Having a woman as their doctor significantly improves the chances that older patients will survive a trip to hospital, and reduces the chances they will need to be readmitted. The differences are proportionately small, but add up to tens of thousands of lives a year. Explaining the reasons could improve clinical practice for everyone.

In the new study, researchers led by Dr Yusuke Tsugawa of the Harvard T.H. Chan School of Public Health and from other Boston medical research centers studied files for a random sample of more than 1.5 million patients aged 65 and over admitted to US hospitals over four years.

From this huge sample, they found that 11.07 percent of those treated by female physicians died within 30 days of admission. Those whose primary doctor was male had a death rate of 11.49 percent over the same time.

Patients of female doctors stood a 15.02 percent chance of being readmitted during the 30 days, which rose slightly to 15.57 percent with a male doctor.

To save a single death it would be necessary to shift 233 patients from male to female doctors, while 182 such changes would prevent, on average, one readmission. Averages can be deceiving. A good male doctor will certainly have better patient outcomes than a bad female one, but taken collectively the differences are too large to ignore. It is, for example, similar in size to the reduction in mortality achieved as a result of improvements in medical technology between 2003 and 2013. If male doctors could rise to match the standards of their female counterparts, 32,000 fewer patients would die each year.

Although the effect of the doctor's gender was not tested on younger patients, and the study was limited to an American context, the implications are larger still if the findings can be replicated elsewhere. The big question, however, is why.

In a paper in the journal JAMA Internal Medicine Tsugawa and co-authors write: “There is evidence that men and women may practice medicine differently. Literature has shown that female physicians may be more likely to adhere to clinical guidelines, provide preventative care more often, use more patient-centered communication and provide more psychosocial counseling to their patients.” Despite this wealth of evidence on differences in practice, only one previous study, using a much smaller sample, has measured consequences for mortality.

The reason doctors' behavior differs by gender was beyond the scope of the study. With about a quarter of the doctors in the study being women, it is possible only the best are making it into hospitals. Alternatively, the greater focus on factors like preventative care may indicate that the way American society socializes girls provides a better grounding for practicing medicine than the upbringing given to boys. 

Irrespective of the reasons, the better performance isn't matched when it comes to pay. A study earlier this year showed that male doctors earn 8 percent more than women with the same qualifications and experience.

Tsugawa wanted to make sure male doctors weren't getting the toughest cases, as might be the case if men are more likely to specialize in diseases with poor prognosis. He dealt with this both by only looking at cases where general internists provided the majority of costs, as billed to Medicare, and by controlling for patient factors such as age, primary diagnosis and race. The age of the physician and medical school from which they graduated were also tested for. Irrespective of the factors considered, patients were still better off with a woman doctor, although the size of benefit varied.