Healthcare Fraud, Waste and Abuse (FWA) is a global problem; and it is a common problem for all healthcare payers. Most of the healthcare purchasers, is facing similar or same challenges, and are aiming to improve FWA detection. Globally, this figure is at an average of 6.99 percent (Gee and Button, 2014). Most global insurance payers perform a post-payment audit claim process; that is, they undertake audits very late in the claims process, as it is often takes place after all reimbursements have already been made to the provider. Also, human resource capacity has often been highlighted as a possible challenge, which can also lead to limited quality and quantity in a claims audit process.
An electronic claims audit process could help identify FWA problems more efficiently. Claims management systems can benefit from the assistance of new affordable information, communication and technology (ICT), such as business intelligence and machine learning tools. Implementation of these tools could dramatically improve early detection of risky claims and enhance the processing of claims audits. This module will apply such tools in the case of FWA detection under the auditing of the claims management system and integrated module of openIMIS claim management functionalities.
This proof of concept project will analyze claim data structure and claim adjudication rules for a fee-for-service based payments model. During the project, the openIMIS claims data structure will be studied and analyzed with the understanding of typical FWA patterns that are able to lead to the development of a customized expert decision system for FWA detection.
AUDENTA Consulting - Audenta consulting firm was founded in 2011 with the goal of consulting related to the business and management in health care and health insurance with a special focus to the application of advanced technologies in the health care system. In this project Audenta Consulting will be responsible for project management, claim management business rules and FWA detection models for claims system.
ZAGREB EXCELLENCE INSTITUTE (ZEI) is NGO network established by the group of senior healthcare consultants with extensive management experience in mandatory health insurance, focusing on business excellence and improvement of national health systems. ZEI in this project will be responsible for Solution architecture and design, Machine learning and integration and project management, FWA detection models and use cases in fee-for-service based payments.
SORSIX is a technology company headquartered in Sydney, Australia, with offices in Skopje, Macedonia and Dublin, Ireland. Sorsix provides bespoke software development services, consulting services, and its proprietary health platform Pinga and proprietary database technologies to customers in the healthcare, infrastructure, and telecommunications industries.Sorsix’s team has extensive experience in serving customers in the healthcare sector, at all levels – from GPs and administrative staff through to insurance providers at the state level. In this project, Sorsix will be responsible for assisting the consortium in solution design, and for providing the software development to build the FWA prevention tools and module into OpenIMIS.
Claims management systems with the assistance of new affordable information, communication and technology (ICT), such as business intelligence and machine learning could dramatically improve detection of risky claims and enhance processes of claims auditing. This module will apply such tools in the case of FWA detection under the auditing of the claims management system and integrated module of openIMIS claim management functionalities.
This proof of concept project will analyze claims data structure and claim adjudication rules in fee-for-service based payments. During the project, openIMIS claims data structure will be studied with the understanding of typical FWA patterns that are likely to lead to the development of customized expert decision system for FWA detection. System and enhanced claims audit process will be piloted and adjusted with claims data model from primary, secondary and tertiary care. By relying on machine learning capabilities, the system/module should be expected to be constantly improved by increasing the number of cases that will be processed and detected as potentially “suspected” cases.
With FWA detection module, it is expected that a few auditors will be able to assess far many claims audit cases than through a random-based process like it is done today in the most payer organizations. A qualitative and quantitative analysis will be carried out at the end of the project to lead to a management decision on the results of this intervention.
A description of how your solution is interoperable with national health information architectures
The claim management process is one of the core business processes in the insurance business and is the core part of IMIS. FWA detection and prevention is essential for effective and successful health insurance business and should be a vital part of national health information architecture. FWA detection module should be incorporated in claim management system of the initial process of receiving and analyzing claims data structure and claim adjudication rules and giving prompt feedback to providers about technical or analytical claim issues and need for possible corrections. If FWA module should be integrated part of openIMIS and therefore interoperable with healthcare providers information systems.
Description of how your solution furthers the development of openIMIS and fits within the priorities and roadmap
We believe that FWA is the common problem for all health insurance organizations and developing FWA detection and prevention module as an integrated module in openIMIS should meet its the priorities. With this proof of concept project, all aspects of integration issues and roadmap priorities would be assessed and defined.
Project management approach
In this project, the market experience and project experience of the consortium members will be partnered with the software development capabilities of Sorsix. Sorsix has extensive experience providing software development in the healthcare and healthcare administration sector to a variety of customers.
Sorsix development teams follow a hybrid Agile approach, enabling adherence to a strong product vision while maintaining flexibility for changes within this vision to ensure development never stalls. As part of the hybrid Agile approach, Audenta and ZEI, with consultation from Sorsix and the OpenIMIS community, will generate a high-level scope of work and requirements. Sorsix teams will break this down into a detailed technical scope and execute this development with an Agile approach, enable fast builds from regular sprints to be demonstrated to the consortium and community for feedback.
The project estimation is 10-12 calendar months, with 2 months of project preparations, 6 months of SW development, 2-4 months of uploading test claim data with machine learning adjustment and improving the scoring algorithms, producing project result reports and recommendations.
Potential obstacles and plan to overcome them
For a successful proof of concept, it is important to have realistic test claim data. Our target is to use actual historical anonymized health insurance claim data, over three to five consecutive years. To effect this, we will engage with the community and establish a project partnership with one or more health insurance organizations. Ideally it will be an organization already using openIMIS, simplifying the transition and use of data for this process, and supporting the openIMIS community. The support of openIMIS community will be important in providing contact and access to the appropriate kind of historical anonymized data. An alternative possibility could be to use data from some other country/payer and to transform the claim data in openIMIS format. Provision for the data engineering necessary in such a step has been made in determining the size of the project.
Another potential obstacle may be enabling the key value-added components of the FWA tool to adapt to different regulatory environments. We plan to mitigate this by adapting to existing configuration options in openIMIS and engaging with the broader openIMIS community to ensure that workflows that have been experienced by openIMIS users have been met.
* Claim management
* Fraud, Waste, Abuse (FWA) detection and prevention
* Machine learning
* Health insurance information system
* Proof of concept