Intelligent Scheduling

The Incrementalist Graphic Jennifer Meller

This week I am talking to Jennifer Meller, MD (@drjen_Navimize), CEO & Founder, Navimize (@navimize )that is focusing on solving the problem of dissatisfaction for patients stuck in waiting rooms around the country.

The problem sounds simple but in healthcare as usual the devil is in the detail. Part of the key ingredients Jennifer found to building a company included the additional education and insights from going back to school for her MBA at Wharton and surrounding herself with a strong team to fill the gaps in knowledge.

Her experience as a primary care physician treating patients informed the concept, especially as patients started rating their doctors across a range of metrics. She would receive a very positive reviews for her care but low ratings for waiting times, something that was difficult to control or manage with all the competing priorities and her desire to deliver the best possible care to each and every patient that came into her office. She always found herself running behind on her appointments much to the frustration of everyone, including herself.

Together with her co-founder they came up with a communications solutions, triggered by her experience waiting for an Uber ride, to start to address the scheduling challenges in doctors offices. Focusing on wait times and building out an automated communication platform that could send out messages to all the stake holders, doctors, clinical staff, support staff and of course the patient, to keep everyone updated on the schedule and any potential delays.

That feature alone proved to be a real crowd pleaser but they went further starting to predict appointment length based on understanding of the individual doctor, the practice, geography, patients condition and beyond. Using Machine Learning and Artificial Intelligence they were able to process data ingested from a variety of sources and multiple EMR’s to build a better schedule for everyone

Listen in to hear how easy the solution is for clinicians to implement and for staff to use and how they are adding features that include automated follow up details, including getting more positive reviews from the many happy patients who had great experiences but rarely provide feedback. And it’s not just easy to implement, it comes with plenty of customization features to account for differences in specialties, practices, patients and offices.

 


Listen live at 4:00 AM, 12:00 Noon or 8:00 PM ET, Monday through Friday for the next week at HealthcareNOW Radio. After that, you can listen on demand (See podcast information below.) Join the conversation on Twitter at #TheIncrementalist.


Listen along on HealthcareNowRadio or on SoundCloud

Raw Transcript

Nick van Terheyden
Today I’m delighted to be joined by Dr. Jennifer Meller. She is the founder and CEO of Navimize. Jennifer, thanks for joining me today.

Jennifer Meller
My pleasure. Great to be here.

Nick van Terheyden
So, tell us a little bit about your background and how you ended up at this position. And what got you to this point?

Jennifer Meller
Sure. So I’m a physician, I’m a primary care doctor based in New York. I lived in New York, my whole life, went to NYU med school, did my residency there was practicing first and faculty practice at the Columbia then moved to establish my own private practice in Manhattan. And it was about 15 years into my career, where I started to see the whole landscape of healthcare changing electronic medical records arrived on the scene, I started to see technology, taking the real stronghold in medicine, and started to see the practice of medicine really shift in a lot of different ways. And I won’t bore everybody with all the personal details, but mid career decided to go back to school. I earned my MBA at Wharton. And while I was there, I had the idea for the company for now, the mice.

Nick van Terheyden
Fantastic. So lots of clinicians go through these sort of experiences. Do you think that opportunity to go back and study and focus on an MBA is critical to this? Or do you think that that was just a supporting act?

Jennifer Meller
Yeah, it’s a really great question. I’ve seen some doctors successfully do things like this without the MBA and enter into different careers and opportunities. But, you know, the MBA really gave me a solid position from where to launch this business, I learned so many things that I wouldn’t have otherwise. And I’ve also watched a lot of my colleagues attempt to start companies thinking that, you know, this will be easy, I understand medicine. But without understanding the business perspective, without having the background of, you know, being taught about how to innovate and how to test before you build and those sorts of things, understanding how to target customers, those are things that are not in the lexicon of any physician when they walk out of training or when they’ve been practicing.

Nick van Terheyden
Right, great points. There’s a part of me that says that maybe it should be part of the medical school curriculum. And I know there’s, there’s 1000s of medical students going no, no, we don’t want to do any more types. But it really does feel that way. Because, you know, business is part of healthcare. And it’s not something that we learned in medical school as a sort of native thing, but you have to learn it as you go along. Right. So I think great that you did that. Tell us a little bit about Navin Mize. And what the genesis of that idea and what the company does. Sure, sure, I’ll

Jennifer Meller
be happy to. So I’m in my practice, I know it will be shocking for many who have ever been to a doctor but I always ran behind schedule and and was late to my next patients. And my patients sat in that waiting room and waited. And it wasn’t because I didn’t care or didn’t want to be on time. There were just so many things happening in the background. When you’re practicing and seeing patient after patient after patient that are out of your control. Sometimes people come in and it takes a lot longer to attend to the things that they’ve come in for sometimes people come in and there’s an emergency, sometimes you get a call from radiology that there’s an abnormal finding on a scan your order two weeks ago. So there’s all this it’s going on in the background. And I felt like I had no control over this. And yet it would make my day miserable day in a day out it would make the patients unhappy, my staff unhappy and make me unhappy. And but really never had a way to fix this problem. And and then the problem got worse when we moved into the era of online reviews. And suddenly my patients were going online and saying Hey, she’s great. I love her five stars bedside manner, one star per wait times and I thought, Oh no, this is not good. And then Urgent Care came onto the scene and my patients instead of coming to my office when they had a fever, sore throat and calling up to be seen. They opted instead for a minute clinics because they could be in and out very quickly. So when I was at Wharton, we were learning about Lean management principles. And it was in the context of that framework that I started to think about waiting in a waiting room as a form of waste. And then one day I walked outside, ordered an Uber, looked at my phone and said, If I can see what’s happening with this car, on my phone, even though it’s not physically in front of me, my patients should be able to virtually peer into my waiting room and know what’s happening before they head to my office and avoid the wedding. And that’s how the idea was born.

Nick van Terheyden
Oh, that’s fantastic. I think, you know, there’s a lot of common reference to Uber rising, I think this is, you know, the, this will resonate with lots of people. And, you know, you talk about a lot of things that go on in healthcare, all of these competing priorities, you know, the seven minutes that’s on average allocated, but of course, it never turns out to be on average. And, you know, if you’re the patient in front of the physician, you want the appropriate amount of time, whether that’s five minutes, or 30. And, of course, that, you know, is very challenging. So it sounds like, you know, this is a really complicated problem to solve, not just by, you know, you can float the information up, and I’m sure that improves things, but will tell us a little bit about the solution and what you started to develop.

Jennifer Meller
Absolutely. So, um, what we do today is we’ve developed a patient communication platform. And the, the uniqueness of our platform is that ability, just as you said, to be able to predict, wait times and push that information out to patients in real time, so that if you have a 2pm appointment, we’re getting you the message at 115 or 130, and then maybe an update at 145, letting you know how things are running in the office, so that you get it straight to your phone, you know what’s happening, you know, when you should be coming in, so that you don’t have to wait. So that’s that’s what we’ve been able to accomplish today, it was not easy, because as you say things are complicated. On the back end, it looks very simple on the front end to the patient. But on the back end, it’s actually quite complex. Since different offices operate differently, different specialties operate differently. And we had to figure out how to gather the information directly from the EMR to electronic medical record to flow into our system so that we could then predict the delay is the kind of vision that we have. And what we’re working on building now is taking that even a step further or several steps further. And what I mean by that is we are starting to look at what drives appointment length, and pulling data to understand how long given appointment is taking. So that the time that a patient goes online or makes that call to schedule their appointment, we can predict how long they’re they will need to see the doctor for the reason they’re coming in for so that we can schedule that five minute visit for five minutes, and that 20 minutes visit for 20 minutes. And we believe that that will have the input impact of really driving efficiencies in the healthcare system, improving throughput for health systems reducing wait times from the bottom on and also improving the experience for everyone.

Nick van Terheyden
Yeah, so that’s really interesting. I mean, not least of all the concept of saying, Hey, I’m going to communicate ahead of time. And you know, gosh, wouldn’t it be better to be waiting somewhere else? Actually, wouldn’t it be better if you didn’t wait at all? It feels a lot like a sort of restaurant waiting process that says, hey, we’ve got your table, and it’s coming in 20 minutes. And we know that you’re 15 minutes away. So we’ll give you a, you know, that amount of time? Is that the sort of foundational pieces of that? Are you getting enough data to be able to sort of attribute that and start to deliver that kind of information to patients? And indeed, the office staff?

Jennifer Meller
Yes, yes, that’s exactly what we do today is we are able to, by pulling certain information from the practice at the time of onboarding about their day to day workflow, and also their preferences, how they like to practice and then also combining that with the daily schedule. So we pull both information that we take from the practice at the time of onboarding, which is specific to their workflow and also information about their preferences, and combine that with data that we get directly from the electronic medical record, including scheduling information, as well as real time tracking information. And all of that is crunched by our algorithms to predict delays and then automates messaging to patients to let them know how the office is running in real time.

Nick van Terheyden
So this isn’t, you know, a manual process for the staff to say, Oh, my goodness, we’re delayed, although I imagine there’s some opportunity for that this is fully automated a bit like the sort of driver heads capability.

Jennifer Meller
Correct. It’s 100% automated, we that’s really cool. Yeah, it’s very cool. We designed it intentionally that way. Because, you know, I know how busy firsthand, I know how busy a doctor’s offices and I know how busy my staff are, and that they don’t have time to stop. And you know, they’re already in the electronic medical record entering so much data they don’t have, there’s there’s no way to add more to what they’re already doing. So that the whole notion of what we designed was with the idea that we would take work off their plate as opposed to adding work to what they were doing.

Nick van Terheyden
Right. So tell us a little bit about the integration because to me, it sounds like data is king in all of this and having as much as possible to be able to process how are you managing to achieve all of this?

Jennifer Meller
Yeah, so we’re working with a variety of electronic medical records and partnering with them to integrate our software into their platform. And through those integrations, we’re able to pull all the data that we need to power our algorithms on the back end. The nice thing about our platform is we don’t have to really write anything back to the electronic medical record. So they don’t have to, there’s not a lot of work from from there. And that needs to be done. They really need to open up their API’s to us, or in certain EMR, in certain cases, with different electronic medical records. It’s done through direct HL seven messaging. But this is all the technical stuff is it’s it’s through those integrations and partnerships with electronic medical records that were able to pull that data.

Nick van Terheyden
So for those of you just joining, I’m Dr. Nick, the incrementalist and today I’m talking to Dr. Jennifer melon, she is the founder and CEO of novomind, we were just talking about the integration and the whole process and the complexity of a doctor’s office that you know, has so many moving parts, all of these elements that contribute to an appointment that, you know, for an individual seems very simple. I said, I was coming in at 230. And I here I am at 245. And I’m still waiting. You talked a little bit about the integration, you know, the need to get that data, I think, you know, helpful that you don’t have to push back because that always creates some problems. We’ve seen some improvements in interoperability with healthcare data, and you know, the new fire interface that I’m sure will start to contribute and make things easier. But how are you processing all of this? What’s going on in the background that sort of provides that insight to say, Hey, I think things are pushing out and also learning by the sounds of it, because you’re doing some predictive analytics.

Jennifer Meller
Yeah, so that’s a little bit goes into our secret sauce. But we’re suffice it to say that we have we spent a lot of time in the early days, really trying to identify what are the key data points that we need to pull? You know, what are the static data points? What are the things about the practice that are consistent over time that our algorithms need to know in order to figure this out? And then what is the what are the dynamic data points that we need? What what do we need to know about what’s happening in the office today? And right now? And how quickly do we need that information? And then how do we pull that all together to make these predictions? And that is what we spent, you know, several years working on and building to get to where we are today.

Nick van Terheyden
So as part of your studies, did you become a data engineer for one to another term? Is that somewhere that you’ve focused in on? Or are you partnering with others that are doing this?

Jennifer Meller
Yeah, definitely partnering, I am definitely not a data scientist. I think that takes a lot of a lot of other training. And after medical school and business school, I would say I’m schooled out. But I have a great partner who I went to actually attended Wharton with her background compete among golf, her background is in tech. And so she’s done large scale technology and implementation started her career as an engineer. And together, we’ve hired other people to help us build this out.

Nick van Terheyden
So again, it comes back to that sort of element of what it takes to be successful, not just with, you know, nevermind, but with other companies. Do you think that partnership is a critical element? You know, that obviously happened as a result of, you know, attending and participating in the same coursework? Is that one of the key pieces of contribution to allow you to launch and come out with something successfully? Do you think?

Jennifer Meller
Yeah, I mean, I think you need a strong team, to to succeed. You know, nobody has there’s no one person that has every talent or every bit of experience needed to succeed as a as a startup. And especially in digital health, I think you do need someone who understands healthcare from the inside out, not just looking at as an outsider that’s enormously valuable. Having somebody who understands technology, who understands data, who understands analytics, that’s another piece, that’s essential. And then we’ve added other team members, so we’ve added a head of bizdev. To our team, the one thing you know, the one the one thing that neither myself nor Kavita had or came into this having was sales experience. You know, we neither one of us had ever gone out there knocking on doors Trying to sell anything. So we really needed we started off doing some of that on our own, and got the early sales on our own, but quickly realized that we would need someone to really kind of take this to the next level.

Nick van Terheyden
Interesting, I think, you know, recurring theme is, you know, the combination of skills that contributes, and also what I always call the adjacent possible the learnings from other places that we can apply, especially in healthcare, which, you know, sometimes feels very backward in terms of some of the approaches. So, you’ve created this, you have a software solution, you’ve got some of the integration, tell us a little bit about the experiences and talk us through the process, you know, from the perspective of perhaps a clinician, you know, back office staff patient.

Jennifer Meller
Sure. So on the patient side, it’s really very simple, they receive a text message. And that text message will let them know how the office is running. On the physician and office side, again, the system is fully automated. So once it’s implemented, they really can go about their daily business and not think about NABBA Mize. What, what we do when we go into these practices, in addition to gathering the data we need to power our algorithms is we’ve also built the platform out and expanded it to allow for just a whole host of customizable messaging, all of which is automated. So oftentimes, you know, my colleagues or the staff would say, Oh, I wish when I click this button that’s scheduled this patient for this procedure, they would just automatically get instructions sent to them, we can do that. Because we can set that up to power the system to send out that message just at the right moment. Or, as soon as a patient is leaving an office, you know, everybody, everybody worries about online reviews these days. And what ends up happening is that the patients are happy with you just go home and the patients who the one in a, you know, 1000, who have a poor experience, for one reason or another, we’ll take the time to go online and leave a review. So we send push, we can push out messages at the time that a patient’s leaving the appointment. So here, we gave them a great experience, they didn’t wait in the waiting room, they were in and out. And then they get a quick text saying, Hey, can you leave a Google review, and, and tons and tons of people will just hit that five stars. So we really are able to help doctors improve their online reviews. So these are all the types of things that we’re able to do for practices. Patient Education is another example. Say a doctor sees a patient, they’re like, Oh, I wish when they left the appointment, I could just send out this information so that they, they learn about healthy diet and lifestyle, every time someone comes for an annual visit to their primary care doctor, well, we can set that up for you. So all of this very robust messaging is what we’ve built in to really make that experience for the doctors and the staff better without them having to think about it or stop their usual workflow.

Nick van Terheyden
So how customizable is that by practice by patient? You know, other different triggers is obviously the manual element that you talked about, you know, for the the physician who says, gosh, I’d really like to do this with this patient. And you know, you enable that, but what about some of the automation is, is that customizable, so that people can get different levels of information, different details? How does that work?

Jennifer Meller
Yeah, that’s all it’s all automated. So so the customizability can actually be down to you, the practice can choose. So it can be in a large organization, they may want to have all of the orthopedist same send out the same messaging, or all their pediatricians set up the same messaging. But the way the system is built, is that for each schedule, or each doctor, they can customize the messaging according to, you know, just however they like so. And even if you have a doctor who says, say, has an office in an urban location, and then an office in a suburban location, they can tweak the messaging for those different locations. And so everything is set up in advance, and then they decide what messages they want to go out and at what times we program everything. And then it just flows once it’s turned out.

Nick van Terheyden
Fantastic. So as you think about the future. Tell us a little bit about where you see this going. I mean, there’s a part of me that says that we might be we might see the end of the waiting room. Is that a possibility? There is no more waiting room I just walk in and I’m there at the exactly the right assuming I got there by traffic and Google got all that right as well.

Jennifer Meller
Yeah, that’s what we aim for. We aim for the patient to be able to walk in the door, you know, register at the desk and immediately be whisked to the Back to be for the exam to begin, and then to be, you know, seen on the way out and to go home and for everything else to be communicated directly to them through this to kind of text messaging, that’s, that’s elegant and set up in advance. That’s that’s what we aim for. And I think with Intel, what we call intelligent scheduling, which is the vision I was talking about earlier, where we can actually predict how long a patient will need for their appointment with this doctor when they’re coming in for this problem, I think that will really drive us towards that reality.

Nick van Terheyden
So part of me thinks that might even be linked to a sort of payment model. I mean, right now, we have this sort of very broad brush, you know, attitude, and no real sort of understanding of the complexity that it takes to deal with some patients relative to others, even though they may have the same condition, you know, well managed diabetes versus poorly managed diabetes, two entirely different things, maybe you’re sort of central to the future that stops penalizing doctors and, you know, allows us to get paid appropriately.

Jennifer Meller
Yeah, it’s very interesting, we’ve thought about that, whether there’s some potential down the line for us to segment these patients and look at, you know, look a little deeper, so we can predict how long this patient is going to need, but then look at the patient’s sort of from a broad perspective and say, Okay, why do these, you know, 500 patients need 20 minutes of their doctor’s time, whereas these other 500 patients only need five minutes? What are the differentiators? Is it an ICD code? Is it their age? Is it something else? And yes, that is something that we we are looking towards exploring down the

Nick van Terheyden
line? Yeah, I think, you know, there’s probably not a physician around that wouldn’t be excited to be just actually doing the work and having appropriate payment for actual activity. And you know, the complexity of care versus sort of this broad brush Level Three versus level two kind of activity, which just, I think frustrates a lot of people. So in the closing minute or so, do you see this going into hospitals and other places? Because it sounds like this is predominantly in practices? Is this even broader?

Jennifer Meller
Yeah, I think it potentially is border applications. So right now we’re in outpatient practices, so ambulatory care, but even within health systems, there are ambulatory care clinics and settings that we are penetrating. I think there’s potential to adapt this for, for the on the inpatient side as well certainly, you know, in places like radiology or infusion centers, or, you know, endoscopy units, there’s lots of potential and then there’s ultimately potential within the ER, although I think the ER is a little bit of a different animal and and a tough nut to crack. But I do think there’s, there’s opportunity within the hospital setting. So we’re eight, we’re starting on the outpatient side, but I believe there are at least segments of the inpatient side where we can we can move to next,

Nick van Terheyden
I’m just gonna say I admire your intent around the Emergency Medicine Department, which, in terms of complexity and prediction, if if you can manage that, I think that would be a fantastic Goldberg. Significantly more challenging, but plenty of opportunity, exciting space to be in, I think, you know, when you think about satisfaction from patients, you know, you captured that really eloquently, you know, five stars for clinical care, one staff a wait time, which, you know, is that really fair in terms of the context that you described? Unfortunately, as usual, we’ve run out of time, so it just remains for me to thank you for joining me on the show. Jennifer, thanks for joining me.

Jennifer Meller
Thank you so much for having me today. I really appreciate it and it was a lot of fun.


Tagged as , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , ,





Search
%d bloggers like this: