The CMS 2024 OPPS Proposed Rule, which includes a section on hospital price transparency, details specific machine-readable file (MRF) reporting requirements for drugs. The requirements include drug unit and type of measurement as separate data elements alongside gross charges, discounted cash price, payer-specific negotiated rate, and de-identified minimum and maximum negotiated charges.
The inclusion of drug prices in price transparency data is crucial to give patients a comprehensive understanding of their cost of care; however, drug reporting poses a number of unique challenges compared to other types of charges such as room and board per diems or surgical rates. In anticipation of the 3/1/24 enforcement date for the updated hospital MRF files, we wanted to offer a blueprint for the inclusion of drug rates in hospital MRFs.
T-Minus seven months!
Inconsistent Charge Descriptions
Uniform reporting of drug unit and type of measurement is transformational for drug price comprehension because, in today’s world, drug charge description inconsistencies are rampant. While this is true across other charge codes within a charge data master (CDM), the problem has heightened significance with drugs, where dosage and unit quantity matter for calculating a baseline per unit negotiated rate.
For example, if we select a single HCPCS, J9271, to study, the official descriptor for the average sale price (ASP) is as follows:
Injection, pembrolizumab, 1 mg
Note the 1 mg quantity in the descriptor above, which, if consistently utilized, would provide a per-unit negotiated rate. Upon inspection of an MRF; however, J9271 yields numerous descriptors, such as:
- inj pembrolizumab
- Keytruda 100mg/4ml SDV
- Pembrolizumab 100mg inj
- Keytruda 100mg/4ml vial
- Pembrolizumab 25 mg/ml intravenous solution
- Pembrolizumab 50mg injection
The varying charge descriptors inconsistently report dosage and administration methods. As a result, the calculation of the baseline per unit rate is muddled. It’s unclear whether it is safe to assume the facility or insurer that published the MRF added a multiplier to the negotiated rate where the units exceed 1 mg. Conversely, suppose the rate reported across all instances of J9271 in the MRF is the same amount, regardless of units stated in the descriptor. What is the best approach to identifying the correct rate on a per mg basis? The requirement in the Proposed Rule tackles that exact scenario.
Converting the Data to Tables
Without clean data, hospitals may be forced to use an interim process to establish a baseline per unit negotiated rate that can be verified across multiple data sources. We did exactly that by focusing on the foundational ASP unit and quantity to target only relevant characters in the text-based charge descriptors. Logic from a Python package can extract pertinent unit and quantity amounts from the varying charge descriptions. For example, results extrapolated from the descriptor examples above could read as follows:
Where the unit and quantity are not clearly defined in the description, the package is unable to make a best-guess assessment.
Results Assessment
We assess the results from the new unit and quantity fields in a series of checks. To pass check one, the as-displayed rate must be within a defined reasonable range created by establishing a set upper bound limit within a multiplier of the ASP price.
If a rate does not pass Check One because it falls above the upper limit, it moves to Check Two and is adjusted based on the description. The reported rate is divided by quantity in the charge descriptor, and the adjusted rate moves back through Check One.
If a rate-adjusted amount still falls out of the reasonable range, it moves to Check Three, dividing the quantity by the standard package unit quantities. Keytruda is typically packaged in quantities of 200 or 400, so each rate is divided by 200, sent back through check one, and, if the results are still above the upper limit, the rate is divided by 400.
Of the nearly 45,000 individual CDM descriptions for Keytruda utilized in this process, almost 80% passed one of the three checks. The remaining 10% could not be assessed, for such reasons as the charge description did not contain any unit or quantity information, or the results from the assessment continually exceeded the set upper bound.
Conclusion
The Proposed Rule confirms that visibility into the cost and price changes a drug incurs prior to getting administered to a patient benefits every stakeholder. Transparency in Coverage had similar MRF requirements for drug reporting; however, the requirement remains delayed. There’s no reason to exclude the publication of drug rates from the 2024 Final Rule until a perfect solution exists. The approach we have taken thus far has already proven to be transformational among beta users, and introducing the data to other stakeholders within the drug ecosystem allows those stakeholders to help fill in gaps in assessment results. The result will be a definitive source of truth regarding the cost and price of drugs. The ramifications of a definitive source of truth extend to patients as consumers of healthcare as well. Consumers will have a much easier time understanding how much a drug will cost given their specific health insurance and provider of choice.
As rates become public, payers and providers will be both accountable for displaying accurate data in their MRFs. The accountability can serve as motivation to streamline drug code descriptions and ensure they are reported with equivalent doses. Curious about what gaps this data could help you fill? Holler at us.