Ha! That jibes with my experience too, my wife prefers a female doctor, and I do too for her. In fact, my mother and sisters have the same preference come to think of it.
> We accounted for patient characteristics, physician characteristics, and hospital fixed effects. Patient characteristics included patient age in 5-year increments (the oldest group was categorized as ≥95 years), sex, race/ethnicity (non-Hispanic white, non-Hispanic black, Hispanic, and other), primary diagnosis (Medicare Severity Diagnosis Related Group), 27 coexisting conditions (determined using the Elixhauser comorbidity index28), median annual household income estimated from residential zip codes (in deciles), an indicator variable for Medicaid coverage, and indicator variables for year. Physician characteristics included physician age in 5-year increments (the oldest group was categorized as ≥70 years), indicator variables for the medical schools from which the physicians graduated, and type of medical training (ie, allopathic vs osteopathic29 training).
They don't say _how_ they controlled for those characteristics though. Presumably, they divided each patient's calculated 30-day risk of death [1, table 10] by the relative ratios in risk of death of every category, calculated from the same data set. That should probably cut it for controlling for this particular difference, although I am not a statistician.
It's a regression model, so "accounted for" usually means that the factors were included in the models.
Ideally, we'd ask a group of male doctors and a group of female doctors to independently diagnose and treat the same set of patients. We can't actually do that--in addition to the cost, you obviously can't treat the same patient twice. However, we can try to estimate this effect statistically.
First, they build a model describing the probability of a patient dying within 30 days. They used a linear probability model, which essentially means that the probability of someone dying is the sum of the "weights" related to the patient, doctor, and hospital. These weights are estimated from the data using ordinary least squares (the same way you may have learned to fit a line to points at school).
Having built the model, they then ask what the marginal effect of the doctor's sex is. In other words, if you hold everything but that constant, how does the probability of dying change? They cite a nice Stata guide (#32 in the paper here: http://www.stata-journal.com/sjpdf.html?articlenum=st0260 ) which gives some background and examples.
The rest of the paper looks at different variations (only hospitalists, different diseases, etc), using a pretty similar approach.
I thought that as more younger doctors are women, then perhaps younger doctors are better. I'm always a bit disappointed that science news doesn't dig that one level deeper. You'd hope they'd cover both the ideas we came up with in 5 seconds in the paper, but if they don't mention it in the story then you're left guessing.
That's an interesting observation. On the one hand younger doctors have been more recently exposed to "current" ideas, research, and stuff. On the other hand younger doctors lack the experience to catch things that are unusual and often discussed as theoretical.
I'd be curious to see if there were research on this topic now!
They only looked at hospital visits, which they assume are pretty randomly distributed based on the physicians' work schedules. The patients don't get to pick their doctors in that case.
On HN, we need to do better than the internet trope of dismissing a study with a single drive-by, usually obvious objection. No doubt there are a lot of bad studies, but bad studies plus bad comments about them is worse.
We should inhibit the impulse to dish out snark for the brief thrill of feeling smarter than an author or researcher, not because they don't occasionally miss the obvious (who doesn't?), but because it lowers the standard of discussion here.
A better way to present such a comment might be as a question: "In my experience, X. Does the study Y?"
I think my criticism is valid. Not only was this particular title not the original one, it still relies on the same false generalization that the article is using to bait people into clicking through.
From the guidelines:
> please use the original title, unless it is misleading or linkbait.
I was about to argue that the HN title should say "Elderly patients..." since there's no guarantee that the study generalizes. However, it looks like NPR also generalized the conclusion in their title.
This is extremely sexist. Not only do they assume genders (and only assume two) but they are relating competency based on genitalia.
I read the paper (1) and they used Doximity, a data gathering company, to provide them with gender + age data. Some of it is self reported but a lot is "gathered" from other sources.
The sort of sexist nonsense needs to go and we need to realize that you can be any gender you like or believe without it effecting your professional competence.
Whooaaaa that was apparently a powerful dog-whistle. Why'd you air-quote "gathered"? Doximity pulls data from NPI, this is not some "flying by the seams of our pants" industry, health care has a lot of controls and thus, a lot of hard data.
I presume you mean heteronormative. And it is somewhat heteronormative, but: the paper has a section, "Limitations", that acknowledges this problem and why including more gender options wasn't needed for (and would have made more difficult) the analysis:
> Third, we used self-reported data to identify physician sex, which requires respondents to categorize themselves as either male or female; therefore, we could not capture respondents who were transgender. It is possible that transgender physicians chose to either leave this question blank or select 1 of the 2 available categories, which may lead to a low degree of misclassification. Any misclassification in self-reported sex would likely bias our estimates toward the null.
Although this last sentence is not totally true (if all transgenders labelled themselves as gender A, and they had non-random performance, it could bias the findings in favor/against gender A), it probably is true in the setting of the study, where you can assume transgender doctors label themselves as one or the other sex with a similar proportion, and have either male or female-level performance. Plus there aren't very many.
Heteronormative refers to the assumption that being heterosexual is the normal or preferred sexual orientation. Since no one has brought up the topic of sexual orientation, you're probably trying to use the word cisnormative here.
I wonder if someone has collected together the timescales on which professions went from "of course women can't do that, don't be ridiculous" to "of course women can do that". I'd guess some would start earlier than others, and some would make the transition faster than others. Our own profession of programming seems to be one of the laggards in both, potentially even going backwards in its youth. Would probably be interesting to see the data laid out graphically if it could be made objective in some way (first admitted to college, first professor, achievement of parity in numbers perhaps?)
NPR has been getting pretty clickbaity recently. They had a segment recently about how scandalous it was that CharityBuzz was auctioning a coffee with Ivanka Trump, but they totally ignored the fact that they were also auctioning lunch with Tony Podesta. I guess with 40% of their funding coming from the public, it's important that they push certain narratives that resonate with their audience, rather than being impartial.
It's a pretty weak signal given all the sorts of things you'd have to correct for. 11.07% vs. 11.49%, a 4% relative difference.
For example, they exclude all data from ICU doctors because they're disproportionately men – suppose only the best doctors work in the ICU, by excluding that you're excluding more of the best male doctors than the best of the female doctors resulting in leaving a better population of women in your study.
They correct for illness but not for doctor specialty, perhaps females specialize in things that happen to have lower mortality rates.
... and on and on and on. There are so many variables to possibly correct for that a study like this (especially having skimmed it) is pretty much useless. It would be trivial to do a study like this and get whatever result you wanted. Just stop doing statistical corrections when you get the result you're looking for. P-value doesn't mean anything here because it doesn't capture the ways the study could be wrong.
Good point about not correcting for medical speciality. I'm neither a doctor nor a statistician, but surely that pretty much invalidates the whole thing?
They report results for "hospitalists", which is its own specialty. It's not a super-specialized field, in the way that some doctors are pediatric neuro-radio-oncologists, but the hospitalists have largely had similar training.
Actually, their methodology seems pretty sound, with robust findings, including in the subgroup of patients cared by hospitalists, where -- as the authors correctly point out -- patients are pretty much randomized to a female or male hospitalist.
The question is "in a given medical speciality -- internal medicine in this study --, are patient's outcomes related to the admitting physician sex?", not if women make better doctors than females. As such, your point about the best doctors not being in internal medicine is pretty a non sequitur.
Basically, I see two potential biais that could explain at least part of the difference in outcomes :
1. Its is "common knowledge" that male doctors tend to refuse admissions more often than females. (I am not aware of any hard data to prove this point) It could be that the patients that they do admit are somewhat sicker that those they let go directly home. That could explain both higher mortality and re-admission rates.
2. It is again, "common knowledge" (again, I'm not aware of any study) that women write better discharge summaries, including all co-morbid conditions. When coded in a database, those patients will appear sicker than similar ones treated by male doctors, and this could skew any analyses.
A final confounder is that patients often have many doctors during the stay, both male and female, and the study assigned as "the" doctor the one who billed most for the patient, which may not be the one who had the most impact on outcomes. It is difficult to say how this would advantage any particular sex, however.
I can find other potential bias that would explain the difference in outcomes, like for example internal competition. If female and male doctors move different within the profession and compete on different aspects (a common finding in many other professions), the outcome will also reflect that. The result might thus simply be explained that male doctors compete more for medical positions in areas outside of elder care.
If that data is true, then we should see a reversed result in other medical areas. It could be that if you wanted the best result of surgery, pick a male doctor. If you want the best result afterward, pick a female doctor.
Ha! If we assume the goal of such research is to improve patient care (rather than grab headlines) then is there any benefit to comparing demographic groups like this? If the hypothesis is that female doctors get better outcomes because they follow clinical guidelines more closely, then surely we should be looking at something like patient outcome vs. how closely clinical guidelines were followed.
This is an interesting study. Men and women are different. We tend to forget that in this pseudo-post-equality world we live in. We forget that equality isn't something that's intrinsic to humanity, but rather it's a social construction we artificially impose upon biological reality:
It is important that this study is limited to elderly patients. It does use a lot of data though from many doctors and physicians, but only over the course of a 30 day window.
I'd like to see studies that do this with younger patients as well, as well as studies done over a longer timespam. If the same percentage present themselves, it be interesting to see if there was some kind of training or something that could be added to the medschool/education process to help male physicians implement the natural practices used by the female physicians to reduce the overall mortality rates.
If studies on younger patients or longer timespans show the percentages converging down to zero, than maybe this study was an isolated result from 30 days in that one facility and the differences aren't really statically significant. Studies like this really need to be reproduced.
49 comments
[ 2.5 ms ] story [ 83.5 ms ] threadIn my experience women tend to go to female doctors and given women live longer than men..
hey presto magico.. patients of women doctors live longer
> We accounted for patient characteristics, physician characteristics, and hospital fixed effects. Patient characteristics included patient age in 5-year increments (the oldest group was categorized as ≥95 years), sex, race/ethnicity (non-Hispanic white, non-Hispanic black, Hispanic, and other), primary diagnosis (Medicare Severity Diagnosis Related Group), 27 coexisting conditions (determined using the Elixhauser comorbidity index28), median annual household income estimated from residential zip codes (in deciles), an indicator variable for Medicaid coverage, and indicator variables for year. Physician characteristics included physician age in 5-year increments (the oldest group was categorized as ≥70 years), indicator variables for the medical schools from which the physicians graduated, and type of medical training (ie, allopathic vs osteopathic29 training).
They don't say _how_ they controlled for those characteristics though. Presumably, they divided each patient's calculated 30-day risk of death [1, table 10] by the relative ratios in risk of death of every category, calculated from the same data set. That should probably cut it for controlling for this particular difference, although I am not a statistician.
Paper link:
https://jamanetwork.com/journals/jamainternalmedicine/fullar...
[1] Supplemental material https://jamanetwork.com/data/Journals/INTEMED/0/IOI160102sup...
Ideally, we'd ask a group of male doctors and a group of female doctors to independently diagnose and treat the same set of patients. We can't actually do that--in addition to the cost, you obviously can't treat the same patient twice. However, we can try to estimate this effect statistically.
First, they build a model describing the probability of a patient dying within 30 days. They used a linear probability model, which essentially means that the probability of someone dying is the sum of the "weights" related to the patient, doctor, and hospital. These weights are estimated from the data using ordinary least squares (the same way you may have learned to fit a line to points at school).
Having built the model, they then ask what the marginal effect of the doctor's sex is. In other words, if you hold everything but that constant, how does the probability of dying change? They cite a nice Stata guide (#32 in the paper here: http://www.stata-journal.com/sjpdf.html?articlenum=st0260 ) which gives some background and examples.
The rest of the paper looks at different variations (only hospitalists, different diseases, etc), using a pretty similar approach.
That's an interesting observation. On the one hand younger doctors have been more recently exposed to "current" ideas, research, and stuff. On the other hand younger doctors lack the experience to catch things that are unusual and often discussed as theoretical.
I'd be curious to see if there were research on this topic now!
The paper reports that they included both physician age AND years in practice, along with a slew of other factors.
We should inhibit the impulse to dish out snark for the brief thrill of feeling smarter than an author or researcher, not because they don't occasionally miss the obvious (who doesn't?), but because it lowers the standard of discussion here.
A better way to present such a comment might be as a question: "In my experience, X. Does the study Y?"
From the guidelines:
> please use the original title, unless it is misleading or linkbait.
We reverted the HN title to the article title. I'm not sure the absence of 'elderly' counts as misleading but s/patients/elderly/ is probably ok.
I read the paper (1) and they used Doximity, a data gathering company, to provide them with gender + age data. Some of it is self reported but a lot is "gathered" from other sources.
The sort of sexist nonsense needs to go and we need to realize that you can be any gender you like or believe without it effecting your professional competence.
1 https://jamanetwork.com/journals/jamainternalmedicine/fullar...
Not necessarily, no.
> ... you can be any gender you like or believe without it effecting your professional competence.
Yep, but if one gender has a natural advantage then we should absolutely research into why, shouldn't we?
This time it is about women doing better, other times it is about men.
How about we stop saying sexism every time? Both when men seems to thrive/do better in certain jobs and when women do?
Edit: yelling->saying
> Third, we used self-reported data to identify physician sex, which requires respondents to categorize themselves as either male or female; therefore, we could not capture respondents who were transgender. It is possible that transgender physicians chose to either leave this question blank or select 1 of the 2 available categories, which may lead to a low degree of misclassification. Any misclassification in self-reported sex would likely bias our estimates toward the null.
Although this last sentence is not totally true (if all transgenders labelled themselves as gender A, and they had non-random performance, it could bias the findings in favor/against gender A), it probably is true in the setting of the study, where you can assume transgender doctors label themselves as one or the other sex with a similar proportion, and have either male or female-level performance. Plus there aren't very many.
Heteronormative refers to the assumption that being heterosexual is the normal or preferred sexual orientation. Since no one has brought up the topic of sexual orientation, you're probably trying to use the word cisnormative here.
Edit: Added a new line for better readability
To be clear, having connections to the Clinton's really is different than sitting in on meetings with foreign heads of state.
(I'm not sure I think that Ivanka having coffee with someone is problematic, but it's still different than a lobbyist doing it)
For example, they exclude all data from ICU doctors because they're disproportionately men – suppose only the best doctors work in the ICU, by excluding that you're excluding more of the best male doctors than the best of the female doctors resulting in leaving a better population of women in your study.
They correct for illness but not for doctor specialty, perhaps females specialize in things that happen to have lower mortality rates.
... and on and on and on. There are so many variables to possibly correct for that a study like this (especially having skimmed it) is pretty much useless. It would be trivial to do a study like this and get whatever result you wanted. Just stop doing statistical corrections when you get the result you're looking for. P-value doesn't mean anything here because it doesn't capture the ways the study could be wrong.
The question is "in a given medical speciality -- internal medicine in this study --, are patient's outcomes related to the admitting physician sex?", not if women make better doctors than females. As such, your point about the best doctors not being in internal medicine is pretty a non sequitur.
Basically, I see two potential biais that could explain at least part of the difference in outcomes :
1. Its is "common knowledge" that male doctors tend to refuse admissions more often than females. (I am not aware of any hard data to prove this point) It could be that the patients that they do admit are somewhat sicker that those they let go directly home. That could explain both higher mortality and re-admission rates.
2. It is again, "common knowledge" (again, I'm not aware of any study) that women write better discharge summaries, including all co-morbid conditions. When coded in a database, those patients will appear sicker than similar ones treated by male doctors, and this could skew any analyses.
A final confounder is that patients often have many doctors during the stay, both male and female, and the study assigned as "the" doctor the one who billed most for the patient, which may not be the one who had the most impact on outcomes. It is difficult to say how this would advantage any particular sex, however.
If that data is true, then we should see a reversed result in other medical areas. It could be that if you wanted the best result of surgery, pick a male doctor. If you want the best result afterward, pick a female doctor.
http://khanism.org/society/created-equal/
It is important that this study is limited to elderly patients. It does use a lot of data though from many doctors and physicians, but only over the course of a 30 day window.
I'd like to see studies that do this with younger patients as well, as well as studies done over a longer timespam. If the same percentage present themselves, it be interesting to see if there was some kind of training or something that could be added to the medschool/education process to help male physicians implement the natural practices used by the female physicians to reduce the overall mortality rates.
If studies on younger patients or longer timespans show the percentages converging down to zero, than maybe this study was an isolated result from 30 days in that one facility and the differences aren't really statically significant. Studies like this really need to be reproduced.
The study covered significantly more than one facility. The data included nearly 60k physicians and over 1.6M patients.
This is why we need to focus on equality of opportunity, not equality of outcome.