British hospitals and different versions of the Glasgow coma scale: telephone survey

BMJ 2003; 327 doi: (Published 02 October 2003) Cite this as: BMJ 2003;327:782
  1. Martin F Wiese, specialist registrar in emergency medicine1 (wiese{at}
  1. 1 King's College Hospital, London SE5 9RS
  • Accepted 3 July 2003


The Glasgow coma scale is a clinical scoring system for objectively assessing how conscious a patient is. Although limited for predicting functional outcome,1 the scale is useful when making decisions about management in the acute setting, particularly for patients with traumatic brain injuries. Patients who score < 15 need imaging or observation, and patients with scores < 9 need to be promptly considered for definitive airways management.

The original Glasgow coma scale, published in 1974, had 14 points.2 Two years later, its authors introduced a distinction between “normal” and “abnormal” flexion (withdrawal to pain and decorticate response) increasing the “best motor response” item by one point.3 This revised scale is central to important clinical guidelines and has been the accepted version for more than 25 years.4

Nevertheless, anecdotal evidence suggests that neurological observation charts based on the original 14 point scale are still being used in British hospitals. This study establishes the prevalence of this practice because it carries the potential for errors in communication and conflict with guidelines.

Participants, methods, and results

I conducted a national observational study of the neurological observation charts used in hospitals which care for adult patients with traumatic brain injuries.

Using the 2001–2 directory of the British Association for Accident and Emergency Medicine, I identified all UK emergency departments which manage patients with traumatic brain injuries. We contacted a sister or charge nurse in each departments. Using a structured telephone interview, we asked staff to determine which version of the Glasgow coma scale they used by checking the neurological observation chart visually. We also asked the nurse to name one ward providing observation for patients with traumatic brain injuries, where we repeated the interview. Finally, we telephoned one ward in every specialist neurosurgical unit in the United Kingdom (table).

Version of the Glasgow coma scale on neurological observation charts in UK hospitals managing patients with traumatic brain injuries. Values are numbers (percentage)

View this table:


The original, 14 point, Glasgow coma scale continues to be used in many British hospital units which manage patients with traumatic brain injuries. Many of the staff that we contacted were not aware that the version of the scale that they were using had been superseded.

The parallel use of two versions of the Glasgow coma scale in the United Kingdom has been virtually unnoticed, possibly because publication of the revised scale in 1976 was not accompanied by an explanation and did not result in a clarifying change of name.

No evidence has been published that the continued use of the 14 point scale may have caused harm to any patient. The practice does, however, lead to difficulties. Staff from one hospital told us about their recurrent problems communicating with the local neurosurgical unit because the two were using different scales. In another trust, the neurological observation charts were changed after a recent coroner's inquest: relatives of a patient who had died of severe traumatic brain injury had raised questions about the logic of two different versions of the Glasgow coma scale being used within the same trust.

Users of the Glasgow coma scale need training to ensure consistency and reliability of scoring.5 The continued employment of two different scales can only add to the confusion. Although the 15 point Glasgow coma scale is not perfect, it should be used by everybody who manages patients with traumatic brain injuries until even better measures of consciousness are devised.


I thank D Wallis for help with the study design and E Glucksman and P Leman for their comments on the drafts. Many thanks also to all the hospital staff.


  • Contributors MFW wrote the paper and is guarantor. Nicola Burger collected some data.

  • Funding None.

  • Competing interests None declared.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
View Abstract