If you’re reading this and you’re not aware of the novel coronavirus that’s either already causing a pandemic or is about to cause a pandemic, I suggest you go back to whatever you’re doing and enjoy the bliss that must be uniquely yours. For the rest of us, I thought it might be helpful to give a quick overview of what your healthcare information technology (HIT) team is likely feverishly working on behind the scenes.
Let me start with some basics: in late 2019, a never-before-seen (aka novel) virus was identified in Wuhan, China as causing respiratory infections. It was identified as a coronavirus, a common virus that causes mostly upper respiratory infections (URIs). Most of us have suffered through a coronavirus infection as they cause between 15-30% of all common colds. This new version, though, is a variant that seems to be easily transmitted between people and can cause serious problems, especially for the elderly and those of us with chronic medical conditions. Different groups around the world have assigned this virus different names, such as COVID-19, SARS-CoV-2, or 2019-nCoV.
From the HIT perspective, one of the most important tasks that a clinical informaticist can undertake at this point is to make sure clinicians using the electronic health record (EHR) have a way to discretely and unambiguously indicate a diagnosis of COVID-19. This seems straightforward, but it’s far from it. To bill an insurance company here in the United States, doctors assign a specific code from a list maintained by the World Health Organization (WHO) called ICD-10. If there is no code, there is no bill as doctors don’t get the opportunity to write free text. Who cares about medical billing? We all should because many of our disease surveillance programs are based on billing ICD-10 codes. Does this make sense? No, not in my opinion, but it’s what we have so we make the best of it.
Today, we have only non-specific ICD-10 codes to describe the sick patients we’re seeing with COVID-19. The code B97.29 represents “other coronavirus as the cause of diseases classified elsewhere.” Not very intuitive or descriptive, but it’s the most accurate available. The WHO has issued a preliminary code identified as U07.1 which will represent “2019-nCoV acute respiratory disease.” This is a temporary emergency code which will undoubtedly be updated in the future. Informaticists want to get this code into our databases as soon as possible so we can accurately report on disease progress to the public health authorities and others who need to know.
Your chief medical information officer (CMIO) colleagues are also working to ensure that physicians can order the right lab tests. As of this writing, tests for COVID-19 are woefully unavailable in the US. To order the test in most EHRs, a doctor has to search for a typically generic-sounding lab test such as “Nasopharyngeal swab – send out.” Depending upon the software vendor, it might be possible to add a synonym to one of these generic orders so when a doc searches on “COVID” or “corona” they’ll discover the proper order. When these tests become more widely available, we’ll be able to properly name them.
Once we have a good COVID-19 lab test order, we need to insert it in the appropriate order sets. For the uninitiated, an order set is just what it sounds like: a set of orders. We use order sets as a tool to allow doctors to enter many orders quickly, but order sets also act as a form of clinical decision support (CDS). If a doctor isn’t certain how to test for COVID-19 or what the proper isolation is for a rule-out patient, an order set can point the physician in the right direction. As a standard of care for COVID-19 is agreed upon, HIT professionals will be creating and modifying order sets to support it.
Let’s turn our attention away from the clinician and toward the patient. You can imagine that someone who thinks they may be contagious with COVID-19 should not spend quality time in the waiting room of your local doctors’ office. Ideally, if they’re not that sick, they should stay home after speaking with a clinical person. However, if they need to be seen, we’d like to have them avoid the registration desk and other public areas. HIT can help with this by modifying the settings of patient portals and mobile apps so that we prevent patients from making online appointments without answering some basic questions. If you want to see your gastroenterologist because your acid reflux is acting up, terrific. If you want to see your primary care doctor because you have fever, a cough, a runny nose, and headache, hold up! We’ll want to think this one through. With the proper forethought, HIT can help direct the right patient to the right physician at the right location.
I was going to finish this post with a discussion of travel screening, but . . . as we see more endemic or community spread of COVID-19, I fear travel screening in the US will soon be moot. At the beginning of this outbreak, it was important to know if a patient had traveled outside the US because that was the way most patients became infected. If you just returned to the United States from Wuhan and now you had fever and a cough, our suspicions were high for COVID-19. As of this writing, many new COVID-19 patients never traveled and couldn’t be connected to a known contagious person. When this becomes the norm, travel screening becomes much less helpful. That said, we can bake such screening into most modern EHRs, so that’s important to remember for the next disease (which is hopefully a long way away!)
I hope this post gives you, my fair reader, a taste of what’s going on in the IT areas of your hospital or clinic. Things are changing fast on the science and epidemiology side, so hold on tight as we try to configure the tools doctors and nurses use to make them as helpful as possible for all involved. Did I miss something? Tweet at me @CraigJoseph.