This transcript has been edited for clarity.
welcome to impact factor, Every week, we provide commentary on new medical research. I'm Dr. F. Perry Wilson from Yale School of Medicine.
Today, it's a battle of the sexes as we work on papers that make us think, “Wow, what an interesting piece of research.'' It will make you think, “Wow, what an interesting study,'' but it will also make you think, “Wow, I wish I didn't have to do that research.'' Because such research always involves some difficulties. They say something about medicine, but they also say something about society. That makes this situation a little unstable. But that has never stopped us. So let's go ahead and answer the question, “Can women be better doctors than men?”
On the surface, this question seems nearly impossible to answer. That's too wide. What does it mean to be a “better” doctor? At first glance, there seem to be too many variables to control, such as type of physician, type of patient, and clinical scenario.
But this study“Comparison of Hospital Mortality and Readmission Rates by Physician and Patient Gender” published in . Annual report of internal medicineuses a very ingenious method to defeat all prejudices by leveraging two simple facts. First, modern hospital care is primarily performed by hospital workers. Second, hospitalist jobs are shift-based, so the hospitalist you are assigned to a hospital admission is almost random.
In other words, if you are hospitalized with an acute illness and a hospitalist attends you, you have no control over whether it is a man or a woman. Is this a randomized trial? No, but it's not bad.
Researchers used Medicare claims data to identify adults 65 and older who had unelected hospitalizations across the United States. The claim revealed the patient's gender and the name of his doctor. By linking to a database of health care providers, it will be possible to determine the gender of the health care provider.
The goal was to confirm the results across the four dyads.
-
Male patient – male doctor
-
Male patient – female doctor
-
Female patient – male doctor
-
Female patient – female doctor
The primary outcome was 30-day mortality.
I mentioned that focusing on hospital workers creates pseudo-randomization, but let's take a look at the data just to be sure. Just under 1 million patients were treated by about 50,000 doctors, 30% of whom were women. Additionally, female and male patients differed, but there were no differences regarding the gender of the admitting personnel. Therefore, by physician gender, patients were similar in mean age, race/ethnicity, household income, Medicaid eligibility, and comorbidities. The authors also created a similar “predicted mortality” score across groups.
Now, female doctors were a little different than male doctors. Female hospitalists were slightly more likely to have an osteopathic degree, had slightly fewer annual admissions, and were slightly younger.
That is, there are roughly similar patients regardless of who the hospitalist was, but hospitalists differ on factors other than gender. Are you okay.
I've graphed the results here. Female patients had a significantly lower 30-day mortality rate than male patients, but those treated by female doctors had even better survival rates than male doctors. For male patients, physician gender did not have a particularly strong effect on outcome. A secondary outcome, her 30-day readmission, showed a similar trend.
Admittedly, this is a relatively small effect, but when you multiply this effect by the millions of hospitalized patients per year, you can begin to calculate the real numbers.
So what's going on here? I think there are four main possibilities.
Let's start with the obvious. On average, women are better doctors than men. I am married to a female doctor, and based on my personal experience, this statement is definitely true. But why is this so?
The authors cite Suggestive data Female physicians are less likely than male physicians to ignore patient concerns, especially those of female patients, and therefore will likely miss fewer diagnoses. However, because this is not possible to measure with administrative data, this study found that female physicians had a lower average height than male physicians, beyond suggesting that their lower average height mediated the advantage. You can't tell if you're even attentive. Perhaps the key is to get closer to the patient?
The second possibility here is that this has nothing to do with the doctor's gender. It has to do with other things related to the doctor's gender. For example, we know that female doctors see fewer patients per year than male doctors, but the study authors adjusted for this in their statistical models. Still, there may be other unmeasured factors (confounders). By the way, confounders do not necessarily change the main results. teeth It is better to see a female doctor.Just not because they are women. This is a useful marker of other qualities, such as age.
A third possibility is that this study represents a phenomenon called collider bias. The idea here is that physicians can only participate in research if they are hospitalists, and the quality of physicians who choose to become hospitalists may vary by gender. Talented medical residents who consider specific lifestyle issues when deciding on a specialty may find hospital medicine particularly appealing. Additionally, the propensity to pursue a more lifestyle-friendly specialty may differ by gender. In some previous studies,. If that's true, female hospitalists may be better than men. This is because a male doctor at that level would not become a hospitalist.
Okay, don't write. I'm just trying to cite an example of how to think about collider bias. Although they cannot prove that this is the case, the authors did, in fact, conduct a sensitivity analysis for all physicians, not just hospital workers, and they show the same results. This is probably not true, but epidemiology is fun, right?
And a fourth possibility: This is just statistical noise. The effect size is incredibly small and only within the statistically significant range. Particularly when working with very large datasets such as this one, great care must be taken to avoid overinterpreting small but statistically significant results.
In any case, this is an interesting study and made me think about how to present it, and of course made me a little worried. I'm sorry if I'm being disrespectful here in dealing with complex issues of sex, gender, and society. But I'm not sure what you're expecting. After all, I'm just a male doctor.
F. Perry Wilson, MD, MSCE, is an associate professor of medicine and public health and director of the Clinical and Translational Research Accelerator at Yale University. His science communication work can be seen on the Huffington Post, NPR, and here on his Medscape.he tweets @fperrywilson and his book, How drugs work and when they don't work, currently available.
