Surgeon Risk of Mortality Is Highest Among All Physician Types

In this episode, host Rick Greene, MD, FACS, talks with Vishal R. Patel, MD, MPH, from Brigham and Women’s Hospital in Boston, about a recent article in JAMA Surgery on Mortality Among Surgeons in the United States. Dr. Patel and colleagues found that, among all physician types, risk of mortality is highest among surgeons. Dr. Greene probes for the common causes of death among surgeons and what mitigation strategies can be put in place to change this somewhat surprising statistics. 

Talk about the podcast on social media using the hashtag #SurgicalReadings

Episode Transcript

(00:06):
Welcome to Surgical Readings,
a podcast brought to youby the American College of Surgeons.
I'm your host, Dr. Rick Greene.
And in this series,we will talk to the authors and experts
featured in relevant and practicechanging articles published in the world's
most prominent medical journals.
As busy professionals,we don't always have time to read the most

(00:27):
current studies.
The goal of this podcastis to bring that information to you
by providing keytakeaways, insights, and perspectives
from leading authoritiesin all surgical specialties
and multidisciplinary areasthat affect the surgical patient.
The opinions expressed in this podcastare those of the participants

(00:49):
and not necessarily those ofthe American College of Surgeons.
Welcome to Surgical Readings.
I'm Dr. Rick Greene,your host for the series,
and it's my absolute pleasure
to welcome as my guest today, Dr. Vishal Patel, who's in the Department

(01:10):
of Surgery at the Brighamand Women's Hospital in Boston.
And he has writtenand his colleagues have helped him
write an absolute great pieceon the mortality
among surgeons in the United States,which appeared in JAMA surgery.
Vishal, welcome.
Hey, Dr. Greene.Thank you so much for having me.
Well, I want to congratulate you onnot only a great piece

(01:32):
in JAMA surgery, but also somethingthat, obviously we've needed
and, I think will be a benchmark,for other publications in this area.
Obviously, we're talking a lot recentlyabout,
health, physicians, burnout,things like that.
And I wonder, if you could just discussbriefly what were your reasons

(01:54):
for undertaking this study?
So I think this the study was motivatedlargely
by a gap that we identifiedin some of our prior work.
Earlier this year,we actually showed that,
using the same dataset that physicians overall,
have lower rates of mortality

(02:15):
than other highly educated professionals.
However, we found several qualifications
to that finding largelyrelated to sex and race.
And what we did not knowwas whether that advantage
extended uniformly acrossmedical specialties.

(02:37):
And surgeons are,
you know, as we know, exposed to distinct
perceptual environmentsafter long work hours.
High sort of physical demands,at times overnight call,
and even occupational hazardslike radiation exposure.
And so in our heads, it was plausiblethat maybe some of these factors

(03:00):
could affect long term health and resultin different mortality patterns.
However, and not really been studiedbefore, and with the recent availability
of this populationlevel occupational mortality data,
on death certificates,
we were able to finally answer this question.
So of course,methodology is so important in this.

(03:22):
I wonder if you could go through
your methodologyand how you, reached your conclusions.
Sure.
So we conducted a populationbased cross-sectional study
using the National VitalsStatistics System.
Specifically from 2023,which is the most recently available year.

(03:44):
This database is,
produced by the CDC.
And what we did is we took this
death certificate data for adults.
Aged 25 to 74 years.
And calculated mortality rates.
The database has newly available

(04:08):
occupational data,
which is sort of the key methodological,
methodological sort of advancethat enabled this analysis.
And it - that variable,
reflects the job in which the decedentspent most of their working lives

(04:29):
and is reported by the informant,which is typically like a family member.
So we took all of the mortality data
from the National Vital Statistics System.
And we used population denominators
from the American Community Survey.
And we calculated rates.

(04:51):
So the number of deaths divided bythe number of people in the denominators.
And we age and sex adjusted these rates.
And then we compared surgeonswith the non-surgeon physicians as well
as with lawyers, engineers and scientists,which is a group that might have

(05:11):
a similar education and income.
And to the general population.
I was just wondering, you know, since
surgeons and maybe physicians in generalare retiring at an earlier age,
is there any way to drill downon the actual number of years
that the individual has spentbeing an active surgeon?

(05:33):
Is that possible at allfrom the data that you looked at?
That's a good point.
Unfortunately, the data only reports the,
the occupation in which the decedent
spent most of their life working. But,
but you're right.
That is that is one sort of factor
that might be different between surgeonsand even other physicians.

(05:57):
The length of time that is sort of worked.
Right.
So I was wondering, now,since you've gone through that,
what did you find as the common causesof mortality among the surgical group?
So, as with most otherprofessional groups,
cancer and heart disease
were the two leading causes of death, in order.

(06:21):
among surgeons.
However,there were several causes of death
that ranked higher among surgeonsthan in the comparison groups.
Motor vehicle crash collisions
were a notably prominent cause, which,
in terms of rank order,ranked fourth among surgeons.

(06:42):
And we also observe higher rankings
for deaths due to hypertensionand assault.
In terms of the actual mortalityrate, surgeons at higher cancer
specific mortality rates than non
surgeon physicians, as well.
The findings related tothe motor vehicle crashes,
were sort of consistentwith prior evidence, which linked -

(07:06):
had linked work hours and fatigue to hazardousdriving events among surgeons.
However, there are several causeslike cancer specific mortality,
which we thoughtwere relatively new and undiscovered.
So I'm wondering again,given the data that you looked at,
are we able to drill down,

(07:27):
into the specialty area of surgeons?
I mean, obviously we're interestedin what general surgeons do,
but there are other, occupational areas,for orthopedists,
for, potentially, other, other areas.
Were you able to look at that at all?
Does the data reflect what kind ofspecialty areas in surgery they performed?

(07:49):
You know, unfortunately, this is anothersort of, limitation of the data.
There's always a trade off between thethe number of data
observations that can be capturedby a data set and the detail,
as, as you, as, you know,and in this situation, while we're able
to capture all of the deathsoccurring in the United States,

(08:13):
we weren't necessarily able toto distinguish
between the type of surgeon.
However, the variable
that characterizes occupation
does seem to be updatedin different iterations of the data set.
Earlier iterations of the data setdidn't even allow us

(08:33):
to differentiate surgeonsfrom non surgeons.
However, this most recent version
in 2023 did and includedseveral other specialties,
including radiologyand emergency medicine.
However, didn't really differentiate
between types of surgeons.

(08:55):
That's very valuable to, to hear you say that.
So now that we have all of thiswonderful data,
what do we do with it?
I mean, what do we do to improve mortalityrates?
What do we do for our trainees,who are just getting into surgery?
How can we how can we help them?
That's an excellent question.
I'd say that

(09:15):
the the primary purpose of a studylike this is you know, epidemiologic
in nature to identify a problemthat we may not have even known existed.
And it's oftenhard to tell that some of these types
of differencesbetween occupations are even present.

(09:36):
Unless,they're looked at at a really large scale.
The population level.
So, so I would say our findings pointto several potential intervention areas.
First, you know, mitigating fatigue,either through scheduling reforms
and call structure redesigns and,
and even post call transportation support,

(09:58):
might reduce injury related mortalityif they're truly related to the patient.
Second, you know, attention to long term
cardiovascular risk through hypertensionscreening and management.
And even cancer screeningmight be warranted in this population.
You know, of course,

(10:20):
these are all very much
postulatory that the data don't truly,
supportthe mechanism behind the differences.
So there's also an importance
to study these questions further.To figure out, you know,
perhaps at the institutional level,if people are more likely to get into,

(10:45):
car accidents and why, specifically,they are.
Yeah, that's the the summary.
Well, certainly that's well said.
And, you know, some years ago,as you well know, we, we identified,
the work hours for surgical trainees,the 80 hour work week, and such.
And that was done to create

(11:08):
a better health environment, less
accidents going to and from the hospital,these kinds of things.
You mentioned, some limitationsto the study already.
I wonder if there are any otherlimitations, to a study like this.
Sure.
The study has several limitations.
The first limitation is that itrelies on a single year of mortality data,

(11:29):
which limits our abilityto assess trends over time.
Because the occupational variable,
was updated to distinguish surgeons
from non physician surgeons only in the mostrecent iteration of the data set.
We we have a limited numberof observations,

(11:50):
which also precludesa more detailed analysis
of the causes that might be contributingto these deaths.
For example,we evaluated cancer-specific mortality,
but we weren't really ableto, evaluate the specific cancer sites.
And we weren't necessarily ableto do a more in-depth

(12:12):
intersectional analysis on ageor sex or race
because of the limited numberof observations.
I’d say the other,
major limitation is that,
occupation is reported by the informants
rather than the, the decedents themselves.

(12:37):
And so there may be some misclassificationthat is possible.
However, it's not super evident to us how
that misclassification would affectsurgeons more than other physicians
or than other than otherhighly educated professionals.
So it's likely that that limitationis not heavily impacting the findings.

(13:00):
Well, your points are excellent.
And I'm sure that the informationthat you've gleaned already
will lead to many other studies.
We've been talking to
Dr. Vishal Patel, who along withhis colleagues, has written a marvelous
article for JAMA Surgery, on the mortalityamong surgeons in the United States.
And, I want to thank you, the listeners,

(13:21):
for being with ustoday on Surgical Readings.
And thank you for listening and learning.
Thank you for joining us on Surgical Readings,
a podcast brought to youby the American College of Surgeons.
I hope you enjoyed the episode.
Please let your friends,

(13:41):
trainees, and colleaguesknow about the podcast.
On social media use the hashtag #SurgicalReadings.
I'm Dr. Rick Greene.
Until next time,thank you for listening and learning.