Premorbid Incidence of Mental Health and Substance Abuse Disorders in Facial Trauma Patients

Abstract

Study Design: A retrospective study. Objective: Facial trauma is a prevalent cause of morbidity and mortality with increasing incidence over recent decades. Few studies have examined the prevalence of mental health and substance abuse disorders at the time of diagnosis. Herein we investigate the psychosocial demographics associated with facial trauma. Methods: The 2016 State Inpatient Database (SID) was used to identify patients with facial trauma from all hospitals in New York, Florida, and Maryland. A non-trauma control group undergoing elective same-day surgeries at ambulatory surgical centers in Florida, Kentucky, Nevada, North Carolina, New York, and Maryland was identified using the State Ambulatory Surgery and Services Database (SASD) from the Healthcare Cost and Utilization Project (HCUP). 777 patients were identified with facial trauma and compared to 500 patients without facial fractures. Results: Patients with facial fractures were statistically significantly more likely to have a substance abuse disorder (OR 34.78, p < .001) or mental health disorder (OR 2.75, p < .001) compared to controls. Patients with facial fractures were significantly more likely to be black than white (OR 4.80, p < .001). Patients with facial fractures were significantly more likely to have Medicaid compared to Medicare (OR 2.12, p = .005). Conclusions: Patients with facial fractures are more likely to have premorbid substance abuse and mental health disorders as compared to controls.

Introduction

Facial trauma is a significant cause of morbidity and mortality in the United States, with its incidence increasing since the early 2000s. There are over 400 000 annual cases with associated healthcare costs of more than one billion dollars.[1] While the development of psychosocial conditions such as major depressive disorder (MDD),[2,3] post-traumatic stress disorder (PTSD), and substance abuse following facial trauma has been well described,[4,5,6] the premorbid presence of such disorders prior to injury has yet to be elucidated in a national population-based study. Should such correlations exist, specific counseling and resources could be provided to appropriate individuals to help reduce the possibility of recurrent injury. Such a focus on whole person care would provide significant benefits for this patient population.[7]

In this study, we use population-based data to analyze the relationship between facial trauma and pre-existing psychosocial conditions, including mental health and substance abuse disorders. We hypothesize that patients undergoing surgery for facial trauma have a significantly higher incidence of comorbid psychosocial conditions than those undergoing elective procedures. We also secondarily characterize the sociodemographic breakdown of patients that suffered facial fractures.

Methods

Data Source

The 2016 Healthcare Cost and Utilization Project (HCUP) State Inpatient Databases (SID) and State Ambulatory Surgery Services Database (SASD) were the databases used in this study. The SID and SASD are comprised of encounter-level data from both hospital- and private-affiliated facilities, and capture patients that underwent any operation during the year examined. Patients with an International Classification of Diseases 10 (ICD-10) diagnosis code of facial trauma were identified in the SID databases of New York, Florida, and Maryland, the most populous states available. The SASD databases of New York, Florida, Maryland, North Carolina, Kentucky, and Nevada were used to identify patients undergoing elective surgery to serve as the control group.

Inclusion criteria for the study population consisted of age at least 18 years old and an ICD-10 diagnosis code of facial trauma. Patients that underwent surgery for facial fractures were included in the study group. No exclusion criteria were present for the study population. Given the database exclusively included surgical patients, the control group consisted of individuals undergoing elective surgery. ICD-10 codes for mental health disorders captured major depressive disorder, bipolar affective disorder, eating disorders, anxiety disorders, behavioral and emotional disorders, and generalized psychotic disorders. Substance abuse captured included alcohol abuse disorder, opioid abuse disorder, cocaine abuse disorder, hallucinogen abuse disorder, and sedative abuse disorder. For the secondary analysis, patients with facial fractures were split into a mandibular fracture and non-mandibular fracture (i.e., maxillary, nasal, orbital, and calvarial) group.

Variables

Baseline patient variables included age, self-reported race, insurance status, residential area (urban vs rural), and median household income (by quartiles). Insurance classes consisted of Medicare (a federal program providing health insurance for individuals over 65), Medicaid (a government program providing health insurance for individuals of limited income), private, and self-pay. Patients’ comorbidity status was measured by the Charlson Comorbidity Index (CCI) and presence of known mental health or substance abuse disorder at the time of fracture diagnosis was recorded.

Statistical Analysis

For bivariate analysis, categorical variables were tested with Pearson chi-squared testing and continuous variables were tested using Wilcoxon rank-sum testing. Multivariable logistic regression modeling was used to identify odds-ratios and 95% confidence intervals of mandibular and non-mandibular fractures compared to a control group. A two-sided alpha of 0.05 and 95% confidence intervals were used to assess for statistical significance. Statistical analysis was performed using STATA 15.1 (STATACorp, College Station, TX, USA).

Results

A total of 777 patients underwent surgery for facial trauma and were included in the study. 570 patients sustained non-mandibular fractures, while 207 patients sustained mandibular fractures. Population characteristics are shown in

Table 1. Facial Trauma Population Study Characteristics.

. A total of 500 control patients were used that did not have facial fractures. Facial trauma patients were more likely to be male (54.3%) compared to the control group (36.9%, P < .001). Additionally, trauma patients were less likely to be white (57.7%) compared to the control (81.0%, P < .001) and had significantly greater rates of medical comorbidities (P < .001). A greater proportion of trauma patients were insured by Medicaid (23.4% vs 9.2%, P < .001) and had lower incomes (P < .001). Trauma patients were more likely to be from large metropolitan centers with greater than 1 000 000 people (74.3% vs 68.2%, P < .001). Preexisting mental health (11.5% vs 4.0%) and substance abuse (9.5% vs 0.4%) disorders were both significantly more common (P < .001) in patients undergoing surgery for facial trauma than those undergoing elective surgery. Substance abuse disorder most commonly consisted of alcohol (n = 54), followed by opioid (n= 18), cocaine (n = 8), and sedative (n = 2) abuse.

Table 2. Multivariate Regression Model for Risk of Facial Fracture.

demonstrates a multivariate regression model identifying variables more likely to be present in patients with facial fractures. Patients with facial fractures were 50% more likely to be male than female (odds ratio (OR) 1.52, P = .002). Facial fracture patients were significantly more likely to be black (OR 4.80, P < .001) or Asian (OR 3.38, P = .007) compared to the white reference patients. Patients sustaining facial fractures were more than twice as likely to have Medicaid compared to Medicare (OR 2.12, P < .005).

Additionally, patients sustaining facial fractures were almost 35 times more likely to have substance abuse (OR 34.78, P < .001) and nearly three times as likely to have mental health disorders (OR 2.75, P < .001) compared to controls.

Table 3. Subanalysis of Mandibular vs Non-Mandibular Fracture Patients.

demonstrates a subanalysis comparing patients with non-mandibular vs mandibular fractures. Patients with non-mandibular fractures were found to have significantly lower rates of comorbidities, with greater than 50% of them having a CCI of 0 (P < .001). Non-mandibular fracture patients also had significantly higher incomes than patients with mandibular fractures (P < .001); 33.6% of non-mandibular fracture patients had an income greater than $71,000 compared to just 18.5% of mandibular fracture patients. These patients also had significantly lower rates of substance abuse disorder (6.7%) compared to patients with mandible fractures (17.4%) (P < .001).

Discussion

Premorbid correlations between psychosocial disorders and facial trauma have yet to be described in a population-based study. Herein, we identify a number of demographic and psychosocial variables that are more prevalent in individuals that sustain facial trauma. Our analysis demonstrates that comorbid substance abuse and mental health conditions independently increase the risk of facial fracture, and that patients with a facial trauma burden are more likely to have fit certain socioeconomic characteristics.

While this work is among the first to describe prevalence of substance abuse in facial trauma in a population-based investigation, other studies have investigated these possible correlations on a smaller scale. The association between facial trauma and alcohol use is well-described. Alcohol and drug use affect decision-making and can increase the risk of head injury.[8,9] In a nationwide survey of trauma managed by oral maxillofacial surgeons in the United Kingdom, 44% of patients over age 15 with facial fractures had consumed alcohol within the previous four hours.[10] Murphy et al[7,11] examined substance abuse and demonstrated that 58% of patients examined in the ED in a single center study met criteria for problem alcohol use and 24% met criteria for problem drug use. In fact, the prevalence may actually be underestimated. McAllister et al reported that less than a third of patients whose urine samples tested positive for illegal drugs endorsed drug usage.[12] Similarly, our study likely significantly underestimates the proportion of substance abuse disorders, as we only identified patients with a documented diagnosis at the time of presentation for assessment of fractures.

Mental health disorders were independent risk factors for facial trauma. Many previous authors have written about the higher rates of depression, anxiety, and PTSD following facial trauma.[4,13,14] Understandably, patients may be bothered by facial appearance, function, and the stress of reliving a traumatic event. However, few authors have examined mental health disorders as a risk factor for facial trauma. In a recent paper, Dugue et al reported that 26.9% of patients that suffered facial fractures had a premorbid behavioral health disorder. The authors found that patients with behavioral health disorders had higher rates of both intentional trauma (self-inflicted and assault) and violent injury. They theorized that these behavioral patterns place these patients at greater risk of facial trauma.[15] Islam et al[10,16] suggested a similar underlying association, but reported a lower incidence (5%) of premorbid psychological disorders. Notably, both papers included substance abuse disorders as part of behavioral or mental health disorder cohorts. In our study, substance abuse and mental health disorders are separated as risk factors, better informing care for these patients. We found patients presenting with facial fractures to have mental health disorders in 11.5% and substance abuse disorders in 9.5% of cases. The prevalence in our study was in between that in the work published by Dugue et al and Islam et al.

We found a variety of sociodemographic factors independently associated with facial trauma. There was a significantly higher rate of facial fracture in patients with lower income compared to our control group. In our subanalysis, we found patients with mandibular fractures had disproportionately lower incomes and higher rates of substance abuse than those with non-mandibular fractures. Other single institution studies have similarly found that mandible fractures were associated with a higher incidence of socioeconomic deprivation, while non-mandibular fractures were not.[17] We hypothesize that the mechanism of injury for mandible fractures is more often assault, which disproportionately affects socioeconomically disadvantaged patients.[18]

Actionable change from our findings would result from screening patients with facial trauma at the time of initial presentation. Early identification of pre-existing substance abuse and mental health disorders can facilitate inclusion of trauma psychiatry in the acute phase of an orofacial injury and improve outcomes.[19,20,21,22,23] Patients identified as high risk could obtain referral to inpatient or outpatient psychiatric or addiction services as appropriate. Additionally, given the high prevalence of postoperative PTSD and MDD following facial trauma, providing avenues for outpatient services that address these issues may prove beneficial for patients.[4,5,6]

This study must be viewed within the context of the study design and has several limitations. This is a retrospective study and causality cannot be determined from the associations presented above. Furthermore, this study draws from a national cohort that spans multiple states. While this provides validity and generalizability, it lacks the ability to identify granular details in the care and hospitalization for these cases. Details regarding fracture severity and mechanism of injury would allow for more direct comparison between cohorts. Patients who were not diagnosed with substance abuse at the time of fracture diagnosis, even if intoxicated at the time of presentation, were not considered as a part of the substance abuse cohort, as we wanted to identify the lower limit of patients with either a substance abuse or mental health disorder. Therefore, our substance abuse rate likely underestimates the true rate. Additionally, we identified a cohort of non-facial fracture patients from the SASD database to serve as a control group. These patients were deemed healthy for ambulatory surgery, had no facial trauma, and were able to undergo elective procedures; while having a healthy cohort in the general population would have been preferable, this was not possible given the limitations of the databases. Given the large population databases used for the study, granular comparisons between the control and study groups were not directly possible. There may have been a skew towards older age in the control cohort, as all patients had elected for surgery. Despite these limitations, this study provides a large sample size of patients in a national cohort that spans multiple states to examine these comorbidities in facial trauma patients. Such limitations could be mitigated and our findings strengthened by altering the study design to prospectively investigate rates of psychosocial disorders in patients presenting with facial fractures at a single center.

Future studies may focus on the impact of early identification of substance abuse and mental health disorders in patients with facial trauma and the ability to link patients with treatment options and therapy. Additionally, potential changes in substance use habits with early identification and treatment could be tracked. Additional work could determine whether patients who receive mental health and substance abuse resources at the time of initial presentation are less likely to incur a subsequent episode of facial trauma.

Conclusions

Patients undergoing surgery for facial trauma are significantly more likely to have a premorbid presence of substance abuse and mental health disorders. This information may be used to identify appropriate candidates for psychiatric resources both during initial hospitalizations and at the time of follow-up.