What was the core problem or set of problems or situation that your team was looking to solve with the solution?
Mandal: There were two key parts to the whole solution. We had done implementations in the past that had helped us think about making this an asset. We saw that there was a missing workflow or interconnected systems, be it in the form of intake or how the intake happens, and bringing in a real dynamic dashboard and metrics which businesses look at to make smart decisions. We felt like based on our past 10-plus years of experience, that’s the missing part. Additionally, how do we really make a timely resolution to this? Once efficiency is implemented, there was a problem of can we really do it on time? This has repercussions directly to CMS star ratings, related to the Medicare and Medicaid lines of business. And 20 percent of the time, appeals reach a second level, meaning the member or provider was not happy on a particular claim, and the denial. So can we really optimize that to a much more optimal value, with the right decisioning, the right and the right documentation? That’s where we found an opportunity to introduce AI to read the data, to read the characteristics of a particular claim, and make a recommendation.
It sounds as though one of the challenges in the compliance area is that everything was exception-based and you needed an algorithm in order to move the process forward more quickly and efficiently?
Mandal: Absolutely, and there are certain augmenting aspects of this; the channel is still integrating. We have someone calling in versus mailing in or faxing; how can we standardize the channels together? We have an NLP-based algorithm that helps us to sweep all the intake into a centralized, seamless way of taking it forward. So yes, a more systemic, algorithmic way of taking those decisions forward.
Swami: And from my perspective, all the payers we are dealing with, do see a big value in this solution. Part of the CMS rating really impacts the payers and impacts the health of the plan member/patient at the time of the review.
Healthcare has been so fragmented and balkanized; one of the things that you've achieved here is pattern detection and categorization for faster, smoother processing. Is that correct? Is that what sets you apart from the competition?
Swami: Yes, those are two key features of the solution that differentiate us. A lot of the time, people are spending time triaging appeals or grievances, trying to figure out whether the clinical review is there, and what the subjective biases might be. Those are two key features of the solution.
How is the landscape going to evolve forward in the next few years?
Mandal: We started this journey almost three years back, laying our product out into an orchestration. The second generation was to introduce NLP and AI into it. We actually have a patient-centered model that’s patented. And the third generation has involved making it modular. We’ve broken it down based on what we’re seeing in the industry, to attack it from a business-value perspective. Right now, we are working around making it more interconnected, in terms of an end-to-end value, introducing for example, blockchains around information around particular members, based on plan details and historical EHR data round the patient.
Swami: The main thing is around a totally interconnected healthcare ecosystem; that’s where healthcare is moving toward. I see this as a solution that will become a digital ecosystem for members, providers, and plans together. They will see a near-real-time resolution of these appeals and grievances. That’s where the industry is going, and this will be a digital ecosystem solution for the healthcare industry.
For more information, visit: www.virtusa.com