Alternative Payment Models Do Not Increase Interoperability

"The policy efforts in the HITECH/MACRA era, from technical requirements in EHR certification to payment incentives in APMs, did not deliver on the goal of national interoperability by 2018."—Journal of the American Medical Association

A new investigation from the Journal of the American Medical Association (JAMA) has posed two questions: how are hospitals progressing on EHR interoperability and do alternative payment models (APM) help facilitate. The answers? Slowly and no.

Less than half of the nearly 4,000 hospitals studied between 2014–2018 have achieved interoperability across all domains. Hospitals cited “technical and governance challenges”—including EHR vendor nuances and none of which APMs alone can solve. Payers must take note given the proliferation of not only APM models but health-plan-driven point-of-care solutions that generate data and require EHR workflow integration to deliver savings and efficiency at scale.

The JAMA study was grounded in existing research and measures—specifically the American Hospital Association (AHA) IT Supplement which, since 2014, has tracked the interoperability domains defined by the HHS Office of the National Coordinator for Health Information Technology (ONC). These are the ability to: query, send, receive, and integrate EHR data. As the JAMA results state: "Congress designated 2018 as the goal for nationwide interoperability, and policy makers hoped that aligning financial incentives via alternative payment models (APMs) would help achieve that goal."

The results

The JAMA researchers compared "interoperability progress … between hospitals that did and did not participate in APMs." The study spanned three APM types: patient-centered medical homes (PCMH), bundled payments, and accountable care organizations (ACO). The study included diverse hospitals to prevent bias based on size, location, EHR vendor, and other factors. The interoperability domains themselves were "functionality-based … [and] technology agnostic."

Only 45% of hospitals demonstrated interoperability across all four domains. And while interoperability was higher by 2018 for APM hospitals (55.4%) versus non-APM hospitals (37.2%), these results were lower than expected and the pace of interoperability growth was similar for non-APMs. While APMs did contribute to a small increase in "interoperability engagement," they fell well short of Congress’ expectations for both nationwide interoperability and the role innovative financial incentives would play to help achieve it.

The ability to query for and integrate data lagged the most—domains associated with unplanned care and more advanced information exchange, respectively. Unplanned care is often the most expensive and the most likely to be not only out of network but beyond the reach of an APM contract.

Why APMs did not accelerate progress

The assumption was that there would be a natural connection between growing interoperability and "risk-bearing population-based and episode-based APMs" (e.g., PCMHs, bundled payments, and ACOs). The JAMA study, however, draws a direct link between weak/non-aligned incentive design and hospital interoperability barriers, specifically:

• Weak APM incentives are not sufficient to drive related interoperability investment or behavior change.
• Competing focus on other initiatives, preventing APMs from having a stronger role in driving improved interoperability.
• While care coordination is common to APMs and interoperability alike, the study notes it doesn't generate sufficient ACO cost savings or hospital revenue to create a meaningful link.

"Technical and governance issues" were the most common interoperability barriers. Examples include cross-vendor data sharing, partners unable to receive data, and inability to match providers and patients across different data sets. The JAMA study concludes: "APM hospitals may encounter these barriers more often, as they have incentives to fill in information gaps for care coordination. This suggests that value-based payment models aligned financial incentives for sharing data, but technical barriers hampered interoperability progress."

How to make progress

Despite the disappointing results, it is unlikely the U.S. healthcare system would have made as much progress without federal mandates. These must continue, argue the JAMA authors, and through "ongoing 21st Century Cures policy making"—including creating application programming interface standards (API) that help reduce friction between different EHR vendor systems.  

API is an example of the patchwork of EHR workarounds that have proliferated since the 2009 HITECH Act (Health Information Technology for Economic and Clinical Health) first mandated EHR adoption and interoperability. And understandably so. In technology, the first solution is rarely the best solution, and minimum viable products are needed to test solutions in a cost-effective way that can eventually be improved upon and scaled.

But just as data alone is not power, connectivity alone is not progress. It is not a given that more accessible information leads to better decisions and thus, downstream improvements to quality, access, affordability, and equity. Similarly, better opportunities to build interoperability through APMs will only be as good as the incentives themselves.