---

name: gcs-calculation description: Extract and calculate Glasgow Coma Scale (GCS) for ICU patients in MIMIC-IV. Use for neurological assessment, consciousness monitoring, or trauma severity scoring. license: Apache-2.0 metadata: author: m4-clinical-extraction version: "1.0" database: mimic-iv category: derived-concepts source: https://github.com/MIT-LCP/mimic-code/tree/main/mimic-iv/concepts/measurement validated: true

Glasgow Coma Scale (GCS) Calculation

The Glasgow Coma Scale assesses level of consciousness through three components: Eye opening, Verbal response, and Motor response. This concept extracts and calculates GCS with special handling for intubated patients.

When to Use This Skill

• Neurological status assessment
• Trauma severity scoring
• Sedation monitoring
• Severity scores (SOFA CNS, APACHE, SAPS)
• Consciousness trajectory analysis

GCS Components and Scoring

Total GCS Range: 3-15 (lower = worse)

Pre-computed Table

CopySELECT
    subject_id,
    stay_id,
    charttime,
    gcs,          -- Total GCS score
    gcs_motor,    -- Motor component (1-6)
    gcs_verbal,   -- Verbal component (1-5)
    gcs_eyes,     -- Eye component (1-4)
    gcs_unable    -- 1 if unable to assess (intubated/sedated)
FROM mimiciv_derived.gcs;

MetaVision Item IDs

Critical Implementation Notes

1. Intubated Patients: When verbal response is documented as "No Response-ETT" (endotracheal tube), the verbal component is set to 0 and flagged with gcs_unable = 1. The total GCS is then set to 15 (assumed normal if only intubation prevents assessment).

2. Component Carry-Forward: If only one or two components are documented at a time, previous values from the past 6 hours are carried forward. This prevents artificially low scores from incomplete charting.

3. Calculation Logic:

   CopyGCS = Motor + Verbal + Eyes
   
   IF current verbal = 0 (intubated) THEN GCS = 15
   ELSE IF previous verbal = 0 THEN use current components only (don't carry forward)
   ELSE carry forward missing components from past 6 hours
   

4. Sedated Patients: Per SAPS-II guidelines, sedated patients should use pre-sedation GCS. In practice, if documented as "unable to score due to medication", this is flagged.

5. Time Series: Each row represents a charted observation, not an hourly aggregate. Multiple observations per hour are possible.

Example: Worst GCS Per ICU Stay

CopySELECT
    stay_id,
    MIN(gcs) AS worst_gcs,
    MIN(gcs_motor) AS worst_motor,
    MIN(gcs_verbal) AS worst_verbal,
    MIN(gcs_eyes) AS worst_eyes
FROM mimiciv_derived.gcs
WHERE gcs_unable = 0  -- Exclude intubated/sedated
GROUP BY stay_id;

Example: GCS Categories

CopySELECT
    CASE
        WHEN gcs <= 8 THEN 'Severe (3-8)'
        WHEN gcs <= 12 THEN 'Moderate (9-12)'
        ELSE 'Mild (13-15)'
    END AS gcs_category,
    COUNT(*) AS n_observations
FROM mimiciv_derived.gcs
WHERE gcs_unable = 0
GROUP BY 1
ORDER BY 1;

Example: First Day Minimum GCS

CopySELECT
    g.stay_id,
    MIN(g.gcs) AS first_day_min_gcs
FROM mimiciv_derived.gcs g
INNER JOIN mimiciv_icu.icustays ie ON g.stay_id = ie.stay_id
WHERE g.charttime BETWEEN ie.intime AND DATETIME_ADD(ie.intime, INTERVAL 24 HOUR)
    AND g.gcs_unable = 0
GROUP BY g.stay_id;

Example: GCS Trajectory

CopyWITH hourly_gcs AS (
    SELECT
        stay_id,
        DATETIME_TRUNC(charttime, HOUR) AS hour,
        AVG(gcs) AS avg_gcs
    FROM mimiciv_derived.gcs
    WHERE gcs_unable = 0
    GROUP BY stay_id, DATETIME_TRUNC(charttime, HOUR)
)
SELECT
    stay_id,
    hour,
    avg_gcs,
    avg_gcs - LAG(avg_gcs) OVER (PARTITION BY stay_id ORDER BY hour) AS gcs_change
FROM hourly_gcs;

Example: Handle Intubated Patients

Copy-- Option 1: Exclude intubated patients
SELECT stay_id, MIN(gcs) AS min_gcs
FROM mimiciv_derived.gcs
WHERE gcs_unable = 0
GROUP BY stay_id;

-- Option 2: Use motor score only for intubated (mGCS)
SELECT
    stay_id,
    MIN(CASE WHEN gcs_unable = 1 THEN gcs_motor ELSE gcs END) AS min_gcs_or_motor
FROM mimiciv_derived.gcs
GROUP BY stay_id;

References

• Teasdale G, Jennett B. "Assessment of coma and impaired consciousness: A practical scale." Lancet. 1974;2(7872):81-84.
• Teasdale G et al. "The Glasgow Coma Scale at 40 years: standing the test of time." Lancet Neurology. 2014;13(8):844-854.