Glasgow Coma Scale: nursing assessment guide

LS
By Lindsay Smith, AGPCNP
Updated March 20, 2026

The Glasgow Coma Scale (GCS) is a standardized tool for assessing level of consciousness in patients with acute brain injury or neurological compromise. Nurses use it to detect deterioration early, communicate findings accurately, and guide escalation decisions. If you are heading into any acute, critical, or emergency care setting, you will use the GCS regularly — and you need to know it cold. This guide covers how each component is scored, what the total means, and the specific mistakes that trip up nursing students on clinical rotations and the NCLEX alike.

What the GCS is and where it came from

The Glasgow Coma Scale was developed by Sir Graham Teasdale and Bryan Jennett at the University of Glasgow and published in The Lancet in 1974. Their goal was to create a simple, reliable, reproducible method for describing the level of consciousness following head injury — something that could be used consistently across different clinicians and settings without requiring specialized equipment.

Before the GCS, descriptions of consciousness level were highly variable. Terms like “stuporous,” “obtunded,” and “lethargic” meant different things to different clinicians. The GCS replaced that ambiguity with a structured, numerical assessment based on three observable behaviors: eye opening, verbal response, and motor response.

The scale has since become one of the most widely used clinical tools in the world, applied not only in traumatic brain injury but in any situation where level of consciousness needs to be tracked objectively — stroke, hypoxia, metabolic encephalopathy, drug toxicity, post-operative care, and more.

The three components: scoring criteria

The GCS has three components. Each is scored independently, and the scores are then summed for the total. The maximum score is 15 (fully conscious, oriented, following commands). The minimum is 3 (completely unresponsive across all three components).

Key principle: Always document each component score separately (e.g., E3V4M5 = 12) rather than the total alone. A total of 10 could reflect several different clinical pictures; the breakdown tells you which one.

Component Response Score
Eye opening (E) Spontaneous 4
To voice (verbal command) 3
To pain 2
None 1
Verbal response (V) Oriented 5
Confused 4
Inappropriate words 3
Incomprehensible sounds 2
None 1
Motor response (M) Obeys commands 6
Localizes pain 5
Withdraws from pain 4
Abnormal flexion (decorticate) 3
Extension (decerebrate) 2
None 1

How to assess each component

Eye opening (E — scored 1 to 4)

Start without stimulation. If the patient’s eyes are already open and they are awake, score 4 (spontaneous). If the eyes are closed, speak to the patient — use their name and a clear instruction like “Open your eyes.” Eyes opening in response to voice scores 3.

If there is no response to voice, apply a peripheral pain stimulus. The trapezius squeeze (pinching the trapezius muscle at the shoulder) or sternal rub are common choices. Eyes opening only in response to pain scores 2. No eye opening despite adequate stimulation scores 1.

Important: Only score 4 (spontaneous) if the patient opens eyes without any prompting. Eyes that are open but not due to arousal — for instance, because the patient’s eyelids are taped open or because of facial trauma — do not count. Score what the patient actually does, not what their anatomy allows.

Verbal response (V — scored 1 to 5)

Ask the patient three orientation questions: What is your name? Where are you right now? What is today’s date or what year is it? A patient who answers all three correctly and holds a coherent conversation is oriented — score 5.

If the patient is clearly engaged and talking but gives incorrect answers or seems confused about their situation, score 4 (confused). If speech is present but consists of isolated words that do not form coherent sentences — especially if those words are random or inappropriate to the context — score 3 (inappropriate words). Moaning, groaning, or grunting without recognizable words scores 2 (incomprehensible sounds). No verbal output despite stimulation scores 1.

For intubated patients: When the patient cannot produce verbal output due to an endotracheal tube or tracheostomy, the verbal component is scored as “T” (not testable) and the total score is written with a T suffix — for example, 7T. The maximum possible score for an intubated patient is 10T. Never assign a verbal score of 1 to an intubated patient as if they chose not to speak — that misrepresents their actual status and can lead to clinical misinterpretation.

For sedated patients: Sedation also suppresses verbal and motor responses. Always document the current level of sedation alongside GCS scores. A GCS of 3 in a deeply sedated patient carries a very different clinical meaning than GCS 3 in an unsedated one.

Motor response (M — scored 1 to 6)

Motor response is the most clinically informative component and the one that requires the most careful technique.

Start with the verbal command: “Hold up two fingers” or “Squeeze my hands.” A patient who follows a command correctly scores 6 (obeys commands). If there is no response to command, apply a central pain stimulus — the trapezius squeeze, sternal rub, or supraorbital pressure are standard. Do not rely on a nail-bed squeeze for determining localization; that produces a spinal reflex that can look like withdrawal or localization when it is neither.

Localization (5): The patient’s hand moves purposefully toward the pain source — for example, reaching up toward a sternal rub or moving a hand toward a supraorbital stimulus. The movement is deliberate and directed.

Withdrawal (4): The patient pulls the limb away from the stimulus but does not reach toward it. Flexion at the elbow and pulling back without any directed reaching behavior. This is a more primitive response than localization.

Abnormal flexion / decorticate posturing (3): Flexion of the arms with wrist flexion and leg extension. The classic picture is forearms curled inward toward the chest. This indicates damage to the corticospinal tracts above the midbrain. The term “decorticate” refers to the cortex being functionally disconnected.

Extension / decerebrate posturing (2): Extension and internal rotation of the arms, with leg extension. Arms extend rigidly outward and rotate inward. This indicates deeper brainstem involvement — a more ominous finding than decorticate posturing.

No motor response (1): No movement in response to any stimulus. Ensure the stimulus was adequate before scoring 1.

Assessment side: Always test the best-responding side. If the patient has a known hemiplegia from a prior stroke, testing the affected side will underestimate their true consciousness level. Test the unaffected limb and document which side you tested.

Scoring interpretation

The three component scores are added together for the GCS total:

Total GCSSeverity category
13–15Mild impairment
9–12Moderate impairment
3–8Severe impairment

GCS ≤ 8 is the critical threshold. A GCS of 8 or below generally indicates the patient cannot protect their airway reliably. This score is associated with a significantly elevated risk of aspiration, airway obstruction, and respiratory failure. In the context of traumatic brain injury, current guidelines and clinical practice treat GCS ≤ 8 as a strong indication for intubation and airway protection. Many emergency and ICU teams use “GCS of 8, intubate” as a clinical rule of thumb, though the decision always involves clinical judgment and the broader presentation.

Limitations of the scale. GCS is not a complete neurological assessment — it does not capture pupil reactivity, limb strength, deep tendon reflexes, or brainstem function independently. It can be confounded by alcohol, drugs, metabolic disturbances, and sedation. Patients with pre-existing cognitive impairment or language barriers may score lower than their true level of consciousness warrants. The GCS was designed and validated primarily for traumatic brain injury; its prognostic weight is strongest in that context. When using GCS in other clinical scenarios, always interpret findings alongside the full clinical picture.

Clinical significance in nursing practice

Serial assessment and trend monitoring

A single GCS score is useful. Serial GCS scores are essential. A patient who scores 14 at 06:00 and 10 at 08:00 is a patient in rapid neurological decline — and that trend is something a single snapshot cannot show you.

In neuro and ICU settings, GCS is typically assessed every one to four hours, with the frequency increasing when scores are falling or low. The Glasgow Coma Scale website (glasgowcomascale.org) developed by Teasdale’s team recommends that the frequency of observation should be guided by the individual patient’s condition and clinical trajectory, not a rigid one-size schedule.

What to document: Time of assessment, GCS components individually (E + V + M), total score, which limb was tested for motor, any confounders (sedation level, intubation, paralytic agents), and any change from the prior recorded score.

Escalation triggers

A drop of 2 or more points in the total GCS, or a drop of 1 point in the motor component, warrants immediate escalation. Most hospitals have RRT (Rapid Response Team) or MET (Medical Emergency Team) activation criteria that explicitly include GCS decline. Know your facility’s thresholds.

When escalating, use SBAR to communicate neurological findings: the Situation is the GCS decline and its timeline; the Background includes baseline consciousness, diagnosis, and relevant medications; your Assessment is your clinical impression of what is happening; your Recommendation is what you think the physician needs to do — come to bedside, order imaging, call neurosurgery.

Airway management

The most clinically significant use of GCS in acute nursing is airway decision-making. A patient with GCS ≤ 8 cannot reliably clear secretions, maintain airway patency, or mount an effective gag reflex. Position the patient appropriately (lateral or semi-prone if not contraindicated by injury), have suction available, elevate the head of the bed, and escalate urgently. Review your ABC mnemonic for the primary survey framework — Airway and Breathing are assessed before GCS scoring, and GCS findings directly inform what airway interventions are needed.

ICU nursing applications

In the ICU, GCS is part of the neurological assessment bundle alongside pupillary response, RASS (Richmond Agitation-Sedation Scale), and CPOT (Critical-Care Pain Observation Tool) for sedated patients. A decreasing GCS in an ICU patient may indicate worsening cerebral edema, new hemorrhage, herniation, or medication toxicity — each with a different management pathway. ICU nurses document GCS every one to two hours for neurologically compromised patients and immediately after any significant clinical event.

Common mistakes

Confusing V2 (incomprehensible sounds) with V1 (none). Any sound the patient produces — even moaning — is V2, not V1. V1 means complete silence. This distinction matters because even incomprehensible sounds indicate a higher level of brainstem activity than total unresponsiveness.

Giving intubated patients V1 instead of VT. Assigning V1 (no verbal response) to a patient who cannot speak because they are intubated misrepresents their status. Document as VT (not testable) and write the total as, for example, E3VTM4 = 7T. Never add a fictitious verbal score to an intubated patient’s total.

Testing motor response on the wrong limb. In a patient with known hemiplegia or focal neurological deficit, testing the affected side will produce a falsely low motor score. Always test the best side and document it.

Confusing withdrawal (M4) with localization (M5). Withdrawal is a reflex — the limb pulls back. Localization is purposeful — the hand moves toward the stimulus. If you are not sure, apply the stimulus again and watch carefully. Students frequently over-score this component by interpreting any flexion movement as localization.

Confusing decorticate (M3) and decerebrate (M2) posturing. Decorticate: arms flex inward toward the chest (think: toward the core). Decerebrate: arms extend and rotate outward (think: extended rigidly away from the body). Decerebrate is the more ominous finding, indicating deeper brainstem involvement. These are not interchangeable.

Failing to account for sedation. A GCS of 3 in a patient on a propofol infusion is not the same clinical picture as GCS 3 in an unsedated patient. Always document the sedation level alongside every GCS score, and report the two together when communicating with the care team.

Reporting only the total. “GCS 9” is less useful than “E2V3M4.” A total of 9 could reflect many different component combinations. When handingoff or escalating, report the components individually.

For nursing students building clinical assessment skills, the GCS connects directly to several other frameworks:

  • ABC mnemonic — the primary survey (Airway, Breathing, Circulation, Disability, Exposure) is performed before and alongside GCS. The D (Disability) component of ABCDE includes a rapid consciousness check using AVPU, with GCS as the follow-up tool for a more detailed assessment.
  • SBAR communication — used when escalating GCS decline to the physician or rapid response team.
  • OLDCARTS mnemonic — when a patient presents with altered consciousness, a symptom history using OLDCARTS (onset, location, duration, character, aggravating/relieving factors, timing, severity) provides the clinical context that makes the GCS score interpretable.
  • Electrolyte imbalances in nursing — severe electrolyte disturbances (hyponatremia, hypoglycemia, hyperammonemia) are common causes of declining GCS in medical patients, separate from structural brain injury.

This article is for educational purposes and reflects current clinical guidelines as of 2026. Always follow your facility’s protocols and the most current evidence-based guidelines in clinical practice. GCS criteria referenced from: Teasdale G, Jennett B (1974) Assessment of coma and impaired consciousness. A practical scale. The Lancet 2(7872):81–84; and StatPearls (NCBI Bookshelf, NBK513298).