Cutting out the Forensic Pathologist
Child Protection Witch Hunts:
The San Diego Experiment
Brenda Buell’s son Phillip had been sickly since birth. He had been treated for pneumonia, bronchitis, an eye infection, and numerous other ailments. In November 1982, when he was twenty-two months old, Phillip’s failing health began to deteriorate rapidly after he bumped the back of his head against a wall. This minor mishap, an event that would be without consequence in a healthy child, caused blood to collect under his scalp. Called Cephalhaemotoma, it revealed that Phillip had a disorder that prevented his blood from coagulating properly.
Seven weeks later, Phillip, pale, clammy, and vomiting bile, was admitted to Kaiser Hospital in San Diego. On his lower abdomen, doctors noted petechia, tiny red dots from small amounts of escaping blood. Phillip was bleeding into his skin, and into his scalp. He was also having nose bleeds and his gums were also bleeding. Three weeks passed before Phillip was examined by a hematologist. The child had been given a blood transfusion which negated the prospects of detecting and diagnosing his bleeding disorder. The hematologist therefore found no evidence of a blood abnormality.
In 1983, when Phillip was undergoing treatment for his undiagnosed illness, and his mother was frantic to find out what was wrong with him, pediatricians in San Diego County were aggressively trying to detect and prevent child abuse. A children’s protection committee had been set up at Children’s Hospital in Kearny Mesa, a town in the center of the county just north of San Diego. This child protection panel consisted of treating pediatricians from the area’s hospitals. Whenever a physician saw an infant or toddler bearing injuries that could have been caused by physical abuse, the committee would be alerted, and the status of that child’s health would be closely monitored. If a pattern of abuse emerged, child protection and law enforcement authorities would be brought into the case. This was a fine idea, motivated by the desire to protect children from abuse. In instances involving infants and toddlers who may have been violently shaken or beaten to death, the coroner of San Diego County was allowing pediatricians to perform the autopsies, to make cause and manner of death determinations. This was not a good idea. It was a bad idea for two reasons: These pediatricians were not forensic pathologists; and many of the physicians performing the autopsies had treated the deceased, a fact that put their objectivity at risk. It was also a dangerous practice because in cases where the manner of death had been ruled homicide, the person who had been with the child when the emergency personnel had been summoned, would end up being convicted of murder ninety-five percent of the time. Having a child die in one’s care, a death later ruled a homicide, was a one-way ticket to death row. It was therefore extremely important that the doctors performing the autopsies were objective and knew what they were doing. A wrong cause and manner of death determination would almost automatically lead to a wrongful conviction and imprisonment. Because there was so much at stake in these autopsies, they should have been conducted by competent and objective forensic pathologists.
A pediatrician at Kaiser Hospital in January 1983, concerned that Phillip Buell’s symptomsthe petechia, his scalp bleeding, the vomiting, and stomach painshad been caused by physical abuse, reported Brenda Buell to the child protection committee at Children’s Hospital. Brenda was notified that Phillip’s condition would be closely monitored for additional evidence of potential abuse. Besides worrying about her son’s health, Brenda now had something else to concern her. These doctors thought she was beating her son. If he didn’t get better, they could take him away and she could be arrested.
Six weeks after being reported to the child protection committee, Brenda received some good news. Another doctor who had treated Phillip at Kaiser, a physician who was on the child protection committee, convinced the panel that Phillip’s symptoms had not been caused by child abuse. Phillip wasn’t getting better, but he was no longer on the list of patients being closely monitored by the committee. Brenda Buell, at least officially, was no longer suspected of abusing her child.
A month after Brenda was taken off the child protection watch list, Phillip, while playing at home, fell and hurt himself badly. It happened at eleven in the morning on April 27, 1983. Phillip, now two and a half years old, was at home with Kenneth Marsh, the man sharing the apartment with Phillip and his mother. Brenda, employed at a wholesale flower market, was at work when Phillip was injured. Kenneth, although home that particular morning, had a job with a Coca Cola bottling plant and owned a small painting business. He was twenty-eight, hard-working, and had never been in trouble with the law. He may not have been aware of it at the time but he was about to have his first brush with the authorities.
Kenneth called 911 and when the ambulance arrived, Phillip was not responsive. He had a pulse, his pupils were reactive and the back of his head, which had landed on a brick fireplace hearth, felt “mushy” which indicated that his blood was not clotting. A paramedic gave him mouth-to-mouth resuscitation as he was being transported to Alvarado Medical Center in San Diego. At Alvarado, Phillip was placed on a ventilator to assist his breathing, then rushed to Children’s Hospital in Kearny Mesa.
Phillip’s brain was swelling and his blood pressure dropping. Brenda arrived at the hospital around one o’clock and was told by the doctor that Phillip was not going to make it. Later that afternoon he died. One of the physicians who had tried to save him, Dr. Roger A. Williams, was on the child protection committee and knew of Phillip’s history. On the chance that Phillip had been killed by someone who had been abusing him, a call went from Kearny Mesa Hospital to the San Diego Police Department. Since Kenneth Marsh had been present when Phillip fell and hit his head, he would be the first person the police would want to question.
Veteran detective Ted Armijo accompanied by two San Diego homicide investigators, went to the apartment where they gathered pieces of a broken ashtray and a towel containing stains of blood. The officers took photographs of the scene, which, except for the towel, was free of blood. The detectives next questioned the family doctor who gave them a rough history of Phillip’s medical problems which included low blood pressure. On two occasions, Phillip’s mother had reported that the boy had fainted, and fallen off his bicycle. After speaking to the pediatrician, Detective Armijo realized that Phillip had not been a healthy child. Now he needed to speak to Kenneth Marsh, and find out how the boy had smacked his head on the fireplace hearth.
Kenneth Marsh told Detective Armijo that a moment after he had stepped out of the living room, he heard a crash. He returned to the room and found Phillip lying unconscious on the hearth with blood on the back of his head. Kenneth said he hadn’t witnessed the accident, but figured that Phillip had fallen off the back, or arm, of the sofa. The boy must have hit the ashtray, or the speaker upon which it sat, on the way down. Kenneth said he applied a bath towel to the back of Phillip’s head to stop the bleeding. Realizing that Phillip was seriously injured, he called 911. It had been a simple but tragic accident. Detective Armijo, having examined the scene, knew that if Phillip had dropped from the sofa’s arm, his head would have fallen four and a half feet onto a hard surface. Had the boy toppled from the back of the couch, the distance would have been between five and six feet. Armijo was a criminal investigator, not a physician, but common sense, and years of death investigation, told him that such a fall, especially if the child was in poor health, could be fatal. He believed Kenneth Marsh’s story, and would write up the case as a tragic household accident. His homicide partners agreed with his assessment. As a result, no one was arrested.
One of the pediatricians who had treated Phillip on admission to Children’s Hospital, Dr. Roger A. Williams, performed the autopsy two days after the child’s death. Dr. Williams should not have conducted this autopsy because, among other things, he wasn’t a board certified forensic pathologist and was therefore not qualified for the task. Dr. Williams had also served on the committee that suspected child abuse in this case, a fact that put his scientific objectivity in question.
Dr. Williams found three small lacerations on the left side of Phillip’s neck; and bruises on his left forehead, cheek, chin, and ear. These were injuries consistent with a child falling off a sofa and hitting his head on a hard surface. Internally, Dr. Williams found bleeding and swelling of the brain, and retinal bleeding, trauma he believed too severe to have been caused by a fall. Dr. Williams concluded that Phillip had either been violently shaken or slammed against something soft. The surface could not have hard because Phillip’s skull had not been fractured. In his ill advised role as forensic pathologist, Dr. Williams ruled the manner of Phillip’s death criminal homicide.
In Dr. William’s report issued May 19, 1983, three weeks after his autopsy and one day after he and other pediatricians met with the district attorney, the doctor neglected to note that Phillip had a blood disorder that might have explained his internal bleeding. On June 30 San Diego County District Attorney Jay Coulter, relying entirely on the opinions of Dr. Williams and his colleagues, charged Kenneth Marsh with the murder of Brenda’s son Phillip. The D.A.’s actions shocked detective Ted Armijo and his partners who believed that an innocent man could be sent to prison.
On November 28, 1983, at the conclusion of one of the most unusual homicide trials ever, the jury found Kenneth Marsh guilty of second-degree murder. The defendant escaped a first-degree conviction on the theory that he intended to hurt rather than kill the victim. The Marsh trial was unusual because of what the prosecution lacked. The jury didn’t hear from a single homicide detective, a coroner, or a board certified forensic pathologist. There was no confession, no evidence or prior bad behavior, no eyewitness or even a jailhouse snitch. Marsh’s defense was equally lacking. His lawyer didn’t challenge Dr. William’s assertion that he was a forensic pathologist, put Detective Amijo on the stand, or offer medical testimony that Phillip’s pre-existing blood disorder had exacerbated the impact of the fall.
The jury found Kenneth Marsh guilty because they heard only one side of the story, a narrative told by Dr. Roger Williams and five other pediatricians who, like Dr. Williams, had served on the child protection committee. Dr. Williams, who identified himself to the jury as a forensic pathologist, testified that Phillip’s injuries were “equivalent to the force of a two to three story fall or an automobile accident at between 25-35 miles per hour.” This conclusion went unchallenged and the jury, in the face of all this testimony from medical experts, did exactly what one would expect them to do. They found the defendant guilty. The judge did his part as well, he sentenced March to fifteen years to life. It was, as they say, a slam-dunk. Kenneth Marsh had, in effect, been convicted by a committee of pediatricians. The Marsh trial illustrates what can happen when the coroner’s office turns the store over to forensic science amateurs.
In 1986, three years after an aggressive prosecutor and six pediatricians sent Kenneth Marsh to prison as a child abuse killer, it was becoming apparent in San Diego County that relying on pediatricians to do the work of forensic pathologists was not a good idea. The causes of death in too many child abuse cases were being successfully challenged by qualified forensic pathologists. In a number of cases, homicide charges had to be dropped. Finally, after five years on operation, the county returned to its former policy of allowing only board certified forensic pathologists to conduct medico-legal autopsies. Had this been the policy in 1983 Kenneth Marsh would not be serving time in prison.
Brenda Buell had not wavered in her belief that Kenneth Marsh was innocent. She was haunted by the realization that had she been home that morning with her son, she would be the one behind bars. In her effort to get Kenneth out of prison, and clear his name, Brenda contacted virtually every defense attorney in San Diego. But without money to finance an appeal, she couldn’t get anyone interested in Kenneth’s case. Finally, in 1994, Brenda found an attorney willing to help, and do it for nothing. David Thompson, an attorney from Carlsbad, California, petitioned San Diego Superior Court for a new trial. In his motion, Thompson challenged the competency of pediatricians to render cause and manner of death opinions in cases like Kenneth’s. Thompson raised the issue of Phillilp’s pre-existing blood disorder, a factor he claimed had been ignored by Dr. Roger Williams and his colleagues. For the first time since his conviction, Kenneth Marsh had a glimmer of hope.
The district attorney decided to fight the appeal by asking two forensic pathologistsDr. Barbara Wolf from Albany County, New York and Dr. Mary Case from St. Louis, Missourito review Dr. Williams’s autopsy report. In May 1995 Dr. Wolf, after assessing Dr. Williams’s work, filed an affidavit in support of his manner of death determination. In response to the possibility that Philllip had mononucleosis at the time of his death (he had tested positive for the disease on January 5, February 10, and March 4, 1983), Dr. Wolf wrote: “The positive mononucleosis spot tests seen in this time frame were likely simply false positive tests.” Referring to Phillip Buell as “a normal healthy child,” Dr. Wolf concluded that “There is no evidence of any pre-existing natural disease in this child.” This conclusion may reflect the fact that Dr. Wolf, by relying on the limited data in Dr. William’s report, was not aware of Phillip’s full medical history. This may also explain Dr. Case’s cursory, four-paragraph endorsement of Dr. Williams’s autopsy results.
New trials, under the best of circumstances (for the appellant) are extremely difficult to obtain, and when prosecutors fight these appeals with expert witnesses, the appellant doesn’t have much of a chance. The appeals judge, influenced by the opinions of two forensic pathologists and the prosecutor’s assurance that the evidence against Kenneth Marsh had been “overwhelming,” denied the appeal. At this point in the case, it looked as though Kenneth’s only chance of getting out of prison was escape or parole. However, before parole was a possibility, he would have to express remorse, and that would require a confession. Unwilling to confess to something he didn’t do, parole would not be his ticket out. If he didn’t get a new trial, Kenneth Marsh would spend the rest of his life behind bars.
In 1996, when it appeared that Kenneth Marsh had no chance whatsoever to regain his freedom or his good name, Brenda Buell met Tracy and Thor Emblem, a husband and wife legal firm out of Escondido, California. Before entering private practice, Tracy had been a prosecutor with the California Attorney General’s Office. Tracy was naturally skeptical of Kenneth’s claim on innocence until she spoke to homicide detective Ted Armijo who told her that he and the other two investigators believed that Phillip’s death had been accidental. Besides Detective Armijo’s opinion the attorneys would have to find medical experts who agreed that Phillip’s death had been caused by a relatively short fall. The Emblems were able to get law professors Justin Brooks and Jan Stiglitz interested in the case. The professors were directors of the California Innocence Project at California Western School of Law in San Diego. The innocence program featured law students who helped defense attorneys and others exonerate people they believed had been wrongfully imprisoned.
The second writ of Habeous Corpus, filed in October 2002, included the opinions of three medical experts who challenged the findings of the prosecutor’s pediatricians. Comprised on 180 pages and 1,300 exhibits, the writ was, unlike the first petition, an impressive document. The appeal, grounded on the proposition that Kenneth Marsh had been doomed form the beginning by “false medical testimony born out of zeal for child abuse protection,” challenged the wisdom of substituting pediatricians for forensic pathologists in cases of infant and toddler death. The medico-legal problem in the Marsh case was exacerbated by the fact the pediatricians saw themselves as combatants in the war against child abusers.
The medical experts who submitted affidavits on behalf of Kenneth Marsh were: Dr. Thomas Schweller, a pediatric neurologist from San Diego; Dr. Michael Innis, an Australian pathologist and hematologist; and Dr. Gregory D. Reiber, a clinical professor of pathology at the University of California, Davis Medical School. Dr. Schweller believed that Infectious Mononucleosis and Hepatitis and a drug administered at the hospital, had made Phillip’s minor accident fatal. Dr. Innis concluded that Phillilp had died from “cerebral edema and Intracranial Hemorrhage following a coagulopathy (bleeding disorder) resulting from Infectious Mononucleosis induced Hepatitis.” Dr. Reiber believed that because of a “flawed and lacking autopsy, the cause of Phillip’s death had never been adequately analyzed from a forensic viewpoint.” Dr. Reiber, quite specifically in his criticism of the autopsy, faulted Dr. Williams for not having the brain examined by a neuropathologist; for not making a microscopic examination of the eyes; for not saving fluid and tissue samples for later review; for not examining the child’s tongue to determine if it had been bitten in a seizure; for not make microscopic examinations of the bone marrow samples; and for denying the obviousthat Phillip had a serious blood disorder that explained the internal bleeding and his brain swelling. In other words, the autopsy had been a pro-forma exercise carried out by a physician who had already made up his mind that Phillip had been an abused child.
The man who had prosecuted Kenneth Marsh was no longer in office. His successor, District Attorney Bonnie Dumanis, decided to fight the appeal. Her spokesperson, Deputy District Attorney Genaro Ramirez, as reflected in statements appearing in the local press, hit back hard and somewhat gratuitously: “[Brenda Buell’s] relentless efforts in championing her former lover’s cause of martyrdom has been the bane of her existence for almost two decades, all because she refuses to confront the somber reality that her very human mistake in loving and trusting [Kenneth Marsh] contributed to the death of her son….Kenneth Marsh is in fact guilty beyond a reasonable doubt of murdering Phillip Buell.”
Three more medical experts, in June 2004, lent their names, reputations and opinions to the Kenneth Marsh appeal. Dr. Jon Edward Leetsma, a veteran Chicago neuropathologist who had testified for the defense in a number of celebrated murder cases (the Massachusetts SBS trial of au pair Louise Woodward, and the case against novelist Michael Peterson accused of bludgeoning his wife to death in Durham, North Carolina), asserted that Dr. William’s testimony that brain injuries like Phillip’s could only be caused by two-story falls, serious car accidents, or violent shaking, was not supported by the medical science. Dr. Kirk Thibault, a pediatric head injury biomechanical engineer from Philadelphia, also a regular expert for the defense in homicide trials, agreed with Dr. Leetsma’s conclusions. Dr. Janic Ophoven, a pediatric forensic pathologist in the Anoka County Minnesota Coroner’s Office, echoed Dr. Reiber’s criticisms of Dr. Williams’s autopsy report. The medical evidence in support of an accidental death in the Buell case was adding up.
District Attorney Dumanis, faced with the opinions of six medical experts who believed that Kenneth Marsh had been convicted on bad forensic science, petitioned the court for the services of an independent forensic pathologist who would look at the evidence and render an “objective” opinion as to the cause and manner of Phillip’s death. Dr. Sam Gulino, the forensic pathologist from Hillsborough County, Florida, accepted the assignment and concluded as follows: “Although I am concerned that the head injury of Phillip Andrew Buell may have been the result of inflicted trauma, I am unable to conclude beyond a reasonable doubt, or to a reasonable degree of medical certainty, that he was a victim of child abuse.”
On August 5, 2004, District Attorney Dumanis informed Superior Court Judge David Gill that: “There is new evidence that points to reasonable doubt in the [Marsh] case. We feel we have an ethical obligation to do the right thing.” The right thing amounted to asking the judge to set aside the conviction and give the appellant a new trial. For a prosecutor this was a rare and difficult move. A week later Judge Gill set aside the conviction.
In Otaymesa, California on August 17, 2004, Kenneth Marsh walked out of Donovan State Prison where he had been living for twenty-one years. His new trial had been scheduled for September 30, but two weeks after his release, District Attorney Dumanis asked the court to dismiss the charges against him. The judge acted on the recommendation which made Kenneth Marsh, if not fully exonerated, a free man.
In February 2005 Marsh filed a claim before the California Compensation Board which can award wrongfully convicted persons $100 for every day served behind bars. Nate Northrup, the Chief Deputy Counsel for San Diego Country, pointing out that no one had proven Marsh innocent, argued that taxpayers didn’t own him a cent. Since Marsh, due to questionable evidence, had not been proven guilty beyond a reasonable doubt, he was, under the law, presumed innocent. In America innocence isn’t something you have to prove. Northrup, not accepting the fact that Marsh had been unfairly tried, said this: “Kenneth Marsh was properly convicted in the death of Phillip Buell and served twenty-one yearsend of story.” A year later, however, the story continued. The California Board agreed to pay Kenneth $756,000, the largest award of its kind in the state.
There is no reason to believe that the pediatricians who sealed Kenneth Marsh’s fate were anything but sincere in their beliefs and testimony. Indeed, if asked, these physicians might still insist that Phillip Buell had been murdered. But what they think or believe is not important. What is important is this: pediatricians without specialized training in forensic pathology should not be conducting autopsies in potential murder cases. This is the lesson in the Marsh case.