Iterative Adaptation of a Tuberculosis Digital Medication Adherence Technology to Meet User Needs: Qualitative Study of Patients and Health Care Providers Using Human-Centered Design Methods

Background Digital adherence technologies have been widely promoted as a means to improve tuberculosis medication adherence. However, uptake of these technologies has been suboptimal by both patients and health workers. Not surprisingly, studies have not demonstrated significant improvement in treatment outcomes. Objective This study aimed to optimize a well-known digital adherence technology, 99DOTS, for end user needs in Uganda. We describe the findings of the ideation phase of the human-centered design methodology to adapt 99DOTS according to a set of design principles identified in the previous inspiration phase. Methods 99DOTS is a low-cost digital adherence technology wherein tuberculosis medication blister packs are encased within an envelope that reveals toll-free numbers that patients can call to report dosing. We identified 2 key areas for design and testing: (1) the envelope, including the form factor, content, and depiction of the order of pill taking; and (2) the patient call-in experience. We conducted 5 brainstorming sessions with all relevant stakeholders to generate a suite of potential prototype concepts. Senior investigators identified concepts to further develop based on feasibility and consistency with the predetermined design principles. Prototypes were revised with feedback from the entire team. The envelope and call-in experience prototypes were tested and iteratively revised through focus groups with health workers (n=52) and interviews with patients (n=7). We collected and analyzed qualitative feedback to inform each subsequent iteration. Results The 5 brainstorming sessions produced 127 unique ideas that we clustered into 6 themes: rewards, customization, education, logistics, wording and imagery, and treatment countdown. We developed 16 envelope prototypes, 12 icons, and 28 audio messages for prototype testing. In the final design, we altered the pill packaging envelope by adding a front flap to conceal the pills and reduce potential stigma associated with tuberculosis. The flap was adorned with either a blank calendar or map of Uganda. The inside cover contained a personalized message from a local health worker including contact information, pictorial pill-taking instructions, and a choice of stickers to tailor education to the patient and phase of treatment. Pill-taking order was indicated with colors, chevron arrows, and small mobile phone icons. Last, the call-in experience when patients report dosing was changed to a rotating series of audio messages centered on the themes of prevention, encouragement, and reassurance that tuberculosis is curable. Conclusions We demonstrated the use of human-centered design as a promising tool to drive the adaptation of digital adherence technologies to better address the needs and motivations of end users. The next phase of research, known as the implementation phase in the human-centered design methodology, will investigate whether the adapted 99DOTS platform results in higher levels of engagement from patients and health workers, and ultimately improves tuberculosis treatment outcomes.


Participant knowledge of the interviewer
What did the participants know about the researcher? e.g. personal goals, reasons for doing the research All providers had been at a day long training to learn the objectives of the study. Patients were read a script and consent form which detailed the overall objective of this sub study.

Interviewer characteristics
What characteristics were reported about the interviewer/facilitator? e.g. Bias, assumptions, reasons and interests in the research topic Each interviewer introduced themselves to the participants including their name, occupation, and country of origin.

Methodological orientation and Theory
What methodological orientation was stated to underpin the study? e.g. grounded theory, discourse analysis, ethnography, phenomenology, content analysis Human-centered design utilized grounded theory and an inductive approach to content analysis.

Sampling
How were participants selected? e.g. purposive, convenience, consecutive, snowball All providers at the training were invited to participate (no sampling). A convenience sample of patients from 2 nearby health centers were included with an effort to purposely sample a range based on age, gender, HIV status, and treatment stage.

Method of approach
How were participants approached? e.g. face-toface, telephone, mail, email Clinic staff, who have a personal relationship with patients, approached participants face-to-face at regular refill visits to ask if they would like to participate.

Sample size
How many participants were in the study? 52 providers and 7 patients were interviewed.

Non-participation
How many people refused to participate or dropped out? Reasons? None.

Setting of data collection
Where was the data collected? e.g. home, clinic, workplace Provider interviews were done in Kampala at a conference center being used for the randomization ceremony. Patient interviews were done at their local clinic in a private exam room.

Presence of nonparticipants
Was anyone else present besides the participants and researchers?
A translator was present at the interviews.

Description of sample
What are the important characteristics of the sample? e.g. demographic data, date The provider participants represented all 18 included health centers -a mixture of men and women of diverse ages. Patients were a mixture of HIV positive, early and late stages of TB treatment, and near even gender distribution men and women.

Were questions, prompts, guides provided by the authors? Was it pilot tested?
A semi-structured interview guide was used for all interviews.

Repeat interviews
Were repeat interviews carried out? If yes, how many? No.

Audio/visual recording
Did the research use audio or visual recording to collect the data?
The research was audio recorded.

Field notes
Were field notes made during and/or after the interview or focus group?
Short-hand field notes were taken during the interviews and focus groups. These were completed immediately following the data collection.

Duration
What was the duration of the interviews or focus group?
The focus groups ranged from 45-60 minutes and the interviews from 15 to 60 minutes.

Data saturation
Was data saturation discussed?
Yes. After interviews were completed researchers gathered to determine if thematic saturation was achieved.

Transcripts returned
Were transcripts returned to participants for comment and/or correction? No.

Domain 3: analysis and findings
Data analysis 24.

Number of data coders
How many data coders coded the data? 2 25.

Description of the coding tree
Did authors provide a description of the coding tree? No.

Derivation of themes
Were themes identified in advance or derived from the data?
Themes were identified from the data in concordance with HCD methodology.

Software
What software, if applicable, was used to manage the data?
No software was used other than word processers.

Participant checking
Did participants provide feedback on the findings? No.

Quotations presented
Were participant quotations presented to illustrate the themes / findings? Was each quotation identified? e.g.

participant number
Yes -quotations were used to illustrate themes for the inspiration phase and quotes were identified.

Data and findings consistent
Was there consistency between the data presented and the findings?
Yes -the insights and themes found during interviews and feedback sessions were used to refine future prototypes.

Clarity of major themes
Were major themes clearly presented in the findings?
Yes -they are present in Table 1.

Clarity of minor themes
Is there a description of diverse cases or discussion of minor themes?
As this paper focuses on the changes made because of the qualitative conclusions, minor themes are not discussed.