Trends and Issues in EMS

Carry a little card with the GCS score on it, and use it!

Author: Dr. Ed Bartkus

When Teasdale and Jennett published their seminal article entitled “Assessment of Coma and Impaired Consciousness – A Practical Scale” in the Lancet in 1974, the Glasgow Coma Scale was born.¹ They reported that different observers were able to elicit the responses using this scale “with a high degree of consistency, and the likelihood of ambiguous reporting appears to be small.” This was demonstrated by having several doctors and nurses examine the same group of patients. “Disagreements were rare” according to their unpublished data. It was originally designed to help in the management of ”recently brain-damaged patients.” Interestingly, their motor component only had 5 (rather than 6) descriptors.

It’s a fairly simple scoring system, and you get three points for just being there. We put a tremendous amount to importance on the scores – they are used to determine need for intubation, a full Trauma Team response, and perhaps entry into (or exclusion from) a field trial or research project. I hear a range given frequently – “they’re about a 10 or 11” – this does not inspire confidence…

Despite the “rare” disagreements initially reported, interrater reliability (the ability of two people to get the same score looking at the same patient) has not been good. There have been ideas proposed to try to make this 15 point scale even easier by only using the 6-point motor score. In 1998, Ross et al looked at 1,410 patients brought in to a Level I trauma center over a 2-year period.² They reported that the motor subscore (GCSM) had exactly the same ability to predict the severity of head injury on the Abbreviated Injury Scale (AIS) as the whole GCS. Interestingly, GCS scores were obtained on only 44% of the 3,235 patients transported during that time.

Flash forward to March, 2015 when Feldman et al published a paper entitled “Randomized Controlled Trial of a Scoring Aid to Improve Glasgow Coma Scale Scoring by Emergency Medical Services Providers” in the Annals of Emergency Medicine.³ Participants were 178 EMTs and paramedics who brought patients into an ED in Cincinnati. Approximately half were paramedics, and they represented 41 diverse agencies – urban, suburban, and rural; paid and volunteer. The mean length of experience was 12 years, and most (70%) of the participants reported that they had been refreshed on GCS score through a formal course or CME within the preceding year. Fifty-six percent reported that they regularly used some type of aid in the field to help determine the GCS score.

Nine standardized brief patient scenarios were modified from three widely-used EMS textbooks. The scenarios had patients with mild, moderate, and severe traumatic brain injuries, and an “expert consensus” was generated by a panel of paramedic instructors. Single, randomly chosen scenarios were placed in an envelope, either with or without a scoring table. Participants were given an envelope and asked to calculate both the total score and the subcomponent scores for eye, verbal, and motor response.

Overall, 41% of the participants gave the correct score – 51/90 (57%) in cases when they had the scoring table as an aid, and 22/88 (25%) if they did not. Overall scores fell within 1 point of the correct GCS score in 69% of the cases. There was an equal likelihood of underestimating and overestimating the score. Using the aid made it more likely to be within 1 point (82.2% vs 55.7% without the aid). Twelve participants gave subcomponent scores that are not possible on the scale. Differences between the two groups were more pronounced in the moderate and severe injury scenarios. Now recall the recommendation to eliminate the total score (and the eye and verbal subcomponents) – believing that the motor score alone discriminates sick vs. not sick patients. In the present study, the motor subcomponent score was the least accurate and reliable.

It’s clear that having a scoring table to use in real time more than doubles the accuracy of the score generated. Make this a part of your everyday practice!

1. Teasdale G and Jennett B: Assessment of Coma and Impaired Consciousness – A Practical Scale.
Lancet July 13, 1974; 81-4.

2.Ross S et al: Efficacy of the Motor Component of the Glasgow Coma Scale in Trauma Triage.
J Trauma Inj Infect Crit Care 1998; 45(1):42-4.

3.Feldman A et al: Randomized Controlled Trial of a Scoring Aid to Improve Glasgow Coma Scale Scoring by Emergency Medical Services Providers.2015 Ann Emerg Med; 65(3):325-31.

To read additional articles please click here