How 'Different Colors of Money' Derail Interoperability Efforts
Everyone who has seen the movie Contact can recall the scene where a wealthy defense contractor declares the "first rule of government contracting" is "Why build one when you can have two at twice the price?"
The Healthcare IT train went off its tracks with a pair of recent columns in industry media outlets. The first addressed the manner in which emergency medical service agencies—and the mountains of useful clinical and situational data they produce—should be able to engage the broader health ecosystem, especially via health information exchanges. A second built on that in perspective piece called "EHRs, HIEs Must Collaborate With EMS for Patient Safety." The authors advanced cogent theses when they said, in the words of the second piece, "EMS organizations must move beyond data collection and reporting toward becoming a formalized part of patient data sharing… EMS services may be so often overlooked in the HIE equation because they are one of the smaller healthcare costs, or [their] voice may not be as well represented in the healthcare conversation." Right on all counts there.
But the second cites a quote from the first article that is inaccurate:
"Data is derived from local EMS providers contributing information to their respective state healthcare registries (generally managed by Health and Human Services in some form), ultimately to be used for analysis to improve care provided by paramedics and emergency medical technicians on the street."
This statement is technically incorrect because while EMS data at the state level may be stored by HHS departments, EMS data models originate at the federal level, where they are not governed by HHS, but rather by NEMSIS, the National EMS Information System, whose project leaders at the University of Utah School of Medicine are funded by the National Highway Traffic Safety Administration (NHTSA), part of the Department of Transportation (DOT).
The distinction between HHS and DOT funding may seem nitpicky, but it is nuanced and critical. The interoperability train among EMS, EHRs and HIEs is sliding down a mountain thanks to "different colors of money" being spent and departmental "siloes" that don’t talk to one another. U.S. taxpayers should be furious because it represents a paragon of redundancy. But it is also enormously important to we who build prehospital technology. Even if hospital systems could access state EMS data repositories (which they generally don’t today, though they theoretically could if they wanted to), hospitals would find the data in those repositories is incompatible with their own health record and analysis systems.
NEMSIS, the data set that forms the backbone of prehospital care records—basically NEMSIS is the EMS version of HL7—has been skewered by emergency department and even EMS medical directors as being deeply concerned with statistics but light on patient care. The data set bears out this claim, as NEMSIS-structured patient care records capture transportation and incident stats (e.g., there are sections dedicated to identifying the vehicles involved in a crash) but nothing about past patient encounters or family history…or even a place to capture a driver’s license, which would help identify the patient but holds no statistical value.
Not only are NEMSIS data largely statistical in nature, as may be expected from a NHTSA-funded academic center, but data on the EMS side are still incompatible with data in the hospital-side EHRs. EHRs are derived from the HL7-deisgned Clinical Document Architecture—formats like the Continuity of Care Document (CCD) and Consolidated Clinical Document Architecture (C-CDA) that are defined and structured by Meaningful Use and very familiar to folks who work with EHRs. Meaningful Use was established ONCHIT (the Office of the National Coordinator for Health Information Technology), which is funded by HHS. NEMSIS was established by NHTSA, a division of the DOT. Different colors of money, different measures of success, and now their data requirements have collided: Hospital systems need prehospital data to track patients and avoid readmission, but they need it in a CCD, C-CDA or other acceptable format. NEMSIS cannot output a CDA without transformation, because it is missing sections like those that pertain to past encounters, family history and mental health. (The NEMSIS-published guide to exporting CDA R2 contained technical errors.)
If EMS runs purely on its own, then EMS data are the industry's problem. We can choose to capture good and useful clinical data, or good and useful statistical data—or neither, or both. The single most efficient way to close the data gap between EMS and hospital systems, or EMS and HIEs, is to let EMS agencies use electronic health records in their ambulances, or to give hospital personnel real-time, secure access to prehospital data. The latter option would mean ambulance-facing electronic patient care record (ePCR) companies have to engineer real-time, secure access to their systems, which is rare in practice today; or that hospitals would have to overhaul their electronic systems yet again, which won’t happen. Alternatively, EMS agencies could switch from ePCRs to EHRs from Cerner, Epic and many others—but EMS regulators nationwide would have to let such a change take place. These regulators currently dictate the structure of data fire and ambulance agencies under their purviews must capture and convey, so they would be shrinking their kingdoms by handing power over to someone else. That has been suggested in the past; it did not go over well.
As soon as EMS purports to touch the rest of the healthcare system, but does so while holding fast to its noninteroperable data set, it faces a serious problem that first reared its head on February 4, 2014, at a Washington, DC, summit convened by the Office of the Assistant Secretary for Preparedness and Response (ASPR). For the summit, focused on health information exchanges and the prehospital environment, ASPR was joined by representatives from the office of Dr. Karen DeSalvo, the National Coordinator of Healthcare Information Technology, and together they made a pitch in favor of a role for EMS in accountable care organizations, including the need for EMS-EHR interoperability.
Why was prehospital data suddenly important to the hospital side of the patient handoff? As it often does, the answer came down to money: Hospital readmission penalties ratcheted up. According to Health Catalyst, “On October 1, 2014, the final payment and policy changes for hospital readmissions from CMS went live. Just weeks into the change, thousands of hospitals across the United States are feeling the financial pressures of the increased penalty.”
By 2015 states started requiring EMS agencies to implement NEMSIS version 3—the latest major release, some 10 years in the making. At the same time, readmission penalties increased and the nationwide use of ICD-10 went into effect in October. Hospital systems like Palmetto Health in South Carolina, Mount Sinai Medical Center in New York City, the University of New Mexico Hospital, Centura Health System in Denver, St. Vincent’s Carmel in Indiana, Essentia Health in Minnesota, and Alameda Hospital near Oakland started asking for prehospital data to track patients and/or keep them from returning to the ED within 30 days. Thus was the birth of “community paramedicine/mobile integrated healthcare”—the deployment of fire and EMS personnel into patient’s homes to proactively keep them from relying on 9-1-1 when they may not need an emergency visit.
ONCHIT, NEMSIS and NHTSA’s Office of EMS must incorporate technologists’ perspectives ASAP to avoid the roadblocks that will persist and get worse if our industry keeps relying on “separate but equal” data systems between prehospital and in-hospital care, when such systems are redundant, crazy expensive and ultimately contradictory to the flow of patient information along a care continuum.
(Disclosure: On November 25, 2015, with support from the California Health Care Foundation, Alameda County EMS validated that Beyond Lucid Technologies became the first technology firm to successfully transform a NEMSIS v3 document to CCD.)
Jonathon S. Feit, MBA, MA, is cofounder and chief executive officer of Beyond Lucid Technologies, Inc., an IT firm committed to connecting first responders with the care facilities they serve using cutting-edge technology. For more see www.BeyondLucid.com.