Editor's note: This AJC investigation was originally published on Oct. 3, 2015, and was recently named a finalist for the Atlanta Press Club’s 2015 Awards of Excellence. Read updates on the subjects in this story here.
Dr. Kris Sperry took the witness stand, exuding the full authority and credibility of the state of Georgia.
Without hesitation, the chief medical examiner testified that Henry Glover died from a bullet to the back, fired by a high-powered rifle. “Any competent forensic pathologist,” Sperry said, would see the evidence the same way.
But Sperry hadn’t examined Glover’s body. He hadn’t studied the bullet, because none was found. And his opinion, like a surprising number of others he presents in court, was far from unanimous.
Sperry wasn’t even testifying in Georgia. On Aug. 29, 2013, he was in New Orleans, appearing as an expert witness against a former police officer accused of murder after Hurricane Katrina. For stating his opinion that day, Sperry earned a fee of $5,000.
It was one of more than 500 cases since 2003 in which Sperry acted as a paid forensic consultant — all while employed full time by the Georgia Bureau of Investigation.
Sperry’s role as expert-for-hire doubles his $184,000 state salary and often takes him out of the medical examiner’s office at GBI headquarters. It also exposes him to conflicts of interest and, at times, undermines his medical and scientific judgment, an investigation by The Atlanta Journal-Constitution found.
The newspaper examined court filings, depositions and trial transcripts from more than five dozen cases. Time after time, lawyers and other adversaries accuse Sperry of tailoring conclusions to suit his paying customers.
“He’s a hired gun,” said Rick Simmons, the defense attorney in the New Orleans case.
“It’s about money,” said George McGriff, another lawyer who challenged Sperry.
Sperry, 60, the chief medical examiner since 1997, oversees investigations into thousands of deaths each year: homicides and suicides, as well as those from accidents or natural causes. An opinion from Sperry or one of his 13 deputies can have profound consequences. Whether a killer faces charges or whether an insurance company pays a deceased person’s beneficiaries may hinge on the medical examiners’ conclusions.
Sperry, though, gives the impression of a detached, somewhat eccentric scientist lost in his work. He indulges an academic fascination with tattoos and sports facial hair invariably described as walrus-like. And yet he is so aware of his status as an expert witness that he can immediately cite how many times he has testified in court (704 on Oct. 31 last year, for instance).
Sperry is “a doctor of national reputation and accomplishment,” said his boss, GBI Director Vernon Keenan. “He operates on an extremely high plane of expertise.”
Sperry declined to be interviewed.
In a memo to Keenan about the Journal-Constitution’s inquiry, Sperry said he remembers few details about his work outside the GBI. When those cases conclude, he told Keenan, he shreds his files.
Keenan dismissed criticism of Sperry as “the back and forth of professionals.”
But in the New Orleans case, for one, four other pathologists attacked Sperry’s conclusions as relying on supposition, not sound forensics. One called his theories on Glover’s death “junk science.”
“Are there people who go out and stretch the truth for the benefit of their private business? Yes,” said Dr. Vincent DiMaio, the longtime medical examiner in San Antonio, Texas, and the author of several influential forensic-science books, who criticized Sperry’s work in New Orleans. “Usually, these are not people who are employed as medical examiners.”
‘A lot of cases’
For reviewing documents and writing reports, Sperry bills his clients $500 an hour. Depositions run at least $1,500. For courtroom testimony, he charges $7,500 a day, (up from $5,000 two years ago), plus travel expenses.
No professional organization or government agency regulates such rates. But interviews with other pathologists suggest Sperry’s fees — like his caseload — rank among the highest in the country.
“Some people make a lot of money because they’re good,” said Dr. Steven Karch, a pathologist in Oakland, California, and a frequent expert witness. “Some make a lot of money because they hustle and do a lot of cases.”
Either way, Sperry’s private caseload rivals that from his state job.
He appeared in court 13 times as the state medical examiner between 2010 and 2014 — and 42 times as a private expert. He performed 208 full autopsies for the medical examiner’s office while accepting 158 outside cases for review.
Sperry is like any other hourly worker in state government, Keenan said: he puts in 40 hours each week, “either actually at work or in a combination of work and leave.”
“After that,” Keenan said, “it’s his free time.”
At times, however, Sperry conducts private business on the public’s time.
The Journal-Constitution examined Sperry’s weekly time sheets for the past five years. On 67 days, Sperry reported working at least eight hours for the state when, according to other documents, he spent time out of the office giving depositions or testifying in court for private clients.
On 13 of those days, Sperry recorded a full day at the GBI but actually was in court out of state.
For example, Sperry testified as an expert witness in Charleston, West Virginia, on Jan. 10, 2013. But his time sheet showed nine hours at his state job: 8:30 a.m. to 6 p.m., with 30 minutes off for lunch.
A GBI review prompted by the Journal-Constitution’s inquiry found numerous inaccuracies in Sperry’s time sheets, Keenan said last week. The agency docked Sperry 226 1/2 hours — about 5 1/2 weeks — of vacation time and other accumulated leave to make up for the discrepancies.
Sperry signed the time sheets beneath a warning that false statements violate the law. The penalty is one to five years in prison.
Keenan said Sperry often filled out the forms late and from memory. “I have no doubt this was a result of sloppy record keeping.”
‘Conflict of interest’
In each deposition and at every trial, one question stands out: Is Sperry an impartial medical examiner, or a paid courtroom advocate?
At least twice, the Journal-Constitution found, Sperry provided clients with opinions that contradicted deputy medical examiners whose findings he had at least implicitly approved.
“That’s procedurally, governmentally, professionally unacceptable,” said Dr. Cyril Wecht, the longtime medical examiner in Pittsburgh. “I’ve never heard of anything like that, ever, ever, ever.”
One case involved Elsie Goedhals, 40, who died shortly after a 14-hour flight from South Africa to Atlanta. Dr. Keith Lehman determined she died of natural causes: a pulmonary embolism resulting from deep vein thrombosis in her leg.
Goedhals’ insurance policy paid only if she died in an accident. Her family sued the insurer, claiming the embolism occurred accidentally because of the long flight. Refuting Lehman’s opinion was critical. So Goedhals’ family hired Lehman’s boss — Sperry — as their expert witness.
Sperry testified that determining the manner of death is “terribly imperfect” and “an opinion situation.”
“It really does depend on the definition of accident,” he said.
A lawyer for the insurance company asked Sperry how often he gets paid to re-evaluate cases that originated in his office.
“It’s very, very rare,” he said.
“Would you consider that a conflict of interest?” the lawyer asked.
“Not unless I was in disagreement with, say, for instance, Dr. Lehman,” Sperry said. “I think he and I are in complete agreement with this.”
In truth, they did not agree.
Asked whether the death could have been accidental, Lehman testified: “I wouldn’t consider it such, based on the criteria we use.”
Sperry told Keenan he recalls nothing about the case.
In February 2005, Sperry completed a report on the death of a jail prisoner in Ocala, Florida. Thomas Duncan, 37, got into a fight with jail officers, who covered his head with a mesh device called a “spit mask” and strapped him into a chair. A doctor said Duncan died after a lack of oxygen caused irreversible brain damage.
Sperry placed no blame on the jail officers. He said Duncan suffered a heart attack because he was “struggling violently and actively resisting.”
Five months later, Sperry finished another report on the death of another prisoner, this one in Stillwater, Oklahoma. Mary Giannetti, 39, got into a fight with jail officers, who restrained her face down on the floor until she stopped breathing.
Sperry’s conclusion: Giannetti’s death was a homicide, caused by “inappropriate restraint procedures.”
In Florida, Sperry was an expert witness for the county sheriff; in Oklahoma, for the dead woman’s family.
As a paid expert, Wecht said, “I’m not bound” to favor a client’s position. “You’ve got to be honest to maintain your credibility.”
An Ohio case in 2013 stretched the limits of Sperry’s credibility.
He was an expert witness for a physician fighting the suspension of his medical license. The doctor had said an elderly patient’s vision was good enough to retain his pilot’s license; in reality, the man was legally blind. A few months later, during a charity event, the man was giving rides in his airplane when, without warning, it crashed. The man died, as did all five passengers.
The National Transportation Safety Board could not determine what caused the crash, but cited a contributing factor: the doctor’s “failure to accurately assess and report the pilot’s visual deficiency.”
Sperry presented an alternate version.
A bad heart, not bad vision, incapacitated the pilot, Sperry testified. The plane crashed, he said, because the passengers couldn’t fly it when the pilot lost consciousness.
Sperry’s opinion drew harsh criticism from a hearing officer for Ohio’s state medical board. He wrote that Sperry had no training in accident reconstruction, did not examine the aircraft, and had no idea what happened in the cockpit. He said he “did not find Dr. Sperry’s testimony credible and, therefore, placed little to no weight on his testimony.”
Sperry’s memo to Keenan said his opinion “had no relationship” to the hearing officer’s decision to uphold the doctor’s suspension.
The same was true, he said, in the New Orleans murder case.
What was left of Henry Glover arrived at the morgue in five red biohazard bags.
Glover, 31, had been burned far beyond recognition in the back seat of a white Chevrolet beside the Mississippi River in New Orleans, straight across from the French Quarter. It was Sept. 2, 2005, four days after Hurricane Katrina struck and the levees broke.
The red bags contained a skull, some body tissue, and a lot of debris.
“Most of it,” said Dr. Dana Troxclair, a medical examiner in New Orleans, “was just charred pieces of bone.”
X-rays revealed what looked like metal embedded in the tissue — bullet fragments, Troxclair guessed. For two hours, she and her supervisor sifted through the remains, but everything crumbled in their fingertips.
A bullet, Troxclair said, would not have deteriorated that much, even in the intense heat of the car fire.
“We came to the conclusion that it was pieces of the car,” she testified. “It could be anything. But we were sure it wasn’t a piece of a projectile.”
Federal prosecutors accused a New Orleans police officer, David Warren, of killing Glover. Convicted in 2010, Warren received a 25-year prison sentence.
An appeals court ordered a new trial, however, and prosecutors called in Sperry to bolster their most damaging assertion: that Warren, armed with a rifle on a second-story balcony like a sniper in a war zone, shot Glover without cause.
Eight years to the day after Katrina hit, Sperry took the witness stand.
As in other cases, Sperry began by reciting his professional experience. As Georgia’s first chief medical examiner, he said, he oversees “all of the homicides and decomposed bodies” and other complicated cases. He claimed particular familiarity with wounds from high-powered rifles because those weapons kill people so often in rural Georgia.
Sperry testified that he reviewed X-rays from the autopsy and four photographs taken before the car was set afire. One picture showed Glover’s body face down in the back seat of the white Chevrolet, with an apparent blood stain on his white T-shirt between his shoulder blades. A larger stain seems to have saturated the right side of the shirt.
The picture, Sperry said, showed that a bullet passed through Glover’s body, back to front — even though his front was not visible.
“At a minimum,” Sperry said, the bullet cut through Glover’s heart, his left lung, and his aorta and other major arteries.
The X-rays, Sperry said, displayed a snowstorm effect of innumerable bullet fragments, appearing white in the reversed image. The “snowstorm,” he testified, “should tell any competent forensic pathologist without any other information that they’re dealing with a high-velocity rifle wound. It’s unique and specific.”
The prosecutor asked whether Glover’s body could have absorbed metal from the car during the fire.
“That concept is preposterous,” Sperry said. “That does not exist in medical science. I mean, in a very simple way, an analogy is if … you order a steak, a pepper-covered steak at a restaurant, the pepper is not down inside the steak. It’s on the outside because that’s where it stays. It doesn’t penetrate. And, the human body, human tissues do not melt and re-form and surround stuff. That’s preposterous.”
Other experts were incredulous over Sperry’s conclusions.
“You can’t make a diagnosis of high-velocity gunshot wound … just on the basis of an X-ray,” Dr. Jerry Spencer, the former chief medical examiner for the Armed Forces Institute of Pathology, testified.
Karch, the Oakland pathologist, agreed. “Any attempt to do so,” he wrote in a statement to the court, “is little more than junk science.”
Perhaps the most damning repudiation came from DiMaio, the retired medical examiner in San Antonio. DiMaio first documented the snowstorm phenomenon in 1985 in his book “Gunshot Wounds,” a definitive pathology text.
DiMaio testified that the X-rays did not show a snowstorm at all. And with the body so decimated, he said, no one could tell whether a bullet killed Glover, much less its path through his body.
“What did the entrance wound look like? You don’t know, because you haven’t seen it. What did the exit look like? You don’t know. Did it actually exit, or was it just under the skin and when the body burned it just fell into all the debris? You don’t know.”
U.S. District Judge Lance Africk, presiding over the pre-trial hearing, asked whether this was a routine disagreement among professionals, or something more fundamental.
“I don’t consider his opinions reliable,” DiMaio said of Sperry. “The thing is, he didn’t have enough objective evidence to reach a conclusion. That’s what I’m saying. I’m saying you can’t reach a conclusion. That’s my testimony.”
Africk ultimately excluded testimony by expert witnesses for both sides. Warren was acquitted in December 2013.
Before dismissing the expert witnesses, the judge mused about the $5,000 Sperry earned for one day in court. He asked Spencer how much he was paid.
Spencer said he charged $200 an hour. “I come pretty cheap.”
“Do you get aggravated,” Africk asked, “after hearing what Dr. Sperry is getting to be here?”
“To use his term,” Spencer replied, “it’s preposterous.”