Depending on how you count, I’ve now been doing emergency medicine for 20 years. This is how I see patients.

The protocol below does not apply to resuscitation patients. For resuscitation patients, see The First Five Minutes of Resuscitation.

I apply this protocol to just about every patient. When you are systematic, you don’t miss stuff.  I see one patient at a time, and I document after each patient. Picking up multiple patients at once and postponing documentation seems like it would make you faster, but mostly it just makes you less effective.

1. Review at the vital signs. Is this a resuscitation patient improperly assigned to a non-resus priority?
2. Review the EMS note and the nursing triage note.
3. Glance at prior ED visits and other relevant notes in the medical record. If the patient has been seen recently or frequently, review those notes.
4. Introduce yourself to the patient and ask everyone at bedside what their relationship is to the patient, say it’s nice to meet you.
5. Before we talk about why you came to the ER today, I’d like to ask a couple questions. What medical problems are you known for? What medications do you take every day? Has there been any change in your medications lately? Do you have any allergies to medications?
6. Who do you live with? Assess functional status if it’s not obvious, and if the patient requires assistance with ADLs, who is providing that assistance? What sort of work do you do? Do you regularly use any substances like alcohol, nicotine, marijuana or street drugs?
7. What brings you to the ER, today? There’s a lot more to say about how to do a history and physical, which I get into here. You are developing your assessment of what is going on with this patient, and which dangerous conditions need to be ruled out. Your further questions specifically target the dangerous differential diagnosis, as does the
8.  Physical exam, leaving the most obvious part of the body for last, because you won’t forget that, but you might forget other areas, which could turn out to be very important.
9. As you walk away from the patient, finalize your dangerous ddx, which drives your orders. Put in your orders before writing the note.
10. Write your note. Do not wait for more information or a better time later on, write it now, with whatever information you have now, documenting the plan that makes sense to you now.

In most environments, an important goal of documentation is billing. Billing is mostly a game, but it is a game we must play effectively and efficiently. Learn how to do this and recognize it as a separate goal than creating a record of care.

Creating a record of care serves to inform downstream providers of what happened on your watch, and also demonstrate that you made good medical decisions for this patient.

Record the history and physical, with the greatest combination of speed and thoroughness that is reasonable. It really helps to get good at dictating, which is its own skill worth working on, because most of us can speak a lot faster than we can type. You emphasize in the history and physical those elements that inform your plan.

The plan is the most important part of the chart because it explains what you were thinking when you put in orders, or didn’t put in orders. The most common mistake in documenting a plan, which is more of a waste of time than a mistake, is charting what you have ordered. “Will send CBC, chemistry, perform CXR, treat with ibuprofen and fluids” is not useful, we already know all that, it’s in the orders. Documenting your plan is for explaining what you were thinking, when it’s not obvious.

I start my plan with what I think is going on. Often I think the patient has a benign condition, and this is where I chart that. I usually use language like “Multiple features for benign viral syndrome,”  Which conveys that there is a good chance that this patient has a benign viral syndrome, but I’m not assuming that’s the diagnosis, because who cares if the patient has a benign viral syndrome. If it turns out the patient has a benign viral syndrome, a migraine, gastroenteritis, or god help you costochondritis, great, the patient is going to get better. But emergency clinicians are not paid big bucks to determine that someone has a benign condition, our job is to identify dangerous conditions.

So now is the time to zoom in on the dangerous conditions that apply to this patient. Know your dangerous conditions.  Dangerous conditions that apply to the patient fall into one of four categories:  Dangerous conditions that you know with certainty the patient does not have (no way diagnoses), dangerous conditions you feel you have adequately excluded without further testing (no test diagnoses), dangerous conditions that you will exclude by further testing (rule out diagnoses), and dangerous conditions that you are concerned enough to treat (rule in diagnoses).

No-way diagnoses do not need to be commented on, and rule out diagnoses require very little if any comment, because you’re ruling them out. If it’s not clear that the test you’re ordering rules out the dangerous condition in question, this should be clarified, for example I often write “given duration of symptoms and reassuring ECG, ACS adequately excluded by single negative troponin.”

Most of your assessment (both in terms of documentation and cognitive space) should be devoted to no test diagnoses, because you are not ruling them out.  The bulk of your mental and written assessment should be devoted to why you don’t think a given dangerous condition requires further testing to rule out. If you focus on this element of the documentation, you will notice many instances where, in the process of justifying why you are not ruling out a certain condition, you realize that actually you should rule it out.

How exactly to document why you are not ruling out a dangerous condition is the hard science and art of emergency medicine, because deciding which dangerous conditions you need to rule out on the well-appearing patient in front of you is the hardest part of emergency medicine. The shortest way is “Doubt PE and dissection.” Just writing “doubt” is much better than writing nothing, but I usually go at least one step further: “No risk factors for or evidence of PE or dissection, will not further pursue.” Even better is to enumerate the risk factors/features of the dangerous condition in question that the patient does not have, “Gradual onset, frontal headache make SAH very unlikely, harms of CT exceed benefits in this 17 year old.”  If the patient does have risk factors or features for the condition but you’re not ruling it out, you should say more about why not, or maybe you should rule it out after all.

There are often dozens of dangerous conditions that your patient might have, but usually you can quickly strike most of them off the list as no way diagnoses, so you’re usually left deciding which dangerous conditions are no test and which are rule out, which is the hard science and art. After 20 years, I still routinely refer to dangerous conditions lists like headache, hypotension, and even chest pain.

Thoughtlessly pasting in a macro does not help you, either from a medical decision-making or defensive/legal perspective, and I routinely review charts with ridiculous macros that are obviously blindly text-expanded and incriminate the clinician as sloppy. However, pasting in a macro and going over it carefully, considering  the likelihood of each listed dangerous condition and editing the macro to apply to the individual patient, is a fabulous way to chart well and provide excellent care.

Finally, chart your plan for negatives. Positive tests dictate care so do not require elaboration, but now is the time to decide what you will do if the tests you order are negative, because you are thinking more about this patient right now than you will for the rest of your shift.

 

Inspired by this discussion–thank you Anton, Mike & Scott.

 

addenda here, on reddit