A, Sex comparison of fracture by decade. B, Overall racial comparison of fracture. Legend details which lines correspond to which racial cohort. C, Fracture risk among men organized by age and race. D, Fracture risk among women organized by age and race. % Fracture indicates the proportion of facial injuries diagnosed as a fracture. Each data point included is an average by decade.
A, Comparison between younger (<60 years) and older (≥60 years) adults. B, Further breakdown by race.
aStatistically significant comparison between the younger and older cohorts for a particular comparison (P < .05).
A, Men. B, Women.
aStatistically significant comparison between the younger and older cohorts for a particular comparison (P < .05).
A, Younger men. B, Older men. C, Younger women. D, Older women.
Customize your JAMA Network experience by selecting one or more topics from the list below.
Hanba C, Svider PF, Chen FS, et al. Race and Sex Differences in Adult Facial Fracture Risk. JAMA Facial Plast Surg. 2016;18(6):441–448. doi:10.1001/jamafacial.2016.0714
Copyright 2016 American Medical Association. All Rights Reserved.
Are there racial and sex difference in adult facial fracture risks?
In this analysis of National Electronic Injury Surveillance System data and adults with facial injury and fracture, there was an increase in the risk of facial fracture among postmenopausal women sustaining facial injuries, particularly among whites and Asians. Black women did not have this increased fracture risk with advancing age.
Mechanisms of injury and facial fracture risk vary by age, race, and sex.
There are well-described racial, sex, and age differences related to osteoporosis and hip and/or extremity fractures. Nonetheless, there has been virtually no inquiry evaluating whether these findings carry over to facial fracture.
To characterize the incidence of facial fractures by patient demographics and injury mechanism, focusing on whether differences are noted with race, sex, and advancing age.
Main Outcomes and Measures
Retrospective analysis of the National Electronic Injury Surveillance System (NEISS) was performed in October and November 2015, specifically evaluating adult emergency department (ED) visits from 2012 to 2014 related to facial trauma. Entries were organized by age groups (both individual decades as well as younger adults [18-59 years] vs older adults [60-89 years]), sex, and race (white, black, Asian, other/unspecified). Incidence of facial fractures and mechanism of injury were also evaluated.
There were 33 825 NEISS entries correlating to 1 401 196 ED (range, 1 136 048-1 666 344) visits for adult facial injury, with 14.4% involving fracture. A greater proportion of facial injuries among younger men (<60 years) were fractures relative to younger women (15.5% vs 12.5%; difference of the mean [DOM], 3.0%; 95% CI, 2.8%-3.1%; P < .001); however, on comparison by sex in elderly populations (≥ 60 years), women had an increased fracture predilection (15.0% vs 14.0%; DOM, 1.0%; 95% CI, 0.8%-1.2%; P < .001). Also, older women had a significantly greater risk of fracture relative to those younger than 60 years (15.0% vs 12.5%; DOM, 2.5%; 95% CI, 2.4%-2.7%; P < .001), a comparison that was significant among whites and Asians. Black women had a significantly decreased risk of fracture in the older aged population. (8.4% vs 9.1%; DOM, 0.7%; 95% CI, 0.2%- 1.3%; P = .001). Both on individual comparisons of younger and older cohorts, white and Asian individuals of either sex had significantly greater rates of facial fracture injury than blacks. Among younger cohorts in either sex, injuries sustained during participation in recreational activities were a significant factor, replaced largely by injuries due to housing structural elements and falls among older cohorts.
Conclusions and Relevance
There is an increase in the risk of facial fracture among postmenopausal women sustaining facial injuries, with these results significant among whites and Asians. In contrast, a decreased risk was noted on comparison of younger and older black women. Mechanism of injuries also varied significantly by age, race, and sex.
Level of Evidence
Numerous analyses have evaluated clinical characteristics of older adults sustaining hip and extremity fractures, facilitating preventive efforts to target populations harboring an increased fracture risk. In our current health care environment, characterized by increasing consciousness of health care costs, understanding these demographic trends is essential for both clinicians and policymakers in an effort to minimize associated morbidity, mortality, and expenditures in our increasingly elderly population.1 Bone architectural and mineral density studies have demonstrated sex-, racial-, and age-related variations in the hip, vertebral body, and distal forearm bone strength, and have linked this information to demographic specific fracture incidences. Furthermore, many analyses have established a lower bone mineral density among whites,2-6 increased structural disorganization among the elderly population,2,4,5,7-11 and an accelerating decline in bone strength for postmenopausal women.7-16 Consequently, these changes in bone health lead to a lifetime hip fracture risk in white women as high as 1 in 6.17
There is a paucity of scholarship examining facial bones and their propensity to fracture with advancing age, despite extensive analysis of fracture characteristics and studies of bone health throughout the rest of the body. In one of the few studies evaluating facial bones and osteoporosis, facial bones were noted to possess age-related deterioration, similar to bones elsewhere in the body.16 Lee and colleagues16 retrospectively examined 96 patients who underwent both facial computed tomographic (CT) scans and dual-energy x-ray absorptiometry (DEXA) studies, noting a significantly reduced bone density among midfacial bones of patients with osteoporosis.
There is a lack of population-based analysis of adult and elderly facial fractures despite the wide body of literature demonstrating an increased postmenopausal prevalence of hip and extremity fractures compared with that of older men, in addition to well-described racial differences. Our primary objective was to determine whether these findings carry over to facial fractures. Specifically, our study aimed to characterize the incidence of adult facial fractures by patient demographics and mechanism, focusing on whether differences in fracture risk are noted with race, sex, and advancing age.
The US Consumer Product Safety Commission’s database detailing visits to a sample of US emergency departments (EDs) was accessed for data collection in October and November 2015. The National Electronic Injury Surveillance System (NEISS) provides information from representative EDs regarding patient demographics, presenting symptoms, sites of injury, and other clinical characteristics for patients seeking medical attention for injuries from consumer products as well as recreational activities and sports. This resource has a population-based algorithm that also allows for calculation of an estimated national incidence for specific injuries, a strength that has been valuable in numerous epidemiologic analyses covering topics of interest to physicians practicing a variety of specialties.18-34
We searched the most recent 3-year block of data available (2012-2014) in an effort to evaluate trends most relevant to contemporary practice. Specifically, the NEISS was queried for adult facial injuries (>18 years). As there were too few NEISS entries for facial fractures among individuals older than 90 years to allow for a reliable estimate of national incidence, we restricted our search to those 18 to 89 years old because we were confident that this search strategy contained ample data to achieve our objectives and evaluate the impact of race, age, and sex on facial fracture risk. Data were analyzed in October and November 2015. This study uses a publicly available nationwide database accessible online and contains no patient identifiers. Because it is accessible to anyone, it is exempt from institutional review board approval; this study qualifies as exempt status per the nonhuman subject research protocol set by the institutional review board of Rutgers New Jersey Medical School.
Facial injuries were individually evaluated for diagnosis, which we classified into “fracture” vs “nonfracture,” and further demographically characterized (race, age, sex) (Table). Furthermore, we also evaluated every entry for mechanism of injury (ie, specific consumer product or recreational activity). Comparison was conducted among age groups and included decade-by-decade analysis (age 18-29 years, then beyond that each individual decade, eg, 30s, 40s). We also conducted comparative analysis on “younger” adults (<60 years) as well as those 60 years or older. For the purposes of this analysis, instances in which we comment on “fracture risk” refer to the proportion of facial injuries that were fractures (eg, as opposed to another diagnosis such as laceration, contusion).
Data were analyzed with SAS statistical software (SAS Institute Inc). Entries were adjusted for cluster variables and sample weights using a Consumer Product Safety Commission–provided PROC SURVEYMEANS coefficient of variation equation for SAS. The coefficient of variation equation was also used to calculate 95% CIs. Categorical data were compared using a χ2, with 2-tailed P values, and a set threshold for significance of P < .05.
There were 33 825 adult facial trauma NEISS entries, extrapolating to an estimated 1 401 196 (range, 1 136 048-1 666 344) emergency department (ED) visits from 2012 through 2014. Fractures accounted for 14.4% of entries, with 55.2% of patients identified as male and the remaining 44.8% as female. Whites accounted for 58.2% of facial injuries and 60.5% of fractures, while black patients accounted for 8.5% of facial injuries and 6.8% of facial fractures. Asians comprised 1.1% of visits for facial injury and 1.4% of facial fractures, while individuals identified as other races or unspecified comprised 32.2% of visits for facial injuries and 31.3% of facial fractures. In all, race was documented in 74.9% of patients.
Overall, fracture risk was significantly higher among individuals 60 years or older than in younger adults (14.6% vs 14.6%; difference of the mean [DOM], 0.2%; 95% CI, 0.1%-0.3%; P < .001) (Figure 1, Figure 2A), and differences were noted on organization by sex and race (Figure 1). With regard to sex, the proportion of facial injuries diagnosed as fractures modestly decreased with advancing age among men, while simultaneously increasing among women, particularly at age 60 years or older (Figure 1). Older women had a significantly greater risk of fracture relative to those younger than 60 years (15.0% to 12.5%; DOM, 2.5%; 95% CI, 2.4%-2.7%; P < .001), a comparison that was significant among whites and Asians (Figure 3). Black women had a significantly decreased risk of fracture in the older aged population (8.4% vs 9.1%; DOM, 0.7%; 95% CI, 0.2%-1.3%; P = .02).
As detailed in the previous subsection, facial fracture risk increased with advancing age in women and surpassed that of their older male counterparts (Figure 1). A greater proportion of facial injuries among younger men (<60 years) were fractures relative to the proportion in younger women (15.3% vs 12.5%; DOM, 2.5%; 95% CI, 2.8%- 3.1%; P < .001); older populations, however, showed an increased fracture predilection in women (14.0% vs 15.0%; DOM, 1.0; 95% CI, 0.8%-1.2%; P < .001).
On organization by race regardless of sex, blacks had the lowest facial fracture risk in both the younger (11.9% for blacks and 14.6% for others; DOM 2.7%; 95% CI, 2.5%-2.9%; P < .001) and elderly cohorts (9.4% for blacks vs 14.8% for others; DOM, 5.4%; 95% CI, 5.0%-5.9%; P < .001) , while whites had the greatest fracture risks (Figure 2B).
The most common mechanisms of injury leading to facial fractures are detailed in Figure 4. Notably, individuals younger than 60 years were more likely to sustain facial fractures as a result of participation in recreational activities, particularly men; greater than 40% of facial fractures in men younger than 60 years occurred as a result of bicycling or playing basketball, soccer, football, and baseball (Figure 4). In contrast, all of the top causes of injury in older women were related to falls involving structural housing elements and furniture, and only 2 of the top causes of facial fracture in older men were recreational activities.
Despite extensive study into fractures sustained elsewhere in the body, facial injury remains an underappreciated concern among older adults. Our findings may represent a valuable addition to previously reported literature. Bone deterioration is a multifactorial entity, having impacts on different body parts at different rates.2,3,5,10 Bone density and composition characteristics have been exhaustively characterized in terms of hip and vertebral body fracture incidence, and the characterization of these through measures such as DEXA scans provide an accurate measure of osteoporosis and bone health.8,9,11,15,35-39 This prior body of literature comprehensively details how declines in bone density and strength are particularly accelerated among elderly white women.2,4-7,9-11,39 These results are consistent with our findings, in which we noted an increased risk of facial fracture after traumatic facial injury in patients older than 60 years, with these findings significant and most pronounced among white women. Furthermore, this finding was observed to a more modest degree or not at all among other demographics (Figure 2, Figure 3).
Numerous analyses have demonstrated a decline in bone strength among aging men, and in our study, men were noted to have a statistically significant modest decline in facial fracture risk with advancing age (Figure 3). One possible reason for the higher facial fracture risk among men younger than 60 years compared with women in the same age group as well as older men stems from the reported mechanisms of injury. Recreational activities were noted to be responsible for most facial fractures in men younger than 60 years; the top 5 recreational activity–related causes of facial fracture alone comprised greater than 40% of these injuries (Figure 4). These findings suggest that physicians should take the time to review the importance of facial protection, as appropriate, in encounters with these younger male patients as a potential strategy for minimizing these injuries. High-energy trauma characteristic of injury sustained from athletic and recreational activities has been previously noted to promote a significant fracture risk to young adult males.31,40,41
In contrast to the significant decrease in facial fracture risk noted among men, women experienced a significantly increased fracture risk after facial trauma in the older cohort (Figure 1A, Figure 3). Although the reasons for this are speculative and beyond the scope of this analysis, these findings are consistent with postmenopausal acceleration in bone loss noted by myriad analyses.5,7-11,13-16,39 Notably, this trend reached statistical significance among white, Asian, and “other/unspecified” women; in contrast, a significant decreased fracture risk was noted on comparison of black women younger than 60 years and 60 years or older (Figure 3). Our findings relating to facial fractures are consistent with the relatively limited bone deterioration noted among elderly black women compared with other races.2-6,8,39
Racial differences in bone strength and fracture prevalence are well documented.2,5,10,15,39 In particular, white women have been demonstrated to have a large increase in hip fracture incidence with advancing age. Our data are consistent with this information, as we demonstrated that elderly white women had a greater risk of facial fracture after facial trauma compared with younger white women. The trends in hip fracture described herein have attributed this increased fracture risk to a high propensity for falls in combination with a weakening skeletal structure, particularly among elderly white women.2,5,10,39 Interestingly, our data supported the propensity for these populations to experience fractures from falls because a significant proportion of injury was caused by falls involving floors and other structural building elements, such as stairs (Figure 4).
There are several possible explanations for the decreased fracture rates noted among blacks, both in previous studies of other anatomic sites as well as the current analysis demonstrating a significantly lower facial fracture risk compared with individuals of other races. Black patients have been noted to have significantly higher bone density, architectural superiority at many locations throughout the body, and more modest rates of deterioration compared with individuals in other races.2,4-6,8,10,11,13,15,35,37,39 In general, facial fracture risk in this analysis correlates well with hip fracture incidence noted in prior studies, and thus may be an alternative reliable predictor of bone deterioration with advancing age. As the US population continues to age, fractures in elderly individuals may increasingly contribute toward national health care expenditures.1,39 In light of this information, our findings reinforce the importance of discussing fall prevention in a targeted manner during visits with health care professionals.
To our knowledge, this is the first analysis to evaluate national incidence and demographic comparisons relevant to adult facial fractures. There are several limitations, however, inherent to our study design. Importantly, the NEISS was originally designed as a resource detailing consumer product–related injury; over the past decade, it was expanded to include recreational activities. Unfortunately, several significant mechanisms of facial trauma are not covered by this resource; notable examples include facial fractures sustained from assaults, as well as those sustained from motor vehicle crashes. However, the mechanisms of injury encompassed by this resource included a considerable sample and likely represents the largest study to date examining demographic related trends in adult facial fracture. We examined 33 825 entries, extrapolating to an estimated 1.4 million ED visits over a 3-year period; however, another limitation deals with the facilities covered by the NEISS because this resource covers only EDs, and facial traumatic injuries not considered significant enough to merit an ED visit may have been evaluated in other venues, including outpatient clinics and urgent care centers. Findings from this analysis could also potentially be enhanced by additional clinical details not available from the NEISS resource, such as fracture management and outcomes. Regardless, intra-institutional analyses may not have allowed for the opportunity to gather enough patients and harbor enough power to detect statistical differences. Consequently, use of a population-based resource such as the NEISS does have value in this regard and also allows for external validity in our conclusions.
There is an increase in the risk of facial fracture among postmenopausal women who sustain facial injuries, with these results significant among whites and Asians. In contrast, a decreased predilection was noted on comparison of older and younger black women. Overall, Asians and whites had significantly greater risks of sustaining a fracture with facial injuries. Mechanism of injuries also varied significantly by age, race, and sex.
Corresponding Author: Peter F. Svider, MD, Department of Otolaryngology–Head and Neck Surgery, Wayne State University School of Medicine, 4201 St Antoine, 5E-UHC, Detroit, MI 48201 (firstname.lastname@example.org).
Accepted for Publication: April 3, 2016.
Published Online: July 14, 2016. doi:10.1001/jamafacial.2016.0714
Author Contributions: Dr Svider had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Hanba, Svider, Carron, Folbe, Eloy, Zuliani.
Acquisition, analysis, or interpretation of data: Hanba, Svider, Chen, Carron.
Drafting of the manuscript: Hanba, Svider, Chen.
Critical revision of the manuscript for important intellectual content: Hanba, Svider, Carron, Folbe, Eloy, Zuliani.
Statistical analysis: Hanba, Svider, Chen.
Administrative, technical, or material support: Svider, Carron.
Study supervision: Carron, Folbe, Eloy, Zuliani.
Conflict of Interest Disclosures: None reported.
Previous Presentation: This study was a poster presented at the AAFPRS at COSM Annual Meeting; March 19, 2016; Chicago, Illinois.