A Cardiovascular Health and Wellness Mobile Health Intervention Among Church-Going African Americans: Formative Evaluation of the FAITH! App

Background In light of the scarcity of culturally tailored mobile health (mHealth) lifestyle interventions for African Americans, we designed and pilot tested the Fostering African-American Improvement in Total Health (FAITH!) App in a community-based participatory research partnership with African American churches to promote cardiovascular health and wellness in this population. Objective This report presents the results of a formative evaluation of the FAITH! App from participants in an intervention pilot study. Methods We included 2 semistructured focus groups (n=4 and n=5) to explore participants’ views on app functionality, utility, and satisfaction as well as its impact on healthy lifestyle change. Sessions were audio-recorded and transcribed verbatim, and qualitative data were analyzed by using general inductive analysis to generate themes. Results In total, 6 overarching themes emerged among the 9 participants: overall impression, content usefulness, formatting, implementation, impact, and suggestions for improvement. Underpinning the themes was a high level of agreement that the intervention facilitated healthy behavioral change through cultural tailoring, multimedia education modules, and social networking. Suggestions for improvement were streamlining the app self-monitoring features, prompts to encourage app use, and personalization based on individuals’ cardiovascular risk. Conclusions This formative evaluation found that the FAITH! App had high reported satisfaction and impact on the health-promoting behaviors of African Americans, thereby improving their overall cardiovascular health. Further development and testing of the app among African Americans is warranted. Trial Registration ClinicalTrials.gov NCT03084822; https://clinicaltrials.gov/ct2/show/NCT03084822.


Introduction
Ideal cardiovascular (CV) health has been defined by the American Heart Association (AHA) as the achievement of optimal levels of 7 key components, including health-promoting behaviors (optimal diet and physical activity [PA], achieving normal weight, and abstaining from smoking) and controlling health factors (blood pressure, cholesterol, and blood glucose). Each of these components is independently associated with lower CV disease risk in a stepwise fashion; however, there are striking racial and ethnic disparities in ideal CV health [1]. Compared with White Americans, African Americans (AAs) have a lower prevalence of ≥5 ideal components (11.8% vs 19.2%) [2] and an 82% lower likelihood of achieving ≥5 of the 7 ideal components [3]. Furthermore, it has been estimated that the components with the greatest potential for improvement in AAs are health behaviors, including diet, PA, and weight management [4]. Even modest shifts in the distribution toward improved CV health in the AA population could result in appreciable reductions in incident CV disease in this group. Thus, innovative, individual, and population-based interventions to promote ideal CV health through lifestyle changes among AAs are desperately needed.
Mobile health (mHealth) [5] lifestyle interventions hold potential for diffusing CV health promotion in AA communities. AAs are embracing mobile technologies with rapid smartphone use expansion and frequent internet searches for CV disease-related health information [6] and positively view mHealth interventions as viable strategies to improve health outcomes [7,8]. However, only a handful of studies have tested mHealth interventions to improve CV health in AAs, with many studies focusing only on health behaviors (eg, diet, PA, weight loss) or single chronic medical conditions (eg, diabetes, hypertension, heart failure) [9][10][11][12][13][14][15][16][17].
Emboldened by the AHA 2020 Impact Goals to improve the overall CV health of AAs, we co-designed a culturally tailored digital app, Fostering African-American Improvement in Total Health App (FAITH! App), to promote CV health within the AA faith community using an iterative participatory approach within an established academic-community partnership [18,19]. The intervention is grounded in behavioral theoretical frameworks, including principles from the health belief model [20], social cognitive theory [21], and the community mobilization model [22,23]. At the heart of the intervention are multimedia education modules that were guided in design by the health belief model, which incorporates an individual's perceived susceptibility, benefits, and barriers to predict behavior change. The social cognitive theory model, or learning through a collective agency to influence health behaviors, was incorporated in specific app features (eg, group sharing board, testimonials on healthy lifestyle) as well as our inclusion of an established social construct within the AA community, the Black Church [24]. To this end, we also embraced the community mobilization model by actively engaging in a strategic community partnership while leveraging the norms, values, and resources it embodied to develop and implement an intervention [25].
To our knowledge, our parent study is among the first to combine a culturally tailored and app-based lifestyle intervention to promote ideal CV health within the context of a primary CV disease prevention program [18,19,26]. Participants within our pilot study achieved improvements in key CV health behaviors and factors and reported overall high acceptability and satisfaction of the intervention [18,26]. To address the dearth of literature integrating formative processes to design and evaluate culturally tailored mHealth interventions, we used a qualitative approach to further explore participant experiences of taking part in the intervention.
This program evaluation aims to gain insight into the impact of an mHealth lifestyle intervention (FAITH! App) on the CV health behaviors of AAs participating in a quasiexperimental behavioral intervention pilot study. We sought to include participants' perspectives to evaluate the intervention with the goal of integrating this information in future iterations of our mHealth intervention to more effectively influence CV health in this population.

Methods
The parent study was approved by the Mayo Clinic Institutional Review Board and registered (ClinicalTrials.gov [NCT03084822]), and participants provided written informed consent before participation.

Research Design and Participant Recruitment
The parent study comprised 3 community-driven phases: (1) app design with the AA community through formative app development [18], (2) app pilot testing [26], and (3) app evaluation using quantitative and qualitative research methods. This analysis (phase 3) evaluates qualitative data on participants' perceptions of app pilot testing, who were subsequently recruited to participate in the evaluation component of our study. Details on the parent study rationale, recruitment procedures, and participant inclusion/exclusion criteria for the overarching parent study have been described [27]. Briefly, we collaborated with 5 predominately AA churches in Rochester and Minneapolis-St Paul, Minnesota, using a community-based participatory research (CBPR) approach to co-design a CV health and wellness digital app-based program (ie, the FAITH! App) [18]. A total of 50 AA church parishioners were enrolled into a single-group pilot study to follow a 10-week intervention centered on the FAITH! App. Its components included 10 core multimedia education modules delivered by health professionals on CV health, interactive diet and PA self-monitoring, and social networking through a group sharing board. Participants completed health assessments, including self-administered electronic surveys of sociodemographic and health behavior information as well as physical examinations and laboratory studies of biometrics at baseline and 28 weeks postintervention. Following the final health assessment, participants from the pilot study were invited to participate in the postintervention focus groups by email. Participants involved in the co-design of the FAITH! App were not invited to participate in the focus groups. There was a 32% response rate (16 out of 50 pilot study participants) to the email invitation. Ultimately, 9 participants were able to participate on the most convenient dates and times chosen by a polling of the interested invitees. Participants received a US $50 cash card for participation in the focus groups. , an assessment of an individual's perception of their ability to understand and apply electronic health information) [28][29][30], were derived from baseline electronic surveys completed at the time of pilot study enrollment.

Data Collection
We developed a comprehensive moderator guide inclusive of neutrally worded, open-ended questions to facilitate open discussion and to obtain feedback on the FAITH! App intervention including the following core topics: app satisfaction, dislikes, and suggestions for improvement. The moderator guide was developed from a literature review [31], our participatory intervention design process [18], informal discussions with participants and church partners as well as quantitative evaluations of the FAITH! App. The primary moderator (CS) had substantial experience conducting focus groups among racial and ethnic minority populations and previously collaborated with the study team during the formative design of the FAITH! App with a similar group of AAs [18]. The study principal investigator (LB) was intentionally not present during the focus groups to minimize social desirability bias and coercion among the study participants. However, the moderator and study principal investigator met before each focus group to review specific strategies to encourage group discussion and culturally tailor questions to place the group at ease to share their valued opinions. Each focus group was 90 min and audio-recorded. The focus group in Rochester was held in a conference room in an outpatient clinical practice in the evening. The Minneapolis-St Paul focus group took place in the evening in a conference room at a community center. Each focus group commenced with a friendly ice-breaker to promote sharing among the group, "What is your favorite thing about Minnesota?" Following each focus group, the moderator compiled summary analysis notes of the most salient participant responses, which were then reviewed by the study principal investigator. All sessions were subsequently transcribed verbatim by an experienced transcriptionist.

Data Analysis
For qualitative analyses, the first author (LB) and a trained qualitative methods specialist (AK) first read the transcripts several times with reference to the focus group moderator guide. The 2 coders collaboratively developed a code book in an iterative process to better organize and categorize the data in accordance with the open coding method, a process that enhances the probability of garnering novel insights [32]. In the code book, each code was described with a concrete definition, and at least one example quote from the data was included. QSR NVivo software version 10 was utilized to organize and manage transcribed data from the focus group sessions. General inductive analysis was used as it aligned with the exploratory and formative nature of the study in the following ways: (1) it condenses raw textual data into a more succinct summary; (2) it links the evaluation (in our case of an intervention) to the summary findings from the raw data; and (3) it cultivates a framework of the underlying structure of experiences or processes evident in the raw data [33]. Two coders (LB and AK) coded all transcripts independently and assigned codes for subsequent categorization around salient themes. To mitigate interpretative biases and to maintain the consistency of coding, the 2 coders met regularly and discussed all coded data until they reached a consensus. Whenever the 2 coders had difficulty assigning a common code, a third team member (CP) assisted with resolving discrepancies or assigned new codes to ensure consensus. Through this collaborative process, the team attained very good agreement on coding [34]. While ensuring intercoder reliability, the 2 coders discussed the main themes in detail, and provisional subthemes also emerged from the coded data [35]. Finally, to further enhance reliability and as a part of our CBPR approach to formative evaluation [18], we reviewed the compiled themes/subthemes with community church partners for member checking and feedback on the themes. We provided the 4 church partners with a preliminary summary of the main themes and subthemes for their review. We then organized a 1-hour teleconference to discuss the preliminary summary (community member checking) among the group with the study principal investigator (LB) and qualitative research expert (AK) moderating. The church partners were in overall agreement with the themes/subthemes and felt that they were in concordance with the general feedback they informally received from study participants at their respective churches. This triangulated process of integrating input from participants, church partners, and the study team provided an additional layer of corroboration to the finalized main themes and subthemes [36,37]. Quotations from participants were extracted from transcripts that best represented participants' experiences. Descriptive analyses were completed for all sociodemographic data with the calculation of frequencies and proportions. Participants were categorized by app frequency of use as follows: high frequency users (viewing at least 50% of the 10 education modules, tracking at least weekly diet/PA, and posting on the sharing board at least once per month) and low frequency users (those not meeting high frequency usage patterns). All quantitative analyses (demographic data) were conducted in 2017 with SAS version 9.4 (SAS Institute Inc).
The average eHealth literacy score was 31.7 (high range by eHEALS ≥26); 7 of 9 participants were classified as high frequency app users. Overall, participants in the Rochester and Minneapolis-St Paul focus groups had similar responses as the vast majority of the generated codes were mentioned by both groups. Thus, data were pooled for both focus groups. Findings of the focus group discussions were grouped into 6 main themes that were divided into corresponding subthemes: (1) overall impression, (2) content usefulness, (3) formatting, (4) implementation, (5) impact, and (6) suggestions for improvement (Table 2).
We present the results by starting with the participants' global perceptions of the FAITH! App that encompasses their overall acceptability and utility of the intervention. We then examine more granular insights regarding app content and delivery modalities, which could inform future iterations of the intervention. Next, we move to the app's impact by highlighting its influence at the individual, interpersonal, and community levels. Finally, we conclude with feedback for app enhancement.

Theme 1: Overall Impression
The focus groups opened with questions about the impressions of the app as a whole. The most common positive perceptions identified by the participants were related to health promotion through the easily accessible health information delivered by the education modules. The most common source of challenges with app use came from technical difficulties with the diet and PA self-monitoring feature.

Promotion of Healthy Behaviors by Self-Monitoring
Participants saw great value of the app self-monitoring features as they compelled them to keep track of their diet and PA patterns. One participant mentioned how this feature provided accountability to meet personal goals to eat healthier and remain active:

Accessible and Instructional Health Content
Most participants reported that the app provided easily accessible and useful information and found the pictorial images and visuals to explain the health information helpful:

Influence on Healthy Lifestyle Adoption of the Church
A fundamental concept from several participants was that the genuine commitment to improving the health and well-being of the study participants conveyed by the study team inspired a commitment from the church congregations to focus on adopting healthy lifestyles. The app intervention as a whole seemed to awaken the church congregations to create an endurable culture of healthy living even beyond the completion of the study:

Technical Difficulties With Self-Monitoring
Participants provided shared challenges with technical difficulties with the app self-monitoring feature, specifically troubleshooting to enter their diet and PA information.

Theme 2: Content Usefulness
Naturally, discussions transitioned to participant perceptions of the core app features. Overall, the participants found the features useful to better understand the impact of CV risk factors on CV physiology and the development of CV disease. They also mentioned that the app provided them with tools to improve their CV health.

Education Modules: General
Thought-Provoking and Self-Reflective One participant felt that the app offered a new awareness of their own health and wellness through the information provided:

Easy to Comprehend Health Information
The participants felt that the information was presented at an optimum level for understanding across different learning styles. They also appreciated that the speakers avoided extensive use of medical jargon but instead relayed information in layman's terms: .

Convenient and Succinct
Participants found the video series format to be a convenient and succinct way to view and access health information both independently and with their families. Several shared how they would watch the videos from a variety of locations (eg, home, work) and revisit the modules for reinforcement. Keeping the videos brief and concise was also viewed favorably. Another participant enjoyed having the summary video at the end of each module to tie together the key concepts: .

Complementary to Video Series
Participants felt that the premodule and postmodule self-assessments (quizzes) were a great addition to the modules and complementary to the video series. They unanimously recommended that they should remain mandatory in order to mark completion of the education module as they were beneficial self-assessment tools. The quizzes also provided an infrastructure to remain on track with the education modules:

Time-Consuming
Although the content was viewed as useful and relevant, some participants found review of this information to be time-consuming and too much to digest as an addition to the other module features: .

Technical Difficulties With Access
Participants noted that access to the brochure content was cumbersome as the content was not housed within the app itself and required redirection to PDF files within a separate screen.

Inspired Action Toward Healthy Eating
Several participants enjoyed heart healthy recipes on the app and shared experiences at the individual and church level of how they incited healthy eating. One participant shared how his family modified their traditional holiday practices toward healthy dietary change through smaller portion sizes: .

Motivational to Healthy Lifestyle Change
Participants used the sharing board posts to connect with others and found the posts to be very encouraging and motivating to maintain healthy behaviors.

Support Network
Participants also found the sharing board to be a support network to better support others through personal struggles. One participant revealed empathy for another participant through prayer:

Preferred Heart Disease Survivors or Those Making Healthy Lifestyle Changes
Participants appreciated the testimonials from the church pastors and past FAITH! participants but initially thought they would feature those with an experience of CV disease or those who have made significant lifestyle changes. They felt that hearing stories from them would inspire them to reflect on their own risk for CV disease. One participant shared: .

Fostered Personal Accountability Toward a Healthy Lifestyle
Overall, participants found the self-monitoring feature useful as it fostered personal accountability toward a healthy lifestyle. One participant shared how the feature helped to "keep track" of diet and PA patterns:

Fulfilled Expectations for Continued Engagement
Overall, the participants enjoyed using the app, with several directly expressing that they would want to continue with the digital formatting of the intervention:

We got enough but we want more! [Focus group 2, participant]
Participants also felt that the interactive multimedia delivery modality was appropriate with the goal of increasing awareness of healthy lifestyle change:

Diversity of Health Care Professionals Within the Education Modules
Participants also took notice of the diversity of health care professionals included within the education module videos that included multidisciplinary individuals (eg, cardiologists, endocrinologists, nurses, dieticians).

Importance of Linking Faith to Health for the AA Community
The participants commended the study team for incorporating biblical scriptures and spiritual messaging within the app as religious involvement is of high importance to AAs and showed humility to this prioritized AA faith community. They appreciated the emphasis on connecting spiritual and physical health for healthy lifestyle changes, as described by this participant:

CV Health Disparities Affecting AAs
The vast majority of participants acknowledged the importance of placing emphasis within the modules on the CV health disparities burden among AAs. They felt that the positive messaging strategies were delivered in an appropriate manner to not place offense or blame on this group but to inform them of their elevated risk for CV disease. This was encompassed in a statement from one participant with many nodding in agreement: ...not pushing one culture over another but still letting you know that this culture is more at risk, and I thought they did that, for lack of a better word, very tactfully, very swoopy where you didn't feel offended or saying that this was speaking against your culture-you people get this and you people are targets for this. I didn't feel like it was being pointed out, but I also understood that you were being addressed. [Focus group 2, participant]

Implications of Visual Representations of AAs
In general, the group felt that the selected photographs of AAs included in the app were accurate reflections of the AA community's experience. This was of significance for the group as this influenced their acceptability of the app: A concern surfaced from a male participant regarding the placement of wording over one of the AA men depicted on the home page visual. The man had a darker skin tone/complexion than the other individuals, and it was felt that this was offensive and, in a sense, reinforced the societal discrimination against and ostracism toward AA men. An AA man in the group exclaimed:

Facilitators for Use
Participants identified several facilitators of their engagement with the app features. They placed emphasis on the education modules and their utility for app use.

Positive Messaging Supported Engagement
A consistent topic among participants was that the intentional use of positive and encouraging messaging throughout the education modules supported their desire to view the entire series: They gave us positives even though we're going through situations, they said "if you just make one change", so it was encouraging. I liked how they encouraged us to make a small change. [Focus group 2, participant]

Focus on Benefits of Healthy Lifestyles
The app as a whole increased awareness about and facilitated healthy behaviors among the participants. Participants reported that having this overarching focus infused into the app motivated them to become more physically active and to make healthier food choices:

Simple Navigation
Furthermore, a simple navigation layout on the app including a homepage and clear tabs (for tracking and the sharing board) was key to using the app features. Participants noted that it was easy to follow the education module curriculum on a weekly scheduled basis:

Visual Display of the Education Module Progress
This was further reinforced by the inclusion of checkmarks to show completion of the education modules as detailed by one participant: ... I also liked the ability to say that you've completed the module by selecting "done". Once you clicked "done," there was a checkmark there, so it already told you, yeah, you did it...you don't remember-did I already look at that (laughs)? [Focus group 1, participant]

Education Module Variety of Activities Appealed to Differing Learning Styles
As a reflection of the initial development of the education module content, participants appreciated the variety of activities within the modules (videos and quizzes) and that they appealed to different learning styles: The education modules and their central feature, videos were viewed as essential app components, as they facilitated learning by the user. One participant shared a positive comment about the succinct "to the point" nature of the modules: Short and sweet but direct to the point. I thought the videos really made the app...to be honest. I mean I thought that because I'm a visual person; so having someone talk to you without you having to read a book of information was convenient. And they were easier than I was thinking they were going to be (laughs). [Focus group 1, participant]

Overview of CV Risk Factors at the Start of Education Modules
The order of the education modules was of importance as participants felt that having the "Introduction to Risk Factors" module first among the series of education modules was key to setting the stage for the purpose of the intervention, which in turn promoted continued engagement:

Consolidation of CV Health Information Through Mobile Technology
It was also advantageous to consolidate CV health information from a trusted source on the app over having to search the internet, which was seemingly overwhelming: ...pooling all the information together in one spot because I know that you could probably search the web and find all this information. I thought that was just awesome because they talked about so many aspects of things that can cause heart problems and it was just all in one spot versus if you have to go out and find this information, you might be in several different websites. So I thought that was great.

Barriers to Use
Participants identified several app features that presented barriers to app usage.

Cumbersome Data Entry and Log-In Process
One of the most commonly mentioned barriers was the cumbersome nature of the self-monitoring/tracking feature, which required manual data entry of daily PA and diet patterns by the user. The app's lack of autosave for the tracking feature presented the greatest challenge. Participants also commented on the multi-step log-in process required by the user to access the app. Although this was intended as a one-time requirement for users at the first long-in, oftentimes, users were required to repeat this process as a security measure or when the user was idle or using different internet networks.

Inability to Download Apps on Other Personal Mobile Devices
Participants were provided with tablet devices for use throughout the intervention phase of the study, which had a direct link to the web-based app. As participants enjoyed having the device available through mobile technology, they also found the inability to download the app on other personal mobile devices as a barrier: The downside was I couldn't really download anything cuz it was on the app, the iPad that you guys gave us...cuz I wasn't able to keep the material or print it off because it wasn't available for me on my own personal iPad. [Focus group 2, participant]

Theme 5: App Impact
The impact of the app on participants' personal healthy lifestyle change and community was intensively discussed.

Positive Changes to Dietary Patterns
Accordingly, participants mentioned specific manners by which the app incited positive changes to their individual dietary patterns: I think about this even today the fruit and the vegetable intake, I really am more conscientious of that and even when I go to the store, I find myself buying more. I go to the fruit and the vegetables, and my cart seems to have more of that in it than it did before. [Focus group 2, participant]

Better Awareness of Long-Term Benefits and Motivation to Make Healthy Lifestyle Changes
Participants also highlighted how the app provided them with a better awareness of the long-term benefits of maintaining a healthy lifestyle. They expressed how they viewed the information provided by the app as vital and practical to apply within their daily lives:

...if I continue this way, I'm going to reap better benefits but that information is so vital that you have that information going into the grocery store that this is what I need to buy. Yes, that's cheaper and I'd get more of that, but this is much healthier and it's going to reap better benefits for me in the long run. [Focus group 2, participant]
Similarly, participants positively expressed that the app provided them with motivation to start healthy lifestyle changes: ...once I started focusing in and how this program showed me just to make slight changes....I started changing my eating habits before I even started going to the gym, and I started losing before I even did anything at the gym. [Focus group 2, participant]

Team-Based Lifestyle Changes Among Couples and Across Generations
Several participants shared how they noticed that there were several team-based lifestyle changes among couples engaged with the app: ... having a spouse doing it along with them, I think that really encourages because both of those couples were really engaged in terms of the physical activity. I think the one couple...tracked the most steps...over a million steps. One of the cascading effects noted from the app was how it inspired an intergenerational healthy lifestyle change among family members. A sense of role modeling of healthy behaviors was observed by younger generations and provoked a desire among them to adopt the same behaviors:

Positive Influence on the Patient-Provider Relationship
Participants identified the positive influence that their participation in the program as a whole had on their relationships with their health care providers. Several participants shared how they were more informed and prepared during their regular check-ups. The information also increased their attention to and understanding of their prior discussions with their health care providers about their CV risk factors. They also were enthusiastically looking forward to sharing their progress toward improving their CV risk factors with their providers:

Health Promotion Within the Church Congregation
Another key element that emerged was that the app enhanced health promotion within the church congregation. This was described as communicating the health information learned from the app to weekly worship services with programming for adults and children:

Theme 6: Suggestions for App Improvement
Participants had a number of valuable suggestions to improve the app with a focus on certain features.

Visuals to See Progress of Diet and PA Self-Monitoring
In particular, participants suggested the use of a visual rewards system or dashboard summary within the app self-monitoring and tracking system to allow users to see their personal progress as it relates to their diet and PA. One participant provided a detailed suggestion:

Automatic Syncing Function From Other Diet and PA Apps
Participants also suggested integrating an automatic syncing function from PA monitors to upload tracking data directly to the app throughout the day. Participants stressed the inconvenience of having to manually input tracking data and the requirement to complete this "all at once." Syncing with other diet and PA apps at any preferred time during the day would alleviate the need to have to recall your routine and specific patterns at the end of the day. To illustrate this suggestion, one participant shared:

Additional Functions for Education Modules and Sharing Boards
Suggestions for additional content to the educational modules included topics related to genetics and CV disease risk. Participants also suggested the inclusion of closed-captioning for viewing content within the education modules, which would also allow them to view videos without sound in the appropriate setting. This feature would also support the needs of those individuals with hearing impairment:

Individual Tailoring of the App to Encourage App Use and Increase its Relevance
Participants also suggested that personally tailored reminders (via email or text messages) be sent to participants when there was a lapse in use of the app but not necessarily to those actively engaging with the app. This would help minimize the need for automated email reminders: ...a little reminder that, hey you haven't cleared your module or you haven't looked at this new content that we have out there...would be good, but that would be intuitive of the app itself and tailored for the person using it. [Focus group 1, participant] It was suggested that the app allowed participants to configure whether they preferred email or text messages and to include an "on/off" setting.
Finally, participants suggested individual tailoring of the app to specific CV risk factors and for women's health. One suggested delivery modality included succinct, personal messages to users outside of the education modules related to major CV risk factors such as hypertension or diabetes. One participant discussed this in the context of heart attack warning signs in women: ... even for heart attacks, women-their symptoms are totally different than men, so maybe having a module about that women, certain ages, watch for certain signs cuz this could be a heart attack for you, where in men it's...and I know a lot of times when you go to the ER, if you present certain symptoms, they don't think about a woman having a heart attack.

Principal Findings
In this formative evaluation, the FAITH! App was perceived as a culturally relevant and acceptable delivery modality for promoting and facilitating positive CV health behaviors for CV disease prevention among AA adults. The main themes that emerged were that the intervention successfully prompted healthy behavioral change through cultural tailoring, multimedia education modules, and social networking. Challenges to healthy behavioral changes related to the intervention primarily revolved around technical malfunctions with the diet and PA self-monitoring features. This formative evaluation contributes to a small but growing body of the literature providing evidence to support mHealth lifestyle interventions to foster sustained behavior change. Taken together, our results provide support for the willingness and eagerness of AAs for the use of mobile devices for health promotion.
A preliminary finding of our formative evaluation was the importance of considering unique cultural influences, as these can be facilitators/barriers to intervention engagement. Few mHealth interventions have adopted tailoring strategies to include content specifically optimized to the sociocharacteristics of AAs [30]. Our parent study is the first to document the development of an mHealth app with faith-based content for CV health promotion to the AA faith community [18]. As maintenance of a healthy religious/spiritual life is central to our prioritized audience [38], our connection of faith and healthy lifestyle through spiritual messaging and biblical scripture was harmonious. It has been previously shown that positive messaging through scripture may heighten the acceptability and use of mHealth interventions by AAs [39]. Our cultural tailoring strategies through both superficial (visual imagery of AAs) and deep (the Black Church) structures also likely increased the receptivity to adapting positive behavioral change strategies [40]. Within our cohort, this was concretely demonstrated by participants' reports of the organization of potluck events within the churches to sample healthy recipes. This has strong implications in activating change in cultural norms within the Black Church around unhealthy eating [41], thereby improving major CV risk factors such as obesity and diabetes, which are highly prevalent within this group [42].
This study adds to previous work showing the importance of understanding how behavioral theory constructs incite behavioral change in mHealth interventions [8,43,44]. Few studies have similarly explored these behavioral pathways through qualitative approaches. CareSmarts, a theory-based, mobile phone-based intervention was associated with improvements in diabetes self-management among AAs through self-efficacy, social support, and health beliefs [44]. Similarly, our participants' expressions of healthy lifestyle change supported our underlying theory-driven intervention development model. In line with the social construct theory model, participants were encouraged by a sense of connectedness or commitment to a group that inspired their adherence to positive health behaviors [45]. Several participants compared themselves with others through their interactions on the sharing board, which increased and maintained their motivation toward healthy lifestyle change. Interestingly, participant commentary also suggested that diet and PA behaviors were learned and reinforced in the context of the family unit (spouses and grandparents to grandchildren) [46]. In accordance with earlier studies, this influence of their behaviors on family members is reflective of the family model of reciprocal determinism [47][48][49]. Furthermore, participants within our study reported that their health beliefs were shaped by the content of the interactive education modules, which in turn enlightened them on their perceived CV risk. This likely further spurred behavior change.
Participants provided several suggestions for improvement to the app that could further strengthen health behavior change. Among these were streamlining the app self-monitoring features, prompts to encourage app use, and personalization based on an individual's CV risk. On the basis of this valuable feedback, there are plans to integrate these suggestions into the next iteration of the FAITH! App and make the intervention accessible on smartphones. Participants' feedback also suggests that efficacious approaches should aim to strike a balance between the need for individualized and collective social support for more meaningful and beneficial user experiences. Similar themes advocating for personalization to the individual's needs were identified by AA gatekeeper stakeholders involved in the development of an mHealth intervention for patients with hypertension [39]. Consistent with our participants, they emphasized the need to include churches in the launch of mHealth interventions. Although these suggestions are closely aligned with a recent systematic review suggesting that personalization and monitoring are effective behavioral change techniques commonly used in mHealth interventions, none of the 21 studies were focused on the preferences/needs of AA end users [43]. This presents a major challenge in drawing conclusions about the most effective features to incorporate into mHealth interventions. The intrinsic advantages of mHealth interventions are their flexibility for delivery in a wide range of settings and their easy adaptability to content to meet the unique needs of specific populations, even at the individual level. Future work is underway to further refine and investigate these app components for AAs within a larger randomized controlled trial.

Strengths and Limitations
This study is among only a few community-based studies that have performed a comprehensive evaluation through mixed methods of a mHealth intervention among AA adults. This study is also novel in that it describes the in-depth formative evaluation process of a CBPR study aimed at promoting CV health using app-based technology among the AA faith community. As such, our CBPR model provides access to valuable perspectives from an often overlooked, underresourced group at high risk for CV disease. Furthermore, our approach to health promotion may be used as a model to support the AHA 2030 Impact Goals to improve the population's CV health and well-being while increasing life span, particularly for AAs [50]. Our qualitative findings align with our previously published quantitative results showing improved CV health behaviors (increased daily fruit and vegetable intake and weekly moderate-intensity physical activity) [26]. Thus, they reinforce the impact that our health promotion strategy had on individuals with the greatest CV disease risk.
Nevertheless, we had a very small sampling (9/50, 18%) of parent study participation in the focus groups, possibly introducing recruitment and selection bias. Furthermore, there is a possibility of overestimation of acceptability of the app by participants due to social desirability bias, or those who did not participate in the focus groups may not have found the app to be satisfactory. We implemented strategies to mitigate social desirability bias through rapport-building techniques, pre-fieldwork training with our data collector, and minimization of power differentials [51]. Finally, our study findings may lack generalizability to other AAs residing in other regions of the United States or to other racial and ethnic minority groups. These limitations must be weighed against the opportunity to utilize this novel approach to health promotion in a community-based cohort of AAs.