Transcription
I I think the main thing about the field of fetal medicine was it was an absolute black box at the time in the early 80s to mid 80s and just started ultrasound. So for the first time you could look at your patient, you could make a diagnosis, you know.
Don't think you can be prescriptive about it. Um, I think we need to teach PhD students, you know, they need to be AI literate to understand the risks but also the strengths of it. I don't think a national system doing all the primary investigation is going to work.
[music]
Welcome to the Denver Sations podcast. Today I am joined by a man who is exceptional in many respects. Professor Nicholas Fisk. I am Professor Nick Fisk. He is an emeritus professor at the University of New South Wales. He is a distinguished fellow of the Australian Academy of Health and Medical Sciences, having had a career as a clinician scientist that has spanned over 30 years. Over that time, he has managed to publish over 400 papers and has held some prestigious, uh, seats including being a professor of obstetrics and fetal medicine at Imperial College London. Professor Nick Fisk is also a highly respected higher education leader. He has held the position of executive dean of medicine at the University of Queensland. He has also more recently been the deputy vice chancellor of research and enterprise at the University of New South Wales. For his contributions to higher education, he was inducted into the Member of the Order of Australia in 2020. Professor Nick Fisk, thank you for joining me at the Denver Asian.
So, thanks Dan. It's, it's, um, real honor to be here and, um, I predict a very bright future for the Denver Sations.
Oh, thank you. Thank you. Thank you. I mean, if I can keep getting people of your caliber to this podcast, then I'm pretty sure we are on the right track. Thank you very much for being here. You have had a very successful career but a very long one as well. So maybe I thought, are you able to walk us through, um, your reflections of what you think have led you to this, um, long and successful career? What habits, what factors do you think have contributed to it?
Sure. So look, um, probably key is, is sort of a resilience for one and appetite for knowledge. Um, you know, these days they say everyone should have three careers and, and you can do that within medicine. Um, but you always got to be looking for new things and, and new opportunities. So, um, you know, there's always a bit of self-reflection about who you are in, in these sort of things. Um, you know, in terms of one's own skill set, um, I know the answer I'd give, but, you know, if you ask my family, they'd say the three drivers for me are insomnia, adult, adult attention deficit hyperactivity disorder, and competitiveness. Um, I had one of those 360 assessments of the people you work with once and reading down and one of the things that stood out is it said competitiveness 99%. [laughter] I asked who got a 100. So, you know, I guess at the more personal level, mental and physical fitness is an important driver. Um, it was the Roman 2nd century poet Juvenal who said, "Mens sana in corpore sano," famous phrase, but he was really guiding people away from the pursuit of power and wealth.
You don't want to take yourself too seriously. Um, it's always important to maintain a sense of humor and work in teams, etc. And, and then, you know, the all important work-life balance. You got to prioritize family, friends, holidays, and, and outside interests. But well, that's just coming back to your competitiveness. Um, what do you think drives that? Is it a will, an inherent will of wanting to do well for yourself? Is it family? What do you think drives your competitiveness?
Must be nature and nurture. It's a combination of both. There'll be some genes there somewhere, won't there? You know.
Yeah. Right. Okay. But one thing that is obvious about you is that you seem to have this adaptability.
Right? You, as you say, you need to have three careers. Could you walk us through?
So that's cuz you get bored after a while. And, and boredom is a, a real thing, you know. Um, if we're going to talk about clinical medicine and academia at research, the beauty is it changes all the time. You know, compared now to five years ago, five years time. Yeah.
It's very dynamic and, and that keeps you going. But,
you know, can you imagine?
I think that, you know, imagining walking into an office and doing the same thing for 35 years, just driving me berserk.
Yeah. It becomes very boring after some time. Yeah. So, for you, it's more so the boredom that gets you moving around and trying new things here and there.
Yeah. Restless. Or is this humility in some sense? Are you being humble?
It's difficult to answer that, but, um, no, the challenges change. You know, what you like doing. I mean, you know, you, you train as a doctor, but then you get interested in just the direction particular specialties going, and then how much that's supported by clinical research, but also by scientific research. And then, you know, naturally, you know, you face the thing as a, as a doctor. Do, do you want to cure the patient? Well, you can try that for a while, or you want to cure the disease, and, and that's where research comes in. And then after a while, do you sort of want to organize teams of people to try and cure the disease? And then, you know, the older you get, the, the importance of dealing with individual patients, although still really important, best left to younger people who are highly skilled at it, and you move on to sort of more senior things. Yeah.
Let's talk about, uh, life work balance, right? Could you walk us through some hobbies that you developed over your career that maybe, as you mentioned, keep you sane?
Um, well, probably two at the moment. One is, um, always been fascinated by, you know, international politics, culture, and travel. Now, I've guess I've been to over a 100 countries. Um, all started as a student. You know, when I trained, you could do a medical course at the age of 18 in five, in five years. And I managed to spend 12 months abroad during five years. Just every time term finished, I'd be off on a plane.
Um, you know, um, election terms, option terms, all that sort of stuff. Um, traveled through, you know, the Eastern block to Siberia, China, through Africa, this sort of stuff, and, and later on to, to other countries. Yeah. But, um, really fascinating to do that, and I used to impoverish myself as a student to be able to afford to do it, and then when I traveled, I used to impoverish myself because it's the only way to spend so much time abroad. I mean, you'd relate to this one. I was, um, I was deported from Ethiopia into Sudan. Um, traveled up through Sudan and was, had to catch an overnight ferry from Wadi Halfa. Wadi Halfa. Um, and, um, of course, you had to get the cheapest ticket possible, but I didn't realize that was third class, sleeping next to camels. Flatulent camels at that. That was quite an experience.
That must have smelled really nice.
Wonderful. [laughter]
So on in all of your travels, I wonder if you have, and this might be a mean question, if you have a favorite destination or a place that you thought you had the most experiences.
This is such a, such a choice. I mean, um, you know, I lived in London for a long time, which was an absolute buzz, um, center of the world, then [clears throat] maybe not so sure now, but, um, you know, that, that, that was a wonderful opportunity. Now, the travel is, um, you know, um, there's quite a bit, there's still work travel, but it's a lot more about cycling adventures. So,
you know, um,
and that is one thing you really enjoy cycling?
Yeah, I do a bit of it.
Big enthusiast of it?
Yes. So I do, um, I, I like longer distance events. You know, I'm probably doing 350k, 300k a week or something at the moment. But, um, you know, there's the brutal long ones like Paris-Brest-Paris.
1250k. I tried to do, well, I went to do the London-Edinburgh-London, that's 1500k this year, but it got called off almost halfway through because of a huge storm with 180 kilometer an hour winds. Um, but yes, you know, Sardinia, Corsica, Colorado, Alps, Dolomites, Norway, they're all great places to cycle.
Yeah. And have you had any race that was your favorite? I mean, you've raced in all of these places, but anyone that thought,
um,
this was an achievement.
Well, the only time, um, I've ever ended up on the podium was when, um, the podium has three people, third, second, first. Only four people turned up and one had to drop out. [laughter] So,
what, what would you say in cycling is your specialty? Is it climbing? Is it sprinting?
Yeah. No, it's definitely climbing. I, I can't sprint. My, um, FTP, functional threshold power is just under 220, which is middle of the range for recreational cyclists. So, I'm not that, that fast. But, um, in climbing, my V02 max is, uh, last weekend was 69 milliliters per kilogram per minute, which is just below professional cyclist level.
Yeah. So that is level.
How much training did you have to do to get to that such level?
I train every day.
Yeah. So are you cycling every day? Is that a?
Virtually every day. Yeah. You know, every now and then you have a day off and run and swim or something.
Wow. Wow. Impressive. And how did you get into cycling?
So, always just cycled a bloody lot in, in the UK. But, um, when we moved back to Australia in 2008, had to separate the family for a year while daughter finished high school and son came here. So I'm here living with a 15-year-old boy. Um, what the hell do you do? You go and buy two bicycles and sort of he now races in a grade and, and, you know, been doing it ever since.
Wow. Impressive. So it's a family thing now.
Yeah.
Okay. Let's talk about, and I don't know how much of this you'd be able to remember, but let's talk then about your medical career and I would like to start with your junior doctor years. Um, so maybe you could walk us through some reflections there.
Sure. So, um, I didn't do very well as a medical student.
Turned out I majored in other things, you know. [laughter] Um, so I didn't get honors or anything like that. Um, actually, I didn't enjoy being a medical student. It, it was, I hope it's changed now, but it was far more about, you know, being an, not even an apprentice, an observer, you know, you're always in the way, you know, you know, nobody wanted you there, you were a nuisance, etc., etc. But once you graduated, it was, it was fun, both intellectually and you could do things and practical stuff was terribly exciting. So then you go to the stage of what sort of specialty are you going to do? And, um, you know, two things really. I've always been fascinated by reproduction, you know, conception, the biology, pregnancy, fetal physiology, birth, etc. So that, that was one. And the other is, um, you got to make this choice. Do you do medicine? And in those days, initially, it was just about, do you give the pink tablets or the red tablets? But all the intellectuals did that. Or do you do surgery, which you cut things open and do heroics, but it's not very, it's not, it's not the most intellectual of sports. So all that's changed now, but obstetrics and gynecology was a mixture of both. You had the medical side and the surgical side, all in one.
Yeah. And this is something that kind of drew you towards it.
Yeah. Yeah.
Yeah. I mean, medicine is still pretty much the same, although people are nicer now, I guess, more so than during your day. But can you walk us through what do you, what was the motivating factor for you doing medicine? And how did this reflect in university, then as a junior doctor and as a clinician, and how did that change?
So, am I under oath? Um, [laughter] look, you know, the desire to help people when you're a kid growing up, you know, that, that's always there, and it, it's one of the healing professions, one of the caring professions. So, that's one. But the other is, um, maybe it's the competitive thingness thing. The most difficult thing to get into was medicine at the University of Sydney, and so that's what you did. [laughter]
And it, it hasn't, it still hasn't changed much. I wonder whether, I mean, this is a conversation we have a lot, the competitiveness actually puts off some really talented people who, um, may be good doctors in the sense that they will be able to follow good instructions, but they never just make it in because they never get the competitive scores that are required.
So there are two sides to that. There are all sorts of different flavors of medical courses now. One's based more rurally, one's, um, you know, with different emphasis on on impact and service and things like that, or one's trying to,
you know, train the next generation of cutting-edge docs. Um, so, yeah, that's changed a bit. I think there's still huge scope. I mean, you know, medicine's so attractive because of the income and status and all those sort of things for many people, but, you know, you don't need to have an IQ of 300 to, to do a lot of the jobs we need people to do, you know, particularly in rural communities where they're the center of the community, that sort of stuff. And, and [clears throat] um, in, in many of the less competitive specialties.
Yeah. Yeah. You ended up practicing as a fetal medicine, um, specialist. Could you walk us through any influences that you may have had here? Any mentors or any intellectual material that you may have consumed that would have driven you down that road?
So many components there. Well, you know, yes, there are a couple of mentors and things, um, you know, both here and in, in the UK. They're sort of important. But in the genre in which I grew up in in hospital medicine, the, the pivotal career was driving a Porsche and being a teaching hospital consultant, and, and that's what you were trained to do. So meeting a few people who were academic and asked questions and, you know, gave you ideas, that, that was just really stimulating. I, I think the main thing about the field of fetal medicine was it was an absolute black box at the time,
in the early 80s to mid 80s, and just started ultrasound. So for the first time, you could look at your patient, you could make a diagnosis, you know, and then you could began to do procedures from them, and, and that led to, you know, the whole of the characterization of their hematology, biochemistry, all sorts of things that you could think about their renal function. Um, a lot of invasive type, you know, procedures and therapies went on, but gradually they, they developed, um, you know, right through to things now like, um, you know, doing cystoscopies on fetuses in utero and try to relieve their posterior urethral valves, or lasering the placenta, that sort of thing. It was just fascinating, and the pace of change was huge.
And a lot of it was centered in London, you know, so more influential to me was my peer group. We had an office with about 10 of us, barely this size, um, with tiny little spaces around the edge, not, not full desks. Every one of those people is now a professor, you know, and has been a very successful professor. It's that sort of stimulating environment, and the world then came to London.
In particular, London, because a lot of fetal medicine is rare diseases, and it, you can, but it's trickier to acquire the numbers for the research cohorts you need, you know, in a lower, in a lesser population, certainly like Australia was then, was then.
Could you, could you walk us through that prospect of moving to London as a young person? I mean, I know that there was not a lot of the work being done in Australia, and maybe it was such an obvious thing, but could you walk us through, was this an easy decision to make? Was it the almost automatic next decision if you had committed to fetal medicine?
So, I really think a variety of career experience is really important, particularly at sort of formative stages, subspecialty training, PhD training, postdoc, that sort of thing. Just working in a completely different environment.
That's really important. It's getting trickier to do. Someone's got to pay for it, and, and there are all sorts of restrictions about licenses and stuff like that. So that's one of them. The, but, you know, at that age, for me, the first five-year trip overseas was when, um, you know, uh, they'd only barely discovered faxes.
You know, the literature came weeks later to Australia, that you just wasn't at the cutting edge. So it was pretty essential then to go abroad for that period of training, and I did. But then, um, I moved back to Australia for a year and a bit, and, um, you know,
I remember, so the first decision was easy, that was exciting. Yeah. Um, the second was tricky because, um, three days after buying the forever house in Sydney, um, I got this amazing offer at the age of 34 of a, a chair at what was then the leading obstetric hospital in the world. Really difficult to say no to. So, um, we went back. Yeah.
Yeah. And how was that moving back to London experiences? Any interesting stories to share? Tricky with two young kids, but it was just such a stimulating environment at the time.
Um, you know, you've got so many major hospitals in London, you've got the population, um, research groups that you establish, you can attract people from all over the place, often coming with their own salary, brightest people possible. Um, we used to almost joke at one stage that, um, the only rule in, we had in our research group, sometimes 10 and sometimes 30 people, is there were hardly any English there. You know, they were all foreigners.
And how did that shape you as a person? The, um, this environment where you're able to work with people from such wide backgrounds, how did that influence you? How did it affect you? Were you able to, did it teach you adaptation in some sense? Were you able to adapt? Was it harder, easier?
I'm always fascinated by different cultures and traveling, and it was so easy to travel, particularly in Europe, etc. Um, you do need to lose your Australian accent to keep going. But, you know, [laughter] these things just happen naturally, you know.
Yeah. I mean, I, I have traveled. I was, my family is originally Sudanese, then we moved to Kenya, then we moved to Australia, and if somebody asked me today, how has that been that journey? It's almost like you forget where you've stayed. But it's good that, I mean, you still have that Australian accent in you.
But, you know, you are working in, in a cosmopolitan London, and London's very different from the rest of the UK. The same working in Sydney, or even Brisbane or Melbourne. You know, you're working in a pretty cosmopolitan international city. Yeah. Um, that's very multicultural.
And it's become, and we're going to talk about academic leadership and academia in a bit, but hubs are like such a big part of research now, right? Precincts and all of these things that attract foreign people, and, and it's, it's almost like you're concentrating information in a hot spot.
Yeah. Yeah.
How does that influence the exchange of ideas? Well, research these days, you know, the old environment was you were a sole trader researcher, whether a clinician or a scientist, and you closed a door and you did your own work and you wrote it up, maybe a couple of PhD students or clinical attachments, etc. You know, research has changed hugely now. For impact, you need big teams, you need multi-institution collaboration, you need interdisciplinary cohesion. Um, so that is great, and that's how you get big things done. Um, and the sole trader type academic will will struggle in the current and, and future environment.
But it, it does come at a bit of a challenging cost for young people though, which is recognition. You know,
people look at CVs and say, well, you're not first author or last author, but hang on, there are 40 authors here, and we've just cracked the gene for this or something like that. It's really important work. Yeah.
The, the, the whole metric basis of research assessments changing.
Yeah. And can you walk us through your experience as a clinician scientist? Um, how did that feel? When was the time you decided I want to pursue this?
So I guess we have to distinguish clinical scientist from clinical academic. A little bit fluid.
Um, you know, that we think of it as someone employed by a university and someone by a hospital. Um, but, um, it's probably the same that, you know, you are bridging the gap between research and development and clinical practice. You know, the clinical scientist is more wet lab, um, more dry lab. For now, you can be a full-time clinician,
and be a clinical academic, but it's tricky. Depends how much pound of flesh your hospital service wants,
um, versus in the UK, the model is more 50/50. Clinical academics, clinical scientists work 50% for the university and 50% for the hospital, standard model.
In Australia, it's trickier. Um, often some of the clinical academics only work two sessions, um, in the hospital, and that makes it trickier to maintain your skills and maintain street cred. Um, you know, there's a downside and an upside to being a clinical academic, clinical scientist. The, the best image I remember of this is the kids' book by Dr. Doolittle had a picture of the push-me-pull-you, this sort of mythical animal that's got a head and four legs facing this way and a head and four legs facing this way, and you're always pushed, pushed, and pulled from one way to the other.
On the other hand, you know, so that, that could be a bit tense, and, and your clinical colleagues all think that you're always in the lab and not doing any work, and the lab people think you're always in the hospital not doing any work for them. But, um, you know,
it's stimulating.
Yeah. If you have a bad day in the hospital, a patient dies, or something's stuffed up, you know, you really feel bad. You can retreat to your research or write something, lab, etc., and that's a, a respite. Conversely, whole series of experiments, samples go off, or don't work. You can go and deal with patients who love you, and you know, it, it's a balance of rewards.
It's a conversation that actually is very prevalent nowadays with most of my peers. I did some wet lab research this year, and I was talking to some of my, um, peers who are actually just scientists and they're PhD students, and they seem to have this perception of a clinician as somebody who commits to something but never completes their side of the, um, bargain. So, is there truly any way to balance the two, or is it just, are you never going to get a good balance, and do you just have to settle for a middle ground, or is there a way where you can achieve the best of both worlds?
So, it's tricky. Time pressures are huge. There is, um, evidence really across the, um, high-income country world that clinical academics are a bit in decline.
But, you know, if you got a tricky problem, you want the best opinion, you go to a clinical academic who's right up to date with the literature at the cutting edge.
It used to be the model that they ran most of the hospital services. That, that's changed now. There's just no time to do that.
But, um, it is tricky. What, what's the alternative? Completely separate them, but then you got to get them to talk together.
Yeah.
Now, there are ways of doing that. There are buildings like the, um, translational research institute in Queensland that really tries to put them together, same coffee shops, you know.
But you also need someone who's bilingual, you know, can understand clinical, clinical stuff and science.
Yeah. And one of the things is, I mean, one of the motivations for people becoming clinician scientists is that you can actually have an impact on patients' lives and on the science of the day. Is that ever something that people achieve? [clears throat]
Oh, I mean, just think how much medical practice has changed, you know. So, um, if, if you want cutting-edge practice,
um, you really should be looked after by, you know, it's just like if you want your operation to work well, go to someone who does a lot of them. You know, they're slick at it. A clinical academic is used to synthesizing literature, to presenting it, how it influences care. Remember that it's not just about new drugs and new treatments and things that are trying to be introduced into practice. They can often be slow because there aren't enough clinical academics. And, and that's this whole thing now called implementation science. But there's a whole area of stuff that either shown to work and just not being implemented. That comes to the previous point.
Mh.
Or things that don't work and people still do. And, and that's this whole area of low-value care.
Yeah.
You know, people who aren't up to the literature, who did this 10 years ago, and no one's questioned them about it. I mean, when, when I first started, um, training and moving up, you know, senior registrar type levels, you'd ring the consultant in the middle of the night, and the treatment would depend hugely on who they were. Consultant A would do this on the left side, and consultant B would completely do the opposite on the right side, and consultant C hated both of them and would do something [laughter] completely different. You know, it's absurd, and we've moved heavily towards meta-analyses for the truth, guidelines, this is what you do, endorsed by national international bodies, etc. That's where clinical science and implementation comes in.
I wonder if I can ask your opinion on this, um, and this is just a rhetorical question, but by pursuing the meta, the meta-analysis pathway, have we almost lost the personalities and the characters in medicine? So the idea that somebody can almost have a twist to the way they practice medicine as an art.
It's a bit like saying, well, let's go back to before seat belts because most of us were okay. Um, well, there is an art to medicine. You, a lot of it's about your decision-making, what's in the best interest of the patient. There is no evidence, particularly in rare event diseases like, you know, the, the one I was in. But you have to make a rational-based decision. Ask yourself, you know, what's the, the right way to do things? What, what could go wrong? Um, you know, this is where, um, multidisciplinary teams, or, or even just team meetings are really important. You know, I was always very keen, um, you know, as professorial team, etc., you gather,
all your junior and senior doctors around once a week, and you debate what went wrong, what went right. And, you know, the approach I've always used is more sort of devil's advocate, Socratic questioning, you know, um, when something goes wrong, it's a team decision. Yeah. You know, it's not, "Why did you do this?" That sort of stuff.
So, you know, a lot of decisions still cannot be based on randomized control trials because there just aren't enough.
Yeah.
And you have to sort of go back to Bayesian type approaches.
Yeah.
I wonder, have you had any clinical trials that you've participated or led, and any stories from them?
I learned very early on that you try and avoid clinical trials because they can take years and years and years, and they can fail. Yeah.
There are people who are really professional at doing it, you know. Um, for instance, people at the George Institute here, you know, absolute, um, formulaic approach to getting the question, getting the community on board, working out the power, getting the funding, and once it starts, they've got all the staff to implement it. And even before it starts, you can almost guarantee it's going to end up in the New England Journal of Medicine.
But you go back 10, 20, or 20, 30 years or so, a lot of trials were started half-baked. Power analyses were overoptimistic, and they fail. Yeah. You know, and that's a waste of money. You know, they're funded for three years, and it's going to take four and a half. No one's going to pay for the extra year and a half. That sort of thing. So, it's become a very professional activity. Um, I've only been involved in a couple of, of randomized control trials. One, two, three, I think. Um, the, um, they're tough, you know. Um, and the ones that really make a difference, um, involve, you know, huge multi-institutional collaborations.
Yeah. Professor, let's move to work that is more relevant to your recent, uh, jobs. First thing, graduate training and AI. So, AI is changing a lot of things, and one of the things that unfortunately it's going to change, or fortunately, is graduate training, and it's going to change it in many ways. One of the ways in which it's going to change it is that it's going to provide a tool that can be used to obviously speed up research, but it's also going to change it in the more archaic, simpler sort of tasks that, uh, graduate students are usually needed to complete. So this includes writing, and this includes analysis. Can I please start with your opinion on that? And maybe could you reflect on your own opinions on what a graduate, um, graduate training program should entail and what the philosophy of graduate training should actually look like?
So AI is everywhere at the moment. It's got scale. It's got momentum. There are some risks, but there's there's huge promise. I mean, even economically, are we in a, a post-winter summer? Are we in a bubble? 43% of the S&P is based on AI companies at the moment.
Very scary, by the way.
Yeah. Very, very scary. So look, PhD training's changing obviously. Um, you know, what's the definition of a PhD? Well, the boring one is you got to produce a significant output that makes an original contribution to knowledge over a period of quite a few years. But, you know, it's really about teaching people about original thinking, critical analysis, independent, independent work. Um, it's all very well to say those things, but AI is pretty disruptive now. Um, sort of PhD training was always academic training. It's like an apprenticeship for academics, but it's really exploded now in that we provide people to industry. We provide, um, people, you know, clinically. Nothing wrong with having a PhD and then a career in both of those arms. Um, 50% of CEOs in Germany have a PhD, you know, very different [laughter] from this country. Um, we have three to two to 3% of our politicians in Australia have PhDs. It's very different in Europe, you know, where it's considered just a baseline intellectual training to go off in many different directions. But, um, yeah, AI is a potential huge disruptor for PhD training, both in terms of the risks and originality,
but particularly in what's the PhD going to look like. I mean, everyone needs to understand, you know, be AI literate and understand its capability, its limitations, and, and its role. Um, look, it's facilitative, but, um, science is a human endeavor. AI is a facilitative tool. But this was most poignantly, um, flagged last week in an article in Nature by Alex Sen Gupta from UNSW here. Really scary. What he said was, um, he asked AI, um, like he would a new PhD student. He, um, gave him a couple of ideas, um, gave AI access to a few databases where he could go and interrogate data, um, suggested just broadly what AI, don't know why we call it, he might want to look at, um, and in one session, answering a few questions, AI went away and, um, religiously analyzed the literature in terms of literature gaps,
came up with hypotheses to test, went and accessed the literature, critically analyzed the results, wrote a paper, suggested referees, and uploaded the data to some open access website, and he was blown away by this. And now it is a database PhD. It's not, you know, going into the laboratory and doing experiments. [laughter] But even still, that's quite amazing.
Yeah.
So two questions arose out of that.
What about assessment? You know, how are we going to assess PhDs? And, and we can come to that, but it's quite a challenge if you only spend half an hour or half a day doing it, you know, even a week. And the other, he asked was, should we shorten the length, um, of a PhD now? You know, it's three and a half years standard at the moment. In the US, it's more like five with coursework. Um, but where do you take it? You know,
two years?
One year? Six months?
And what change or what effect does that have?
Well, I don't think it's a good idea. I, I think, you know, serious training, three and a half years out of a career now of, you know, people going to work till they're 70, is, is not a huge impost for the top-level people who are going to be driving all our industries and hospitals and academia, etc. Yeah.
Um, look, the argument I'd use is, you know, um, we had an article in the, in the early 90s in the New England Journal of Medicine where we did PCR gels, and, and that was the output, you know, and we perpetuated these things and altered the temperatures in the PCR and went and, you know, analyzed all the sort of electrophoresis things, all that sort of stuff we used to do. Now, hugely robotically processed. Even PCR, free genomics, computers run it all, analyze the data. The human bit is interpreting it.
So you wouldn't say that, um, you know, there are some people that say, oh, well, you really need to do it yourself. AI is really just a tool that's facilitative that will help us. And in many ways, I think the same level PhD training is appropriate with with some guardrails around how you assess it.
And, you know, you're assessing three to three and a half years of knowledge progress. And if you just think on the exponential curve, how fast knowledge is going to progress, it's going to take off, and I think the PhD needs to be part of that.
Yeah. And I mean, I wonder, because we are talking about three years of assessments here, I wonder whether AI actually then makes a big difference in the assessment realm with that regard, because it almost becomes then that you can't fake it, right?
So it's interesting. We can come to fakery a bit later, but, you know, um, there is a risk of fakery for a lot of things, image manipulation, plagiarism, but even more nefarious stuff, um, AI produces those.
Yeah. But AI is also a detection tool, you know, and, and there's some pretty sophisticated detection tools now. And you look at the rate of retractions, it's going up and up and up, and, um, people reporting doctored images, etc., you know. So there are two sides to that coin. In Australia, typically, we've assessed the thesis just on the book, the output, um, and that's assessed, and it's assessed without a viva, typically in Australia, and that's very different in the rest of the world. Why did that happen in Australia? The tyranny of distance, the difficulty getting the right number of examiners who know enough about that field, um, even if they travel interstate. Now, that all changed a bit with COVID and Zoom, but there is a recognition that you don't just want regurgitated stuff that someone else has written. You want to really probe the intellectual contribution and the understanding of the field of the candidate being assessed for the PhD.
Now, by the time you finish your PhD and hand it in, you should know more about your subject than anyone else in the world.
Yeah.
Um, you know, I, I know this having examined PhDs. I've got to read 500 pages, and the candidate knows everything, and I'm going to look stupid if I ask the wrong question about what's on page 370 when it's also on on page 210.
So look, I did, um, but oral vivas are now have now come, or are coming to Australia. It started in Adelaide. Um, the major universities in Sydney do it, um, and, you know, it's routine now for all new PhDs in the leading universities, but it's still got quite a role to filter down. But, you know, the, the European and US model of of the viva, um, I was asked, um, to examine a PhD in Leiden. It's one of the oldest universities in the world.
Yeah.
It's in Germany, I think, is it? It's,
in Holland. Yeah. It's in Holland. Yeah. Um, and so I went there, suit, tie, all, and they said to me, you can't examine a PhD like that. And I [laughter] said, what? You got to wear a white shirt, not a blue shirt. So they went and bought me a white shirt. I had to get dressed in Calvinist robes to be one of 15 examiners before an audience of 400 people, with the candidate in white tie and tails, and also the, um, he had a philosophy and a scientific assistant on either side. Now, that was a, a show trial almost, you know, they would never get to that stage, but you really couldn't probe them. It began in Latin, ended in Latin, in that sort of context. So, you know, I think the, the British style, and now the Australian style viva, of two, three hours of rigorous interrogation of the student's own contribution to the thesis is really important.
That's actually quite impressive. I mean, I, I have always had the impression that the PhD training system has become very mercantile in its training today, in the sense that we're almost producing people to contribute to industry more than to be academics. Um, and Gandela Bosch at John Hopkins has an interesting term for this. She says, "Putting the philosophy back into the doctorate of philosophy." So she is studying this movement. She believes, and she thinks this is because we've almost kind of stopped, uh, trying to produce intellectuals and we're too focused on specialization.
I wonder what your thoughts may be on that.
Interesting. So, you know, PhD training's one of the tensions is how much coursework you throw in. Now, the US throws a lot of coursework in, almost two years full-time.
The Australian model, very little. Um, UK model moved towards, you know, several units of of coursework within it. But it, it depends how streamed your PhD can be, because people going into industry don't really need to know about the finer points of academia, and,
um, the same academics maybe don't need to know so much about entrepreneurship or commercial stuff.
Yeah.
Um,
but, you know, I think streaming a PhD and a limited amount of coursework is great. The moment you add a lot of coursework, that's pretty tricky in three and a half years if you're supposed to come up with significant original contributions to knowledge. Yeah. And I don't think I favor, um, five or six-year PhDs. It just gets ridiculous.
Quite long.
Particularly as for academics, you then need to go and do a postdoc of two to three years, and for some people, two or three of those before you're really grounded at the cutting edge of a particular discipline.
Yeah. So it's almost pointless because they're going to always pick up these skills elsewhere if they do take the route of academia.
Horses for courses is the way. Yeah.
Yeah. And we were talking about AI just then, and it being also on the other end of, uh, checks, as a tool that we can use to obviously check for plagiarism, for example, check for, uh, the originality of the sources and, um, the product that has been produced by students. I wonder, could you, and this may be a bit mean, could you come up with a system where you have integrated AI into the PhD training system? How does that look like for you?
Well, so PhDs can be so hugely different, can't they? You know, to someone who's primarily doing, you know, observational laboratory work, to someone who's doing some very complex process data or throughput work in, in, in the laboratory, to arts and social sciences, you know, which could involve,
you know, um, historical things for which there's a big role for, um, you know, digital humanities, etc. So, I, you know, I don't think you can be prescriptive about it. Um, I think we need to teach PhD students, you know, they need to be AI literate to understand the, the risks, but also the, the strengths of it. Um, clearly, it's got huge roles in searching the literature, and there are now tools now that, you know, forget about PubMed and Scopus and things like that, tools that put it all together from the peer-reviewed literature, and,
um, you know, when you published a lot of articles previously, you had to begin with saying you've done a, a, a PubMed search, and you had to put in the search terms and the date so that someone could repeat it if ever that was controversial. AI will sort of take over from that. Um, so that's one example of, of where there are tools we need to teach them about. Um, the, you know, one of the controversies is, um, you know, like at an undergraduate level, um, you know, no, sorry, postgraduate coursework level, you know, um, my daughter tutors, um, half the students use AI, and you can tell a mile away. Um, and she says, "Well, I'm, I'm not allowed to use AI to assess them." [laughter] So,
that's one of the big tensions right now.
Is one of the big tensions, you know, should you be allowed to have AI write? I use AI many times a day, but we'll always primarily write my own stuff.
Yeah.
Um, but, you know, particularly say, people who don't speak English as the first language, why shouldn't AI write their first draft if they sort of tell them bullet form how to do it? I guess the argument against that would be that they should do their due diligence to try to learn these things because they, they are important skills to have at the end of the day.
I think it's the end product that matters, you know, and what will be important, you, you know, and when you submitted your PhD, you had to put down, well, so and so was done at this other laboratory by such and such, and I certify that this is what I've done. You're going to have to do that with AI a lot.
Um,
and also in terms of publications. But, you know, there are some no-go areas. Um, patents shouldn't involve AI. Patents have to have human endeavor. You can use AI to get there.
Yeah.
Um, you know, the Chinese at the moment are churning out patents in AI drug design, hugely so, but with individuals listed.
I don't think you should give prizes to AI. You can give a prize to a, a device or a technique, but it's somebody's developed it. That's the key.
Sorry, could I just cut you off there? You mentioned that, um, we shouldn't give prizes to AI. So, I wonder what you may think of the Nobel Prize in chemistry last year that was given to AlphaFold.
So, it wasn't just chemistry last year. It was chemistry and physics.
Physics were for the neural networks and, and how it all worked. Chemistry was for protein design and, and how you, you know, understand the structure of proteins. Both hugely important. And interestingly, um, more on the chemistry side, you know, bells are often awarded for something that happened in the '90s, you know, so the, the particularly the chemistry one was, was very interesting. But it wasn't awarded to AI, it was awarded for people using the technique of AI
to, you know, with AlphaFold to work out protein structure. And,
protein structure is fascinating. We tended to think of it years ago as a bit old-fashioned.
Um,
old-fashioned.
Well, so there are 20 amino acids in your body, proteinogenic amino acids, but there are about 500 in nature, largely non-proteinogenic ones.
So if we can harness those to make new drugs, just think of the possibilities.
It's, it's quite mind-boggling. Um, I, I really enjoy these things. This and developmental biology gets me excited. But let's
Talk about okay AI and integrity, right? And maybe the role of institutions here. What sort of system is required to be able to regulate and to be able to oversee these particular worries that we have right now?
Well, so a good system, but what that looks like is really tricky. We suffer from imperfect information, and the field is changing so much. [clears throat] I think prevention is a, you know, prevention and and useful tools. Um, that PhD students, for instance, understand the limitations of AI, but the exciting possibilities that they don't share data too widely in terms of IP and ownership, etc. They use, they use, you know, institutional proprietary systems, etc. So, I mean, this is a challenge for universities, how to how to do it. It's a challenge for governments. You know, they want to set guardrails on AI, but there's no national policy really on on AI. There will be one. Um, you know, government puts its toe in the water internationally on the social media thing for kids under 16. That that's absolutely unique, but yet, you know, we have advertising, betting all over the place. Um.
It's it's interesting. There's there's been a lot of push recently to have a national board that oversees the regulation of and as like a watchdog of um integrity within universities. And I think this almost is a natural outcome of the fact that anytime something comes out of universities, there's always going to be an effect that people do do feel like this was mishandled.
Um, I wonder what your thoughts on that maybe.
Okay, so good point. You know, um, integrity regulation, I mean, a bit like AI, key component here is prevention, education, etc., but you always get get some bad eggs. So, um, what's Australia's integrity regulation system like in in universities? Well, it's pretty good compared to internationally in many places. We have pretty robust national guidelines, um, national code of responsible conduct of of research. We have an oversight body that has its issues, but is um, run by the ARC and the NH and MRC. Um, so, and it's largely institutionally based. However, um, whenever there's anything of substance, it usually goes to external inquiries, etc.
So that that that's the the the glass half full side of integrity. Um, on the glass half empty side, um, there are occasional scandals, you know, and and you get rallies of cries of people wanting to put miscreants in the stocks and behead them. Um, so that that that's one push. Um, it is an area of sort of self-regulation, and and that's going out the window. You know, the royal colleges for specialties can do the training, but they can't regulate, um, you know, professional malpractice, that sort of stuff, which is all done state and now national level.
Um, so, and and there's no national data. So we know from the top universities that there can be lots of complaints, um, about minor misdemeanors, as you know, bit of jealousy and interpersonal things that go on, um, that really dominate the numbers. We know from our top universities is that there's one or fewer cases per leading university per year of what we call serious research misconduct. So FFP, falsification, fabrication, or plagiarism. Yeah, you know, that's that's significant. I'm not talking about two ands in a sentence instead of one.
Um, so there's a bit of a hawks and and doves things going on there at the moment. Um, the one of the tensions is around smaller universities and medical research institutes that just don't have the capacity to deal with these things. And they're largely the people who've been sort of clamoring for some form of national body.
There is no national body abroad which runs the whole show, and and so a lot of the smaller institutions and MRIs just want to wash their hands of things and and someone else will deal with it and [snorts] and pay for it and please. But it's not really going to work. And if you look at the figures out of the US where they have an office for research integrity, the number of cases that they eventually deal with would equate to about one case of research misconduct over the whole country per year in Australia, and it's clearly far more prevalent than that. Um, all sorts of fusses when they tried to introduce it in Sweden. So, um, I don't think a national system doing all the primary investigation is going to work. You need to know far more about local knowledge, all sorts of people, um, you know, procedural fairness, confidentiality, enterprise bargaining agreements. It it's really quite a complex thing, particularly cases that involve multiple institutions and lots of people together. Um, the oversight body at the moment is is limited in their remit in that they can only review procedural cases afterwards and and just in terms of the compliance with the guidelines. So, you know, you've got a hawks versus doves issue, big calls for a national body from some people, and um, it's all okay from others. I think the answer is um lies somewhere in between in that we really could do with a national body that would give us national data.
Yeah.
Um, that would reassure the public, um, in particular, could oversee university and medical research institute investigations, and that means approving say terms of reference of any external inquiries. They don't need to get into the weeds of, you know, um, arguments, you know, I should have been author on this paper, not this one, that sort of stuff, but um, they should approve terms of reference for external inquiries, approve the result. And I think from institutions' point of view, what would be crucially valuable is maintaining a database of suitable experts. So people have to be almost an expert in research integrity, cognizant of the code, but an expert in that in that discipline, and they're really difficult to find sometimes and can lead to delayed or or even sub-optimal processes. So, I sort of sit in between, you know, I think I wrote once that I I see a role for an, you know, a national integrity body, um, that's stronger than one we currently have, you know, that that has teeth rather than dentures.
Yeah. And how do we apply this whole thing to the AI then?
Um, and how does that affect your position, for example, or do you think it's still the same condition that should be applied?
Um, it's obviously going to need to adapt. It'll it'll probably be based. So, you know, there are criteria for authorship and contribution and stuff in the code. They're probably reasonable. They will need to be modified in time to cope with AI, but.
Right at the moment, I don't think you can alter that. We don't quite understand enough about what AI is going to do.
Yeah, true, true. Well, professor, still on the topic of graduate traineeship, um, you've been a big advocate for PhD living wages.
Could you walk us through how that came about?
And where you think we are with that at the moment?
So, PhD students, higher degree research students, um, really important, you know, they produce maybe 50% of a top university's research output. Um, there has been a bit of a um history of regarding them as slave labor, you know, which is is not appropriate. And I guess really at the time of COVID, a lot of them got into trouble, you know, financially, international students, all that sort of stuff. PhDs disrupted. And around that time, it just became obvious that um cost of living, the wage they were being paid was was actually below poverty level, let alone the minimum wage level. So, um, you know.
One of the the drivers was it was sort of felt they was just part of academic training and almost like teenage apprentices. In Australia now, the average age of a student starting a PhD is over 30, often with families, and and that's just not a sort of slightly longer degree that that is, you know, almost midlife. So, um, you know, I think the arguments arose that morally this is wrong, ethically this is wrong, um, to have PhD students running off and working one or two jobs, you know, for delivery or whatever, um, is distracting them from the thesis and their own mental health, you know, so on productivity grants, that's another thing that, you know, we needed to do something about that. Now, the scholarship levels had been set at, you know, what was then $28,000 a year or something, and it needed to go up by at least 10, and arguably more, to equate to the minimum wage for 35 hours a week, adjusted for the fact that PhD stipends are tax-free. The problem was the RTP, the research training program allowance to universities, set the PhD rate as between 28 and 40, and so all universities went for the lowest rate possible as a price volume thing. We want as many PhD students as possible, and they were allowed to do that, and the RTP rate set the standard for all the other local university scholarships and and subsidized ones. So we really thought this is not appropriate anymore on all those grounds, and we need to up it to equivalent of the minimum wage for a 35-hour week, adjusting for tax. And so there was a lot of resistance to do so. It cost UNSW tens and tens of millions of dollars to do so. And there was a lot of argument about do you do it for all scholarships, and and the answer was yes, you have to do it for all scholarships. So it's been very expensive. Um, it's been very popular, and it it was written up in I think it was the Times Higher as saying, uh, what did they say? It sort of started a chain reaction amongst other universities to to roll this out. Did we assess whether PhD students would approve or not? We didn't think that was necessary.
Yeah, I think a lot of them got it with quite high experience. Um, let's move to medical research then. This is probably the most exciting part. So I thought I'll just get your views then on the direction of medical research. What do you think is going to be the new frontiers in the next 5 to 10 years with regards to medical research?
So look, practically at a clinical level, it's going to be prevention. It's going to be hospital in the home. It's going to be community-based care. It's going to be high-value care, all driven by big data. They're likely to make a big difference at the community level. At cutting-edge research level, I mean, there'd be so much going on in drug discovery at the moment, you know, from I mean all the antibody-based stuff, CAR-T therapy, particularly for solid tumors, um, you know, genomics is making a huge difference in cancer already. Almost 40% of advanced cancers have a mutation that dictates which treatment, and that's just so different from three years ago.
Um, we've got the rise of medtech wearables, um, AI is going to have a huge influence. So, um, you know, how would I advise a young student now or, you know, a young doctor, what area to get into? Well, I think I'd say, um, pick something that's important, you know, that that, um, it's not a trivial side effect or a rare disease.
And think about your own skill set, your own toolbox. Um, so acquire skills, probably in a new area that that's got scope to develop, but, um, you know, they're in an area, sorry, they're in an era of lifelong learning. So you got to be able to adapt. And, you know, our current research system and funding system is such that it's quintessentially, and perhaps deservedly, Darwinian. You know, um, things die natural attrition, and and you go for other areas. So you also want to be aware of opportunity. You know, whatever areas you set out your career on will change and adapt by what's out there.
Personally for you, what excites you the most about molecular biology or about medical research at the moment? What are some fields that you feel like these have some great promise?
Everything. [laughter]
So look, I think precision medicine is going to be the big deal. You know, we'll all have our genomes sequenced, um, at if not well before birth. Um, you know, that that will be a huge driver. And I think, um, protocol-based care, um, sounds a boring term, is going to make such a difference. So, you know, at the moment, you have a heart attack or a stroke.
Yeah.
You want to have a catheter, femoral artery, maybe a radial artery, um, as quick as possible, but certainly within four hours. Um, you know, we're almost at the stage that if if you have a a stroke, and a quarter of the Australian population will, um, you'll get a CT scan in the ambulance before you get to hospital. I think those sort of streamlined things, yeah, joined-up care is going to make a big difference.
Yeah, there's quite a lot. Talking about precision medicine, there's been a lot of investment recently into it. Um, what about precision medicine do you think excites you the most?
So I, you know, it precision medicine can mean stratified medicine, just sorting people into groups, but it, it's largely genomically driven by, you know, um, rarer components.
And, um, you know, I see that we've mentioned how important it's going to be in cancer, and, you know, um, a lot of us are going to get cancer, and that's going to make a huge difference to treatment. But at at more at my end of the spectrum, um, you know, rare developmental conditions. Um, it's the thing that's leading to diagnosis, and in some cases already, you know, um, pediatric treatments tailored, um, developed after birth, um, some quite exciting things there from, you know, the the the earlier ones like, you know, spinal muscular atrophy, but to lots of different rare diseases of which there are, you know, hundreds and thousands, not hundreds and thousands, but hundreds, almost towards the thousands. Yeah.
Um, so I think, yeah, that's going to that's the areas that I think will make a huge difference.
All right, Professor Fisk, thank you very much for joining me. Please be back. Thank you.
Great. Thank you.