Sex differences in OSA and implications for diagnosis

Q: You’ve done much research on women and OSA. How did you become so engaged in this topic?

Dr. Morris: I didn’t start my career in research. I was a clinical nurse for years, and when my kids were grown, I decided to explore something new. I came to the University of Pittsburgh with an interest in women’s health and chronic disease.

What struck me early on, working on a hospital floor, was that men often had caregivers—wives, daughters—looking out for them, while women frequently didn’t have that same sort of support, possibly because women are often the caregivers, but not always the ones receiving care. That observation really shaped my desire to improve care for women. When I got to the University of Pittsburgh, I met my mentor, Eileen Chasens, whose expertise was in OSA, type 2 diabetes and daytime function. That aligned with my interest in chronic disease, so I jumped in.

My first sleep conference was a decade ago in Seattle. I remember there was a presentation about a woman who’d had a severe car accident caused by undiagnosed OSA. It took 10 years for her to get the correct diagnosis. That story really resonated with me. I knew this was the path I wanted to follow.

Since then, I’ve published, earned grants and worked with some great colleagues. Women’s health and sleep remain my passion. Coming from a background where women’s health wasn’t prioritized, I find this work incredibly meaningful.

Q: In 2021, you and your colleagues published work in Sleep Medicine focused on symptom subtypes among people with mild OSA and related sex differences. Can you tell us about your findings and the implications for clinicians?

Dr. Morris: Yes. Prior research on symptom subtypes often lumped women into broad categories—by race, age, obesity—without unpacking how symptom presentation differs by sex. I wanted to focus specifically on mild OSA because studies tend to focus on moderate to severe cases, where women are underrepresented.

We used the same dataset Diego Mazzotti used for his work on subtypes and cardiovascular outcomes. When we re-ran the analysis to include those with mild OSA, we saw similar subtypes: disturbed sleep (insomnia-like symptoms), excessive daytime sleepiness, moderate daytime sleepiness and minimal symptoms.

The difference between the moderate and excessive daytime sleepiness subtypes was striking. The moderate group was largely captured by behavioral questions from the Epworth Sleepiness Scale (ESS) like “Do you doze off watching TV?”, while the excessive group endorsed more global statements like “I feel fatigued or sleepy all the time.”

Women were much less likely to fall into the moderate group. I think that reflects how women talk about sleepiness. They don’t typically frame it as ‘I fall asleep when I'm sitting down’. Instead, they are more likely to endorse global omnipresent statements. Consequently, more men fall into the moderate sleepiness group.

I also observed \ in the qualitative work on my dissertation project women often describe pushing through sleepiness because of their responsibilities, whereas men tended to approach their daytime sleepiness more pragmatically by taking naps.

In another recent paper I worked on using the same dataset where we looked at symptom subtypes in mild OSA, we saw that about 50% of patients changed subtypes over five years. Many transitioned out of the excessive sleepiness group, possibly because they became accustomed to the symptoms. Notably, women were more likely than men to move from moderate to excessively sleepy or conversely experience minimal symptoms. It’s difficult to pinpoint why they transitioned from the moderate sleepiness subtype, but it may be related to the Epworth Sleepiness Scale-type questions that primarily define that subtype.

Q: Your work has uncovered differences in the perception and presentation of impaired sleep between the sexes, such as in your publication about the symptoms of people with type 2 diabetes and OSA. Can you describe the differences you found?

Dr. Morris: In that study, we looked at people with type 2 diabetes at high risk for OSA, recruited using the Multivariable Apnea Prediction (MAP) index. We assessed mood using the Profile of Mood States (POMS) and found that while a higher apnea–hypopnea index (AHI) was associated with worse mood, sex didn’t moderate the effect.

Even though women are more likely to be diagnosed with depression—which can overlap with symptoms of OSA—we didn’t find a clear mechanism linking AHI to depression. It may just be a coincidental association, but we can’t say for sure.

Q: What OSA-related factors should women be aware of as they move through menopause?

Dr. Morris: Postmenopausal women have OSA rates nearly equivalent to men, even if they’re not obese. Risk rises sharply after menopause, but most screening tools don’t account for hormonal status.

Q: There are many questionnaires used to assess sleep, including the Pittsburgh Sleep Quality Index (PSQI), which evaluates sleep quality across diverse populations. Your research indicates men and women may interpret the meaning of “sleep quality” quite differently. Knowing this, what should clinicians keep in mind when using tools like the PSQI?

Dr. Morris: The PSQI asks questions like “How was your sleep last month?” and “How would you rate your sleep quality?” Men’s responses to these questions demonstrated that they may define sleep quality in terms of duration and efficiency.

For women, sleep quality aligned with questions about how disrupted their nights are. It’s about how they feel the next day. So, clinicians need to listen carefully to how women describe their sleep and consider that context when interpreting PSQI results.