*2.2.9 HCP is unable to integrate app data into their own EMR for analysis or follow-up or share the data in their EMR with their patient's apps*

The current state of interoperability in most jurisdictions has not matured to the point where this type of digital data exchange can occur easily or quickly. HL7's Fast Healthcare Interoperability Resources (FHIR) standard is evolving rapidly and is likely to be readily available over the next 5–10 years. As of late 2021, it is relatively immature and not ready for production use with apps.

## **2.3 Identifying and designing for a behavior-theoretical model**

Through a process of discussion, reflection and mapping the experience of patients with MG to various behavior change theories, we determined the Selfdetermination Theory (SDT) [90] to be appropriate for patients with MG. SDT posits that people have a need for personal growth and that there are 3 intrinsic motivators that drive that growth: 1) A need for mastery over life's situations (Competence); 2) a need to feel in control over our lives (Autonomy); and 3) a need to feel connected to others (Relatedness).

SDT has a strong connection to the lived experiences of people with MG. Many patients with MG want to have more control over their lives and be able to do that through more knowledge and skills about their disease. They also want to gain greater connectedness to their HCP to get better guidance on how to manage their disease and faster time to treatment. The incentive for using the app should be rooted in a behavior theory that is compatible with improvement of the disease experience.

We incorporated the 3 elements of the SDT by providing more facts and information that the patient can use to manage their disease, by providing visualizations of the data they collect to help them see how well they are doing and what they can do to improve their disease experience and providing them with a concise and easy to understand report suitable for sharing with their HCP, thereby promoting control and relatedness at the same time. These factors can be refined over time, as patients start interacting with the app in the real-world.

#### **2.4 Identifying and designing for app-related behaviors**

App-related behaviors were considered at various stages of app use: 1) during on-boarding, 2) when capturing baseline data, 3) during routine use. The Elaboration Likelihood Model and the Information-Motivation Behavioral Skills model are excellent behavior change models for use in mhealth app design and development as the focus on app-related behaviors [54].

The Elaboration Likelihood Model (ELM) states that behavior change will usually arise after a change in one's beliefs and attitudes and that a change in beliefs and attitudes can be triggered by a cognitive stimulus, i.e., a message or fact. The ELM proposes several mechanisms for delivering the cognitive stimulus, including *personalization*, *verifiability*, *expertise* and *surface credibility*.

The Information-Motivation Behavioral Skills (IMB) model states that new behaviors are the result of information, motivation and skill in executing a behavior. The IMB proposes the following mechanisms for delivering relevant information, motivations and skills to patients: *reduction* of effort needed to do something, *tunneling* to breakdown large goals into smaller, more manageable goals, *reminders* that help the patient remember to do a task and *macro tailoring* which means adapting messages to the needs of a specific group or segment.

During on-boarding, we want to make it fun and easy for users to start using the app. The first few screens do not require any data entry (*reduction*). They inform the patient about the app, explain the benefits of using the app from the perspective of all the different personas (*macro tailoring*) and start educating the patient about how to use the app and what to expect from using the app. The app's design is very professional and has a clean and visually appealing look and feel,

giving the app *surface credibility*. The patient also learns that the app has been developed by experts in the field (*expertise*).

When capturing baseline data, the patient is offered some control over what they would like to track and share with their HCP (*personalization*). Although all patients are asked to enter the minimum data that will generate a report for the HCP, they have the option of only capturing relevant information about their symptoms. Once they have entered a small amount of data, they are offered a visualization of their information (*tunneling*). Baseline capture occurs on the first day of use. The aim is to keep the session relatively short and pleasant, even though the patient is likely to be excited about using the app for the first time and would be willing to spend a lot more time with the app. Allowing the patient to get too excited could lead to overinflated expectations and disappointment.

Once the patient has entered their baseline information, they enter the routine use phase. During this phase, they can explore the app, learn about their disease and learn about the provenance of the information made available to them (*verifiability*). They also receive notifications to enter data periodically (*reminders*) and can change the settings to personalize the schedule of notifications.
