Effect of the Coronavirus Disease 2019 Pandemic on Suicide-Related Trauma Burden at a Level 1 Trauma Center.

Introduction: In March 2020, the first coronavirus disease 2019 (COVID-19) cases were reported in New York and a stay-at-home order was enacted soon after. Social isolation combined with pandemic-related stressors profoundly affected mental health. We hypothesize that there was an increase in violent suicide attempt during the COVID-19 pandemic lockdown compared to previous years.
Methods: We queried our institutional trauma registry for total number of trauma activations and identified adult patients with International Classification of Diseases-10 diagnosis of intentional self-harm. We compared incidence during the lockdown to corresponding time periods from previous years. Demographic and injury characteristics were compared, as were outcomes such as mortality.
Results: We observe a significant uptrend in patients requiring trauma intervention after suicide attempts from July 2019 through July 2020 (r = 0.8, P < 0.001) despite a significant downtrend in trauma volume at our institution during the same period (r = ‒0.7, P = 0.003). Although not statistically significant, patients attempting violent suicide during lockdown were more likely to have preexisting psychiatric diagnoses, to live alone, to have injury severity score >9, and to require surgical intervention. Three COVID-period patients died in the emergency room compared to zero in the comparison group.
Conclusion: Our data show a rise in violent suicide attempts during the pandemic lockdown despite an overall decrease in trauma volume. The ramifications of a stay-at-home order seem to have the most profound impact on individuals with preexisting mental health disease. Early establishment of mental health outreach programs may mitigate the reverberating psychosocial consequences of a pandemic. Copyright:

INTRODUCTION

In January 2020, the first confirmed case of coronavirus disease 2019 (COVID-19) in the United States was reported in Washington State; within the following weeks, New York City emerged as an epicenter of the pandemic. On March 20, a New York State executive order was signed, outlining considerable restrictions on social gatherings as well as mandating closure of all nonessential businesses. The disease itself along with the multifactorial ramifications of new policies to curtail its spread had an abrupt and dramatic impact on people's daily lives. The economic repercussions of COVID-19 and the associated city-wide shutdown have been profound, resulting in major job loss and financial insecurity. Similar circumstances have historically been associated with an increase in suicidality.[123] Loneliness and social isolation in general, as well as in the context of infection prevention, have been established as important risk factors for mental health disorders including suicidal ideation and suicide attempts.[45] Furthermore, unclear transmission patterns and strain on public health care have led to a decrease in accessibility to both physical and mental health providers.

The World Health Organization has identified certain individuals who might be particularly vulnerable to a decline in mental health under the conditions of quarantine, including those with preexisting mental or physical health conditions and the elderly.[6] There has been a much-needed emphasis on identifying other at-risk groups, with a recent study finding young adults and women to have the greatest increase in rates of psychological distress during April 2020 compared to April 2018.[7] The negative social impact of a stay-at-home order is not limited to those living alone; an increase in domestic violence has been observed during this period, as well.[8] In totality, the social, economic, and health-related impact of the COVID-19 pandemic and the reactionary consequences have created a proverbial “perfect storm” environment for suicidality. While the mental health impact of the pandemic has been postulated, scarce data exist regarding the objective toll that COVID-19 has had on the population of the United States.

Queens County, New York, was affected profoundly by the outbreak.[9] At our level 1 trauma center located in Queens, New York, we noticed unusually high number of patients requiring trauma team intervention due to suicide attempts. We hypothesize that a high COVID-19 caseload is associated with increased incidence in violent suicide attempts compared with previous years. We set out to quantify the effect of psychosocial destabilization on rates of violent suicide attempts at our institution.

METHODS

Data source/selection of patients

We performed a 3-year (2018–2020) retrospective review of trauma patients presented to our institution with a diagnosis of intentional self-harm. We queried our institutional trauma registry to identify patients presenting with International Classification of Diseases-10 diagnosis of intentional self-harm (X71-X83) from January 2018 to July 2020. We excluded patients below the age of 15. We also excluded patients who were not injured by intentional self-harm via chart review. We collected patient data by reviewing electronic health records. We defined a violent suicide attempt as one requiring evaluation by the trauma team.

To reduce the effect of seasonality on suicidal behavior, we compared patients admitted for suicide attempt in the time period of March 1 to July 31 in 2020 to those during the same periods of 2018–2019, specifically studying demographic and injury characteristics. Demographic variables included age, gender, preexisting comorbidities, preexisting psychiatric diagnosis, social history, and previous suicide attempts. Clinical data collected included method of suicide attempt, emergency room disposition, operative intervention, injury severity score, hospital length of stay (LOS), intensive care unit LOS, and hospital disposition including mortality. Operative intervention included bedside surgical procedures such as chest tube placement, emergent thoracotomy, and complex wound closures.

Statistical methods

De-identified data were stored using Microsoft Excel (Microsoft, Redmond, WA, USA), and statistical analysis was performed using MedCalc for Windows, version 19.1.0 (MedCalc Software, Ostend, Belgium) for parametric and nonparametric tests. We reviewed the monthly incidence of violent suicide attempts compared to overall monthly trauma incidences in our institution from July 2019 to July 2020. Incidence of violent suicide was plotted against month of the year in chronological order. A best-fit trend line was generated using locally estimated scatterplot smoothing at 50% span. Pearson's correlation was performed to evaluate the trend. The Chi-square test was used to compare factors. Student's t-test was used if continuous variables were normally distributed; otherwise, the Mann–Whitney U-test was used. Normally distributed continuous variables are summarized as mean and standard deviation and nonnormally distributed continuous variables as median and interquartile range.

RESULTS

Incidence and trend of suicide attempt

Before the pandemic lockdown (July 2019 to February 2020), the monthly incidence of violent suicide attempts ranges from zero to two victims per month. During the pandemic lockdown (March 2020 to July 2020), the incidence ranges from one to five victims per month. Lockdown period has significantly higher incidence than prelock down period (median 3.5 vs. 1 incident/month, U = 3.50, P < 0.01). During the same time periods, the monthly trauma activations in our institution were lower (143 vs. 276, P < 0.001).

We observe a significant uptrend in the number of patients requiring trauma team activation after violent suicide attempts from July 2019 to July 2020 (r = 0.8, P < 0.001) and a significant downtrend in total number of trauma seen at the institution (r = ‒0.7, P = 0.003) [Figure 1].

Figure 1

Number of suicide victims and total trauma volume patients admitted monthly from July 2019 to 2020. Black triangles representing total trauma volume (×100) and circles representing number of patients presenting after traumatic suicide attempt by month

Injury characteristics and emergency department disposition

When adjusting for seasonality, during the months of March through July, we identified 5, 6, and 18 violent suicide attempts that required trauma team activation during the observed period of 2018, 2019, and 2020, respectively. The mean age of victims and gender distribution are similar between all groups. Although not statistically significant, patients admitted due to suicide attempt during the pandemic were more likely to have preexisting psychiatric diagnoses (67% vs. 28%, P = 0.12), to live alone (33% vs. 6%, P = 0.10), to have prior suicide attempts (22% vs. 18%, P = 0.71), to jump from height (50% vs. 11%, P = 0.09), to have injury severity score >12 (50% vs. 18%, P = 0.09), and to receive surgical intervention (73% vs. 46%, P = 0.16) [Table 1]. Three victims during the COVID period died in the emergency room compared to zero in the comparison group.

Table 1

Baseline Characteristics of the Study Population

Variables	Pre-Pandemic		Pandemic		P
Number of Patients	11		18
Gender (Male)	8	73%	14	78%	0.76
Age (Standard Deviation)	46	6.7	43	4.1	0.7
Age Group
< 45	7	6	11	78.1%	0.89
>= 45	4	3	7	21.9%
Trauma Activation Level
Level 1	7	64%	11	61%	0.96
Level 2	2	18%	3	17%
Consultation	4	36%	2	11%
Pre-existing Psychiatric Diagnosis	5	45%	12	67%	0.12
Lives Alone	1	9%	6	33%	0.1
Previous Suicide Attempt/Ideation	2	18%	4	22%	0.71
Modality of Trauma
Stab/Slash wound	8	73%	9	50%	0.13
Jumped from height	2	18%	9	50%
Hanging	1	9%	0	0%
Injury Severity Score >12	2	18%	9	50%	0.09
ED Disposition
General Floor	5	45%	7	39%	0.65
Step Down	3	27%	5	28%
ICU	2	18%	4	22%
Operating Room	3	27%	3	17%
Died	0	0%	3	17%

ED: Emergency department; ICU: Intensive care unit.

Hospital course and outcomes

Of the patients admitted to the hospital, there were more patients who required surgical intervention during the pandemic period (80% vs. 45%) [Table 2]. Of the victims who presented during the pandemic, no patient was detected to have COVID-19 infection on admission. Only one patient was found to be infected with the virus 2 months after discharge from the hospital. There was no significant difference in LOS and hospital disposition.

Table 2

Clinical Outcomes of Admitted Patients

Variables	Pre-Pandemic		Pandemic		P
Number of Patients	11		15
Requires Surgical Intervention	5	45%	12	80%	0.27
Surgical Exploration	3		3
Orthopedic Surgery	1		4
Neurosurgery Procedure	0		3
Chest tube placement	0		1
Wound Closure	1		1
Median Length of Stay (IQR)	5	3 to 9	7	3 to 22	0.60
Hospital Disposition
Home	2	18%	5	33%	0.72
Inpatient Psychiatric Facility	6	55%	8	53%
Skilled Nursing Facility	1	9%	1	7%
Died During Admission	2	18%	1	7%

IQR: Interquartile Range

DISCUSSION

Several articles projected an increase in suicidality during the COVID-19 pandemic and the associated lockdown period. Leading theories suggest that the COVID-19 pandemic created an environment of fear, uncertainty, social isolation, and economic instability.[1011121314] The dynamic interplay between these aforementioned stressors creates a self-perpetuating loop where those who are at the most risk, including individuals with mental health conditions and previous suicide attempt, are also the ones most profoundly impacted. While there are plenty of scholars predicting widespread destabilization of mental health, offering insight into its root cause, and identifying at-risk populations, there remains little to no empirical evidence in the current body of literature supporting an increase in suicidality at global on a national level. Despite a lack of national uptrends in suicides, certain enclaves may exist that are more significantly affected due to cultural predisposition, underlying mental health issues, or greater exposure to COVID-19.

National trends and international trends in death by suicide have not shown an increase during the 1st year of the pandemic.[151617] These population-level trends are in direct opposition to our experience. Our institution is located at the early epicenter of the pandemic in the United States which may have contributed to significantly higher emotional distress due to disproportionately high disease mortality, lack of understanding of disease transmission leading to mass hysteria, and severely prohibitive lockdown regulations leading to isolation and financial hardships. Due to gradual spread, improving disease understanding, and variable lockdown policies, levels of emotional distress caused by the pandemic may have varied greatly across the nation. Thus, a lack of national increase in suicidality may inaccurately reflect the destabilization of mental health at a local level in certain hard-hit enclaves of the population such as ours.

The pinnacle of destabilized mental health is reflected by suicidality. At our level 1 trauma center in Queens, New York, we noted an unusual increase in trauma activation secondary to suicide attempts by violent means during the COVID pandemic. A review of our trauma registry database was consistent with our hypothesis, demonstrating a statistically significant increase in violent suicide attempts during and immediately following the COVID-19 lockdown when compared to rates from previous years. During the pandemic, our trauma center experienced a roughly 50% reduction in overall trauma activations (143 vs. 276, P < 0.001) while seeing a 300% (n = 18 vs. n = 6 and n = 5) increase in violent suicide attempts compared to the same time frame in the previous 2 years. Our study only included patients attempting self-harm by violent means at one trauma center, representing a microcosm of the population experiencing COVID-19 pandemic-related psychological distress. When extrapolated to all self-harm attempts, the trend we observed is likely to reflect a more profound problem in the community at large.

We noted trends toward higher injury severity scores (50% vs. 18% had an international space station >12) in patients presenting after violent suicide attempts during the pandemic. In addition, we saw an increased rate of mortality in the emergency department (3 vs. 0). These trends, though not statistically significant, are suggestive of a heightened severity of psychological distress culminating in more serious suicide attempts. Studies have shown that objective measures of suicide intent and relative mental suffering are correlated with the lethality of the attempts.[181920] Thus, our findings may reflect a higher level of psychological distress and destabilization of mental health during the COVID-19 pandemic in our community, manifesting in more medically serious suicide attempts.

While not reaching statistical significance, a higher percentage of patients presenting after violent suicide attempts during the pandemic lived alone (33% vs. 9%), had a preexisting mental health condition (67% vs. 45%), or had previously attempted suicide (22% vs. 18%) when compared to those from the two preceding years. These attributes are frequently cited as predisposing factors for increased rates of suicidality.[1112] Identifying and targeting these at-risk subgroups of the general population for preventative intervention programs is of utmost importance.

Trauma surgery departments have an opportunity to play a role in suicide prevention through outreach to at-risk populations.[21] Trauma registries can be utilized to identify at-risk patients such as those who have a history of mental illness or previous suicide attempts. These patients can then be targeted for trauma community outreach with wellness phone calls and early intervention services. Our data show that at minimum, 67% of patients who attempted suicide during the pandemic could have been identified as vulnerable patients based on the history of suicide attempts (20%) or mental illness (67%). It should also be noted that patients attempting suicide by non-violent mechanism as well as those patients who were successful in their attempt were not captured in our cohort. More research is needed on the use of trauma registries for identifying and targeting at-risk populations.

Our study has several limitations. The retrospective single-center study design limits our study's external validity as our results are unique to our institution and may be limited in relevance to other hospitals, depending on various demographic, socioeconomic, and other factors. Our sample size was too small to power meaningful trends in our data, namely the increase in suicide attempt among individuals with preexisting mental health problems, previous suicide attempts, and individuals living alone. Our population was limited to patients with violent suicide attempts excluding a much large subset of patients treated for suicidality. These limitations present important avenues for future research to evaluate the incidence of all suicide attempts in our population as well as predisposing factors in this patient population.

CONCLUSION

Our study provides quantitative evidence of an increase in violent suicide attempts during the COVID-19 pandemic at our institution. Further studies are needed to elucidate specific population at risk for suicide due to the effects of COVID-19 pandemic from a geographic, ethnographic, and psychiatric perspective. Additional research will help further develop and tailor prevention measures targeting populations at greatest risk for mental health destabilization in the setting of a public health crisis.

Research quality and ethics statement

This study was approved by the Institutional Review Board/Ethics Committee (Theresa and Eugene M. Lang Center for Research and Education. IRB# 13030720). The authors followed applicable EQUATOR Network (http://www.equator-network.org/) guidelines during the conduct of this research project.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.