How AI and Data Mining Expose Healthcare Fraud
When most people hear the word fraud in healthcare, they probably picture something fairly obvious. A provider bills for a service they never performed. Someone submits a claim for a patient they never treated. An organization knowingly receives money for something it never provided.
Those situations certainly happen. Healthcare fraud can involve intentional misrepresentation of services, medical necessity, billing, or other information in exchange for financial gain.
The reality is that fraud in healthcare compliance can become much more complicated when you look at how healthcare organizations actually operate.
Fraud Does Not Always Begin With an Obvious Scheme
Healthcare professionals make decisions in environments filled with competing priorities, complicated regulations, staffing challenges, changing payer requirements, and enormous amounts of documentation.
COVID-19 demonstrated this particularly well.
Healthcare providers were treating patients during an unprecedented public health emergency. Rules changed. Telehealth expanded rapidly. Hospitals created temporary treatment environments. Providers worked under extraordinary pressure while trying to keep patients safe.
n some situations, organizations later discovered that claims did not accurately reflect what happened.
That can include situations such as:
Billing for more time than a provider actually spent with a patient
Using codes that did not accurately reflect the service provided
Documenting services after the fact based on memory
Applying diagnoses or billing practices based on information that later changed
Failing to meet requirements contained within payer or government contracts
The circumstances behind an error matter. Intent matters. The surrounding facts matter.
That does not eliminate the compliance issue, but it does demonstrate why organizations need to understand how a situation developed before deciding how to respond.
Data Analytics Has Changed How Fraud Is Identified
One of the biggest changes healthcare organizations need to understand involves data.
Claims generate enormous amounts of information. Payers, government agencies, and investigative organizations can analyze that information for patterns that may indicate unusual activity.
Data mining can identify things such as:
Billing patterns that fall outside expected norms
Unusual increases in claim volume
Services that frequently appear together
Providers whose billing patterns differ significantly from their peers
Changes in utilization or reimbursement over time
Artificial intelligence and increasingly sophisticated analytics have made these processes even more powerful.
This matters because you should assume your organization's activity can be analyzed at a level that may not have been possible several years ago.
A provider may look at an individual claim and see nothing unusual. An organization analyzing thousands of claims may see a completely different pattern.
That is where fraud in healthcare compliance becomes an organizational issue rather than simply an individual billing issue.
Your Contracts Can Create Compliance Risk
Another area organizations sometimes overlook involves contracts.
Healthcare organizations enter into agreements with Medicare, Medicaid, commercial health plans, government programs, and other entities. Those agreements can contain specific compliance obligations.
You may agree to:
Conduct exclusion screenings
Maintain certain policies and procedures
Follow HIPAA requirements
Complete credentialing activities
Meet specific reporting obligations
Maintain particular documentation
Follow applicable billing and coding requirements
If you sign an agreement stating that your organization will perform those activities, you need to understand what happens if those processes do not actually exist or consistently operate as required.
This is particularly important for smaller and mid-sized organizations. You may think of your payer contract primarily as an agreement governing reimbursement, while the contract may contain much broader compliance obligations.
From a compliance standpoint, you need to know what you have agreed to do.
Compliance Teams Need to Build Trust While Addressing Risk
There is another practical challenge here that does not always receive enough attention.
Compliance professionals have to identify risk while maintaining relationships with the people responsible for delivering care and operating the organization.
If every mistake immediately becomes a conversation about fraud, people may become afraid to bring problems forward.
That creates its own compliance risk.
You want employees to feel comfortable saying, "I don't know if we handled this correctly."
That gives the compliance team an opportunity to investigate, research the applicable requirements, involve legal counsel when appropriate, and determine what happened.
Sometimes the most responsible answer is simply, "I don't have enough information yet. Let me find out."
The important part is following through.
Ask Questions Before a Problem Becomes a Pattern
Fraud prevention requires organizations to pay attention to patterns.
Consider asking:
Has billing activity changed significantly?
Are certain services being documented differently than before?
Do your actual operations match the requirements of your contracts?
Are employees relying on outdated guidance?
Have new services or technologies created compliance gaps?
Are unusual billing patterns being reviewed?
Does leadership understand the organization's highest compliance risks?
These questions can help identify problems before they become much larger issues.
The goal is not simply to have policies sitting in a compliance folder. You need processes that people understand, follow, monitor, and adjust when circumstances change.
Healthcare will continue to evolve. Billing systems will change. Technology will change. Regulatory expectations will change.
Comments