Guidelines are wonderful. Guidelines are dangerous – KevinMD.com – ROBERT CENTOR, MD | July 2017
Over the past decade, I have thought often about the benefits and the problems of clinical guidelines. The first concept that attracted my attention was reading about conflicting guidelines.
Given the same data, different guideline committees would have significantly different recommendations. At the least, this problem raises questions about guideline validity.
It makes clear that committee perspective could influence recommendations.
This is exactly what happened when the CDC opioid prescribing guidelines were ghost written by addiction specialists from PROP.
Guideline recommendations sometimes are clear and demonstrably evidence based, but too often recommendations reflect the committee’s view of the problem.
The problem here is a long tail problem. As physicians, we strive to know when a patient’s presentation is not routine. Our challenge comes from knowing when we should switch from system 1 (or automatic) thinking to system 2 (deliberate) thinking. What clues must we consider prior to using a guideline?
Too often we see patients with diagnostic labels and “perfect” treatment for those labels — e.g., CHF or COPD.
But when we consider the patient more carefully we see that the patient does not carry to correct diagnosis. The treatment (guideline directed) is wrong because the diagnosis is wrong.
Perhaps guidelines should start with a very careful inclusion definition.
- So for acute pharyngitis perhaps we would require short duration (at most 3 to 5 days), and no red flag symptoms.
- For systolic dysfunction, we might define an inclusion ejection fraction.
- For COPD we should require full PFTs (to define obstruction and exclude restrictive lung disease).
I submit this is not a trivial problem.
Guidelines become recipes, but recipes do not work with the wrong ingredients.
Like opioid prescribing guidelines for addicts don’t work for pain patients.
Guidelines should not suppress the physician’s thought process.
The CDC opioid guidelines are being codified into strict dosage limits that restrict a physician’s options and preclude careful thought and balancing of benefits and harms.
Perhaps a great guideline would define the warning signs (or “red flags”) that should induce more careful thinking.
As an academic perhaps I worry too often about zebras, but then working at a community hospital and a VA hospital, I sure seem to see them.
I often say the diagnosis is job #1.
We need our guidelines to clearly define the relevant patients for that guideline.
The guideline should direct us to return to the diagnostic process when the patient’s problem representation does not fit the illness script that the guideline defines.
When patients are not addicted to opioids, they should not be prevented from using them at whatever dosage is effective.
Author: Robert Centor is an internal medicine physician who blogs at DB’s Medical Rants.