0:00

Well, welcome everybody.

0:02

We have heard Michael Bruno's talk on the sources of errors

0:07

for radiologists and physicians in general,

0:09

and now we will talk a little bit about what we can do

0:13

to reduce the likelihood that we will commit those errors.

0:17

And I've put it in a sort of a fun top 10 list,

0:21

and hopefully that will be useful to you.

0:25

So the top 10 hints that I'm going

0:27

to give are read fewer cases, strategize against misses.

0:31

Do not miss big ones. Do not read when tired.

0:34

Read prior studies and reports. Do not trust others.

0:39

Be curious. Get smarter peer review

0:43

and know the standards.

0:44

So let's start going through these.

0:47

So the first thing that we know is that

0:51

it's basically a numbers game.

0:54

The more cases that you read, obviously the more likely

0:58

that one of them will have an error in them.

1:01

So it really is the truth that

1:04

as one gets more senior

1:06

and has more years of experience in reading radiology cases,

1:11

you just accumulate so many cases that

1:15

the numbers catch up to you.

1:17

So how can we read fewer cases?

1:20

Well, this is sort of one of the arguments

1:22

for being in academia

1:24

or being, having a role in administration

1:27

or having a, uh, lecture teaching career in

1:32

that you have days off of the schedule

1:35

where you're not reading the cases.

1:37

The the people who are reading six days a week

1:39

with a large number of cases, you know, it catches up.

1:43

Whereas those who are only reading three days a week,

1:46

basically there's 50% less likelihood that there will be a

1:50

air, an error that occurs.

1:53

So, um, the other thing is, you know, start late in life

1:58

and retire early, make, make a good a good amount

2:00

of money in a short amount of time.

2:02

Um, over fewer cases.

2:05

There are some modalities where you read less overall cases.

2:09

The people who have to read a large number of plain films,

2:12

particularly for something like lung cancer screening, um,

2:16

again, the numbers add up over time.

2:19

Whereas someone who is doing, for example, uh,

2:22

20 PET scans a day, uh, has an advantage, if you will,

2:26

over the person who is reading 120 plane films over the day

2:30

just because of numbers.

2:32

And similarly, the interventional radiologists have

2:34

that advantage as well.

2:36

So I try to avoid playing films

2:38

and screening mammograms if you can.

2:40

I'm joking. Of course we have to do that

2:43

and it's a service to our patients,

2:45

but this is the, the nature of the beast.

2:48

I will give a personal example.

2:50

Here I am with, uh, when I, um, am reading my cases.

2:55

So I read about 60 to 70 studies per day, and

2:58

I do that four days a week.

3:01

And let's just say I'm darn good. I'm 99.9% accurate.

3:06

Well, with simple math, you realize that if I'm reading 60

3:10

to 70 cases for four days a week, I'm reading about 240

3:14

to 280 cases a a week.

3:17

And I'm doing that four weeks in a month.

3:19

So around a thousand cases each month.

3:23

And if I'm 99.9% accurate, that means

3:27

one big miss per month.

3:30

And I've been reading for 35 years.

3:32

So over the course of my lifetime, it's not that

3:37

surprising that I might have 420

3:41

major misses so far in my career that are just waiting

3:44

to catch up to me.

3:46

But this is the nature of it that, that

3:48

because we do high volume, um,

3:50

even if we're 99.9% accurate,

3:53

even if we're 99.99% accurate,

3:58

it would equate to 42 major misses in my career thus far.

4:03

So the more cases you read, the more chance for mistakes

4:06

and the more cases that you read per shift,

4:09

you know you're going to make more mistakes

4:12

at the end of the shift.

4:14

When you tire, you make more mistakes.

4:17

The more shifts that you're on per week,

4:19

the more likely you are to have, um,

4:22

a medical malpractice suit the longer the shifts.

4:25

So shifts happen, you know, and this is, um, Dr.

4:28

Krupinski who is at Emory, actually a PhD radiology faculty,

4:33

found a nearly 5% drop in diagnostic accuracy, um,

4:38

at the end of a, an eight hour shift.

4:41

And she looked at eye tracking data as well

4:45

as ophthalmologic exam

4:46

and said, noted that there is in induced myopia

4:50

and opia, which is another fancy name for eye fatigue.

4:55

And she equated that to being the cause

4:59

for why more errors are made at the end

5:01

of a shift than at the beginning of the ship.

5:04

So you just have to accept that it's going to happen.

5:06

The more cases you read, the more likely you are

5:08

that you are going to eventually make an error

5:11

that shows up in a malpractice case.

5:15

So know your stopping point.

5:16

Obviously once you see that you're getting fatigued,

5:19

it's important to make that stopping point and take a break.

5:25

When you look at, uh, articles on radiologists

5:28

and errors, one of the things that is interesting is

5:31

that the radiologist age is equated

5:35

with the higher risk of making mistakes.

5:39

You think that maybe the new people,

5:40

the newcomers are making more mistakes,

5:43

they make more mistakes in interpretation error

5:46

because of lack of fund of knowledge,

5:48

whereas the old professor tends to have more detection risk.

5:53

So the relative risk is about 1.235 for each

5:57

Decade to make an major error.

6:00

And, you know, I'm entering my fourth decade of,

6:04

or fifth decade of work here.

6:06

Um, you can see that those numbers eventually catch up.

6:10

The minor error is relatively smaller.

6:12

Radiologists age was positively associated with the odds

6:15

of making a major error as opposed to a minor error,

6:19

which occurs relatively equally between young,

6:22

young radiologists and, and senior radiologists.

6:25

So my risk factor is about 35.7 times higher

6:29

for a major error accumulated over the course of my career.

6:37

So how would we safely

6:41

dictate a scan that might mitigate our potential

6:45

for making an error or being sued?

6:48

So here you go. The, there appears

6:50

to be a possible borderline in indeterminate, equivocal,

6:54

suspected pixel probably of questionable significance,

6:58

clinical correlation needed maybe.

7:00

So a lot of people as they get more tired and

7:05

or more wary of medical-legal cases in reading this, uh,

7:09

CAT scan, um, do a lot more hedging

7:13

and more recommendations of additional studies.