Part I, Section I

Prosecutors build murder cases on disputed Shaken Baby Syndrome diagnosis


Mother acquitted of shaking a baby: ‘I would have committed suicide’

Part I, Section II

Doctors who defend shaking diagnosis dismiss scientific challenges


Prosecutor: ‘The public should know that this can kill’


The unsettled science of Shaken Baby Syndrome

Part I, Section III

Mother released from prison: ‘It hurts too bad to remember’


After five years in prison, a young mother is set free

Part I, Section IV

In Maryland, a baby collapses and a babysitter is blamed

Part I, Section V

Prosecutor to jurors: ‘Healthy babies just don't die’

Part I, Section VI

But that year, the case took a significant turn.


Babysitter once sentenced to 20 years in prison: ‘I’m not gonna say that I killed somebody’


A breakdown of the dismissed cases


Map: Parents and caregivers accused of shaking nationwide


Former medical examiner: ‘Alleged cases of pure shaking are unusual’

Part II, Section I

Doctors who diagnosed Shaken Baby Syndrome now defend the accused


Doctor who helped introduce the hypothesis behind Shaken Baby Syndrome: ‘I don’t think innocent people should be in jail’

Part II, Section II

A child abuse diagnosis raises questions, doubts


Accident reconstruction specialist: ‘We need to apply the science’

Part II, Section III

Engineers: Falls could be more dangerous than shaking


A biomechanical look at shaking


How much acceleration can be generated by shaking a 22-pound crash-test dummy

Part II, Section IV

Doctor: ‘What could be right about getting it wrong?’

for story updates

Your browser does not support the <video> element.

Shaken science

A disputed diagnosis imprisons parents

Prosecutors build murder cases on disputed Shaken Baby Syndrome diagnosis

You are reading one part of an investigative series on the disputed Shaken Baby Syndrome diagnosis that has put people behind bars.

Six years ago, in a tidy home day care on the edge of a cornfield, with angel figurines in the flower beds and an American flag over the driveway, 9-month-old Trevor Ulrich stopped breathing. He had contusions on his scalp and bleeding on the surface of his swollen brain.

Within weeks, day-care owner Gail Dobson was charged with killing the baby in a fit of frustration at the business she had run for 29 years along a rural stretch of Maryland’s Eastern Shore. The Sunday school teacher and grandmother of three was convicted of second-degree murder in 2010, eight months shy of her 54th birthday, and sentenced to 20 years in prison.

“To me, she is a monster,” Trevor’s mother told a local television reporter in 2013. “She is a cold-hearted killer.”

But what prosecutors called a clear-cut case of child abuse is now mired in doubt. Two doctors working on Dobson’s appeal last year argued that the scientific testimony used against her was fundamentally flawed. A judge overturned the conviction and ordered a new trial, finding that a jury hearing that argument could have had “a reasonable doubt” about her guilt.

Doctors for the prosecution said Trevor had been a victim of Shaken Baby Syndrome, a 40-year-old medical diagnosis long defined by three internal conditions: swelling of the brain, bleeding on the surface of the brain and bleeding in the back of the eyes. The diagnosis gave a generation of doctors a way to account for unexplained head injuries in babies and prosecutors a stronger case for criminal intent when police had no witnesses, no confessions and only circumstantial evidence.

It has also led to more than a decade of fierce debate: Testing has been unable to show whether violent shaking can produce the bleeding and swelling long attributed to the diagnosis, and doctors have found that accidents and diseases can trigger identical conditions in babies.

Challenges to the diagnosis have spilled into courts on two continents. In 2005, Britain’s Court of Appeal found that the head and eye injuries alone were not absolute proof of abuse and, in Sweden last year, the Supreme Court ruled that the scientific support for the diagnosis had “turned out to be uncertain.”

In the United States, 16 convictions have been overturned since 2001, including three last year. In Illinois, a federal judge who recently freed a mother of two after nearly a decade in prison called Shaken Baby Syndrome “more an article of faith than a proposition of science.”

Despite the uncertainty, prosecutors are still using the diagnosis to help prove criminal cases beyond a “reasonable doubt” against hundreds of parents and caregivers.

“You can’t necessarily prove [Shaken Baby Syndrome] one way or another — sort of like politics or religion,” said forensic pathologist Gregory G. Davis, the chief medical examiner in Birmingham, Ala., and the board chairman of the National Association of Medical Examiners. “Neither side can point to compelling evidence and say, ‘We’re right and the other side is wrong.’ So instead, it goes to trial.”

The Washington Post, in an investigation with journalists at Northwestern University’s Medill Justice Project, carried out the first systematic examination of dispositions in Shaken Baby cases since doctors started disputing the science behind the syndrome.

Reporters used court records and newspaper reports to track down murder or abuse cases involving shaking that have been filed or dismissed since 2001. The year-long study unearthed about 1,800 resolved cases nationwide, finding some of the heaviest concentrations in counties in Ohio, Pennsylvania, Wisconsin and Nebraska.

About 1,600 cases resulted in a conviction, a rate that is higher than that for other violent crimes. In hundreds of cases, there were reports of shaking along with more obvious forms of violence that left extensive bruises and broken bones, with prosecutors alleging that babies also had been slammed, thrown or beaten.

The study for the first time identified about 200 cases in 47 states that ended when charges were dropped or dismissed, defendants were found not guilty or convictions were overturned.

Among them: a 39-year-old software entrepreneur in California who mourned his infant son while locked in an isolation cell; a 13-year-old babysitter in Washington who was charged with second-degree murder; and a 46-year-old grandmother in Arizona who spent nearly 2½ years facing capital murder charges.

Kelly Kline, acquitted in 2012 of shaking a baby to death in her home day care in rural Ohio, met her 6-year-old daughter for a parents’ lunch at school the day her mug shot flashed on the local news.

“If I wasn’t married and didn’t have kids, I would have committed suicide because of the hell, the embarrassment,” said Kline, a mother of three.

Kelly Kline at home with her family April 20 in Shreve, Ohio. Kline was charged with killing a child by shaking. Kline was found not guilty. (Photo by Bonnie Jo Mount/The Washington Post)

In four of the cases, doctors who had diagnosed shaking later revised their opinions, saying they were uncertain about the timing or cause of the injuries. One of the revisions helped free a Sacramento father after 3½ years in jail.

In four other cases, new medical examiners found that their predecessors had made mistakes by diagnosing shaking in babies who likely died from conditions that had nothing to do with violence. One doctor in Tennessee found a 10-week-old diagnosed with shaking appeared to have suffered from a series of strokes while he was in the womb.

Forensic pathologist George Nichols is among the doctors who once diagnosed Shaken Baby Syndrome and no longer believes in the science.

“Doctors, myself included, have accepted as true an unproven theory about a potential cause of brain injury in children,” said Nichols, who was the chief medical examiner of Kentucky for 20 years before retiring in 1997. “My greatest worry is that I have deprived someone of justice because I have been overtly biased or just mistaken.”

Continue story below

Expand to read more  Close article