Methods: Agile iterative development methods are being used to incorporate end-user, and design and domain expert input to develop the TB support app. This study phase was conducted in the spring of 2018 in a pulmonary specialized hospital within the national public health system in Argentina in the Province of Buenos Aires. Individuals in active TB treatment were invited to participate in focus groups, field testing with a beta-version of the TB support app, and follow-up focus groups or individual working sessions. The focus groups, prior to field testing, involved semi-structured and open-ended questions, demonstrations, and discussions to understand their experiences being diagnosed, and their information and treatment adherence related needs and challenges. The PI and an experienced qualitative researcher oversaw and lead the focus groups. Following the initial focus group and a sample demonstration of the app use, participants were asked to field test each of the components of the app for 5 to 10 days. The components of the app tested included: (1) report taking medication daily (2) report and/or review list of side effects (3) take drug metabolite test (4) read educational material. After field testing, the participants returned for focus group/working group sessions to give feedback on design or functionality issues, content needs or refinement, as well as technical challenges, and barriers to using the tool. All focus groups and working sessions were audio recorded and transcribed verbatim for analysis by a local, native Argentinian Spanish speaking transcriptionist.
Thematic analysis of the focus group/working group transcripts were conducted by researchers in the native language. The process was iterative and coding schemes were generated using the Information-Motivation-Behavioral Skills (IMB) Model (Aliabadi et al., 2015) and the Mobile Application Reviews Scale (Stoyanov, Hides, Kavanagh, & Wilson, 2016). Three researchers individually read transcript for initial themes, then coded transcripts using coding developed schemes. Dedoose was used for data management. Coding together, developing testing codes to assess inter-rater reliability, and discussion and consensus were used to ensure confirmability. The remaining transcripts were coded individually. The data were summarized and participant recommendations were collated to guide the next programmer iterations of the app.
Results: The major code categories identified by the participants in the IMB model included: (1) adherence information, (2) adherence motivation, and (3) moderating factors. Within adherence information, the largest number of participants had specific questions about their diagnosis/treatment. For example, “How long do I have to wear the mask? When can I go back to work?”). Within adherence motivation, participants mainly indicated that their primary source of motivation was personal, in particular, to prevent transmitting the infection of others. One participant stated that it is “super important to use a mask for you and for everyone else to prevent” transmission. Additionally, many participants provided recommendations for other people living with TB. Often reassuring others about confronting the diagnosis of TB, “don’t worry too much because there is treatment, don’t be frightened.” Moderating factors mainly focused on psychological, emotional, or illness-related experiences with many surrounding fear of the diagnosis and the impact of the diagnosis on their lives. One participant stated, “When it starts, you face a disease that you don’t know much about. What it is? The fear of not knowing whether it is serious or chronic. If there is a cure, you fear the long treatment and the amount of pills.”
Using the uMars scale, the major categories identified by participants included: (1) engagement, (2) functionality, and (3) information. Within the main theme of engagement, many of the participants identified features or functions that needed to be added to improve customization of the app. For example, many participants requested that an alarm as a reminder for medication administration, calendar view to see their progress, and the option of consultations be added. In functionality, some of the concerns expressed were slow connectivity issues, test strip upload issues, and recommendations for instruction and education modifications. For information, many participants felt that the app’s goal for assistance in treatment was good. Additionally, they felt that the quality of the information provided was very useful. In regard to the urine testing component of app, many of the participants noted that photo quality for the testing strips could be influenced by environmental factors such as lighting.
Conclusion: Overall, the participants found the app to be a very useful and relevant tool. The participants provided invaluable feedback that will inform further iterations of the app for testing among a larger sample of participants in Argentina. Considering the lack of patient-focused TB medication adherence mobile apps, the mobile app in this study potentially serves a vital role in improving positive health outcomes. In order to meet the 2030 deadline for the UN’s sustainable development goals continued collaborations of experts and end-users are critical. It is only through sustained global partnerships among all relevant stakeholders that progress can be made.