Challenge

Kaiser Permanente’s legacy enrollment system relied heavily on the usage of MUMPS code to handle the API layer. The legacy UI screen navigation was very cumbersome and take a lot of command line inputs to get to the appropriate fields. The enrollment process involved parsing the fixed-length files, via MUMPS code and loading the data into the central repository called TMS Database after the data validations are applied.

Solution

EHRE lead the project to modernize the InterSystems MUMPS based Member enrollment process and provide a scalable architecture with InterSystems Ensemble and Caché as integral part of the solution to enable Kaiser Permanente to accommodate growing volumes of member enrollments and provide analytical capabilities. A common-object Model was developed to ensure a consistent flow of data to and from different components involved in processing a document. This common-object model has been extended to handle a consistent data flow between the validations and filer operations along with the dispatcher workflow. The capability to identify the changed data elements and the corresponding filer operations was developed to improve the overall system performance and the data updates to only the changed data elements. We proposed Self-contained Cross-cutting Pipeline Architecture, an architecture that is self-contained in the sense that the implementation contains code for all the necessary business requirements and does not depend on any shared code from other layers. The self-contained code can be easily built and deployed without having to rebuild the whole software code base. The architecture supports a pipeline to be used to support Asynchronous message workflows (message queue implementation) and will be able contain business layer and data access layer code as needed, and thus cuts across several layers of the architecture.

EHRE designed the new system with a scalable architecture for the modernized system utilizing Ensemble as the core technology. Numerous native features including Complex Record Maps to parse the fixed length records, Business Rules for data validations, Cache DB as the data repository, Data Transformation and Business processes and the Operations for outbound interfaces were utilized to implement the new system.

Additional features were implemented for data integrity utilizing the Cache object scripts. The user-friendly UI was developed using Angular JS as the front-end with the API layer built using Cache Object script. High performance ECP connectivity is established for the API and for the batch processing systems to communicate with the centralized database. Additionally, prioritization of certain batch jobs over others, halting and resuming in-flight processing with all the required data integrity checks at the transaction and at the batch level was implemented for over 9 data providers. The modernized system is developed to support Enrollments, Terminations, Demographic updates, Group Transfers, Family Transfers, Eligibility Edits, Cancellations and Implied Terminations for certain data sources.

A common-object Model was developed to ensure a consistent flow of data to and from different components involved in processing a document. This common-object model has been extended to handle a consistent data flow between the validations and filer operations along with the dispatcher workflow. The capability to identify the changed data elements and the corresponding filer operations was developed to improve the overall system performance and the data updates to only the changed data elements.

The new system was a modernized version of the legacy and was built to accommodate future business growth, fluid, user-friendly navigation. Extensive reengineering of some of the modules was performed in the design phase. The new solution supports state of the art encryption technologies and the client team manages the keys and the certificates.

EHRE implemented another feature to handle the fallouts during processing. This is integrated with the dashboards wherein the authorized users can make necessary updates to the data and resubmit the transactions to start from the point where it failed. This feature allows users to resubmit one or more transactions after fixing the “bad” data. The overall system performance and various improvements to ensure the online and the batch processing systems are not constrained to process the data. Additionally, dashboards and reports using DeepSee analytics were developed to provide the management a birds-eye-view of what is in the pipeline, what is currently being processed and what is completed are built as part of the overall solution.

This solution also covered non-functional areas like, security management, auditing and logging. EHRE aligned its development tools and processes to Kaiser Permanente’s organizational processes, as much as feasible. This included full compliance with Kaiser Permanente’s document formats for project artifacts including creation and distribution, software code repositories and other processes.

The code packages were version controlled through Deltanji systems and configured to push the changes to the respective environments at the click of a button. Quality Management included utilizing an automated test tool (UFT) and also manual test data based on the input requirements. Additionally, programs were developed to verify the data quality and integrity at various steps to ensure the requirements are completely met. An additional level of testing was performed by utilizing the developed User Interfaces to identify any data quality issues.

Some of the key features implemented in the robust modernized system are:

• User-friendly UI
• Scalable architecture to support future growth and new service lines
• Performance with Quality
• Extensible Visual Business rules
• Complex maps to handle additional data sources easily
• Data transformation layer to handle the prioritization and “start-stop” processing
• Error handling and resubmitting of the transaction