Mental Health Software Revamp to Improve Security and Performance for NGO Serving 15,000+ Patients

About the Client

The Client is a US-based NGO dedicated to conducting research in behavioral sciences, educating the public about mental disorders, and delivering evidence-based mental health care. Since its foundation, the organization has been engaged in clinical research on neuroscience and has published over 2,000 papers. The Client’s team of psychologists, psychiatrists, social workers, and speech therapists provides mental health care services both in person and through video calls, serving over 15,000 patients.

The Client’s mental health software suite was presenting issues related to poor performance and security concerns. Trusting our decades-long experience in healthcare IT, the Client engaged ScienceSoft for an audit of its software ecosystem. Later, the Client contracted ScienceSoft to redesign and enhance the app with new capabilities, relying on the insights and evolution plan we delivered as part of our audit.

20+ Healthcare IT Experts to Conduct Software Evolution

ScienceSoft assembled a project team of over 20 senior healthcare IT specialists, including a team lead, project managers, business analysts, front-end and back-end developers, open-source engineers, functional testing engineers, and test automation engineers.

To promote continuous dialogue with the Client’s stakeholders, our team adopted an Agile project management approach. The initial 4-week software audit and evolution planning phase was divided into one-week analysis cycles dedicated to the key software components:

• A cross-platform mobile app for patients (for iOS and Android).
• A web app for clinicians.
• An admin panel.
• A library with mental health exercises and assessments.
• A physician tool for custom assessments, exercises, and research surveys.

During the modernization phase, which involved refactoring and redeveloping software components, we used the time-tested iterative approach employing the Scrum framework and AAMI TIR45 guide. This enabled timely stakeholder feedback, allowing us to quickly adjust the project priorities to the changing needs and avoid budget overruns.

At the project’s onset, ScienceSoft interviewed the Client’s stakeholders to understand the existing care processes, as the mental health software lacked sufficient documentation. ScienceSoft manually explored the functionality of the software modules and analyzed the solution’s architecture and source code.

UX audit

ScienceSoft analyzed the existing mental health solution to identify the primary user scenarios (for clinicians, admins, and patients) and assess the transparency of software navigation and logic. Based on the UX research findings, we prepared a report documenting the identified problems, a UX refactoring strategy, and the updated UX wireframes for the software.

Code audit

ScienceSoft’s team reviewed the current software architecture and source code and assessed its quality, security, maintainability, readability, formatting, performance, and configurability. Our team delivered a comprehensive code audit report to the Client and performed a QA documentation analysis to assess the software testing strategy, test coverage, and bug reports. To ensure PHI security, we analyzed the data protection measures implemented in the software, assessed the security controls, and identified potential code vulnerabilities.

Software evolution planning

Based on the Client’s needs and the audit results, ScienceSoft’s business analysts defined the goals for software refactoring and evolution, designed a detailed project roadmap, and estimated the project budget. Working together with the organization’s stakeholders, ScienceSoft planned the new mental health software features. They were to improve the patient assessment workflows, the accuracy of the assessment results, and the reporting processes.

Software refactoring and rearchitecting

The audit results revealed that the software required significant refactoring, including architecture and infrastructure changes. Moreover, several software components needed to be redeveloped from scratch.

Following the software evolution plan, ScienceSoft’s team:

• Engineered new code in React for several faulty application modules.
• Refactored the existing code to enhance its readability, formatting, quality, and configurability.
• Eliminated code duplication.
• Fixed software defects.
• Improved the performance of the existing software functionality.
• Updated software libraries and deleted duplicated libraries.
• Implemented and configured the tools for automated code validation to be used during the development and continuous software evolution.
• Improved the software architecture and the split into functional modules and components.
• Updated the software infrastructure using AWS services.

In parallel with refactoring, ScienceSoft created the documentation for the existing and new software functionality to facilitate future solution maintenance.

Data migration

The process of data migration was particularly challenging due to the extensive history of research studies, which spanned many years. Ensuring the integrity of this data was paramount, as any compromise could impact the Client’s work and reputation.

ScienceSoft dedicated significant time and resources to execute the migration meticulously. However, the team encountered data inconsistencies in the existing database, stemming from previous structural changes and the addition of new tables without proper oversight. Since immediate migration of that data was not feasible, ScienceSoft advised the Client on the alternative data management scenarios that could resolve the problem in the future. In the meantime, the team managed to ensure that all critical research and patient data was preserved for the newly modernized system.

UX/UI updates and new feature delivery

Leveraging the insights from the software UX audit, ScienceSoft’s team upgraded the UX flows and redesigned the user interfaces for the clinician web app and the admin panel.

ScienceSoft’s business analysts acted as proxy product owners, taking the initiative to suggest timelines and roadmaps for new features. This allowed us to introduce system improvements for the Client’s review and approval, forming the backlog for our sprints.

Based on the mental health software evolution plan, we introduced the following new functionality:

• Support for partial assessment completion. The patients can submit their answers for assessment even if some of the questions are skipped. Additionally, if a patient starts an assessment but does not complete it within the scheduled time, the system automatically submits their responses after the deadline.
• A drawing tool with improved accuracy and usability.
• New assessment question types.
• An enhanced data visualization tool that supports the analysis of all question types as well as allows researchers to view patient responses and conduct assessments directly within it.
• Patient activity menu with all assigned assessments.
• Patient report generation based on a set of filters (e.g., by mental health condition, assessment type, patient activity).
• Integration with a partnering data analytics platform to streamline the review of assessment results.

Support request handling

Throughout the engagement, ScienceSoft managed change requests from the Client' priority partners and the primary user support portal. Our QA specialists triaged incoming requests, categorizing them as bugs, new feature requests, and feature update requests. Bugs were subsequently addressed by our developers, while new feature requests were handled by the business analysts, who collaborated with the Client to evaluate the necessity of new features, document the requirements, and plan the implementation. This streamlined approach ensured timely issue resolution and smooth delivery of system updates.

Knowledge transfer to the new vendor

After the successful completion of core project phases, the Client decided to transition the project to a different team for ongoing software maintenance and evolution. To ensure a seamless handover, ScienceSoft conducted a comprehensive knowledge transfer, providing full project documentation and a detailed system overview to help the next team continue the work without disruption.

Key Outcomes for the Client

• Improved system stability and performance thanks to eliminated software defects, refactored code, and enhanced scalability.
• Boosted security achieved by analyzing data protection measures, assessing security controls, and identifying code vulnerabilities.
• Preservation of vital research and patient data and improved data integrity, supporting ongoing and future studies and personalized patient care.
• Elevated user experience for patients, clinicians, and administrators due to streamlined assessment workflows and accelerated report generation.
• Comprehensive software documentation and direct knowledge transfer facilitating easier software maintenance and smooth project transition to the new team.

Technologies and Tools

React.js, React Native, TypeScript, Python, Firebase, Redis, PostgreSQL, AWS.