evidence [10], results of this study found that current smoking cessation approaches are limited in supporting people with SMI to quit as they do not account for their mental health–related needs [49], nor do they address the symptoms of mental illness and nicotine dependency simultaneously. The need for a dual approach to smoking cessation that provides quit strategies to address both the symptoms associated with smokers’ psychiatric diagnoses and nicotine addiction was identified many years ago [50]. Another limitation of current smoking cessation approaches is that they do not offer real-time assistance regarding experiencing the urge to smoke. For example, they do not address stressors arising in the context of their daily lives in the moments [33] when those stressors are heightened, and the person is at high risk of reaching for a cigarette to alleviate their distress. The delivery of novel smoking cessation approaches that offer assistance in the here and now to support people with SMI to prevent smoking relapse and quit smoking has been established as important in this study and warrants further investigation. The Kick.it app’s smoking cessation interventions may provide the solution to address the limitations of current smoking cessation approaches and support people with SMI to quit.

This study also provided valuable insight into the effects of stigma and a possible solution to the perpetuated entrenched marginalization and social disadvantage among smokers with SMI [11,49]. Gaining a deeper understanding of participants’ lived experiences of stigma and how those experiences had impacted on their self-worth highlighted why it was important to them to have a caring app. Therefore, the features of the Kick.it app may be different from generic smoking cessation apps in that it could focus more on building their self-esteem and self-efficacy. There is also potential for the Kick.it app’s caring features and social network to reduce the effects of stigma, social isolation, and loneliness by enhancing social inclusion, and a sense of belonging to a social support network for smoking cessation. The benefits of using a social network to gain peer support for smoking cessation among people with psychosis has been established [51].

An adapted Kick.it app has the potential to become an all-encompassing solution, a virtual friend that offers around the clock support to help people with SMI to quit, and address stigma which is a by-product of SMI [49]. The specific app features and their adaption that participants perceived would be feasible, useful, and acceptable in assisting SMI populations to quit smoking included the smoke and crave profile, the supportive messaging, the tracking device, and social networking.

Limitations of the study included the participants’ self-reported [38] smoking status and psychiatric diagnoses. However, smoking status was obtained using a screening tool adapted for this population [39], and all but 1 participant was recruited from CMHS, which provides ongoing case management support to people with existing SMI. Another limitation relates to the participants’ sampling a prototype of the Kick.it app at the stage 2 interviews which provided users with a brief window of 1.5 hours to view the app and advise on its features in an interview setting rather than a naturalistic setting [37].

This study provides evidence for innovative smoking cessation approaches to support people with SMI to prevent smoking relapse and successfully quit. We contributed to the limited knowledge on designing smoking cessation apps for SMI populations by using a co-design principal based on the IM framework to explore their lived experiences of smoking and quitting and their perception of the Kick.it app features to guide the tailoring of the app. Through the lens of people with SMI, this study provides insight into the smoking behaviors and personal struggles they encounter in their endeavor to quit smoking. By confronting some of the major barriers to smoking cessation for this population, this study contributes to possible solutions for important mental health–related issues, including stigmatization and social isolation [11,49]. The next stage of research planned by the authors of this study involves tailoring the Kick.it app in accordance with the findings and then conducting a quantitative study to gain a representative sample to assess the effectiveness, utility, and acceptability of the app among people with SMI in relation to smoking cessation [14].