STOP-BANG Score Calculation and Interpretation
The STOP-BANG questionnaire consists of 8 dichotomous (yes/no) items with scores ranging from 0-8, where a score ≥3 has excellent sensitivity (>90%) for detecting moderate-to-severe OSA, while scores ≥5 identify patients at high risk with significantly increased probability of severe disease. 1, 2
Score Calculation
Each "yes" answer receives 1 point for the following 8 items: 1
- Snoring: Loud snoring (louder than talking or heard through closed door)
- Tiredness: Daytime tiredness, fatigue, or sleepiness
- Observed apnea: Witnessed breathing pauses during sleep
- Pressure: High blood pressure or currently treated for hypertension
- Body mass index: >35 kg/m²
- Age: >50 years
- Neck circumference: >40 cm (approximately 16 inches)
- Gender: Male
Risk Stratification Algorithm
Low Risk (Score 0-2): 1, 3
- Probability of moderate-to-severe OSA: 18%
- Probability of severe OSA: 4%
- A score <2 has 95% probability of excluding moderate-to-severe OSA 4
- These patients can proceed with standard perioperative management 1
Intermediate Risk (Score 3-4): 1
- Requires additional stratification criteria
- If STOP-BANG ≥2 plus BMI >35 kg/m², classify as high risk 1
- Consider adding overnight pulse oximetry to further risk-stratify these patients 4
- The "BANG" items (BMI, Age, Neck circumference, Gender) contribute most significantly to predictive accuracy 4, 5
High Risk (Score 5-8): 1, 2
- Score 5: Probability of moderate-to-severe OSA increases to 47%, severe OSA to 26% 2
- Score 6: Probability of moderate-to-severe OSA reaches 51%, severe OSA to 30% 2
- Score 7-8: Probability of moderate-to-severe OSA reaches 60%, severe OSA to 38% 1, 2
- Odds ratio for severe OSA with score ≥6 is 11.6 compared to lower scores 2
- Specificity of 91% for moderate-to-severe OSA at score ≥6 4
Clinical Performance Characteristics
Sensitivity and Negative Predictive Value: 1, 3
- Score ≥3 for moderate-to-severe OSA (AHI >15): Sensitivity 93%, NPV 90%
- Score ≥3 for severe OSA (AHI >30): Sensitivity 100%, NPV 100%
- This makes STOP-BANG excellent for ruling out OSA when scores are low 3
Specificity Considerations: 4, 3
- Score ≥6 provides specificity of 69% for moderate-to-severe OSA 4
- The trade-off is that high sensitivity comes at the cost of lower specificity, resulting in more false positives 3
- In surgical populations, prioritize sensitivity to avoid missing cases that could lead to perioperative complications 3
Geographic and Population Variations
The STOP-BANG performs consistently across most geographic regions with AUC >0.80, except in East Asian populations where accuracy drops to 0.52. 6
- Neck circumference cut-offs may need adjustment for different populations 7
- Latin American populations may require lower NC thresholds (36.5 cm for women, 41 cm for men) compared to the standard 40 cm 7
- The questionnaire maintains high discriminative power globally with negative predictive values of 77% for moderate-to-severe OSA and 91% for severe OSA 6
Critical Caveats
Neck circumference is the single most independently predictive variable for severe OSA (OR 2.20), followed by observed apnea, blood pressure, and BMI. 5
- The "STOP" items (Snoring, Tiredness, Observed apnea, Pressure) alone have lower predictive value than when combined with "BANG" anthropometric measures 4, 5
- Patients with intermediate scores (3-4) should not be dismissed—add pulse oximetry or proceed with formal sleep study if clinical suspicion remains high 4
- A score of 0-2 nearly excludes moderate-to-severe OSA, but clinical judgment should prevail if other concerning features exist 8, 4
- For perioperative risk assessment, the ASA guidelines recommend treating patients with ≥2 categories of clinical symptoms as having at least moderate OSA even without formal sleep study 8