Third year medical students at the University of Miami attend a class where they work with Harvey, the cardiopulmonary patient simulator. (Joe Raedle/Getty Images)

Some first-year doctors are working 30 hours in a row at dozens of hospitals around the country in a test of work-hour limits that were imposed in 2011 because of fears that inexperienced, sleep-deprived physicians might jeopardize patients.

The 30-hour shifts, which were banned four years ago, are one element of a $9 million research project partly funded by the National Institutes of Health to determine the best way to train novice physicians while maintaining patient care.

The arrangement is reviving one of the oldest debates in the education of physicians: whether they learn best when subjected to brutally long and difficult working conditions, whether their patients have a right to know about it, and whether physicians and patients can be kept safe.

Critics of the study, some young doctors among them, argue that exhausted physicians can make mistakes that cost lives. But the researchers and some experts say the work limits force more handoffs of patients from physician to physician, which are more dangerous than leaving them in the care of sleep-deprived doctors.

To change the doctors’ work hours, the researchers obtained permission from the organization that regulates physician training. And because it involves actual patients, the study also needed approval from the ethics panel at the University of Pennsylvania, which is leading the research. That panel deemed the study’s risk to patients and new doctors minimal, and said hospitals did not have to inform either.

David Harari and Jeff Clark, first-year residents in psychiatry at the University of Washington School of Medicine, said that the risk is more than minimal. They have told the school’s ethics panel they were not informed that they could be working as much as 30 ­­­hours at a time when they interviewed for their jobs and that their patients are unaware of it.

(Psychiatry residents study internal medicine during their first post-graduate year. A University of Washington official said internal medicine residents were notified about the trial, but psychiatry residents were not.)

“It’s really hard to function for 30 hours in a row,” said Harari, who was interviewed after a 30-hour shift during which he was responsible for six patients and got about 2  1/2 hours of sleep. “I’m not aware that I’ve made a mistake yet, and I hope that never happens.”

Sidney M. Wolfe, co-founder of Public Citizen’s Health Research Group, which in 2001 tried to persuade the federal government to regulate new doctors’ work hours, agreed there are risks in longer hours. “From my perspective, it seems both patients and doctors have to be involved [in consenting], and both patients and doctors need to be able to opt out.” He said that is unlikely or impossible because of the way this study is designed.

But Mildred Solomon, president of the Hastings Center, a nonpartisan research institute that studies ethical questions in health care, strongly endorsed the research. “We haven’t really studied whether [restrictions on work hours] made a difference or not, not in this kind of rigorous way,” she said. “And I think we need to find out.”

Long confined to academic medical circles, the issue burst into public view after the 1984 death of Libby Zion, the 18-year-old daughter of influential New York lawyer and journalist Sidney Zion. Libby Zion was being treated for an undiagnosed problem by an overworked first-year doctor at a New York hospital when she went into cardiac arrest and died.

A grand jury took up the case but declined to issue criminal charges. New York state eventually passed a law limiting residents’ work hours. As a result, no New York hospital can participate in the current study.

In 2011, after a study by the prestigious Institute of Medicine, the Accreditation Council for Graduate Medical Education banned 30-hour work shifts for all trainees and prohibited first-year physicians from working more than 16 hours consecutively, with a few minor exceptions.

But some involved in training newly graduated physicians — known as “residents” at hospitals — say the work-hour limits have made staffing needlessly difficult. They said they believe that a novice learns best by following a case in the critical 36 hours after a patient is first admitted and that patients are better served, as well.

“Duty hour rules were born out of a concern for patient safety. . . . But the pushback is they have greatly increased handoffs,” said David Asch, director of the Center for Health Care Innovation at the University of Pennsylvania’s school of medicine and the study’s lead researcher. “And we already know that handoffs affect patient safety.”

The study involves 6,000 internal medicine trainees at 63 programs across the country. About 2,400 are first-year physicians. The rest are second- and third-year doctors. It began July 1 with the incoming crop of residents, with results expected in 2019. It is similar to a year-long study of longer work days for new surgeons that concluded in June.

In the current study, 31 programs will stick to current rules on resident work hours. Along with the 16-hour limit for first-year trainees, second- and third-year doctors are restricted to 24 consecutive hours of work. Residents may not work more than 80 hours a week, must have one day off every seven days and may not work overnight more often than every three days — all averaged over one month.

In the other 32 programs, directors who schedule residents’ work hours must observe most of those rules. But they can allow the young doctors to work as long as 30 hours consecutively or, in rare cases, even more.

The “minimal risk” designation by the University of Pennsylvania’s Institutional Review Board was critical to the decision to waive the usual requirement to inform patients and doctors that they would be participating in an experiment. “We did a pretty intensive review just because the design of the study is a little bit different,” said Tracy Ziolek, the director for human research protection for the university panel.

She said she also determined that it would be impractical to notify every internal medicine patient at all the participating hospitals.

Adam Zion, Libby Zion’s brother, said in an interview that he views any attempt to lengthen young doctors’ hours with caution. “Let’s see what their data shows on the back end of it all. The simple reality of it is that there continues to be significant problems” with patient safety for many reasons, he said.

Long before the study’s results are available, data from the 2014 trial among 4,400 surgical residents at 119 programs will shed light on these issues. The data will not be available until February, but Karl Bilimoria, the principal investigator, said some residents in general surgery have worked 30 hours or more to finish long operations and stabilize patients. General surgeons train for seven years.

Bilimoria, director of the Surgical Outcomes and Quality Improvement Center at Northwestern Feinberg School of Medicine, said that if the study shows no impact on patient health and safety, his group will ask the accreditation association to loosen its work-hour restrictions.