ROI of Social Determinant Programs with Alex Trahey, VP of Data Science at Algorex Health: Audio automatically transcribed by Sonix

ROI of Social Determinant Programs with Alex Trahey, VP of Data Science at Algorex Health: this mp3 audio file was automatically transcribed by Sonix with the best speech-to-text algorithms. This transcript may contain errors.

Saul Marquez:
Hey everyone, Saul Marquez here. Have you launched your podcast already and discovered what a pain it can be to keep up with editing, production, show notes, transcripts and operations? What if you could turn over the keys to your podcast busywork while you do the fun stuff like expanding your network and taking the industry stage? Let us edit your first episode for free so you can experience the freedom. Visit Smoothpodcasting.com to learn more. That’s smoothpodcasting.com to learn more.

Saul Marquez:
Welcome back to the Outcomes Rocket everyone. Saul Marquez is back and today I have the privilege of hosting the excellent Alex Trahey. He is a health care data scientist and technologist with 10 years of industry experience with health plans and pharmaceutical companies. He has published research on the clinical and cost effectiveness of a variety of products and interventions and brings a deep expertise in using modern statistical methods, combined with claims and clinical data to study interventions in a real world setting. After studying for his MBA at MIT Sloan, he built the analytics and data science function at Wolfram as its first data science hire. And he’s passionate about building and validating scalable data products in health care organizations. When Alex is not working on analysis, he’s almost always playing ultimate Frisbee and the work that he’s doing at Algorex Health is fascinating. You guys have probably heard our interview with Jacob Luria, the CEO of Algorex Health. If you haven’t listened to it, it’s episode 536. But Alex is doing some extraordinary things with data science. And today we’re going to be diving into those today and just diving into his thoughts around some of the opportunities for bridging the gaps in data. So, Alex, such a privilege to have you with us today.

Alex Trahey:
Thanks, Saul. I’m really looking forward to the conversation.

Saul Marquez:
Likewise. Likewise. Before we dive into to the work that you’re up to at allegorically health and the many things you’ve done in your career, talk to us a little bit about what inspires your work in health care.

Alex Trahey:
Yeah, so I kind of got into health care a little bit by accident early in my career. But I continue to be inspired by thinking about bringing modern technology to a place where we can actually help people. I see a lot of really interesting work in the ad tech space right on the just pure technology side. And sometimes it’s tough to think how do we spend spend our lives in the ad tech space just to make sure we can optimize that cycle of getting money into the system and health care? I’m I hate to say this, but I’m inspired by the mess in some ways. There’s a lot of people who are really trying to do the right thing, trying to help people, but they’re hamstrung by not really having the right tools, not really having the right methods, having to deal with privacy constraints and the kind of ongoing pressure from their organization to fit into that box. So just a lot of opportunity to improve both the system in people’s lives and bring keep learning on modern methods and technology and keep pushing things forward that way, too, for sure.

Saul Marquez:
And there is, like you mentioned, there’s all this effort and execution on bringing money into the door, bridging the gap on those inefficiencies and thinking through how. And the best ways to get there is really the I think, the most rewarding work. So talk to us what you’re doing there at Algorex Health and how you guys are adding value to the health care ecosystem.

Alex Trahey:
Yeah, so we are generally focused on analyzing social determinants of health. So if you listen to the conversation with Jacob, you probably heard some of this background. But I’ll give a quick refresher. We work with large health plans and ACOs and anyone in the Value-Based Health Care space will go out and acquire as much data as we can from the kind of consumer marketing world on individual patients and then use that to run a series of models to understand what are the social determinants of health that the members are barriers of the members are facing. So things like food access, unstable housing, social isolation, and we predict those outcomes and then design, run and evaluate campaigns or interventions around addressing those outcomes? So, for example, we did a program in upstate New York with Blue Cross Excellence, where they sell us Blue Cross Blue Shield that identified a couple of thousand people who would really, really benefit from a food intervention and then gave them a monthly gift card for just twenty five dollars a month to a mobile food market. So kind of identifying where the best locations for mobile food markets to be. Who are the patients who would benefit the most? All kind of using the non-clinical data from the consumer marketing world. So trying to bring in one sentence, it’s really bringing consumer marketing techniques to the world of health care to try to address social determinants of health.

Saul Marquez:
Yeah, it’s really great. And the problem of food scarcity, man, I mean, it’s just like it is. We totally underestimate it.

Alex Trahey:
Yeah, it’s crazy. I mean, it’s one thing we often say this to our clients that really we work with a lot of Medicaid clients and would say pretty much everyone needs some help with food. It’s just such a pervasive problem and it has such huge downstream impacts that we we talk a lot about being financially sustainable and running small scale interventions to understand how can we scale them up, understand the value, that kind of thing, but we probably believe most people would benefit from this kind of program. Twenty five dollars a month for food, but it’s crazy to us, just the depth of it that we see everywhere and wish there was. It’s exciting to be in the space. Is it fun to think about those problems? But I wish that there was more work on it. A lot of small scale stuff and hard to do is.

Saul Marquez:
So then you’ve got you’ve got these trucks, these food trucks are. So who sponsors the food trucks? Just curious.

Alex Trahey:
Yeah. So we’re actually working with Healthify as another partner here to think about looking for local organizations and building a network where we can identify kind of CDOs that the academic community benefit organizations identifying local partners who can help and then kind of designing the contract with the health plan around that. And then we’re more the glue of can we identify people, make sure we’re facilitating data transfer back and forth, but then leveraging the expertise of the local organization to kind of deliver the actual intervention. So in this case, the organization is called Food Link and they’re doing great work and we’re kind of taking data back from them into our system to do the evaluation, the downstream evaluation. Can we match if you went to the market and purchased your twenty five dollars plus additional or not, how often did you go? Did that have any impact downstream on potentially PXP attachment or reduced visits over the long term? We don’t have to see thousands of people to really see a big impact on E.R. visits, but with several hundred you can see things like increased PCP visits and increased engagement with helpline. So local organizations largely and in this case kind of facilitated by Healthify. But in general, we can bring national partners to bear where where that makes sense, something like no health for a transportation benefit. But most often there’s great organizations doing work in the area. And it’s kind of a question of how do you connect the dots between the big health plan and the small local organization and knowing what their capacity might be? Yeah, that’s super interesting.

Saul Marquez:
Thanks for thanks for diving into that. I mean, it is just an opportunity. And Alex, you mentioned Right.. I mean, we so much data is collected on all of us. I mean, even recently got the thing that’s happening with WhatsApp and their new privacy center and the like, the flow to the other platforms.

Alex Trahey:
Unfortunately, I don’t have access to all of your message history to analyze your health. They might be where we go someday.

Saul Marquez:
But the point being like the information’s being collected and it’s being used to commercialize to us. But what you guys are doing is saying, OK, we get that. But here’s the way to do well, like do good and do well. And let’s collect those things to help nudge people to get the food that they need to get checked out to go get that COVID inoculated. Right. like these are the things you guys are trying to get done.

Alex Trahey:
Yeah, it’s a really interesting one, because most people, if you ask them, are there health plans looking out for them, they’d probably say absolutely not. Their health plan is just trying to milk me for money. In my experience, worth working at both Wellframe and then digital care management space, as well as at Alaric’s doing analytics and data science and social determinants health plan is trying their best to help people like they. They really do want people to have better health, but they have this massive organization they’re trying to fit into the constraints of. So it’s a fun place to be in that we really are trying to improve people’s lives. And the health plan is funding that, really trying to improve people’s lives. But it is a tricky one of, you know, if you design a campaign and me from Blue Cross Blue Shield is calling you up saying here, can you please answer the survey? We often joke health and really want to do is we’re here about your social determinants. And what we want to come with is an offer first, rather than saying we bought all this data on you and we understand you have a food access challenge. It’s more we have a food program. Would you like to participate? But building that trust is hard. And that’s one thing I think that may take years to undo in the health care systems we have. We have just seen so much problem. But it’s fun to use all that data where we are trying to help. But you have to toe that line carefully. If we were to give just all the data to the health plan, would they go in a different direction over time? Maybe. And that’s a hard one to manage, but fun place to be for sure.

Saul Marquez:
So, Alex, what would you say makes you guys different than other approaches in the market?

Alex Trahey:
I think primarily it’s the trying not to build a another workflow tool or SAS platform, particularly the venture capital capital world wants to creates this pressure to create a workflow tool, bring people onto your platform, have a standardized data model. You know, can we put more activity of the health care system in the four walls of my company? And we try to bring more of a data analysis and data science approach that is the glue between other players. So we work with companies like Arcadia and Healthify and Unite us and others to kind of feed data back and forth to vendors who might need all that kind of combined with, I would say, fairly straightforward approach to analyzing value, but one that I don’t see a lot, making sure we run things at small scale and have an effective evaluation plan. So a lot of my background is in running cost effectiveness analysis. Both for digital interventions and for pharmaceutical interventions and a kind of simple approach to can we study one population versus another with a statistically valid approaches, can we approximate a randomized controlled trial in any way to understand what are the real outcomes? And sometimes that’s a more clever statistical approach, like a propensity for matching or sometimes it truly is randomization where we say, here’s five hundred people who got the intervention, here’s 500 who didn’t. Let’s study the outcomes we think this might impact and then get that full turn of data on the pilot to know did this work? Can we scale it up? How do we go from here?

Alex Trahey:
But how we’re different, I think, is mostly that we’re we’re trying to help other players do this and make sure we provide the data and the expertise to facilitate that process and therefore to collect that what we call the full turn of data from identifying a member to delivering an intervention, to tracking and evaluating the impact and seeing can that inform our better targeting. We want to be that engine more than we want to be the workflow tool we want. We know health care organizations have spent millions of dollars and months of training on their care management tools, in their clinical workflow, tools and otherwise. And we want to help them realize more value out of that rather than to replace it with another vertically integrated one.

Saul Marquez:
Yeah, that’s critical. And the difference is key, sort of like you’re aiming to be sort of that insights engine rather than just another tool.

Alex Trahey:
Yeah. And it’s I mean, it’s easy to say and harder to execute because we see all of your listeners will will know the kind of mess that is health care data. So we have built our platform to try to take random CSV files and text files and Excel files and turn that into a coherent data model, make sure we have enough of a standardization process that we can do that work. But that’s now that’s not sexy. Where to take an analysts pull from a reporting system and turn that into a roster that we can go acquire data on and understand the political impacts? So it’s kind of why I see people haven’t done this. But it’s it’s a battle constantly to know how can we create open data standards. I’d love to see more agile seven or fire or similar to make this easier. But there’s just so much going on that we have to do that hard kind of manual work of standardizing data.

Saul Marquez:
Yeah. And, you know, that’s that’s where the magic happens, right? I mean, that’s where the standardization of it, the organization of it, if you’re able to do it reproducibly, that’s the huge value that you guys are adding.

Alex Trahey:
Yeah. And I’m not sure it can be done in a purely automated way. And I think that’s the interesting part. We tend to think about it is how can we put constraints or tools around just encouraging standardization over time? We’re not going to create one data model to rule them all. We’re going to create a set of tools that allows me to show up with a client and say, let me collect all this chaotic data, turn it into a slightly more standardized version where at least the memory is the same format everywhere and use that to drive value one cycle at a time. It’s not about installing the perfect tool. It’s about getting one cycle of iteration, learning a little bit from it, scaling up the interventions that work and kind of going from there.

Saul Marquez:
And that makes sense, Alex. And so give us an example of something that you guys have done to improve outcomes.

Alex Trahey:
Yeah, I mean, so I mentioned the food truck or the mobile food market one, but I’ll mention another one around COVID. We’ve seen a decent amount of work with delivering what we call care kits or care boxes to about one hundred, one hundred and fifty thousand members in the Midwest with a big blues plan of thinking about how do we identify people who are really in high risk areas for COVID and where we’re kind of bringing expertise and external data where the world is changing too fast to keep up with claims data. In fact, we saw a massive drop in claims data as a whole, just any utilization during COVID, as you might expect, and bring in consumer marketing data to say who are the people who are at risk from either being low income or in areas where there’s high traffic that are at risk of contracting COVID from their prior comorbidities and say, what can we provide as a health plan to encourage them to stay home? So we spent about one hundred and fifty thousand care boxes that include masks and hand sanitizer and a little bit of canned food that kind of encourage people to just stay home for a couple more weeks while the early phases of it happened. And that’s not a really unusual approach. I think a lot of players in the system are doing something similar, but we’re coming in and helping, just helping get that off the ground quickly. A lot of people know this is often something we see. Organizations know that they want to do something, but they don’t quite know where to start. So it’s easy to spinning your wheels for too many weeks. And in the case of COVID, those weeks pile up quickly in terms of exponential growth. So we were able to come in with this with this blues plan and within about a week get a target list of one hundred and fifty thousand people and send out these care boxes and make sure that we could reduce the impact of the spread of the virus.

Saul Marquez:
That’s really, really awesome. And just having the expertise to to quickly scale and and implement a project like this is as important as as we saw through months of twenty twenty how quickly things could escalate. Awesome that you guys were able to do that.

Alex Trahey:
Yeah, it worked in. A favor that we’re very lightweight. I think we talked about the not being a SAS platform and there’s some there’s some real amount of power of being a platform of having control over the whole system because we’re just lightweight and we have the expertise to kind of bring in that data and no real tools to install or train or anything like that. It allows us to go fast. Obviously, that has its downsides of we can go too fast and get a little bit more chaotic. Sometimes that works. I think in COVID it’s really played to our advantage. In other cases, a more elaborate, comprehensive approach can be can be right. So it’s been fun to see. Well, fun is the right word, but it’s been fun to try to stay agile. The world changing fast for reasons that we don’t like has been a challenge. But it’s been our kind of smallness has played to our advantage here.

Saul Marquez:
That’s wonderful. And so right now, especially with the pandemic, we turn the year. I feel like social isolation is one of those things that that sticks out any anything around that and things that you guys have done.

Alex Trahey:
Just curious. Yeah, we think about this one a lot. We had been running some programs that were like togetherness nights or kind of in-person events for seniors who were at risk of social isolation. And we run a set of models to try to understand social isolation if we can. It’s a hard one to predict mental state of someone, but at least to know you live alone. You don’t have access to a vehicle. You live in a low density area. Those are kind of factors that we know are correlated in the pandemic. That’s much harder to address. Right. we it was much easier to say. We know you’re socially isolated now. Let’s bring you in for a bingo night or something like that. Pretty lightweight interventions now, very difficult. So we’re working with an organization called Papà Pals, which you may have heard of, which is kind of interesting to pair younger people with seniors to to kind of help address both technical challenges and social isolation, at the same time thinking about can we do more digital interventions to get people together. So on my workout classes that are pretty lightweight, none of the solutions are perfect, even with all the resources available, being socially isolated as a real problem in the pandemic and only one that’s going to get worse. We do run models on it and think a lot about different kinds of interventions to help address that where we can work, where we can find it.

Saul Marquez:
And I think that’s great. Yeah, all these things need attention and you guys are doing such an interesting job, effective job and approaching it in an interesting way to get results. What would you say has been one of the biggest setbacks, Alex, and a learning that came out of that?

Alex Trahey:
Yeah, I won’t mention the organization, but this is it’s happened in various degrees a few times of kind of providing data science for the sake of data science. I’ve talked a lot about getting to an intervention and our company is built around trying to get value in the hands of patients. But so often we’ve we try to start with an analytics project or a population summary or similar. And a couple of times we’ve shown up and done this whole analysis and what generally we produce a slide deck and a strategy, but don’t ever get anything done. And that’s a very frustrating place to be. I think early we thought we could provide great data and a great set of models. And because they’re excited about social determinants and wanted to pay us to do this analysis, they clearly we’re going to take it and run with it. And we really found that if we show up, it doesn’t matter how great the targeting list is, unless we’re the ones driving the intervention or really pushing them to get often twenty five yeses in an organization to get an intervention out the door. And we needed to bring that expertise in-house to say if they’re ready to run with it, great. But most of the time they know social determinants are a thing and they would like to do something, but they don’t know what it is. So it isn’t enough just for us to give a recommendation.

Alex Trahey:
We also have to be ready to find an organization locally like Food Link in Rochester and do the work to kind of get that contract signed between the health plan and your organization and make sure that it’s getting off the ground. These things always take longer than we’d like. So it was definitely early on in the company was a frustrating experience to have a few. Here’s this ten thousand ten thousand member target list. These people really need a food intervention. Let’s go. And then they would just spend six months sitting on it so we would love for them to be. And I think we see the industry getting a little bit more sophisticated here, but it goes a little slower than we’d like sometimes. Yeah, you’re like, here’s gold and do something with it, melt it into something useful. Yeah. And people are often excited about where in the organization you get to. People want to look good and they say this is a great thing. We have this opportunity, but it’s lower priority. So put it on the backlog. Maybe we’ll get to it next year. And now we’ve spent all this time identifying members, identifying need and not doing anything. And that’s just the reality of the industry where there’s just so much happening all at once. So how do we how do we keep moving is the real learning that we’ve had to figure out?

Saul Marquez:
Totally. And so you guys have been able to create partnerships with companies and organizations, local and national, that help with operationalizing the the insights and and the things that that you provided like this list, right?

Alex Trahey:
Yeah, absolutely. And sometimes that’s we think about how do we bring on national vendors so we can be prepared to bring transportation or food or even just financial support to anyone out of the box, but so much more often the local organizations are the ones that do the best work. And you look at companies like Healthify and Unite US that are doing kind of a network building. So they’re building a big tool and thinking about how do we get all of the local organizations organized in a referral platform around it. And those are really valuable pieces. So I think we’re excited to see the progress of that in the industry. We can bring a lighter weight interventions to go ready to go in a few days or a few weeks rather than a few months. But we do believe fundamentally that that’s that scale is necessary over the long term.

Saul Marquez:
Yeah, that’s super interesting. And so are you guys looking for more folks to collaborate? Are you always open minded because there might be an organization listening that says, wow, maybe there’s something we could do to collaborate with you guys?

Alex Trahey:
Yeah, absolutely. I mean, we love to hear from people, especially in the food access space is one that we’ve we found a lot of local organizations in the Northeast that we’ve loved working with wholesome wave and speak to in particular, that we’ve enjoyed working with. But there are lots of people doing great things around housing that we’re interested in learning more about in the food space and the transportation space. That is the only thing we really ask is that you’re able to send individual data. There’s we can close a referral or just kind of know that someone has been provided a service. We’re thinking about how to connect that with the health plan or the ACO and make sure we can evaluate the intervention and know how do we facilitate finances there, too. So I would love for people to get in touch with me. You can look us up on AlgorexHealth.com and find me or really anyone. My colleague Erin Dynarski is a food systems expert and she brings a lot of that to how do we find partners and build a sustainable approach with them. And that’s really exciting work she’s doing. Another person you may want to reach out to, but it’s so if people are listening and want to get in touch, would love to talk to them about it.

Saul Marquez:
Love it. Thanks for that, Alex. And if you’re listening and something resonated with you today of the amazing work that Alex and team are doing at Algorex Health, by all means, we’ll leave links to how to get in touch with him in the show notes. So take advantage of the opportunity. It’s about doing don’t just think about it. Do it. What are you most excited about today, Alex?

Alex Trahey:
Yeah, I’m really excited about kind of more integration of these approaches. You know, we we focus a lot on social determinants now. I tend to think that’s a the most cost effective and I guess health effective choice for interventions right now. But there’s a lot of good work happening in the industry around care management and concierge services and otherwise that are all going to fit together. Right. We’re seeing an increasing trend towards patient first on the whole patient health care. That’s just really exciting to see. It’s I know we’ve been talking about this maybe for ten years, but the momentum does seem to be increasing. And I’m excited to see where that goes around having technology that can collect all the data available and make sure we know what are the things that people need and kind of provide them across the board, whether that’s food access or a doctor’s visit or medication or otherwise. Looking at all that together will make the whole system go better.

Saul Marquez:
I love it. Yeah, that is exciting. Alex, I agree with you. And so, you know, there is a question that I had. So there’s the populations that are covered by insurance, whether it be through an employer or directly with the insurance company. And then there’s the Medicare Advantage. But how about what happens to Medicaid? What happens to the Medicaid population? Feel like they’re the ones that need the most help?

Alex Trahey:
Absolutely. And I would say about 70 percent of our work is with the Medicaid population. Oh, it is here. And often it’s Medicaid CEO or Medicaid managed Medicaid plans. But that is absolutely where people need the most help. Yeah, it’s an interesting one because so much of the problem we see is the continuous enrollment. So that churn problem makes it very hard to make it financially sustainable to invest in people’s long term health care. So it’s an interesting place to be. I would love to see the Medicaid be expanded in some way, more states to pick that up. But there’s a lot of really great work happening in the Medicaid world of trying to address the issues. That’s often when I say that the most cost effective and truly effective across the board interventions are going to be in for a lower income population that doesn’t have those basic resources, food or transportation or otherwise. So we love working that space. I think we’d love to see more sustainable enrollment. So we’re not just worried about a lot of the effort is can we keep them for more than one year, keep them eligible, make sure if they move multiple times that the paperwork has been filled out, stay with the plan so that it is financially justifiable to spend hundreds or thousands of dollars to to improve someone’s health. And that’s a that’s a hard problem, but it is where we see the biggest area of need.

Saul Marquez:
Fascinating. Well, maybe that’s an angle we cover with you guys next time. Thank you, Alex. Well, listen, this has been super interesting and insightful. You guys are making a big difference. So thank you for all you do. What closing thought. Would you leave us all with Alex? And then what’s the best place that the listeners to get in touch with you?

Alex Trahey:
Yeah, I mean, I would say pretty. People in the health care industry out there just to get started. I think the approaches that that we’re bringing are relatively simple, but most of the time it’s about getting started and evaluating how it goes and learning from there. We often want to have the perfect solution to a problem, but really we see the best option is to do something and then see how it goes and learn from it. The more we can build those learning systems, the better off we’re all be. If you want to get in touch and learn about social determinants or data science that we do agree health dotcom is a great choice. My email is, Alex@algorexhealth if you want to send me marketing materials or just get in touch to have a conversation, we’d love to chat with anyone who’s listening.

Saul Marquez:
Thank you so much, Alex. That’s awesome. And folks, take Alex up on it. And I love your message, Alex, because it is about just doing it, you know, and you’re never going to get it right. Maybe you might get it right the first time. I get lucky, but it’s about iterating and trying new things, but just doing it now. And and so you’ve left us with a really strong message. Alex, I can’t thank you enough for spending time with us and really sharing the amazing work that you and the team at Algorex Health are up to. Thanks so much.

Saul Marquez:
Thanks, Saul. I really enjoyed the conversation.

Saul Marquez:
Hey, Outcomes Rocket listeners, Saul Marquez here. I get what a phenomenal asset a podcast could be for your business and also how frustrating it is to navigate editing and production, monetization and achieving the ROI you’re looking for. Technical busywork shouldn’t stop you from getting your genius into the world, though. You should be able to build your brand easily with a professional podcast that gets attention. A patched-up podcast could ruin your business. Let us do the technical busy work behind the scenes while you share your genius on the mic and take the industry stage. Visit smoothpodcasting.com to learn more. That’s smoothpodcasting.com to learn more.

Sonix is the world’s most advanced automated transcription, translation, and subtitling platform. Fast, accurate, and affordable.

Automatically convert your mp3 files to text (txt file), Microsoft Word (docx file), and SubRip Subtitle (srt file) in minutes.

Sonix has many features that you’d love including powerful integrations and APIs, share transcripts, automated transcription, automated translation, and easily transcribe your Zoom meetings. Try Sonix for free today.