CMS has released a proposed rule that will require Medicaid managed care plans and other insurers offering products on the ACA Exchanges to support the sharing of patient data amongst themselves.
The announcement follows the release of the Interoperability and Patient Access final rule in July, which requires all health plans doing business with the federal government patients’ data readily available to the 85 million patients in CMS-regulated health plans.
The agency had already released Blue Button 2.0, an HL7 FHIR-based API making Medicare claims data available to patients across multiple devices. To date, CMS estimates, Blue Button 2.0 has delivered data to more than 60 million Medicare beneficiaries.
Under the terms of the final rule, payers do not need to use Blue Button 2.0 technology, but they are required to deliver some kind of functionality that addresses this purpose. The key requirement is that the payers implement a FHIR-based API, the use of which will give new plans access to patient claim data almost as soon as they enroll. In addition to opening up data sharing amongst payers, the API can be used to send patient claims, encounter data and clinical data directly to providers’ EHRs, according to a CMS blog post outlining the proposal.
This final rule, in turn, flows from CMS’s MyHealthEData initiative, which focused on giving patients easy and direct access to their medical data. The agency has attempted to not only make patient health data more accessible, but has also worked to educate patients on the value of engaging with that data.
All well and good. However, while this stepwise approach to expanding health data interoperability seems relatively sound, particularly given that patients were given a seat at the data buffet table, one shouldn’t forget that there are still some serious interoperability issues remaining unresolved.
As anyone who has followed industry battles around interoperability closely knows, one of the trickiest rabbits CMS needs to pull out of its hat is to see that providers and vendors cooperate with the final rule prohibiting information blocking. Early last month, the ONC agreed to delay the compliance date for these provisions from November of this year to April of next year.
Of course, the omnipresence of the pandemic and the impact it has had on providers certainly must’ve played a part in its making the decision to move up the date. But it’s hardly a secret that both providers and vendors have good reason to avoid complying with this rule for as long as possible, and ONC has to have been under pressure to move up the date due to long-term concerns about these requirements.
The reality is that while there are good clinical and administrative reasons to make health data available to all appropriate stakeholders at all times, the same providers and vendors still have a proprietary interest in avoiding sharing the data that fuels their business.
Even under these circumstances, there has been some incremental progress. For example, the agency does not seem to have gotten too much pushback on its new Condition of Participation requiring all Medicaid- and Medicare-participating hospitals to ping other providers when a patient is admitted, discharged or transferred.
Still, requirements like this one don’t involve asking the hospital to turn over its core data assets upon demand. In other words, the fact that this requirement doesn’t seem to have made huge waves may not mean a lot.
Think of it this way. While there are of course important differences between, say, the financial industry and healthcare, it’s still worth noting that financial institutions would never share data of this type and sensitivity if they had the slightest choice.
The bottom line here is that while CMS can more or less force payers to share data, other players in the health data accumulation loop will be harder for the agency to reach. Not impossible, but harder.
Ultimately, it seems to me that all these efforts over the years have managed to scrape by without addressing the key questions hanging over all of these discussions. To what extent, if any, are medical records proprietary assets and if so, how should we compensate those from whom we are asking the data? If we analyze the interoperability problem in those terms, we might find providers and vendors to be more cooperative.