Designing the mobile app for the CONNECT study was a collaborative process that emphasised the importance of user-centric design. The app, which plays a central role in the study’s goal of predicting and preventing psychosis relapse, was developed through a series of co-design workshops with the study teams Lived Experience Advisory Panel (LEAP).

Co-design

The design process began with a series of co-design workshops. The goal of these sessions was to ensure the app was intuitive, accessible and met the needs of its users.

In the initial workshops, we focused on understanding the daily challenges individuals face when using technology. Participants shared their experiences, highlighting the importance of engaging interface design, accessible content, non-intrusive monitoring and the need for privacy and data security. These insights were invaluable in shaping the app.

Figure 1: The iterative user research process

Iterative Development: Rapid Prototyping and Feedback

With a clear understanding of user needs, we moved on to prototyping. Early versions of the app were created using prototyping software and shared with workshop participants for feedback. This iterative process allowed us to refine the app continuously. For instance, participants emphasised the need for an engaging interface, leading to the development of visually appealing design elements and progress-tracking features.

Feedback also highlighted the importance of accessibility, in relation to reducing cognitive load. Participants wanted the ability to use the app without having to concentrate for long periods of time. Readability, in particular, was highlighted as potentially problematic. This input guided the integration of features to aid readability. These included support for screen readers and the development of a ‘dark mode’ for the app.

Figure 2: Evolution of prototype

Balancing Functionality and User Experience

One of the significant challenges was balancing the app’s functionality with an optimal user experience. As well as collecting passive data via their phone and wearable, the app needed to collect comprehensive amounts of active data, including digital questionnaires. Users are prompted via notifications to respond to the questionnaires, which provide vital data to the CONNECT study. After speaking with participants, we created a notification management system, so that the participants could easily control the frequency of notifications they received. This approach ensures that users remain engaged without feeling overwhelmed by the data collection process.

Privacy and Data Security

Given the sensitive nature of the data collected by the app, we created a series of diagrams to explain the ‘data journey’ of participants’ data. The diagrams were used to explain complex flows of data in a transparent and understandable way.  This would provide participants the information they needed about how their data was being used.

 

Software Development for the CONNECT project

Once the graphical designs of the app were finalized by our designer, our software team started working to build the software for CONNECT. We systematically transformed the designs into a fully functional application. For the CONNECT app, we employed a cutting-edge technology that facilitates simultaneous development for both Android and iOS platforms. This dual-platform approach significantly accelerated our release timeline, allowing us to deliver the app for the project more swiftly than the traditional method of building for each platform independently.

Despite our best efforts and planning, it is inevitable that challenges and deviations will arise. To effectively manage these uncertainties, we adopted an Agile development approach. This involved segmenting the app development process into three-week sprints, each focused on achieving specific milestones. At the conclusion of each sprint, we conducted comprehensive testing of the developed features and implemented necessary adjustments based on the feedback received from the project, testing and lived experience teams. This iterative approach not only allowed us to refine and enhance features progressively but also ensured that the application’s functionality aligned with real-world user requirements and expectations.

Figure 3: Agile development methodology

Passive data, challenges and privacy

Our most significant challenge was collecting data from various phones and wearable devices. CONNECT supports numerous devices, including Apple Watches, Fitbits, and Samsung Watches, each requiring individual system integration. Furthermore, we needed to obtain approvals from Apple and Google to access data from these phones and watches; these approvals required completing documentation and responding to technical and research queries.

Given the vast amount of data that the CONNECT project will generate, we needed to ensure efficient data storage and transfer. To address this, we implemented advanced compression techniques to minimize the load on devices. Additionally, we utilized background processing to facilitate seamless data transfer to our servers and allow the app to be used normally while transfer is taking place. This approach ensured that we could handle the extensive data efficiently without compromising device performance or user experience.

Given the nature of the app and its handling of personal information, security was a paramount concern throughout the development process. We implemented stringent measures to ensure that user data remains completely isolated from other aspects of the application. All personal data is transmitted to highly secure vaults, which are protected with multiple layers of security protocols. Access to this sensitive data is restricted to a very limited group of authorized personnel. This robust security framework guarantees the protection of user information and upholds the highest standards of data privacy and security.

Each phase of the development was followed by extensive testing – both automated and manual user testing – to ensure that the app was defect-free and fit for use in the study. We are delighted that that the CONNECT app is now in live use in the study and we look forward to seeing the results of the research.

Figure 4: Screenshots from the final CONNECT app