The Best Conversations Happen At Happy Hour, Welcome to Ours!
Feb. 19, 2026

AI Won’t Save Your Clinic (But This Might)

AI Won’t Save Your Clinic (But This Might)
AI Won’t Save Your Clinic (But This Might)
PT Pintcast
AI Won’t Save Your Clinic (But This Might)

AI Is Not a Silver Bullet

Recorded live at APTA CSM, Todd Norwood joins the show to talk about AI, digital health, and why physical therapy clinics need to fix their data before chasing the next shiny tech solution.


What We Cover:

  • The difference between good data and bad data in PT
  • Why “ish” measurements don’t scale in an AI world
  • How to evaluate AI scribes and clinic tools
  • Imposter syndrome in leadership and tech transitions
  • How PT skills translate into digital health roles
  • Using AI to assess your resume against job descriptions
  • Why investing in yourself beats any market investment


Key Takeaway

Good data is foundational to making the most of AI and digital innovation in physical therapy.


Guest

Todd Norwood

PT in Digital Health

<p>This episode is part of the <a href="/physical-therapy-business-podcast">Physical Therapy Business Podcast</a>, where physical therapists and clinic owners learn how to get more patients, improve operations, and grow their clinic without relying only on referrals.</p>

<p>Looking for more episodes? Explore the full <a href="/physical-therapy-business-podcast">Physical Therapy Business Podcast</a> for real-world strategies on physical therapy marketing, patient acquisition, and clinic growth.</p>

Transcript

0:01 how tall are you six four yeah i'm not i'm not that tall although my brother's six three i'm five nine and we look vastly different in terms of height which is what you get on a live stream so we're live all right uh let's do this live from apta csm technology isn't coming to health care people it's already here we're doing a live stream from a conference of course technology is already here Our next guest lives at the intersection of physical therapy, data, and digital health, helping clinicians use technology without losing the human side of care. Welcome back to the show, Todd Norwood. Todd, welcome back to the show. Thanks for having me. Nice to meet you in person and be here together in person. We're also matching our shirts. We did seafoam green it up.

0:42 You were from California. Are you still in California? I'm from California. Still in California. I live a mile from where I grew up. So you were on the show a few weeks ago. We did like a CSM prep episode because you're speaking tomorrow. Yep. There will be candy tomorrow morning. That was my first question. All right. If you're going to do a presentation on or near Valentine's Day, you have to have a little bit of the bribery. Give people who might be watching the live stream here, because we have had people walking around listening. It's still crazy. Technology. It's like watching the game while you're at the game. I don't know. We do it. Just give people just a quick reminder of what you're doing tomorrow. Yeah. We're going to be talking about how your skills as a PT translate readily to becoming a leader and how you can leverage those skills to lead and have a larger impact.

1:27 You did that, like you said, you were in, like, give like your sort of like, how'd you go from a physical therapist to what you do now? The origin story, right? The superhero origin story. Right, so I was in a clinic, I was practicing orthopedic outpatient physical therapy. I met a patient, I work in Silicon Valley, so I met a patient who was in technology. You're bumping into tech people. That's right, you can't throw a stone and hit one, right? So I was out there, I met a patient, and we started talking about technology. He was quizzing me one night, he'd come in during his visit, like, hey, how'd you use technology differently? I was up for it. I love technology. So we started talking and that was Phizera ultimately. And then we built that up, grew a nationwide network of PTs doing telehealth before COVID, which is crazy when I think about it.

2:10 And we got acquired by Omada back in twenty twenty and I'm still with Omada, still doing the same thing and and a few more things with what Omada does in the chronic disease space. I feel like every like third booth here now has some sort of technology component in it. But I want to talk to something more basic and emotion. When you were now leaving the clinic and going to this tech thing, did you feel impostery or like, hey, maybe I don't, do I belong here? Like, did you feel that at all? Oh my goodness, yes. And then, all right, so how'd you get over it? Or did you? How'd you get past it? Different question. It's so interesting because I think there's still moments today where I'll be in a room or I'll be in a venue and it's like, oh, do I belong here?

2:51 Do I have the right to be here? And I think the thing about imposter syndrome is it comes and goes throughout your career. You have moments where you're like, yes, I earn my right to be here, right? I am the expert in the room and I own that. And sometimes as you grow, right, as you develop as a professional, you're going to be in rooms where you may question things Do I belong here? And I think for me, it's been an opportunity to look back on past experience. And I had a lot of experience as an athlete growing up and going through different ranks. And each time you got to that new level, it's like, oh, I was pretty good, right? I was a big fish in a small pond before, but this is a much bigger pond. And now am I a small fish here?

3:31 And so I had that background. And I think that helped me a lot going through this experience. You're talking about that sort of at a macro level. You're describing a day for me. Like, nine a.m., I am a badass. By eleven, I'm like, I don't know what I'm doing. An hour later, I'm pretty okay. By three o'clock, I'm all right again. And hearing people say that makes me at least feel like I'm not the only crazy guy in the room. No, I mean, I think that's true. That can happen within a day, for sure. For me as well. Rollercoaster. Yeah, I'll have those days where that first meeting, like, I am the man. I am the guy. And then later in the day... You get knocked out a few pegs by yourself. Am I the guy? Am I the guy? Should I even be on the team anymore? Let's get into some tech then.

4:13 Why is good data foundational for modern healthcare? But before you answer that, explain to me what the difference is between good and bad data is. You gotta be able to trust your data. So for me, that's the difference between good and bad. Is it reliable? So I think, let's take something in PT that's common. Range of motion. We measure range of motion. many of us eyeball it and then what's the plus or minus even if we're doing it right and if we do with a goniometer and we write it down do we do we write if we did a goniometer do we write down to the degree or do we do round to five or ten this is where you go from having potentially good and precise data right having it to the degree or if you were in a biomechanics lab right it's to the degree it's the tenth of the degree even two in the clinic It's one-thirty-five, I think.

4:57 Ish? Maybe. You can't really put ish into a spreadsheet. No, and AI is not going to like ish. Yeah, that's true. If AI is processing the data, it's not going to do okay with ish. It needs to be precise. Well, it'll handle the ish and you'll get garbage on the output. Garbage in, garbage out, right? That's right. That's the idea. That's right. So why is that so important? So if we're moving into this world, maybe as PTs we're moving into this world, the world's moved there already with AI and big data. If the data's not accurate, then the interpretation and the outputs are not gonna be accurate. And I think it's interesting from a patient and a care perspective is like either you get an ineffective intervention, an ineffective recommendation, or potentially a harmful recommendation if you don't have sound foundational data.

5:46 And that's my fear. We have a lot of data in PT. one we may not all document uniformly and so that's a whole data cleansing issue we got to make sure that it's all uniform and then do we all agree that one thirty five is one thirty five or is your you know like your one thirty five is one thirty two mine's one forty the next person is one thirty That makes it challenging. It makes it really challenging. Especially when you get to start, when you start to use, when things get larger, when, you know, more and more clinicians and larger and larger. That's right. So can you imagine if you took data from clinician, let's go five, fifty outpatient clinics in fifty different states with fifty different clinicians.

6:29 It's a lot. It's a lot of variability, potentially. Yeah. So what would you say to people? There's still a few people who, not brag, but say, I have not touched artificial intelligence yet. What would you say to that person? This isn't going, it's not going to infect your brain. It's not going to change how you do things. What would you say to someone who's still hesitant about leaning into, maybe not even just AI, but technology in a clinical setting? Well, I mean, do you have an iPhone? Because you've touched... Correct. If you have a mobile device, a smart device now, you've interacted with AI. It may not be the chat GPT or the perplexity or clot or the newer models that are out there, but you've certainly interacted with AI in some way, shape, or form.

7:14 So I think there's that. I think there's this... threat around AI. I don't know, there's this belief that we're gonna get the Terminator, and AI's gonna rule the world. Okay, that's scary. I think that's a low probability, at least in the near term, of that happening. And then there's the, I think it feels comfortable AI, like our phones. We use these devices. Whether or not you've used them in the clinic yet, I don't know if we're there where I'm thinking, gosh, you're really behind if you haven't. I think we have some progress. I think all the adopters are there, but the middle of the bell curve isn't. Sure. Maybe. And I think that's maybe where we were with telehealth, right? Like COVID was a fire starter for that.

7:56 It was like pouring fuel on the fire on telehealth. So I'm not concerned if you're like, I haven't used it. I'm concerned if you're saying... I'm not ever interested in using it. I won't. I won't use this technology. I think the clinic owners, of course, were the first people who were approached by everybody who was selling a scribe two, three years ago. And the smart ones were being good clinicians and saying, like, show me, show me. Don't give me the, I don't want the brochure. I want you to show me on your screen right now. And then show me ten people who've done it already. And if you were selling that three years ago, you're like, I don't have ten people. You're the first people I'm calling. But now we're sort of past that. Yep. And I think as a business owner, what I want to see is, does it work?

8:40 So can I do a mock case right now? Don't show me the fancy demo. That's cool. We can polish the demo. But let's do a mock case right now. And let me take your subjective and your history on a condition. And let's see what happens. And does it do what I expect it to do? Right. I was working for a small tech company. as sort of a consultant for a little while, and I was gonna shock you. When you go to conferences, all of a sudden you wind up in an Uber with three people you don't know and two people you do. And someone was like, who are you and what do you do? I'm like, I'm doing this thing with a tech company. And they're like, well, you should give me a demo. And I was like, great, let's do it right now. And they're like, well, there's a lot of loud noise in the Uber. I go, great, there's a lot of loud noise in your clinic.

9:20 If I can't do it at the back of the Uber with my knees in my chest, and we did. We did a subjective evaluation right there, and I said, I even threw in some curveballs. I was like, what was your last movie you saw or last concert or like how old are your kids? And he was like, why are you asking me this? And when I showed him the output, he was like, I said, where's the thing about the Garth Brooks concert and your nine year old in the movie? And he's like, oh, it's not there. I go, right. That's right. Because it has to filter that out. It's not clinically relevant. Correct. And it understood that. But their clinic owners are right for showing. Show me. I want you to be. I need to beat on this thing. Yep, absolutely. I think that's true of any technology. Any new technology that you're going to adopt in your clinic, whether it's AI based or anything else, you've got to test it. You've got to push through its paces. Absolutely.

10:01 Yeah. What are some of the things that get people past that skeptical phase? I mean, this would be the million dollar question. Just how many? Because I'm going to get past that as fast as possible if I'm selling some of that stuff. For sure. I think you got to see it work. That's a big piece of it. And you've got to see it work in that case where you as a skeptic don't believe it's going to work. Right. So if you're like, oh, well, I can't do it in a noisy clinic. And then you show me in the middle of the exhibit hall. It's noisy. I'm like, OK. All right. Or, you know, it's not going to filter out if we have chit chat about the baseball game last week. Right. Oh, okay. Although I do think it's interesting as a clinician, some of those things I would always jot down somewhere, right?

10:43 Like, oh, you have a nine-year-old daughter that likes softball. Ah, good point. Because those are good social interaction pieces for us to build that relationship with the patient. And so it's an interesting balance, right? If you're letting the AI do too much and it's filtering out too much of that. Yeah, yeah. Last thing I'll ask, roles. You stepped into a technology role, maybe not on purpose, but one sort of appeared. What would you tell, you said earlier, you probably as a clinician have the skills already. If someone, and I'm in Facebook groups, for better or for worse, the comments section of Facebook groups are interesting. But people are asking those questions, like, hey, I'm thinking about, or, and it is nice to see people who've stepped into those roles say, I will have a phone call with you.

11:26 I will show you what I did, I don't know if that's gonna be relevant for you, but what would your advice or suggestion be for someone who's like, I do like this, this is fast-paced, and I can see working in technology, instead of helping one person at a time, I could help a thousand people at a time. What would your step one or two and a half be? I think you have to understand what the role is that you may want to pursue and try to understand what that looks like. Because I have a lot of folks reach out, it's like, oh, can I do customer success? Yeah, you could. Do you understand what customer success means? What's entailed in the day to day? Sounds like customer service to me. Well, yes and. So do your research, I think would be my first recommendation.

12:06 Understand what it is that you're looking at. mine was sort of a happened by chance and i had the right interaction that person was probably sizing you up though that's right that's right and i think it's it's so interesting that's the thing that in retrospect i recognized at first i was like oh gee i had i had ready i had great fortune to meet this person and like yes it's true there's some luck involved right but he wouldn't have called me back had i not answered the questions in the right way and put the pieces together so so Yes, luck and. You have to have the skill and the knowledge. Ask yourself, what are your competencies from a technical standpoint? What do you know? I did a prompt once yesterday.

12:46 Okay, that's good. Keep digging. Good start. Also, AI is a fantastic tool. I do it all the time now. But ask it, right? I'm a physical therapist with five years of clinical experience in outpatient orthopedics. I'm looking to transition into a technology role in customer success. What skills should I develop? What competencies may I already have? I go one step ahead. I download my resume from LinkedIn. I download the job description and I go, match me, give me the five main skills a great this could have and rank me from zero to a hundred on those. And then help me understand if I should apply. Yeah. Use the tool. That's right. And that's the thing, I think. AI is an incredible tool.

13:27 I think we're all still learning how to use it best. Pilot it, right. You're handed an infinite information machine. We're not all using the same inputs. And somehow you're like, how did you get that great output? Just the same as data. I gave it a better input. That's right. You gave it a better prompt. And sometimes... you can ask it how to prompt for something. I think that's the hack that a lot of newcomers don't understand or don't see. It knows how to get the best out of itself, so ask it. I use that a lot for images. Because if I liked a way that I accidentally got to an output, I'm like, whoa, I like that. I want that more. But I don't want to have to sit here and explain it to you. So I'm like, give me the prompt that will get that type of image in the future.

14:09 And I save that in Microsoft Notes. And then I come back to that and I have that as my image prompt. That's right. That's your default. And then you plug in the specifics of what you want. Yes. Perfect. All right. Last thing we do want to show is called the parting shot. Mic drop moment, soapbox statement. What do you got? You know, I think I told you on the last time I was here, I'm going to go with the same thing. I think it's so important and it's Warren Buffett quote, but he says, invest in yourself, right? Like if you invest in the stock market, as he did very successful, there's going to be some taxes you can lose. But if you invest in yourself, you take courses, you learn, you gain knowledge. There's no tax on that and you can't lose it. Nobody can take that away from you. So good. So invest in yourself, whether it's courses or reading, whatever it may be.

14:53 Invest, make yourself better than you were yesterday. Todd, appreciate it. Absolutely. Thank you, Jimmy. Good luck tomorrow. Make sure you hand out the candy here live from CSM in