January 1, 2023, is around the corner, and we’ll enter the new year older, wiser, and, here at Turquoise, armed with another year’s worth of data shenanigans. 2022 brought us payer machine-readable files, Good Faith Estimates, and further clarifications on specific components of the No Surprises Act. Back by popular demand, let’s take a trip down memory lane to see what insights we uncovered.
Transparency Data Hit a Payer Growth Spurt
You may be sitting there thinking, “Wow, all these people do is talk about how huge the payer data set is. They’re probably being hyperbolic for likes and subscribes.” We’ll let you be the ultimate judge of that. First, a baseline measurement: since 1/1/2021, when hospital transparency data first became publicly available, we’ve amassed about 3 terabytes of the stuff. “That’s real cute,” scoffed the payer data when it burst onto the scene on July 1, 2022.
The floodgates opened rather rapidly as our friends at AWS were quick to notice via our server activity. Five months into our extraction and parsing, we’ve more than doubled the number of payers in our data set to the tune of 163 total. That very casually translates into 630 terabytes of payer data and counting…which is a modest 20,090% increase in size from the hospital data.
For you visual learners, think of the hospital data as a bowling pin and the payer data as a giant sequoia tree. The results housed in that sequoia tree shake out to 78 billion price records. If price records were miles, that’d be the equivalent of traveling from Earth to Neptune and back. Fourteen times. Bon voyage!
The Transparency in Coverage Final Rule requires payers to refresh their data monthly, so Turquoise ingests payer data every 30 days. Once completed, the ingestion results clock in at about 4 trillion rows each refresh. Suppose you’re feeling young and up for a challenge and decide to count those rows one at a time. At a cadence of counting 10 records per minute, you’d be done in a crisp 76,104 decades. At that age would you still qualify for Medicare?
So yeah. That payer data set is pretty huge.
Honey, I Shrunk the Files!
Now that we all agree the payer files are mammoth, is there anything to be done to keep them as trim as possible? All payer MRFs must be published within set schemas CMS published, which could hypothetically lead a person to think all the files would have the same process to parse and download. That hypothetical person thought wrong. Within the CMS schemas, there’s flexibility in how the files are structured…and when we say flexibility, we mean we saw file sizes that varied 50-100x. The takeaway: payers can drastically reduce their file sizes by taking measures to ensure smart architecture. For example, using references to provider groups within rates, as opposed to duplicating the provider group data within each rate shrinks the overall file sizes. TL;DR: Bigger isn’t always better.
If You’re Not First, You Don’t Get Your Own Section of the Blog
Give it up one time for the first payer we detected a live MRF for, Central California Alliance for Health!
Can You Hear Me Now? Good.
There are a handful of major insurance payers in the American market that the industry sometimes refers to as BUCAH (gesundheit): Blue Cross (or Anthem), United Healthcare, Cigna, Aetna, and Humana. Curious to see if your local provider reported rates for each of the BUCAHs? There’s a coverage map for that. (Scroll up).
Spelling is Hard…
Pick a company name that’s easy to read, they said. Make it memorable, they said. Use as many vowels in a row as possible, they said.
Okay, so it’s possible they didn’t say that third one, but we self-sabotaged and did anyway. The result? Inboxes full of bungled spelling attempts, including but not limited to:
- Turqoise
- Turquise
- Turkoise (this one hurts)
- Turqose
…but Pronunciation is Harder
If you’ve ever been on a call with us, listen closely for the use of “Trice Pransparency.” It’s there, we’re sure of it.
Wishing you all a very happy, healthy, and transparent new year!