What’s your best path to adopting advanced FWA technology?

What’s your best path to adopting advanced FWA technology?

Take this interactive self-assessment to evaluate your current SIU successes and determine your health plan’s need for advanced technology to level up your fraud, waste and abuse mitigation efforts. 

More advanced technology solutions for healthcare fraud detection, analysis and management are now readily available on the market. But, if you’re used to traditional rules-based tools siloed from other cost containment functions, the mindset shift required to think a little differently about how best to augment the SIU’s expertise can be daunting. How can your health plan properly prepare to take advantage of an advanced FWA technology solution? Take the interactive self-assessment to find out. 

Self-Assessment: Evaluate Your Readiness for Advanced FWA Technology 

The first step when considering upgrading enterprise technology solutions – for any area of your health plan – is thoughtfully considering your current state. While there are no wrong answers, this self-assessment will help you level set on some key areas of evaluation and decision making to better prepare you for successfully adopting a modern FWA solution.

Your Results: How Did You Score? 

Now that this assessment has provided a framework to evaluate your current efforts to mitigate FWA, your scored answers can help you determine a suggested path to successfully adopting advanced technology for fraud detection, analysis and management. 

5-13 points: Time to access outside expertise 

At this level, your SIU may not have access to sufficient personnel and data science resources to make significant gains in your FWA mitigation efforts. The time spent running down false positives and manually compiling reports to satisfy regulatory compliance requirements only exacerbates the issue by further constraining over-taxed clinical and investigative experts. If you’re like many teams being asked to do more with less, how can you get ahead? Next steps: 

  1. At this level – especially if you struggle to maximize your value from current tools – the very idea of adopting advanced technology can seem an insurmountable obstacle. You’re not alone in this feeling, and it’s possible to catch up.
  2. An achievable goal is working on tackling overpayments in whatever form they come – namely, wasteful billing practices. Working with the audit function at your health plan and shoring up payer-provider relationships are excellent foundational best practices that can make your FWA function more effective. 
  3. While advanced technology, clinical audits and investigative capabilities are the three pillars of preventing and mitigating FWA, it’s okay if your SIU isn’t well-resourced in all of those areas. Outside expertise can help you shore up any gaps until you are ready to internalize additional functions. 

14-17 points: Integrating FWA and Payment Integrity yields dividends 

You’ve been able to make a great deal of progress on your own up to this point, and that is to be commended. But your current siloed tools that keep your SIU team disconnected from other areas of cost containment may be holding you back from real gains. Next steps: 

  1. Are your current tools for fraud detection, analysis and management as advanced as you think? Most are based on decades-old technology. These three tell-tale signs can help you investigate the differences. 
  2. Your FWA efforts may be stymied by a lack of key information if you’re not well aligned as an integrated, total payment integrity function. Watch this 10-minute presentation any time for a look at the consequences of these data silos and the five benefits of eliminating them. 
  3. Start thinking about how advanced technology like deep learning models can help you overcome perennial challenges like false positives and manual processes. This on-demand webinar is a great educational resource. 

18-23 points: Ready to take your FWA efforts to the next level 

Relying solely on the rules-based tool your SIU has worked with for years is no longer the sophisticated strategy your innovative organization has historically found success with. And advanced data science teams that are still relying on predictive analytics understand that applications of artificial intelligence have progressed to an ideal and achievable solution for mitigating healthcare fraud and abuse. Next steps:

  1. When you go looking for an advanced technology solution for FWA, it pays to understand that all A.I. isn’t created equal. Here are three questions to ask to determine if the solution you’re investigating is the real deal. 
  2. Connecting disparate data sources – geographic, weather, financial, audit data and more – that provide context to aberrances enables your SIU to work smarter. Here’s how exactly deep learning models help address the complexity of healthcare data that foils simpler tools – while overcoming the interpretability hurdles presented by A.I. results. 
  3. While you don’t have to give up your reliable rules-based tool to take on an integrated fraud detection platform like Pareo, which is powered by A.I. based on deep learning models, ferreting out potential fraud and abuse without rules is a mindset change. This article helps explain the difference in the approaches, including how, by working with a modern FWA solution, your internal experts continue to provide a key competitive advantage for your health plan.

Bottom Line: Setting Expectations is the Key to Success 

No matter how sophisticated your current FWA operations are, you can successfully adopt advanced technology that helps you maximize avoidance and recoveries at the most optimized cost. Along with the stage-specific recommendations we offer above, here are three additional pieces of advice: 

  1. Start by setting realistic goals based on what is it that you want to accomplish. How do you want to use this technology in order to address the core challenges you’re facing? 
  2. Determine if your staff is organized appropriately. If your goal is to increase your case level or increase your savings, does that mean that you’re going to need additional staff? How would you organize (e.g., by line of business or product line)? 
  3. Prepare your SIU for the change. How will advanced technology affect your workloads in your day-to-day operations, and how do you make the most of your role? 

 Do you need help preparing for and navigating this change? Don’t hesitate to turn to us for fear that it’s too early to engage a partner in this processYour SIU can benefit greatly from advanced FWA technology that features true A.I. powered by deep learning models and seamlessly connects fraud detection, analysis and management. Those benefits are compounded greatly when they’re natively integrated on a platform like Pareo that also engages with overall payment integrity activity. This integration offers unparalleled visibility into the progress your team is making toward their goals and helps prove their sizable impact on cost containment at the health plan as a whole.



Talk to ClarisHealth about how Pareo®, a total payment integrity platform, is driving innovation at health plans. 

You Don’t Need Rules to Mitigate Healthcare FWA

You Don’t Need Rules to Mitigate Healthcare FWA

Health plan SIUs find deep learning models for fraud detection provide advantages over rules-based tools.  

The onset of the novel coronavirus pandemic highlighted a glaring weakness in rules-based FWA systems: the need to develop rules well ahead of their actual use in order to attempt to catch coronavirus-related FWA in a timely manner. At best, these rules are educated guesses at schemes that health plans are only beginning to pursue. But, as we’ve mentioned before, a fraud detection solution powered by deep learning eliminates the need for specific rules to be created. Even during an unprecedented pandemic, an A.I.-based solution natively surfaces aberrances that indicate fraud, waste and abuse. While the benefits of that change are significant, it’s also a huge mindset shift. How does fraud detection without rules even work?   

But first, let’s discuss why rules-based fraud detection is no longer the ideal tool for the job, especially in the current environment of stressed providers and increasingly complex novel fraud schemes.   

Why Not Rules?

If your health plan has been working to mitigate fraud, waste and abuse for any length of time, you likely have invested a great deal of resources in a rules-based system. But the novel coronavirus pandemic has exacerbated some concerns around relying solely on this traditional approach to fraud detection. 

Continued telehealth

From the beginning of the pandemic-created healthcare crisis, there have been concerns that the rapid adoption of telehealth would usher in a wave of fraud designed to exploit the relaxed rules that made its expansion possible. While that outcome remains to be seen, virtual care encounters certainly go against the conventions of in-person provider visits that fraud rules have been structured for. Legitimate providers conducting majority telehealth visits could easily trigger rules designed to catch outliers in place of service, impossible days and numbers of visits. 

increase medical savings clarishealth

False positives risk provider abrasion

How are you distinguishing legitimate providers whose billing patterns have changed for good reason over the last few months from a few fraudulent opportunists? With the fallout from COVID-19 battering providers, health plans have been understandably reluctant to overly burden their providers with aggressive pursuit of overpayments.  

Unfortunately, traditional rules-based systems – even those updated with coronavirus-specific queries – are prone to creating false positives that overwork investigators and risk provider abrasion. Providers have already raised the alarm that COVID-19 diagnosis codes don’t accurately reflect some of the more severe complications of the diagnosis, which could worsen the false positive rate of rules-based tools that can’t evaluate multiple approaches to the data simultaneously and fare poorly with unstructured clinical information. 

Novel fraud schemes

As we presented in our short on-demand webinar on the topic, a unique healthcare crisis like this one can absolutely give rise to bad actors gaming the system for personal gain. But traditional rules-based fraud detection tools are primarily designed to look for fraud schemes you already know about. As a result, the SIU could easily miss novel fraud schemes featuring more subtle aberrances that nonetheless add up over time. The success of these more sophisticated approaches depends on the assumption that health plans have limited their fraud strategy to the same tools in use for decades. 

Rules vs. Deep Learning Models

If rules-based tools are increasingly outmatched by the growing complexity of healthcare data and fraud schemes, what kind of solution features the flexibility and advanced technology to help health plans meet their FWA goals? In a previous article, we covered just how deep learning is uniquely suited to overcoming many of the common challenges faced by the SIU. But, how does fraud detection without rules operate? 

To help illustrate how fraud detection differs in a rules-based environment and with a deep learning model, let’s explore a typical COVID-19 claim you might encounter in the SIU. 

A 55-year old patient calls his Internal Medicine provider with complaints of chest congestion and a sore throat. The patient has underlying bronchial asthma and sees his Allergist for this condition once per quarter. The Internal Medicine provider schedules a telehealth visit and, based on his findings from the visit, prescribes a full lab workup for COVID-19. The payer is billed the following:  

  • Dx:  COVID-19 
  • Office visit 99214 
  • Lab codes 87635, 86328, 86769 

How does this new claim make it through the SIU to be evaluated for potential fraud? 

COVID-19 scenario: typical rules-based environment

If the SIU works primarily with a rules-based system, it’s possible the tool has been updated with a query for Level 4 office visits with a lab panel based on a diagnosis of COVID-19. Otherwise, the rules will need to be updated accordingly with the help of the IT or analytics team. Then, the SIU would query other Internal Medicine providers billing the same combination and manually compare rates and paid amounts looking for clusters of outlier claims. 

These findings may lead the SIU to additional queries to increase the likelihood of understanding if there is fraud or not: 

  • Query patient for other visits?  
  • Query Internal Medicine provider for past 6 months on COVID-19? 
  • Is this a trend for this provider or a one-off? 
  • Is the provider’s concentration of COVID-19 diagnoses higher because of his geography? 

COVID-19 scenario: deep learning model

On the other hand, if the SIU is working with a fraud detection solution powered by deep learning models, much of the work related to historical claims and peer-to-peer provider comparison information has already been accounted for in the system. In this scenario, the patient’s history of asthma, peer provider comparison based on block of dates, spikes vs. trends, geography, patient acuity and more are present in the models, which continuously update as new information comes in. 

From there, the deep learning models in the system push likely provider fraud leads to the SIU. With Pareo Fraud: Detection, the billing provider could be highlighted with a high Super ScoreTM based on high model scoring for Unusual Active Days, Ratio of Px/Codes and High Claim Volume. The investigator could then one-click drill down to the provider for more supporting detail on these scores and optionally conduct additional peer-to-peer comparison. 

This workflow supports the investigator’s expertise while greatly reducing the manual effort and eliminating the need to “pull” unprioritized query results from the system. 

Benefits of Deep Learning Models for Fraud Detection 

Deep learning models for fraud detection allow the SIU to go beyond the Boolean, if-then logic of rules-based tools. Freed from these restrictions, investigators are better able to discern the true potential fraud offenders and realize several differentiators that offer unique value: 

  • Eliminates need for multiple queries and manually layering those results 
  • Continuously learns and evolves so that  the more investigators work with it, the more precise it becomes 
  • Adds specificity to results to reduce false positives and bring focus to providers with high likelihood of fraudulent or abusive billing practices – including the amount of dollars at stake 
  • Digs deeper and wider into the data in multiple dimensions to find the subtle changes in provider behavior that may indicate FWA 
  • Extends capabilities of identifying relationships in the data to find hidden, previously unknown schemes 

Pareo Supports the SIU in Transitioning to Fraud Detection Powered by Deep Learning 

The benefits provided by deep learning models go a long way towards your compliance goals, including FWA prevention. But making a move like this, along with its associated mindset shift, can also feel like a gamble. The perceived risks associated with upgrading your FWA solution are largely centered around the discomfort of change for users and concerns about wasted resources. But conserving resources – time, effort and expertise – is exactly what a deep learning-powered fraud detection solution helps accomplish.   

And, for those SIU teams that have invested heavily in a rules-based tool and value its reliability and familiarity, that effort is not lost. As an integrative platform, Pareo allows you to easily connect this resource as another useful data input that Pareo Fraud: Detection can leverage to improve the precision of results more quickly. 

Finally, as the only solution for the SIU that seamlessly integrates overall payment integrity efforts, Pareo Fraud can help you quickly covert detection results to leads or cases, communicate relevant information back-and-forth with the audit function, and refer honest billing mistakes for provider education. This ability to maximize recoveries and avoidance at the most optimized cost end-to-end is unique to Pareo and ensures this solution proves valuable for all stakeholders – both internal resources and network providers. 


Learn more about the ClarisHealth 360-degree solution for total payment integrity and FWA, Pareo Fraud: Case Management and Detection.

ClarisHealth Launches A.I. Powered Healthcare Fraud Detection Solution for Health Plans

ClarisHealth Launches A.I. Powered Healthcare Fraud Detection Solution for Health Plans

NASHVILLE, TENN. (PRWEB) SEPTEMBER 09, 2020 – – ClarisHealth, the company transforming health plan operations with its comprehensive payment integrity technology platform Pareo®, launched today an artificial intelligence-based (A.I.) solution for healthcare fraud detection. Pareo Fraud Detection works seamlessly with Pareo Fraud Case Management for an integrated, 360-degree approach to mitigating fraud, waste and abuse (FWA).

According to Mark Isbitts, Vice President of Program Integrity for ClarisHealth, who led the project, Pareo Fraud Detection provides significant advantages over conventional tools that have been in use for decades. This traditional approach involves sifting through massive amounts of data and defining thousands of rules to detect fraud, which limits its scope of effectiveness.

“Currently, the SIU deals with overwhelming numbers of false positives and manual processes and still may miss some aspects of fraudulent activity and new emerging criminal schemes,” says Isbitts. “Pareo Fraud Detection is based on our proven payment integrity technology platform and features a transparent and efficient user interface that provides a view of why a provider is marked as fraud and enables investigators to visualize different patterns and scenarios in just a few clicks – without inputting multiple rules. ”

Pareo Fraud Detection models were developed in partnership with data science experts from the College of Business Administration at the University of Illinois-Chicago (UIC). Kyle Cheek, PhD., the Director of the Center for Applied Analytics and a clinical associate professor of Information and Decision Sciences at the university, has worked in various payer fraud organizations. He helped ensure the solution escaped a fundamental drawback of A.I.: lack of explainability.

“An A.I.-based technique is not usually transparent about how it generates its outputs and is often referred to as the ‘black box,’” says Cheek. “However, Pareo Fraud Detection harnesses the power of deep learning applications of A.I. informed by domain expertise so the SIU will get information on which providers to investigate first, why a provider is suspicious, and how valuable the investigation can be – in real-time, efficiently, and with limited manual efforts.”

The release of Pareo Fraud Detection continues the ClarisHealth strategic direction of expanding the native functionality of the Pareo platform to maximize health plan cost avoidance and recoveries at the most optimized cost. According to ClarisHealth CEO Jeff McNeese, this approach is termed Total Payment Integrity™ and it transforms engagement with internal and external stakeholders.

“Just like there had been little innovation in payment integrity before Pareo, the same has been true for FWA solutions,” says McNeese. “Being able to solve payment integrity and FWA challenges end-to-end – from prepay to post-pay – on a single platform fulfills the Total Payment Integrity promise of Pareo.”

About ClarisHealth

ClarisHealth is the answer to the health plan industry’s siloed solutions and traditional models for identification and overpayment recovery services. ClarisHealth provides health plans and payers with total visibility into payment integrity operations through its proprietary advanced cost containment technology platform Pareo®. Pareo enables health plans to maximize avoidance and recoveries at the most optimized cost for a 10x return on their software investment. For more information please visit  https://www.clarishealth.com.

View source version at: https://www.prweb.com/releases/clarishealth_launches_a_i_powered_healthcare_fraud_detection_solution_for_health_plans/prweb17373183.htm

Source: ClarisHealth

3 Ways MCOs Can Prevent Fraud, Waste and Abuse

3 Ways MCOs Can Prevent Fraud, Waste and Abuse

A robust fraud, waste and abuse program at MCOs and MAOs includes three key areas: Technology, Clinical Audit and Investigative capability.

As your health plan grows its fraud, waste and abuse initiatives, there are three areas a strong prevention plan should address. Rather than relying on piecemeal components to combat Medicare and Medicaid fraud, as many health plans do, a comprehensive approach is best practice.

To truly be successful with a fraud, waste and abuse program, you must have three key pieces of the puzzle in place: Technology, Clinical Audits and Investigative capabilities. With all three of these tactics in place, health plans are more likely to increase their recoveries as a percentage of total claim spend. Pareo®, a total payment integrity solution, supports each of these fraud, waste and abuse capabilities.

Risks of Overlooking Common Types of Fraud, Waste and Abuse

Fraud, waste and abuse are three classifications of improper payments, which is a payment made or received in error in a government healthcare assistance program (like Medicare and Medicaid). Improper payments are more broadly combated by a comprehensive payment integrity and FWA program, as simply addressing fraud, waste or abuse alone will still miss some areas of fraudulent or wasteful activity.

According to this article, fraud, waste and abuse investigators typically focus on “two general areas: corruption and asset misappropriation.” They do this by analyzing the large amounts of big data generated by healthcare transactions. But are MCOs and MAOs doing enough in this area?

A recent report found that Medicaid insurer efforts to root out fraud, waste and abuse were disappointing. Major oversights found in their report include:

  • Failure to report offending providers to the state (allowing them to defraud other Medicaid insurers)
  • Failure to recover millions of dollars in overpayments, which could lead to increases in Medicaid rates that are based on fraudulent numbers 

Meridith Seife, a co-author of the report, said, “We are concerned anytime we see evidence that managed-care organizations are not [finding fraud and abuse and sharing it with states] in a rigorous way.” 

And a 2020 GAO watchdog report found that at least 63% of MAOs encounter data is missing national provider identifiers, data essential to tracking provider ordering, prescribing and billing habits. As the report states, “Both CMS and OIG rely on NPIs for ordering providers to conduct oversight and pursue fraud investigations.” 

As stewards of taxpayer dollars, MCOs and MAOs have a duty to thoroughly combat fraud, waste and abuse. To more effectively manage FWA in your organization, you need to be sure your solution has the following three key components:  

1. Technology 

According to CMS, Medicare improper payment rates have been steadily decreasing while Medicaid’s have been rising. The robust fraud initiatives in the Medicare program have been given the credit for the progress there, and their strategy was recently updated with the aim of further improving the effectiveness through 5 pillars:

  1. Stopping bad actors
  2. Preventing fraud
  3. Mitigating emerging programmatic risks 
  4. Reducing provider burden
  5. Leveraging new technology 

Steps to ensure proper payments have been taken in the Medicaid program as well. In 2018, under the third pillar of its reform initiative – integrity and accountability – CMS announced audits of state programs to review Medicaid enrollee eligibility as well as if programs are correctly reporting medical loss ratios. It also promised to leverage increased data sharing and analytics and step up provider education of proper billing practices. In 2019, it released a new rule intended to prevent known fraudulent providers from billing government insurer programs. 

The continued focus of CMS on eliminating Medicaid and Medicare fraud, waste and abuse means that health plans need to properly utilize technology to gain transparency into their process. Health plans and managed care organizations have to connect large volumes of data in order to comply with fraud, waste and abuse regulations. Advanced technolog– especially a solution that leverages applications of A.I. like deep learning – can integrate and process large amounts of data and to identify anomalies and patterns more effectively than people can do alone.

Technology-enabled fraud, waste and abuse solutions can quickly turn things around for MCOs – especially if they are plugged into a larger, more integrative payment integrity system. Documentation, risk identification, referrals, audit preparation and reporting are all capabilities that a robust fraud, waste and abuse technology solution can provide to MCOs and health plans.

 2. Clinical Audits 

The second element a preventive fraud, waste and abuse program needs to have is the ability to perform clinical audits. This gives MCOs and other payers and health plans the ability to review claims that have been flagged as potential fraud, waste or abuse cases. Clinical audits determine if diagnoses, prescriptions, encounters, procedures, and more are worthy of further investigation or not. An internal audit program is beneficial to a health plan, as it can often be used to prevent improper payments from occurring in the first place. 

While retrospective provider audits can unnecessarily stress the payer-provider relationship, that doesn’t have to be the case. By connecting real-time data between providers and payers, streamlining the medical records request process, seamlessly coordinating with vendors to prevent overlap, and transitioning more efforts prepay – all supported by Pareo – you can work to mitigate provider abrasion while satisfying compliance requirements.

Notably, the ability to analyze claims data and use predictive modeling allows a health payer, plan or MCO to effectively safeguard against fraud, waste and abuse. Tech-enabled clinical audits, like those performed within Pareo by a services vendor, can either complement or supplement the internal efforts of a health payer as well as those performed by Recovery Audit Contractors (RAC) for outlier billers. However, RAC audits have reduced significantly in recent years as payers move away from fee-for-service arrangements.

 3. Investigative Capability 

Increasingly, health plans, payers and MCOs should arm their FWA programs with a strong investigative arm in order to protect against non-compliance. If fraud is suspected, investigative capabilities allow direct reporting of fraud schemes to the appropriate authorities, providing evidence that supports (and protects) health payers and taxpayer funds. Should a claim be taken to court, both the evidence and a documented FWA process within a health insurance organization prove invaluable.

The amount of big data collected by healthcare organizations presents incredible opportunities to those invested in fraud, waste and abuse prevention. Governmental agencies are now using big data to investigate and prosecute FWA offenders. Mike Cohen, an operations officer with the OIG’s Office of Investigations, explains that “data…creates a pyramid effect, and we can go to the top of that pyramid.” And as fraud schemes grow increasingly sophisticated, the evidence data must evolve along with it. 

Pareo® supports investigative activities across a healthcare organization’s data ecosystem, and ClarisHealth staffs a team of expert-level investigators, a benefit which also supports cyber threat intelligence efforts. As with clinical auditing, the investigative component of a fraud, waste and abuse program can be  partially or completely outsourced to ClarisHealth.

Request a Fraud, Waste and Abuse Presentation 

Does your organization have what it takes to effectively prevent Medicaid and Medicare fraud, waste and abuse? Find out by requesting an FWA presentation from ClarisHealth, where a member of our team will discuss your specific needs.

Learn more about the ClarisHealth 360-degree solution for payment integrity and FWA, Pareo Fraud: Case Management and Detection powered by A.I. here.


Talk to ClarisHealth about how Pareo®, a total payment integrity platform, is driving innovation at health plans. 

Deep Learning and You: The Answer to the Complexities of Healthcare Fraud Detection

Deep Learning and You: The Answer to the Complexities of Healthcare Fraud Detection

When your health plan looks for an A.I.-powered solution for the SIU, here are 4 reasons deep learning should top your list. 

If your health plan is looking to improve your fraud detection capabilities – by reducing false positives, finding more missed opportunities or reducing manual processes – you’re probably considering adopting more advanced technology. When exploring your options, no doubt you’ve found mentions of “artificial intelligence” in this area intriguing. We’ve written before about how you can determine if these claims are valid or not, and while we have provided some insights on what different applications of A.I. may offer to healthcare, we want to get more specific about the ideal match for fraud detection: deep learning. 

So, what is deep learning, and why is it particularly suited to the complexities inherent in predicting instances of potential fraud?  

The Confusion Around A.I.

With the significant challenges facing the SIU, it makes sense that progressive health plan fraud organizations would be interested in emerging technologies that harness the power of A.I. But the exaggeration of the technology in pop culture leads to many misperceptions of what it is and how it is used. We tend to think of A.I. as a sort of a ubiquitous solution to a wide variety of problems without a lot of thought to how it works.  

Instead, our focus should pivot to the underlying methods that drive A.I. This wide variety of methods handles decision-making problems – primarily classification and prediction. And no application of A.I. is better at handling the complex situations presented by healthcare fraud than deep learning. 

What is Deep Learning?

Deep learning is a subset of A.I.-based technology methods. It relies on advanced technologies like machine learning, neural networks and supervised/unsupervised methods to process structured and unstructured data in a way that mimics human intelligence. Deep learning methods are used for detecting financial fraud, predicting healthcare costs and outcomes, and for smart home devices translating the signal that is your voice into recognizable inputs.  

We are not yet at the point where we can ask a voice assistant, “Is this claim fraud?” But the methods that allow the device to recognize the questions we ask are some of the very same methods that we can apply to fraud, waste and abuse detection in the highly complex healthcare data space. Though, at this point, it still requires a fair amount of subject matter expertise to accomplish that effectively (more on that below). 

Deep learning for healthcare fraud detection: peer-to-peer provider comparisons

Fraud detection models utilize raw claims, audit data, and demographics of providers and patients. The models use many derived features that involve peer-to-peer comparison of providers: 

  • Overutilization or misuse of services 
  • Provider share of volume and claim dollars in their peer neighborhood 
  • Provider share of patient costs 
  • Discrepancies between the diagnosis and the treatment 
  • Patient and provider geography 
  • Recency of billing to the payer 

The Complexity of Healthcare Fraud Detection Meets Its Match 

As we think about the FWA detection analytics problem – and the benefits that this broad class of deep learning methods and technology bring to that problem – it may be helpful to contrast healthcare FWA against other industries that have had success with the application of deep learning analytics to the detection problem. As you might recall from our previous article on the topic, the comparison that most often comes to mind is credit card fraud, despite a huge difference between the two: complexity.  

Challenge: Very little structured, historical claims data labeled “fraud” 

Even if financial institutions didn’t detect credit card fraud at the moment of the transaction, the fraudulent activity would still be found upon consumer evaluation of the monthly statement. That confirmation data, along with the exponentially larger numbers of credit card transactions vs. healthcare claims, provides financial institutions with reams of structured, labeled historical data to train fraud models. 

On the other hand, even upon close inspection of EOBs provided to consumers, these documents lack the specificity and understandability to root out healthcare fraud at this level. This difference in the historical data leads to the fundamental problem with advanced analytics, which is the problem of making correct classifications. How accurately can you identify fraud as a fraud, and a non-fraud as a non-fraud? 

Solution: Deep learning can handle vast amounts of complex, unlabeled, unstructured data 

While some information useful for fraud detection comes in the form of relatively structured claims, most data must be mined from free text clinical information. It’s incredibly inefficient for investigators to read through pages of information and reference several applications while looking for the needle in the haystack that is potential fraud. One of the best use cases for deep learning is taking unstructured clinical documents (notes, transcripts) and converting it to structured concepts with actionable information.  

Moreover, the unsupervised deep learning acts in a way human expertise cannot. We know there’s probably fraud in the data, yet we’re not sure what it might look like when we find it. Which leaves us in a position of being unable to say, “I’ll know it when I see it.” But rather, “I’ll say and do it after I know that I’ve seen it.” Making these connections is what deep learning excels at, and it performs even better at this task with massive amounts of relevant data – beyond what limited historical claims information can provide. 

Challenge: False positives are expensive to handle

Even with enviable amounts of structured, labeled data, credit card fraud detection still averages a 90% false positive rate that no health plan could reasonably accommodate. But, with financial institutions’ largely hands-off approach to resolution, they are easily able to keep their costs low.  

SIU teams are far too under-resourced to sift through that many red herrings in search of legitimate fraud. Plus, in the effort to prove or disprove a fraud lead, the investigator likely will need to request additional information from the provider at some point – and risk unwanted abrasion in the process. Health plans understandably want to maintain these valuable relationships, and the vast majority of providers submit legitimate claims with the best of intentions, so false positives are a business risk for healthcare more so than other industries. 

Solution: Deep learning processes data in multiple layers and dimensions to fine tune results 

To offset the risk of false positives, healthcare fraud detection solutions must go beyond the “outlier analytics” version of deep learning that financial institutions can leverage. It’s a tradeoff between sensitivity (catching more potential fraud) and specificity (catching only very specific fraud). 

We don’t have a way to directly measure the accuracy of the analytics that we apply in unsupervised learning methods, so we have to use some indirect measures to make sure that the algorithms that we apply are accurate enough to not exacerbate the cost considerations of false positives. Deep learning is able to transform the space of the underlying data and go beyond provider anomalies through multiple layers of processing. As a result, you can better balance between false positives and false negatives. 

Challenge: Impractical to develop rules for all possible scenarios 

If your SIU has been working to mitigate fraud for any length of time, you likely have a trusty rules-based system at your disposal. These systems usually come pre-populated with a number of fraud-focused rules or concepts that are run against incoming batches of claims. You may also have a data science organization – or access to resources shared with IT or other areas of payment integrity – that work to augment this native functionality with your own proprietary rules. 

This competency is foundational technology essential to any healthcare fraud organization. But the process has its limitations – especially if your data science resources are limited. Theoretically, we can build as many rules as there are examples of fraud, waste, and abuse – but at a certain point, that becomes impractical. Additionally, FWA schemes tend to stay a step ahead of the rules we build, despite our best efforts. The unfortunate result of an analytics-only system is one that can only surface known schemes.  

Solution: Deep learning evolves to “get smarter” over time without explicit programming 

To the extent that we can build a base of rules is always useful, but advanced analytics based on applications of A.I. are what modern fraud data science organizations need to accelerate their capabilities. Fraud detection methods based on deep learning can, for instance, scrutinize links between multiple providers to see if an unethical or illegal activity associated with one provider may also be practiced by another. These advanced analytics can also explore demographic, geographic and other data about members and providers to double check or justify claims that may raise a red flag – all without being explicitly programmed to do so.  

Deep learning fraud detection solutions are also not limited to specific points in time. Historical information augments and informs the new data arriving and updating daily to get smarter over time. As a result, you do not need to update models every time new information comes in to identify short-term spikes and longer, more subtle trends in the data. This is where advanced technology excels for the SIU.  

Challenge: Healthcare fraud too complex to eliminate the human element 

Like all areas of health plan cost containment, fraud organizations are being charged with doing more with less. But with the high stakes of a fraud investigation and the complexity of fraud data, decisions often come down to value judgments. How can you make the most of your expert resources so they know which leads to focus on and why?  

Solution: Deep learning can leverage domain expertise 

Deep learning won’t replace your fraud team. Rather, it informs them by reducing the need for manual, repetitive work. It gives your team the ability to prioritize leads, focus on strategic analysis and improve accuracy. Moreover, the domain expertise that investigators and fraud organizations have spent years developing is an excellent input to more finely calibrate the results of algorithms based on deep learning methods. When investigators work alongside a deep learning system, their reviewing and labeling results improves and optimizes the analytics.  

Deep Learning is Key to Pareo Fraud: Detection 

For all of these reasons, deep learning is the core application of A.I. selected by ClarisHealth to power our Pareo Fraud: Detection solution. It’s able to dig deeper and wider into your data to find those very subtle changes and nuances in provider and member behavior that may indicate fraud, waste or abuse that even expert investigators may not be able to find on their own. Our 4-tier scoring methodology – configurable by health plan priority – allows the models to self-learn and evolve so that the false positive rate reduces with each round of provider scoring and data ingestion. 

Pareo Fraud: Detection is also able to transcend the explainability issue that creates confusion in many A.I.-based systems. It was designed to address this “black box” gap through a specialized Super Score output – a separate deep learning method – that provides insights into results to engender trust by users. And because it’s built onto the Pareo platform, it creates an end-to-end payment integrity solution that fully eliminates gaps in a health plan’s cost containment continuum, from prepay to post-pay. 


Talk to ClarisHealth about how Pareo®, a total payment integrity platform, is driving innovation at health plans. 

How can the SIU find value with A.I. for fraud detection?

How can the SIU find value with A.I. for fraud detection?

In partnership with NHCAA, ClarisHealth is hosting a live, educational webinar on Tuesday, August 11 to talk with health plan fraud teams about how advanced technology – specifically solutions that employ artificial intelligence – can offer value to the SIU. 

We sat down with the experts who will be presenting the webinar – Mark Isbitts, vice president of program integrity for ClarisHealth, and Kyle Cheek, the director at the Center for Applied Analytics and Clinical Associate Professor of Information and Decision Sciences at the College of Business Administration at The University of Illinois-Chicago – to get a quick preview of the session. 

LIVE Webinar Tuesday, August 11 – 1 pm CDT | 2 pm EDT 

Read the Transcript

AB: In partnership with NHCAA, ClarisHealth is hosting a live, educational webinar on Tuesday, August 11 to talk with health plan fraud teams about how advanced technology – specifically solutions that employ artificial intelligence – can offer value to the SIU. 

We sat down with the experts who will be presenting the webinar – Mark Isbitts, vice president of program integrity for ClarisHealth, and Kyle Cheek, the director at the Center for Applied Analytics and Clinical Associate Professor of Information and Decision Sciences at the College of Business Administration at The University of Illinois-Chicago – to get a quick preview of the session. 

Both Mark and Kyle have worked with and in health plan fraud organizations for years. Mark provided insights into three major challenges facing SIU teams: 

04:17-04:29 Mark “The bigger issue is they need better leads. They need to reduce the false positives and get more accuracy around what they’re facing.” 

05:13-05:30 Mark “The whole process of … or a true trend.” 

08:33-08:54 Mark “Classic example is … not getting any more resources.” 

10:54-11:05 Mark “Only addressing the big dollar cases. And that’s really at the heart of it. Spending a lot of time without getting much in return.” -11:10 “That’s why getting the false positives down is so critical.” 

AB: With these seemingly insurmountable challenges of too many false positives, not enough resources to dedicate to hitting escalating targets, and a lot of manual processes to connect the dots on leads and weave together a workflow, investigators have an idea of how their day-to-day could improve. According to Mark, it’s about prioritizing the data and presenting it in context: 

12:00-12:28 Mark “Ideally, they want to be told what to look for: here’s a provider, here’s why they’re abnormal, here’s how big a deal it is, here’s how sure it is a real case.”  

AB: SIU teams are concerned about providing value and proving that value, but with no way to trend the data, they might be overlooking potentially big cases: 

15:05-15:35 Mark “If someone’s really good at gaming the system, they’ll just be upcoding from a level 3 to a level 4 office visit every other patient. What’s that, $50? Not enough to notice over a short period of time, but it really adds up over a year.”         

AB: And, as Kyle reminds us, it’s not just ROI that’s at risk when investigators are making value decisions on whether or not to pursue a case. 

22:20-22:59 Kyle “Health insurance company has a counter-balancing interest in maintaining goodwill in its provider network. Fraud mostly driven from the provider side. SIU challenge is not only do they have to build a case strong enough for law enforcement but strong enough to sell internally and consider how it affects other business units.” 

AB: Healthcare data – both in terms of quality and quantity – has traditionally been a huge hurdle in fraud teams’ abilities to take advantage of advanced technology, which sets it apart from other industries’ advances in nearly hands-off fraud detection. 

23:48-24:05 Mark “Healthcare fraud is so grey where something like banking fraud is black and white. Those grey areas are why you need different detection systems than what they have.” 

26:00-26:22 Kyle “It’s just a function of the way the business works that … If CMS would just put a checkbox on the 1500 form that says is this claim fraud, yes or no, that would make things a lot easier, but it doesn’t exist.” 

26:50-27:23 Kyle “It’s the difference between being able to measure the accuracy of your models and not being able to directly measure … Healthcare falls as far on that spectrum as they can be towards not having labels on the data and having to do unsupervised analysis and really being an exercise of feeling around in the dark as best as possible without having traditional tools to measure analytic performance by.” 

AB: Because healthcare data presents so many challenges with its complexity and lack of consistencyadvanced technology offers distinct advantages, but A.I. is easily misunderstoodWith his expertise in data science, Kyle has some theories as to why. 

32:05-32:33 Kyle “Everyone focuses on the “I” and I think the emphasis should be on the “A.” The key qualifier is artificial. We’re trying to build processes that emulate something, and the thing we’re trying to emulate is intelligence. And most of us don’t have a good definition of intelligence – cognitive scientists don’t have a good definition of intelligence.” 

AB: And it’s not just common definitions that are lacking. The way applications of A.I. work, they’re really good at producing superior results, but they can resist simplistic explanations.  

36:25- Kyle “Where we think of traditional models, we tend to think of those in a linear context. It’s easy to say, ‘If a provider’s use of a certain procedure goes up, the probability they’re committing fraud goes up. In the case of neural nets – of multi-layered, sequential, recursive, iterative processes – we can’t simplify the causal links that way.” 

41:21-41:35 Kyle “It’s pretty easy to make improvements in the output of the analytic engine, but it’s hard to do so in a way that maintains the transparency in why the result was produced.” 

AB: How can models for healthcare fraud detection that take advantage of A.I. overcome the challenges with healthcare data, avoid provider abrasion and offer explainability for investigators so they can focus on their best work, trust the results and prove value for the SIU? Mark and Kyle will be discussing these hot topics and showing some real-world examples in this interactive session. 

Please sign up today to attend the live, educational webinar Demystifying A.I. for Healthcare Fraud Detection: Finding Value for the SIU on Tuesday, August 11 at 1 pm CDT. Find the registration link in the description below.