As the Chief Product Officer at Cohere Health, Gina Kim‘s career has taken a very mission-driven approach, where she focuses on solving problems related to patient access, health in the home, and prior authorizations. In the episode, she takes the time to explain what prior authorization is in laymen’s terms and shares about her work in helping create more transparency around the patient’s best path.
Transcription
Joy Rios: Welcome! Are here with the Hit Like a Girl Podcast again, my name is Joy Rios. I’m joined by my esteemed colleague, Shereese Maynard. Today as our guests, we have Gina Kim.
Gina, I am so excited to get to know you a little bit better. Can you please take a moment to help our listeners understand who you are and what is your piece of the health IT puzzle?
Gina Kim: Sure! Thank you for having me.
So my little corner of the healthcare world is really digital startups and product management to really improve healthcare using technology. I’d say my focus for the last 10 years has really been improving how patients experience healthcare.
I had an early career in consulting and mechanical engineering, but for the last little while I’ve been in healthcare IT. I started at Castlight Health, which was focused a lot on cost transparency and consumerism. That really spoke to me because it’s about how if you’re a family and you’re trying to make a decision about healthcare, like, of course you need to know how much it costs and what the quality is.
And so, since then I ‘ve been on a series with mission-driven organizations – looking at problems like patient access, health in the home, and, most recently at Cohere, around the way the prior authorization system works.
All of these sound like they’re challenging, complex systems, but at the end of the day, they have some human impact on the way that people make decisions about their care.
So, that’s my little corner.
Joy Rios: So we’ve been talking a lot about insurance lately on our journey and trying to help people even understand it and take a grasp of like what goes on behind the scenes.
So if somebody doesn’t understand what prior authorization is, can you just talk about that? Like what does that even mean?
Gina Kim: So, healthcare is complicated and I think one of the things that makes it really complicated is the way that healthcare is paid for. Unlike most things where you make money, you have a budget, you figure out how you’re going to pay you, take care of it yourself, and maybe sometimes you’ll get a loan from the bank – healthcare doesn’t work like that. Healthcare is about risk.
What you have is a health plan that is looking across a whole population of people and trying to to ensure that the people who need the care, there’s money to pay for that care and that care is being delivered at a really high quality. You also have providers who are trying to do the same thing for patients, right? You go to your doctor and that provider says, “Hey, this is the course of treatment that I believe that you need to be on based on my expertise and training.” And prior auth is a process where the two of these come together – where the doctor and at the provider side says, “this is what the patient needs.”
There has to be an agreement over with whether or not that service will be paid for. So there’s a process called prior auth, prior authorization, where the doctor sends over all that information to the health plan and the health plan reviews it and decides whether or not they will pay for it and gets that determination back to the doctor and the patient.
I know the way that I just expressed it makes it sound like, “oh, of course, like this is how it happens!” But it is a really friction-full process. Nobody really likes it and it has a lot of challenges because of some systemic issues.
The approach that we’re trying to take at Cohere is really that there are a lot of stakeholders – the patient, the provider, the doctor and the health plan – that all need to make this work because at the end of the day, the way that we pay for care requires that there’s this kind of alignment so that the patient can get the best care that they need.
Joy Rios: Knowing what you know, and I know that there’s a lot of patients that might get stuck and say, “Hey, I think I need this care, but then the insurance company says I disagree” is there any advice you could give for patients to navigate that system? Like how to know either what questions they should be asking of their doctor or what they should know beforehand, maybe before saying yes or seeking care.
Gina Kim: It’s such a great question and I think the unfortunate thing about our system a little bit is it puts so much of that already on the patient, right? Even when I was talking about cost transparency, like this idea that you have the system, it doesn’t quite talk to each other, it’s not well coordinated, and then at the end of the day, it’s like the patient who has to pick up the pieces or maybe you are doing that on behalf of someone in your family, like your mom or your dad or something like that.
It’s already unfortunate that the patient kind of has to bear so much of this, but I think the advice I would give is really around is you have to know what the policies are, and you have to get down back to the policy level- it is going through the paperwork, unfortunately.
So every health plan will post the documents that say what they will cover, at least like the guidelines for that. And they will try to make that as friendly as possible – obviously at Cohere, we believe there are ways that we could make that even more accessible to patients – but that’s the place to start to understand what is it that they’ll cover. And those policies are based on evidence and so understanding the evidence behind that is helpful.
Then I think it’s like really working with the doctor, and often, provider groups now have a lot of expertise on what we’ll get, prior auth, what we’ll not. You have to kind of understand, too, like, “okay, why is it that I need this procedure? What Is the evidence behind this?”
And I think the patient can also decide based on what they’re learning, “Hey, do I really need this procedure? Do I not?” and then if you really need it, you can work with the doctor and the health plan. Often, you can also call the health plan and, get more information and things like that.
Ultimately I think you have to kind of educate yourself on that policy, educate yourself on the evidence, and then you have to be able to advocate for yourself. And it’s hard. Like I’m not saying it’s easy, right? Like it is something that is challenging, even for people who are incredibly medically literate and is a part of the system. We’ve talked to people who are nurses, who, you know, suddenly someone gets sick and they have to navigate this and it’s not easy for them either.
So I’m not saying there’s a silver bullet here, but you do the best you can.
Joy Rios: So it’s not necessarily sexy work, but it’s something that needs to be done. And so, your work at Cohere, can you talk about products that you are developing? Are you making that process easier?
Gina Kim: Yeah, we are making that process easier. The thought behind Cohere is, “let’s really align all of the stakeholders in this whole puzzle to the patient’s journey.”
And I think it’s the thought that the current system today is really built up on this transactions, right? Like you go for a visit and that’s the visit; you go for a surgery and that’s a surgery; you go for a procedure; you go for a lab, right? All of these things are disparate, small, services that are happening, but as a patient, it’s one journey to you that you are going through and trying to navigate your way through this maze.
And if we could just get all of the health plan, the provider to align on the optimal care paths for people, and then collaborate on that, you could stop a lot of the paperwork and the administrative burden that the providers have and the health plans have. You could also just make it a lot easier for the patients and everybody to see and have that transparency as to what’s the best path.
And so, I think the work we’re doing here is really around the administrative burden to the providers and the health plans, trying to fix that with, with using, using, uh, technology. Then the second is okay now, how do we actually, you know, use analytics and use behavior change with patients and with providers and with health plans to really start to drive more collaboration on, on these cases.
Which are all based on evidence from, you know, like leading medical societies and, and other places like that.
Shereese Maynard: So the question I actually started out in pediatrics and I worked on both sides where I had to explain to parents of medically fragile children, how their benefits work. Then I also worked with providers where I had to explained to them things like the EOP and all that kind of stuff, which I totally did not get, but of course it also, we had the problem of ELBs being snail mailed and that type of thing.
So we are moving into an area where we do have CPAs or electronic prior auths and that type of thing, which makes things a little bit easier and can improve, um, the problem of the actual inefficiencies, but going forward, do you think that part of coheres mentioned is probably educating providers on how the process is changing and how we’re shifting from, um, this kind of paperworld into this digital world and how that all works and how it can simplify the whole
Gina Kim: process.
Yeah, absolutely. And I think it, you know, I would even go a step beyond education, but like, let’s just make the process better. So I think the technology that we use. For example, if people put in a service that we have it coded up so we can understand all the evidence. We have evidence-based guidelines.
And we are able to just, as soon as somebody submits the authorization, we know a lot about the patient and we know a lot about their history. And we know we have, you know, their prior claims. We have the clinical note that’s coming over as part of this process. Let’s actually treat this as a data problem to them.
Should this be approved and we can do it instantly. So our median time to approval is zero seconds because as soon as we can see that it’s aligned to the evidence we’re able to approve. And so that takes a lot of the burden. Out of it for four providers and it’s a digital experience we of course we support phone and fax and more traditional ways of doing it.
But, you know, we believe this is one of these cases where technology and data are able to really. This problem that that was so manual before and so painful before. So yes, we are excited to educate people about you know, new ways of doing this. And we have great teams that, you know, work with our provider groups and help them understand.
But we also, we have a great design team that has designed software, which is really— it’s easy to use. We have really amazing satisfaction scores and things from our users and they say, this is so easy, right? Like people don’t design software for, um, for healthcare often that that has that kind of design mentality for users as you, I’m sure you will know.
And so it’s an area we’ve made an investment because we believe if you can make it easy for people to do the right thing and, you know, have that be transparent, good things happen just from there.
Joy Rios: Can I ask him, I don’t think it’s a dumb question. It’s just a genuine question. How are people interfacing with your product?
Is it like an API that’s connected to the EHR and also the billing system? Is it a totally separate thing that somebody has to log into? Like how is it. Integrating all of these different pieces of information. Yeah. The interact with it.
Gina Kim: Yeah. So we are, we have a product which.. so every practice actually has portals and things like that, that they go to.
So we do have a portal and that’s the way that most people still engage in. Actually, you know, some still go to facts, so I’m still going to phone, but the portal is the main thing, but then we also, we’ve built in a lot of inoperability. So, um, you know, we have APIs, we also have the ability to kind of reach into an EMR and, um, and get the note where a provider, you know, kind of, uh, authorizes us to do that.
And the harder part is actually like, it’s not so much like the connection to the, to the EMR. I think it’s more actually. We are already getting the clinical note manually as part of this. And so how do you actually work with that data? Right? Like it’s, you know, something comes in on a fax or something comes in, it comes in as a PDF.
How do you work within, so our team has built up a lot of the. You know, machine learning models and other technology that actually helped to extract the information from that. And then help us to use that as part of this prior auth process to make it much more seamless and easy for the provider. I think interoperability and EMR integration, that’s part of that.
I think the world is headed towards, you know, systems where people almost invisibly are able to get the off, but what we have actually heard from a lot of our customers is like, please, don’t get me involved in another EMR project.
They’re really taxing actually. And practically speaking, there’s a division between work that a clinician does and work the administrative people do. And that’s the reality. And then healthcare and stuff. You know, I think it’s a little bit naive to say, oh, once everything’s in the EMR, it’s going to be amazing.
Right? Like everything will happen from there, but actually like those of us with scars on our back working with, you know, you do one EMR integration, you’ve done one like, and you know, and a single health system may have eight instances of, of an EMR and you have to figure out a way to, to work with that.
So for us, I think the question is like, what’s a solution that’s gonna work for everybody off the bat already, you know, improves their workflow already gets patients seen faster because they can be scheduled faster because you’ve gotten them the right approvals, um, instantly.
And if you can solve that, the EMR piece is like, you know, cherry on the cake and the next piece that I think we can solve, but I think we feel really good about kind of the approach that we’ve taken that can improve things for a hundred percent of providers.
Shereese Maynard: Okay. Well, I don’t know if you saw the last on AMA survey, but for that survey, 30% of the physicians that they surveyed said that their patients had negative outcomes due to the PA requirements.
I’m wondering, are you guys able to actually show how your product impacts those scores or how you’re able to change outcomes for people in a way that doctors can actually see and then.
Gina Kim: Yeah, absolutely. So I think there’s a couple pieces to that. And that AMA survey is, you know, like they, they run that every year and every year I’m like, oh, I wish this problem were getting better.
Right. And so I’m happy that, you know, we can, we can be part of that solution. So one is that we have, uh, just by authorizing care instantly to be aligned to the evidence, what you’re able to do is get that patients seen faster. And I think like one of the questions on that AMA survey is like, you know, how much has the patient been delayed?
And have you seen delays in care? So delays in care is a quality metric that you can measure. But the second piece is around the appropriateness of the care, right. So, you know, and this is a harder problem. It’s how do you identify the appropriate variation in care? Like where is it that you have a patient that is part of a cohort that should be on a different care path, right?
There’s a good example around something like joint surgery. There are some patients for whom conservative therapy will not help them. A lot of prior models of step therapy, essentially, where to say, well, people have to go through this path, right?
That was before you had analytics and a data approach that would allow you to have more personalization given a cohort. So, then if you’re able to say, Hey, this patient maps to this cohort and therefore they should have a different decision. And therefore they are getting the right services, um, that they need, according to the evidence that has a quality outcome.
And so we’re looking at ways to, you know, there are more standard quality outcomes, but looking at things like joint mobility, right? Like, and at the end of the day, like, what does the patient care about? Like, Hey, I can’t actually get out of bed. I can go, I can walk. I can get to the bathroom by myself.
Like I can return to the activities of daily living. And so we’re looking at ways that we can actually measure that. As a way to prove more of those outcomes, but certainly on process outcomes, like time to decision and then you can also look at more standard claims-based outcomes.
I think we can, we can show, we have, we have good evidence around the impact we have.
Shereese Maynard: So I always say that regulation never keeps up with innovation. I’m wondering when you look at things like time to approval and that type of thing, do you see those, not only the innovation changing in the next 18 months, do you see regulation changing in a way that benefits to patient changing much in the next 18 months to five?
Gina Kim: Yeah. So I think what’s really interesting here is, you know, this has been such a pain point for everybody that I think prior auth has really re like come to the top of a lot of regulatory agendas. And so you see it on the federal level. The previous administration actually announced a rule that was around interoperability and prior auth for example, and also had things in there.
You know, the PA the prior auth must be done within three days for an expedited case or seven days for a standard case. And you see also state legislature is taking the signs that you have Texas and other, other folks that are looking to exclude the the gist of the law.
If a provider is proven to, for like 80% of the time, get a prior approved services, they should be exempted from the process for one or two years. Right. So it’s really kind of get to the burden. When I look at these approaches, I think it’s really a lot of all that, you know, smart people everywhere are trying to solve this problem.
And so, I think that part is. I do think like we could aim higher, right? Seven days for prior auth, we’re talking about? Let’s do 90% of ours instantly. Like that should be the bar, right? Like let’s find ways that it’s not rooted in the past where this was thought of as a manual process, an administrative process.
Let’s think of this as a clinical, you know, an, a data problem. And if you do that, I think it can really unleash kind of the, the creativity that, that people can bring to a problem like this. And so, I think the legislature, the regulatory approach is absolutely moving in the right direction.
I do think this is an area that the private sector can bring a lot as well.
Shereese Maynard: Do you think that AI can solve some of these problems? And I, I asked you this because you know, in the past, and it’s not the distant past we had CMOs sitting in an office deciding on care.
Do you think we could develop an algorithm that says, Hey, this should be approved in real time. Because that, to me, I have always thought that that would solve the problem. If we could build out a model that says, Hey, if these parameters are met and alleviate some of the biases associated with algorithms, if we could do that, then a simple equation could solve the problem of pre-auth don’t you think?
Gina Kim: I do. And actually, I think what you said there, that last point simple things can solve the problem of pre-op. I think simple things get us 80% of the way there. So AI is really, so I’ll just say simple rules-based approaches get you pretty far. Right. And, and they’re easier to implement and they, they’re not hiding behind the black box and.
I know people throw around AI, like, you know, it’s going to solve world hunger, but it’s just not true. Right. Like, um, and you’re, you also raised the problem of bias. That’s a really important one. I’m sure you’ve read a lot of the literature around the bias that goes into minority populations and picking up features that may not really be correlated.
So, to me, AI is a little bit of a smoke screen. It’s an incredibly at machine learning is a very, very powerful technique, but you have to use it responsibly and judiciously, and like really pointed at a problem where it actually can be solved. Um, we had coherent. So, you know, again, like we want to solve this for a hundred percent of, of people.
AI is really good for smaller problems, and an exploration, things like that, but at the end of the day, these are medical decisions being made. They are things that this impacts the care like this could be your father, your child, your mother, right? Like whose whose livelihood could depend on this decision?
So do you really want to outsource that to a machine? My personal philosophy is, not so much, like, I really still feel that, we need to have AI or other approaches be supportive of physicians who can really make the decision, so it can be an aid, but it should never replace the models themselves. They should be explainable. They should be things that you can understand the logic that goes behind them. Right? Do you understand which variables and what the weightings are and things like that, so that people understand the decisions or the recommendations that are being made from these.
I think AI has a way to go before really being responsibly used in such a substantive and really important process. But we have a great ML team. They are focused on like some problems that we think are really interesting and where, where we believe it can make a difference.
But we’re not, you know, the white horse that comes in and like saves everything. Like I think, like we there’s a lot we were able to do and then, kind of supplements that approach.
Joy Rios: So I feel like I’ve already learned a lot just in, can I talk, I talk really fast. No, no, I totally love it. And I’m like, okay, let me take it all in. This is amazing, but I want to learn a little bit more about you specifically, and I’d like to know. A little bit more about your particular career journey and your path.
And if somebody were to follow in your footsteps, like, has it been linear? Has it been all over the place? Like, did you know when you were 10, what you wanted to do? Can you maybe talk about what is it like to have your job and how did you get it? If somebody wanted your job, what would they have to do?
Gina Kim: I can talk about my job. So I’m Chief Product Officer at a really exciting, growing startup and growth stage company. We’ve raised our Series B not, to get too insider lingo on the investing side. But I helped build products right then, and I also helped build companies.
And so I think my job is a combination of strategy. So, you know, where, what should our company do? What should we invest in? How do we, and then there’s also the people side of things. Build a team with a great culture. Let’s, you know, get in the best people we can like make this like a really great place to work.
And then there’s product. How do we actually launch products that actually solve problems for people? Solve the right problems and do it in the right way. Then there’s process for me, which is how we work is almost as important as like what we work on and why we work on it.
My job is looking across all of those elements and being on a leadership team and working with our board and talking to customers and working with users. I mean, so it’s a pretty fun job. I can’t lie. I think my path here has been, you asked if it were linear.
It was completely non-linear. I like whatever footsteps I have, if people were actually following them around, it would probably look like a giant model. So I started life. I will say, my mom is a nurse, so that’s how I got interested in healthcare in the first place. And when I was in high school, one of my first jobs was taking medical records out of a basement and putting them on a cart.
And, you know, that would get that cart would get shipped up to the floors and then they would get shipped back then I would go file them again. So that was like my, probably my first exposure to actually I guess I can say I’ve been in medical records for a really long time. So always have had an interest in, in healthcare.
When I went to college, I thought I wanted to be a biomedical engineer. So I went and I studied mechanical engineering at MIT. And I realized when I was there that. Just, I was more interested in kind of the people in the systems of things, as opposed to like working on designing a specific part of a, you know, I’ve had great internships and research opportunities.
But at the end of the day, like I think I was more cut out for the business world. And so then I went into management consulting and did that for five years and had a great experience there. I think learning a lot about different organizations, how they work, and it was a great experience.
And then, then I kind of got itchy to go back to building things as opposed to just advising. And so that’s when I would really say like this this last run of my career has been working in different roles, and successively. The tackling different problems in healthcare and in digital health.
I’m always looking for like an interesting problem and then a great team. And then product management has been the thing that’s been really fun for me building software, because it’s agile, you can find a problem. You can test stuff.
I love the design side of things as well. We have a great design team. I think I’ve already mentioned and a great product team here and a great engineering team. So like you get all this collaboration, you got the clinical people coming in and like it’s a really, it’s a, it’s a really fun, fun place to be.
So I don’t know if I have any guidance for people who like want to follow my path, but I feel like product is one of these jobs. There’s no traditional path. Like I have worked with people who have been engineers done have come into it. I’ve worked with people. Who’ve been consultants, business consultants.
Who’ve come into it. People who like one of my product directors has a painting background and then did operations and like live in Japan for three years and came back. I mean like people have these really interesting paths, but at the end of the day, I think like what makes someone interested in product is like, somewhat technical.
They really care about people and then they have business sense. And if you bring those three together, it’s like a magical combination.
Joy Rios: Okay. I love that. Okay. Thank you for that. So it sounds like you get very stimulated by all sorts of intakes and inputs, but what do you do to calm yourself down?
What do you do to stay balanced when you’re not working? How do you spend your time to kind of just keep your own internal?
Gina Kim: I’m not sure that’s ever been anything I’ve been very good at. So I will say, what you see as kind of what you get. I get, I get enthusiastic about a lot of things.
But I think on the personal side, so I have, I’m married. I have two kids, they’re young. They’re six and four and I love him to pieces. And so I think a lot of my home life is actually like corralling them, you know, like going to the park, doing things like that. And then I’m also a musician.
And so, playing music is something that just very spiritual for me. You get into the flow of something. It’s like, it’s art, it’s something very outside of the day to day. And, you know, it gives you a perspective, I think when you have an outlet like that. I think getting outside, playing with the kids and then getting a chance to play music when I very rarely get a chance to do that is really fun for me.
Joy Rios: I totally get you. So what’s your instrument?
Gina Kim: I play violin and piano and lately I’ve been playing European. I love piano. Oh, what do you do? You play something?
Joy Rios: I play some piano and I, I don’t share this super widely, but I learned how to play the bass so that I could be my own in my own wedding band.
Gina Kim: Oh my gosh. That’s so amazing. That is great.
Joy Rios: I played just for a year so I could like practice the songs that I was going to play. And then I essentially haven’t played the bass since, but I know what you mean. It’s, it’s like using a whole different side of your brain and it’s very calming. Yeah, but yeah, I have a piano at home and I play it as often as I can.
Gina Kim: Yeah, that’s awesome.
Shereese Maynard: Okay. Gina, one of the things we ask everybody is what’s one mindfulness thing you do every day to keep you focused
Gina Kim: Mindfulness? I probably could say I could be more mindful, but one thing I do is at the beginning of the day, I do kind of like try to take a step back.
So like, no matter how stressful things are, or, you know, what is going on that day or what I’m nervous about, just like try to take a breath and like, just appreciate, you know, and be grateful for the opportunities I have. And for the life I have, I think, I have a lot of gratitude for my parents and my kids and my husband, and like all of that and the great people I work with.
And so I think that that helps me stay grounded because even if the day kind of goes south or it gets away, you know, like I think starting the day with something that really. You know, grounds you and keeps you positive. I think that that’s helpful for me. At the end of the day, I think this is a little bit less meditative, but I just really enjoy having dinner with my kids.
I mean, even if they’re screaming, family time is also really grounding for me.
Joy Rios: I love it. Well, thank you for spending your time with us today. If somebody wanted to follow you or work with you or understand more about cohere, how would they get in touch?
Gina Kim: Yeah. So I think on the Cohere side, so we have coherehealth.com and we have a Twitter feed and we’ve got LinkedIn and everything. You can find us there. And for me personally, LinkedIn is probably the place to go.
Joy Rios: Beautiful. Well, thank you again for taking the time and letting us get to know you a little bit better. This has been a real pleasure and.
Gina Kim: Yeah, I just appreciate it. Yeah. Thank you for taking time. Great to chat with you. Have a great day.