Intimate Partner Violence and Other Trauma Exposures in Females With Traumatic Brain Injury

Abstract

We examined whether females with a history of traumatic brain injury (TBI) and intimate partner violence (IPV) have greater exposure to lifetime trauma relative to females with TBI but no IPV history. Further, we assessed the effects of lifetime trauma on psychological outcomes after TBI. Female participants (n = 70; age M [standard deviation–SD] = 50.5 [15.2] years) with TBI (time since injury median [interquartile range –IQR] = 10.2 [5.3–17.8] years) completed a structured assessment of lifetime history of TBI, including an IPV module to query head injuries from physical violence by an intimate partner. We characterized lifetime trauma exposure with the Adverse Childhood Experiences (ACEs) questionnaire and Survey of Exposure to Community Violence (CV). We evaluated psychological functioning with self-report questionnaires of post-traumatic stress disorder (PTSD), depression, and anxiety symptoms. Compared with those with no IPV history (n = 51), participants reporting IPV-related head injuries (n = 19; 27.1%) reported more ACEs (M[SD] IPV: 4.5[2.9]; No IPV: 1.6[1.8], p < 0.001, d = 1.08) and greater CV (IPV: 17.5[8.4]; No IPV: 7.6[6.1], p < .0001, d = 1.26). Within the full sample, ACEs (β = 0.21, 95% confidence interval [CI] = 0.04–0.39) and CV (β = 0.07, 95% CI = 0.01–0.13) predicted worse PTSD symptoms, while IPV alone did not. Exposure to all three sources of trauma (ACEs, CV, and IPV) was associated with worse PTSD symptoms relative to fewer traumas. The results highlight the scope of traumatic exposures among TBI survivors and the importance of considering IPV and other lifetime trauma exposure in assessing and managing TBI. Trauma-informed interventions that are modified for TBI-related impairment may offer improved outcomes in managing psychological symptoms.

Introduction

Intimate partner violence (IPV), including physical violence, sexual violence, emotional abuse, and stalking by a current or former partner, is a significant public health concern.^1,2 All genders experience IPV, but existing research indicates that physical violence is more common and consequential in women.^2–4 Physical IPV increases the risk of traumatic brain injury (TBI)^5,6 from repeated physical assault directed toward the head, face, and neck.^7,8

Both IPV^3,9,10 and TBI^11,12 have severe effects on psychological functioning, and when experienced together, the consequences can be profound. The IPV-related TBI contributes to various mental health problems, including depression, anxiety, and post-traumatic stress disorder (PTSD),^6,13–16 as well as structural and functional neural changes.^17–19

Evidence suggests an association between certain early life experiences and an increased risk for IPV in adulthood. Adverse childhood experiences (ACEs), such as abuse, neglect, and household dysfunction,²⁰ are associated with worse health outcomes²¹ (e.g., depression, anxiety, PTSD^22,23) and predict an increased risk of physical IPV victimization in adulthood.^10,24,25 These ACEs, typically experienced within the home environment, may contribute to a cycle of trauma outside the household.²⁶

Caregiver maltreatment is associated with a 2.5x increased risk of witnessing violence and 2.4x increased risk of other indirect violence exposure, including community violence (CV).²⁷ The CV refers to interpersonal violence in community settings (e.g., neighborhoods, local streets), which may involve experiencing or witnessing assaults, use of weapons, or other threats of injury.^28,29 The CV uniquely contributes to early life adversity beyond conventional ACEs³⁰ and predicts a higher incidence of IPV victimization in adulthood, even when controlling for other person-level ACEs.³¹ The CV exposure, like ACEs, increases the risk of depression and PTSD.^23,30

Exposure to multiple types of trauma over a lifetime, often referred to as polyvictimization,²⁶ predicts various negative outcomes in adulthood.^32,33 It is important to understand how these often-overlooked life course events may interact with TBI to better assess and understand outcomes. The combination of ACEs, CV, and IPV, both individually and cumulatively, may exacerbate the psychological symptoms experienced by survivors of TBI.^11,12

The current study aimed to (1) examine whether females with TBI and IPV-related head injuries have greater exposure to ACEs and CV relative to females with TBI and no history of IPV-related head injuries and (2) determine the effects of lifetime trauma on psychological outcomes in females with remote TBI (time since injury median [interquartile range – IQR] = 10.2 [5.3–17.8] years). We use the term “lifetime trauma” to indicate exposure to ACEs, CV, and IPV spanning a person's lifetime. We acknowledge, however, that our operationalization of this term may not encompass all lifetime trauma that people may experience.

Methods

Participants

We included female participants who completed the IPV module of the Brain Injury Screening Questionnaire (BISQ) as part of the ongoing multi-center prospective longitudinal study, the Late Effects of Traumatic Brain Injury (LETBI) project.³⁴ The LETBI participants met inclusion criteria of having at least one TBI (defined by either loss [LOC] or alteration [AOC] of consciousness based on Department of Defense (DoD) criteria³⁵), being at least 1 year from injury, aged ≥18 years, and English speaking. We collected essential demographic data including sex at birth, current gender identity, age, education, race/ethnicity, employment, and marital status (Table 1); all participants reported they were assigned female at birth and self-identified as women during testing. The local Institutional Review Boards approved the study protocol.

Table 1.

Sample Characteristics by Intimate Partner Violence Exposure

	Total (N = 70)	IPV (n = 19)	No IPV (n = 51)	p
Demographic Characteristics
Age, M ± SD	50.5 ± 15.2	50.1 ± 11.8	50.7 ± 16.4	0.863
Education, n (%)				0.086
Some High School	2 (2.9%)	2 (10.5%)	0 (0.0%)
High School/GED	3 (4.3%)	1 (5.3%)	2 (3.9%)
Some College	12 (17.1%)	4 (21.1%)	8 (15.7%)
College and Beyond	53 (75.7%)	12 (63.2%)	41 (80.4%)
Race, n (%)				0.516
White	63 (90.0%)	16 (84.2%)	47 (92.2%)
Black/African American	1 (1.4%)	0 (0.0%)	1 (2.0%)
Other	6 (8.6%)	3 (15.8%)	3 (5.9%)
Ethnicity, n (%)				0.253
Hispanic/Latino	5 (7.1%)	2 (10.5%)	3 (5.9%)
Employment, n (%)				0.490
Working	20 (28.6%)	3 (15.8%)	17 (33.3%)
Unemployed	6 (8.6%)	2 (10.5%)	4 (7.8%)
Retired	14 (20.0%)	3 (15.8%)	11 (21.6%)
Disabled	25 (35.7%)	10 (52.6%)	15 (29.4%)
Student	1 (1.4%)	0 (0.0%)	1 (2.0%)
Other	4 (5.7%)	1 (5.3%)	3 (5.9%)
Marital Status, n (%)				0.670
Never Married	24 (34.3%)	5 (26.3%)	19 (37.3%)
Married	19 (27.1%)	5 (26.3%)	14 (27.5%)
Domestic Partnership	3 (4.3%)	2 (10.5%)	1 (2.0%)
Divorced	15 (21.4%)	5 (26.3%)	10 (19.6%)
Widowed	5 (7.1%)	1 (5.3%)	4 (7.8%)
Injury Characteristics
Non-IPV TBIs, Median (IQR)	2.0 (1.0, 3.0)	3.0 (1.5, 4.0)	1.0 (1.0, 3.0)	0.101
Years Since Most Recent Injury, Median (IQR)	10.2 (5.3, 17.8)	12.9 (7.1, 18.4)	7.7 (4.5, 17.4)	0.286
Lifetime Trauma
ACEs Total, M ± SD	2.4 ± 2.5	4.5 ± 2.9	1.6 ± 1.8	< 0.001
CV Total, M ± SD	10.3 ± 8.1	17.5 ± 8.4	7.6 ± 6.1	< 0.001
Psychological Outcomes
PTSD Total Screen, M ± SD	1.4 ± 1.7	2.7 ± 1.7	0.9 ± 1.4	< 0.001
NeuroQol Anxiety T-Score, M ± SD	52.3 ± 8.4	55.6 ± 6.8	51.0 ± 8.7	0.051
NeuroQol Depression T-Score, M ± SD	48.1 ± 8.1	50.5 ± 8.4	47.2 ± 7.9	.144

IPV, intimate partner violence; SD, standard deviation; TBI, traumatic brain injury; IQR, interquartile range; ACEs, adverse childhood experiences; M, mean; CV, community violence; PTSD, post-traumatic stress disorder; NeuroQoL, Quality of Life in Neurological Disorders. Two participants (no IPV) were missing time since injury. Five participants (1 IPV, 4 no IPV) were missing anxiety and depression scores. Bold text indicates statistically significant at p < 0.05.

Measures

TBI and IPV

We used the BISQ to characterize lifetime history of head trauma, including exposure to IPV-related head injuries.^36,37 The BISQ provided 20 cues to facilitate recall of events causing head injuries, with LOC and AOC duration recorded for each event endorsed, and has previously demonstrated good construct and criterion validity.^36,38 Modules on head impacts from contact sports, military service, and IPV preceded the 20-item BISQ, and respondents were reminded not to rereport the same injury.

The IPV module queried seven events of IPV-related head and neck injury (e.g., pushed/shoved, shook violently, strangled/“choked”) since age 18.³⁷ We defined IPV-related head injury exposure using the IPV module. Specifically, all participants in the current sample had experienced at least one TBI, and the IPV-exposed group consisted of women who reported having sustained blows to the head from an intimate partner, regardless of whether the LETBI study-qualifying TBI was related to IPV exposure.

ACEs

Participants completed the ACEs Questionnaire,^20,39,40 a 10-item measure assessing self-reported childhood experiences including abuse, neglect, and household dysfunction. It is important to note that while the ACEs questionnaire is used as a measure of lifetime trauma in this study, some items on the ACEs, such as neglect and household dysfunction, may reflect adversity rather than trauma per se. In the current study, respondents answered yes or no for each experience, and we created a single summary score reflecting exposure to adverse experiences in childhood, per previous work.³⁹

Previous research has reported strong validity and reliability of the ACEs questionnaire, with associations to numerous health outcomes,²⁰ moderate to substantial test-retest reliability,^40,41 and acceptable to good internal consistency,³⁹ including in our current sample (Cronbach alpha = 0.81).

CV

We used an abbreviated 18-item version of the Survey of Exposure to Community Violence^28,42 to assess frequency of lifetime exposure to violent or threatening events, such as being chased or witnessing drug activity. Responses for each event ranged from Never (1) to two or more times (3). A total score was calculated by summing the responses, with higher scores indicating greater lifetime CV exposure, which we defined as direct and indirect experiences of violence in one's immediate environment through the lifespan. The Survey of Exposure to Community Violence has previously demonstrated concurrent validity⁴³ and acceptable to excellent internal consistency,²⁸ with a Cronbach alpha of 0.90 in our sample.

Psychological outcomes

The five-item Primary Care PTSD Screen for DSM-5 (PC-PTSD-5)^44–46 was used to examine PTSD symptoms. The measure asked about exposure to potentially traumatic events. If negative for exposure, we assigned a score of 0; if endorsed, we asked five questions about PTSD symptoms in the past month and summed items as a measure of PTSD symptoms. The PC-PTSD-5 has demonstrated acceptable to excellent internal consistency⁴⁷ (current sample Cronbach alpha = 0.77) and high test-retest reliability,⁴⁸ as well as good to excellent agreement with clinician rated diagnosis.⁴⁵

We used the Quality of Life in Neurological Disorders (Neuro-QoL) Anxiety and Depression short form scales^46,49 to assess symptoms of anxiety and depression. Each scale consisted of eight items answered on a Likert scale ranging from Never (1) to Always (5). We summed items to produce a total score for each scale. One participant was missing an item for the anxiety subscale, and four participants were missing no more than two items each for the depression subscale. Therefore, the total scores were prorated before converting to T-scores according to the Neuro-QoL Scoring Manual.⁵⁰ Higher T-scores indicate more symptoms. Both Neuro-QoL measures have demonstrated concurrent validity and excellent internal consistency⁴⁹ (Cronbach alpha in current sample: anxiety = 0.95; depression = 0.97).

Data analysis

Of 87 females who completed the BISQ with IPV module, four reported head injuries related to assault or physical abuse on the core BISQ, but did not endorse IPV-related head injuries. We excluded these participants to have a comparison group without physical violence-related head injuries. We also excluded two participants who did not meet DoD criteria for TBI, 10 participants with incomplete ACEs and/or CV measures, and one participant with incomplete PTSD and Neuro-QoL assessments. The final analytic sample was n = 70.

The included and excluded participants did not significantly differ in age, education, race, ethnicity, and marital status. The groups significantly differed, however, in employment status (p = 0.01), with the included group being more likely to be currently employed and less likely to be a student, keeping house, or categorized as other. Data for marital status and employment were missing for eight participants in the excluded group.

We compared demographic characteristics between those with and without reported IPV exposure using chi-square or Fisher exact tests for categorical variables and t tests for continuous variables. To determine the relationship between lifetime trauma and IPV, we employed t tests to compare ACEs and CV between groups, with corrections for homogeneity of variance violations as appropriate. We then conducted a cross-sectional analysis to investigate the association of ACEs and CV with IPV exposure using logistic regression. Both ACEs and CV were entered as predictors, while IPV served as the outcome variable in the model.

We conducted independent sample t tests to compare psychological outcomes between individuals with and without IPV exposure. We then used multiple linear regression to assess the relationship between ACEs, CV, and IPV with psychological outcomes while controlling for the total number of non-IPV TBIs and time since most recent injury. ACEs and CV were continuous variables in the regression models.

To capture the breadth of lifetime trauma experiences, we calculated a polyvictimization score by dichotomizing ACEs and CV using the sample median as the threshold (ACEs: ≥2; CV: ≥9) and summing the resulting binary variables with IPV. Given the lack of consensus on cut-points for the association of ACEs^51–53 and CV²⁸ with negative outcomes across studies, we opted to use the median because the cut-point provided a sample-specific threshold. We compared psychological outcomes across the four groups using analyses of variance and Tukey honestly significant difference (HSD) post hoc tests. We performed analyses in R version-4.2.1.⁵⁴

Results

Participant characteristics

Among the 70 females with TBI, 19 (27.1%) reported IPV-related blows to the head, with 15 (78.9%) meeting DoD criteria for TBI.³⁵ Demographic and injury characteristics did not differ between those with and without IPV exposure (Table 1).

Predicting IPV exposure

Females with TBI who reported IPV had higher exposure to ACEs (t[23.37] = 4.00, p < 0.001, d = 1.08) and CV (t[25.24] = 4.69, p < 0.001, d = 1.26). The logistic regression model revealed significant associations with IPV exposure. For each unit increase in CV, the odds of IPV exposure increased by 13.7% (odds ratio [OR] = 1.137, 95% CI [1.037, 1.268], p = 0.010). The ACEs showed a trend toward significance, suggesting that for each unit increase in ACEs, the odds of IPV exposure increased by 34.2%, although this effect was not significant (OR = 1.343, 95% CI [1.000, 1.857], p = 0.057).

Psychological outcomes

Females with TBI who reported IPV had more PTSD symptoms (t[68] = 4.53, p < 0.001, d = 1.22), but showed no differences in anxiety or depression symptoms.

We examined whether continuous ACEs, CV, and dichotomous IPV exposure predicted psychological outcomes. Greater ACEs and CV exposure, but not IPV, predicted more PTSD symptoms when controlling for the number of non-IPV TBIs and time since most recent injury and accounted for more than a third of the variance in reported PTSD symptoms (Table 2). The ACEs, CV, and IPV exposure did not significantly predict anxiety or depression symptoms.

Table 2.

Multiple Linear Regression Models Evaluating Association of Lifetime Trauma With Psychological Outcomes

					95% Confidence interval
		β	SE	p	Lower	Upper
PTSD
	ACEs	0.21	0.09	0.019	0.04	0.39
	CV	0.07	0.03	0.015	0.01	0.13
	IPV	0.58	0.46	0.213	-0.34	1.49
F(5, 62) = 8.79, p < 0.001, adjusted R2 = 0.37
Anxiety
	ACEs	0.90	0.56	0.113	-0.22	2.02
	CV	0.20	0.18	0.277	-0.16	0.56
	IPV	0.53	2.85	0.853	-5.18	6.25
F(5, 57) = 2.07, p = 0.08, adjusted R2 = 0.08
Depression
	ACEs	0.05	0.56	0.931	-1.07	1.17
	CV	0.28	0.18	0.129	-0.08	0.64
	IPV	0.69	2.86	0.811	-5.04	6.41
F(5, 57) = 0.99, p = 0.43, adjusted R2 = 0.00

SE, standard error; PTSD, post-traumatic stress disorder; ACES, Adverse Childhood Experiences; CV, Survey of Exposure to Community Violence; IPV, intimate partner violence; TBI, traumatic brain injury.

Models predicting PTSD, anxiety, and depression controlling for number of non-IPV TBIs and time since injury.

We examined the impact of polyvictimization scores (comprising dichotomized ACEs and CV scores and IPV presence/absence variables) on psychological outcomes. The PTSD scores differed based on polyvictimization scores (F[3, 66] = 11.21, p < 0.001). Participants with exposure to all three traumas had significantly higher PTSD symptoms (M[SD] = 3.07[1.59]) compared with those with one (M[SD] = 0.50[1.04]) or no trauma sources (M[SD] = 0.68[1.32]; Tukey HSD p's < 0.001; Supplementary Figs. S1 and S2). Anxiety and depression scores did not differ significantly by polyvictimization score.

Discussion

This study examined lifetime trauma exposure in females with TBI who did and did not report IPV and investigated diverse trauma exposures associated with psychological outcomes after TBI. Among females with TBI, those with a history of IPV reported greater exposure to ACEs and CV, which predicted worse PTSD symptoms than those with no IPV history. These findings suggest cumulative effects of trauma exposures and TBI on psychological outcomes, particularly symptoms of PTSD. Findings highlight the importance of considering lifetime trauma exposure in treating females with TBI.

Our results align with previous research showing that ACEs and CV independently predict increased risk of IPV victimization,^24,25,31 and ACEs and IPV are individually associated with all-cause TBI^55–57 in adulthood. The current findings extend this work by highlighting the interconnectedness of multiple types of trauma and violence. The logistic regression analysis indicated that experiencing one type of trauma is associated with increased risk of experiencing another, perhaps reflecting a recurrent cycle of victimization.

Notably, exposure to CV was identified as a significant predictor of IPV exposure, thus highlighting CV as a crucial, potentially modifiable factor associated with IPV exposure. Addressing and preventing CV through public health and community-based interventions could help to break the cycle of violence^25,31,58,59 and improve access to care and resources for TBI survivors. Moreover, addressing CV can also help to reduce community distrust that can create barriers to reporting IPV. Lifetime trauma predicted worse psychological outcomes in this TBI sample, consistent with past studies showing that trauma exposure contributes to psychological symptoms in community samples^22,23,30,32 and that pre-injury psychiatric symptoms predict worse symptoms and psychiatric diagnosis after TBI.^12,60,61

In this study, experiencing ACEs and CV alone or with IPV was associated with more PTSD symptoms, suggesting that each additional trauma may add to the overall burden of PTSD symptomatology. As such, assessing lifetime trauma, in addition to assessing psychological health, and providing trauma-informed care⁶² are crucial to improve TBI interventions. The IPV alone did not predict worse psychological symptoms in this study, but contributed to the cumulative trauma effect on PTSD symptoms. Ultimately, considering trauma exposure history and integrating trauma-informed approaches may improve the effectiveness of TBI interventions and enhance recovery from TBI among those with lifetime cumulative trauma.

Limitations

We used brief self-report measures to assess psychological outcomes, which may have recall and reporting biases. In addition, while the measure of PTSD symptoms used assessed symptoms within the last month, it should be noted that symptoms may vary across time and were not aligned with timing of most recently experienced traumatizing event. Multiple assessment time points alongside clinician-administered assessments would permit a more thorough evaluation of psychological functioning. Further, it is essential to note that our CV measure encompassed lifetime exposure without specifying whether CV preceded or followed ACEs or IPV exposure. Hence, future research should investigate the temporal relationship between CV, IPV, and ACEs exposure.

The small sample and cross-sectional design prevented establishing causal relationships between the variables investigated. Further research is necessary in larger and more diverse (e.g., race, education) longitudinal birth cohort samples to investigate the impact of broader lifetime trauma exposure, including events not studied in this research (e.g., combat, non-IPV–related assault, natural disasters) on TBI outcomes over time. Given that all women in the IPV-exposed group also had TBI, the findings may not generalize to women with a history of IPV who have never had a TBI.

Conclusion

Females with TBI and IPV history reported higher exposure to lifetime trauma than those without IPV. Lifetime trauma predicted worse psychological outcomes in TBI. The study emphasizes the significance of considering lifetime trauma in TBI assessment and treatment, and highlights the need for trauma-informed interventions to improve psychological outcomes in survivors of TBI.

Transparency, Rigor, and Reproducibility Summary

The study methods are reported in Edlow et al., 2018 (https://doi.org/10.1089/neu.2017.5457). The analysis plan was not formally pre-registered, but the lead author, Nicola de Souza, certifies that the analysis plan was pre-specified unless otherwise stated (i.e., described as an exploratory analysis initiated in response to results of pre-specified analyses). A sample size calculation was not performed as this study was a preliminary investigation of secondary data. 379 participants underwent study procedures, and data meeting current study inclusion criteria were obtained from 70 participants. Participants were blinded to the results of other assessments throughout the study. Data collectors were aware of relevant participant characteristics (i.e., IPV exposure) and data were labeled using codes linked to participant identifying information. The data analysis was performed by investigators who were aware of relevant characteristics of the participants. Data were acquired between October 2019 and August 2022. All data were analyzed at the same time and performed in R version-4.2.1 with methods specified in the Material and Methods. Specific measures used to acquire data are freely available online or by contacting the corresponding author of the measures. The key inclusion criteria and the primary clinical outcome measures are emerging standards in the field. The statistical tests used were based on the assumptions of normality, homogeneity of variance, and independence. Corrections for violations of homogeneity of variance were applied as appropriate and are indicated in the manuscript. Effect sizes and confidence intervals have been reported in the abstract for primary outcomes and main text for all outcomes. Statistical analysis and review were performed by the authors (NDS, RK, and KDOC) with qualifications including training in logistic mediation analysis and applied regression modeling. Correction for multiple comparisons was not performed. No replication or external validation studies have been performed or are planning/ongoing at this time to our knowledge. De-identified data from this study are not available in a public archive. De-identified data and analytic code from this study will be made available (as allowable according to IRB standards) by emailing the corresponding author. The authors agree to provide the full content of the manuscript on request by contacting the corresponding author.

Authors' Contributions

KDOC, JH, and BE are principal investigators of the study and developed study concept and design. AP, EB, WS, AM, and PL contributed to project administration including participant recruitment, data collection, and data management. NDS formulated manuscript research questions, led statistical analyses, and wrote the first draft of the manuscript. NDS, RK, and KDOC contributed to the analytic approach and data interpretation. All authors reviewed, provided critical feedback, and approved final manuscript.

Funding Information

The LETBI Project is supported by the National Institutes of Health (NIH)/National Institute for Neurological Disorders and Stroke (NINDS) Grant # 1U01NS086625, NIH/NINDS Grant # 1RF1NS115268, and NIH/NINDS Grant # 1RF1NS128961.

Author Disclosure Statement

No competing financial interests exist.

Supplementary Material

Supplementary Figure S1

Supplementary Figure S2