Interview with Orbita CEO Patty Riskind

This episode focuses on the impact of Artificial Intelligence in the healthcare sector. We  explore how AI is improving patient experiences and streamlining healthcare processes. My guest, Patty Riskind, CEO of Orbita,  discusses her journey in the healthcare industry and the experiences that have influenced her leadership approach.

Patty explains the role of generative and conversational AI in healthcare, examining its contribution to more empathetic patient interactions and its role in reducing administrative tasks for healthcare providers. 

We discuss advancements in AI-driven product development, particularly in speeding up chatbot deployment. The capabilities of virtual healthcare assistants, including appointment scheduling and post-visit follow-ups, are also covered.

Host David E. Williams is president of healthcare strategy consulting firm Health Business Group. Produced by Dafna Williams.

Transcript

0:00:10 - David Williams
Today's guest, Patty Riskind, has decades of experience in healthcare analytics and IT, so it's the perfect preparation to lead Orbita, which leverages generative and conversational AI to help customers increase revenue, automate workflows and improve the patient experience. Hi everyone, I'm David Williams, President of Strategy Consulting from Health Business Group and host of the Health Biz podcast, a weekly show where I interview top healthcare leaders about their lives and careers. If you like the show, please subscribe and leave a review. Patty, welcome to the Health Biz podcast. 

0:00:42 - Patty Riskind
Hello David, great to be here. 

0:00:44 - David Williams
Absolutely. So we're going to talk about Orbita, but before that I'd love to hear about your background, your upbringing, a particular you know what was your childhood like and any childhood influences that have stuck with you. 

0:00:56 - Patty Riskind
Okay, let's think about this I had a happy childhood, to be honest. I grew up in the suburbs of Chicago and basically, you know, was a hardworking student, but I can't say well, I feel like I was very lucky, I'm the oldest of four and it kind of blazed the trail. I was always the one who there was a high bar, I had to hit it, that kind of thing. 

0:01:29 - David Williams
That's good. Well, you know the interview, the questions are about you, so, like I'm not, there's no like second and third or fourth question there, so that's what you're comfortable with. That's good. 

0:01:38 - Patty Riskind
And I think it's good to be that. 

0:01:39 - David Williams
You know I'm the oldest child myself, so you know I don't usually look back too much. But yeah, so then you went. It looks like you went to Brown after that, is that right? 

0:01:50 - Patty Riskind
Yeah, so I graduated and I went to Brown University and I studied international relations at Brown. I did a semester in Geneva, switzerland, nice. I studied at the UN and realized that the UN couldn't get anything done, and so that's what pivoted me to business. So I ended up graduating and the first job I took was with a division of Dunn and Bradstreet called plan services. Okay, it is now called health plan services, but it was a third party administrator of small group health insurance. So what does that mean? 

I worked for a company that worked with a number of health insurance companies and I actually did a rotation through every department so from sales to underwriting to claims management and ended up being a customer service ombudswoman and organized their first service guarantee. It was the time of Deming and guarantees were a big deal and I was the one who had to go and tell the underwriters and the claims people that we were doing a service guarantee. But it was very interesting because I got a sense of how cultures can change and how you, by putting the customer at the center of things, you really do change perceptions, and there's both internal customers and external customers. And what that job did was not only teach me how to work in a professional environment, but, ultimately, how important it is that you treat both your internal employees with a level of respect and with a level of service, as well as you treat your external employees or not employees. Customers, yeah, customers. 

0:03:42 - David Williams
Well, it makes good sense. And in a sense, like from your time at Brown, you learned, you know, obviously, international relations, un is a certain thing you don't really know until you kind of get there and say, wow, Geneva it's. You know it's very fancy place and there's a lot of, a lot of good talk happening there. But beyond that, and then I think, to be in a first job where you get to rotate around different areas is really good. I think some people just don't get that experience. They just found out hey, I don't like this, whatever it is whereas you had more of an opportunity to find out what was a fit. 

0:04:08 - Patty Riskind
Well, and I've actually employed that approach with every job I've ever had. So I would highly encourage others to when you're hired into an organization, ask your manager to follow the customer journey and to meet with people. If not do stints in each department, but meet with people who can tell you about their job and what they do and how it relates to the bigger picture. And ever since that first job, every job I've had since, by doing that, it allows me to understand, number one, who's who in each department so I have a contact and two, what their role is and how it relates to the bigger picture. So when I'm doing my job, I have a context for all the touch points that a customer might have and, internally, who to go to for what. So so it was a great learning experience. 

0:05:03 - David Williams
And what happened after D&B? Where did you go from there? 

0:05:06 - Patty Riskind
So after D&B that was in Tampa, Florida, and I'm originally from Chicago and I realized I missed the seasons and so I moved back to Chicago and I got a job with a company called the SACS Group, which was a small company. I was employee number 36. And they it was a software company. They had created a software tool that was predominantly used by hospitals to forecast inpatient and outpatient demand for services, and so I was initially hired to start a telemarketing division or unit in this small little company. So it was focused on sales and moved from that into running the help desk, into running operations, into running their product development effort. So again, it was a great job. I got a bunch of different experience over the course of six years and ultimately, I think when I left the company was around 500 employees. So got to see a little company grow. 

And then I ended up going to a company called Doran Fest and Associates which was a market research company, and I was the vice president of operations and oversaw 30 market research telemarketer types who called hospital CIO and CTO types and asked them about their inventory of hardware and software. And it was a tough job, one, getting the attention of these CTO and CIOs and two getting them to actually give you the information. It was like pulling teeth, but ultimately we provided consulting for these hospitals in terms of what to do next and what to think about next in terms of their software purchases, and we used the content that we collected and compiled a massive database that we sold to vendors who could understand what that institution would need. So that was. That was an interesting job. 

And then from there I went to a division of the American Hospital Association called Health Forum, and Health Forum focused on education as well as publications as well as databases, and I was in a business development role, and the beauty of working at the AHA was I got to interact with all kinds of hospital CEOs, so really grew my network and, in addition, representing the AHA, everybody would take my calls, so was able to sell the AHA annual survey and forms of it, and there was hospitals and health networks magazine and trustee magazine. So it was a really it was a great job. 

0:07:59 - David Williams
Then it sounds like after that you were so done with the association and then like to a smaller company and then onto, I think, sold to Press Ganey eventually. Is that right? 

0:08:07 - Patty Riskind
So you're still on this, yes, Eventually I moved from to a consulting firm, realized I'm not a consultant, I like to make money when I'm sleeping and started a company called Patient Impact, which was the first electronic survey company in healthcare. 

And we took what SurveyMonkey was doing. I hired a developer off of Craigslist, yeah and then built my own electronic survey solution which ultimately Press Ganey acquired in 2009. So when to Press Ganey was there for almost eight years, ended up being their chief experience officer, helped take the company public in 2015 and back private in 2016. And at that point I was I needed a break and ended up doing some board work, went and did a stint at an early, early stage virtual reality software company and then ultimately, qualtrics came calling and I ended up going to Qualtrics and was the head of global healthcare at Qualtrics. Establishing their healthcare division had an incredible run and then Qualtrics went public and it was at that point I said, huh, maybe going back to early stage wouldn't be so bad and I ended up at Orbita Great Well, that's quite a tour to get to Orbita, which is where we're starting In this discussion. 

0:09:33 - David Williams
So that's excellent. I know that is. It's really helpful to hear that background and that that career path it's, it's good and sometimes it doesn't always look, I don't know, sometimes In I said in the intro that it's like the perfect background to run Orbita, which I think it is. The path makes a lot of sense as you describe it. It doesn't make more sense, like as you look back on it, like versus. At the time did you feel you were you're sort of going in this, in this direction, so I stumbled into healthcare. 

0:10:00 - Patty Riskind
Kind of unknowing and started on the payer side and then when I moved back to Chicago, I was serving more providers, yeah. And so, looking at it, I think the consistent theme was I predominantly worked for small companies that grew. So the SAX group, Dorinfest, Tiber Group and others grew and I saw the entrepreneurs that were doing it and I think I got my MBA from the, from the, from the, from the, from Kellogg at Northwestern. And when I was at Kellogg I was in an entrepreneurship class where I had to write a business plan and the business plan was starting a consumer online consumer reports for women who needed to pick an OB-GYNE Because I was pregnant. 

When I went to Kellogg I had my son after I graduated and, bottom line, I took that business plan, dusted it off and realized I can do this. I'm not necessarily starting a consumer reports, but I can do electronic data collection. That would help not just consumers make decisions based on their feedback and publicly available feedback through a hospital or through a medical group, but ultimately it helps that group improve. So that's a long way of saying I saw these small businesses grow and I thought I can be an entrepreneur myself and then I jumped into the fray in 2004. 

0:11:26 - David Williams
Sounds good. So orbit is a little bit different from some of these other players. Some of the same customers maybe, and I'm sure there's elements that cross over. But you know how did you get interested in in orbita? What was exciting about it? 

0:11:38 - Patty Riskind
So my whole career was in large part measuring patient experience, measuring employee experience and understanding what needed to be addressed, and the appeal of orbit is that actually can solve the problem. So complaints about wait times or complaints about communication using a digital virtual assistant to help you get from point A to point B was very exciting to me. You know, you think about open table or Uber or Amazon even. Why don't we have something like that for health care? Why don't we have something that makes it super easy for a consumer to find care, to find the right doctor, to get the right information? And then ultimately, on the flip side, for those that work inside of health care and health systems, how do you reduce the administrative burden, excuse me, how do you help them serve their patients more efficiently so that they have time to focus on the meat or the good stuff in terms of the actual patient interaction, versus having to do all the administrative stuff? 

0:12:46 - David Williams
Got it. So I met orbit I think about three years ago, was four years ago, before the pandemic. I remember having some discussions there and as I was looking at it more recently in preparation for this conversation, I see the positioning around AI, more of that, and so one question broadly about AI is you know where is AI a good match with health care versus where is it sort of like not ready for prime time? How do you think about that? 

0:13:11 - Patty Riskind
So, or, but originally started as more of a voice oriented company using Alexa and Google home to predominantly help at the bedside. So a patient could say I need a glass of water, alexa, and instead of the nurse call button, the patient could get served by just using a voice request. We've since evolved to use chat bots as well as email texts, along with voice in the form of IVR in terms of phone calls. And where the application and health care makes a lot of sense is automating workflows making it easy for a patient to find care, to schedule an appointment, to fill out their pre visit paperwork, their post visit follow up and reduce that administrative burden on both clinicians as well as administration. So what Orbita does, and where I see a natural fit in health care, is eliminating that administrative burden and making it easier to navigate through a patient journey. 

We recently did an acquisition where we now have care pathways for complex surgeries. So if I'm getting a hip replacement or I'm having bariatric surgery, oftentimes there's several steps to one qualifying for that procedure and then to getting all the paperwork and the labs and the imaging, everything done, and you need someone to hold your hand and what we've created is a virtual assistant that does that and that means it's reducing the number of phone calls that patient has to make and the number of phone calls a nurse or a clinician has to answer or has to address. And then it helps ultimately a patient to be prepared for the procedure or the surgery that they're about to have and it allows that patient to show up and then more often than not, could get the surgery and have good outcomes. 

0:15:19 - David Williams
So I, you know, I see the value proposition because, from the standpoint of the provider organization, they have a hard time just even staffing people. So you know, never mind getting to the meat of what they want to do. It's just that you can't even. You know, if you get through, it's going to be very difficult. And you notice this not just with providers but in general. When you make a call now they're basically telling you you know, please don't talk to me, please go to the website. You could do it there. You could get all the information and, you know, could I text you and your wait time will be long. You know longer if you don't, and and I'm a kind of a technophile and I don't need to talk to somebody if I can find it somewhere but I'm almost invariably disappointed with my chat experience. Chat bot, live chat search for it on the website, because what I'm calling it's generally I got to talk to a person. Are we getting closer to being able to bridge that? Am I just particularly fussy? 

0:16:11 - Patty Riskind
No, I don't think you're fussy. I think that you're discerning, which is good. There's nothing worse than a bad bot, and I think the key is leveraging the technology that's out there in order to create an empathetic and human like experience which actually gets you answers. And the good news is that generative AI has allowed for data ingestion to happen super fast, which in turn creates a knowledge base that can be tapped to answer all kinds of questions. And then generative AI also can learn through machine learning. Where are the holes? So what other content do we need to ingest? 

And then Orbita compliments generative AI with conversational AI, where we're taking what someone asks and chunking it down into pieces that you know. We call it an utterance and intense. So what is somebody asking for and what is an appropriate response? Do you need to ask more questions in response to their initial request in order to give them a sufficient answer? Or do you need to upfront say I don't know, you've got to talk to a human being, so is the answer more complex, which really requires a human being to be in the loop? So I think there's technology today that allows for a better bot experience. I do think chat GPT has broken down some of the old perceptions that chat bots were super annoying, because now chat bots can actually be great and there's part of the population that's all they want to do. They don't want to talk to a human being, they only want to interact with a bot. 

0:18:05 - David Williams
So I saw a demo of some of your latest offerings and I will say that I'll give it the benefit of the doubt, and I think it actually it does go beyond what my experience has been. I think, in order to get at some of the issues that people are concerned about with AI, like making up an answer, you've got it that it's constrained by the actual content that's on there and it is, in a sense, a better way to navigate whatever existing content there is. Now I jump right ahead and say, OK, well, now I'm going to see where are the limitations of that content or what else would I want to do with it, now that I know I can interact with it this way. Talk about your product development pathway, what's deployed now and where you see it's going Sure. 

0:18:45 - Patty Riskind
So I don't know, if you want, about writing prescriptions. So there are certain things that and I actually think in many cases there's always going to be the need for a clinician because it's a human to human business. But where we've evolved is we've been in the digital front door business for some time and we used to do data ingestion and it would take months to get enough content, so that, say, a chatbot was knowledgeable enough to be helpful. And now we use generative AI to do ingestion and it takes us minutes, if not hours, to be prepared in order to roll something out. Where there's work to be done is probably anticipating where is this going to go. 

We know that we can do Q&A chatbots very easily. We integrate with electronic health record systems or scheduling systems in order to hold the hand of the patient to actually schedule that appointment. We then can get post-visit follow-up. So, once they are discharged, we know to do outreach and we use text, email and voice to do that. So where I see this going is that these tools will become more and more sophisticated and we will have enough data that we can start to predict what someone's going to need or what they're going to ask, and so we can be proactive in terms of team things up. The care management pathways are probably a step in that direction where we already know this hip replacement patient is going to need X, y and Z. I think with more data and over time with machine learning we're going to see. Oh, if they answer this question a certain way, we know to push them towards physical therapy or tell them go to the emergency room, and we can probably better predict how the path might change and then how to support them in that. 

0:20:53 - David Williams
Do you anticipate that you ought to also take into account, let's say, broader data sets? So if I'm on the website and I'm trying to navigate, is it going to be relevant what insurance I have and maybe even where I am in my deductible, my personal medical record? Would those things be taken into account? And how far down the road is that, or is that now? 

0:21:18 - Patty Riskind
So that is now, and what we can do is check someone's insurance. So, based on the name and demographics that they provide to us, we can go out and check if they have insurance, let alone. They might not know, or they can say I've got this plan, but they really have something else, or they might not realize oh, I'm on Medicare or whatever it may be. So, one, we have the ability to check and then, two, we can do pre-authorizations so we can understand is the procedure they're inquiring about? Is it covered? And then what is their deductible? And then there are ways of finding out how much of their deductible has been satisfied. 

So I would say that's here and now, and it's only going to get more and more robust over time. So, using that data source, the other sources, we can use Google Analytics where did someone come in from? Were they on the website or did they go direct to a specific landing page? So there's different data sources that we can use to complement, to get a more holistic picture of who's interacting with this virtual assistant and, based on that, knowing the language that they speak or their medical history, who they saw before or what pre-existing conditions they might have, or whatever it may be in order to respond to that patient in a more holistic way. 

0:22:51 - David Williams
So I have another question, which is about tone. So when I get things and this isn't always AI, but marketing things it's very much I think I'm in my mid-50s and I get stuff and it definitely seems like it's written for, I guess Gen Z and I was buying a key from my kayak rack and I abandoned the cart and I came back and they said, hey, we won't tell anyone, but here's this discount for you and it's very. Or with my watch you'll say you rocked your move ring and it's like I closed it. It didn't rock it, but how do you think about the tone? So there's partly you're dealing with a curmudgeon like myself, but there's also like, is it like, hey, you're having a baby, that's one thing. Versus you have terminal cancer and we're getting you into hospice. So like, how do you adjust that? And also, how do you do that without making inappropriate assumptions? Just cause like, maybe I'm old, but I like to be, I like to act like I'm 20. How do you deal with those sort of things? 

0:23:52 - Patty Riskind
Well, it sounds like they missed on their target audience. 

Yes, in that case, I do think that you know there's demographics, there's psychometrics, there's socioeconomic indicators that the more data that you can take into account, the more you can customize the response. 

And we're seeing, you know, gen AI in terms of avatars, like video avatars can reflect their user, so you can ask specific questions and then they can spin up something that's gonna resonate. So, you know, I would talk to a middle-aged woman type of thing, but I think it's gonna get more and more advanced and sophisticated over time. I think today it's probably still fairly generic and it's a decision that say that kayak company is taking into account of who they see their target audience and thus they're putting messaging out there or creating an interaction that reflects what they think. The ideal is, when it pivots to, it doesn't matter whoever's visiting it should adopt to that person's profile, and that can be language. That could be use of slang, that could be, you know, colors, it's probably. However, a chatbot is designed, it should, it will be, I'm guessing, nimble enough to adjust based on who's interacting with it. I don't think we're quite there yet. 

0:25:27 - David Williams
No, I know so clearly at you know AHA, prescany, qualtrics, you got the sense of like, how do you measure experience, what matters, and now you're reflecting some of that in what you're doing at Orbita. How does that loop get kind of closed or tightened? In other words, as you see what you're rolling out, what kind of an impact would you expect to see from you know, prescany or Qualtrics feedback? How is that all gonna get worked in? What is the sort of loop that will occur? 

0:25:57 - Patty Riskind
So our customers are very interested and concerned with return on investment. So if it's a I'll call it a digital front door type application, where someone's going to a website to look for something, you want a sense of closure. Did they end up scheduling an appointment? So that's a very clear metric. They expressed interest and they booked, so that's easy to measure. And then in terms of care pathways, did the patient see it through and have the surgery? And then, post-surgery, were there any complications? So those are very clear metrics that you can measure and hold up to say this saved this much time and it got this type of result. 

I also think from a patient experience perspective, which is the majority of my background, asking the question did you, you know, on a scale of zero to 10,? How likely would you be to recommend this interaction or this bot to others? I think measurement becomes also very easy when you're having a digital exchange or even an audio voice exchange, by asking the questions did we meet your needs today, yes, no? Or on a scale of, you know, one to five or zero to 10, whatever it may be. I think that becomes part of part for the course you know, were your requests fulfilled? And if the answer is yes, then you know that you've accomplished something. 

0:27:22 - David Williams
I can see the the initial value proposition is being reasonably straightforward. Like, if you are reducing calls and that has a big cost if there's a call, so if you know, reduce calls by 20%, 40%, whatever you do like, that's a clear ROI. If there's a fee for service patient and they get the hip surgery versus not, you know there's a good ROI. The next step in terms of within value based care, where it's not just the patient engagement, of course the patient engagement score matters, even directly for the reimbursement. But if we put that aside, if you have just sort of better engagement, can we get to better outcomes. That it would be measured in such a way that, like if you've got a Medicare advantage or other value based kind of approach, that you could actually see a difference. How important is what you're doing in that? 

0:28:06 - Patty Riskind
So I would say, very important in that, especially with value based care, you're trying to squeeze the costs out of the system and out of the processes, so the more efficient you can be by making sure that the patient understands what's happening and takes the appropriate action in a certain time period and reduces the back and forth and the potential complications that might pop up, I think that there is a significant cost savings in time as well as in resources, as well as in just basic dollars. 

So, knowing that you have a care pathway, virtual assistants that are powered by Gen AI and conversational AI and machine learning can help ensure that the patient stays on that pathway and, if there is something that goes wrong, an escalation so that it can be handled quickly and reduce the potential for someone to end up in the emergency room, where it becomes costly post-discharge, making sure they stay out of the hospital. 

I think there's a lot of components that automating that workflow and building in appropriate touch points or nudges whatever you want to call it to ensure that patient's okay. So we've had a customer that does. We're part of their pacemaker app. We'll call them and every day we reach out with five questions to ensure that patient's on course, and the five questions rotate through a bank of 150 different questions, so it doesn't become annoying, but it helps ensure that that patient's doing okay and if something is awry, if something's weird, then there's an escalation so that an intervention can happen before that patient ends up in a dire situation. So that's just an example of how the technology can be used that doesn't require people but can result in better care and better outcomes Great. 

0:30:18 - David Williams
So for my last question, I'm going to go back to something a little more personal and ask if you've read any good books lately, anything you would recommend or anything you would recommend we avoid. 

0:30:29 - Patty Riskind
So I'll be honest with you. I haven't had time to read any good books lately, but I've been watching things on Netflix and Prime that counts yeah. 

Yeah. So you know I try and lose myself a little bit in thrillers and I'm watching the fall, which is really intense, so I recommend it's one of those. It kind of sucks you in. You can't stop watching it. 

And then, in terms in general, in the healthcare industry and in the AI world, I would say it's important for everybody to know what's going on and be up to date in terms of what's going on with chat, gpt and what are the ways that conversational AI and generative AI can be used to streamline workflows and make people's jobs easier while at the same time ensuring that there's guardrails. So you know, anything you put out or put into chat GPT then becomes publicly available. So, especially in healthcare, it's very important that privacy and security is, you know, table stakes, and so how do we ensure, when employees are using chat GPT, that they're not putting anything out there that is related to PHI or, you know, could, could be cause problems? But also, how do we take advantage of it and how do we put the appropriate guardrails in place so that it can be a source of truth, it can be authenticated and validated and used in a way to make life easier? Well, great. 

0:32:08 - David Williams
Well, patty Riskind from Orbita, Thank you so much for joining me today on the Health Biz podcast. Thank you, you've been listening to the Health Biz podcast with me, david Williams, president of Health Business Group. I conduct in-depth interviews with leaders in healthcare, business and policy. If you like what you hear, go ahead and subscribe on your favorite service. While you're at it, go ahead and subscribe on your second and third favorite services as well. There's more good stuff to come and you won't want to miss an episode. If your organization is seeking strategy consulting services in healthcare, check out our website, healthbusinessgroupcom.