How to Talk to Clinicians: Forget Workflows, Just Tell Us How Things Work

Workflows are all the rage with EMR people. But doctors, NPs and PAs are smart. Nothing burns us out as fast or as completely as being told how to do things instead of why. We are not circus animals.

Let me explain:

If we had no professional education at all, we would have clinical workflows memorized instead of clinical knowledge. For example, two weeks after starting an ACE inhibitor like lisinopril, order a basic metabolic profile. That sounds pretty straightforward, but if you add up all the possible clinical workflows we would need if we didn’t know medicine at all, that would be a huge burden – a massive amount of seemingly random and senseless rules.

But, of course, the clinical knowledge we acquired in our training is that ACE inhibitors can act like a stress test for patients with undiagnosed renal artery stenosis and a BMP drawn soon after starting such medications ensures we aren’t causing kidney damage with our prescription.

Such clinical knowledge makes us not only order the blood test, it guarantees that we will always remember to do it because it makes sense. It is like memorizing a beautiful poem instead of a long string of random letters.

So when it comes to how to interact with our computers, give us the benefit of the doubt and explain what the computer can do or can not do, rather than just ordering us around seemingly useless detours and side roads from point A to point B

Especially since computer workflows are often counterintuitive to clinicians, it helps us to know why they require us to seemingly do the wrong thing. As I have written before, EMRs are workarounds because today’s computer programs can’t replicate how the clinical mind works. Some of us might even take a certain pride in becoming expert at using a less than brilliant tool when we have the admission of our IT people that it is the EMR that is stupid and not us.

Seriously, we deal with worse challenges than that every day, just trust our intelligence and let us know when and why our computers can’t yet do what every reasonable clinician would expect them to.