#435: Hidden barriers to healthtech adoption with Saira Ghafur from Prova Health

17:42Dr Saira Ghafur:definitely see myself it's people relationships and behavior yeah all of that and actually and if you're not good at influencing those things then um it's a difficult place to be yeah like

17:57James:like slow moving corporate politics i don't like but flip that into a high stakes fast moving like then it started starts to become yeah i don't know there's a there's a political uh comedy called the thick of it very old yeah but um i don't know if you've ever seen the thick of it but if it's like that i'm in like you give me like a malcolm tucker character to to mess around with

18:21Dr Saira Ghafur:like yeah cool i'm in but um anybody listening to who's an mp james is out i'm interested i'm

18:29James:interested i'm interested i'll explore it i'll explore it again well as i say that suggests that i'd have any sort of skill required um which is probably a far leap at this point giving people dedicate their careers to it party british about this if you're in the u.s

18:48James:it's playing out in real time all of this isn't it oh god yeah in the u.s it's crazy isn't it i mean they had reagan they got trump they've got i mean now it's just if you've got any sort of level of celebrity schwarzenegger like you've got any sort of level of celebrity you're sort of like do you want to sort of govern a state do you want to sort of do you've got your own podcast so i've common podcast yeah yeah the next logical step is to run for office yeah absolutely although you know health secretary i mean i'm would i be a credible anyway let's not go down this route um back to let's talk about you um fellowships you did some fellowships and then what

19:26Dr Saira Ghafur:uh so it went back 11 months back into medical training to get my cct and i think that was probably the longest hardest stretch was that it's gonna be hard to go back to clinical and yeah be spoken to by tpd yeah okay and then kept getting asked what i was going to do when i cct'd

19:45James:and when was i going to get a proper job the belittling is like never ending in medicine

19:51Dr Saira Ghafur:i was like nobody's going to take you seriously if you do this silly of course yeah yeah if you do this if you do this what silly tech stuff and you know get real and yeah nobody really quite understood what yeah well the health policy stuff it was like yeah that's fine but um you know of course right you can write an audit another time and i was like that's not what health policy is

20:23Dr Saira Ghafur:wow but yeah it was yeah it wasn't particularly taken seriously so um and at that point i was people apply for another fellowship okay so then i applied to do a heartness fellowship um that's a proper fellowship proper policy fellowship which was really um so that was um so then i applied got it and then ended up when i did cct so i did make it to cct which was great um and then went to and then as before i cct acted up as a consultant for six months which was quite fun um and then i remember getting uh bollocking off the medical director on my last on call and then responded as to this is why i did it and they were like oh that's actually a really good point and uh do you want to join this committee uh i was like no because i'm flying to new york next week and i'm out of here goodbye nhs and so that was a nice way to leave yeah okay and then went to new york to do a fellowship in health tech um and policy which was amazing um i was a really actually i think that was probably like the pivotal year i would say okay why in terms of uh changing career direction right um again it was when you kind of like go back into clinical the noise kind of starts again and you're like oh what am i going to do now um and job wise and everything else but actually more than anything it was a complete you know escape from kind of nhs clinical what's next and a proper break from all of that but also what's next and what i do and just super interesting in a very different context environment learning about our new health system a different way to see how things are done because i think And, you know, one of the amazing things of working in the NHS is, you know, you know a system very, very well.

22:24Dr Saira Ghafur:And doing all of that, all of the different policy fellowships, getting to know the kind of system as a whole. And going to explore somewhere completely different where you're the newcomer, you don't know anything about it, asking stupid questions, learning new things, meeting new people. It was absolutely amazing. and then again so my project for the year was kind of looking at health tech consumer facing health tech and learning from different industries and so this was about nine years ago and how that landscape changed is amazing and actually I'm in Boston next week I'm going to meet one of my mentors um who suggested that we write a follow-up paper from what we did at the start and when it was all like fairly kind of nascent all the kind of consumer facing stuff and now literally we've got chat gpt and open ai and unrecognizable all of these things um basically they are your health tech advisor and on hand so how things have changed um yeah crazy and closing

23:33James:the loop though to respect your audit origins um ideal ideal um one thing i do just want to point out though is that you are still clinical do you still work in is it an nhs clinic that

23:46Dr Saira Ghafur:you still do yeah yeah so came back when i finished the um harness fellowship came back to um the uk um and then applied for a job to work at the institute of global health innovation uh imperial yeah um and lead our health tech policy unit um and work still work there uh with ara darsey um and at that point um affiliated with saint mary's and um ended up uh doing clinics there um so an honorary consultant at saint mary's um and then when covid happened i did actually go back in as an inpatient consultant as well, which was interesting to say the least after a couple of years out. But no, still practice, do clinics, respiratory clinics.

24:37Dr Saira Ghafur:And then we also look after the Grenfell population as well. So do that once a month. So it's, yeah, it's brilliant. Actually, I'm very, very lucky to be able to still do that clinical because I do still love it. Love it. an incredible um team uh the spiritual team there are brilliant super helpful and just um very good colleagues so it's great to still be able to do it and i think it just does keep it real yeah i mean you can write health tech policy you can do health tech but actually when you walk into a clinic and no bloody clinic room in the nhs is fully functioning wi-fi and somebody is trying to show you their results on their nhs app and we can get onto it it's really basic things like that where actually you're kind of like here we're trying to bring like the latest ai technology nhs but literally we can't get the basics right so yeah you're right it keeps it running it keeps

25:36James:you it keeps you humbled by what the problems on the ground floor actually are i think the i think the other thing as well is that i think i think fairly sometimes like i can get criticized for like glorifying leaving and like you know we've done a bit of nhs bashing here is you know the nhs isn't just one organization it's you know 100 000 different organizations and everyone's got their own problems and as any big corporate employer will have they'll have slow moving stuff they'll have people that try and beat you down in terms of promotions or is it like it's just another employer that that obviously has its issues as well and i think it is worth pointing out though that things do change when you're a consultant in terms of your time in terms of your sway in terms of your ability to have impact on the way that things are done in the NHS and actually even your optionality within health tech will I imagine look very different as a consultant as it would do in other places and so I don't think it is about glorifying leaving I don't think leaving is a good idea for people for which it isn't a good idea i think for some people it definitely is for everything that they want to do individually and what they think they can do in the system but obviously for you it it wasn't it was it was more important for you to stay so can you talk me through that decision because i imagine you know although you've know you just come back from the heartless you've got 11 months still to do or whatever like i get i get that that seems like a short amount of time and I also know people that have gone at STA and not and not cct'd and so you mentioned it was you know the hardest bit to do and so I imagine you know many people do still leave at that point what do you think the advantages are of getting to consultant for your options within health tech and and for everything that you've done and what does that decision mean to you personally to stay until consultancy and still practice so i mean i would say one of the

27:32Dr Saira Ghafur:i mean just going back earlier it was always a what else can i do alongside i i could never so that was always your frame yeah but it was kind of you know i still loved the job and you know it's such an incredible thing that you're you know you've been training for and it's not even training through medical school and everything else beyond is you know all the way through your schooling to you know get your a levels or hires or whatever is to be able to get to medical schools all of that um but actually just as a job it is incredibly rewarding and it's an amazing thing to do and did always you know the good bits were always good uh reflecting on that and still being able to do it i think and being able to do it safely i would always say um is you know it's an incredible privilege um and love doing it i think in terms of what it adds to my role now i think a keeping it real in terms of what i was saying in terms of you know whether it's policy whether it's actually health tech delivery any of these things and being able to have that insight but also from working with clients and things being able to reflect that experience on the ground but also your clinical network as well so the people that you work with you know all of that bringing all of that it's that richness of information the clarity that you get all of those different things and back to you know the policy writing if you don't appreciate that kind of behavior and the mindset that people are working in and you know the last couple of years have been so incredibly tough for everybody on the front line in terms of the NHS whether it's been covid workforce shortages you know junior doctor strikes all of these different things so being able to like appreciate that and how that impacts everyone that you're working with and then all of a sudden it's we're implementing ai in the nhs which is like a completely like bizarre way to kind of go about it it's just um you know it's it's it's tricky um like appreciating that kind of context um that we're working in and having said that you know i have my frame of context whereas you know if you're a full-time nhs clinician but if you're working as an inpatient um clinician all of that is very very different if you're working as a gp all of them have different challenges um i have my insight and i think it's very very helpful for all the work

29:59James:that I do yeah that's great with that in mind you spend some of your week on being a businesswoman as you wanted to when you were younger so you you did the thing so congratulations um tell me about

30:15Dr Saira Ghafur:that side of your life oh yeah so um so actually uh two businesses um so co-founder of um a startup called Saima which I actually run with my sister um and it's a mental health technologies um it's online therapy um and we started this um just as I was coming back from the US and my sister she's a psychologist so she's actually the subject matter expert so it was all pre the whole um online time um it was very quickly bashed down um and it was it's been a super interesting journey and i always like laugh about our kind of like early um fundraising uh tries uh that neither of us in terms of the whole kind of vc pitches it was very well versed in so um nothing like shame

31:11James:being an excellent teacher absolutely absolutely but i tell you what but i tell you what though i think getting yourself out there to to do the thing and be in the position to feel that based on some feedback good for you good for you because the amount of people that will just think about this and have the idea and write the deck but never actually push for getting the meetings even doing the work like how many how many people did you have to message or email or cold email to get the the intro to the intro to the intro that actually got the meat like if you're starting from zero like good for you for getting as far as you did no matter how shameful like those early

31:51Dr Saira Ghafur:days were i mean it it's an incredible achievement was it the the best feedback was um yeah word of

31:57James:advice uh don't give up your day job yeah wow wow and that just also goes to show you the point and

32:04Dr Saira Ghafur:And I definitely would take that on board.

32:06James:Well, you did action it. You did action that. So fair enough. But I don't know. I don't know.

32:12Dr Saira Ghafur:Actually, many years on and through COVID and everything else, we are doing very well. And it's grown. But that, I would say, is a tech-enabled business. We run that very well. Excellent evidence, excellent outcomes. um it's not something that we would probably not kind of have the all ai all kind of yeah

32:37James:losing startups so psychologists telemedicine solving access like yeah yeah i get it i get it

32:44Dr Saira Ghafur:and this this works well and then the other startup is prova health uh we work on evidence generation for innovation so it's basically helping to accelerate innovation um in health care and this was as a result of work that myself and two co-founders Gianluca Fontana and Jack Halligan we all met at Imperial where I do my policy work and we had started doing some of that work at Imperial. I was looking at more pragmatic ways of being able to generate evidence for tech and as we all know it's very very difficult to sit and do a randomised control trial for any new technologies. So it was thinking of better, faster, but more kind of equally as robust ways, showing evidence for these technologies and how you work on implementations with all types of evidence, economic evidence, and helping to accelerate all of that in the health tech space.

33:50Dr Saira Ghafur:And I think this has become even more relevant nowadays, and especially post-COVID where we saw such a massive increase in terms of the number of technologies on the market um so that is the other startup um great well I'll ask you about Prover actually because

34:09James:that value proposition has been I would say exactly the same since I very first entered what was healthcare IT at the time, where there were people developing new technologies that were convinced as the founding team would be amazing, but they couldn't get a pilot because they needed evidence. They couldn't get evidence because they needed a pilot. And that chicken and egg hasn't really been solved other than through relationships. and actually that was the advice that i used to give people i remember when i was at digital health.london i mean goodness knows how long ago that was now that um i found a presentation on my old laptop the other day and it was literally one of the slides was a chicken and an egg and one one line was trial the other one was there was evidence and so it was like to get as a new trial and like i can remember the only way to break at the time i'm presenting on it the only way to break this really is to is to build trust with a human being as a human being and actually go and spend time with that department those people understand them listen to them try and get them to trust you and eventually something's going to give where you're you're you're given an inch and you've got to try and take a mile and and try and get something done but then even then defining what the something is or what the something should be became difficult because then it was is that what like a 10 person patient reported thing is it a is it a double blind thousand person thing like what's the because ultimately what's going to be enough to then convince someone else to buy it and it was i and i don't think that has changed like i i really don't but what's your what's your impression of the lie of that land has anything changed and what broadly do you have sort of big big sort of headline advice for people at different stages or doing different things as to how to how to actually approach that i think i completely agree with that and actually my

36:18Dr Saira Ghafur:master's dissertation that i did was actually looking at where's ever it was looking at primary care apps that were all in the app store and literally none of them had any real evidence or medical backing for them so that was like and that was so then you know fast forward a couple of years on doing that work at imperial and then fast forward another five years on we celebrated our five-year anniversary at prova um a couple of months ago and you know we're still in that space and actually looking at the new llms like open ai and looking at um claude it's still that whole space where is evidence to show that these llms work that are being regulated um and it's always that sweet spot isn't it is like obviously regulated health technologies have got a very clear process but it's everything that falls in between um that don't have the evidence and more than anything you know if you're actually an innovator if you want to be on the market for more than you know six months before people catch you out and launch every two of your product you need to be able to make sure that you've got the right evidence for it and it's exactly i'd say exactly that you need those relationships from the get-go and I think that's where that whole you know it's behavior change relationships and influence that you're going to be able to do that pilot but my advice would be thinking about evidence right from the start um because actually the landscape has changed that every single person that is going to be involved in your journeys whether it's your VC, whether it's the payer, the provider, any of those people, whoever you're working with are going to ask you for that evidence.

37:59Dr Saira Ghafur:And you've got to quickly think right at the beginning, you've got to think, who's this evidence actually for? Is it for VC? Is it for the funder? Is this for the payer who's going to, you know, take on this interview? Is it for a provider? is it for a healthcare professional is it direct to consumer all those different people what type of evidence are you going to need and that's going to vary what stage is your product at and again that will change so if you're early on in the the kind of design of your innovation it's going to be user feasibility usability all those things whereas if your later stage is going to be clinical outcomes it's going to be you know safety it's going to be security all those different things um and then you know once you're actually out in the market it's the real world evidence and it's the the monitoring but also one of the huge things now you know everywhere that you look um cash is king but also cash is strapped everywhere so you need to be able to illustrate economic evidence for your solution as well where is the value for money for anything that you're creating and is it any better than the incumbent so yes it might be bright new shiny amazing but actually is it better than the kind of day-to-day um and beyond that is how are you going back to that behavior change side of things how are you actually going to implement this innovation and have you considered that so it's all of these different things um and it sounds like a huge you know like undertaking undertaking the beginning but once you start breaking it down into those like blocks as to who why what and when for um then it's much easier to start kind of collecting all the different types of things um and you know you're not going to go straight into a trial one's going to be you know where are you going to pilot it what kind of early user information do you have what kind of pilot information do you have all of these things and then when you do approach a pilot site um then you know you've at least got something to present to them

40:04James:sure what benefits so do you consider the health economic study and by the way i know that that can mean multiple things i'm actually quite interested in what those multiple things are actually like what's the what's the cheapest you know quick and dirty health economic study versus what is a really in-depth nice ready health economic study but do you consider the health economics as part of your evidence stack then is that something that needs to be thought about you know just as early and if yes which it is so that's a yes i think then one thing i know people struggle with is how do i get this done at a reasonable cost meaningfully that i can meaningfully communicate a message of cost saving cost reduction cost efficiency whatever how do i do that meaningfully that's but i'm early stage and so there's that tension i don't have the money to do a nice ready massive health economic study but what what is that first step actually that's not

41:02Dr Saira Ghafur:what you're expected to produce at the beginning it's going to be you know simple numbers that you be you're able to confidently present that will show and a lot of that is you know secondary research based so it's already numbers and figures out you can base your assumptions on um but in terms of kind of earlier stage um startups um what we usually recommend there's lots of kind of grant funding that's out there to help support some of this and then working with the health innovation networks as well so all of these resources that are available to people to definitely you know get in touch and then pretty much help signpost advise in that kind of early stage bit as well there's lots of resources that innovators are often not fully aware of the kind of help and support that's out there but also all the different networks that people belong to i mean you and i probably overlapped in so many different you know innovation groups and things but there's some like amazing people out there um who will always kind of come and help you and advise and all these different things but certainly i think that economic evidence you know in uh at this time and point i think is something that everybody or every single startup needs to be thinking about if they're not thinking about it um you mentioned open ai

42:26James:anthropic or you know you you mentioned what they're doing at the minute and the lack of evidence around it we don't even know we barely know what they're planning at this point by the way this is probably going to wait a few weeks where the whole world has changed again and there's probably new stuff come out but as we record this open ai and anthropic have released their health versions and that's kind of what we're talking about here the chat gpt health the claude health um it's really interesting i'm interested in your in your take on it because i my feeling on it is that i get what people are saying that okay everyone's going to upload all their stuff um you know i'll go with the open ai one because that's one i know better i haven't looked as much into the anthropic announcement but um you know they've said they've said it's not going to diagnose it's not going to treat but it's going to help you in your journey um we know that chat gbt is going to go some way to suggesting potential investigations you should prompt your physician to ask for and all this stuff so you can definitely get around it like it is gonna it is definitely a clinical decision support so i mean surely like it just is acting that way and you can use chat gbt for it now it's not as if they're about to release this thing you can do this now it's just they're clearly going to segment part of chat gbt to be purely for this stuff and and hone it in um so i understand people saying yeah they're acting as a medical device but they're not a medical device and this is not okay um i saw in their report that they brought out i even said in my commentary on it in in the newsletter health division i was like look look at page 13 they've literally just fired shots at the fda they've said we urge you to consider how you're regulating us or something like something along those lines like we we urge you to sort out your regulatory policy like because they're what they're basically saying is this is a freight train that is just bulldozing right through so either catch up and find a way or this is just going to be released and you're going to have to do something about it and we're going to tie this up in court for longer than you can like it's already going to be on version five by the time this even gets to to be judged so like that's the sort of vibe that i got from it and and in part like part of me there is a part of me that looks at their announcement looks at the video that they put out where they've got all these case studies of people in rural the rural us with chronic disease and they've not got access to available health care anywhere near them or anytime soon and they're using chat gpt to help them with a flare-up of a chronic disease and all this stuff and it is useful and it contextualizes those and and like the the big part of me is like i know you're not stopping this because actually it does add loads of value and i get the concerns and i get the issues and i get it but i'm playing that i'm playing both sides in my own mind and i find it very difficult because i evangelize a lot of tech i'm an optimist i i'm also pragmatist and realist where i'm like you're not stopping it so let's just rearrange and reassess what we're doing here based on the fact it's coming let's not waste too much energy saying it shouldn't be here because let's more talk about okay how do we now make it as safe as possible how do we do this how do we do so i don't know i'm interested because you've got a really good view of this you're still clinical you see patients you are in the you know regulatory and evidence base as well and so what are your thoughts let's be honest the horse is bolted

46:15Dr Saira Ghafur:yes this is and you know who doesn't use it i mean i'm the biggest hypondriac and i'll put in

46:25James:my symptoms as well and actually more from that data privacy side of things it's something that

46:30Dr Saira Ghafur:we touched upon earlier i mean if people find something that is useful and they get the utility from something people don't think twice about entering their data in no not at all because it

46:42James:also builds trust because it speaks it speaks natural language so you build you build a

46:46Dr Saira Ghafur:relationship with it but it's not it's utility more than anything else if i'm going to get a 10 percent discount off x then i'm going to put in some of my details to get that 10 discount it's utility but also if you're going to get help and you know that the trusted language all of that and it takes you x amount of time to see your doctor then of course you're going to use it but actually this is nothing new i mean how often have you googled your medical symptoms and actually google tracks everything that you put in as into a search and that's how they define their search settings and so this is nothing new this is just a kind of you know a build onto all of that and that's not regulated and at the end of the day what they're saying is a recommendation not here is what we're telling you to do however is then you know in the NHS you will be the gatekeeper is the clinician in terms of like what investigations that you will have you can also go privately but But also, again, you will have a clinician in other jurisdictions.

47:48Dr Saira Ghafur:Is it somewhat different potentially where you can say, I've read this on whichever LLM that you've been using and this is what I want done. Is that the most sensible thing? Is it the safest thing? No. So then it's that kind of, you know, how do you regulate that bit in between? And I think as that evolves as well, because this is, you know, version one. This is also the thing, yeah. you know give it six months what's going to be the next version is upload your ct scan and we'll give you a diagnosis based on that or here's a second opinion you know this is where we're then moving into very different territory um and it does need the checks and balances and the safety and it does okay so this is early on yes you know building on and based off stuff that already exists but actually the next step the next step after that what else can you do and i think you know that shift in the space of you know how long have llm's been out there for you know a year 18 months um and that huge shift in terms of what we're able to do um so i think it's

48:57James:that is so interesting isn't it what it let's sensibly assume that most so so many millions of people are going to use chat gbt health and anthropic health flawed health whatever they call it um let's sensibly assume millions of people are going to do that millions people are going to upload all of their medical notes they're going to

49:21Dr Saira Ghafur:upload their scans let's let's let's assume sensibly they upload all their radiology

49:29James:what do you think then about the potential for open ai to start acquiring chiron cured ai like all those like they'll they'll just start acquiring the algorithms that can that can actually diagnose does it does it then become a healthcare system of which the front door is the chat gbt app is that literally where we're going do you think i mean 100 percent or

50:00Dr Saira Ghafur:why would they why wouldn't they do that i mean at the end of the day that's delicious what's just

50:05James:come to mind for me why why wouldn't you do that of course all of these companies their actual value

50:10Dr Saira Ghafur:proposition they're tanking billions into it but actually how are they making money from it it's actually thinking of that like flipping it right into how are they then going to generate revenue from any of these things and is it your pocket doctor that you know you put in your medical data and it gives you for a monthly fee or whatever else how are they then going to utilize that or how are they going to utilize the data that they are that's been input or do they sell that back to x startup or whatever else or how how does that work but it's definitely going to be monetized in some way at the end of the day um because they need to actually show that value at the end of the

50:52James:day like thought experiment what if this is a good thing is there a world where this is a fantastic thing that's incredibly cheap health care that doesn't have long human defined waiting lists

51:06Dr Saira Ghafur:i mean it is democratizing health care to some extent um in terms of access if you're a you can't you don't have the deep pockets we are incredibly lucky low middle income countries etc we have a safety net if you're in the u.s where you don't have health insurance and medicaid and medicare and everything else is a bit of a disaster right now then actually you are going to use that if you are in low middle income countries and access is poor um you know again paying for something if that is the first thing that you've got and actually it's a reason doing you know it's obviously not as good as what you can now but if it's doing a reasonable effort then should we not be working with these companies to think how can we make it safer effective you know use it as a first point of contact for people that need it or for anyone that needs it but it's working with as opposed to actually let's talk we need to it definitely does need checks balances 100 % but how do you work with them to be able to deliver that better care as i said the horse is bolted so there's no point yeah i completely agree and it's just going to get more and more sophisticated um and be able to do more things but actually as our for people who are you know we are privileged for people who are not um how does that work one of the challenges as well i

52:32James:think with it is regulation is at odds with the race that is happening internationally as well people talk about the US versus China in this and for the US to clip its own wings by introducing a load of regulatory stuff that slows everything down does not help its cause in winning the race against china as well and so there's a tension we talked about on the pigeon podcast today that there seems to be this there's probably going to be this tension with okay open ai open ai is firing shots at the fda but you'll know this more than me i mean how close is the fda to government is it arm's length is it like an nhs england type organization is it oh it's if it is

53:30Dr Saira Ghafur:more as closer than the arm's length so so it's influence organization right and so at some point

53:36James:if if the fda is the blocker to open ai doing this thing with health and we know that china is building it is is there a moment where an administration like the current one all of a sudden a few people disappear in the fda you're like where have they gone um but all of a sudden it's all okay like i don't know there's the there's these really interesting very real we talk it's been a theme in this conversation you know the link between making change and politics how close is politics to health care and we've you know been in those meetings as national medical directors fellows like yeah we've been in those meetings where we've seen how close that is that will be the equivalent in the u.s there'll be there'll be that they'll be surely

54:18Dr Saira Ghafur:these conversations it's it's kind of doing it and then doing it properly isn't it at the end of the day they are going to have to be if you want something to work really effectively yeah um and people to trust it and to use it well i mean which one would you use oh here's you know the unlicensed version but actually here's the licensed version that's got the checks balances security safety in it you're going to go for it aren't you um i think it's with that building with that mindset um and whether that you know you push our responsibility responsibility back to the innovators uh or whether it's a kind of you know we're going to implement and impose this on you yeah and and it's working out that side of things but actually anyone and everyone you would use the tried and trusted version and if this is being adjunct to you know healthcare seeing a clinician at some point you know you need to make sure that it's something that is your you know in effect your licensed version of something um so i think that's the kind of lens you see it and not see regulation as a bad thing regulation is a good thing when done properly um and when it's not a barrier to innovation so it's kind of making sure that you're balancing that and i say this as a you know balancing this and you know done very easily effectively it's super tricky um and actually we're dealing with technologies how we don't that we don't know how they're going to evolve and what they're going to be capable of doing in six months and 12 months so that's the tricky bit is we are um regulating for the unforeseen yes at some points stuff that we don't even understand for all these black box models everything else like what are we actually regulating and black

56:05James:box strategy so i think this is one of the key points here that i'm glad that we've got to actually because this unlocked a thought of mine i think i think because of us both doing that fellowship and knowing the link between policy and politics and media and all the rest of it i think this is blowing this up so i think one of the big issues here i started health tech pigeon the last week on sunday i started it with like beginning like a dystopian film chat gbt health has been released and it was this very vague very very very vague announcement of oh this is what it is this is what it does and no chat around exactly what you've just said about clearly this is the next stage and this is the next stage and this is the next stage there is no transparency and that gives it the vibe of these dystopian films which start with oh here's this really cool new technology and ends with everyone's organs being harvested it's it's like what we have is like we have we have the very first step there like but we don't have a national ai strategy or an international ai strategy or even a this is what we want health care to look like and therefore this is the beginning but here's how it plots to that like we don't have that like it does it just doesn't exist and that's why this can feel very dystopian because look at us here just trying to guess as to what is going on they're obviously going to have loads of radiology well that means they're probably going going to acquire some algorithms that are regulated in which case well that means they can start diagnosing if they can diagnose then maybe that's going to be good maybe that's going to be a bit terrifying like what like we don't and we're trying to like figure this out and it's branching further and further out in terms like the limits of possibility well i mean that's

57:49Dr Saira Ghafur:politics i mean it's all geopolitics now isn't it and it's you know versus us chase but this is also geopolitics but also the cross-regulation of ai um in different countries it's just not happening so even that's part of the problem that's just turning into the most cumbersome you know like it's almost as a good thing the uk's slightly outside able to do it in their own way because actually to be able to do that in a way that's going to be agile and be able to move with the rapid pace of change is going to be tricky um at the end of the day everybody is looking to the fda because actually everybody goes to the us with their ai stock at the end of the day if you're going to monetize anything that's where you go um and waiting for the fda and what is the uk's role in it um you know are we the ones that kind of say we're safety and you know agile regulators but actually nhs data asset as well to train things on you know that is much more yes you know it's much better curated it's from a much more homogenous oh sorry not homogenous heterogeneous population yeah and clean curated working that way so actually is a safety and testing environment is this where the uk's um space is it's looking at it but it's going to be interesting we definitely

59:24James:do have value if we can uh if we just zip all of that into one folder and send it to people

59:30Dr Saira Ghafur:it's we're not competitors to the us or china but actually we've got a very different role to play this yeah and being able to you know the the health data research um service is you know a real you know this could potentially be the cap and the feather for the uk's ai strategy if done well um but and as a plug we've written a blueprint for well i was gonna say i was gonna say it does

59:58James:kind of lead us on nicely we've got 10 minutes left so like just to talk about the report obviously cyber security is is incredibly important here so you want to talk about what you've written

1:00:06Dr Saira Ghafur:Yeah, so this is, I mean, two research areas at Imperial that I work on. So it's the kind of health data work. So we've been working on that for the past, I think, eight years or so. This is the third in our white papers. And really focused it on basically looking at the UK's or the English NHS's health data strategy. and this white paper is really focused on kind of putting a blueprint for the health data research services and potential strategy in the way to go forward so um we've been working with a huge range of stakeholders um from across the nhs and academia and kind of commercial companies and patients public practitioners all these people bringing it together and putting out a recommendations and then the other bit of work I do is cybersecurity research which is again everyone's seen that the number of cyber attacks have happened in healthcare but globally in healthcare and again a geopolitics tool as well is like how can you hit a country hit its healthcare system in terms of a cyber attack to cause near-end disruption and basically extortion of

1:01:28James:money so um it's a blueprint for that so yeah as you said the health data research service which is this new national sort of front door for research access to uk health data um backed by 600 million half a billion from the government 100 million from the welcome trust aiming to go live this year in 2026 um problem being obviously the data setup's very fragmented um primary care data huge missing piece for what would be amazing longitudinal insights and so essentially the core bet here is that if this health data research service is designed as critical national infrastructure rather than like another portal it's going to unlock this in inverted commas triple dividend which is better care stronger nhs and life sciences competitiveness but only if it earns trust with the transparency the governance the visible public benefit and things like that um how you know aiming to go live this year how how this might be this might sound like a really stupid question to ask you but like how realistic is it that this

1:02:57Dr Saira Ghafur:happens and that it works oh it's i mean it needs to happen we can't afford for it not to happen

1:03:04James:So I would agree, yeah.

1:03:06Dr Saira Ghafur:And this is like, I mean, the NHS data, I mean, the reason that this needs to happen is, I mean, we religiously collect data across the board, whether it's like primary care, social care, secondary care, every single patient interaction, but not even patient interaction, every single interaction in the NHS. Unbelievable data. It's crazy. And if we're not using it to its most effective potential, then that's unethical. So why are we collecting the data in the first place? And if it's going to improve patient outcomes or find the cure for cancer, why are we not using it to its maximal potential?

1:03:46Dr Saira Ghafur:Obviously, safe, ethical, legal, all of those things, secure ways to do it. But we need to be using it. And in terms of what I was saying, we're not in competition with China and the US and everyone else. We have a very different role to play. This is a critical national infrastructure. Also, first and foremost, to be able to improve the health outcomes of people and patients in the NHS. You know, if we're not doing that, then or that should be the kind of basics and building up from that, using it as, you know, game changer in terms of delivery of AI and delivery of services and all of like the potential but also a net income generator for the NHS as well or for the UK and you know that is where we've always been and the UK has always been a front runner in terms of life sciences we're kind of losing advantage because it's just not being invested in as heavily enough as it should be um and to be able to get back to the glory days that's what we need to to really focus

1:04:58James:on yes agreed um one thing that i just want to pull up on this is that there's there's like a quite an exciting implementation roadmap and i don't think that sentence has been said by anyone

1:05:12Dr Saira Ghafur:in their entire lives but um i sort of regret saying that but it it is in the sense of there's

1:05:19James:there's a proposed timeline to 2031 yeah with with wins as close as 2026 this year um which is to appoint ceo team and board this year and to launch an mvp this year uh which would target 25 research teams um scale it deploy data sets into primary care and prescribing all this year that should happen and then with the longer term plan to fully integrate with genomics england and trial matching capabilities that's phase two and population scale coverage including gp prescribing mortality by 2031 that's phase three so it feels it feels very real like it feels very much like hold on a minute we're gonna have incredibly useful revenue generating nhs data capabilities this year i think it's not even revenue generating it's basically this is

1:06:36Dr Saira Ghafur:what it should be i mean caveat that with this is um our strategy not the hdrs strategy this is absolutely rubbish so um but they did appoint uh the chief exec yesterday was announced and so actually the roadmap is working perfect um but the mvp is there because that mvp is you know all of this is achievable because it's not designing something from scratch this is using existing data capabilities across the NHS and actually we've got a wealth of data expertise people across the country the geographical spread and it's making sure that everything is being utilized to its most effective capabilities and taking people along with and so actually that is all the potential to deliver is absolutely there and more than anything I think it's something that's not very nhs is having that ambition um and having a really strong ambition to be able to deliver this and you know that phase three by 2031 is completely achievable it's you know at the end of the day you need to make sure that you're investing in it the momentum's there you're training people up and you're pushing because otherwise you know if we're kind of oh here's another nhs endeavor where we tank a bit of money into and let's see what happens yeah that push has constantly got to be there otherwise you know you lose it halfway through that that five six year period and uh

1:08:07James:to to comment on a theme that's run through this that it stays out of politics as well

1:08:11Dr Saira Ghafur:yeah but you need the politics actually i i disagree in terms of the politics needs to be there because actually if the current government the politics isn't championing it as part of critical national infrastructure also in terms of that kind of it has to be cross-party though

1:08:30James:it has to be cross-party because if someone's champing it too much they'll be like oh we started it therefore if they're going to get the credit we're going to tank it so yeah that's where

1:08:38Dr Saira Ghafur:you're going to fair enough fair enough you're going to come in and you're going to our point

1:08:43James:of agreement would be cross-party then in that case yeah yeah um the cross-party agreement that

1:08:47Dr Saira Ghafur:we're keeping this yeah that would be nice this is a good idea this is a definite good idea but honestly we need this more than anything we all need this to work and you know we're all patients in the public at the end of the day indeed i hope it will benefit every single one of us absolutely correct in full support to the team who have the the excellent task of delivering now

1:09:09James:indeed um sorry this has been an absolute pleasure thank you um thank you i've learned a great deal i think you've got a fascinating career i think you're doing incredibly interesting things um You're involved in so much and yeah, fair play to you for constructing the career you have done. And I celebrate the fact you are still a clinician and a consultant at that. So congratulations to you. And I hope that inspires people to construct careers that are similar. If people want to get in touch with you and ask you any questions or just to connect, what's the best way for them to do so? I assume LinkedIn.

1:09:50Dr Saira Ghafur:LinkedIn.

1:09:52James:perfect we will put that in the show notes um syra thank you for joining see you soon thank you 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