Merging AI and human intelligence for big recovery results

May 27, 2020 Dan Iantorno
Merging AI and human intelligence for big recovery results

Technology plays a key role in health plans’ transition to more proactive, data-driven payment integrity results. We sat down with Dan Iantorno, Chief Information Officer at Discovery Health Partners, whose team received the FutureEdge 50 award from IDG/CIO magazine for Discovery’s work with machine learning and AI. We discussed how technology is driving a transformation in payment integrity and what Discovery is doing to help clients access new technologies to improve revenue, avoid costs, and enhance their members’ experience. 

Data and analytics are transforming many industries, including healthcare. What are the biggest challenges health plans face when implementing these new technologies? 

We know that health care costs are skyrocketing, driven by administrative complexity, fraud, and abuse. It’s estimated that as much as $935 billion, or nearly 25% of total spending, is wasted in the US healthcare system every year. As a result, providers are under intense pressure to manage costs and ensure payment integrity, while at the same time continuing to provide quality care for their members.

Technologies that leverage artificial intelligence, machine learning, and analytics can enable plans to implement process efficiencies and dramatically increase recovery rates, while reducing member abrasion. But many health plans lack the internal tools and resources to identify and pursue recovery opportunities for high-cost, complex claims. Discovery is partnering with health payers to support data-driven payment integrity solutions and help them identify and pursue the highest-value cases to drive bottom-line results.

What are some of the ways Discovery is innovating to help clients transform their payment integrity approaches?

Since the company’s inception, Discovery has been at the leading edge of analytics-powered technology solutions to help clients address payment integrity challenges. Today, we are using machine learning and predictive analytics to tap into the potential of more than a decade of case outcome data to improve results, drive efficiency, and guide our customers to more proactive payment integrity strategies. Last year, we unveiled our Case Open Logic solution, an initiative that uses machine learning as part of our claims ranking process in our Subrogation practice. 

Rather than relying on human logic to prioritize the 2% of cases that result in 90% of recoveries, our solution uses machine learning to augment human intelligence by selecting the cases with the highest likelihood of success, doing in seconds what would previously have taken hours of manual work. These enhancements help validate subrogation claims faster and more accurately and identify cases that otherwise might be missed. The process also helps health plans reduce member friction because there’s no need to contact members unnecessarily on claims that are not of substance. This solution has delivered immediate results for our clients, and the technology shows great potential to enhance solutions across business lines. We’re very proud that our Case Open Logic solution was honored with IDG/CIO’s FutureEdge 50 award that recognizes cutting-edge applications of emerging technologies to advance business goals. 

What’s in store for the future of payment integrity and how will Discovery support health plans as they evolve their payment integrity strategies?

Discovery’s data science team has a history of blazing new trails in the payment integrity industry. In 2020 and beyond, we will continue to innovate with solutions that drive results for our health plan clients and provide best-in-class models for the industry. For example, our investment in machine learning with our Case Open Logic solution has already helped us boost subrogation recoveries for our customers by 20%. Based on that overwhelming success, we intend to roll out similar machine learning capabilities to other lines of business such as Data Mining, Clinical Audits, and Coordination of Benefits (COB). The application of machine learning based on a decade of data enables Discovery to provide services that are immensely valuable to our health plan clients. Combining technology innovation with our team’s decades of experience in the health payer space is how we will continue to deliver game-changing profound client value.

Find out how Discovery Health Partners can help strengthen your payment integrity initiatives in 2020. Contact us today!

Dan IantornoMerging AI and human intelligence for big recovery results
read more

5 Key roles your company needs for data analytics success

September 26, 2017 Discovery Health Partners
5 Key roles your company needs for data analytics success

What roles are necessary in your healthcare organization for successful application of your data analytics? Whatever the size, culture and maturity level of your company, there are five key functions your organization needs to develop models that can help drive solutions to real-world problems.

5 key roles necessary for creating analytics that will drive value for your organization

  1. Data Liaison
    The person in this critical role is someone who really understands your business goals—and can straddle the discussion between business and data. They participate in discussions with your business around your real-world problems and understand enough of the data to realize when a particular problem might be something that can be addressed with the data you can access. In fact, this person is so familiar with your business they might even be able to generate their own list of real-world problems you face that could be addressed with available data knowledge.
  2. Data Architect
    This key technical leadership position understands your big picture—they know what data you have, where it is, and how it fits together. They are current in their understanding of data technologies and can apply that knowledge to your organization’s plans on how it will leverage data.  They help create the blueprint for the environment(s) you need for data science and analytics.
  3. Platform Architect
    Many organizations don’t have data set up in a way that’s really conducive to analytics or big queries. In this IT role, your platform architect will work closely with the analytics team to create the infrastructure needed for effective analytics. They are the person who makes sure your organization has enough “horsepower” for the job at hand.
  4. Data Analyst
    As the extractor of data, this is the person you’re most likely to already have in your organization. The data analyst is often your go-to person for analyzing data sets and reporting results. The data analyst understands SQL, SAS statistical software, and your business goals to manipulate healthcare databases and produce analytic findings.
  5. Data Scientist
    For more advanced analytics against your data sets, the data scientist works to understand real-world problems and writes the models. They work with big data, using various technologies to develop models that convert data into actionable insights. They may also help identify new data sources and work with the data and platform architects to fuse them with other enterprise data sources. This role collaborates with the data analyst to get access to usable data and works with the data liaison to understand what the real-world problem is and build the models that ultimately help drive your value.

You don’t necessarily need five people to fulfill these functions since some of these data analytics roles can be combined. You may have an organization where your data analyst and data liaison roles are filled by the same person, or one person may serve as both data and platform architect. The key is to understand that you’re checking each of these boxes so your company is able to take a singular real-world problem and help turn it into the model that’s going to help drive value.

 

Discovery Health Partners5 Key roles your company needs for data analytics success
read more

Calculating cost avoidance: A closer look at one of 2017’s top payment integrity trends

August 30, 2017 Discovery Health Partners
Calculating cost avoidance: A closer look at one of 2017’s top payment integrity trends

This post is part of an ongoing series about trends happening within the payment integrity space for healthcare payers. This series features contributions from Discovery Health Partners payment integrity experts discussing these trends, why they’re happening, and how they affect health plans. To learn more about all of the top trends, download our 2017 Payment Integrity Trends whitepaper.

Making the business case for prepayment cost avoidance

As health plans more aggressively adopt cost avoidance as a payment integrity tactic, many struggle with the business justification. There simply is no industry-standard method of quantifying cost avoidance.

With pay-and-chase models of recovery, it’s usually pretty simple – you calculate the recovery and if you’re using a vendor, you subtract a percentage contingency fee. It works nicely in a spreadsheet formula and the extra cash looks great in your P&L. But if you’re avoiding—not recovering—dollars, how do you measure the return on investment? How do you calculate the costs avoided?

Health plans have been left to their own devices to determine the right method to quantify the business case for cost avoidance. And to compound this issue, the method of measuring cost avoidance and the business case isn’t consistent across all types of payment integrity. The calculation and return on investment will differ depending on whether you’re looking at coordination of benefits, subrogation, claims analytics, etc. Based on my experience, even among the largest health plans, there is incredible diversity of opinion on how to measure and value prepay. Read on to learn about some examples that I’ve come across.

Claims cost multiplied by estimated months of savings

This large commercial plan with over 40 million members uses average claim cost per member to calculate potential savings from cost avoidance. The plan first identified “leads,” or members suspected of having other coverage, and sent them to Discovery Health Partners to verify other coverage.

Of those leads, 10% have been confirmed to have other primary coverage. The plan estimates that it would have paid claims for those members for 6 months before catching the error. By multiplying the 6 months times a monthly claims cost per member, the plan figures it avoids more than $7 million in erroneous payments.

This method provides a general sense of the value of cost avoidance, which allows this plan to justify the cost of using a vendor as a partner for some of its prospective COB processes. Not all buy into this method, though. Some might argue that not all members would incur the average claim cost in all 6 months, and some of the costs, had they been paid up front, likely would have been recovered on the back end. This method doesn’t account for that.

On the other hand, it accounts for neither the administrative cost avoided by not having to recover on the back end nor the fact that a percentage of recovery efforts are unsuccessful. In the end, this plan felt that these balance each other out and the methodology works for now.

In another example for COB cost avoidance, one of our clients uses the average cost of claims for each member over the previous 12 months and applies that value over the next 12 months.

Costs to consider when calculating ROI on cost avoidance

Once you have identified a method of calculating the value of cost avoidance, you need to understand the costs that are involved in developing cost avoidance capabilities.

  • Vendor fees. How vendors make revenue will depend on the method the health plan uses to calculate cost avoidance. Options could include contingency, transactional (per validation), monthly, and fixed fees.
  • Resources. Subject matter expertise and operational expertise will help ensure you avoid the right costs at the right time with minimal member and provider abrasion.
  • Technology. Software and other programs allow you to integrate the data from correct sources into your systems so you can make timely pre-payment decisions. This could include applications to manage the workstream.

The move to prepay cost avoidance requires a set of skills that health plans need to develop or acquire in order to be successful. These should be considered when calculating the cost. See our infographic for a list of these capabilities.

 

 

 

 

 

 

 

 

Discovery Health PartnersCalculating cost avoidance: A closer look at one of 2017’s top payment integrity trends
read more

 

 

 

Subscribe to our blog

Receive the latest payment integrity insights and perspectives

Recent posts

Resource pages

Contact request