Guest Bio:
Prof. John Dillon – Clinical Professor (Teaching and Research) & Professor of Hepatology and Gastroenterology at University of Dundee
List of question themes and full question:
- Medical school – What was medical school like for you? Did you always want to pursue hepatology?
- Confidence – Where do you get the confidence to tackle these big challenges?
- Dreaming big – Eradicating a virus is a big dream (And perhaps has never been more topical than now). How do you think we can encourage doctors to dream big?
- Shape of healthcare – Since you moved away from traditional model of the hospital as a treatment center (e.g. for IV drug users & Hep C), do you think medicine as a whole would be taking that sort of an approach a lot more? In a broader sense would healthcare provision get more centralised (i.e. hospitals) or decentralised?
- Eradication vs therapeutics – Your Hep C project highlighted the importance of researching prevention strategies as opposed to the more exciting realm of therapeutics. How important do you think the former are in the wider context of research?
- Problem solving – Do you try and solve problems by looking at examples e.g. similar problems that were solved in the past and try to adopt some useful strategies? Or do you start from a blank slate and use reasoning to lay out a strategy?
- New/underestimated diseases – Do you think there’s a disease that we are grossly under-estimating and could potentially make a big comeback or manifest in unexpected ways (such as Hep C)?
- Reflexive testing – Do you foresee reflexive testing becoming a more core part of medical tests? Where do we draw the line between treating numbers and treating symptoms?
- Automation – What part(s) of medicine do you think are still safe from automation?
- iLFTs – When you implemented iLFTs, did you have to change the actual processing systems in the lab?
- Integrating with existing systems – You mentioned the lab machines involved in iLFTs. I assume those were manufactured by a different company – how did you manage to integrate your system with them? Did you get an engineer from that company to come over or did you get their permission to alter their hardware?
- Intuition vs Data – Moving on to innovation in research questions, did the prompt for you to work on iLFTs come from intuition? Or was it a conscious, data-driven approach?
- Progress in medicine – Based on personal observation, it looks like, apart from hepatology, immunotherapy and some other hotspots where there seems to be a lot of exciting new work, the other areas of medicine seem to be relatively stagnating? Could you offer us your thoughts on whether it’s because these specialties somehow have kind of low hanging fruit that’s still left to be picked? Or is it something more to do with the approach to research and maybe some of these approaches can be adopted by other stagnating areas as well?
- Long time horizons – You’re quite ambitious in your work, yet at the same time, it looks like a very long time before you finally got to see results. How do you cope with the pressure of trying to make big changes where you don’t get quick outcomes and you don’t want to seem like you’re just promising big things?
- Personal influence – Do you have any personal influences be it scientists or non-scientists from different fields?
- Management – Could you share with us some crucial lessons you’ve learned about managing teams, especially in these kinds of large-scale initiatives?
- Wandering vs Aiming – Some more personal questions now: how important is it to ‘wander around’ as opposed to having a fixed focus goal in research?
- Applied vs theoretical research – You mentioned the distinction between applied and theoretical science. What are your thoughts on whether research must have a utilitarian application in the end or whether research for the sake of it is good too?
- Advice – Do you have any advice for people starting off right now or within next few years who are keen to pursue an academic career?
- Automation – With the rise in automation, do you think doctors at some point will take on the supportive and emotional care of patients while machines do all the algorithmic work behind the scenes? Or do you think there’ll be edge cases where doctors still do some of the decision making?
Podcast Transcript
Medical school
Q1: What was medical school like for you? Did you always want to pursue hepatology?
A1: I had been a fairly ordinary medical student. I hadn’t troubled the examiners on either end, either with prizes or with failures. One first encounter with a retake, but that turned out fine. I got through and then was a junior doctor floating around various hospitals in around London and then up through the Midlands etc. Was not working in teaching hospitals for most of my early training and so wasn’t exposed to research and had done very little of it at medical school. I then ended up in Edinburgh and decided I wanted to be a gastroenterologist/hepatologist because I was good with my hands and liked doing things. Surgery I found a bit boring, and I wanted to be a physician. So I applied to Edinburgh and got a job there as a trainee registrar and there was a newly appointed charismatic hepatologist, called Peter Hayes. He just had this enthusiasm for research. It made life more interesting if you understood why you did what you did and he would always ask those “why” questions on the ward rounds.
And so within a couple of months, I was doing research projects for him. I then became a research fellow, did a high degree and started writing MRC grants and things. So it started from this inspiration of sorts, to where I’d got comfortable knowing what was known about liver disease. So you’ve moved from that stage where everything is terribly advanced. When you’re a young medical student or junior doctor, you think everybody knows more than you do. And you get to the stage where “Oh, hang on a minute, no one knows the answer to this question. But, why not”? And that’s the bit about trying to get that question answered. And that makes the whole aspect of what you do much more interesting. Hepatology’s also an amazing specialty. I hadn’t had much exposure to it up till then and fell in love with it because it had so many questions. I probably have a touch of attention deficit disorder so the fact you can swap diseases and go from doing stuff on genetically modified cells and animals to humans and all spaces in between appeals to me as well. But I think that whatever you get interested in, there is a research aspect to what you can do.
Confidence
Q2: Where do you get the confidence to tackle these big challenges?
A2: When faced with an unanswered question on your own, it’s terribly daunting. I mean, why would you have the arrogance to think that you can answer an unanswered question? We go through medical school and medical careers standing on the shoulders of others and think that we’d never be worthy to take on that mantle. Then you realize a relatively ordinary step is the next step that you should make. When you see people around you in an enthusiastic group, who present to you their next breakthrough that no one else in the world knows about, and it’s not a huge step forward, but it answers a bit of the question of what’s another plank down on the bridge that you’re trying to build on and it becomes doable, you think “oh, they can do that. I could have a crack at that.” It becomes as ordinary as going away to learn anatomy or physiology or biochemistry or huge amounts of pharmacology. Designing the experiment to answer that question becomes as ordinary as that. So it becomes that sort of confidence. When you’re putting your first venflon in, it’s terribly daunting but then becomes entirely ordinary is because you see other people do it, and you get that confidence to tackle those things.
Dreaming big
Q3: Eradicating a virus is a big dream (And perhaps has never been more topical than now). How do you think we can encourage doctors to dream big?
A3: It’s about taking it on and thinking, “This is mine, I’m going to do something for it”. Hepatitis C was discovered after I left medical school. When I started formally training as a gastroenterologist having done sort of my general internal medicine training, a paper got published that they discovered what we call with non-A non-B Hepatitis (up till that stage). At that stage, we knew that it was a transfusion-related hepatitis, it was a mild illness and it didn’t cause any problems. It was a bit boring. At August 1991, I’d been a GI registrar for about five months and we had a waterfall of cirrhotics. One morning, someone said, “Oh, you should check them for Hep C”. Almost by mistake, one of our house officers was a bit too enthusiastic and requested it on everybody and a third of our cirrhotic patients who we thought were dying of alcohol related liver failure, had Hepatitis C. We suddenly thought, “Oh shit, this isn’t, you know, a mild disease. There’s a lot here that we don’t know about.” We started testing and discovered that 1% of the Scottish population was carrying it. At the same time, a paper about 20-year mark follow-up of a cohort of transfusion hepatitis from the early 1970s who didn’t have Hepatitis B (when Hep B got discovered) found that they were dying of cirrhosis. So it suddenly wasn’t a benign disease and we had very personal ownership of that disease. When I arrived up here, as a consultant, there was no one doing any Hep C work and Dundee had a particular problem with Hep C because of the drug use, etc. I thought we should do something about it and it spiralled from there. This was now my bread and butter clinical workload because was the ward was filling up with patients with it, and so the question of how we tackle it. They weren’t your usual patients. I had to go and learn a whole new vocabulary, new places to go and deal with people in very different ways. I had to take advice from colleagues in addiction psychiatry and advice from patients because I was going to chat to the people who are coming from a very different lifestyle choice. These brought a whole new realm to my research, which is ironically, what got me noticed first, because of the unusual approach. But it involved using those same research skills I developed (physiological measurement and wet bench chemistry, etc.). It was applying these to a whole new group of patients where it was as much about the drugs and the pharmacology as it was about the patients and how you design services to interact with it. Doing that sort of research is actually quicker and easier. And it’s the sort of research that everybody can do; asking yourself that question when you’re sat in your outpatient office or your surgery or wherever and then people don’t turn up, and then ask yourself why don’t they turn up and finding out from them and then answering those questions and taking it from there.
And then there was a moment in time where we were about to do a series of lectures about Hepatitis C, and I was chatting to David Goldberg, who is a Professor of Public Health and works in Public Health Scotland. He just reviewed a paper from a group in Bristol, who are mathematical modellers. And he said, “I’ve just seen this paper and I can’t…I can’t believe it’s that simple. But he’s talking about eliminating Hepatitis C, by treating active drug users. And they reckon you need to treat about 20 per 1000, to get the thing eliminated in 10 years.” Shortly afterwards, it became available online. I was reading the paper and my hairs were standing on the back of my head, because we were already treating that number of drug users that the paper said was impossible unless there were new ways of accessing these patients and things. But we’d done it. And we were accessing the patients, largely because we started using dried blood spot testing – literally stabbing the side of a finger to squeeze more blood out – because up until then the problem has always been venepuncture. Guys who are actively injecting, if they had a vein, were not going to let you anywhere near it, because they had a good use for it and they didn’t want you taking blood and causing a hematoma etc. so we had developed dried blood spot testing. This also meant instead of doctors and nurses, we could train needle exchange programme workers to do it. They were dead enthusiastic, because it meant they could change their relationship with the people that were there from “Here’s your needles clear off, give me your old needles, etc.” to “Have you thought about other health problems? You might have Hepatitis C”. The guys who were injecting were well aware of HIV, Hep B but to a lesser extent Hep C. So they were chatting about that. Suddenly, all the bits of the jigsaw were lying on the table. And then I was down in London at another meeting, chatting and someone pointed out to me Matt Hickman and Natasha Lawson, who were the two people that have written the Hep C mathematical modelling paper. So I got a cup of coffee, and went up and sat myself beside them, and started a conversation. 10 years later, the three of us have eliminated Hepatitis C in Tayside. We’ve got multinational grants where we’re trying to do it; in Scotland, across England, and we’re in the process of trying to replicate that in countries across southern Africa. Those random conversations and just suddenly realizing what you’ve got makes a difference. And there’s a rush, I suppose, in getting the results. There are some scientists who are incredibly focused on having the next 10 steps worked out. But sometimes you’ve got to see which way the pieces are falling, recognize the patterns and think “Ah!” It may not end up the way you thought because we were still using interferon in those days and then Gilead arrived with Sofosbuvir, and suddenly, we could cure 97% of people in eight weeks. So rather than, 12 months of interferon injections, and only curing 50% of the patients, we could suddenly do it, which is why we accelerated from thinking we might have it done by 2025 to having done it in 2020. It’s sometimes having that flexibility, seeing the opportunities and the lucky accidents with these things. That makes you realize that research is actually easy, because sometimes it’s just you spotting your opportunity, and it arrives in your lap. I suppose that goes back to the old saying that “fortune favours the prepared mind”. The more I practice, the luckier I get. It’s those things that you recognize the opportunities on what can be done and suddenly see that it is easy, and it’s actually a series of small steps. But once you do it once it becomes easier to do it again, as you start to see the commonalities of how you can do things.
Shape of healthcare
Q4: Since you moved away from traditional model of the hospital as a treatment center (e.g. for IV drug users & Hep C), do you think medicine as a whole would be taking that sort of an approach a lot more? In a broader sense would healthcare provision get more centralised (i.e. hospitals) or decentralised?
A4: I think you’re touching on a very important point. What is the shape of medicine and medical practice to come? We’ve shaped medicine, because what we were offering was a limited resource and comprehensive, and therefore we concentrated in certain areas. So we’ve moved away from having lots of peripheral hospitals. You’ll be aware as you go through the towns, in Tayside, for instance, every place had its own Royal Infirmary or its own formal workhouse that became a cottage hospital of some kind. Things were very local. And that was because people could travel three or four miles and that was their distance if you had a slow railway train etc. It became complicated now because of motor transport and we have people who are doing distances in Tayside for a single five minute outpatient appointment of 150 miles. If you think of the people who are coming from Crieff and Comrie, if they’re doing a round trip to Ninewells and back, that’s a big distance. Many people, who are ill/frail, or not feeling great, wouldn’t really want long journeys. There’s also an element of fear, coming to a big strange hospital; you don’t know your way around, you’re stressed, you’re coming to see the big doctor who’s going to potentially tell you bad things or good things. There are loads of TV stereotypes of doctors, some good, and lots of them pretty awful. So people have that perception of not feeling great. Having got a little older, and on occasion being on the other side of the table it’s very different even though I know my way around, and I know I’m going to be treated differently because of who I am. It’s scary for people. So I think that’s the situation that we’re in.
If you then think about the diseases that we’re commonly dealing with; obesity and alcohol and smoking account for the vast majority of all preventable disease we deal with. Now, bringing someone up to hospital to deal with obesity, alcohol, or smoking in the first instance is not the answer to anything. I think taking that back to a community solution in terms of diet education, exercise education, smoking cessation helps and we’ve got the public health measures in place for that as well. Intelligent liver function testing, for instance, is about putting the pathway back into primary care and putting it into pre-primary care, which is something I think we need to start developing and thinking about. E.g. how do you just identify people rather than medicalizing it and giving it a pill/operation, which is one direction of travel for obesity? Isn’t it criminal that we’re inventing operations to make people mal-absorb food so they can carry on eating it when a third of the population is still starving? And equally in our own society, we’ve got footballers having to tell us how to look after our kids because they’re starving during their school holidays because they don’t get a free school meal. And we’re bypassing bits of people’s stomach so they can lose weight. So I think we do need to change that model and think about the research of how we do that and how we break those moulds that we’ve set up . Now, that doesn’t necessarily mean it’s the wrong model of care. For the rare and the unusual, it’s probably the best model of care we’re going to have where you do need to refer to a specialist sitting in a big ivory tower. But we need to think about how we improve the general health of the population (i.e. that pre-primary care). I think a lot of what we’re doing now is trying to de-medicalize risk factors so that we prevent those presentations and social prescribing is one of those interesting things where you get your diagnosis and the labelling from your general practitioner, and perhaps the stern warning of, if you don’t do something about this, this is where you’re headed. You then get referred to a social prescriber who will guide you to the things within the community that you can engage with, which will help you change your lifestyle, e.g. exercise classes, educational resources about food, etc. Because it’s amazing how ignorant people are about nutrition. I mean, let’s be honest as medics, we get precious, precious little teaching about it. But we’re hopefully quick enough to pick up what should be healthy. But most of our colleagues in the general public don’t get that level of education, don’t really know what it is they should or shouldn’t be eating. You can imagine doing research in that area. You don’t need a laboratory. You don’t need deep science. You are experts at it, because you’ve sat there and watched those consultations happen. Seeing people being dragged up and knowing how far they’ve travelled, and thinking about all the things that are going through their head, and you’ve seen the end stage consequences of not intervening earlier. It’s how we can then come up with better ways of delivering it. Now, what we know is that presenting to a general practitioner, getting a shouting out or a telling off from a general practitioner will change behaviour and alcohol or obesity in about 25% of people need to find something that helps those other 75% of people who can’t, or want to strive but fail and need supporting along the way.
Eradication vs therapeutics
Q5: Your Hep C project highlighted the importance of researching prevention strategies as opposed to the more exciting realm of therapeutics. How important do you think the former are in the wider context of research?
A5: Quite important, but if we did it, we will cure more cancers than the cancer drugs cure, because you’ll have prevented them from happening – obesity, and alcohol and smoking or all pre-malignant conditions and prevent those. If we reduce the average BMI of Britain by 25%, we would halve the incidence of breast cancer and colon cancer. People just don’t realize it, but let’s be honest curing cancer before you got it is a much better option than having a yard of colon taken out or being exposed to some multi-kinase inhibitor. Now we’ll still need those things. But there is this glamour thing about the research. It’s interesting, the intelligent liver function testing stuff, which is all about new pathways of care, we’ve picked up 11 awards for that and have travelled around the world to pick up some of the awards and it was just nitty gritty, reprogramming a lab management system that got all those things. So there are recognitions for that applied nitty gritty research, rather than rushing back from the lab with a test tube. I mean, I’ve had the joys of rushing out of the lab with a test tube that’s just given me a positive result. But I think about 10 people read that paper. So there are rewards in it all and it does get glamorous and new. So more people have come up to me and said, “Wow, that iLFT stuff’s brilliant” because it makes a difference to them in their day to day practice, rather than this sort of obscure enzymatic stuff of having found an inhibitor for NRF-2 or something which I think is really important, but hasn’t attracted quite the same enthusiasms etc. If you go to that concept of the doctor as a researcher, or the student as a researcher, all of us can do that type of research. All of us can pose those questions as to how we should do things better. “How should we improve things?” If you have every single one of us thinking in that way of “How can I help?” “What can I do to change and improve my service?” That’s research. That’s what we should all be thinking about. Because I don’t think any of us would accept that what we do is perfect. And sometimes you come away from having had a consultation thinking, “Gosh, I really didn’t help that person” or “I didn’t add much to that. But I put them through a lot” or “The system put them through a lot. There must be a better way of doing that”. And I’m sure you’ve had those experiences as medical students already. And you’ll have a lot more of them as a practitioner, and it’s about having a response to that. “Right… What can I do to make this better? How can I improve the system?” Because, as a doctor if you treat the patient in front of you, you make that patient better, hopefully in some way. If you change the system to make it better, you make it better for a whole lot more people, people that you may never see.
Problem solving
Q6: Do you try and solve problems by looking at examples e.g. similar problems that were solved in the past and try to adopt some useful strategies? Or do you start from a blank slate and use reasoning to lay out a strategy?
A6: I’ve done both. So I’ve sat there on a blank sheet of paper and worked it out. And about half the time I discover someone else had already done it. So I’ve got better at searching the literature for other people’s ideas, using that and adapting them. Looking sideways where if you have a problem, thinking of a different disease that might have similarities and looking to how they solved that can work too. Between the different blood-borne viruses, for instance, it’s very common practice that you move between two of them. I strayed outside my field fairly commonly looking for other things (e.g. into psychiatry and psychology and addictions, medicine, etc.) because of trying to look at patients from the behavioural and drug usage point of view. That’s changed the way I look after people who have obesity issues and alcohol-related liver disease because it changes your constructs about the way you deal with them. At big international conferences, I also make a habit of going to the sessions that I know nothing about. So that at the end of it, I come away usually with some ideas about things I already know about and different ways of tackling them and a little bit more knowledge about that unfamiliar area.
There are a lot of bright people around who are thinking about these things. Trying to keep up with all the literature that flows around them now is easier with the internet but equally the internet has led to a lot more stuff being published. So it’s trying to find someone who’s already done the thing you want to do. But at times you’re truly standing on the edge of something completely unique. So it’s a mixture of both. But it’s really worth being good at literature searching and adopting lateral thinking.
New/underestimated diseases
Q7: Do you think there’s a disease that we are grossly under-estimating and could potentially make a big comeback or manifest in unexpected ways (such as Hep C)?
A7: So within gastroenterology when I was a medical student and was training there, we were told that ulcer disease was due to stress. Huge numbers of people were in hospital with GORD and gastric ulcers and that’s what the “cause of it was”. We were told that non-A non-B Hepatitis was a disease that was mild and didn’t cause a problem. And we were told that there was alcohol-caused liver disease and there was no other liver disease associated with anything. So we found H. pylori, duodenal disease disappeared, and it was all due to a bacterial infection, which made us look pretty stupid. Hepatitis C, 71 million people around the world dying of it, but we’ve got a cure coming. With non-alcoholic fatty liver disease we didn’t believe our patients when they said they didn’t drink. But you’ve got a liver biopsy that looks like alcoholic liver disease. And alcoholics lie. “So you’re a liar, you’ve got it”. And for 20 years, we didn’t make the diagnosis as the obesity epidemic took off, and we didn’t see the liver disease coming.
So I’m not going to tell you that there aren’t some other things that we’re going to find. There’s one theory going around that obesity and the inflammatory obesity that we see is caused by a virus infecting adipocytes. Not a theory that I’m particularly convinced of yet. But they’ve got a putative viral sequence they’ve extracted from adipocytes causing the hyperplasia reaction in the adipocytes, they produce more adipokines, that causes the pro-inflammatory condition that leads to type 2 diabetes and fatty liver disease. If obesity was a virus and we’ve been telling them you have to have eaten stuff, perhaps we might have been wrong. “It was a virus after all”. So think of all the rubbish we’ve probably taught you at medical school, and how much it’s going to be rubbish in the future. In terms of my 35 years since I graduated, all of the drugs I prescribe on a weekly basis, weren’t licensed by when I left medical school. So that’s what’s going to happen for you lot in the next 35 years. I mean, it’s just incredible the potential for what you might find. So how we split the diseases up, e.g. the irritable bowel syndrome that we lump things together with may become multiple different syndromes of autoimmune destruction of various nerves in your gut. How else will things change over that timeframe? Very different ways of diagnosing and treating patients I suspect, and I think that’s something that will have to be rigorously evaluated to make sure that we don’t get wrong, because the way we practice medicine at the moment, has worked for 200 odd years, and has been effective. And we shouldn’t throw the baby out with the bathwater. It needs adapting and it needs changing to cope with different things. But it’ll still have that fundamental, taking decent history, examining your patient and thinking about what has to be part of that core, “how we think about things”. But how else we can automate diagnosis, automate some parts of it all and take it from there? And things like COVID. If anyone thought the one man couldn’t make a difference, they’ve never met someone who ate a bat.
And it’s not only the changes in dealing with COVID, it is changing the way that we’ve had to adapt to working on the front line with a dangerous infection. I had colleagues who dealt with Ebola in Africa, and they had to deal with a much more lethal virus than COVID. But it’s still risky. You’re going in and doing aerosol-generating procedures on somebody who’s COVID positive. That needs research and answers about how we move things further forward. And then if we’re not seeing patients as regularly as we want to, or how we have done in the past, how can we improve the quality of care? Do we need to see them face to face? Teams and zoom? Can we safely see many of our patients via that route? We’ve started to develop that and we need to evaluate it and take it forward. So those challenges are going to be out there in terms of how we get used to COVID. Hopefully we’ll have an effective vaccine early next year as the trials come through on that. But I don’t think we should necessarily lapse back to the old normal and we should be thinking about what we’ve learned from this.
Reflexive testing
Q8: Reflexive testing – Do you foresee reflexive testing becoming a more core part of medical tests? Where do we draw the line between treating numbers and treating symptoms?
A8: Yeah, I think there’s a line in terms of the ER, TV American approach where you have the patient arrive in the door, and it’s the chem-7, FBC, CT scan, then I’ll think about the patient, rather than the approach you’re taught; your history examination, select your investigations. So I think there’s a lot to be said for that. But once you’ve thought about that investigation, so in iLFTs’ case, you’ve got abnormal liver function tests, you know what you’re going to do next, but you then send the result back to the GP, who has to take more blood from the patient, further trip etc. And then the old algorithm was you repeat the LFTs again, to see if the abnormalities persisted. So six trips for the patient assuming the consultation and blood taking are separate, which they often are. But you’re all going to end up doing that screen. We then discovered that when you looked at industrial levels of liver function testing, say you’re living in Tayside, in the last 20 years, you have a 66% chance that someone’s checked your liver function tests, I mean, we have managed to screen the whole population almost. And it then became common sense to think about reflex testing. Because you have an abnormality, the abnormality is going to persist in 80% of patients if you repeat it in a couple of weeks. And even if it doesn’t persist, and has returned to normal, you could have set loads that return to normal still have the disease, but it’s just going up and down like the waves on the sea, and you’ve just caught a low trough. So actually, that patient needs investigation. And so if you can reflex it on the train track systems that we have and use the technology of these existing systems by reprogramming to do it, that’s a real step forward. And so there’s lots of other diseases where there’s a set of biochemical investigations, which will dramatically change the way that you follow the patient up that you’re going to do anyway. And why not just do them on that first set of bloods, rather than waste the time of the practitioner and of the patient going backwards and forwards. I think, if it makes sense, we should bundle these things together, if you’re going to do them anyway. And you’ve already fulfilled the indication. I think routinely bundling everything together and doing the whole lot is something we need to tread more carefully. The big challenge is going to come when we can now genome sequence. We now have multiple genes that clearly are associated with illness that you can screen for, but there are many more genes that moderately change your risk of developing certain diseases. But if you have 20 of the different SNPs that are concerned with susceptibility to type II diabetes, etc. And you’ve had that genome measured once, and it’s stored in a central lab, and you present with other abnormalities in your blood tests, you could then use that bank of genetic information to change the probability of you having X or Y diagnosis. So that’s an interesting way of going forward where we start to use all of that information. And as humans, we think in yes, no. we can probably deal with an algorithm that has a series of three or four outs from it rather than 15, or 20. Our brains can’t cope with algorithms that complicated. But computers can and if they can then give you, rather than those diagnoses that we come through is through algorithms where you’ve got three or four steps that computers can go through 20 steps. And at each step, there is 20 different options to give you a probabilistic outcome of what the diagnosis might be based on the subtle variations in the biochemistry in the interaction. So at the moment, if we take liver function tests, as example, you look at the alt, the alp, the bilirubin and the albumin, and you think of them in isolation. But if you start to think of them as interactions together, where they change their significance by if the ALT’s a certain level, in combination with a certain level of ALP, there’s a few in one direction, you start to use machine learning and look at those patterns that are too complex for us to recognize, or that we sort of recognize but can’t quite get into hard rules. If you talk to an experienced hepatologist and get them to look at LFTs they’ll start to suggest things Oh, that subtle combination. That’s up and this is down so I might think about this being more likely. And they’re normally right because they just sort of seen that patterns often. But they don’t have the hard numbers for it to be laid down as a sort of above 55, below 55. This is the cutoff. It’s a trend and that sort of subtlety that they see which machine learning can help sort out. You can then put genetic influences across that as well. You start to make that biological feed-out look much better. And if you’ve got previous blood tests on our patients, and start to look at deltas and deviations from standard deviations of the previous results, you get a whole different level of evidence. I think we’ll start to see that sort of thing coming to fruition. People have been talking about artificial intelligence for the last few decades to my knowledge and there’ve been lots of false starts. But we are getting more sophisticated about the way we think about it, and that’s the sort of stuff I would expect. You’ll have much more interpretation. Now where the doctor fits into interpreting all of that? At the moment, we’ve largely been sorting out those grey areas and hopefully, I think that there will still be more shades of grey and more need for us to interact with that, but there’ll be interesting changes to our role in diagnostics.
Automation
Q9: What part(s) of medicine do you think are still safe from automation?
A9: Well, deciding what a patient’s description of symptoms translates into in terms of your differential diagnosis will still be our role. I think we’re safe on that one. I can guarantee you at least 40 years of safety on that one.
iLFTs
Q10: When you implemented iLFTs, did you have to change the actual processing systems in the lab?
A10: Yeah. So there’s the laboratory Information Management System, which is a bit of software that sits between ICE, which is where you order the tests. Then there’s the train track system that controls the samples, tracking around all the machines, etc. And then there’s the limbs, which sort of take all the results and move them backwards and forwards and presents them back in ice so you can see the results. It’s that limbs system, that was the bit that we put all the algorithm into.
We designed the algorithm in terms of logic maps for a human to look at, and then had to change that into logic for the computer to work with. We are fortunate that the limb system is very old in our hospital, and so it’s entirely hackable. While we don’t like to say we hacked our system, we basically reprogrammed it because the system is primitive. If you moved a machine around inside the lab, you had to go in and manually reprogram the system to tell it where the machine had moved to. You could then tell it what to do with the samples and where to direct them. This meant that you could then start to feed back into it because it saw the results as well and that’s how we managed to manage to create dummy results: the positive blood results came back into the system, it triggered a whole series of other things and we reprogrammed that cascade. In more modern systems, you’d have to pay the software engineer that designed those systems about 30,000 pounds to do the reprogramming. So we can share the logic and the algorithm with others, but the more advanced computer systems need reprogramming because their software is more sophisticated. We’re now trying to make that happen across the world.
Different companies and healthcare systems have different reasons for why they’d find this useful. In the UK, primary care clinicians love it because it diverts workload away from the screens and towards the patients who are most in need of being seen, and it can appropriately highlight those patients for lifestyle intervention etc. The hospital’s like it because it also reduces workload. The labs like it because it makes them feel a bit better that they’re doing stuff. if you’re in a country with a privatized healthcare system, the laboratory companies love it, because suddenly they can offer a lot more tests to their customers. To every patient coming in they can say, we can do all this for you, we can save you several visits, you don’t have to pay for each time you go to see your clinician. We’ve had large laboratory companies in Russia, North London, etc, who are keen to set this up. In certain other countries in the world, doctors hate this, because that’s six clinic visits that they are no longer able to bill for. So we’ve changed millionaires into not being millionaires at all. There are some countries where I’ve been invited to speak where you see that they don’t want to hear anything to do with that intelligent liver function test.
So it’s interesting how the different healthcare systems and the different beneficiaries work, etc. Obviously I’m exaggerating slightly, but on the whole people are positive about iLFTs because it improves the quality of investigation. The big problem is that most people don’t get investigated after their first set of LFT abnormalities and get lost in the system – iLFTs is a quality improvement tool, because it makes sure all of the patients are investigated, and flags them up in terms of what needs to happen to them. So it’s a win win situation.
Having done the research and proved it works, now our challenge is to find the right levers to get it implemented widely, which is a completely different skill set and requires thinking about different ways of doing things. In the beginning, it was me and Ellie Dow, the clinical chemist who I developed all of this with and we were working together in our two services, which meshed closely together, and so we overcame all our problems. Now, we’re having to tell somebody else in another hospital in another lab, this is what you should be doing. And then they’ll have to find a company that will do the donkey work for them, trying to persuade them, writing the business case to get the money out there, etc. Of course, there’s also other people who will say they can come up with a better system because they want to come at it from the research angle, and that’s another challenge.
There’s a completely different set of challenges in terms of that implementation component, and I’m learning new stuff as to how to make that sort of thing happen. It’s that more entrepreneurial skill, and perhaps we need more doctors with that slightly different mindset. Perhaps some with less focus on the why questions and instead more focus on the how questions. How do I make that happen? There are plenty of doctors who have gone on to become excellent businessmen, in terms of the way they run things, and perhaps we need someone with a bit more of a take on that aspect of it that will make iLFTs really take off.
Integrating with existing systems
Q11: You mentioned the lab machines involved in iLFTs. I assume those were manufactured by a different company – how did you manage to integrate your system with them? Did you get an engineer from that company to come over or did you get their permission to alter their hardware?
A11: With regards to the machines that do the analysis, and the program that moves the track systems around, we had to put one extra bit of software to link with haematology because we needed the platelet count and that was on a separate part of the system. So, we needed that fed back into our limb system here, because it was a different management system that was running that. The software that we were reprogramming didn’t belong to Siemens, but the machines it controlled did, and we’re now in negotiations with them to try and build it into their machines. If it becomes a standard offering for their systems around the world, this will be a big breakthrough if we can pull that off. That project has unfortunately become a casualty of COVID but hopefully, when this madness settles down, we’ll get back to the other part of the work.
There are other systems too, like ABA, which make decision support software and laboratory instruments – they’re very keen on the decision support software to help make the most out of that laboratory information in a sort of way we just talked about. They’ve got a more entrepreneurial approach to laboratory management, where they’re identifying ways that you can save your hospital costs by bundling tests and changing the way the test results are fed back to the clinician. In fact, they make the claims that if you buy their decision support software, they will reduce your hospital costs by 25% by virtue of moving patients through the system more rapidly and getting them out of hospital more quickly. Our algorithm could fit within their decision support software, but they wanted us to buy their software.
There’s lots of places that the algorithms could sit into because there’s different ways that people have organized those different systems. It could be the machine, or it could be the software more closely related to the machines or it could even be the software that controls the whole lab. Through this project, I’ve learned a huge amount about the fantastic technology that sits within NHS laboratories that don’t really think at all about, you know: what happens to the blood that zooms up the tube and then comes back half an hour later on the computer screen in front of us as digital results?
Intuition vs Data
Q12: Moving on to innovation in research questions, did the prompt for you to work on iLFTs come from intuition? Or was it a conscious, data-driven approach?
A12: iLFTs started from a realization I had one evening when I was in the high dependency unit waiting to take a bleeder to theatre. It was eleven o’clock at night and the surgical case that was in theaters before me was overrunning and so in the meantime I decided to go back through the results as this was the first presentation of cirrhosis and what turned out to be a variceal bleed. As I was looking back through the trail of liver function tests, I could see the abnormalities and that nothing had been done. It was then that I had the realization that I’d seen this type of case dozens of times, where there’s all of these abnormalities being recorded and opportunities are being missed where we could have intervened. I have often had that conversation with patients too, where they say “If only someone had told me earlier, I’d have stopped drinking Doc”, or even with the relatives often after the patient had died.
Following on from that there was an opportunity to look at the epidemiology of liver disease and to get a better feel of what the epidemiology was. As a side output from that project, we went looking at abnormal liver function tests and what happened to them. We looked through the databases for the sequence of LFTs before first presentation in hospital. There were years of them, and every year or two the repeat LFTs would be still abnormal, and no one would do anything about them. They’d have that abnormality, and nothing happened in over 50% of the cases. That’s where the idea for iLFTs came from.
It took us several years to get our head around strategies to tackle it and algorithms to help, because liver disease is complicated. In the end we had 32 possible outcomes from combinations of abnormal liver function tests and screens, etc. so it was a case of locking enough hepatologists in a room with hot coffee and chocolate biscuits to get them to agree on ‘if you had this, this and this in your LFTs, what would the diagnosis be?’. We had to build alcohol and obesity and the metabolic syndrome into the equation as well. The hepatologists were allowed three clinical facts: (1) BMI, (2) presence of Type 2 Diabetes, hypercholesterolemia and high blood pressure, and (3) alcohol intake. We gave them different combinations of blood results and other corroborative tests, and then recorded what diagnosis were they prepared to make. It was a Delphic processes with mass disagreements to start with but everyone finally came to a consensus that if there were certain levels, this was safe, and that was safe – it was a matter of pulling it all together and then converting that into an algorithm that could be computerized which was another challenge.
It all started from noticing a problem my patients faced, so I was very driven by that. I know I can make a difference to people if I can make that diagnosis early on. It became clearer later on in our research that the problem with liver clinics where we had year long waits to get to clinic just to get an evaluation done, is that most of the evaluation was actually dead simple and could have been conducted in general practice or through iLFTs, giving us early diagnoses and reducing the need to bring patients through to clinic, which would create more capacity for us to see people who are sicker. The downside is that this makes clinics really heavy now because whereas before, we used to have lots of people coming through who were very well, now every single patient is in liver failure. Clinics have suddenly become really hard work, but that’s a good thing. So that’s you can see that that’s the pathway. It’s that being having those research skills to I’ve got a question here.
It’s all about having the research skills to take a question and unpick it, and then having the confidence to take on the opportunity in the moment when you realise that there’s a problem, and you can do something about it.
Progress in medicine
Q13: Based on personal observation, it looks like, apart from hepatology, immunotherapy and some other hotspots where there seems to be a lot of exciting new work, the other areas of medicine seem to be relatively stagnating? Could you offer us your thoughts on whether it’s because these specialties somehow have kind of low hanging fruit that’s still left to be picked? Or is it something more to do with the approach to research and maybe some of these approaches can be adopted by other stagnating areas as well?
A13: I’m not in a position to comment about the innovation in other areas as I’m very much a specialist hepatologist-gastroenterologist with more on the hepatology than the gastroenterology. There is plenty of innovation happening, but there are other areas where things haven’t changed and things are sort of just carried on in the same old fashion. I think it is about the mindset, and it’s about the people who appear and think about things differently. The grey-haired old bloke who walks in is not normally expected to be the guerrilla innovator. Certainly within hepatology circles, during the early days of Hepatitis C, there were the infectious disease doctors, there were addictions doctors, and then there were the hepatologists, and we were regarded as the old stick in the mud, because the rest would all have been terribly trendy with HIV and alcohol and all the rest of it. It’s about breaking down those moulds and thinking differently and being revolutionary about what we’re doing.
I remember I was accused of being a guerrilla hepatologist. At one meeting where I was asked to speak, a very old and eminent Austrian hepatologist went down the panel and commented rather than asking a question on the three presentations. The other two had been talking about the pharmacology of HCV, drugs, etc. and he got to me and said that we’ve been complimentary and he came to me and just said “You are mad”. I’ve taken that as a compliment ever since.
You do need to have the mindset that things can evolve, that things need revolution. Yes, innovation allows revolution to happen more quickly, but at the end of the day it’s having that data around which tells you what’s good and what’s bad. I think the first stage of those stagnating specialties is to think about what are you doing that you’re doing? Is it all perfectly done, or can you make it better? In Dundee, you’ll get the Dundee take on things. In other areas, other departments and different specialties will be more dynamic. It’s just the nature of the personalities of the people that you have, and you need all sorts of new people get things going.
You constantly need to ask these questions and a good example of this is how COVID has created a lot of opportunities for us to innovate. Things that were set in stone suddenly had to change overnight, because we either had to stop doing them or find a different way to do them as we couldn’t carry on the way we were because of COVID. People got inventive, very quickly, about how to solve things. I think there is going to be more flexibility in the system, about how we cope with the different things and clearly, as we enter the second wave, hopefully the rest of the services are going to be innovative about the way we cope with it. This illustrates, again, the point about that need for innovative thought and how to do things. Whether you call that research or innovation or the entrepreneurial approach to things, it’s that flexibility of thinking how do you change the system? Why shouldn’t you keep changing, keep improving things?
Long time horizons
Q14: You’re quite ambitious in your work, yet at the same time, it looks like a very long time before you finally got to see results. How do you cope with the pressure of trying to make big changes where you don’t get quick outcomes and you don’t want to seem like you’re just promising big things?
A14: Where I am now it’s easier because I’ve got some credibility about those things, so if I now promise something, people will listen. Back in the day, 10 or 11 years ago, when we first started to say we’re going to eliminate Hepatitis C, there were lots of people that were doubtful, and you had to persuade them. I had the data, I had the plans, I had the evidence, and I was evangelical about it. That helped. Communication skills become important in that realm, and you use your small victories as a way of showing each of the steps we spoke about before in terms of being able to reach and diagnose people earlier, present that at meetings and writnig that up as a paper etc. Having started the narrative and showing what the steps would be, you tick off step one and move on to step two. Because these small wins are happening in parallel, you then start to build a momentum around it, and start to attract a team of people who believe you and follow you, and that keeps you going as well. The leadership is often quite distributed and other people might start to take point and help along so that on the days when you’re feeling like crap, someone else will be having an ‘up-day’ and will carry you through it.
When I started, I didn’t think we’d be talking about making waves in 2025, 2030 roughly at the time I’d be retiring; I didn’t even think we’d make it in 2020. But then the field started to change, and other enthusiasts came on board and we started to make those steps forward. Each time we did get funding from a grant that allowed us to put the next plank down, it started to get easier because it was snowballing. That’s the thing with research: I mean, it’s unusual to have a big game like eliminating Hepatitis C and I’m sure people were laughing at me when I first started. Did I actually believe we were going to eliminate Hepatitis C? I thought it was possible, but I thought we should give it a go rather than not try. I’m a great believer that the only thing that ensures that you won’t make a change is if you don’t try. That was the ethos throughout it and we started to attract people around us who had the same thought – it helps being in a teaching hospital or in a team where you can get those people who have been with you for the whole ride, and quite a lot of enthusiastic medical students along the way as well!
Personal influence
Q15: Do you have any personal influences be it scientists or non-scientists from different fields?
A15: Widely! I like reading a fair amount of history, both medical and non-medical, and I always find inspiring people from there. My daughter is an English graduate and she has always criticized me for not having enough quotes in my writing, and there’s often inspiring stories behind the people who have uttered these quotes. There are phenomenal scientists, the three guys, for example, who have been attributed to discovering Hepatitis C won the Nobel Prize last week richly deserved it from the different aspects that they put together. The thing about all of them is that they were nice guys as well. The fact that you can see that movement all the way through, there’s lots of inspirations of people who have done amazing things. Sometime, less so, you suddenly see a research field where someone has made a fundamental breakthrough that has changed everything radically and they haven’t really got the credit for it that they deserve. It’s important that you are aware of what’s gone before and remembering that these people, they weren’t super men or women, they were the same as you and I. They’ve had the same education, and we can take our steps forward, standing on their shoulders. It should be all of our duties to think about doing that and to help move medicine forward. Be that through audits, laboratory bench research, or changing pathways of care, etc.
Management
Q16: Could you share with us some crucial lessons you’ve learned about managing teams, especially in these kinds of large-scale initiatives?
A16: Everyone has their strengths and weaknesses and it’s important to recognize both and support people through those jobs. With some roles, you get to pick the people that you work with, whereas in other roles you have to make do with what’s there already, because you inherit staff and things. It’s all about getting to know them, making yourself approachable, and using appropriate amounts of the jokey humor skillset as well as spending quite a bit of time explaining to people what it is you’re trying to do and what it is you need them to do to make it all happen. That’s the one thing that’s worth investing a lot of time in, getting that explanation pitch right in a manner that suits the people that you’re trying to explain it to. Particularly so if you’re doing pathways of care etc. and there are going to be people who are trying for PhDs as well as people who left school without standard education around and you need to explain to all of them what you’re trying to do and get them on board. I think that if you’ve got a story to tell as to the difference you’re going to make, what the end point is and how you’re going to get there, or what they need to do to make it happen, that story is going to make a big difference. That’s how you get really effective teams, both in research and in NHS clinical practice. I can get away right now by being the eccentric old professor, but when I was younger I had to sort of adopt different personas. It’s about adjusting the way you interact with the person to suit that person so that you get the most out of them. It’s also about knowing what’s going to keep them happy because, if they’re happier, motivated, and feel comfortable, then they can do what you’re expecting them to do. That’s the key to good successful management.
Wandering vs Aiming
Q17: Some more personal questions now: how important is it to ‘wander around’ as opposed to having a fixed focus goal in research?
A17: Within my work, I wander a fair bit as hepatology has the advantage of being a broad specialty with lots of different aspects. I am frequently accused of being unfocused and it is a criticism I have faced of my research by universities throughout my career. From a university perspective, my research would be criticized as being hopelessly unfocused and doomed to failure. They don’t say that quite so loudly now that I’ve got 250 publications and 30 million pounds in grant money! I’ve wandered through several areas and looked at different diseases although there have been some similarities across the board as to what I focus on. It’s unusual to find someone who’s quite as unfocused as me from having research with genetically modified cells, to animals, to big gene/informatics databases, to pathways of care in active drug users – a very broad spectrum of research. I wouldn’t recommend anyone following quite my pathway. But I’ve enjoyed that journey, although it hasn’t been the most direct route to getting a chair. Had I been more focused, I might have gotten to things earlier, but then I suppose I might not have had the inspirations to do the things I did.
The unfocused route is not encouraged in the current career structures – to do things the way I did would be very difficult now because there was much more flexibility in the old systems that I came up through and the flexibility was starting to disappear by the time I became more senior, which is why I was getting quite a lot of criticism. There were still a few mavericks left in the system that sort of gave me the odd helping hand when I really needed it. But my career is not a typical career path. I really admire those people who have focused on their idea and gotten to where they wanted to be in a very efficiently manner, but sometimes they then get a bit boxed in once they’ve gotten there.
I came into research later than others. I was first in the university of Edinburgh, then spent 10 years as an NHS consultant here before the university bought me and I sort of worked my way back up the tree again as a junior senior lecturer. I think nowadays, the pendulum is perhaps swinging back towards people being a bit more applied, and therefore a bit more unfocused. Whereas when I’d gone through my career development, we were swinging very firmly to the clinician scientist who was predominantly a single-method, methodology scientist with a disease who brought the two things together, as that was the model that people thought was going to solve everything. I’ve got a little aspect of that, and particularly my collaboration with Professor John Hayes here is to thank for that. A lot more of it has been applied in terms of the data science, things that I’ve done with folks, and I’ve been very collaborative, which has been one of the fun things about this.
I’m not an intrinsic expert in the techniques, but I’ve found people who are and so I’ve collaborated with lots and lots of different specialists across the whole spectrum from molecular biologists, biologists, data scientists of every shape and form, epidemiologists, mathematical modelers, to psychology and clinical psychologists as we tried to design pathways of care. I always take the approach of ‘I have a problem, let’s find someone who can solve it’. Knowing enough about the field and being honest enough about it, helps when you go to approach people that you want to collaborate with. It’s always been fun to find new people who have very different ways of thinking. Now, sometimes, you’ve formed a team with someone where it didn’t work out and we parted company, but on the whole, people are keen to collaborate and particularly if you’re coming with a health problem to a non-clinician, they’re keen to see their stuff being applied to healthcare, because it’s important to them as well. I think if you are going to follow the sort of career of research that I’m doing, then it’s important to learn to collaborate, knowing who it is you need to find and having the ability to go and buy people coffee and encouraging them about your idea while they’ve got to at least spend 10 minutes with you as they drink their coffee before they can go away. Sir James Black, who has sadly departed, did his training here in Dundee and probably should have gotten two Nobel prizes, for the work he did on beta blockers and h2 antagonists. He said the most important place is the coffee room where you meet people who you don’t normally talk to. Those points of collaboration are how you find your inspiration. It’s often through those random conversations whether it be in coffee rooms or big international meetings (which sadly, we can’t have at the moment) where you’re wandering down the corridor, you chat to someone, and then someone comes up and joins you, and as you’re chatting you discover that the person who just arrived is the same guy who wrote the paper that you read late yesterday that you think is the greatest thing since sliced bread. And now you’ve suddenly got 10 minutes with him or her to chat – that often leads to joyous collaborations.
Applied vs theoretical research
Q18: You mentioned the distinction between applied and theoretical science. What are your thoughts on whether research must have a utilitarian application in the end or whether research for the sake of it is good too?
A18: Within a medical school, research tends to be very applied because of the nature of what we do, whereas if you go to the School of Life Sciences, they’re working out how stuff works, that might become useful someday. Yes, there is an increasing focus that all research has to be applied, because the funding authorities are wanting the health of the nation improved, and to have clear feedback, rather than funding what would be called ‘Blue skies’ research. I’ve worked with some blue skies researchers, who were trying to become a little more applied by seeing whether the protein that they were working on had any relevance to disease, treatment etc. Once they discovered it, it became applied research so there’s that part where they meet I suppose
It’s not as focused as saying, for example, we think this molecule is vital in cancer, and therefore we’re going to look at all these cancers, or as applied as saying, we think this molecule will be a diagnostic for rhubarbs syndrome and we’ll measure it in 1000 patients with it. By applied I mean you’re intervening in patients rather than being one step away where you’re seeing if the mechanism works in a cell or an animal model, and moving on from there. Then there’s the bit behind that, i.e. fundamental blue skies research, where you’re looking to work out how things work, rather than working towards applying them to a disease as such. All of that feeds towards applied research though.
Advice
Q19: Do you have any advice for people starting off right now or within next few years who are keen to pursue an academic career?
A19: Because everything in medicine is becoming more pressurized in terms of the training period shortening, and the struggle you have to face of deciding what specialism you’re most interested in, I think you have to make use of the opportunities that are now available as medical students, dipping your toe in the water and deciding if you like research or not. The common misconception that research involves standing by a bench moving a pipette etc., for those of us that have actually done that, is very boring. What’s more accurate is the mindset if you want to get involved in research, about asking the why question: why does this happen this way? It’s that thinking of why, rather than the how and learning to ask the why question and not giving up until you have an answer is the first piece of advice.
As young clinicians, there are a whole series of huge hurdles in front of you. But they’re not hurdles, they’re steps up and, once you’re on the other side of them, you can’t see them, they’re actually very tiny and quite easy. Keep going and don’t give up. I’ve failed plenty of postgraduate exams, and stuff along the way. The journey’s not going to be perfect, but it’s about doing what’s going to make you happy. If you’re going to ask that why question wherever you end up doing your career, you’re going to spend 35 to 40 years doing it, so you need to find something that’s going to keep you stimulated for that whole time. If you’re going to end up in the same job for that whole period of your time, and it doesn’t change, and you just do the same old thing every day, that’s not going to help your work life balance. But if you’re asking the why question, and your job is changing, because you keep changing, and improving it and making it better, then that’s going to keep you interested.
That’s what I would recommend to anyone thinking about research. Don’t be put off, don’t think it’s too hard or too difficult. And remember you’ve got to keep asking that why question all the way through your career.
Automation
Q20: With the rise in automation, do you think doctors at some point will take on the supportive and emotional care of patients while machines do all the algorithmic work behind the scenes? Or do you think there’ll be edge cases where doctors still do some of the decision making?
A20: I think your for your career, you’re safe. I think there will still be that component of diagnostic logical reasoning, and the assessment of patients’ symptoms is still going to be very important. You can’t put that into a machine, because you can’t get the definition of that. A lot of the problems with machine learning arise when you put a garbage definition of a symptom in and then get garbage out. Assessing someone’s aches and pains requires a sophisticated conversation to form an opinion about what that is in the context of the patient and comparing that to the generality of everybody, in terms of where that level of pain fits in and where the diagnostic areas of it are, etc. so that that’s always going to be there. I think that the decision support tools and the like will ensure that all the right things get done all the time for the appropriate patients, once they’re in their pigeonholes. I think they will also be of some help in pushing people into the appropriate pigeonholes but I think it’ll still be a human interaction at the end that will be sorting out the final placement etc.
A lot of medicine is very simple and it becomes a bit binary. Decision support tools will help sort that out a little bit. But there are still going to be patients with the more complicated cases that need help and the worst possible place for a patient to be is in the wrong pigeonhole. I think if we have too much mechanization and we simplify too much, we can end up putting patients in the wrong pigeonhole. If you look at our semi-automated systems now, say on the various call help lines for patients, almost all the disasters and the problems that have happened from them are due to the way in which the symptomatology was interpreted, which led to them being in the wrong pigeonhole. That’s where the disasters happen and I think we’re a long way from having a perfect system yet. We’re gonna have to work with those systems, though, because we humans aren’t perfect either and there have been plenty of disasters because people were having an off day and didn’t think about things clearly, or they were lazy, or just weren’t very good at their job. Hopefully, some of these support systems take that variability out. You’re still going to need somebody to interpret symptoms in the context of the patient, though. Having said that, I think that the human aspect of our interaction with people in terms of how we interact with them, how we give them a diagnosis, how we advise them about their prognosis, and their management plan is very important. We need to recognize that as being more therapeutic than we give it credit for, and that’s probably an important part of the therapy that that patient needs which only we can provide.
Finally some quickfire yes or no questions.
Q: Paper or iPad?
A: If you’d asked me about two years ago, I’d have said paper but now I’m moving towards iPad.
Q: Look to the past or look to the future?
A: If we ignore the past, we will wreck the future, but we need to look at both because the future is where we’re going.
Q: Telescope or microscope?
A: Telescope
Q: Tea or coffee?
A: Tea, although I quite like a cup of coffee as well.