Using Artificial Intelligence to Expedite the Electronic Benefit Verification Process

Consumer technology continues to evolve at an incredible rate—from virtual reality experiences that drop users into exotic locales across the universe to artificial intelligence (AI)-enabled, self-driving cars that reduce auto accidents. The healthcare industry, too, is at the forefront of innovation, and is looking to apply these latest and greatest consumer technologies to enhance our understanding of disease and streamline how we approach patient care. Investment in healthcare technology is at an all-time high, with private equity and corporate venture capital in the digital health sector growing 74% in the first three quarters of 2016 compared to the same time period in 2015.1 In healthcare, consumer technologies coupled with the right expert resources can help tackle some of the industry’s toughest challenges and provide high-quality care.

As the industry shifts towards personalization, medication is becoming more complex and treatment plans are increasingly utilizing specialty prescriptions. Due to the relatively high cost and intricacies associated with distributing, storing, and administering specialty medications, insurers often scrutinize these prescriptions more closely. The ability to verify a patient’s insurance coverage for these medications in a timely and accurate fashion is an ongoing challenge for both prescribers and patients. While the need to verify benefits is not new, the demand for a more efficient method is growing.      The benefit verification (BV) process is a critical step in a patient’s treatment journey. Occurring after a provider gives a prescription, but before the patient is able to receive medication, the BV checks a patient’s eligibility, determines product benefits under that specific insurer, looks at how benefits are coordinated, kicks off a quality review, and provides qualitative guidance. While several methods are available today for verifying pharmacy benefits electronically, millions of benefits under the medical plan in the United States are still verified manually—through phone calls and paperwork exchanged among provider offices, payers, and third-party hub support services providers. This process is cumbersome and inefficient.

The Flaw with Current eBV Systems

To streamline the BV process, increase accuracy, and reduce costs, manufacturers and their partners look to technology. Referred to as electronic benefit verification (eBV), the goal is to accomplish each of the core competencies of the BV process without reducing their effectiveness. Traditional eBV systems are logic-based and rely on a set of specific coverage rules that are determined and programmed into the system by people. These parameters reflect the governing rules and other factors that guide decision-making by hundreds of private and government medical health insurance plans related to any given medication.

Logic-based eBV approaches rely on humans to manually—and correctly—program product coverage on a payer-by-payer basis and there is often a heavy reliance on the manufacturer or hub to provide coverage rules. However, the complexities associated with specialty medications result in an insurance landscape that is difficult to navigate with changes to coverage rules and requirements often changing without notice. Logic-based solutions represent a static snapshot of the coverage landscape at a specific point in time and can quickly become outdated and inaccurate.

Until now, no eBV system fully addressed each part of the medical BV process across all products and payers. There are two major reasons for this. The first is the level of complexity for both payers and products. Products in new treatment classes or introduced in new sites of care create uncertainty in the BV process. The second is heavy reliance on custom integration between point solution providers that do not aggregate all components into a single view. As a result, manufactures have the experience-added costs for hub service counselors to pull additional information and aggregate it by hand, which is time-consuming and leaves room for error.

How to Determine the Best eBV Solution for You

One innovation that shows major promise for eBV is AI. Instead of relying on manual input from humans, an eBV system powered by AI technologies with machine learning capabilities uses massive amounts of historic data to “train” the models, which, in turn, produces a coverage answer. If the system isn’t confident in the coverage information, it triggers intervention from a benefit counselor to phone the payer to reconcile the coverage information, the results of which are captured by the system and used to retrain the models with new data. The algorithms in the machine learning system continuously analyze new input, identifying trends, recognizing the emergence of new rules, and eliminating the need for manual reprogramming.

As eBV solutions evolve and become relevant for a wider array of products and patients, they become increasingly attractive for manufacturers to invest in. When evaluating an eBV solution, manufacturers need to consider three major factors:

  1. Can my eBV solution consistently produce trusted results? Uncertainty around treatment costs often result in delayed or abandoned therapies, and knowing how quickly the eBV technology reacts to plan changes is crucial in giving patients and providers the confidence they need. The best solution will prioritize quality over automation by leveraging human-machine interaction to force manual verifications when it knows it needs to learn more.
  2. How do I incorporate eBV into my overall reimbursement offering? No matter how great the technology, remember that a fragmented BV experience will result in poor adoption. It’s critical to ensure a seamless integration of the eBV solution into existing workflows so that there is no difference between the automated and manual benefit summary. Consider including prior authorization support, appeals management, and billing and coding support as part of the offering as well.
  3. What’s my ability to monitor ongoing BV quality? Even though AI-enabled eBV systems learn adaptively on their own, they are only as good as the data they receive. We know that humans make mistakes, and it’s essential that the eBV system includes built-in mechanisms for quality monitoring of both automated and manual BVs within a structured data environment.

Looking ahead, a variety of ongoing advancements will help strengthen AI-based eBV systems’ ability to accurately and reliably predict complex benefit-related outcomes. These innovations promise to drive improvements in the speed and precision of the predicted outputs. Technology will undoubtedly improve a variety of healthcare processes, including BV, but it is equally important to remember the indispensable value of the human element. By blending high-tech processes with the smart touch of human intervention, we can transform the healthcare landscape and create healthier futures for all.


  • Amy Jones

    Amy Jones serves as Director, Product Strategy e-Technologies for Lash Group (Fort Mill, SC), an AmerisourceBergen company. She leads business efforts to create next-generation patient support services that leverage artificial intelligence and machine learning. 


You May Also Like

Gadgets to Help Make Your Summer a Blast

It’s time to have some summertime fun. Whether to you that means beaches, camping, ...

Optimize Content Quality Globally

With anticipated launches of novel therapies and increasing sales for currently marketed prescription products, ...