I want to start a practical discussion about OSA screening, particularly in the GLP-1 patient population. As a PCP, I'm seeing more patients for weight management, and I suspect I'm under-screening for OSA. Here are the screening tools I currently use: STOP-BANG Questionnaire: - Snoring? (loud enough to be heard through closed doors) - Tired? (daytime fatigue or sleepiness) - Observed apneas? (has anyone seen you stop breathing) - Pressure? (treated for hypertension) - BMI >35? - Age >50? - Neck circumference >40 cm (men) or >38 cm (women)? - Gender male? Scoring: 0-2 low risk, 3-4 intermediate, 5-8 high risk Epworth Sleepiness Scale (ESS): Rate likelihood of dozing (0-3) in 8 situations Score >10 suggests excessive daytime sleepiness My questions for the sleep medicine community: 1. What percentage of my obese patients (BMI >35) should I expect to have undiagnosed OSA? 2. Should I screen ALL patients starting GLP-1 therapy? 3. Is a home sleep test adequate or do I need to refer for in-lab polysomnography? 4. If a patient screens positive but refuses CPAP, does the GLP-1 they're already on count as "treatment"?

Excellent questions. Let me take them one by one. 1. Prevalence in your obese population: At BMI >35, OSA prevalence is approximately 60-70% in men and 40-50% in women. Yes, those numbers are real. The majority of your obese patients likely have some degree of OSA. Among T2D patients (who are a huge portion of GLP-1 prescriptions), prevalence is 55-85% depending on the study. The under-diagnosis rate is staggering — estimated at 80-90% of moderate-severe OSA cases in the US are undiagnosed. You're almost certainly sitting on dozens of undiagnosed cases in your panel. 2. Should you screen ALL GLP-1 patients? My strong opinion: YES. Here's why: - The pre-test probability is extremely high (most GLP-1 patients have obesity, T2D, or both) - Untreated OSA undermines the cardiometabolic benefits of GLP-1 therapy - OSA is independently associated with insulin resistance, which opposes GLP-1's metabolic effects - Identifying OSA creates a baseline for monitoring improvement on GLP-1 STOP-BANG takes 2 minutes. If score >=3, order a home sleep test. This is low-cost, high-yield screening. 3. Home sleep test vs. in-lab PSG: For uncomplicated suspected OSA in adults without significant cardiopulmonary comorbidity, a home sleep test (HST) is appropriate and has sensitivity ~85-90% compared to in-lab PSG. HST is cheaper ($200-400 vs. $2,000-3,000) and more accessible. Refer for in-lab PSG when: - HST is negative but clinical suspicion remains high - Suspected central sleep apnea, narcolepsy, or parasomnia - Significant heart failure or neuromuscular disease - Pediatric patients 4. Does GLP-1 "count" as OSA treatment? This is a nuanced question. Currently, GLP-1 therapy is NOT an FDA-approved treatment for OSA. However, SURMOUNT-OSA data may change this. In practice, if a patient has moderate OSA, refuses CPAP, and is losing significant weight on GLP-1, I consider it partial treatment. I'd still recommend a follow-up sleep study at 9-12 months to document AHI improvement.

I want to share my story as someone who went YEARS with undiagnosed OSA. I'm 42M, had all the risk factors — BMI 39, neck circumference 44 cm, hypertension, T2D. I mentioned to my PCP multiple times over 5 years that I was "tired all the time." Each time, the response was: - "You're overweight, that's probably why" - "Your blood sugar isn't controlled, that causes fatigue" - "Try exercising more" - "Maybe you're depressed" (prescribed an SSRI) Nobody ever asked if I snored. Nobody ever gave me a STOP-BANG. Nobody ever suggested a sleep study. In 2024, I started tirzepatide and my new endocrinologist — bless her — asked me three questions: "Do you snore? Does your partner say you stop breathing? Do you wake up with headaches?" I answered yes to all three. She ordered a home sleep test that night. Results: AHI 47. SEVERE obstructive sleep apnea. Lowest SpO2: 76%. I had been walking around with severe OSA for at least 5-7 years. My blood pressure had been "resistant" to three medications — turns out untreated OSA causes resistant hypertension. My HbA1c was "hard to control" — turns out OSA worsens insulin resistance. My fatigue was blamed on everything except the actual cause. After starting CPAP + tirzepatide simultaneously: - BP went from 148/94 on 3 meds to 122/78 on 2 meds - HbA1c dropped from 8.4% to 6.1% - Weight dropped from 276 to 218 lbs - I feel like a different person To the PCPs reading this: SCREEN YOUR OBESE PATIENTS FOR OSA. It takes 2 minutes and can literally save lives. My years of unnecessary cardiovascular risk and poorly controlled diabetes were entirely preventable.

your story is exactly why I started this thread. I suspect I've been that doctor who attributes fatigue to weight and diabetes without investigating further. Starting this month, I'm implementing universal STOP-BANG screening for: - All patients with BMI >30 - All patients with T2D - All patients starting GLP-1 therapy - Any patient reporting daytime fatigue, regardless of BMI I'm also adding two questions to my intake form: 1. "Has anyone told you that you snore loudly or stop breathing during sleep?" 2. "Do you wake up feeling unrested despite adequate time in bed?" For my sleep medicine colleagues: what's the most efficient pathway for getting home sleep tests for my patients? Currently I have to refer to pulm/sleep which takes 6-8 weeks, then they order the HST, then it takes another 2-3 weeks. A 2-3 month delay from screening to diagnosis is unacceptable.

The referral bottleneck is real and it's a major reason OSA stays undiagnosed. Here are some solutions: 1. Direct-to-consumer HST services: Several companies (WatchPAT, Lofta, SleepImage) allow you to order a home sleep test directly as a PCP. No sleep medicine referral needed. The test ships to the patient's home, they wear it for one night, data is auto-scored and sent to you. Cost: $150-300, often covered by insurance with proper coding (CPT 95806). 2. Your own HST program: Many PCPs are now stocking HST devices in their offices (WatchPAT ONE is disposable and costs ~$200 per test). You hand it to the patient, they take it home, and you interpret the results using the automated scoring algorithm. If AHI >5, refer to sleep medicine for treatment initiation. If negative, no referral needed. 3. Telemedicine sleep medicine: Several sleep medicine practices now offer telemedicine consultations with 1-2 week wait times. They can review your HST results remotely and initiate CPAP therapy through online DME providers. For your screening protocol: Consider adding neck circumference measurement. It takes 10 seconds and is the single best physical exam predictor of OSA. Neck >43 cm (17 inches) in men or >38 cm (15 inches) in women has a positive predictive value of ~60% for moderate-severe OSA. Also worth noting: STOP-BANG has lower sensitivity in women, younger patients, and non-obese patients. In women particularly, OSA presents differently — more insomnia, fatigue, and morning headaches rather than classic loud snoring and witnessed apneas. Keep a high index of suspicion even if STOP-BANG is low.