Factors for Supporting Primary Care Physician Engagement With Patient Apps for Type 2 Diabetes Self-Management That Link to Primary Care: Interview Study

Background The health burden of type 2 diabetes can be mitigated by engaging patients in two key aspects of diabetes care: self-management and regular contact with health professionals. There is a clear benefit to integrating these aspects of care into a single clinical tool, and as mobile phone ownership increases, apps become a more feasible platform. However, the effectiveness of online health interventions is contingent on uptake by health care providers, which is typically low. There has been little research that focuses specifically on barriers and facilitators to health care provider uptake for interventions that link self-management apps to the user’s primary care physician (PCP). Objective This study aimed to explore PCP perspectives on proposed features for a self-management app for patients with diabetes that would link to primary care services. Methods Researchers conducted 25 semistructured interviews. The interviewer discussed potential features that would link in with the patient’s primary care services. Interviews were audio-recorded, transcribed, and coded. Framework analysis and the Consolidated Criteria for Reporting Qualitative Research checklist were employed to ensure rigor. Results Our analysis indicated that PCP attitudes toward proposed features for an app were underpinned by perceived roles of (1) diabetes self-management, (2) face-to-face care, and (3) the anticipated burden of new technologies on their practice. Theme 1 explored PCP perceptions about how an app could foster patient independence for self-management behaviors but could also increase responsibility and liability for the PCP. Theme 2 identified beliefs underpinning a commonly expressed preference for face-to-face care. PCPs perceived information was more motivating, better understood, and presented with greater empathy when delivered face to face rather than online. Theme 3 described how most PCPs anticipated an initial increase in workload while they learned to use a new clinical tool. Some PCPs accepted this burden on the basis that the change was inevitable as health care became more integrated. Others reported potential benefits were outweighed by effort to implement an app. This study also identified how app features can be positively framed, highlighting potential benefits for PCPs to maximize PCP engagement, buy-in, and uptake. For example, PCPs were more positive when they perceived that an app could facilitate communication and motivation between consultations, focus on building capacity for patient independence, and reinforce rather than replace in-person care. They were also more positive about app features that were automated, integrated with existing software, flexible for different patients, and included secondary benefits such as improved documentation. Conclusions This study provided insight into PCP perspectives on a diabetes app integrated with primary care services. This was observed as more than a technological change; PCPs were concerned about changes in workload, their role in self-management, and the nature of consultations. Our research highlighted potential facilitators and barriers to engaging PCPs in the implementation process.

Would this be a useful feature? How would you use this information in a consultation? c. Prompt for patient to see their doctor (for example, if they had several very high blood glucose levels) and periodic reminders for check-ups with PCP and eye and foot checks with specialists. [

can include manually implemented and automatic prompts]
How useful would this feature be? What problems/risks do you think there might be?
6. GoShare Healthcare is an interface that allows PCPs to 'bundle' videos and educational material from a database of materials, and then either email or text these resources to patients. Each patient can receive their own specific bundle for their needs. It has been used in Victoria by a PCPs as a way to improve communication with patients. How would you feel about using this kind of resource to communicate with your patients? 7. I'd like to go back to the patient scenarios we discussed before. I will show you examples of app content or output from three different patients. This information will be similar to what you would receive as a 'summary report' of the patients' self-management and blood glucose readings.
[provide participant with example output for a patient with several low blood glucose readings, one with mostly on-target blood glucose readings, and one with several high blood-glucose readings]. At the time of the interviews, JA was a PhD candidate.

Gender
Was the researcher male or female?
The researcher was female.

Experience and training
What experience or training did the researcher have?
JA was trained in qualitative methods. There were no participants who dropped out. Some PCPs were approached and indicated interest but ultimately decided not to participate because they had limited time do so. Setting 14. Setting of data collection Where was the data collected? e.g. home, clinic, workplace Interviews were conducted face-to-face in the PCPs consultation room.

Presence of nonparticipants
Was anyone else present besides the participants and researchers?
Only the participant and researcher were present at the time of the interview.

Description of sample
What are the important characteristics of the sample? e.g. demographic data, date All interviews except one were audiorecorded and transcribed verbatim by an independent transcription service. One interview was not audio-recorded as the PCP had requested that only notes be taken (although these were transcribed verbatim as much as possible). 20. Field notes Were field notes made during and/or after the interview or focus group?
Field notes were only made during one interview (see above), for which the participant had requested no audiorecording be made. 21. Duration What was the duration of the inter views or focus group?
Interviews lasted an average 25 minutes.

Data saturation
Was data saturation discussed? Analysis took place concurrently and iteratively as interviews were conducted. This meant that saturation was discussed with the research team throughout the analysis process. The project team concluded that theoretical saturation was reached after 25 interviews, where variation in PCP attitudes could be adequately explained through three main overarching themes 23. Transcripts returned Were transcripts returned to participants for comment and/or correction? Transcripts were not returned to participants for comment and/or correction.