Five changes to make OSA diagnosis and screening more inclusive for women.
By Sree Roy
Emerson Kerr, MBA, RRT, FAAST, sounds fed up. It’s the second time I’ve spoken with him about the sexism in obstructive sleep apnea (OSA) diagnostic pathways, and, if it’s even possible, he seems more worked up today than when we initially broached the topic.
“The rules were built for guys like me,” he says, “White, BMI greater than 30, men, period.”
His simmering frustration bubbled over after seeing the struggle of a family member, a thin woman who, despite four months of advocating for herself with her primary care physician, continues to wait for a diagnosis. She wants a name—and any therapy, please—for what is causing her restless sleep, daytime fatigue, and, the clincher, the apneas during her sleep that her bed partner has repeatedly witnessed.
“Are we screening, identifying, diagnosing, and treating people the way they should be? And the answer to that right now is no,” says Kerr, AAST president and director of clinical affairs at Nox Medical. “One group is benefiting, while the rest are marginalized.”
I agree with Kerr that the OSA screening and diagnosis pathways must become more inclusive to women. I don’t know if he—or the other sleep professionals I spoke to about sexism in sleep—agree with the recommendations that follow. But here are the changes I believe sleep medicine should make.
1. OSA Screeners Need to Change.
OSA screeners fail women by not including insomnia as a symptom and by giving men an extra point at the outset.1
“Women tend to wake quite quickly when there’s any respiratory instability,” says sleep researcher Alison Wimms, PhD, director of medical affairs, incubation, and growth at Resmed. “With this lower arousal threshold, they wake up, remember being awake (and how they couldn’t stay asleep), and they think, ‘I’ve got insomnia.’” That means screeners like the Epworth Sleepiness Scale don’t flag women with OSA as well as they flag men.
Worse, too many OSA screeners give men points just for being male. One is the Multivariable Apnea Prediction index, which Jonna L. Morris, PhD, RN, happens to have scrutinized. “We looked at the associations with their actual diagnosis of obstructive sleep apnea. And we found it was underestimating women’s risk for obstructive sleep apnea by a lot,” says Morris, an assistant professor at the University of Pittsburgh, School of Nursing, Health and Community Systems and faculty at the Department of Medicine at The Center for Sleep and Circadian Science.2
Peer-reviewed literature tells us women’s sleep apnea is different from men’s.3 Real-world data does too. More than 160,000-plus users of the SleepScore Labs app have added to the evidence of sex differences: Compared to men with OSA, women with OSA report more headaches, dry mouth, and nasal congestion. Women are also 44% more likely than men to be unaware of their snoring status.4
Solveig Magnusdottir, MD, MSc, MBA, chief medical officer at SleepImage, comments, “It is universally accepted by the field of sleep medicine that ‘children are not small adults,’ and now the field also needs to accept that ‘women are not men.’”
I agree. In my opinion, OSA screeners should add women’s symptoms, or their cutoffs should be adjusted based on sex. (Table 4 in Morris’ paper details the cut probabilities for the Multivariable Apnea Prediction index among the sex and menopausal status groupings at each apnea-hypopnea index cut criteria for OSA diagnosis.2)
In the meantime, I like the tactics R. Nisha Aurora, MD, MHS, incorporates. In addition to screening every patient with the Epworth Sleepiness Scale, Aurora, an associate professor in the Department of Medicine at NYU Grossman School of Medicine and director of Women’s Sleep Health Initiatives in the Division of Pulmonary, Critical Care, & Sleep Medicine, screens for OSA with the Insomnia Severity Index. Plus, when taking a history, she goes beyond “sleepiness.” “I ask about things like nocturia, fragmented sleep, daytime fatigue, if they are ‘feeling tired,’” she says. “I try to use more of the language and symptoms that seem more representative of their manifestation of sleep apnea.”
2. Declare Differences in HSTs.
Analogous to the growing awareness of the health equity problem of oximeters, sleep medicine needs awareness about the sensitivity and specificity of every home sleep testing (HST) device in key subgroups, which could include sexes, races, and endotypes. Only then could physicians make reliable calls on which HST to give to which patient.
“The effectiveness of any home sleep apnea test depends on its ability to provide clear and accurate insights with a sufficient level of diagnostic sensitivity that help clinicians make informed decisions—especially for patient groups more likely to experience hypopneas with arousals and without significant oxygen desaturations (women, non-obese individuals, and younger patients),” says Heidi Riney, MD, chief medical officer at Nox Health. Her colleague Jon S. Agustsson, PhD, vice president of AI and data science at Nox Medical, says the company is developing better ways to detect arousals—in lab, this is done with electroencephalography—particularly for populations underserved by type III and IV HSTs. “By enhancing sleep state and arousal detection, Nox BodySleep 2.0 aims to provide more conclusive home sleep tests,” he says of the investigational technology that could ultimately perform better in women than current options.
For Kerr’s family member, she took a single-night HST—and it came back negative. I would love to see data on whether that HST technology—and others—systematically misses women with OSA. “That’s sort of the mystery, and it’s something I’ve been asking different doctors about,” Kerr says.
For now, a potential workaround is to plan for multinight HST upfront (realistically with the same device each night, but it could also be via different technologies), having the patient test nightly for a week to increase the likelihood that an AHI greater than 5 is recorded.
3. Use AASM Recommended Hypopnea Scoring.
Even in-lab sleep testing misses women’s OSA. This is especially the case in labs using an outdated hypopnea definition that only counts flow limitations with oxygen desaturations, not arousals.5
But it can also occur when physicians do not examine the raw data. “What we often see in female patients is a lot of flow limitation during the night—sometimes they can have hours of flow-limited breathing but not necessarily clear events,” Wimms says. “Their sleep study may look kind of normal if you’re just glancing at the AHI, and you send them away because there’s ‘no problem.’”
Also, while sleep physicians know the risk of OSA increases in postmenopausal women, they don’t often account for the less subtle effects of hormones. Though an uncommon cause of a false-negative in-lab sleep test, there is evidence that women undergoing sleep testing during the follicular phase of their menstrual cycles have significantly lower AHIs than those in the luteal phase.6 “I suspect not many people have read that paper or factored that in,” says Andrew R. Spector, MD, first author of a paper on the impact of menstrual cycle phase on sleep apnea and an associate professor of neurology at Duke University School of Medicine. “But having written it, I can’t ignore the possibility.” If a woman with classic sleep apnea symptoms tests negative, consider redoing even a “gold-standard” in-lab study.
Overall, sleep physicians could help more women by scrutinizing sleep studies for abnormalities easily missed at first glance.
4. Coin a Better Disorder Name Than ‘Mild OSA.’
Many physicians (and entire countries, actually) assume mild OSA does not need to be treated. But the problem lies in the under-recognized category of mild OSA with a high symptom burden that responds well to CPAP. What sex is commonly in that category? Women.
The MERGE clinical trial of about 200 people with mild OSA emphasizes this.7 “We treated everyone with CPAP, and we found that the women improved significantly—not just above the control group, but also above the men on CPAP,” says first author Wimms. “We’re definitely not saying CPAP for everyone or to throw it out to the masses. But if people are turning up to the clinic because they’re so symptomatic, they should be offered CPAP treatment.”
Because many of these women never make it to the sleep clinic, losing their care battle in the primary care practice, I take this a step further: There needs to be a separate diagnosis for people who today are dubbed as “mild OSA” with severe symptoms—that adds credibility to the seriousness of their condition. And I would love to see corresponding metrics added to sleep study reports, say, percent of flow limitation or weighted REM AHI.
5. Hasten the Clinical Pathway.
Sleep specialists must also educate primary care providers about OSA in women and facilitate other means to hasten the clinical pathway to a specialist, as well as lessen wait times and costs upon their arrival.
“I’ve had patients who self-refer after being refused a referral by their primary doctor saying ‘You’re a woman; you can’t have sleep apnea,’” Spector says. “I don’t know any sleep doctor in the country who’s ever going to say that. We know better.”
So by the time a woman gets to a sleep practice, often “they’ve been through the wringer,” Wimms says, experiencing misdiagnoses from depression to thyroid problems to, of course, menopause (“Anything and everything is ‘menopause,’” she says.)
Then, if a sleep specialist supplies an HST that records an AHI less than 5 (see points 2 and 3), women face a wait—not to mention the additional expense—for an in-lab sleep test.
The wait for an in-lab bed is where Kerr’s family member has been waylaid for months. Many sleep physicians don’t see the problem in her having a negative HST followed by an in-lab study. After all, this is the current protocol for negative and inconclusive HSTs for people with high pretest probability of a sleep disorder. But I don’t think this protocol is good enough.
As Kerr says, “We’re all hardwired for the clinical pathway. Instead, we should say: Something is wrong with the clinical pathway if it’s marginalizing people.”
References
1. Morris JL, Mazzotti DR, Gottlieb DJ, Hall MH. Sex differences within symptom subtypes of mild obstructive sleep apnea. Sleep Med. 2021 Aug;84:253-8.
2. Morris JL, Orbell S, Scott PW, et al. Risk stratification by sex and menopausal status in the multivariable apnea prediction index. Sleep Breath. 2023 Oct;27(5):1695-1702.
3. Pihtili A, Kiyan E, Balcan B, et al; Turkapne Study Group. Sex differences in clinical and polysomnographic features of obstructive sleep apnea: The Turkish sleep apnea database (TURKAPNE) cohort. Sleep Med. 2025 Feb;126:228-234.
4. Email correspondence with Elie Gotlieb, PhD, lead applied sleep scientist, SleepScore Labs. 2025 Mar.
5. Berry RB, Abreu AR, Krishnan V, et al. A transition to the American Academy of Sleep Medicine-recommended hypopnea definition in adults: initiatives of the Hypopnea Scoring Rule Task Force. J Clin Sleep Med. 2022 May 1;18(5):1419-25.
6. Spector AR, Loriaux D, Alexandru D, Auerbach SH. The influence of the menstrual phases on polysomnography. Cureus. 2016 Nov 9;8(11):e871.
7. Wimms AJ, Kelly JL, Turnbull CD, et al; MERGE trial investigators. Continuous positive airway pressure versus standard care for the treatment of people with mild obstructive sleep apnoea (MERGE): a multicentre, randomised controlled trial. Lancet Respir Med. 2020 Apr;8(4):349-58.
Dive Deeper:
ID 102519580 © Skypixel | Dreamstime.com
The good news is that research from Prof Braem, scheduled to be presented at Sleep and Breathing next week, reports excellent success treating moderate and severe OSA, and no difference in success between men and women.