April 15, 2025 • 44 mins
Dr Katharine Halliday, President of the Royal College of Radiologists and a leading voice in UK radiology, joins host Sam Menter to discuss how collaboration, authentic leadership, and ground-up innovation drive meaningful change in complex healthcare systems.
In this episode, they explore:
- Balancing AI with human expertise: Exploring how artificial intelligence can complement rather than replace clinical judgment.
- Collaboration as a catalyst for innovation: Breaking down silos and integrating diverse perspectives to solve pressing healthcare challenges.
- Harnessing hidden expertise through co-design: Revealing and leveraging frontline knowledge to spark effective change.
- Redefining healthcare leadership: Rethinking how clinical leaders are supported, enabling them to lead effectively.
- Improving patient communication through user-centred design: Closing the gap between clinical language and patient understanding.
Filled with practical insights and compelling examples, this conversation provides inspiration for anyone aiming to deliver impactful change by combining human insight with innovation in healthcare.
Problems Worth Solving is brought to you by Healthia, the collaborative service design consultancy for health, care and public services.
Find out more about our work at healthia.services.
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
SPEAKER_01 (00:00):
Hello, this is Problems Worth Solving, the
podcast where we meet peopletransforming health and care
through human-centered designand digital innovation.
I'm Sam Mentor, ManagingDirector at Healthier, the
(00:21):
collaborative service designconsultancy.
If you enjoy listening, you cansubscribe to this podcast and
the accompanying newsletter athealthier.services.
Today's guest is Dr.
Catherine Halliday, a leadingfigure in UK radiology and
president of the Royal Collegeof Radiologists.
She's led national work toimprove diagnostic services,
(00:44):
shape policy and leadresponsible innovation.
Today, we're going to talk aboutleadership, AI, service
redesign, and how ahuman-centred approach can
unlock change in one ofhealthcare's most pressured
areas.
Thank you so much for joining ustoday, Catherine.
It's great to have you here withus.
SPEAKER_00 (01:01):
It's really nice to be here.
Thanks, Sam.
Thanks for inviting me.
SPEAKER_01 (01:05):
So I'd like to start with a little bit about you as a
person so that we can get abackground and understand what
drew you into this space.
So I'm curious about what youwere interested in as a young
person and how that led you intoradiology.
SPEAKER_00 (01:20):
Well, I always wanted to be a doctor.
I can't remember a time when Ididn't want to be a doctor.
Initially, actually, I wanted tobe a psychiatrist.
I've always been very interestedin what makes people behave the
way they do.
It took me a while to get intomedical school.
I didn't get in straight awayand I kept trying.
And then I met lots ofinteresting people when I
finally did make it.
(01:40):
So my career's had lots of upsand downs, but I've never
regretted being a doctor for asecond.
SPEAKER_01 (01:46):
And what type of person tends to go into
radiology?
UNKNOWN (01:50):
Thank you.
SPEAKER_00 (01:50):
Well, that is a very interesting question, actually,
because the role of theradiologist is really changing.
And traditionally, I mean, thereare a number of role models.
I don't know, maybe many peoplelistening to this will be too
young to remember Green Wing onthe television, where the
radiologist is a person wholives in a sort of dark dungeon
(02:10):
down below and doesn't reallyspeak to anyone and actually
gave himself coffee enemas, Ithink.
It was a very, very odd person.
And that is are traditional typethough I've never met anybody
like that I must say buttraditionally a radiologist is
somebody I think The stereotypeis that it's somebody who
(02:31):
doesn't really like to interactwith patients or people, likes
to live in the dark and just getthrough a load of tasks.
But actually, in reality, ourjob is very much about people.
It's very much about dealingwith our colleagues, about
understanding patients and thecontext in which their imaging
(02:52):
has taken place.
And of course, I'm a paediatricradiologist, so even more, I'm
very sort of hands-on in termsof dealing with children and
their parents.
So it's not quite as withdrawn,I think, as you might think.
And of course, we get all sortsof different people, but it does
attract and particularly suitpeople who have a real attention
(03:13):
to detail.
SPEAKER_01 (03:14):
You mentioned kind of there's a perception of
radiology happening behind thescenes.
And I wondered, from yourperspective, what's it really
like doing the work?
SPEAKER_00 (03:23):
Oh, it's, I mean, it's amazing.
It's so varied.
Honestly, every week, I've beendoing it for 30 years and every
week I see something that I havenever seen before or come across
a situation I've never seenbefore.
So it's incredibly varied.
And also it's developing soquickly.
We've been so lucky to alwaysreally be at the forefront of
(03:46):
innovation and development, newimaging techniques.
When I was training, there weresome people who had not really
seen much ultrasound when theywere training.
You know, ultrasound came insort of before I started, but
then MRI was really new when Istarted.
And now we do hundreds andthousands.
(04:07):
And now we've got AI.
So it's terribly varied on a dayto day.
Sometimes you're scanningpeople.
Sometimes you're looking atimages.
Sometimes you're interactingwith your colleagues.
And also in terms of innovationand tech, it's got something for
everybody there.
SPEAKER_01 (04:23):
For you personally, what do you find most rewarding
about the work?
SPEAKER_00 (04:28):
So I love the variation.
I love dealing with children.
I love...
Working in a team, that is sogreat.
And again, it's one of thethings that people don't really
realize about radiology.
I think the times when we addmost value to patient care is
when the pediatric surgeons orsome of the pediatricians might
(04:51):
come and talk to us and say,we've got this child and this is
their issue.
This is their socialcircumstance.
These are the images.
Let's put all that together andsee what's best to do for that
person now or that family.
And then you're working with areally focused team trying to do
(05:11):
the right thing for somebody whooften can't really stand up for
themselves.
And that feels like propermedicine.
SPEAKER_01 (05:24):
So I think one of the things that I was interested
in talking to you about is thefact that you're focusing on
solving some of the most complexissues in diagnostics.
There are some pretty urgentissues that you're working on.
We know that over 65,000patients are waiting too long
for scans.
And I wondered if you had a viewon what's driving these delays
and what are some of the ways wecan tackle these problems?
SPEAKER_00 (05:47):
The demand for imaging will keep on increasing.
And that's right.
Imaging is good for patients.
It's much better to diagnosethings quickly and early.
So more imaging overall...
is a good thing.
And probably we do less.
Well, we definitely do less thanthey do in other countries.
(06:08):
So the demand will keep onrising.
But as you say, people arewaiting too long.
And even though we areincreasing the workforce, we're
never going to be able toincrease it fast enough to keep
up with that increase in demand.
So we do have to innovate.
We do have to do somethingdifferently.
(06:28):
And there are many, many ways inwhich we can do that.
But what our focus should be onnow is in creating capacity in
the system.
So AI can help us, definitely.
Tech can help us.
A lot of the time in the NHS atthe moment, digital solutions
actually don't help us.
(06:49):
They often take a lot ofclinical time.
One of the effects of the sortof increasing digitization in
the NHS has been to move quite alot of administrative tasks from
low-paid workers to highly paidworkers because now most
consultants will input their owndata and we're often pretty slow
(07:10):
at that.
And the systems are often clunkyand you log in lots of times and
then you forget your passwordand it logs you out and the sort
of everyday experience.
So digital systems aren'tworking for us and we need to
change that.
We need to make sure that thosepeople who are highly trained
clinicians, not just doctors,but radiographers, nurses, those
(07:32):
people spend most of their timedoing what they've been trained
for.
So that's the first thing.
We need to concentrate on AIhelping us with flow through the
system.
We also need to look at whowe're imaging.
So as I've said, overall, moreimaging is better for patients,
but we still are not necessarilyselecting the people who could
(07:56):
benefit most.
And in a resource-constrainedsystem, which we're all in one
way or another, but we have tomake sure that we are, as it
were, getting the biggest bangfor our buck in terms of
imaging.
And we can use AI to do that,but we also have other systems
in place.
We also have a system...
(08:18):
called iRefer, which is now aclinical decision support, which
actually helps people to choosethe right sort of imaging.
So if you're a primary caredoctor, for instance, you might
go onto your computer and think,well, I want to request this
person's got a headache.
I might want to request, andthey've had it for ages, I might
want to request a CT.
So they'll go and request thatCT, but then the system will
(08:41):
say, well, actually, no, an MRIis better for that.
So instead of wasting that CTslot, we've gone straight to the
most appropriate imaging.
And sometimes the clinicaldecision support software will
say, actually, imaging is reallynot going to help here.
It doesn't really move the dial,but it does mean that some of
the studies don't take place.
(09:01):
And it also means that thosestudies that do take place are
ones which are the bestpractice, you know, most likely
to make a difference for thatpatient.
So that's the sort of thing weneed to concentrate on.
And we need to also ensure thatwe've got the data to understand
which tests really are making adifference to patients.
(09:24):
And at the moment, we're notvery good at linking that up.
So we know we've got a lot ofdata around what tests we do,
but we're not terribly good atlinking up which ones really
make a difference to patientswith clinical data, with outcome
data.
So we need to get better atthat.
SPEAKER_01 (09:39):
And what about workforce?
Is there a lack of radiologistsin the system?
SPEAKER_00 (09:44):
Huge lack.
At the moment, there's about a30% shortfall.
And actually, that's due to goup to 40% in the next five
years.
So, I mean...
That's what's preventing us fromkeeping up with the wait.
And it's also terribly stressfulfor people.
No one goes into medicine orradiology to provide a bad
(10:06):
service.
Everybody wants to do the rightthing for patients.
And so people feel a hugepressure.
The staff...
It's, you know, when in COVID wecalled it moral injury, you
know, staff are trying so hardto do their best for patients,
but they're so constrained bythe volume that's coming at them
that they actually can't do it.
(10:26):
And of course, the difficultywith that is that feeling of
being out of control, of notbeing able to do what you always
wanted to do.
drives people away.
And so they retire early, theygo into some other branch or
whatever.
And so it has a sort ofspiraling downward effect.
(10:49):
So people are leaving ourprofession much earlier.
The retirement age, average,well, the average age of leaving
the service is around 54 now,which is so young.
So we have to make things betterfor people.
So as we keep these highlytrained people in the service.
SPEAKER_01 (11:06):
And how do we create better working conditions that
are going to retain thosepeople?
Similarly, where does servicedesign come into this?
SPEAKER_00 (11:13):
Oh, well, this is the big question, isn't it?
And I think there's been a lotof good work on this, but it's
something that we haven't reallymanaged to get to the bottom of
in the NHS, I would say.
And it has to be about lookingafter the people and getting the
best out of the people.
So I think there are...
(11:35):
many ways in which we don'treally make the most of our
staff.
And very good work by MichaelWest about caring for doctors,
caring for patients, a great,great publication there, which
talks about autonomy, the thingsthat people value about their
work, autonomy, belonging, andcontrol.
And of course, a lot of thosethings, control, we've talked
(12:00):
about lack of control, you can'tmanage the workload, Belonging,
we've lost that a little bit.
We're not really creatingcommunities into which people
belong.
And some of the technicaladvances since COVID, where
quite a lot of us are remote,not getting together, doing
stuff on teams, that actuallyundermines belonging even
(12:21):
further.
And the autonomy, people feeltoo much top-down control,
right?
I mean, it's important that weuse data, but sometimes we use
data as a sort of weapon againstour staff.
And we've got to be very, verycareful about that.
So it is about looking after thepeople.
(12:43):
And I believe in the NHS, wedon't really have the structures
in place to do that.
So If you're a clinical leader,you often get very, very little
time to look after your staff.
So you have maybe 12 hours aweek for a department with maybe
70 consultants.
(13:04):
So then you won't really be ableto reach out to your staff.
We don't really have a structureof having one-to-ones with
people.
In any other company, when Icame here to work at the Royal
College, for instance, the CEOwill have one-to-ones each week
with his, you know, the peopleon the sort of tier below and
they will have one-to-ones withtheir team and they will have...
(13:25):
So there is a kind of web, anetwork that spreads through the
organisation.
We don't really have that in theNHS.
And so it's very hard to seehow...
culture and values can reallyeffectively spread down through
the staff.
We've concentrated so hard onkeeping people working, doing
(13:48):
their clinical work.
SPEAKER_01 (13:50):
What's the peer community like in radiology?
SPEAKER_00 (13:54):
Well, it's very strong.
We have some really good thingsin radiology, which have made
our peer community very strong.
So one of those is looking atmistakes or discrepancies, as we
call them.
The thing about radiologists isthat if you miss something on a
scan or misinterpret something,it's there forever for everybody
(14:15):
else to have a look at.
And so you might miss a noduleon an x-ray or interpret it as
benign when in fact it ismalignant, tragically.
And those things will happen.
And then Ever after, it's thereand you can see it and you feel
absolutely terrible about it.
(14:36):
When you go back to look at it,you think, how could I have
misinterpreted that?
How could that happen?
But because of that, we have awhole system in radiology called
radiology events and learning.
And what we do is we, any caseswhere there's a discrepancy of
some sort, we submit thoseanonymously and have a weekly
meeting or monthly meeting andwe go over those.
(14:58):
So we'll all as a team look atthem and say, oh gosh, that was
difficult or, oh yeah, maybe.
But what happens is one, welearn from those mistakes.
So, you know, we're taking thatas a opportunity to learn.
And the other thing is that eventhe most junior person
understands that other peoplemake mistakes too.
(15:20):
And some of those other peopleare really senior and really
experienced.
And as a community, that teachesus about error, about supporting
each other, about using error asa learning opportunity.
And That makes us a very, verystrong community.
And I do think that otherbranches of medicine could learn
(15:44):
a lot from that because it's notonly good for the learning from
your craft, but it's good forthe culture of departments.
SPEAKER_01 (15:52):
Does it feel like your peers have your back when
you're doing this work?
SPEAKER_00 (15:55):
Definitely.
Yeah, definitely.
When within our profession, weall understand, we all recognise
that we don't get things rightall the time.
But I mean, that's true ofmedicine, of all aspects of
health care.
We don't get things right allthe time.
What's important is that welearn from it.
(16:16):
But I think we don't prepare ouryoung health professionals very
well for the fact that they willmake mistakes.
And sometimes those mistakeswill have very bad outcomes for
people.
That is the nature of ourbusiness.
SPEAKER_01 (16:31):
It's impressive the way medically trained people are
taking on so much responsibilityand so much risk with their
work.
And that's just normal in a day
SPEAKER_00 (16:42):
to day working lives.
not let that mistake go towaste.
SPEAKER_01 (17:08):
So I'd like to move on to talk a bit around AI
diagnostics and how we keep ithuman.
There's lots of excitementaround AI as well as complexity.
It's part of the diagnosticinfrastructure and has become
over the last few years.
What do you see as the realpotential here?
SPEAKER_00 (17:23):
Well, it's very exciting.
That's the first thing.
And very welcome because we'vetalked about the fact that we
have capacity problems.
And so we need all the tools wecan get to improve, you know, to
maximize the time that staff canspend doing the things they've
been trained to do.
AI should help our productivityor efficiency, whatever you want
(17:45):
to call it.
At the moment, we've beenconcentrating.
I mean, about 80% of all thealgorithms available are in
radiology.
So it's really in ourwheelhouse.
At the moment, a lot of the techis concentrating on helping us
with accuracy, which is alwayswelcome.
But it's not our biggestproblem.
Our biggest problem is capacity.
(18:06):
So we might need to sort of moveour attention a little bit from
things that help us withdiagnosis to things that help us
select the right patients forimaging and also smooth the
patient's pathway through theirjourney through the imaging
department and also just make usa bit more efficient so there
(18:29):
are things for instance likereporting solutions which can
quickly record so you can justtalk to a computer and it can
reorder your stream ofconsciousness into a structured
report and also makerecommendations.
So you might see some sort of athing on a renal cyst, for
(18:51):
instance, a fluid filled thingon a kidney.
Now there's guidance about whichcysts should be followed up and
which you don't need to worryabout.
And I don't have that in myhead.
So I would have to look that upon a computer.
I think, well, I know there'sguidance, but I'm going to have
to look this one up.
But, you know, the AI can youthat as you're doing the report.
So it just makes you muchquicker and more efficient.
(19:13):
So we need to concentrate on thesort of admin side of things to
help us do our thing morequickly.
In terms of the diagnostic AI,we've still got quite a long way
to go, I think, in deciding howto use that best.
We have systems, we'redeveloping systems to look at
(19:35):
the accuracy of the algorithmitself, the accuracy of the
output of the computer.
But actually, what reallymatters is what is the
combination of the computer andthe human?
Is that better?
Is it worse?
Sometimes it can be worse.
Do patients who have had thattechnology used, do they do
(19:55):
better?
And those questions we stillactually don't really know the
answers to.
So there's a lot of work to bedone.
SPEAKER_01 (20:04):
And where does the responsibility then lie?
Because, you know, theconversation we were talking
earlier about that risk thatyou're taking on and trying to
make decisions and getting itwrong at certain times.
Does that then mean that ifyou've done the scan with an AI
device, And you've had the samedecision as the AI.
It's almost abdicatingresponsibility to the
technology.
Or does that responsibilityalways lie with the radiologist?
SPEAKER_00 (20:26):
Well, it always lies with the radiologist.
And so one really important partis to educate people about what
the technology can do and whatit can't do.
Because the technology, likehumans, is good at some things
and not good at others.
So the technology does missthings.
For instance, I heard of a casewhere it can actually blind you
if it's telling you to looksomewhere You concentrate on
(20:49):
that and you don't look at quiteobvious things that you might
have seen somewhere else.
So I heard of a case the otherday where the AI would only pick
up lumps of a certain size orbelow a certain size.
And actually that meant that theradiologist missed a much bigger
lesion Because they were sort ofdistracted by the AI and that
(21:11):
can happen.
So we need to, first of all, toeducate people what the AI can
do and what it can't and whatyou still need to look for.
And then we need to monitor itreally carefully and we need to
describe what those monitoringsystems are.
But at the moment, it's alwaysthe radiologist's responsibility
to There are some algorithms outthere now that say that they can
(21:35):
manage without a radiologist.
For instance, there are somethat say that they can
effectively identify normalchest X-rays or a proportion of
normal chest X-rays, whichwouldn't need to be reported at
all.
So if that were to happen,there'd need to be some changes
in the law.
We'd need to describe verycarefully what sort of quality
(21:56):
assurance needs to be in place.
to protect patients whose scanshave gone through that.
And the responsibility forsetting that system in place
would be with the hospital andthe radiologist.
SPEAKER_01 (22:10):
You hear the phrase human in the loop a lot when
we're talking about medicine andtechnology.
What's the implication of thatin this area?
SPEAKER_00 (22:17):
So at the moment there always has to be a human
in the loop.
I think we need to think aboutthat moving forward, whether
that is always the right way.
As I've said, sometimes thecombination of the human and the
computer can mean that it can beworse, actually, than either one
alone.
So we need to understand that.
(22:39):
And I think we also need todescribe the systems that need
to be in place for some thingsthat could be done without a
human in the loop.
What our polling showed aboutpublic attitudes towards AI was
that people are happy for AI andtechnology to be used, but they
(23:02):
do trust doctors and expectdoctors to implement that
safely.
So I think that is where ourresponsibility lies for making
sure that these are happeningsafely.
And I think if patients gettheir results quicker, they're
really happy.
SPEAKER_01 (23:17):
I mean, there's a balance, isn't there, between
speed, safety and empathy?
Totally.
How do you get that balanceright?
SPEAKER_00 (23:24):
Well, we've got a lot of work to do, I think is
the answer.
But that is entirely what...
doctors and particularlyradiologists have been doing for
many years.
So that's absolutely 100% wherewe can contribute.
And we've got lots and lots ofexperts who can work on that.
And it's very exciting work.
And it promises to be able torelieve some of the capacity
(23:47):
issues that we've got and helpus to deliver a better service.
So it is a lot of work, butwe're the right people to do it.
And we're very happy to takethat on.
SPEAKER_01 (23:58):
We're talking a lot about kind of technical
innovation in the space.
I wondered what your thoughtsare around innovation around
service experience and, youknow, what it's like to go
through radiology.
SPEAKER_00 (24:10):
Well, yes.
And I think that's a veryinteresting question.
And I think one of the thingsthat we're very poor at in
radiology is communicating withpatients.
Often Patients don't reallyunderstand where or when they're
going to get the results anddon't even understand that once
(24:31):
you've had your scan, there'sanother step.
You know, the other step ishaving that scan reported and
analyzed by a radiologist.
So patients could really do witha bit more information.
So as they had a bit morecontrol over what's happening, I
think had previously had a roleas the GERF lead, the Getting It
(24:51):
Right First Time lead forradiology, where I went and
visited 143 radiologydepartments in England.
It was an NHSE project.
I really saw people doing greatthings in this field, people who
had patient navigators.
So people employed withinradiology to work with patients
and say, yeah, this is what'sgoing to happen now.
(25:13):
And this is where you're goingto get your results.
And Sometimes concentrating onjust getting more examinations
done and getting more reportedmeans that we don't concentrate
on patients as we should.
Some people are doing good work,but it's sporadic.
SPEAKER_01 (25:29):
And how much work goes into the communication
following the diagnostic.
So I was talking to JonathanGregory in the previous podcast
who was talking about workthey've been doing with
Macmillan around end oftreatment communication and
actually simplifying the lettersand using not medical terms but
terms that your average personcan understand in the letters.
Is that normal practice inradiology?
SPEAKER_00 (25:50):
Well no it's not and actually you know most people in
radiology get their result fromanother specialty.
So the way it happens is I willwrite a report and that report
will go to either the GP or thepediatrician or the pediatric
surgeon or whoever.
And that person will be theperson who communicates the
results to the patient.
(26:11):
So my reports are not writtenfor patients.
They're written for otherclinicians.
But now, a lot of reports arenow visible on the NHS app.
Which is the right thing, butthey're not written in a way
that is necessarily accessibleto all patients.
(26:33):
And they can be alarming.
You know, you might say there'sa possibility of cancer because
there is a possibility, but it'snot very likely.
And I mean, it depends who youare, but some people find this
very stressful reading.
So this is an issue.
SPEAKER_01 (26:48):
But if you've written that letter for another
medical professional, it's goingto be a very different letter
from the letter you would writeif it was going to be viewed by
a patient.
SPEAKER_00 (26:56):
Absolutely.
So we've had these conversationswithin radiology.
So should we change the way wereport?
But actually, that's probablynot the right thing to do.
The right thing to do is to putin some other step which helps
patients understand or enablesthem to access health.
(27:16):
So either you could usetechnology.
And in the States, they do thisa lot so that all patients have
access to their results.
So the things that they do inAmerica now are they have little
videos and things embedded intheir reports.
So you might say you have a tearof your medial meniscus in your
knee, and then there'll be asort of link onto that of this
(27:38):
is what a normal meniscus lookslike, and this is what yours
looks like.
And that's sort of technologythat does that.
And then you have a link to saywhat normally happens.
And so They also are employingpeople in radiology departments.
So if a patient's worried abouttheir report, they can then ring
up and discuss it with aradiologist.
So they have a whole system inplace to look after it.
(28:01):
And there are no shortcuts forthis.
If we're going to do this, weneed to do it properly because
we do need to look afterpatients properly.
SPEAKER_01 (28:09):
And that all comes down to funding, I assume.
SPEAKER_00 (28:10):
Absolutely.
And recognising that it's anissue.
SPEAKER_01 (28:14):
So just going back to the getting it right first
time work that you were leading,which was a massive effort to
understand and improve practiceacross the country.
What were the biggest challengesand lessons from leading that
redesign at scale?
SPEAKER_00 (28:26):
Oh, well, it was the most amazing experience, the
most amazing experience.
One, Tim Briggs, who started thewhole program.
I mean, it was visionary andamazing.
He has worked tirelessly to getthis program off the ground, not
only in orthopedic surgery, butin everything.
He's done really great thingsfor medicine in the UK.
So with that program, andactually that was real
(28:50):
leadership.
He just did it and took otherpeople with him.
He saw there was a problem andhe did it.
So that's one lesson.
If you think there's a problem,no matter who you are, just go
for it.
And I learned that so many timesthrough that trip.
So during the process, you takedata, whatever data you have,
and you go and discuss that withthe different departments.
(29:12):
It works best when you can geteverybody in the department into
that.
So it'd be like the chiefexecutive or the medical
director or both, and then allthe helpers and the porters and
the PACS people.
And then you can start aconversation and actually you
can just unlock things in theroom.
The chief executive would say,well, why haven't you got
anywhere to do that?
(29:33):
And they say, well, nobody'sfound me.
And they say, well, we'll getthat now.
You know, so that's amazing.
And it was so I met so manypeople who were just working
really hard in really difficultcircumstances to do the right
thing for their localpopulation.
Inspirational, totally.
SPEAKER_01 (29:52):
And how important was the collaboration in there?
SPEAKER_00 (29:54):
Well, the collaboration goes all the way
through.
I mean, one with your otherclinical leads, the other
clinical leads from the otherspecialties were incredible.
And it was great to talk tothem.
Two, the collaboration and theteam are quite small.
I got great advice from one ofthe past presidents of the Royal
College of Radiologists, who ismy co-lead, Giles Maskell, and
(30:17):
had an absolutely wonderfulproject manager, Gail
Roadknight, who ran the wholething, Elaine And Quick was a
radiographer who came with usand a service manager and Lucy
Proling, all of these people.
That was a great team.
And then when you get there,often the helper will never have
been in the same room as themedical director or the CEO.
(30:40):
And so to get people in the sameroom and have that conversation,
that's the thing that is magic.
SPEAKER_01 (30:47):
How did you do that in a meaningful way?
People are so busy in thesystem.
SPEAKER_00 (30:51):
Well, our project manager did a lot of work on
that.
And it depends on the leadershipin the trust as well.
Those really good trusts willsee this as an opportunity for
them to get to know more aboutwhat is really happening on the
ground, led by someone else.
And it's free.
You know, really good leadersembrace this process.
(31:12):
Those ones that are not led sowell, it's really hard.
And we've flogged off to variousplaces and there'd be like two
people there.
And so, you know, and you walkinto a department immediately,
immediately you can tell aboutthe leadership.
A hundred percent.
Leadership is so important.
I mean, everybody says this, butactually you can tell the minute
(31:34):
you go into a department if it'sgot a strong leader.
SPEAKER_01 (31:38):
I'm a big advocate for co-design when you're
thinking about change andbringing people together to
think about how those changescan be made in a way that works
for them.
It sounds like this was aco-design approach you were
taking with these teams.
Can you tell me a bit more abouthow that made the changes stick
or examples of how it worked?
SPEAKER_00 (31:55):
Absolutely.
I mean, you know, the innovationthat goes on is unbelievable.
I think people very rarely getthe chance to be together to
talk about the innovations thatthey've done.
I mean, one brilliant innovationwas around paediatric MRI
scanning, so scanning toddlers.
(32:16):
You know, for an MRI, you needto keep still for quite a long
time.
And of course, toddlers can't dothat.
But most of the time, we need ananaesthetist and, you know, a
whole load of equipment that youcan take into the MRI scanner
and things like that.
So it's a very expensive,labour-intensive process that
can't be done in a districtgeneral hospital.
It has to be done in a bighospital.
(32:37):
In one place I went to, thehelpers there had said, toddlers
sleep really heavily.
Why don't we do them at bedtime?
So they got the toddlers intothe department around about
seven o'clock and they bringthem in.
They had a travel cot.
They bring their kind of blankieand their bottle or whatever and
play with them for a littlewhile, put them to bed in a cot.
(33:00):
And then they'd fall asleep.
Then they'd put some dear littleheadphones on, lift them up, put
them in the MRI scanner, do a 45minute MRI scan.
no medication, no nothing, thenput them in their car seat and
take them home.
SPEAKER_01 (33:13):
So no anaesthetic?
SPEAKER_00 (33:14):
No anaesthetic.
SPEAKER_01 (33:16):
And they would lie still enough for that?
SPEAKER_00 (33:17):
Yeah, they're fast asleep.
And, you know, this, I've been apaediatric radiologist for 20
years by the time, I'd neverthought of that, you know, and
this was, and in fact, it's goneon and it's other people have
had the idea as well around thecountry, but now this is now
much more standard practice.
And this was a helper.
This was a very low band personsaying, your toddler sleep
(33:38):
really heavily normally.
It's just, and that sort ofthing is just light bulb.
SPEAKER_01 (33:46):
It sounds really obvious once you've talked about
it.
Of course it makes sense.
SPEAKER_00 (33:50):
I could not believe that I'd never thought of that.
And I was just, oh, oh yeah, ofcourse.
SPEAKER_01 (33:56):
Easier for you and easier for the toddlers as well.
SPEAKER_00 (33:58):
Better for them.
I mean, general anesthetics, youdon't want to give little
children general anesthetics.
We don't think it's much goodfor them.
SPEAKER_01 (34:04):
So the changes that came out of this process that
you were leading, were theytop-down changes or was it more
kind of participatory and peoplewere making the changes
themselves?
SPEAKER_00 (34:13):
It's more participatory then.
And I think that's the thingabout health service
improvement.
I mean, it's a complex system.
So you just edge things forwarda little bit each time.
But what it did was, first ofall, it showcased what people
were doing in radiology.
So in previous times, radiologyhas often been regarded as a
problem, you know, a blocker.
(34:35):
We can't get things throughradiology.
And when the more seniormanagers in the hospital had it
showcased to them all the thingsthat radiology All the different
departments in the hospital orprimary care, the many different
sorts of things they did.
And they would suddenly think,wow, that's something actually,
that is something to be dealingwith all of that.
(34:58):
And just as you say, co-design,just discussing some of the big
issues they had.
Sometimes, one department Iremember, they couldn't get as
many people through the CT scanbecause when they needed to
cannulate the patients, theydidn't have a separate room to
do it.
They had to do it on thescanner.
So that was wasting scan time.
And they said, well, we'veasked, but nobody's given us
(35:19):
one.
The chief executive was justable to say, oh, right, we're
doing that now.
And so getting people togetherand making time to do that is
absolutely crucial.
We are squandering the expertisewe have in the health service
because we are not reallylistening to the things that
people know could be improved.
SPEAKER_01 (35:45):
I'd like to move on and...
speak about leadership a bitwith you now.
So you've written about theurgent need to value leadership
in medicine more.
What does great leadership looklike to you in the system today?
SPEAKER_00 (35:58):
Well, I mean, there is something about authenticity,
about people, you know, it'sactually not about necessarily
learning or it's just aboutbeing yourself and trying to
make things better and otherpeople seeing that you are
trying to do the right thing.
So I think you just have to dowhat you do.
(36:22):
So I don't think there'snecessarily a right way or a
wrong way.
I do think we really hamperpeople.
When we talk about leadership inthe NHS, we tend to take people
away, give them a learningcourse, do their Myers-Briggs
score or whatever.
But actually, we make it reallydifficult for people because we
give them no time.
We give them no support.
(36:42):
So actually, as a clinicaldirector in a radiology
department, what you need isdata.
You need to be able to reallysee what's happening.
And you need a person whounderstands healthcare data to
help you.
You need good secretarialsupport.
You need time to work with yourstaff and look at ideas and just
hear how they are.
(37:02):
And If you don't have that, nomatter how much training you
have, you won't be able to besuccessful.
It's a bit like sort of victimblaming in a way in that people
are really trying hard and we'resaying, oh, you need more
training.
You know, it's because you don'tknow how to do it.
It's not because they don't knowhow to do it.
It's because we're making itimpossible.
SPEAKER_01 (37:25):
How do you personally lead when time and
energy are in short supply?
UNKNOWN (37:29):
Yeah.
SPEAKER_00 (37:29):
Well, I honestly, I don't really think of myself as
a leader.
I just rely on other people.
I see other people who I get alot of help.
I mean, in this job, I get lotsand lots of help, which is
great.
So one of the things is when youget more senior, it all becomes
more easy because you get a lotof help.
(37:51):
And also people listen to whatyou say, which is something that
You know, for most people, thatjust doesn't happen.
So it gets easier.
And also, I surround myself.
I'm lucky to be surrounded withpeople who've got a lot of good
ideas.
SPEAKER_01 (38:04):
You've highlighted the lack of time, support and
recognition for clinical leaderspreviously.
What's the impact of that?
SPEAKER_00 (38:12):
The impact is very bad on the person who's trying
to do the leading.
So we don't have many peoplestepping up for medical
leadership.
It's seen as impossible.
Sometimes people think, oh,you're going over to the dark
side, you know, because medicalleaders don't have enough time
to develop their staff and towork with their staff.
(38:34):
It can be seen as quiteconfrontational.
So people who step up who aretrying to do the right thing, it
takes a huge toll on them.
And it means that people arescared to step up.
And it means we are not makingthe most of the talent that we
have.
So we've got to really changethat because nothing's going to
change unless we can persuademore people to do that.
SPEAKER_01 (38:58):
And what are the things that you need to change?
SPEAKER_00 (39:00):
So I think we need to support the people more who
are doing those leadership jobs.
Give them time.
Give them resource.
You know, to economize onadministrative staff and have
the people who are the clinicaldirectors doing all the admin
work, that's ridiculous.
The number of doctors who arewriting rotas, I mean, that's
(39:22):
ludicrous.
We need to give people supportand time.
And Often we use our data tosay, well, if everybody just
works a bit harder, we can do abit more.
And that's just not the way todo it.
We need more fundamental change.
So we need to give people thetime to innovate, to make more
time.
But that's quite a brave step.
SPEAKER_01 (39:42):
Mentorship is important.
And I think it's played a rolein your own career.
Can you tell me a bit more aboutthat?
SPEAKER_00 (39:48):
A hundred percent.
I have had some wonderful peoplewho have...
faith in me and supported me andtaught me so much.
When I started in medicine, itwas a much more hostile
environment for women,particularly.
I still do think we have anissue with women's voice not
(40:09):
being heard.
We have a lot of data aroundthat.
I've had a lot of very strongmale mentors who have helped me
and supported me throughout.
Tim Briggs, I've mentioned, myhusband, Giles Maskell, who I
mentioned in Girfed.
A lot of people who've reallysupported me.
and taught me to just have faithin myself.
(40:30):
And I would like to try and passsome of that on to other people.
SPEAKER_01 (40:34):
What advice would you give someone who's thinking
about maybe stepping into aleadership role in the health
system?
SPEAKER_00 (40:41):
I'd say do it.
Do it.
There are huge rewards.
That's the way to make adifference.
Get yourself some good support.
Don't blame yourself too much.
Learn from mistakes, but don'twallow in them.
SPEAKER_01 (40:56):
Before we wrap up, I'd like to look ahead to the
future a little.
What are you most optimisticabout right now?
SPEAKER_00 (41:03):
Well, one of the things I get in this job is I
get to look around other healthservices.
We're very fond of tellingourselves how everything's
terrible and there's definitelya lot of room for improvement.
But actually, healthcare in theUK is not as bad as we think.
There are a lot of places in theworld where if you're poor, you
get no access to healthcare.
And that's not the way it is inthe UK.
(41:25):
I mean, there are tremendoushealth inequalities, we could
definitely improve, but at leastwe have healthcare that is free
for people.
So that's good.
We have a hugely wonderful staffand we have great education,
great systems in place.
We're very lucky in that regard.
You know, the Royal Collegesreally support education and
(41:49):
standards.
So our healthcare system isactually not too too bad.
And we've got the buildingblocks to do a great deal more.
SPEAKER_01 (42:00):
How do you see radiology evolving over the next
five, 10 years?
SPEAKER_00 (42:04):
It's going to change, no question.
We can't continue to doeverything ourselves.
Technology is really going tohave a big effect.
We need to lead that change andmake sure it's having a good
effect and not just makingthings more complicated.
We will still be needed.
When you're dealing with humansand You need another human to
(42:24):
understand the complexity.
We'll be doing more clinicalliaison with other groups, but
patients are getting morecomplicated.
People have the multimorbidity.
People have more than onecondition.
More people have more than onecondition.
So you need to understand theinterplay as well as with social
factors and human factors.
(42:45):
And that's where radiologistswill really have a lot to offer.
SPEAKER_01 (42:49):
If you could wave a magic wand and change one thing
about the system now, what wouldyou change?
I
SPEAKER_00 (42:55):
think I'd change the sort of structure.
I'd put that one-to-one systemin.
So I'd make sure that everybodyin the health service had a
regular one-to-one with theirline manager, which was not
necessarily sort of judgmental,but it was kind of developmental
with their manager.
So as we have that web goingthrough the health service of
(43:18):
good ideas being able to come upfrom the bottom and culture
being spread down from the top.
I think it's about making surewe have time to develop our
staff and every staff membershould have regular time with
their manager.
SPEAKER_01 (43:32):
What's next for you, Kaf?
SPEAKER_00 (43:33):
Well, I'm not quite sure.
I mean, it's been an amazingexperience being here at the
college.
I've loved it.
Some wonderful people here.
And I would quite like to keeptrying to do some service
improvement work.
So anybody out there want to...
I'm available.
SPEAKER_01 (43:52):
Amazing.
Kath, thank you so much fortaking the time to talk to me
today.
SPEAKER_00 (43:56):
Thank you.
It's been a real pleasure, Sam.
One of the perks of the job.
UNKNOWN (44:01):
Thank you.
SPEAKER_01 (44:09):
Problems Worth Solving is brought to you by
Healthier, the collaborativeservice design consultancy for
transformation in health careand public services.
Find out more about how we canhelp you deliver user-centered
change at healthier.services.
Hello, this is Problems Worth Solving, the
podcast where we meet peopletransforming health and care
through human-centered designand digital innovation.
I'm Sam Mentor, ManagingDirector at Healthier, the
(00:21):
collaborative service designconsultancy.
If you enjoy listening, you cansubscribe to this podcast and
the accompanying newsletter athealthier.services.
Today's guest is Dr.
Catherine Halliday, a leadingfigure in UK radiology and
president of the Royal Collegeof Radiologists.
She's led national work toimprove diagnostic services,
(00:44):
shape policy and leadresponsible innovation.
Today, we're going to talk aboutleadership, AI, service
redesign, and how ahuman-centred approach can
unlock change in one ofhealthcare's most pressured
areas.
Thank you so much for joining ustoday, Catherine.
It's great to have you here withus.
SPEAKER_00 (01:01):
It's really nice to be here.
Thanks, Sam.
Thanks for inviting me.
SPEAKER_01 (01:05):
So I'd like to start with a little bit about you as a
person so that we can get abackground and understand what
drew you into this space.
So I'm curious about what youwere interested in as a young
person and how that led you intoradiology.
SPEAKER_00 (01:20):
Well, I always wanted to be a doctor.
I can't remember a time when Ididn't want to be a doctor.
Initially, actually, I wanted tobe a psychiatrist.
I've always been very interestedin what makes people behave the
way they do.
It took me a while to get intomedical school.
I didn't get in straight awayand I kept trying.
And then I met lots ofinteresting people when I
finally did make it.
(01:40):
So my career's had lots of upsand downs, but I've never
regretted being a doctor for asecond.
SPEAKER_01 (01:46):
And what type of person tends to go into
radiology?
UNKNOWN (01:50):
Thank you.
SPEAKER_00 (01:50):
Well, that is a very interesting question, actually,
because the role of theradiologist is really changing.
And traditionally, I mean, thereare a number of role models.
I don't know, maybe many peoplelistening to this will be too
young to remember Green Wing onthe television, where the
radiologist is a person wholives in a sort of dark dungeon
(02:10):
down below and doesn't reallyspeak to anyone and actually
gave himself coffee enemas, Ithink.
It was a very, very odd person.
And that is are traditional typethough I've never met anybody
like that I must say buttraditionally a radiologist is
somebody I think The stereotypeis that it's somebody who
(02:31):
doesn't really like to interactwith patients or people, likes
to live in the dark and just getthrough a load of tasks.
But actually, in reality, ourjob is very much about people.
It's very much about dealingwith our colleagues, about
understanding patients and thecontext in which their imaging
(02:52):
has taken place.
And of course, I'm a paediatricradiologist, so even more, I'm
very sort of hands-on in termsof dealing with children and
their parents.
So it's not quite as withdrawn,I think, as you might think.
And of course, we get all sortsof different people, but it does
attract and particularly suitpeople who have a real attention
(03:13):
to detail.
SPEAKER_01 (03:14):
You mentioned kind of there's a perception of
radiology happening behind thescenes.
And I wondered, from yourperspective, what's it really
like doing the work?
SPEAKER_00 (03:23):
Oh, it's, I mean, it's amazing.
It's so varied.
Honestly, every week, I've beendoing it for 30 years and every
week I see something that I havenever seen before or come across
a situation I've never seenbefore.
So it's incredibly varied.
And also it's developing soquickly.
We've been so lucky to alwaysreally be at the forefront of
(03:46):
innovation and development, newimaging techniques.
When I was training, there weresome people who had not really
seen much ultrasound when theywere training.
You know, ultrasound came insort of before I started, but
then MRI was really new when Istarted.
And now we do hundreds andthousands.
(04:07):
And now we've got AI.
So it's terribly varied on a dayto day.
Sometimes you're scanningpeople.
Sometimes you're looking atimages.
Sometimes you're interactingwith your colleagues.
And also in terms of innovationand tech, it's got something for
everybody there.
SPEAKER_01 (04:23):
For you personally, what do you find most rewarding
about the work?
SPEAKER_00 (04:28):
So I love the variation.
I love dealing with children.
I love...
Working in a team, that is sogreat.
And again, it's one of thethings that people don't really
realize about radiology.
I think the times when we addmost value to patient care is
when the pediatric surgeons orsome of the pediatricians might
(04:51):
come and talk to us and say,we've got this child and this is
their issue.
This is their socialcircumstance.
These are the images.
Let's put all that together andsee what's best to do for that
person now or that family.
And then you're working with areally focused team trying to do
(05:11):
the right thing for somebody whooften can't really stand up for
themselves.
And that feels like propermedicine.
SPEAKER_01 (05:24):
So I think one of the things that I was interested
in talking to you about is thefact that you're focusing on
solving some of the most complexissues in diagnostics.
There are some pretty urgentissues that you're working on.
We know that over 65,000patients are waiting too long
for scans.
And I wondered if you had a viewon what's driving these delays
and what are some of the ways wecan tackle these problems?
SPEAKER_00 (05:47):
The demand for imaging will keep on increasing.
And that's right.
Imaging is good for patients.
It's much better to diagnosethings quickly and early.
So more imaging overall...
is a good thing.
And probably we do less.
Well, we definitely do less thanthey do in other countries.
(06:08):
So the demand will keep onrising.
But as you say, people arewaiting too long.
And even though we areincreasing the workforce, we're
never going to be able toincrease it fast enough to keep
up with that increase in demand.
So we do have to innovate.
We do have to do somethingdifferently.
(06:28):
And there are many, many ways inwhich we can do that.
But what our focus should be onnow is in creating capacity in
the system.
So AI can help us, definitely.
Tech can help us.
A lot of the time in the NHS atthe moment, digital solutions
actually don't help us.
(06:49):
They often take a lot ofclinical time.
One of the effects of the sortof increasing digitization in
the NHS has been to move quite alot of administrative tasks from
low-paid workers to highly paidworkers because now most
consultants will input their owndata and we're often pretty slow
(07:10):
at that.
And the systems are often clunkyand you log in lots of times and
then you forget your passwordand it logs you out and the sort
of everyday experience.
So digital systems aren'tworking for us and we need to
change that.
We need to make sure that thosepeople who are highly trained
clinicians, not just doctors,but radiographers, nurses, those
(07:32):
people spend most of their timedoing what they've been trained
for.
So that's the first thing.
We need to concentrate on AIhelping us with flow through the
system.
We also need to look at whowe're imaging.
So as I've said, overall, moreimaging is better for patients,
but we still are not necessarilyselecting the people who could
(07:56):
benefit most.
And in a resource-constrainedsystem, which we're all in one
way or another, but we have tomake sure that we are, as it
were, getting the biggest bangfor our buck in terms of
imaging.
And we can use AI to do that,but we also have other systems
in place.
We also have a system...
(08:18):
called iRefer, which is now aclinical decision support, which
actually helps people to choosethe right sort of imaging.
So if you're a primary caredoctor, for instance, you might
go onto your computer and think,well, I want to request this
person's got a headache.
I might want to request, andthey've had it for ages, I might
want to request a CT.
So they'll go and request thatCT, but then the system will
(08:41):
say, well, actually, no, an MRIis better for that.
So instead of wasting that CTslot, we've gone straight to the
most appropriate imaging.
And sometimes the clinicaldecision support software will
say, actually, imaging is reallynot going to help here.
It doesn't really move the dial,but it does mean that some of
the studies don't take place.
(09:01):
And it also means that thosestudies that do take place are
ones which are the bestpractice, you know, most likely
to make a difference for thatpatient.
So that's the sort of thing weneed to concentrate on.
And we need to also ensure thatwe've got the data to understand
which tests really are making adifference to patients.
(09:24):
And at the moment, we're notvery good at linking that up.
So we know we've got a lot ofdata around what tests we do,
but we're not terribly good atlinking up which ones really
make a difference to patientswith clinical data, with outcome
data.
So we need to get better atthat.
SPEAKER_01 (09:39):
And what about workforce?
Is there a lack of radiologistsin the system?
SPEAKER_00 (09:44):
Huge lack.
At the moment, there's about a30% shortfall.
And actually, that's due to goup to 40% in the next five
years.
So, I mean...
That's what's preventing us fromkeeping up with the wait.
And it's also terribly stressfulfor people.
No one goes into medicine orradiology to provide a bad
(10:06):
service.
Everybody wants to do the rightthing for patients.
And so people feel a hugepressure.
The staff...
It's, you know, when in COVID wecalled it moral injury, you
know, staff are trying so hardto do their best for patients,
but they're so constrained bythe volume that's coming at them
that they actually can't do it.
(10:26):
And of course, the difficultywith that is that feeling of
being out of control, of notbeing able to do what you always
wanted to do.
drives people away.
And so they retire early, theygo into some other branch or
whatever.
And so it has a sort ofspiraling downward effect.
(10:49):
So people are leaving ourprofession much earlier.
The retirement age, average,well, the average age of leaving
the service is around 54 now,which is so young.
So we have to make things betterfor people.
So as we keep these highlytrained people in the service.
SPEAKER_01 (11:06):
And how do we create better working conditions that
are going to retain thosepeople?
Similarly, where does servicedesign come into this?
SPEAKER_00 (11:13):
Oh, well, this is the big question, isn't it?
And I think there's been a lotof good work on this, but it's
something that we haven't reallymanaged to get to the bottom of
in the NHS, I would say.
And it has to be about lookingafter the people and getting the
best out of the people.
So I think there are...
(11:35):
many ways in which we don'treally make the most of our
staff.
And very good work by MichaelWest about caring for doctors,
caring for patients, a great,great publication there, which
talks about autonomy, the thingsthat people value about their
work, autonomy, belonging, andcontrol.
And of course, a lot of thosethings, control, we've talked
(12:00):
about lack of control, you can'tmanage the workload, Belonging,
we've lost that a little bit.
We're not really creatingcommunities into which people
belong.
And some of the technicaladvances since COVID, where
quite a lot of us are remote,not getting together, doing
stuff on teams, that actuallyundermines belonging even
(12:21):
further.
And the autonomy, people feeltoo much top-down control,
right?
I mean, it's important that weuse data, but sometimes we use
data as a sort of weapon againstour staff.
And we've got to be very, verycareful about that.
So it is about looking after thepeople.
(12:43):
And I believe in the NHS, wedon't really have the structures
in place to do that.
So If you're a clinical leader,you often get very, very little
time to look after your staff.
So you have maybe 12 hours aweek for a department with maybe
70 consultants.
(13:04):
So then you won't really be ableto reach out to your staff.
We don't really have a structureof having one-to-ones with
people.
In any other company, when Icame here to work at the Royal
College, for instance, the CEOwill have one-to-ones each week
with his, you know, the peopleon the sort of tier below and
they will have one-to-ones withtheir team and they will have...
(13:25):
So there is a kind of web, anetwork that spreads through the
organisation.
We don't really have that in theNHS.
And so it's very hard to seehow...
culture and values can reallyeffectively spread down through
the staff.
We've concentrated so hard onkeeping people working, doing
(13:48):
their clinical work.
SPEAKER_01 (13:50):
What's the peer community like in radiology?
SPEAKER_00 (13:54):
Well, it's very strong.
We have some really good thingsin radiology, which have made
our peer community very strong.
So one of those is looking atmistakes or discrepancies, as we
call them.
The thing about radiologists isthat if you miss something on a
scan or misinterpret something,it's there forever for everybody
(14:15):
else to have a look at.
And so you might miss a noduleon an x-ray or interpret it as
benign when in fact it ismalignant, tragically.
And those things will happen.
And then Ever after, it's thereand you can see it and you feel
absolutely terrible about it.
(14:36):
When you go back to look at it,you think, how could I have
misinterpreted that?
How could that happen?
But because of that, we have awhole system in radiology called
radiology events and learning.
And what we do is we, any caseswhere there's a discrepancy of
some sort, we submit thoseanonymously and have a weekly
meeting or monthly meeting andwe go over those.
(14:58):
So we'll all as a team look atthem and say, oh gosh, that was
difficult or, oh yeah, maybe.
But what happens is one, welearn from those mistakes.
So, you know, we're taking thatas a opportunity to learn.
And the other thing is that eventhe most junior person
understands that other peoplemake mistakes too.
(15:20):
And some of those other peopleare really senior and really
experienced.
And as a community, that teachesus about error, about supporting
each other, about using error asa learning opportunity.
And That makes us a very, verystrong community.
And I do think that otherbranches of medicine could learn
(15:44):
a lot from that because it's notonly good for the learning from
your craft, but it's good forthe culture of departments.
SPEAKER_01 (15:52):
Does it feel like your peers have your back when
you're doing this work?
SPEAKER_00 (15:55):
Definitely.
Yeah, definitely.
When within our profession, weall understand, we all recognise
that we don't get things rightall the time.
But I mean, that's true ofmedicine, of all aspects of
health care.
We don't get things right allthe time.
What's important is that welearn from it.
(16:16):
But I think we don't prepare ouryoung health professionals very
well for the fact that they willmake mistakes.
And sometimes those mistakeswill have very bad outcomes for
people.
That is the nature of ourbusiness.
SPEAKER_01 (16:31):
It's impressive the way medically trained people are
taking on so much responsibilityand so much risk with their
work.
And that's just normal in a day
SPEAKER_00 (16:42):
to day working lives.
not let that mistake go towaste.
SPEAKER_01 (17:08):
So I'd like to move on to talk a bit around AI
diagnostics and how we keep ithuman.
There's lots of excitementaround AI as well as complexity.
It's part of the diagnosticinfrastructure and has become
over the last few years.
What do you see as the realpotential here?
SPEAKER_00 (17:23):
Well, it's very exciting.
That's the first thing.
And very welcome because we'vetalked about the fact that we
have capacity problems.
And so we need all the tools wecan get to improve, you know, to
maximize the time that staff canspend doing the things they've
been trained to do.
AI should help our productivityor efficiency, whatever you want
(17:45):
to call it.
At the moment, we've beenconcentrating.
I mean, about 80% of all thealgorithms available are in
radiology.
So it's really in ourwheelhouse.
At the moment, a lot of the techis concentrating on helping us
with accuracy, which is alwayswelcome.
But it's not our biggestproblem.
Our biggest problem is capacity.
(18:06):
So we might need to sort of moveour attention a little bit from
things that help us withdiagnosis to things that help us
select the right patients forimaging and also smooth the
patient's pathway through theirjourney through the imaging
department and also just make usa bit more efficient so there
(18:29):
are things for instance likereporting solutions which can
quickly record so you can justtalk to a computer and it can
reorder your stream ofconsciousness into a structured
report and also makerecommendations.
So you might see some sort of athing on a renal cyst, for
(18:51):
instance, a fluid filled thingon a kidney.
Now there's guidance about whichcysts should be followed up and
which you don't need to worryabout.
And I don't have that in myhead.
So I would have to look that upon a computer.
I think, well, I know there'sguidance, but I'm going to have
to look this one up.
But, you know, the AI can youthat as you're doing the report.
So it just makes you muchquicker and more efficient.
(19:13):
So we need to concentrate on thesort of admin side of things to
help us do our thing morequickly.
In terms of the diagnostic AI,we've still got quite a long way
to go, I think, in deciding howto use that best.
We have systems, we'redeveloping systems to look at
(19:35):
the accuracy of the algorithmitself, the accuracy of the
output of the computer.
But actually, what reallymatters is what is the
combination of the computer andthe human?
Is that better?
Is it worse?
Sometimes it can be worse.
Do patients who have had thattechnology used, do they do
(19:55):
better?
And those questions we stillactually don't really know the
answers to.
So there's a lot of work to bedone.
SPEAKER_01 (20:04):
And where does the responsibility then lie?
Because, you know, theconversation we were talking
earlier about that risk thatyou're taking on and trying to
make decisions and getting itwrong at certain times.
Does that then mean that ifyou've done the scan with an AI
device, And you've had the samedecision as the AI.
It's almost abdicatingresponsibility to the
technology.
Or does that responsibilityalways lie with the radiologist?
SPEAKER_00 (20:26):
Well, it always lies with the radiologist.
And so one really important partis to educate people about what
the technology can do and whatit can't do.
Because the technology, likehumans, is good at some things
and not good at others.
So the technology does missthings.
For instance, I heard of a casewhere it can actually blind you
if it's telling you to looksomewhere You concentrate on
(20:49):
that and you don't look at quiteobvious things that you might
have seen somewhere else.
So I heard of a case the otherday where the AI would only pick
up lumps of a certain size orbelow a certain size.
And actually that meant that theradiologist missed a much bigger
lesion Because they were sort ofdistracted by the AI and that
(21:11):
can happen.
So we need to, first of all, toeducate people what the AI can
do and what it can't and whatyou still need to look for.
And then we need to monitor itreally carefully and we need to
describe what those monitoringsystems are.
But at the moment, it's alwaysthe radiologist's responsibility
to There are some algorithms outthere now that say that they can
(21:35):
manage without a radiologist.
For instance, there are somethat say that they can
effectively identify normalchest X-rays or a proportion of
normal chest X-rays, whichwouldn't need to be reported at
all.
So if that were to happen,there'd need to be some changes
in the law.
We'd need to describe verycarefully what sort of quality
(21:56):
assurance needs to be in place.
to protect patients whose scanshave gone through that.
And the responsibility forsetting that system in place
would be with the hospital andthe radiologist.
SPEAKER_01 (22:10):
You hear the phrase human in the loop a lot when
we're talking about medicine andtechnology.
What's the implication of thatin this area?
SPEAKER_00 (22:17):
So at the moment there always has to be a human
in the loop.
I think we need to think aboutthat moving forward, whether
that is always the right way.
As I've said, sometimes thecombination of the human and the
computer can mean that it can beworse, actually, than either one
alone.
So we need to understand that.
(22:39):
And I think we also need todescribe the systems that need
to be in place for some thingsthat could be done without a
human in the loop.
What our polling showed aboutpublic attitudes towards AI was
that people are happy for AI andtechnology to be used, but they
(23:02):
do trust doctors and expectdoctors to implement that
safely.
So I think that is where ourresponsibility lies for making
sure that these are happeningsafely.
And I think if patients gettheir results quicker, they're
really happy.
SPEAKER_01 (23:17):
I mean, there's a balance, isn't there, between
speed, safety and empathy?
Totally.
How do you get that balanceright?
SPEAKER_00 (23:24):
Well, we've got a lot of work to do, I think is
the answer.
But that is entirely what...
doctors and particularlyradiologists have been doing for
many years.
So that's absolutely 100% wherewe can contribute.
And we've got lots and lots ofexperts who can work on that.
And it's very exciting work.
And it promises to be able torelieve some of the capacity
(23:47):
issues that we've got and helpus to deliver a better service.
So it is a lot of work, butwe're the right people to do it.
And we're very happy to takethat on.
SPEAKER_01 (23:58):
We're talking a lot about kind of technical
innovation in the space.
I wondered what your thoughtsare around innovation around
service experience and, youknow, what it's like to go
through radiology.
SPEAKER_00 (24:10):
Well, yes.
And I think that's a veryinteresting question.
And I think one of the thingsthat we're very poor at in
radiology is communicating withpatients.
Often Patients don't reallyunderstand where or when they're
going to get the results anddon't even understand that once
(24:31):
you've had your scan, there'sanother step.
You know, the other step ishaving that scan reported and
analyzed by a radiologist.
So patients could really do witha bit more information.
So as they had a bit morecontrol over what's happening, I
think had previously had a roleas the GERF lead, the Getting It
(24:51):
Right First Time lead forradiology, where I went and
visited 143 radiologydepartments in England.
It was an NHSE project.
I really saw people doing greatthings in this field, people who
had patient navigators.
So people employed withinradiology to work with patients
and say, yeah, this is what'sgoing to happen now.
(25:13):
And this is where you're goingto get your results.
And Sometimes concentrating onjust getting more examinations
done and getting more reportedmeans that we don't concentrate
on patients as we should.
Some people are doing good work,but it's sporadic.
SPEAKER_01 (25:29):
And how much work goes into the communication
following the diagnostic.
So I was talking to JonathanGregory in the previous podcast
who was talking about workthey've been doing with
Macmillan around end oftreatment communication and
actually simplifying the lettersand using not medical terms but
terms that your average personcan understand in the letters.
Is that normal practice inradiology?
SPEAKER_00 (25:50):
Well no it's not and actually you know most people in
radiology get their result fromanother specialty.
So the way it happens is I willwrite a report and that report
will go to either the GP or thepediatrician or the pediatric
surgeon or whoever.
And that person will be theperson who communicates the
results to the patient.
(26:11):
So my reports are not writtenfor patients.
They're written for otherclinicians.
But now, a lot of reports arenow visible on the NHS app.
Which is the right thing, butthey're not written in a way
that is necessarily accessibleto all patients.
(26:33):
And they can be alarming.
You know, you might say there'sa possibility of cancer because
there is a possibility, but it'snot very likely.
And I mean, it depends who youare, but some people find this
very stressful reading.
So this is an issue.
SPEAKER_01 (26:48):
But if you've written that letter for another
medical professional, it's goingto be a very different letter
from the letter you would writeif it was going to be viewed by
a patient.
SPEAKER_00 (26:56):
Absolutely.
So we've had these conversationswithin radiology.
So should we change the way wereport?
But actually, that's probablynot the right thing to do.
The right thing to do is to putin some other step which helps
patients understand or enablesthem to access health.
(27:16):
So either you could usetechnology.
And in the States, they do thisa lot so that all patients have
access to their results.
So the things that they do inAmerica now are they have little
videos and things embedded intheir reports.
So you might say you have a tearof your medial meniscus in your
knee, and then there'll be asort of link onto that of this
(27:38):
is what a normal meniscus lookslike, and this is what yours
looks like.
And that's sort of technologythat does that.
And then you have a link to saywhat normally happens.
And so They also are employingpeople in radiology departments.
So if a patient's worried abouttheir report, they can then ring
up and discuss it with aradiologist.
So they have a whole system inplace to look after it.
(28:01):
And there are no shortcuts forthis.
If we're going to do this, weneed to do it properly because
we do need to look afterpatients properly.
SPEAKER_01 (28:09):
And that all comes down to funding, I assume.
SPEAKER_00 (28:10):
Absolutely.
And recognising that it's anissue.
SPEAKER_01 (28:14):
So just going back to the getting it right first
time work that you were leading,which was a massive effort to
understand and improve practiceacross the country.
What were the biggest challengesand lessons from leading that
redesign at scale?
SPEAKER_00 (28:26):
Oh, well, it was the most amazing experience, the
most amazing experience.
One, Tim Briggs, who started thewhole program.
I mean, it was visionary andamazing.
He has worked tirelessly to getthis program off the ground, not
only in orthopedic surgery, butin everything.
He's done really great thingsfor medicine in the UK.
So with that program, andactually that was real
(28:50):
leadership.
He just did it and took otherpeople with him.
He saw there was a problem andhe did it.
So that's one lesson.
If you think there's a problem,no matter who you are, just go
for it.
And I learned that so many timesthrough that trip.
So during the process, you takedata, whatever data you have,
and you go and discuss that withthe different departments.
(29:12):
It works best when you can geteverybody in the department into
that.
So it'd be like the chiefexecutive or the medical
director or both, and then allthe helpers and the porters and
the PACS people.
And then you can start aconversation and actually you
can just unlock things in theroom.
The chief executive would say,well, why haven't you got
anywhere to do that?
(29:33):
And they say, well, nobody'sfound me.
And they say, well, we'll getthat now.
You know, so that's amazing.
And it was so I met so manypeople who were just working
really hard in really difficultcircumstances to do the right
thing for their localpopulation.
Inspirational, totally.
SPEAKER_01 (29:52):
And how important was the collaboration in there?
SPEAKER_00 (29:54):
Well, the collaboration goes all the way
through.
I mean, one with your otherclinical leads, the other
clinical leads from the otherspecialties were incredible.
And it was great to talk tothem.
Two, the collaboration and theteam are quite small.
I got great advice from one ofthe past presidents of the Royal
College of Radiologists, who ismy co-lead, Giles Maskell, and
(30:17):
had an absolutely wonderfulproject manager, Gail
Roadknight, who ran the wholething, Elaine And Quick was a
radiographer who came with usand a service manager and Lucy
Proling, all of these people.
That was a great team.
And then when you get there,often the helper will never have
been in the same room as themedical director or the CEO.
(30:40):
And so to get people in the sameroom and have that conversation,
that's the thing that is magic.
SPEAKER_01 (30:47):
How did you do that in a meaningful way?
People are so busy in thesystem.
SPEAKER_00 (30:51):
Well, our project manager did a lot of work on
that.
And it depends on the leadershipin the trust as well.
Those really good trusts willsee this as an opportunity for
them to get to know more aboutwhat is really happening on the
ground, led by someone else.
And it's free.
You know, really good leadersembrace this process.
(31:12):
Those ones that are not led sowell, it's really hard.
And we've flogged off to variousplaces and there'd be like two
people there.
And so, you know, and you walkinto a department immediately,
immediately you can tell aboutthe leadership.
A hundred percent.
Leadership is so important.
I mean, everybody says this, butactually you can tell the minute
(31:34):
you go into a department if it'sgot a strong leader.
SPEAKER_01 (31:38):
I'm a big advocate for co-design when you're
thinking about change andbringing people together to
think about how those changescan be made in a way that works
for them.
It sounds like this was aco-design approach you were
taking with these teams.
Can you tell me a bit more abouthow that made the changes stick
or examples of how it worked?
SPEAKER_00 (31:55):
Absolutely.
I mean, you know, the innovationthat goes on is unbelievable.
I think people very rarely getthe chance to be together to
talk about the innovations thatthey've done.
I mean, one brilliant innovationwas around paediatric MRI
scanning, so scanning toddlers.
(32:16):
You know, for an MRI, you needto keep still for quite a long
time.
And of course, toddlers can't dothat.
But most of the time, we need ananaesthetist and, you know, a
whole load of equipment that youcan take into the MRI scanner
and things like that.
So it's a very expensive,labour-intensive process that
can't be done in a districtgeneral hospital.
It has to be done in a bighospital.
(32:37):
In one place I went to, thehelpers there had said, toddlers
sleep really heavily.
Why don't we do them at bedtime?
So they got the toddlers intothe department around about
seven o'clock and they bringthem in.
They had a travel cot.
They bring their kind of blankieand their bottle or whatever and
play with them for a littlewhile, put them to bed in a cot.
(33:00):
And then they'd fall asleep.
Then they'd put some dear littleheadphones on, lift them up, put
them in the MRI scanner, do a 45minute MRI scan.
no medication, no nothing, thenput them in their car seat and
take them home.
SPEAKER_01 (33:13):
So no anaesthetic?
SPEAKER_00 (33:14):
No anaesthetic.
SPEAKER_01 (33:16):
And they would lie still enough for that?
SPEAKER_00 (33:17):
Yeah, they're fast asleep.
And, you know, this, I've been apaediatric radiologist for 20
years by the time, I'd neverthought of that, you know, and
this was, and in fact, it's goneon and it's other people have
had the idea as well around thecountry, but now this is now
much more standard practice.
And this was a helper.
This was a very low band personsaying, your toddler sleep
(33:38):
really heavily normally.
It's just, and that sort ofthing is just light bulb.
SPEAKER_01 (33:46):
It sounds really obvious once you've talked about
it.
Of course it makes sense.
SPEAKER_00 (33:50):
I could not believe that I'd never thought of that.
And I was just, oh, oh yeah, ofcourse.
SPEAKER_01 (33:56):
Easier for you and easier for the toddlers as well.
SPEAKER_00 (33:58):
Better for them.
I mean, general anesthetics, youdon't want to give little
children general anesthetics.
We don't think it's much goodfor them.
SPEAKER_01 (34:04):
So the changes that came out of this process that
you were leading, were theytop-down changes or was it more
kind of participatory and peoplewere making the changes
themselves?
SPEAKER_00 (34:13):
It's more participatory then.
And I think that's the thingabout health service
improvement.
I mean, it's a complex system.
So you just edge things forwarda little bit each time.
But what it did was, first ofall, it showcased what people
were doing in radiology.
So in previous times, radiologyhas often been regarded as a
problem, you know, a blocker.
(34:35):
We can't get things throughradiology.
And when the more seniormanagers in the hospital had it
showcased to them all the thingsthat radiology All the different
departments in the hospital orprimary care, the many different
sorts of things they did.
And they would suddenly think,wow, that's something actually,
that is something to be dealingwith all of that.
(34:58):
And just as you say, co-design,just discussing some of the big
issues they had.
Sometimes, one department Iremember, they couldn't get as
many people through the CT scanbecause when they needed to
cannulate the patients, theydidn't have a separate room to
do it.
They had to do it on thescanner.
So that was wasting scan time.
And they said, well, we'veasked, but nobody's given us
(35:19):
one.
The chief executive was justable to say, oh, right, we're
doing that now.
And so getting people togetherand making time to do that is
absolutely crucial.
We are squandering the expertisewe have in the health service
because we are not reallylistening to the things that
people know could be improved.
SPEAKER_01 (35:45):
I'd like to move on and...
speak about leadership a bitwith you now.
So you've written about theurgent need to value leadership
in medicine more.
What does great leadership looklike to you in the system today?
SPEAKER_00 (35:58):
Well, I mean, there is something about authenticity,
about people, you know, it'sactually not about necessarily
learning or it's just aboutbeing yourself and trying to
make things better and otherpeople seeing that you are
trying to do the right thing.
So I think you just have to dowhat you do.
(36:22):
So I don't think there'snecessarily a right way or a
wrong way.
I do think we really hamperpeople.
When we talk about leadership inthe NHS, we tend to take people
away, give them a learningcourse, do their Myers-Briggs
score or whatever.
But actually, we make it reallydifficult for people because we
give them no time.
We give them no support.
(36:42):
So actually, as a clinicaldirector in a radiology
department, what you need isdata.
You need to be able to reallysee what's happening.
And you need a person whounderstands healthcare data to
help you.
You need good secretarialsupport.
You need time to work with yourstaff and look at ideas and just
hear how they are.
(37:02):
And If you don't have that, nomatter how much training you
have, you won't be able to besuccessful.
It's a bit like sort of victimblaming in a way in that people
are really trying hard and we'resaying, oh, you need more
training.
You know, it's because you don'tknow how to do it.
It's not because they don't knowhow to do it.
It's because we're making itimpossible.
SPEAKER_01 (37:25):
How do you personally lead when time and
energy are in short supply?
UNKNOWN (37:29):
Yeah.
SPEAKER_00 (37:29):
Well, I honestly, I don't really think of myself as
a leader.
I just rely on other people.
I see other people who I get alot of help.
I mean, in this job, I get lotsand lots of help, which is
great.
So one of the things is when youget more senior, it all becomes
more easy because you get a lotof help.
(37:51):
And also people listen to whatyou say, which is something that
You know, for most people, thatjust doesn't happen.
So it gets easier.
And also, I surround myself.
I'm lucky to be surrounded withpeople who've got a lot of good
ideas.
SPEAKER_01 (38:04):
You've highlighted the lack of time, support and
recognition for clinical leaderspreviously.
What's the impact of that?
SPEAKER_00 (38:12):
The impact is very bad on the person who's trying
to do the leading.
So we don't have many peoplestepping up for medical
leadership.
It's seen as impossible.
Sometimes people think, oh,you're going over to the dark
side, you know, because medicalleaders don't have enough time
to develop their staff and towork with their staff.
(38:34):
It can be seen as quiteconfrontational.
So people who step up who aretrying to do the right thing, it
takes a huge toll on them.
And it means that people arescared to step up.
And it means we are not makingthe most of the talent that we
have.
So we've got to really changethat because nothing's going to
change unless we can persuademore people to do that.
SPEAKER_01 (38:58):
And what are the things that you need to change?
SPEAKER_00 (39:00):
So I think we need to support the people more who
are doing those leadership jobs.
Give them time.
Give them resource.
You know, to economize onadministrative staff and have
the people who are the clinicaldirectors doing all the admin
work, that's ridiculous.
The number of doctors who arewriting rotas, I mean, that's
(39:22):
ludicrous.
We need to give people supportand time.
And Often we use our data tosay, well, if everybody just
works a bit harder, we can do abit more.
And that's just not the way todo it.
We need more fundamental change.
So we need to give people thetime to innovate, to make more
time.
But that's quite a brave step.
SPEAKER_01 (39:42):
Mentorship is important.
And I think it's played a rolein your own career.
Can you tell me a bit more aboutthat?
SPEAKER_00 (39:48):
A hundred percent.
I have had some wonderful peoplewho have...
faith in me and supported me andtaught me so much.
When I started in medicine, itwas a much more hostile
environment for women,particularly.
I still do think we have anissue with women's voice not
(40:09):
being heard.
We have a lot of data aroundthat.
I've had a lot of very strongmale mentors who have helped me
and supported me throughout.
Tim Briggs, I've mentioned, myhusband, Giles Maskell, who I
mentioned in Girfed.
A lot of people who've reallysupported me.
and taught me to just have faithin myself.
(40:30):
And I would like to try and passsome of that on to other people.
SPEAKER_01 (40:34):
What advice would you give someone who's thinking
about maybe stepping into aleadership role in the health
system?
SPEAKER_00 (40:41):
I'd say do it.
Do it.
There are huge rewards.
That's the way to make adifference.
Get yourself some good support.
Don't blame yourself too much.
Learn from mistakes, but don'twallow in them.
SPEAKER_01 (40:56):
Before we wrap up, I'd like to look ahead to the
future a little.
What are you most optimisticabout right now?
SPEAKER_00 (41:03):
Well, one of the things I get in this job is I
get to look around other healthservices.
We're very fond of tellingourselves how everything's
terrible and there's definitelya lot of room for improvement.
But actually, healthcare in theUK is not as bad as we think.
There are a lot of places in theworld where if you're poor, you
get no access to healthcare.
And that's not the way it is inthe UK.
(41:25):
I mean, there are tremendoushealth inequalities, we could
definitely improve, but at leastwe have healthcare that is free
for people.
So that's good.
We have a hugely wonderful staffand we have great education,
great systems in place.
We're very lucky in that regard.
You know, the Royal Collegesreally support education and
(41:49):
standards.
So our healthcare system isactually not too too bad.
And we've got the buildingblocks to do a great deal more.
SPEAKER_01 (42:00):
How do you see radiology evolving over the next
five, 10 years?
SPEAKER_00 (42:04):
It's going to change, no question.
We can't continue to doeverything ourselves.
Technology is really going tohave a big effect.
We need to lead that change andmake sure it's having a good
effect and not just makingthings more complicated.
We will still be needed.
When you're dealing with humansand You need another human to
(42:24):
understand the complexity.
We'll be doing more clinicalliaison with other groups, but
patients are getting morecomplicated.
People have the multimorbidity.
People have more than onecondition.
More people have more than onecondition.
So you need to understand theinterplay as well as with social
factors and human factors.
(42:45):
And that's where radiologistswill really have a lot to offer.
SPEAKER_01 (42:49):
If you could wave a magic wand and change one thing
about the system now, what wouldyou change?
I
SPEAKER_00 (42:55):
think I'd change the sort of structure.
I'd put that one-to-one systemin.
So I'd make sure that everybodyin the health service had a
regular one-to-one with theirline manager, which was not
necessarily sort of judgmental,but it was kind of developmental
with their manager.
So as we have that web goingthrough the health service of
(43:18):
good ideas being able to come upfrom the bottom and culture
being spread down from the top.
I think it's about making surewe have time to develop our
staff and every staff membershould have regular time with
their manager.
SPEAKER_01 (43:32):
What's next for you, Kaf?
SPEAKER_00 (43:33):
Well, I'm not quite sure.
I mean, it's been an amazingexperience being here at the
college.
I've loved it.
Some wonderful people here.
And I would quite like to keeptrying to do some service
improvement work.
So anybody out there want to...
I'm available.
SPEAKER_01 (43:52):
Amazing.
Kath, thank you so much fortaking the time to talk to me
today.
SPEAKER_00 (43:56):
Thank you.
It's been a real pleasure, Sam.
One of the perks of the job.
UNKNOWN (44:01):
Thank you.
SPEAKER_01 (44:09):
Problems Worth Solving is brought to you by
Healthier, the collaborativeservice design consultancy for
transformation in health careand public services.
Find out more about how we canhelp you deliver user-centered
change at healthier.services.
Popular Podcasts
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.
The Joe Rogan Experience
The official podcast of comedian Joe Rogan.
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Special Summer Offer: Exclusively on Apple Podcasts, try our Dateline Premium subscription completely free for one month! With Dateline Premium, you get every episode ad-free plus exclusive bonus content.