Without evidence of benefit, an intervention should not be presumed to be beneficial or safe.

- Rogue Medic

Motor Vehicle Intrusion – EMS Research Episode 7

ResearchBlogging.org

Also posted over at Paramedicine 101 (now at EMS Blogs) and at Research Blogging. Go check out the rest of the excellent material at these sites.

This study looks at whether passenger compartment intrusion is an effective MOI (Mechanism Of Injury) assessment criterion. This study also looks at whether replacing passenger compartment intrusion with extrication would be better. This is discussed on the EMS Research podcast at Motor Vehicle Intrusion – EMS Research Episode 7.

We chose to define the “use of trauma center resources” as one of our outcome measures, but there is no consensus on what constitutes appropriate use of a trauma center.[1]

If there is little agreement on what is appropriate use of a trauma center, how well can we say that any of these criteria predict which patients are appropriately triaged to a trauma center?

In summarizing these three studies, the CDC expert panel states that there is little evidence to support the use of the intrusion criteria, yet it is included in the latest set of guidelines.[1]

Why let reality get in the way of a protocol for trauma triage?

After all, we have never let reality interfere with trauma treatment before.

From the prehospital PCRs and ED records, two researchers (DI and DCC) abstracted patient demographic data, MVC characteristics (e.g., rollover and intrusion), final ED diagnoses (as determined and documented by either the emergency physician or the trauma surgeon responsible for ED charting), and patient disposition.[1]

One problem is that the authors then tried to figure out what the predictive value would be for criteria not used.

How much attention is devoted to documenting information that is not used to make trauma triage decisions?

We want EMS to pay attention to the assessment of the actual patient, rather than the assessment of the possible cost of repair of the vehicle.

We calculated the sensitivity, specificity, and positive predictive value (PPV) for intrusion for each of the two outcome measures. Based on our observations, we made a post hoc adjustment where we recalculated the sensitivity, specificity, and PPV for entrapment in place of intrusion. We defined entrapment as any use of tools, other than simply opening (“popping”) a jammed door with simple hand tools, to extricate a vehicle occupant.[1]

I do not spend much time documenting mechanism criteria that are not backed up by assessment findings. My chart is a medical record, not an estimate of auto repair costs.

Without any need for documenting extrication information, is extrication information documented accurately?

If only the patients ending up at the trauma center are evaluated for specificity of trauma triage criteria, how specific can we say these triage criteria are?

If 90% of these patients are transported to the trauma center (and triage criteria have excellent sensitivity, i.e. no critical trauma patients are missed), then we should have accurate information to assess specificity.

If 50% of these patients are transported to the trauma center (and triage criteria have excellent sensitivity), then we do not have accurate information to assess specificity.

If 10% of these patients are transported to the trauma center (and triage criteria have excellent sensitivity), then we do not have even remotely accurate information to assess specificity.

Another thing to consider is whether the triage criteria being evaluated are superseded by criteria that are more impressive to EMS, or given more attention by the QA/QI/CYA department. For example, the impressive criterion documented might be that the A post was cut during extrication, rather than that there was greater than a foot and a half of intrusion to the passenger compartment.

Does this mean that the extrication was necessary?

That would be difficult to determine without seeing the car before extrication was begun.

Does this mean that the extrication is relevant, while the intrusion is not?

Nobody knows.

How much does this tell us about what are the best MOI triage criteria?

Not a lot.

Why do we treat STEMIs (ST segment Elevation Myocardial Infarctions) with the opposite approach?

The dichotomy is that with trauma triage, we accept a 1,000% to 2,000% overtriage rate, while with STEMI triage, we consider a 5% overtriage rate to be unacceptably high.

What about cardiology has led us to focus exclusively on specificity, but to ignore sensitivity?

What about trauma has led us to focus exclusively on sensitivity, but to ignore specificity?

What these criteria do is discourage the use of critical judgment, which should be more sensitive and more specific than any set of trauma criteria.

If we were to spend more time on assessment than on memorization of criteria, would patients be more appropriately triaged to trauma centers?

Go listen to the podcast.

Footnotes:

[1] Motor vehicle intrusion alone does not predict trauma center admission or use of trauma center resources.
Isenberg D, Cone DC, Vaca FE.
Prehosp Emerg Care. 2011 Apr-Jun;15(2):203-7. Epub 2011 Jan 12.
PMID: 21226551 [PubMed – in process]

Isenberg D, Cone DC, & Vaca FE (2011). Motor vehicle intrusion alone does not predict trauma center admission or use of trauma center resources. Prehospital emergency care : official journal of the National Association of EMS Physicians and the National Association of State EMS Directors, 15 (2), 203-7 PMID: 21226551

.

Trackbacks

  1. […] Medic weighs in with his opinion on evaluating vehicle intrusion as a predictor of injury, which was the subject of a recent episode of the excellent EMS Research Podcast: In his post, he […]