Three billion dollars in annual uncompensated Emergency Medical Services — that is, care provided to patients by prehospital personnel but that will not be reimbursed due to documentation errors or omissions — coupled with costs tied to inappropriate, avoidable activations of emergency resources in ambulances and hospitals, represents a measurable but addressable risk to public health and homeland security. [i] As Accountable Care graduates from buzzword to bona fide in pursuit of the so-called Triple Aim — better health, better care, lower cost — the emergence of EMS-side concepts like Community Paramedicine / Mobile Integrated Health asks ambulance operators to benchmark against quality metrics that justify the expense of their services. Do prehospital interventions facilitate care, patient throughput or healthcare quality? (If so, how frequently and to what degree?) Do readmission avoidance, post-discharge follow-up, and patient tracking improve the flow from prehospital to in-hospital to post-hospital contexts? We talk about data, but few ask a question that is central to critically assessing our discipline: How can a data company make care providers better?
Electronic patient care reports (ePCR) are not diagnostic systems; they are digital documents that collect, store, and transmit data inputted by clinicians or imported from devices and repositories. Troubling statistics emerge when practitioners rely too heavily on automation to make clinical decisions despite the “open secret” that medical devices produce incomplete or incorrect data. We are in an age reminiscent of the film Wall-E, seeking middle ground between manual, laborious methods that make it tough to track information, and the Holy Grail of artificial intelligence where we trust the machines that are becoming like appendages.
What resources are wasted on mistakes, like activation of a catheterization lab in anticipation of an ST-elevated myocardial infarction (STEMI) patient even though the incoming patient is not experiencing one? Might access to “perfect” prehospital data — in real-time — reduce such waste? How would EMS agencies benefit? My late colleague Bruce Graham, former Vice President of the Ohio EMS Chiefs Association and Partner-Client Development Manager at Beyond Lucid Technologies, has eloquently noted that “an EMS agency [does] not make a dime more or less if the hospital does or does not pull the trigger on a cath team. Plain and simple…I get nothing from what the hospital does. And if the hospital doesn’t like what I do, their answer is to go for tighter controls. This was one of the points of the ACA and the creation of ACOs: to work together to share the savings.” What savings would justify investment in “perfect prehospital data”?
Bruce see training as the bridge from a manual system to one that is technology-enabled. Too much technology can make people dependent, as we have seen from medicine to airplanes. Digital tools are just that — tools; they are meant to help, not replace. In the words of Mark Wittman, MD, MBA, MPH, critical care physician and Chief Medical Officer at Beyond Lucid Technologies: “Healthcare is a combination of data and brains. If data can make brains better, then you have real value.” Technology represents progress, and the ability to collect, connect, and analyze Big Data — instantly — finding trends that heretofore would have been missed. But software cannot supplant the EMS provider; it merely supports his or her readiness to respond. In Bruce’s words: “Software can be set up to allow the entire incident and encounter to be captured, thus allowing others to go back over the
incident with the providers, to debrief them. I’m talking about closing the loop: the fact that you get the 12-lead wrong…well, maybe what you thought you saw isn’t what you printed out. Technology will allow us to go back and look at it in the context of the entire call. In the future the entire patient care report will be embedded in the electronic health record in the hospital. Now the emergency medicine physician, internist, interventional cardiologist, intensivist, and even the primary physician can look at the same record. You are indeed closing the loop but most importantly, you are getting feedback, in real-time, so you can go back and say, ‘This is what you missed.’”
If EMS providers cannot benefit from field data in real-time, all they are left with are noise and toys. Validating “good data” into the ePCR is essential, so making workflows better, faster, easier and more cost-effective is the ePCR company’s prime objective. There are tangible clinical and financial benefits to be gained from improving documentation quality: For example, obtaining a 12-lead for assessment is S.O.P. when examining a cardiac patient. But without a comparative baseline — a review of the current strip in light of previously captured ones — crews are hard-pressed to deduce whether they are investigating something serious versus a quirk of the individual body. I have seen crews in places like Pittsburgh ignore alerts from monitor-defibrillators because “the patient has had a funky heartbeat for years, and he’s not having a heart attack.” Insight about a patient’s condition over time is rare, especially in emergency contexts, so access to clinical history — putting the current incident in context — is vital to accountable, informed care, whether the patient is suspected of STEMI, stroke, a diabetic event, allergy, seizure, or most complex syndromes.[ii]
Without context, many conditions look alike, which helps explain why Ramanujam et al. (2008) challenged the reliability of decision-making by field medics.[iii] Cardiac research shows enormous regional variation among false positive cath lab activations. (It would be worth investigating why these rates vary so widely, whether due to matters of public policy, medical direction, technology, or training.) Rokos et al. (2010) found that “recent surveillance of ‘unnecessary’ cath Lab activations…by ED physicians demonstrated a 5% rate from a single-center experience (n = 249 activations) in Virginia and a 6% rate across 14 hospitals (n = 2,213 activations) in a North Carolina STEMI system.” In Los Angeles and Orange Counties, false cath lab activation rates based on paramedics’ interpretation of the ECG were 20% and 23% respectively.[iv]
Contra Costa County EMS describes a false-positive as “a paramedic tell[ing] the STEMI Center that a STEMI has been detected on the 12-lead ECG, but upon arrival at the hospital it is determined that the patient’s 12-lead does not show a STEMI. Most of these patients do not need the urgent availability of the catheterization lab.”[v] JAMA says “the frequency of false-positive cardiac catheterization laboratory activation for suspected STEMI is relatively common in community practice, depending on the definition of false-positive.” False activation is also an expensive error: Patricia Frost, CCEMS Director, wrote in 2013 that 26–41% of STEMI activations in her county were false, at a cost of $5000 per incident. Eliminating those could save $480,000 per year.[vi] (The range reflects variation across the county’s hospital system that includes six STEMI receiving centers.) The insurance value of prehospital technology to provide “perfect information,” and to connect those data to the rest of the care continuum, thus justifies its investment by eliminating the cost of wasteful line-items.
Accountable Care — and its EMS corollary, Community Paramedicine / Mobile Integrated Health — have inspired motivations to align clinical, operational and financial efficiencies. EMS has a once-in-a-generation chance to centralize its role as the web that connects the ACO. But what are (a) the risks to the patient, (b) to every provider along the healthcare value chain, from EMS to ED to nurses to physicians, and (c) to the healthcare facility if “bad data” wends its way into longitudinal records that follow the patient over time? At a rising number of hospitals, complex codes like stroke and STEMI are being called by EMS crews with authorization to bypass the ED, theoretically speeding time to treatment while relieving congestion and overcrowding. How reliable are the data being used to call those codes, and how good are the medics at distinguishing true positives from false ones? New health information exchanges and data management files make it feasible to search and import patient clinical histories while at the patient’s bedside — improving prehospital care while lowering costs, and leveraging telemedicine for tailored, informed medical direction.
First published at the 2017 Texas EMS Conference, November 16–22, 2017
[i] Hagen T. The Value of EMS. EMS World. 1 Sept 2012.
[ii] O’Donnell D, Mancera M, Savory E, Christopher S, Roumpf S, Schafer J. “The Availability of Prior ECGs Improves Paramedic Accuracy in Identifying STEMIs.” IU School of Medicine at the 2014 NAEMSP Annual Meeting. Accessed online 16 Apr 2015 < http://www.naemsp.org/Documents/2014%20Annual%20Meeting%20Handouts/E-POSTERS%20145-153%20Combined.pdf >
[iii] Ramanujam P, Guluma KZ, Castillo EM, Chacon M, Jensen MB, Patel E, Linnick W, and Dunford JV. “Accuracy of Stroke Recognition by Emergency Medical Dispatchers and Paramedics — San Diego Experience.” Prehospital Emergency Care. (2008) 12(3):307–313.
[iv] Rokos, IC, French WJ, Mattu A, Nichol G, Farkouh ME, Reiffel J, Stone GW. “Appropriate Cardiac Cath Lab activation: Optimizing electrocardiogram interpretation and clinical decision-making for acute ST-elevation myocardial infarction.” American Heart Journal. December 2010. 160(6):995–1003.
[v] Contra Costa EMS Agency. “Contra Costa 60 Day STEMI System Review: A Great Launch with Challenges Ahead.” Accessed 15 Jan 2015. < http://cchealth.org/ems/pdf/stemi_1st_60days.pdf > See also: “Contra Costa County EMS Data Infrastructure Project.” p.11. Published 28 Dec 2013. Accessed online 2 Feb 2015. < http://cchealth.org/ems/pdf/QI-HIE-grant-report-2013.pdf >
suspected ST-segment elevation myocardial infarction.” J American Medical Assn. 2007 Dec 19. 298(23):2754–60.