Elderly hospitalized patients treated by doctors who graduated from non-US medical schools are slightly less likely to die within 30 days than those treated by graduates of US medical schools, according to a study published Thursday.

The study arrives amid the furor over President Trump’s 90-day ban on people from seven Muslim-majority countries entering the United States — an order that prompted concerns that it would block some foreign medical students from training and practicing here. About 1 in 4 US physicians were born abroad, and among internists, the primary care doctors who were the focus of the study, 44 percent graduated from foreign medical schools.

The journal where it was published, the BMJ, said the timing is a coincidence. The researchers launched the study a year ago and submitted it in September, and a BMJ spokeswoman said it “was always scheduled to go into next week’s print journal.” (Papers run online ahead of print.)


Dr. Ashish Jha of Harvard T.H. Chan School of Public Health and his colleagues found that 11.2 percent of Medicare patients cared for by foreign graduates died within 30 days of being admitted to a hospital, compared with 11.6 percent of those cared for by US graduates.

America “attracts the very best and the very brightest,” Jha said. “Certainly that’s true in science, and our paper says that’s true in medicine as well.” But Trump’s entry ban “sends a signal that America is not going to be as open anymore,” he added. “The ones who will lose will be American patients.”

The researchers analyzed data on 1.2 million hospital admissions of Medicare patients aged 65 and over between 2011 and 2014 and for 44,227 internists. The average age of patients was 80, and the most common causes of death were sepsis, pneumonia, congestive heart failure, and chronic obstructive pulmonary disease.

This is the first nationwide study to compare the patient outcomes of US- and foreign-educated doctors; previous research has compared test scores or adherence to quality measures. The researchers excluded from their analysis physicians who graduated from US colleges but attended medical schools in the Caribbean or Central America, which usually happens because they were not accepted by US medical schools. Those graduates “are known to be less qualified and would have biased our analysis,” said the study’s lead author, Dr. Yusuke Tsugawa, who graduated from a medical school in Japan.

One possible reason for the lower mortality among patients of foreign graduates is that requirements for winning a hospital residency, which is required to practice in the United States, are more rigorous for non-US graduates. “That might select for physicians who are, on average, more knowledgeable and skilled,” Tsugawa said. Another possible explanation is that foreign graduates have, in the past, edged US graduates on standardized test scores, though it is not clear if that’s still the case.

More likely: Most foreign graduates do two residencies, one in their home country and one in the United States, an “intensive and prolonged training” that might make them better doctors, the researchers wrote.

There’s a cultural factor, too. Because foreign doctors often face prejudice, “they might be extra cautious and ask for more consultations so as to avoid mistakes,” Tsugawa said.

A strength of the study is that nearly half the doctors were foreign medical graduates, which makes the statistics more solid than if only a small percent were, said Patrick McKnight of George Mason University, an expert in research methodology and statistics.

But a result that is statistically significant — there was less than a 0.1 percent probability that the difference the researchers found would happen by chance — might not be meaningful in real life, he cautioned. In this case, the mortality difference works out to roughly 1 fewer death per 250 patients treated by foreign-educated internists. That means 2,200 of the 66,250 deaths among the patients treated by US-educated physicians in the study might have been avoided if they were treated by foreign-educated doctors.

Whether that is a significant reduction is in the eye of the beholder: The researchers call it of “modest clinical significance,” comparable to the decline in rates of death that hospitals achieved from 2000 to 2010 by implementing quality improvements and better treatments.

Another concern is that the researchers’ focus was narrow. One could also look at one-year mortality, at younger patients, at physicians in other specialties, and many more. “Any time I see an outcome specified so narrowly I wonder what others might look like,” McKnight said. That is, it’s not clear whether the reported mortality edge for foreign-educated internists applies to other measures and other kinds of patients and doctors, too, or is an outlier.

“We don’t know whether the results will generalize,” said Michael Lavine, professor of statistics at the University of Massachusetts, Amherst.

study last year suggests the internists’ patient outcomes are not an anomaly, however. When researchers led by Dr. Rachel Kelz of the University of Pennsylvania compared outcomes of 972,718 operations by 4,581 surgeons, they found that the rates of mortality in patients operated on by foreign medical school graduates and US graduates were almost identical: 2.1 percent and 2.2 percent, respectively. So were rates of complications: 14.3 percent vs. 14.5 percent. US-educated surgeons performed a higher number of complex procedures, however: 13.7 percent vs. 11.1 percent.

Outside experts generally praised the BMJ study for looking at many factors that differed between foreign and US graduates and might affect patient mortality: length of hospital stays, how much care patients received (a measure of how sick they were), how many chronic conditions they had, and what kind of hospitals (teaching or non-teaching) doctors worked in, among others.

The foreign graduates’ patients were more likely to be non-white, poor, covered by Medicaid, and sicker with congestive heart failure, chronic obstructive pulmonary disease, and diabetes. All of those factors increase the chance of dying and so should have given the US graduates an edge. But the mortality difference in favor of foreign-educated internists held up after accounting for the confounding factors.

The study, said Dr. Andrew Gurman, president of the American Medical Association, “provide[s] additional reassurance about the high quality of care that international medical graduates deliver in the United States.”

This story was updated on Feb. 3 with comment from the American Medical Association.

Leave a Comment

Please enter your name.
Please enter a comment.

  • The US public education system is so under funded it’s no laughing matter! Then add all wasted education money and time spent on useless football and other sports. It’s pretty clear that our students are way behind India, China, and especially Western Europe.

    The best and the brightest and hardest working are absolutely not white Americans – across the board.

    America will die because of white nationalism and Trump and the Republicans!

    Germany will be the new leader of the free world!

  • It’s not only about the recent withhold/ban. Trump has already signaled to reduce H1B visas, which allow non US-IMGs to join medical residencies and work afterwards. The J1 visa opens doors to residency, but imposes a 2 year period of living outside US, which can be bypassed by a waiver given to those going to work in medical underserved areas under H1B. Holding back H1Bs will impact US medical service in both ends. There are not enough US graduates and US-IMGs to fulfill all residency spots and even less are willing to work where they’re needed the most. Medical knowledge is widespread and the greatest difference between US and other countries lies on structure and resources, not on clinical knowledge.

  • I wish the media would stop spreading the stigma that practicing US-IMG’s are somehow “less qualified.”

    Findings from previous articles, including Norcini 2010 (http://content.healthaffairs.org/content/29/8/1461.long), state there is no significant mortality difference between all IMGs and their US counterparts. There have been similar findings for IMGs in Canada and their Canadian counterparts.

    The article itself states, “Approximately 73% of US citizens who are international medical graduates attend medical schools in Central America and the Caribbean. Excluding physicians from these countries in our sensitivity analysis did not affect our findings.”

    I understand the focus of this article is on foreign-trained medical graduates, but practicing US IMGs should not be openly criticized without recent, peer-reviewed evidence.

  • I would say that a 0.4% difference is well within noise/variability.
    On the political aspect: aside from the moronic actions of the new government with complete disregard for the emotional, economic damage caused and the obvious illegality of those actions, I am pretty confident that the number of MDs/scientists affected by the ban is pretty much irrelevant in the contest of the whole US healthcare/scientific system. So, I do not think that overblowing things, and providing “alternative” facts like the other side does, is the way to go

    • You have to look at the data to determine if 0.4% difference is noise or is statistically significant. Depending on how much sample data is taken and standard deviation there is 0.4% can easily be statistically significant, I haven’t read the paper but I would assume is passed the <0.05 P value test

    • Dear, you should know as well that a p value is an arbitrary threshold, and 0.05 just happen to be considered an acceptable confidence value, not an absolute statement of significance (assuming that the statistical analysis was correct, and by experience I can say it is very often not the case). In particular, being the study based on observational reported data and not on a experimental lab work under controlled conditions, I feel fairly/absolutely confident that the difference is within noise. cheers

  • They should not have published this. The reason is because if the study had turned out differently and showed that foreign-educated doctors have worse outcomes, they would have buried the result out of concern for being racist.

    • Sorry to be late to this exchange, but your comment reminded me of an incident from about 25 years ago, when my division chief noted that all of our residents in medicine were FMGs, and he wondered if there would be communication problems with patients. A reasonable question, but our IRB refused to approve a formal study of the issue, and refused to offer any reason for their refusal. Turned out that most of the IRB members were themselves FMGs! We worked on the issue informally and found no problems whatsoever, but we were not able to publicize those results, since our IRB had nixed the project….

  • A moratorium was placed on immigrants from7 nations that the Obama administration indicated the governments were unstable and reliable information was not available for the immigrants. Comey said we have no way of knowing who these people are. Does anyone remember the doctors in London who drove a car into the airport.
    Can’t we wait 3 months to see if we can vet these people?

  • It seems to me that the true losers are the people in the countries these bright MD’s are leaving behind. A man I know who comes from Haiti told me “We don’t want to be here. If you really want to help me, help change things in Haiti.” Just a less selfish point of view.

  • First, this story starts out with a non-sequitur in that it states foreign educated doctors have a lower rate of patient mortality than Dr’s trained here but then the author worries that less students may come here for training. Second I’d bet my house that of all the foreign trained Dr’s coming here statistically zero came from Syria, the other 6 countries on the list are not on a ban but a temporary hold.

    • Good thing you didn’t bet your house, or you’d be looking for new digs – if you follow one of the links in the article (https://www.statnews.com/2017/02/02/match-day-trump-medical-students/)
      “Grover said the AAMC has identified more than 1,000 non-US citizens applying for residency programs who are from one of the seven countries or who have listed contact or family addresses there. Many of them have visas or green cards and are already in the United States…”

Sign up for our Daily Recap newsletter

A roundup of STAT’s top stories of the day in science and medicine

Privacy Policy