clinical practice
The GP thought: “Everything seems normal, you seem fit, but I’m
just going to check your tummy.”
He thought he felt an aneurysm there so he sent him off for
an ultrasound. There was no aneurysm there, but it looked like
there was something funny with the pancreas, so he sent him off
for a CT scan of the pancreas. That was normal, but they spotted
something on the liver. So they said: “Oh, we have to biopsy that
because it looks like it might be a tumour,”
So they biopsied it. It turned out to be a congenital
hemangioma, which is a vascular abnormality of the liver. It was
harmless, except if you put a needle in it. He bled severely and he
could have died. He spent several weeks in hospital as a result of
this, all from a series of tests that were to do with nothing.
Now that’s an extreme example but that sort of thing is
happening all the time with those thyroid incidentalomas I was
talking about. People are now having thyroid surgery and getting
the complications from it. You can get local infections afterwards.
You can also get a nerve problem that causes you to lose your
voice from it. That’s probably the most serious one. If we keep
doing more of these, it can get worse.
To give you an extreme case from South Korea, there has been an
almost 20-fold increase in the number of thyroid cancers recorded
there because they’re doing so much use of ultrasound to look. The
three-fold increase in Australia is small compared with that, but if we
kept going the way of South Korea, we could end up there as well.
This is occurring in the whole body, with all sorts of things and
with all sorts of tests that we do. The tests are good, but we have
to learn to use them more wisely in order to avoid all of these
downsides. It’s not just a cost problem. It’s this follow-on problem
of the sequence of testing and complications and labelling that
occurs with people we’re concerned about.
This is a five-year project starting in January 2017. Over those
five years, what would success look like?
There are two things we need to do. One is that we need a better
understanding of this problem. This has been an under-recognised
and under-researched problem. The second is we then need ways
of intervening with that problem to reduce the dangers of the
tests, while keeping the advantages of the tests, so there’s this fine
balance we need to work out.
For example, we don’t know the exact pathways by which those
excessive thyroid nodules, I’ll call them, are being detected. So
we need to check what the medical records are, talk to people
who’ve had the cases, etc, to find out what was the pathway, how
this happened and what they were being investigated for, to have a
better understanding of that. Based on anecdotes, we think much
of it’s due to the ultrasounds being done on the neck arteries, and
then the ultrasonographers just having a look. But we don’t know
all the pathways and how common they are – we need absolute
understanding.
Then, we want to do a whole lot of intervention studies. For
example, one of the things we want to try for imaging of the lower
back to analyse back pain is to use what’s called delayed imaging.
We know that delayed prescribing is effective for the doubtful
cases of antibiotic usage, but nobody’s tried it for imaging.
[With] imaging of the back early on in acute back pain, unless
there’s what’s called a red flag – something that tells you there
might be something seriously wrong like a tumour or an infection
or something – we’ve got very, very good data to say that imaging
does not help and that it can lead to harm.
Despite that, lots of people get x-rays, MRIs and CT scans done
early on [when experiencing] back pain. Partly, that’s due to the
expectations of patients who feel like they want to know what’s
going on; that expectation comes through to the doctors who
order it as well. By using delayed imaging, we say [in effect] most
people will be getting better in a couple of weeks, but if you’re not,
then you can go and get this imaging.
We expect that will roughly halve the rate of imaging and,
therefore, stop a lot of the incidentalomas being detected.
Further down the track, we want to use shared decision-making
for patients. When results come back, there’s a dilemma about
whether you should treat for some cases. Patients and doctors
have quite false expectations about what the benefits of some
treatments are. If we inform particularly the patients about the
benefits and harms of some of the treatments, we think we can
better guide the patients who will benefit to get treatment and
those who won’t benefit so much not to get the treatments.
what are the risks of taking no action with regard to the spread
of incidentalomas?
We think this is addressing probably the most important problem in
the future of medicine. Because our technology keeps improving
all the time, but there’s always a downside to that technology and
we’ve got to work out how to optimise the benefits and harms from
it. Otherwise, we think this is a genuine threat to the sustainability of
medicine. Most of the increases occurring in the cost of medicine
and the harms of medicine are due to us not finding that correct
balance at the moment. This is vital work to be done now, so we’re
looking forward to getting on with it. n
Lecturer/Senior
Lecturer/Associate
Professor
Applications are invited f ܈HX