APACHE II Score Calculator — ICU Severity & Mortality

How to Use This Calculator

1. Enter the worst value in the first 24 hours of ICU admission for each of the 12 physiology variables.
2. Enter patient age and select the chronic health category.
3. APACHE II score and predicted ICU mortality display instantly.
4. Use the vs SOFA tab to compare scoring systems.
5. Use the SMR Benchmarking tab to calculate your unit standardised mortality ratio.

Formula

Example

Frequently Asked Questions

• What does the APACHE II score measure? ▼
  APACHE II (Acute Physiology and Chronic Health Evaluation II) quantifies illness severity in ICU patients using 12 acute physiology variables — temperature, mean arterial pressure, heart rate, respiratory rate, oxygenation, arterial pH, serum sodium, potassium, creatinine, hematocrit, white blood cell count, and Glasgow Coma Scale — each scored 0–4 based on deviation from normal. An age adjustment (0–6 points) and a chronic health penalty (0–5 points) are added. The composite score ranges from 0 to 71; higher scores correlate with increasing predicted ICU mortality, from under 5% at scores 0–4 to over 85% above 34. Developed by Knaus et al. at George Washington University and published in Critical Care Medicine in 1985, APACHE II became the most widely adopted ICU severity tool in the world and remains a benchmark for clinical trial stratification and quality benchmarking.
• How accurate is APACHE II for predicting ICU mortality? ▼
  In the original 1985 validation cohort of 5,815 ICU patients across 13 hospitals, APACHE II demonstrated an area under the ROC curve of approximately 0.86, indicating good discrimination between survivors and non-survivors. Subsequent studies across different populations generally confirm AUROCs of 0.75–0.90. However, accuracy degrades over time as case-mix and treatments evolve; modern cohort studies show APACHE II overestimates mortality due to improved supportive care. The tool performs better for medical patients than surgical or trauma patients. Calibration — how well predicted percentages match actual observed mortality — can be poor in specific subgroups. APACHE IV, published in 2006, improved predictions using updated regression coefficients from 110,558 patients, though APACHE II remains institutionally embedded due to its simplicity and widespread familiarity.
• How do I score the chronic health component? ▼
  The chronic health component adds 0, 2, or 5 points based on pre-existing organ insufficiency or immunocompromise. Score 0 points if the patient had no significant chronic disease or was admitted for elective surgery. Score 2 points if the patient had documented chronic organ insufficiency — including liver cirrhosis with portal hypertension or encephalopathy, COPD with functional limitation, chronic dialysis, cardiovascular NYHA Class IV, or immunocompromised status — and was admitted for non-operative or emergency post-operative reasons. Score 5 points for the same organ insufficiency criteria in a non-surgical patient. The higher penalty for non-surgical patients reflects data showing that elective surgical patients tolerate chronic disease better than medical patients presenting acutely. Documentation of prior diagnoses from the patient record is required for accurate scoring.
• What is the difference between APACHE II, III, and IV? ▼
  APACHE II (1985) uses 12 physiology variables and is calculated at 24 hours post-ICU admission. Its simplicity made it globally adopted. APACHE III (1991) expanded to 17 physiological variables, added lead-time bias corrections, and incorporated 41 diagnostic categories for admission reason — providing better calibration but far greater complexity. APACHE IV (2006) updated regression coefficients using a contemporary 110,558-patient database, incorporated ventilation status, urgent admission diagnosis categories, and prior ICU stay, achieving AUROC approximately 0.88 across all patient groups. Despite the availability of APACHE III and IV, many institutions continue using APACHE II because the data are available in most EMRs, staff are trained on it, and published clinical trials use APACHE II for enrollment and stratification, maintaining its relevance as a comparative benchmark.
• Can APACHE II be used for individual patient prognosis? ▼
  APACHE II should not be used to make individual patient-level treatment decisions or prognosis discussions with families. It is a population-based statistical tool: a score of 25 predicts roughly 40% mortality across thousands of similar patients, but any individual patient may survive or die regardless of that estimate. Wide confidence intervals around individual predictions make them clinically unreliable for single patients. The score is most appropriately used for: comparing illness severity across patient groups in clinical research; stratifying patients in randomised controlled trials; benchmarking ICU performance via standardised mortality ratios (observed divided by predicted deaths); and communicating illness severity within clinical teams. Patient families should receive individualised prognostic discussions based on the full clinical picture, not APACHE II scores alone.

Related Calculators