Blogs

One of the items we need better clarity on is the whole issue of allergies. Using the strictest definitions, much of what is currently reported as allergies are not.

The first question that occurs to me is why we focus on allergies when there are a variety of reasons for patient intolerance:

1. Allergy- the drug as a chemical entity invokes IgG or IgE mediated allergic response that may include vasomotor problems up to an including anaphylactic shock.
2. Idiosynchrasy- the drug as a chemical entity produces an unexpected or even paradoxical response.
3. Adverse reaction- the drug produces known adverse reactions that may be severe enough that the patient cannot tolerate the drug. Back when Procainamide was commonly used, 33% of patients who took it had to be taken off because of the severity of GI adverse effects.
4. Low tolerance - some patients simply overreact to specific drugs. The effects are expected (but at higher doses) but patients may be intolerant of a drug simply because their therapeutic response occurs to the far right on the bell curve.

The reality is that any of these may be a contraindication to the use of a drug in most conditions, but there may instances where the risk of low tolerance or known adverse reaction is a rational tradeoff to the needed therapy, where managing a true allergy would be intolerable.

The second question revolves around why supposedly competent trained healthcare professionals report such nonsense "allergies" as "I am allergic to all mycins..." Our patient intolerance management cannot possibly be properly automated when we blindly accept such reports without further investigation about what is really behind that report.

In a previous blog, I related a case where a patient who was reportedly allergic to Codeine had successfully taken Acetaminophen with Codeine for pain. Pretty clearly, whatever the patient's previous experience had been with codeine, it was not an allergy. If we are going to automate intolerance management in our healthcare systems, we are going to have to have more accurate, more rational reporting than we currently get.

Corollary to that is that we are going to have to be willing to modify patient-reported allergies in patient medical records when it becomes clear that the patient report is false, or ill-informed.

So my thought on this is that we should stop using the term allergy, and should talk about patient intolerance. When we receive the information about these intolerances, we should carefully vet the patient reports so that they represent truly actionable items. And when we learn that the report is false (85% of reported Penicillin allergies cannot be confirmed with skin testing), we must have the courage, and professionalism to be willing to update it and maintain it in a usable form.

If we are not willing to do these things, we will not successfully automate intolerance management. 

What do you think?