To test or not to test.001

I am just going through my PILE OF GUILT (read: stack of EM literature) that tends to accumulate over time… At the very least I try to browse through the journals and magazines as they trickle or pour in, but sadly enough I end up playing a catch-up game.

The August 2013 issue of the Annals of Emergency Medicine ended up waiting for its review until December. Such a 4-month delay is nothing to be proud of :-(. Especially because of this excellent editorial by Dr. Steven Green on clinical decision rules. He describes common pitfalls and important caveats concerning these clinical tools. They are omnipresent in our daily practice and we tend to apply them as a sort of bandaid or anti-lawsuit remedy. The problem is that quite a few of the rules add little or nothing to our sound judgement (read: GESTALT). They are far from being the Holy Grail of emergency medicine, so before you choose a given rule and potentially cause more harm than benefit to your patients, consider the following factors (after Dr. Green):

  1. Relevance of clinical question – simply don’t bother with rules that answer trivial or unimportant questions.
  2. Derivation – best rules meet rigorous derivation standards.
  3. External validation – crappy decision rules perform great in the derivation sample but fail if applied to a new patient sample.
  4. 1-way versus 2-way application – most rules are designed as 1-way tools, and lead to negative consequences if applied in a 2-way fashion. Let’s take PERC (Pulmonary Embolism Rule-out Criteria) as an example. You are supposed to forego further testing if the patient is „PERC negative”. It is not meant to indicate the need for a PE evaluation if patient is „PERC positive”. If applied in 2-way fashion this rule would lead to increased testing and potential harm.
  5. Implication for current practice – ask yourself if the rule improves your clinical gestalt. In a multitude of medical conditions gut feeling seems to perform better, e.g. pulmonary embolism.
  6. Applicability to your patient population – certain rules do not factor in modern practice patterns such as use of bedside ultrasonography, which honestly makes them redundant and obsolete to begin with.
  7. Ease of use – who on earth has the capacity to remember all those multi-step rules? Availability of automated calculators and/or drop-down charting add-ons certainly increases applicability.

The bottom line is – majority of the rules are more of a double-edge sword than a protective shield. It is unwise to apply them in an automated fashion. Once you’ve built your clinical judgement it is often superior to any rules. Nevertheless, these tools probably help to unwrap your gestalt from its immature cocoon.