Health care fraud, waste, and abuse (FWA) cost the industry billions each year—but what if we could outsmart it? Enter SAS Payment Integrity for Health Care: Detect and Prevent, a groundbreaking solution that is redefining how we can tackle FWA. As the first to offer models-as-a-service in a Commercial Off-The-Shelf (COTS) solution, SAS is setting a new standard for accessibility and speed in advanced analytics.
This innovation is part of our broader mission to democratize AI. We are empowering business users and analysts alike with low-code/no-code tools that put powerful machine learning capabilities at their fingertips. The SAS Applied Artificial Intelligence Modeling team leads this effort, building targeted, intelligent solutions that drive real-world impact.
From faster fraud detection to smarter prevention strategies, SAS is making it easier than ever to protect health care systems—and the people who rely on them.
Background
Everyone should care about FWA. It impacts all of us, not just the profits of private health care companies or the overall costs for Medicaid and Medicare programs. The extra costs incurred by FWA, including their investigations and efforts to recover lost money, are passed on to private insurance users in the form of higher premiums. The costs are also passed on in the form of more complex processes for obtaining services for private insurance users, as well as Medicaid and Medicare enrollees.
As health care payers implement controls to prevent and detect FWA, they increase providers’ administrative workload. These would include things such as prior authorizations and record requests. This can result in patients not receiving the necessary care, despite having health care coverage. Delaying care might lead to patients becoming sicker and requiring more expensive services in the long run. Skipping needed diagnostic testing can let illnesses like cancer go undetected and spread. This would increase the cost of care. Or even enable the cancer to advance to a late stage and become untreatable.
Health care spending in the U.S. is at historic highs and shows no signs of slowing down. The Centers for Medicare and Medicaid Services (CMS) recently reported a 7.5% increase in national health care spending to $4.9 trillion in 2023. This trend continues a decades-long pattern, as shown in Chart 1. This chart is based on an NHE Fact Sheet.
Experts at the National Health Care Anti-Fraud Association (NHCAA) estimate that 3 to 10 percent of annual health care spending in the US is due to fraud and abuse. That equates to a loss of between $135 billion and $450 billion in 2022. That’s BILLION, with nine zeros… The upper-end estimate of $450 billion is larger than the total GDP of all but 35 countries worldwide. Experts estimate that wasteful billing behavior—including fraud and abuse—accounts for as much as 25% of total spend. This is shown in Chart 2, which is based on the article “Waste in the US Healthcare System”.
The solution? SAS Models for Health Care Payment Integrity
Fighting health care FWA is complex and requires advanced analytics and AI to be effective. SAS has spent over 15 years working with health care organizations to develop tools that identify FWA. These tools are user-friendly, configurable to a given customer’s specific situation, and designed to provide a holistic view of an organization’s data when used together. SAS models offer users a range of options, from quick starts in smaller bites where they need help most to complete solutions with a narrow focus. Our data scientists actively help health care payers fight FWA, continually updating and improving these models.
SAS Models for Health Care Payment Integrity are designed to fulfill the needs of a variety of business and analyst users. Many health care payers lack staff with both analytics and health care fraud expertise. Some are smaller or even niche health care plans that have evolved from emerging quality-driven payment models. They require a low-code/no-code FWA solution that aligns with their risk and compliance needs.
SAS Models for Health Care Payment Integrity includes three current offerings:
- Peer Grouping – With outlier detection being one of the most effective methods for identifying improper billing behavior, accurate peer groups are a necessity. The SAS Peer Grouping model enables your data to drive peer group determination, facilitating apples-to-apples comparisons.
- Outlier Detection and Reporting – Identify specific claims, claim lines, or entities that are flagged according to advanced anomaly detection algorithms. Use leading industry reports to quickly identify anomalies that warrant further investigation.
- Rules – Rules, both complex and straightforward, from CMS, CDC, and SAS’ own experience, identify known patterns of improper billing as well as simple impossibilities, such as the unbundling of bundled procedure codes or surgeries that are not possible.
SAS Viya and what’s next
Our market-leading advanced analytics platform, SAS Viya, supports these models. SAS Viya operates as a cloud-native platform, offering high scalability, flexibility, and the ability to handle large volumes of data. It integrates a wide range of analytics capabilities. This includes data management, advanced analytics, machine learning, artificial intelligence, and data visualization, all within a single platform. In addition, SAS Viya supports open standards and integrates with popular open source tools, such as Python and R. This enables interoperability with third-party tools and data sources. It enables organizations to leverage their existing technology stack and easily integrate SAS’s advanced analytics with other systems.
Users can purchase a SAS Model for Health Care Payment Integrity in the same way they would other SAS software. Government customers can get these AI models through the General Services Administration pricing at a discount. Users can expect ongoing model improvements, as SAS AI Models have versions and updates, just like other SAS software products.
And there is more to come! Our next models for delivery will be:
- Analytic Data Mart – a data mart of pre-summarized data about providers and patients, ready for reporting or further analysis.
- Comprehensive Risk Scoring – A method of combining all potentially bad behaviors of a provider into a single risk score that can be used to rank order providers for investigation.
- Link Analysis and Reporting – A tool to let business users investigate linkages between providers to identify potential collusion.
- Claim Risk – A model to quantify the risk of FWA for individual medical and pharmaceutical claims.
Summary
The SAS Health Care Payment Integrity products provide end users responsible for payment integrity with a comprehensive set of analytical tools. Results from these products enable end users to identify the most likely cases of fraud, waste, and/or abuse, allowing for targeted investigation.
