8 Dr. Smith and the Practice of Pediatric Forensic Pathology As required by my terms of reference, I have conducted a systemic review of the policies, procedures, and practices of pediatric forensic pathology in Ontario from 1981 to 2001. In this chapter, I report on that review and assess the systemic failings it revealed. They provide the basis for the policy recommendations I make in Volume 3 to improve the practice of pediatric forensic pathology and to ensure, so far as possible, that history will not repeat itself. This chapter examines the practices used in pediatric forensic pathology in individual cases, and how those practices could and did fall short of what is required. In Chapter 9, Oversight of Pediatric Forensic Pathology, I do the same when considering the mechanisms of oversight of that work, again to expose the failings that could and did occur. In both contexts, it must be remembered that what was happening with pediatric forensic pathology reflects in very large meas- ure what was happening with forensic pathology generally. The practices used, the oversight mechanisms available, and the shortcomings were common to both. In this sense, pediatric forensic pathology is a subset of forensic pathology. Before I turn to a detailed report on the troubling aspects of pediatric forensic pathology as practised in Ontario from 1981 to 2001, several things must be said. First, although the Inquiry heard considerable evidence of a general kind about the practice of pediatric forensic pathology in those years, our review for the most part focused on Dr. Charles Smith and the way he did his work. This focus reflects the reality that the errors he made were a primary cause of the significant loss of public confidence that made the review necessary. It is important that these errors be identified because my recommendations must address them if public confi- dence is to be restored. Second, although much of what we heard dealt with Dr. Smith, the evidence also showed that, in a number of instances, other pathologists were involved as well. Some made the same errors he did. Many, and in some instances most, fol- 116 | INQUIRY INTO PEDIATRIC FORENSIC PATHOLOGY IN ONTARIO: VOLUME 2 lowed some of the same practices. In all these instances, however, the serious errors that were made, whether by Dr. Smith or others, exemplify grave systemic problems with the practice of pediatric forensic pathology in Ontario at that time. The troubling problems were not confined to Dr. Smith. Without correction of these systemic failings, these errors could well occur again. These were not merely the isolated acts of a single pathologist that could be fixed by his removal. Third, the evidence of Dr. Smith’s mistakes in individual cases is derived largely from 18 of the cases that were the subject of the Chief Coroner’s Review ordered by the Office of the Chief Coroner for Ontario (OCCO) and that were examined in detail in our hearings. These cases were selected for the Chief Coroner’s Review because they involved Dr. Smith and they engaged the criminal justice system. Although the evidence about Dr. Smith’s work in these cases paints a stark picture of the grave errors he made, it is not my role to determine whether or to what extent his mistakes might have led to a wrongful conviction. Whether or not that occurred, the errors were nonetheless serious. They represent ways in which the practice of pediatric forensic pathology in Ontario in Dr. Smith’s time could and did go badly wrong. Fourth, these cases were part of a complete review of Dr. Smith’s work in criminally suspicious cases between 1991 and 2001. They provide little basis, however, on which firm conclusions can be drawn about his work in hospital pathology or his work for the OCCO in cases that were not criminally suspicious. Finally, it is important to remember that the troubling aspects of the practice of pediatric forensic pathology that occurred in Ontario during this time took place within a setting larger than that of the individual pathologists. As my review of the oversight of pediatric forensic pathology in Ontario in these years later describes, the senior officials who oversaw the death investigation system must also be held responsible for the tragic events about which I have heard. I turn then to the various aspects of Dr. Smith’s work that I have found want- ing and that demonstrate systemic failings in the practice of pediatric forensic pathology from 1981 to 2001. I will begin with the training and experience that Dr. Smith brought to his work. TRAINING AND EXPERIENCE In this section I address three questions. First, what was Dr. Smith trained and certified to do, and what training did he lack? Second, how did he become the dominant figure in pediatric forensic pathology when he had no formal training or expertise in the core discipline, forensic pathology? And third, how did this deficiency affect his work in the cases before me? DR. SMITH AND THE PRACTICE OF PEDIATRIC FORENSIC PATHOLOGY | 117 Dr. Smith’s Training Dr. Smith is a pediatric pathologist, not a forensic pathologist. He has neither for- mal forensic pathology training nor board certification in that field. Nevertheless, the OCCO and the Hospital for Sick Children (SickKids) appointed him director of the Ontario Pediatric Forensic Pathology Unit (OPFPU) in 1992, and, with time, he came to be known as the province’s leading expert in pediatric forensic pathology. Dr. Smith now acknowledges that his forensic pathology training was “woefully inadequate” and that this gap contributed significantly to his mistakes in the cases examined by the Commission. In the 1980s and 1990s, no formal forensic pathology training or certification was offered in Canada. That remains the case today. The Royal College of Physicians and Surgeons of Canada (Royal College) does not yet offer specialty training or certification in forensic pathology. By contrast, the United States and the United Kingdom have offered specialty examinations in forensic pathology since the 1960s. In the absence of well-defined postgraduate training programs, pathologists doing forensic work in Canada have traditionally been self-taught or have resorted to informal training networks. A small number have obtained qualifica- tions outside Canada. Few of the pathologists who performed post-mortem examinations for the OCCO in the 1980s and 1990s were formally trained and certified in forensic pathology. Those who did receive formal training and certifi- cation did so in the United States or the United Kingdom. Indeed, in the 1980s and 1990s, the prevailing Canadian view was that pedi- atric pathologists were best situated to perform forensic autopsies on infants and children. As a result, expertise in pediatric pathology was emphasized over train- ing and qualifications in forensic pathology. Dr. Smith graduated from the University of Saskatchewan medical school in 1975. He then spent the first two years of his anatomical pathology residency there, and his final two years with the University of Toronto. His fourth and final year was spent at SickKids in pediatric pathology. During his residency, he per- formed some coroner’s autopsies; however, none was in a criminally suspicious case. After completing his residency in anatomical pathology, Dr. Smith remained at SickKids from July 1980 to July 1981 to train further as a Fellow in pediatric pathology. During his fellowship year, he performed some forensic autopsies, but, again, none were in criminally suspicious cases. In November and December 1980, Dr. Smith passed the examinations in anatomical pathology offered by the American Board of Medical Specialties (American Board) and the Royal College, respectively, and was certified as an 118 | INQUIRY INTO PEDIATRIC FORENSIC PATHOLOGY IN ONTARIO: VOLUME 2 anatomical pathologist. At that time, examinations in the subspecialty of pedi- atric pathology were not offered in either the United States or in Canada. Nineteen years later, however, that had changed in the United States, and in 1999, he passed the American Board examination and also became certified in pediatric pathology. In 1981, after completing his fellowship, Dr. Smith started working full time at SickKids. He had no forensic pathology training, and only limited exposure to criminally suspicious cases and death investigations. Because of his strong inter- est in autopsies, however, he began to perform more of them than did his pathol- ogy colleagues at SickKids, who were primarily interested in clinical pathology. By the 1990s, most of his autopsy work was forensic pathology, that is, autopsies per- formed under coroner’s warrant. Despite his increasing concentration on forensic work, Dr. Smith did not take any forensic pathology training. His continuing medical education, which con- sisted of attending conferences and reviewing the available literature, focused pri- marily on pediatric pathology. He told us that at that time he did not view forensic pathology as a separate discipline that could inform his work. He received no training in either injury identification or the appropriate role of the forensic pathologist in the criminal justice system. He had no exposure to any certified forensic pathologists and did not appreciate that there was any value in obtaining knowledge about forensic pathology. As Dr. Smith admitted, “[t]hat thought didn’t cross my mind, and certainly no one suggested it ...” Instead, he picked up his limited understanding of forensic pathology on the job. Over time, however, Dr. Smith’s reputation grew. In the mid-1980s, he began lecturing on pediatric forensic pathology, particularly about issues relating to the criminal justice system. By the 1990s, he was lecturing on the subject to Crown counsel and police officers and had become a regular participant at edu- cational courses offered for coroners. There is no doubt that he became an effec- tive speaker to these audiences. At the Inquiry, Dr. Smith testified that these speaking engagements helped to build his experience and comfort level in both pediatric pathology and forensic pathology. His growing reputation seems to have been based more on these speaking engagements than on his work in crimi- nally suspicious cases. It certainly was not based on any formal training in foren- sic pathology. Dr. Smith’s Experience In 1992, as mentioned above, the OCCO and SickKids agreed that Dr. Smith should become the first director of the OPFPU. The evidence at the Inquiry sug- DR. SMITH AND THE PRACTICE OF PEDIATRIC FORENSIC PATHOLOGY | 119 gests that the OCCO and SickKids did not select him on the basis of his forensic pathology expertise. He had only limited experience with criminally suspicious pediatric cases, which are often the most difficult in pediatric forensic pathology. To that point in his career, he had been involved in only 10 to 15 such cases, by his own estimate. Many of those did not involve giving evidence – another aspect of forensic pathology in which he had no training. Dr. M. James Phillips, the pathol- ogist-in-chief at SickKids who formally appointed Dr. Smith to the director’s position, was not a forensic pathologist and not in a position to evaluate Dr. Smith’s forensic training, skills, or expertise. The OCCO wanted to have someone who would specialize in pediatric forensic pathology and appears to have been moved more by Dr. Smith’s reputation and interest than concerned about his lack of training. Equally important, Dr. Smith was the only pathologist at SickKids who had the time and inclination to take on the role. He filled a void that no one else wanted to fill. As director of the OPFPU and with the active support of the OCCO, Dr. Smith became the dominant pathologist for child abuse and homicide cases in Ontario. He brought with him an impressive title and a growing reputation and, relatively quickly, came to be perceived as the authority in pediatric forensic pathology. Dr. Smith also presented himself in this way. When he testified in September 1994 in Valin’s case, for instance, Dr. Smith told the court that, as director of the OPFPU, a “unique” unit in Canada and indeed North America, he probably performed more pediatric forensic autopsies than anyone else in the country. In April 1998, he told the court in Sharon’s case that, given his vast experience with pediatric cases, he was more qualified to assess a child’s pene- trating wounds than a forensically trained pathologist, whose primary experi- ence would have been with adults. We now know, as Dr. Smith himself admitted, that he was self-taught and his forensic pathology education and training were “minimal” and “woefully inad- equate.” He simply did not have the specialized professional skills necessary for the work. He acknowledged that his lack of training and expertise contributed significantly to the mistakes he made. This problem was especially true in crim- inally suspicious pediatric cases, particularly the more difficult ones involving, for example, identification of injury or the timing of the infliction of injury. The consequences are best illustrated in three cases – the cases of Valin, Sharon, and Jenna – where he committed basic forensic pathology errors, with tragic consequences. 120 | INQUIRY INTO PEDIATRIC FORENSIC PATHOLOGY IN ONTARIO: VOLUME 2 1 Consequences in the Cases of Valin, Sharon, and Jenna In Valin’s case, Dr. Smith did not perform the post-mortem examination but was consulted for a second opinion in August 1993 and testified in court in September 1994. Because he did not understand that normal post-mortem changes can include bruising of the neck and dilation of the anus, Dr. Smith wrongly concluded that Valin had died of manual strangulation and that she had been sexually assaulted. Other pathologists agreed with his opinion, to varying degrees. At the time he provided his consultation report and testified at the trial of Valin’s uncle, William Mullins-Johnson, Dr. Smith had never before been involved in a post-mortem examination of a sexually abused child. Many years later, qualified forensic pathologists who reviewed the case, including those who conducted the Chief Coroner’s Review, concluded that the cause of Valin’s death was unascertained and that there was no evidence of sexual abuse. Dr. Smith’s observations of “ulceration, laceration, and hemorrhage in the anus” were prop- erly attributable to the dissection of tissue or its preparation for microscopic work. The dilation of the anus, and much of what Dr. Smith described as bruising to Valin’s body, represented post-mortem artefacts – that is, post-death occur- rences that have no pathological significance. Dr. Smith’s basic mistakes in interpreting the autopsy findings reflect his inad- equate training in forensic pathology. However, he compounded them by failing to recognize the limits of his own expertise. When the pathologist who conducted the post-mortem examination consulted him on the case, Dr. Smith had neither the training nor the experience to provide that opinion. He ought to have recog- nized his limitations. Dr. Smith’s lack of training and experience, and his failure to recognize his lack of experience, had serious consequences. The Court of Appeal for Ontario has concluded that Mr. Mullins-Johnson was wrongly con- victed of first-degree murder, yet he spent more than 12 years in prison. In Sharon’s case, Dr. Smith performed the post-mortem examination in June 1997 and concluded that the cause of death was blood loss due to multiple stab wounds. He testified at the preliminary hearing that the wounds were consistent with having been caused by scissors. Dr. Smith was wrong. The stab wounds that Dr. Smith observed at the post-mortem examination were in fact dog bites. When Dr. Smith performed the autopsy, however, he had virtually no training or previ- ous experience with either stab wounds or dog bites. His inexperience with wound interpretation led to this very significant misdiagnosis. At the Inquiry, Dr. Christopher Milroy, a forensically trained and certified pathologist with experi- 1 See Appendix 28 at the end of Volume 4 for summaries of the cases. DR. SMITH AND THE PRACTICE OF PEDIATRIC FORENSIC PATHOLOGY | 121 ence in the area, pointed to several basic errors in Dr. Smith’s interpretation of Sharon’s wounds. First, Dr. Smith misinterpreted the edges of the wounds, which displayed sig- nificant abrasion and contusion. According to Dr. Milroy, this condition indi- cated that the injuries were not caused by a sharp penetrating weapon, such as scissors or a knife. Contrary to Dr. Smith’s testimony at the preliminary hearing, the appearance of the injuries was actually inconsistent with stab wounds. Second, Dr. Smith failed to recognize that there was a canine bite mark pattern on some of the injuries and on the skull. At the preliminary hearing, Dr. Smith testified that Sharon’s wounds and the marks on her skull did not reveal any dog teeth marks. According to Dr. Milroy, Dr. Smith was wrong on both counts. In at least one of the photographs, the injury displayed a patterned abrasion that was highly suspicious of the arch of a dog’s teeth. And, on Sharon’s skull, there was an almost circular area of indented penetrating fractures, a characteristic of animal tooth bite marks. Third, Dr. Smith misinterpreted an injury to Sharon’s scalp, part of which was torn off during the attack. Dr. Smith testified at the preliminary hearing that the scalp appeared to have been “cut or incised” with some crushing or tearing, indicative of the use of scissors. Dr. Milroy, however, found that the scalp had a lacerated wound edge suggesting that it had been torn or ripped away, not cut. Fourth, Dr. Smith misinterpreted the wound tracks. He testified at the prelim- inary hearing that some of the injuries had a double-pointing mark and, in some instances, there were two tracks to the injury. In his view, both characteristics were consistent with the use of scissors. Dr. Milroy told us that one could not be sure that some of the wounds displayed double-pointing marks. In any event, scissors have two blades, with blunt outer edges. Thus, in a penetrating wound caused by scissors, one would typically observe two blunt edges to the wound, not two pointed edges. Fifth, Dr. Smith failed to consider that the distribution of the injuries to Sharon’s body weighed heavily in favour of a dog attack, not a stabbing. The wounds were largely to Sharon’s upper arms and neck, with little injury to her trunk. According to Dr. Milroy, a dog would tend to clamp onto the neck and arms with its jaws, but not the trunk – the latter being more difficult to grip. In a stabbing, by contrast, the trunk is typically the target, and there are usually defen- sive wounds to the hands and forearms as the victim attempts to fend off the attacker. The injuries to Sharon’s arms were not defensive wounds. The relative absence of injury to Sharon’s trunk and the lack of defensive wounds were strong evidence that Sharon was not stabbed. In his closing submissions at this Inquiry, Dr. Smith noted that other experts, 122 | INQUIRY INTO PEDIATRIC FORENSIC PATHOLOGY IN ONTARIO: VOLUME 2 including a forensic odontologist, Dr. Robert Wood, also misinterpreted Sharon’s wounds. This is true, but does not excuse Dr. Smith’s errors in the case. Dr. Smith expressed his opinion that Sharon’s injuries were stab wounds several months before Crown counsel consulted Dr. Wood. The reality is that each of the errors identified above contributed to Dr. Smith’s misdiagnosis in the case. Had he properly interpreted the wounds, he would not have arrived at the conclusion that he did, regardless of Dr. Wood’s opinion. In my view, with appropriate foren- sic pathology training and expertise, he likely would not have made the basic errors that he did. Dr. Smith’s mistake in Jenna’s case was not with respect to the cause of death. Instead, he erred in his interpretation of the timing of Jenna’s multiple abdominal injuries. Cases involving the timing of multiple injuries causing death are extremely difficult and require sound forensic pathology knowledge. It is evident from a review of the events of Jenna’s case that Dr. Smith lacked the training and experience to take on such a difficult task. Because the body reacts to injury over time, the forensic pathologist may assist in establishing when a child suffered an injury by examining the progress of the body’s healing reaction to the injury at the time of death. For instance, once an injury is inflicted, inflammatory cells rush into the tissue to repair the damage. That process stops when the injured person dies. So, the level of inflammation in the tissue helps the forensic pathologist assess how long before death the damage was sustained. Dr. Smith performed the post-mortem examination on Jenna’s body in January 1997 and correctly determined that Jenna had suffered multiple abdom- inal injuries. Where Dr. Smith erred, however, was in his ultimate diagnosis of when Jenna suffered her injuries. His error was significant because timing was a central issue in the death investigation, since establishing the timing of the injuries had the potential to exclude one or the other of the two suspects: Jenna was in the care of her mother, Brenda Waudby, before 5 p.m. on the evening of her death, and with her babysitter after 5 p.m. An added dimension of the problems caused by Dr. Smith in this case is that, over the course of his involvement, he provided several different opinions on the timing of Jenna’s injuries. First, after the autopsy, he told the police that there was no evidence to suggest that the injuries to the duodenum, pancreas, and liver had begun to heal, indicating that they had occurred within a few hours of death. This suggested that Jenna’s fatal injuries were inflicted while she was in the care of her babysitter. Subsequently, however, Dr. Smith’s opinion appeared to change. One month after the autopsy, in February 1997, Dr. Smith informed representatives of the DR. SMITH AND THE PRACTICE OF PEDIATRIC FORENSIC PATHOLOGY | 123 OCCO and the police that, although he could not determine the exact time of Jenna’s fatal injuries, all injuries took place within 24 hours of death. The police understood Dr. Smith’s opinion to mean that Jenna’s fatal injuries had occurred approximately 24 hours before her death. Because Ms. Waudby was the only one who had care of her daughter during that time, the police charged her with second-degree murder. In October 1998, Dr. Smith testified at the preliminary hearing in the case. His evidence on the timing of the injuries can only be described as extremely confus- ing. His testimony could be understood to say that the healing reactions to Jenna’s abdominal injuries suggested that the injuries occurred at different times. However, in all the confusion and apparent discrepancies in timing, Dr. Smith appeared to arrive at a final conclusion by assuming that all Jenna’s abdominal injuries occurred at the same time, which could have occurred some 24 or 28 hours before death. Ms. Waudby had care of her daughter during this window of time. The preliminary hearing judge thus committed Ms. Waudby to stand trial on the charge of second-degree murder. After the preliminary hearing, Dr. Smith’s opinion appeared to change once again in the face of contrary opinions. The defence retained a clinician, Dr. Sigmund Ein, a staff surgeon at the Division of General Surgery at SickKids, to examine the timing of Jenna’s fatal injuries. In December 1998, Dr. Ein spoke to Dr. Smith about the issue. Both agreed that the fatal injuries occurred on the evening of Jenna’s death, which was clearly contrary to the thrust of Dr. Smith’s evidence at the preliminary hearing. Then, in April 1999, during a meeting with Dr. Ein, Crown counsel, defence counsel, and the police, Dr. Smith again agreed with Dr. Ein’s opinion that Jenna sustained her fatal injuries after 5 p.m. on the evening of her death. Ms. Waudby did not have care of Jenna at that time. Instead, these opinions implicated Jenna’s babysitter as the perpetrator. At the April 1999 meeting, Dr. Smith and the other experts noted, however, that there were also healing rib fractures that happened earlier than the“after 5 p.m.” time frame, likely in the days before death. Although they did not cause Jenna’s death, they were relevant to the question of whether Jenna had previously been abused. In June 1999, the Crown withdrew the second-degree murder charge against Ms. Waudby. Before that withdrawal, however, Ms. Waudby pleaded guilty to a charge of child abuse under the Child and Family Services Act, RSO 1990, c. C.11, in relation to an incident that occurred in the one to three weeks before Jenna’s death. The healing rib fractures, which the experts opined were older than Jenna’s fatal injuries, served as the pathology evidence that supported her plea. In other words, according to the factual basis for the plea, although Ms. Waudby was not responsible for the fatal blows, she had abused Jenna in the past. 124 | INQUIRY INTO PEDIATRIC FORENSIC PATHOLOGY IN ONTARIO: VOLUME 2 Dr. Milroy and Dr. Michael Pollanen, Chief Forensic Pathologist, have since reviewed the case. At the Inquiry, Dr. Milroy testified that the pathology findings indicated that the fatal abdominal injuries were likely less than six hours old. There was no inflammation in Jenna’s abdominal injuries, suggesting that they had just been inflicted and that a healing reaction had not yet commenced. In addition, the information that Jenna appeared fine when Ms. Waudby handed Jenna to her babysitter at 5 p.m. supported the conclusion. If Jenna had already sustained her injuries by that time, she would have been in obvious pain. Dr. Pollanen also noted, after a review of the histology, that none of the rib fractures that Dr. Smith observed at the autopsy showed a healing reaction. Instead, they occurred at or around the time of death. As a result, the pathology evidence that, along with Ms. Waudby’s plea, formed the basis for the child abuse conviction also could not be confirmed on review. According to the expert forensic pathologists who reviewed the case, there was actually no pathology evidence to support Dr. Smith’s opinion at the preliminary hearing that Jenna’s fatal injuries could have occurred some 24 or 28 hours before death. Although there was an older liver injury that could have occurred up to several days before Jenna’s death, that was not the immediate cause of her death. In Dr. Pollanen’s view, Dr. Smith erred by grouping the abdominal injuries together and finding that they all occurred in one period of time. Dr. Smith failed to recognize that the apparent discrepancies in timing suggested that the injuries were inflicted at two different times. Dr. Smith’s misdiagnosis of the timing in Jenna’s case had significant conse- quences for the criminal and child protection proceedings. The criminal case against Ms. Waudby rested primarily on Dr. Smith’s opinion. Once it became clear that Dr. Smith’s opinion implicating Ms. Waudby was incorrect, and that Ms. Waudby did not have care of Jenna at the time of her fatal injuries, the Crown properly withdrew the second-degree murder charge. At the Inquiry, Dr. Smith was asked to explain his various opinions on tim- ing. He said that, after the autopsy, when he told the police that the injuries were only a few hours old, he had not yet conducted a microscopic examination of the wounds. Once he reviewed the histology, he reached the opinion that the injury to Jenna’s liver had a more advanced healing reaction, suggesting that it occurred in the range of 24 to 48 hours before death. He said that he never believed that the fatal injuries were all 24 to 48 hours old but always recognized, based on the healing reactions he observed microscopically, that some of the injuries were much more recent than that. In his view, however, his opinion on the timing of the fatal injuries needed to take the older liver injury into account, as that could have contributed to Jenna’s death. As a result, he extended the time period in