Study design

This study employed a modified HCD process focusing on formative qualitative input to inform the contextual adaptation of the MASI mobile app for AYALHIV in Benue State, Nigeria [12]. HCD emphasizes participatory processes in which end users actively contribute to the identification of needs and the co-creation of solutions. The formative phase of the project included a series of FGDs with AYALHIV aimed at identifying adherence barriers, assessing user experiences with an early version of the MASI app, and generating insights to guide app customization.

Following thematic analysis, and to ensure systematic prioritization of suggested features, we applied the Must have, Should have, Could have, and Won’t have (MoSCoW) method which allowed us to balance user preferences with technical feasibility and resource constraints [16,17]. This dual framework ensured that the adaptation process remained both user-driven and practically implementable.

Study setting

The study was conducted at three high-volume HIV treatment facilities in Makurdi Local Government Area, Benue State. These sites are supported by APIN Public Health Initiatives under the PEPFAR program and collectively provide ART to over 19,000 clients, including a significant adolescent and young adult population.

Participant recruitment

Participants were recruited through consecutive recruitment at the three HIV treatment clinics in Benue State, Nigeria. Individuals were eligible to participate if they were aged 15–24 years at the time of recruitment (adolescents aged 15–19 years and young adults aged 20–24 years), were currently receiving ART at one of the participating sites, owned an Android smartphone, and demonstrated basic digital literacy. Demographic data were collected at the time of participation. To ensure confidentiality, participants were assigned anonymous pseudonyms for all transcripts and study records.

Ethical considerations

Written informed consent was obtained from all adult participants, and assent was obtained from adolescents under 18 years, along with guardian consent. Confidentiality and data security were maintained throughout the study. Ethical approval for the study was obtained from the National Health Research Ethics Committee of Nigeria (NHREC) with approval number NHREC/01/01/2007–15/06/2024D.

Pre-FGD app exposure

To support informed and meaningful participation in the FGDs, the existing version of the MASI app, originally developed for adolescents living with HIV in South Africa, was installed on participants’ Android smartphones approximately one month prior to their scheduled FGD. The recruitment and installation commenced on 4^th September 2023 and ended on 2^nd October 2023.

At the time of installation, participants received a brief orientation on how to access and navigate the app. Over the next four weeks, participants were encouraged to explore all app features independently at their own pace. No formal training, usage goals, or structured engagement requirements were imposed during this period. This approach allowed users to develop familiarity with the app’s content and functionality in a naturalistic manner, thereby enabling more grounded feedback during the FGDs.

Focus group design and implementation

Three FGDs were conducted to gather feedback on the contextual adaptation of the MASI app. In total, 22 participants took part in the discussions. Groups were stratified by age and gender to promote comfort and peer comparability: (1) one mixed-gender adolescent group aged 15–19 years (n = 6), (2) one male young adult group aged 20–24 years (n = 8), and (3) one female young adult group aged 20–24 years (n = 8). Participants were recruited consecutively during routine clinic visits at the participating study sites.

All FGDs were conducted in private rooms within the participating health facilities to ensure confidentiality and participant comfort. Discussions were facilitated using FGD interview guides developed for this study. The discussion guide was developed in English and administered in English; no translation was required. To support participation, each individual received a transport stipend of 5,000 naira. Lunch was also provided during the sessions.

Data management

Participant information, including names and mobile phone numbers, was stored in a separate, password-protected file accessible only to designated study personnel. This information was used solely for scheduling FGDs and administering participant compensation. All physical data (e.g., printed transcripts or handwritten notes) were stored in locked file cabinets at the study coordination offices and accessed only when in use. Digital data, including audio recordings and typed transcripts, were stored on secure, limited-access network drives maintained by the lead research institution. To safeguard confidentiality, participants were assigned pseudonyms in all transcripts and written materials. Participants were also reminded not to disclose any identifying information or details discussed during the FGDs to individuals outside the group.

Data analysis and feature prioritisation

All focus group discussions were conducted in English. This was appropriate because participants were required to be sufficiently literate to operate the Lu Dedoo/MASI app, which was available in English, and to participate meaningfully in group discussion. The focus group discussion guide was therefore developed and administered in English and was not translated into local languages. In addition to audio-recording, field notes were taken during each session to document non-verbal expressions, group interactions, and contextual observations that could support interpretation during analysis.

All FGDs were audio-recorded with participants’ consent and transcribed verbatim by a trained research assistant familiar with the study protocol. To ensure accuracy, a second trained research assistant and the first author reviewed each transcript against the corresponding audio file and corrected any discrepancies. All transcripts were de-identified before analysis and imported into NVivo® v15 for qualitative data management and coding.

We used thematic analysis following Braun and Clarke’s six-step approach [18]. To develop the coding framework, three transcripts were initially coded independently by two researchers. The team then compared codes, discussed areas of overlap and divergence, and agreed on a preliminary coding framework. A combined deductive and inductive approach was used. Deductive codes were derived from the study objectives and the major domains covered in the FGD guide, including app usability, engagement, reminder functions, barriers to use, and suggested modifications. Inductive coding was then applied to identify additional concepts emerging directly from participant narratives that were not fully captured in the initial framework. The coding framework was refined iteratively and then applied across the full dataset. Coding differences were resolved through discussion and consensus. Field notes were used to support interpretation of the transcripts but were not analysed as standalone data sources. The final coding framework comprised three overarching themes, each with related subthemes derived from recurrent codes identified across the dataset (Table 1).

Download:

• PNG
  larger image
• TIFF
  original image

Table 1. Coding framework used for thematic analysis of focus group discussions.

https://doi.org/10.1371/journal.pone.0350279.t001

In line with HCD principles, emergent themes were translated into design requirements and feature suggestions. These were then organized and prioritized using the MoSCoW method: Must have: essential features for functionality and user acceptance (e.g., offline access, medication reminders), Should have: important but not critical features (e.g., localized multimedia content), Could have: desirable features to improve user experience (e.g., customizable avatars, gamified reminders), and Won’t have (for now): features not feasible in the current development cycle [16].

App Adaptation and Feature Refinement Process

The Nigerian research team synthesised key thematic findings from the FGDs, with particular attention to participant feedback on usability, content relevance and preferences. These findings were prioritised and translated into specific, actionable recommendations for app modification.

Proposed changes were shared with the app development team through a series of structured virtual meetings. This iterative process ensured that adaptations were both user-informed and technically viable, maintaining the core functionalities of the original platform while enhancing its relevance for AYALHIV in Benue State.

User engagement

Participants responded positively to the app’s engaging features, particularly the interactive and gamified sections. Elements such as quizzes, badges, and games were frequently cited as primary drivers of interest and regular app usage. Several users reported that these elements made the experience enjoyable and rewarding, as one adolescent participant noted:

“The quiz was interesting. I always checked to see if there was a new one. It felt good when I got answers right.” (Male, 20 years old)

The gamification features, such as badges, enhanced emotional engagement, especially among adolescents aged 15–19. For instance, a female participant shared her experience:

“I liked the winning badges. It made me feel like I was doing something good for myself.” (Female, 20 years old)

The self-tracking features, which allowed participants to visualise their adherence behaviours, were also highly valued. These features helped users stay on track with their medication regimen and provided a sense of accountability. One participant explained:

“I checked my medicine tracker every morning. It helped me stay on time with my drugs.” (Female, 21 years old)

However, participants suggested diversifying the reward system, such as incorporating new levels or customizable avatars, to sustain long-term engagement. Participants’ preferences regarding user engagement are summarised in Table 2.

Download:

• PNG
  larger image
• TIFF
  original image

Table 2. Participants’ Preferences Regarding User Engagement.

https://doi.org/10.1371/journal.pone.0350279.t002

App functionality

Participants generally found the app easy to navigate (Table 3). One participant shared:

Download:

• PNG
  larger image
• TIFF
  original image

Table 3. Participants’ Preferences for App Functionality and Usabili.

https://doi.org/10.1371/journal.pone.0350279.t003

“I didn’t need help to understand where to go or what to do.” (Participant 3, Male, 23 years old)

However, some participants pointed out challenges related to internet data availability, with one female participant highlighting:

“It’s a good app, but without data, I couldn’t watch the videos when my data finished.” (Female, 20 years old)

While the navigation and layout were deemed intuitive, a few participants faced difficulties, particularly with features like setting reminders and accessing the chat function. One participant remarked:

“I needed help the first time. But after that, it was simple.” (Male, 22 years old)

Suggested modifications for improvement

Participants suggested several actionable improvements to increase the app’s accessibility, relevance, and overall functionality (Table 4).

Download:

• PNG
  larger image
• TIFF
  original image

Table 4. Participants’ Suggested Modifications for Improvement.

https://doi.org/10.1371/journal.pone.0350279.t004

Participants also requested more frequent updates to multimedia content, including local-language videos, animations, and peer testimonials. These additions were seen as crucial for enhancing both informational content and emotional support. A participant stated:

“If someone like me talks about taking ARVs in a video, it would encourage others like me. We want to hear from people we can relate to.” (PFemale, 21 years old)

Customizable medication reminders were another frequently mentioned improvement, with users suggesting features like personalized reminder tones, adjustable timing, and motivational messages. One male participant shared:

“They should include personalized sound or message.” (Male, 19 years old)

Finally, social features such as in-app forums or chat rooms for anonymous peer interaction were proposed to foster a sense of community, reduce stigma, and facilitate the sharing of coping strategies. Additionally, the inclusion of links to local support resources like clinics or helplines was recommended.

Recommendation was made for offline access, especially for features like the medication tracker, daily quizzes, and educational content. One participant expressed:

“I want to open it anytime, even without data.” (Female, 20 years old)

Feature Prioritization Using the MoSCoW Method

To translate participant feedback into actionable design requirements, we applied the MoSCoW prioritization framework (“Must have,” “Should have,” “Could have,” “Won’t have”) [16]. This method provided a structured approach for balancing user preferences with technical feasibility during the adaptation of MASI. This process ensured that the most critical features identified by adolescents and young adults were prioritised for immediate development, while secondary or less feasible suggestions were reserved for later iterations.

Lu Dedoo app development progress

The Nigerian research team reviewed the MoSCoW prioritisation with the app development team and implemented key modifications. Regular updates of prioritised features were added, including medication tracking and quizzes. Additionally, the team incorporated more multimedia content, including peer-driven games and localised educational materials.

Limitations

This study’s generalizability is limited by its geographic scope and participant criteria. However, the focused scope of this study could also be a strength rather than a limitation. The purpose of the formative phase was to ensure that the MASI app was meaningfully adapted to Benue State’s unique sociocultural and infrastructural context. Recognising that the original South African version of the app may not fully reflect the experiences of Nigerian youth, the localised feedback obtained through these FGDs allowed for precise, context-specific design modifications. While the resulting adaptations may not be directly generalizable to other regions such as Abuja or rural settings, this reflects the very essence of human-centred design, creating digital tools that are responsive to the needs and lived realities of a defined user community.