It’s back in the news – again. That’s right, just when you thought it was safe to go outside, the “debate” regarding shaken baby syndrome, and its use in criminal courtrooms throughout the country, is back. On February 2, 2011, Emily Bazelon, a senior editor at Slate and the Truman Capote law-and-media fellow at Yale Law School, published a balanced, well researched, and thorough summary of the most recent state of the controversy surrounding shaken baby syndrome forensic evidence in the United States —Shaken-Baby Syndrome Faces New Questions in Court, (NY Times, 2/2/11).
Shaken baby syndrome is a term used to describe the idea that a set of three symptoms – subdural hemotomas, retinal hemorrhages, and edema (brain swelling – commonly known as the shaken baby triad, are diagnostic of intentional abuse by shaking to the exclusion of any other possible cause.
Other common tenants advanced by doctors staffing child abuse centers in pediatric hospitals across the country called by prosecutors as witnesses in these cases include: 1) that shaking, of an otherwise healthy child, causes retinal hemorrhages and subdural hematomas; 2) short distance falls cannot cause a severe or fatal injury; 3) chronic subdural hematomas do not rebleed spontaneously or with lesser trauma; 4) a child suffering from a serious and potentially head injury cannot experience a lucid interval between the inception of the subdural and his or her death; 5) and retinal hemorrhages are exclusively diagnostic of non- accidental trauma.[i]
When examining shaken baby syndrome allegations, it is important to note that the “True” Scientific Process takes a hypothesis, and conducts experimentation to disprove that hypothesis. To prove a theory using the scientific process, you must try to disproveit. You do not prove a theory true by looking for facts that support it, as such a method is fraught with what scientists term confirmation bias.[ii] The advantage of falsifiability is that it avoids confirmation bias, that is, the phenomenon that one finds that for which one is looking.
In contrast, the Junk Science commonly seen behind the forensic science that appears in courtrooms across the country is distinguished from “True Science” in that it relies on faulty scientific methodology. Researchers take a hypothesis (frequently that they have prejudged to be correct), and attempt to confirm or “prove” the hypothesis is correct (using anecdotal evidence), again running the risk of confirmation bias. As noted in the seminal US Supreme Court decision Daubert v. Merrell Dow Pharmaceuticals, 509 U.S. 579 (1993), “Scientific methodology today is based on generating hypotheses and testing them to see if they can befalsified; indeed, this methodology is what distinguishes science from other fields of human inquiry.” Daubert, 509 US 579, at 593.
SBS suffers from a severe case of failing to adhere to these important principals. The theory of shaking as the mechanism or cause of SDHs and retinal hemorrhages, as compared to SDF’s, has never been verified by testing. The theory has, however, been falsified. See, Duhaime, et al., 1987, supra.
From the mid 1950’s through the 1970’s, a pioneering pediatric radiologist named Dr. John Caffey (and others) developed the hypothesis of SBS based initially on a single case study – that of Virginia Jespers, a child nurse who had “confessed” to shaking children.[iii] Dr. Caffey described four cases of children who had subdural hematomas (SDH) and long bone fractures. See, Caffey, J., “Multiple Fractures in the Long Bones of Infants Suffering from Chronic Subdural Hematoma,” Am. J. Roentgenol and Radium Therapy, 1946, Vol. 56, No. 2, p 163. Dr. Caffey – at that time – drew no conclusions about the way in which the four children sustained their respective intracranial injuries; yet, this article is frequently cited as “proof” for the existence of this syndrome.
Dr. Henry Kempe published an article in 1962 in the Journal of the American Medical Association that concluded that all children with subdural hematomas and retinal hemorrhages had been abused by shaking – with no science, evidence, or data to support this conclusion. In 1971, Guthkelch hypothesized – based on only two cases of children who had SDHs, with no external evidence of impact – that the SDHs had been caused by human manual shaking. See, A.N. Guthkelch, “Infantile Subdural Hematoma and its Relationship to Whiplash Injuries,” BMJ, 1971; 2:430-31. Guthkelch did not, however, devise a way in which to test his hypothesis. Yet, once again, the article has been cited numerous times by physicians in numerous case series studies.
SBS is born out of reason and inference and not scientific testing. No data or research to determine if people could physically cause such injuries, and no attempt to adhere to the scientific method and falsify the hypothesis. The SBS theory of causation has never been validated by biomechanicians or even by physicians. Based only on other limited, individual anecdotal cases, other doctors joined the SBS bandwagon and the medical literature about SBS began to grow. Again, little or no research done, data collected, or experimentation conducted to support it. A theory born out of reason and inference, SBS has been confirmed solely by further reason and inference.
It is on this point – it’s origin – that SBS first should have demonstrated its flaws. As noted by one expert commentator:
“The medical literature upon which the diagnoses of SBS is based is replete with more than half a century of confirmation bias . . . . It is composed of a patchwork of “studies” each often consisting of less than a handful of cases which include suspect “confessions” to shaking and inconsistent methods of analyzing the ‘science’.”
“The data that have been cobbled together to support the hypothesis that shaking alone causes brain injury to children has a statistical significance of zero. And, ‘bad data’ that outright disprove or challenge shaking as the cause of brain injury in children are routinely discarded and explained away because they do not fit the prevailing misdiagnosis of SBS.”[iv]
The testimony that a child can only present with this “constellation” of three (3) symptoms has been has been strongly rejected by numerous other experts:
However, research for this Note uncovered no studies showing that cases of child abuse can be separated from other accidental or medical reasons for harm. Medical evidence is accumulating that other reasons for the possible harm, previously disregarded, should be given weight. If a caretaker says that a baby’s injuries were caused by a fall from a low height, they are assumed to be lying. Courts and medical authorities often state uncategorically that a fall would have to occur from the top of at least a two-story building to cause the SBS symptoms. However, a study published in 2001 indicates that a “fall from less than 3 meters (10 feet) . . . may cause fatal head injury and may not cause immediate symptoms,”] with the injury including the classic SBS pattern of retinal hemorrhaging and subdural hematoma. Eighteen observed deadly falls by children from heights varying from two to ten feet were examined. Thirteen of the children had subdural hematoma, five had skull fractures, and four (of only six examined for eye injuries) had retinal hemorrhaging. [Furthermore, twelve of the children had a lucid interval before expiring.
Gene Lyons, Shaken Baby Syndrome: A Questionable Scientific Syndrome And A Dangerous Legal Concept, 2003 Utah L. Rev. 1109, 1110 (2003).[v]
Similarly, the notion that subdural and subarachnoid hematomas are consistent with child abuse and are not consistent with non-accidental trauma has been seriously called into question:
Subdural and arachnoid hematomas as diagnosed in SBS are said to be produced by rotational forces that shear blood vessels as the brain is shaken and battered against the skull. These hematomas can also be produced by natural causes such as a ruptured berry aneurism,] congenital bleeding problems, osteogenesis imperfecta (brittle-bone disease), external hydrocephalus (water on the brain), and rebleed of prior hematoma. Some researchers suggest that vaccine reactions could also cause the symptoms of SBS. Recent medical research shows that lack of oxygen to the brain can, by itself, cause subdural hematoma.
Notwithstanding this, prosecutor’s continue to go forward with cases where the only evidence of ID (who did it based on the timing of the trauma), causation (the injuries could only be caused by intentional abuse), and intent (the trauma had inflicting the injuries had to be severe) is ALL supplied by doctors looking at X Rays, CT scans, and MRI’s and seeing the big three – subdural hemotoma, retinal hemorrhages, and edema (brain swelling).
Mark A. Satawa practices in the area of criminal defense, specializing in forensic evidence and shaken baby cases. He is a member of the board of directors of the National Association of Criminal Defense Lawyers, and is a frequent continuing legal education speaker on shaken baby cases, most recently on April 3, 2009, at the NACDL annual forensic science seminar in Las Vegas.
See Amicus Curiae Brief of Physicians and Bio-Scientists on Behalf of Appellant Alan Yurko, Yurko v State, CR 98-1730 (9th Judicial Cir, FL) (attached).See, Foster, R.K., and Huber, P.W., “Judging Science – Scientific Knowledge,” Ch.3, p 38, et seq.
One of Caffey’s “factual” premises for his 1972 article was what he termed the “shake prone nurse.” Never once did Caffey refer to the actual medical data of the children the nurse was supposed to have shaken. Instead, he relied on a Newsweek article. See“The Boys Jeered Her,” Newsweek, Sept. 10, 1956. Nurse maid Virginia Jaspers was said that have admitted to shaking several infants in her care. As part of the factual “basis” for his 1972 article that the children in the nurse’s care were, in fact, shaken, Caffey quoted the confession the nurse allegedly made “that she shook some children.”
Elaine Whitfield Sharp, “The Elephant on the Moon.” The Warrior, Fall 2003.See also, J.F. Geddes & J. Plunckett, The Evidence Base for Shaken Baby Syndrome, 328 Brit. Med. J. 719 (2004); P.E. Lantz et al., Perimacular Retinal Folds from Childhood Head Trauma, 328 Brit. Med. J. 754 (2004); Michael T. Prange et al., Anthropomorphic Simulations of Falls, Shakes, and Inflicted Impacts in Infants, 99 J. of Neurosurgery 143 (2003; Glenda Cooper, Doubts Grow on Shaken Baby Syndrome, The Sunday Times, Dec. 26, 2004, available here; Sandra Laville, Doubt Cast on Scores of Child Death Cases, The Guardian, June 13, 2005; Lee Scheier, Shaken Baby Syndrome: A Search for Truth, Chicago Tribune Magazine, June 12, 2005, at 10.