Artificial Intelligence in Medicare Audits: Part II

Providers are deploying their own AI to fight back against the tyranny of algorithms.

In October of last year (2019), President Trump signed the “Executive Order on Protecting and Improving Medicare for Our Nation’s Seniors.” Section 9 of the Order entitled “Eliminating Waste, Fraud, and Abuse to Protect Beneficiaries and Taxpayers.” states that “The Secretary shall undertake all appropriate efforts to direct public and private resources toward detecting and preventing fraud, waste, and abuse, including through the use of the latest technologies such as artificial intelligence.”

For years, computers and algorithms have been used to conduct Medicare audits. A number of approaches have been developed. Each requires a set of rules that can be applied to the ocean of claims. These rules will then help identify potential problems.

But Artificial Intelligence (AI) is different. Using AI allows the computer to create its own algorithms, and test them over and over, and then refine them.

Soon after the Executive Order was signed, the Centers for Medicare & Medicaid Services (CMS) announced its AI Health Outcomes Challenge. The goal of this challenge is to develop AI technology that will change the “pay and chase” model of auditing to a predictive system. Rather than employing the old system of using algorithms to comb through data ferreting out fraud and abuse, CMS wants AI to prevent making bad payments in the first place. If the technology is developed, then deep learning and neural networks will be used to predict adverse events and unplanned admissions to skilled nursing facilities and hospitals.

We can then expect that AI in auditing will be expanded to work in other areas of Medicare.

Healthcare providers endure much frustration with current audit practices. Between 5 and 25 percent of claims billed are audited and subjected to a post-payment claim audit (PPCA). After all, it is merely a matter of policy to simply turn on the computers and ask them to spew out a tsunami of audit attacks against providers. There is little a provider can do to respond. As they respond to audits, almost all are forced to use a cumbersome and disruptive manual process.

For the time being at least, this represents a type of “asymmetric warfare” in which it costs much more to defend against an audit than to make the audit in the first place.

Providers Use AI to Fight Back

Providers are not standing still. If they are under attack from a wave of audits based on faceless algorithms, the answer is to build defensive systems to respond. In 2016, a hospital in the Midwest let out a contract asking engineers to build an intelligent system capable of responding to post-payment audits. It needed a system that would allow harnessing a single process simultaneously across its 16 facilities. It needed a system that would meet all of the strict HIPAA security requirements, and could easily be kept up to date as new patches and upgrades were released.

The result was i.pass™, a cloud-based application that streamlines and simplifies the entire process of post-payment review audits. The software works by allowing the health care provider to securely upload into the cloud their audited claims. A sophisticated workflow application handles the flow of claims and documentation as the provider compiles its responses to the audit. The i.pass™ application also comes with built-in “dashboard” technology that gives both claims processors and management an overview of the entire process. A critical element of the system is its scheduling control which prevents losing counter-claims because of missing deadlines.

The software worked so well that its parent company, Health System Applications, was approached by the State of Michigan. The State offered to provide any health care provider based anywhere in the United States funding for a pilot test use of the i.pass™ system. This funding is part of the State of Michigan’s efforts to accelerate innovation.

As the auditing processes goes on, the i.pass™ system will use Artificial Intelligence (AI) to learn which responses best allow the Provider to recover their claim money.

The use of AI by healthcare providers responding to a wave of audits initiated by AI systems is an escalation in the cyber arms race. The government will continue to deploy AI systems in Medicare audits, and providers will respond with their own counter-measures, also using AI.