Enhancing Patient Care and Operational Efficiency: A Digital Transformation Case Study

About the customer:

A large, acute teaching hospital providing access to over 20 medical and surgical specialties, delivering child-health, adult, psychiatric and age-related healthcare on one site. The hospital is a public, voluntary teaching hospital funded by the nation’s Health Service Executive. It manages approximately 420,000 patient attendances annually. Its Emergency Department is one of the busiest nationally, serving a catchment population of around 450,000 and is staffed by over 3,000 people encompassing 58 different nationalities.

Challenge:

The hospital set out a five-year strategy (2019-2024) to improve access to services and create a 'hospital without walls'. Its aim was to ensure that all staff had access to the latest patient information across the hospital and its multiple locations in the community. The strategy included a combination of care pathway redesign and capacity investment actions aimed specifically to improve wait times. A critically important element of achieving this was digital enablement, harnessing all elements of the strategy to consolidate the Hospital’s position as a leader in integrated care. To realise these aims, it sought to develop a new integrated Electronic Patient Record (EPR) system to address several critical healthcare challenges:

• Fragmented Information: Patient data was scattered across different systems and paper records. This fragmentation made it challenging for healthcare professionals to access a comprehensive view of a patient’s medical history.
• Efficiency and Accuracy: Manual data entry and reliance on paper records were time-consuming and error prone. Administrative tasks hindered healthcare professionals from focusing on patient care.
• Integrated Care: Communication gaps existed between different hospital departments and services. Lack of integration hindered collaboration and continuity of care.
• Patient Safety: Incomplete or outdated records posed risks to patient safety. Without accurate information, healthcare decisions could lead to adverse events.
• Outdated Technologies: End of life business critical legacy systems presented an operational risk and inhibited service transformation.

Solution:

Kainos was engaged in mid-2021 to develop the new EPR solution based on our Evolve Cloud platform. We architected and delivered the solution using AWS Lambda alongside the Simple Queue Service (SQS), Elastic Kubernetes Service (EKS), Relational Database Service (RDS) and Elastic Compute Cloud (EC2). Following modern Twelve-Factor App methodologies, we designed and implemented the bespoke business applications in Java, leveraging industry microservice architectural and operational patterns to maximise solution availability and performance. The platform’s core is compliant with the Fast Healthcare Interoperability Resources (FHIR) standard, storing data in RDS PostgreSQL clusters and S3 storage. Following our standard DevOps methodology, our team adopted GitOps practices from the outset, leveraging ‘everything as code’ to ensure environmental consistency was maximised by:

• Using multi-disciplinary delivery teams responsible for code changes all the way to production.
• Fostering transparent and open communications by facilitating access to shared dashboards.
• Automating platform infrastructure provisioning/configuration via Terraform and OS-level bootstrapping and configuration via Ansible/Packer.
• Defining desired state Kubernetes configurations within EKS via Helm charts and manifests.
• Best practice code review and automated testing processes, with all changes applied exclusively through CI/CD pipeline automation.

One of the project’s many challenges was to build a robust, scalable, and secure way of exchanging data between on-premises datacentres and the AWS platform. After analysing the technical requirements and datacentre operational constraints, our team deployed AWS Storage Gateway with secure VPC endpoints to provide seamless, high-capacity data layer access and low-latency bi-directional data replication. To provide a scalable file ingestion process, we architected an asynchronous mechanism leveraging AWS Lambda, S3 and SQS. At a high level, the process includes:

• Validation: Checks the input file is well-formatted.
• Split: Divides large files into smaller chunks and submits to the SQS queue for processing.
• Load: Multiple Lambdas execute in parallel to ingest data into the RDS PostgreSQL cluster.
• Report: Generates a summary of actions and the process result Monitoring.

Throughout delivery, we ensured compliance with NCSC Cloud Security Principles and industry standard frameworks and CIS Benchmarks. Early in the build phase, we designed and implemented a comprehensive set of overlapping AWS monitoring and security services such as, AWS CloudTrail, Config, Security Hub, WAF, Shield, Amazon CloudWatch, GuardDuty, Inspector and Simple Notification Service (SNS). Supporting a zero-trust approach, these services provide a comprehensive suite of security tooling enabling end-to-end auditing, compliance, and a real time performance monitoring solution.

In 2024, we made several large infrastructure-related changes to the AWS platform. The first concerned moving microservices from EC2 instances to ECS containers. This had a significant, positive impact on many levels of the platform. For example, it greatly improved the development and deployment speed – dealing with containers is much faster than with virtual machines. It also gave our team a new level of elasticity while working on new platform features. Further, it enhanced platform security – ECR repositories with AWS Inspector checks let us know about any vulnerabilities before we deployed them, which was not possible with EC2 instances.

We made further important changes to our logging setup – instead of using a third-party solution with agents, pushing the logs out and manage them in an external service, we moved everything to CloudWatch. Together with Container Insights it gave us all the information we needed to know about. As a bonus, we also got the whole logging configuration setup (metrics, dashboards, and alerts) in our Terraform, where previously it was just a manual configuration of the external service.

Results and benefits:

The solution furthers the clients’ strategic goals by providing healthcare professionals with quick and easy access to a single view of a patient's health: their test results, medications, discharge summaries and more, across the multiple departments that can be involved in a patient's healthcare journey. In doing so, it has helped to address several critical healthcare challenges:

• Fragmented Information: consolidates all relevant information into a single EPR. Now, doctors, nurses, and other staff can quickly retrieve accurate data, leading to better decision-making and patient care.
• Efficiency and Accuracy: digitises records, streamlining workflows. It reduces administrative overhead, minimises errors, and ensures accurate data entry. This efficiency translates to improved patient outcomes.
• Integrated Care: facilitates seamless communication. All relevant healthcare providers have access to the same up-to-date information. This integration promotes collaboration and ensures consistent care across departments.
• Patient Safety: enhances patient safety by providing real-time, comprehensive records. Healthcare professionals can make informed decisions, reducing the risk of errors and improving overall safety.
• Outdated Technologies: replaces multiple legacy systems, reducing licensing, administration, and support costs.

In addition, the solution provides a modern secure, extensible, and maintainable solution. While reducing operational overheads and risk through the GitOps everything-as-code approach, it provides the ability to develop and deliver application enhancements into production within hours instead of weeks or months. This has significantly enhanced the hospital’s ability to adapt to evolving healthcare challenges.