Detection and Referral of Orbital and Ocular Injuries Associated With Intimate Partner Violence Following an Educational and Screening Initiative in an Emergency Department

Key Points

Question

Is an intimate partner violence (IPV) educational campaign with standardization of emergency department screening protocols associated with improved referral of patients with IPV-associated orbital fractures and ruptured globes to social work and law enforcement?

Findings

This quality improvement analysis revealed an increase in referral of patients with IPV-associated orbital floor or zygomaticomaxillary complex fractures to social work and law enforcement following an education and screening initiative.

Meaning

These findings suggest the combination of an educational campaign on IPV injury patterns and enhanced emergency department screening protocols can improve referral patterns of patients affected by IPV and may potentially be lifesaving.

This cohort study examines the number of adult female patients with intimate partner violence–associated orbital floor fractures, zygomaticomaxillary complex fractures, and ruptured globes referred to ancillary services after an educational and screening intervention for health care professionals.

Abstract

Importance

Intimate partner violence (IPV) is a substantial cause of morbidity and mortality in the US. Previous studies indicate gaps in identifying and referring female patients with IPV-associated orbital and ocular injuries to ancillary services.

Objective

To determine the number of IPV-associated orbital floor fractures, zygomaticomaxillary complex (ZMC) fractures, and ruptured globes referred to ancillary services in adult female patients following an educational and screening intervention to health care professionals.

Design, Setting, and Participants

This single-center retrospective quality improvement analysis examined electronic medical records of adult female patients seen in a single level 1 trauma center emergency department and ophthalmology clinic between January 2015 and February 2019, after the initiative began. Female adults who sustained orbital floor fractures, ZMC fractures, or ruptured globes were included. Preinitiative data were previously collected between January 1995 and January 2015 on adult female patients and published. Data analysis for this study occurred from May 2020 to September 2020.

Interventions

A 2-part, ongoing initiative began January 2015. First, enhancement of IPV screening protocols in the emergency department was conducted. Second, an educational campaign on IPV injury patterns was presented to residents and faculty in ophthalmology, emergency, otolaryngology, and trauma departments.

Main Outcomes and Measures

Comparison of ancillary service involvement preinitiative (January 1995 to January 2015) and postinitiative (January 2015 to February 2019).

Results

A total of 216 adult female patients (mean [SD] age, 55.0 [22.7] years; age range, 18-99 years) sustained orbital floor or ZMC fractures postinitiative. A total of 22 of 216 (10.2%) sustained fractures from IPV compared with 31 of 405 (7.6%) preinitiative (95% CI, −2.2% to 7.3%; P = .28). Documented social work referrals (11 of 31 preinitiative vs 20 of 22 postinitiative; difference, 55% [95% CI, 35%-76%]; P < .001), homegoing safety assessments (1 of 31 preinitiative vs 18 of 22 postinitiative; difference, 79% [95% CI, 61%-96%]; P < .001), and law enforcement involvement (7 of 21 preinitiative vs 16 of 22 postinitiative; difference, 50% [95% CI, 26%-74%]; P < .001) were higher in patients who presented after the initiative with orbital floor and ZMC fractures. A total of 51 adult female patients (mean [SD] age, 57.7 [20.8] years; age range 20-93 years) sustained ruptured globes postinitiative. A total of 5 of 51 patients (9.8%) sustained injury due to IPV postinitiative, compared with 5 of 141 (3.5%) preinitiative (95% patients, −2.5% to 15.0%; P = .08).

Conclusions and Relevance

Following the start of the initiative, referral patterns of adult female patients with IPV-associated orbital fractures improved. Targeted IPV screening of patients with orbital and ocular injuries is essential for effective intervention.

Introduction

Intimate partner violence (IPV) is a substantial cause of morbidity and mortality in the US. More than 1 in 3 women (36.4%; 43.6 million) and 1 in 3 men (33.6%; 37.3 million) experience IPV in their lifetime, with more severe injuries occurring among women.¹ Forty-five percent of female homicides in the US are perpetrated by an intimate partner.² Injuries caused by IPV disproportionally affect the head, neck, upper torso, and upper extremities, with 1 study reporting ocular injuries in 45% of female patients affected by IPV.³ A case-control study revealed an increased likelihood of IPV among female patients with orbital fractures.⁴ In a small study of individuals with orbital fractures, one-third of female patients sustained their injuries through IPV, compared with 0 male patients.⁵

A previous retrospective case series at the study institution included 405 female patients with orbital fractures and determined IPV to be the third leading causative mechanism, with exceedingly low rates of documented ancillary service involvement in patients who had experienced assault (1.7%).⁶ A follow-up retrospective case series of 190 female patients with ocular trauma revealed assault as the fourth leading causative mechanism, with IPV accounting for nearly one-third of assaults.⁷ After these investigations, the study institution implemented an IPV initiative with the goal of improving referral patterns.

Methods

Intervention was initiated in January 2015. The emergency department standardized IPV screening protocols, requiring all adult patients be screened by nursing staff, who were cued by the electronic medical record. The educational arm of this intervention consisted of grand rounds on IPV injury patterns and screening delivered to the residents and faculty members. Further details on the initiative can be found in the eMethods in the Supplement.

Postinitiative data were collected through a retrospective review identifying adult female patients 18 years and older with orbital floor fractures, zygomaticomaxillary complex (ZMC) fractures, or ruptured globes between January 2015 to February 2019. Data extraction was initiated in April 2019. Preinitiative data collected between January 1995 to January 2015 had previously been extracted and published (details in the eMethods in the Supplement).^6,7 Male adults were excluded from this study because their likelihood of substantial morbidity from IPV is lower than that of female adults.² Previously published preinitiative data were also only collected for adult female patients.^6,7

International Classification of Diseases, Ninth Revision (ICD-9) and Tenth Revision (ICD-10) codes were used to identify patients with orbital floor fractures, ZMC fractures, or ruptured globes. The following data were extracted from the medical records of identified patients: age at time of injury, mechanism of injury, clinical course, final best-corrected visual acuity, and assailant, if applicable. For patients whose assault was secondary to IPV, referral to ancillary services, homegoing safety assessment, and law enforcement involvement were determined. The study received institutional review board approval at the University of Iowa, adhered to the standards of the Declaration of Helsinki, and was compliant with the Health Insurance Portability and Accountability Act. Informed consent was waived in accordance with institutional review board policy due to the inability to consistently and safely communicate with participants.

Statistical analyses were performed on preinitiative and postinitiative data using Pearson χ² tests or Fisher exact tests, with differences given with 95% CIs. The significance threshold was set at P < .05, 2 tailed, and statistical analysis was completed with SAS version 9.4 and SAS/STAT version 14.3 (SAS Institute).

Results

A total of 216 adult female patients (mean [SD] age, 55.0 [22.7] years; age range, 18-99 years) were identified with orbital floor fractures (176 of 216 [81.5%]) or ZMC fractures (40 of 216 [18.5%]). The mechanism of injury was documented in every case, and leading mechanisms included falls (118 of 216 [53.5%]), motor vehicle crashes (33 of 216 [15.9%]), non-IPV assault (29 of 216 [13.4%]), IPV assault (22 of 216 [10.2%]), and unintentional injury by inanimate object (7 of 216 [3.2%]). The mean (SD) age of patients treated for IPV assault was 39.4 (9.8) years (age range, 25-63 years). In the IPV assault group, ZMC fractures accounted for 8 of 22 injuries (36%) compared with 5 of 29 (17%) in the non-IPV assault group. Perpetrators of IPV included current nonspousal male partners (16 of 22 [73%]), husbands (2 of 22 [9%]), ex–nonspousal male partners (2 of 22 [9%]), and unspecified intimate partners (2 of 22 [9%]). In 7 of 22 IPV assault cases (32%), surgical repair of the orbital fractures was required, compared with 3 of 29 (10%) in the non-IPV assault group, with a risk difference of 22% (95% CI, −2% to 45%; P = .08).

A total of 51 patients with ruptured globes were identified (mean [SD] age, 57.7 [20.8] years; age range, 20-93 years). Leading documented mechanisms of injury were falls (30 of 51 [59%]), IPV assault (5 of 51 [10%]), non-IPV assault (5 of 51 [10%]), accident by inanimate object (5 of 51 [10%]), and animal-associated injuries (3 of 51 [6%]). In the IPV assault group, the mean (SD) age was 43.6 (9.8) years (range, 33-57 years). The perpetrator of IPV was a current nonspousal male partner in 4 of 5 cases and an ex–nonspousal male partner in 1 of 5 cases. Four of 5 patients (80%) had a final best-corrected visual acuity of hand motion or worse. Of the patients affected by IPV, 1 required enucleation, and 3 of 5 were offered enucleation but elected to undergo observation.

The rates of IPV in the preinitiative and postinitiative groups are in Table 1.^6,7 Comparisons of documented social work referrals, safety assessments, and law enforcement involvement preinitiative and postinitiative are in Table 2.^6,7 In patients with orbital floor fractures and/or ZMC fractures, documentation of social work referrals (11 of 31 preinitiative vs 20 of 22 postinitiative; difference, 55% [95% CI, 35%-76%]; P < .001), homegoing safety assessments (1 of 31 preinitiative vs 18 of 22 postinitiative; difference, 79% [95% CI, 61%-96%]; P < .001), and law enforcement involvement (7 of 21 preinitiative vs 16 of 22 postinitiative; difference, 50% [95% CI, 26%-74%]; P < .001) increased following the start of the initiative. Further details on the injury patterns and chronicity of IPV can be found in the eResults in the Supplement.

Table 1. Rates of Orbital Floor Fractures, Zygomaticomaxillary Complex (ZMC) Fractures, and Ruptured Globes Associated With Intimate Partner Violence.

Intimate partner violence–associated injury	Patients, No./total No. (%)		Difference, % (95% CI)	P value
	Preinitiative	Postinitiative
Orbital floor and/or ZMC fractures	31/405 (7.6)a	22/216 (10.2)	2.5 (−2.2 to 7.3)	.28
Ruptured globes	5/141 (3.5)b	5/51 (9.8)	6.3 (−2.5 to 15.0)	.08

^aData from January 1, 1995, to April 29, 2013, previously published by Clark et al.⁶

^bData from January 1, 1995, to January 1, 2015, previously published by Cohen et al.⁷

Table 2. Rates of Documented Ancillary Service Involvement.

Intimate partner violence–associated injury	Patients, No./total No. (%)		Difference, % (95% CI)	P value
	Preinitiative	Postinitiative
Orbital floor and/or zygomaticomaxillary complex fractures
Social work referral	11/31 (36)a	20/22 (91)	55 (35 to 76)	<.001
Safety assessment	1/31(3)a	18/22 (82)	79 (61 to 96)	<.001
Law enforcement involvement	7/31 (23)a	16/22 (73)	50 (26 to 74)	<.001
Ruptured globes
Social work referral	4/5 (80)b	5/5 (100)	20 (−34 to 72)	>.99
Safety assessment	3/5 (60)b	4/5 (80)	20 (−43 to 73)	>.99
Law enforcement involvement	2/5 (40)b	4/5 (80)	40 (−30 to 87)	.52

^aData from January 1, 1995, to April 29, 2013, previously published by Clark et al.⁶

^bData from January 1, 1995, to January 1, 2015, previously published by Cohen et al.⁷

Discussion

Following the initiative, referral rates of patients with IPV-associated orbital floor or ZMC fractures increased. This combination of educational interventions and institutional changes has been shown to improve referral patterns of patients treated for IPV in the primary care setting.^8,9,10 Emergency department–based electronic screening protocols have been shown to increase the likelihood of discussion and disclosure of IPV and use of domestic violence services.¹¹ To our knowledge, this is the first study demonstrating improved referral patterns following enhanced screening protocols in the emergency department and education on ophthalmic and orbital IPV injuries.

This study supports previous findings that IPV injuries tend to be more severe.^6,7 Fractures of the ZMC result from greater impact force than isolated orbital floor fractures.¹² These ZMC fractures accounted for 40 of the 216 (18.5%) fractures but were nearly twice as many (8 of 22 [36%]) within the IPV cohort.

Previous studies indicate that physicians often underestimate IPV rates in their patient populations, highlighting the importance of education on IPV injury patterns and screening protocols.¹³ When patients in the emergency department disclose IPV to their health care professionals, the likelihood that they will use an intervention increases 4-fold.¹⁴ With such a substantial portion of orbital floor, ZMC fractures, and ruptured globes resulting from IPV, screening could be vision saving and could potentially redirect a fatal course. Targeted screening of female patients with orbital floor fractures, ZMC fractures, or ruptured globes of unknown causative mechanism is strongly encouraged, and sample phrasing is provided in the Box.¹⁵

Box. Recommended Intimate Partner Violence Screening.

Introduce the Topic With the Patient Unaccompanied

• “Because violence is common in many people’s lives, there are some questions that I ask every patient.”

Disclose Mandatory Reporting, If Applicable

• Screen directly:

• “Have you ever been physically, sexually, or emotionally abused by an intimate partner?”

• “Are your current injuries a result of this kind of abuse?”

Respond to Positive Screening Results

• “I am glad you shared this with me, and I am sorry that this happened to you.”

• “This is not your fault.”

• “You are not alone.”

• “Help is available.”

• “Are there children in the home who are at risk?”

Refer to Ancillary Services

• “Would you like to speak with a social worker to learn about local or national resources?”

Assess Homegoing Safety

• “Do you feel safe going home?”

• Endorse patient’s wishes on how to proceed.

Limitations

Physicians were not individually tracked; therefore, we cannot directly attribute improved referral patterns to clinicians who attended the lectures. Improvement could be attributed to increased general awareness of IPV and efforts to include IPV in medical education. Another plausible explanation could be cultural shifts that have allowed women to feel more comfortable disclosing IPV to their health care professionals. This sample was limited to female adults, but it is reasonable to infer that screening would be prudent in male patients with orbital fractures or ruptured globes of unknown causative mechanism. Other limitations include the retrospective nature of this study, particularly with regard to paper medical records. Finally, teenagers also sustain injuries from intimate partners; however, only patients 18 years or older were included.

Conclusions

Intimate partner violence is a major cause of morbidity and mortality among female people. A high index of suspicion is essential when evaluating female patients with orbital and ocular trauma. In this study, education on IPV injury patterns and enhanced screening protocols were associated with improved referral patterns.

Supplement.

eMethods. Details of pre-intervention and intervention

eResults. Intimate partner violence injury patterns and chronicity