In the near future, two industry trends may converge, followed by a major federal push to strengthen investigations into Medicare Advantage (MA) programs for fraud, waste, and abuse.
The first is the rapid popularity of the MA program, in part due to patient demand for it low premium And innovation, new benefits. What followed was a flood of Covid-19-related claims, as patients, payers and healthcare providers sifted through medical bills and clinical records to reconcile billions of dollars in medical bills.
Taken together, these factors will create a new reality where payers will be forced to sift through an increasing number of clinical records to identify potential fraud, waste and abuse (FWA) and confirm billing accuracy to properly reimburse supply business.
MA Program: Greater Visibility, Greater Audit Risk
Over the past decade, MA (private plan alternatives to traditional health insurance) enrollments have grown more than doubledIn 2021, more than 26 million Americans are enrolled in MA plans, representing 42% of the total Medicare population and $343 billion (46%) of total Medicare spending (net of premiums). By 2021, the average Medicare beneficiary will have access to 33 MA plans, the most in the past decade.
The MA program contracts with the Centers for Medicare and Medicaid Services (CMS) to provide enrollees with a variety of benefits and Headquarters, where the plan receives a predetermined dollar amount per enrollee from CMS each month.Importantly, the monthly capitation repayment is risk adjustment Let all enrollees reflect their health status and develop an appropriate level of monthly spending for Medicare-covered services.
To help ensure the accuracy and completeness of monthly payments to the MA plan, CMS conducts Risk-Adjusted Data Validation (RADV) audits to recover any improper payment for a risk-adjusted diagnosis that is not supported by data in medical records.
In 2021, the OIG announced its intention to release a series of Multiple RADV Audits for MA Program Over the next few years, “tens of millions of dollars” in funds could be recovered from improper payments. RADV audits help ensure that MA Advantage does not accidentally or knowingly submit a payment claim that is not supported by a patient risk-adjusted diagnostic code.
In April, 21 people were charged in multiple schemes involving $150 million. False Billing and Theft of Federal Covid-19 Assistance ProgramsIn one case, defendants allegedly offered Covid-19 tests by asking people to provide their personally identifiable information and saliva or blood samples, which were then submitted to Medicare for fraudulent claims for unnecessary, More expensive testing or service. The other defendants allegedly took advantage of telehealth policies implemented during the pandemic, misappropriated funds intended for front-line health care providers, and created and distributed fake vaccination record cards.
To be sure, these won’t be the last cases of Covid-19-related fraud to lead to federal charges.
Improve audit efficiency
Payers have traditionally relied on expensive and time-consuming chart reviews to find and extract important unstructured data from patient records, such as information indicating whether various Covid-related tests are medically necessary. Now, however, as an alternative to graph reviews, payers are increasingly looking to natural language processing (NLP), an artificial intelligence-based technology that enables computers to “read” by simulating a human’s ability to interpret language and understand text, but without the constraints of human bias and fatigue.
NLP enables organizations to retrospectively analyze longitudinal health data to find specific clinical information about individual patients or to identify subsets of the population that require further exploration. As CMS continues its MA program for FWA audits, NLP will play an increasingly important role in helping payers pinpoint FWA instances before entering the audit phase.
How NLP can improve audit performance
To excel in RADV audits, or avoid them in the first place, MA programs need solutions that allow users to efficiently and accurately identify the details that support accurate risk adjustments. To assess risk and identify potential FWAs, auditors must access and gather critical information from thousands of pages of medical records. Here are three ways MA programs can leverage NLP to improve FWA audit performance:
- Detection mode: One of the hallmarks of fraud is a repeatable pattern of data, such as a large number of patients meeting the same prior authorization requirements. NLP helps payers detect these patterns, which lack the natural variability found in legitimate patient records.
- Identify outliers: Likewise, payers can use NLP to uncover unusual data that could indicate fraud, such as expensive tests on records that have no medical justification. By analyzing unstructured data to identify anomalies in patient records, NLP can quickly verify the existence of critical data.
- increase scale: Humans have a limited ability to perform a large number of chart reviews in a very short period of time, but NLP automates the process, allowing for huge growth in scale. With some medical records running into thousands of pages, advanced NLP tools can save time and money significantly.
Demand for Medicare Advantage plans among seniors shows no signs of slowing anytime soon, which means CMS will likely continue to expedite its FWA audits of MA plans, especially given the prevalence of Covid-19-related fraud. MA programs can now prepare for upcoming claims and audits by leveraging AI-based tools such as NLP to increase efficiency and accelerate accurate identification of potential FWAs.
