Using AI to help patients breathe easier
Browse Blog Topics

Using AI to help patients breathe easier

Welcome back to What’s NEXT, a podcast from Samsung NEXT exploring the future of technology. In this episode, I talk with CEO Melissa Manice about how Cohero helps patients breathe easier by tracking their lung capacity and medication adherence over time.

You can listen to the full episode in the player below, or subscribe through Apple PodcastsGoogle PlayRSS, or your favorite podcast app of choice. New episodes will be released every other week.

Ryan Lawler: Welcome to What’s NEXT, Melissa.

Melissa Manice: Thanks for having me.

Ryan Lawler: So how about just to get started you tell me what is CoheroHealth, and what do you do?

Melissa Manice: Yeah, so CoheroHealth is a smart respiratory care company and we’re transforming respiratory care through smart mobile technology with the goal of optimizing diagnosis, care coordination, and care management and improving overall patient health for those impacted by a pulmonary condition.

Ryan Lawler: Okay. How do you do that? What are the pieces that come together that enable you to improve care?

Melissa Manice: Absolutely. So when we say sort of smart mobile technology, that’s comprised of two proprietary wireless devices, one that’s able to diagnose and screen lung function, all mobile handheld, can be used at the point of care, so in a physician practice or in the home. Then the other are our proprietary medication adherence solution, also wireless, Bluetooth energy powered device, and those are sort of universal fit to fit all the different shapes and sizes of medication, inhalers for those with asthma and COPD with the end goal of really trying to drive proactive compliance to treatment plan as well as giving real time insights on response to therapy via lung function.

Ryan Lawler: Okay. So just to break that down into different parts, you have the hardware.

Melissa Manice: Absolutely.

Ryan Lawler: So what’s that hardware look like? Where are the actual pieces of hardware that people are using?

Melissa Manice: Yeah, so the first piece of hardware we call the Hero tracker is a flexible piece of silicon. Wraps around the patient’s medication, has a little sensor that captures medication utilization and then pairs wirelessly with the patient’s smart phone of any kind, and then the mobile spirometer is really trying to take what is currently sort of a large phone booth type system in a pulmonary practice and reduce it to something that is smaller than your smart phone, and so miniaturizing that, but that’s a device that’s also wireless and a patient exhales into a spirometer and captures a full sort of set of pulmonary function parameters, but really with the end goal of connecting patients in an unprecedented way to their entire care team.

Ryan Lawler: Okay. So you have these different pieces of hardware. What types of data are they collecting?

Melissa Manice: We call them digital biomarkers. The types of data they’re collecting are what are critical to understand whether patients responding well to therapy but also the types of data that are really needed to drive sort of compliance to treatment plan. Some other pieces of data that we capture that are really critical are what we call sort of disease triggers, so environmental data and other elements, but again, sort of with this holistic snapshot for the pulmonary patient community, so the types of biomarkers that are sort of critical to driving optimized outcomes for that patient population.

Ryan Lawler: So you are collecting all of this data and then on the back end what do you do with it?

Melissa Manice: We like to say that digital health in general, but really CoheroHealth’s mission is really allowing for a shift from sort of a reactive care model to a proactive care model, and what that means is that you can sort of intervene all of a sudden in real time if a patient is at risk of an acute event, so an acute event being hospitalized for example with a patient with asthma, and so what we do with that data is intervene or allow the patient to be engaged on the front end, educate them about sort of how to optimize their treatment plan, but really importantly also connecting them to care givers and their care team with really the intention of allowing for real time insights so that a provider, if necessary, can intervene before a patient has been hospitalized.

Ryan Lawler: Okay. Do you have mobile apps for patients or how are you engaging with them?

Melissa Manice: Currently it’s all mobile app based on android and iOS, also tablet based and so we really are trying to be agnostic to age, be able to connect and visualize data, and so we’ve decided that a smart phone is something that most patients don’t leave their home without but we are not currently offering sensor based connectivity to a non-smart phone based system.

Ryan Lawler: So when you talk about patient engagement on these mobile devices, what are the regulations that you have to keep in mind with health information and how do you make sure that you’re HIPAA compliant for example?

Melissa Manice: I think, first and foremost, you hit it that HIPAA compliant is something that we all think about but ensuring … and ensuring that it’s not just HIPAA compliant from the connectivity of device to app, but the entire ecosystem, so we like to think that sometimes the key to many of these solution sets is to be interoperable and so when we think about HIPAA compliance, that might be from the device connected to an app, connected to our back end, pushed into the electronic medical record. We sort of think end to end data security, so that’s sort of critical and then there’s emerging both US based and global requirements for data security, GDPR is one, so those are the kinds of things that I think we like to think about from the point of view of PHI, personal health information, and ensuring that we stay compliant from Cohero front end all the way to back end systems as well.

Ryan Lawler: Okay. We’ve been talking a lot about how the patient uses this data and these devices. What happens on the physician or care provider end of things?

Melissa Manice: Yeah, so the types of data that we generate from the patient facing what we call our BreatheSmart application, all of the data is what is grounded in clinical guidelines as being able to profile whether patients at risk of an acute event. On the back end, what we do is we basically allow for risk profiling based on an aggregate view of a patient cohort, etc., but all of this patient generated data, and so what a provider sees is, in our case, red, yellow, green zone based on a patient’s risk threshold of all of this data. We try and make it very clinician centered. Ours is sort of razor focused on the data that we collect on our application and pushing that into a really easy to digest dashboard for them.

Ryan Lawler: So, help me understand the size of the problem that Cohero is trying to solve. How many people are affected by these conditions, and how big is this health problem?

Melissa Manice: In the US, it’s about 50 million Americans with asthma or COPD and about 300 million worldwide with prevalence increasing every year. It is a leading chronic condition in both this country as well as globally, and I say leading being both asthma and COPD. Environmental factors being both indoor and outdoor air quality and other aspects have really both are seen to drive both incidents of these conditions but also exacerbate these conditions. When we think of what are the opportunities with an application on a patient engagement side, it’s to educate on what are your triggers, what are the kinds of things that you can do in your indoor home environment to remediate or improve risk of an exacerbation? One of those is roaches, dust mites, smoke exposure, etc. in the indoor home environment, and then in the outdoor air environment some of the types of data that we capture are pollen is a huge trigger, but in urban environments things like proximity, a patient’s proximity to things like bus depots and so forth.

Ryan Lawler: Got it. How do patients actually find out about Cohero? Is this something where you’re doctor recommended or are people searching for things online? How do they come across your products and how do you start to build a relationship with that?

Melissa Manice: They’re informed or finding us through their physicians and some of those other trusted stakeholders in their ecosystems, so that includes programs in our case with pharmacy benefit managers, so PBM programs, health systems, payers, we work directly with life science and pharmaceutical companies powering clinical research, so generally patients are finding us through our own B2B partners.

Ryan Lawler: Let’s actually talk a little bit more about the technology and the data that you’re collecting. Can you break that down into the types of data that you collect and what it’s telling you about the patient and about their triggers and respiratory health?

Melissa Manice: Yeah. When we … a patient presents themselves in clinic and they’ve had … been hospitalize din the last year and they have an asthma diagnosis, we try and assess what we would call a refractive patient, we’re trying to assess whether they’re non-compliant to their treatment plan or whether they’re not benefiting therapeutically from their treatment plan, and the way in which we currently isolate that clinical decision tree is to ask a whole bunch of self reported surveys on medication compliance, etc, and then we collect lung function sporadically during their clinical touchpoints in clinic.

What we do at Cohero is basically take that same sort of data structure, but make it much more objective, not based on patient’s self report, so collect medication utilization data, collect lung function data, but collect it in real time so collect it passively as the patient is living their daily life, and with the same intention for a provider of detecting that signal, and so that signal being whether they’re at risk of an event based on lung function decline. We know that patients in our case are generally about 30% compliant to their treatment plan, to their daily therapy, and what is needed to prevent an exacerbation is to be about 80% compliant.

So what we really try and do is again take that data we’re collecting and remind the patient proactively to comply with their treatment plan, educate them on why that’s important but ultimately a key piece is really again to sort of drive both real time insights for the providers but give a patient sort of a meaningful tool to drive compliance.

Ryan Lawler: Right. When you talk about ensuring compliance, you mean actually instead of having the patient self report, having the device on their medication to tell you when they’ve taken it?

Melissa Manice: Absolutely. That’s the objective piece is to know in real time the patient’s taking it as they’re living their life administering medication rather than sporadically relying on self report.

Ryan Lawler: On the flip side, if they’re not compliant, you can send them a push notification or something?

Melissa Manice: Absolutely. That’s a key piece, is really in the same way that I don’t remember to make a meeting if it’s not in my calendar in my phone, we’re really trying to drive sustain longitudinal behavior change by reminding, tracking, engaging in real time.

Ryan Lawler: What sort of data do you have to show that this works in terms of improved health outcomes?

Melissa Manice: Yeah. Maybe it’s because of my background coming out of clinical medicine, but we have really evolved the BreatheSmart platform being sort of grounded in showing outcomes and so some of the types of outcomes that we’ve been able to demonstrate are about a 45% lift on average in daily medication compliance. What that correlates to is a significant reduction in hospitalization, so we’ve done that in both clinical trials, we’ve done that in what we would call our real world evidence base, which is just our patient users living their daily lives.

And then another key piece is that we’re tracking their decline in their emergency medication, so what we call rescue use and what we’re seeing is about a 95% drop in rescue use for patients that are using the BreatheSmart platform. Those are the kind of types of outcomes data that we’ve demonstrated in a variety of clinical and commercial programs and then another key piece, a portion of both our current evidence base and as we are growing is to connect that to cost data, so really trying to connect to cost containment data.

Ryan Lawler: You know, when we determine what works over time in medicine, a lot of it’s done through a certain amount of limited clinical trials over time, but when you’re collecting data on an ongoing basis for a number of patients over an extended period of time, then can you abstract out that anonymized data and create better health outcomes over time using it than you might through sort of that limited clinical trial?

Melissa Manice: One hundred percent. One of the really exciting applications of Cohero’s technology is just that, it’s this idea of sort of a real world clinical trial, the notion that instead of having the current clinical trial environment where like you said, it’s sort of sporadic pieces of data that it’s much more powerful to have continuous longitudinal data that’s captured remotely in a patient’s home. What we’ve found empowering that type of … that sort of new model of research is that it also improves elements like patient retention, because patient satisfaction on a clinical trial is much easier to stay on a clinical trial, for example, when you’re doing it in the home rather than having to go every four, eight weeks to a site for data collection.

But I think that that’s exactly it is that it can absolutely advance innovation, new drugs coming to market, etc.

Ryan Lawler: Okay. What are the biggest challenges to adoption?

Melissa Manice: I would say in terms of a big challenge for us to adoption, certainly reimbursement is critical. We like to think that we’re in this adolescent years of sort of a shift from a fee for service model to a changing, at least in the US, changing care delivery model towards a value based care model, and so I think that can be challenging when you’re between payment structures, if you will.

Ryan Lawler: Right. When you talk about being in this space that is so highly regulated, it’s gotta be scary just in terms of whether or not something will work and then going through all of the clinical trials and everything that you need to do to be compliant and then have to make sure that you’re HIPAA compliant, so it just there seems to be all these additional layers of challenges that may be a consumer facing mobile app would have to go through.

Melissa Manice: Yeah. It’s not just about being passionate when you’re trying to innovate in health care, but it’s really trying to understand that at the end of the day the regulatory requirements are there for a reason but they can be very challenging when you’re trying to grow and evolve and you’re doing all this on often very limited capital. I do agree that I think some of the challenges to growing, scaling, getting adoption, etc, are so, so different than maybe a consumer tech company, but certainly is sort of a fun chicken, egg game often for that reason. Because at the end of the day when you’re raising your first round, you need to show revenue traction. Well, you can’t show revenue traction unless you have regulatory clearance and you can’t get regulatory … so it is certainly a fun challenge in the health care ecosystem.

Ryan Lawler: So let’s talk about you. What were you doing before Cohero? How did you get into the tech world?

Melissa Manice: Oh man. My PhD is in pulmonary medicine and bioinformatics and so I am certainly a first time entrepreneur, founded Cohero because I felt that instead of spending a lifetime studying a well documented and costly problem in circles, that I felt that there was a need to build and scale a meaningful solution, and so I, I guess, made the leap from academic medicine to founding Cohero.

Ryan Lawler: Well I’m kind of curious how you go from academics to entrepreneurship and how you meet your co-founders and build a team and how you’ve gone across that entrepreneurial journey.

Melissa Manice: Yeah, I think in many ways … gosh. That entrepreneur journey. I think in a way I was quite lucky. I did my PhD at Mount Sinai and was also working full time at Sinai as I was finishing my PhD and I think had some phenomenal colleagues and collaborators at Sinai who sort of urged me to try and build an early MVP and then I was lucky enough to then apply for a fellowship to basically then test this kind of what you’d call version … I guess I’d call version ugly point O, but certainly V.0 of this sort of BreatheSmart platform. Then I came back to Sinai to test and validate it through a fellowship.

In many ways, I think then trying to build a team, it’s challenging when you’re bootstrapping because of course you’re trying to raise your first round of capital to build a team and so you have … I was fortunate enough to have a co-founder who was willing to make the leap independent of us having raised our seed route. It’s been an exciting ride, both in the building a team but certainly when we were in our early days when it was sort of myself and my co-founder, Dan.

Ryan Lawler: Was there an aha moment where you said, “This is something that’s missing in this space,” or, “We could treat these conditions better if we had more data and these are ways in which we could collect that data”? What was the trigger point for you to determine that this is something that needed to come to market?

Melissa Manice: Yeah, I think it was a combination of academic and professional experiences, both in a clinical encounter setting from the pain point shared by colleagues, etc. one of the aspects that I would say that I sort of urge those who come out of seeing patients each and every day is that they really understand the pain points quite well. So I think that sometimes whether it’s a physician or a researcher or what have you or a patient, that sometimes those can be the best people to innovate, because they’re sort of living those aha moments every day. Certainly I would say in my case it was grounded in the pulmonary pain points that I had experienced in my training.

Ryan Lawler: So if you weren’t doing this and I think that this is a big question for you, because I feel like you are … what you’re doing now comes so much from your clinical experience.

Melissa Manice: Yeah.

Ryan Lawler: But if you weren’t solving this problem, is there something else that you would be working on?

Melissa Manice: Yeah. I think … you know, in my early days I would have said, “Well, I really only have domain expertise in this one little spot, and so me solving for this is really probably the best place for me,” and so maybe that’s still true a little bit, but if you were to ask if I weren’t doing Cohero each and every day, I would say this whole journey has taught me just how much I love being a problem solver. So really mentoring and jumping in to a company that’s in its infancy at what I call napkin stage is where I love to be. That to me is what fuels my fire. I love that no two days are the same and it’s scrappy and requires a lot of persistence and passion so whether another venture would be joining on an existing team or starting another company, gosh, I don’t know, but certainly love being a problem solver.

Ryan Lawler: Right. How will the future be different if Cohero becomes ubiquitous?

Melissa Manice: Yeah. Gosh, i like think back to our mission statement on our wall of our office that we wrote in our early days that has rung true, and that at the end of the day is really to improve patient health for those impacted by a pulmonary condition. I think how will the future be different, for us, it’s earlier disease detection, better outcomes, better response to treatment, care coordination is improved and that’s from a provider to patient, that’s from care giver to patient. That’s the sort of future that we’d like to see.

Ryan Lawler: So what’s one controversial opinion you have that’s very strongly held?

Melissa Manice: I think sometimes we, in digital health, for example, I’ll stick with that, think so much about the product offering and the solution set and sometimes when I’m mentoring or speaking with colleagues and reflecting on the challenges to adoption, the issue of payment structure, who pays, it comes back to that. I don’t know how controversial that is, but I think that leading with understanding that as a driver to adoption is critical.

Ryan Lawler: Is that something that you had in mind when you set out on this venture, or is it something that you had to learn the hard way?

Melissa Manice: No, and I guess that’s probably why it’s controversial, because I really would rather try and spend a lot of time building, being strategic about, thinking about the elements of a product offering that are sort of meaningful to our end users, but instead sometimes that’s not the reality of day to day growing a business, so I think when I set out, I didn’t think just how important that is, but at the end of the day, that is key to your survival is understanding how to get adoption by getting revenue, and so sort of focusing on a business model in a complex reimbursement world is certainly important.

Ryan Lawler: Okay. Kind of curious because I feel like I’ve been thinking about this in the context of the US insurance system.

Melissa Manice: Right.

Ryan Lawler: And the health care system here. Is it easier or harder in other markets? Are you looking at other markets or available in other markets? And how do you think about international availability?

Melissa Manice: Our go-to market was to start in the US and then expand OUS, but I would say in single payer systems, adoption can be harder, let’s just say, to get buy in from the NHS, in the UK for example, but it is single payer, and the US is so fragmented system by system, state by state, that I do think in many ways OUS can be easier but I think also health care again, coming back to the questions you asked around data security, PHI, HIPAA compliance, the regs are completely different in Europe with GDPR, so sometimes I think that also early on as a company you have to sort of take a thoughtful approach to your go-to market and not be reactionary to random inbound opportunities if you will, and so we have systematically tried to go regionally and then nationally in the states and then OUS and part of that is also because of the regulatory data security and other requirements, but also business model being pretty core to that that are quite different outside the American health care environment.

Ryan Lawler: So I’m kind of curious, do you see any future trends that you think will be impactful in a health care or your problem area, let’s say?

Melissa Manice: Gosh, yes. I would say data science, innovation, AI, machine learning, or something that are absolutely not just getting traction but are really changing, I think changing health care for the better. The key piece is really recognizing that at the end of the day, we need to ensure that medicine stays human so that providers don’t feel disintermediated by all these new, modes of enhanced data analytics. Certainly I think they are taking hold. We have to make sure they’re empowering providers and making their clinical decision making care delivery lives easier, but that there’s this human side sort of stays rather than feeling like they’re being replaced by machines, if you will.

Ryan Lawler: Okay. Well let’s leave it there. Melissa Manice, thank you for being on What’s NEXT.

Melissa Manice: Thanks so much for having me, Ryan.

Related Stories