Some medical experts have decided that shaking can’t kill babies.
These experts regularly testify for the defense when murder charges are brought after an infant’s death has been attributed to shaking.
I acknowledge that shaken baby syndrome has been vastly over-diagnosed and that innocent people have been convicted and imprisoned. I’m awfully glad I never certified shaking as the cause of an infant’s death. But I don’t believe shaken baby syndrome is a complete myth.
Doctors once viewed with too little skepticism the anecdotal reports and inadequate studies underlying the discredited dogma that specific medical findings proved fatal shaking. I think doctors now view with too little skepticism experiments cited as proof that death by shaking is impossible.
In one study, college football players shook infant dummies equipped with force sensors as hard as they could. The published report says forces recorded by those sensors were insufficient to cause brain injury. That’s hard to believe.
Anybody who has seen a prosecution expert shake a doll until its head snaps back and forth would be hard pressed to believe that a college football player couldn’t shake an infant to death. I doubt that any of the experts who cite the football study as evidence against fatal shaking would volunteer one of their own babies for a real-life experiment.
I think the football model has at least two flaws: No dummy comes close to approximating the anatomical and physiological complexity of a human infant and shaking deaths may not be caused by direct physical injury to the brain.
Consider another model of shaking that supports a different conclusion: the dog.
Duffy, my Yorkie, came out all wrong. His sister, and littermate, is a dainty, 5-pound beauty. Duffy is twice the size of the breed standard, black where he should be gray, with a fluffy, cottony coat and big, floppy ears. He loves the little stuffed lamb that was his prize for winning a puppy obedience contest.
Duffy drops Lamb at my feet, grinning expectantly. I throw it. With a few bounds, a pounce and fwapita-fwapita-fwapita, he shakes Lamb to death and drops it triumphantly at my feet.
Duffy and Lamb accurately model canine killing – the reality of which my daughter once observed. Alison was walking through my grandfather’s barnyard with Dakota, her 70-pound Akita, when an unwary groundhog crawled sleepily from under the barn and wandered across their path. Dakota leaped on it. Fwapita-fwapita-fwapita, and she dropped the dead groundhog at my horrified daughter’s feet.
I like dog/toy interactions and real-life canine killing as models for shaken baby syndrome. Shaking a toy or a rodent looks just like shaking a doll. I’m inclined to dismiss the football player experiment as a flawed model.
I know the dog model is also seriously flawed as a representation of infant shaking. Pound for pound, dogs are much stronger than humans. The anatomy of adult rodents and toy lambs differs greatly from that of infants.
Dog behavior only proves that forceful shaking can kill small animals and tear the stuffing out of toys.
So why do I gravitate toward the dog model even while knowing its inconsistencies and limitations? The answer is clear: The dog model predicts a result that supports my pre-conceived belief that shaking can kill. The football model contradicts that opinion and is thus harder for me to accept.
My preference is a clear example of confirmation bias, a psychology that causes all of us to be too certain of what we know.
Dr. Carol J. Huser, a forensic pathologist, served as La Plata County coroner from 2003-12. She now lives in Florida and Maryland. Email her at firstname.lastname@example.org.