For four years, Jed Meltzer studied communication disorders at the National Institutes of Health, using brain-imaging technology to pinpoint the impact of strokes on speech. His postdoctoral training, he wrote on his blog, comprised “some of the most scientifically satisfying years of my life.

“I got to collect amazing, irreplaceable data, and I got to learn from the best and work with unparalleled resources. Most importantly, I got to publish several papers that established my scientific reputation and positioned me to move into a faculty position in 2010.”

But now that data is useless for Meltzer and about a dozen other scientists caught in a dispute that is unusually fierce, even for the highly competitive world of elite biomedical research.

The leadership at the National Institute on Deafness and Other Communication Disorders, where Meltzer worked, has banned the use of data collected over 25 years from more than 1,000 volunteers in the lab of neurologist Allen R. Braun, citing “serious and widespread” record-keeping errors, all of them clerical matters related to forms used for matters such as screening volunteers or logging physical exams.

But there have been no allegations that data was altered, plagiarized or fabricated, and no one’s safety was threatened — the kind of misconduct that usually leads to such severe penalties in scientific research. Many people say the harsh punishment stems, instead, from a long-standing conflict at the institute, whose leadership has forced numerousscientists like Braun to leave in recent years.

Critics contend that millions of dollars’ worth of research has been squandered at a time when NIH faces the prospect of sharp budget cuts from the Trump administration.

The penalty is “absolutely bizarre,” said David Poeppel, a professor of psychology and neural science at New York University who has followed the controversy in his field. “It’s actually unheard of. It’s also unclear who’s being served by that. Certainly not the taxpayer.”

NIDCD Director James Battey and other leaders of the 29-year-old institute — one of the smallest parts of NIH — declined to comment. In letters to Meltzer and others, however, Battey contended that a February 2016 audit conducted by a contractor hired by NIDCD concluded that the work in Braun’s lab was “irretrievably compromised and we felt that the only course was to close” the studies.

But those affected say the audit widely mischaracterized the procedures in Braun’s lab.

Violations in Braun’s lab were “like a low-grade fever,” said Nan Bernstein Ratner, a professor in the University of Maryland’s Department of Hearing and Speech, who has worked with some of the affected researchers and has lost data that might have been used in two papers. “We are not talking about something that compromises the publishability of the data. It doesn’t impact the interpretation of the data. It doesn’t impact the veracity of the data.”

Braun, 71, studied language and communication disorders. He and other researchers used imaging technology in an attempt to determine the brain’s role in aphasia, stuttering and other conditions that affect communication.

He was at least the sixth scientist forced out since Andrew J. Griffith took over in 2009 as science director of NIDCD, which has just 16 laboratories, according to people familiar with the conflict.

Braun was forced to retire in June 2016, just one day before an “institutional review board,” which oversees research conduct at several institutes, was scheduled to issue a decision that might have allowed researchers to continue their work. Griffith’s alleged interference with the oversight process is one of Braun’s many complaints against NIDCD.

“The NIH has procedures in place for resolving allegations involving its scientists. That Dr. Griffith could apparently short-circuit these procedures and unilaterally try to terminate the career of a distinguished researcher is nothing short of vindictive and should not have been tolerated,” said George M. Chuzi, Braun’s attorney.

Some of the scientists have filed complaints with the U.S. Equal Employment Opportunity Commission that allege various forms of discrimination, according to interviews and documents reviewed by The Washington Post. In one case, a researcher reached a legal settlement after filing an EEO complaint. Others did not fight their departures.

After Griffith closed the lab of Konrad Noben-Trauth in 2011, he required the tenured scientist to sit in an office with the door open each day so that administrators could check on his whereabouts, according to numerous accounts from people aware of the punishment. Noben-Trauth also had to check in and out with Griffith’s office any time he left the building, they said.

Noben-Trauth, 57, finally quit and went to law school. He graduated, but before he could begin a new job, he was diagnosed with terminal brain cancer.

An NIH spokeswoman said that other NIDCD and NIH leaders, including Griffith; Carter Van Waes, NIDCD’s clinical director; and Michael Gottesman, NIH’s deputy director of intramural research, could not comment on the various complaints. They would not address Braun’s case because it “is in litigation,” an apparent reference to the 2016 complaint he has filed with the federal Merit Systems Protection Board.

In letters to Meltzer and Ratner, NIDCD has made clear that it considers the matter closed. The institute released a statement to The Washington Post on Saturday that emphasized the importance of rigorous record-keeping and data collection in studies that involve human subjects.

Van Waes said in the statement that if researchers “can demonstrate that all of the data for all of the participants was collected in compliance with the approved protocol, then the NIDCD will consider additional requests for publication based on the data.”

Meltzer said that offer was made previously but that the standard is impossible to meet, because inconsequential records from years ago may no longer be available. Ratner also said it is unlikely to change anything.

“They’re not going to allow anyone to publish anything,” Meltzer said.

Braun and Noben-Trauth also declined to comment.

The scientist who almost certainly has lost the most to the publication ban works in the tiny field of stuttering research, most of which is funded by NIH. He lost data that would have been used in about 15 papers, according to people familiar with his work. The scientist declined to discuss his situation.

“They’ve essentially vaporized five years of his life,” Ratner said. “The irony of doing that is that these were the five years promised him to build his career. That’s what the NIH postdoc program is.”

The researchers have repeatedly asked NIDCD to consider alternatives to the publication ban, including transferring custody of the information to another academic institution or allowing peer-reviewed journals to decide whether the results were contaminated. (The NIDCD did not order that articles already published be retracted.)

They have complained about the wasted effort of volunteers with strokes, stutters and other disorders — as well as comparison volunteers without hearing or speech conditions — who took part in the tests. Meltzer estimated that more than 1,000 volunteers had taken part in various studies over the years, many of which have been published.

“It’s an enormous waste of taxpayer dollars to conduct 25 years of research and flush it all down the toilet,” Meltzer said.

The battle over publication of the research was first reported by Science magazine.

Before researchers at NIH or universities can launch studies that involve human subjects, they must spell out how they intend to conduct their work. These written plans include everything from the specific language in consent forms to the ways researchers will protect volunteers. Once approved by an institutional review board, the formal plans become known as “protocols.” Numerous researchers can conduct a variety of studies under one protocol.

Braun’s plan was first approved in 1992, leading to years of published research. It was amended from time to time over the years.

In 2013, outside experts were brought in to evaluate Braun’s work, a periodic review that all researchers undergo. The panel, known as a Board of Scientific Counselors, gave Braun an outstanding review — a score of 2 on a descending scale of 1 to 9 — and recommended he receive an additional staffer.

Instead, the final report was changed by NIDCD, and Braun’s resources were slashed, according to Poeppel, one of the experts who conducted the review.

“We’re more than a little bit annoyed to do the work and then be summarized as saying something completely different,” he said in an interview. “When you give someone a score of 2, it’s incompatible with saying ‘and your research program should be cut.’ It’s just not logical. Honestly, don’t waste my time.”

In March 2015, a new volunteer in Braun’s study was mistaken for another volunteer with the same name and underwent an MRI, according to documents related to the case obtained by The Post. No harm came to the volunteer; in fact MRIs are considered almost no threat to anyone in good health.

Braun reported the mistake to Van Waes. His lab was closed the next month and its work suspended.

Braun launched his own review and found 107 instances in which medical histories and physical exams could not be located for volunteer control subjects — the healthy people who are used for comparison to those who have hearing or language disorders. However, all but two had been given those tests during the previous two years when they volunteered for other experiments at NIH. And every person underwent additional safety screening before having an MRI.

In May 2016, the institutional review board found “serious, continuing noncompliance” in Braun’s record-keeping and agreed with the suspension of his research. But it said the lab might be able to resume its work if Braun submitted an acceptable remedial plan.

He did so. But NIDCD forced Braun to retire in June 2016, a day before the review board was scheduled to consider his plan.

Those who lost the data for their research remain in limbo.

“Those people are looking at black holes in their [resumes],” Ratner said. “They were employed to do something and that something was to do research and publish it. And they can’t do that.”