Intimate Partner Violence,Clinical Indications,and Other Family Risk Factors Associated With Pediatric Abusive Head Trauma

Abstract

Over half of fatal pediatric traumatic brain injuries are estimated to be the result of physical abuse, i.e., abusive head trauma (AHT). Although intimate partner violence (IPV) is a well-established risk for child maltreatment, little is known about IPV as an associated risk factor specifically for AHT. We performed a single-institution, retrospective review of all patients (0–17 years) diagnosed at a Level 1 pediatric trauma center with head trauma who had been referred to an in-hospital child protection team for suspicion of AHT between 2010 and 2016. Data on patient demographics, hospitalization, injury, family characteristics, sociobehavioral characteristics, physical examination, laboratory findings, imaging, discharge, and forensic determination of AHT were extracted from the institution’s forensic registry. Descriptive statistics (mean, median), chi-square and Mann–Whitney U tests were used to compare patients with fatal head injuries to patients with nonfatal head injuries by clinical characteristics, family characteristics, and forensic determination. Multiple logistic regression was used to estimate adjusted odds ratios for the presence of IPV as an associated risk of AHT while controlling for other clinical and family factors. Of 804 patients with suspicion for AHT in the forensic registry, there were 240 patients with a forensic determination of AHT; 42 injuries were fatal. There were 101 families with a reported history of IPV; 64.4% of patients in families with reported IPV were <12 months of age. IPV was associated with a twofold increase in the risk of AHT (Exp(β) = 2.3 [p = .02]). This study confirmed IPV was an associated risk factor for AHT in a single institution cohort of pediatric patients with both fatal and nonfatal injuries. Identifying IPV along with other family factors may improve detection and surveillance of AHT in medical settings and help reduce injury, disability, and death.

Keywords

violence exposure child abuse physical abuse domestic violence prevention of child abuse

Introduction

Traumatic brain injury (TBI) is associated with the highest risk of death from accidental and nonaccidental trauma (NAT) in infants (Springer et al., 2007a) and toddlers under 4 years of age (Palusci & Covington, 2014). One study estimates that over half of fatal pediatric TBIs are the result of abuse (Snyder et al., 2014). The common presentation of TBI from an inflicted injury for children <5 years old has led the American Academy of Pediatrics to recommend health researchers use the specific term “abusive head trauma” (AHT) (Christian et al., 2009) replacing other characterizations such as “NAT,” “physical abuse,” or “physical abuse involving head trauma.” AHT can be defined as an inflicted injury to the brain or skull due to intentional impact or violent shaking (Paul & Adamo, 2014). Injuries that do not meet this definition, such as those sustained during a motor vehicle collision, unintentional acts such as an infant rolling off furniture or other injuries arising from negligent supervision, are not included in this definition. Estimates show that AHT occurs in 10–40 per 100,000 infants and young children (age <5) per year (Barlow & Minns, 2000; Duhaime et al., 1998; Eisele et al., 2006; Ellingson et al., 2008; O’Meara et al., 2020; Jayawant et al., 1998; Keenan et al., 2003; Minns et al., 2008; Sieswerda-Hoogendoorn et al., 2012). However, these estimates may be low owing to limitations of using billing codes to track occurrence (Scott et al., 2009). Previous reports also highlight the critical clinical challenge of accurately detecting AHT since many head injuries in nonverbal children may be well-appearing or patients present with symptoms similar to other conditions such as lethargy, vomiting, or fussiness (Berger et al., 2017; Keenan et al., 2004; Letson et al., 2016; Scott et al., 2009). Research suggests as many as 31% of children with AHT may be misdiagnosed on initial visit (Letson et al., 2016). As a result, milder forms of AHT are seriously under surveilled (Jenny et al., 1999; Oral et al., 2008). Failure to detect AHT on the initial presentation substantially increases the risk of repeated injury and death (Dudley et al., 2015; Erickson et al., 2005; Sheets et al., 2013).

An estimated 18%–25% of children diagnosed with AHT die, and of those that survive, up to 80% are left with significant lifelong physical, developmental, and emotional sequelae (Currie & Widom, 2010; Damashek et al., 2013; Glaser, 2000; Springer et al., 2007b), or endure injuries that may continue to impede functional domains later in life (Currie & Widom, 2010; Glaser, 2000; Springer et al., 2007b). Young survivors often face long-term cognitive impairments in domains such as communication, social behaviors, and problem solving (Keenan et al., 2019). Head trauma in early life increases the rate of major depressive disorder, drug use, suicide, sexually transmitted infections, and high-risk behaviors (Norman et al., 2012). Moreover, there are long-term financial and economic consequences that ensue from early head trauma, including high rates of unemployment, low income potential, and fewer assets in adulthood (Currie & Widom, 2010). One study estimated that two-thirds of survivors were moderately or seriously disabled (Narang & Clarke, 2014). Another recent study estimated that the lifetime cost of a single patient with AHT can exceed $3 million in medical costs, lost wages, disability, and government and educational services (Miller et al., 2018).

Research on AHT-related fatalities has characterized several demographic and clinical factors associated with death such as gender, hypoxic ischemic brain injuries, and complex subdural hematoma (SDH) (Gencturk et al., 2019). Risk factors associated with AHT (not specific to death) include patient and caregiver demographics, clinical and radiographic indicators, and socioeconomic status (Berger et al., 2011; Chen et al., 2019; Duhaime & Christian, 2019; Hymel et al., 2014; Hymel et al., 2019; Niederkrotenthaler et al., 2013; Piteau et al., 2012; Rolfes et al., 2018). Family-related risk characteristics for AHT have been less well-described. One recent study assessing perinatal factors found that the risk of AHT increased 13-fold when hospital interviewers were unable to obtain the combined data for prenatal care, partner status, social concerns and substance abuse (Kelly et al., 2015). Of family factors, violence in the home has been well established as a risk factor for child maltreatment, injury, and fatality (Evans et al., 2008; Kitzmann et al., 2003; van IJzendoorn, et al., 2019), as well as a risk factor for head trauma in adults (Wilson, 2009). Less is known about violence in the home as a specific risk for AHT. In one of the only studies to assess family violence and AHT, 11% of substantiated AHT patients also had previous exposure to family violence (Bonkowsky et al., 2008).

Intimate partner violence (IPV) is the term commonly used to describe any pattern of violence between intimate partners or spouses; IPV focuses on spousal and partner violence, historically referred to as domestic violence. IPV does not include violence between a spouse and the spouse’s extended family, among siblings or between parents and children (Ali et al., 2016). Research suggests IPV may set a behavioral norm for the parent–child relationship (van IJzendoorn et al., 2019). Part of the behavioral norm may be parental dysregulation of difficult emotions and negative affect, which has been associated with physical and emotional abuse on a child (Drake et al., 2006; Stith et al., 2009). Recent studies have also shown that IPV may be a multigenerational pattern of family violence, meaning children who were raised in a home with IPV are likely to be perpetuators of violence themselves (Pears & Capaldi, 2001). IPV indicates a general risk of injury for children and may further suggest a distinct injury risk for young children, namely AHT. Identifying violent family patterns that increase the risk of injury for infants and young children may also aid in the rapid and accurate detection of AHT. Early recognition of and intervention for IPV and its associated effects on pediatric head trauma is critical to prevent child injury and death (Letson et al., 2016; Tiyyagura et al., 2018).

This study assessed the association of IPV in the family structure with AHT in a pediatric sample of patients with head trauma who presented to the emergency department (ED) in a major United States metropolitan Level I pediatric trauma center. These patients were referred to an in-hospital Child Protection Team (CPT) for suspicion of AHT between 2010 and 2017. This study addresses a gap in our understanding of the potential family demographics that may increase the risk for AHT. In addition, by including both fatalities and nonfatalities, a more accurate characterization of risks across a wide spectrum of injuries can be made. Screening for IPV as an associated risk for AHT has important implications for early detection as well as timely coordination of medical and therapeutic interventions to protect infants and young children (King et al., 2006).

Methods

Subjects and Data Collection

We performed a single-institution, retrospective review of a cohort of pediatric patients who presented to the emergency department (ED) or were admitted to the hospital with any fatal (in-hospital deaths only) or nonfatal injuries, where physical abuse or neglect was suspected between 2010 and 2017. The nonfatal patients were drawn from the Forensic Registry at Phoenix Children’s Hospital (PCH); the fatally injured patients were drawn from the Forensic Registry with supplemental data from the trauma registry and autopsy reports. All patients where abuse or neglect was suspected were referred to an institution-based CPT for further evaluation and forensic determination. The hospital CPT consisted of attending physicians, nurse practitioners, social workers, and other clinical providers. For each patient, medical providers performed multiple fully disrobed physical examinations and ordered specialty consults, (e.g., ophthalmology, endocrinology, neurology), laboratory tests, and imaging to evaluate the extent of the patient’s injuries. In addition, a combination of clinical providers and social workers from the CPT performed clinical interviews of patient families as part of the standard of care when abuse was suspected to determine if additional social services were needed, or if a report should be made to Child Protective Services (CPS) for suspected abuse or neglect. The full CPT included the hospital CPT members, state, country, tribal and local investigative agency representatives, members of law enforcement, and child protection agency personnel to evaluate patient status and make a final forensic determination of definitive/presumptive physical abuse. The CPT met weekly during the time frame of the study.

A definite/presumptive determination of AHT was reserved strictly for patients where an adult perpetrator with agency and intent, could be specifically linked to a head injury supported by medical data. When the evaluation yielded definitive evidence of accidental injury or a head injury where intentional infliction of injury could be ruled out, these injuries were classified as non-AHT. Non-AHT-related injuries primarily included accidental falls from furniture. Injuries where inflicted injury could not be definitively eliminated or could not be adequately supported by the medical evidence were classified as undetermined under the institution’s CPT determination scheme. The forensic determination was documented in the forensic registry but not the patient’s electronic health record (EHR). Approximately a third of eligible patients were classified as undetermined (N = 181) and four patient fatalities were undetermined in the forensic registry. Using information obtained from the EHR and clinical interviews, data on patient demographics, hospitalization, injury, laboratory tests, imaging, and self-reports regarding the presence/absence of other family characteristics (e.g., substance use/abuse, diagnosis/treatment for mental illness, spousal/intimate partner violence, prior involvement with police or CPS or evaluation in ED) were entered into the forensic registry weekly during the time frame of the study.

Inclusion Criteria

All pediatric patients who presented in the ED between 2010 and 2017 with a diagnosed head or brain injury alone, or in combination with other injuries or conditions, and were referred to the institution’s CPT for evaluation were included in this study. Relevant data were extracted for all patients who met the inclusion criteria from the institution’s Forensic Registry. Fatality data was supplemented with data from the Trauma Registry and autopsy reports. There were 762 injured and 42 fatally injured patients who met the inclusion criteria. This study was completed under PCH IRB 09-055 and PCH IRB 17-015.

Statistics

Descriptive statistics, Mann–Whitney U tests, chi-square test statistics (with Yates correction) as appropriate were used to compare fatal vs. nonfatal injury forensic registry patients and forensic registry patients by case determination. For significant differences using the chi-square test statistic, one-tailed (directional) z-tests for differences in proportions were then used to determine direction. Chi-square test statistics, z-test statistics, and simple logistic regression were used to identify covariates of interest for inclusion in a multiple logistic regression model for AHT. Multiple logistic regression was then used to estimate adjusted odds ratios (AOR) for main effects while controlling for any potential confounders (a priori, from univariate or from appropriate collinearity tests). Final model fit was evaluated based on the Akaike Information Criterion (AIC) and the Hosmer–Lemeshow goodness-of-fit statistic. All patients with incomplete data were excluded from analyses on a case-wise basis based on the respective variables. SPSS version 26 was used for all data analyses. Statistical significance was set at p < .05 for all statistical tests.

Results

The computed tomography (CT) findings for all the eligible forensic registry patients compared the nonfatally injured patients with fatally injured patients (Table 1). Of the 762 nonfatalities, the head CT for 164 patients yielded normal or unremarkable results, and 2 patients were imaged using alternative methods. For the remaining 596 nonfatal patients with abnormal CT findings, there were 957 total findings. Findings were as follows: 45.6% skull fracture, 5.1% epidural hematoma, 20.2% SDH, 6.6% extra-axial hemorrhage; 3.9% subarachnoid hemorrhage, 5.7% other intracranial hemorrhage and 13.0% other brain injury. There were 60 abnormal CT findings among the 42 patients with fatal injuries. Two of the fatal brain injuries occurred from submersion, and the remaining 40 (90%) were inflicted from blunt force trauma. Table 1 shows that skull fractures occurred in 10% of fatal head injuries and 45.6% of nonfatal head injuries (z [1, 437] = –5.46, p = .001, one-tailed). Other intracranial hemorrhage occurred in 15.0% of fatal head injuries, and 5.7% of nonfatal head injuries (z [1, 437] = 2.86, p = .002, one-tailed). There were no statistically significant differences between fatal and nonfatal head injuries found for SDH (χ² [1, 437] = 1.09, p = .30), extra-axial hemorrhage (χ² [1, 437] = 0.4, p = .52); or other brain injury (χ² [1, 437] = 1.86, p = .17). The remaining imaging results occurred infrequently to compare statistically.

Table 1.

Abnormal Computerized Tomography (CT) Findings for Nonfatally and Fatally Injured Head Trauma Patients With Suspected Accidental Head Trauma (AHT).

Head Imaging by CT N = 804	Nonfatally Injured Patients N = 762 (%)	Fatally Injured Patients N = 42 (%)
Patients with normal findings	164 (21.5)	0 (0.0)
Patients unable to be imaged	2 (0.3)	0 (0.0)
Patients with abnormal findings^*	N = 596 (78.2)	N = 42 (100)
Total number of abnormal imaging findings	957	60
Specific findings (% of total findings)
Skull fracture	436 (45.6)	6 (10.0)
Epidural hematoma	49 (5.1)	1(1.7)
Subdural hematoma	193 (20.2)	15 (25.0)
Extra-axial hemorrhage	63 (6.6)	6 (10.0)
Subarachnoid hemorrhage	37 (3.9)	5 (8.3)
Other intracranial hemorrhage	55 (5.7)	9 (15.0)
Other brain injury, bleeding, hypodensity, swelling, and edema	124 (13.0)	12 (20.0)
Hypoxic ischemic encephalopathy	0 (0.0)	6 (10.0)

Note. ^*Patients may have more than one abnormal finding.

Characteristics of AHT

The patient, clinical, and family characteristics of all 804 fatal and nonfatal patients included in the study are outlined in Table 2. Of the 42 fatal patients, 96% were under the age of 5; the oldest fatality was 6 years old; the oldest nonfatally injured patient was 14 years old. The median [IQR] for age was 7 [3–10] months for nonfatal injuries which was considerably younger than fatally injured patients: 21 [2–42] months (Mann–Whitney U = 6,327; p = .006). Median length of stay (LOS) did not significantly differ between fatal (42 hours) and nonfatal (39 hours) injuries (Mann–Whitney U = 8,782; p = .67). Few other clinical differences were noted between patients with fatal head injuries vs nonfatal head injuries. Of the differences noted, bruising rates were much higher for nonfatally injured patients than fatally injured patients (z [1, 632] = –2.41, p = .008, one-tailed). The proportion of patients with severe TBI (Glasgow Coma Score [GCS] 3–8) was significantly larger for fatally injured patients (z [1, 632] = 16.46, p = .001, one-tailed). The proportion of fatal patients presenting with a symptom of difficulty breathing (z [1, 640] = 6.16, p = .001, one-tailed) or unresponsive/lethargy (z [1, 640] = 5.17, p = .001, one-tailed) was greater than the proportion of nonfatal patients presenting with these same symptoms. None of the fatalities had presented previously in the ED for a suspicious injury or had been referred to the CPT for prior forensic determination.

Table 2.

Family, Injury, and Family Characteristics of Definite/Presumptive AHT, Not Inflicted and Undetermined Head Trauma for Fatally and Nonfatally Injured Patients.

	Nonfatally Injured Patients N = 762			Fatally Injured Patients N = 42
Characteristic	Definite/Presumptive AHT	Noninflicted	Not Determined	All Determinations
	N = 240 (%)	N = 341 (%)	N = 181 (%)	N = 42 (%)
Patient age
Median [IQR] months	7 [3–10]	7 [3–16]	8 [3–12]	21 [2–42]
0–12m	170 (70.8)	281 (82.4)	134 (74.0)	17 (40.5)
13–24m	31 (12.9)	31 (9.1)	24 (13.3)	7 (16.7)
2–5y	29 (12.1)	23 (6.7)	20 (11.0)	14 (33.3)
≥5y	10 (4.2)	6 (1.7)	3 (1.7)	4 (9.5)
Patient gender
Male	131 (54.6)	208 (61.0)	110 (60.8)	23 (54.8)
Female	109 (45.4)	133 (39.0)	71 (39.2)	19 (45.2)
Patient race/ethnicity
Caucasian	90 (37.5)	113 (33.1)	58 (32.0)	16 (38.1)
Hispanic	82 (34.2)	165 (48.4)	77 (42.5)	12 (28.6)
African American	19 (7.9)	16 (4.7)	16 (8.8)	1 (2.4)
Native American	38 (15.8)	27 (7.9)	20 (11.0)	9 (21.4)
Asian	1 (0.4)	2 (0.6)	2 (1.1)	0 (0.0)
Other	10 (4.2)	18 (5.3)	7 (3.9)	4 (9.5)
Insurance
Public	188 (78.3)	242 (71.0)	151 (83.4)	30 (71.4)
Private	46 (19.2)	88 (25.8)	28 (15.5)	5 (11.9)
None/unknown	6 (2.5)	10 (2.9)	2 (1.1)	7 (16.7)
Caregiver marital status
Married	56 (23.3)	93 (27.3)	77 (42.5)	6 (14.3)
Divorced	4 (1.7)	0 (0.0)	2 (1.1)	0 (0.0)
Single	58 (24.2)	37 (10.9)	52 (28.7)	3 (7.1)
Live-in partner	50 (20.8)	24 (7.0)	43 (23.8)	28 (66.7)
Other/unreported	72 (30.0)	187 (54.8)	7 (3.9)	5 (11.9)
Transfer
Yes	147 (61.3)	90 (26.4)	99 (54.7)	28 (66.7)
Method of injury
Unknown	109 (45.4)	49 (14.4)	22 (12.2)	19 (45.2)
Fall	59 (24.6)	118 (34.6)	69 (38.1)	13 (31.0)
Other	107 (44.6)	176 (51.6)	90 (49.7)	10 (23.8)
Glasgow Coma Score
Severe (<8)	47 (19.6)	11 (3.2)	13 (7.2)	42 (100)
Moderate (8–12)	20 (8.3)	3 (0.9)	7 (3.9)	0 (0.0)
Minor (≥13)	173 (72.1)	327 (95.9)	161 (89.0)	0 (0.0)
Altered/loss of consciousness
Yes	83 (34.6)	25 (7.3)	24 (13.3)	22 (52.4)
No	48 (20.0)	240 (70.4)	67 (37.0)	0 (0.0)
Unknown	109 (45.4)	76 (22.3)	90 (49.7)	20 (47.6)
Bruising
Yes	126 (52.5)	53 (15.5)	43 (23.8)	5 (11.9)
Presenting symptoms*	N = 494	N = 533	N = 309	N = 75
Fever	12 (2.4)	11 (2.1)	12 (3.9)	2 (3.3)
Vomiting	58 (11.7)	62 (11.6)	34 (11.0)	6 (10.0)
Unresponsive/lethargy	78 (15.8)	51 (9.6)	28 (9.1)	23 (38.3)
Swelling to head	82 (16.6)	202 (37.9)	99 (32.0)	4 (6.7)
Difficulty breathing	24 (4.9)	6 (1.1)	8 (2.6)	12 (20.0)
Poor eating	13 (2.6)	9 (1.7)	10 (3.2)	2 (3.3)
Fussiness/crying	66 (13.4)	109 (20.5)	56 (18.1)	3 (5.0)
Not using extremity	7 (1.4)	3 (0.6)	2 (0.6)	0 (0.0)
Swelling to extremity	6 (1.2)	2 (0.4)	5 (1.6)	0 (0.0)
Seizures	53 (10.7)	15 (2.8)	16 (5.2)	8 (13.3)
Skin concern	60 (12.2)	25 (4.7)	19 (6.2)	11 (18.3)
Dizziness/nausea	0 (0.0)	0 (0.0)	2 (0.6)	0 (0.0)
Other	35 (7.0)	38 (7.1)	18 (5.8)	4 (6.7)
Head CT findings^a (N = 958)	N = 309	N = 435	N = 213	N = 60
Skull fracture	105 (34.0)	229 (52.6)	102 (47.9)	6 (10.0)
SDH	104 (33.7)	53 (12.2)	36 (16.9)	15 (25.0)
EDH	12 (3.9)	22 (5.0)	15 (7.0)	1 (1.7)
EAH	21 (6.8)	24 (5.5)	18 (8.5)	6 (10.0)
SAH	15 (4.9)	20 (4.6)	2 (0.9)	5 (8.3)
Other bleed	9 (2.9)	39 (9.0)	7 (3.2)	9 (15.0)
HIE	0 (0.0)	0 (0.0)	0 (0.0)	6 (10.0)
Other brain injury	43 (13.9)	48 (11.0)	33 (15.5)	12 (20.0)
Posterior rib fracture
Yes	18 (7.6)	2 (0.6)	1 (0.6)	2 (4.7)
Metaphyseal fracture
Yes	19 (8.0)	7 (2.0)	2 (1.3)	2 (4.7)
Multiple fracture
Yes	41 (17.0)	13 (3.8)	16 (8.9)	5 (11.9)
Length of stay
Median [IQR] hours	72 [43–192]	24 [18–43]	40 [23–66]	42 [18–76]
1–12	8 (3.5)	38 (11.0)	12 (6.4)	1 (2.4)
13–24	34 (14.1)	170 (49.8)	55 (30.2)	13 (31.0)
25–36	13 (5.3)	35 (10.4)	14 (7.6)	3 (7.1)
37–48	40 (16.4)	51 (15.0)	45 (25.0)	6 (14.3)
49–60	5 (2.3)	2 (0.6)	6 (3.5)	1 (2.4)
61–72	24 (9.9)	14 (4.0)	21 (11.6)	5 (11.9)
73+	116 (48.5)	31 (9.2)	28 (15.7)	13 (31.0)
Number adults in home
1–2	102 (42.5)	115 (33.7)	78 (43.1)	13 (31.0)
3–7	17 (7.1)	42 (12.3)	22 (12.2)	1 (2.4)
Unknown	121 (50.4)	184 (54.0)	81 (44.7)	28 (66.6)
Alleged perpetrator
Mother	34 (14.1)	3 (0.9)	12 (6.7)	8 (19.0)
Father	61 (25.4)	0 (0.0)	6 (3.3)	7 (16.7)
Significant other	33 (13.8)	1 (0.3)	2 (1.1)	9 (21.4)
Relative	10 (4.1)	1 (0.3)	2 (1.1)	3 (7.1)
Caregiver (unrelated)	11 (4.7)	1 (0.3)	2 (1.1)	2 (4.8)
Other/unknown	91 (37.9)	4 (1.2)	11 (6.1)	5 (11.9)
Not applicable^b	0 (0.0)	331 (97.0)	146 (80.6)	8 (19.1)
Substance use
Yes	22 (9.1)	7 (2.0)	6 (3.3)	5 (11.9)
Mental illness
Yes	10 (4.2)	15 (4.4)	5 (2.8)	0 (0.0)
Intimate partner violence
Yes	46 (19.2)	24 (7.0)	23 (12.7)	8 (19.1)
Prior CPS report
Yes	83 (34.4)	45 (13.1)	54 (30.0)	22 (52.4)
Prior police involvement
Yes	59 (24.6)	30 (8.7)	26 (14.4)	9 (21.4)

^aMultiple findings/symptoms per patient.

^bNot applicable—“alleged perpetrator” is not applicable for noninflicted injuries.

IPV was reported in approximately 20% of all families and this rate did not differ significantly between fatally (19.1%) and nonfatally (19.2%) injured patients (χ² [1, 357] = 3.38, p = .066). For patients where AHT was determined, IPV was more than double the rate of IPV reported by patient families where AHT was ruled out (19.2% vs. 7.0%), (z [1, 198] = 4.42, p = .001, one-tailed). However, it must be noted that the sensitive nature of these issues when there is a fatality or a potential criminal act, could result in underreporting and/or dissembling. There was no difference between fatally injured patients and nonfatally injured patients for injuries most frequently associated with physical abuse: posterior rib fracture (χ² [1, 640] = 0.64, p = .43); metaphyseal fracture (χ² [1, 633] = 2.24, p = .14); and multiple fractures in different body locations (χ² [1, 633] = 1.62, p = .20). For the 42 fatal head injuries, 36 (85%) were ruled AHT, 2 (5%) ruled out AHT and 4 (10%) were undetermined. Multiple fractures occurred in 12% (5/42) of fatal patients.

AHT vs. Non-AHT Patients

The comparison of patients by forensic determination (definite/presumptive AHT, non-AHT, and not determined; Table 2) showed that of the 762 nonfatally injured patients, 240 (31.5%) were ruled AHT, 341 (44.8%) ruled out AHT, and 181 (23.7%) were undetermined. Seventy-one percent of nonfatal AHT patients were < 12 months; 84% < 2 years; 96% < 5 years. Fifteen patients (19.7%) presented to the ED a second time during the study period with a subsequent head injury. Ten patients (66.7%) presented within 30 days; four within 90 days, one within the year. Of these 15 patients, 4 had CT findings on initial presentation for fracture in combination with SDH, 2 patients had findings for SDH alone, 3 had other bleeds including extra-axial hemorrhage, and the remaining 6 patients had unremarkable/no findings.

Less than 25% (23.2%) of patients with AHT had married parents, a proportion that was not significantly different from patients where the injury was non-AHT (27.4%) (χ² [1, 355] = 0.29, p = .59). The majority of patients lived with a biological parent who was either single or had a live-in partner. Unreported parental marital status was highest among patients with noninflicted injuries (55%). Patients who lived in a household where marital status was reported as “live-in partner” represented 67% of fatal injuries. AHT patients had higher rates of public insurance compared to non-AHT (z [1, 580] = 1.99, p = .02, one-tailed), and lower rates of private insurance (z [1, 580] = –1.87, p = .03, one-tailed). AHT patients were more likely to have been a transfer from an outside hospital (z [1, 499] = 8.42, p = .001).

Median [IQR] length of stay (LOS) for AHT patients (72 [43–192]) was significantly longer than that of patients without AHT (24 [18–43]) (Mann–Whitney U test = 16.56, p = .001). Other clinical characteristics where the rates for AHT were higher than non-AHT in Table 2 include: Glasgow Coma Score < 8 (z [1, 501] = 6.48, p = .001, one-tailed); altered or loss of consciousness (z [1, 335] = 8.31, p = .001, one-tailed); SDH (z [1, 577] = 7.03, p = .001, one-tailed); and other intracranial bleeding. By contrast, rates of skull fracture (z [1, 577] = –5.62, p = .001, one-tailed) and other intracranial bleeding (z [1, 577] = –3.31, p = .001, one-tailed) were lower for the AHT patients compared to non-AHT. Symptoms that occurred in significantly higher rates for AHT patients included: unresponsive/lethargy (z [1, 577] = 5.01, p = .001, one-tailed); difficulty breathing (z [1, 577] = 4.42, p = .001, one-tailed); seizures (z [1,577] = 6.53 p = .001, one-tailed); and skin concern (z [1, 577] = 5.93, p = .001, one-tailed). Swelling to the head (z [1, 577] = –5.95, p = .001, one-tailed) occurred in higher proportions for non-AHT patients. AHT patients presented with higher rates of several injury characteristics: posterior rib fracture (z [1, 577] = 4.51, p = .001, one-tailed); metaphyseal fracture (z [1, 458] = 3.37, p = .001, one-tailed); multiple fractures in different body locations (z [1, 559] = 5.42, p = .001, one-tailed); and bruising (z [1, 479] = 9.52, p = .001, one-tailed). Almost 40% of patients with AHT did not report a method of injury (MOI), did not observe the injury or did not know the MOI, while only 14% of patients where AHT could be ruled out had an unknown MOI (z [1, 577] = 8.28, p = .001, one-tailed). Falls were the most frequently reported MOI for noninflicted injuries (34%).

The majority (53%) of individuals that inflicted the injury were parents or cohabitants. Of this 53%, the father was the most frequently identified alleged perpetrator for inflicted head trauma (25%), with mother (14.1%), and significant other (13.8%), being identified in roughly equivalent proportion (χ² [1, 270] = 0.02, p = .90). IPV was reported in 19% of patients with AHT, compared to 7% in patients where the injury was noninflicted (z [1, 198] = 4.42, p = .001, one-tailed). Other family factors where differences were noted between AHT and non-AHT included: reported prior CPS contact (z [1, 526 ] = 6.12, p = .001, one-tailed); prior police involvement (z [1, 294] = 5.20, p = .001, one-tailed) and substance use (z [1, 270] = 3.88, p = .001, one-tailed).

IPV as an Associated Risk for AHT

The adjusted odds ratios for clinical and family risks associated with definite/presumptive AHT were calculated (referent = noninflicted; Table 3). Only significant factors from Table 2 based on unadjusted odds rations (ORs) were included in the final model. The results confirm that IPV was an associated risk for AHT (Exp(β) = 2.3, p = .02) adjusting for other family risk factors: substance abuse (Exp(β) = 5.7, p = .001); unknown number of adults in the home (Exp(β) = 4.1, p = .001); prior police involvement (Exp(β) = 5.9, p = .001); and clinical risk factors: the presence of multiple fractures (Exp(β) = 9.9, p = .001); bruising of any kind (Expβ = 5.7, p < .001); SDH (Exp(β) = 5.3, p = .001); and seizures (Exp(β) = 4.9, p = .02). Unknown MOI was also an associated risk (Exp(β) = 3.9, p = .001). Skull fracture was protective (Exp(β) = 0.5, p = .011).

Table 3.

Adjusted Odds Ratios (AOR) for Clinical Risk Factors Associated With Definite/Presumptive AHT.

	N = 619
Risk/Protective Factor	AOR	p Value	95% CI
Skull fracture SDH Seizures Multiple fracture	0.5 5.3 4.9 9.9	.011 .001 .02 .000	0.3–0.9 2.6–10.5 2.1–11.5 4.1–23.7
Bruise	5.7	.000	3.1–10.5
Substance abuse	5.7	.001	31.7–19.9
Unknown # of adults	4.1	.000	2.2–7.4
MOI unknown	6.4	.000	2.9–14.2
Prior policeIPV	5.92.3	.001.018	2.8–12.11.0–5.2
Constant	0.06	.000	–

Note. CI = confidence interval; MOI = method of injury; IPV = intimate partner violence; SDH = subdural hematoma; CPS = Child Protective Services.

We also estimated a logistic regression to assess any risks specifically associated with a subset of AHT patients where IPV was reported (referent = AHT where IPV was not reported). The results are reported in Table 4. Associated risks for AHT in this subcohort included: seizures (Exp(β) = 3.0, p = .001; bruising (Exp(β) = 4.3, p = .001); and prior police involvement (Exp(β) = 6.4, p = .001). Swelling to the head was associated with non-AHT (Exp(β) = –0.39, p = .031).

Table 4.

Adjusted Odds Ratios (AOR) for Definite/Presumptive AHT Where IPV Was Reported (Referent = Definite/Presumptive AHT but IPV Not Reported).

N = 53
Risk/Protective Factor	AOR	p Value	95% CI
Swelling to head	0.4	.031	0.2–0.9
Seizures	3.0	.001	1.4–6.8
Bruising	4.3	.001	2.2–8.
Police involvement	6.4	.001	3.2–12.6
Constant	0.03	.001	–

Discussion

IPV, defined as any pattern of violence between intimate partners or spouses, has been described as a common risk factor for subsequent injuries in very young children (Office on Child Abuse and Neglect, 2003) and for TBI in adults (Wilson, 2009). To date, IPV-associated risk for pediatric TBI has not been well-established. In this study, we assessed the risk of head trauma associated with IPV for pediatric. We compared patients where AHT had been determined compared to patients where AHT was ruled out using data from a single institution’s forensic registry. Using this definition of IPV, we found families with reported IPV had twice the likelihood of AHT. We also found seizures, bruising and prior police involvement were associated risks for AHT when IPV was reported. Our results suggest that in addition to clinical indicators, comprehensive family histories could aid in the early detection of AHT.

Clinical Indicators of AHT

Using confirmed radiologic indicators (e.g., SDH) and presenting symptoms (e.g., seizure) of head trauma, this research assessed clinical indicators associated with AHT. Seizures have been widely recognized as a clinical indication of an underlying AHT (Christian, Committee on Child Abuse and Neglect, 2015; Duhaime & Christian, 2019). In this study, a clinical presentation of seizures was associated with a three-fold elevation in the detection of AHT. Research suggests most SDH found in infants and toddlers is associated with a mechanism of child abuse (Feldman et al., 2001). In our study cohort, we also found SDH was an associated risk for AHT as compared to noninflicted TBI. SDH has been identified as a radiologic finding often associated with AHT when no prior history of TBI is reported, or when the history did not match the injury pattern (i.e.) fall from a low height (Kemp et al., 2011). We found SDH was an associated risk for AHT but when we restricted the definite/presumptive AHT cohort to families with reported IPV, SDH was not significant and head swelling was a negative predictor (Exp(β) = 0.4 [p = .03]).

Taken together, the results for SDH and head swelling for the IPV subcohort suggest IPV-related TBI may be unintentional. Both intentional and unintentional injury can be associated with IPV (Notrica et al., 2020). Unintentional injuries can occur in the context of a neglect or inattention, a domestic dispute such as a toy or phone thrown at an adult, hitting the child, or physical force such as one partner pushing the other causing the caregiver to drop an infant or fall while holding an infant (Office on Child Abuse and Neglect, 2003). Falls constitute at least 37% of all injury-related pediatric ED visits (Cherry & Wang, 2016; Damashek et al., 2013). Falls can result in serious injuries and fatalities as a result of head trauma (Cherry & Wang, 2016; Damashek et al., 2013). Falls made up 30% of all injury mechanisms in our study cohort. Our results provide additional support for research suggesting mild SDH may be unrelated to AHT (Gencturk et al., 2019). We also found isolated head fractures were associated with a 50% reduction of AHT as a mechanism for the injury. Research suggests skull fractures are more likely associated with noninflicted injury than AHT because the fracture is often associated with fall as the MOI (Hymel et al., 2014; Notrica et al., 2020).

Similar to prior research, bruising was an associated risk of AHT in our study cohort (Morad et al., 2010). While there were significantly higher rates of skeletal injuries in AHT patients (e.g., posterior rib fractures, metaphyseal fractures), these injuries were not risks associated with AHT in our study cohort. Their infrequent occurrence likely limited the utility of skeletal injuries for detection of AHT. GCS was not an associated risk for AHT in our study cohort, but GCS has been described as more reliable in severe head injury and/or older patients (Kochar et al., 2020; Simpson et al., 1991). Eighty-two percent of our patients had a GCS score >13 reflecting mild head trauma. Many presenting symptoms were not associated risks for AHT. When patients present in the ED without outward manifestation of a head injury (head swelling, head bruise), imaging may not be ordered since it may not be clinically indicated and it may expose the patient to unnecessary radiation. Clinical diagnosis of AHT can also be complicated by prior head injuries that went untreated.

Reported substance abuse was also an associated risk according to previous research on child maltreatment (Stith et al., 2009). Our results suggest a pattern of family violence may better highlight a family structure where the risk of AHT can be elevated in priority over prior involvement with CPS. In our study, prior CPS reports were not an associated risk for AHT although research on general child maltreatment supports CPS as an AHT associated risk (van IJzendoorn et al., 2019). It should be noted that for both prior CPS reports and prior police reports, we were able to confirm self-reported information against CPT-provided background on the patient’s family from police and CPS records. We speculate that the ages of the cohort may have limited the extent of prior reporting, or past reporting may have included reporting for neglect and unsubstantiated reports as opposed to substantiated abuse reports.

For 31%–43% of the families in our study cohort, 1–2 adults were reported in the home (Table 2), while 44%–67% of families had missing data for the number of adults in the household or reported an unknown number of adults. Income inequality, the lack of affordable housing, and material conditions in the home, including overcrowding have been linked to disorganized households and impermanent housing (Featherstone et al., 2019). Children who live in homes with a greater total number of individuals are more likely to be victims of fatal and nonfatal injury in general (Featherstone et al., 2019). Kelly and colleagues (2015), found the absence of information was associated with AHT. An unknown number of adults in the home and an unknown MOI were both significant associated risks for AHT in our study. It is widely accepted that caregivers often provide no or inaccurate injury histories (Berger et al., 2017). A recent qualitative analysis of caregiver narratives for MOI suggested that narratives supporting unintentional injuries tend to provide detail on the MOI and the detail more accurately matched the observed injury features (Notrica et al., 2020). By contrast, narratives provided to support injuries from physical abuse were limited in detail and were more frequently associated with caregiver lack of knowledge regarding how the injury occurred (Notrica et al., 2020).

Fatal vs. Nonfatal AHT

Few studies have investigated differences in fatal and nonfatal AHT (Ferguson et al., 2016). In this study, we investigated differences in demographic and injury patterns for fatal and nonfatal injured patients. Overall, there were few significant differences in imaging, demographics, or clinical characteristics between nonfatally injured and fatally injured patients in our forensic registry, a result somewhat consistent with Ferguson et al. (2016). However, our fatally injured AHT patients were considerably older than our nonfatal patients. This age difference may be attributable to a pattern of abuse (targeting) by unrelated adults in the home and caregiver neglect regarding the child’s health (Jenny & Issac, 2006; Jonson-Reid et al., 2007). Jenny & Isaac (2006) also found that an unrelated adult in the home not only increased the risk of child death but that stepfathers target older children (Jenny & Isaac, 2006). Fatally injured AHT patients demonstrated more contemporaneous bleeding on CT scans as previous research has found (Rolfes et al., 2018). Fatally injured patients in our study cohort presented with clinical conditions linked with AHT in previous clinical reviews (Chen et al., 2019; Duhaime & Christian, 2019; Hymel et al., 2014, 2019), e.g., GCS < 8, bruising and altered or loss of consciousness. Our results suggest that contemporaneous characteristics of fatal and nonfatal AHT patients other than age are similar, but data that track the visit histories of AHT patients will better understand injury patterns over time and healthcare utilization by families with histories of IPV. Additional research is needed to improve our understanding of family violence and deploy interventions to address preventable injuries.

Limitations

The current study has limitations that affect the interpretation of the results. This was a single-institution study and the results may not generalize to other settings. There could be a significant risk of type I error, namely by excluding children who should have been identified. CTs are somewhat limited in detecting older accumulations. IPV, prior police involvement, prior CPS reports, substance use and mental illness may have been underreported given the sensitive nature of this information despite a thorough background assessment performed by the CPT. We had a very small number of patients with repeat visits for suspected abuse and the cohort of patient deaths did not allow us to track patient transfer histories. Additionally, a third of patients who suffered serious head trauma had characteristics fitting both AHT and accidental determinations, suggesting that our institution’s current clinical pathway for AHT lacks sufficient sensitivity and specificity to detect milder forms of AHT. In addition, it is likely that the rate of fatal AHT may be much higher than reported because many child fatalities may go unreported or are classified as accidental.

Conclusions

IPV is a family risk that warrants additional attention in pediatric medical settings to prevent head trauma. Failure to document histories of IPV in families with very young children may miss opportunities to identify violent family patterns and subsequently protect minors at high risk for injury and death. TBI constitutes the overwhelming majority of fatal and nonfatal head injuries for children under the age of 5 years. Among known and potential risk factors for AHT, this study has confirmed IPV as an associated risk for a cohort of 804 children screened for suspected abuse at a single institution. Early identification of family and clinical risk factors associated with AHT may help prevent further injury, disability and death in very young children.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Abbreviations

ORCID iD

Lois W. Sayrs

Author Biographies

Lois W. Sayrs, PhD, is the senior clinical research scientist/biostatistician in pediatric surgery and trauma research at Phoenix Children’s Hospital in Phoenix, Arizona and research assistant professor in child health at the University of Arizona College of Medicine—Phoenix. Her monograph, Pooled Time Series (1996), is in its ninth edition and has been cited in over 800 published papers and books.

J. Bryce Ortiz, PhD, is a postdoctoral researcher at the University of Arizona College of Medicine—Phoenix and Barrow Neurological Institute at Phoenix Children’s Hospital. Bryce’s research focuses on how hormones are affected following pediatric traumatic brain injuries and how these changes affect sleep and development.

David M. Notrica, MD, FACS, FAAP, is Trauma Medical director at Phoenix Children’s Hospital. He is a professor of child health and surgery at the Arizona College of Medicine Phoenix and associate professor of surgery at Mayo Clinic, where he is also the pediatric surgery fellowship program director. His research is in pediatric trauma as well as trauma systems.

Lisa Kirsch, MD, is the division chief of the Child Protection Team at Phoenix Children’s Hospital. She is also a clinical assistant professor in the Department of Child Health at the University of Arizona, College of Medicine—Phoenix. She is board certified in both general pediatrics and child abuse pediatrics. Her research interests include child maltreatment and interpersonal violence.

Cara Kelly, PhD, MSW, is an associate faculty member in the School of Social Work at Arizona State University and a collaborator with Phoenix Children’s Hospital trauma research. Her research focuses on child maltreatment prevention and intervention, child welfare risk and safety assessments, child well-being, and the application of epidemiological methods to improve the surveillance of child abuse and neglect.

Rachael Stottlemyre, BS, is a clinical research assistant in Surgery and Trauma at Phoenix Children’s Hospital. She completed her undergraduate degree in psychology at Barrett, The Honors College at Arizona State University in 2018 and is currently applying to medical school.

Aaron Cohen, BS, is an endoscopy technician at Valleywise Health Medical Center. He completed an undergraduate degree in psychology at Barrett, The Honors College at Arizona State University in 2018, where he was a research assistant in the Memory and Attention Control Lab. Aaron is currently applying to medical school.

Shivani Misra, MD, MPH, was a Flynn scholar at the University of Arizona. She is a graduate of UACOM-P and is currently a resident in neurodevelopmental disabilities at UT Southwestern, Dallas, TX.

Tabitha R. Green, BS, completed an undergraduate degree in Biology from the University of Bath, UK. She is currently a graduate student in medical neuroscience at the University of Arizona College of Medicine—Phoenix.

P. David Adelson, MD, is the director of Barrow Neurological Institute at Phoenix Children’s Hospital and a professor at the University of Arizona College of Medicine—Phoenix. Dr Adelson is a globally recognized expert in the areas of pediatric neural injury, surgical management of pediatric epilepsy, brachial plexus and peripheral nerve, and brain tumors.

Jonathan Lifshitz, PhD, is professor of child health at the University of Arizona College of Medicine Phoenix and a senior research scientist at Phoenix Children’s Hospital. He leads research on the incidence, pathology, and treatment of traumatic brain injury, with specific focus on translational opportunities to generate new knowledge, develop devices, and inform practice-based decisions.

Rachel K. Rowe, PhD, is a research assistant professor at Barrow Neurological Institute at Phoenix Children’s Hospital and the Department of Child Health at the University of Arizona College of Medicine—Phoenix. Her research focuses on physiological homeostasis of traumatic brain injury, aging, and neurodegenerative disease. Specifically, the disruption of sleep and endocrine homeostatic regulation as mediated by inflammation.

References

Abuse, Office on Child Neglect Children’s Bureau. (2003). Child protection in families experiencing domestic violence . U.S. Department of Health & Human Services.

Ali

P. A.

, Dhingra

, & McGarry

(2016). A literature review of intimate partner violence and its classifications. Aggression and Violent Behavior , 31, 16–25.

Barlow

K. M.

, & Minns

R. A.

(2000). Annual incidence of shaken impact syndrome in young children. Lancet , 356(9241), 1571–1572. https://doi.org/10.1016/S0140-6736(00)03130-5

Berger

R. P.

, Fromkin

J. B.

, Stutz

, Makoroff

, Scribano

P. V.

, Feldman

, Tu

L. C.

, & Fabio

(2011). Abusive head trauma during a time of increased unemployment: A multicenter analysis. Pediatrics , 128(4), 637–643. https://doi.org/10.1542/peds.2010-2185

Berger

R. P.

, Pak

B. J.

, Kolesnikova

M. D.

, Fromkin

, Saladino

, Herman

B. E.

, Pierce

M. C.

, Englert

, Smith

P. T.

, & Kochanek

P. M.

(2017). Derivation and validation of a serum biomarker panel to identify infants with acute intracranial hemorrhage. Journal of the American Medical Association Pediatrics , 171(6), e170429. https://doi.org/10.1001/jamapediatrics.2017.0429

Bonkowsky

J. L.

, Guenther

, Filloux

F. M.

, & Srivastava

(2008). Death, child abuse, and adverse neurological outcome of infants after an apparent life-threatening event. Pediatrics , 122(1), 125–131. https://doi.org/10.1542/peds.2007-3376

Chen

C. C.

, Hsieh

P. C.

, Chen

C. P. C.

, Hsieh

Y. W.

, Chung

C. Y.

, Lin

K. L.

, Abuse, Prevention Protection Against Child, & Group., Neglect (PCHAN) Study (2019). Clinical characteristics and predictors of poor hospital discharge outcome for young children with abusive head trauma. Journal of Clinical Medicine , 8(3), 390. https://doi.org/10.3390/jcm8030390

Cherry

, & Wang

(2016). The link between male employment and child maltreatment in the US, 2000–2012. Children and Youth Services Review , 66, 117–122. https://doi.org/10.1016/j.childyouth.2016.05.008

Christian

C. W.

, & Block

Committee on Child, A., Committee on Child Abuse and Neglect, Pediatrics. (2009). Abusive head trauma in infants and children. Pediatrics , 123(5), 1409–1411. https://doi.org/10.1542/peds.2009-0408

10.

Christian, Committee on Child Abuse and Neglect. (2015). The evaluation of suspected child physical abuse. Pediatrics , 135(5), e1337–1354. https://doi.org/10.1542/peds.2015-0356

11.

Currie

, & Widom

C. S.

(2010). Long-term consequences of child abuse and neglect on adult economic well-being. Child Maltreatment , 15(2), 111–120. https://doi.org/10.1177/1077559509355316

12.

Damashek

, Nelson

M. M.

, & Bonner

B. L.

(2013). Fatal child maltreatment: Characteristics of deaths from physical abuse versus neglect. Child Abuse and Neglect , 37(10), 735–744. https://doi.org/10.1016/j.chiabu.2013.04.014

13.

Drake

, Jonson-Reid

, & Sapokaite

(2006). Re-reporting of child maltreatment: Does participation in other public sector services moderate the likelihood of a second maltreatment report? Child Abuse and Neglect , 30(11), 1201–1226. https://doi.org/10.1016/j.chiabu.2006.05.008

14.

Dudley

, Ackerman

, Brown

K. M.

, & Snow

S. K.

American Academy of Pediatrics Committee on Pediatric Emergency Medicine; American College of Emergency Physicians Pediatric Emergency Medicine Committee; Emergency Nurses Association Pediatric Committee. (2015). Patient-and family-centered care of children in the emergency department. Pediatrics , 135(1), e255–e272.

15.

Duhaime

A. C.

, & Christian

C. W.

(2019). Abusive head trauma: Evidence, obfuscation, and informed management. Journal of Neurosurgery Pediatrics , 24(5), 481–488. https://doi.org/10.3171/2019.7.peds18394

16.

Duhaime

A. C.

, Christian

C. W.

, Rorke

L. B.

, & Zimmerman

R. A.

(1998). Nonaccidental head injury in infants—The “shaken-baby syndrome.” New England Journal of Medicine , 338(25), 1822–1829. https://doi.org/10.1056/NEJM199806183382507

17.

Eisele

J. A.

, Kegler

S. R.

, Trent

R. B.

, & Coronado

V. G.

(2006). Nonfatal traumatic brain injury-related hospitalization in very young children-15 states, 1999. Journal of Head Trauma Rehabilitation , 21(6), 537–543. https://doi.org/10.1097/00001199-200611000-00008

18.

Ellingson

K. D.

, Leventhal

J. M.

, & Weiss

H. B.

(2008). Using hospital discharge data to track inflicted traumatic brain injury. American Journal of Preventative Medicine , 34(4 Suppl.), S157–162. https://doi.org/10.1016/j.amepre.2007.12.021

19.

Erickson

S. J.

, Gerstle

, & Feldstein

S. W.

(2005). Brief interventions and motivational interviewing with children, adolescents, and their parents in pediatric health care settings: A review. Archives of Pediatrics and Adolescent Medicine , 159(12), 1173–1180. https://doi.org/10.1001/archpedi.159.12.1173

20.

Evans

S. E.

, Davies

, & DiLillo

(2008). Exposure to domestic violence: A meta-analysis of child and adolescent outcomes. Aggression and Violent Behavior , 13(2), 131–140.

21.

Featherstone

, Morris

, Daniel

, Bywaters

, Brady

, Bunting

, Mason

, & Mirza

(2019). Poverty, inequality, child abuse and neglect: Changing the conversation across the UK in child protection? Children and Youth Services Review , 97, 127–133. https://doi.org/10.1016/j.childyouth.2017.06.009

22.

Feldman

K. W.

, Bethel

, Shugerman

R. P.

, Grossman

D. C.

, Grady

M. S.

, & Ellenbogen

R. G.

(2001). The cause of infant and toddler subdural hemorrhage: A prospective study. Pediatrics , 108(3), 636–646. https://doi.org/10.1542/peds.108.3.636

23.

Ferguson

N. M.

, Shein

S. L.

, Kochanek

P. M.

, Luther

, Wisniewski

S. R.

, Clark

R. S.

, Tyler-Kabara

E. C.

, Adelson

P. D.

, & Bell

M. J.

(2016). Intracranial hypertension and cerebral hypoperfusion in children with severe traumatic brain injury: Thresholds and burden in accidental and abusive insults. Pediatric critical care medicine: A Journal of the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies , 17(5), 444.

24.

Gencturk

, Tore

H. G.

, Nascene

D. R.

, Zhang

, Koksel

, & McKinney

A. M.

(2019). Various cranial and orbital imaging findings in pediatric abusive and non-abusive head trauma, and relation to outcomes. Clinical Neuroradiology , 29(2), 253–261.

25.

Glaser

(2000). Child abuse and neglect and the brain—A review. Journal of Child Psychology and Psychiatry , 41(1), 97–116.

26.

Hymel

K. P.

, Armijo-Garcia

, Foster

, Frazier

T. N.

, Stoiko

, Christie

L. M.

, Harper

N. S.

, Weeks

, Carroll

C. L.

, Hyden

, Sirotnak

, Truemper

, Ornstein

A. E.

, Wang

, & Investigators., Pediatric Brain Injury Research Network (2014). Validation of a clinical prediction rule for pediatric abusive head trauma. Pediatrics , 134(6), e1537–1544. https://doi.org/10.1542/peds.2014-1329

27.

Hymel

K. P.

, Wang

, Chinchilli

V. M.

, Karst

W. A.

, Willson

D. F.

, Dias

M. S.

, Herman

B. E.

, Carroll

C. L.

, Haney

S. B.

, Isaac

, & Investigators., Pediatric Brain Injury Research Network (2019). Estimating the probability of abusive head trauma after abuse evaluation. Child Abuse and Neglect , 88, 266–274. https://doi.org/10.1016/j.chiabu.2018.11.015

28.

Jayawant

, Rawlinson

, Gibbon

, Price

, Schulte

, Sharples

, Sibert

J. R.

, & Kemp

A. M.

(1998). Subdural haemorrhages in infants: Population based study. BMJ , 317(7172), 1558–1561. https://doi.org/10.1136/bmj.317.7172.1558

29.

Jenny

, Hymel

K. P.

, Ritzen

, Reinert

S. E.

, & Hay

T. C.

(1999). Analysis of missed cases of abusive head trauma. Journal of the American Medical Association , 281(7), 621–626.

30.

Jenny

, & Isaac

(2006). The relation between child death and child maltreatment. Archives of disease in childhood , 91(3), 265–269. https://doi.org/10.1136/adc.2004.066696

31.

Jonson-Reid

, Chance

, & Drake

(2007). Risk of death among children reported for nonfatal maltreatment. Child Welfare , 12(1), 86–95.

32.

Keenan

H. T.

, Presson

A. P.

, Clark

A. E.

, Cox

C. S.

, & Ewing-Cobbs

(2019). Longitudinal Developmental outcomes after traumatic brain injury in young children: Are infants more vulnerable than toddlers? Journal of Neurotrauma , 36(2), 282–292. https://doi.org/10.1089/neu.2018.5687

33.

Keenan

H. T.

, Runyan

D. K.

, Marshall

S. W.

, Nocera

M. A.

, & Merten

D. F.

(2004). A population-based comparison of clinical and outcome characteristics of young children with serious inflicted and noninflicted traumatic brain injury. Pediatrics , 114(3), 633–639. https://doi.org/10.1542/peds.2003-1020-L

34.

Keenan

H. T.

, Runyan

D. K.

, Marshall

S. W.

, Nocera

M. A.

, Merten

D. F.

, & Sinal

S. H.

(2003). A population-based study of inflicted traumatic brain injury in young children. Journal of the American Medical Association , 290(5), 621–626. https://doi.org/10.1001/jama.290.5.621

35.

Kelly

, John

, Vincent

A. L.

, & Reed

(2015). Abusive head trauma and accidental head injury: A 20-year comparative study of referrals to a hospital Child Protection Team. Archive of Disease in Childhood , 100(12), 1123–1130. https://doi.org/10.1136/archdischild-2014-306960

36.

Kemp

A. M.

, Jaspan

, Griffiths

, Stoodley

, Mann

M. K.

, Tempest

, & Maguire

S. A.

(2011). Neuroimaging: What neuroradiological features distinguish abusive from non-abusive head trauma? A systematic review. Archive of Disease in Childhood , 96(12), 1103–1112. https://doi.org/10.1136/archdischild-2011-300630

37.

King

W. K.

, Kiesel

E. L.

, & Simon

H. K.

(2006). Child abuse fatalities—Are we missing opportunities for intervention? Pediatric Emergency Care , 22(4), 211–214. https://doi.org/10.1097/01.pec.0000208180.94166.dd

38.

Kitzmann

K. M.

, Gaylord

N. K.

, Holt

A. R.

, & Kenny

E. D.

(2003). Child witnesses to domestic violence: A meta-analytic review. Journal of Consulting and Clinical Psychology , 71(2), 339–352.

39.

Kochar

, Borland

M. L.

, Phillips

, Dalton

, Cheek

J. A.

, Furyk

, Neutze

, Lyttle

M. D.

, Hearps

, Dalziel

, Bressan

, Oakley

, & Babl

F. E.

(2020). Association of clinically important traumatic brain injury and Glasgow Coma Scale scores in children with head injury. Emergency Medicine Journal , 37(3), 127–134. https://doi.org/10.1136/emermed-2018-208154

40.

Letson

M. M.

, Cooper

J. N.

, Deans

K. J.

, Scribano

P. V.

, Makoroff

K. L.

, Feldman

K. W.

, & Berger

R. P.

(2016). Prior opportunities to identify abuse in children with abusive head trauma. Child Abuse and Neglect , 60, 36–45. https://doi.org/10.1016/j.chiabu.2016.09.001

41.

Maguire

S. A.

, Kemp

A. M.

, Lumb

R. C.

, & Farewell

D. M.

(2011). Estimating the probability of abusive head trauma: A pooled analysis. Pediatrics , 128(3), e550–564. https://doi.org/10.1542/peds.2010-2949

42.

Miller

T. R.

, Steinbeigle

, Lawrence

B. A.

, Peterson

, Florence

, Barr

, & Barr

R. G.

(2018). Lifetime cost of abusive head trauma at ages 0–4, USA. Prevention Science , 19(6), 695–704. https://doi.org/10.1007/s11121-017-0815-z

43.

Minns

R. A.

, Jones

P. A.

, & Mok

J. Y.

(2008). Incidence and demography of non-accidental head injury in southeast Scotland from a national database. American Journal of Preventative Medicine , 34(4 Suppl.), S126–133. https://doi.org/10.1016/j.amepre.2008.01.016

44.

Morad

, Wygnansky-Jaffe

, & Levin

A. V.

(2010). Retinal haemorrhage in abusive head trauma. Clinical and Experimental Ophthalmology , 38(5), 514–520. https://doi.org/10.1111/j.1442-9071.2010.02291.x

45.

Narang

, & Clarke

(2014). Abusive head trauma: Past, present, and future. Journal of Child Neurology , 29(12), 1747–1756. https://doi.org/10.1177/0883073814549995

46.

Niederkrotenthaler

, Xu

, Parks

S. E.

, & Sugerman

D. E.

(2013). Descriptive factors of abusive head trauma in young children—United States, 2000–2009. Child Abuse and Neglect , 37(7), 446–455. https://doi.org/10.1016/j.chiabu.2013.02.002

47.

Norman

R. E.

, Byambaa

, De

, Butchart

, Scott

, & Vos

(2012). The long-term health consequences of child physical abuse, emotional abuse, and neglect: A systematic review and meta-analysis. PLoS Med , 9(11), e1001349. https://doi.org/10.1371/journal.pmed.1001349

48.

Notrica

D. M.

, Sayrs

L. W.

, Kirsch

, Kelly

, Nickoles

, Greenberg

, Valdeze

, Fernandez

, Krishna

, & Misra

(2020). Inflicted versus unintentional injury: Towards improved detection and surveillance of supervisory neglect. Journal of Pediatric Surgery , 55(2), 341–345. https://doi.org/10.1016/j.jpedsurg.2019.10.010

49.

O’Meara

A. I.

, Sequeira

, & Ferguson

N. M.

(2020). Advances and future directions of diagnosis and management of pediatric abusive head trauma: a review of the literature. Frontiers in neurology , 11, 118.

50.

Oral

, Yagmur

, Nashelsky

, Turkmen

, & Kirby

(2008). Fatal abusive head trauma cases: Consequence of medical staff missing milder forms of physical abuse. Pediatric Emergency Care , 24(12), 816–821. https://doi.org/10.1097/PEC.0b013e31818e9f5d

51.

Palusci

V. J.

, & Covington

T. M.

(2014). Child maltreatment deaths in the U.S. national child death review case reporting system. Child Abuse and Neglect , 38(1), 25–36. https://doi.org/10.1016/j.chiabu.2013.08.014

52.

Paul

A. R.

, & Adamo

M. A.

(2014). Non-accidental trauma in pediatric patients: A review of epidemiology, pathophysiology, diagnosis and treatment. Translational Pediatrics , 3(3), 195–207. https://doi.org/10.3978/j.issn.2224-4336.2014.06.01

53.

Pears

K. C.

, & Capaldi

D. M.

(2001). Intergenerational transmission of abuse: A two-generational prospective study of an at-risk sample. Child Abuse and Neglect , 25(11), 1439–1461. https://doi.org/10.1016/s0145-2134(01)00286-1

54.

Piteau

S. J.

, Ward

M. G.

, Barrowman

N. J.

, & Plint

A. C.

(2012). Clinical and radiographic characteristics associated with abusive and nonabusive head trauma: A systematic review. Pediatrics , 130(2), 315–323. https://doi.org/10.1542/peds.2011-1545

55.

Rolfes

M. C.

, Guerin

J. B.

, Brucker

, & Kalina

(2018). Pediatric abusive head trauma: Review of standard imaging evaluation and typical findings. Journal of Pediatric Neurology , 16(02), 039–055.

56.

Scott

, Tonmyr

, Fraser

, Walker

, & McKenzie

(2009). The utility and challenges of using ICD codes in child maltreatment research: A review of existing literature. Child Abuse and Neglect , 33(11), 791–808. https://doi.org/10.1016/j.chiabu.2009.08.005

57.

Sheets

L. K.

, Leach

M. E.

, Koszewski

I. J.

, Lessmeier

A. M.

, Nugent

, & Simpson

(2013). Sentinel injuries in infants evaluated for child physical abuse. Pediatrics , 131(4), 701–707. https://doi.org/10.1542/peds.2012-2780

58.

Sieswerda-Hoogendoorn

, Boos

, Spivack

, Bilo

R. A.

, & van Rijn

R. R.

(2012). Educational paper. European journal of pediatrics , 171(3), 415–423.

59.

Simpson

D. A.

, Cockington

R. A.

, Hanieh

, Raftos

, & Reilly

P. L.

(1991). Head injuries in infants and young children: The value of the Paediatric Coma Scale. Review of literature and report on a study. Childs Nervous System , 7(4), 183–190. https://doi.org/10.1007/bf00249393

60.

Snyder

C. W.

, Muensterer

O. J.

, Sacco

, & Safford

S. D.

(2014). Paediatric trauma on the Last Frontier: An 11-year review of injury mechanisms, high-risk injury patterns and outcomes in Alaskan children. International Journal of Circumpolar Health , 73, 25066. https://doi.org/10.3402/ijch.v73.25066

61.

Springer

K. W.

, Sheridan

, Kuo

, & Carnes

(2007a). Long-term physical and mental health consequences of childhood physical abuse: Results from a large population-based sample of men and women. Child Abuse and Neglect , 31(5), 517–530. https://doi.org/10.1016/j.chiabu.2007.01.003

62.

Springer

K. W.

, Sheridan

, Kuo

, & Carnes

(2007b). Long-term physical and mental health consequences of childhood physical abuse: Results from a large population-based sample of men and women. Child Abuse and Neglect , 31(5), 517–530. https://doi.org/10.1016/j.chiabu.2007.01.003

63.

Stith

S. M.

, Liu

, Davies

L. C.

, Boykin

E. L.

, Alder

M. C.

, Harris

J. M.

, Mary

A. S.

, McPherson

, & Dees

(2009). Risk factors in child maltreatment: A meta-analytic review of the literature. Aggression and Violent Behavior , 14(1), 13–29. https://doi.org/10.1016/j.avb.2006.03.006

64.

Tiyyagura

, Christian

, Berger

, Lindberg

, & Ex

S. I.

(2018). Occult abusive injuries in children brought for care after intimate partner violence: An exploratory study. Child Abuse and Neglect , 79, 136–143. https://doi.org/10.1016/j.chiabu.2018.02.003

65.

van IJzendoorn

M. H.

, Bakermans-Kranenburg

M. J.

, Coughlan

, & Reijman

(2019). Annual research review: Umbrella synthesis of meta-analyses on child maltreatment antecedents and interventions: Differential susceptibility perspective on risk and resilience. Journal of Child Psychology and Psychiatry , 61(3), 272–290.

66.

Wilson

S. R.

(2009). Traumatic brain injury and intimate partner violence. In Intimate partner violence: A health-based perspective (pp. 183–199). Oxford University Press.