Hospital price transparency data assessments vary greatly. Let’s dig into why.

If you’ve followed the hospital price transparency saga since January 1st, 2021, (What’s that, you haven’t? No worries, Turquoise was up at midnight downloading machine-readable files and singing a rousing rendition of “Aude Lang Syne.”) you may have noticed several different answers to the same question: “Just how compliant are hospitals with this new rule?” We’ve seen a wide range of statistics associated specifically with the percent of hospitals that are compliant. On this episode of Turquoise Investigates, we review why numerous third party reports yield such varying results.

CMS: The Purveyor of Compliance

Before we jump into the details, a reminder: ultimately, the Centers for Medicare and Medicaid Services (CMS) is the only true gauge of hospital compliance, and as such, only CMS can deem a hospital compliant or noncompliant. We like to think they’ll even issue participation trophies at some point. Although CMS alone is the compliance arbiter, thus far they have been slow to publish a complete pass/fail hospital list. Given that delay, interested parties have been left to their own devices to make judgment calls. The lack of a centralized location showing compliance/noncompliance has paved the way for others to create narratives around what’s available within the published data.

Here’s a cross section of headlines from different groups we’ve seen:

  • “...6% of facilities covered by the rule were totally compliant" (Axios via Patient Rights Advocate, June, 2022)
  • “65% of hospitals have posted a machine readable file....1.8% of hospitals in true compliance with the rule.” (Healthcare Dive via BRG, October 2022, data from a March 2022 study)
  • “65% of hospitals have posted a machine readable file...3,234 (52%) of hospitals have published complete or mostly complete files” (Turquoise Price Transparency Impact Report, October 2022)

There’s clearly variance in what exactly the headlines are reporting. Are hospitals publishing useful data? Are all requirements for compliance met? Let’s explore some variables that drive these interpretations.

Home, Home on the Range…of Differing Requirement Interpretations

In the CMS mandate, hospitals must follow strict guidelines to be fully compliant. The placement of the patient estimate tool on the hospital’s website, the naming convention of machine readable files (MRF), and the use of plain language (language that patients can easily understand) to describe a healthcare service are all defined and required within the rule. Because of the vast array of guidelines included in the rule, certain studies, such as the PRA study, lean more on the discoverability and patient experience than the contents of the MRFs. On the other hand, Turquoise Health and BRG studies lean more on the contents of the MRFs and less or not at all on the patient estimate tool. That variance in the specific areas of focus leads to different estimates of compliance based on differing factors.

Are All Hospitals Subject to The Price Transparency Final Rule?

The question of how many total hospitals are subject to the rule is more nuanced than one might think, which we’ve found to be a very on-brand response to most questions related to the cost of healthcare. Most recently, CMS specified that state-owned psych facilities are exempt from the requirements. There are other categories of government owned hospitals that are also subject to different payment methodologies than standard acute care hospitals. In addition, there are free-standing emergency departments and hospital-owned clinics that confound requirements and leave researchers looking for clarity. To complicate things further, hospitals close, consolidate, change ownership, and sometimes have multiple locations that bill under one provider identifier. Hospitals change names and affiliations more than James Harden has in the past few NBA seasons.  

As a result, the denominator differs from study to study. For example, the BRG and Turquoise studies cite a denominator in the range of 6,200 to 6,700. Generally, all studies reference at least 5,500 hospitals, and in some cases, such as this Guidehouse report in early 2021, studies only analyze a subset of hospitals (1,000 in this case). The extrapolated findings are then used to represent all hospitals.

Inconsistent Data Updates and Schemas

Hospital MRF publishing progress over the past 7 quarters

When the press first started reporting on hospital transparency compliance in early 2021, initial findings showed hospitals were sluggish to publish charge and rate data. Although hospitals are only required to update their data once a year, Turquoise reviews the MRFs posted by all hospitals on a quarterly basis. It’s easy to see the progress in the overall number of MRFs available for download when viewed across the past seven quarters.

Many academic publications must await peer review and reference an antiquated snapshot of the data by the time reports are released. It’s helpful to check the timestamp on an article or report next time you see a hospital compliance assessment in the wild to know if you’re comparing two reports using different data from different points in time.

In addition, due to the lack of a standard for the hospital MRF, it’s extremely difficult for non-engineers to assess the contents of a new or updated file. Since there’s no standard schema for the MRF data, Turquoise writes custom python ingestion code to translate the file contents into a common database ontology. Once all the files are through the ingestion and parsing process, our team calculates attributes of the files to give them a transparency score. Attributes include reporting the full payer mix, the presence of inpatient DRG, per diem, and charge information, as well as charge and price information for ancillary services, drugs, devices, and surgery.

A standard schema for hospital MRF data is the key to unlocking innovators’ ability to create useful data products. Standardizing the schema would also likely result in increased compliance and monitoring.

Looking to The Future

We’ve seen CMS issue its first fines for non-compliance to Northside Hospital, but we have not yet seen an exhaustive naughty or nice list. An official list denoting compliance/noncompliance from CMS would lift the burden from third parties unofficially evaluating compliance. Publication of a list would also help answer the outstanding question of exactly which hospitals are subject to the rule.

This year we saw CMS issue a Request For Proposal and subsequently award a contract to begin evaluating compliance in earnest. Because of that, we have reason to be optimistic that CMS will be more vocal in 2023. With this additional help to trudge through hospital data, CMS should ultimately emerge as the definitive arbiter on hospital compliance, which would ideally lead to standardized assessments.