Development and initial testing of mindful journey: a digital mindfulness-based intervention for promoting recovery from Substance use disorder

Abstract

Background/Objectives

There is a great unmet need for accessible adjunctive interventions to promote long-term recovery from substance use disorder (SUD). This study aimed to iteratively develop and test the initial feasibility and acceptability of Mindful Journey, a novel digital mindfulness-based intervention for promoting recovery among individuals with SUD.

Patients/Materials

Ten adults receiving outpatient treatment for SUD.

Methods

Phase 1 (n = 5) involved developing and testing a single introductory digital lesson. Phase 2 included a separate sample (n = 5) and involved testing all 15 digital lessons (each 30- to 45-minutes) over a 6-week period, while also receiving weekly brief phone coaching for motivational/technical support.

Results

Across both phases, quantitative ratings (rated on a 5-point scale) were all at or above a 4 (corresponding with ‘agree’) for key acceptability dimensions, such as usability, understandability, appeal of visual content, how engaging the content was, and helpfulness for recovery. Additionally, in both phases, qualitative feedback indicated that participants particularly appreciated the BOAT (Breath, Observe, Accept, Take a Moment) tool for breaking down mindfulness into steps. Qualitative feedback was used to iteratively refine the intervention. For example, based on feedback, we added a second core mindfulness tool, the SOAK (Stop, Observe, Appreciate, Keep Curious), and we added more example clients and group therapy videos. In Phase 2, 4 out of 5 participants completed all 15 lessons, providing initial evidence of feasibility. Participants reported that the phone coaching motivated them to use the app. The final version of Mindful Journey was a smartphone app with additional features, including brief on-the-go audio exercises and a library of mindfulness practices. Although, participants used these additional features infrequently.

Conclusions

Based on promising initial findings, future acceptability and feasibility testing in a larger sample is warranted. Future versions might include push notifications to facilitate engagement in the additional app features.

Keywords:

• Mindfulness-based relapse prevention
• mindfulness meditation
• digital interventions
• smartphone application
• substance use disorder

Introduction

Substance use disorder (SUD) is a prevalent and disabling condition associated with great costs to society and individuals [1]. Current treatments for SUD are suboptimal given that relapse (i.e. return to problematic substance use) remains a common outcome within the first year after receiving treatment for SUD [2–5]. SUD, particularly when individuals have severe symptoms, is considered a chronic relapsing condition [6]. Major neurobiological models of addiction posit that relapse is a defining feature of addiction [7].

Efforts to curb relapse rates and promote long-term recovery are critically needed. Recovery has been defined as a multi-dimensional, dynamic process of change involving remission from SUD, reduction in substance use, and improvements in biopsychosocial functioning [8]. Novel interventions are needed to promote improvements in broader life functioning, rather than only reductions in substance use. Moreover, interventions are needed that target core processes associated with SUD relapse and recovery. Notably, meta-analyses and reviews suggest that negative affect [9–12] and difficulties regulating negative affect [12–15], substance craving [12,16–19], and anhedonia [10,20,21] are core processes that predict recovery outcomes, such as substance use and substance-related problems.

Mindfulness-based treatment (MBT) holds great promise as a continuing care approach for promoting multi-dimensional recovery from SUD. Mindfulness is defined as paying attention to present moment internal and external experiences with openness, curiosity, and acceptance [22]. MBT for SUD [23–25] involves systematic training in mindfulness to improve awareness and acceptance of challenging experiences (e.g. negative affect, craving), reduce automatic reactions to these challenging experiences, and increase flexible regulation of attention towards present-moment experiences, including those linked with positive affect. Meta-analyses show that MBT significantly reduces frequency of substance use [26,27] and increases the likelihood of abstinence [12]—in some cases greater than active treatments such as cognitive-behavioral therapy [27–29]. Meta-analyses show that MBT has a significant effect on other recovery outcomes, such as substance-related problems [30] and psychological functioning [12,31].

Moreover, research indicates that MBT has significant effects on putative mechanistic processes, including regulation of negative affect [32,33], craving [26,31,34,35], anhedonia [36], trait positive affect [37,38], and autonomic responsiveness to non-drug rewards [36,39–41]. Indeed, MBT is designed to reduce SUD symptoms and promote broader functioning. Specifically, MBT aims to reduce SUD symptoms by enhancing mindful regulation of substance craving (i.e. noticing craving with acceptance and letting it pass). Further, MBT aims to promote broader functioning via mindful regulation of negative emotions [23] (i.e. noticing and nonjudgmentally accepting negative emotions) and positive emotions [24] (i.e. paying attention to sensory experiences and emotions with curiosity during pleasant activities).

Despite promising efficacy of MBT for SUD, implementation remains a challenge. Notably, there are a limited number of clinician training opportunities and a lack of effective implementation strategies [42]. Furthermore, manualized MBIs for SUD are offered as eight sequential 2-hour closed-cohort group sessions [43]; this delivery format has been highly challenging to implement in addiction treatment settings [42].

Digital interventions, such as web-based programs or smartphone-based apps, have great potential for increasing access to evidence-based treatments [44]. Currently, 90% of American adults use the Internet and 83% own mobile smartphones [45]. Digital interventions are viable and effective treatments for many disorders [46], including SUD [47]. Yet, it is important to note that engagement in digital interventions can be low, particularly when there is no concurrent clinician-based support [48]. Web-based cognitive-behavioral therapy has been shown to reduce substance use in many trials [47,49]. However, existing digital cognitive-behavioral therapies for SUD have modest effects on substance use [49], and importantly, there is a lack of evidence that these therapies impact other recovery outcomes, such as psychological functioning.

Given the promise of MBT for SUD, expanding access to this treatment approach—via methods like digital delivery—may fill an important gap in efforts to treat SUD and support long-term recovery. Nevertheless, there are currently no existing digital interventions offering MBTs among individuals with alcohol and other drug use disorders such as cocaine, opioid, and cannabis use disorder.

Research suggests that digital MBTs are feasible, acceptable, and efficacious in clinical populations. For example, digital MBTs are efficacious for stress, anxiety, and depression [50,51]. While there are no digital MBTs for individuals with alcohol and other drug (e.g. cocaine, opioid, cannabis) use disorders, there are digital MBT s for tobacco use disorder [52]. Notably, there is considerable empirical support for both web-based and smartphone app-based acceptance and commitment therapy (ACT), a mindfulness-based behavioral therapy, for facilitating smoking cessation among individuals with tobacco use disorder [52,53].

Overall, research shows that MBTs may be highly promising for facilitating multi-dimensional recovery SUD, via targeting core mechanisms implicated in recovery. However, there remains a need to develop scalable implementation approaches. Accordingly, this study reports on the development and initial testing of acceptability (e.g. ratings of usability, visual appeal, and helpfulness) and feasibility (intervention engagement) of a smartphone-delivered MBT, called Mindful Journey, for reducing relapse among individuals in early recovery from various types of SUD, such as alcohol, cocaine, opioid, and cannabis use disorder.

Methods

Overview of mindful journey and the development process

To provide context for the entire development process, here we begin by briefly describing the final version of Mindful Journey that was ultimately developed at the end of the three-year development period. Mindful Journey is based on several existing in-person MBTs for SUDs, particularly Mindfulness-Based Relapse Prevention [23] (MBRP). The final version of Mindful Journey we developed was a smartphone-based app that included: (a) 15 sequential digital lessons (∼ 30–45 min each), with each lesson comprised of several animated videos (typically 1- to 3-min each), one or two audio-guided mindfulness meditation practices (typically 5- to 10-minutes), and several question prompts displayed within the digital lessons on the device; (b) a ‘Cope Now’ section with 2-min ‘on-the-go’ audio-guided mindfulness practices; (c) an ‘All Practices’ section with a library of meditation practices that can be repeated, and a subsection with each of the 15 lessons that can be repeated as desired; and (d) a ‘Track’ section with options to self-monitor and record recovery milestones. Figure 1 provides examples of animations that are in the video clips of Mindful Journey. Figure 2 provides visuals of the how the app itself looks on a smartphone. Additionally, access to the app was paired with brief phone calls and texts focused on supportive coaching and technical support (described in more detail in the procedures section).

Figure 1. Example animations in the video clips of Mindful Journey.

Figure 2. Example visuals of the screens on the app.

Next, we outline the steps of developing Mindful Journey and provide details on the various prototypes that were developed over time during different phases.

Key elements informing intervention development

Twelve key elements played an important role in the content that was developed for the app. These elements are elucidated in Table 1.

Table 1. Key Elements Informing intervention development.

Element	Details
In-person mindfulness treatments for SUD	Several existing in-person mindfulness treatments for SUD have demonstrated efficacy, and therefore can inform content development of the app These include Mindfulness-Based Relapse Prevention (MBRP) [23], Acceptance and Commitment Therapy (ACT) [25,53], Mindfulness-Oriented Recovery Enhancement (MORE) [24], Regulation of Craving Training (ROC-T) [57] and Dialectical Behavior Therapy (DBT) [56] When developing the content for the app we drew upon these treatments Key features drawn from MBRP included formal meditation practice, inquiry after meditations, using an acronym to promote mindfulness in daily life, applying mindfulness to craving, difficult emotions, thoughts, and sensations, being curious and present in daily life, self-compassion, exploring needs underlying urges to use substances, and self-care (i.e. lessons 1 to 10) Key features drawn from ACT include acceptance of craving (lessons 3 and 4), seeing thoughts as thoughts (lesson 6), and values clarification (lesson 14) Key features drawn from MORE include savoring of pleasant experiences and positive emotion regulation (lessons 10–13) Key features drawn from DBT include using multiple methods to teach mindfulness (e.g. acronyms, brief exercises, catchphrases, and modeling) (integrated throughout all the lessons)
Clinical experience	Clinical experience offering mindfulness treatment to individuals with SUD can be very valuable in informing ongoing refinement of treatment Our team drew upon our extensive clinical experience delivering MBTs for SUD For example, many of the authors have had clinical experiences in which they recognized that patients were satisfied with acronyms for learning tools and frequently used acronyms to apply skills. Importantly, research data also indicates that participants with SUD who have received MBT frequently use acronym-based tools [56,59]. Therefore, we included acronyms in the app Additionally, many of the authors have had clinical experiences in which the inquiry process following meditation practice (i.e. reflecting on and talking about experiences one noticed during a meditation) was a valuable process to pique participant’s curiosity in their own experiences and the process of relating mindfully to these experiences. Therefore, we included inquiry-based questions (e.g. what did you notice?) after every meditation practice, and we included therapist-client videos involving inquiry-based discussion.
Predictors of relapse	Research shows that negative affect, difficulties regulating negative affect, substance craving, and anhedonia predict relapse The app includes content focused on these core processes (e.g. directly explaining to the user how these processes predict relapse)
Mechanisms of recovery	Research suggests that improvements in mindfulness, reductions in craving, improvements in negative emotion regulation, and increases in savoring may be mechanisms underlying the effect of mindfulness-based treatments on recovery outcomes Table 3 provides details on key content in the app that is intended to mobilize each of these mechanisms. The app also directly explains to users that mindfulness, reductions in craving, improvements in emotion regulation, and savoring are key factors that promote recovery
Multi-dimensional recovery	Recovery has been defined as a multi-dimensional, dynamic process of change involving remission from SUD, reduction in substance use, and improvements in functioning and quality of life We included content in the app designed to target each of these dimensions of recovery (e.g. discuss mindfulness as a general skill used in range of life situations, not just those related to substance use) The app also defines multi-dimensional recovery to users and directly explains to users that the app is designed to promote multi-dimensional recovery
Digital interventions for SUD	Developing Mindful Journey was informed by existing digital interventions for SUD, particularly Computer-Based Training in Cognitive Behavioral Therapy (CBT4CBT) [60] For example, our team chose to develop multiple digital lessons that are 30- to 45-minutes in length (described further below) because prior research testing digital intervention like CBT4CBT have shown that digital interventions in the format of multiple 30- to 45-minutes lessons are acceptable, feasible, and efficacious among individuals with SUD [49]. Additionally, consistent with CBT4CBT, we chose to: (a) deliver the content primarily via video clips and audio-guided exercises, rather than text-based information, (b) have all language in program be at a 6^th grade reading level or lower, (c) include narrators who guide participants through the program, and (d) include example clients who share their experiences and model skills Key features drawn from ROC-T include combining distal substance cues (e.g. cash) with mindfulness instruction (lesson 3)
Digital mindfulness treatments	Developing Mindful Journey was also informed by existing digital mindfulness treatments This includes iCanQuit a smartphone app-based on acceptance and commitment therapy [53] that aims to promote cessation from tobacco use. Similar to iCanQuit we included example clients, animations, and broke down lessons into smaller segments. Moreover, similar to iCanQuit, Mindful Journey has a targeted focus on acceptance of craving This also includes web-based mindfulness-based cognitive therapy (MBCT) for depression [51]. For example, similar to MBCT we include sequential series of lessons that are approximately 45-minutes long, and we include inquiry-based reflection questions after each meditation practice Similar to other mindfulness apps, such as Headspace [58], we used animations to teach mindfulness concepts and skills
Brand identity	Brand identity is the unique ‘personality’ of the app and how the app is designed to make people feel. The brand identity is ideally suitable for the target audience, which is individuals in early recovery from SUD. A cohesive and suitable brand identity will increase engagement in the app The key personality descriptors of the app are calming, encouraging, playful, scientific, professional. The app is designed to make users feel at ease, welcomed, supported, hopeful, curious, and trusting of the content The brand identity can be expressed through multiple elements, including the name of the app, color schemes, logo, other visual content, and the tone and language used in the meditations and in the videos with narrators/therapists and the example clients The color scheme is three shades of blue to connote calmness and ease. The logo is a blue wave that connects with the metaphor of craving as a temporary wave that one can ride out. This logo is intended to provide encouragement that craving and other challenges are not insurmountable, but instead one can ride out or surf the wave and continue moving in the direction of recovery. The font used for the word mark (i.e. the word Mindful Journey next to the logo image) and the font used in the videos is a style that balances professionalism and playfulness The app tries to avoid taking a too serious and formal tone that may come across as cold or stale. Instead, the app strives to sometimes exude an attitude of playfulness, lightness, and curiosity. For example, the BOAT and the SOAK are somewhat playful acronyms. The catchphrase ‘Use the BOAT to stay afloat’ is somewhat silly. The inclusion of animations, instead of video clips of real people, is intended to deliver the content in a lighter and more playful way The visual content of the videos (e.g. illustration, moving animations) are modern, simple, and professional looking, and include charts and images of a brain to connote scientific credibility. The app is scientific by clearly explaining—in simple language—how mindfulness is theorized to improve recovery. The narrators/therapists talk in a gentle and supportive tone while also being professional and direct. The example clients express hope and encouragement of how mindfulness can improve their recovery
Simple layout and navigation	Simple and noncomplicated layout and navigation will enhance engagement The sequence of 15 lessons involves very simple linear navigation. All users have to do is press continue again and again to navigate to the next section within a lesson and to advance to the next lesson after finishing a lesson A bottom navigation bar that stays on the screen is commonly used in apps so users can easily access core sections of the app. We included a bottom navigation bar with 4 icons so users can easily access the core sections
Diversity and inclusivity	We designed the app to include individuals with a diverse array of identity dimensions and individual differences (e.g. gender, race, age, occupation, type of SUD) so that individuals using the app feel welcomed, valued, and respected For example, we chose to have two narrators varying in gender and race. And we chose to have example clients that varied in gender, race, age, socioeconomic status and occupation, as well as the primary substances they were using
Low-cost app development resources	Our team strived to leverage low-cost resources for building the digital intervention. This was driven in part by a limited budget and a desire to create a digital tool that can be sustainable over time and does not require exorbitant costs to maintain it Details on the resources we used are provided in Supplementary Table 1
Evidence-based educational principles	Our development approach was informed by several evidence-based educational principles, which are explained below
Active learning	Active learning is learning by doing and reflecting on what one experienced The app includes meditation exercises in every lesson and reflection question after each exercise (e.g. What did you notice?) The app aims to facilitate ongoing active application of mindfulness in daily life and prompts users to reflect on what they experience when applying mindfulness
Spaced learning	Spaced learning is repeatedly engaging in similar learning activities with intervals of time in between each session The app includes numerous lessons (e.g. 15) that individuals complete over time (e.g. 8 weeks) and the app can recommend spaced practice (e.g. 2–3 lessons per week)
Mnemonics	Mnemonics are techniques to improve memory such as metaphors, imagery, acronyms, and rhymes The app includes two key acronym: the BOAT the SOAK The app has a coherent theme with nautical imagery and language, including the BOAT, recovery as a journey or voyage, craving as a wave and riding out the wave with the BOAT, changing weather patterns, and soaking in the sun as savoring Thre app includes ‘Use the BOAT to stay AFLOAT’ as a catchy rhyme. The app includes other metaphors such as emotions as guests or visitors, and mindfulness as a muscle that can get stronger
Vicarious learning	Vicarious learning is the process of learning by observing other people’s experiences and behaviors Participants learn about the experiences of two core characters (Maria and Jamal) and two side characters (Jen and Steven) who re-occur throughout the lessons The app includes several types of videos with the characters that are intended to provide an array of vicarious learning opportunities For example, the characters provide testimonials (i.e. share past or recent experiences, such as challenges and successes during recovery, and application of mindfulness skills) The app includes ‘Inside the Mind’ videos in which participants get an inside look at characters’ internal thought processes while applying mindfulness in various situations The app includes individual- and group-based therapy videos with discussions between therapists and the characters. These include inquiry-based discussions about momentary experiences after meditations, as well as discussions about the process of recovery and relapse and key concepts and application of mindfulness skills
The multimedia principle	The multi-media principle posits that learning with visuals and words is more effective than words alone. We designed the program to be primarily video-based, instead of audio only Carefully designing videos with moving animations, still graphics, and text-based animations that complement the words spoken by the narrator/therapists or characters
Minimizing cognitive load	Minimizing cognitive load is the process of offering learning content in a manner that does not overload working memory capacity The lessons broken into numerous ‘bite-size’ segments The app introduces new content at a relatively slow pace and includes lots of repetition and review The app includes a simple, linear process of navigating through lessons The app includes many semi open-ended questions (‘What did you notice?’) with examples of potential answers (warmth, tingling, pressure), instead of fully open-ended questions
Positive reinforcement	At the end of lessons, a funny congratulatory GIF is presented The narrators provide ongoing warm and genuine verbal praise as users advance through the content

The first element we drew upon is existing in-person mindfulness treatments for SUD. Given the efficacy data on MBRP [28,29,54,55] and given the team’s predominant expertise in MBRP (particularly authors CRR, MK, ES, SB, KW, and HK), we opted to base the content of Mindful Journey primarily on Mindfulness-Based Relapse Prevention (MBRP) [23]. However, given efficacy of related MBTs, we also drew from similar mindfulness- and acceptance-based approaches including Acceptance and Commitment Therapy (ACT) [25,53], Mindfulness-Oriented Recovery Enhancement (MORE) [24], and Dialectical Behavior Therapy (DBT) [56]. We decided to draw from several different MBTs because these different treatments, while overlapping, also have unique areas of focus, and therefore valuable content can be drawn from a range of approaches to create a wholistic mindfulness training experience. Table 1 has details on which specific intervention features were drawn from each of these treatments.

The second key element we drew upon was our extensive clinical experience delivering MBTs, both in group-based and individual format, to individuals with SUD. As explained in Table 1, the authors have had clinical experiences in which patients seem to benefit from acronym-based tools and engaging in inquiry-based discussion about direct experiences following a meditation. Thus, in the app we include acronyms and question prompts similar inquiry-based question asked during in-person MBT groups.

The third key element is research on predictors of relapse among individuals with SUD. As noted in the introduction, negative affect, difficulties regulating negative affect, substance craving, and anhedonia are established predictors or relapse. We made sure to have lessons focused on applying mindfulness to these challenges.

The fourth key element is research on mechanisms of recovery, which of course are closely related to the abovementioned predictors of relapse. As noted in the introduction, studies suggest that mindfulness, reductions in craving, improvements in negative emotion regulation, and increases in savoring may be mechanisms of recovery in the context of in-person MBTs. Table 2 provides in-depth details on key content that is designed to target these core mechanisms.

Table 2. Target mechanisms and content intended to mobilize these mechanisms.

Target mechanisms	Key content
Mindfulness: paying attention to present moment experience with openness, curiosity, and nonjudgmental acceptance	One or two audio-guided mindfulness meditations for each lesson, typically around 5- to 10-minutes each Reflection questions immediately after mindfulness meditations to facilitate nonjudgmental describing of subjective experiences (e.g. thoughts, emotions, sensations) with words Brief (about 3 min) audio-guided exercises of the BOAT (Breathe, Observe, Accept, Take a Moment) and SOAK (Stop, Observe, Appreciate, Keep Curious) tools to facilitate mindfulness in daily life Didactic narrator-led video clips explaining mindfulness, key related concepts (the nature of thoughts, emotions, and sensations) and the application of mindfulness in daily life Video clips with characters providing testimonials or characters having a discussion during an individual or group-based therapy session to model key processes, such as nonjudgmental describing of present moment experience with words, as well as the application of mindfulness in daily life A variety of additional questions prompts related to mindfulness and its applications, including quiz questions, as well as both open-ended and mixed open-/closed-ended questions eliciting personal perspectives and understandings on the material and plans for applying skills in daily life
Substance Craving: the subjective desire to use substances	Audio-guided mindfulness exercises combining imaginal exposure to a situation eliciting substance craving with mindfulness instruction Exercises combining presentation of images to elicit craving (food and distal substance cues such as cash) and audio-guided mindfulness instruction A variety of question prompts related to craving and coping with craving, such as quiz questions, question to identify triggers of craving, creating a plan for coping with triggers and craving Audio-guided exercises, video clips, and question prompts focused on key themes, such as: (a) using the BOAT to cope with craving, (b) exploring personal needs underlying craving for substances, and (c) the role of self-compassion in recovery
Negative emotion dysregulation: difficulty managing the experience of negative emotions and how one responds to these emotions	Audio-guided mindfulness exercises combining imaginal exposure to a situation eliciting negative emotions with mindfulness instruction Reading of the ‘Guest House’ poem [23] by Rumi to facilitate reflection of negative emotions as temporary guests or visitors Audio-guided exercises, video clips, and question prompts focused on key themes, such as (a) using the BOAT to cope with negative emotions, (b) the nature of emotions, (c) the role of thoughts in challenging situations, (d) mindfulness of thoughts, (e) the role of the body in managing emotions, and (f) mindfulness practice as self-care in daily life
Savoring: the ability to appreciate pleasurable features of daily experiences and positive emotions that arise from these experiences	Audio-guided mindfulness exercises combining imaginal exposure to pleasant situations with mindfulness instruction Exercises combining presentation of images to elicit positive emotions and audio-guided mindfulness instruction Audio-guided exercises, video clips, and question prompts focused on key themes, such as (a) using the SOAK during daily activities, (b) the role of curiosity in savoring, (c) appreciating the full range of positive emotional experiences (including contentment, calm, meaning, connection, and love), (d) clarifying personal values and meaning-making, and (e) appreciating positive emotions while also allowing them to come and go instead of ‘chasing’ them or ‘clinging’ onto them

The fifth key element is multi-dimensional recovery. Our intention was for the app to target multi-dimensional recovery as the outcome, which includes remission from SUD, reduction in substance use, and improvements in functioning and quality of life [8]. Hence, the app includes content closely related to SUD symptoms, as well as content that is broader and intended to facilitate general well-being.

The sixth key element is existing digital interventions for SUD. In particular, we drew from Computer-Based Training in Cognitive Behavioral Therapy (CBT4CBT) for SUD. For example, similar to CBT4CBT we included 30- to 45-minute multi-media lessons. Table 1 provides further details on how we drew from CBT4CBT. We also drew from Regulation of Craving Training (ROC-T) [57].

The seventh key element is existing digital mindfulness treatments. We drew from several of these treatments, including a smartphone app-based acceptance and commitment therapy intervention for smoking cessation [53], a web-based MBT for depression [51], and publicly available apps such as Headspace [58]. Further details are provided in Table 1.

The eighth key element is brand identity, or the unique personality of the app and how it is intended to make people feel while using it. A cohesive brand identity provides an overarching framework for making decisions regarding many intervention components, including overall design, logo, visuals, therapeutic content and how this content is delivered. And importantly, a strong brand identity can increase app recognition and engagement. Table 1 describes several ways in which we defined a brand identity for Mindful Journey and made development decisions based on this identity. Notably, the key personality descriptors of the app are calming, encouraging, playful, scientific, professional.

The ninth key element is having a simple layout and navigation on the app. For example, individuals primarily go through the program in a simple linear way, advancing from one section to another within a lesson, and from one lesson to the next by simply pressing continue again and again.

The tenth key element is diversity and inclusivity. The narrators/therapists and characters in the app have a range of identity dimensions (e.g. gender, race/ethnicity) to make the app as accessible as possible to participants. The gender and skin color of the narrators/therapist (authors CR and AH) match the skin color of the animated avatars. Additionally, the race/ethnicity and skin color of the characters matches the individuals who did the voiceovers. We hired professional actors to do the voiceovers of the characters.

The eleventh key element is using low-cost app development resources. Our team hopes that the Mindful Journey can be sustained over time, and we reasoned this would be more likely if it does not require high costs to maintain the app or make further updates. Relatedly, low-costs for maintaining the app may increase the possibility of making the app available at no-cost or low cost in the future. Details on the low-costs app development resources that we used are presented in Supplementary Table 1 and discussed further below.

The twelfth key element is incorporating evidence-based educational principles. An app-based intervention is after all an educational tool. From script writing to visual design, our team aimed to leverage as many educational principles as possible to optimize learning and memory consolidation. These principles include active learning, spaced earning, mnemonics, vicarious learning, the multi-media principle, minimizing cognitive load, and positive reinforcement. Table 1 provides details on these principles and how features of the app aimed to leverage each principle.

Digital resources used to create the app

Supplementary Table 1 provides details on the low-cost digital resources we used to create Mindful Journey. In short, we used Audacity to record and edit all audio recordings, Fiverr to find freelance voiceover artists for the characters, Powtoon to create animated videos, Vimeo to host and stream videos, Qualtrics to build digital lessons, and Thunkable to develop and publish a smartphone app-based version of Mindful Journey. These resources proved to be tremendously valuable and drastically reduced the costs for developing the app.

Steps in the development process

Our process for developing and refining the app over time was informed by user-centered design procedures [61] that involve iteratively developing an app while collecting and utilizing information about participants’ experiences, preferences, and ideas related to prototypes of the app. Consistently, we had participants test and provide feedback on several prototypes of Mindful Journey. Our development approach was also informed by the stage model of intervention development [62], which prioritizes intervention refinement, modification, and pilot testing before efficacy testing. We prioritized obtaining in-depth nuanced feedback from a small set of participants so we could refine Mindful Journey. Efficacy testing will be conducted at a later stage after feedback is obtained, refinements are made, and feasibility and acceptability are demonstrated. Additionally, our development approach was informed by the theoretical framework of acceptability (TFA), which emphasizes the importance of assessing numerous dimensions of acceptability for interventions and utilizing this feedback to refine interventions. During the development process, we obtained feedback on many acceptability dimensions, including usability, overall satisfaction, appeal of visual content, understandability, desire to continue the program, satisfaction with the name of the program, perceived helpfulness, and satisfaction with the pace and length of the program.

Figure 3 provides a schematic of our development process, which includes two core phases. In Phase 1, we focused on developing and testing a preliminary prototype in the form of a sample digital lesson (∼1 h). In Phase 2, we focused on developing and testing two versions of an advanced prototype with all 15 digital lessons. After each step in the development process, our team reviewed the audio-recordings and written feedback and then discussed participant feedback and how to apply this for updating the intervention. Members of the team voted on whether or not to make a given change, and then author CRR took the votes into consideration, and he had the final say on whether to make the change or not. Of note, even if only one or a few participants provided a piece of feedback, in some cases our team deemed this feedback valuable and incorporated that feedback when updating the intervention.

Figure 3. Visual schematic of the steps in our app development process.

Developing a preliminary outline of the intervention

As shown in Figure 3, in the ‘Prepare’ Step of Phase 1, our team created an initial outline of the entire intervention, which served to establish a vision of what the intervention might look like and to guide script writing. We decided that the initial lessons would focus on mindfulness of craving and negative emotions given these topics are central to MBRP and are covered at the beginning of MBRP. We decided to have the latter half of the program focus on mindful engagement in pleasant activities and regulation of positive emotions, which are topics particularly relevant to MORE. Additionally, we theorized that improving regulation of craving and negative affect first would ‘create space’ for individuals to subsequently pay attention to pleasant activities.

As described in Table 1, we drew upon other features in various MBTs for SUD. For example, drawing from other topics in MBRP, we included lesson content on mindfulness of thoughts, self-compassion, self-care, exploring wholesome needs underlying craving for substances, and developing a regular meditation practice. Drawing from ACT, we included lesson content on acceptance of craving, seeing thoughts as thoughts, and clarifying one’s personal values. Drawing from ROC-T, we included lesson content that combined presentation of images with mindfulness instruction. Finally, drawing from DBT, throughout the lesson, we used multiple methods to teach mindfulness, such as acronyms, brief exercises, catchphrases, and modeling of skill application. These additional lesson topics were included to add variety to the content of the program, while also having individuals continue to practice core mindfulness skills related to varying topics.

The details of the preliminary outline of the lessons were largely similar to the final set of lesson. The key revisions to the preliminary outline were changing the order the lessons slightly (e.g. moving the Exploring Your Needs lesson to be right after self-compassion given the parallels between these two lessons) and breaking up longer lessons into two separate lessons (e.g. having two different lessons focused on regulation of craving).

Our team opted to have the primary focus of the program be the multi-component lessons and to have the lessons completed one after another, instead of including intervening sessions that only included audio-guided meditations. This is somewhat different than standard MBTs that includes meditation homework in between sessions. However, we designed the program this way to facilitate optimal engagement in the context of a digital intervention without a high dose of in-person support. We surmised that multi-component lessons with varying topics and varying meditation practices would pique interest and promote engagement.

Finally, our team initially decided that we would include other resources in the program besides the lessons to facilitate additional and ongoing mindfulness practice, such as a library of mediations and brief exercises that could be done on-the-go. This way individuals could choose to do extra meditation practice while doing the lessons and also after completing the lessons.

Developing the initial sample lesson and initial full version

As shown in Figure 3, Phase 1 also included developing a sample digital lesson. When developing the sample lesson, our team strived to create a prototype with multiple features that participants could provide feedback on. This included inclusion of narrators, two audio-guided meditations (∼ 6-min each), didactic animated videos teaching the BOAT, quiz questions about the BOAT, and open-ended reflection question. The initial sample lesson was similar to the final first lesson. However, there were some key differences. As compared to the final first lesson described in Table 3, the initial version of the first lesson tested in phase 1 was slightly longer, only included one example client, and did not include instructions on how to continue using the app.

Table 3. Overview of the 15 core lessons in Mindful Journey.

Lesson number and name	Core learning objectives	Audio-guided mindfulness meditation practices
Getting Started	Define mindfulness Identify ways in which mindfulness in plays a role in recovery Practice guided mindfulness exercises Explain the steps of the BOAT (Breathe, Observe, Accept, Take a Moment) Describe how to use the Mindful Journey app	Body and breath practice BOAT practice
Responding with Awareness	Define automatic pilot Describe the role of automatic pilot in substance use Identify and define the three components of the triangle of awareness (thoughts, emotions, sensations) Practice mindfulness after thinking about challenging personal situation Explain how to use the BOAT in daily life	Brief BOAT BOAT with imaginal exposure to challenging situation
Coping with Cravings	Define craving Practice mindfulness while viewing images that may elicit craving Describe the similarity of cravings with ocean waves Explain how mindfulness can be applied to a situation that elicits craving	BOAT Exercise – Food Cravings (with visual display of food images) BOAT Exercise – Substance Cravings (with visual display of distal cue images)
Riding the Wave	Define a trigger and describe the roles of triggers in substance use Identify personal triggers Practice mindfulness after thinking about a situation that triggers craving for substances Create a plan for coping with triggers	BOAT with imaginal exposure to substance craving
Emotions as Visitors	Identify types of difficult emotions and describe how they play a role in substance use Practice mindfulness of one’s current emotions before and after listening to a poem about emotions as guests Explain how difficult emotions are similar to guests or visitors Describe the role of mindfulness, particularly the component of acceptance, in the process of coping with difficult emotions	Mindfulness of emotions with reading of the Guest House [23]
Seeing Thoughts as Thoughts	Define thoughts Differentiate thoughts from body sensations and emotions Describe the role of thoughts in the relapse cycle [23] Practice mindfulness of thoughts after hearing about an imaginary ambiguous situation Practice mindfulness of current thoughts while using the breath as an anchor for attention to return to	Walking Down the Street Exercise [23] Mindfulness of Thoughts
Practicing Self-Compassion	Describe the role of self-criticism in the relapse cycle Practice mindfulness while offering compassionate words to other people and oneself Explain how self-compassion can facilitate recovery Create a brief self-compassion reminder statement Practice mindfulness while mentally reciting one’s self-compassion reminder	Compassion Practice Self-Compassion reminder exercise
Exploring Your Needs	Describe types of needs that human beings have in life Explain how substance use is related to trying to meet one’s needs Practice mindfully reflecting on one’s own personal needs that may underlie craving and substance use Explain how the BOAT can be used as a tool to check-in with one’s needs in the moment	Exploring Your Needs Practice Brief BOAT
Taking Care of Your Body	Explain the importance of self-care in recovery, including taking care of one’s body Practice mindfulness while engaging in gentle movements and stretches, as well as while intentionally resting one’s body	Mindful Movement Mindful Rest
Being Curious	Explain the steps of the SOAK (Stop, Observe, Appreciate, Keep Curious) Describe the role of the SOAK in recovery. Define curiosity	SOAK practice Brief SOAK exercise with visual display of vivid photos.
Savoring Positive Experiences	Explain how the SOAK can be used in daily life Practice mindfulness after thinking about a pleasant situation Describe how the BOAT and the SOAK are both mindfulness tools one can use in different situations Creating a specific plan for engaging in a pleasurable activity over the next week and planning to practice the SOAK in this situation	Brief SOAK practice SOAK practice with imaginal exposure to pleasant situation
A Healthy Relationship with Positive Emotions	Explain how a relentless pursuit of positive emotions can be problematic and trigger substance use Describe the difference between letting positive emotions come and go versus chasing them or clinging to them Practice mindfulness after thinking about a situation that triggers intense desire for more positive emotions Explain how not allowing oneself to feel positive emotions can also be problematic Practice mindfully reflecting on ways of welcoming positive emotions and recognizing one is deserving of them	Letting Positive Emotions Come and Go (with imaginal exposure to situation triggering intense desire for more positive emotions) Welcoming positive emotions
Exploring New Positive Emotions	Identify other types of positive emotion words besides happy, such as love, interest, calm, amusement, confidence, or meaning Write about a specific activity that might bring about a one of these other types of positive emotions Practice mindfulness after writing and thinking about this specific activity Explain how connecting with a variety of positive emotions is important during recovery Explain how positive emotions and negative emotions can be felt at the same time	Brief SOAK Getting to Know Positive Emotions (includes SOAK)
Connecting with Your Values	Define values and describe their role in the recovery process Practice mindfully reflecting on personal values in key domains of life Explain the connection between mindfulness and making choices guided by our values Practice mindfully reflecting on values after thinking about a situation that triggers craving	Values Practice Brief BOAT with imaginal exposure to situation eliciting craving to use substances.
The Journey Ahead	Mindfully reflect on one’s recovery journey so far and the journey ahead Describe the impact of mindfulness practice on the brain Explain the importance of a regular mindfulness practice routine Describe how the BOAT and the SOAK are both mindfulness tools one can use in different situations Practice mindfulness after listening to the Serenity Prayer	Reflecting on Your Journey Serenity Prayer Exercise

The initial full version of the intervention tested in Phase 2 Wave 1 was similar to the final version. However, key differences included the inclusion of only one example client, the BOAT as they core acronym and not the SOAK, and more BOAT-related content. The Results section below clarifies how we incorporated participant feedback to progressively build and revise prototypes of the intervention.

Details on the final version of Mindful Journey

At the end of Phase 2, we finalized the app. Here we elaborate further on intervention features in the final version of Mindful Journey. For example, instead of using the SOBER Space acronym in MBRP [43] (i.e. Stop, Observe, Breath, Expand, Respond), we created a similar acronym called the ‘BOAT’ which stands for ‘Breath, Observe, Accept, Take a Moment’. This acronym does not have a word such as ‘sober’ that directly relates to substance use, and therefore may be able to be applied more broadly to negative emotion and physical pain. The BOAT also explicitly includes the step ‘Accept’ to emphasize the importance of adopting an attitude of nonjudgmental acceptance when paying attention to momentary experiences. The word BOAT was deemed useful as it ties into the metaphor of surfing or riding out waves of craving and difficult emotions, as well as the overarching ‘nautical’ theme of Mindful Journey. That is, recovery is discussed as a journey or voyage one might take on a boat—one in which the storms and strong waves can emerge, with one staying on the boat to ride out these storms and waves.

We ultimately chose to include an additional core acronym the ‘SOAK’ which stands for ‘Stop, Observe, Appreciate, Keep Curious’ and is intended as a tool for savoring pleasant experiences and cultivating a range of positive emotions, such as contentment, gratitude, calm, and meaning. The word SOAK was also deemed useful because it relates to ‘soaking’ in the rays of the sun while one embarks on a journey on a boat. Hence, the ‘Mindful Journey’ involves using the BOAT to ride the waves (i.e. cope with challenges) that arise during recovery and using the SOAK to ‘soak in’ the rays of the sun (i.e. savor pleasant experiences that don’t involve substances).

Table 3 provides details on each of the 15 lessons in the final version, including learning objectives and mindfulness practices in each lesson. The first lesson is intended to take about 40 to 45-minutes to complete and the rest about 30 min to complete. Table 2 provides key examples of intervention content designed to mobilize each of these putative mechanisms. Participants accessed the lesson on the app via the ‘My Journey’ button on the bottom navigation bar. When returning to his section they could pick up from where they left off, even if it was in the middle of lesson.

Videos include animated avatars of individuals. This includes two narrators/therapists (CRR and AH) who guide individuals through the program and serve as the therapists in video clips of therapist-client discussions. This also includes several example fictional clients, which are voiced by professional actors. The animated avatars’ mouths and bodies move as they speak. Drawing upon our team’s extensive experience conducting MBRP and ACT with individuals with SUD, we created fictional clients (2 core characters and 2 side characters) for the app, and we scripted all client testimonials and dialogues with therapists. Notably, the videos interweave a variety of other visuals, including moving animated illustrations and scenes, still graphics, and presentation of text, diagrams, charts. As noted above, all animations were created using the Powtoon animation creator program [63]. Figure 1 provides examples of animations that are in the video clips of Mindful Journey. Figure 2 provides example visuals of the how the app looks on a smartphone. Additional information on the final version of the app is provided in the supplementary materials.

Details on the phone coaching paired with mindful journey

For both waves in Phase 2, access to Mindful Journey was paired with brief manualized phone coaching that included: (1) once per week phone coaching calls, (2) 3–4 text reminders per week to engage in the app, (3) one additional 10-minute phone call per week and two additional texts per week for motivational support as needed for participants who did not access the program for 6 days in a row, and (4) additional technical support as needed by phone or text. The purpose of the phone coaching is to provide a minimal level of general motivational (e.g. praise for engagement and encouragement to keep using the app) and technical support (e.g. answering questions about how to use the app or navigate any technical issues that arise) [51] to optimize intervention engagement. The phone coaches were Bachelor’s-level research staff who were trained by CRR, a licensed clinical psychologist, during a single 1-h training session, and who also received ongoing supervision by CRR (20-minute meeting per week) during the study.

Importantly, the phone coaching did not involve any specific therapeutic intervention related to mindfulness training. The initial phone coaching call during week 1 was 30-minutes and involved the coach explaining their role as a coach, providing an overview of the app and encouraging daily use of the app, briefly inquiring about participants’ recent experiences with substance use and recovery, inquiring about what participants hope to get out of the program, and helping participants find a time to use the app. Subsequent weekly phone coaching calls were 5 to 15 min and involved a brief check-in about recent engagement with the app, praise for any engagement, validation of challenges in finding time to use the app, asking what a participant has learned so far, and encouraging continued use of the app. Text reminders included general reminders and encouraging statements to promote engagement in the app.

Participants were encouraged (by the narrators in the app and by their phone coaches) to prioritize completion of the 15 digital lessons and complete about 3 lesson per week, while also using these additional features as they see fit, such as for assistance using skills in the moment, and for engaging in additional formal mindfulness practice either while still completing the lessons or after completing all the lessons. Participants were able to go at a slower or faster pace when completing the lessons if needed or desired. Our team decided that 3 lessons per week (about 1 lesson every other day) was a reasonable pace and is consistent with MBT emphasis on repeated practice of skills over time and engaging in multiple days of practice per week.

Phase 1 Methods

The aim of the Phase 1 study was to test the acceptability and feasibility of a single introductory lesson of Mindful Journey, which was completed during a single study visit, and to obtain feedback that would inform further development.

Phase 1 participants

Participants (n = 5) were recruited from the Substance Abuse Treatment Unit at the Connecticut Mental Health Center in New Haven, Connecticut between April and September of 2021. This treatment program offers outpatient-based weekly group and/or individual therapy focusing on general support and motivation enhancement, 12-step principles and skills, psychoeducation about addiction and recovery, and relapse prevention skills (e.g. avoiding triggers). Participant inclusion criteria included being treatment for SUD at an outpatient SUD clinic, 18 years of age or older, and English-speaking. Eligibility criteria was broad for this phase to facilitate easier recruitment amidst challenges (e.g. the clinic providing mostly phone-based, not in-person treatment) during the COVID-19 pandemic. Participants were recruited via digital flyers emailed to clients and word-of-mouth from clinical staff. A brief phone-based screener was used to determine preliminary eligibility. Then participants were invited to attend the zoom-based study visit in which the study was described in more detail and eligibility was confirmed. The phase 1 study received ethical approval by the Institutional Review Board (IRB) at Yale School of Medicine (IRB Protocol# 2000028537). Informed consent was obtained for participation in the study and also for audio-recording the interview portion. Descriptive statistics for the demographic data of participants in Phase 1 are shown in Table 4.

Table 4. Demographic data for participants (n = 5) in Phase 1.

Measure
Age, mean (SD)	52 (10.95)
Gender, N (%)
Women	4 (80%)
Men	1 (20%)
Race, N (%)
African American	2 (40%)
Asian	0 (0%)
Multiracial	0 (0%)
Native American	0 (0%)
Pacific Islander	0 (0%)
White	3 (60%)
Other	0 (0%)
Ethnicity, N (%)
Hispanic/Latino	0 (0%)
Education, N (%)
Less than high school	0 (0%)
High school or GED	2 (40%)
Some college or  Associate degree	2 (40%)
Bachelor’s degree  or higher	1 (20%)
Employment, N (%)
Working	3 (60%)
Not Working	1 (20%)
Unable to Work	1 (20%)

Note. SD: standard deviation.

Phase 1 procedures

The sample introductory lesson was created using Articulate Rise, an online course builder program [64]. Participants completed the sample lesson on a laptop, tablet or smartphone during a zoom-based meeting with research staff. After completing the lesson participants took part in a qualitative interview about their experiences completing the lesson, and then completed questionnaire items, via a Qualtrics-based survey [65], about demographics and their experiences completing the lesson. Participants were compensated with a $50 gift card.

Phase 1 measures

Table 5 shows the core questions that were asked during the qualitative interview. The interview was semi-structured [66] and the interviewer asked additional questions as needed to further elicit, clarify, and understand participant’s experiences with the lesson. Age, gender, race/ethnicity, education level, and employment were assessed with a demographic questionnaire. Usability of the lesson was assessed with the System Usability Scale (SUS) [67], a validated 10-item measure (e.g. ‘I thought the program was easy to use’.) with responses ranging from 1 = strongly disagree to 5 = strongly disagree. Drawing upon the theoretical framework of acceptability [68], we created quantitative assessment items to measure various additional dimensions of treatment acceptability, including usability, overall satisfaction, appeal of visual content, understandability, desire to continue the program, satisfaction with the name of the program, overall helpfulness, helpfulness of the BOAT, and satisfaction with the pace and length of the program. These items (shown in Table 6) were each rated from 1= strongly disagree to 5 = strongly disagree.

Table 5. Qualitative feedback from Phase 1 single lesson testing.

Qualitative Queries	Themes
What did you learn?	Steps of the BOAT and BOAT as a recovery tool (5)
	Awareness of feelings and not reacting (3)
	Focusing on one’s breathing (3)
	Slowing down, being present, and focused (2)
	Accepting things as they are (2)
	Staying sober is just as important as getting sober (1)
What was most helpful?	BOAT was catchy and useful (4)
	Program was easy to follow and easy to navigate (4)
	Breaking down mindfulness into detailed steps (2)
	Observing clips of therapy sessions (1)
	Quiz questions helped for memorization (1)
	Brief length of each lesson component helped with engagement (1)
	Open-ended questions gave chance to reflect and relate to material (1)
	Liked how characters were moving and not just still images (1)
	Voices in meditations were soothing (1)
What was least helpful or difficult?/Feedback for improvement	Lesson was somewhat too long (2)
	Hard to concentrate during meditations (1)
	Practical challenges of meditating (1)
	Confused about whether to close eyes or not during meditations (1)
	Volume could be louder (1)
	Too much typing. Prefer to click answers (1)
	Hairstyles of characters out of date (1)

Note. The number in parentheses next to each theme indicates how many participants (out of 5) reported on a given theme in their qualitative feedback.

Table 6. Acceptability ratings for the phase 1 single lesson.

	Mean (SD)
Usability	4.20 (0.48)
I am satisfied with the program.	4.20 (0.44)
The program did a good job at keeping me engaged.	4.60 (0.54)
I thought the visual content of the program was appealing and high quality.	4.20 (0.83)
The material covered in the online program was clear and understandable.	4.75 (0.50)
I would want to continue using the program.	4.20 (1.30)
I think ‘Mindful Journey’ is a good name for the program.	4.00 (0.71)
I think the program would be helpful for me and my recovery.	4.60 (0.55)
The BOAT was a helpful tool for learning mindfulness.	4.6 (0.89)
The program was too long.	1.40 (0.55)
The pace of the program was too slow.	1.80 (1.30)
The pace of the program was too fast.	1.80 (0.45)

Note. Usability was assessed with the SUS, which included 10 items each rated 1= strongly disagree to 5 = strongly agree. The SUS score in the table is the average score per item. All other items were also rated from 1= strongly disagree to 5 = strongly agree.

Phase 1 analyses

We computed descriptive statistics (e.g. means and standard deviations) in Statistical Package for Social Sciences (SPSS) Version 28 for demographic data and quantitative ratings of acceptability dimensions. We conducted thematic analyses [69] to derive key themes emerging from the qualitative interview data. First, the audio-recorded interviews were transcribed. Then we generated initial codes and collated these codes into initial themes. Subsequently, we reviewed the themes again, discarded, combined, and redefined themes as needed, before finalizing the themes and selecting vivid example quotes to report for given themes, particularly in the case that the quote provided an enriched description of the theme. Themes were reviewed and discussed by two coders (CRR and MJ). Any disagreements were discussed among the two coders and resolved. Given the small sample size, we did not use a software program for qualitative analyses. Instead, we recorded how many participants endorsed a given theme.

Phase 2 methods

Phase 2 involved testing a full version of Mindful Journey with all 15 digital lessons. There were 2 waves of data collection. Briefly, for the first wave, participants completed 15 lesson on a web browser on their smartphones. For wave 2, participants completed a revised version of the 15 lessons on a smartphone app, which also had additional features (e.g. library of meditation practices). For both waves, participants had access to the program for 6 weeks and also received phone coaching.

Phase 2 participants

From August 2022 until February 2023, participants (n = 5) were recruited via flyers posted at an outpatient clinic that is part of the Midwestern Connecticut Council of Alcoholism (MCCA) in New Haven, Connecticut between. The outpatient clinic offers weekly group and/or individual therapy focusing on similar themes as the clinic described in Phase 1. Participant inclusion criteria included: (1) 18 years of age or older; (2) English-speaking; (3) meet DSM-5 criteria for a current substance use disorder; (4) are in an early phase of recovery as demonstrated by (a) completing 1 month or more of SUD treatment in the past 4 months; (b) reporting use of their primary substance in past 6 months; (c) are not currently enrolled in residential/inpatient treatment; (5) are willing and able to participate for the entire study period (6 weeks); (6) are willing to provide locator information for follow-up; and (7) own a working, WIFI-enabled smartphone. Exclusion criteria included: (1) current untreated psychotic symptoms; (2) suicidal ideation with intent; (3) homicidal ideation posing imminent danger to others; (4) pending legal case, imminent incarceration, or a planned move that results in inability to commit to procedures during the entire study period; and (5) participation in the Phase 1 study.

A brief phone-based screener was used to determine preliminary eligibility. Then participants were invited to attend the zoom-based study visit in which the study was described in more detail and eligibility was confirmed. The phase 2 study received ethical approval by the Institutional Review Board (IRB) at Yale School of Medicine (IRB Protocol# 2000028537). Informed consent was obtained for all participants. Descriptive statistics for the demographic and clinical data of participants in Phase 2 are shown in Table 7.

Table 7. Demographic and clinical data for participants (n = 5) in Phase 2.

Measure
Age, mean (SD)	36 (5.98)
Gender, N (%)
Women	3 (60%)
Men	2 (40%)
Race, N (%)
African American	3 (60%)
Asian	0 (0%)
Multiracial	0 (0%)
Native American	0 (0%)
Pacific Islander	0 (0%)
White	2 (40%)
Other	0 (0%)
Ethnicity, N (%)
Hispanic/Latino	0 (0%)
Education, N (%)
Less than high school	0 (0%)
High school or GED	3 (60%)
Some college or Associate degree	1 (20%)
Bachelor’s degree or higher	1 (20%)
Employment, N (%)
Working	1 (20%)
Not Working	3 (80%)
Unable to Work	0 (0%)
On public assistance	3 (60%)
Referred by criminal justice system	5 (100%)
Type of Substance Use Disorder
Alcohol Use Disorder	2 (40%)
Cocaine Use Disorder	2 (40%)
Opioid Use Disorder	1 (20%)
Hallucinogen Drug Use Disorder	1 (20%)
Marijuana Use Disorder	2 (40%)
Sedative Use Disorder	1 (20%)
Two or more SUD diagnoses	3 (60%)
Primary Substance Used
Alcohol	2 (40%)
Cocaine	1 (20%)
Heroin	1 (20%)
Phencyclidine (PCP)	1 (20%)
Lifetime number of inpatient treatment episodes for SUD	3 (3.93)
Lifetime number of outpatient treatment episodes for SUD	3.4 (2.61)
DSM-5 SUD Symptom Count (out of 11 criteria)	10.6 (0.54)

Note. SD: standard deviation.

Phase 2 procedures

Phase 2 involved testing all 15 digital lessons (∼ 30–45 min each). Participants first attended an in-person baseline visit with research staff. This visit involved completing Qualtrics-based assessment questionnaires, completing lesson 1, and receiving instructions about continuing to use Mindful Journey and being provided support from their phone coach (described further below) over the next 6 weeks. Participants then attended a post-testing visit after the 6-week testing period, which involved completing Qualtrics-based assessment questionnaires. Participants were compensated up to $150.

Phase 2 measures

To obtain qualitative feedback, the same set of questions (see Table 8) was used. However, for Phase 2, we did not conduct interviews, but instead obtained written feedback from text-based question prompts, which were administered after each of the 15 lessons (in reference to the lesson just completed) and at the post-testing visit (in reference to the entire program). Demographics (e.g. age, gender, race/ethnicity) and clinical history (e.g. treatment episodes) were assessed with a demographic and clinical history questionnaire. SUD diagnosis was assessed with the Mini International Neuropsychiatric Interview (MINI) [70].Single-item quantitative assessments (Table 9) were used to measure various dimensions of treatment acceptability. Of note, for Phase 2 we did not use the full SUS and instead used a single item to assess usability (i.e. ‘I thought the app was easy to use’). Additionally, for Phase 2 we used three items to assess acceptability of the phone coaching (see Table 9).

Table 8. Acceptability ratings for the phase 2 full program with all 15 lessons.

	Both Wave 1 and 2 (n = 4)	Wave 1 Only (n = 3)	Wave 2 Only (n = 1)
	Mean (SD)	Mean (SD)	Mean (SD)
I thought the app was easy to use.	4.5 (0.57)	4.67 (0.57)	4 (N/A)
The material covered in the program was clear and understandable.	4.25 (0.50)	4.33 (0.57)	4 (N/A)
The program did a good job at keeping me engaged.	4.5 (0.57)	4.67 (0.57)	4 (N/A)
I thought the visual content of the program was appealing and high quality.	4.25 (0.96)	4.33 (1.15)	4 (N/A)
I found the program helpful for me and my recovery.	4.25 (0.50)	4.33 (0.57)	4 (N/A)
I learned valuable skills after participating in the program.	4.25 (0.50)	4.33 (0.57)	4 (N/A)
I am confident in my ability to apply these skills in my daily life.	4.50 (0.57)	4.67 (0.57)	4 (N/A)
The phone coaching calls were helpful	4.00 (0.82)	4.00 (1.00)	4 (N/A)
The phone coaching calls motivated to use the app	4.00 (0.82)	4.00 (1.00)	4 (N/A)
My phone coach was supportive and encouraging	4.25 (0.50)	4.33 (0.57)	4 (N/A)

Note. For wave 2, one participant dropped out of the study (started an inpatient treatment program) and did not complete the post-testing survey. All items were rated from 1= strongly disagree to 5 = strongly agree.

Table 9. Qualitative feedback from full program testing in Phase 2.

Qualitative Queries	Themes and example participant quotes (italicized)
What did you learn?	Awareness and not reacting automatically^c (5)
	Accepting difficult emotions and craving^c (4)
	Noticing thoughts as thoughts^c (4)
	Being kind and compassionate towards oneself^c (3)
	Being curious and appreciating the little things^c (3)
	Importance of regular mindfulness practice^c (3)
	Reflecting on needs underlying desire for substances^c (2)
	Importance of self-care^c (2)
	Values-based action^c (2)
What was most helpful?	BOAT and how it breaks it down into steps is helpful^c (4)
	Easy to use the program^c (4)
	Engaging^c (3)
	Accessibility of the program^c (2)
	Good amount of variety in the content^c (2)
	How it was mostly videos/audio and not too much text^c (2)
	All the videos helped for staying engaged^c (2)
	Use the BOAT to stay afloat is a catchy phrase^a (1)
	SOAK as a new and different way to practice mindfulness^b (1)
	Mindful movement as a new way to practice mindfulness^a (1)
	Repetition/reminders of concepts and skills were helpful^a (1)
What was least helpful or difficult?/Feedback for improvement	Wanted more characters^a (2)
	Some of the videos with Maria were boring or unrelatable^a (1)
	Some videos with narrators are very short and could just be text^a (1)
	First lesson is too long^a (1)
	Want more movement with the animations^b (1)
	Concern about lack of motivation^b (1)
	Wanted more techniques and less BOAT^a (1)
	Content was repetitive towards the end^a (1)

Note. *Indicates that the theme was derived from data in wave 1 only, not wave 2. ^bIndicates that the theme was derived from data in wave 2 only, not wave 1. ^cIndicates that the theme was derived from both wave 1 and 2 data. The number in parentheses next to each theme indicates how many participants (out of 5) reported on a given theme in their qualitative feedback.

Phase 2 analyses

We computed descriptive statistics in SPSS Version 18 for demographic and clinical data and for quantitative ratings of acceptability dimensions. Using the written feedback provided after each lesson and at the end of the program, we conducted thematic analyses [69] using the same procedures described previously in the Phase 1 Analyses.

Results

Phase 1 intervention engagement

All 5 participants completed the single introductory lesson during the study visit in Phase 1.

Phase 1 acceptability ratings

As shown in Table 6, ratings of usability, overall satisfaction, how engaging the program was, appeal of visual content, understandability, desire to continue the program, satisfaction with the name of the program, overall helpfulness, and helpfulness of the BOAT were all at or above a mean score of 4, which corresponded with ‘agree’ on the 5-point response scale. Ratings about whether the program was too long, too slow, or too fast were all at a mean of 1.8 or below, which was below the score of 2 corresponding with ‘disagree’.

Phase 1 qualitative feedback

Table 5 provides a summary of the qualitative feedback from Phase 1, including example quotes for themes. Overall, the qualitative feedback indicated understanding, satisfaction, and acceptance of the intervention. Participants reported that they learned skills involving both awareness and acceptance of their experiences. They reported that they understood and were able to remember the steps of the BOAT, and they particularly appreciated how the BOAT broke down mindfulness into detailed steps to follow. One participant said she liked the clips of therapy sessions because it was helpful to see another person’s perspective. Overall, participants found the program easy to use.

However, participants did note some challenges while using the program. For example, one participant said it was hard to concentrate during the meditation, multiple participants said the lesson was too long, and another participant said there was too much typing and clicking answers would be preferable.

All feedback, including critical feedback, was considered when refining Mindful Journey for Phase 2. As examples, prior to testing in Phase 2, we shortened the length of lesson 1 and included less open-ended questions and more mixed open-/closed-ended question such asking an open-ended question (‘What did you notice?) and then providing many examples of potential answers (e.g. sensations of tingling, my mind wandering, sadness, boredom) that one can select with the option of selecting as many that apply, as well as an option to additionally write their response in their own words. Additionally, given participant satisfaction with the BOAT, we included repeated review and practice of the BOAT across multiple lessons for Phase 2.

Phase 2 intervention engagement

For wave 1 (n = 3), all participants completed all 15 lessons. The first enrolled participant in wave 1 completed all the lessons over 20 days and completed 4 phone coaching calls. The second participant completed all the lessons in 11 days and completed 2 phone coaching calls. The third participant completed all the lessons in 16 days and completed 3 phone coaching calls. Thus, in wave 1 the average timespan for completing lessons was 16 days and the average by number of phone coaching calls completed was 3.

For wave 2 (n = 2), one participant completed 4 lessons and 2 phone coaching calls over 20 days, but then dropped out of the study and was unavailable to complete assessments at the post-testing visit. The other participant completed all 15 lessons and 3 phone coaching calls over 33 days. Additionally, this participant practiced the ‘Brief BOAT’ in the Cope Now section of the app two times total. Neither participant in wave 2 completed any of the meditations in the All Practices section of the app.

Across both waves 1 and 2, the average number of lessons completed per day was 0.74 week lesson per day, and thus 5.18 lessons per week. Of note, participants had access to the lessons for 6 weeks, but completed it in less than the time allotted.

Phase 2 acceptability ratings

When rating the Mindful Journey program as whole after the 6-week testing period, participants reported ratings with a mean score of at or above 4 (corresponding with ‘agree’) for all acceptability dimensions, including usability, understandability, how engaging the program was, appeal of visual content, overall helpfulness, as well as the degree to which they learned valuable skills and were confident in applying these skills (see Table 9). Additionally, participants reported ratings with a mean score at or above 4 for items assessing acceptability of the phone coaching, including items about overall helpfulness, the degree to which phone coaching calls motivated them to use the app, and supportiveness of the phone coach. All acceptability ratings were at a score of 4 or above when averaging across wave 1 and 2, and when considering each wave separately.

Phase 2 qualitative feedback

Table 9 provides a summary of the qualitative feedback from Phase 2, including example quotes for themes. Similar to Phase 1, the qualitative feedback in Phase 2 generally indicated understanding, satisfaction, and acceptance of the intervention. Participants reported that they learned skills and concepts related to many topics, including awareness and not reacting automatically, accepting emotions and craving, noticing thoughts as thoughts, self-compassion, self-care, being present and appreciating daily activities, and values-based action. Similar to Phase 1, participants in Phase 2 were particularly satisfied with the BOAT as a tool to apply in daily life. However, one participant wanted more techniques and less of the BOAT. This participant was in wave 1, which included the version of Mindful Journey that only had the BOAT, not the SOAK. Based in part on the feedback of this participant, we added the SOAK to the refined program in wave 2. Interestingly, the participant in wave 2 who completed the program with the SOAK reported that he was satisfied with the SOAK and appreciated how it was a new and different way to practice mindfulness.

Many participants said they liked how the program as mostly videos and audio, and not too much text. Participants also appreciated that there was adequate repetition of key concepts and skills, while also finding that the program had enough variety in content. However, one participant in wave 1 (which again included only the BOAT not the SOAK) found the content to be repetitive towards the end. This same participant said he found some of the videos of Maria boring and unrelatable and that he wanted more characters. Accordingly, when refining the program for wave 2, we added additional characters, including a core male character, and we included clips of group-based therapy instead of only individual therapy. Moreover, we abbreviated some of the repeated BOAT content towards the end of the program.

One participant in wave 1 said the first lesson was too long and that some of the videos with the narrators talking were very short and could just be text. Hence, we shortened lesson 1 even more and removed some of the very short narrator clips and replaced them with either text or brief comments embedded at the beginning of an audio-guided meditation.

One participant in wave 2 noted that she was worried about having enough motivation to use the skills on her own. Based on this feedback, we added additional question prompts throughout the lesson that focused on helping individuals identify personal motivations for applying skills in daily life.

Finally, it is notable that participants wanted more movement with the animations in the video clips.

Discussion

This study aimed to develop and evaluate the preliminary acceptability of Mindful Journey, an adjunctive digital mindfulness-based intervention for promoting long-term recovery among individuals with SUDs. Mindful Journey is primarily based on Mindfulness-Based Relapse Prevention (MBRP) [23] and consists of 15 sequential digital lessons. In phase 1, we tested the introductory digital lesson. In phase 2, we tested a full version of the program with all 15 lessons. Overall, quantitative and qualitative results from both phase 1 and 2 indicated high acceptability of the intervention.

In phase 1, all 5 participants completed the 45-minute introductory lesson. Quantitative acceptability ratings reported after completing the lesson were high for all dimensions assessed, including usability, overall satisfaction, how engaging the program was, appeal of visual content, understandability, desire to continue the program, satisfaction with the name of the program, overall helpfulness, helpfulness of the BOAT, and satisfaction with the length and pace of the program.

Additionally, in phase 1 we interviewed participants after they completed the introductory lesson. They reported that they found the intervention easy to use, learned both awareness- and acceptance-based skills, appreciated the BOAT and how it broke down mindfulness into clear steps, and liked hearing new perspectives in the video clips of the characters having a discussion in a therapy session. Participants also reported some challenges and suggestions for improvement. Even though quantitative ratings about the length of the program indicated high acceptability, in the qualitative interview several participants said the lesson was somewhat too long. Based on the qualitative feedback from phase 1, we revised lesson 1 and further developed the full program which consisted of 15 lessons. For example, we shortened the length of lesson 1, included less open-ended questions, and incorporated the BOAT across multiple lessons.

For Phase 2, in both wave 1 and 2, quantitative acceptability ratings reported at the post-testing visit were high for all dimensions, including usability, understandability, how engaging the program was, appeal of visual content, overall helpfulness, the degree to which they learned valuable skills and were confident in applying these skills, overall helpfulness of the phone coach, the degree to which phone coaching calls motivated them to use the app, and supportiveness of the phone coach. Additionally, participants in Phase 2 provided qualitative written feedback after each lesson and at the post-testing visit. Overall, written feedback across both waves indicated understanding, satisfaction, and acceptance of the intervention. In both waves, participants reported that they learned skills related to key intervention topics, such as not reacting automatically, accepting emotions and craving, and noticing thoughts as thoughts. Further, in both waves, participants particularly appreciated the BOAT tool and how the program was mostly video-based.

Findings from this study are consistent with research demonstrating that digital delivery is a viable option for mindfulness-based interventions for psychiatric disorders, such as depression and anxiety disorders [50]. Indeed, mindfulness-based interventions may be suitable for digital delivery because mindfulness meditation—the primary therapeutic component—can be offered via audio-guided exercises. Moreover, mindfulness-based concepts (e.g. accepting craving) can be conveyed via imagery and metaphors (e.g. a boat riding a wave), which can be visually displayed in animated videos.

The relatively high rates of intervention engagement in this study may be attributable to the weekly phone coaching support that was provided over the 6-week testing period. This aligns with research showing that low-intensity clinician- or coach-based support improves adherence to digital interventions [48]. Of note, all participants completed less than the six available weekly phone coaching calls. This was likely because most participants completed the 15 lessons well before the end of the 6-week period.

Interestingly, participants generally completed the lessons at a faster pace than the pace that was suggested to them by the app and their coaches. Whereas the suggested pace was 3 lessons per week, participants on average completed the lessons at a pace of 5.18 lessons per week. This finding suggests that participants may be willing and able to complete lessons at a relatively fast pace, and perhaps may be willing to complete more than 15 lessons. It is also possible that participants sped through the lessons without fully comprehending the material. However, it is important to note that participants could not advance to the next section of a lesson unless they answered quiz questions correctly.

Prior studies of digital interventions for SUDs have had participants complete digital lessons in a private, quiet space at a clinic [60,71]. In this study, however, participants completed the digital lessons outside of the clinic environment (e.g. at home, work, on the bus) with remote phone-based voice and text support from a coach. The high rates of intervention engagement in this study suggest that completing lessons outside of the clinic with remote support is a feasible model for delivering digital interventions to individuals in outpatient treatment for SUDs. This model may be valuable in clinical practice because it does not rely on participants being at a clinic in-person, which may be challenging for some individuals because of constraints related to available time outside of work and transportation options.

For the final version of Mindful Journey tested in wave 2 of phase 2, the program was delivered as part of a smartphone app, which in addition to the lesson had other features including a Cope Now section with brief audio- and text-based on-the-go tools and an All Practices section with a library of meditations exercises. Overall, participants used these additional features infrequently. Future versions of the app may need to include push notifications to remind participants to use these features. Regarding potential improvements to the app, a future version might also include more movement with the animations, as participants in wave 1 and 2 of phase 2 expressed a desire for this in the qualitative feedback.

This study included a small sample of participants, which may limit the generalizability of findings. The sample only included individuals identifying as African American or White, and not other races/ethnicities. One limitation is that in Phase 1 we did not collect details on the type of SUDs individuals had. In Phase 2, individuals met criteria for a range of different SUD types. However, some SUD diagnoses (opioid, hallucinogen, and sedative) were only present for one participant. Future research is needed to examine whether Mindful Journey is feasible, acceptable, and efficacious among individuals with different SUD types. Importantly, given the small sample size in this study we did not test the effect of the intervention on putative mechanisms and recovery outcomes. The current data only provide evidence of preliminary acceptability and future feasibility testing with a larger and more diverse sample is needed.

Conclusions

Altogether findings provide preliminary evidence of the acceptability of Mindful Journey, an adjunctive digital mindfulness-based intervention for reducing relapse among individuals with SUDs. Mindful Journey has the potential to expand the reach of MBTs for SUD because individuals with SUD may find it more accessible to engage in an app rather than attend in-person therapy groups. However, apps such as Mindful Journey may also have limitations. For example, without direct in-person support, participants may feel less accountability to consistently engage with the intervention The next step is to conduct a pilot randomized controlled trial of Mindful Journey with a larger sample. This study will provide further data on the feasibility and acceptability of Mindful Journey. Additionally, future studies are needed to test the effect of Mindful Journey on mechanisms and outcomes.

Author contributions

CRR and HK played a lead role in all aspects of the study, including conceptualization and study design, funding acquisition, development of the digital intervention, data collection, data analysis, and writing the paper. BK and KMC played a leading role in conceptualization and study design and funding acquisition, and a supporting role in development of the digital intervention and writing the paper. MS, MK, ES, MJ, CJM, RP, JF, AH, DB, LF played supporting roles in conceptualization and study design, development of the digital intervention, and writing the paper. JB played supportive roles in all aspects of the study, including the grant application, intervention design, providing intervention content, study design, interpretation of study results, and writing the paper. SB and KW played leading roles in conceptualization and study design, funding acquisition, development of the digital intervention, and supporting roles in writing the paper.

Acknowledgements

We would like to acknowledge Iya Hanik, Daniel Wang, and Kaiqi Zhang for their assistance developing the animated videos for the program.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

Data from this study are available by request from this first author CRR.

Additional information

Funding

CRR is supported by a grant from the National Center for Complementary and Integrative Health [K23AT011342].