In short
Podcast Notes: The Healthtech Podcast - Episode #418
Episode Overview
- Title: Sonia Szamocki, CEO at 32Co: What if every GP had a specialist on speed dial?
- Host: Dr. James Somauroo
- Guest: Dr. Sonia Szamocki, CEO of 32Co
- Description: The episode discusses how 32Co is revolutionizing healthcare by enabling general practitioners (GPs) to provide specialist-level care through a unique platform that connects them with specialists in dentistry and sleep medicine.
Key Themes and Discussions
Sonia's Journey
- Background:
- Grew up in the Netherlands, studied medicine at Oxford.
- Worked as an A&E doctor, later transitioned to management consultancy with BCG.
- Shift to Entrepreneurship:
- Initially focused on ophthalmology but found fulfillment in A&E due to the human interaction aspect.
- Gained experience in building a small education company, which sparked an interest in entrepreneurship.
Role of Technology in Healthcare
- Importance of Information Flow:
- Modern treatments require seamless information flow for success.
- Lack of communication and trust between generalists and specialists can hinder effective patient care.
- Building Trust:
- Emphasized the need for trust in healthcare interactions between patients, GPs, and specialists.
- Solutions must consider emotional and structural barriers to clinician adoption.
The 32Co Platform
- Vision:
- To enable GPs to deliver specialist-level care without patients needing to travel long distances or wait for appointments.
- Initial Focus:
- Started with dentistry, addressing the gap in specialist orthodontic care.
- Expansion:
- Currently expanding into sleep medicine to address the prevalence of sleep disorders, including sleep apnea.
Challenges in Healthcare Adoption
- Emotional Barriers:
- Clinicians may be hesitant to adopt new technologies or treatments due to fear of inadequacy or lack of training.
- Complex Healthcare Structures:
- The healthcare system is often fragmented, making it difficult for new solutions to gain traction.
The Human Connection vs. AI
- Role of AI:
- Recognized AI as a tool that can help but emphasized that human connection and empathy are irreplaceable in healthcare.
- Building Relationships:
- Highlighted the importance of clinician-patient relationships in facilitating better healthcare outcomes.
Key Takeaways
- Patient-Centered Care: The approach must prioritize patient outcomes, building a system that integrates specialists into primary care.
- Education and Training: Continuous education for clinicians is vital to ensure they feel confident in using new technologies.
- Future Vision: The model is scalable beyond the UK, with potential applications in various healthcare sectors globally.
Conclusion
- Final Thoughts: Sonia Szamocki emphasized the need to innovate healthcare delivery by empowering primary care providers with the tools needed to improve patient outcomes, highlighting the balance between technology and human interaction in achieving this goal.
Additional Resources
- Website: [32Co](https://www.32co.com)
- Host's Links:
- [Healthtech Podcast](http://www.thehealthtechpodcast.com)
- [James Somauroo](http://www.jamessomauroo.com)
Listening Information
- Listen to the full episode: Available on major podcast platforms.
Written by AI. May contain mistakes. Listen to the episode to check what was said.
Transcript
Automatic transcript. May contain errors.0:00Welcome to the Health Tech Podcast. Here we talk about everything healthcare and technology. and I'm your host James Someru. Hey everybody, this week delighted to be joined by Sonia Shomotsky from founder and CEO of 32co. Sonia, welcome to Health Tech Podcast, how are you doing? Very well, thank you. Good, nice to have you. Whereabouts are you speaking to us from today? We're in London, in Shoreditch. Ah, very nice, very nice. The cool side of London. Yes. Cool, Sonia, yeah, absolutely pleasure to have you on. Looking forward to getting into it. as you know from listening the first question I ask is you tell your story so yeah let's go for it where does it where does it all begin for you it it begins probably um in the Netherlands which is where I grew up um again yeah I came to the UK for for university um I went to Oxford and read medicine and um moved to to London to to do my clinical years most of that was spent in A &E, much like yourself.
1:01And I was in the London deanery. I loved it. I really enjoyed my time in A &E. I originally thought I was going to become an eye surgeon. Ophthalmology, I really, really loved. Why? I liked the microsurgery. I liked the intricacy of it. I liked the science of it. I just thought it was a beautiful specialty. But I had a chance to spend some time in in a and e and i realized that i really craved or enjoyed the human interaction side of a and e um ophthalmology is is quite dark as a specialty uh lights are off quite a lot of the time i was gonna say literally as in there's less light yes i don't mean dark as in you know difficult and morbid um you know it's um but the room is dark and and you know you see a patient at a time and I really like the buzz of A &E I found it really thrilling and worked in a fantastic A &E department with an amazing consultant body so I really I really got hooked and I realized that maybe I needed to look a bit further afield what made it a good A &E department out of interest the reason the reason I ask is I did a lot of locuming around London and like I remember I almost wrote a blog post about this about what would make the perfect a and e because i just experienced so many and it's something i never actually did although when i later worked at um the bmj and a few other places they actually wished i'd done that for loads of different reasons but it would it would have been such a good idea at the time i can't remember a thing about it now but what made it a good a and e i think that as a as a junior doctor what i appreciated most was a consultant body that was incredibly supportive and very knowledgeable and took the time to to make you feel like you were a welcome and be useful and never let you feel like you were on your own and I think this is why it's quite an important part of the journey because ultimately where I've ended up building is to create that exact same feeling and that environment for clinicians that don't work in that kind of setting.
3:09And so that's the first time that I really started to think about, well, what does work look like for me that is enjoyable? I love medicine. I love being a doctor. I love patients. I think it's all a lot of fun. But what else could I do? I had set up an education company at the time. I was quite interested in teaching. So I taught STEM subjects or essentially helped students who were interested in getting into STEM subjects at university and that gave me a bit of a taste of working outside of medicine and being a doctor and yeah I think I picked up a lot of very difficult lessons early on around how to build a company and this was not a massive company but what I did have to do is learn how to market and learn how to sell and learn everything that that we don't get taught as doctors in in medical school And so it was a it was not a soft landing.
4:05It was a hard landing, but it was great. It was a lot of fun. And I realized that I could I could do more with that learning. And so I took the decision to apply for a role in management consultancy. So I went to BCG for about four years. And again, a complete career pivot. But again, really enjoyed it and learned again there for the second time that when you do something new, you're going to be very bad at it for a little while. And that's OK. BCG is an amazing place where you're surrounded by incredibly intelligent people working on interesting problems in an environment that is in some ways similar to A &E in that you're all piled in together.
4:47You're moving fast. You're working on lots of different projects. But in other ways, very different. And it's a corporate environment where you have big clients that are spending a lot of money. And I suppose the setting feels very different. A &E can be very mucky and rushed and you're wearing scrubs. And it's all kind of, you know, it's not polished. Whereas in consultancy, it was the complete opposite. So I learned a lot there. I worked on healthcare projects. I did say to them when I first joined, look, I don't want to work in healthcare projects. I've left to learn lots around. A lot of friends have said the same thing.
5:29And I understand, you know, when you first join, you're very useful in those settings because you know what you're talking about. And so you can be deployed quite quickly and add value. But I spend a lot of time doing other things as well. So consulting in mining. And then the next week it will be luxury fashion. And then it would be airlines. and then it would be banking. And so I really got exposure to a lot of different industries, which was immensely helpful, a lot of fun. So I spent, yeah, that's about four years there. And then I managed to work or managed to get a place in a place called Digital Ventures, which is wholly owned by BCG.
6:09But it's essentially a startup incubator. And it was sort of a separate entity. And it's a venture building studio. but because it's linked to a lot of the work that BCG does it's it does it on behalf of corporate clients and so I spent a year building an app for patients with hemophilia so at the time yeah obviously hemophilia rare disease and expensive to treat clotting factor is expensive now obviously there's newer newer treatments but what was being what people were finding is that tracking of the crossing factor was very very difficult um you're finding patients you know logging it on bits of paper and you know kind of by gut and and that's that's hard because everybody has a different half-life and so could you build an app or a medical device that helps patients with with their with their treatment and monitoring things and reporting back to the hematologist as well so that they knew what was going on and so that was my first foray into med tech and really understanding how technology is built and, you know, what it takes to get something from an idea or a problem to here is a fully functioning medical device that you can now give to patients and they can use it in the wild and get hopefully a huge amount of value from it, but also see their health outcomes improve.
7:31So, you know, that exposed me to, I guess, the cutting edge of health care. so the things that are happening at the frontier of innovation and development of tools and drugs and protocols and things that we know will have really beneficial impacts on patients. But I found myself at that point always being the sort of, not the naysayer, but reminding people that this is all very well. These are amazing things that we're seeing being invented and coming to market. but what you're forgetting is that if you want this to be adopted there's a clinician in the way right and and they have to feel confident confident and comfortable um that what you've just invented is actually going to be helpful to patients you know it doesn't carry any risks and and that they can they can crack on so when you think about um you know how you're trained my medical knowledge probably quite out of date if you think about it because i graduated when most of the things that we're treating patients with now, well, there are new things that are being brought to market that we can now treat patients with, and that wouldn't have been covered in our syllabus.
8:48So you're placing a huge expectation on your clinical workforce to remain up-to-date and to remain up-skilled, but I don't remember having any time really in the diary for that other than sort of evenings and weekends. And you might argue that that's fine. I mean, everyone has to do that. everyone has to stay up to date in terms of in every profession but I think particularly important given how fast healthcare is moving that that that bit is understood and taken seriously and I think a lot of money is spent on trying to convince doctors to do stuff to use stuff I mean you know the bag of M &S sandwiches you know never went amiss in terms of getting an audience but that's quite a big gap from from that from being introduced to something to saying do you know I actually feel really comfortable and confident injecting this into a patient or, you know, offering this treatment.
9:39And so I think I started to think a lot about that problem and how health care is becoming increasingly specialized. And we are bringing amazing new things to market. But I don't think we're spending enough time talking to clinicians about it and helping them feel confident and giving them the tools that they need to deliver that sort of care locally. And so I left and I said, I'm going to go and do something in this space. I think I've now got a good enough perspective on what it's like on the front line. I understand health care systems because of my consulting experience. And I've spent time actually building tech.
10:21So if anyone now has the sort of ingredients to do something, it's me. and I should probably go and find out how to actually apply this in the real world. And so I looked around a lot and I think my, you know, I looked at medicine, I looked at areas where I felt like there were bottlenecks. And some of these are really well documented. So, for example, you know, we know that there are not enough dermatologists to look after every patient that has a dermatological problem. There are bottlenecks in fertility. There are lots of areas where I thought, you know, if only we could do more to get our clinical workforce really, really up to speed with things that they can deliver to patients.
11:07And but I also started talking a lot to dentists and we're sort of coming off the back of COVID at this point. And dentistry was massively impacted by COVID, as you can imagine. You know, you're very close to the mouth. and I was looking around and learning about dental specialties and the sorts of treatments that I think lots of people wanted but perhaps there wasn't as much access as you would like to think and that's where the first vertical came from so for 32co we're essentially an end-to-end platform that helps clinicians in primary care provide more specialist treatments so it's it's as simple as that so we talk about locally deliverable specialist health care so these are the sorts of treatments where you really do want specialist input but you don't need that in person you know you could have that locally if you got for example your GP to work with that specialist virtually you know you didn't have to travel to the hospital or to whatever a center to see that specialist and sit in a year-long waiting list.
12:17And we talk about that specifically because I think there are areas of healthcare where you absolutely must go and see a specialist in person for practical reasons, as well as for kind of, look, we're operating on you and I need to see the surgeon or the surgeon needs to operate on me. And then on the other end of the spectrum, there are treatments where I think you probably can do it direct to consumer. You probably can deliver high quality level of care without actually physically seeing a doctor. But I think those use cases are quite small. And I think we've seen lots of direct-to-consumer companies struggle.
12:55I think that you can probably tackle a proportion of the population, low acuity, low risk. And I think that's great. But there's this massive, massive group of patients where, again, if we could get that specialist care delivered locally, we would increase access an enormous amount. And so, yeah, so starting to talk to dentists and seeing that this problem existed there too. At the time, there was a very big direct consumer company that was selling orthodontic devices online. They were doing incredibly well. And at the same time, dentists were very worried because of the amount of, well, quite honestly, terrible outcomes that they were seeing from patients just giving it a go.
13:44And orthodontics, I think most people don't necessarily realize that orthodontic is a dental specialty. And it's an additional three years of training that you need to go through to move teeth in the mouth. And there are only a thousand orthodontists in the UK. So if you want good quality orthodontic treatment, going online might seem like the easy thing to do, but it carries with at risk. And lots of patients did have bad outcomes and ultimately the company went bust. So my conversation with these dentists was, look, well, you all agree that online only treatment isn't the best way forward for this specific use case.
14:20But why aren't they just coming and seeing you? Because you're a dentist, you're a trained expert. Why can't they come and see you? And it came back to the same thing as I mentioned originally, which was, look, I'm just not super confident that I can deliver the highest quality care at the standard of an orthodontist. And so, you know, I'm not going to just give it a go unless I'm sure that I can do an amazing job. And I think that's great, right? I mean, it just shows that we're doing something right, that clinicians are making sure that they don't just, they're not just gung-ho and give it a crack because they can make some money.
14:58You know, that's not what I was seeing anyway. So I said, well, look, what would it take for you to feel confident in doing this? You know, how could we create technology in an environment whereby you would say, do you know what, I actually now can offer this to my patient base? and the basic ingredients were, look, I want to be trained. I want good quality education. I want to learn. Then I would like to be able to talk to an orthodontist. Very simple, right? I would like to be able to communicate with them virtually on a case so that I can be 100 % sure that I'm going to deliver the same quality of care as they would.
15:36And then finally, I need a place to design the treatment and get hold of the devices and procure things. and then a whole list of other requests came through. But those are the fundamental pillars, really. And if I now take it back to my time in A &E and why I think this felt so important, it was because when I was in A &E, if a patient came, and I'm sure you're the same, if a patient came through the door and you had a head injury or a dislocated shoulder or someone's vomiting blood you know you knew that there was a red or a consultant neurosurgeon um you know orthopedic surgeon just around the corner and you can grab them and you can say look i'm not entirely sure i'm just going to check my management with you what do you think am i missing anything very normal for us to do that in a &e almost forget we always forget about it now you transport yourself to primary care and you realize that none of that support exists right it's just you and you know your your patients expect you to know everything obviously they don't really understand that you are not an encyclopedia of knowledge in every specialty um and and so you're you're trying to do the best that you can with with the sort of information available to you without a ton of diagnostic tools um and without the latest in for example, management of some complex dermatological issue.
17:09But maybe we could create that. Maybe we could create the same kind of environment because all it would take is for me to patch you into a dermatologist. You could be looking at the patient together and saying, okay, well, I think you've done everything right here. I would actually just recommend that you take part A, B, and C. That shouldn't be that difficult. It turns out it is quite difficult to do, but that's what building companies is about. It's taking something that seems simple and then actually getting really stuck in and building it out. But that was our first use case. And so we're operating across the UK now.
17:43You know, there's a 32-code clinic within 30 minutes of 90 % of the population. So there's plenty of access now. And what I'm most proud of is the fact that we are not compromising one iota on the standard of care. Because you basically get two for the price of one. And you get your own family dentist who you trust, who's got a very broad set of skills. And then you've got a specialist. And now we're rolling that out to our next vertical, which is in sleep, which is a little bit closer to, I suppose, what I know. Because I am not a dentist, but I have learned an awful lot of dentistry from doing this.
18:22But sleep is another area entirely where we are not, I don't think, delivering the greatest quality care across the board. and that's not to blame anyone but we do not have that many sleep specialists in the UK and there is more and more and more awareness of the importance of sleep and the prevalence of sleep apnea and snoring which is one of the sort of treatments that is included in this vertical. So it's the same principle really. Can you get your local care provider to help you with your sleep problems without you having to physically go and see a sleep specialist? Wow, what a story. um so much that i want to talk to you about um and it's it's it's interesting to me how people often try and solve the issue that they had in the world as well and they you know they want to create what they lacked and it's somewhat in your story there about you know being in a and and you know the best part of a and has been supported but then you know you can end up as a sole clinician completely isolated you want to do this new thing you're completely capable but you just don't have the knowledge you'd love to just pull some around the corner but it doesn't exist it's an incredibly relatable issue for like all clinicians i think anyone that's trained in definitely any system that i've been part of incredibly relatable um i also think there's something interesting in your story even before that that you talked about at bcg um and this element of needing to be the fool and that's okay this element of starting at zero before you're competent there's always a zero moment you're always at the bottom of a mountain to climb of becoming competent and that being okay it's the that being okay part which i think is interesting because i think as overachievers medics can often struggle with this and not just medics but i've got friends in law in loads of other things that that definitely struggle with this in sort of moving over to something new and being willing to be the fool being willing to accept they know zero and getting to a level of of competence and i think that's also interesting because that needs to be explained and taught and people need to come to terms with that in order to adopt this model that you're talking about so rather than just outsourcing to the expert it's actually this willingness to be the fool that's that sort of baked in there that has to actually exist first and the willingness to be the fool is in part knowing that you are capable and having confidence and knowing that you can be skilled up and you can teach an old dog new tricks and all that sort of stuff is all kind of baked into this model but it is super interesting and the other thing is you've had investment as well from like boulder turn and eco venture like all the very credible like huge investors to really make this work which is it definitely adds a level of gravitas to that with to this with you know seasoned investment firms like that going like hey this is actually a really interesting model to to take forwards so really keen to talk more about it i want to go back to the the problem itself here and it's something that we all know to exist but to one of the final points you made about this seems really simple but it's actually really complex what's actually the issue here is it the structure of medicine and dentistry that pigeonholes people early into a certain generalist or specialist and then you can't do anything exterior is it partly that and something else is it learned behaviors of us that we don't want to be the fool and don't want to do that other stuff is it the there's a lack of a reward system for that is it the the financial system of something like dentistry lends itself more so than medicine is it like what's actually the structure of the problem here and like why is it so messy to solve yeah I think you've touched on lots of reasons why it's challenging and I say there's not one thing but if I take a couple of them in turn so I think that um not wanting to look like you don't and what you're doing is such a fundamental part of being a doctor or a dentist right you've got to oneself and to the patient to the patient yeah and all sorts people people put their trust in you um to look after them and so to sort of take down that mask and say look i don't actually know is not something here's my helper for the day exactly um so i think that that's that's kind of that's quite core and fundamental to being a healthcare professional.
23:13And I don't think that I'd be advocating to remove that entirely. You know, we do need to build trust with patients and we do need to show that our advice is to be taken seriously and that, you know, we will be there to look after them and, you know, help them make the right decisions for their care. I think that the second thing is information flows. I think that there is so much to know now in healthcare, different to the NHS when it was set up all those years ago. There is not, one person could not know everything that there is to know. And if you accept that, then we need to be building information flows to allow the right information to be accessed in the right place at the right time.
24:04And obviously technology plays a big role in that, and I'm sure we will talk about AI. My fundamental belief is that when you're dealing with humans, AI can't solve everything. It can do a lot, but I think the human connection and building relationships between humans, whether it's patient to patient, patient to doctor, doctor to doctor, I think is fundamental to trust building. And a crucial actually to, sorry to jump in and stop you, but I just want this for listeners because we say things like this all the time on here, but I just want to highlight, what you're actually saying there is that the human to human interaction you know separate from ai is important in facilitating the flow of information optimally that's a really important thing to highlight because it's a it it's also something i've talked about before in something as simple as history taking because there's so much evidence that there is a direct correlation between the level of empathy felt between a patient and a clinician and the quality of that history that is being taken.
25:13And I think this is a really, really, really important point. As we move into a future of academically analyzing where humans are best and where AI is best, this is a really important point you're touching on. And I would encourage you to go into more detail on Because like this thing about humans building trust improves the flow of information and therefore AI cannot deliver a perfect system. It needs humans because we need the most information in order for it to be perfect is so, so, so important before we start removing humans from all context and parts of healthcare. So that I think is an important distinction.
25:54Yeah, completely agree. I mean, if we take the patient to doctor interaction and we said, OK, well, option one is that you can have a seasoned health care professional who's done thousands of cases before, talk to you, talk you through what they've seen historically and explain to you why, given what you've said, actually, this is the approach that we should take. And then the alternative being, well, here's an AI that's been trained on thousands and thousands and thousands of patients. Just talk to it and it will tell you the answer. I think most people when they're dealing with healthcare, especially when you're talking about a loved one, you say, look, I actually just want to make sure that I'm dealing with someone who I can see in their eyes that they know what they're talking about.
26:39And I'm not suggesting that this will never, ever happen and that AI will never get to this point. But I think today we're not quite there yet. I think a patient wants to know that this person is going to have their best interests at heart. No, it's true. I find this fascinating though, because i almost wonder if there is like a fundamental truth um a different conversation whether you believe in objective truth whether it's a relative but like i i think is there a truth to the fact that we as humans actually value a human telling us that's the 98 good human is better than a 100 ai is there some fundamental truth in that and if so like we need to study that we need to figure that out we've been in a situation where that's always been there it's always been a human delivering information always in healthcare and we're only now approaching a time where it's optional so like what is the benefit and i think if if what we assume is that there is a feeling of being cared for that ranks higher than actually the quality get we've known for we've known for decades you trained in medicine we we've known for decades that in inverted commas, good doctors get away with murder, quite literally, unfortunately, sometimes, because they just have amazing bedside manner.
27:56So we know there's actually evidence for this, there is a genuine truth to this, that bedside manner has always counted for so much. And it's just interesting to me that I think there is a fundamental truth that we rank that higher than perfect diagnosis and treatment, like perfect patient experience is a better definition of perfect healthcare than perfect diagnosis and treatment is and that's i don't know how you feel about that but like that's something that i'm playing with at the minute yeah i think it's it's a fascinating area that will need a lot more research and i think you're right um i think that we've known for a long time that you know even in person healthcare provision that there is something about it that allows you to pick up more information on both sides to be honest you know you know a patient is looking at the doctor as much as the other way around to try and figure out whether they can trust this person but if I give you an example on the clinician to clinician side of what we build um you know one of the one of the things that we've said and I think you touched on this earlier is there is no such thing as a stupid question right this technology this environment that we've created encourages you to ask the question that you didn't think you could um because at the other end is going to be a specialist who's going to take the time to to walk you through something um and often it's not a stupid question it's just you know it's you're trying to get reassurance that you're on the right track.
29:18So, you know, we, for example, offer, we've built an AI coach tool that allows clinicians to ask questions that they can get real-time answers to and that has been trained on thousands and thousands of questions that we've been asked over the years and that we have meticulously curated the answers to. now we're very deliberate about where we use that so what we are not saying is um you don't use that tool to ask questions about a patient so these are questions where look i just want to know just what does the evidence say about this or remind me what is the best technique for this you know it's those questions where you just think actually i'm i'm quite happy to get the textbook answer for that I'm actually just trying to learn yeah that's very different to the other type of interaction which is what should I say to this patient about this thing yeah and and you know we see it every day we see you know we see dentists they don't want to use the coach when they're like look I've got patient coming in I'm just not entirely sure about this in your experience you know how how do you see this progressing if it's something tricky or something challenging and this happens you know all the time.
30:34I mean, not every, not every patient has the smoothest ride through, um, through treatment, you know, orthodontics is a relatively low risk treatment, but patients have high expectations and, um, you know, there's a, there's a compliance piece as well. They have to wear these, these, these devices themselves as well. So you just want a real human being that has done this a thousand times before to say, don't worry, you're completely fine. I've seen this will resolve itself, no problem. Or one in a hundred times saying, do you know what, you are actually right? I think we need to pause things here.
31:07And that is where the trust gets built. I'll give you another example. When we first started working on this, I said, one fundamental principle, if I'm going to be building this business, is that patient outcomes will always be number one. It's not even a question about whether we compromise on patient outcomes. And that isn't just a sort of airy fairy, you know, oh, I've got these kind of high minded ideas. It's, it's because I think it's good for business. Because when you show the clinicians that you work with, that you are unwilling to compromise, they trust you as well, they trust you more.
31:43So when we first launched the orthodontic vertical, and we said to dentists, look, you can train with us, you can start providing this treatment, you're going to collaborate with literally the top, top experts in the country on your on your cases. um the first thing we said is okay well we need to build in the mechanism for the specialist to say i wouldn't treat this patient if i were you yeah and initially we were quite worried about it because we thought we've gone to all this effort you know we've trained these people we've got them set up on their platform they're really excited they've got a patient who's um who's sort of agreed to go ahead with treatment and now the specialist is saying look i wouldn't do this if i were you this is going to be a bad bad experience for everybody and actually it was the complete opposite um the specialist explains you know here's something that i've seen um you know i don't really like the look of this i think we should probably you know let's do let's do a couple more scans let's let's see where we're at and and but i don't want you to rush into this one because uh i think that there's a risk that we could get not a great outcome here it built so much trust and it wasn't a kind of an ai computer said no it was a real human being that said i'm looking out for you I care about you I care about your patient it's not worth it you know send this to see a real specialist in person because it's something that I think is tricky and so that's another example of where a real human being watching out for you has such a big difference and that's the difference between that dentist saying I'm confident I'm confident now making this part of my toolkit I'm confident offering this to every patient that comes through the door because I know there's a safety net.
33:18This is something dodgy that I haven't spotted. I'm working with a team. We're going to do it together. So yeah, I agree with you on the AI piece. I think it's a very interesting area to figure out where it's useful and where we'd like it to be useful, but actually we're trying to do horn in some tech where it's not yet welcome. I like just how deliberate you've been there like about where what communication is being done what is that spread of information and and a human is better here definitely it's one thing i like about the uk health system broadly as well if you're going to get involved in something like you're doing which is spanning a few different layers and actually bridging the gap between a few different people and really it's it's facilitating relationship building is is is what you're doing i love the fact that the uk health system will hold you to account massively and anyone else that decides to enter this space the UK system will hold you to account like without any shadow of a doubt because as you quite rightly point out physicians are never going to compromise on that and if you try and push a commercial agenda over a patient care patient outcome quality of care agenda you're just going to get ratted out and also and ultimately like you're you use the word the trust the trust's gone once the trust's gone it's over like there's just there's just literally no point so having that fail safe in there which is just a it's just making sure every every door is two-way rather than any door being one way towards any kind of commercial benefit like it's so so so essential um in order to build the trust which i think which i think is just so necessary one question i did have is to put some color into this could you talk through like maybe maybe an example a couple of examples of the types of the types of treatments and i know it's in dentistry we talk a lot in health care here it is still broadly health care you're moving into sleep maybe talk about it there as well but so i imagine the vision's like far beyond this right like this is going to be a system where you talked you said it like you know this is this is a system where primary care can then get you know the expertise to do more in primary care that's more specialist i think that's a really interesting vision which fits into so many 10-year plans and all this kind of stuff So it's important, I think.
35:36So yeah, what sort of stuff, what would be just following a case through? What would that actually look like? Yeah, maybe I'll take sleep as an example because it's an extremely multidisciplinary challenge. And so where we've started in sleep is to build the infrastructure so that dentists can provide sleep apnea devices. So what most people don't know, and I didn't know as a doctor either, is that there are, you know, CPAP is not necessarily first line for every patient who's got mild to moderate sleep apnea. There are some patients. CPAP, by the way, for listeners being a machine, was it continuous positive airway pressure?
36:18That's testing my anesthetic knowledge. Yeah, I still remember it, still got it, et cetera. yeah the loud machine that you wear to mask the that keeps a positive pressure in your airway so they don't collapse exactly exactly and sleep apnea is a condition that probably affects about 10 million people in the uk although very very underdiagnosed at the moment as you can imagine um symptoms very non-specific you wake up feeling rubbish i mean that that does not tell you very much right so lots of people live with this condition and don't know i think snoring is a little bit of an alarm bell so snoring goes hand in hand with sleep apnea sometimes but um other than that it's not well diagnosed but c-pad masks are you know it's a relatively invasive treatment and not everyone gets on with them it looks deeply unpleasant and that's coming from an anesthetist from the partner as well the sleep partner they've got to listen to it and so um what most people don't know is that there are devices that can be worn in the mouth that they're custom-made devices they're nice approved they have been for a long time but the problem is that you need a dentist to make it because it's a custom made it fits around the teeth and it the mechanism of action is slightly different it still opens up the air but it does it by pushing the lower jaw forward slightly during sleep yeah and if this is are interested try try and tilt your head back and push your jaw forward and see if you can snore or see if you can create that collapse it's very difficult and so they're very very effective devices now obviously patients are not turning up to the dentist to solve snoring or um sleep apnea problems it's just not known that that's a channel and and when i speak to sleep specialists they say look we know that they have that these devices work we just don't have any dentists near us that offer this interesting um and so that's right for the for the diagnosis of that would one go to a gp to then be referred to a dentist or would one have to explain those symptoms to a dentist it depends yeah what's the link there that seems like a gap well that's that that's exactly where i think our technology comes in is that we can right bring different specialties together virtually to work on behalf of a patient which is why i thought this was such an important use case and the other thing is i i did a neuroscience degree as part of my my training and um sleep was one of the areas that i that i was interested in um and i'd say that in the 15 longer now years since that have passed sleep um knowledge has come a huge huge way and awareness of the problem has come a huge way but the treatment availability has not come very far at all um we find we see sleep centers with two-year waiting lists to get a c-pack device that's that's pretty pretty standard so that the pathway might be patient um feels terrible goes through gp we rule out other causes of fatigue obviously um and they send them for a sleep test they can be done overnight in 24 hours and you can get a pretty clear indication of whether there's evidence of sleep apnea and if so then the patient has a choice they can be referred to a sleep service or well and they can also pursue other areas of of care for example going and seeing if they can get one of these devices they're called mandibular advancement devices now the other channel is if you're just if you're someone who is you doesn't have a diagnosis but you're you find you're snoring terribly and it's causing problems for you it's causing problems for your sleep partner you can get one of these devices as well without for snoring alone without a diagnosis of sleep apnea as in right okay ruled out so i guess the the the interesting thing about that use case is that number one there's very little patient awareness of the solution yeah there's evidence huge amounts of evidence that these devices do work um doctors recognize it um but we just we they just haven't got access.
39:59And it's this knowledge and information flow challenge again. So we apply the same model as we did in orthodontics. We say to dentists, you know, we're going to train you. We're going to give you everything that you need in one place to be able to really, really look after one of these patients from the treatment planning to choosing the device, to monitoring the patient remotely, to communicating with the patient. But crucially, the specialist that is going to be involved here is going to be someone who knows something about sleep right it's not going to be another dentist it's going to be a medical professional and i think we've not really seen medicine and dentistry being bridged in that way and sleep is such an important use case for that because i mean even i mean anesthetist ent surgeons can be involved in in airway um in some way shape or form it's probably the most multidisciplinary area of the body if you think about it so it's a it's a fantastic use case and then when i think about how do we apply this beyond and using those same pillars and those same principles.
40:58I look at anywhere where there's bottlenecks in care, where I look at the number of specialists on the register and you think that is not nearly enough for the population of the UK or the US. Fertility is high on that list. It's not easy to get access to good quality, well, to fast and comprehensive fertility advice. You're usually being bounced from one specialist to another. And it's not like we're doing complex surgery in the majority of cases, right? It's not like you're having to go to a center to do something really complex and invasive. You could imagine a GP working with a network of fertility specialists and working as almost like a virtual MDT on your case.
41:46And it means that rather than you as the patient having to being bounced around and you traveling to those specialists they're actually coming they're coming to you and from a health economics perspective you can imagine obviously that this is also a more cost effective way of doing things right you reduce the wait list because the scaling you know one of the other ways that I describe the business is that I'm scaling specialist experts and so if you think about you know outpatient clinic how many how many patients can a rheumatologists see in one session right even if you overbook them and you know make them do it late into the evening that that's that's really the number of patients they can impact whereas if you get them working with a network of gps or asynchronously or whatever you can scale them massively right you know creating capacity in the system yeah exactly so so these are the sorts of use cases that i'm looking at so chronic disease is diseases are are one but then there's also more um i don't want to call they're not lifestyle things but they are treatments that patients want that are life-changing um but not you know these are not conditions that are immediately life-threatening so this is something i was going to actually mention or talk about because like that you're bridging this gap in in areas that i think a gap exists so we for whatever reason we say we have a public health care system in the uk yet anything in the mouth or teeth broadly that's privatized you want an eye test privatized hearing test private like there there for whatever reason there are just these like majorly important organs that we've just decided like yeah if they're failing if the if the lens in your eye is failing that's actually a you problem that's actually put your hand in your own pocket problem rather than anything else so it's like it's a there's this and i know that there are obviously nhs examples of that but broadly privatized in a big way so there are lots of these kind of gaps that these as you quite rightly points out important to me as a patient things fall particularly where it becomes now you know a lot more onus on me to organize it and arrange it then if i've got multiple people involved i've got to coordinate my own care um you know it's it's funny like even my own dentist journey that i'm thinking about like like having to coordinate all of that stuff myself and then there's this like incidental oh you might have a saliva stone which isn't a salivary gland stone that's not really a problem but like yeah i'm gonna have to refer you like do you want it private you want it this and like are the appointments not for like a year and a half but just remember that that's okay like there's all this like stuff and it's just like no it's not gonna kill you don't worry about it like it's you could just do nothing well well now i'm just in this like no man's land like no one cares about this like the gp doesn't care the oral surgeon or whatever the hell doesn't care like there's this weird like there's just loads of stuff like going on seemingly that this helps to bridge because it gives people an ability for like someone to care like if it give if it hands primary care the keys to all of this stuff to go we'll coordinate this for you now like this is fine and i think that helps because you're right things like fertility also fall into this bucket of like is it privatized is it not like for whatever reason this is a crucially important thing for life and for me to pass that on and yet doesn't fall into the nhs remit as much as i'd kind of want it to or like it to but that's perhaps idealistic so it feels like a model for this stuff is i'm surprised i haven't heard about this before in all honesty i'm surprised like i haven't seen more of this like happen or be tried to fix but i guess it's because the problem is so messy and and diagnosing what the problem actually is i think is important let alone trying to figure out how to actually solve it so yeah i just think you've come at it in in quite a unique way i guess and I just wonder like how broad that future is for this and the question being really like who pays there's a lot in that I think that we would all love for the NHS to be able to fund everything but as I said right at the top is we are inventing so many amazing things every day that can change health you know the budgets are just not keeping up and it's a decision that I guess we have to make as a country that you know we either do want to fund everything So you get an MRI of your knee tomorrow if you want it.
46:12At the moment, what I think is still fantastic about the NHS is that if you're very, very sick, you will be looked after and it will be the highest quality care. And I think that is very much still worth protecting. But then it means that the sort of tail end, exactly as you say, of things that are not killing you today immediately, but they would probably increase the quality of your life by quite some margin. those are the things that are dropping off into the private sector and I'm not going to make a judgement of whether that's good or bad it's just it is a decision that needs to be made and I think the reason why look I don't think when I explain this to people people usually are like oh that actually makes a lot of sense and I feel like maybe this should exist already the reality of actually building this is it's really really hard there are so many parts that have to work to make this all sync up perfectly.
47:07And because, you know, it's a multi-multi-sided environment, right? You've got, you know, in our platform, for example, you know, we've got the diagnostic side, you've got the clinician, you've got the experts, you've got the manufacturers, you've got the patient, you've got the clinic that they work in, and you have to tie them all together. That is a big, big technical build, and you have to get everyone on board along that journey as well. You know, if one person in that journey is not, you know, incentivized or maybe they don't have the tech that makes this very, very easy for them to do, then you won't see adoption.
47:42And the reason why using orthodontics was so useful for us is because we had to build absolutely everything just purely based on stakeholder feedback. You know, we didn't go into this with a sort of diagram of, no, no, this is the end state. This is how it should look. That wasn't it. we just looked at the problem so what would it take what would it take to overcome this challenge what would it overtake to come this time you end up with the platform that you have but that's a huge amount of earned learnings right I think a lot of learnings you know nothing there's nothing in a book that tells you how to do this you you learned it by doing it and feeling the problem and seeing the problem um and so the next question for us is well how much of that is now reusable for the next vertical and i think that we were very positively surprised by how much was just directly transplantable onto the next vertical which is exciting for us as a business because it means that the next verticals will come very very quickly and especially with ai now supporting the the speed of build um you know you can take you can you can bring new you can increase access to different treatments quickly.
Read the full transcript
48:50And so it is an exciting time. I think for me, the question of scale, you know, there is a question around this problem is not a UK only problem, right? We've started to work with the US, with US dentists as well. And by and large, it's exactly the same. We're hearing exactly the same thing. You know, healthcare in some ways is very universal, right? The mouth is the mouth, whether you are in the US, the UK, or Papua New Guinea, right? That is constant. And actually how clinicians work, even if some systems are maybe more insurance-based than others, I think by and large it's the same flow. The patient flows are the same.
49:31You're trying to achieve the same outcome. So scaling this internationally is obviously an interesting area for us as well as scaling verticals. And then you come back to the challenge that all startups have, which is that, you know, where do you deploy resource to make sure that you're focused enough to nail it and that you're not spreading yourself too thin. You mentioned AI before. Where does AI fit into this whole thing for you? Huge. I think for us, it's been, you know, we've made a very deliberate shift to making the whole company completely AI first. And the first thing I'll say is that that is not just the tech team.
50:07That's everybody needs to understand the AI tools that are being used. So, you know, when we first, for example, launched the infrastructure for the AI coach, everyone used it from HR to operations. You know, it wasn't a case of, oh, this is a nice feature. This is now embedded into the DNA of the company. And it's actually more important that the non-techies adopt it than the techies. The techies will get there, right? You know, this is their day-to-day. This is their bread and butter. And so the AI has done a few different things. It's A, it's allowed us to build faster. I think a lot of people have experienced that.
50:45I think that it has allowed us to build things for our customers that we know will either reduce workload or increase quality or preferably both. And those are the parameters that we use. That is where we will deploy AI, where we think that there's genuine gains to be made and where a human couldn't do a better job. and then I think that from a I suppose just a mindset perspective I think we we have to accept that we're in a completely new world now when it comes to how you build and that again coming back to you know going back to the basics and and being the fool this technology is changing every five minutes so the other thing that we've definitely infused is this idea that you don't get comfortable with one piece of software because we're changing it every every two weeks there's another one that comes out that's better and so you have to build to undo things that you've built and be completely okay with that if you want to stay at the cutting edge you have to do that you can't just say well we built this last year I mean we've done that and we've been like well it's already you know it's it's way worse than it could be if we just updated everything so you you're not building the same way right you're not building in this kind of deterministic um i build it now and then it's just going to keep running you know these things need to be maintained they need to be um you need to run regular evaluations and its performance you need to be constantly looking for new new software i think this is like a lesson for entrepreneurs everywhere anyone with a business anyone leads a team in fact i think there is a there is a line on the p &l and the literal money p &l but also your kind of overall resource and time of your staff p &l that that is experimentation as well yeah you have you have to like i've i've had to just accept that i'm going to spend 100 quid a month buying random ai platforms i'll never use again like i mean easy in the hundreds per month probably um in the hope that and in the knowledge that when we get the one that's right for us it will deliver 100x on that value in the time before the next one comes out so there's something about all the gains to be made by being at the cutting edge of this and being in the early adopter group will only be realized by companies that are experimenting and trying and in that there is going to be a lot of loss a loss of time a lot of a loss of financing to pay for it but in the grand scheme of things the roi is so unbelievably large there's no other option there really isn't i mean i've spent i've spent days like just on lovable and like cintra and like or just like anyone that you can mention that has just like made any sort of waves like i've definitely tried and bought and like tried to experiment with my custom gpts the amount of this amount of that that you've just tried to build just to try and like get something done and eventually something clicks and you're like oh my god now this whole podcast process is easier because all you do is export the transcript and then blah is done like whatever it is but then all of a sudden it starts to pay back and it's so so so so so important like it really is and yeah good luck getting investment with that sort of stuff now and I think it's a real really good point about the grand scheme of things because I'm reading a lot now on LinkedIn around how companies that have invested in AI yet to see a return you know hasn't made any meaningful difference you know hasn't generated ROI and I think that of course not this is a completely this is a complete shift those gains will come but there's this horrible embedding process and learning process that you have to invest and you have to pay for it um and it means that resource will be dedicated to that rather than other things but i agree with you i don't think we have a choice i really don't especially if you're building the frontiers um you know you are going to be left behind if if you don't do that but and it comes back to the earlier point of get comfortable being rubbish at something for a bit totally agree it's it it's it's that is just what it takes and i also think as well like just to put just just to be really clear on that like there are some days where like i'll i'll take like a week off looking at these platforms and the rest of it and so and then so much has happened within a week that i'm then reticent to go into this and like i'll fall into the same traps as most people i'll end up scrolling past these videos of people that are now talking about this i saw one okay i'll get really real here i saw one which made me think i'm having an existential crisis about oh my goodness the the entire video generation side of what we do is now like needs to move this thing and just prompted this whole existential thing this guy literally just went on n8n.com and built this really complex system like automated system of n8n's like a turbo zapier for anyone that doesn't know what that is listening but built this thing which used nano banana the new google image generation video generation thing it used uh like a load of them a load of stuff in level labs like blah blah to the point where he would literally just describe a video to his phone and then he would refresh youtube and there was like a video of like star wars star troopers talking about some ridiculous thing in ancient greece that and he literally just spoke this into his phone seconds ago and now he's built this this this thing that i'm like oh my goodness this is and you can look at that stuff and you can be so unbelievably overwhelmed until you realize that probably broke five seconds later with the most minor update on the back end of nano banana or any one of the hundred things that you've probably done and actually turning gimmick into practical utility is a whole nother ball game and there's there's a load of stuff that's that goes around that but what my point being there are these things that are possible and people are always pushing the frontier of the possible from a technological standpoint and that's what that person represented that's what that creator represented there is a whole other frontier going on the frontier of adoption that is difficult turning that stuff from gimmick into utility is one part of that convincing other people to buy it is another actually deploying it in a company for which it actually delivers ROI and value is another and the collection of all of those things is kind of where you need to be at that's taking some of the stuff that's at the technological frontier not all of it nowhere near all of it like this creator was doing but some of it is there some small part of that that could redeploy a client's product into a different environment that makes it look this to fit this narrative of this post in this way and build this story well yes we can probably do that cool does it need to be sent into a phone show oh not really but the point is you can take some of that inspiration but rest assured if you just experiment with it in your own way and get some utility from it get comfortable with doing that then you're going to be on the road so that when the new things come in you'll feel more comfortable but you're right like that's the journey that i've had to become completely okay with that i'm never going to be a person building those complex and 8n things with brand new models and api and stuff at the back end like to high heaven and i'm not going to be doing that but i am going to be taking some of that with a real focus on how do i turn gimmick into utility that's the bit that i'm hyper focused on right now because that's the bit that no one's done i think anyway particularly in my niche of health tech marketing so that's where i've that's where i've come like full circle on it anyway but it's taken you the exploration of all of these tools to find it right i think that people correct too often look for the magic formula and they go on wherever it is linkedin or substack and yeah someone says here's my magic formula for x and they say right I'm going to adopt the formula actually what you've done is you've tried everything and found and tasted a bit of everything and then you found the two things that you that you like you put them together and now as now there's a use case that you can really see it working on I really like the um from gimmick to utility thing I haven't heard that before but I that that was what I was going to say is that not to discourage playing around because I've done the same thing playing on weekends with every tool that I can to build something crazy just to see how far I can push myself.
59:18And I'm not from a technical background, but I think it is incredible what you can do without a technical background, with Cursor, with the lovables, with the sales of the world. But I think that the jump from prototype to real world application is so much bigger than people realize. and again not to discourage the building of the prototype you should do it but to get that into the hands of real customers stable secure and you know again AI is not deterministic you know every update you have to go and recheck it you have to build infrastructure that allows you to understand whether it's still performing like you wanted it to there are so many more layers but the front end stuff is what gets people excited right the lovables as well oh my god I built a website in five minutes that that is what gets people excited back and integrations not so sexy so yeah i would agree before i let you go something last thing i want to ask you is is more about your sort of entrepreneurship journey um going from bcg obviously management consultant company it's it's a very different environment but we do i do see a lot of people start companies from management consultancy like mark jenkins that we've had on here and i mean there's loads right there's loads of people that go from that it it seems to me though that management consultancy is a very different environment to entrepreneurship although someone actually said said this to me i can't remember who it was i'll try i'll try and remember but um they did say that at a management consultancy firm you learn how much work can possibly be done in a day so that when you do become an entrepreneur it actually feels like a bit of a you get a lion and actually you get all things even though you're working your ass off actually actually you you do know how much work can be done in a day which is an incredible advantage once you do start a company so that being one advantage um but how do you think about that do you consider yourself someone that was always going to build a company and that you had that risk appetite and you wanted to create and and you wanted to express yourself in terms of building a company have you always been that sort of person or was it more that you went to bcg and learned a lot of the skills to be a founder and a CEO, like which side of the fence do you kind of lean on there?
1:01:31It's a good question. I think, unfortunately, I'm going to say both, but I'll try and justify it. So I did think I would end up building companies. I don't really know how, but I started my first one when I was, you know, while, you know, even before I was a doctor, to be honest. I just... What was that? That was the education company. And I don't really know where it came from. I just knew that I could do it and that I wanted to, I just had a certain amount of drive towards that. And not from an ego perspective, I'm a pretty low ego person, not from like, oh, I want to be able to say that I run a company.
1:02:07I just thought it would be cool to build stuff and that I wanted to give it, I wanted to give it a go. To see something that doesn't, didn't exist and then now does exist is quite exciting. but BCG taught me a lot about how everything else in the world works right when you're a medic you learn a lot about medicine you go very very deep but you don't have a lot of exposure to I mean how does a company work I didn't know what a marketing department was and I didn't know how sales you know those sorts of things are very different I've always I think that the quote that you you raise is quite fair I've always worked very hard and so it wasn't like BCG was a shock but I worked I worked hard there um I work harder now but but it doesn't feel like it right I I love what I do I I'm it's it's never a chore to be to be working on something where you can see that it's making a difference and where every single week you look back on the previous week and you think wow we actually done a lot that's crazy and also the kinds of people that you can hire and that you can you can decide to work within startups is very very motivating so you can choose to bring in the brightest the most ambitious the craziest the most risk-taking people and it creates um it just creates energy and it creates momentum that it's very hard to I guess stop even um if you wanted to so my journey is is definitely not one it's it wasn't one that I laid out deliberately where I said right I'm going to achieve these um learnings and then I'm going to be ready to be an entrepreneur I think it was a bit more organic I think I always had it but I did need the the time outside of medicine doing something completely different to to broaden my mind as to how work can you know how work is done um how hard you can work yeah and how other industries work outside of health care as well and final final question for 32co um who are you looking to get in touch with now like what's what's on the agenda and who that might be listening to this do you want to speak to I think one of the things that we're trying to do is is raise a lot of awareness for patients who have problems with snoring and sleep apnea one thing that we did probably underestimated was just how many people have this problem and how desperate they are for solutions i mean just look at reddit forums and you'll see but either the partners of the person or the person themselves so we've um so aerox is aerox health is the is the brand that that is our is the sleep brand um and that's hopefully a destination where people that have this problem can at least see if they'd be suitable right again coming back to we're never going to compromise on a patient's outcome if you think that you could benefit from it you will be told honestly and if not we'll help you find where you can be treated but um more awareness about how important it is to get treated and a diagnosis for something like sleep apnea and not to accept it as just well it's fine um maybe i'll just be tired and not really find out why these are life-changing treatments when you get them right so that was probably my main one for now for everyone listening website is 32co.com so three two the numbers and then co.com um sonia it's been an absolute pleasure it's been great learning about this i think you're tackling such a complex messy problem and you presented the solution in such a way that i think will have so much value i think now yes but down the line even more so um i haven't heard of many if any people certainly not doing it the way that you're doing it so it's a really interesting focus i think it's a it's a problem that you know you and i both felt throughout our careers that that same like i could just i could just solve this myself if someone showed me um and actually the coordination of care thing at the other end putting far more control into primary care i think these are all fantastic directions for healthcare so yeah I think it's awesome wish you all the best amazing thanks for having me James it's been a lot of fun hey everyone thanks for listening and making it all the way to the end of this episode remember to subscribe rate us and leave a review and you can head to the description of this episode to follow me on all of my social media so you don't miss out on any of the latest health tech content
From the publisher
In this episode, James is joined by Dr. Sonia Szamocki - former A&E doctor, BCG consultant, and now founder & CEO of 32Co - to explore how healthcare can finally unlock the full potential of its clinical workforce. Sonia shares her journey from frontline medicine to building a platform that helps generalists deliver specialist-level care, starting with dentistry and now expanding into sleep medicine.
Together, they dig into:
- Why modern treatments fail without the right flow of information
- How 32Co builds trust between generalists and specialists
- The hidden emotional and structural barriers to clinician adoption
- Why AI isn鈥檛 the whole answer - and where the human connection still matters
Chapters
00:00 Sonia's Journey: From Medicine to Entrepreneurship
02:59 The Role of Technology in Healthcare
05:56 Building a MedTech Solution for Patient Care
08:56 Challenges in Healthcare Adoption
12:03 Creating Access to Specialist Care
14:56 The Importance of Human Connection in Healthcare
18:08 Expanding into Sleep Health
20:50 The Complexity of Healthcare Structures
23:58 Trust and Information Flow in Healthcare
32:47 Building Trust in Healthcare
35:51 Innovations in Sleep Apnea Treatment
41:10 Bridging Gaps in Healthcare
46:01 The Role of AI in Healthcare
01:00:40 Entrepreneurship Journey and Insights

