Women more likely to die if surgeons are male

[Allo V Psycho]

Guardian:
https://www.theguardian.com/society/202 ... geon-study

They found that for women, treatment by a male surgeon was associated with a 15% increased likelihood of a poor outcome than if they were treated by a female surgeon. However, men experienced no differences whether they were treated by a male or female surgeon.

Similarly, women who were operated on by a male surgeon had a 32% higher risk of death than those whose surgery had been performed by a woman. For example, while 1.4% of women who had a cardiothoracic operation with a male surgeon died, fewer – 1% – did so when a female surgeon was involved. In both brain surgery and vascular surgery, while 1.2% of women who underwent either type of operation with a male surgeon died, again that proportion was much lower among those whose surgeon was female – 0.9% – giving a 33% higher risk of death.

Overall, female patients also had a 16% greater risk of complications and an 11% greater risk of readmission and were 20% more likely to have to stay in hospital longer.

Women had a higher risk of death, readmission or complications when a man performed the operation across many of the 21 types of surgery analysed.

For example, while 20.2% of women who had cardiothoracic (chest) surgery by a male surgeon suffered some form of adverse reaction, a lower percentage – 18% – did so if their surgeon was female. The same pattern was seen in general surgery, brain surgery and orthopaedic surgery.

Technical differences between male and female surgeons are unlikely to explain the findings “as both sexes undergo the same technical medical training”, said Jerath.

“Implicit sex biases”, in which surgeons “act on subconscious, deeply ingrained biases, stereotypes and attitudes”, may be one possible explanation, she said. Differences in men’s and women’s communication and interpersonal skills evident in surgeons’ discussions with patients before the operation takes place may also be a factor, she added. And “differences between male and female physician work style, decision-making and judgment”.

Fiona Myint, the vice-president of the Royal College of Surgeons of England, highlighted that 86% of consultant (senior) surgeons in Britain were men.

Abstract:
https://jamanetwork.com/journals/jamasu ... .2021.6339

Results Among 1 320 108 patients treated by 2937 surgeons, 602 560 patients were sex concordant with their surgeon (male surgeon with male patient, 509 634; female surgeon with female patient, 92 926) while 717 548 were sex discordant (male surgeon with female patient, 667 279; female surgeon with male patient, 50 269). A total of 189 390 patients (14.9%) experienced 1 or more adverse postoperative outcomes. Sex discordance between surgeon and patient was associated with a significant increased likelihood of composite adverse postoperative outcomes (adjusted odds ratio [aOR], 1.07; 95% CI, 1.04-1.09), as well as death (aOR, 1.07; 95% CI, 1.02-1.13), and complications (aOR, 1.09; 95% CI, 1.07-1.11) but not readmission (aOR, 1.02; 95% CI, 0.98-1.07). While associations were consistent across most subgroups, patient sex significantly modified this association, with worse outcomes for female patients treated by male surgeons (compared with female patients treated by female surgeons: aOR, 1.15; 95% CI, 1.10-1.20) but not male patients treated by female surgeons (compared with male patients treated by male surgeons: aOR, 0.99; 95% CI, 0.95-1.03) (P for interaction = .004).

Link: https://ps-vascular.ch/wp-content/uploa ... tcomes.pdf

The statistical methods don't seem very complicated, so I would be surprised if there is a subtle hidden stats error. (I will note that Odds Ratio is not the same as Relative Risk, and Relative Risk is not the same as Absolute Risk, but I don't think either issue is significant here).

My qualitative research buddy proposes to me (quite frequently...) that the numbers can tell you what happens, but not why it happens. Although the authors make various suggestions, I would really value any comments or suggestions from forum members. It could make a difference to how I go about the day job.

There are similar studies out there.

Same lead author:

Wallis, C.J., Ravi, B., Coburn, N., Nam, R.K., Detsky, A.S. and Satkunasivam, R., 2017. Comparison of postoperative outcomes among patients treated by male and female surgeons: a population based matched cohort study. Bmj, 359.

Results 104 630 patients were treated by 3314 surgeons, 774 female and 2540 male. Before matching, patients treated by female doctors were more likely to be female and younger but had similar comorbidity, income, rurality, and year of surgery. After matching, the groups were comparable. Fewer patients treated by female surgeons died, were readmitted to hospital, or had complications within 30 days (5810 of 52 315, 11.1%, 95% confidence interval 10.9% to 11.4%) than those treated by male surgeons (6046 of 52 315, 11.6%, 11.3% to 11.8%; adjusted odds ratio 0.96, 0.92 to 0.99, P=0.02). Patients treated by female surgeons were less likely to die within 30 days (adjusted odds ratio 0.88; 0.79 to 0.99, P=0.04), but there was no significant difference in readmissions or complications. Stratified analyses by patient, physician, and hospital characteristics did not significant modify the effect of surgeon sex on outcome. A retrospective analysis showed no difference in outcomes by surgeon sex in patients who had emergency surgery, where patients do not usually choose their surgeon.

Hospital medicine rather than surgery:

Tsugawa, Y., Jena, A.B., Figueroa, J.F., Orav, E.J., Blumenthal, D.M. and Jha, A.K., 2017. Comparison of hospital mortality and readmission rates for Medicare patients treated by male vs female physicians. JAMA internal medicine, 177(2), pp.206-213.

Findings In this cross-sectional study, we examined nationally representative data of hospitalized Medicare beneficiaries and found that patients treated by female physicians had significantly lower mortality rates (adjusted mortality rate, 11.07% vs 11.49%) and readmission rates (adjusted readmission rate, 15.02% vs 15.57%) compared with those cared for by male physicians within the same hospital.

Greenwood, B.N., Carnahan, S. and Huang, L., 2018. Patient–physician gender concordance and increased mortality among female heart attack patients. Proceedings of the National Academy of Sciences, 115(34), pp.8569-8574.

We examine patient gender disparities in survival rates following acute myocardial infarctions (i.e., heart attacks) based on the gender of the treating physician. Using a census of heart attack patients admitted to Florida hospitals between 1991 and 2010, we find higher mortality among female patients who are treated by male physicians. Male patients and female patients experience similar outcomes when treated by female physicians, suggesting that unique challenges arise when male physicians treat female patients. We further find that male physicians with more exposure to female patients and female physicians have more success treating female patients.

[Imrael]

female surgeons might be on average slightly better due to having to overcome adverse social pressures? Admittedly that wouldnt explain why male patients see no difference.

Also wondering, depending on type of surgery, if later presentation to primary care by men is somehow involved.

[Woodchopper]

The two most important variables remain unobserved: the skill of the surgeon and the infirmity of the patient. The newspaper article linked to above suggests that female surgeons may be better (eg better at communicating and more sound judgement). But before concluding that we should consider whether male surgeons are more likely to operate on more difficult cases.

[Bird on a Fire]

Why would either of those factors affect women patients but not men?

[sTeamTraen]

Why would either of those factors affect women patients but not men?

Broadly: Because (a) surgeons are not randomly assigned to surgical specialities by gender and (b) patients are not randomly assigned to conditions requiring surgery by gender. Indeed, we can expect both of those to be not only not random, but systematically biased. (We can probably assume that, within a speciality, the assignment of a male or female surgeon to a patient is effectively random or at a minimum unbiased, at least in Canada, although I can think of some parts of the world where even that might not be the case.)

I don't have the paper to hand, but I think I recall that there was quite an imbalance in the types of surgery being carried out by male and female surgeons (for example, I think that male surgeons were carrying out more orthopaedic procedures). I also seem to recall that the study also said that the reported effect (i.e., that male surgeon/female patient had the most deaths) was larger in older women.

So for example, you might find that a greater proportion of the operations being carried out on older women could be (a) more complicated/higher risk operations and (b) in specialities dominated by men. The latter point there could, of course, reflect historical sexism and/or other differences between men and women in terms of the surgical specialities that they either choose to go into and/or are [not] allowed to go into; my feeling is that that will explain quite a lot of the differences.

It might also be that women are more likely than men to be offered surgery in the first place. An 80-year-old woman already has considerably greater life expectancy than an 80-year-old man. So perhaps out of 100 of each gender for a given condition, 70 women get surgery from a man of whom 10 die within a year "from the operation", while only 30 men get the surgery from the same man, of whom 2 die from the operation. But of the 40 other men who weren't operated on, perhaps 15 die within a year, or whatever the criterion period is, because they weren't operated on. The latter point is intended to show that there are many outcomes that are not looked at in a study of this kind (for example, people who were not candidates for surgery are not represented in the denominator).

All of those factors need to be looked at before we can draw conclusions about what is going on, or indeed if anything is going on at all. Retrospective observational studies are hard to interpret because you really have to make sure you have accounted for the sources of variation that you can't control. And despite the number of patients, the study was not especially well powered, because power for this kind of study depends on the number of events of interest (i.e., deaths), and modern surgery is very safe. This means that the authors probably did not have enough cases to start to tease apart the subgroups. The headline p values of 0.02 and 0.04 tend to suggest that too.

(tl;dr: What Woodchopper said)

[Bird on a Fire]

Well the paper is here https://jamanetwork.com/journals/jamasu ... .2021.6339

N=1,320,108 so "not enough cases" definitely isn't a problem.

But I don't have access to see what other kinds of matching they did. I'd assume a published paper in a proper journal would think of obvious things like age and procedure if we can come up with them here in five minutes' spitballing, but hopefully someone can take a look and confirm rather than guessing.

ETA just noticed AvP linked a pdf.
Methods:

Patient age, sex, geographic location (local health integration
network's), geographically derived socioeconomic status,
rurality, and general comorbidity (Johns Hopkins aggregate
disease group16) were obtained. We also collected data
regarding surgeon sex, years in practice, specialty, and surgi-
cal volume. Surgical volume was determined for each sur-
geon and the specific procedure by identifying the number
of identical procedures the operating surgeon performed in
the previous year, operationalized in quartiles. Hospital in-
stitution identifiers were used to account for facility-level
variability.

Results:

Baseline characteristics of the 4 groups are
provided in Table 1; female surgeons in both relevant dyads
were younger and had lower annual surgical volumes than
male surgeons. Similarly, female surgeons treated younger
patients with less comorbidity than male surgeons. Overall,
189 390 patients (14.9%) experienced an adverse postopera-
tive outcome: 22 931 (1.7%) died, 88 132 (6.7%) were read-
mitted, and 114 421 (8.7%) had significant complications in
the 30-day following surgery.
We first considered the association of surgeon-patient sex
discordance while accounting for both patient and surgeon sex
independently as well as other procedure-, patient-, sur-
geon-, and hospital-level factors. Sex discordance between the
operating surgeon and the patient was associated with a sig-
nificantly increased likelihood of a composite adverse post-
operative outcome (adjusted odds ratio [aOR], 1.07; 95%
CI, 1.04-1.09).

In stratified analyses according to surgeon, patient, pro-
cedural, and hospital characteristics while assessing the pri-
mary composite adverse postoperative outcome, we found sig-
nificant heterogeneity in the association of sex discordance
with development of adverse postoperative outcomes by pa-
tient sex: sex discordance was associated with worse out-
comes for female patients (aOR, 1.11; 95% CI, 1.06-1.16) but
better outcomes for male patients (aOR, 0.96; 95% CI, 0.93-
0.99) (P for interaction = .004).

So a quick butchers at the paper suggests that the authors weren't totally incompetent.

I wonder if they go on to suggest any more plausible explanations for the observed results?

[Millennie Al]

One other possiblility is that it's not the surgeons who are causing this effect but the patients. It says:

A retrospective analysis showed no difference in outcomes by surgeon sex in patients who had emergency surgery, where patients do not usually choose their surgeon.

Maybe women who feel more ill are more likely to choose a male surgeon, or maybe women communicated better with a female surgeon before the operation, or some other such reason introduces the bias.

Without randomisation it is extremely difficulty to tell what's going on.

[sTeamTraen]

N=1,320,108 so "not enough cases" definitely isn't a problem.

J'insiste: Yes, it could be.

For example, the death rate was around 1% (source: eyeballing Table 2), but for plastic surgery it was either 0.1% or 0.0% for each M/F-M/F pair (damn you, single-digit precision!) out of 45,409 operations, so perhaps there were 40 deaths across 4 categories during 13 years. You can't do very much with those numbers. For general surgery you have 554,988 patients with death rates a little under 1%, so perhaps 4,500 total deaths. You can do something with that, of course.

Interestingly, Table 2 shows that male patients had substantially worse outcomes of all kinds (death, readmissions, complications, and stay length) than female patients, more or less independent of the gender of the surgeon, and that effect of patient gender appears to be several times larger than the difference in outcomes for women by surgeon gender. Now of course that might well be expected (men are probably less healthy than women; that's why they die earlier), although again there can be counter-intuitive effects if that difference in health makes men no longer candidates for surgery. The point is that there are lots of factors at play in a non-randomised study, and researchers have to be very careful to justify why they are focusing on X and not Y. Interpreting smaller interactions in the presence of a larger main effect is a risky business.