Editor’s note: Francoise A. Marvel, MD, is an assistant professor of medicine within the Division of Cardiology at Johns Hopkins Hospital, codirector of the Johns Hopkins Digital Health Innovation Lab, and the chief executive officer (CEO) and cofounder of Corrie Health. We spoke with her about the emergence of application-based tools used for healthcare and the patient safety issues surrounding the use of such tools.

Francoise Marvel (FM): Sure; thank you for having me for this conversation. My name is Francoise Marvel. I am an assistant professor in cardiology at Johns Hopkins Hospital. I am core faculty of the Johns Hopkins Ciccarone Center for the Prevention of Cardiovascular Disease. I am also codirector of the Digital Health Innovation Lab at Johns Hopkins, along with Dr. Seth Martin. Outside of my position at Johns Hopkins, I am a cofounder and CEO of Corrie Health, which is a company cofounded with Dr. Seth Martin and Dr. Mathias Lee, with a mission of improving patient and clinician engagement and experience, and delivering guideline-directed medical management using health technology. In my roles at Johns Hopkins and with Corrie, I champion for preventive cardiology and health equity with a focus on healthcare transformation with digital health.

SM: That’s great; thank you. How did you become interested in digital health and app-based tools for healthcare?

FM: For me, it’s a very personal journey. I was always interested in how digital health and technology tools can be a part of our everyday life as clinicians caring for patients and to help patients be more empowered and engaged in their care. I was a medical student when I started exploring the potential of smartphone apps. I wondered, could these apps actually benefit us while we’re on rounds with patients? And how, instead of scribbling little notes on notepads in hard-to-read handwriting, can we really get this translated into a digital and useful experience? That was my first adventure into creating a smartphone application in medical school. The app I created helped translate clinical guidance and tips or “clinical pearls” from some of the top experts in internal medicine on how to best diagnose and manage patients.

The turning point for me and digital health was during my intern year at Johns Hopkins. At that time, my father was diagnosed with cardiovascular disease. It was a very challenging time. I was at the bedside with my father experiencing what patients go through every day. I realized that there was a major gap in how we care for our patients. There was a lost opportunity to engage with our patients more on what’s going on with their health condition and to do it in a way that’s simple, helpful, and inspiring—for patients to be part of their journey and not just part of a passive process that’s occurring to them—but that they are driving that process forward. That was the real turning point for me. Unfortunately, I lost my father during my intern year of medical training. From that point on, I really dedicated myself to creating something actionable—to have digital health make a positive impact on patients and their families. In a way where patients can see what they really need to be doing, digital technology can empower them with understanding how their health that could be better and how to be successful. Digital health apps are not just for those people who have devices or have internet access but for people who don’t have those things; it is for people who might fall through the cracks or feel like they might not know how to get started. I carried that passion through my training and after I started here at Johns Hopkins. I started working in digital health with a focus on cardiovascular patients but also, more broadly, with health equity in mind and how we can leverage digital technology to reach more patients in a safe way. In fact, on the point of safety and digital health, I did an Armstrong Institute Patient Safety and Quality Improvement Fellowship that really focused on how to create these tools in a way that champions patient safety. That’s part of my start with Corrie Health. I also did a Johns Hopkins Technology Innovation Center program called a HEXCITE Fellowship1 that focused on how to make these tools in a way that’s scalable and to go through the regulatory and design process successfully.

SM: Yeah, that’s great. You have started to touch on some of the things that I’d like to explore a little bit more in the interview. I wonder though, if maybe we back up for just a second, and start with a really basic question for some of our less tech-savvy readers. How would you define an app? Is it something that is just on your phone? Is it something that can be on your computer? How do accessories and devices come into play or fit in the picture?

FM: This is fundamental to our conversation, and honestly, I want to let readers know that I want to keep our conversation simple with definitions and terminology. You want to think about an app as being a software program. These software programs can run on just about any device or computer. If that device happens to be a smartphone like iPhone or Android, you can call that software program a mobile application or a mobile app. You might also have an iPad or some other type of tablet that can similarly run these apps. Many of us actually have televisions that can run these apps as well, so you can use an app, like Amazon Prime, on your television to watch a movie or TV show. So more and more, we are interacting with these apps on our phones or TVs that are part of everyday life. Apps, in turn, do things that can make your life a little simpler. Smart watch devices, whether it’s a Fitbit, an Apple Watch, or a Garmin, are like mini computers on your wrist. They also can run apps that, for example, send you reminders, check your daily step count, or check your heart electrical activity (called an electrocardiogram or ECG)—that’s actually a small application that’s running on your smartwatch.

SM: That is really helpful. I wonder if you could give us some examples of apps used in healthcare and what they’re used for, whether it’s to direct care or collect information that’s used later.

FM: Yeah, that’s a great question, Sarah. I think one of the parts I always focus on is who’s the audience? Apps are made for different types of users. For example, if you’re a patient who has diabetes, and part of your everyday life is to monitor your blood sugar levels, you might benefit from an app called One Drop2 that connects with your blood glucose monitor. It serves as sort of a central command on your phone, where you can get a sense of where your daily blood sugar is trending. That’s an example of a patient-facing app, meaning the user is the patient. There are also applications that provide information to the patient. We can provide patients with recommended apps that help them learn more about their health condition, such as high blood pressure or high cholesterol, or to learn more about that recovery experience after they suffered a heart attack. These are opportunities where patients receive education from the applications. Another example is an iPhone, which has a health app already installed. This app helps you to track your activity and tracks your step count. The Apple Health app3 can also track labs or amount of sleep. That’s just another useful example of a patient-facing application.

Now, switch to someone like myself, such as a cardiologist or a doctor. We might be using different apps for different things. One way doctors can use an app is as a clinical reference, for example, Epocrates4 or other apps on your phone that can be used to look up a medication. You can check to see if there’s any problems with compatibility with different medications; that’s a useful reference guide. Apps can be used for clinical and diagnostic assistance. For example, if we’re going to start a patient on a blood thinner medication to prevent stroke and want to calculate their CHA₂DS₂-VASc score,5 an app can perform that calculation. These are examples of apps doctors can use to help make clinical decisions. There are other apps for remote monitoring or even diagnosing. Those are just a few examples of patient-facing or clinician-facing apps, but there are a lot of them out there, and I encourage app users and readers to explore those options.

SM: Those are great examples, thank you. Now that we’ve talked about what apps are and how they’re used in healthcare by patients and healthcare providers, I’m wondering what your overall perspective is on the current state of app-based healthcare tools.

FM: I’m an enthusiast of digital health and applications. I’m quite a believer that this technology is transformative in terms of the patient journey and the way that we interact and engage with our patients. I think it’s really poised to transform the way we provide medical care, especially when it comes to guideline-based recommendations for care. So, I think now is a good time for mobile applications, and as a healthcare system, we are ready to make more of a formal translation of what we already know is working to the bedside. For example, evidence-based clinical applications are becoming increasingly mainstream with everyday use by patients and clinicians. I think that’s something that is in the near future because over the last several years, we’ve been challenged with the increased demand for more virtual care settings. We’ve accepted that challenge, and I think we’ve done quite well in responding to relying on virtual care models when traditional center- or hospital-based services were limited. Now is the time to really take that momentum we’ve created over the last 2 years and make this digital health transformation more official and sustainable. Whether you’re in cardiology and using applications and devices to help diagnose atrial fibrillation or prevent stroke, or you’re using applications to assist with diagnosing early signs of heart failure and keeping patients safe and out of the hospital, or improving cholesterol control to reduce the risk of heart attack or stroke, or you’re using an app to help keep patients educated and engaged with chronic cardiovascular conditions—these solutions are all available through digital health. However, we still need the infrastructure to create sustainability to make this part of our everyday practice. Right now, use of these types of digital health solutions is limited.

SM: I agree that there is a lot of potential with digital health. I wonder if we could shift the conversation to thinking about safety implications. The apps are designed to improve care, for patients often through self-management. I don’t think patient safety is part of the design process. Are you aware of any safety issues that have surfaced especially as use has become more widespread?

FM: Absolutely. I think this issue is critical because, ultimately, we want to have only the safest and the most user-friendly technology make it into the patient’s hands. So, I think there are at least four main issues when we are considering safety and implementing health apps.

The first issue is operability, or the ease with which a person can effectively use an app. A digital health app is different from a medication. A medication has a defined mechanism of action. It is typically swallowed or injected, and simplistically, that’s the end of the therapy delivery. With an application, we have to install it, load it, run it, update it, and provide a reasonable user experience as part of the delivery. Also, many different types of user experiences—such as digital health literacy and health literacy in general—go into operability. I’m emphasizing this point because our team, and here at Johns Hopkins, we’ve devoted extensive time and energy and conducted human-centered sessions with a focus on health equity as a key piece of improving the onboarding and end-user experience across groups. You can create technology at many different levels very quickly, but this aspect of making it easy to use, making it reliable, and ensuring it has a delightful user experience all requires quite some attention from a multidisciplinary and diverse team.

The second issue is privacy. It’s critical that we protect the user’s information and are in full compliance with all applicable HIPAA, confidentiality, and privacy laws. This sensitive patient health information must be safeguarded.

The third issue is security, which is related to privacy. Privacy and security often go hand in hand, but security is more about protection from external threats, like hackers. It involves keeping the backend system of the app secure.

The fourth issue is the content of the app. The content of an app, especially in the realm of healthcare-related apps, has to be peer-reviewed, relevant, updated, and constantly curated and evaluated. If there are new guidelines, new therapies, changes to the clinical management practices, that content must be updated in consultation with experts in the field.

So again, regarding the safety implications of healthcare apps, I think it requires a balanced approach. We’re trying to innovate, but we need to elevate the key app components of operability, privacy, security, and content to make sure that we’re keeping our digital health interventions safe for patients, accurate, and reliable.

SM: That’s great, and I can understand how all those components are very relevant. Regarding the content, I can see how that would be really challenging to keep it updated and current, even if the app content were up to date when it was first built and launched. Fairly quickly, that content could end up with incorrect or incomplete information if it’s not rigorously curated and maintained.

FM: That’s correct. Science changes rapidly and updates come through on a daily basis. For nonclinical teams who are creating these medical apps, that presents a patient safety challenge because that type of nonclinical team is not providing medical care at bedside, not interacting with patients, and not going to professional medical conferences to be educated on the latest clinical updates in patient care. Staying current with the latest clinical information is a requirement for maintaining reliable app content.

SM: Yeah, I absolutely agree with that statement. You touched on this topic a little bit earlier, but what do you think could be some of the consequences to patients when it comes to these kinds of core safety issues?

FM: If you’re not consistently following evidence-based guidelines and you’re providing inaccurate information and education to a patient through an app, the repercussions may lead to patient harm. It’s concerning to know health application development and design can occur by individuals without clinical training and experience to guide product development. I’ll give you an example because I think this is a good one. Drs. Timothy Plante and Seth Martin, both renowned Johns Hopkins physicians, conducted a landmark validation study on a blood pressure smartphone app.6 They evaluated an instant blood pressure app from AuraLabs that was in the top 50 most downloaded apps in the Apple App Store. This particular app came up as a concern because they were using a very nontraditional technique of capturing blood pressure. Basically, the top edge of the smartphone is placed on the left side of the chest while the patient then places their right index finger over the smartphone camera to capture the blood pressure measurement. The validation study conducted by Drs. Plante and Martin found that the app had an extremely low sensitivity for hypertensive measurements, meaning that 77% of individuals with hypertensive blood pressure were falsely reassured that their blood pressure was in a normal range and not hypertensive. From a public health perspective, this shows the importance of clinical validation. Fortunately, Drs. Martin and Plante were able to provide that to protect patient safety.

SM: That’s a great example of a really significant potential safety risk to patients that they probably can’t tease out by themselves when using such an app. Thank you for sharing that. You touched on this already when you talked about lack of expert involvement in app development. I wonder if you have thoughts of what other factors are potentially contributing to patient safety issues with healthcare apps.

FM: I do think that in addition to a lack of physician and healthcare clinician involvement in clinical tools, there’s a gap in evidence generation from clinical studies and research studies. Evidence generation is a major issue contributing to safety issues and health apps. Ultimately, if we were thinking of apps being as powerful as medications, we need to support end-user evaluation and validation. Also, encouraging human-centered design will ensure all of our patients connect with the intervention, so it is not just for a small, selected group of people. So, I think there’s a lot that we can do to promote patient safety in health app design and development: involving clinicians, creating a framework around research or clinical investigation, and using human-centered design before products go to implementation.

SM: That’s great. Who else should be playing a role in maximizing patient safety during app development and use?

FM: I think patients need to feel empowered to explore evidence-based health apps on app marketplaces like the Apple App or Google Play Stores. I encourage and empower my patients to take a critical look at the background of the apps they want to use or the devices they are considering. For example, if a patient is going to download an app, they should consider key questions: Is it supported by scientific evidence or nationally accredited organizations like the American Heart Association and American College of Cardiology? I encourage my patients to bring the app or the device in with them during a visit and show me directly. I am confident about any device that’s FDA [U.S. Food and Drug Administration] approved for the intended purpose. For example, there is clear FDA approval for some features in different smartwatches like Apple Watch for atrial fibrillation detection. I use a similar approach with other clinicians who might be interested in finding out more about health apps or devices.

SM: We’ve spent a good amount of time talking about some of the threats to patient safety. Unfortunately, there are many. I wonder if you can comment on whether you see any improvements to patient safety as a direct result of using some of these apps.

FM: There are plenty of use cases for smartphone applications in the cardiology space demonstrating that secondary prevention measures, like supporting medication adherence and blood pressure control, lead to positive outcomes. Currently, we’re working to generate more evidence to support use of things like virtual cardiac rehabilitation. A primary area of focus for my research over the last several years is to improve the safety around 30-day readmissions for patients who survived a heart attack. When we look at national statistics, about 20% of patients return to the hospital after having a heart attack, and about 75% of those hospital readmissions are preventable.

So, we thought, what would happen if we reengineered the process of being discharged from the hospital after a heart attack with paper-based instructions and we enhance it by providing a digital health experience, including a smartphone app as well as wearables? The digital health experience was designed to help patients engage with clinical guideline-directed management, including taking important cardiac medications, learning to check their vital signs like blood pressure and heart rate, wearing a watch so they can monitor their steps and activity, and wearing a blood pressure cuff. So, we thought, what if we do this for patients who just come into the hospital with a heart attack, and they get this digital health platform that uses app and wearables and we check on them in 30 days and check whether it made any difference in hospital readmissions? We conducted the MiCORE study across four hospitals: Johns Hopkins Hospital, Johns Hopkins Bayview Medical Center, Massachusetts General Hospital, and Reading Health. The MiCORE study showed that patients who used Corrie had a 52% lower risk of going back to the hospital within 30 days of discharge for any health issue than the historical group.7 The findings, published inCirculation: Cardiovascular Quality and Outcomes, also suggested that those using the app were more motivated to improve their quality of life. More than 90% of users said they felt prepared to manage their health at home. This is an example where we reengineered a system using app technology combined with wearables. It made a significant difference in the safety for patients recovering from a heart attack and was cost-effective saving nearly $7,500 per patient using Corrie compared to what the standard care would be without using the app.8

Another highlight about health equity and the MiCORE study is we created a loaner program called “iShare,” which provides health technology for underserved, underrepresented individuals. We shared the story of a middle-aged woman, Tammy, who had several challenges, from being incarcerated to having multiple health issues including heart attacks, hypertension, being a cigarette smoker, and working late nights as a shift manager at Walmart. She had her second heart attack when she was young—in her 50s. She represents a patient population that often gets left behind in traditional medicine. In order to reach individuals like Tammy, extra support must be provided. We created this loaner program because she did not have an iPhone and she did not have an Apple Watch. This program, called iShare, enabled us to share the technology and then recirculate it, just like we do with halter monitors and other things in cardiology. This particular patient, and the patient population from the study, benefitted from this program. It’s been about 6 years now, and she has not had a recurrent heart attack. She quit smoking, she’s working as a manager at both Amazon and Walmart and living her best life. I think this speaks to one of the big potential advantages of digital health—extending the reach of healthcare through technology.

SM: I think that’s a fantastic story of being able to deliver care in a really unique way to a patient who potentially wouldn’t have received that care otherwise. It does make me think about my parents, for example, who between the two of them have one shared cell phone and it’s a flip phone. So “tech savvy” is not at all a term I would use to describe them. I wonder whether you think apps will work for people like them. Do you think there are other ways to support those types of users?

FM: That’s a great question. Speaking of the patient in the iShare program—she had a flip phone too. She didn’t have an iPhone. We also worked with Tom Bauer, who’s the director of education here at Johns Hopkins, around the educational level of the content. We established it at the sixth-grade level and lower so the comprehension level was appropriate for everyone in our community. We try to make things simple so people aren’t overwhelmed from the beginning.

To answer your question, “How do we make it simple to start interacting with the technology?” One, we get industry leaders who are already incredibly successful in that space, like Apple who has world-class leaders in design. Two, we make it accessible by having loaner programs that include everyone. Three, we have many different ways to get patients to start using it. For example, we have a video that gives patients a glimpse at it before we actually have the human interaction. We have a Corrie navigator connect and essentially coach the user through on how to get started. That takes about 20 to 30 minutes. Then, we have weekly coaching calls that we use for both a health check-in and a tech check-in.

So, technology cannot exist on its own, and human interaction complements the technology. For your parents, it would be a conversation to get to know them and what’s important to them. It might be they just need a little extra time. Perhaps they would rather have an in-person session to walk them through it and provide them with a paper handout that shows the features in large print. On that note, we worked with Apple to figure out the optimal contrast of colors and the size of buttons, so that it’s built for visual acuity for all ages.

SM: That’s great that you’re thinking about it from a health literacy standpoint as well as a technology literacy standpoint and bringing both of those together in order to provide something that’s, like you said, personalized to the patient—thinking about each unique individual and how to best meet their needs when using this technology. You mentioned developers, and I wonder if you have any suggestions for nonclinical teams and developers about how to integrate a patient safety focus into app development as they move forward with healthcare-based apps.

FM: For developers of health apps, it all starts with regulatory discussions with the FDA. In our case, even as clinicians and cardiologists, we started with discussions with the FDA to understand if we were under the purview of being a medical device. It ended up that our product, the Corrie app, was considered a medical device for which the FDA intended to exercise enforcement discretion. It’s really important when creating a health app that it’s understood which specific profile it meets—medical device or not a medical device—because each has a different regulatory pathway to keep patients safe. Developers need to be in compliance with the policies for devices, software functions, and mobile medical applications that came out in 2019 per the FDA.9

I think it’s also important to connect with academia because ultimately you need a leader on your team who will provide a pathway toward a clinical evaluation of the app or final product. Whether it’s a quality improvement study or a randomized controlled trial, you need clinical evidence to support the use of your technology and assess safety. This is something that we’ve recently been undertaking with Dr. Nino Isakadze, who’s part of our lab here at Johns Hopkins. She is a future electrophysiology doctor and star here at Johns Hopkins Hospital. She has been leading human-centered design, which is essentially bringing in feedback from a diverse group of end users that helps us to craft and design something from the patient’s and clinician’s perspective, which is a human-centered design.

In every health specialty or general care setting, there are going to be academic bodies like in Cardiology the American Heart Association (AHA), American College of Cardiology (ACC), that guide what we do every day in our practices by setting national guidelines for management of patients. It’s important to engage with these leaders within the clinical topic area of interest for the app that you’re creating for a collaborative, up-to-date, and scalable solution.

SM: I would imagine that would really help you stay up to date in terms of changes in practice.

FM: Right, you’re raising an excellent point around content there. Taking it a step further, both AHA and ACC offer opportunities to take their content and bring it into your application.

SM: That’s great. Well, I know we’re coming up on our time, and I would like to pause and see if there’s anything that we didn’t discuss that you’d like to cover.

FM: Thank you so much for having me and for having this discussion. It’s exciting. I’m so glad that AHRQ is focused on patient safety around digital health tools and smartphone applications and devices. I think this really speaks to where we’re headed. App-based medicine is a critical part of bringing better care to our patients and more accessible care to our patients. There are challenges that I wanted to highlight with sustainable use of digital health: As a community, we need to find more thoughtful ways to provide better access to data and Wi-Fi connections in our communities. Our loaner program, the iShare study that we published, started to work in that space, but it’s something that I think has a lot of potential. Health equity and health literacy are a key part of success around the topic of patient safety because content must be built in a way that connects with every one of our patients. If it does not, then we need to reengineer it, rewrite it, and build it better. Scalability really depends on how we structure reimbursement models and how we can create ways to make this a part of everyday healthcare. I think there’s a lot of opportunity here for policy change and advocacy. Overall, we talked a lot about smartphone apps, devices, and wearables. The big picture here is that all of this information creates a lot of health data that are very valuable from a population standpoint and an individual health insight standpoint. I think the potential for machine learning and artificial intelligence has the potential for major positive impact but will also require close monitoring, safety, and ethical considerations. There are so many great teams and companies out there working on the integrated technology with apps, and machine learning, and analytics. It’s so exciting, but we want to make sure that excitement is balanced by safety, diligence, and health equity.

References