What data tells us about first responders dealing with COVID-19 cases
Data sharing between emergency medical services (EMS) and hospitals is being stress tested as first responders are handing off COVID-19 patients. And once the pandemic and crisis clear, there will likely to be a new set of best practices and data lessons to ponder.
We caught up with Dr. Brent Myers, chief medical officer, of ESO, a data and analytics company focused on first responders, to talk how data is informing the COVID-19 handoff from EMS workers to hospitals.
ESO was founded by a paramedic and computer programmer to tackle data sharing issues between EMS and clinicians, hospitals and other partners.
Tableau makes Johns Hopkins data available for the rest of us | As COVID-19 data sets become more accessible, novel coronavirus pandemic may be most visualized ever
Among the key takeaways:
There's a need for better data and information sharing between EMS and hospitals. "We have become one of the largest providers of data aggregation, so we provide an electronic medical record for firefighters and paramedics, and then we now aggregate that at the hospital side, not only for patient care in the emergency department, but for registries, repositories, research, and data across the United States," explained Myers.
COVID-19 or suspected cases need to be tracked during that first mile when first responders enter the picture. Myers said:
We gave the medics the ability to say, 'we are either suspicious that this is COVID or this patient has already been tested and we know that they are COVID positive at the time of the encounter.'
So, from there, we've been able to aggregate data around the care that these patients receive, and it is very consistent with some of the projections that we've seen around the country, and then it's also very consistent with data that we've seen internationally from China, Italy, and elsewhere, meaning that only about 10% of the patients that we encounter actually require EMS intervention. Most of them are actually quite stable and do very well, and then about 2% to 3% are critically ill across the country.
First responder data can inform personal protective equipment needs. "We're trying to provide realistic data to help the federal policy makers understand how much EMS is going to need versus the other components of the healthcare system," said Myers.
EMS data is mostly structured. Hospital data often unstructured. Myers explained:
The great thing about EMS across the country is, years ago the National EMS Dataset was established, so there is a not free text, but structured data elements that are required for every EMS call, and those become more sophisticated with iterations.
The structured data is going to be much more important, and the beauty that we've learned here is how nimble you can actually be with adding structured data elements. Some of these we added to the software and deployed across the country within 24 hours of requests.
The outbound (hospital) data back to EMS is typically unstructured, but ESO is using natural language processing to bridge the gap. "We're really working that fine line between structured and unstructured, because the hospital language is just different than EMS," he said.
5G is likely to improve the EMS to hospital handoff and enable telemedicine and in-ambulance collaboration. Myers said:
There are two major ways that 5G's going to help us. The first of those is that when you have a lower bandwidth or a completely non-connected environment. Obviously, some of the data has to live temporarily on whatever device the medic is entering the data. And so when we want to push an update, such as the reuse dropdown menu, we have to wait on connectivity. Our ability to be nimble and move quickly with updates pushing out to the field. It also really has difficulty about which types of devices the medic can use. Can you go all the way to a pure tablet device when we have to be able to store some of the data in a non-connected environment? So 5G helps with that tremendously.
The other piece that we're really excited about is the telemedicine component to share things like an EKG tracing in real time, or an ultrasound in real time. And for the physician to be able to do an exam on, for example, a stroke patient, in a remote manner to determine the best hospital destination for that patient. The ability to do that without worrying about bandwidth is going to be tremendous for us.