Tailored Care For Sleep Apnea & Insomnia

Sahil Chopra, MD

Thank you, Dr. Wells. Thanks for having me. I am excited to talk to you today.

Audrey Wells, MD

Yes. I wonder if you could kick us off with your motivation to start a program called Empower Sleep. You are one of the co-founders, and I imagine your experiences in sleep medicine, a traditional model, had something to do with you doing something different and with Empower Sleep.

Sahil Chopra, MD

Yes, I know it is a great question, and I think I am blessed in the sense that I never had the opportunity to work in a traditional model. I finished my Sleep Medicine Fellowship in Boston at Harvard and had the opportunity to work with very brilliant and amazing people and the care that they deliver. The idea had always been, How do we deliver Harvard-grade sleep care online? As a fellow, we would see patients from all over the country who would want to get care from Beth Israel Deaconess Medical Center, MDH, or Boston Children’s Hospital. They delivered very personalized sleep care. But in the community, I do not think that was happening. Otherwise, they would not have come over there to see us. The idea has always been: how do we deliver? How do we replicate what they are doing and just use the first principles approach to providing sleep care, then try to do it in an online, scalable way?

Audrey Wells, MD

Sleep is very amenable to this online platform. I wonder if you can help everyone understand that when a person comes to Empower Sleep, what do they experience in terms of their evaluation?

Sahil Chopra, MD

Yes, I think that to get a better sense of how we are different, it makes sense to maybe talk about what the traditional model is and what typically happens if someone has a suspected sleep problem, either because they think something is wrong or because they have some co-morbidities that are linked with a sleep disorder. Their primary care pediatrician tells them that they need to go get a sleep checkup or they are wearable. I have a Whoop, and I have a garment. That either of these sorts of wearable tools tells us something is not healthy about your sleep. 

Once patients become aware of that, what typically happens is that they will do a sleep test. This is a one- or two-night test that happens either at home or in a lab. That test is treated as a binary tool in the sense that either you have sleep apnea or you do not have sleep apnea, and there is this binary outcome. If they have sleep apnea, then that patient ends up getting a CPAP device or a dental appliance. There is usually a very poor aftercare program for these patients. But if they do not have sleep apnea but do not sleep well, unfortunately, these patients get lost in the system. This is the current standard of care today. From the first referral to finally getting your CPAP or finally getting acclimatized to a sea bed or dental appliances, it is very easy for that to be a six-month journey that while also not have a very good outcome. 

That is the standard of care, what we do is fundamentally different in the sense that we use this idea of continuous sleep testing. If you are familiar with these continuous glucose monitors, these little patches that go on your skin measure blood glucose in a very dynamic fashion, and one starts to realize that blood sugars are very dynamic, and how you walk, what you eat, how much exercise you do, and how much sleep you get influence blood sugars in a fundamentally different way. The same thing is true for our sleep. If one starts doing continuous sleep testing over time, we can learn that alcohol. Alcohol influences my sleep, exercise influences my sleep, and eating too late influences my sleep. Then, if you take that a little bit further and say, How does this treatment influence my sleep? Then that unlocks this whole dimension of being able to provide people with personalized sleep care while continuously doing sleep testing to see what is working and what is not working. 

What happens with us is that the patient will sign up, and they will see one of our doctors. They will do two weeks of sleep testing using these medical-grade sleep testing devices. They can look at their sleep data every morning in a very patient and patient-friendly way and understand what is happening. This two-week trend is way more useful than having a single-night snapshot of someone’s sleep health. We then use those two weeks of data to come up with a very personalized care plan. Maybe it is CPAP, maybe it is body position, or maybe it is something as simple as a nose strip. If they have mild sleep apnea, it could be a nose dilator or nasal spray. I saw your talk on CPAP alternatives. There are so many different things that can be used for people who have, say, sleep apnea or want to improve their sleep health. But if you continue to test over time with these different modalities, then you can figure out what the best solution is for that individual. 

We do that already. If we think about the management of hypertension, if someone has high blood pressure, we start them on a diuretic, a beta blocker, or some medication. We rechecked it. We tell the patient, Here is a blood pressure cuff. Check your blood pressure every couple of times a week. Then let us circle back in a couple of weeks to see if this is working or not. If it is not, let us do something else. Let us not leave this disorder or this problem untreated for months or decades. Let us quickly figure out what is working and what is not working and get you on a program that is working for you. Empowers Sleep, does exactly that. Except for sleep, specifically sleep apnea and insomnia.

Audrey Wells, MD

Yes. It makes so much sense because, in my practice, I cannot tell you the number of times someone has reported back to me that their sleep on the night of the test was different than what they usually experience. That does not only take place in the sleep lab, where you would expect that night to be quite different than at home. It also takes place in a home setting, and I know that there is new data sort of quantifying that better and bringing to light the idea that if we test sleep on multiple nights, we can get a more comprehensive picture of what a person is experiencing because there may be major differences, even from a weekday to a weekend night, for example, or different things influencing sleep, as you pointed out: exercise, alcohol, and food. This is a way to get a full-body personal experience of somebody’s sleep. The testing modality is the ring that you just showed, which tends to be very unintrusive and something that a person could easily handle for two weeks.

Sahil Chopra, MD

Yes. Then also, continue to use it over time, making the in-lab experience more of a one-night experience. There is quite a bit of data published on this already about how much night-to-night variation there is in sleep health over, say, three nights. But it is even more true when you start getting more nights of data. It is also exciting because if one can start figuring out what it was that they did that gave them bad sleep, there can be a feedback mechanism to not do that or, of course, time. If we figure out what it is—what is it that I did that gave me healthier sleep?—I can try to replicate that over time. It creates this closed feedback loop system that trains a healthier behavior towards at least trying to get higher quality sleep over time, for sure.

Audrey Wells, MD

Yes. Sleep is so unique. In that the person who is experiencing it is unconscious. That is sort of a blind spot in their lives. With this data that you are capturing, you are allowing them to make connections between what is important to them during the day and how it affects their sleep. You brought up the idea of continued testing even after a treatment intervention. This is key for both patients who are struggling with insomnia and patients who are moving forward with sleep apnea treatments, especially when sleep apnea treatments do not always fix the sleep, even if they fix the apnea. Can you talk about that more?

Sahil Chopra, MD

Yes, I know. That is, I think, something that we do not talk about enough. The first assumption that I think the clinicians, including the former version of myself, made is, Let us say you have someone who has sleep apnea and you put them on CPAP. We sometimes assume that the happy face that we get from the CPAP device is a reflection of happy sleep, but it is not. It is just a reflection of better breathing or happy breathing. It is not a reflection of sleep; it can be a reflection of sleep, but not necessarily. Because sleep and breathing are two separate things. There can be overlap when breathing is bad, and it can disrupt sleep. But that is not always the case. Sometimes, if you try to improve breathing with, say, a CPAP device, it is disruptive to sleep. The only way to do that is if you measure it. Doing a sleep test on a treatment modality, say something, CPAP is very critical to see if it is the cost of fixing my breathing coming at the cost of breaking down my sleep. If that cost is too high, then we need to think of a different solution, because if the cost is too high and then I am not going to feel substantially better the next day, it is going to be harder for me to keep using this therapeutic modality over a course of time. In medicine, we use this idea called compliance. It is silly, but the overall compliance or adherence to that treatment is going to be lower, and who does not want to wake up rested? If something helps people, the feeling of waking up rested is so powerful that people try to go back to that frequently. If you can help somebody wake up rested, there will be a natural gravitation towards adherence to that treatment. Retesting on CPAP is critical to figure out: one, is it fixing breathing, and two, is it improving my sleep? If it is not, then we need to think of other potential solutions.

Audrey Wells, MD

Yes, I agree. I am glad that you brought up this idea of compliance, a word that I do not use, because it suggests some degree of submission to this treatment imposed by the health care delivery system. I think that it is very unsatisfying when patients struggle to use a treatment CPAP, even though all of their numbers from the machine look good. I think it does damage to the relationship with their sleep medicine physician when there is a discrepancy between the good numbers that the CPAP is reporting and how the patient is experiencing their sleep or even their daytime alertness.

Sahil Chopra, MD

Yes. We see that often the machine will underrepresent the burden of residual disease, meaning if someone has it, the machine will tell you that you are having 0.1 happiness per hour, for example, or, I think, respiratory events per hour. But then, when you do a sleep test on that, on CPAP, you see that there is more than what the machine is telling you. Then too, if it destroys your sleep, then we need to think of a different solution, and you can see it now objectively.

If someone was having way more deep sleep without CPAP and then on CPAP, they had less deep sleep, or what we call stable sleep, then this is not a good solution for that individual. This sleep disruption over the coming decades will have its downstream consequences. It just tells us how little we know about sleep medicine in general. It is sort of like our field is in its infancy from how much we understand about it.

Audrey Wells, MD

I find it humbling, to be honest. I think that there are probably still things we do not understand very well about sleep quality. I am wondering, with the device that you are using to measure multiple nights of sleep, what are the components captured that feed into this sleep quality index?

Sahil Chopra, MD

Yes. The way that these recorders work is by capturing what is called the polysomnographic signal. It is just a smart way of saying it is the oxygenation signal throughout the night, and through that signal, because we know you and I are talking now, I am awake. You are awake. My autonomic nervous system is behaving in a specific way. I am breathing, and my heart rate is behaving in a specific way. But when I am in, say, REM sleep or deep sleep, in N3-sleep, my autonomic nervous system will behave in a very fundamentally different way. REM and Wake are very similar, but deep sleep is very specific, and it is hard to replicate that pattern from the autonomic nervous system. The way these records work is by looking at what is happening to heart rate and what is happening to breathing and matching those patterns with what happens to what is supposed to look like in deep sleep, REM sleep, and being awake. That is how these recorders work. They work with a pretty high level of accuracy. 

Now, answering your question about sleep quality, what we can also start figuring out is that the higher the amount of deep sleep, the healthier that individual’s sleep will be. This idea of sleep quality is heavily weighted by how much deep sleep one has as well as how much-unwanted arousal an individual has. Those unwanted arousals are called fragmentation. Fragmentation can either be what is called macro fragmentation, where you are awake and you have these long periods of awakenings in the middle of the night, or micro fragmentation, where it is just these tiny awakenings that are disruptive to the process of sleeping. 

This idea of sleep quality, at least from these recorders, is looking at how much deep sleep this individual has. The higher, the better. How much fragmentation or arousal does this person have? The lower, the better. It does not take duration into account. It just tells us that when this individual is sleeping, what is the health of that sleep for that period? But you and I know as clinicians that sleep quality is timing and duration; true sleep quality is timing, duration, and quality of sleep. That is how one predicts better how one feels when they wake up in the morning. But at least with these recorders, one can figure out that when I do sleep, what is the quality of that sleep?

Audrey Wells, MD

Yes. To bring this home for the people watching, I wonder if you can describe what sort of interventions you might suggest if a person has their sleep apnea treated, for example. However, the recording device is still showing fragmented sleep or periods of sleep disruption that are resulting in your patient feeling they are not well rested.

Sahil Chopra, MD

It is a complex question because what we need to ask is: is CPAP the solution for that individual? If, say, a patient has, I am just going to use this as an example of severe sleep apnea. They have prolonged periods where they have very low oxygen levels throughout the night. CPAP appears to be the best solution then for that individual. It is all. The devil’s always in the details, and it is all, so it is purely obstructive sleep apnea and not central sleep apnea. It is worse during REM sleep than during non-REM sleep. In those situations, it is quite easy to help somebody get acclimated to CPAP by modulating the pressures throughout a couple of weeks to find what the pressure pressure for that individual. If it is, we do not want it to be too low, because if it goes too low, it will cause those obstructive or impending obstructive events to cause somebody to wake up. We do not want it to be too high, because if it goes up too high, the high pressure will cause that individual to wake up. Finding that balance of pressure over time, using the data that we get from the ring, and then also finding the mask is critical. 

Thirdly, do we need to use a sedative or a hypnotic medication to help consolidate sleep? Is that making it harder for that individual to wake up? It is called the arousal threshold. We want to raise that as best we can. Then what I think becomes interesting is: let us say I use CPAP with these settings. What my care team thinks is best for me is that, if I combine side sleeping with these pressures, what does that do to my sleep? Or if I combine a CPAP with these pressures plus my functional therapy, what starts happening to my sleep? Then it just opens up this world. There are so many different permutations of things that can be done that we can sort of objectively figure out if they are helping us or not. It is a complex journey. We do not understand sleep in a way that you can do, say a month’s worth of testing and then tell them this is going to be the best solution for you, which I think will be there in maybe 5 to 10 years from now. But today, there is a little bit of trial and error and troubleshooting to determine what the best solution is and what permutation of solutions is best for that individual.

Audrey Wells, MD

I want to highlight a couple of things that you are saying, just to underscore some things that are different from what people’s experiences might be in a typical sleep clinic. One is the idea that there are an array of treatments available for sleep apnea. It is not just CPAP for everybody. I think it is important to highlight that because, in my experience, there are individuals who will avoid getting tested because they think that CPAP is going to be at the end of that road and there is no other path. The other thing that you are saying is that you can combine different treatments and start to work out for a particular person what different modalities in combination would result in sleep optimization. This idea of doing repeat testing is what lets you get there. But I am also hearing a lot of you are alluding to a sort of high-touch situation where you are collaborating with the patient. I am wondering: do you see the patient more frequently after they get started with treatment to map out that path just for them?

Sahil Chopra, MD

Yes, I know, 100%. This map is very dynamic because we do not know how one is going to respond to what treatment. It is a very high-touch care system where we see the patients at least every two weeks for the first few months, and we then also see that with the app, they can communicate with their care team. Hey Doc, I tried this, and this is what it is resulting in. Can you help me make sense of this? There is unlimited asynchronous communication between the patient and their care team. But we try to see patients at least every two weeks because it is complex. It is a mix of objective numbers and subjective feelings. Those subjective feelings and emojis can help. But sometimes you just have to have a conversation with the other individual to figure out, What is going on.

Audrey Wells, MD

What is the care team like? Who is participating in the care of a particular patient?

Sahil Chopra, MD

Our pods are led by a clinician, a physician who is a Board-Certified Sleep Specialist that I vetted out. I have a good relationship, and I trust. Then that physician will oversee anywhere between 3 and 4 nurse practitioners in that pod. Then, within that pod, there is also a medical assistant for each nurse practitioner. Each patient will be designated a pod, which consists of a sleep specialist, a nurse practitioner, and a medical assistant. On all the visits, the doctor is involved in the care; the nurse practitioner, the mid-level, is the point of contact, but the doctors are kept in the loop. We do these huddles every day, either at the beginning of the morning or in the evening, to discuss the patients that we saw that day and the patients that we plan to see tomorrow. It is a very robust system that I am proud of that we have been able to build so that we can plan for the patients that we are seeing the next day. We have sort of a closed-loop system for the patients that we did not see that day. The reason that we designed it that way is because we are designing the future of what I think is sleep medicine. Because, even many times I have seen over 100,000 hours of medical-grade continuous data over the last year, and it is confusing sometimes. I do not know what the best solution is for that individual. 

Even though I have had the opportunity to train with brilliant people. There is sometimes a little bit of experimentation involved. Having the physician, the nurse practitioner, the medical assistant, and the patient on the same team allows us to figure out what is going to be the right solution for that individual. Then also, for us to learn how this individual sleeps so differently than the other people that we have taken care of. It is quite humbling to be very candid with you, to look at it this way, and then to learn what happens to people’s sleepover time with different treatment options.

Audrey Wells, MD

Yes, I get it. I imagine that even already in your mind, you are thinking of specific situations where a patient is presented with a challenging picture. Can you give an example of a way that your high-touch system was able to overcome their sleep barrier and get them moving on to sleeping healthily?

Sahil Chopra, MD

Yes, that is happened to so many. It is sort of the routine now. but I can share many examples if I just use, for example, somebody who has insomnia. Let us say, you have somebody who has insomnia with low sleep quality. This person is not doing well. They have tried cognitive behavioral therapy for insomnia, plus maybe adding a sedative like Trazodone or Ambien. I will use Trazodone in this example because I just saw this person a couple of days ago, and when we put them on Trazodone, you can objectively see that they have more deep sleep. That is exciting and that is interesting, both from a scientific standpoint and then when you talk to the patient the next day, they feel better too. 

Then the question becomes, We have established that Trazodone is a drug that works for you, but now it is fair. What is the lowest effective dose of that medicine to help you? Like as close to physiologic sleep as possible, but not having any hangover effect or feeling groggy the next day. That is and so that will be an example where having these high touch points allows high touch points to get subjective feedback, and then high touch points to look at the data frequently allows us to figure out what is the best solution for what individual. If we use the example of, say, mild sleep apnea and snoring in someone who is minimally symptomatic, it is a tough situation because, in those situations, one of the things is that we do not want the treatment to be worse than the disease.

Audrey Wells, MD

I have had that experience where I prescribe a treatment for mild sleep apnea, but the person is feeling worse.

Sahil Chopra, MD

Yes, and it is, so the treatment becomes worse than the problem itself. Then the question becomes: what are other solutions out there that are less cumbersome, that are better than the disease itself, and that are not cumbersome over time? They can be done consistently. Many times, in these patients who have mild sleep apnea, relatively preserved sleep quality, and minimal daytime symptoms. Well, let us start. We could call it a mild sleep apnea kit which has lip tape and a nose dilator. It has a special straw. You may have seen this REM blemish therapy straw, so many times off with that. We will see. When will they ask the patient, Hey, try lip tape? This is a very controlled environment. You are going to keep testing. Your sleep apnea is mild. It is not going to kill you. Just try, and then you are probably right: the physics and physiology of keeping the mouth open during sleep are conducive to having apneic events versus if someone’s mouth is closed. 

If someone can help promote nasal patency by keeping the tongue in the right position by keeping the mouth closed and or minus some body positioning intervention, over six weeks, we can figure out whether these conservative interventions, while doing continuous testing, solve the problem of mild sleep apnea and snoring for that individual. Maybe that is another example. But these examples are given every single day. This is just how our clinic operates on a day-to-day basis, and I think we are just used to it. However, in a traditional care model, I do not think these are very frequent scenarios.

Audrey Wells, MD

Totally. I am thinking of a place that I have practiced at in the past. The truth was, that there were thousands of people on the waiting list to see a doctor. I would see a person for a new patient consultation and order testing for them. I may or may not ever see them again. Chances were, I would not. Because they turned up with a different provider they would be lost to care or any number of different outcomes. But it was extraordinarily unsatisfying because I did not feel that there was care. What you are describing is more in line with a healthcare model. I applaud your innovation in founding and running Empower Sleep. I think the title is appropriate, both for the patient and the physician running these pods. I wonder if you can tell us, Dr. Chopra, where people would go to find out more or even to connect with Empower Sleep.

Sahil Chopra, MD

First of all, thank you. It is very humbling to hear from another fellow clinician who is sort of embarking on building something different. Thank you for that. For patients, they can just go to EmpowerSleep.com. It is pretty self-explanatory if they click on this. I think it is a Fix My Sleep button in the upper-hand corner, and it is pretty self-explanatory is ready to schedule a visit and sign up, and we would love to be able to take care of you and help you find a solution that works for you.

Audrey Wells, MD

I think clicking the button that says Fix My Sleep is a great start to the end journey of that path. 

Dr. Chopra, I thank you for talking about your program and your wisdom with us today. It has been a pleasure.

Sahil Chopra, MD

Thank you so much.