Morning Report: 7/5/2012

Today’s Morning Report is thanks to Dr. Joshi!

Clinical Scenario:

19 yo woman was walking down the street and twisted her ankle and now is in the ED for evaluation. The next patient you are supposed to see is a 90 yo man who slipped on some ice and is complaining of knee swelling and tenderness after the fall.

Your burden:

How will you work up these two patients? Who do you decide to xray? Can you send either of them home without xrays?

HOW DO YOU KNOW WHAT TO DO??????

Lucky for all of us, there are clinical decision rules for this. What are CDRs? CDRs are ways to help GUIDE clinical management.

Objective of Morning Report:

– How are CDRs created

– HOW do I use CDRs?

– How do I know if the CDR is good?

Stiell 1999 helped write the definitive answer on how CDRs should be created, and what to base judgment upon when deciding if the CDR is good or not. Gallagher 2004 helped to provide further clarity when understanding how and when to use particular CDRs.

Considerations of CDRs, quick summary:

1. Definition of Outcome

The outcome of the CDR should be clinically important, clearly defined, blindly assessed. For example: Ottawa ankle and knee rules seek fractures as clinical outcome; Wells Criteria seeks PE

2. Definition of Predictor Variables

The predictor variables should all be clearly defined and fully disclosed. Prior to creating CDRs, the investigators should have gone through an extensive list of predictor variables before compiling the predictors that are in the CDRs. The reader should be fully aware of all of these. For example, not simplying stating if a patient had a fever, but clearly defining fever as a temperature greater than 101.

3. Did they test reliability of the predictor variables?

The inter-rater reliability should be checked for the variables. It should be very easy and objective to decide upon the variables, it should not be experience based, etc. For example, it is easy to say whether or not there is pain at the base of the 5th metatarsal, it is a very specific area that doesn’t require too much room for variability.

4. Patient selection

The study should state prior to the start of the project how many patients they want to enroll based upon mathematical analysis. Patients should also reflected your patients. For example, if you want to study whether to LP meningitis patients in the ED, you want a CDR that examines ER based patients, not neuro clinic patients, not patients who were told to come to the ER.

5. Sensibility

Does the rule make sense? If the rules don’t make sense, if they are confusing, then the likelihood they will be used is much lower. The rule should make economic sense and be easy to utilize, it should not cause excessive harm.

Stiell IG, Wells GA. Methodologic standards for the development of clinical decision rules in emergency medicine. Ann of Emerg Med. 1999. 33:4, 437-447.

Gallagher EJ. Shooting an elephant. Ann of Emerg Med. 2004. 43:2, 233-237.