We developed the prototype in three phases (see Figure 1). This process occurred in collaboration with a research team, an expert advisory panel (EAP), and a technology team. The EAP consisted of 3 transgender women of color, a Latinx woman, a Filipina woman, and an African American woman, each of whom holds a high-level position in an organization providing services to transgender women. They provided consultation and feedback throughout the prototype design and development process and were compensated as project consultants. The technology team included a technology director, an app developer, a database engineer, and a user experience and graphic designer and was responsible for the design and programming of the prototype. The ETR Institutional Review Board approved the study protocols and provided oversight. In phase one, the research team conducted formative research with potential users of the mobile app to understand the health needs of this population and how an app might support their health and to gather specific suggestions for the development and content of the app. Phase two included prototype design and development of the overall app structure and three activities (ie, a vision board, a pre-exposure prophylaxis [PrEP] education activity, and an interactive map). In phase three, a group of potential users used, rated the usability and acceptability of, and described their experiences with the prototype.

The research team conducted four focus groups and 20 in-depth qualitative interviews with transgender women between the ages of 18 and 59 years living in urban and rural areas in every region of the United States [39]. Briefly, women were recruited through partnerships with community-based organizations, fliers in community spaces and on social media, and word of mouth. Focus group participants received US $100 in compensation; interview participants received US $50. Topics included overall well-being and connectedness, transgender health, sexual health, use of internet and social media, and recommendations for a trans-specific app. Focus groups and interviews were audio-recorded and transcribed. We used the grounded theory open coding methodology and ethnographic methodologies to classify key themes [40,41].

On the basis of the findings of the formative research, the research team developed a preliminary curriculum outline, which was discussed with and refined by the EAP and, at times, additional transgender women from their networks. The technology team then determined the optimal formats for conveying the curriculum in an engaging manner, in small self-paced segments, with sufficient levels of interactivity, and customization. The technology team next created wireframes and storyboards to mock-up the layout, workflow, and interaction of all app elements. Through an iterative process of review and feedback, the EAP guided the designer to create a final set of imagery and style guide (see Figure 2 for the overall look). Following final approval of the storyboards, the app developer created the prototype framework using Adobe Animate. Using this cross-platform development method, the app could be built for several mobile platforms, including Android and iOS, and would allow the utilization of integrated mobile device (eg, mobile phone and tablet) features such as GPS and camera.

Guided by the formative research and EAP, the technology team created three prototype interactive activities. The first activity, entitled Create Your Own Vision (Board), provides participants with the opportunity to imagine how they would like their lives to be in eight domains: health, living situation, school/work, sex/love, making a difference, people, free time, and spirituality (see Figure 3). This activity seeks to help transgender women situate their well-being and sexual health within the social determinants of health and structural factors that emerged as the dominant theme of the focus groups and interviews.

This interactive vision board allows participants to use their existing pictures, take pictures, create doodles/artwork, select quotes, and add their own text or other images to create a vision for each of the eight life domains. These eight collages are merged into a final scrollable collage that represents the user’s vision for their future. The activity also has the user identify a primary goal for their life today, lay out the steps needed to reach that goal, and identify who can help them execute each step. This goal setting activity provides a roadmap to reaching their goal that users can return to anytime. In addition, the vision board can be printed or shared with other app users—an option that all usability testers said they would use.

The second activity—Is PrEP Right for Me or My Partners?—presents a basic overview of PrEP and guides participants through a benefits and risks assessment to help them decide if PrEP might be appropriate for them or their partners (see Figure 4). The activity primarily presents information through a series of videos featuring a charismatic African American transgender woman who is getting ready for and enjoying a night out with friends, including 2 transgender women who tell personal stories about using PrEP (see Figure 5). Interactive elements were included through questions and a poll that allows the user to see what other users are thinking about PrEP and a map that identifies PrEP providers in the users’ area.

Responding to the need for helping transgender women find culturally competent and welcoming services, the staff developed the third activity: Interactive Resource Map. The map displays providers/organizations in five service categories: medical, support, educational, employment, and housing/living. On the basis of GPS location, users see icons for each entry and can customize the map to view all or certain categories. When users choose an icon, they see details about the resource including name, contact information, and user ratings and reviews (see Figure 6). In addition, the map allows users to add locations to the map as they identify transgender-friendly providers in their area.

We conducted a 3-phase process to develop a prototype HIV prevention mobile app for and with transgender women. In the first phase, we completed formative research, which indicated that HIV prevention for transgender women must address the larger context of women’s lives. In the second phase, we developed and refined the overall app framework and three prototype activities based on feedback from the EAP. In the third phase, we conducted usability testing. Participants rated the app highly on usability and likeability. There was a significant increase in self-efficacy in finding LGBTQ-friendly services, intention to seek online social support, and PrEP knowledge. Overall, the participants were supportive of the app, with many saying they would use it regularly with no additional incentive beyond the information and opportunity to connect with other transgender women.

There are a number of strengths of the Trans Women Connected prototype mobile app and research process. The 3-phase development process allowed us to collect formative data about the health priorities of transgender women and incorporate this information into the app prototype. We developed these activities because they addressed issues or requests observed in the formative research. This approach, which is considerably more extensive than the typical app development approach, helps to ensure that the selected prototype activities were addressing real needs of the community. The extensive formative research and the rigor of the analysis in this project, when combined with an iterative usability testing approach, are unique among mobile app developers.

Although researchers have called for digital health interventions to (1) be grounded in behavioral theory [44,45], (2) utilize an in-depth qualitative understanding of the intended population [46], and (3) be iteratively developed with multiple stages of user feedback [47,48], the process has been rarely implemented consistently, with variations in the development process between academic researchers and technology industry developers. Most mHealth interventions designed by those within the industry do not yet incorporate theory-based strategies known to drive changes in health behaviors or undergo systematic testing to demonstrate their effectiveness [45,49,50]; however, they typically employ a rapid iterative development process of the technology. In contrast, interventions developed within an academic environment are more often grounded in behavioral theory [51], but researcher-driven mobile interventions often do not benefit from rapid iterative prototyping and standardized usability testing with iterative feedback [52].

The development of this app integrated the strategies used in both academia and the industry and, in some instances, used methodologies that were more rigorous. For example, rather than use a local convenience sample for focus groups, as is typical of most software and app developers, the project conducted focus groups with transgender women in four regions and interviews with transgender women throughout the United States, with key transgender community stakeholders actively assisting in recruitment activities. By using a community-driven, grounded theory approach, the research team was able to identify and delve deeply into themes that would not have been detected through simply examining focus group transcripts from convenience samples.

Furthermore, input from transgender women was integrated throughout the prototype development in coordination with the EAP and women from their networks. These iterative feedback loops began earlier than typically in app development and included gathering feedback on colors and image representations of transgender women and several other functional and aesthetic elements of the app before moving on to feedback on the app interface and activity content, function, and appearance. The inclusion of members of the LGBTQ community as part of the development process extended beyond formative research into voice talent/narration, video development, and design, with the designer being an LGBTQ-identified woman of color, the narrators being transgender women of color, and all video participants being transgender women. These multiple forms of community engagement not only resulted in a highly tailored app but also further provide a foundation for dissemination and active utilization once the Trans Women Connected mobile app is finalized. As suggested by other researchers, community engagement increases the likelihood the knowledge gained and the intervention, in this case an app, will benefit the community [53-55]. We recognize and appreciate the importance of the community’s priorities and solutions, which enhances the relevance and use of data and quality and validity of research [56-58].

Additional strengths of the usability testing were the multiple types of data and an evaluation of impact. We gathered both quantitative and qualitative data about the experience of using the prototype mobile app; the quantitative data allowed us to verify that we were meeting numeric targets for usability and acceptability (as is typical for usability testing), whereas the qualitative data gave participants the opportunity to provide more in-depth feedback and suggestions. However, measuring the impact of the prototype on knowledge, attitudes, and intentions during the development process rather than when the app is completed goes considerably beyond standard usability testing, which is typically limited only to identifying problems within the user experience [59]. This approach helps determine whether the learning and engagement activities are performing as intended before building out the entire app.

Finally, a central strength of this mobile app is the emphasis on addressing HIV holistically using a strengths-based perspective. This approach recognizes the resilience of the transgender community and the social and structural barriers that transgender women face [60]. Developing an effective intervention requires a comprehensive approach that intentionally targets community strengths and challenges [60-63]. Our formative research confirmed that attention to the larger context of trans women’s lives was key to HIV prevention [14], [61,64,65], and this understanding guided our development process.

A few limitations of our study need to be considered. The usability testing sample was limited to 16 participants in two different cities in the West Coast. Research has determined that 7 to 10 usability participants will uncover 80% to 90% of usability problems [66,67], and we were able to test the mobile app with a diverse group of 16 participants. This sample allowed the examination of some of the different experiences and identities of transgender women. We likely identified functional problems with the app and the app was well received as usable and acceptable; however, it is possible that more geographically diverse participants or participants from rural areas would have qualitatively experienced the app differently.

The inclusion of pre- and posttest knowledge and intention measures during a usability test is unusual and presents some limitations. Although significant increases in knowledge and intent were found on some variables, the sample size, although adequate for usability testing, is small for such testing and again may not represent the effect that would have been found with a more diverse or larger sample. However, the results are promising given that there were positive changes in key measures after viewing only three prototype activities, and this would seem to bode well for the efficacy of the complete mobile program. For several measures, there was no significant increase between pre- and posttest scores; we may have observed a ceiling effect. Given the short duration of app use, we were unable to measure behavior change. However, we measured antecedents of behavior change according to behavior change theories, and in the full trial when participants have a longer period to use the mobile app, we expect to observe increases in these antecedents and more protective behaviors.

A final limitation of the project was that all members of the research team identify as cisgender. Although the input of transgender women was included throughout the development process, no transgender individuals participated directly in the interpretation and analysis of data. As a result, there may be aspects of the data that were missed or misinterpreted because of the absence of this perspective.

Given the positive results from the prototype development and usability testing, our next steps are to complete the development of the mobile app, including the storyboards and curriculum/content for the remaining activities, and conduct a cluster randomized controlled trial to assess the effectiveness of the Trans Women Connected mobile app. The expansion of the app to include additional activities will be based on a theoretical framework of gender affirmation, resilience, and cognitive behavioral theory and will be implemented using an agile development process with iterative testing. Through the entire process, we will continue to work with an expanded expert and community advisory group of transgender women who will provide feedback on the app; they will review materials to ensure the content reflects their experiences and those of their community. The project team, in collaboration with the expert advisors, will identify the most essential topics, which will then be reviewed and revised through an iterative process of refinement. A newly formed community advisory board, consisting of 10 geographically and racially diverse transgender women, will provide feedback on the activity plans and storyboards. The project team will then revise the documents, with the process continuing until the content is approved by the community advisors. In addition, the community advisory board will test and review completed interactive activities throughout the project.

Strong formative research, followed by an iterative review from members of the intended audience, resulted in a usable and acceptable mobile app that has been well received by transgender women. Usability testing findings provided important insights toward refining and the further development of the Trans Women Connected mobile app. The results of this study suggest that an mHealth intervention can address critical structural factors that shape transgender women’s lives and support positive changes in the knowledge and attitudes about social connection and health access that are antecedents to increasing protective and health promotion behaviors.