Long-range research to be first U.S. published database
By Mary McGrath
Medill Innocence Project
Published: Sept. 4, 2012
The Medill Innocence Project is creating the first national and publicly available database of infant injuries and deaths linked to shaken-baby syndrome to shed light on an increasingly controversial criminal-justice issue.
Over three months, the Medill Innocence Project, which examines potentially wrongful murder convictions, gathered information from 1,469 criminal cases in all 50 states and Guam with the aim of analyzing patterns and trends and to examine the medical and legal dimensions of shaken-baby syndrome cases.
After the information is confirmed, database findings will be published months from now in an interactive format so the public can search for information. Among other things, the data will include the official cause of death, the defendant’s relationship to the infant, the experts who testified at trial, as well as other basic case information. The Medill Innocence Project hopes its findings will serve the public, including medical and criminal justice professionals and academics as well as individuals and families affected by shaken-baby syndrome injuries and convictions, said Northwestern University Professor Alec Klein, director of the Medill Innocence Project.
“There’s a sense of urgency about pursuing these issues, because children are so vulnerable,” Klein said.
Experts say much remains unknown about shaken-baby syndrome, including the number of cases and whether they’re on the rise or decline. It is also unknown how many convictions have been overturned on the basis of new scientific and medical understanding; nations like Britain and Canada have reexamined how the medical establishment considers shaken-baby syndrome cases.
The dearth of public information is one reason the Medill Innocence Project has undertaken research on these cases. “If we could do something that could help one child or one caregiver, I think we have done some good,” Klein said.
Over the past 30 years, shaken-baby syndrome generally has been identified by a triad of symptoms: bleeding behind the eyes, bleeding in the brain and brain swelling, according to the U.S. Centers for Disease Control. The term “shaken-baby syndrome” appears as far back as in radiologist John Caffey’s 1972 clinical article in the American Journal of the Diseases of Children. Caffey referenced a paper published the previous year in the British Medical Journal by English pediatric neurosurgeon Norman Guthkelch, who observed that shaking infants could lead to subdural hematomas, or brain bleeds.
More than 40 years later, many experts have questioned whether shaken-baby syndrome can be identified solely by the triad of symptoms. Some doctors assert the possibility of a lucid interval, or period of time between the injuries and onset of symptoms, which casts doubt on whether the last adult caring for the child could be criminally responsible, as often alleged. In addition, experts disagree about whether it is biomechanically possible for humans to shake an infant forcefully enough to cause internal injury, without leaving external signs of abuse, such as bruising or broken bones.
Citing privacy concerns, many organizations and individuals were reluctant to share information, at times even public records, with the Medill Innocence Project. While gathering data, the Medill Innocence Project reached out to the National Center on Shaken Baby Syndrome, a Utah-based non-profit organization whose stated aim is to prevent shaken-baby syndrome. The organization declined to contribute any information to the Medill Innocence Project database. Toni Blake, an attorney and trial consultant specializing in the defense of individuals accused of shaking infants, also declined to share her datasets with the Medill Innocence Project.
Drawing from 27 other sources, including lawyers, doctors, family members and friends of both victims and those accused of child abuse, Northwestern sophomores and Medill Innocence Project fellows Mihir Boddupalli and Mary McGrath, this article’s author, gathered information about the cases with assistance from Medill Innocence Project research associate Alison Flowers and under Klein’s supervision.
Using spreadsheets and database software, the team examined cases through the lens of 38 categories to distill key judicial and medical information, such as the criminal charges faced and whether the infant experienced the triad of symptoms.
This past spring, the Medill Innocence Project, through the Medill investigative journalism class it supports, began looking into its first shaken-baby syndrome case; it involved Illinois caregiver Jennifer Del Prete who says she was wrongfully convicted of murder. The Medill Innocence Project had never previously taken on such a case; the lack of witnesses and crime-scene evidence makes reexamining shaken-baby syndrome cases particularly challenging, but the director this year eliminated its policy prohibiting the examination of such cases. “The students did a remarkable job digging into the case,” Klein said. “As we were going through the term, I became increasingly convinced that there was a need not only to look at specific cases, but to look at shaken-baby syndrome as a systemic issue in the United States.”
Caught in the middle of the debate over how shaken-baby syndrome is viewed is Norman Guthkelch, the pediatric neurosurgeon who is credited as the first to observe that shaking could lead to brain bleeds. “It is the worst-tempered medical controversy I have ever encountered,” Guthkelch said. Now 96 and retired, he regrets the polarization surrounding the medically and legally entangled issue. “What conclusions are you justified in drawing when an infant is found dead or horribly injured?”
To investigate the science behind the shaken-baby debate, a 1987 University of Pennsylvania study subjected models of infants—dolls with stuffed heads—to various shaking and impact. Biomechanical researchers concluded that shaking alone is unlikely to result in the triad of symptoms. Rather, researchers found major external injuries, such as broken necks, would accompany the internal injuries. But Frederic Bruhn, a California-based retired pediatrician and child-abuse expert who has testified for the prosecution and the defense in shaken-baby syndrome cases, disagrees. “There’s absolutely no way to know for certain without testing on human infants,” said Bruhn. “And that is absolutely out of the question.”
In 2009, the American Academy of Pediatrics, a professional organization that works to improve children’s health, suggested that doctors identify cases as “abusive-head trauma” rather than shaken-baby syndrome. In a policy statement, the academy reported the term shaken-baby syndrome is “sometimes used inaccurately” to describe infant injuries, because it implicates a perpetrator and disregards the possibility of the infant’s head striking against something. Nonetheless, the term “shaken-baby syndrome” still is widely used in the medical community.
Chicago neuropathologist Jan Leestma, who has testified as an expert witness for the defense in many shaken-baby syndrome cases, said he is dubious about the method physicians use to identify shaken-baby syndrome. “People say that the triad is only seen in shaken-baby syndrome, but that’s completely wrong. You can see them in accidental head injuries, meningitis and the list goes on.”
Pathologists have observed lucid intervals in children, when after a traumatic brain injury, their condition temporarily improves, only to suddenly deteriorate. Wisconsin pathologist Robert Huntington III reversed his 1996 testimony in a shaken-baby syndrome case after observing a lucid interval in a similar case. In 2008, an appeals court ruled the shift in medical opinion meant convicted Wisconsin daycare provider Audrey Edmunds should get a new trial. Prosecutors dismissed Edmunds’ charges nearly six months later.
Witnesses in shaken-baby cases are rare, said Robert Parrish, a child-abuse prosecutor in Salt Lake City. At trial, the judge or jury must decipher complex medical testimony from experts who have opposing interpretations of the evidence. “One of the most difficult parts is presenting medical information to the jury,” said Dermot Garrett, a Maryland assistant state’s attorney specializing in child-abuse prosecution.
But it is especially challenging for the defense to overcome the emotions attached to the tragedy of a child’s injury or death, said Deborah Tuerkheimer, a former New York County assistant district attorney who teaches law at DePaul University in Chicago. In 2009, Tuerkheimer published a controversial paper in the Washington University Law Review, contending the criminal justice system is not keeping up with new scientific discoveries in the United States and abroad. Tuerkheimer said: “Almost implicitly, the defendant is put in the position to defend their innocence.”