In April 2024 the Centers for Medicare and Medicaid Services (CMS) released their Hospital Inpatient Prospective Payment Systems for Acute Care Hospitals and the Long-Term Care Hospital Prospective Payment System and Policy Changes and Fiscal Year 2025 Rates; Quality Programs Requirements; and Other Policy Changes for public comment. Most interesting to public health, however, is a Request for Information (RFI) Regarding Public Health Reporting and Data Exchange (starting on p. 36377). To understand this fully, one needs to have pretty good familiarity with these programs. We will try to address the major RFI questions from a public health perspective.
The first section of the RFI focuses on how hospitals attest to meeting a public health measure, currently done with just a “yes/no” response. The RFI wonders whether a numerical response with a numerator/denominator, and whether FHIR queries by public health to healthcare systems also measured with a numerator/denominator would be more useful.
We wonder why there is no reference to the newly-introduced “Insights Conditions” for certified electronic health record technology. These are measures of interoperability that are incumbent on CEHRT vendors to submit to ONC every six months in aggregate for all users of their certified technology – a single set of metrics reported across all customers, all products, and all versions of the CEHRT (assuming contractual requirements do not prevent this). For public health, the Health Data, Technology, and Interoperability: Certification Program Updates, Algorithm Transparency, and Information Sharing Final Rule (HT-1) defined a set of measures for immunization reporting, where determining an appropriate numerator and denominator for measurement are straightforward (see our earlier blog post on this). It seems almost redundant to have the same data reported once by a healthcare organization and again in aggregate by EHR vendors. Perhaps CMS should more explicitly seek to align their measures with these Insights Conditions.
That being said, aside from immunization reporting, the calculation of denominators for other types of public health reporting (e.g., electronic laboratory reporting [ELR], electronic case reporting [eCR], syndromic surveillance [SS]) is problematic. You do not know the true occurrence of the disease in your patient population, for instance, to establish a baseline against which you might compare the number of cases reported (the likely numerator). This gets even murkier for, say, syndromic surveillance which may not even involve reporting of identified health data. The measures for SS will likely yield a score of 100% (all encounters sent) or 0% (no encounters sent) and nothing in between. Perhaps an initial focus on the use of standardized coding would be more impactful on the quality of reporting in these areas.
As for FHIR, we are a bit confused about what CMS is asking. FHIR is not yet widely used for public health reporting, so we have very limited experience and understanding of how FHIR APIs can improve public health reporting. It is likely that FHIR will initially be enabled in public health to support gaps in reporting rather than to replace existing reporting transactions. Examples include the use of bulk FHIR query to get data out of public health registries, and the use of FHIR-based query from public health systems to clinical care for case investigation or follow-up. Since FHIR is not yet a part of any established public health reporting it seems premature to discuss it here. Query of clinical systems by public health is equally far off, though HL7’s Project Helios does have a stream of work actively developing and testing standards for this.
CMS also asks whether it should expand public health reporting to new areas, and how it can achieve better completeness for electronic case reporting. It seems that clinical care already has a sufficient amount of reporting to do, and the best way to improve completeness of all public health measures – including eCR – is to stay focused on the currently-supported transactions and push for greater compliance with those rather than introduce new ones.
Next, CMS asks how it can incentivize or support more flexibility in how EHR technology can adapt to changing reporting requirements such as those during COVID pandemic and Mpox outbreak. This is a tough problem as EHR technology is complex and sometimes even convoluted, making it difficult to alter quickly and support adequately. But the best way to improve this nimbleness would be to continue to invest in and leverage central resources and services like the Reportable Conditions Knowledge Management System (RCKMS), which provides decision support services regarding reportable conditions, by expanding the reach of RCKMS to include other disease and condition domains. Other shared infrastructure, like the APHL Informatics Messaging Services (AIMS) and Immunization Gateway (IZG) also deserve continuing and even expanded investment.
CMS next asks a question related to certification of public health systems as a way to improve the capability of public health agencies to achieve interoperability with clinical care. One example of such infrastructure is the Measurement and Improvement Initiative that has been underway under the auspices of the American Immunization Registry Association (AIRA) and funded by CDC. With participation by NIST, this initiative tests Immunization Information Systems compliance with a variety of standards and functional requirements in a voluntary, non-obtrusive way to promote standardization and data quality improvement. A similar initiative could be initiated in other domain areas based on this successful model, though there is danger that certification will focus too much on interoperability to the exclusion of focusing on actual system functionality for agencies..
But CMS has to beware that establishing certification requirements without the concomitant funding would be a disservice to public health agencies and clinical care alike. It is also possible, if not likely, that some public health system vendors will simply choose not to certify their systems at all given the time and expense to do so. With very few choices available to public health agencies this may not improve the overall landscape at all.
Questions follow regarding incentivising early adoption of FHIR (how can you do that now when the whole CMS program is one of disincentives?), and allowing providers to receive credit for the health information exchange (HIE) objective through participation in TEFCA (why would it be excluded as currently written?). Finally, CMS wonders whether these measurement programs unduly increase administrative burden on clinical care organizations, and what can be done to minimize this burden. From a strictly public health perspective, while public health reporting is at the core of public health surveillance, it’s not clear that the measurement of this reporting in more detail helps public health in any way. CMS wonders whether a shift to public health query of clinical care might reduce the burden on health care organizations. But until the ecosystem around query of EHRs by external organizations is more mature (security, authorization to access data, capacity planning to be able to respond to queries) and public health systems are able to issue queries, this desire is purely aspirational.
Responses to the RFI Regarding Public Health Reporting and Data Exchange are due on June 10, 2024. HLN submitted its own comments as well as participated in the comment submission by HIMSS and the Pew Foundation.