Data Driven Healthcare

 

Welcome to transform now, the podcast brought to you by robotic process automation Pioneer S S and C Blue Prison. Digital transformation has the potential to reshape the way companies service their customers, engage their employees and manage their operations. Whether you’re looking to develop strategies, tactics or best practices to positively impact the future of work, or you’re curious to see how other companies have successfully navigated their digital transformation programs, than this podcast is for you. We’re here to help you transform now. Hello Everyone, I’m your host, Michael Marcha. Welcome to the transform now podcast. Our guests on this episode is Mesh Schie Vastava, chief technology and digital innovation officer at Clever Care Health Plan. We’ll be chatting today about data driven healthcare and the role of automation, AI and machine learning. Welcome to Mesh. Thank you, Michael, thanks for having me on your podcast. I’m excited that you’re here. Tell us a little bit more about yourself and share a little more about your background. Absolutely my name is on Myshrivastava, currently serving as chief technology and digital innovation officer for Clever Care Health Plan. I’ve been in healthcare for I’d say close to fifteen years. Have worked in both large and small organizations. Have worked with United Health Group, optim Kaiser, permanente and then worked with startups like p three health partners, which grew into a public organization, and now with clever care health plan. My educational background is I’m under my Undergrad is electronics and communication engineering. My masters was in engineering management. I’m an avid learner, so every year I go back to school and try to learn something new. And the health care is near and there to me. So very excited with all the work that I’ve done so far and what is ahead of us at clever very nice, very so about the education. The only thing is missing like French…

…literature or something like that. You know that you covered quite the basis here. I know some voights to like. So we give English today, so I appreciate you sticking with that one language. So you’re a big advocate for data driven patient care. How can you leverage data to support the clinicians care for their patients? Data is, at this point in the lives that we are living today and the times we are living today, is probably the most critical. Like it’s I think it’s more valuable than gold or anything else out there, because data can inform better decisions. Data can also predict a lot of what is going to happen based off on trends and models. So I think data is super critical now in context of health care, I think it’s even more important. I think if we can use data and convert that into the right information and push it at the point of care, then it can really help clinicians to pretty much make more evidence based, data based clinical decisions. Example, if I’m a patient and I’m in front of my PCP and you can pop up a view which shows last three years of my clinical history, my historical diagnosis codes, my e Er visits that I’ve had in the last six months to a year, all the medications I’m taking, all the pharmacy like, all the labs I’ve been to, my lab results, my vitals. Right there’s so much data that is being generated on a day to day basis regarding my health and if you can combine that in a simplistic view and present to my primary care or my specialist or for my next procedure, the physician will have a little bit more context around meat as a whole person.

So that they can then align whatever care, their care delivery, they’re going to provide to me in a better way. So I really feel like data is super critical to more informed decisions. And then data is also, outside of this clinician purview, can also be used for a lot of predictions on healthcare, on on disease progression, on social interventions and things like that. So I feel like data is probably the most critical part of healthcare economy at this point. That makes sense, but, as you mentioned, a wash in data, whether it’s from the health information exchanges, PHARMACIS labs, hospitals, your primary care physician, specialist health plans and wherever they come in, there’s a lot right of data out there, data sources out there. How can organizations use automation or the tools to be able to pull that information in, to be able to do some of that review or consolidation? You know, that’s a great point you bring Michael, and I can share my experience. The good thing these days is you’ve got standards like fire four point Oh. We are using fire four point oh sort of data models which act as our native data models in which we can ingest data from multiple sources, whether it be clinical whether it be claims, whether it be biometrics and so forth. So I think the need of a proper data operating system or data platform is critical. Now, as you said, there’s different types of data in health care. You’re going to receive data from pcps, specialists, hospital skill nursing facilities, l tax long term acute care facilities, a lot of ADP feeds you get from hospitals to no admit, distance transfer prior. There’s tons of data that that we are generating on a day to day basis and there’s definitely a need to ingest it. Their platforms like aws, Azure,…

Microsoft, azure G CP. Then there are specific platforms around snowflake and talent and a lot of these. So I think technology is there, but in in healthcare you need to have a team that really understands healthcare to really deploy these technologies. So in the past, the way I deal with it is you’ll use a combination of you and just no. Thirty percent of healthcare data is unstructured, so you will get a lot of images and PDFs and and soap notes in which there will be a lot of intelligence hidden. So from a data strategy perspective, you go about building a strong data operating system, data warehouse, whatever you call it. Put the right team of resources who can automate those feeds coming in and, I think, more importantly, if you are aggregating data from all these sources, you want to have a very clear error logging mechanisms built into your et l jobs so that because there’s a lot of dirty data that you’re see so your in as tekeys, our job is to make sure we put the right barriers and right sort of checkpoints that bad data doesn’t enter our gold layer or our integration layer warehouse, and that’s what we’ve worked towards building out. But it’s super critical to have that kind of data to really drive analytics and other informed decisions within the organization. That makes a lot of sense. So you’ve given the fact that some of this data is either hidden, a lot of its own structured, as you mentioned. Some of it could be either inaccurate or mistyped, dirty data, as you mentioned. HOW ARE AI and ML getting access to this data to be used for provider care and where you see that next wave of innovation? I think interoperability. As you said, there is unstructured data sitting so from a technology perspective, AI and m L. Specifically, there’s natural language processing, an LP. N L U is growing pretty fast. We use Amazon a…

WS product called t extract, which is at the point of intention where we are ingesting unstructured data, and the great thing is it works out very handy to convert a lot of unstructured data into structured data. Are Extracting core intelligence that we need from that. So the extracted is the tool we are using, but there’s a lot of other NLP technology that’s out there to deal with unstructured data. So that was first first part of your question. How do you deal with unstructured data? The second part was the future of health care. Is that what you were looking for? Sorry, I missed that. Yeah, the next wave of innovation, because you mentioned stuff that’s related to sort of text, pulling text out. But as we look at AI and ml they’re not just being used from a text and notes perspective but also, as you mentioned, really something images and imagery that can we leverage either by radiologists or whoever they are coming as part of these data packages, if you will, for that particular patient. As we look at that, what kind of way? What’s the next wave of innovation that you see that you’re coming and the heels of the things that are already been built. There’s a lot going on in health care right now, Michael, and I think data and and just ai is growing pretty fast. Right there’s a lot of disease progression models that that we are building out and that the health care world is building out, where you can really predict how you’re how the diseases will progress outside of just the clinical factors. Right there’s a lot of social determinants data, there’s a lot of clinical data and there’s a lot of historical claims data combined and when you run through the models you can really predict if the current condition is not managed, then where can this lead to? So a lot of disease progression models, a lot of dynamic risk stratification, which is already going on, but I really see like in the future, from a data and analytics perspective and Ai Perspective specifically, it really will be used two to have, because there’s such a clinician shortage in our country right now and I truly feel like ai will help do a lot of like triaging of the right patients,…

…of the conditions that need to be addressed as opposed to looking at a population as a whole. So I think that’s where there will be a lot of a lot of movement. But in general, like there’s a lot going on in healthcare. Interoperability, as an example. Right you can do ai if you don’t have data, and interoperability has been a big challenge in healthcare. So I personally feel in the future interoperability will be mandatory. That’s the thing. If you’re a health system, you have to share the data. A lot of that is already out and it’s already moving in a positive direction, but in future interoperability will be mandatory for organizations. The big part of health care right now is lack of transparency. You get all these bills and you don’t know who to talk to, whether it’s the insurance whether it’s a provider. Who Do like patients or members or individuals are confused because of lack of transparency in healthcare, which leads also there’s we are, like spending a lot of money in health care. So I think price transparency is going to be something which will be central and it will be just like any other industry. I think constant, constant diagnosis, variable devices, is going to become more mainstream. Right, you will have a patch that you can put on your body which is constantly diagnosing you, constantly taking your key vitals, your data, and then pushing it back into some level, some kind of AI mechanism, which can really predict how you’re doing and then what kind of interventions are required, whether it be from a nutrients perspective, whether it’s going to be from a clinical perspective. In general, right, a continuous monitoring will be more mainstream in the future. Cancer. Last decade was really all about diabetes and CHRONIC CCHF heart disease. Cancer is, I really feel this decade will really be dedicated on on fixing a lot of problems with cancer, right,…

…proactive elimination of cancer cells, correcting gene mutations, things like that, manufacturing antibodies, I think that will be something which will become more centralized. And also, I feel like genomics right now is great stuff that we all hear about, but it’s not democratized. But I think in the next decade genomics will be a little bit more mainstream. So yeah, there’s a lot going on. Michael. I don’t know if I can tell you one area, but there’s just so much going on in healthcare in very positive direction and I’m very hopeful that the future looks bright. That’s that’s very interesting. There’s like two of those things that you mentioned, because there are so many. Start of the first one from from a the perspective of democrazyation and sort of lack of clinicians. Within the pandemic we’ve certainly seen an increase Intel medicine. How are these clinicians using chat paths and remote technology to support patients and increase their level of care? Yeah, I think that’s a great point. Interoperability in in this pandemic, all the folks who were involved talking about and trying to improvement tell health. What the pandemic did is it brought telehealth mainstream necessities, mother of all inventions. We had telehealth around for a long time, but it became really mainstream when pandemic hit. It was learning on all sides, learning on side of the health systems to really deploy telehealth at scale, learning on the side of consumers. Are Folks who want the health care service because you know, healthcare, like having your physician on a call, is much different than the touch, the touch and feel healthcare that you will get in a brick and motor setting, and then this was quite a change for the physicians as well. I still feel like telehealth is evolving. Just you can do basic visits on telehealth cold golf, body pain stuff, which is…

…it’s basic in terms of your needs. But I think telehealth is also evolving in models where I’m seeing a lot of like hybrid telehealth models where you know m mas and maze and other type of resources are going out to patients homes and connecting to a telehelp because now you’re they’re also getting the vitals, doing the screenings and then connecting the physician who’s at a central line directly. There’s a lot of telehelp in mobile medical units in remote areas that folks are doing, but there’s a lot of remote patient monitoring lead telehealth when you need more longitudinal care as opposed to episodic like that. A lot of people who are on care management programs, disease management programs, Diabet, C K D, C Chf, C op d. We send devices out to them where they are going to use these devices on ongoing basis. We are constantly the data is constantly being sent into the central console, which is letting a team off. Nurse Practitioners as well as physicians know that what’s going on and if there are any anomalies on that data, that’s when you know folks who work on intervention. So I think they’re different for different use cases. There are different sort of modes of telehealth which are being used and I really feel in future it’s really going to be more hybrid as opposed to virtual only, where, if the intervention is required, you can send a clinician to patients home, versus most of the Times it will be virtual and interventions can be led through a remote video. Is that yeah, that makes sense and I guess to my second question, which related to you mentioned wearables and the fact that quite a few of us have apple watches, fitbits, that sort of thing, which are giving some level of vitals in some minor way or certainly tracking things over time, depending on what the capabilities of those devices are. What do you see as the next step for these wearable devices, because you just mentioned some disease management where you have these long term health issues that can be remotely…

…monitored. Do you see these wearables coming into playwork? It’s not just gonna be long term health folks that will be having these kind of more sophisticated data management things coming from the wearable devices they have. Yeah, I really think wearables are really gonna be central to all of our to consumer healthcare in general. Everyone is going to have a device which will be constantly tracking, if nothing, then the vitals of the body and then, based off on your gene score or based off on what kind of sort of prediction we’ve got on your healthcare and future, what is that? Your risk on risk for these variables are going to be tracking those particular vitals particularly. So I think the future off wearables, at least from my perspective, is it’s going to be continuous and it will be Democrat diast for everyone. So we will have variables which will constantly pull data on all the consumers and then feed that into some level of I don’t know if that is that by big data or lockchain or whatever, but this data will be continuously fed and then, anormally will be the ones which where the clinical interventions will be triggern too, and at least that’s how I see this in the future. Right now, it’s still getting there. There’s a lot of work going on, but I don’t think wearables are central to how folks are managed remotely, but I have a feeling in the next ten years it’s going to be more mainstream. Gotcha. Okay, so here’s the opportunity for you to give us your bold prediction for that next wave of healthcare that providers need to be aware of and you plan for to be able to care for their patients in the future. And you mentioned some things about wearables, go bowls. What’s the next wave? Where do you see this going? I think data and analytics is going to take a very…

…strong seat and it will enable the physicians. I think what we see at the ground level is there are early adopting physicians who are early adopters of tech. They’re ones who are buying the idea as they go. They’re all burned with the e Mrs Right, which were implemented, which we’re supposed to help them, but turned into this massive administrative burden of the documentation. So I truly feel in the future data, especially artificial intelligence, which includes n LP, where you really don’t have to type it all, the goal everyone in technologies is one of their goals is to reduced physician burnout. So I really feel that’s going to be central a lot of NLP technology so physicians can have regular conversation with their patients. Data will be recorded, stored and they don’t have to type it all up at the end of the day, spending three hours. I think all of that will go away. So N LP is going to be super critical. Disease Progression, ification of diseases much earlier than today will be centralized, obviously, with both bringing genomics centrally, bringing genomics making it more cost efficient and democratize that, as well as a combination of wearables will lead to this continuous monitoring of large populations. So I think that’s something that I truly feel in next decade will be real I truly gene mutation works and gene mutation is there’s a lot of work, a lot of investment going on in that area and that’s going to be centralized. At least I can see that in the next decade where you can you’ll be able, when the kid is born, you would know, based off on the gene patterns and things like that, what what are the diseases that at which this particular person is can get, and then that gene mutations will be centralized or will be done at a at a more democratized way. Yeah, and then, I don’t know if this is bold, but I truly feel if, once…

…interoperability centralized, a lot of art is mandatory across the board and we have systems to support it, it will it will really impact healthcare with the flow of data across the ecosystem. So at least these are some of the poor things that I feel are going to be critical in the next ten, ten to fifteen years. That makes sense and I think as well. And Miss Thank you so much for your insights into the future healthcare and the data the technologies that are be central to achieving better patient outcomes. So I’m excited to see how healthcare is going to change over the next next number of years. But your viewpoints and where things are now and where things going are very helpful. So I really want to thank you again for joining us today. Thanks for having me on your podcast. Very exciting conversation, Michael. Thanks for tuning in too. Transform now, for more insightful discussions on digital transformation and more, check out our podcast channel, where you’ll find all of our previous episodes to make sure you never miss an episode subscribe to the show on your favorite podcast player and, if you like what you heard, please leave us a review. For more information about digital transformation and the future of work, CHECK OUT BLUEPRISM DOT COM to learn how blueprism’s digital workforce is enabling enterprise transformation now. Thank you.