Maryland’s longtime medical examiner wouldn’t call George Floyd’s death a homicide. It has triggered a reckoning.

BALTIMORE — As the trial approached for Derek Chauvin in Minneapolis, the pathologist Dr. Emily Duncanson wondered who would testify in defense of the ex-cop.

The world had watched cellphone video of Chauvin kneeling on George Floyd’s neck. What doctor wouldn’t call that homicide?

Duncanson expected one of the celebrity doctors known more for TV appearances than medical opinions. The name on the witness list surprised her: Dr. David R. Fowler.

“Dr. Fowler was not one of those,” she said. “He was very well respected.”

Indeed, the retired chief medical examiner for Maryland was highly regarded in the small world of forensic pathologists. Maryland’s office ranks among the leading institutions in the specialty field; Fowler ran it for nearly two decades. He’s published more than 100 scientific journal articles and lectured around the world on everything from the risk of crib bumpers to homicides by kitchen knife.

Though defense attorneys challenged Fowler’s rulings over the years, his peers elected him president of the National Association of Medical Examiners in 2015. Then he served as chairman of the board.

With one day of testimony, his reputation was in tatters.

“Everyone who I was talking to was taken aback by how baseless his testimony was,” said Duncanson, of St. Paul, Minnesota. “Like, ‘Have we ruled out aliens?’ I mean, it was so far out there.”

In Baltimore, Dr. Joshua Sharfstein, the former Maryland health secretary and city health commissioner, watched with confusion.

“It’s hard for me to square the Dr. Fowler who I worked with on child mortality issues with what I saw during the trial,” Sharfstein said.

The Minnesota medical examiner, Dr. Andrew Baker, performed the autopsy of Floyd and found severe heart disease. During the trial, he testified that Chauvin’s knee “was just more than Mr. Floyd could take.” His conclusion: “homicide.”

On the witness stand, Fowler blamed a mixture of Floyd’s bad heart and restraint by police, all of it worsened by drugs in the man’s system. Fowler testified that the tailpipe of the police car may have leaked carbon monoxide into Floyd’s face and contributed to his death.

When asked to classify the manner of Floyd’s death, Fowler said he would “fall back to ‘undetermined.’”

That set off a firestorm.

Commenters branded Fowler a cop apologist, a racist, a quack scientist. A former forensic examiner of Washington, D.C., circulated a petition asking authorities to investigate Fowler’s old cases. Maryland Attorney General Brian Frosh agreed to take up the matter and assemble a review panel.

Seldom heard was the opinion of medical examiners who did not find Fowler’s conclusion outlandish. Some told The Baltimore Sun his testimony was grounded in science, not bad faith. Forensic pathologists often reach different conclusions on the same evidence, some observers noted.

“Medicine is more of an art than a science,” said Dr. Randall Frost, a former chief medical examiner in San Antonio. “If you ask 10 forensic pathologists about a cause of death on anything other than a pretty straightforward case, you’re going to get 10 different answers.”

In short, it’s complicated. And considering the history of the Maryland Medical Examiner’s Office, Fowler’s testimony was no surprise. The office has classified deaths “undetermined” for decades — more so than just about anyplace else.

Only this time, the world was watching.

More right than wrong?

Open-air carbon monoxide poisoning doesn’t seem far-fetched to Dr. James Claude Upshaw Downs, a forensic pathologist in Georgia. He recalled the death of a 12-year-old boy who inhaled carbon monoxide while riding on the back of a boat through Mobile Bay.

“It’s not uncommon, because of the nature of what we do, that sometimes people aren’t happy with the results,” he said.

Fowler’s testimony in the Chauvin case represented the consensus opinion of 14 medical specialists — in pathology, pulmonology, toxicology, cardiology, emergency medicine — recruited by The Forensic Panel, a New York consulting firm. It was hired by Chauvin’s defense attorneys to review the evidence for trial.

Prosecutors provided the defense with test results showing just insignificant levels of carbon monoxide in Floyd’s system — but did so only after Fowler’s testimony, said Dr. Michael Welner, chairman of The Forensic Panel.

In Virginia, former chief medical examiner Dr. Marcella Farinelli Fierro considers Fowler a professional friend.

“He wasn’t out there on a limb,” she said. “He was more right than wrong. You’re dealing with a video, OK. And that video looks awful. But if you get down to the physiology of it and you say to yourself, ‘What’s operational here in terms of physiology?’ there are a lot of questions.”

Still, Fowler’s testimony has become a flashpoint in the debate about bias in forensic pathology. In February, the Journal of Forensic Sciences published a study finding racial bias in death investigations. The authors presented 133 forensic pathologists with the hypothetical case of a toddler with a fractured skull.

About half the doctors were told the toddler was Black and the mother’s boyfriend was the caretaker. For the other half, the toddler was white and the grandmother was the caretaker. Researchers found the doctors five times more likely to rule the Black child’s death a “homicide” than an “accident.”

The results caused a stir. More than 90 forensic examiners, including Fowler, signed a letter to the journal editor, demanding the study be retracted, said Dr. Brian Peterson, the Milwaukee County medical examiner who wrote the letter.

The doctors who signed included Dr. Christopher Milroy, a forensic pathologist in Canada and former chair of the Pathology/Biology Section of the American Academy of Forensic Sciences. He said key scenarios in a study must be identical: white child with mother’s boyfriend; Black child with mother’s boyfriend.

“Boyfriends are more likely than grandmothers to kill. Non-biological relations are more likely than biological relations to kill,” Milroy said. “I signed it not because I don’t believe there’s bias; I signed it because it’s a terrible piece of science.”

The controversy was smoldering when Fowler took the witness stand two months later and called the manner of Floyd’s death undetermined.

Critics accused Fowler of obfuscating. Despite the bad heart and drugs, Floyd would be alive but for 9 1/2 minutes of the cop’s knee on his neck. How can that be classified as anything other than homicide?

“Many medical examiners for a long time have been ignoring police-related deaths, looking for the most minute, insignificant findings to explain a death like George Floyd,” said Dr. Cyril Wecht, a nationally recognized forensic pathologist from Pittsburgh.

Alumni of Fowler’s university in South Africa petitioned administrators to condemn the doctor’s testimony and censure him. University of Cape Town faculty issued a statement saying they support a review of his old cases.

Fowler’s testimony led to scathing comments online.

In a racially charged murder case, the people asked, who trusts a white doctor trained in apartheid South Africa?

Reputation for independence

The U.S. has a patchwork of state and local offices responsible for death investigations. In rural Texas, for example, elected justices of the peace fill out death certificates after a few weeks of training.

Maryland is among a handful of states without county offices, but one statewide medical examiner. The office investigates deaths from the Eastern Shore to Western Maryland. The chief is a board-certified forensic pathologist.

A native of the African land once called Rhodesia, Fowler attended medical school in South Africa and came to Baltimore in 1991 for a residency at the University of Maryland Medical Center. He completed a two-year fellowship in the medical examiner’s office and worked his way up to deputy chief.

In 2001, Chief Medical Examiner Dr. John Smialek died suddenly in his office. The oversight board promoted Fowler to succeed him. He would go on to serve as Maryland’s chief medical examiner until stepping down in 2019.

Fowler took over one of the busiest offices in the country. At that time, staff performed more than 4,000 autopsies a year, according to a history written by Bruce Goldfarb, the office spokesman. The count has since climbed to more than 5,600 autopsies a year, according to office reports.

Goldfarb and the current chief, Dr. Victor W. Weedn, declined to comment for this article. Fowler declined, too.

Fowler became well known as a researcher and lecturer. He co-authored a study in the Journal of Forensic Sciences that found 202 people died in the custody of police, prisons and psychiatric hospitals in Maryland from the late 1930s to the early 2000s.

His later research into the suffocation hazards for infants helped persuade state lawmakers to ban crib bumpers. Fowler, meanwhile, had a reputation for being fiercely independent and insulating his staff from interference by government officials.

In September 2010, the state opened a new $50 million, 120,000-square-foot medical examiner’s office on Baltimore Street — one of the largest, free-standing forensic medical centers in the U.S.

Medical students coveted a residency there. About 10% of all medical examiners working in the U.S. trained in Baltimore, according to the office.

“They often get the cream of the crop,” said Dr. James Gill, Connecticut’s chief medical examiner and president of the national association. “There are only a handful of offices around the country that do that amount of training and research.”

When investigating a death, medical examiners may conduct interviews, visit crime scenes and perform autopsies. They want to determine the cause of death and classify the manner, whether homicide, natural, accidental, suicide or undetermined.

Most offices can’t determine a manner in fewer than 5% of cases, Gill said. New Mexico, Connecticut, D.C. and Los Angeles settled for “undetermined” in less than 2% of cases, according to office reports there.

Maryland medical examiners classified 13% of deaths “undetermined” in 2018, the most recent year available.

“We’ve always kind of had a mandate that you want to, to the best of your ability, answer questions to the way a person died. If you have a large number of ‘undetermined,’ that’s not helping anybody,” said Dr. Joye Carter, a former chief medical examiner of Washington, D.C., Indianapolis and Houston. She consulted with the prosecution in Chauvin’s case.

The Maryland tendency predates Fowler, but he’s explained the rationale, particularly in cases of drug overdoses. Medical examiners usually don’t know whether the deceased deliberately injected a deadly amount of heroin to commit suicide, accidentally took a bad dose, or had someone else inject the drugs. Without the facts, they settle on undetermined.

Drug overdoses caused nearly 90% of deaths classified “undetermined” in the office’s annual report for 2018, the most recent available. Excluding these deaths, the rate of “undetermined” cases drops from 13% to 1.3%.

The method is “intellectually honest,” Fowler has said. In police killings, however, the ruling has big implications.

The national association issued guidelines three years ago to classify in-custody deaths because of inconsistency among states.

The deceased may have drugs in his or her system, causing some medical examiners to classify “accidental.” Others consider the mix of factors — drugs, police restraint, underlying health conditions — and settle for “undetermined” rather than elevating one factor. The association recommends “homicide” for “death at the hands of another.”

That’s how Fowler’s office classified the death of Freddie Gray in 2015, concluding he suffered a “high-energy injury” to his neck and spine. The ruling presented Baltimore State’s Attorney Marilyn Mosby with probable cause to file criminal charges against the officers.

The office ruled “homicide” in the death of Anthony Anderson two years earlier. He was tackled by police in a vacant lot in East Baltimore and died of internal bleeding. That time, however, State’s Attorney Gregg Bernstein chose not to prosecute the cops.

The Office of the Chief Medical Examiner ruled “undetermined” in the death of Tyrone West in 2013. He died after a traffic stop and violent struggle with police, according to witnesses and officers. Medical examiners concluded his death was caused by a heart condition complicated by dehydration and police restraint. No officers were charged.

Medical examiners ruled the 2016 death of Tawon Boyd “accidental.” He died after a fight with Baltimore County police. The examiners listed the cause of his death as organ failure brought on by his use of drugs known as bath salts. None were charged.

In both cases, attorney A. Dwight Pettit represented the families and sued police. He hired private forensic investigators who in each case reached a different conclusion than the medical examiners: chiefly, asphyxiation caused by police restraint. City officials settled the lawsuit over West’s death for $1 million; county officials settled with Boyd’s family for $1.15 million.

The fiercest criticism of the Maryland office during Fowler’s tenure came after the death of Anton Black in 2018. His family has filed a federal lawsuit accusing the office of misrepresenting the teen’s death to protect police.

Black died after a struggle with officers who pinned him under their bodies. Medical examiners concluded he suffered “sudden cardiac death.”

They classified his death as an “accident.”

Calls for a review

Fowler’s testimony last month in Minneapolis renewed attention to the Maryland cases. Former Washington, D.C., medical examiner Dr. Roger Mitchell circulated a letter calling for authorities to review all deaths in custody during Fowler’s tenure. Mitchell serves as chair of the pathology department at Howard University.

“Dr. Fowler’s stated opinion that George Floyd’s death during active police restraint should be certified with an ‘undetermined’ manner is outside the standard practice and conventions for investigating and certification of in-custody deaths,” Mitchell wrote.

Mitchell said 431 other doctors signed his letter online. But the men and women self-reported their professions. About 25 called themselves forensic pathologists. The list included names such as “Poop Marmelade” and the infamous Nazi physician Josef Mengele.

Many medical examiners say they have an important duty to serve as expert witnesses for the defense. They make sure prosecutors prove a case beyond a reasonable doubt. In San Antonio, Frost worries the backlash against Fowler has sent a chill through the profession. Will doctors feel pressure to rule deaths in custody as “homicides?” he asked.

And who will be willing to testify in the next police killing?

———

(Baltimore Sun librarian Paul McCardell contributed to this report.)