The Shaken Baby Problem

Friday, February 11th, 2011

Emily Bazelon has a long, well-reported feature in the New York Times Magazine on new doubts about the diagnosis of shaken baby syndrome.

I wrote about this issue in 2009, and my column then inspired some spirited email responses. There is a small but growing part of the medical community that is skeptical of the diagnosis, and a very adamant larger group that says there’s no legitimate debate, here—the diagnosis is sound, and the skeptics are either nuts or are guns for hire.

I’m obviously not a doctor, but it strikes me that there’s something tantalizingly easy about the shaken baby diagnosis. It is based on just three symptoms, all internal, and can be made even when there are no external signs of abuse. Some experts and prosecutors claim that the diagnosis is enough by itself to prove (a) a crime has been committed, (b) who committed it (conventionally, the diagnosis implicates the last person who was alone with the child before the death or injury), and (c) the suspect had the requisite intent (the diagnosis includes the conclusion that the injury could only be caused by intense, vigorous shaking, which prosecutors usually argue in court shows anger and intent to harm).

If doctors find the triad of symptoms, there’s really no defense, unless the suspect attempts to show that someone else was also alone with the child shortly before the symptoms began to appear. (The symptoms are bleeding at the back of the eye, bleeding in the protective area of the brain, and brain swelling.)

The emerging group of skeptics attack both the diagnosis itself and how it’s used in court. They argue the triad of symptoms can be caused by incidents or medical conditions other than shaking, and that the injury itself could occur days before the symptoms begin to appear, instead of the hours or minutes often claimed in court. If true, both of those claims would destroy the half to two-thirds of shaken baby diagnoses in which the child showed no other signs of abuse.

One other note: It’s interesting how quickly the skeptics are dismissed as defense experts for hire. I’m sure there are no shortage of quacks offering their services to criminal defense attorneys. But regular readers of this site have seen enough horror stories by now to know that there’s nothing about testifying for the state that cleanses an expert of bias, either. If they’re outside consultants, they too are paid for their services. And if they actually work for the state as a medical examiner or in a state crime lab, the biases are built into the system.

Digg it |  reddit |  del.icio.us |  Fark

31 Responses to “The Shaken Baby Problem”

  1. #1 |  Danny | 

    For obvious reasons, this can’t be reproduced in a lab. I suppose we need to cobble together a “scientific” sample from which there is conclusive evidence of a shaken baby apart from the symptoms of the “syndrome,” and see what circumstances obtained in those cases.

  2. #2 |  Wade | 

    “It’s interesting how quickly the skeptics are dismissed as defense experts for hire. I’m sure there are no shortage of quacks offering their services to criminal defense attorneys.”

    In my experience (practicing law for eleven years, criminal defense for six) there is a bit of a shortage of experts (quacks or not) willing to testify for the defense. Criminal defendants are not known for being a group that has a lot of money available to spend on expert witnesses, so it is often hard to find experts in many relevant areas who aren’t already employed by the government.

  3. #3 |  Dave Krueger | 

    The question you have to ask yourself here is whether a child was injured. If the answer is yes, then the accused is guilty. It’s as simple as that. If it weren’t for child haters who defend those animals by using such sinister defense strategies as the “accidental death” loophole, the state could save the cost of a trial and execute the accused on the spot.

  4. #4 |  terraformer | 

    “The question you have to ask yourself here is whether a child was injured. If the answer is yes, then the accused is guilty. It’s as simple as that. If it weren’t for child haters who defend those animals by using such sinister defense strategies as the “accidental death” loophole, the state could save the cost of a trial and execute the accused on the spot.”

    Yeah, because kids don’t do stupid things to injure themselves…

  5. #5 |  Richie | 

    Hmn. I’m not a doctor or a lawyer, but it seems to me that if there were other conditions that could cause the three symptoms, there would be cases in which the symptoms occurred where there was a negligible possibility of recent violent shaking (e.g. the child was in a public space with lots of impartial witnesses or video surveillance, in a hospital, etc). Are there?

  6. #6 |  EH | 

    With the way the system is set up, I can tolerate some defense quacks for the time being. People getting away with rape or murder is not the worst problem the legal system has.

  7. #7 |  Burlyman78 | 

    The money disparity for hiring “expert witnesses” strikes me. In the first case from Fairfax, Va. the prosecution called nine expert witnesses to testify that they thought the child was injured due to SBS. The defense called just one. Later in the article, it says the defense in that case also sought the expert testimony of a second well-known SBS skeptic and doctor. That doctor looked at the child’s MRIs from the hospital and concluded that it was not SBS also. But would two vs. nine expert witnesses have really made a difference in the outcome of the case? The state will *always* be able to find more “experts” to testify on its behalf because it will always have more money to hire “experts” than any individual private citizen. Not sure if there is a solution to this problem, other than for jurors to start actually presuming innocence, and questioning “expert” testimony more skeptically.

  8. #8 |  Joe | 

    (The symptoms are bleeding at the back of the eye, bleeding in the protective area of the brain, and brain swelling.)

    These symptoms are not the sort of thing that are common. Clearly if a baby is dead, something killed him or her. This is something that should be subject to scientific forensic determination. Obviously, if a child dies of say, something that mimiced these symptoms–such as meningitis–and then a prosecutor goes after a grieving innocent parents or caretakers, that is very very scary. That said, a lot of children get abused and it is not quack science to say shaking an infant can lead to fatal complications.

    To convict it has to meet a “beyond a reasonable doubt” standard. That is the level of proof the prosecutor has to meet. And that is what courts need to emphasize to juries.

  9. #9 |  Joe | 

    I am all for requiring the state to match expert to expert dollars for defense. Fair is fair.

  10. #10 |  MikeZ | 

    “For obvious reasons, this can’t be reproduced in a lab. I suppose we need to cobble together a “scientific” sample from which there is conclusive evidence of a shaken baby apart from the symptoms of the “syndrome,” and see what circumstances obtained in those cases.”

    Well technically, I’d doubt whether SBS is peculiar to the human skull and brain configuration. So the experiment could be reproduced. However A) The prosecution experts have no need of performing more testing and B) A defense expert who had shaken a bunch of puppies/baby chimps to death would be demonized in court no matter what his findings are.

  11. #11 |  cApitalist | 

    Mr. Balko,

    I’d be interested in reading what the skeptics have to say. Do you have any links you could share? Scholarly works from peer reviewed journals would be especially appreciated, although sadly, there may be substantial publication bias working against these authors. Animal models seem to be the only possible way to conduct prospective, double blinded, controlled trials. Interesting stuff.

  12. #12 |  Joe | 

    A defense expert who had shaken a bunch of puppies/baby chimps to death would be demonized in court no matter what his findings are.

    What if they use raccoons?

  13. #13 |  Joe | 

    Then again you could use possums. No one really likes possums.

  14. #14 |  cApitalist | 

    Unfortunately, few of the skeptics articles that I looked at are more than position statements. Animal models more appropriate than crashing adult primates into walls could be really useful. Fascinating topic.

  15. #15 |  johnl | 

    Cap read Bazelon.

  16. #16 |  johnl | 

    Cap read Bazelon.

  17. #17 |  Nathan | 

    There is a very interesting case out yesterday from the Ontario Court of Appeal setting aside a guilty plea from 1992 which was originally based on the “triad” of symptoms. Based on the state of medical knowledge now, the Crown and defence experts all agreed that it could not be proven that the baby had been shaken.

    http://www.ontariocourts.on.ca/decisions/2011/2011ONCA0120.htm

  18. #18 |  Me (Not I) | 

    Hi I.

    Who is I.

    It doesn’t say in this article.

    Either be unanimous or be anonymous.

    What? It’s Radley Balko?

    Well, he should have said so. I like Radley.

    Maybe it’s OK now, but if he’s going to use “I”, shouldn’t we have some idea who “he” is?

    MeThinks this is one of those unsigned editorials.

    Hmm..I’ve heard about those…

  19. #19 |  Dan | 

    Using animals, even young animals, may not mimic the trio of the so called SBS symptoms. Human infants have a unique skull and brain morphology, thanks to our evolution of big noggins. I have worked at a ME Office for several years now, and we have not had a death listed as SBS that I can remember (same w/ SIDS) since the new guard of pathologists.

  20. #20 |  johnl | 

    Right Dan. It is hard to understand how you could shake a human baby enough to cause head trauma without snapping its neck.

  21. #21 |  zendingo | 

    great, not only do we hate cops we hate babies now too. one step closer to somalia…..

  22. #22 |  johnl | 

    Cap this is the very first article referenced by Bazelon.
    http://www.ncbi.nlm.nih.gov/pubmed/3819836
    “””
    It was concluded that severe head injuries commonly diagnosed as shaking injuries require impact to occur and that shaking alone in an otherwise normal baby is unlikely to cause the shaken baby syndrome.
    “””

  23. #23 |  Aaron | 

    Yeah, because kids don’t do stupid things to injure themselves…

    In which case clearly the kid needs to be executed. I doubt he’ll ever ever injure himself again after a lesson like that.

  24. #24 |  PermaLurker | 

    Actually I was researching SBS just the other night. Studies that have been done on baby pigs show that a human shaking cannot produce sufficient force to generate the injuries. In fact, the trio of injuries cannot be induced without a sudden stop, ie slamming the head into a wall or other hard object ( ex:a fall or car accident). This however leaves an external injury, and SBS is never listed as a cause of death when there is an external injury.

  25. #25 |  Buddy Hinton | 

    This article suggests to me a pervasive legal problem, which is that the law will not quantify the threshold for “beyond reasonable doubt.” The law stubbornly refuses to say whether it is 85% or 90% or 95% or 99% or something else. When pressed, legal scholars seem to give two reasons about why the law refuses to provide this number.

    One reason is that this threshold should be left to the jurors. However, it makes no sense to have the same person (that is, the juror) deciding both the objective threshold number and also the issue of whether the person subjectively believes that this (unstated) objective threshold was met under the totality of circumstances in a given case.

    The other reason they give is that usually a juror cannot precisely quantify the probability that the defendant did the crime. It is true that a juror usually cannot precisely quantify the probability that a defendant did the crime, but as an excuse for the law to fail to give a numeric threshold for “beyond reasonable doubt” it is nonsense. It is nonsense even in the run-of-the-mill case where there are all kinds of subjective credibility determinations, and other irreducibly subjective factual-probabilistic determinations to be made, but in cases where the scientific evidence is essentially the only evidence then it is double-plus-nonsense because forensic scientific experts (if they are good) speak in terms of numeric probabilities and the confidence levels that can be attached to them.

    In shaken infant cases where the only substantial evidence is non-lay-observable medical symptoms it is ridiculous that the court does not tell, in numerical terms, how sure the prosecution doctors have to be (if they are believed) to prove their case as a legal matter. Frankly, if the only evidence is the prosecution doctors, their testimony should not be allowed in if: (i) it cannot be quantified; or (ii) the quantification number does not meet the numerical reasonable doubt standard that the law should have set a long time ago.

  26. #26 |  Buddy Hinton | 

    see also:

    http://www.thebigquestions.com/2010/11/10/reasonable-doubts/

  27. #27 |  johnl | 

    Buddy the other problem is junk science that claims to provide 100% certainty (always a red flag). In this case, it seems like the triad + no external or neck injury, which was considered certain proof of SBS is actually proof of long term injury.

  28. #28 |  Trent McBride | 

    Just a small point on the details…findings like those described in SBS are appropriately termed, in medical parlance, as “signs”, not “symptoms”. Symptoms are medical findings that are experienced by and reported by the patient (or parent, in the case of a child). Things that are found on exam, physical or laboratory, are signs.

    Now back to your regularly scheduled outrage…

  29. #29 |  Mannie | 

    The law stubbornly refuses to say whether it is 85% or 90% or 95% or 99% or something else.

    Because the number would be meaningless. What is it 85% of? If you sat on a hundred juries with identical fact sets, 85% of the accused would be guilty? The prosecutor spoke 1,000 sentences and you believe 850 of them?

    If the doctors examined 1,000 dead babies (and no one has done that) that 850 of them were determined to have been shaken to death? How would you even determine that?

    What does it mean when there is “60% chance of rain? Nothing.

    Law is not a science.

    One of the biggest things I took from the article is that there is sometimes a delay of as much as 2 days between the injury and the onset of signs. Formerly it was believed that the bleeding began instantly. Now the question is not whether the baby was killed, but who did the deed. That’s even less quantifiable.

  30. #30 |  Buddy Hinton | 

    Because the number would be meaningless. What is it 85% of? If you sat on a hundred juries with identical fact sets, 85% of the accused would be guilty? The prosecutor spoke 1,000 sentences and you believe 850 of them?

    It is good to know how sure you need to be numerically even if it is subjective.

    Me: How sure are you?

    You: Pretty sure, but it is complicated, not absolutely sure, can’t quantify it exactly?

    Me: Are you 90% sure.

    You: I think it is more than 90% likely.

    This is an example where 90% is a useful number, even tho you cannot quantify your sureness.

    Then there are the medical type cases.

    Me: the murderer left her DNA at the scene.

    You: is it Ms. Balko’s DNA.

    Me: The scientists tell us it is 90% chance it is Ms. Balko’s.

    You: any other good evidence?

    Me: No.

    You: so its 90% — does that mean that Ms. Balko is guilty or innocent under the standard of certainty required by law?

    Me: law doesn’t know.

    You: whaaaaaat?

    Me: The law is a ass.

  31. #31 |  supercat | 

    The idea of “reasonable doubt” shouldn’t be regarded as mysterious. It’s really pretty simple, and I don’t know why judges try to obfuscate it.

    Reasonable doubt exists if a reasonable person who observed the evidence would believe that there exists a plausible scenario which is consistent with both (1) the defendant’s being innocent of the particular charge before him, and (2) that reasonable person’s observations of the evidence and sense of credulity. A decent defense attorney will suggest such scenarios, but a jury need not limit itself to those scenarios supplied by defense.

    Suppose numerous eyewitnesses, video evidence, and DNA evidence implicate a man in a crime, and he claims that perhaps an identical twin brother did it. Absent some evidence of the existence of an identical twin brother, a juror should probably not regard such a notion as believable. On the other hand, if the defendant were to introduce evidence showing that the he did have a twin brother who strongly resembled him and was likely genetically identical, such a claim should be a basis for reasonable doubt unless the prosecution can show some particular reason the twin brother couldn’t have committed the crime.

    Note that to acquit someone, a reasonable person would not necessarily have to believe that any scenario involving the defendant’s innocence was true, but would merely have to find that such a scenario was plausible. By contrast, to rightly convict someone, a reasonable person would have to decide that there was no way a reasonable person could believe everything he’d have to believe in order to believe the person didn’t commit the crime.

Leave a Reply