Women are on average more neurotic than men, therefore physicians will be more skeptical of their claims; psychosomatic symptoms are more common for females.
from https://en.wikipedia.org/wiki/Neuroticism
Edit: this isn't to say that there isn't a bias... maybe doctors know this already, but assume that the effect is bigger than it is? Alternatively, they don't know this, and could treat their patients better if they had known. In any case, nothing good comes from denying the truth.
Balance.
Sexist: "I'm going to be more skeptical of this particular woman's experience because she's a woman and probably is just hysterical."
Group differences are real, and burying or ignoring that truth is doomed to failure, since it is obvious. However, we have a moral responsibility to evaluate every person as an individual - to give each the benefit of the doubt, regardless of how people who look like them tend to behave.
I myself think this is a two way street. If you belong to a group with a particular characteristic and you're an exception, you may need to make an effort to make that explicit - with a statement, via behavior, via dress. And it is not evil when people make assumptions about you; it's understandable. At the same time, we should try to be fair to everyone, to keep in mind that exceptions exist to every rule, and to allow others to change our first perceptions of them easily. I think this approach walks the line between truth and justice, though I am certainly open to hearing a better strategy.
Suppose 99% of the time a set of symptoms is harmless and 1% of the time it's deadly. A 1% risk of death is extremely high and needs to be taken seriously even if ignoring it would be correct 99% of the time.
The other time I went to the doctor to complain about back pain, and get an appointment for a physio. The doctor asked me to do some movements that didn't seem directly related to my issues. Then she explained that they have to check for a specific condition (like a spinal bulge or something) that, albeit very rare, absolutely needs to be operated on as soon as possible.
Doctors and patients have little reason to care about odds directly they care about Risk which is ~= odds * harm. Odds going from say 10 to 15% or 0.015% to 0.02% don't have big impacts on your net Risk.
Not without having very detailed information on exactly how big the difference is and, equally important, how these differences are distributed within each sub-population.
Basically people will treat information that should realistically cause them to update their priors from 50-50 to 51-49 as if it was information that should cause them to update to 80-20 (and then they round off too 100-0).
I fully agree they exist, but it is very hard to separate which are just culture from those that are real differences. You can measure hormone differences of course, and there are other measurable things, but you quickly get into measurable differences but no idea if they are nature or nurture.