Abstract
Background/Purpose: We explored the complex relationship between the presence of metastatic cancer and undergoing cancer chemotherapy in adults with moderate to severe trauma and risk of reported physical abuse/assault-related trauma as defined in the American College of Surgeons Trauma Quality Programs Participant (ACS-TQP) dataset. We analyzed how these specific health conditions are associated with vulnerability to assault-related trauma, addressing a significant knowledge gap in existing literature.
Methods: We utilized retrospective data from the ACS-TQP from 2017-2019 (N=27,531). The outcome was the report of physical abuse/assault-related trauma (no/yes). We focused on adults (aged ≥ 18 years) diagnosed with moderate to severe physical trauma (Injury Severity Scale: 9-75) who either had metastatic cancer (exposure) or were currently receiving chemotherapy (intervention), adjusting for predisposing factors and need for care factors based on Anderson's Model of Health Service Utilization. We used descriptive statistics, Fisher's exact test, chi-square analysis, and logistic regression using Stata.v18, with a statistical significance of P≤0.05.
Results: We found 0.19% of 27,531 overall patients reported assault-related trauma, with 16,261 (0.16%) among those with moderate to severe trauma. Among these, a substantial majority with metastatic cancer reported assault-related trauma compared to those without metastatic cancer (84.62% vs. 15.38%; P value: 0.040). Patients receiving chemotherapy reported less frequent assault-related trauma than those not receiving it (26.92% vs. 73.08%; P value = 0.045). In the adjusted model, patients with metastatic cancer had significantly higher odds of reporting assault-related trauma than those without metastatic cancer (OR:7.847; 95% CI: 1.021-60.337; P<0.05). Chemotherapy was associated with a lower adjusted odds of assault-related trauma (OR 0.31, 95% CI: 0.08–1.14), but this did not reach statistical significance.
Conclusion: In this large trauma cohort, metastatic cancer was associated with higher odds of assault-related trauma, while chemotherapy showed a reduced risk only in unadjusted analyses. Our findings highlight the higher vulnerability of patients with metastatic cancer and support further investigation into potential protective associations with active treatment.
Existing literature indicates that individuals with serious health conditions face higher vulnerability from the pervasive nature of domestic violence, including physical abuse/assault.1–3 However, trauma’s intersection with metastatic cancer and physical abuse/assault (a major form of domestic violence) remains largely underexplored. Researchers also acknowledge that socioeconomic factors play a significant role in health outcomes, but their specific risk-adjusted association in the trauma, cancer, and physical abuse/assault intersection remains unclear.4–6 Our study aimed to shed light on these interrelationships, particularly for adults with moderate to severe trauma (Injury Severity Scale/ISS: 9-75) who are undergoing chemotherapy. Our findings could potentially inform future research to guide targeted interventions, health policies, and social support systems.
Previous studies have reported individuals with heightened stress, financial strain, and social isolation have higher vulnerability to physical abuse/assault.1–3 Additionally, evidence suggests trauma patients with preexisting health conditions, particularly cancer, experience disparities in health outcomes, which may be associated with systemic inequities in healthcare access and psychosocial support.4–6 While much of the existing literature has focused on the relationship between domestic violence as a whole and chronic diseases such as cancer,7–9 fewer studies have examined whether patients with metastatic cancer have a different likelihood of experiencing physical abuse/assault compared to those without the disease. Additionally, limited research has examined whether trauma and abuse are associated with disparities in the material and financial resources available for oncology and trauma care services.5,6,10 This study sought to address some of these gaps by assessing the association between metastatic cancer and physical abuse/assault in trauma patients, exploring whether chemotherapy is linked to differences in reported physical abuse/assault, and adjusting for socioeconomic factors. Findings may inform hypotheses for future screening efforts, and multidisciplinary approaches. It is important to note that in this study, the outcome variable ‘physical abuse/assault’ was operationally defined according to the American College of Surgeons Trauma Quality Program Participant Use File (ACS-TQIP) coding rules and may capture both intentional interpersonal assaults and certain trauma mechanisms when documented in the context of suspected or confirmed abuse. Accordingly, we interpreted this outcome as ‘reported abuse/assault-related trauma,’ which may not fully align with conventional legal or psychosocial definitions of abuse. This framing allows us to examine abuse-related vulnerability within the constraints of trauma registry data while acknowledging these limitations. Cancer-specific vulnerability mechanisms (caregiver burden, functional dependence, symptom distress, and financial toxicity) are described in the oncology literature; however, these constructs are not captured in ACS-TQIP and were not directly measured or modeled in our analyses. We hypothesized that metastatic cancer would be associated with higher odds of reported abuse/assault-related trauma in adults with moderate–to–severe physical trauma, and that current chemotherapy would be associated with lower odds due to intensified clinical contact and monitoring.
Our study presents the association between metastatic cancer and cancer chemotherapy in adults with moderate to severe trauma and the likelihood of their exposure to physical abuse/assault. We adjusted for potential confounders (socioeconomic factors) compared to controls (those without metastatic cancer and not undergoing chemotherapy). We aimed to generate evidence that may inform more holistic patient care strategies by analyzing the relationship of these variables. Our findings may be useful for clinical practice, policy, and the creation of support mechanisms aiming to reduce physical abuse/assault incidence and enhance patient outcomes.
Methods
Study Design and Data Source
We employed a retrospective cohort study and utilized the nationwide ACS-TQIP Public Use File (PUF) from 2017 to 2019.13 We selected the 2017-2019 period for our analysis because, as of August 2023, this was the only timeframe for which data on the physical abuse/assault related trauma variable (our study’s outcome) were available. The ACS-TQIP-PUF dataset comprises anonymized research data from more than 700 trauma centers across the United States, encompassing Level I-V or undesignated centers, and includes all records transmitted to the National Trauma Data Bank.13 The Multidisciplinary Trauma Research Operations Committee approved data usage. The TQP PUF Data Use Agreement adheres to the data protection guidelines of HIPAA and the ACS TQIP Hospital Participation Agreement. Access to TQP PUF data requires adherence to the Data Use Agreement terms and Institutional Review Board approval.
Population and Eligibility Criteria
The study population included adults aged 18 years and older diagnosed with moderate to severe trauma (ISS Scores: 9-75), metastatic cancer, undergoing cancer chemotherapy, and who had documented incidents or reports of physical abuse/assault related trauma in the United States between 2017 and 2019. Our inclusion criteria were: (1) age ≥18 years, (2) moderate to severe physical trauma diagnosis, (3) metastatic cancer, (4) currently receiving chemotherapy for cancer, (5) documented physical abuse/assault related trauma, and (6) complete data on relevant demographic, socioeconomic, and clinical factors. Exclusion criteria included patients aged < 18 years, those with mild or no trauma, patients without metastatic cancer and not currently undergoing chemotherapy for cancer, patients with incomplete or missing data, and patients with a history of cognitive or psychiatric disorders that might interfere with the study outcomes. Initially, our study encompassed 27,653 trauma patients, but after excluding 122 children under age 18, the final analysis focused on 27,531 trauma patients. Among these, 16,406 had moderate to severe trauma, and 17,351 trauma patients had metastatic cancer. Among those with moderate to severe trauma (n=16,406), 10,758 had metastatic cancer.
Outcomes of Interest and Definitions
The dependent variable or study outcome was reports of physical abuse/assault related trauma (yes/no), derived from ICD-10-CM external cause codes for injury events. For physical abuse, external cause codes for child and adult abuse take priority over all other external cause codes, ensuring accurate reporting in medical and legal records. There were 655 (2.38%) cases with missing data for this variable. In this study, ‘physical abuse’ reflects the operational definition used in the ACS-TQP dataset, which may include both intentional assaults and certain injury mechanisms (e.g., falls, transport-related, animal-related) when coded in the context of possible abuse. As a result, this measure does not fully align with the conventional clinical or psychological definition of intentional physical abuse. Our analyses, therefore, interpret ‘physical abuse’ as trauma cases identified by these codes, recognizing that both intentional and injury-related mechanisms may be represented. Because ACS-TQIP prioritizes ICD-10-CM abuse/assault codes, some mechanisms (e.g., burns, traffic injuries, falls) may be coded within these cases when the injury occurred in the context of suspected or confirmed abuse; thus, our outcome reflects ‘reported abuse/assault-related trauma’ rather than strictly intentional abuse.
The independent variables in this study were the presence of metastatic cancer (yes/no) and current cancer chemotherapy (yes/no). The comparison group comprises patients without metastatic cancer and not receiving chemotherapy for cancer. Metastatic cancer (yes/no) was defined using the TQP PUF comorbidity flag ‘Metastatic cancer’ and/or ICD-10-CM secondary malignancy codes (C77–C79) present on admission. Current chemotherapy (yes/no) was defined as documentation of active systemic antineoplastic therapy during the index encounter, identified by the TQP treatment flag and/or the ICD-10-CM diagnosis code Z51.11 (‘encounter for antineoplastic chemotherapy’).
Confounders were modeled based on Anderson’s Model of Health Service Utilization, which encompasses predisposing factors, enabling factors, and need-for-care factors.14 This behavioral model identified those variables in need of adjustment for the analysis of factors influencing patient utilization of health services (such as chemotherapy for cancer). Predisposing and enabling factors consisted of age categories (18-89 years), with a subdivision into Younger and Middle-Aged Adults (18-55 years) and Senior and Elderly Adults (56 years and above). Sex was categorized as Male or Female, and Race and Ethnicity were classified as non-Hispanic White (White) and non-White. The primary methods of payment included Medicaid, Self-Pay, Private/Commercial Insurance, Medicare, Other Government, or Not Billed. Facilities were stratified by bed size, with categories including ≤ 200 beds, 201-400 beds, 401-600 beds, and > 600 beds. Furthermore, facilities were classified by their ACS Verification Level as Level I, II, or III Trauma Centers, as well as by State Designation into categories I, II, III, IV, or Other/Not applicable. The dataset covers the years of discharge as 2017, 2018, and 2019. Need-for-care factors were categorized by injuries based on their intent, which included accidents such as falls, collisions, and animal-related incidents; assaults or abuse indicating intentional harm by others; self-harm or suicide attempts; and other causes such as legal, medical, or environmental events. The mechanisms of injury were classified as transport-related, which encompassed motor vehicle occupants, motorcyclists, cyclists, and pedestrians; accident and environment-related, such as falls, natural bites, and being struck by or against objects; and intentional and self-inflicted, including firearm injuries, pierces, and fire-related traumas. Trauma types were recorded as blunt, penetrating, burn, or other/unspecified. The ICD-10 Primary External Cause Code specifies injuries, poisoning, transport accidents, man-made environmental accidents such as jumping or diving, exposure to smoke, fire, heat, and natural forces, and causes of morbidity, including assaults and unspecified causes. Places of injury were coded as home and residential areas, sports and athletic areas including schools, commercial and service areas such as farms and construction sites, and other specified and unspecified non-institutional places. The total Intensive Care Unit (ICU) length of stay was noted as short (1-3 days), moderate (4-9 days), or long (10+ days). Emergency Department (ED) discharge disposition included inpatient admissions, outpatient or home, transfers to other facilities, and special circumstances such as deceased patients or those who left against medical advice. Hospital discharge disposition involved further care requirements, home and self-care, specialized non-acute care, legal or law enforcement transfers, patient self-directed discharge, and deceased patients. The presence of any comorbid condition was noted as either yes or no. The Injury Severity Score was categorized as minor for scores less than 9, and moderate to severe trauma for scores between 9 and 75. Finally, the Glasgow Coma Scale/Score (GCS) for Traumatic Brain Injury severity was classified as mild for scores between 13 and 15, and moderate to severe trauma for scores between 3 and 12.
Statistical Analysis
First, we carried out a bivariate analysis using Stata version 1815 to identify predisposing, enabling and need-for-care factors associated with reports of physical abuse/assault among trauma patients from 2017 to 2019, using chi-square and Fisher’s exact (for small samples <5) tests to determine the significance of associations between the outcome and various factors. After the bivariate analyses, we respectively conducted adjusted logistic regression analyses to assess the impact of metastatic cancer and chemotherapy on the outcome among patients with moderate to severe trauma, guided by Anderson’s conceptual model. In the first model (Model 1), we adjusted for predisposing and enabling factors, while in the second model (Model 2), we included all factors: predisposing, enabling, and need-for-care. This approach enabled us to discern the individual and combined associations of these factors on the odds of physical abuse/assault among patients with moderate to severe trauma. Finally, we reported the frequencies, percentages, P values, and adjusted odds ratios (OR) along with their 95% confidence intervals (CIs), with P values ≤0.05 considered statistically significant.
Results
Physical Abuse/Assault Reporting Among Trauma Patients
In our nationwide analysis of trauma patients from 2017 to 2019 (Table 1), we found 50 out of 27,531 patients (0.19%) reported physical abuse/assault. In contrast, among those with moderate to severe trauma (ISS, between 9-75), 26 out of 16,261 patients (0.16%) reported assault. These findings underscore a concerning—yet relatively low—frequency of reported physical abuse/assault related trauma in both groups.
Predisposing and enabling factors associated with physical abuse/assault reporting among trauma patients, 2017-2019
Among all trauma patients, those with metastatic cancer reported higher incidents of physical abuse/assault compared to those without (72.00% vs. 28.00%; P value: 0.271). This trend was more pronounced in patients with moderate to severe trauma, where a substantial majority with metastatic cancer reported physical abuse/assault (84.62% vs. 15.38%; P value: 0.040). In the context of receiving chemotherapy for cancer, a lower percentage of all trauma patients undergoing chemotherapy reported physical abuse/assault compared to those not receiving chemotherapy (38.00% vs. 62.00%; P value: 0.199). This finding was more evident among those with moderate to severe trauma, where patients receiving chemotherapy reported less frequent physical abuse/assault than those not receiving it (26.92% vs. 73.08%; P value: 0.045).
Senior and elderly adults (aged 56 and above) reported higher incidents of physical abuse/assault in overall trauma patients than younger and middle-aged adults aged 18-55 years (83.33% vs. 16.67%; P values: 0.077). Similarly, senior and elderly adults reported higher incidents of physical abuse/assault in those with moderate to severe trauma compared to younger and middle-aged adults (83.33% vs. 16.67%; P values: 0.118). Females reported higher incidents of physical abuse/assault than males in overall trauma patients (60.00% vs. 40.00%; P value: 0.192). Similarly, females reported higher incidents of physical abuse/assault in those with moderate to severe trauma compared to males (46.15% vs. 53.85%; P values: 0.671). Non-Hispanic white individuals reported physical abuse/assault more frequently than non-white individuals in all trauma patients (67.35% vs. 32.65%; P value: 0.001) and those with moderate to severe trauma (69.23% vs. 30.77%; P value: 0.020). Among all trauma patients, those using Medicare reported the highest rates of physical abuse/assault (55.10%), followed by private/commercial insurance (18.37%), Medicaid (12.24%), self-pay (10.20%), and other government or not billed (4.08%) with a P value of 0.001. In the moderate to severe trauma group, similar trends were observed, with Medicare users again reporting the highest incidence of physical abuse/assault (64.00%), followed by private/commercial insurance and self-pay (both 8.00%), Medicaid (5.02%), and other government or not billed (4.00%) with a P value of 0.009.
Focusing on the injury intent (Table 2), among all trauma patients who reported physical abuse/assault, the most common intent was assaults (56.00%), followed by accidents (42.00%) and other intents (2.00%) with a P value of ≤0.001. For those with moderate to severe trauma, assaults and accidents remained prevalent (both 50.00%), with a P value of ≤0.001.
Need for care factors associated with physical abuse/assault reporting among trauma patients, 2017-2019
In the mechanism of injury category, among all trauma patients who reported physical abuse/assault, accident and environment was the most reported mechanism (87.76%), significantly higher than intentional and self-inflicted injuries (10.20%) and transport-related injuries (2.04%), with a P value of 0.001. Among patients with moderate to severe trauma, this trend persisted, with accident and environment injuries accounting for 92.31% of physical abuse/assault reports, followed by intentional and self-inflicted (7.67%), yielding a P value of 0.016. In our analysis of trauma type among all trauma patients who reported physical abuse/assault, blunt trauma was predominant, reported in 87.50% of incidents, which was significantly higher compared to penetrating trauma (8.33%), burn injuries (2.08%), and other/unspecified injuries (2.08%), with a P value of 0.003. In patients with moderate to severe trauma, a similar pattern was observed, with blunt trauma accounting for 88.46% of cases, followed by penetrating (7.69%), and other/unspecified injuries (3.85%), yielding a P value of 0.012.
Regarding the ICD10 primary external cause code, among all trauma patients reporting physical abuse/assault, the highest category was causes of morbidity (assault by unarmed brawl/fight) at 46.00%, followed by man-made environmental accidents such as jumping, diving, and sharp leaves (38.00%) and exposure to smoke, fire, heat, and natural forces (10.00%). In contrast, injuries, poisoning, and other consequences were only 4.00%, and transport accidents and related injuries were 2.00% (P value: <0.001). Among patients with moderate to severe trauma, the pattern was similar, with man-made environmental accidents constituting 50.00%, causes of morbidity at 42.31%, and exposure to smoke, fire, heat, and natural forces (7.67%), with a P value of <0.001. In terms of the ICD-10 place of injury code, home and residential areas were the most common locations for physical abuse/assault among all trauma patients, reported in 84.00% of cases (P value: 0.003). This percentage increased to 96.15% of patients with moderate to severe trauma, with a P value of 0.034. Other specified and unspecified non-institutional places accounted for 8.00% of the incidents, while sports, athletic areas, and commercial service areas were less frequently reported.
In trauma patients who reported physical abuse/assault, we examined the ISS and GCS Scores. Regarding the ISS, among all trauma patients reporting physical abuse, 52.00% experienced moderate to severe trauma (Scores 9-75), compared to 48.00% among those with minor trauma (Scores <9), with a P value: 0.183. In the assessment of GCS Score, among all trauma patients who reported physical abuse/assault, 80.00% had a mild traumatic brain injury (TBI) (Scores 13-15), which was significantly higher compared to those with moderate to severe TBI (Scores 3-12) at 20.00% (P value < 0.001). This significant difference was also evident in patients with moderate to severe trauma, where 69.23% had a mild TBI compared to 30.77% with a moderate to severe TBI (P value <0.001).
Metastatic Cancer and Physical Abuse/Assault Risk in Trauma Patients
In our adjusted analyses, we examined the independent association of metastatic cancer on the risk of physical abuse/assault in patients with moderate to severe trauma, after accounting for various predisposing, enabling, and need-for-care factors (Table 3). In the first adjusted model, which considered predisposing and enabling factors (age, gender, race and ethnicity, method of payment, facility levels: bed size and state designation level, ACS verification levels and year of discharge), we found patients with metastatic cancer experienced substantially higher odds of reporting physical abuse/assault compared to those without metastatic cancer (OR: 7.847; 95% CI: 1.021-60.337). When we added/adjusted for need-for-care factors in our second model (ICU total length of stay, any comorbid condition, and GCS Score), this association remained significant, though the magnitude slightly lowered. In this model, the odds of physical abuse/assault for patients with metastatic cancer were 6.637 times higher than those without metastatic cancer (95% CI: 0.857 - 51.411). These findings confirm that while trauma characteristics are important descriptive features, the central pattern is a significantly heightened vulnerability to abuse/assault among patients with metastatic cancer, which remained robust after adjustment for multiple covariates.
Metastatic cancer and physical abuse/assault risk in trauma patients
Receiving Chemotherapy for Cancer and Physical Abuse/Assault in Trauma Patients
We focused on the adjusted impact of currently receiving chemotherapy for cancer on the risk of physical abuse in patients with moderate to severe trauma, employing two adjusted models and incorporating various factors (Table 4). In the first adjusted model (which included predisposing and enabling factors such as age, gender, race and ethnicity, method of payment, facility levels, ACS verification, and year of discharge), patients currently receiving chemotherapy for cancer showed a lower odds of reporting physical abuse compared to those not receiving chemotherapy (OR: 0.395, 95% CI: 0.122 - 1.277), suggesting a trend toward lowered risk, although this did not reach statistical significance. When we further adjusted for need-for-care factors in our second model, which included ICU total length of stay, any comorbid condition, and GCS Score TBI Severity, the association remained in the same direction. The odds in this model indicated an even stronger trend toward a lowered risk of physical abuse/assault in patients undergoing chemotherapy, albeit, still not reaching conventional levels of statistical significance (OR: 0.308 95% CI: 0.084 - 1.138). However, the small number of abuse/assault cases produced wide confidence intervals, and this trend did not meet statistical significance, limiting the ability to draw firm conclusions.
Receiving chemotherapy for cancer and physical abuse/assault risk in trauma patients
Discussion
This study directly addresses the research question by demonstrating that metastatic cancer was associated with higher odds of reported abuse/assault in adults with moderate to severe trauma, while patients in chemotherapy showed a lower proportion/association than those not in chemotherapy at the bivariate level. These findings highlight the dual and contrasting associations of disease burden and treatment on abuse risk. Our findings contrast with prior work, which emphasized cumulative injury risk among women experiencing different forms of abuse without considering cancer or treatment-specific contexts.2 In our study, patients actively receiving chemotherapy reported lower physical abuse/assault-related trauma at the bivariate level, which may reflect the frequent clinical contact and structured monitoring inherent in oncology care, which could plausibly facilitate earlier detection of abuse or greater monitoring. Our analysis incorporates the dual burden of metastatic cancer and chemotherapy in a nationwide trauma population, unlike studies that focus solely on psychosocial vulnerability to domestic violence.1,3 Moreover, socioeconomic and structural inequities known to be associated with both health outcomes and abuse risk4-6 were accounted for in our adjusted models, strengthening the interpretation that healthcare engagement during chemotherapy may buffer abuse risk. Although chemotherapy appeared protective, this association was not statistically significant in our adjusted models, and the wide confidence intervals underscore the need for cautious interpretation and replication in larger datasets. Thus, while earlier literature underscores vulnerability, our study highlights a potential protective factor linked to treatment-related healthcare oversight. The higher risk associated with metastatic cancer may be explained by mechanisms described in prior oncology and trauma research, including social isolation, caregiver dependency, functional decline, and financial toxicity, potentially contributing to higher observed odds of abuse/assault. These factors can exacerbate household stress and power imbalances, potentially increasing the likelihood of abuse/assault. Although these mechanisms were not directly measured in our dataset, their presence in the broader literature provides context for our observed association and underscores the need for integrated psychosocial support in this population.
Three key takeaways emerged based on the adjusted models. Firstly, the study revealed a significantly higher risk of physical abuse/assault related trauma with metastatic cancer, as evidenced by the adjusted odds ratio of 7.85, highlighting the vulnerability of this group to abuse (OR: 7.847; 95% CI: 1.021-60.337). Secondly, the analysis indicates patients undergoing chemotherapy for cancer were less likely to report physical abuse/assault related trauma (OR: 0.395, 95% CI: 0.122 - 1.277), though not statistically significant. Finally, the study underscores the importance of considering a range of predisposing, enabling, and need-for-care factors, as these are significantly associated with the likelihood of physical abuse/assault in trauma patients, pointing towards the need for a comprehensive, multifactorial approach in patient care and protection strategies.
Our study’s exploration into the intersection of metastatic cancer, chemotherapy, and physical abuse/assault in trauma patients reveals additional insights that both align with and diverge from existing literature. Trevillion et al.1 highlight the higher vulnerability of individuals with mental health conditions to domestic violence, underscoring the broader context of vulnerability among those with serious health conditions, including cancer. Our findings extend this understanding by specifically illustrating how metastatic cancer heightens the risk of physical abuse/assault (a major form of domestic violence) in trauma patients, a connection less explicitly made in previous research. Additionally, our study contrasts with Mechanic, Weaver, and Resick,2 who focus on risk factors for physical injury in battered women, by providing a detailed analysis of how chemotherapy may be protective against physical abuse, a novel area of investigation. Furthermore, the work of Macy, Ferron, and Crosby,3 which discusses the chronic health problems stemming from partner violence, provides a backdrop for our study’s emphasis on the need for targeted interventions and support systems. Unlike previous studies that broadly address health outcomes following domestic violence, our research contributes specific evidence regarding the intersection of trauma, cancer treatment, and abuse, advocating for integrated care approaches that address the unique vulnerabilities of this patient population.
To contextualize the robustness and potential limitations of our findings, it is essential to consider both the methodological strengths and the inherent constraints of our study. The study’s strengths lie in its use of nationwide trauma registry data from the TQP PUF provided by the American College of Surgeons. The registry ensures a substantial and diverse sample size, enhancing the findings’ generalizability. It featured comprehensive inclusion and exclusion criteria, delineating a specific patient population and minimizing potential confounding. A multifaceted analytical approach was employed, incorporating descriptive statistics, chi-square analysis, Fisher’s exact test, and adjusted logistic regression models to provide a thorough analysis of the relationships between variables. The study also adjusted for various confounding factors using Anderson’s Model of Health Service Utilization, which helped to reduce potential biases and solidify the reliability of outcomes. Lastly, a rigorous statistical analysis was conducted using Stata version 18, adhering to strict statistical standards to ensure significance, reinforcing the study’s commitment to precision.
On the other hand, a key limitation is the definition of ‘physical abuse’ within the trauma registry, which is derived from ICD-10 coding and may capture both intentional assaults and injury-related mechanisms (e.g., falls, environmental accidents). Consequently, the outcome should be interpreted as abuse- or assault-related trauma events as coded in the dataset, rather than narrowly as interpersonal violence. We acknowledge our study could not identify the perpetrator or source of abuse (e.g., spouse, caregiver, stranger), which restricts the ability to infer patterns of domestic violence specifically. Although the background highlighted the role of domestic violence, the ACS-TQP dataset does not specify perpetrator type; therefore, our outcome captures reported abuse/assault-related trauma broadly, not exclusively domestic violence. Also, since abuse, metastatic disease, and chemotherapy are ascertained from ICD-10 coding and registry fields, under-reporting and misclassification are possible; such errors are likely non-differential with respect to the outcome and would bias estimates toward the null. Our study was based on retrospective data from 2017 to 2019, which may not accurately reflect current practices or trends within the patient population. The exclusion of patients with incomplete or missing data could lead to selection bias, limiting the generalizability of the findings. Focusing only on documented incidents of physical abuse/assault might result in an underestimation of the actual prevalence, as many incidents go unreported. Additionally, there was a potential for residual confounding, as not all predisposing, enabling, and need factors could be adjusted for. The nature of observational data used in the study further constrained the ability to establish causality between the presence of metastatic cancer, cancer chemotherapy, and the study outcome, although this was not the intended purpose of the study.
Study Implications
Despite the above limitations, our study unveils critical implications for clinical practice and multidisciplinary considerations in the care of trauma patients with metastatic cancer. The higher vulnerability to physical abuse/assault in this population necessitates a more vigilant approach to trauma and oncology care. Current practices should integrate routine screening for abuse, especially among patients with metastatic cancer, as they demonstrated higher odds in this study. This integration can lead to early detection and intervention, potentially improving patient outcomes and safety. The importance of this practice change is underscored by the need to address not only the physical aspects of patient care, but also the psychosocial dimensions that significantly impact health outcomes.3 In addition to healthcare providers, social workers and mental health professionals should develop targeted support and intervention strategies for patients and families dealing with the compounded stress of cancer, trauma, and abuse. Policymakers should consider these findings when allocating resources for healthcare settings, with a focus on ensuring the safety and well-being of populations with compounded vulnerabilities. This study highlights the necessity for an interdisciplinary approach in healthcare, where medical treatment is informed by psychosocial support and policy frameworks, acknowledging the observed associations between health, abuse-related trauma, and socioeconomic factors.2
Conclusion
Our study sheds light on the critical intersection of metastatic cancer, chemotherapy, and physical abuse/assault within the trauma patient population, revealing significant associations that warrant further investigation and consideration by clinical and social support communities. Our analysis, grounded in a comprehensive retrospective cohort study, contributes valuable insights into the existing literature by elucidating the factors associated with higher odds of physical abuse/assault among patients with serious health conditions. Through rigorous statistical analysis and careful consideration of a range of predisposing, enabling, and need-for-care factors, we offer a foundation for future research, policy formulation, and clinical practice improvements. Ultimately, our work emphasizes the need for a multifactorial approach in patient care, advocating for enhanced measures, targeted support, and a deeper understanding of the vulnerabilities faced by trauma patients with metastatic cancer and those undergoing chemotherapy.
Acknowledgements
We are particularly grateful for the use of the Trauma Quality Programs Participant Use File (TQP PUF) provided by the American College of Surgeons. The TQP remains the full and exclusive copyrighted property of the American College of Surgeons. The American College of Surgeons is not responsible for any claims arising from works based on the original Data, Text, Tables, or Figures. Special thanks to the trauma centers and their dedicated staff across the United States for their invaluable role in data collection and patient care, which formed the foundation of our research. We also acknowledge the support of the Marshfield Clinic Health System Multidisciplinary Trauma Research Operations Committee for their approval and oversight of our use of the data.
Footnotes
Funding Disclosures: This project was funded in part through philanthropic support of Marshfield Clinic Research Institute, led by the Marshfield Clinic Health System Foundation (255800-00-RES SUPPT TRAUMA).
Author Contributions: Dr. Shour made substantial contributions to the conception and design of the study, played a leading role in the analysis and interpretation of data, and the drafting of the work. Dr. Rhodes contributed significantly to the design of the study, the acquisition and interpretation of the data, and the drafting of the work. Dr. Puthoff was involved in drafting the manuscript and made substantial contributions to the data interpretation. Dr. Sharma provided clinical insights for data interpretation and critically reviewed the manuscript. Dr. Onitilo played a pivotal role in the conception and design of the study, the interpretation of data, and drafting the manuscript. All authors critically reviewed the manuscript for intellectual content, approved the final version to be published, and agree to be accountable for all aspects of the work. All authors meet the International Committee of Medical Journal Editors criteria for authorship for this manuscript, take responsibility for the integrity of the work as a whole, and have given their approval for this version to be published.
- Received February 7, 2025.
- Revision received August 26, 2025.
- Accepted October 17, 2025.
References
- 1.↵Trevillion K, Oram S, Feder G, Howard LM. Experiences of domestic violence and mental disorders: a systematic review and meta-analysis. PLoS One. 2012;7(12):e51740. doi:10.1371/journal.pone.0051740.
- 2.↵Mechanic MB, Weaver TL, Resick PA. Risk factors for physical injury among help-seeking battered women: an exploration of multiple abuse dimensions. Violence Against Women. 2008;14(10):1148-1165. doi:10.1177/1077801208323792.
- 3.↵Macy RJ, Ferron J, Crosby C. Partner violence and survivors’ chronic health problems: informing social work practice. Soc Work. 2009;54(1):29-43. doi:10.1093/sw/54.1.29.
- 4.↵Yu MY, Sarri R. Women’s health status and gender inequality in China. Soc Sci Med. 1997;45(12):1885-1898. doi:10.1016/S0277-9536(97)00127-5.
- 5.↵Sekine M, Chandola T, Martikainen P, Marmot M, Kagamimori S. Socioeconomic inequalities in physical and mental functioning of British, Finnish, and Japanese civil servants: Role of job demand, control, and work hours. Soc Sci Med. 2009;69(10):1417-1425. doi:10.1016/j.socscimed.2009.08.022.
- 6.↵Cain BT, Horns JJ, Huang LC, McCrum ML. Socioeconomic disadvantage is associated with greater mortality after high-risk emergency general surgery. J Trauma Acute Care Surg. 2022;92(4):691-700. doi:10.1097/TA.0000000000003517
- 7.↵Kim B, Royle M. Domestic Violence in the Context of the COVID-19 Pandemic: A Synthesis of Systematic Reviews. Trauma Violence Abuse. 2024;25(1):476-493. doi:10.1177/15248380231155530
- 8.Smith SG, Zhang X, Basile KC, Merrick MT, Wang J, Kresnow M, Chen J.; National Center for Injury Prevention and Control (U.S.). Division of Violence Prevention. The National Intimate Partner and Sexual Violence Survey: 2015 data brief – updated release; 2018. Available at: https://stacks.cdc.gov/view/cdc/60893 Accessed December 12, 2023
- 9.↵Sardinha L, Maheu-Giroux M, Stöckl H, Meyer SR, García-Moreno C. Global, regional, and national prevalence estimates of physical or sexual, or both, intimate partner violence against women in 2018. Lancet. 2022;399(10327):803-813. doi:10.1016/S0140-6736(21)02664-7.
- 10.↵Carlson LE, Zelinski EL, Toivonen KI, Prevalence of psychosocial distress in cancer patients across 55 North American cancer centers. J Psychosoc Oncol. 2019;37(1):5-21. doi:10.1080/07347332.2018.1521490.Medline
- 11.Mason TM, Tofthagen CS, Buck HG. Complicated Grief: Risk Factors, Protective Factors, and Interventions. J Soc Work End Life Palliat Care. 2020;16(2):151-174. doi:10.1080/15524256.2020.1745726.
- 12.Carr JB, Packham A. SNAP Schedules and Domestic Violence. Journal of Policy Analysis and Management. 2021;40(2):412-452. doi:10.1002/pam.22235.
- 13.↵American College of Surgeons. Trauma Quality Programs Participant Use File. Published 2023. Available at: https://www.facs.org/quality-programs/trauma/tqp/center-programs/ntdb/datasets Accessed December 12, 2023.
- 14.↵Andersen RM. Revisiting the behavioral model and access to medical care: does it matter? J Health Soc Behav. 1995;36(1):1-10. doi:10.2307/2137284.
- 15.↵StataCorp. Stata Statistical Software: Release 18. College Station, TX: StataCorp LLC. Available at: https://www.stata.com/support/faqs/resources/citing-software-documentation-faqs/ Accessed December 12, 2023.
