Missed Opportunities in Orthopaedics for Intimate Partner Violence Identification

Abstract

Background:

Intimate partner violence (IPV) commonly manifests as a musculoskeletal injury, yet the majority of orthopaedic surgeons estimate IPV to be rare in “their” orthopaedic patients. This work aimed to (1) provide education on the prevalence and manifestations of IPV, (2) investigate departmental referral patterns to Domestic Abuse Intervention Programs (DAIP) at 2 high-volume quaternary-academic centers, and (3) characterize IPV cases identified by orthopaedic surgery providers.

Methods:

The DAIP registry from 2000 to 2024 was queried for patients reporting IPV. Referrals were categorized by department and provider type. The electronic health record of patients referred by orthopaedics was further investigated for case characteristics. Findings were summarized as count/percentages and referral rates compared via the 2-proportion z-test with alpha set at 0.05.

Results:

A total of 11,227 patients were referred to DAIPs. The most common referrals were from the emergency department (ED) (29.3%; n = 2,393), behavioral health (18.2%; n = 2039), and obstetrics/gynecology (8.4%; n = 939), while only 0.3% (n = 30) patients were referred by orthopaedic surgery providers (p < 0.001). Patients referred by orthopaedics were commonly female (83.3%; n = 25) and identified during an inpatient encounter (76.7%; n = 23). Half (53.3%; n = 16) presented with an injury sustained from abuse, while the remaining patients presented for scheduled care. Injuries included high-energy injuries such as subtrochanteric, open tibia, bicondylar tibial plateau, and nongeriatric elbow fractures. Orthopaedic referrals to DAIPs relied on social workers (93.3%; n = 28). In the inpatient/ED setting, patients initially disclosed to bedside nurses (56.0%; n = 14), residents (20.0%; n = 5), and advanced practice providers (16.0%; n = 4), while patients disclosed to attendings (60%; n = 3) and fellows (40%; n = 2) in the outpatient setting. The majority (76.7%; n = 23) of patients reporting IPV-only interacted with orthopaedic providers in the 6 months preceding IPV disclosure. IPV identification led to patient safety coordination (16.7%; n = 5), alternative care plans (23.3%; n = 7), and resource assistance applications (23.3%; n = 7). Patients referred to DAIPs remained engaged in services a median 6.8 years after referral.

Conclusions:

Orthopaedic surgery referrals to DAIPs are significantly lower than other specialties, highlighting missed opportunities for intervention in both inpatient and outpatient contexts. Enhancing IPV awareness and screening in orthopaedics could improve patient safety and long-term support. Integration of artificial intelligence has the potential to facilitate efficient targeted screening within existing practice models.

Level of Evidence:

Level III (retrospective cohort). See Instructions for Authors for a complete description of levels of evidence.

Introduction

Physical, psychological, and/or sexual violence by an intimate partner, known as intimate partner violence (IPV), is the most widespread form of violence against women globally. The World Health Organization (WHO) and United Nations (UN) reported in 2018 that up to 1 in 3 women of reproductive age will experience IPV over their lifetime, equating to approximately 641 million women globally and 24.1 million women in the United States^1-4. Of these, up to 1 in 3 women (307 million globally and 8 million in the United States) are recent or active survivors/victims^1-3,5.

Musculoskeletal injuries are the second most common manifestation of IPV⁶. However, efforts to improve IPV identification and intervention have largely focused on primary care, emergency/urgent care (ED), and obstetrics/gynecology (Ob/Gyn) settings⁷. The PRAISE study, a multinational study across 12 orthopaedic clinics, found that 1 in 6 women recently experienced IPV with 1 in 50 presenting with an injury directly related to IPV. Furthermore, physical abuse severity was higher in orthopaedic settings^8-11.

Orthopaedic providers are often the first-contact providers for IPV survivors/victims^5,6,8,11-15. The United States Preventive Services Task Force and American Academy of Orthopaedic Surgeons (AAOS) encourage IPV screening in musculoskeletal care, yet there remains a discrepancy in current orthopaedic practice patterns^16-20. A recent survey reported only 4% of orthopaedic trauma providers routinely screen high-risk patients^16,17. Misperceptions are highly prevalent in orthopaedics such as underestimating IPV prevalence (i.e., <1% of patients experiencing IPV), concern about causing harm through screening, unfamiliarity with IPV response, and weak connections to local Domestic Abuse Intervention Programs (DAIPs) which contribute to current IPV screening patterns in orthopaedics^17,18,21-25.

Despite the clear relevance of IPV to orthopaedics^9-11,26,27, there remain several barriers to optimizing care for IPV patients in orthopaedic settings. It is imperative to first define current orthopaedic practices to evaluate and improve subsequent practices for IPV patients. Thus, this study aimed to (1) provide education on the prevalence and manifestations of IPV, (2) investigate referral patterns to DAIP at 2 high-volume quaternary-academic centers, and (3) characterize IPV cases identified by orthopaedic surgery providers.

Methods

Data Sources

Following institutional review board approval (#2016P002096), the DAIP registries from 2 high-volume level 1 trauma centers were reviewed to identify patients who reported IPV. The DAIP database spans 2000 to 2024 and is separately maintained from the electronic health record (EHR). Referrals from all encounter types (including emergency, inpatient, and ambulatory settings) were included.

Patient Selection

All DAIP referrals undergo an initial evaluation by licensed social workers who document referral source, IPV characteristics (emotional, physical, and sexual abuse), and support needs. Patients not reporting IPV on intake are referred to alternative resources and are not enrolled in the DAIP registry.

Referral Sources

Referrals were categorized by department and provider type (see the supplementary materials). Provider roles included social workers, physicians (attendings, trainees), advanced practice providers (APPs; nurse practitioners, physician assistants), nurses (RNs not in care management roles), administrators (front desk staff, practice managers, human resource employees), care management (case managers, patient navigator advocates), affiliated providers (midwives, medical assistants, physical/occupational therapists, etc), and unspecified/unknown if there was insufficient information to identify the referral source on DAIP intake. Patients who directly contacted the IPV program were classified as self-referrals.

Orthopaedic Referrals to IPV Program

Patients referred by an orthopaedic provider were further investigated to garner patient, abuse, injury, and referral context characteristics from the institutional EHR. Records were reviewed to identify which provider first documented IPV concerns and received IPV disclosure.

Historical Considerations

Key historical factors include (1) the DAIPs included in this work were established in 1997, (2) IPV was declared a public health emergency within the state of study in 2008, (3) a comprehensive IPV support pathway was developed in 2010 within the DAIPs to deliver equitable trauma-informed care regardless of gender identity and/or race, and (4) the institutions of study transitioned from paper to an electronic EHR in 2015. At this time, specialty departments at the institution of study could opt-in to enable EHR prompts to complete routine IPV screening. During the period of study, the orthopaedic department did not have this feature enabled; however, the results of this work prompted an institutional change.

Statistical Analysis

Broad referral patterns were summarized as absolute rates across the full patient cohort identified and relative rates within departments (counts/percentages). Relative referral rates were compared between orthopaedic and other specialties via the 2-proportion z-test with alpha set at 0.05. There were no statistical adjustments applied as the universal null hypothesis was that orthopaedic providers had similar referral patterns to other specialties. All statistical analysis was conducted in R.

Results

Institutional Referral Patterns to DAIPs

From 2000 to 2024, 11,227 patients were referred to DAIPs from a total cohort of 10,899,807 patients (Fig. 1). Most referrals originated from the ED (29.3%), behavioral health (18.1%), and self-referrals (14.8%). Ob/Gyn and primary care contributed 5% to 8% of referrals each. Medical and surgical specialties accounted for 1% to 2% each. Orthopaedic referrals comprised 0.3% (n = 30), the lowest relative rate across all departments (p < 0.001, Table I).

Fig. 1.

Longitudinal trends of patients reporting IPV referred to DAIPs at 2 high-volume level 1 trauma centers by specialty. DAIPs = Domestic Abuse Intervention Programs, ED = emergency department, IPV = intimate partner violence, OB/GYN = obstetrics and gynecology, and Ortho = orthopaedic surgery.

TABLE I.

Referrals to DAIPs by Department

Referring Department	N (%)
Total referrals*	11,227
Emergency department	3,292 (29.3)
Behavioral health sciences	2,039 (18.1)
Self-referral	1,668 (14.8)
Other†	1,522 (13.5)
Obstetrics/gynecology	939 (8.4)
Primary care/medicine	588 (5.2)
Oncology	221 (2.0)
Medical subspecialty	187 (1.7)
General surgery and subspecialties	129 (1.2)
Orthopaedics	30 (0.3)

DAIPs = Domestic Abuse Intervention Programs.

^*Other includes referrals from administrators and affiliated providers.

^†Not included above are 612 patients with insufficient information on intake documentation to identify the department of the referral source.

Referrals to DAIPs by Provider Role

Social workers made over half of referrals (56.5%, n = 6,356), significantly more than any other provider role (p = 0.03). Physicians and self-referrals each accounted for about 10% to 15%. Nurses and APPs referred similar numbers of patients. Administrators, care coordinators, and affiliated providers made few referrals. In orthopaedics, most referrals (93.3%, n = 28) were made by social workers, with a small number from orthopaedic nurses (8%; n = 2, p = 0.90) (Table II).

TABLE II.

Referrals to DAIPs by Provider Type

Provider Role	All (n = 11,227)* N (%)	Orthopaedic (n = 30) N (%)
Social work	6,356 (56.5)	28 (93.3)
Self-referral	1,712 (15.2)	-
Physician	1,178 (10.5)	-
APP	439 (3.9)	-
Nurse	370 (3.3)	2 (8.0)
Administrator	143 (1.3)	-
Care management	146 (1.3)	-
Affiliated providers	56 (0.5)	-

APP = advanced practice provider, and DAIP = Domestic Abuse Intervention Programs.

*Not included above are 827 patients with insufficient information on intake documentation to identify the referring provider's role in the care team.

IPV Support Direct Referrals by Department and Provider Role

In departments with high referral rates (ED, behavioral health, primary care), physicians contributed 13% to 14% of referrals with high referrals from physicians in medicine/primary care (22%, p = 0.86). In surgical specialties (Ob/Gyn, general surgery), physicians and APPs made about 8% of referrals. In orthopaedics, providers overwhelmingly relied on social workers (93.3%, p = 0.93) for referrals, similar to oncology teams (91.4%, p = 0.99).

Referrals to IPV Support by Orthopaedic Surgeons

Thirty patients (median age 49 years, interquartile range [IQR] 35, 56) mostly female (83.3%) were referred to DAIPs by orthopaedic providers. Most referrals occurred during an inpatient encounter (76.7%), while a minority were referred during an ED (6.6%) or outpatient (16.7%) encounter (Table III).

TABLE III.

Characteristics of Patients and Orthopaedic Care for Patients Reporting IPV

Patient and Care Characteristics	N = Number of Patients (%)
Female	N = 25 (83.3)
Average age (IQ)	46 yrs (35, 36)
Encounter type
ED	2 (6.6)
Inpatient	23 (76.7)
Outpatient	5 (16.7)
Musculoskeletal injury at time of IPV disclosure	16 (53.3)
Subspecialty of scheduled care	14 (46.7)
Spine	7 (50.0)
Arthroplasty	3 (21.4)
Trauma	1 (7.1)
Upper extremity	3 (21.4)
Foot and ankle	-
Seen by primary care within 6 months of IPV disclosure	7 (23.3)
Formal preop clearance before IPV disclosure	4 (13.3)

IPV = intimate partner violence.

Categorical data reported as counts/percentages and continuous data reported as median/interquartile range.

The most common providers to first identify IPV were nurses (46.6%), followed by residents (16.7%), fellows (6.7%), attendings (10.0%), and APPs (13/3%) (Table IV). Social workers facilitated most referrals (93.3%, n = 28) but were rarely the initial point of disclosure (6.7%, n = 2) (Table IV).

TABLE IV.

Characteristics of IPV Disclosure and Characteristics for Orthopaedic Patients Disclosing IPV

IPV Case Characteristics	N = Number of Patients (%)
High-energy injury due to abuse	8 (26.7)
Provider identifying IPV
Social worker	2 (6.7)
Nurse	14 (46.6)
APP	4 (13.3)
Resident	5 (16.7)
Fellow	2 (6.7)
Attending	3 (10.0)
Documentation in EHR of IPV inquiry	7 (20.0)
Characteristics prompting IPV inquiry
Reported mechanism of injury inconsistent with severity of observed injury	9 (30.0)
Interpersonal dynamics interfering with care	9 (26.7)
Patient concerns with postoperative recovery	5 (16.7)
Explicit IPV disclosure	7 (23.3)
Abuse type
Verbal	27 (90.0)
Physical	23 (76.7)
Sexual	2 (6.7)
Immediate impact of IPV identification
Security/police involvement for immediate safety	5 (16.7)
Alternative plan of care	7 (23.3)
Facilitation of application to governmental assistance	7 (23.3)
Active engagement in DAIP after referral (years)	6.8 (5.2, 8.6)

DAIP = Domestic Abuse Intervention Programs, EHR = electronic health record, and IPV = intimate partner violence.

Categorical data reported as counts/percentages and continuous data reported as median/interquartile range.

Nurses (56.0%, n = 14), resident physicians (20.0%, n = 5), and APPs (16.0%, n = 4) identified the majority of IPV in the inpatient (n = 23) and ED (n = 2) setting. In the outpatient setting (n = 5), patients unanimously disclosed to either attending surgeons (60%, n = 3) or fellows (40%, n = 2).

Mutual IPV Patients

In the 6 months preceding IPV identification during an orthopaedic encounter, a minority (23.3%) of patients were seen by their primary care provider (PCP) and more than half were only seen by orthopaedic providers. More than half of patients had no prior documentation of any home life concerns. A minority of patients (13.3%) required preoperative clearance before elective surgery which was typically completed in preoperative clinics rather than by their PCP (Table III).

Patient and IPV Manifestation Characteristics of Patients Reporting IPV Referred to IPV Resources by Orthopaedic Surgery Providers

More than half (53.3%, n = 16) of patients presented with a musculoskeletal injury directly related to IPV, while the remaining patients presented for scheduled orthopaedic care such as shoulder/hip/knee arthroplasty (n = 4), spine decompression and/or fusion (n = 6), implant removal (n = 1), and chronic pain (n = 3) (Table III). Among those injured, half (n = 8) had high-energy injuries (e.g., subtrochanteric fractures, open tibia fractures, bicondylar tibial plateau fractures, nongeriatric distal femur fractures, tibiotalar fracture dislocations, and nongeriatric elbow fracture dislocations) (Table IV). The remaining patients had moderate energy injuries such as unstable ankle fractures, stable pelvic fractures, vertebral body compression fractures, proximal humerus fractures, first-time shoulder dislocations, isolated distal ulna fractures, and radiographic-negative soft tissue injuries to the wrist.

IPV identification was triggered by reported injury mechanisms inconsistent with the severity of injury patterns (30.0%; n = 9), explicit patient disclosure (23.3%; n = 7), abuser interference in care (26.7%; n = 9), and concerns about postoperative recovery (16.7%; n = 5). The most common initially reported mechanism for high-energy injuries sustained due to abuse was a fall down several stairs. Postoperative recovery concerns involved neglect and caregiving burdens as primary caretakers for dependents, while their functional status was altered during orthopaedic recovery. Abuser interference included verbal altercations and controlling behaviors such as frequent phone calls, visits, and dictating interactions with children/family/friends (Table IV).

Most patients reported verbal (90.0%; n = 27) and physical (76.7%; n = 23) abuse, while a minority (6.7%; n = 2) disclosed sexual abuse. IPV identification led to safety interventions (16.7%, n = 5), altered care plans (23.3%; n = 7), and/or government resource applications (23.3%; n = 7). More than half (53.3%, n = 16) remained engaged in DAIPs over a median 6.8 (IQR 5.2, 8.6) years (Table IV).

Discussion

This 24-year retrospective review of 11,227 patients found that orthopaedic providers referred significantly fewer patients to DAIPS (0.3%, p < 0.001) compared with providers in other specialties despite practicing within a high-risk specialty. Most orthopaedic referrals occurred during inpatient encounters with many patients having only seen an orthopaedic provider in the prior 6 months. More than half of patients initially presented for elective care, highlighting a critical opportunity for routine IPV screening in orthopaedic clinics even in the absence of overt musculoskeletal injury. While social workers facilitated 93.3% of referrals to DAIPs, they were rarely the care team member patients initially confided in (<5%). Patients most commonly disclosed to bedside nurses, residents, and APPs in the inpatient/ED setting, whereas in the outpatient setting, patients disclosed to attendings and fellows. Identification of IPV had both a direct influence on patient’s active care plans and long-term psychosocial support with more than half of patients remaining longitudinally engaged in DAIPs. There is a clear need for routine and standardized IPV screening in orthopaedic care.

Orthopaedic surgery providers had the lowest referral rate to DAIPs. Compared with other surgical fields with similar clinical schedules/volume, orthopaedic DAIP referrals were one-third the rate. Established IPV screening and DAIP referral practices in the ED (29.3%), behavioral health sciences (18.1%), and obstetrics/gynecology (8.4%) contributed to the higher referral rates observed for patients presenting in these care sites with high pretest probability of IPV^21,26,28,29. Recent evidence supports that orthopaedic sites should also be considered high-risk, yet screening remains infrequent and unstandardized in orthopaedics^6,8,14,30-32.

Provider referrals to DAIPs were more frequent than self-referrals, emphasizing the need for clinician engagement in IPV screening. Orthopaedic physicians and APPs had the lowest direct referral rates. These results highlight both the importance of provider education in effective IPV screening and the clear deficiency in relying solely on visible IPV messaging in patient-facing areas to facilitate self-referrals to DAIPs²⁵. The vast majority of orthopaedic referrals to DAIPs were in the context of an inpatient or emergency department encounter. This distribution could be explained by routine IPV screening protocols in the emergency department and by nursing during intake after inpatient admission. On the other hand, only 5 DAIP referrals were made in an outpatient orthopaedic context despite many patients having no other recent medical contact. This is consistent with prior reports of orthopaedic providers being key first-contact providers^5,8-10,33. Together, these findings highlight the importance of improving provider education on effective and appropriate IPV screening, as well as the clear insufficiency of relying on “other” providers to identify IPV in orthopaedic patients.

Barriers to IPV identification in orthopaedic settings are multifactorial with both system-level and provider-level factors contributing. On a systems level, high patient volumes (5-12 patients/hour), compared with internal medicine patient volumes (2-3 patients per hour), leave little opportunity for IPV inquiry during encounters^17,34,35. Fragmented care across specialties dilutes IPV warning signs and decreases the likelihood of identification⁵. Provider-level barriers include misconceptions about IPV prevalence^9,17,18, discomfort with the topic projecting onto patient perceptions of screening practices²⁴, lack of trauma-informed are training^23,33,36,37, limited awareness of local DAIP pathways, and varying bandwidth to navigate the additional cognitive load of IPV interpretation and discussion^34,38 all contribute to the rare occurrence of IPV screening during orthopaedic encounters^8,22,39.

Orthopaedic providers often lack up-to-date knowledge of community resources, contributing to reluctance to screen. Thoughtfully designed sociotechnical tools integrated into the EHR can reduce cognitive load, consolidate IPV risk indicators, streamline screening, and guide clinicians through trauma-informed care using regionally relevant structured pathways. By improving the generalizability and efficiency of IPV screening, these thoughtfully designed systems can allow providers to focus on delivering valuable first-line care for patients care rather than deliberating whether to initiate an IPV inquiry.

Targeted screening for IPV in orthopaedic surgery care settings is a high-yield opportunity to standardize IPV practices and is not perceived by patients to be an invasion of privacy^17,18,24. Current guidelines from the WHO⁴⁰, USPSTF²⁰, Center for Disease Control⁴¹, and AAOS¹⁹ are unanimous in highlighting the importance of IPV recognition and first-line intervention. In addition to a systems-based approach with community partnerships^25,42, incorporation of technological tools such as leveraging digital health records and artificial intelligence can augment the effectiveness and scalability of targeted IPV screening initiatives in orthopaedic surgery. The first critical step in improving patient safety and wellbeing is IPV identification.

Orthopaedic surgery providers have several inherent advantages to identify IPV compared with other specialties with well-established IPV screening practices. These include expertise in injury pattern assessment, longitudinal patient relationships that build trust³⁹, familiarity managing patients in pain or distress, and workflow opportunities to privately engage patients in an inquiry (e.g., provider-directed radiographs, immobilization fitting) where patients can feel safe discussing what would otherwise not be disclosed. Altogether, orthopaedic providers are uniquely poised to identify survivors/victims of IPV who otherwise are not presenting routinely for preventive care in other sites.

Not all IPV survivors/victims are ready to take steps to leave an abusive relationship at the time of disclosure. In the case of disclosure, it is imperative for orthopaedic providers to briefly mentally separate from “immediate solutions” and rather offer validation (e.g. “This is not your fault” and “No one has the right to hurt you”), assess safety, and provide support^15,16. While this approach may not seem “enough” for the gravity of the situation⁴³, offering first-line support and referral to a DAIP can empower patients to feel able to help him/herself and may be the only opportunity to offer a path to these important resources⁴⁰. In addition to improving education on appropriate IPV screening and first-line care practices during individual encounters, broad initiatives such as incorporation of AI tools to facilitate targeted screening, direct referral pathways to DAIPs, and policy changes to compensate for delivery of IPV preventive care in orthopaedic settings are all high-yield objectives to globally improve the value of care we deliver to our orthopaedic patients.

Conclusions

Orthopaedic surgery referrals to DAIPs are significantly lower than other specialties, highlighting missed opportunities for intervention in both inpatient and outpatient contexts. Enhancing IPV awareness and screening in orthopaedics could improve patient safety and long-term support. Integration of artificial intelligence has the potential to facilitate efficient targeted screening within existing practice models.

Appendix

Supporting material provided by the authors is posted with the online version of this article as a data supplement at jbjs.org (http://links.lww.com/JBJSOA/A993). This content was not copyedited or verified by JBJS.