Short Fall Arguments in Court: A Probabilistic Analysis

University of Michigan Journal of Law Reform Volume 50 | Issue 3 2017 Short Fall Arguments in Court: A Probabilistic Analysis Maria Cuellar Carnegie Mellon University Follow this and additional works at: http://repository.law.umich.edu/mjlr Part of the Evidence Commons, Juvenile Law Commons, and the Medical Jurisprudence Commons Recommended Citation Maria Cuellar, Short Fall Arguments in Court: A Probabilistic Analysis, 50 U. Mich. J. L. Reform 763 (2017). Available at: http://repository.law.umich.edu/mjlr/vol50/iss3/11 This Symposium Article is brought to you for free and open access by the University of Michigan Journal of Law Reform at University of Michigan Law School Scholarship Repository. It has been accepted for inclusion in University of Michigan Journal of Law Reform by an authorized editor of University of Michigan Law School Scholarship Repository. For more information, please contact . SHORT FALL ARGUMENTS IN COURT: A PROBABILISTIC ANALYSIS Maria Cuellar* I will be talking today about how statistical arguments are used in court, specifically in cases of Abusive Head Trauma in which the defendant has claimed that an accidental short fall, and not shaking or child abuse, has caused the child’s injuries. So actually the Johan case1 that Peter Aspelin was talking about leads perfectly into this. In particular, I will be talking about one specific paper by David Chadwick et al. from 2008.2 In this paper, he and his colleagues calculate the risk that a child, a young child, will die from a short fall.3 They find that the risk is less than one in a million.4 In fact, 0.48 in a million.5 I will be providing some criticism of how this quantity gets used in the court; then I will provide an alternative method for figuring out whether a short fall could have caused this specific child’s injuries, and I will close by talking about the challenges that we have today with respect to data. What I will call the Chadwick paper is a study that was written by David Chadwick, from the University of Utah, with seven co-authors in 2008.6 It has the unambiguous title of “Annual Risk of Death Resulting from Short Falls Among Young Children: Less than One in a Million.”7 And as I said, 0.48 in a million is the quantity they calculate.8 So to calculate this value, they used a database called the EPIC database, which stands for the Epidemiology and Prevention for Injury Control database. I will talk in detail about how they found the value shortly. * PhD student at Carnegie Mellon University; joint program in Statistics and the Heinz School of Public Policy and Management. BA, Reed College; MS, Carnegie Mellon University. 1. Joseph Shapiro, Dismissed Case Raises Questions on Shaken Baby Syndrome. NAT’L PUB. RADIO (Dec. 21, 2012), http://www.npr.org/2012/12/21/167719033/dismissed-case-raisesquestions-on-shaken-baby-diagnosis. 2. See generally David L. Chadwick et al., Annual Risk of Death from Short Falls among Young Children: Less than 1 in 1 Million, 121 PEDIATRICS 1213 (2008) (describing results from a systematic review that found only six possible fall-related fatalities of young children in a population of 2.5 million young children over a five year period). 3. Id. at 1214. 4. Id. 5. Id. 6. See generally id. 7. Id. at 1214. 8. Id. 763 764 University of Michigan Journal of Law Reform [VOL. 50:3 This paper is very careful in defining what a short fall is, specifically. They define it as a fall from a height of less than five feet in which the horizontal velocity is no faster than that which a child could have achieved alone.9 They define their population very specifically. They choose the population of infants, ages zero to five.10 This is the only paper of which I am aware that provides a numeric estimate for how likely it is for a short fall to cause death. For this reason, it is used widely both in the literature and in court. A few other papers have also looked at the issue of short falls. I will talk about two of them here. The first one is John Plunkett’s paper from 2001.11 In this paper, he reviewed eighteen cases of deaths in infants that had been caused by short falls,12 and he reviewed each case in detail. What is important about this paper is that he showed that deaths can indeed be caused by short falls.13 But he does not provide a specific numeric estimate that will tell us how common it is for a child to die from a short fall. The paper by David Moran, Keith Finley, Patrick Barnes, and Waney Squire from 2012 titled “Shaken Baby Syndrome, Abuse Health Trauma, and Actual Innocence: Getting it Right”14 has a short section that criticizes the Chadwick paper.15 The Moran et al. paper also does not provide a numeric estimate for how common it is for a child to die from a short fall.16 To arrive at their estimate of 0.48 in a million, Chadwick et al. used the EPIC database from California, which has death records from various medical examiners, in addition to other sources.17 They used data from 1999 to 2003,18 and they counted the number of infants who had died from a short fall in California in this time period.19 Just six were not disproven as short fall deaths.20 Then 9. Id. 10. Id. 11. See generally John Plunkett, Fatal Pediatric Head Injuries Caused by Short-Distance Falls, 11 AM. J. FORENSIC MED. & PATHOLOGY 1 (2001) (concluding that an infant or child may suffer a fatal head injury from a fall of less than three meters (ten feet)). 12. Id. at 2. 13. See id. at 10 (concluding the falls from less than three meters may be fatal). 14. See generally Keith A. Findley, Patrick D. Barnes, David A. Moran & Waney Squier, Shaken Baby Syndrome, Abusive Head Trauma, and Actual Innocence: Getting It Right, 12 HOUS. J. HEALTH L. & POL’Y 209 (2012) (providing a history of SBS, its evidence, and calling for more collaboration between medical and legal communities to “get it right”). 15. See id. at 247–48. 16. See generally id. 17. About the Fatal (Death) Data, Nonfatal Patient Discharge (Hospitalization) Data, and Nonfatal Emergency Department (ED) Data, CAL. DEP’T OF PUB. HEALTH, http://www.cdph.ca.gov/ HealthInfo/injviosaf/Pages/EpiCenterdata.aspx#fatal (last visited Jan. 13, 2017). 18. Chadwick et al., supra note 2, at 1214. 19. Id. 20. Id. SPRING 2017] Short Fall Arguments in Court 765 they divided six by the number of all infants in California during this time period.21 So if you wanted to convert this to a probability statement, you might say this is the probability of an individual having a death and short fall, given that this individual is an infant in California, in the period of 1999 to 2003. They found that this value was 0.48 in a million.22 What I claim is that in court, implicitly, what is being argued is that this value of 0.48 in a million is equivalent to the probability that the child had a short fall given the evidence in that criminal case, the evidence being this is an infant with head trauma and death. Assuming that the 0.48 in a million is i (...truncated)