We recently discussed value sets – the lists of codes that are used to define data elements for CQMs. You’re all probably familiar with the use of CPT codes to define services and procedures, and ICD-9 and ICD-10 codes to define diagnoses, but things get a bit trickier with medications. While procedure and diagnosis codes benefit from widespread standardized adoption, numerous medication terminologies used in a variety of contexts complicates the situation.
Most of the data we retrieve for DRVS has medication records identified by NDC codes, which are regularly published by the FDA. They represent “all drugs manufactured, prepared, propagated, compounded, or processed by it for commercial distribution.” The format of these NDC codes is often problematic: NDC codes are displayed/stored in a variety of formats that Azara normalizes to an 11 digit, no hyphen format on the way into DRVS. Sometimes medication records lack NDC codes; these records are often identified by brand or generic name from one of the several commercially available drug databases.
Getting back to value sets. When we build a measure that looks for specific medications, we must create a value set that defines those medications. For Meaningful Use eCQMs, the value sets are defined in the VSAC using RxNorm codes. Per the CMS eCQM logic guidance, these value sets should be defined with “SCD” (Semantic Clinical Drug) RxNorm codes, which means the value set will only contain generic drug codes. Don’t expect to find “Flovent” in the “Preferred Asthma Therapy” value set; instead, you’ll find RxNorm codes for things such as “Fluticasone propionate 0.22 MG/ACTUAT Inhalant Powder.”
So we have these reporting requirements that use generic drug RxNorm codes to define medication value sets and that use NDC codes and generic/brand names data to define medications. (Note: We receive some medication data already identified with RxNorm codes, though we’ve had mixed results with this data. This is a topic for a different post). Translating between NDC, generic/brand name, and SCD RxNorm codes is an interesting problem that our engineering team constantly works to improve. In our March release, we overhauled the mapping process to produce more accurate medication based reporting. We continue to develop new features to further improve the mapping accuracy for medications, and we look to create a more transparent process to manage and expose value set content.
As we’ve noted in the past, value sets can get complicated, and it’s important to note that they’re not set in stone! We encourage DRVS users to submit support tickets regarding value set content, especially related to medications, so that we can relay feedback to the appropriate value set authors. This not only improves value sets for Azara and our users – It benefits the wider community of providers that use Meaningful Use eCQMs.
Eric Gunther is an engineer at Azara Healthcare.