Health insurance companies have been removing prior authorization requirements and implementing new technologies to speed up the precertification process, while facing growing calls for regulation on how carriers use medical management tools and artificial intelligence.
Many companies have announced they are removing some preapproval requirements with the stated goal of relieving clinicians’ administrative burden.
Related: Are insurers using tech to automate claims denials?
Florida Blue has taken a different tack. The nonprofit insurance company in early 2022 deployed AI software from the vendor Availity to automate some preapproval requests from providers. Florida Blue has since relied on AI to review more than 1 million prior authorization requests related to advanced imaging and hip, knee and other musculoskeletal services for its 6 million Medicare and commercial enrollees. The company says it does not use AI to deny prior authorization requests, but rather to speed up the process for the most common prior authorization requests that are most approved.
In an interview, Chief Customer Officer Peter Serio discussed how the company uses AI and whether it is anticipating greater regulation of the technology’s use in healthcare. The interview has been edited for length and clarity.
How does Florida Blue use AI?
Prior authorizations are a big deal, as we all know, and the process can be manual, and there can be back and forth between us and the providers. If it doesn’t get resolved in a timely fashion, it can get in the way of care.
Through our partnership with Availity, 75% of the prior authorizations in that program are authorized within 90 seconds, in the doctor’s office, in real-time. That leads to improved speed, efficiency and accuracy.
What have you learned through this program?
A lot. It shows how we can use AI to augment what we’re already doing and free up capacity to focus on more important things. There are a lot of win-win-wins to be had when [AI is] applied to the right problem.
The satisfaction from the providers is through the roof, and from the members as well. Part of the way we can measure that is through the number of appeals that we have, and then how many of those appeals are overturned. Those numbers are dramatically down.
Keep up with one the industry’s fastest growing sectors. Sign up for Digital Health Intelligence.
The cycle time is another big piece of feedback we get; just being able to do it there in the office makes a world of difference.
The ultimate goal isn’t to just implement AI. It’s to improve those key performance indicators, including whether there are any issues that arise downstream that we have to resolve related to an authorization. We’re seeing those numbers continue to mature.
How does Florida Blue ensure it’s using AI responsibly?
There’s a lot of oversight that we have within our enterprise to ensure we’re protecting our customers’ information, as well as providing real transparency into how we’re using these tools. We take protecting our customers very seriously.
There’s a lot of governance. A leader sits on top of the program and runs a cross-functional steering committee, where we talk about all the things that are needed to incorporate AI into our business, like what technologies we are using, how we are securing data and whether we are doing it appropriately. We talk about which use cases we want to work on.
We have to use AI carefully and thoughtfully. We’re being very intentional about how we’re using it.
Do you feel like people are more skeptical of the use of AI in insurance, particularly health insurance, than they are in other industries?
I think there’s natural skepticism about the use of new technologies, and … there’s always an opportunity to either use them for good or use them for not good purposes.
I think there’s some skepticism around, “Whoa, this is a big, scary new tech.” Then, combined with just the general view of insurance, what are people’s perceptions of our motivation?
Not a Modern Healthcare subscriber? Sign up today.
What advice do you have for other insurers that want to use AI to automate prior authorizations?
Finding the right partners is a big one.
There’s a whole host of things we use to evaluate partners—like their technical chops, for lack of a better term. How good are they at artificial intelligence, or the machine learning, or the underlying technology? What is their roadmap? Are they going to stay advanced? And just like any other partner, are they really a partner that is working with you to solve your problem? Or are they just selling you a piece of technology? Find somebody that you’re going to be able to work alongside.
Ensure that you’re outcome-driven and design it from a customer’s perspective and a provider’s perspective. That’s really our North Star.
We’re not trying to implement AI just to implement AI. We recognized a problem from all stakeholders’ standpoints, and AI is a way to solve that problem. Start with your customer and your problem, and think of AI as a tool in your toolkit.
What does the future for AI look like at Florida Blue?
Streamlining and personalizing customer service is a big area of focus we have right now. For example, we’re using AI to understand customer sentiment in real time and help guide the conversation, as well as gather information from our different data sources and present it to the member or our service advocate in a logical language.
We’ll continue to use AI to synthesize big data to better understand an individual’s health status, their needs and their preferences. How do we get the care to be more personalized for that person?
What do you think the future of AI oversight looks like? Do you think it needs to be more regulated?
That’s going to be a factor. I’m not sure exactly where it will land, but there will be some. It seems like regulation is even wanted by those creating AI.
I could see the merit in doing it. It could be the Wild West without it. But how you do it is a brilliant question.
