Decision Rules for Students

This forum made possible through the generous support of SDN members, donors, and sponsors. Thank you.

Perfect Hair Day

Grapes of Wrath, Chocolate Ice Cream, Johnny Cash.
7+ Year Member
Joined
Aug 16, 2016
Messages
375
Reaction score
276
Trying to put together important list of Clinical Decision Rules to keep handy during the clerkship rotations in EM. Would appreciate if people would add - students, and especially attendings who want to see students using the clinical decision rules, if something isn't already listed on this post, please add!

HEART Score
Wells for DVT
PERC
Wells for PE
CURB-65
PECARN
Alvarado Score
Ottowa Ankle
 
Tooth:Tattoo ratio and Law of Inverse Value.
I prefer the law of inverse value, as it's more broadly applicable. But the T:T is as good a first blush patient evaluation as the allergy list is...and probably more useful than actually talking to, or examining them.
 
Canadian CT head and c-spine imagine rules.

CURB-65 is worthless in the ER. It was designed for primary care physicians to use in the office to refer patients for admission. It's never been validated in an ER to my knowledge (granted I'm not the most up on research related to CURB-65). The pneumonia severity index (i.e., PORT score) is way better.
 
CURB-65 is not my favorite at all. Lowest risk group still 1% death at 30 days (this is for pneumonia!). Meh.

Also not a huge fan of San Fran Syncope Rule.

Obvious the key is to learn how to truly apply the rules, and understand when not to.
Wells, PERC
Nexus / Canadian
Ottawa ankle / knee
HEART Score
Would be the first wave I would teach a student-- all pretty well validated, not hard to apply, and COMMON enough to use everyday.
 
Fibromyalgia score
- Does the patient have fibromyalgia? 0 for no, +1 for yes

Only one I would add is LRINEC
 
Trying to put together important list of Clinical Decision Rules to keep handy during the clerkship rotations in EM. Would appreciate if people would add - students, and especially attendings who want to see students using the clinical decision rules, if something isn't already listed on this post, please add!

HEART Score
Wells for DVT
PERC
Wells for PE
CURB-65
PECARN
Alvarado Score
Ottowa Ankle

Regarding the Wells criteria for DVTs: are any of you actually sending out d-dimers to rule out DVTs? I can't say I've ever seen it; we either ultrasound or not.
 
CURB-65 is not my favorite at all. Lowest risk group still 1% death at 30 days (this is for pneumonia!). Meh.

Also not a huge fan of San Fran Syncope Rule.

Obvious the key is to learn how to truly apply the rules, and understand when not to.
Wells, PERC
Nexus / Canadian
Ottawa ankle / knee
HEART Score
Would be the first wave I would teach a student-- all pretty well validated, not hard to apply, and COMMON enough to use everyday.

Ottawa ankle/knee rules are useless.

Does it hurt? Yes —> X-ray.
 
Regarding the Wells criteria for DVTs: are any of you actually sending out d-dimers to rule out DVTs? I can't say I've ever seen it; we either ultrasound or not.

What my ED seems to do is order the d-dimer and the CTA Chest at the same time. Why both? I have no clue.
 
Canadian C-Spine...
Seriously, Canadian C-Spine all the way. That is, unless you like CT scanning the neck of basically every young MVC patient who complains of midline neck tenderness after getting rearended at 10 MPH.
I hated NEXUS as a resident.

As a resident, I also hated when attendings misinterpreted Canadian C-spine when they let me use it. Some would insist on the CT in a low risk patient because they could range the neck 45º both ways but had pain when doing so. Irrelevant. The study explicitly states that you can clear their neck if they can rotate regardless of pain.

OK, off my soapbox now.
 
Fibromyalgia score
- Does the patient have fibromyalgia? 0 for no, +1 for yes

Only one I would add is LRINEC
LRINEC had an impressive original paper, however the validation study showed that it was significantly less reliable than originally claimed and that it was easy to miss early cases of Nec Fasc by using it.
Validation of the laboratory risk indicator for necrotizing fasciitis (LRINEC) score for early diagnosis of necrotizing fasciitis - ScienceDirect
 
Something to keep in mind about decision rules:

They are invented to help you when there's a gap in your clinical judgement. But, none apply to all patient populations (the studies usually exclude certain categories of patients, which for all you know, your patient could be in). Also, none are 100% accurate. Most importantly, all have (or a t least should have) a disclaimer saying something like this, "Clinical decision rules are not intended to take the place of clinical judgment." Make a point to search the text of these decision rules for these "don't rely on me" disclaimers. You might have to go to the original journal text.

So what good is a rule to help us when there's a gap in my clinical judgement, if the rule ultimately relies on my clinical judgement?

I think they're a good learning tool. And they're a good way (sometimes) to work through a difficult case, to give one confidence, if the rule agrees with your pre-decision rule assessment. But ultimately, it's up to us to know how and when to apply these rules, and it's up to us to know when the rule is wrong. And if you're relying on the rule to be smarter than you, you're falling into a trap. It's not like a diagnostic study with another physician's name, judgement and license attached to it, like a radiologic study, or heart cath.

Can you blame the MD who's lead author on the paper that invented the decision rule, if it turns out to be wrong?

Can you or the patient sue those who developed the decision rule, if it turns out your patient was in the excluded patient types and you miss something?

No.



Decision rules can be good learning tools.
They do not replace clinical judgement.
Ultimately, it's all you.
 
Last edited:
p = 1 / n

let p = probability of disease
let n = number of family members + friends in the room
 
My ED director had asked me about the following at various points in the rotation
HEART score
Canadian CT head and C Spine
WELLS for PE (not the DVT one)
PERC
San Francisco Syncope Rule
Centor Criteria

I had MD Calc installed on my smartphone.
 
Something to keep in mind about decision rules:

They are invented to help you when there's a gap in your clinical judgement. But, none apply to all patient populations (the studies usually exclude certain categories of patients, which for all you know, your patient could be in). Also, none are 100% accurate. Most importantly, all have (or a t least should have) a disclaimer saying something like this, "Clinical decision rules are not intended to take the place of clinical judgment." Make a point to search the text of these decision rules for these "don't rely on me" disclaimers. You might have to go to the original journal text.

So what good is a rule to help us when there's a gap in my clinical judgement, if the rule ultimately relies on my clinical judgement?

I think they're a good learning tool. And they're a good way (sometimes) to work through a difficult case, to give one confidence, if the rule agrees with your pre-decision rule assessment. But ultimately, it's up to us to know how and when to apply these rules, and it's up to us to know when the rule is wrong. And if you're relying on the rule to be smarter than you, you're falling into a trap. It's not like a diagnostic study with another physician's name, judgement and license attached to it, like a radiologic study, or heart cath.

Can you blame the MD who's lead author on the paper that invented the decision rule, if it turns out to be wrong?

Can you or the patient sue those who developed the decision rule, if it turns out your patient was in the excluded patient types and you miss something?

No.

Decision rules can be good learning tools.
They do not replace clinical judgement.
Ultimately, it's all you.

Along these lines (from Annals, Sept 2017):

Structured Clinical Decision Aids Are Seldom Compared With Subjective Physician Judgment, and Are Seldom Superior

http://www.annemergmed.com/article/S0196-0644(16)31520-7/fulltext

Props to Fox for the Weingart link. Hadn't seen that but completely agree - a valuable assessment tool, but I would expand it to include more than just massive transfusion. In fact, given the findings above, it may be the most valuable clinical decision aid ever.

FWIW, I document HEART scores, Centor criteria, and NEXUS in my charts all the time, but I actually use the PORT score more than I thought I would.
Sadly, while our Canadian friends have valuable lower extremity imaging rules, they aren't terribly helpful, since everyone gets x-rayed in triage anyway.
 
LRINEC had an impressive original paper, however the validation study showed that it was significantly less reliable than originally claimed and that it was easy to miss early cases of Nec Fasc by using it.
Validation of the laboratory risk indicator for necrotizing fasciitis (LRINEC) score for early diagnosis of necrotizing fasciitis - ScienceDirect
I like a very high LRINEC score to convince a surgeon. If it's low and my suspicion is very low, I feel good about that. If it's low and my suspicion is still high, that's different.


OP, the test for a DVT is an ultrasound. I've used the Wells DVT score exactly no times. There is some utility for a d dimer in r/o DVT only if you're doing 2-3 point compression DVT studies, but completely unnecessary for a full DVT study.
 
San Francisco syncope isn't helpful, but you might get asked about it.

I think it's a very helpful tool for new residents (with an easy to remember mnemonic) and a great teaching opportunity for syncope cases. Less helpful as an attending and after you've developed a certain degree of clnical gestalt, pattern recognition and mastery of core concepts/broad ddx but I still put "San Francisco Syncope/OESIL negative" in many of my syncope charts today. If you let them go home, it looks good that you've applied a well known and validated CDR regarding syncope IMO.
 
I think it's a very helpful tool for new residents (with an easy to remember mnemonic) and a great teaching opportunity for syncope cases. Less helpful as an attending and after you've developed a certain degree of clnical gestalt, pattern recognition and mastery of core concepts/broad ddx but I still put "San Francisco Syncope/OESIL negative" in many of my syncope charts today. If you let them go home, it looks good that you've applied a well known and validated CDR regarding syncope IMO.
It failed external validation.
 
It failed external validation.

Did it?

Prospective validation of the San Francisco Syncope Rule to predict patients with serious outcomes. - PubMed - NCBI

San Francisco Syncope Rule to predict short-term serious outcomes: a systematic review

https://onlinelibrary.wiley.com/doi/pdf/10.1111/acem.12130

However, the San Francisco Syncope Rule, derived by Quinn and colleagues in 2004,3 is the only prediction rule for serious outcomes that has been validated in a variety of populations and settings.
 
Last edited:
Top