How does healthcare pricing data move the industry forward?

Key takeaways

• Domain knowledge is crucial: without it, analysts build tools that don't address the real business problem
• MRFs are just data files with frustrating gaps—claims data is needed to validate pricing and model plans around utilization
• Claims data currently runs on a six-month lag, while price transparency data is released monthly—real-time claims would transform accuracy
• The industry needs to shift from CPT codes to episodes of care, bundling codes to represent full care pathways

In healthcare, being technically sound isn't good enough: Domain knowledge is crucial. Sean Trainor, Handl Health’s lead data analyst, learned that lesson early on in his career.

“Without domain knowledge you don't really know what problem you are trying to solve,” he explained recently. “When I was new to the industry, I ended up building things that didn’t really address the problem or business need, simply because I didn’t understand why these things were being built in the first place.”

His deep experience in healthcare analytics has also led him to another universal truth about machine-readable healthcare pricing data: “Price transparency is great, but MRFs are just data files,” he said. “And a lot of the files have frustrating gaps.”

In his day to day at Handl, Trainor is focused on how he can enrich that healthcare pricing data, putting that information into action to drive down healthcare costs by steering patients toward high-value providers.

Learn more about how Sean is all about the data:

How do you enrich MRFs when they aren't complete?

While price transparency MRFs have been transformational for the industry, and the level of detail in the required data has been steadily getting better, pricing data alone can’t drive down the cost of health care.

Claims data is a crucial piece of the puzzle that helps validate pricing, correctly model plans around utilization rates, and predict future plan effectiveness. But the claims information we have now is on a six-month lag time. Price transparency data is released monthly, but because of the lag in claims data, we don't have a good way to validate the accuracy of those prices until months later.

It’s true that Electronic Health Record (EHR) data is a suitable stand-in for claims information, as that data tends to be more real time, but the information provided within those data sets isn’t as meaningful.

To truly help push forward industry transformation, we need real-time claims data. This will take a united effort across multiple parties throughout the industry, and speeding up the submission and processing of claims will be no easy task, but this would go a huge way toward increasing the accuracy of healthcare pricing and correctly tying costs to quality.

Why aren't MRFs as groundbreaking as they once seemed?

I think the industry doesn’t correctly use price transparency data. That might be because the data taken on its face is informative, but you can’t really do much with it. It's certainly helpful in understanding the costs charged by certain doctors for given procedures, but that lacks a lot of context.

Price transparency data alone does not answer:

• Does the provider actually do the service they have a rate for?

• Does the doctor provide high-quality service?

• How much market share does a health system have in a given area? If it's too large, it's helpful to know why prices are high, but I really want to know if anything can be done about those high costs.

Price transparency data has gotten better over the last four years, but it’s still not perfect. It’s hard to find rates sometimes, and a lot of the data is incomplete and inconsistent. To move the industry forward, we really need to shift our thinking: MRFs are a foundation to build upon, not the goal.

How is AI used in healthcare pricing analysis?

AI will not change the healthcare ecosystem overnight, and we’re far off from it making a seismic impact in the industry. Why? Healthcare needs a human touch. AI should be used as an augmentation tool, not a replacement for the very human decisions that need to be made for effective plan management.

That said, we have been employing LLMs to speed up processes and enable us to react quicker to market shifts and customer demands.

How will healthcare data analysis change next?

I always consider where the data is leading the industry. We need to move toward assessing patients not by CPT codes but by episodes of care. In my job, I’m focused on figuring out how to move toward bundling codes so they are more representative of the full care pathway, paving the way for industry-wide growth of alternative healthcare plans.