Skip to main content
Patient Care

Coma Scales: The Good, the Bad and the Ugly

It’s 0718, the “nursing home hour” for EMS. The day shift has just come to work at Reliable Manor, a facility in your district, and sure enough found a nonresponsive patient. If you’ve been in EMS for any length of time, you know that early calls to nursing homes are frequent as the new shift arrives, checks on patients and often finds problems.

Arriving 10 minutes later, you find an 86-year-old female lying in bed with snoring respirations. She is unarousable by either verbal or painful stimuli; her eyes don’t open spontaneously but do with painful stimuli, which causes her to moan.

She does not move her hands or arms on command or in response to pain. You note there’s a Foley catheter in place, and you touch her skin, which is warm and dry but not hot. 

The nurse is new to this nursing home—this is her first shift. She’s not familiar with this patient or her history. The patient’s vital signs are pulse 80 and irregularly irregular, respirations 15 and BP 128/92. Her temperature is 98.2°F. 

As you wait for the charge nurse to find the patient’s chart, you remember the adage “If GCS is less than 8, you must intubate.” You start trying to figure her GCS. You know she gets a 2 for eye opening because she opens to pain, but what motor response does she get? She doesn’t move her hands or arms but opens her eyes to pain. Is that a motor response or not? In a way it is, but that doesn’t fit the GCS chart’s descriptions. So does she get a score above 1? You’re unsure. She doesn’t speak but moans to pain, so you decide to give her a 2 for verbal response. This makes her GCS 5. So do you intubate her? 

Gene Gandy: Every EMT, AEMT and paramedic has been taught the Glasgow Coma Scale (GCS). It is described variously as a tool to objectively assess the degree to which a person is conscious or unconscious, a way to track changes in level of consciousness and a way to establish a baseline for awareness. It was developed in 1974 by Drs. Graham Teasdale and Bryan J. Jennett at the University of Glasgow in Scotland. Teasdale and Jennett were neurosurgery professors working in the field of head traumas. Their paper outlining the GCS was published in 1981 and since has become the gold standard for evaluation of responses from patients with traumatic brain injury (TBI). Kelly, can you describe the GCS, how it’s supposed to work and any problems you see with it?

Kelly Grayson: The Glasgow Coma Scale relies on numeric rating of three parameters: eye opening, verbal response and motor response (Figure 1). Generally, the higher the score, the greater the level of consciousness. But over the years the Glasgow Coma Scale has succumbed to that old military malady known as “mission creep.” We use it for too many things it was never designed for. Teasdale and Jennett intended it solely as a bedside scoring system for comatose head-injury patients. It was never intended as a catch-all descriptor of level of consciousness for patients without head injuries, nor was it intended to have each parameter assigned a number and tabulated into a cumulative score. Yet that’s how we use it, and it fails in those tasks to varying degrees.

One argument against the GCS is its variability between individual raters. Simply put, one medic’s “purposeful movement” may be another’s “localizes pain.” There is even variability in the verbal parameter. For instance, a patient with mild dementia may score as “confused” on the verbal component even though that’s been her baseline for years, while a confused patient with a traumatic brain injury is a different kettle of fish entirely. 

Prognostically it’s far too variable. I shudder to think of how many mildly sick bedbound patients I intubated back in the day under the doctrine of “GCS less than 8, intubate” when there was an entire wing of patients in the same nursing home with identical scores who had been that way for years.

Gandy: Jason, as an emergency physician, what importance do you give a GCS score from the medic during their first report to you when you receive a patient?

Jason Kodat: Absolutely none, and I don’t recall having ever dictated a GCS into a patient record, either. I’ve just submitted a proposal to speak at our regional EMS conference I titled “GCS and Other Wastes of Your Time.” It’s so confusing that a survey discovered that one-fourth of British hospitals were using the original form of the GCS—rather than the modern 15-point form, which has one additional item—in 2003, when the GCS was 29 years old.1 

Gandy: Steve, there’s a new coma score, called FOUR (for full outline of unresponsiveness) Score, now being touted as a better tool than the GCS. Can you describe it for us?

Steve Cole: The idea was to produce an assessment scale that was more sensitive for predicting severity and mortality than the GCS. As discussed, the GCS was originally developed as a simple assessment tool in 1974 and expanded to detect mortality in TBI in the ’80s. Since then it’s come to be used in a wider variety of conditions and now almost every patient record in EMS and emergency medicine. Despite this widespread acceptance, it has had a few problems: It doesn’t always predict mortality well, and despite its simplistic design, many providers assess it incorrectly. Attempts to modify or improve the GCS (e.g., the SMS) have been varied and troubled too. In 2005 several physicians from the Mayo Clinic scrapped the GCS entirely and developed a new scale. While at first glance it seems more complicated (it has four categories instead of three), it also seems more intuitive and easier to use. This results in a net positive, especially for paramedics and EMTs. It also has been validated extensively in multiple languages and settings and seems to be a better predictor of severity and mortality than the GCS or Simplified Motor Score (SMS) are. 

Each category is graded on a scale of 0–4, and unlike the GCS, on which dead people have a GCS score of 3, they will now get a score of 0. The four categories are eye, motor, brain stem reflexes and respiration (Figure 2). You will note that the “verbal response” of the GCS is not included, because this is the criterion most variable and affected by extrinsic or chronic factors.

Also interesting is the way the FOUR Score accounts for patients who are intubated or ventilated. It distinguishes between the completely apneic patient and one who is ventilated but still has some respiratory effort or response. This makes a lot of sense to me!

Gandy: Jason, what are your thoughts about the FOUR Score versus the GCS? Would you learn anything different from one of them versus the other in the first report from a medic?

Kodat: A medic giving me a FOUR Score report is telling me they’re keeping up with the literature, and since the recommended test for blink reflex is dropping saline into the eye from 4–6 inches, successfully doing so tells me more about the medic’s eye-hand coordination than it does anything else. But I don’t want to know about the medic right then—what I really want to hear about an individual patient is a decent neurologic exam. 

Gandy: I agree. I teach my students to do a neurologic exam in less than one minute. It’s easy to do, and it says volumes more than a numerical scale. And in fact, when I say to an emergency physician, “Doc, my patient is awake and talking to me appropriately,” doesn’t that tell her what she needs to know in 10 words? If I were to say, “Doc, my patient is talking to me but not making any sense whatsoever, and she doesn’t follow commands,” does that not tell him more than “Doc, my patient is a GCS 11!”? 

I think GCS is for long-term tracking, not for immediately meaningful information at the time of injury or illness. To me it’s much better to say, “Doc, when we got on scene, she was talking to me and making sense, but now she only responds when I pinch her trapezius, and that is with a groan.” I have painted a picture for the physician, not given a score. Which is more meaningful?

Kelly, as a medic, do you think it’s important to calculate a GCS early in your evaluation and treatment of a trauma patient, or can it wait?

Grayson: I think it’s important for risk managers and people who run QA reports without reading the narrative section. But since my run reports get returned if GCS isn’t documented and my yearly raises are partly based upon my ePCR accuracy, it is important to me.

But clinically? Not so much. I’d venture to say any competent EMT can tell when a patient is in extremis without stopping to calculate a GCS. Assessment of GCS is one of those automatic and intuitive things where you note a finding mentally without giving it much thought and recall it for documentation purposes later, in much the same way that you don’t really need a blood pressure cuff when you’re kneeling over a critical trauma patient on a scene—all you have to know is what the GCS isn’t. You can calculate what it is on the way to the hospital or later when you’re doing your PCR.

Gandy: Steve, how does the FOUR Score work with medical patients such as stroke patients or those with acute subarachnoid hemorrhage or epidural or subdural bleeds?

Cole: A 2012 study assessed the use of FOUR Score in stroke patients. It compared favorably with other scales such as the GCS. Another 2015 study compared the FOUR Score to the GCS in more than 1,000 CCU/ICU patients, also with favorable results. Most interesting is a 2016 study of 80 critically ill TBI patients in which FOUR Score not only compared to GCS but seemed to outperform the more intensive APACHE II score to predict early mortality. I think FOUR Score is more reflective of modern medicine and critical care than the GCS, while remaining simpler to implement compared to other scores. 

Grayson: Honestly, I don’t see FOUR Score changing the way I do things in the field, simply because coma scoring in general doesn’t much dictate my treatment. However, since it is more accurate and reliable than GCS, I’ll use it. For one thing, as Jason pointed out, using a scale like FOUR Score may up a medic’s street cred in the physician’s eyes, and greater respect and trust from the ED physician is a benefit not lightly discounted. I foresee the most utility in critical care transports. A great many of those patients are intubated and/or sedated, and the GCS is a pretty blunt instrument for monitoring them.

Kodat: There’s no reason to spend much time on these in the field, since they were never intended to be used in the field. There’s even less reason when the scales have an accuracy that isn’t particularly impressive. I’m going to pick on the GCS simply because it’s older and has much more research behind it.
In one study, emergency physicians assessing the same patients independently got the same GCS score only 38% of the time and were within one point of each other only two-thirds of the time—they got one-third of all scores at least two points different!2 

The very people who created it didn’t intend the numbers to be added, either, but rather used as three separate scales.3 They also once wrote: “We have never recommended using the GCS alone, either as a means of monitoring coma or to assess the severity of brain damage or predict outcome.”4 

Gandy: Looking back at the history of GCS, it’s apparent it was never intended to be used in emergency medicine; rather it was a way to track the progress of neurologic patients in the ICU. How did we get this into EMS? Is there any science that validates its use in emergency medicine? 

Kodat: The GCS makes up one component of the Revised Trauma Score, which has been shown to be predictive of mortality but isn’t particularly useful on its own. This makes sense when you understand that a GCS of 4 can predict a fatality rate of 19%, 27% or 48%, based simply on which of the three scales was ranked a 2.5 In fact, the motor component of the GCS actually outperforms the summed GCS in virtually every aspect that has been tested.

A lot of the push to use it comes from researchers. Real patients and real charts are messy. Researchers want to study things that are easy to type into SPSS (the most common statistical analysis software): yes or no, present or absent, alive or dead, and numbers. Creating a numerical scale makes messy data look very clean, regardless of whether it is and regardless of whether the scale is actually useful to the clinician at the bedside.

Grayson: In EMS we often fall victim to a cognitive bias whereby we believe that if something can be measured, it should be measured. In my estimation, the greatest use of numerical rating systems like GCS and FOUR Score is in research. The more precisely and reliably we can quantify data points, the more assured we are that we’re comparing apples to apples in medical research. They won’t much alter my clinical practice at the point of care, but I can certainly see their utility for data mining.

Cole: The GCS was not the first attempt to use a numerical scale in emergency medicine, but it has been perhaps the most widely accepted. I think the widespread use of the GCS was reflective of the growth of emergency medicine and trauma care in the ’70s and the increase in the need/desire for data. When the GCS was introduced in 1974, physicians were trying to improve communication on a patient’s condition between facilities and newly developed specialty centers. In that regard, the GCS did its job. It was simple, practical and (relatively) easy to perform and reproduce among multiple providers. 

Even as early as 1976, though, difficulties surfaced in interrater reliability. A 1991 study reaffirmed these difficulties. In some cases scoring a patient at all may have been impossible. A 1993 European study showed that in severe injuries, up to 61% of patients were “unscorable,” meaning that other factors (such as intubation, chronic medical conditions or sedation) prevented accurate scoring, something not accounted for in the original GCS. Finally, a 1998 study showed that incorrect assessment by “eyeballing” the patent was common (51%). Despite this, almost every EMS ePCR requires a GCS to be entered. 

As medicine and trauma care evolved, the GCS did not. The FOUR Score’s value is that it reflects current practices and assessments more than the GCS while keeping the practicality of a simple tool any paramedic can use. 

The real question is, should we use coma scales at all? I think Jason touched on this well. After all, the impact of these scales at the point of care is negligible. The only reason to continue this practice is in documentation, where EMS can participate in the overall research effort in emergency medicine. Otherwise, perhaps coma scales have outlived their usefulness altogether. 

Gandy: To sum up, GCS was never designed for use on the street but has been adapted, somewhat clumsily, to street use. It is useful for data miners but of little use to prehospital caregivers and emergency physicians. It may be more useful to the hospitalist who takes care of the patient in the ICU if she or he can compare the patient’s current GCS to their score on arrival in the ER. 

In my mind reliance on a numerical score minimizes the ability to actually learn the fine art of patient assessment. FOUR Score uses more components than GCS, and in order to learn how to use it, one must become adept at using more assessment parameters. This is a good thing. 

Neither GCS nor FOUR Score will tell you much about a suspected CVA patient, but a one-minute survey of the cranial nerves surely will. 

As EMS rumbles along the rocky road toward “professionalism,” it is incumbent upon all medics, of whatever level of training, to continue to learn, improve skills and stay up with changes in concepts. While announcing a FOUR Score to your ED doc may not tell her or him all that much about the patient, it may say a lot about you and enhance the trust you are extended. 

Conclusion

And with that, we return to our opening scenario.

Just as you’ve almost decided your patient must be intubated, the charge nurse finally arrives with the patient’s chart, which reveals that she is 10 years post-CVA with residual neurological deficits. She has essentially been in this same condition for the last 10 years but is usually more responsive. Reading her medication list, you see she received her nightly dose of alprazolam (Xanax) at 0430 instead of 2100, as prescribed. There is no explanation for this, but you decide intubation is not appropriate and decide to transport to rule out sepsis. Follow-up with the ED staff reveals that she was observed for four hours and then released back to the nursing facility. There was no sign of sepsis, and her level of consciousness improved with time. 

You are left to ponder her GCS and how it helped you. You talk it over with several colleagues, but they each arrive at different scores. You are puzzled.

We leave you with this scenario and ask you to decide for yourself what this patient’s GCS should have been and how spending time figuring it on scene was of any benefit to either you or the receiving nurses and physicians. We welcome your feedback.  

Figure 1: Glasgow Coma Scale
Response                                                                                        Score

Eye opening    
Opens eyes spontaneously                                                                   4
Opens eyes in response to speech                                                       3
Opens eyes in response to painful stimulation                                      2
Does not open eyes in response to any stimulation                              1

Motor response    
Follows commands                                                                                6
Makes localized movement in response to painful stimulation              5
Makes nonpurposeful movement in response to noxious stimulation   4
Flexes upper extremities/extends lower extremities to pain                  3
Extends all extremities to pain                                                               2
No response to pain                                                                               1

Verbal response    
Oriented to time, place and person; responds appropriately                 5
Converses but is confused                                                                     4
Replies with inappropriate words                                                           3
Makes incomprehensible sounds                                                           2
Makes no verbal response                                                                     1

Figure 2: FOUR Score—Full Outline of UnResponsiveness
Response                                                                                            Score

Eye opening    
Eyelids open or opened, tracking or blinking to command                      4
Eyelids open but not tracking                                                                   3
Eyelids closed but open to loud voice                                                      2
Eyelids closed but open to pain                                                               1
Eyelids remain closed with pain                                                               0

Motor response    
Thumbs-up, fist or peace sign                                                                  4
Localizing to pain                                                                                      3
Flexion response to pain                                                                          2
Extension response to pain                                                                      1
No response to pain or generalized myoclonus status                             0

Brain stem reflexes    
Pupil and corneal reflexes present                                                           4
One pupil wide and fixed                                                                          3
Pupil or corneal reflexes absent                                                               2
Pupil and corneal reflexes absent                                                            1
Absent pupil, corneal and cough reflex                                                    0

Respiration    
Not intubated, regular breathing pattern                                                   4
Not intubated, Cheyne-Stokes breathing pattern                                      3
Not intubated, irregular breathing                                                             2
Breathes above ventilator rate                                                                  1
Breathes at ventilator rate or apneic                                                         0

References

1. Weise MF. British hospitals and different versions of the Glasgow coma scale: telephone survey. BMJ, 2003; 327: 782.

2. Gill MR, Reiley DG, Green SM. Interrater reliability of Glasgow Coma Scale scores in the emergency department. Ann Emerg Med, 2004 Feb; 43(2): 215–23.

3. Teasdale G, Jennett B, Murray L, Murray G. Glasgow coma scale: to sum or not to sum. Lancet, 1983 Sep 17; 2(8,351): 678.

4. Teasdale G, Jennett B. Assessment of coma and severity of brain damage. Anesthesiology, 1978 Sep; 49(3): 225–6.

5. Healey C, Osler TM, Rogers FB, et al. Improving the Glasgow Coma Scale score: motor score alone is a better predictor. J Trauma, 2003 Apr; 54(4): 671–8.

Additional Resources

• Bruno MA, Ledoux D, Lambermont B, et al. Comparison of the Full Outline of UnResponsiveness and Glasgow Liege Scale/Glasgow Coma Scale in an intensive care unit population. Neurocrit Care, 2011 Dec; 15(3): 447–53.

• Cohen J. Interrater reliability and predictive validity of the FOUR score coma scale in a pediatric population. J Neurosci Nurs, 2009 Oct; 41(5): 261–7.

• Gill M, Steele R, Windemuth R, Green SM. A comparison of five simplified scales to the out-of-hospital Glasgow Coma Scale for the prediction of traumatic brain injury outcomes. Acad Emerg Med, 2006 Sep; 13(9): 968–73.

• Hosseini SH, Ayyasi M, Akbari H, Heidari Gorji MA. Comparison of Glasgow Coma Scale, Full Outline of Unresponsiveness and Acute Physiology and Chronic Health Evaluation in prediction of mortality rate among patients with traumatic brain injury admitted to intensive care unit. Anesth Pain Med, 2016 April; 6(2): e33653. 

• Jalali R, Rezaei M. A comparison of the Glasgow Coma Scale score with Full Outline of Unresponsiveness scale to predict patients’ traumatic brain injury outcomes in intensive care units. Crit Care Research Practice, 2014; 289803. 

• Schnakers C, Giacino J, Kalmar K, et al. Does the FOUR Score correctly diagnose the vegetative and minimally conscious states? Ann Neurol, 2006; 60(6): 744–5.

• Settervall C, de Sousa R. Glasgow Coma Scale and quality of life after traumatic brain injury. Acta paul enferm, 2012; 25(3), 364–70. 

• Stambrook M, Moore AD, Lubusko AA, Peters LC, Blumenschein S. Alternatives to the Glasgow Coma Scale as a quality of life predictor following traumatic brain injury. Arch Clin Neuropsychol, 1993 Mar; 8(2): 95–103. 

• Wijdicks EF, Bamlet WR, Maramattom BV, Manno EM, McClelland RL. Validation of a new coma scale: the FOUR score. Ann Neurol, 2005 Oct; 58(4): 585–93. 

Steve Cole is an educator and training captain for a third-service EMS agency in the Northwest. He has been in EMS for over 26 years, and his background includes military medicine, technical rescue and EMS education. His EMS passions include FOAMed, street toxicology and mentorship.

William E. “Gene” Gandy, JD, LP, has been a paramedic and EMS educator for over 30 years. He has implemented a two-year associate degree paramedic program for a community college, served as both a volunteer and paid paramedic, and practiced in both rural and urban settings and in the offshore oil industry. He lives in Tucson, Ariz.

Steven “Kelly” Grayson, NRP, CCEMT-P, is a critical care paramedic for Acadian Ambulance in Louisiana. He has spent the past 22 years as a field paramedic, critical care transport paramedic, field supervisor and educator. He is a frequent EMS conference speaker and author of the book En Route: A Paramedic’s Stories of Life, Death, and Everything In Between and the popular blog “A Day in the Life of an Ambulance Driver.”

Jason Kodat, MD, EMT-P, has been in EMS for more than 15 years. He has reviewed EMS textbooks and the USFA’s Handbook for EMS Medical Directors, and lectures at regional EMS conferences regularly. He currently works as an emergency physician and associate EMS medical director at hospitals near Pittsburgh, Pa.

Back to Top