Highmark has been considering its next era of AI-driven member engagement, putting a spotlight on what members may say in passing.
Emily Iacolo is Highmark’s vice president of customer data, insights and guidance product management. She told Becker’s that the insurer has been focused on defining an “enterprise source of truth” with its data. She said one year ago, it was common to point fingers at the quality of the AI model itself when something would go awry. Now, Ms. Iacolo said there has been a pivot to data integrity and reliability.
Ms. Iacolo said “absolute data integrity” is largely unachievable, but Highmark is focused on ensuring strong confidence and trust in its internal data, then leveraging that for a more “proactive” member experience.
“We have many sources of truth for foundational information — name, address, eligibility — and so we are trying to figure out how to stitch those together and get the right source of truth established,” she said. “… [If] we want to deliver a more proactive experience, we know that we can count on the foundational information being correct, and then rinse and repeat with, I call it, ‘more complex information,’ like diagnoses, medications, communication preferences.”
Highmark’s near-term AI efforts are centered on facilitating outbound communications, something health systems have been getting in on as well.
“We’re trying to figure out what are the right pieces of information that we need to better serve a member in the moment where they are, and give them the communication that’s most relevant for them in a given moment,” she said, rather than sending irrelevant or incorrect messaging.
Looking ahead, Ms. Iacolo sees an untapped opportunity with AI picking up on the “hidden context” in member calls or chats. For example, a member could call about medication adherence issues or trouble getting to an appointment. Ms. Iacolo said Highmark has historically not been able to extract that information and look at it in relation to the company’s data products. She said there is potential for this information to be reviewed alongside other member insights, including social determinants of health, plan eligibility and prescription patterns.
While Ms. Iacolo acknowledged that some people may think extracting that information would not be difficult, she said the real challenge is referring back to that source of truth and acting upon that issue in near real time. For example, she said a proactive next step could be calling an Uber for the member. Ms. Iacolo noted that Highmark has not committed to that particular use case but emphasized the underlying sentiment.
“It really gives us a way to let AI be more deterministic instead of probabilistic,” she said.
At the Becker's 5th Annual Fall Payer Issues Roundtable, taking place November 2–3 in Chicago, payer executives and healthcare leaders will come together to discuss value-based care, regulatory changes, cost management strategies and innovations shaping the future of payer-provider collaboration. Apply for complimentary registration now.
