This has been widely reported over the past few days:
Doctors from Harvard have an intriguing suggestion for saving 32,000 lives each year: Make sure all senior citizens who wind up in the hospital are treated by female doctors.
After examining the medical records of Medicare patients from across the country, the Harvard researchers calculated that 10.82% of those treated by physicians who were women died within 30 days of being admitted to the hospital. Among patients treated by male physicians, the 30-day mortality rate was 11.49%, according to a study published this week in JAMA Internal Medicine.
That gender gap persisted even after the researchers accounted for factors like the age, gender and income of patients, how sick those patients were when they first checked into the hospital, the resources of the hospitals and the experience of the doctors. In that analysis, the Harvard team found that 11.07% of patients treated by women died within 30 days of being hospitalized, compared with 11.49% of patients treated by men.
I haven’t had time to check out the original study, but I can’t see any particular reason why the effect being reported wouldn’t be true.
I’d be interested in seeing the data by age cohort.
My understanding is that a lot of things were controlled for (also haven’t read it), so presumably they controlled for this.
Post hoc ergo propter hoc?
An awful lot of things were not controlled for. For example, they looked at how sick they were on admission, but not what happened if they got sicker. Did they get transferred to another doc? As usual, a prospective study would be the way to go. (Would not totally be surprised if this was confirmed. Running the stats for my own department I have found that our female docs are generally more willing to follow protocols.)
It is safe to assume you know a lot more than I do about this field. I do note that your comment indicates that you have noticed that female doctors are more likelty to follow protocol, and that in itself is likely to different outcomes.
In a number of fields, it seems that women slightly outperform men on average, but the distribution for male performance has much wider tails (i.e., many more at the bottom and many more at the top) leading to a greater percentage of the superstars in the field (and a greater percentage of the lowest performers in the field) being male. If two populations were exactly identical in all regards except willingness to follow protocol, the higher average & lower variance is what you’d expect from the group that follows protocol more closely, assuming of course the protocol has any logic to it.
I think it was Public Citizen or some other place I was reading. There is a protocol for patients or a checklist of things to check out on an incoming patient. If you do not follow I, you are bound to miss something along the way. In hospitals this has been proven to eliminate many mistakes being made. Myself, I went to the doctor in 2012 just after Thanksgiving complaining of chest pain. Thought it was pneumonia. Did the EKG and my heart was fine according to it. Runner, backpacker, low cholesterol, etc; what could be wrong? On Monday in Mansfield , Ohio where I worked, I went to Med Central hospital ER and complained of pain in the upper part of my chest.
Two vials of blood drawn and an imaging. No EKG and it was determined my heart was fluttering. Triple bypass Thursday of that week probably saved my life and my heart as I had no heart attack damage to it. There is a protocol to stuff and following it is important. My knowledgeable Cardiologist told me that even with an enzyme test, it might have been missed. Last time I checked in to the ER, I had a reoccurrence of ITP (platelets drop to zip). No need to test for that as I knew. Blood blisters in my mouth and purpura all over me. I was show and tell for the residents that morning. Think the USMC poisoned me with the water at Lejeune.
Maybe doctors and male doctors do not follow protocol as well. Not sure; but, protocol was one of the issues in eliminating hospital mistakes.
I heard an interview on CBC radio with one of the lead authors this morning. After explaining the methodology for the study he was asked why the 0.5% difference in outcome based on the sex of the doctor. He said that wasn’t what the study was meant to study but he offered a few ideas based on different studies from others. Following protocol was one that he then divided to include the subset of better practicing evidence based medicine. Another was better bed side manner that allowed for more patient input. to incorporate in diagnosis.