When I was 15 and going through cancer treatment at the NIH, having my blood drawn was a regular and miserable part of the routine. After a while I established a single rule that eventually became inviolable: Never let a doctor draw my blood.

I had spent too many minutes with physicians searching fruitlessly for my skinny, scarred-over veins in my skinny teenaged arms. Doctors were as sure as science when it came to predicting where the few viable vessels were, leaving a groundhog’s trail of holes from my elbows to my wrists as they disproved themselves, time and again.

The blood would be found, inevitably. By nurses. Phlebotomy is a subtle hunt, guided as much by instinct as textbook biology. The oncology nurses, well schooled in patience, brought ritual taps and skin caresses to the art. They, too, had medical training but came to the task with a soft knowing, not a certainty. It worked.

Later I wondered if I had miscast my judgment as a distinction between castes (doctor vs. nurse) rather than genders. The NIH clinic doctors, way back then, were mostly men; the nurses on the 8th floor of Building 10 were mostly women.

I recalled this decades-old memory when I read an investigation published online yesterday in JAMA Internal Medicine entitled, “Comparison of Hospital Mortality and Readmission Rates for Medicare Patients Treated by Male vs. Female Physicians.” The study, by Yusuke Tsugawa and colleagues at the Harvard T. H. Chan School of Public Health, Harvard Medical School, and other institutions, examined the record of a large random sample of Medicare patients, 65 years or older, who were hospitalized from January 2011 to December 2014.

All told, the researchers reviewed 1,583,028 separate incidents of hospitalization and 1,540,797 occasions when patients were sent back to the hospital within 30 days of their initial discharge. What they found was just plain striking: Patients treated by female internists had lower rates of mortality (within 30 days of hospitalization) and lower rates of readmission than those treated by male physicians.

That was true using various statistical tests and controlling for any number of variables, from length of patient stay to the number of years the physician had been practicing. It was also true, importantly, regardless of the primary condition the patient had going into the hospital. The researchers examined outcomes for eight common medical conditions—from sepsis, pneumonia, and congestive heart failure to renal failure and GI bleeding. In each case, those treated by female physicians had lower mortality and readmission rates, though the differences were relatively modest all around.

The Harvard study joins a growing collection of investigations in the academic literature that attempt to determine whether a physician’s gender plays a role in health outcomes. Some studies suggest it does; some the opposite. When it comes to diabetes care, for example, this German research team found “female physicians provide an overall better quality of care”; this U.S.-based team found “patients of female physicians received similar quality of care compared with patients of male physicians.”

Ashish Jha, the senior author of the JAMA Internal Medicine paper, points out in a blog post that there are reasons why women doctors might well deliver better care, citing research to suggest that they might adhere more closely to established clinical practice guidelines and that patients often report better personal experiences (something that ultimately might lead to better outcomes). But he, too, acknowledges that the evidence is both thin and anecdotal—or at least it was until the study he and his colleagues published yesterday.

Still, there is a rich vein of understanding to be drawn from continuing to ask the question. Researchers Debra Roter and Judith Hall have suggested that differences in patient communication and attentiveness may play a role: In one analysis, in fact, female physicians spent two minutes (10%) longer with patients than male doctors did.

One has to wonder, could the answer be as simple as that? Could patience, in other words, be the secret to helping more patients?