Moderate to Severe Asthma Podcast

AI Is Here. Will It Change Asthma Treatment?

Michael Wechsler, MD; Anthony Gerber, MD, PhD


September 13, 2023

This transcript has been edited for clarity. For more episodes, download the Medscape app or subscribe to the podcast on Apple Podcasts, Spotify, or your preferred podcast provider.

Michael Wechsler, MD: Hello. I'm Dr Mike Wechsler, and welcome to Medscape's InDiscussion series on moderate to severe asthma. Today, we'll be discussing digital technology in asthma treatment and management with Dr Tony Gerber. Dr Gerber is a pulmonologist, a professor in the department of medicine in the Division of Pulmonary Critical Care and Sleep Medicine at National Jewish Health, where he's also the director of pulmonary research. Welcome everyone to InDiscussion. Tony, it's great to have you. What can you share with listeners that not many people know about you?

Anthony Gerber, MD, PhD: I spent six summers at a live-in math camp at Ohio State between the ages of 12 and 18.

Wechsler: I suppose that nerdiness is what led you to eventually go to MIT?

Gerber: That probably had a lot to do with it. I majored in math at MIT, and I'll say the math camp was nerdier than MIT was.

Wechsler: I think you're the perfect person to talk about digital technologies in asthma. We've come so far in asthma management. However, there's still a fair number of unmet needs. Patients still have exacerbations. There are still a lot of patients who have poor asthma control. In this day and age, we have so much technology that's emerged, and it's important to see whether or not we can utilize some of that technology to try to advance our asthma management plans for our patients and improve outcomes.

What are your perspectives in general on the use of digital technologies in asthma? What are some of the unmet needs, and what can we offer our patients as well as some of our providers?

Gerber: There are people who have asthma that is just bad asthma and is very hard to control. Then there are people who have asthma where you would think that their asthma should be well controlled, but for a variety of reasons like adherence, inability to access medical care, a variety of things, their asthma is not well controlled, and they have a lot of symptoms.

That is a group that you would think would be primed for digital interventions. With some kind of digital technology, you can pick up their symptoms, help them understand what medications they should be taking, maybe what their triggers are and have this system where the patient's getting constant information from the digital technology. In turn, it is helping them manage their asthma. I'm not sure that we have made as much progress in that problem as I would think that we could have.

Wechsler: There is technology everywhere. Everyone has a smartphone. So many people have home devices, like a Nest device or Alexa device. There's all these devices that are out there, and there's so many advances in terms of Bluetooth technology, even wearable devices. You'd think that we would be implementing this a lot more than we have. Why do you think that is? What have been some of the delays in terms of why we have not implemented some of the technology that we have available to us?

Gerber: I think there are a couple of things. The first is, how good is the technology? Is there a technology that is demonstrated to really solve this problem? When someone uses it, does their asthma control get reliably better through that interactive technology? That's the first question.

The second question is, even if that technology exists, how do we make sure that it's getting to the providers and getting to the patients, and that there's a closed loop? Part of the problem is that there are hundreds of different apps that are all purported to address this gap, and it's a little bewildering to both a provider and a patient to figure out which technology is the best, how to use it, and how to integrate it with the health system. There are a number of barriers that create difficulty in getting technology into the hands of patients.

Wechsler: Not to mention cost; that's probably another barrier. Who's going to pay for some of these advanced technologies to get implemented? Is it going to come out of the patient's pocket? Is it going to be the payer? How much effort is the provider going to have to utilize in order to implement these strategies with their patients?

Physicians are already overburdened in terms of dealing with phone calls, texts, portal messages, and all sorts of tasks in their electronic health record. You're well aware of that. How do you expect healthcare providers to deal with this onslaught of newness, and perhaps more complexity, in their day-to-day management? Or will it simplify their day-to-day management because maybe patients will be able to address some of their asthma concerns on their own?

Gerber: That's a big, complicated question. There's the value proposition. Who's benefiting? Is it just merely you're getting free work from the providers that are constantly responding to this interactive technology, and they're not going to like that? On the other hand, if you're not getting some level of sophistication in the recommendations, we all tune out apps that are constantly sending you information. I walk down the street, and I may get a notification on my map that tells me that there's a restaurant nearby that I might want to eat at. Eventually you just learn to ignore that.

One of the challenges with these apps is that they have to solve the Goldilocks problem? They can't do too much or too little, because either thing doesn't work. They can't constantly bombard you with information. They have to fit within an ecosystem where it isn't just creating more work and more expense for the provider under the guise of an application that should actually be saving time and effort. That's another barrier.

Wechsler: I agree with you. I think there are many barriers, and we're going to have to figure out how to overcome some of these barriers, whether it's the busyness of the physician or it's cost. Why don't we step back for a second and talk about all the different types of technologies that are potentially available?

You've been mentioning apps, but there are also smart inhalers. There are also passive devices that can listen to people's breathing. There are also artificial intelligence technologies that can help guide patients. Why don't we talk a little bit about some of those different approaches? And what are you aware of that's currently available or is going to be coming down the pipeline?

Gerber: I view the app as sort of the hub, and the different technologies that you're bringing up are sort of spokes which lead into the hub. The app might not have any monitoring. It might not even have a lot of great artificial intelligence. It might be as simple as putting in a couple of your parameters, and it spits out a recommendation every day to take the medications your doctor prescribed. Or, as you're suggesting, there are increasingly sophisticated information sources that can go into that app.

There are smart inhalers where, every time you use the inhaler, it's recorded and that would then flow into the app. The app would presumably be smart enough to do something with the information. When you talk about artificial intelligence, that can range from the inner workings of the app being simple — just taking very simple data and giving recommendations — to being very complicated, where it's integrating patient vital signs, patient activity, and maybe air pollution levels and then trying to spit out some kind of recommendation either to the provider or the patient about what to do.

Now when you mention different technologies and different platforms, that's part of the problem. You're hard pressed to say, " I'm going to use Asthma X as my asthma app." There are literally dozens of different apps that are out there. In terms of the technologies that feed into the app we could start with smart inhalers.

Wechsler: Some of those smart inhalers will not only tell you how often you're using it, but some of them will also tell you when you're not using it, perhaps reminding you to use a controller — sort of like a pill reminder that many patients rely on. But then there are also some smarter technology inhalers that can also evaluate inspiratory effort. I think that could be useful. In fact, there was a study done at our institution, National Jewish Health, demonstrating that you could actually improve asthma control by using some of these smart inhalers, which will educate the patient about how deep a breath to take and what type of effort the patient needs to put into using the inhaler.

What are your thoughts on some of those?

Gerber: Going back to what you said earlier, who's going to pay for it? I think that there are a lot of opportunities. One of the problems as a pulmonologist is that it's not like taking a pill. People might not take the pill, but at least if they take the pill, they usually know how to take the pill. You don't have to spend 20 minutes with the patient saying, put it in your mouth and drink water. With an inhaler, there's a lot of technique and different ways you use the inhaler that can affect whether or not the medication is getting where it's supposed to go. Technologies that give some insight into that are useful. Who's going to pay for that is a separate question.

We all know that using an AeroChamber is a big help with many inhalers. Providers don't even give an AeroChamber to the patient if they don't use it. There's a number of barriers here that are parallel to the use of technologies that also need to be broken down. Technologies that can give those sorts of assessment and provide the chance to improve asthma control without necessarily jumping to an injectable biologic are important. I think they make patients better and reduce costs. That's the promise of the technology, whether we're there.

Wechsler: Much more research needs to be done to demonstrate the utility of these kinds of devices in terms of improving asthma control, preventing exacerbations, improving symptoms, and improving other outcomes. I think as these technologies are developed, we will be doing more and more research in that regard.

One of the other issues that we talked about are issues of adherence and reminders. I think one of the other key components of poor asthma control relates to the poor perception that patients have of their symptoms. Many times, patients aren't aware of how bad their asthma is, and several studies have demonstrated some dissonance between patient perception and physician perception.

Sometimes patients are underperceivers, sometimes they're overperceivers, and that can also pose a problem. What about utilizing some type of passive measurement device — some type of digital technology that can be in the background, like a smartphone or a home device to evaluate people's respiratory patterns? Those technologies are being developed. We're doing a study here in National Jewish in that regard using one of those devices. What are your thoughts about that type of technology?

Gerber: It reminds me of a patient I had when I was training who had terrible asthma, really reduced lung function. I said, "Do you have any limitations?" And he said, "No, I feel pretty good." And I said," Well, how about going upstairs?" He said, "I can't walk up more than three steps before panting." It had been so long since he'd been able to breathe normally that his reporting of abnormal breathing had reset to a level that was grossly abnormal. Now, the technology that you bring up has a chance to set that patient's baseline and then potentially catch deviations from the baseline.

One of the problems is that maybe you're a big football fan, and so during the playoffs you're going to sit on your couch on Saturday and Sunday watching the playoffs and not move. Then the app might say, look, this activity has dropped significantly and give a false alarm.

There are problems, of course, with that kind of technology. The more sensors you can get and the more that you measure that patient's baseline, the more accurate the recommendations and the data.

Wechsler: Are you concerned at all that by using some of these digital technologies, we're going to remove the physician from the equation and put more reliance on smart technologies to guide our patients? Are you worried about that in any way?

Gerber: Earlier, we talked about who's going to pay for this. I had long thought that insurance would pay for this. They could say, look, your asthma control is bad; if you want to go on a biologic, you need to try, Dr Wechsler's app for 3 months. If you're no better then, you're going to go onto a biologic. That hasn't happened. Part of the reason it hasn't happened is that without the full power of artificial intelligence, the apps don't seem to be filling the need for patient-physician interaction. The most effective apps seem to prompt teleconference visits or direct communication with the physician.

It's true that maybe artificial intelligence will eventually be able to be just as convincing and say all the things that you want to say; there is something about human-to-human interactions that I think can be more effective and more convincing. Physicians can hone their message based the audience and on experience with that patient in a way in which I don't think artificial intelligence is there yet.

One more thing: A lot of these questions aren't really at the level of needing artificial intelligence. If you're not taking your inhaler, take it. If you're going to be visiting a house that has a cat and you're allergic to cats, be careful. We don't need a huge amount of sophistication for some of these problems. It's not clear to me that there's that big a need for artificial intelligence, given that we haven't solved some of the simple problems yet.

Wechsler: You've used the term artificial intelligence a lot, and that term has been bandied about increasingly for the past couple of years. How do you define artificial intelligence? How do you view its potential in asthma?

You're the MIT guy, so tell us about artificial intelligence and its potential in asthma.

Gerber: Artificial intelligence takes a lot of data and finds underlying patterns in those data. Then it gets a new example of the data, and it tries to match the new data to one of the patterns that it found and then comes up with an output. It's not really intelligence; it's a lot of overlapping patterns and recognition. That's what it does.

The drawbacks in medicine are when the pattern isn't that difficult to recognize. You don't need sophisticated artificial intelligence, but what you need is that physician-to-patient interaction to get the right kind of care implemented.

There are examples when the pattern recognition is very complicated, and as we get better at that with artificial intelligence, the apps will improve. But for a lot of the bread and butter, the problem isn't the pattern recognition; the problem is the interface between the physician, the app, and the patient. That's not a problem that artificial intelligence necessarily is going to address.

Wechsler: I view the potential for artificial intelligence to enhance the care delivered by providers — particularly primary care providers, who may not be as up to speed with all the latest nuances, guidelines, and therapies that are emerging and the biologic therapies.

One way to think about digital technology and artificial intelligence is to prompt physicians. For example, " If you know your patient has asthma, ask them these questions: How often are you using your rescue inhaler? How often does asthma interfere with your daily activity? How often is your sleep interrupted?" That might provide some impetus to either order spirometry or refer to an asthma care specialist. What are your thoughts in that regard?

Gerber: That's a dramatically underused part of what I'd call assistive technology. I think that would be very useful for primary care physicians, who have to treat 20, 30, 40 different diseases routinely. It's challenging to switch into severe asthma mode when you might only see it once a month. It could prompt a discussion about whether this is severe asthma and ask, "Have you thought about these things?" and maybe recommend referral. That would be extremely useful. That also then allows the physician to help communicate with the patient. One of the things that we have learned, and we'll learn over time, is that direct communication can often be more effective than an app or something spitting out canned information to the patient.

I think it's a great idea, but again, I'm a little surprised it hasn't been implemented in a lot of electronic medical records (EMRs) yet. I do worry that the electronic records that the physicians might end up tuning out recommendations, because the EMR might be providing so many of them, that you tend to throw the baby out with the bathwater. The challenge is that there's so much going on, healthcare providers are going to ignore it all. Interface problems, I think, are one of the things we need to work on, in addition to the underlying functionality.

Wechsler: We often get bombarded by all sorts of tasks, and I think people tend to ignore most of those alerts that come up. Maybe there's a way to prioritize the most essential ones. You're the director of pulmonary research; is there a way to implement some of these strategies in terms of guiding research strategies, like identifying new targets and helping guide research priorities in asthma?

Gerber: I know it's ironic, but insurance is a potential big player here. They are most incentivized to prevent patients getting inappropriate care. You might have a single EMR, which would be ideal to collect data and use to implement some of these events. Unfortunately, many of the patients within that EMR will have different insurance. Who's going to pay for these sort of pragmatic research programs? I think is a challenge, but there's a missed opportunity to not do more real-world intervention studies that could potentially be facilitated and made easier through the EMRs. There's a lot of opportunities there and I'd like to think that in the next few years, some of those research opportunities will be realized.

Wechsler: Do you think it's possible in the future where patients will have some of these digital technologies and be guided by the technology and not have to see a provider at all? They app could say, your asthma is poorly controlled; check your nitric oxide level and if it's high, go up on your inhaled steroid. Don't even bother calling your prescriber.

Gerber: For the patients who are extremely comfortable with digital technology — and younger folks are more comfortable — there may be a window of opportunity for relatively straightforward iterations of a number of diseases. Patients will come to rely on a digital interface is really informing their decisions while recognizing that if they have anything that doesn't go according to what they're expecting out of their disease, they might lose faith in the application in a way that they may not lose faith in their providers. I do think that people may be able to opt into more digital interventions, which may be more convenient, particularly if they're very comfortable with technology.

Wechsler: I think there still is a lot of room for improvement in asthma management. There's still a huge burden for our patients. We still are faced with significant economic burden, both direct and indirect costs, for patients with asthma. A lot of that, I think, can be addressed by using digital technologies to improve adherence, which is a major issue; improve inhaler technique; and advise physicians about the optimal strategies to manage their patients.

Today we've talked about devices, technology, artificial intelligence, and the potential to support our patients with disease self-management and allow for remote management by healthcare providers and to optimize asthma strategies. I'm really excited for the prospects of the future, with digital technologies.

Tony, I'll give you the last, word. What are your concluding thoughts, and what are your key takeaways from this discussion?

Gerber: I think that there are enormous opportunities for patients with severe and unusual disease who are going to need close care from physicians. That's just the way it goes. I think there's a huge group of patients who may have limited time for healthcare appointments. Maybe it's a kid who has asthma and both parents work, and it's a real hardship to get them into the doctor. Mindful technology that can help with those management decisions. We can make sure the patients are getting something that's reasonably close to the quality of care but might fit into different lifestyles better. It's really exciting as a potential to democratize care and make it more efficient and ultimately improve outcomes and save costs. It's a very exciting times as we move forward.

Wechsler: This has been a great discussion. I want to thank our guest again today, Dr Tony Gerber, a national expert and a professor of medicine at National Jewish Health. We had a great discussion talking about technology and asthma treatment.

I want to thank you, our guests, as well for joining us today. This is Mike Wechsler for InDiscussion. I look forward to our next episode.

Listen to additional seasons of this podcast.


Digital Inhalers for Asthma or Chronic Obstructive Pulmonary Disease: A Scientific Perspective

Mobile Health and Inhaler-Based Monitoring Devices for Asthma Management

Apps for Asthma Self-Management: A Systematic Assessment of Content and Tools

Effectiveness of a Digital Inhaler System for Patients With Asthma: A 12-Week, Open-Label, Randomized Study (CONNECT1)

Artificial Intelligence: Exploring the Future of Innovation in Allergy Immunology

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.