I hear, and personally experience instances where the insurance company does not understand (or at least can explain to us…
Readers Write: Payers Are Approaching a Moment of Reckoning on Fraud, Waste, and Abuse
Payers Are Approaching a Moment of Reckoning on Fraud, Waste, and Abuse
By Ketan Patel, MD
Ketan Patel, MD is chief medical officer of SyTrue of Stateline, NV.
Payers are poised to face a new operating environment with significantly more scrutiny over fraud, waste, and abuse (FWA) in the wake of COVID-19.
Two years ago, the federal government created the Medicare Advantage (MA) Risk Adjustment Data Validation (RADV) program to beef up audits of MA insurers. For 2022, CMS also doubled its budget for fraud, waste, and abuse (FWA) investigations, and the Department of Justice just announced charges against 21 defendants accused of various healthcare fraud schemes involving the COVID-19 pandemic. Meanwhile, payers are working to reconcile billions of dollars in COVID-related medical expenses and correctly identify risk for the surging number of long COVID patients.
These factors have converged to generate significant potential headwinds for payers and will create the following two new realities:
- Payers will be forced to sift through increasingly huge volumes of clinical records to identify potential fraud and waste, as well as confirm bill accuracy to properly compensate providers.
- At the same time, as we head into the third year of the pandemic, payers will uncover an unprecedented amount of FWA related to COVID-19.
How successfully payers manage these challenges will be determined by their ability to replace time-consuming and expensive manual processes with artificial-intelligence-based tools that comb patient records to identify potential fraud, assess patient and population risk, and confirm payment accuracy.
In the past, payers depended on expensive and time-consuming chart reviews to find and extract key unstructured data from patient records, such as information that reveals the need (or lack thereof) for a patient to undergo various COVID-related tests. More recently, though, payers have turned to natural language processing (NLP) as an alternative to manual chart reviews. NLP is an AI-based technology that enables computers to “read” and understand text by simulating humans’ ability to interpret language, but without the limitations of human bias and fatigue.
With NLP, payers can retrospectively analyze longitudinal health data to find a particular piece of clinical information about a single patient or identify subsets within populations that require further exploration. Given today’s environment of increased FWA scrutiny, NLP is poised to play an increasingly important part in helping payers pinpoint instances of FWA.
The following are three ways payers can leverage NLP to improve FWA detection:
- Detect patterns. In cases of FWA, there is often a pattern of repeatability in the data, such as a large number of patients meeting the same prior authorization requirements. NLP helps payers detect these patterns that lack the natural variability found in legitimate patient records.
- Identify outliers. In the same respect, NLP can help payers spot unusual data that may be representative of fraud, such as expensive tests for which there is no medical necessity. With its ability to accurately analyze unstructured data to identify anomalies within records, NLP can quickly verify the presence, or lack of, critical data.
- Improve scale. While even the most hard-working humans possess limitations on their ability to perform a high amount of chart reviews in a narrow timeframe, NLP automates the process, enabling substantial improvements in scalability. Because some complex medical records may consist of thousands of pages, NLP can drive significant savings in time and money in reviews.
For payers, the time to prepare for increased FWA scrutiny is now.
So, are payers on board with two separate claims being billed for a single well-visit, if the patient reports something new during the visit? Someone told me about this, and I was like “nah, not a thing, that’s what an annual visit is for” until my last annual visit, when I was told that when I checked in. I was told that my well-visit had no copay (correct) but if I brought up anything *else* it would be billed as a separate visit and I would be charged a copay for it.
How is this not double-billing? Isn’t the purpose of an annual visit to not only monitor any existing chronic conditions, but identify any possible new conditions? Charging me money for reporting new symptoms that might result in new diagnostics and new treatment seems like a disincentive. If I’m in a financial bind, I’ll just keep my mouth shut and hope I don’t get sicker.
Preventative “well” visits are strictly to address preventative care – the CPT E/M code guidelines even call this out by clarifying this by outlining the following:
“The following codes are used to report the preventive medicine evaluation and management of infants, children, adolescents and adults.
The extent and focus of the services will largely depend on the age of the patient.
If an abnormality is encountered or a preexisting problem is addressed in the process of performing this preventive medicine evaluation and management service, and if the problem or abnormality is significant enough to require additional work to perform the key components of a problem-oriented evaluation and management service, then the appropriate office/outpatient code 99202, 99203, 99204, 99205, 99211, 99212, 99213, 99214, 99215 should also be reported. Modifier 25 should be added to the office/outpatient code to indicate that a significant, separately identifiable evaluation and management service was provided on the same day as the preventive medicine service. The appropriate preventive medicine service is additionally reported.”
Preventative well care is not the same as acute medical problem care (i.e.; new symptoms or new condition) and the guidelines for documenting, coding and receiving reimbursement are different based on the coding guidelines set forth in the CPT book.
Right, healthy patients don’t get charged, sick patients do. What an amazing incentive structure. If I were an insurance company I’d be asking why my members keeps getting sicker instead of having chronic conditions well-maintained. The answer would be: well, my doctor charges me a copay every time I say that my condition has gotten worse or I have a new pain I’ve never had before, so I stopped saying that.
AI should play a role in detecting fraud. Since I put the first AI-based system in the US in 1991, I have been watching the developments. The author’s approach also assumes that the data input by the provider office is legitimate. This is a much of a flaw as assuming claims have proper information, or Medicare Advantage plans all have sicker patients.
Any analytic tool will require independent verifiable sources of information other than claims or medical records (which can be easily manipulated) in order to be effective. Time and location, missing from today’s systems, go a long way in identifying fraud. Multifactor authentication at the point of care will also help. Right now there is zero authentication that the patient was at the doctor’s office when the claim stipulates. all that’s needed to push a claim through is an insurance number and some CPT codes. That is why stolen health insurance identities are worth $250-$1000 in the underground market.