At the beginning of October, one of the big topics online was a case from 2008 where an EMS crew misdiagnosed a chest pain patient with acid reflux, recommended Maalox and left the patient at home. The next day, the patient called 911 again with chest pain and shortness of breath. This crew transported the patient to the hospital where the patient ultimately died of a pulmonary embolism. The new turn in this story was a judge rejecting immunity from liability for the first crew since the original situation was “non-emergent.” A discussion of the call, as well as the court document for the ruling, can be found over at Fire Law.
In response to the case, the most common refrain online was “EMTs don’t diagnose,” to which I say phooey. EMS providers DO diagnose, in fact it’s a requirement to do the job properly. They just aren’t taught to use clinical judgment properly when making a diagnosis. So, the big question is, “What is a diagnosis?” and “How exactly does the process work?”
The first misconception that needs to be laid to rest is the concept of a diagnosis as being final. There are many types of diagnoses (working, differential, admitting, discharge, etc) and the diagnosis changes as more information is learned about the patient and as other providers become involved. A patient in the emergency room may have one diagnosis by the emergency physician, a more specific diagnosis by the hospitalist, and an even more specific diagnosis from an infectious disease specialist due to more information being gained over the course of the patient’s stay and differing levels of education. Similarly, a patient with chest pain may be diagnosed with “chest pain, possible MI” by an EMT (symptoms can be a diagnosis when the provider at any level is unable to narrow it down at that time), STEMI, including a general location, by a paramedic, and the cath lab team will be able to diagnosis the specific branch of the vessel with the amount of occlusion. Given the knowledge of the provider and the tools available, all three are valid diagnosis for the same patient.
A second misconception that seems to be present is that there is only one diagnosis that explains everything. A patient can have multiple problems that haven’t been found before. One thing to think about if a presentation doesn’t make sense is, “Is this one problem or several?” Over at Street Watch: Notes of a Paramedic, Peter C related a case where he caught (diagnosed) a patient who was having a concurrent CVA and STEMI.
Bates’ Guide to the Physical Examination and History Taking lays down a 5 step process for “clinical reasoning: developing hypotheses about patient problems.” Feel free to substitute “diagnosis for hypothesis.”
1. “Start with the most critical finding.” The example given is a patient with the “worst headache of the patient’s life,” nausea, and vomiting, think increased intracranial pressure, not GI problems.
2. “Match your finding with all possible causes.” This is essentially the formation of differential diagnoses (DDx). All a DDx says is, “Given what I know right now, these are the possible causes.” This should start as soon as you hear the patient’s chief complaint. Take a second and think of all of the causes of chest pain? Trauma, pericarditis, MI (STEMI –and- NSTEMI), pulmonary embolism, pneumonia, pleuritis, and probably many more. These are your differential diagnoses.
3. “Eliminate the diagnostic possibilities that fail to explain the findings.” As you interview, perform a physical exam, and perform point of care testing (most common in EMS is a 12 lead EKG, pulse oximetry, and blood glucose testing), some things you didn’t think of based on the chief complaint may come up. Those get added to your list of DDxs. Say the patient doesn’t complain of trauma and there are no signs of trauma. Ok, mark trauma off of the list. You acquire a 12 lead EKG and no ST elevation is noted. Mark off STEMI (but not NSTEMI).
4. “Weigh the competing possibilities and select the most likely diagnosis.” Ok, what makes sense out of the list given what the signs, symptoms, and lab results are while taking into account the course of the disease? When did the pain start? Did it start 5 days ago or 2 hours ago? Has this happened before? Given what you know about the patient and your knowledge base (taking into account the limits of your knowledge. This is sound advice for everyone from the EMR through physician), what makes sense? Is this a patient that I need to refer (for EMS, push hard for transport)?
5. “Give special attention to potentially life-threatening and treatable conditions.” This is, of course, where the original crew failed. Is gastric reflux a valid possibility? Sure, but so is NSTEMI (which, given the limited information provided about what the crew knew, I’d argue this is more likely than a PE. Of course limitations in diagnostic tools is an important consideration) and pulmonary embolism, both of which cannot be ruled out in the field.
In other words, if you take a history, perform a physical examination, and use diagnostic tools to gather information, incorporate that information into a conclusion about what the patient is experiencing, and use that to guide treatment, then you are making a [working] diagnosis. If you are considering that multiple diseases can give the same symptoms, you are making a [differential] diagnosis. How close does this sound to describing how you actually practice?
If you are truly not making a diagnosis then what exactly are you treating? I’d argue that the vast majority of the physical exam is useless and that provider should be content to following cook book, symptom relief protocols with zero deviation. After all, if you deviate even a little, the only true basis is that you made a diagnosis. I sincerely hope that most EMS providers diagnose, even if they call it something else.
1. Bickley, Lynn S., Szilagyi, Peter G. “Bates’ Guide to Physical Examination and History Taking.” (10th edition). Philadelphia: Wolters Kluwer/Lippincott, Williams, & Wilkins. 2009. Pg. 28-29.