Perpetrator Accounts in Infant Abusive Head Trauma Brought about by a Shaking Event

Abstract

To analyze perpetrator and medical evidence collected during investigations of infant abusive head trauma (IAHT), with a view to (a) identifying cases where injuries were induced by shaking in the absence of any impact and (b) documenting the response of infant victims to a violent shaking event.
A retrospective study was undertaken of IAHT cases investigated by the Queensland Police Service over a 10-year period. Cases of head trauma involving subdural and/or subarachnoid hematoma and retinal hemorrhages, in the absence of any evidence of impact, were defined as shaking-induced. Perpetrator statements were then examined for further evidence to support the shaking hypothesis and for descriptions of the victim's immediate response to a shaking event.
From a total of 52 serious IAHT cases, 13 (25%) were found to have no medical or observer evidence of impact. In 5 of those 13 cases, there was a statement by the perpetrator to the effect that the victim was subjected to a shaking event. In several cases both with and without evidence of associated impact, perpetrator accounts described an immediate neurological response on the part of the victim.
The study confirms that IAHT resulting in death or serious neurological impairment can be induced by shaking alone. In cases where the infant's medical condition was adequately described, the symptoms of head injury presented immediately.

Full text

Child Abuse & Neglect 29 (2005) 1347–1358
Perpetrator accounts in infant abusive head trauma
brought about by a shaking event
Dean Birona,∗, Doug Sheltonb
aState Crime Operations Command, Queensland Police Service, Brisbane, Qld, Australia
bCommunity Child Health, Gold Coast Health Services, Bundall, Qld, Australia
Received 11 May 2004; received in revised form 18 April 2005; accepted 27 May 2005
Abstract
Objective: To analyze perpetrator and medical evidence collected during investigations of infant abusive head
trauma (IAHT), with a view to (a) identifying cases where injuries were induced by shaking in the absence of any
impact and (b) documenting the response of infant victims to a violent shaking event.
Method: A retrospective study was undertaken of IAHT cases investigated by the Queensland Police Service over a
10-yearperiod. Cases of head trauma involvingsubduraland/orsubarachnoid hematoma and retinal hemorrhages, in
the absence of any evidence of impact, were defined as shaking-induced. Perpetrator statements were then examined
for further evidence to support the shaking hypothesis and for descriptions of the victim’s immediate response to a
shaking event.
Results: From a total of 52 serious IAHT cases, 13 (25%) were found to have no medical or observer evidence of
impact. In 5 of those 13 cases, there was a statement by the perpetrator to the effect that the victim was subjected to a
shaking event. In several cases both with and without evidence of associated impact, perpetrator accounts described
an immediate neurological response on the part of the victim.
Conclusion: The study confirms that IAHT resulting in death or serious neurological impairment can be induced
by shaking alone. In cases where the infant’s medical condition was adequately described, the symptoms of head
injury presented immediately.
© 2005 Elsevier Ltd. All rights reserved.
Keywords: Physical abuse; Head injury; Inflicted injury; Shaken baby syndrome; Perpetrator accounts
∗Corresponding author address: School of English, Communication & Theatre, University of New England, Armidale, NSW
2351, Australia.
0145-2134/$ – see front matter © 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.chiabu.2005.05.003

1348 D. Biron, D. Shelton / Child Abuse & Neglect 29 (2005) 1347–1358
Introduction
Infant abusive head trauma (IAHT) constitutes a serious problem in contemporary society. In the
United States alone, it has been estimated that annually upward of 250 infants die after being subjected to
a violent shaking event (Lazoritz & Palusci, 2001); including impact-related and misdiagnosed fatalities
would likely lead to a much higher figure. Many more abused infants present with nonfatal head injuries
that result both in significant costs for society generally and poor long-term medical outcomes for the
victim. It has been determined that victims of IAHT require longer periods of hospitalization, at far
greater expense, than those suffering from noninflicted head trauma (Libby, Sills, Thurston, & Orton,
2003;Reece & Sege, 2000). The high level of morbidity in those infants who survive an abusive head
injury has been well documented (Bonnier, Nassogne, & Evrard, 1995;Gilles & Nelson, 1998;King,
MacKay, Sirnick, & The Canadian Shaken Baby Study Group, 2003).
Shaking-type injuries
The inherent dangers that arise from the shaking of infant children were first stated in the literature over
30 years ago (Caffey, 1972; Guthkelch, 1971). By the mid 1980s, medical professionals had become con-
fident in diagnosing shaken baby syndrome (SBS), noting physical findings that included increased head
circumference, intracranial hematoma, cerebral edema and retinal hemorrhages (Dykes, 1986;Ludwig
& Warman, 1984). More recently, the American Academy of Pediatrics (Committee on Child Abuse and
Neglect, 2001) classified SBS as “a clearly definable medical condition.” Infant shaking has been closely
linked to both subdural hematoma (Tzioumi & Oates, 1998) and retinal trauma (Levin, 2000), and in
cases of fatal inflicted head trauma these markers of abuse are commonly identified at autopsy (Case et
al., 2001). Several studies have described an absence of any impact-type injury in a significant number
of IAHT cases (Alexander, Sato, Smith, & Bennett, 1990;Brown & Minns, 1993;Gilliland & Folberg,
1996).
However, in 1987, Duhaime, Gennarelli, Thibault, Bruce, Margulies, and Wiser published a paper
that threw considerable doubt upon the existence of nonimpact SBS. Through a combination of clinical
analysis and laboratory experiments using a doll model, they concluded that “shaken baby syndrome, at
least in its most severe form, is not usually caused by shaking alone” (p. 414). The study has since been
updated, with a similar verdict reached (Prange, Coats, Duhaime, & Magulies, 2003).
A subsequent small group of critics have relied largely upon the findings of Duhaime and co-workers
to support the argument that the clinical data on shaking-induced brain injury are lacking (Leadbeatter,
James, Claydon, & Knight, 1995;Plunkett, 1999;Taff, Boglioli, & DeFelice, 1996). Some medical
witnesses have stated in courtroom testimony that shaking has no relationship to brain injury (Block,
1999). Elsewhere, it has been claimed that everyday household falls have the potential to mirror the
symptoms seen in IAHT and SBS (Plunkett, 2001; Root, 1992). The strength of the link between violent
shaking and intracranial hematoma has been questioned (Geddes et al., 2003;Howard, Bell, & Uttley,
1993;Wilkins, 1997), with the latter group also arguing that the relationship between shaking and retinal
hemorrhages remains unproven.
Some debate also persists as to the likely response of an infant to a shaking event. The neurological
reaction in a victim who has received an ultimately fatal head injury involving a subdural or subarachnoid
hematoma has been estimated to be rapid, with no lucid interval (period of normal behavior postinjury)
expected to occur (Gilles & Nelson, 1998;Reece, 2001;Willman, Bank, Senac, & Chadwick, 1997). In

D. Biron, D. Shelton / Child Abuse & Neglect 29 (2005) 1347–1358 1349
1997, over 50 specialist child protection physicians published a letter to challenge the implication, made
in a well-publicized American court case, that an infant could respond normally after receiving a life-
threatening abusive head injury (Alexander et al., 1997). Gilliland (1998, p. 724) notes blunt trauma to be
“notnecessarilyasimmediatelydisruptiveofthenervoussystem and brain functioningasviolentshaking.”
Plunkett (1998) in particular disputes the contention that severe or fatal infant head injuries involving
acute intracranial hemorrhage do not involve a lucid interval. In a letter responding to the findings of
Willman et al. (1997), he questions the validity of extrapolating data on accidental head injuries to IAHT,
claiming that the lucid interval question remains unanswered. In his own study of fatal pediatric head
injuries caused by short distance falls, Plunkett (2001) finds evidence of a lucid interval in 12 cases
(although the 3 alleged lucid intervals in infants under 2 years of age are documented as occurring over
no more than 15minutes). Mindful of the ongoing debate about the appearance of symptoms in lethal
infant shaking, Nashelsky and Dix (1995) lament the lack of available data in the scientific literature and
suggest that more specific research should be conducted.
Specifying whether trauma has been induced by shaking, impact, or a combination of both proves
problematic in many cases. The important issue in IAHT, with regard to the ongoing welfare of the infant
and any subsequent criminal inquiry, is that an abusive act has occurred (Block, 1999). Nonetheless,
a statement such as the one made by Geddes et al.—that SBS symptoms can arise “without impact or
violencebeingnecessary” (2003, p. 20)—can onlyobfuscatethe investigationand subsequentprosecution
ofIAHTcases,suggestingasitdoesthepossibilitythatatraumaticbraininjurymaynotnecessarilysignify
violence in the absence of any rational explanation. This is just one example why it is of paramount
importance that the potential mechanisms for abusive head injuries continue to be subject to research.
It is equally vital that a diagnosis of IAHT is not confirmed until the appropriate multidisciplinary
investigationhasbeen conducted, one involvingboth medical andlawenforcement authorities.Regardless
of the debate attention needs to be paid to all aspects of IAHT investigations, so that the witness and
circumstantial evidence gathered by law enforcement agencies can be closely analyzed and compared
with accepted medical knowledge in a research setting. Perpetrator statements can form an important part
of this approach. Broad statements concerning the inducement of confessions by the authorities (David,
1999) are not supported by evidence. Perpetrator statements obtained by law enforcement agencies are
routinely tape recorded, with appropriate cautions administered beforehand.
It is important to recognize that child abuse almost inevitably occurs in the absence of independent
adult witnesses. There is clearly an inherent unreliability in the content of perpetrator statements, yet
they remain the only available window through which the abusive act can be directly observed. Treated
with caution and in conjunction with the available clinical data, these statements can increase the existing
knowledge on IAHT and help future investigators piece together what took place in cases where no such
evidence is forthcoming.
This study aims to combine medical evidence with statements of perpetrators and other witnesses in
order to examine two commonly debated themes in shaking-type IAHT: (a) can shaking alone cause
serious injury or death and (b) how rapid is the neurological response to a violent shaking event?
Methods
Queensland Police Service investigation files pertaining to serious IAHT cases over the 10-year period
between July 1993 and June 2003 were examined. Case selection was limited to those investigations

1350 D. Biron, D. Shelton / Child Abuse & Neglect 29 (2005) 1347–1358
which constituted homicide or grievous bodily harm assault involving infant victims up to 2 years of age.
(Grievous Bodily Harm is defined in the Criminal Code of Queensland as an injury which if left untreated
would endanger or be likely to endanger life, or cause or be likely to cause permanent injury to health.)
Injuries were confirmed through coordinated medical assessment at a major children’s hospital and/or
via postmortem examination. Each case was the subject of notification to multidisciplinary Suspected
Child Abuse and Neglect (SCAN) teams consisting of child protection pediatricians, welfare, and law
enforcement professionals.
The use of data from Queensland Police Service records was approved by the State Crime Operations
Command Project Board and the Ethical Standards Command. Data contained in investigation files
included medical and other witness statements, transcripts of tape-recorded interviews with witnesses
and perpetrators, and (where applicable) full postmortem reports.
The initial goal was to identify instances of shaking in the absence of any impact evidence. Cases
were classified as shake-only by the presence of the following criteria: (1) subdural and/or subarachnoid
hemorrhage, (2) retinal hemorrhages, and (3) the absence of any medical (skull or scalp injury) or witness
evidenceof impact. Remaining cases wereclassified as impact-only,shake/impact and indeterminate. The
criteria required for cases to be classified as impact-only were: (1) skull or scalp injury, (2) perpetrator or
witness evidence of an impact event without any associated shaking, and (3) the absence of retinal hem-
orrhages. Shake/impact assaults involved some combination of the shake-only and impact-only criteria.
Cases with insufficient available evidence to identify a method of assault were classified as indeterminate.
All cases meeting the shake-only criteria were then examined for evidence of a confessional statement
by the perpetrator. Where such statements were available, transcripts of the accounts provided were
reviewed and the specific admission describing how the shaking occurred was extracted. The original
tape recordings were reviewed to confirm the accuracy of the transcripts.
In addition, all cases in the study, regardless of whether or not they qualified under the shake-only
criteria, were further examined for perpetrator statements alleging how an infant victim responded to
a shaking event. Again, when such evidence was present, the relevant passage of the statement was
extracted.
Results
Atotal of 52 cases were identified as fitting the overallcriteria,20(38%)ofwhichinvolvedfatalinjuries.
The infants involved were subjected to a disturbing array of violent and unprovoked attacks, as evidenced
both by the accounts of perpetrators and witnesses and by the range of injuries chronicled. Victims were
shaken,thrown,punched,head-butted,and attackedwithobjectssuchas lumps of wood.Insomeinstances,
the acute head injury was the only evidence of trauma present; other victims presented with both chronic
and acute injuries to the head and/or body. In one case, medical staff initially thought an 11-month-old
infantwith massivetrauma had been run over by a motor vehicle. The head injuries recordedincludedskull
fractures, cerebral edema, subarachnoid and subdural hematomas, and ocular trauma including retinal
hemorrhages, subhyaloid hemorrhages, detached retinas, retinal folds, subconjunctival hemorrhages, and
optic nerve hemorrhages.
The mean age of the infant victims was 5.98 months (median 3 months). There was a reasonably even
spread of fatal cases across the range (mean 7.6 months; median 3 months), although it was noted that of
the seven victims over 12 months of age, only two survived. Of the 52 victims, 27 were male.

D. Biron, D. Shelton / Child Abuse & Neglect 29 (2005) 1347–1358 1351
Table 1
Method of assault
Method of assault Number of cases Number of fatalities
Shaking only 13 5
Impact only 3 1
Evidence of shaking and impact 25 10
Indeterminate 11 4
Total 52 20
The identified mechanisms of assault are summarized in Table 1. Of the 52 cases, 13 (25%) were
defined as shaking alone; 5 of those infants did not survive. On 5 of the 13 occasions identified as shaking
mechanism only (2 fatal), tape-recorded perpetrator confessions were obtained by investigating police.
Those five cases are now summarized.
Case 1
Ambulance officers called to the family residence observed the victim (a female infant aged 11 weeks)
to have “labored breathing and a mottled appearance.” Approximately 70minutes earlier, the infant’s
mother had left the residence, leaving her in the sole care of the father. Prior to departing, the infant’s
mother fed her and confirmed her condition to be normal. Upon the mother’s return to the residence, the
father approached her in a distressed state, saying that he did not like the way the baby was breathing.
She was transported to the hospital, but her condition deteriorated over the following 24hours, and life
support was discontinued. The father later admitted to the offense. He stated that when changing the
infant’s nappy she was crying so “he gave her a bit of a shake.” When the infant continued to scream
he shook her again, at which time “she went numb ... there was no movement at all with her hands
and legs.” According to the father’s version, there was only a period of about 5minutes between the
assault and the mother returning home. At autopsy, the victim was found to have brain swelling with
global hypoxia, acute subdural and subarachnoid hematomas, multilayered retinal hemorrhages, and
some optic nerve hemorrhaging. The scalp was reported as free from bruising and there were no skull
abnormalities. The father subsequently pleaded guilty to unlawful killing and was sentenced to a term of
imprisonment.
Case 2
Thevictim (a male infant aged 8 weeks) was subjected toongoingabusebyhisfather,includingmultiple
shaking events, scalding in a bathtub, and a fracture to the left femur. The infant was hospitalized after
his mother called an ambulance to the residence because she witnessed him to be “limp like a rag doll
and having trouble breathing.” Radiological examination revealed chronic bilateral subdural collections,
but with fresh blood also present. Bilateral retinal hemorrhages were diagnosed. When interviewed, the
father discussed the events immediately leading up to the infant’s hospitalization. He stated that he had
shaken the infant “out of anger” for 2–3minutes, supporting his head with his hands, although it still
“rocked back and forth” (he later indicated that the shaking may not have taken that long). He stated the
infant “went flop in his arms,” after which he “woke up to himself and stopped.” He immediately called

1352 D. Biron, D. Shelton / Child Abuse & Neglect 29 (2005) 1347–1358
his wife into the room and showed the infant to her. The father subsequently pleaded guilty to assaulting
the infant and was sentenced to a term of imprisonment.
Case 3
The victim (a male infant aged 7 weeks) was transported to the hospital after allegedly being found
unresponsive by his father. Upon initial presentation, the infant was described by medical staff as being
“blue, floppy and cold, with large pupils.” He was subsequently found to be suffering from extensive
cerebral edema, acute subdural and subarachnoid hematomas, “extensive” bilateral retinal hemorrhages,
and a left-sided subhyaloid hemorrhage. He was intubated and ventilated for a period of 7 days. The
father, who had sole care of the infant, later admitted to shaking him. Having awoken in the middle of
the night to feed the infant, he stated that he picked him up and “gave him a little shake” because he was
screaming. The infant “went quiet for a short period and then started to cry again.” When he stopped
crying the father put him back to bed, returning some 4hours later to find the infant unresponsive and
with “white stuff coming out of his mouth.” The father then took the infant to hospital. The infant died 10
months later from complications related to the original brain injury. The father pleaded guilty to a charge
of unlawful killing and was sentenced to a term of imprisonment.
Case 4
The parents of the victim (a female infant aged 3 weeks) presented her at a medical center, where she
was found by the doctor to be “pale and lethargic, with a slow rate of respiration and periods of apnea.”
Subsequent hospital examination revealed the infant to be suffering from cerebral edema, left-sided acute
subdural hematoma, and left-sided retinal hemorrhages, with “no external manifestations of injury to
the skull.” Radiological examination revealed the infant had received a fractured clavicle some 2 weeks
earlier. The infant’s father later admitted to the offense. He stated that he held her out in front of him and
gave her “a bit of a shake to stop her crying.” He stated that it was “a fairly vigorous shake,” but that
he immediately cuddled her after realizing what he had done. He was not questioned about the infant’s
immediate response. The infant was seen in her crib by her mother some 3hours later and noted to be
“pale blue and cold to touch,” whereupon she was taken to the medical center. The father was convicted
by a jury and sentenced to a term of imprisonment.
Case 5
The victim (a female infant aged 4 months) was presented at a local hospital by her father, who stated
that she was drowsy and not feeding well. The infant arrived in the late afternoon, the father having had
sole care of her for the preceding 10hours; the father stated that the symptoms appeared approximately
1hour before he took the infant to the hospital. The first doctor to see the infant described her as “asleep
but screaming when aroused.” A subsequent Computed Tomography (CT) scan revealed evidence of both
acute and subacute subdural and subarachnoid blood. The infant was transferred to a major children’s
hospital, where she developed seizures and required ventilation, remaining under intensive care for 2
days. Ophthalmic investigations revealed bilateral retinal hemorrhages. When interviewed, the father
claimed that he often vigorously bounced the infant on his knee after feeding, causing her head to move
“up and down and back and forth.” He further stated that when performing this action, the infant would

D. Biron, D. Shelton / Child Abuse & Neglect 29 (2005) 1347–1358 1353
“get upset and cry.” The father demonstrated this action on video tape using a doll model. The video
was subsequently reviewed by a consultant pediatrician and a pediatric ophthalmologist, both of whom
found the bouncing/shaking action sufficient to have caused the injuries received. No skull or external
head injuries were noted. The father pleaded guilty and received a wholly suspended sentence.
In two other cases, detailed perpetrator statements outlining shaking-type events were obtained; how-
ever, these cases did not qualify under the shake-only criteria due to the presence of impact injuries. The
details of the two cases are described below:
Case 6
Ambulance officers attended the family residence and found the victim (a male infant aged 6 weeks)
to be “extremely pale and cyanosed.” The infant had been in the sole care of his father for a period of
approximately 1hour. He was transported to hospital but died 2 days later. The father later admitted to
the offense. He stated that he became upset at the infant’s crying and shook him “vigorously ... three
hard shakes.” He further stated that in response the infant’s arms went straight up in the air and he “just
lay there, staring at the ceiling.” The infant did not respond to attempts to revive him (by blowing air on
Table 2
Summary of seven cases discussed
Case Infant age Reported infant
response to shaking Reported delay in
seeking medical
assistance
Initial symptoms
observed upon medical
intervention
Injuries documented after
full medical assessment
1 11 weeks Became numb &
unresponsive Nil Labored breathing; mottled
appearance Cerebral edema; subdural
and subarachnoid
hematomas; retinal
hemorrhagesa
2 8 weeks Floppiness;
difficulty breathing Nil Not available Subdural hematomas;
retinal hemorrhages
3 7 weeks Stopped crying, then
resumed 4 hours Blue; floppy; cold; pupils
enlarged Cerebral edema; subdural
and subarachnoid
hematomas; retinal
hemorrhages
4 3 weeks Not available 3 hours Pale; lethargic; slow
respiration rate; periods of
apnea
Cerebral edema; subdural
hematoma; retinal
hemorrhages
5 4 months Not available 1 hour Asleep; screaming when
aroused Subdural and
subarachnoid hematomas;
retinal hemorrhages
6 6 weeks Arms stiffened;
became
unresponsive
Nil Pale and cyanosed Cerebral edema; subdural
hematoma; retinal
hemorrhages; scalp
bruisinga
7 4 months Stopped crying; fell
asleep Not applicable Not applicable Subdural hematoma;
retinal hemorrhages;
skull fracture
aInjuries observed at postmortem examination.

1354 D. Biron, D. Shelton / Child Abuse & Neglect 29 (2005) 1347–1358
his face and patting him on the back) and fluid began to flow out of his nose. The father then called an
ambulanceandcommenced cardiac compressions.At autopsy,the infantwas found to haveboth acuteand
chronic subdural blood, widespread cerebral edema, and right-sided retinal hemorrhages. Although two
areas of periosteal hemorrhage were noted below the scalp, these were determined to be consistent with
impact at an earlier stage. The skull was found to be intact. The infant’s father made no statement about
the impact trauma present. Radiological survey revealed numerous healing fractures to ribs and legs. The
father subsequently pleaded guilty to unlawful killing and was sentenced to a term of imprisonment.
Case 7
The victim (a male infant aged 4 months) was presented to a children’s hospital as irritable and
vomiting regularly. A CT scan showed evidence of both acute and chronic bilateral subdural collections.
Skeletal survey revealed a recent skull fracture. Bilateral retinal hemorrhages were also documented.
The mother of the infant admitted to shaking him approximately 10 days prior to the hospital admission.
She stated that the infant was crying so she held him under his arms and shook him “just a few times,
not very many.” She further stated that his head “was thrown back” and “when it went back he stopped
crying and then fell asleep.” The mother did not make any admissions in relation to the recent skull
fracture. Over the 10 days between the shaking event and hospitalization, the infant was presented to
medical centers on at least two occasions suffering from irritability and vomiting; however, he was
discharged without a CT scan being ordered. The mother pleaded guilty and received a wholly suspended
sentence.
The seven cases discussed above are summarized in Table 2.
Discussion
From a total of 52 cases of IAHT under review, 13 revealed the intracranial and retinal hemorrhages
characteristic of a shaking mechanism, with no medical evidence of impact trauma. In five of those cases,
offenders who were ultimately convicted and sentenced by the courts admitted to causing the injuries by
shaking their infant victims. Together, the medical and perpetrator evidence pertaining to all five incidents
provides strong evidence of shaking in the absence of any type of impact trauma, resulting in very serious
(on two occasions, fatal) brain injuries being incurred.
In Cases 1, 2, and 3, the perpetrator made a specific statement about the response of the infant victim to
being shaken. In Cases 1 and 2, the response is described as immediate. In Case 3, an immediate response
was described (the baby went quiet), even though the major symptoms were not confirmed as present
until some 4hours later. In Case 5, the perpetrator described the symptoms as appearing 1hour prior to
hospitalization. He also made a general statement about the infant becoming upset and crying whenever
he bounced her rapidly on his knee. Case 4 did not provide evidence of the infant’s immediate response,
although it can be shown that when seen 3 hours later by her mother she was in an obvious state of illness.
Cases 6 and 7 do not strictly meet the shake-only criteria; the perpetrators, however, confessed to
shakingthe infant victim at some stage. In bothinstances, there are grounds for arguing that their evidence
is relevant in assessing how infants respond to shaking events. In Case 6, the medical evidence pointed
to the impact damage occurring some time prior to the acute injuries that resulted in death, such impact
apparently being insufficient to require immediate medical intervention. The perpetrator stated that after

D. Biron, D. Shelton / Child Abuse & Neglect 29 (2005) 1347–1358 1355
being shaken the infant became unresponsive immediately, and thereafter required assistance to continue
breathing; the deterioration was both rapid and ultimately fatal. There is no evidence that the infant was
unwell in the period immediately preceding the shaking. In Case 7, the presence of chronic bilateral
subdural collections indicates a previous traumatic event, and may well correspond to the perpetrator’s
statement that she shook the infant 10 days before. The perpetrator stated that the infant responded to
being shaken by falling asleep immediately, although the injuries inflicted at that stage were apparently
not sufficient to require immediate medical intervention.
Symptoms indicating the presence of an underlying traumatic brain injury caused by rotational forces
may include alterations in respiration and or/temperature, unresponsiveness, poor feeding, irritability,
lethargy, apnea, posturing and coma (Reece, 2001). The various described responses in the seven cases
reviewedfeatureoneormoreoftheseclinicalsigns.Thetwoproximately fatal cases (1 and 6) involvestrik-
ingly similar perpetrator descriptions of an infant becoming instantaneously unresponsive and comatose.
Overall, the perpetrator evidence as described in Cases 1, 2, 3, 5, and 6 strongly supports the finding of
Gilliland (1998) that the progression from injury to symptoms requiring medical intervention in severe,
shaking-type IAHT is almost always rapid. Although there is a lack of evidence of immediate response
in Case 4, the infant presented as unwell within 3hours. All seven cases suggest that shaken babies
experience an early neurological response of some kind, although sometimes the infant becomes unwell
without requiring urgent medical assistance to maintain respiration. It may be worth noting that Case 7,
the sole example in the seven of an infant not appearing to require immediate medical attention, was also
the sole case where a female perpetrator was identified.
Anyrelationship between duration andvelocityofshakingand injuries receivedisdifficultto determine,
particularly with the relatively small number of cases in this study. The medical outcome from a shaking
event may also be influenced by, among others things, the size and weight of the perpetrator, the size and
weight of the infant, or whether the infant’s head was supported in some way. In addition, even when
perpetrators openly admit their actions they may well minimize them in some way. The perpetrators in
Cases 1, 3, and 4 use vague descriptions—such as “a little shake” and “a bit of a shake”—which alone
would not be sufficient to account for the severity of injuries received. It appears that when questioned
they have downplayed the extent of the shaking. Conversely, the perpetrator in Case 6 is more specific
in describing “three hard shakes.” As long as experts are unable to speak definitively about duration and
velocity in shaking-type assaults, any action that can be described as shaking, no matter how brief, must
be regarded as potentially hazardous.
The value of conducting multidisciplinary investigations into suspected cases of IAHT and SBS has
been widely emphasized. At the same time, there has been a paucity of interagency and nonmedical
research into these phenomenon. One of the few interagency studies published has been that of Ricci,
Giantris, Merriam, Hodge, and Doyle (2003), which explores issues such as injury types, evidence of
prior trauma and parental risk factors. More recently, Starling et al. (2004) analyzed the relationship
between perpetrator admissions and IAHT, although that study did not specify the exact details of those
admissions, the circumstances under which they took place, or if they were made to law enforcement,
welfare or medical authorities. Nevertheless, they also concluded that shaking can cause death or serious
injury and that the symptoms of inflicted head trauma are likely to be immediate. In particular, the results
of their study were comparable in that the only cases involving an apparent delay in symptoms appearing
were those where the victim was not closely observed postassault.
The importance of conducting research involving all forms of evidence in IAHT cases should not be
underestimated. For example, law enforcement officers are likely to have knowledge of crime scenes and

1356 D. Biron, D. Shelton / Child Abuse & Neglect 29 (2005) 1347–1358
histories obtained posthospital not readily available to medical researchers. Juxtaposing clinical data with
other forms of physical and witness evidence can only enhance the scientific literature.
One limitation of the present study was the lack of detailed evidence recorded in some of the 52
investigationfiles,particularly those from the earlier part of the reviewperiodwhen investigativeresources
and techniques were less developed. It is nevertheless clear that a great deal of relevant material, in the
form of witness and perpetrator accounts, can be acquired from law enforcement investigation files. It
is hoped that this study might stimulate more interagency research into IAHT. If conducted in larger
catchment areas, such research would have the potential to uncover further valuable evidence concerning
the mechanisms of, and responses to, abusive head injuries in infants.
Conclusion
The abusive shaking of an infant by an adult caregiver can result in death or serious neurological
impairment even when no associated impact takes place. In all cases in this small study in which the
infant’simmediatemedical condition was adequately described, the symptomswereseento be immediate.
Further interagency studies, combining medical, perpetrator and other witness evidence, would have the
potential to increase understanding of the mechanisms and outcomes of IAHT.
Acknowledgments
The authors thank Dr. Andrea Quinn from the University of Queensland (UQ) for her review of the
manuscript and general support, Justin McNamara of UQ for his review of the manuscript and assistance
with data collection and coding, and Erin Hitzke of UQ for her assistance with data collection and coding.
References
Alexander, R., Amaya, M., Bays, J., Bechtel, K., Beltran-Coker, R., Berkowitz, C., et al. (1997). Shaken baby syndrome—A
forensic pediatric response (Letter to the editor—Electronic version), http://child-abuse.com/sbsletter.shtml, accessed 22
November 2003.
Alexander, R., Sato, Y., Smith, W., & Bennett, T. (1990). Incidence of impact trauma with cranial injuries ascribed to shaking.
American Journal of Diseases of Children,144, 724–726.
American Academy of Pediatrics: Committee on Child Abuse and Neglect. (2001). Shaken baby syndrome: Rotational cranial
injuries—Technical report. Pediatrics,108, 206–210.
Block, R. (1999). Child abuse-controversies and impostors. Current Problems in Pediatrics,29, 253–272.
Bonnier, C., Nassogne, M., & Evrard, P. (1995). Outcome and prognosis of whiplash shaken infant syndrome: Late consequences
after a symptom-free interval. Developmental Medicine and Child Neurology,37, 943–956.
Brown,J., & Minns, R. (1993). Non-accidentalhead injury,with particular reference to whiplash shaking injury andmedico-legal
aspects. Developmental Medicine & Child Neurology,35, 849–869.
Caffey, J. (1972). On the theory and practice of shaking infants. American Journal of Diseases of Childhood,124, 161–
169.
Case, M., Graham, M., Handy, T., Jentzen, J., Monteleone, J., & the National Association of Medical Examiners Ad Hoc
Committee on Shaken Baby Syndrome. (2001). Position paper on fatal abusive head injuries in infants and young children.
The American Journal of Forensic Medicine & Pathology,22, 112–122.

D. Biron, D. Shelton / Child Abuse & Neglect 29 (2005) 1347–1358 1357
David, T. (1999). Shaken baby (shaken impact) syndrome: Non-accidental head injury in infancy. Journal of the Royal Society
of Medicine,92, 556–561.
Duhaime, A., Gennarelli, T., Thibault, L., Bruce, D., Margulies, S., & Wiser, R. (1987). The shaken baby syndrome: A clinical,
pathological and biomechanical study. Journal of Neurosurgery,66, 409–415.
Dykes, L. (1986). The whiplash shaken infant syndrome: What has been learned? Child Abuse & Neglect,10, 211–221.
Geddes, J., Tasker, R., Hackshaw, A., Nickols, C., Adams, G., Whitwell, H., & Scheimberg, I. (2003). Dural hemorrhage in non-
traumatic infant deaths: Does it explain the bleeding in ‘shaken baby syndrome?’. Neuropathology & Applied Neurobiology,
29, 14–22.
Gilles, E., & Nelson, M. (1998). Cerebral complications of nonaccidental head injury in childhood. Pediatric Neurology,19,
119–128.
Gilliland, M. (1998). Interval duration between injury and severe symptoms in nonaccidental head trauma in infants and young
children. Journal of Forensic Sciences,43, 723–725.
Gilliland, M., & Folberg, R. (1996). Shaken babies—Some have no impact injuries. Journal of Forensic Sciences,41, 114–
116.
Guthkelch, A. (1971). Infantile subdural hematoma and its relationship to whiplash injuries. British Medical Journal,2, 430–
431.
Howard, M., Bell, B., & Uttley, D. (1993). The pathophysiology of infant subdural hematoma. British Journal of Neurosurgery,
7, 355–365.
King, W., MacKay, M., Sirnick, A., & the Canadian Shaken Baby Study Group. (2003). Shaken baby syndrome in Canada:
Clinical characteristics and outcomes of hospital cases. Canadian Medical Association Journal,168, 155–159.
Lazoritz, S., & Palusci, V. (Eds.). (2001). The shaken baby syndrome: A multidisciplinary approach. Binghamton, NY: The
Hayworth Maltreatment & Trauma Press.
Leadbeatter, S., James, R., Claydon, S., & Knight, B. (1995, June 17). The shaken infant syndrome: Shaking alone may not be
responsible for damage (Letter to the editor). British Medical Journal,310, 1600.
Levin, A. (2000). Retinal hemorrhages and child abuse. Recent Advances in Paediatrics,18, 151–219.
Libby, A., Sills, M., Thurston, N., & Orton, H. (2003). Costs of childhood physical abuse: Comparing inflicted and unintentional
traumatic brain injuries. Pediatrics,112, 58–65.
Ludwig, S., & Warman, M. (1984). Shaken baby syndrome: A review of 20 cases. Annals of Emergency Medicine,13, 104–
107.
Nashelsky, M., & Dix, J. (1995). The time interval between lethal infant shaking and onset of symptoms. American Journal of
Forensic Medicine and Pathology,16, 154–157.
Plunkett, J. (1998). Response to Willman and colleagues (Letter to the editor). Child Abuse & Neglect,22, 943–944.
Plunkett, J. (1999). Shaken baby syndrome and the death of Matthew Eappen: A forensic pathologist’s response. The American
Journal of Forensic Medicine & Pathology,20, 17–21.
Plunkett, J. (2001). Fatal pediatric head injuries caused by short-distance falls. The American Journal of Forensic Medicine &
Pathology,22, 1–12.
Prange, M., Coats, B., Duhaime, A., & Margulies, S. (2003). Anthropomorphic simulations of falls, shakes, and inflicted impacts
in infants. Journal of Neurosurgery,99, 143–150.
Reece, R. (2001). Controversies in shaken baby/shaken impact syndrome. In S. Lazoritz & V. Palusci (Eds.), The shaken baby
syndrome: A multidisciplinary approach (pp. 367–388). Binghamton, NY: The Hayworth Maltreatment & Trauma Press.
Reece, R., & Sege, R. (2000). Childhood head injuries: Accidental or inflicted? Archives of Pediatric and Adolescent Medicine,
154, 11–15.
Ricci, L., Giantris, A., Merriam, P., Hodge, S., & Doyle, T. (2003). Abusive head trauma in infants: Medical, child protective
and law enforcement analysis. Child Abuse & Neglect,27, 271–283.
Root, I. (1992). Head injuries from short distance falls. American Journal of Forensic Medicine and Pathology,13, 85–
87.
Starling, S., Patel, S., Burke, B., Sirotnak, A., Stronks, S., & Rosquist, P. (2004). Analysis of perpetrator admissions to inflicted
traumatic brain injury in children. Archives of Pediatric and Adolescent Medicine,158, 454–458.
Taff, M., Boglioli, L., & DeFelice, J. (1996). Commentary on “Controversies in shaken baby syndrome” and Gilliland, M. G.
F. and Folberg, R., “Shaken babies—Some have no impact injuries” (Letter to the editor). Journal of Forensic Science,41,
729–730.

1358 D. Biron, D. Shelton / Child Abuse & Neglect 29 (2005) 1347–1358
Tzioumi, D., & Oates, R. (1998). Subdural hematomas in children under 2 years. Accidental or inflicted? A 10-year experience.
Child Abuse & Neglect,22, 1105–1112.
Wilkins, B. (1997). Head injury—Abuse or accident? Archives of Disease in Childhood,76, 393–397.
Willman, K., Bank, D., Senac, M., & Chadwick, D. (1997). Restricting the time of injury in fatal inflicted head injuries. Child
Abuse & Neglect,21, 929–940.
R´
esum´
e
French-language abstract not available at time of publication.
Resumen
Spanish-language abstract not available at time of publication.