Pediatric Brain Injuries Are Associated With Intimate Partner Violence-Related Brain Injuries Among Women in Adulthood

Abstract

Objective:

Prior studies have reported associations between childhood physical abuse and intimate partner violence (IPV). However, there has been limited literature examining the relationship between pediatric brain injuries and IPV-related brain injuries later in life. We examined the association between childhood brain injuries and brain injuries sustained from IPV in adulthood.

Setting:

Participants were recruited through Mass General Brigham’s “Rally,” social media, and information sheets distributed in locations likely to serve women in current or past abusive relationships (eg, domestic violence shelters, and IPV support programs).

Participants:

Women who have experienced IPV with and without brain injury (n = 129), ages 19 to 69.

Design:

Retrospective, cross-sectional study design.

Main Measures:

We administered the Ohio State University Traumatic Brain Injury Identification Method and the Brain Injury Severity Assessment to assess brain injury history, and the Childhood Trauma Questionnaire (CTQ). A logistic regression was used to examine the association between pediatric brain injuries and IPV-related brain injuries, adjusting for childhood trauma, years of education, and age.

Results:

A majority of participants (64.3%) reported experiencing at least 1 IPV-related brain injury, and 30.2% experienced at least 1 childhood brain injury. Child abuse-related brain injuries occurred in 10.9% of the sample, whereas childhood brain injuries from other causes occurred in 19.4% of participants. Participants who sustained at least 1 childhood BI were significantly more likely to sustain a future IPV-related brain injury as an adult (OR = 4.652, P = .004). For each additional childhood BI sustained, participants were more likely to sustain an IPV-related BI (OR = 3.198, P = .005).

Conclusion:

This study illustrates a high prevalence of childhood brain injuries among women who have experienced IPV and identifies an association between pediatric brain injuries and brain injuries due to partner violence in adulthood. Our findings speak to the urgency of increasing our understanding of childhood brain injuries and whether they or other related factors might confer an increased risk for brain injuries or possibly IPV later in life.

The National Coalition Against Domestic Violence found that approximately 1 in 3 women in the United States experienced some form of intimate partner violence (IPV) in 2020.¹ Women in abusive relationships are often subjected to severe physical harm, such as beating, strangulation, and high-impact forces to the head, and are thus at high risk of sustaining brain injuries (BIs).² IPV-related BIs include both traumatic brain injuries (TBIs) resulting from an external force (eg, fist, foreign object, etc) to the head and strangulation-related brain injuries, due to strangulation-related oxygen deprivation.³

Prior literature has indicated a clear association between IPV and prior experiences of abuse as a child.^4–8 For example, McLindon and colleagues⁹ found that those subjected to IPV were 2 to 3 times more likely to report experiencing prior physical, emotional, or sexual abuse as a child in comparison to participants who did not experience IPV. Additionally, children living in a family environment in which IPV is perpetrated have been shown to report an increased risk of sustaining pediatric head trauma.¹⁰ With respect to IPV-related BIs, adverse childhood experiences (including abuse, neglect, and household dysfunction) were significantly more common among IPV survivors who sustained BIs compared to matched controls.¹¹

Pediatric brain injury, both due to abuse and unintentional injury, serves as a leading cause of mortality and permanent disability among children globally.¹² Several of the most common causes of childhood BIs include falls, unintentional blunt trauma (eg, hits from a baseball), motor vehicle collisions, and physical abuse.¹³ Furthermore, childhood BIs have been linked to a host of long-term adverse health and social outcomes including learning disabilities, mental health conditions (eg, depression, anxiety), psychiatric inpatient hospitalizations, lower educational attainment, and premature mortality (below the age of 41).¹⁴ More generally, adverse childhood experiences, including physical abuse, have been previously associated with poor neuropsychiatric health and a greater risk of developing dementia.¹⁵

The literature to date supports associations between childhood physical abuse and IPV. However, there have been limited studies examining the relationship between BIs sustained in childhood and IPV-related BIs in adulthood. The current study examined the association between childhood brain injuries (at or below the age of 12) and brain injuries sustained from intimate partner violence later in life. We hypothesized that participants who sustained at least 1 brain injury in childhood would be more likely to have sustained a brain injury in adulthood due to IPV than participants who sustained no brain injuries in childhood.

METHODS

Participants

While partner violence affects individuals of all genders, the current study focuses on women who are disproportionately affected by IPV.¹ Study participants (n = 129) were recruited using “Rally” with Mass General Brigham (MGB), a site which shares open clinical research studies with all MGB staff (eg, physicians, nurses, custodian staff, researchers) as well as the general public. We also recruited participants using other social media platforms (eg, Twitter, Instagram, Facebook), information sheets distributed in locations that were likely to have women in current or past abusive relationships (eg, domestic violence shelters, IPV support programs, etc), as well as presentations delivered by the principal investigator and study staff to possible referral sources. Nearly all women who were enrolled were community dwelling (n = 127) and only 2 participants resided in domestic violence shelters.

A telephone screening ensured inclusion and exclusion criteria were met. To be included, women needed to: (i) be at least 18 years of age, (ii) have experienced at least 1 physically abusive incident from a current or past intimate partner, (iii) have the ability to provide informed consent, and (iv) have access to a computer and a mobile electronic device. Participants were not eligible if they exhibited current symptoms of schizophrenia, bipolar disorder, or autism spectrum disorder that would interfere with remote testing as determined by the clinical opinion of the principal investigator. There were no women who were considered ineligible for the study, and there were no missing data for any participants. This study was approved by Mass General Brigham’s Institutional Review Board and consent was obtained verbally and in writing. In addition to assessing whether participants met inclusion/exclusion criteria, screening procedures were conducted to examine participants’ ability to communicate clear consent and proceed safely with study protocols. All participants were ultimately deemed capable of providing informed consent by the principal investigator and trained personnel.

Brain injury history

The Ohio State University Traumatic Brain Injury (TBI) Identification Method (OSU TBI-ID)¹⁶ was used to identify participant histories of brain injury. The OSU TBI-ID is a standardized procedure for eliciting lifetime history of TBI via a structured interview. The instrument is based on Center for Disease Control and Prevention¹⁷ case definitions and recommendations for TBI surveillance. The OSU TBI-ID prompts participants to recall injuries to the head or neck due to motor vehicle collisions, falls, and sports- or military-related incidents across an individual’s lifetime. This interview also includes 1 question that would capture IPV-related BIs: “In your lifetime, have you ever injured your head or neck in a fight, from being hit by someone, or from being shaken violently? Have you ever been shot in the head?” In addition, the OSU TBI-ID asks, “Have you ever had a period of time in which you experienced multiple, repeated impacts to your head (eg, history of abuse, contact sports, military duty)?” The interview gathers information regarding loss of consciousness and symptoms of daze and memory loss following these injuries. Prior studies have supported the high reliability and validity of the OSU TBI-ID among vulnerable populations, including IPV survivors.^18,19

Brain injury history was also assessed using the Brain Injury Severity Assessment Interview (BISA) in order to more comprehensively gather participants’ experiences of IPV-related BI.^20–29 The BISA is a semi-structured interview explicitly designed to identify all IPV-related brain injuries, including those caused by incidents of head trauma and strangulation, and has since been used to assess IPV-related brain injury in many reports nationally and internationally.^20–29 It elicits the number, severity, and time course of potential BI incidences (eg, hit, shaken, strangled) experienced by people subjected to IPV.²⁵ Brain injuries were determined according to the diagnostic criteria provided by the American Congress of Rehabilitation Medicine Special Interest Group (ACRM) on Mild TBI, which also overlap with the criteria outlined by the BISA and OSU TBI-ID.³⁰ According to the ACRM, people are considered to have sustained a brain injury if, after a “traumatically induced physiological disruption of brain functioning,” they reported experiencing at least 1 of the following symptoms: “(i) loss of consciousness for any period of time, (ii) any loss of memory for events immediately before or after the incident, (iii) any alteration in mental state at the time of an incident (eg, feeling disoriented, dazed, or confused), or (iv) focal neurological deficit(s) that may or may not be transient.”^{30(p 86)} In this report, women who experienced any of the aforementioned alterations in consciousness following incidents of strangulation-related anoxia or hypoxia were also considered to have sustained an IPV-related BI, in line with prior studies.^{23–25,28,29}

IPV-related brain injuries

We examined responses to both the BISA and the OSU TBI-ID regarding brain injury signs and symptoms following incidents of partner violence. If either questionnaire captured at least 1 instance of IPV-related brain injury, the participant was categorized as part of the IPV-BI group. Participants who did not sustain an IPV-related BI according to either the BISA or the OSU TBI-ID were categorized as the no-IPV-BI group.

Childhood brain injuries

Using both the OSU TBI-ID and the BISA assessments, participants who were determined to have sustained at least 1 non-partner-related BI at the age of 12 or younger were classified as having experienced a childhood BI. Types of childhood brain injuries were further divided into the following categories: (i) child abuse-related, and (ii) non-abuse-related. The Child Welfare Information Gateway defines childhood physical abuse as “any nonaccidental physical injury to the child and can include striking, kicking, burning, or biting the child, or any action that results in a physical impairment of the child.”³¹ In line with this definition, child abuse-related brain injuries included in our study were those intentionally perpetrated by surrounding individuals (ie, adult relatives, caretakers, strangers, traffickers, etc). Childhood BIs caused by unintentional circumstances (eg, falls, sports injuries, etc) or other mechanisms of injury not related to abuse were also identified.

Childhood trauma questionnaire

Participants’ recall of childhood trauma was measured using a 28-item Childhood Trauma Questionnaire (CTQ).³² The CTQ measures childhood trauma using 5 subscales: physical abuse, emotional abuse, sexual abuse, emotional neglect, and physical neglect. Responses are measured on a 5-point Likert scale (1 = never true, 5 = very often true) to statements such as “I got hit so hard by someone in my family that I had to see a doctor or go to the hospital” or “People in my family said hurtful or insulting things to me.” Total scores for this scale range from 25 to 140, with a value of ≥35 indicating a significant history of childhood trauma.^33,34 This scale was used descriptively (see Table 1). The CTQ has previously been shown to exhibit high validity and reliability among diverse participant demographics.^35,36

TABLE 1.

Study sample characteristics

	N	Range	Mean or %(a)	SD
Total sample	129	-	-	-
Demographics
Age (years)	-	19–69	39.7	12.6
Education (years)	-	9–23	15.6	2.6
Shelter-dwelling (%)	2	-	1.6%	-
Race (%)
White	78	-	60.5%	-
Hispanic	19	-	14.7%	-
Black	13	-	10.1%	-
Asian	7	-	5.4%	-
Native Hawaiian/Pacific Islander	1	-	0.8%	-
Multiracial	9	-	7.0%	-
Other	2	-	1.5%	-
Childhood trauma (CTQ total score)	-	25–115	57.7	21.8
BI characteristics
IPV-related BI	83	-	64.3%	-
Childhood BI	39	-	30.2%	-
Child abuse-related BI	14	-	10.9%	-
Childhood BIs from other causes	25	-	19.4%	-
Number of childhood BIs(b)	-	0–25	0.8	3.2

Abbreviations: CTQ, Childhood Trauma Questionnaire; N, sample size; SD, standard deviation.

^aMeans provided for continuous variables. % provided for dichotomous variables.

^bNumber of brain injuries (BIs) were capped at 25 due to estimation difficulties beyond that.

Statistical analyses

A Pearson chi square with associated odds ratio (OR) and 95% confidence interval (CI) was used to examine the primary research question and hypothesis relating to the association between childhood BI and future IPV-related BI. Logistic regression was also used to examine the association between childhood BI and IPV-related BI in adulthood, adjusting for age, childhood trauma, and years of education. Model 1 uses a dichotomized predictor variable, childhood BI, to examine whether experiencing at least 1 childhood BI from any mechanism is associated with experiencing IPV-related BIs. Model 2 uses a continuous predictor variable, number of childhood BIs, to predict whether childhood brain injuries from any mechanism are associated with experiencing IPV-related BI. Model 3 is an exploratory analysis using 2 dichotomous predictor variables, abuse-related and non-abuse related childhood BI, to examine whether the etiology of childhood BI has distinct associations with the likelihood of experiencing IPV-related BI. This analysis was considered conceptually important for future hypothesis generation. In all models, we adjusted for childhood trauma (CTQ total score), years of education, and age. Age was binned, because the continuous age variable did not satisfy the linearity assumption of logistic regression. All analyses were conducted with IBM SPSS Statistics 29.0.2.0.

RESULTS

Participant demographics are presented in Table 1. All participants were women. They had an average age of 39.7 years old (range = 19–69; SD = 12.6) and a mean of 15.6 years of education (SD = 2.6). The majority of participants identified as White (60.5%). There were 14.7% who identified as Hispanic. A large percentage of participants (64.3%) reported experiencing at least 1 IPV-related BI, and 30.2% experienced at least 1 childhood BI. Child abuse-related BIs occurred in about 10.9% of the sample, while childhood BIs from other causes occurred in 19.4% of participants. On average, participants received a CTQ score of 57.7 (range = 25–115; SD = 21.8). Participants who sustained at least 1 childhood BI from any mechanism (abuse-related or non-abuse related) were statistically significantly more likely to sustain an IPV-related BI in adulthood (84.6% vs 55.6%; χ2(1) = 10.0, P = .002, OR = 4.40, 95% CI = 1.68–11.54).

Table 2 presents the logistic regression results, adjusting for age, education, and childhood trauma. Participants who sustained at least 1 childhood BI from any mechanism (abuse-related or non-abuse related) were statistically significantly more likely to sustain a future IPV-related BI as an adult (OR = 4.652, P = .004, Model 1). As seen in Model 2, when childhood brain injuries were included as a continuous variable, there was also a significant association with IPV-related BI (OR = 3.198, P = .005). As Model 3 shows, sustaining at least 1 abuse-related childhood BI was not significantly associated with a higher likelihood of sustaining IPV-related BI in adulthood (OR = 7.616, P = .067). Participants who sustained at least 1 childhood BI from causes other than abuse were statistically significantly more likely to sustain IPV-related BI (OR = 3.980, P = .020). All of the above associations remained significant after adjusting for childhood trauma, years of education, and age. Both education and age between 50 and 59 years old were also significant predictors of IPV-related related BIs (Table 2).

TABLE 2.

Logistic regression results

	Model 1					Model 2					Model 3
	OR	S.E.	Wald	p	95% CIs	OR	S.E.	Wald	p	95% CIs	OR	S.E.	Wald	p	95% CIs
Predictors of interest
Childhood BI(a)	4.652	0.533	8.325	.004	1.637–13.218	-	-	-	-	-	-	-	-	-	-
Number of childhood BIs	-	-	-	-	-	3.198	0.417	7.781	.005	1.413–7.236	-	-	-	-	-
Abuse-related childhood BI(a)	-	-	-	-	-	-	-	-	-	-	7.616	1.106	3.367	.067	.871–66.604
Non-abuse related childhood BI(a)	-	-	-	-	-	-	-	-	-	-	3.980	0.592	5.453	.020	1.249–12.688
Control variables
Childhood trauma (CTQ)	1.003	0.010	0.088	.766	.983–1.024	1.002	0.011	0.037	.848	981–1.023	1.003	0.010	0.059	.809	.982–1.023
Years of education	0.770	0.093	7.938	.005	.642–.924	0.771	0.092	7.906	.005	.643–.924	0.773	0.093	7.662	.006	.645–.928
Age (reference category: 19–29)	-	-	6.324	.176	-	-	-	7.115	.130	-	-	-	6.184	.186	-
30–39	2.356	0.574	2.230	.135	.765–7.257	2.680	0.581	2.878	.090	858–8.369	2.340	0.572	2.208	.137	.762–7.179
40–49	2.012	0.555	1.585	.208	.678–5.972	2.293	0.566	2.148	.143	756–6.954	1.998	0.555	1.558	.212	.674–5.926
50–59	5.027	0.751	4.62	.032	1.153–21.920	5.500	0.757	5.067	.024	1.247–24.266	4.825	0.755	4.340	.037	1.098–21.213
60–69	0.896	0.806	0.019	.891	.185–4.348	0.969	0.822	0.001	.970	194–4.853	0.850	0.820	0.039	.843	.170–4.242
(Constant)	36.221	1.694	4.489	.034	-	33.925	1.693	4.334	.037	-	35.455	1.691	4.453	.035	-

Abbreviations: BI, brain injury; CI, confidence interval; CTQ, Childhood Trauma Questionnaire; OR, odds ratio; S.E., standard error.

^aDichotomous variable (1: BI, 0: no BI); P < .05 in boldface for independent variables.

DISCUSSION

A large minority of women in this study who have experienced intimate partner violence reported a personal history of childhood brain injuries (BIs) from any mechanism (abuse- or non-abuse-related). Approximately 30% of our sample reported sustaining at least 1 childhood BI, including those related to abuse and those related to other mechanisms of injury. Nearly 2 out of 3 (64%) of the women in this study reported experiencing a brain injury during adulthood resulting from intimate partner violence (IPV). Those who sustained at least 1 BI at or before the age of 12 were significantly more likely to report at least 1 IPV-related BI later in life. This study is the first to report specific associations between pediatric BIs and BIs due to partner violence in adulthood. This association remained after adjusting for childhood trauma, education, and age.

Numerous studies have explored how women’s prior experiences of child abuse (including physical, emotional, and sexual abuse) are associated with increased risk of experiencing IPV later in life.^4,37,38 Pediatric moderate to severe traumatic brain injury in the general population has been linked to a wide range of long-term behavioral, medical, and cognitive outcomes including future mental health problems (eg, depression and anxiety), greater impulsivity, communication difficulties, and internalizing behaviors,^14,39,40 which may place an individual at an increased susceptibility to unsafe and abusive environments.⁴¹

We conducted an exploratory analysis to examine whether the type of childhood BI (ie, child abuse-related versus not abuse-related) was associated with sustaining a future IPV-related BI. Participants who sustained at least 1 BI from causes other than abuse at or before age 12 were significantly more likely to report IPV-related BI as an adult (OR = 3.98, P = .020). Participants who sustained at least 1 abuse-related childhood BI may have been more likely to report IPV-related BI, however this association was marginally above the common significance threshold of .05 (OR = 7.62, P = .067). We consider this exploratory finding to be an important hypothesis-generating result. Thus, further research should examine these associations using larger sample sizes to determine and quantify a possible association between child abuse-related BIs and future IPV-related BIs. Interestingly, both education and age between 50 and 59 years were also associated with IPV-related BIs. These findings could serve as a foundation for future studies examining specific age and education related issues.

Implications for primary and secondary prevention

Our study findings provide additional support for the importance of enhanced pediatric brain injury prevention strategies. Given that the most common causes of childhood BIs are falls, motor vehicle collisions, and recreational activities,⁴² clinical and educational institutions should implement robust brain injury prevention programs teaching proper safety practices to parents, caregivers, and children (eg, using child safety seats in vehicles, installing window guards and stair safety gates, wearing helmets while biking, etc).

Though we did not identify a statistically significant association between child abuse-related BIs and later IPV-related BIs, this study highlighted a high prevalence of nearly 11% of childhood BIs from abuse. The prevalence of abuse-related pediatric BIs among the general population is challenging to quantify, as a 2021 systematic review found that rates of inflicted BI in children ranged from 4.6% to 82%.⁴³ Nonetheless, these findings speak to the importance of longitudinal parental and caregiver education that may lower the risk of perpetrating abuse-related pediatric head and brain injuries. Current educational programs on abuse-related BIs have been primarily conducted with parents during the neonatal period (e.g., management of stress related to infant crying), supported by research focusing on abusive head trauma among infants (or shaken baby syndrome).^44–47 Similar initiatives conducted in the early years following childbirth may also reduce the prevalence of abusive head trauma among children beyond infancy.⁴⁸ Moreover, it may be helpful for health care professionals to receive training on detecting the risks of child abuse and neglect as a means of strengthening frontline prevention of abuse-related head trauma.⁴⁹

Early educational initiatives relating to intimate partner violence are also important for promoting wider discussion and prevention of IPV-related brain injuries later in life. Such targeted interventions, particularly among children and adolescents who may have already sustained brain injuries, may help prevent future IPV-related BIs in adulthood. While training and education are important for increasing awareness surrounding the impacts of IPV, community-wide efforts to translate such awareness into sustained, intentional action are needed to foster meaningful change.

Limitations

Our results should be interpreted within the context of this study’s limitations. First, given the cross-sectional research design, we cannot estimate risk or infer a causal association between the exposure (childhood brain injury) and the outcome (future adult IPV-related brain injury). Second, as done in previous reports, we utilized a retrospective design to gather brain injury history, rather than relying on medical records to confirm these injuries. This design was chosen intentionally, given that many women in abusive relationships are often hesitant to and/or prevented by their partners from seeking medical attention for abuse-related injuries. There is also a possibility of recall bias, as all brain injuries (including those from childhood) were self-reported.¹¹ Third, our sample was, on average, highly educated and mostly White—thus, future studies should include women of various races and ethnicities who have experienced IPV. However, it is important to note that the women who participated in this study were recruited from communities beyond exclusively domestic violence/emergency shelters and therefore, serve as a more representative sample than those drawn entirely or mostly from women’s shelters. Finally, some of our results were limited by a relatively small sample size and low statistical power, specifically for the exploratory analysis investigating the association between child abuse-related BIs and adult IPV-related BIs.

CONCLUSION

Many women subjected to intimate partner violence have a childhood history of abuse, neglect, and brain injury. A high proportion of these women also experience IPV-related brain injuries as adults. The current study reports an association between brain injuries in childhood and IPV-related brain injuries during adulthood. Future studies utilizing larger samples are needed to determine if, and the extent to which, abuse-related brain injuries during childhood are associated with IPV-related brain injuries in adulthood. Nonetheless, our findings speak to the importance of addressing and preventing pediatric brain injuries through the enhancement of life course and child abuse interventions.

Transparency, rigor, and reproducibility statement:

The original contributions are presented in the study. The study design and analytic plan were not preregistered. The total sample size was 129. Participants were not informed about the hypothesis for this study. All authors had access to the raw data and to the statistical analyses. Data analyses were conducted using IBM SPSS Statistics 29.0.2.0.