Epidemiology and Characteristics of Women with Facial Fractures Seeking Emergency Care in the United States: A Retrospective Cohort Study

Abstract

Study Design: Facial bone fractures in women are less common than in men in the United States. However, little is known about the epidemiology of women who sustain facial fractures. Objective: Our aim is to describe the patient population of women seeking emergency care for facial fractures in the United States and they type and cost of care received in this setting. Methods: This is a retrospective cohort study using the 2019 National Emergency Department Sample. The inclusion criterion was diagnosis of facial fracture. The primary outcome was patient characteristics. The secondary outcomes are emergency department (ED) characteristics, discharge disposition, total visit charges, and most common cause. Diagnoses and procedures were identified using ICD10-CM codes. Outcomes were compared to men. Results: Thirty-seven percent of ED facial fractures were encountered in women. Both women and men were most likely adult, Caucasian, from the lowest median income quartile, sustained nasal bone fractures, and presented to a southern, metropolitan, private nonprofit, non-trauma ED. Conclusions: Women were older, more likely insured by Medicare and less likely by private insurance, discharged home, and had lower ED charges than their male counterparts. However, the financial burden of emergency care for facial fractures among women was $1.6 billion.

Key Points

Question: What characterizes U.S. women’s emergency care experience for facial fractures, especially in comparison to the general data?

Main Finding: While U.S. women sustain fewer facial fractures than men, 29% are nasal bone injuries. The average age of affected women is 53. 40% use Medicare for coverage. The majority (93%) are adult Caucasians, treated in metropolitan EDs (53%) and teaching hospitals (67%). The discharge rate post-treatment for women is 71% with an average ED charge of $9734.

Implication: Even with lower individual costs and overall favorable outcomes, the financial burden of emergency care for facial fractures among women is substantial at $1.6 billion/year.

Introduction

Fractures of the facial bones in women are known to be less common than in men in the United States although there appears to be a recent rise in violence directed towards women based on data during the early pandemic [1]. There is a bimodal distribution of facial fractures in women at 25-35 years and greater than 65 years [1,2]. The specified etiology of these injuries can be subdivided into those caused by physical aggression, falls, motor vehicle accidents, sports, work accidents; however, most common etiologies differ by age. In the postmenopausal elderly there is an almost 2-fold increased odds for facial fractures compared to younger women [2,3]. Incidence of facial fractures are increasing more in elderly individuals, likely due to factors that are conducive to falls such as osteoporosis, decreased coordination, and reduced muscle strength [4]. On the other hand, assault is the most common mechanism of injury in younger women presenting with facial trauma, with many maxillofacial injuries stemming from intimate partner violence [5,6]. In fact, compared to women with injuries limited to other areas of the body, young women presenting to the ED with head, neck, and facial injuries are 7.5-times more likely to be victims of intimate partner violence [7]. Other characteristics regarding women with the highest incidence of facial fractures include those with a low socioeconomic status, unemployment, and residing in a major city [2,5].

These injuries are of particular concern because they can have long-term functional, aesthetic, and psychosocial impacts [5,8,9]. Facial fractures result in long-term mental and physical health issues for patients, including deficits in mastication, speech, swallowing, and vision. Facial injuries can also result in disfigurement and affect self-image. Additionally, other serious sequelae of these injuries include unrelenting epistaxis, nasal airway compromise, secondary deformity, concomitant traumatic brain injury, open facial wounds, and an increased 1-year risk of stroke [10,11,12].

Maxillofacial trauma is a major challenge for healthcare services due to the high incidence and financial cost in the United States. The management of facial fractures often requires surgery with long-term follow-up and possible revisions, which can overwhelm healthcare systems and increase costs. There is a need to ascertain more about the epidemiology and characteristics of women who sustain facial fractures, both for primary prevention initiatives to decrease rates of facial fractures, and to enable clinicians to better understand and address the role of disparities and social factors. Therefore, we sought to describe the patient population of women who seek emergency care for facial fractures in the United States as well as the fracture distribution, costs of care, and discharge dispositions.

Methods

Data Source and Study Design

This study is a retrospective cohort study based on data from the 2019 Nationwide Emergency Department Sample (NEDS). The NEDS is a comprehensive database produced and maintained by the Healthcare Cost and Utilization Project, a collaborative effort involving the Federal-State-Industry partnership sponsored by the Agency for Healthcare Research and Quality [13]. The NEDS is the largest publicly available ED database in the United States, with over 28 million unweighted ED visits annually, representing an estimated 123 million weighted ED visits. It is a nationally representative sample sourced from the State Inpatient Databases and State Emergency Department Databases, providing information about patients seen in the ED who are either discharged or admitted to the same hospital. Weights were assigned to each discharge using the AHA universe as the standard to assess nationwide estimates [14]. Participating in the NEDS are 40 states and the District of Columbia. It is an approximate 20% stratified sample of US hospital-owned EDs. Additionally, the NEDS includes data on principal and secondary diagnoses, along with in-hospital procedures. The study was determined exempt from the Fox Chase Cancer Center Institutional Review Board approval IRB number 23-9921, as it involved an analysis of de-identified existing data, ensuring patient privacy and confidentiality.

Study Population

In this study, the study population comprised patients with a primary diagnosis of facial fracture. Notably, male patients were excluded from the main study population, but their data were still utilized for comparison purposes. The comparison involved adjusting for potential confounding factors using multivariate regression analysis. The specific ICD-10 CM facial fracture codes were used to identify eligible patients. These codes can be found in Appendix A of the study.

Outcomes

The main focus of the study was to analyze the total number of ED visits in 2019 using the NEDS database. In addition to this primary outcome, the study also examined several secondary outcomes, which included the in-hospital admission rate, total ED charges, and the most common Current Procedural Terminology (CPT) codes used in the ED. These outcomes were specifically compared between females and males who presented with facial fractures. To ensure accurate comparisons over time, all costs were adjusted for inflation using the consumer price index.

Study Variables

The database contains both patient-level and hospital-level information. Patient-level information includes age, sex, insurance carrier, discharge disposition, as well as cost-related measurements such as total emergency department charges and length of stay. Hospital-level information includes hospital teaching status, urban-rural designation, and region. Furthermore, the NEDS provides data on hospital trauma status, total emergency department charges, and the patient’s race. The in-hospital mortality rate is coded as the patient’s vital status at the time of discharge. Total charge and length of stay are directly recorded in the NEDS. The total emergency department charges represent the amount of money the hospital billed for the services provided during the patient’s encounter, and the total emergency department costs represent the amount of money the hospital paid for the services provided [14].

Statistical Analysis

The statistical analysis was conducted using STATA MP, version 14.0 (STATACorp, College Station, TX). STATA’s svyset and svy commands were utilized to account for the sampling design complexities, generating nationally indicative variance estimates, unbiased results, and P-values. Unadjusted likelihood ratios were calculated for primary and secondary outcomes through univariable logistic regression analysis. Multivariable logistic regression analysis adjusted for potential confounders identified during the univariate analysis, selecting variables with a P-value cut-off of .2 and strong correlation with the outcome. Additionally, clinically significant variables were incorporated into the model, irrespective of their P-values from univariate analysis. Proportions and continuous variables were compared using Fisher exact and Student t test, respectively, with a statistical significance threshold of .05.

Results

Patient Characteristics

In 2019, a total of 180,407 women presented to the ED with facial fractures, accounting for 37% of all facial fracture encounters. The fracture location was the nasal bones (29%), mandible (11%), maxillary/malar/zygoma (9%), orbit (8%), and skull and facial bone combination (5%), similar to men (P = .55) (

Table 1. Facial Fracture Distribution.

Total Facial Fractures	% (180,407)
Nasal bones	29% (52,318)
Orbit	8% (14,433)
Maxillary/Malar/Zygoma	9% (16,237)
Mandible	11% (19,845)
Skull and facial bone combination	5% (9020)
Unspecified	38% (68,555)

). Facial fracture was the principal diagnosis in two thirds of encounters, 31% of which were on weekends (

Table 2. Patient Characteristics.

Patient Characteristics	% (n)
Race
Caucasian	71% (128,089)
African American	13% (23,453)
Hispanic	10% (18,041)
Asian	2% (3608)
Other	4% (7216)
Median income in the patient’s zip code
$1–$45,999	30% (54,122)
$46,000– $58,999	24% (43,298)
$59,000–$78,999	23% (41,494)
$79,000 or more	22% (39,690)
Primary payer
Medicare	40% (72,163)
Medicaid	19% (34,277)
Private	27% (48,710)
Self-pay	10% (18,041)
Other	4% (7216)
Weekend encounter	31% (55,926)
Encounter season
Winter	24% (43,298)
Spring	25% (45,102)
Summer	26% (46,906)
Fall	25% (45,102)
Disposition from ED
Discharge home	71% (128,089)
Transfer to short-term hospital	5% (9020)
Other transfers, e.g., skilled nursing facility, intermediate care	3% (5412)
Admitted as an inpatient to this hospital	20% (36,081)
Died in ED	<1% (<1804)
Against medical Advice	<1% (<1804)
Total charge in ED	$9734
Adult	93% (167,779)
Intent of injury
Assault	31% (55,926)
Unintentional	60% (108,244)
Intentional	<1% (1804)

). Notably, facial fractures were rare during pregnancy but were attributed to physical assault in 31% of cases. The majority of fractures not related to assault were unintentional (60%, intentional <1%). The mean age was 53 years, significantly older than men (42 years, adjusted mean difference (aMD): 11.6 (11.0-12.0), P < .01). Regarding insurance coverage, most women had Medicare (40%), followed by private insurance (27%), Medicaid (19%), self-pay (10%), and other forms of insurance (4%). Conversely, most men were privately insured (29%), followed by Medicaid (23%), self-pay (20%), Medicare (18%), and other forms of insurance (8%) (P < .01).

Similar to men (P > .05), the majority of women were adults (93%), from the lowest income quartiles ($1–$45,999: 30%, $46,000 $58,999: 24%, $59,000–$78,999: 23%, $79,000 or more: 22%) and Caucasian (71%, African American 13%, Hispanic 10%, Other 4%, Asian 2%). Females presented evenly throughout the year compared with males (Winter: 24%, Spring: 25%, Summer: 26%, Fall: 25%) (P = .52).

Hospital Characteristics

This study provides a comprehensive overview of the distribution of facial fracture encounters in women based on various characteristics in the United States (

Table 3. Hospital Characteristics.

Hospital Characteristics	% (n)
Hospital urban-rural location
Large metropolitan areas with at least 1 million residents	53% (95,616)
Small metropolitan areas with less than 1 million residents	33% (59,534)
Micropolitan areas	8% (14,433)
Not metropolitan or micropolitan	5% (9020)
Hospital control
Government or private	15% (27,061)
Government, non-federal, public	12% (21,649)
Private, nonprofit, voluntary	51% (92,008)
Private, invest-own	10% (18,041)
Private, collapsed category	12% (21,649)
Hospital region
Northeast	19% (34,277)
Midwest	23% (41,494)
South	40% (72,163)
West	19% (34,277)
Hospital trauma center level
Nontrauma center	45% (81,183)
Trauma level 1	24% (43,298)
Trauma level 2	18% (32,473)
Trauma level 3	13% (23,453)
Teaching status of hospital
Metropolitan nonteaching	19% (34,277)
Metropolitan teaching	67% (120,873)
Nonmetropolitan	14% (25,257)

). The majority of these encounters were observed in hospitals located in large metropolitan areas (53%), defined as regions with at least 1 million residents. Additionally, 33% of patients sought treatment at hospitals situated in small metropolitan areas with less than 1 million residents, and 8% in micropolitan areas. Regarding hospital control, the largest proportion of hospitals (51%) where patients sought treatment fell under the category of private, nonprofit, voluntary institutions. Government or private hospitals constituted 15% of the total, while government, non-Federal, public hospitals accounted for 12%.

The South had the highest concentration of admissions, making up 40% of the total. The Midwest followed with 23%, while both the Northeast and West regions had an equal share of hospitals at 19% each. In terms of trauma center encounters, 45% of hospitals that females presented to with facial fractures were classified as non-trauma centers, indicating that they did not specialize in trauma care. Trauma Level 1 centers accounted for 24% of encounters, while Trauma Level 2 centers represented 18% of the total. The teaching status of the hospitals where patients presented also varied significantly, with metropolitan teaching hospitals having the most admissions, comprising 67% of the sample. Metropolitan nonteaching hospitals accounted for 19% of encounters, while nonmetropolitan teaching hospitals constituted 14%.

In-Hospital Admission Rate

This study examined the mortality rate and discharge outcomes for women with facial fractures (Table 2). The mortality rate, which represents the percentage of patients who died during their hospital stay or in the ED, was found to be 1% which is similar to men (P = .84). However, the study did identify a notable difference in the discharge outcomes between men and women. Specifically, a higher proportion of women (71%) were discharged home compared to men (65%). This difference in discharge disposition was found to be statistically significant, with a P-value of .02.

Healthcare Resource Utilization

The average total ED charges were $9,734, which was less than men (aMD: $650 ($865 - $434), P=<.01). The mean total charge in the ED for males was $9993 (CI:$ 9280-$10,587). The overall ED healthcare cost was $1.6 billion for women in 2019 (Table 2). The total financial burden for males is $2.87 billion.

Most Commonly Used CPT Codes

The most commonly used CPT codes during the ED encounters for women are summarized in

Table 4. Most Common CPT Codes for Female Facial Fracture Patients.

	CPT Code	Description	No.
1	70450	CT scanning of the head or brain, without contrast	21,725
2	70486	CT scanning of the maxillofacial area, without contrast	20,314
3	99284	ED visit for evaluation and management level 4, moderate MDM	19,089
4	72125	CT scanning of the cervical spine, without contrast	12,266
5	99285	ED visit for evaluation and management level 5, high MDM	11,099
6	99283	ED visit for evaluation and management level 3, low MDM	10,317
7	36415	Collection of venous blood by venipuncture	4389
8	12011	Simple repair of superficial wounds to the face, ears, eyelids, nose, lips, and/or mucous membranes, that are 2.5 cm or less	4159
9	74177	CT scanning of the abdomen and pelvis, with contrast material(s)	2899
10	70160	XR of the nasal bones, three views	2298

. The most common procedure was a CT of the head or brain, without contrast, which was obtained in 12% of female patients who presented to the ED with facial fractures. In male patients, the most common procedure was a CT scan of the maxillofacial area, without contrast. Level 4 ED visits were the most common for females with facial fractures, followed by Level 5 and then Level 3 ED visits.

Discussion

This study provides valuable insights into the epidemiology and characteristics of women who seek emergency care for facial fractures in the United States. In 2019,180,407 women presented to the ED with facial fractures, representing 37% of all facial fracture cases. Women who present with facial fractures were older than their male counterparts, more likely to have Medicare insurance and less likely to be privately insured. Women are potentially more likely to have fractures related to falls due to bone density differences between sexes, with fall-related injuries occurring more often in females [15]. Most female patients were Caucasian, from the lowest median income quartile, and sustained nasal bone fractures, which was similar to men. Interestingly, this study found that facial fractures rarely occurred during pregnancy. Physical assault comprises 31%, but there is no delineation between domestic vs other forms of assault. We found that women presented to the ED at similar rates across all seasons in the US, despite other international studies demonstrating seasonality with more falls seen in the winter related to adverse weather [16]. Additionally, most of the women seeking emergency care for facial fractures presented to large metropolitan areas, teaching hospitals, and non-trauma EDs in the southern region of the United States. The increased ED encounters in the south were the highest possibly due to encompassing more of the US population in the NRD [14]. The outcomes of emergency care for facial fractures in women showed that the mortality rate was comparable to that of men, but women were more likely to be discharged home. The average total ED charges for women with facial fractures were numerically lower than those for men, indicating potentially more cost-effective care for women in this context. However, despite lower individual ED charges and total ED encounters, the financial burden of emergency care for facial fractures among women was substantial, amounting to $1.6 billion for the year 2019.

The higher likelihood of women being discharged home indicates that they were more likely to have a favorable recovery and did not require further hospitalization or transfer to other care facilities after their medical treatment. Conversely, men may have required more complex care needs or had conditions that necessitated further hospitalization or transfer to short-term hospitals or other care facilities. This difference could suggest more cost-effective care in women or it can be considered that men may be involved in higher energy trauma, resulting in more extensive injuries or complex care needs. It is important to note that while the study found a statistically significant difference in discharge outcomes between men and women, the mortality rate was not significantly different. This suggests that, despite the discrepancy in discharge dispositions, both sexes had a similar risk of mortality during their hospital stay or ED visit. By acknowledging the most common causes and social situations of patients, more effective care can be offered and steps could also be potentially taken to prevent those fractures. For example, when a young woman presents with facial trauma, a complete facial skeletal survey should be performed. Additionally, it should be noted the likelihood that this patient may be a victim of assault. Conversely, an older patient who presents with head trauma from a fall may get a CTscan of the head to rule out intracranial bleeds, but maxillary/facial CT should be included in the workup as well to assess for facial fractures. The same data can inform potential interventions aimed at preventing those fractures. We propose that such interventions can include awareness campaigns against domestic violence to encourage and empower women to confidentially report abuse. They can also include the provision of phone hotlines and safe houses that women at imminent risk of abuse can use before the assault happens. Among elderly women, these measures can potentially include awareness campaigns aimed at increasing osteoporosis screening and treatment rates as well as physical rehabilitation for the women most at risk for falls. Further research is needed to validate those suggestions and uncover others that might be beneficial in this setting. The economic burden on women is also notable; this study has shown that the majority of these patients belong to the lowest median income bracket, and use Medicare. With the average charge of visit being $9,734, this places an enormous financial burden on individuals already in strenuous financial situations, therefore additional care must be taken to avoid extraneous use of resources and materials.

Two important articles were reviewed, focusing on risk factors and sex differences in adult facial fractures. The first study by Pham et al involved a retrospective analysis of the National Trauma Data Bank (NTDB) database, which is managed by the American College of Surgeons [9]. The NTDB encourages voluntary participation and includes patients who presented as trauma activations, encompassing over 7.5 million records from 850 trauma centers across all 50 states [9,17]. The second significant study, conducted by Hanba et al, utilized the National Electronic Injury Surveillance System, which provides details on visits to a sample of US EDs and collects data on consumer product-related injuries occurring in the United States [3,18]. In our project, we opted to use the NEDS, the largest all-payer emergency department database in the United States, to obtain national estimates of hospital-owned ED visits. This unweighted database contains data from over 28 million ED visits annually, covering 995 hospitals. Since it approximates a 20-percent stratified sample of U.S. hospital-owned EDs, it provides a more nationally representative snapshot of the United States population. One notable difference from these studies is our ability to examine resource utilization in ED encounters, which we have explored in this project specifically in relation to facial fractures [9].

Compared to Pham et al, our study design encompasses a wider range of facial fractures and focuses primarily on women. Similarly, Hanba et al found that the most commonly affected race was Caucasian, but they found that Asians were also significantly more at risk, whereas our study found this population to be the least at risk out of all races identified. This discrepancy could be due to the larger sample size of the NEDS, giving our study an advantage which might more closely represent the true population. Neither study included the cost of health care, compared the discharge dispositions or cost of healthcare among different sexes, income demographics, or used CPT codes to quantify diagnoses and interventions.

The weaknesses of this study primarily emanate from its retrospective review using NEDS, as it is an administrative database. The reliance on ICD-10 CM codes and procedures, rather than direct clinical and laboratory assessments, may have led to misclassification or the omission of a small subset of patients or procedures [19]. However, it is worth noting that previous studies have confirmed that ED coding demonstrates high agreeability and reliability [20].

Additionally, misclassification is an error rather than a bias, and it does not necessarily alter the direction of association between the exposure variable and the outcome. Nonetheless, it might be more challenging to establish statistical significance. It is essential to acknowledge that most of the presented outcomes are positive, which minimizes the likelihood of a beta error. Another limitation of the study is the omission of records regarding medications or laboratory information. Consequently, these factors were not included in the analysis. Instead, healthcare resources were measured using total hospitalization charges and costs.

Our study possesses several noteworthy strengths. Firstly, the utilization of the NEDS as the largest ED database helps to minimize the risk of beta errors. The national representation of the NEDS is another advantage, as it includes patients from 40 states, encompassing various locations, teaching statuses, and ownership types. This highlights the valuable data available in the NEDS, in which its broad representation enhances the generalizability of our findings. Moreover, the substantial sample size provided by the NEDS grants us ample statistical power to identify and analyze factors such as patient characteristics, ED charges, and estimates of household income. This stands in stark contrast to the limitations of single-center studies, which often suffer from smaller sample sizes and reduced statistical power. These strengths contribute to the robustness and reliability of our study, providing a more comprehensive understanding of the factors under investigation and supporting the validity of our conclusions.

Conclusions

These findings provide valuable insights into the distribution and characteristics of encounters for facial fractures in women in the United States. The findings underscore the significance of providing appropriate care to this patient population to ensure optimal outcomes and to address the considerable economic burden associated with emergency care for facial fractures. Further investigation into the reasons behind the differences in discharge outcomes between men and women are needed to determine if any specific factors or medical conditions played a role in these discrepancies. Understanding such factors could help healthcare providers tailor their treatment and discharge planning to optimize patient outcomes for both sexes.