STOP-BANG — How to Tell If You Should Be Tested for Sleep Apnea
Obstructive sleep apnea (OSA) is one of the most under-diagnosed conditions in adult medicine. Estimates suggest 80% of moderate-severe cases go undiagnosed — partly because the worst symptoms (loud snoring, witnessed apnea events) require a partner to notice, partly because daytime tiredness gets blamed on a hundred other things, and partly because the screening question rarely gets asked in a 15-minute primary-care visit.
The fix is the STOP-BANG questionnaire — 8 yes/no questions, validated in tens of thousands of patients across surgical, sleep-clinic, and general populations. It's the most-used clinical screening tool for OSA worldwide.
What the acronym stands for
The first four letters cover symptoms: - S — Snoring: loud, louder than talking, audible through closed doors - T — Tired: persistent daytime fatigue or sleepiness, even after a full night - O — Observed: anyone witnessed you stop breathing during sleep - P — Pressure: high blood pressure, or treated for it
The last four cover physical / demographic risk factors: - B — BMI > 35: obesity is the single biggest modifiable risk factor - A — Age > 50: prevalence rises sharply after 50 - N — Neck > 40 cm (~16 in): neck circumference correlates with airway crowding - G — male Gender: OSA is roughly twice as common in men, though post-menopausal women approach the same prevalence
Each yes = 1 point. Total 0–8.
The risk thresholds
- 0–2: low risk
- 3–4: intermediate risk
- 5–8: high risk
At the ≥3 cutoff, sensitivity for moderate-to-severe OSA is around 90% — meaning if you have moderate-severe OSA, the screen is very likely to catch it. Specificity is lower: many people score 3+ without having OSA. That's intentional — this is a screen, designed to catch cases. A positive screen leads to a sleep study, not directly to a diagnosis.
Why "male gender" is in the acronym
The original validation found that male sex was an independent risk factor — about 2:1 vs female. That's preserved in the validated tool because changing it would re-validate it. Real-world clinical use has caught up: post-menopausal women lose the protective effect and approach equal prevalence, so the question is gradually getting more nuanced. Most modern uses treat it as "sex assigned at birth" and apply clinical judgment for women over ~55.
The B-A-N-G subscore
Some validated variants of STOP-BANG note that the physical half (BMI, age, neck, sex) carries more weight than the symptom half. If your B-A-N-G subscore alone is ≥3, that's also typically considered high risk regardless of total — physical risk factors predict OSA strongly even without the user knowing they have symptoms.
This matters because symptoms depend on a sleeping partner reporting snoring or apnea episodes. People who sleep alone often don't know they snore at all. The B-A-N-G subscore catches them.
Why undiagnosed OSA matters
Untreated moderate-severe OSA contributes to: - Cardiovascular disease — chronically elevated nighttime BP, increased atrial fibrillation risk - Stroke — independently elevated risk - Type 2 diabetes — insulin resistance correlates - Cognitive issues — impaired concentration, memory, mood - Daytime accidents — falling asleep at the wheel, reduced reaction times - All-cause mortality — modestly elevated
CPAP and mandibular advancement devices are highly effective when treatment is appropriate. The bottleneck is diagnosis — most people who get treated wonder why they didn't do it sooner.
What "talk to your doctor about a sleep study" actually means
Two main paths to diagnosis:
- In-lab polysomnography (PSG): the gold standard. You sleep overnight in a sleep lab with electrodes monitoring brain waves, breathing, oxygen, heart rhythm. Detailed but expensive and inconvenient.
- Home sleep apnea test (HSAT): a simplified device you wear at home for one or two nights. Measures airflow, oxygen, breathing effort, and pulse. Less detailed than PSG but adequate for most cases of suspected moderate-severe OSA in otherwise-healthy adults. Usually less expensive and much easier to schedule.
Most insurance covers HSAT for STOP-BANG ≥3 with appropriate symptoms. Your provider's referral pattern depends on local availability and your specific risk profile.
Skip the clipboard
The Sleep Apnea Screening tool walks you through STOP-BANG, scores it instantly, and surfaces the appropriate clinical-handoff language at intermediate or high risk. History saving is opt-in (off by default) so you can take it without persisting anything. Same on-device privacy as the rest of /health.
The short version
8 yes/no questions, score ≥3 means talk to a doctor about a sleep study. Sensitivity ~90% for the conditions worth catching. Don't dismiss it because you "don't snore that loudly" — the physical factors (BMI, age, neck, sex) catch a lot of cases the symptoms wouldn't. CPAP and oral appliances are highly effective when treatment is appropriate; the bottleneck is the test you haven't taken yet.