Putting Data and Metrics to Work
Information overload is not a new thing. The Old Testament writers of Ecclesiastes astutely observed that “of making many books there is no end.” While the struggle to manage, analyze, and act on data is not new, the computer age accelerated this challenge.
In 1997 two NASA engineers coined the popular term “big data” to refer to the challenges they were having processing and visualizing the large amounts of data produced after running airflow simulations around aircraft. The computer power they had readily available simply wasn’t enough.
The digital age we live in now only compounds the issue. Consider for a moment that about 90% of the world’s data has been produced within the last 2–3 years—that’s more than the last 5,000 years combined. And with more and more Internet-connected devices across households and businesses, the creation and accumulation of data will only increase. No industry is immune.
But here’s where it starts to get interesting: The evolution of big data has led to the advent of smart data. Big data often requires mathematical PhDs combing through mounds of information to make any sense of the numbers. Smart data, on the other hand, is about creating platforms and systems that empower individuals or organizations to succinctly understand the landscape and act quickly—more quickly than they could act without this type of insight.
In other words, smart means access to the right data at the right time to make highly informed decisions that lead to a greater chance of a successful outcome. That really is a matter of life or death when it comes to patient outcomes and healthcare.
The 2018 ESO EMS Index
So, what does all this have to do with the 2018 ESO EMS Index we created? Well, everything. We designed this first iteration of the index to provide smart, objective, data-driven insights around a handful of key metrics EMS agencies and providers experience almost daily. This is objective information to which we’ve never before had access on a scale like this.
This is important for a couple of reasons: 1) The index showcases the value and promise of data for EMS agencies specifically and across the healthcare ecosystem more broadly, and 2) it provides something real and concrete that EMS agencies across the country can use to compare their efforts to other agencies’—a reference point for benchmarking of sorts.
The metrics we explored are:
- Aspirin administration in adult chest pain;
- 12-lead performance in adult chest pain;
- EtCO2 after advanced airway procedures;
- Stroke assessment performance.
The data set used to create the index is based on 5.02 million patient encounters at more than 1,000 EMS agencies across the country from January 1–December 31, 2017.
Setting the Scene
We’ve all been there. It’s controlled chaos: The call comes in, we rush out—often with the scantest of information. It could be a motor vehicle crash, an acute stroke, an ST-elevation myocardial infarction (STEMI), an overdose, or a patient utilizing 9-1-1 to access the healthcare system for a less-time-critical condition. In any case, for an increasing number of conditions, we now have evidence-based best practices that should be employed in treatment. Below we’ll look at the four metrics in context.
Aspirin administration—We know that aspirin, and particularly its early administration in adults with acute coronary syndrome, improves outcomes and may reduce deaths for these patients by as much as 23%. The index indicates that in only 56% of cases was aspirin administration documented in patients over age 35 with a primary impression of nontraumatic chest pain.
This insight from the index also exposes much of what we don’t know that is critical to know: Did providers only follow protocol 56% of the time? Or was it a combination of not following protocol and not documenting aspirin administration?
Regardless, we know the magnitude of aspirin’s benefit far outweighs the risk of administration except in the case of known allergy, and thus we have an excellent opportunity for improvement. It may be that more aspirin needs to be given, or it may be that administration by first responders or the patient themselves prior to EMS arrival needs to be more thoroughly documented—either way we can do better.
12-lead EKG—While monitoring of a cardiac rhythm via three, four, or even five leads can be lifesaving in that significant cardiac arrhythmias may be detected early, there is absolutely no substitute for obtaining 12-lead EKGs to evaluate for cardiac ischemia.
This is the gold standard, and there are measurable and reproducible reductions in mortality associated with appropriate acquisition and interpretation of EMS EKGs.
The data from the index show, however, that 12-lead EKG was obtained and documented in just 75.9% of cases after nontraumatic chest pain was identified as a primary impression in individuals over 35. That means either no EKG or some other variation was used almost a quarter of the time.
While there are always other factors to consider—including proper capture and documentation of information, patients with clear noncardiac etiology of their chest pain, and the possibility of an EKG obtained from a referring facility—a 76% documented compliance rate with 12-lead EKG in this particular scenario may indicate room for improvement.
EtCO2—End-tidal carbon dioxide monitoring is the evidence-based method to mitigate risk associated with decompensation due to a misplaced or dislodged advanced airway. Based on data from the index, EMS providers recognize this best practice, with EtCO2 monitoring in place 94.5% of the time.
While the index focused specifically on the use of monitoring for advanced airways, EtCO2 is a powerful tool that has many practical uses beyond that scope, including detecting the presence and severity of bronchospasm in asthma and COPD patients and in the detection of sepsis.
Stroke assessment—Assessment and care of stroke have changed dramatically of late. This can’t be emphasized enough, as the industry as a whole is still processing this information. Two recent studies (the DAWN and DEFUSE 3 studies) highlight findings that provide new guidelines for assessment and transport of stroke patients. Specifically these studies indicate the appropriateness of extension of the treatment window to 16 or even 24 hours after the onset of symptoms for a subset of patients with large-vessel occlusion acute ischemic stroke (LVO-AIS).
The treatment recommended for these patients is mechanical thrombectomy, not the traditional intravenous lytic medications utilized by many stroke-capable and primary stroke centers. Thus it appears not all strokes are created equal, and EMS will be called upon to help sort out which patients may still be appropriate for IV lytics and which will need thrombectomies and route them accordingly.
To do this we must develop a more complete picture regarding the performance of our currently available stroke screening and severity tools.
Based on data from the index, when the primary EMS impression was stroke, only 50.5% of patients had documentation of a complete stroke screen. This is not a complete surprise, as EMS treatment and hospital destination decisions have historically not been affected by how many elements of a stroke assessment tool were positive, but rather only if any of the elements were positive.
In other words, there was no reason to complete any other elements of the Cincinnati Prehospital Stroke Scale once any of the elements were positive. Given the new treatment recommendations, however, it is more important now than ever to complete the entire screening tool to assist with determining severity.
While the nearest hospital may have seemed a good destination in the past, the new research indicates transport to the nearest thrombectomy-capable or comprehensive stroke center may be preferred for those with evidence of LVO-AIS, even when it is not the nearest facility.
Just as a regional approach to the treatment of trauma and STEMI has evolved, the same now needs to happen with stroke. A key element of this approach is completion of the stroke assessment tool.
Overdose and Opioids
Beyond the four metrics listed above, the scope of the EMS index focused on one irrefutable fact that needs to be mentioned, if only briefly (since there will be another article on this topic coming): Overdose is a major problem in this country. It has reached epidemic proportions. Overdose encounters accounted for 1.65% of the 5.02 million patient encounters in the 2017 data set used for the EMS index. There were 12% more overdose encounters in 2017 than stroke encounters.
Recent data for a six-month period show the problem is only getting worse, with overdose encounters for EMS providers doubling during this period. The scope and severity of the problem cannot be overstated, including the impact on EMS agencies in terms of resource strain, cost to transport, provider safety, and other factors.
Big data, smart data, and predictive data are all here to stay. This evolution will shape the way EMS agencies and providers assess patients, respond to encounters, and interact with hospitals and other organizations across the healthcare spectrum. Tighter integration and sharing of information across multiple systems will continue to develop, as will a more evidence-based approach to patient care.
The EMS index is a part of this conversation and provides a snapshot into what is happening at the national level for agencies to use to assess their own performance.
This is just the beginning of a very important dialogue. As we continue to compile important data in this digital age, we can get even smarter and more refined in the way we react, respond to, assess, and care for patients to ensure the most positive outcome.
Agencies that acknowledge and embrace the data-driven world in which we now live will be primed for success. As the CEO of a company called Mixpanel said about the value of data: “Most of the world will make decisions by either guessing or using their gut. They will be either lucky or wrong.”
Brent Myers, MD, is chief medical officer for ESO Solutions and one of the authors of the EMS Index. He is president of the National Association of EMS Physicians. Contact him at firstname.lastname@example.org.