Containing health care costs: Analytics paves the way to payment integrity
Health care organizations look to data-driven analytics to target payment integrity issues
By John Maynard, Principal Solutions Architect, Global Security Intelligence Practice, SAS
To contain costs, health care organizations (HCOs) are changing the way they pursue claims overpayments. Traditionally, they focused on fraud prevention and detection. But today – in a process more broadly referred to as “payment integrity” – health plans are looking to uncover a broader range of fraud, waste and abuse in claims processing.
Data-driven analytics is making that possible.
The past: Simple payments simplified healthcare payment integrity
Healthcare payers historically paid providers on a fee-for-service (FFS) schedule for claims. Claims coding became even simpler with electronic claims processing. However, complexity increased with successive revisions of the International Classification of Diseases (ICD) into ICD-10 and other claims codes. Payment integrity in health care was less of a priority than keeping track of these evolving codes.
Yet social and political pressures for healthcare cost containment continued to mount as healthcare costs increased. Managed care was introduced to help – but healthcare costs still grew too quickly. When the concept of payment integrity arrived, it often existed only in healthcare payer operations and was focused on claims payment accuracy, third-party liability and claims subrogation.
From a data analytics perspective, the hardware in this era made analyzing large data volumes difficult and expensive. Most payers struggled to clean and combine data sets and make them useful for healthcare payment integrity. Robust tools for integrating, cleansing and managing data made it easier for payers to prepare their data for advanced analytics.
SAS Viya, a cloud-native data analytics platform, allows programmers to code in both SAS and open source in a single environment. And with no-code/low-code data analytics capabilities, users can simply point and click for robust analytics and data visualizations.
The present: Healthcare payment integrity meets value-pay models
Over time, healthcare payers and those managing government programs learned that FFS payment models resulted in more services and higher costs without improved health outcomes. This trend created the shift to a growing number of value-pay models based on quality of care and cost containment.
These value-pay models focus on services clustered around particular healthcare needs (i.e., knee replacement surgery) and related health outcomes. At the same time, point solutions and wrap-around services (for patients with health care issues like diabetes) paid under an at-risk capitation model are growing more common. Such models require complex analytics to determine patient health acuity and projected costs of care for patient risk pools – along with actuarial science for rate setting.
As healthcare delivery grew more complex, complex payments followed. Medical claims coding options and requirements increased and included coding specific to the quality of care. Private healthcare payers developed more nonstandard provider contracts. US public payers, like Medicaid and Medicare, moved more beneficiaries into managed care under these private payers.
Countries with national healthcare systems often fail to share data between the national plan and private payers that offer wrap-around plans for the public plan’s non-covered services. These healthcare payers continue to struggle with coordination of benefits and, as a result, payment integrity.
US healthcare organizations, including payers, determined integration would help control cost and quality of care. Today, we see a convergence between large healthcare payers, pharmacy benefit managers and clinical providers – for both physical and mental health – reflecting the movement toward whole person care.
The emergence of the cloud for data storage changed how we use big data. Cloud-based virtual machines, containers and Kubernetes were born, allowing data architecture to catch up to software capabilities. Open source tools are still evolving but may continue to struggle with handling data volume and velocity along with model management.
To support comprehensive payment integrity and more, we will see rapid evolutions in the use of computer vision, document vision and text analytics across healthcare. Expect to see even more automation and intelligent decisioning capabilities to drive efficiency in the future.
The future: Healthcare payment integrity and advancing health-tech
Both advancing health technologies and the COVID-19 pandemic changed how healthcare is delivered.
For example, smartphones and wearable devices have gradually changed consumer expectations. Value-pay models are still evolving and expanding. High data volumes and the need for healthcare clinical data integration and interoperability will continue to increase complexity. Health equity, an emerging concern, is becoming a more common discussion among leaders of US public health plans.
From a data analytics perspective, today’s healthcare organizations have a deeper understanding of the value of unstructured data. But as service delivery and payment models evolved, data capture and storage trailed. As a result, large volumes of unstructured data like care coordination and clinical data that were rich in information were often inaccessible for data analysis.
To support comprehensive payment integrity and more, we will see rapid evolutions in the use of computer vision, document vision and text analytics across healthcare. Expect to see even more automation and intelligent decisioning capabilities to drive more efficiency in the future.
Want to learn more?
SAS combines three disciplines − behavioral analytics, claim analytics and clinical targeting − into a comprehensive healthcare payment integrity solution available through multiple deployment options.
Learn how we’ve implemented leading-edge technologies across the healthcare ecosystem.