Abstract
Survivors of intimate partner violence (IPV) are at heightened risk of sustaining a brain injury (BI). Problematically, a high overlap between BI and trauma symptoms leads to difficulties in identifying when an IPV-related BI has occurred. This paper investigated differences in symptom reports between survivors with (n = 95) and without (n = 42) probable IPV-related BI. Chi-squared analyses isolated a constellation of symptoms found to be specifically associated with BI status. These symptomatic markers may assist professionals in discerning BI from other comorbid conditions present in IPV, and thus help survivors access BI-specific treatments and resources.
Background
Intimate partner violence (IPV) refers to sexual, psychological, or physical harm caused by a former or current intimate partner. It is a significant public health issue that disproportionately affects women, who are more likely than men to experience more severe forms of physical violence and with greater frequency (Smith et al., 2017). A majority of the physical injuries sustained in IPV are to the head and neck, thus heightening the risk of brain injury (BI; Corrigan et al., 2001; Sheridan & Nash, 2007). IPV-related BI may result from a traumatic force inflicted to the brain (e.g., hit/punches to the head, whiplash from being shaken, falls from being pushed) and/or reduced/lack of oxygen to the brain (hypoxic/anoxic injuries) from strangulation. Previous reviews of the literature have found traumatic BI prevalence rates of 19–75% in survivors of IPV (Haag et al., 2022; Kwako et al., 2011), indicating that it is a frequent occurrence in IPV.
Sustaining an IPV-related BI can exacerbate the medical, psychiatric, and behavioral symptoms common to IPV, creating additional, obstructive, and long-lasting consequences (Iverson et al., 2019; St. Ivany & Schminkey, 2016). Studies have found that individuals who have sustained an IPV-related BI are more socially isolated, have greater risk of developing mental health issues, have a harder time sustaining gainful employment, and experience more difficulty accessing health resources compared to survivors of IPV without BI (Iverson et al., 2017; St. Ivany et al., 2018; St. Ivany & Schminkey, 2016). Furthermore, these may become chronic issues, as individuals with IPV-related BI have reported poor psychosocial outcome even 18 months after their initial injury (Iverson et al., 2019). Similarly, epidemiological research has found that survivors of IPV-related BI reported poorer health status, greater perception of disability, and lower rate of return to work a year after injury compared to individuals who sustained BI from other violence (Gabbe et al., 2022). Taken together, these findings indicate that BI occurring in the context of IPV has additive, serious, and long-lasting implications for the health and well-being of survivors.
Although the evidence supports an unequivocal need for treating IPV-related BI, the actual practice of identifying BI in IPV can be challenging. Formal diagnosis of BI is typically determined by a professional and requires the careful, assessment of multiple factors including duration of alterations/loss of consciousness (e.g., using Glasgow Coma Scale scores), periods of posttraumatic amnesia, and neuro-imaging results (Levin & Diaz-Arrastia, 2015). BI screenings, on the other hand, are based on self-report and can be easily administered by individuals without medical training. While not used to officially diagnose a BI, screenings may offer a more accessible and viable option for recognition of a possible BI. However, screenings can still be difficult to accomplish and interpret in this population. Often, personal and safety factors may prevent an individual in an abusive relationship from seeking care for the injury (Corrigan et al., 2001; Joshi et al., 2012; Robinson et al., 2021). Furthermore, even when survivors do come forth in reporting an injury, the high overlap between symptoms of BI and other mental/behavioral health issues experienced in IPV creates diagnostic challenges leading to the BI being overlooked or misdiagnosed (Campbell et al., 2018; Haag et al., 2022). However, emerging research supports the possibility that BI symptoms might be distinguished from more general trauma symptoms. Recently, researchers found that BI severity was significantly correlated to a specific pattern of cognitive and arousal state symptoms endorsed by survivors with IPV-related BI (Smirl et al., 2019). The findings offer promise in identifying symptomatic markers to better identify IPV-related BI, but generalizability of the results was limited by a small sample size and lack of a control group.
The identification of IPV-related BI is a critical first step in ensuring that survivors obtain appropriate and timely injury-specific services. However, difficulty differentiating between symptoms of BI and other comorbid conditions greatly hinders the diagnostic process. The goal of the present study was to identify differences in symptom presentation between survivors of IPV who screened positive versus negative for probable IPV-related BI. Comparing the symptoms endorsed by these two groups facilitates the identification of unique markers that may be useful in discerning between BI and trauma symptoms.
Method
The current study involves a secondary analysis of data from earlier research reported elsewhere (Rajaram et al., 2021). In brief, staff at community based organizations serving women who have experienced IPV offered a modified version of the HELPS BI screening tool (Picard et al., 1991) to clients. The original HELPS screen was developed to determine occurrence of BI by querying about: (a) event(s) when head injury could have occurred, (b) use of emergency medical services, (c) alteration in mental status, (d) symptoms of head injury, and (e) other significant sicknesses. Criteria for screening positive on the HELPS screen are consistent with guidelines for defining mild traumatic BI established by the American Congress of Rehabilitation Medicine (Mild Traumatic Brain Injury Committee of the Head Injury Interdisciplinary Special Interest Group of the American Congress of Rehabilitation Medicine, 1993) and the tool has been used in previous research to identify BI in IPV populations (Jackson et al., 2002; Manoranjan et al., 2022; Saleem et al., 2022). The screen is considered positive for possible BI if the following three criteria are met: (a) there is a reported event that could have caused injury (endorsing having been hit or strangled, or if an individual was seen by a medical professional because of an injury to the head), (b) there was a period of loss or altered consciousness after the injury, and (c) endorsement of two or more symptoms that were not present before the injury (Picard et al., 1991). The screen in this study was slightly modified to obtain more information about injury characteristics, such as inquiries about mechanisms of injury (including strangulation), history of when and how many times injury has occurred, as well as expanding the symptom checklist to obtain a more comprehensive profile of commonly reported symptoms after BI (see Table 1). At the request of the organizations, an item querying about history of other illnesses was omitted. Women who were ≥19 years old, had experienced IPV, and were verbally fluent in English were invited to complete the screening tool. Data collected from the original project were approved for analysis by the Institutional Review Board at the lead author's institution.
Difference in Endorsement of Symptoms by Probable BI Status.
Note. List of symptoms included in the modified HELPS screening tool used in the study. % Endorsed for each group (positive or negative screen for probable BI) was calculated by dividing the number of individuals who endorsed the symptoms by total number of individuals in that group.
*Significant after accounting for multiple comparisons using p < .003.
To examine differences in commonly reported and overlapping symptoms on the modified HELPS tool, screens where no symptoms were endorsed were excluded. Frequency of symptom endorsement by each group (individuals screening positive versus negative for probable BI) was determined by calculating the percentage of individuals who endorsed that symptom in each group. The relationship between probable IPV-BI status and symptom endorsement was examined using chi-square tests of independence. Multiple comparisons were corrected for by applying a Bonferroni correction, whereby p < .003 was the new alpha for significance testing.
Results
A total of 171 women completed the screenings. After exclusion of screens where no symptoms were endorsed, data from 137 women remained (positive screen: n = 95, negative screen: n = 42). All participants who screened positive reported having sustained at least one injury from an assault (including injuries sustained from being pushed into/against something; for additional injury characteristics, see Table 2). Forty-one out of the 42 women who screened negative also endorsed sustaining at least one injury from a fight, and 6 of these women reported some alteration in consciousness (but endorsed less than 2 symptoms from the injury, thus resulting in a negative screen based on the developer's screening criteria). Mean age of the women was 35.8 years old (SD = 10.8), with no significant differences in age between groups.
Available Injury Characteristics of Participants Who Screened Positive for Probable BI.
Note. Information about injury characteristics is available from the modified HELPS screen. Percentage of participants reporting was calculated out of the sample of individuals who screened positive for probable BI (n = 95). The sum of some characteristics exceeds 100% because more than one option was sometimes marked for an individual (e.g., an individual who sustained multiple injuries reported an event 1–2 years ago, and another event longer than 4 years ago).
Women who screened positive for probable BI reported on average a total of 11.8 symptoms (SD = 5.0), while those who screened negative reported a total of 7.1 symptoms (SD = 4.2). An independent samples t-test showed that women screening positive reported a significantly greater number of symptoms than peers with negative screens, t(135) = −5.43, p < .001.
The frequency of endorsement for each symptom is presented in Table 1. Symptoms were endorsed by 72.48–88.24% of survivors with probable BI with the most commonly endorsed symptoms being: changes in smell/taste, numbness in limbs, dizziness, sensitivity to noise, and difficulty with problem-solving. In comparison, symptoms were endorsed by 32.14–64.84% of survivors without BI, with the most commonly endorsed symptoms being: nausea, changes in smell/taste, difficulty with problem-solving, challenges returning to work/school, and sleep disturbances.
Results from the chi-square analyses showed several symptoms were significantly related to BI status. Symptoms that remained significantly related to BI after correcting for multiple comparisons included experience of: headaches (OR = 4.3, p < .001), dizziness (OR = 7.1, p < .001), sensitivity to noise (OR = 4.3, p < .001), numbness in limbs (OR = 5.4, p < .001), lack of balance (OR = 4.0, p < .001), anxiety (OR = 7.8, p < .001), and difficulty with memory (OR = 3.7, p < .001; all χ2 statistics, phi coefficients, and p values reported in Table 1). None of the symptoms were found to be significantly related to a negative BI screen.
Discussion
This study examined the profile of symptoms endorsed by survivors with and without probable IPV-related BI. In this sample, survivors with probable IPV-related BI reported a greater total number of symptoms and at a higher frequency than their peers who had screened negative for probable BI, suggesting that BI may compound problems commonly present in IPV. Results also showed that certain symptoms were more likely associated with experiencing IPV-BI. Specifically, women screening positive for IPV-related BI were more likely to endorse the experience of headaches, dizziness, sensitivity to noise, numbness in limbs, lack of balance, anxiety, and difficulty with memory. Potential added distress caused by these additional symptoms may explain in part why survivors with IPV-related BI experience worse outcomes than survivors without BI.
The type of symptoms experienced by survivors in this study differed significantly depending on BI status. That is, survivors who sustained a probable IPV-related BI were more likely to experience headaches, dizziness, sensitivity to noise, numbness in limbs, lack of balance, anxiety, and difficulty with memory compared to survivors without an IPV-related BI. Some of these symptoms, such as headaches, dizziness, and difficulty with memory, have been identified in previous studies in other BI sub-populations as being more “neurogenic” in nature (Mounce et al., 2013) and potential markers of BI (Vanderploeg et al., 2009). The results of the current and previous studies suggest that these symptoms may serve as reliable indicators of BI across various sub-populations, including IPV-related BI. Identification of these markers has significant clinical implications for those working with IPV survivors. First, it may help service professionals better distinguish between symptoms associated with BI versus other trauma, and thus taken together with information from a more basic BI screener, can improve the chances of recognizing when a BI has occurred. Further, this knowledge may help guide the development of more refined and focused screeners specific to this population which would facilitate more efficient identification of BI. More accurate and timely identification of IPV-related BI is a critical first step in helping survivors to access BI specific resources that could improve outcomes and quality of life. Future research should be conducted to further explore and confirm the predictive utility of such symptomatic markers.
While high endorsement of mood symptoms was expected of both groups, feelings of anxiety, but not depressive symptoms, were significantly associated with BI status. This result was somewhat surprising; given the high prevalence of anxiety disorders in IPV (Chandan et al., 2020; Lagdon et al., 2014), it was not expected that endorsement of anxiety would differ based on BI status. One possible explanation is that IPV-related BI further compromises the neural resources needed to appropriately interpret or cope with anxiety symptoms, resulting in a heightened experience or increased susceptibility to these symptoms. For example, studies have shown altered patterns of functional activation (Wang et al., 2017) and network connectivity (van der Horn et al., 2016) in circuits associated with emotion processing and regulation following injury. Indeed, neuroimaging study findings have alluded to neurobiological contributions in the development of anxiety (Mallya et al., 2015) and posttraumatic stress disorder (Williamson et al., 2013) following injury. These findings, taken together with evidence that BI affects the same neural networks underlying posttraumatic stress disorder (Weis et al., 2022), suggest that sustaining an IPV-related BI could result in alterations in neural functioning that increase risk for stress and anxiety disorders. An additional consideration is the possibility that long-term survivors of IPV may have preexisting histories of anxiety prior to their BI, which place them at increased susceptibility to anxiety postinjury. While some of these cases of anxiety after BI are novel diagnoses for the individual (Alway et al., 2016), studies have found a preinjury history of diagnosis to be a strong predictor of postinjury disorder (Alway et al., 2016; Scholten et al., 2016). The intersection of anxiety and IPV-related BI should be further explored; it should be noted that the screener only included one inquiry about the experience of anxiety, thus, more detailed investigations of the different facets of anxiety are needed. Additionally, it is noted that anxiety is commonly comorbid with depression (Gorman, 1996); this complex relationship should also be further explored in future research of IPV-BI.
Much research is still needed to better understand the nuances between symptoms that are specific to IPV-related BI and more generalized trauma responses. This can be an especially daunting but critical task as many BI symptoms are not unique to injury alone (Donnell et al., 2012), and not properly assessing for the source of these symptoms may lead to misdiagnoses that interfere with treatment programming (Rosen & Ayers, 2020; St. Ivany & Schminkey, 2016). Thus, a thorough and comprehensive assessment of the symptoms experienced, whenever possible, can be helpful to better distinguish whether they can be attributed to a potential BI or psychiatric status (McMillan, 2001). Furthermore, while identifying symptoms associated with IPV-related BI may help improve the precision of screening and diagnostic tools, it should be noted that this information should complement the use of existing screeners that query about BI history (Goldin et al., 2016). In fact, an additional application of surveying symptomology is to better identify subjective areas of distress, which can then be used to guide treatment planning. For example, endorsements of cognitive problems may suggest a referral for a neuropsychological evaluation, while consultation with a neuroopthamologist may be sought for vision problems. In this vein, interdisciplinary collaborations between survivor advocates, medical and mental health professionals, as well as BI experts will be critical in creating the best care for survivors with IPV-related BI.
Study Limitations
There are some limitations to the current study due to the nature of the original project from which these data were collected. Namely, the collection of more comprehensive psychosocial history would have been helpful in better characterizing the sample. Also, BI status was determined by self-report on the screening tool; other measures to corroborate and better detail BI history would have strengthened the findings. Relatedly, it is acknowledged that there is some discrepancy between screening criteria between the HELPS measure used in this study compared to others (e.g., Ohio State University-TBI ID), namely the inclusion of symptoms in the determination of BI status. Thus, our categorization of BI status may have been conservative (i.e., must have experienced symptoms in addition to alteration in consciousness). Further, we limited our sample to individuals who had reported at least one symptom, thus, our group of individuals who screened negative for BI was a select sub-group and may not be representative. In particular, all but one individual from our included sample screening negative for BI had all been exposed to violence (endorsed event of blow to head or strangulation), whereas the majority of the individuals excluded from the study (30 out of 34) did not endorse such an experience. Another shortcoming was that the symptoms investigated in this study were limited to those presented on the HELPS and focused on BI (although there is overlap with trauma-related symptoms, e.g., sleep issues, fatigue). While the results did identify specific symptoms are more highly associated with a positive screen for probable BI, a survey of additional trauma-related symptoms would have been helpful to further help distinguish IPV-BI-specific symptoms. Lastly, future research would benefit from additional comparison groups (e.g., women who have sustained a BI but have not experienced IPV) that would allow for further study of the additive and interactive effects of BI in IPV.
Conclusions
BI sustained from physical IPV is common, yet often overlooked and misdiagnosed. Substantial overlap between symptoms of BI and other mental/behavioral health diagnoses complicates the identification and diagnosis of BI in IPV. Preliminary results of this study suggest that survivors with probable IPV-related BI do indeed experience more symptoms compared to those without BI. Further, a specific profile of symptoms was isolated as being significantly associated with the experience of IPV-related BI. These findings show promise in the determination of symptomatic markers that can better guide professionals in more accurate screening and diagnosis of IPV-related BI.
Footnotes
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported in part by the Women Investing in Nebraska.
