The Information Exchange: Galaxy Brains & Gardening Tips

Well, well, well. March came in hot. I was out thanks to the unending vortex of daycare-induced illness that plagued my home all February. So Pryce, Brad and Ryan had to carry the load - and carry it they did.On top of that, they tackle:* Jack Dorsey’s Block layoffs: 10,000 down to 6,000 at a profitable, growing company, because the AI tools made the headcount unnecessary* The SaaS-mageddon question: are we watching the first domino fall, or is the “AI replaces everyone overnight” narrative still overblown?* FDA deregulation and Harrison AI’s push to skip premarket review: speed vs. safety in a domain where the stakes are human lives* Glyphosate, Roundup, and the MAHA contradiction: how do you push preventive health on one hand and subsidize empty calories on the other?* The ACCESS Model’s brutal reimbursement math: $15/patient/month doesn’t exactly scream “invest in behavioral health infrastructure”* A galaxy brain segment on whether AI opens the door to a thousand little EHRs (spoiler: the interop team would never sleep again)Relevant Articles* Steve Posnack’s LinkedIn post: Brad mentions the episode was “inspired by” it.* Jack Dorsey’s post on X: About Block’s reduction in force from ~10,000 to ~6,000 employees, specifically citing AI/agentic tools as the reason despite the company being profitable and growing. Ryan describes it in detail.* An Epic Anthropic Alliance: Pryce references something Brendan “just wrote” about discovering Epic is leaning into Claude usage on campus for internal code development.* THE 2028 GLOBAL INTELLIGENCE CRISIS by Citrini Research: A hypothetical 2028 scenario where AI-driven unemployment triggers economic collapse, mortgage market failure, etc. Pryce notes there are “cracks in its argument” and it’s somewhat sci-fi in tone.* AI Transformation Is a Workforce Transformation: Brad cites a stat breakdown: 10% is the model’s capabilities, 20% is what you’re introducing it into, and 70% is operationalization/teaching people how to use it.* Harrison AI’s FDA exemption request: Ryan discusses Harrison AI (and possibly others) seeking exemptions from the FDA’s pre-market review process for AI devices.* Trump’s executive order on phosphorus/glyphosate production: Pryce references a signed order increasing domestic glyphosate (Roundup) production, and RFK’s subsequent cosigning of it.Chapters* Introduction and Conference Takes (0:00 - 4:30): Brad recaps ViVE, where every booth pitched identical agentic AI solutions. Ryan and Price weigh in on the real value of industry conferences.* Block Layoffs and the AI Employment Shockwave (4:30 - 9:00): Ryan breaks down Jack Dorsey’s decision to cut 4,000 employees from a profitable, growing company because AI tools made them redundant. The group asks: is this the first domino?* AI Hype vs. Economic Reality (9:00 - 14:00): Price and Brad debate whether AI will pull the rug out from the economy or diffuse slowly enough to avoid catastrophe. Brad drops a BCG stat — 70% of AI implementation is workflow, not the model — and predicts an explosion of custom software.* AI in Healthcare: Scribes, Code, and Patient Safety (14:00 - 19:30): Ryan draws a line between AI that removes friction (scribes) and AI that replaces humans writing safety-critical code. The group reflects on Epic’s patient safety escalation culture and whether AI-generated code can meet that bar.* FDA Deregulation and Harrison AI (19:30 - 24:00): The group examines the FDA’s loosening of premarket review for AI devices, distinguishing between probabilistic recommendations that need scrutiny and interface-level AI that may not.* Glyphosate, Nutrition, and the MAHA Contradiction (24:00 - 30:00): Price goes full gardener, connecting Roundup subsidies to nutrient-depleted industrial farming and questioning how HHS can push preventive health while enabling cheap empty calories.* A Thousand Little EHRs? (30:00 - 33:00): Brad asks whether AI enables a proliferation of bespoke EHRs. Ryan sees job security for interop folks; Price would rather see a million provider groups than a million EHRs.* Competition, Single Payer, and the ACCESS Model (33:00 - 37:00): Ryan puts on the socialist brain to argue healthcare competition doesn’t optimize for health outcomes. Brad closes hot on the ACCESS Model’s $15/patient/month reimbursement math and what it means for behavioral health.TranscriptWe ran the transcript through an LLM to smooth it out. So it’s a rough approximation of the conversation (and in many cases significantly clearer than our rambling), but notably diverges from the word-by-word blows quite a bit.Brad Thorson [00:00]welcome to another edition of the information exchange sponsored by HDD Health. Can you believe our employer is sponsoring us? I am back from Vive this week and this is a very special edition inspired by Steve Posnack’s LinkedIn post, no Brendan Keeler today. So, boys, you got a lot of explaining to do for me today.Pryce [00:27]This is a facial hair only podcast today. Sorry, Brendan.Ryan Tucker [00:32]that’s right.Brad Thorson [00:34]I wanted to kick off just with I think some things I saw at Vive are gonna be Well, I missed all the big announcements because everybody does their press announcements around Vive, but if you’re at five You don’t have time to check your RSS reader or linkedin But you know I would say my takeaway it was great to see my good friends got to meet some new people But because we’re gonna be spending a lot of time talking about AI My concern is every booth was about agentic AI And the marketing copy is essentially the same from booth to booth to booth and so, know, one of the problems with using AI is it can pull everything towards a mean and I don’t know if anybody wants to hire me to help them run their conference. I probably don’t want that job but I’m a little concerned that our conferences are starting to look all the same and that’s gonna make it tough. If you gotta go on the conference circuit and there’s not a whole lot that changes, it’s not gonna pull you in. But maybe that’s the direction we’re heading because this week felt like it was all about outside of Vive, the AI,Ryan Tucker [01:55]get into the ARR again, which is very relevant to this, I just wanted to ask you, since you went to Vive, we also went to the ASTP annual meeting event very recently. I’m sure we’re sending folks to HIMS. Do you feel like it’s really about the content that’s shared?Brad Thorson [02:11]we are.Ryan Tucker [02:19]during that event. course, somewhat the conversation stems around there. Do you think it’s really just an excuse for folks who are in this industry to get together in person and have some conversation? Like that’s the most value I’ve gotten out of these things is most of us work remote nowadays. We don’t really come together outside of like our internal company workings, our working with our client. And to me, it’s like a good excuse for everybody to get together in a room, have some drinks, you know, go over what’s on their mind from that time period, not necessarily what the latest and greatest announcements that came out put on by the conference and the organizations attending.Brad Thorson [03:04]Yeah, I mean, think they’re depending on the type of conference. mean, the ASTP having an annual conference is a nice forcing function for for them to like put together everything that they want to communicate. Same way I feel like when we went to open an epic, that was like, you know, a singular organization had a North Star and that felt like a forcing function. Industry conferences. mean, God, I don’t know how many have been to at this point. I think that there are three types of cohesions that they create that are valuable. One is like, yeah, I can see you guys, I can see other people that I’ve worked at our previous companies or have partnered with and like that brings us closer, it reminds us that we could work together. That’s the positive thing that I get out of it. I think it’s also, you know, for executive level attendees, it’s an easy way to smash a lot of You know, I don’t think they are not having all high value meetings, but they can have more opportunity to have a high value meeting. And so I think that continues to bring in major exhibitors and they get an opportunity to talk about their But for people that are earlier in their career, it can be really tough. I think it’s important if you can go to one and you can talk to people. But so many people that are early in their career are tied to their booth and are really given like, you know, they only have one talk track or one thing to speak about. Whereas I feel like when we were seeing people at ASDP, one of the last things we talked about was interoperability or HCD. Like we spent so much more time just getting to know those people. I think there’s tons of value in that. It’s just tough to convince an employer to spend money to send their employees to just, you know, conversate.Pryce [05:01]I mean, I’m nodding my head because I think everything you all are saying is ringing true. I know, Arc. too, thanks to the HHS.Brad Thorson [05:04]The only three times we’ve seen each other, We’re at- Yep. Thank you, Steve.Pryce [05:14]And thanks. Yeah, thanks, Dr. Keene. And where was the other one? no, in New York. Yeah. I’ve been in DC in New York for the past two weeks. But it’s so high. I love Ryan’s take there. I think it was Ryan that just said, it’s hard to the value to your employer to send you to these things. Yeah. You know, like, I think about all the badges that you’re supposed to scan and all the leads that you’re supposed to show, turn into converted into revenue dollars. And I’m like, I’m like, yeah, that stinks because of course that’s how you have, that’s how you have to convince a business. But, the real value is meeting people. mean, I was just thinking like meeting, meeting competitors and realizing we’re actually pretty similar and we’re both humans and we both want to make people healthier. And then you kind of realize that maybe we should talk to each other more often and not be, you know, so close minded with our thought leadership or, or advice and things like, so getting a person to me is hugely valuable. Conveying the value of it is really hard. And I think I was nodding, particularly thinking about how we have a similarly challenging problem in healthcare, because today I’ve been thinking about how valuable it would be to sort of. force healthy habits upon people when they’re my age so they’re not as sick when they’re on Medicare but how the hell are you gonna convince someone to throw money at that because it’s way too long of a horizon to see value on and so instead we have a lot of the programs that we’re gonna talk about today which hopefully will still still do good things but that’s my quick quick take on conferences I love people I don’t really care about Vegas never been actually soBrad Thorson [07:07]I’ll drag you there someday.Pryce [07:09]I’ll sure you will.Brad Thorson [07:11]this is not going to be surprising to anybody who even wasn’t at Vive, but every every single software solution is talking about their agent to capabilities. Right. And I think I have not read Jack Dorsey’s announcement at block. I know of it. I saw the headline.Ryan Tucker [07:32]so speaking of trying to justify company spending and basically proving your worth as an employee, Jack Dorsey, former founder of Twitter, multi-entrepreneur, founded many companies, but Latus’s Blocks, he just tweeted out, X’d out, whatever we call it nowadays, last night, a message that he had sent to his employees for a massive reduction in force where basically they had about 10,000 employees, they’re moving down to 6,000. And he specifically spoke of the company is actually doing well. The company is growing, the company is profitable, but he is making this difficult decision to do this now versus later because he’s seeing that the agentic tools, the artificial intelligence tools that they’re using at the company are getting to a point where it’s not justifying the number of employees that they have and that they’ve gained enough efficiencies that actually to become even more profitable, grow even faster, they can reduce that force and don’t feel those folks are needed. Which if you’re looking at the business sector and you’re an investor, their stock actually went way up because you obviously when you reduce by 4,000 people, that’s a big saving on the payroll and they’re still doing well. So it looks great if you’re interested in the stock price. Not so great if you’re those 4,000 people who are now out of a job. And of course also not so great if you’re just in this general industry, more of the white collar workers that really anybody could be impacted in the near future here. And I’m wondering if this is just the start, if that’s the first domino to kind of fall here. And we’ve all heard it’s coming, it’s coming, it’s coming for years, but this is, I think the first I’ve seen that’s such a clear strike and clear saying, we are growing, we are profitable, we’re doing well, but we’re deciding to lay these folks off just because we don’t need them anymore.Pryce [09:52]Was there a sort of classification of, I guess, like the types of roles that were particularly impacted?Ryan Tucker [09:58]I don’t think it went into that detail. We could definitely, I don’t know if we have show notes, but we can link to the post. But basically I think it’s more on the development side. I think the coding tools that they’re using nowadays, I suspect also more of the entry level or more junior developers would be impacted versus the ones who are reviewing the code that’s output by these agents. But yes, it’s a little bit scary.Pryce [10:12]sense. Mmm. didn’t Brendan just write about how he somehow found out that like Epic is really leaning into Claude use on campus. And I mean, it’s no doubt these tools are useful. These the Cetrine research article, which I will admit I haven’t read in full, but I’ve heard a ton about. was also news this week because it wrote about this possible future in 2028 where everything, the economy’s collapsing because everyone’s out of a job and then if everyone’s out of a job, then they can’t prop up the mortgage market. it’s, from what I have heard, you know, there are some cracks in its argument and it’s very sort of hypothetical and almost like sci-fi sounding, but it does make me think over and over again, especially as we, As we consult about product strategy and how to use AI, how to make people healthier, et cetera, et You know, I’m really starting to try to map out on a whiteboard, like where do agents remove friction or support humans versus replace them and where can they not go at all? Like, I think Brendan and I talked about the last time I was on like an agent can’t. recreating a trusted network, right? Like you can’t recreate without humans getting on and saying, I have a house and here’s my verified identity and come and stay here if you’re another human, things like that. You can’t just create sort of two-sided economies, but you can complete so many tasks. So I’m... Very impressed with AI, still a skeptic as compared to what I would say is like market frenzy. It’s gonna disrupt everything just the same way that the internet did in my lifetime.Brad Thorson [12:30]I I vibe code on the weekend. I love it. Nothing that I make is should be usable, should be used by anybody else. But I’ve noticed that the tools have gotten a lot better. I think that like it’s terrible that these people are losing their jobs, especially in a company that is increasing their revenues. But we have seen it. I can’t make a proclamation that history will repeat itself. This will be a diffusion of people that were in some ways forced to use AI energetic tools that will likely be seeking employment at like not Amazon or Google, but finding their ways into enterprises that are not necessarily AI native or enterprises that will benefit from really talented product managers or engineers and everything is tech now. But I think the diffusion of that talent, the diffusion of that knowledge takes time and I just, see so many new applications getting built just the way that like, you know, I want to track my cat’s behavior. I can spin up a little application in hours that kind of works. we will likely see that same pattern across industries, which is part of why people talk about SAS-Mageddon. But we’re just allowing people, there’s a BCG, I think it’s BCG study that AI implementation is like 10 % the model’s actual capabilities, 20 % the actual, like, what are we introducing it into, and 70 % like, operationalizing, teaching people how to use it. And we refer to that as workflow. And so I think one of the reasons that we’ll see a ton of these applets or these tiny apps is because software, SaaS software has not been responding to workflow. has been enforcing workflow. where that doesn’t work, where two softwares don’t align, we throw a human in between them and force them to make it work or we define some sort of ETL application. I don’t know. I think we’re just gonna see an explosion of custom software that allows us to explore a million different versions of workflow. And then over time, we’ll start narrowing the use cases that are, or tightening around workflows that are actually effective.Pryce [15:19]You know, it’s funny, you saying all that almost made me think like, yes, the speed, the velocity at which AI is improving and like the use cases that you could leverage it for is evolving so fast. But there are other bottlenecks that are going to prevent like an economic collapse where we all lose our jobs overnight. For example, humans are slow to learn or, gosh, you said something else that made me think like. completely different reason that it would be just hard to introduce. There’s going to be a concentration gradient, if you will, of AI users and non-AI users, and it will slowly diffuse into, we all use AI in some form or fashion. I don’t think it’ll happen quickly enough to pull the rug out from half of the economy, but maybe that’s just what I have to think to self-preserve and not quit my job and go start a farm right now. don’t have enough land forRyan Tucker [16:20]Maybe we could all buy a farm together if this just doesn’t work out. But yeah, I think, and I obviously don’t have the expertise on the full macroeconomic side of this.Pryce [16:22]The HDD part.Ryan Tucker [16:34]I’ve actually been reading lately, there’s some kind of thought that you almost hope it’s fast and not a slow burn because that might force some function to help the massive people that are, you know, unemployed all of a sudden talk about UBI, things like that. Versus if it’s more of a slow burn, there’s less likely of a response to help those folks. So that’s one interesting thing. But I think tying it more back to health care, and what we do know about. I think it’s been mainly AI recently like everything else, I think IPryce [17:07]Is this a healthcare podcast? I’m kidding. huh. huh.Ryan Tucker [17:16]a major pro and major con that the AI tools, what Brad was talking about, like being able to customize to your own workflow and also things like the AI scribes where they’re actually just totally getting out of the way of the workflow. So when we get really great scribes where now you can go back to how it used to be with a provider where you’re just having a face-to-face conversation and the AI is still doing all the charting it needs to do, still doing all the tasks that the provider actually doesn’t want to do and making the experience better for both the provider and patient, those are fantastic use cases for AI. And that’s what I’m very excited about the usage for. The negative side. I think is what we’ve talked about with the potential to actually not replace and make the workflow better, but replace the actual human being that’s doing the job. And then I also fear the jobs we’re replacing, what we talked about with blocks where they’re replacing some of their engineers. You know, when you’re moving fast and breaking things in FinTech, the the major catastrophe that could happen there is you get something wrong and there’s a massive decrease in money and how it’s tracked and things like that. But in healthcare, and we talked about Epic using quad more in their workflows, and I think there’s some positive to that, right? To get stuff out there faster, get some of these beneficial software. Yes, internal code development, yeah. And like I said, to make beneficial software,Pryce [18:49]Internal, just to be clear, this is like they’re using cloud internally. Yeah. Yeah. Yeah.Ryan Tucker [18:59]That’s great. Make that faster, get those out to the users. But the downside, FinTech dollars are affected, healthcare lives are affected. And when we were at Epic, there was a very, very strict system around code development. And we had what were called patient safety escalations, which were things that have gone wrong with code and things that have actually been a risk to the patient potentially. And there were huge procedures around that. And it was taken very seriously. The software is not sexy, but it was incredibly safe and had to be because we were in healthcare. And that’s where I start to worry as we start to replace people and the cogs in the wheel. Healthcare, especially, I think there’s a risk when there’s not as much review, there’s not as much personal code written. And of course, also maybe in the end, 50 years from now, it’s actually much less of a risk. like self-driving cars, eventually they may be safer than a human doing it. Eventually the code written may be safer than a human doing it, but there’s, I think, a threshold that still needs to be met. And right now I trust a senior engineer versus collaborating the code at this point.Pryce [19:59]Yeah, exactly. And we have no idea where that line is because we haven’t really like, we haven’t had enough at bats to see like when things go haywire. Well, speaking of, I guess, patient safety and regulating AI, Ryan, I think you had linked out something about, remind me what it was, are we moving to the FDA now with regards to topics? Because they have released something about.Ryan Tucker [20:23]Absolutely.Brad Thorson [20:45]Harrison AI just let us let us throw some AI at patient problems patient populations looking goPryce [20:46]Yeah. What was the story there?Ryan Tucker [20:54]Yeah, and so basically, so with this administration, we saw this in HDI-5 as well. Reduction in regulation. And it’s interesting because that’s not everywhere, right? But at least when we’re talking about health care, at least some of the latest things to come out, HDI-5, reducing the scope of the regulation in order to hopefully, the thought is, make it more efficient, get things out there quicker. remove some of the red tape, literally, like the red tape that existed. Same thing over here, FDA, they are implementing a path, and I will admit, I didn’t really delve into exactly what this path looks like, but a path that if you already have a solution that you’ve taken through the regulatory process with the FDA, and you have, it’s a, AI device, Harrison AI and I think some others have been looking for exemptions from that pre-market review process, which is very strict, especially around medication development, procedure development, things like that with the FDA, more traditional things we’ve seen in the past. Say for AI devices, let’s skip that. And I think it makes sense in some ways because it’s not like bringing a drug to market. But at the end of the day, it’s usually a patient-facing thing if you’re going through the FDA. And I think there needs to be some guardrails still. And I think there is nuance there. But just interesting them trying to of screw around certain things. We’re trying to make things less regulated for the sake of speed, which I think is good in some ways. But also we need some guardrails. We need some transparency, those types of things. You know, we don’t have to spend too much time on that fact, if you have any kind of thoughts on that. I think, you know, talking about the FDA, I know Price, you want to talk about making America healthy again as well.Pryce [22:49]No. Yeah, well, so let me touch on that real quick. I don’t know very much about the FDA. I would probably loathe to work at the FDA because it feels like a no-win situation. Either you’re like regulating things, everything in the market is safe, and we can all thank the government for paternalism. And really we wouldn’t thank them. We would like not think about it or maybe we would complain about how hard it is to get through regulation or how slow America is to adopt whatever, whatever. The flip side of that coin is like, we do have a worker shortage in healthcare. We have an overabundance of data that’s so underutilized. It’s incredibly difficult to access, not necessarily from a technical perspective, but definitely from like a business relationship perspective or privacy and security perspective or even just like a governance and what does this data actually mean when I get it perspective. And so this current administration keeps saying, yeah, well, let’s throw technology at it. Let’s let the private market figure it out. Let’s take away certification criteria as proposed in HTI-5. allow new payment models to exist like CMS or the CNMI Innovation Center’s access model, where they’re specifically saying, look, reimbursement rates are pretty darn low for what we’re asking for, which is to maybe lower a patient’s blood pressure. But they’re like, we don’t expect a human to do it. For the most part, we expect you to keep an eye on it. We expect you to leverage tools and AI to actually help your patient change their behavior or coach your patient to becoming healthier. And so, yeah, I’m a little bit conflicted there. I don’t know nearly enough about clearance and how software or hardware goes through the FTA to have a strong opinion, but I am like, it’s not great in either direction. If we never get good tools in healthcare or if we get them tomorrow, like there are risks to both. I don’t know, Brad, do you have any other thoughts on that front?Ryan Tucker [25:15]Yeah. Price, think just for quick price with what you said before with, you know, these paths and how strict or how open we should be. I think we need more at bats to really determine, like you said, and it, and it would be interesting to compare this beyond AI, like any crazy new innovations that have come through these processes, what that looked like back then. That’d a good comparison for a blog post maybe written by the guy that’s not here right now. But yeah, it just seems very early. And Brad, please, your take would be great.Brad Thorson [25:56]Well, we actually have an expert in the company on the FDA approval process in Veronica. we’ll have to get her on here. Time would not have worked to have tried to loop her inRyan Tucker [26:03]That’s right.Brad Thorson [26:08]The introduction of AI, when I hear AI, think probabilistic algorithms, and I think they have like great use cases. The usage of them to like generate text or to be able to take a patient query that says like, what are these lab results mean? And be able to turn that into something that a computer can execute upon, get something from the computer and turn it into language is great. But part of me feels like if the translation is not too a deterministic model, I didn’t read enough about the Harrison AI thing. So somebody from Harrison should reach out and get us updated. If somebody is bringing a probabilistic, not just like text generation, but probabilistic recommendations. I think that is something that should absolutely go through a strict process. But if the ask is like, we have a new interface that is using something that hallucinates sometimes, like, I don’t know. Well, there was some stat that like 4 % of queries in Claude’s free model include, I need to go find the actual number, but it’s an insanely high number of queries into a free chat bot asking for health information, oftentimes including their own health information and not using the HealthX extension or the B-Well extension, ChatGPT, like people are just grabbing a PDF and throwing it in there. It feels just because Pandora’s box is open or the cat is out of the bag doesn’t mean that the FDA or other regulatory bodies should completely cave. But I do think that there needs to be some response to, know, if it takes companies forever to get through the FDA process, they’re just going to find ways to introduce products outside of that.Pryce [28:07]Hmm. I love what you said. Just to like frame that within a sort of thought process that we have a lot within HTT. We run a workshop called the why what how workshop. And we might have mentioned this on the podcast before, but lots of folks come to us and say, here’s how I want to integrate. Like, is that right? Or I want it to be, it has to use fire because that’s modern. And That’s the wrong way to think about it, right? Before you think about how you do something, you sort of need to think about what you’re doing. Maybe before you think about what you’re doing, you need to think about why would your product exist, right? And if AI is the tool, it’s just a hammer, right? Like, yeah, let it drive nails. Let it, like if this request is to allow a device to sort of say, your blood sugar is low. Go get some insulin and here’s how much we think you should do. Double check or, or I don’t know, like here’s what you should eat tomorrow as recommended by the NAI dietitian. That feels very different than like we’ve prescribed a medication for you. Go pick it up. Right. and one of those needs a hammer right now. And if you took a hammer and then said, like, I’m going to knock down a wall with this tiny hammer, it’s going to be inefficient. It’s going to be dangerous. You’re going to. yourself swinging it too hard. I am like, we have to treat it like a tool. And right now it’s good at certain things. I think it’s easy for us to imagine that it’s good at everything because it is good at such a broad swath of things. But it was designed for certain things right now. We need to get used to what those things are and then implement it responsibly, I think. FDA, please, please do well. I don’t have a U like I said, but please do well. Can I jump to the other news from DC this week? I think it was this week. I guess Trump signed, IBrad Thorson [30:28]I don’t know what time is, so I trust you.Pryce [30:33]I think Trump signed some order that was meant to increase the production of maybe phosphorus in the US or something to that effect. Ultimately, the headline was that the US needs more glyphosate, which is also commonly known by its brand name Roundup from Monsanto. And this is an interesting. take to have in this administration, especially because of the installation of RFK and the Maha movement and all the people who probably think glyphosate is worse than the devil are now sort of coming to terms with, do they agree with this administration? Do they not agree with this administration? And then RFK sort of cosigned on what Trump had done and said, it’s going to help us feed ourselves, et cetera, et cetera. I just want to say I barely studied things like this in college and I now I’m a gardener, almost exclusively organic. I do use sulfentrazone on my nutsedge.Brad Thorson [31:48]This is why we have to have the podcast because we don’t go to conferences together enough and I didn’t know you were a gardener.Pryce [31:56]I love to garden. I love to garden. But here’s the thing. I actually don’t think, I mean, maybe glyphosate is a carcinogen. Maybe it’s actually proved to be a carcinogen. I don’t think it’s like, you know, actually there’s lead at my house right now. You know, we’re working through a lead issue and nobody’s going to die. There’s lead everywhere and we need to get lots of it in our system. And the same thing could be true for herbicides or, know, pesticides and things like glyphosate. But what I do think glyphosate allows is industrial farming, like huge volumes of farm on the same volume of land. And inherently, that’s going to create this sort of imbalance in the nutrients that actually make its way into food, right? So if you can just crush your corn crop every year because there are no weeds. Oh, that’s great. You’re growing corn, I guess. But every year it’s growing in soil that’s less and less nutrient dense. And then you pump fake nutrients into the soil, and then we pump more fake food. And so all this to say, this is sort going back to my point where I’m like, yes, we need to keep 65 and older healthy. We should also be thinking about how to make 32-year-olds healthy for the next 60 years, or longer, because who the hell knows what’s going to happen with our lifespan. And that probably starts with not feeding the country just empty calories that come from soy and corn and super industrial operations in farming. If there are any farmers listening to this, I apologize. I know nothing about how hard it is. But it’s just a bummer to me. I think it seems straightforward that there should be more farms that are smaller and organic and creating nutrient-dense foods. I don’t know if this was a corporate decision or a corporate. I mean, surely it was in the interest of some corporations. But I don’t know why the decision was made or if there’s actually a food insecurity crisis or a glyphosate crisis. But it’s hard for me to see the HHS put out models that are like, we have to prevent people from going to the hospital. Let’s catch it early. Let’s lower their blood pressure. We’ll incentivize you to lower their blood pressure. And then on the other hand, they’re like, don’t stop buying Doritos because those are super cheap and they’re SNAP eligible. And I love Doritos, but. I shouldn’t have almost free access to them because they’re subsidized by glyphosate production. So curious what you all think about, I guess, nutrition and health and the long-term plan here.Brad Thorson [35:03]What do I think about nutrition and health?Pryce [35:06]Took us way down a gardening route. Ellen Brown would love this conversation.Brad Thorson [35:10]That was a very like, well I don’t know what I think Price. I think I wanted GLP-1 so that I don’t have to worry about that. when you were talking, I actually had a, I want the pharmaceutical industry to cure my desire to eat.Pryce [35:26]Yeah, you know what? And you can get that with AI now probably. You don’t even need to talk to your doctor.Brad Thorson [35:32]man, perfect. I have, this is like, maybe this is not a good corollary, but since you’re both alumni of Verona, Wisconsin, do you think that we need a little bespoke EHRs everywhere? And if you guys want me to do a little galaxy brain thing like I did last week, doesn’t AI open up the opportunity for us to have all sorts of little EHRs running everywhere? To be clear, I don’t think we need 7,000 EHRs.Ryan Tucker [36:10]We need a Brad. This is a we need an official Brad Galaxy Brains segment of the not a podcast, I think, at this point. Well, you know, I think there’s more positives to the thousands of smaller farms than there are to the thousands of smaller HR’s. There’s a lot of different things at play on both of those sides. Working on Interop though, thousands of smaller EHRs, sounds like good job security because there’s so many headaches. Yeah. But does not sound good in terms of make sure everybody’s connected and getting, meeting the goals we’d actually like to hit. So that’s, that was way, way left field from prices, nutrition and food system.Pryce [36:49]for us. Yeah.Brad Thorson [36:50]Certainly for all the CN.Ryan Tucker [37:10]But Tidewell, I think.Pryce [37:13]I have a take. Finish your thought, Ryan.Ryan Tucker [37:17]just going to say I am not the best on the food and nutrition side. Personally, I love to be better with that. My understanding is that more localized food production would definitely be a net positive, but the incentive structure that’s been set up with subsidization and also just inherently how mass production in a company that can handle shipping anywhere versus a company that has trouble getting it 10 miles down the road. It’s way easier for me as somebody who has a decent income and is relatively economically stable to say, I want to buy local and great food from around here because it’s generally much more expensive and something that is not as accessible to somebody on a lower economic spectrum. That’s really...Pryce [38:06]Hmm. 100%.Ryan Tucker [38:14]I’m not an expert in this field at all. yeah, Price, I’d love to hear what thought sparked there.Pryce [38:18]Well, there’s a million variables, which is why I’m not claiming to have a solution to the problem. just sort of perpetually bummed with some of the changing opinions in Washington. The thought that had sparkedRyan Tucker [38:35]And we’re all aspiring farmers if we lose our jobs. So it’s a good, it’s very relevant.Pryce [38:39]Exactly, Yeah, sassmageddon will lead to the actually to the organic farm revolution. I was going to say, do I want a million little EHRs? Probably not. I think I would like more that are competitive with each other because over the past few weeks, know, Kat McDavitt, Lisa Berry, Brendan have been getting into my, you know,Ryan Tucker [38:48]That actually doesn’t sound so bad.Pryce [39:09]brain and talking about how American it is to have all this competition and how competition drives down prices. And while that might not be great for Epic, it would be great for providers who might be paying less for their SaaS or for their EHR platform or whatever. And then I stopped and realized I am always making the argument that EHRs are not what’s making healthcare so expensive right now. I hate when someone points to Epic and says, that’s why healthcare is broken or that’s why healthcare is so expensive because that is a blip on the radar of healthcare costs. I would love if there were a million different provider groups instead of one provider group that had a million providers and there’s only one that does. So you can go look it up if you want to figure it out. I think that would be way better for healthcare and for healthcare costs. And I think the same thing is true of payers. So yeah, obviously we’ve got the FTC and we’ve got info blocking and the OIG working on creating this more competitive, price friendly healthcare ecosystem, which is not easy to do. But that’s my, that’s my little take there. I would rather there be a million provider groups and a million payers than a million EHRs. Call me crazy. Any other thoughts that you all have?Ryan Tucker [40:40]I agree with that.Pryce [40:44]from the week, weekendRyan Tucker [40:46]just spurring on.Brad Thorson [40:46]Hopefully sleeping, I don’tRyan Tucker [40:50]I think all of us can use some more more sleep. Just spurring on the American comment and the competitiveness. So Brad’s got his galaxy brain. If if Ryan could put on his socialist brain, which might shock everyone real quickly here, generally not.Pryce [40:51]Love it. yeah.Ryan Tucker [41:14]Socialist overall, I would say. But when it comes to healthcare, I think competition is generally good in a lot of different segments when business is the goal. And healthcare, when treatment and the health of the people is your goal, I don’t think those economic models are necessarily the best way to go. There’s a lot of inherent disincentives that are set up there. We see that with the pharmaceutical industry and the pricing that they set, the payer industry, health insurance and how expensive that can be. And so I think if health and wellness of your population is the goal, it seems like there’s other countries who are less about 100 % capitalism that have done a little bit better job there. That being said, we’re in the system as it’s set now. And I do agree with you Price that If we are going to take a capitalist take at healthcare and health tech, then it does help to have that competition between those EHRs, between pharmaceutical industry companies, between health insurance companies. Because ultimately, if there’s choice among amongst those companies that are provided versus a government provision or a single payer or something like that, that is purposely driven to as low cost as possible. and we’re in this capitalistic side, then yes, we need that competition to help the consumers.Pryce [42:46]Mmm. I don’t want to take us down another rabbit hole, but I love that take. I’ve heard smarter people than I say when they all get in the room and talk about how they would fix healthcare if they could do anything. They’re it’s like single payer all the way. And I was thinking about today how, you know, I keep on harping on this. Oh, I’m 32 right now. please help me make good decisions to stay healthy so that I don’t become a cost burden to society. Well, that’s really hard to prove an economic model on when I change insurance companies and the risk bearing entity or the profiting entity of carrying my risk changes every time I change employers or perhaps changes every year if I’m on the marketplace. so, if imagine if there was one benevolent well-oiled machine that was caring for all of us and thinking on the, you know, 100-year time horizon. It would be a lot easier to solve for some of these problems.Ryan Tucker [43:56]Definitely. Yeah, the tide of employment in general is interesting, right? Like if you lose employment and you have a serious health concern, you’re screwed.Ryan Tucker [44:07]at this point, right? Like, yes, we have some Medicare options and. That’s a problem in itself, but then you’re right. The payers, the health insurance companies will sell to these employers with a one-year contract. And so when you have your open period to switch your benefits around, that is because that health insurance is sold and said, I’m just bearing the risk and modeling the financials off of just this year. And so their actuaries are only looking at that time horizon. Versus if they pick that up and extended that beyond, they’d be much more into that preventative care side, the value-based care side, that kind of thing, as you mentioned.Pryce [44:52]Well, I probably won’t solve that this weekend. I will probably watch the... Yeah, I’ll get a clod of Ico. I will probably watch the Spurs who are 11-0 in their last 11 games. so Vosper is Va. And then we’re going to go... My son is at the zoo today for a field trip. And we’re going to go to the zoo tomorrow as a family.Brad Thorson [45:20]Does that mean he’s a tour guide?Pryce [45:21]Double Zoo. He’ll be the tour guide. It’s for his birthday. He’s turning eight next week. So yeah, we’ve got to get him started right now on the right program, the right access model. Get him a bio-wearable and keep his cholesterol down.Ryan Tucker [45:27]Happy birthday.Brad Thorson [45:40]Guys, the Axis model reimbursement rates, I haven’t written about it, man, I... 180 bucks a year for behavioral health.Pryce [45:53]Is it a month or a year?Brad Thorson [45:55]Now a year. Like, do you know how big of a patient population you already have to have? Not that we need to, listen, we gotta wrap this up, we all gotta go do our day jobs, but the cost of implementing a program, a patient monitoring program, like you have to make a significant investment at the beginning of that program, and we’re talking about $15 a month, is that right? I don’t do public math, so I shouldn’t have tried to do that calculation.Pryce [45:58]That’s tough. Yeah.Brad Thorson [46:26]Like it’s not worth it to make that investment for a lot of behavioral health providers if they don’t already have an enormous population. so like, you know, good for the vendors that are already enormous, they can get another $15 per patient per month. But if anything, the access model is just going to lead to greater consolidation amongst provider groups.Pryce [46:38]Well, often. Yeah, yeah.Brad Thorson [46:53]We can’t swing all the way back around to competition, capitalism, socialism, but let me tell you, I got thoughts. They’re rooted in transaction cost economics and I do not like the reimbursement rates tied to the act. was so excited about the access program and I’ve just been out on it guys. I’m upset. So.Pryce [47:12]We’ll have to talk to Abe Sutton from the CMMI and get educated on how this could work so we can coach folks to make people healthier.Brad Thorson [47:22]I’m just a random guy on the outside.Pryce [47:26]There we all.Brad Thorson [47:27]That’s what I’m going to do this weekend. I’m going to write about reimbursement rates in the access model.Pryce [47:34]I’ll root you on from the zoo. All right. Well, thanks, fellas. I’ve got to run.Ryan Tucker [47:34]I look forward to reading that. Well, the breadth of this podcast was incredible. I think my favorite quote might be, I don’t do public math. I’m definitely going to steal that in the future. But yeah, it was great. Catch it up with you guys. Agreed. Thank you all. Get full access to Health API Guy at healthapiguy.substack.com/subscribe