pulmonology & sleepquestionnaire

STOP-Bang Questionnaire

The STOP-Bang questionnaire is an 8-item screening tool for obstructive sleep apnoea. Each item scores 1 point: Snoring, Tiredness, Observed apnoeas, blood Pressure, BMI >35, Age >50, Neck circumference >40 cm, and male Gender.

questionnaire

when to use

Use for OSA screening in adults in primary care, pre-operative assessment, and sleep medicine clinics. Particularly important in perioperative settings where undiagnosed OSA increases anaesthetic and post-operative risk. Also useful for screening patients with treatment-resistant hypertension, atrial fibrillation, and type 2 diabetes, where OSA prevalence is high.

when not to use

STOP-Bang is a screening tool, not diagnostic. A positive screen requires confirmatory sleep testing (polysomnography or HSAT). Sensitivity is high but specificity is moderate, particularly at the ≥3 threshold — many patients will screen positive without having OSA. Not validated in children or adolescents.

clinical pearls

  • In perioperative settings, a STOP-Bang ≥3 should prompt consideration of OSA-specific precautions: avoidance of excessive opioids, lateral positioning post-operatively, continuous pulse oximetry, and possible CPAP in recovery.
  • At the ≥3 threshold, STOP-Bang is highly sensitive but not specific — it over-refers. Using ≥5 as the threshold for 'high risk' improves specificity and is more appropriate for targeting diagnostic sleep studies in resource-limited settings.
  • Neck circumference >40 cm (16 inches) is measured at the level of the thyroid cartilage. This is often omitted in practice but is one of the strongest individual predictors of OSA severity.
  • STOP-Bang was designed for screening, not severity assessment. Once OSA is confirmed, the AHI (apnoea-hypopnoea index) from polysomnography determines severity (mild 5–15, moderate 15–30, severe >30).
  • Consider OSA screening in patients with resistant hypertension (≥3 antihypertensives), unexplained atrial fibrillation, or metabolic syndrome — the prevalence of undiagnosed moderate-severe OSA in these populations exceeds 30%.