Columns, Patient Care, Special Topics

Study Shows Inaccuracies in GCS Score Among All Emergency Care Providers

Issue 11 and Volume 40.


Bledsoe BE, Casey MJ, Feldman J, et al. Glasgow Coma Scale scoring is often inaccurate. Prehosp Disaster Med. 2015;30(1):46–53.


In an attempt to determine how accurately EMS and ED professionals can determine the Glasgow Coma Scale (GCS) in simulated patient encounters, the authors of this study produced 10 video scenarios and had the providers determine the GCS.

There were 217 providers including EMTs, advanced EMTs, paramedics, critical care paramedics, ED nurses, ED physicians and emergency medicine residents. Nearly half (49%) of them had 1–10 years of experience while just over 34% had more than 10 years.

The subjects were shown videos with actors playing the patient, then given 10 seconds to measure the GCS and provide scores for each of the three components (eyes, verbal and motor). They weren’t allowed to use any aids such as printed GCS scoring sheets. The correct GCS for each scenario was determined by two board-certified neurologists.

A total of 2,084 GCS observations were made. The overall accuracy for all providers was 33.1%. Only 69.0% of verbal components scores were accurate, while 61.2% of eye-opening scores and 59.8% of motor component scores were correct.

Residents were the most accurate at 51% with nurses being the least accurate at 29%. The majority of inaccurate scores were for patients with a true GCS of 9–12.

The authors concluded, “Glasgow Coma Scale scoring should not be considered accurate. A more simplified scoring system should be developed and validated.”


I’m so happy this study was published. It echoes what several others have found: We need more training in how to calculate the GCS on the fly. How can we provide high-quality care to our patients when that care is so often linked to the proper GCS?

Study after study on the proper care of patients with traumatic brain injury (TBI) and multiple trauma are based on the GCS of the patients enrolled in these huge, multimillion dollar studies. If we as a group of professionals can’t calculate the correct GCS, how in the world will we ever be able to determine if the time-honored EMS canon of “Less than 8? Intubate” is true or just fun to say?

As important as it is for our patients to have clear do-not-resuscitate paperwork, it’s equally so for us to be able to quickly determine a simple score between 0 and 15. Or is it 3 and 12? Perhaps we need to require that all EMS providers have the GCS tattooed to their arms. But how hard can it be? Eyes, verbal, motor—4, 5, 6. Right? I mean, it has its own website for crying out loud:

So let’s get with it folks. The GCS is here to stay. The AVPU (alert, voice, pain, unresponsive) scale is just for first responders. How can we ever be taken seriously if we can’t add up three simple numbers?


You say “GCS is here to stay”? Doc, it’s almost as old as you are. The GCS is a time-consuming score that’s most often incorrect in both prehospital and emergency medicine. So why use it?

I think a clearer picture of patient status is better described with AVPU. There’s little room for error with this basic approach.

Other than a description of loss of consciousness and length of this state at the scene, I don’t think any neurosurgeon looks to our guesstimate of the GCS number. They want to know mechanism of injury and mental status on scene. The determination for recovery from a TBI isn’t made on the prehospital report of the GCS.

Personally, because the electronic record requires it, I usually determine GCS once I’m writing my report long after having delivered the patient to you and your colleagues. I don’t want to waste the time it takes to break out a chart in the field, just to give my patients a number. I’m pretty sure they would much rather have me provide the care they need, and give a concise, professional verbal report at the ED.

Trying to train us to use any scoring scheme in the prehospital setting is just a waste of time—time that would be better spent educating providers on the science of oxygenation, early shock intervention and monitoring of our TBI patients.

I think the authors of this study nailed the conclusion. The GCS is unreliable not only because no two paramedics can come to the same answer, but the actual number is irrelevant. It’s the change, or lack thereof, in the neurological status over time that’s the true predictor of outcome. In current form the GCS is a waste of valuable time, but I will continue to document it because I’m required to.

Keep GCS in the prehospital setting, Doc? Next you’ll have us doing the “Rancho Los Amigos Scale.” Yes, it’s a real score. Google it.