If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Given the early surge of COVID-19 in New Jersey (NJ), a statewide executive order (EO) to stay-at-home was instituted on March 22, 2020. We hypothesized that the EO would result in a decreased number of trauma admissions, length of stay, and resources utilized in trauma patients at NJ trauma centers.
In an institutional review board-approved, retrospective, multicenter study, trauma registries at three level one trauma centers in NJ were queried from March 22 to June 14 in 2020 and compared to the same timeframe the year prior. Epidemiological and clinical data were obtained including demographics, select preexisting conditions, mechanism of injury, injury severity score, resources utilized, and outcomes.
1859 trauma patients were evaluated during the EO versus 2201 the year prior. During the EO, trauma patients were less likely to be transferred from another hospital (21% versus 29% P < 0.05), more likely to have a penetrating mechanism (16% versus 12% P < 0.05), were equally likely to require a procedure (P = 0.44) and had similar injury severity score (5 [interquartile range [IQR] 1-9] versus 5 [IQR 1-9], P = 0.73). There was no significant difference in ventilator days (0 [IQR 0-1] versus 0 [IQR 0-2] P = 0.08), intensive care unit days (2 [IQR 0-4] versus 2 [IQR 0-3] P = 0.99), or length of stay (2 [IQR 1-5] versus 2 [IQR 1-6] P = 0.73). Patients were more likely to be sent home than to rehabilitation or long-term acute care hospital during the EO (81% versus 78%, P = 0.02).
The 2020 COVID-19 EO was associated with a significantly different epidemiology with a higher rate of penetrating injury during the EO, and similar volume of injured patients evaluated.
The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) was initially declared a global pandemic on March 11, 2020. Given the resource-limiting constraints imposed by this pandemic, surgeons were redeployed to care for COVID-19 patients.
Elective surgeries were canceled, and many routine medical appointments were postponed indefinitely. New Jersey (NJ), in close proximity to a large epicenter of disease in New York City, issued an executive order (EO) from March 22 until June 14 of 2020, mandating that only essential workers were permitted to leave their homes for work. The three American College of Surgeons and NJ Department of Health-verified level one university trauma centers led the regional COVID-19 response and coordinated the efforts of other facilities in the region. These include University Hospital in northern NJ, Robert Wood Johnson University Hospital in central NJ, and Cooper University Hospital in southern NJ (Fig. 1). Several studies have suggested that the epidemiology and resource utilization of the injured patient changed during the pandemic. Notably, some regions saw a trend toward increased rates of intentional injury and suicide, with reduced injury-related deaths and intensive care unit (ICU) utilization.
We aimed to evaluate the association between the EO in NJ and volume and clinical outcomes of the injured patient. We hypothesized that the EO would result in a decreased number of trauma admissions, decreased hospital length of stay (LOS), and decreased resource utilization in these patients at academic trauma centers in the state of NJ.
This was a retrospective, trauma registry-based study. The study included trauma admissions at the three American College of Surgeons and NJ Department of Health-verified level one university trauma centers: University Hospital in Newark, Robert Wood Johnson University Hospital in New Brunswick, and Cooper University Hospital in Camden. Data were collected from March 22 to June 14, 2020 and compared to data from the same period in 2019. The following data points were analyzed: demographics, clinical data, and outcome measures such as hospital LOS (HLOS), ICU LOS (ICULOS), and ventilator days.
The comparison of categorical variables between periods was carried out using Fisher's exact test or chi-square. Continuous variables between periods were compared using t-tests or Wilcoxon rank sum test. Single and multivariable linear regression analyses were used to analyze continuous outcome variables (i.e., HLOS, ICULOS, and ventilator days). Single and multivariable logistic regression analyses were used to analyze categorical outcome variables (i.e., blood product usage and discharge destination). A P-value <0.05 was used for all analyses to determine statistical significance. Test of Skewness (any value within the -1 to 1 range is considered normally distributed while data outside these limits are considered to be nonparametric). Analyses were carried out using SAS v9.4 (SAS Institute, Cary, NC).
The study was approved by the local Institutional Review Boards at University Hospital, Robert Wood Johnson University Hospital, and Cooper University Hospital with a waiver of informed consent due to the retrospective nature of the review.
Demographics and preexisting conditions
1860 patients were identified during the EO in 2020, and 2201 patients during the same period in 2019 across the three institutions. Two institutions showed ∼25% reduction in patients while one institution demonstrated a 1.8% increase in patients (Supplementary Table 1). There was no difference in age, although a trend toward older patients was noted during the EO. The percentage of patients who presented as a transfer from surrounding institutions was 21% during the EO, as compared with 29% in 2019 (P < 0.05). With regard to preexisting conditions, injured patients presenting during the EO were more likely to suffer from alcohol use disorder (P = 0.01), dementia (P = 0.01), and functional dependence (0.04) but there was no difference in those with cancer, psychiatric history, or substance use disorder (Supplementary Table 2).
During the EO, penetrating trauma was observed at a higher rate (16% versus 12% P ≤ 0.05). There was no change in rates of assaults, transport-related accidents, nontransport accidents, burns, falls, self-harm, or work-related traumas. There was no change in suspected abuse. There were significantly higher rates of gunshot wounds (29% versus 27%) and stab wounds (26% versus 17%) but a lower rate of assault with blunt instruments (43% versus 54%) (P < 0.05) (Supplementary Table 3).
Overall, patients who presented during the EO had a similar injury severity score (ISS) to those that presented the year before (5 [interquartile range [IQR] 2-9] versus 5 [IQR 1-9] P = 0.73). 87% of patients in both periods had an ISS less than 9. During the EO, patients had a comparable ISS, with 1.9% of patients having an ISS >25 while 1.6% had a similar injury the year prior (Supplementary Table 4). Patients who presented during the EO had reduced percentage of head/neck injury (Abbreviate Injury Scale 0 [IQR 0-2] versus 0 [IQR 0-2], P = 0.001) and facial injuries (0 [IQR 0-2] versus 0 [IQR 0-0], P = 0.046), but no changes in chest (0 [IQR0-0] versus 0 [IQR 0-0], P = 0.65), abdomen (0 IQR [0-0] versus 0 [IQR 0-0], P = 0.59), or pelvic injury (0 [IQR 0-2] versus 0 [IQR 0-2], P = 0.96) (Supplementary Table 5).
Mortality was similar during the EO (4.3% versus 3.4% P = 0.1). Univariate analysis demonstrated that the EO was not associated with a change in mortality (odds ratio 1.3, 95% confidence interval 0.949-1.8). There were similar HLOS (2 [IQR 1-5] versus 2 [IQR 1-6), P = 0.73), ICULOS (2 [IQR 0-4] versus 2 [IQR 0-3], P = 0.99), and ventilator days (0 [IQR 0-1] versus 0 [IQR 0-2], P = 0.08). A similar fraction of patients required tracheostomies (0.59% versus 0.68%, P = 0.71) and received blood products (11.6% versus 9.5%, P = 0.28). When evaluating discharge disposition, more patients were discharged home than to rehabilitation or long-term acute care hospital during the EO (81% versus 78%, P = 0.02) (Supplementary Table 6).
This study aimed to evaluate associations between the 2020 EO on locoregional trauma epidemiology and resource utilization in planning for future pandemics. NJ and its health infrastructure were gravely affected by the COVID-19 pandemic, resulting in a mandate to halt elective surgeries to circumvent shortages in ICU beds, ventilators, and extracorporeal membrane oxygenation circuits. Resident and attending physician staff were reassigned to care patients outside of their area of expertise. The EO, passed to reduce transmission of the disease and preserve resources, afforded the majority of the population to remain homebound. During this time of quarantine, we expected to observe reduced utilization of resources utilized to care for the injured patient.
In evaluating care of the injured patient across NJ, the most densely populated state, we discovered that volumes of trauma patients were lower during the EO in two of the three centers included in this study. The centers in northern and central NJ observed a 24-25% decrease in trauma volume, while the southern NJ center saw an increase of 1.8%. This finding suggests a trend toward decreased trauma volume during the EO, while highlighting specific locoregional variation throughout the state. This variation may be attributed to the fact that the northern and central regions of the state experienced a more concentrated burden of SARS-CoV-2 during the lockdown period (Fig. 2), and, therefore, perhaps increased adherence to the EO. Other contemporary studies demonstrate greater decreases in volumes of injured patients. One such study of trauma and emergency general surgery volume at a New Hampshire trauma center found a 57.4% decrease of trauma volume during that period as compared to a previous timeframe (P < 0.001).
The three centers in our study noted a significantly reduced number of trauma transfers during the EO (21% versus 29%, P < 0.01), a likely contributor to the reduced volume seen in these busy trauma centers who serve as a tertiary care facility for surrounding hospitals. A study at Cooper University Hospital specifically evaluating maxillofacial trauma during the COVID-19 pandemic attributed this finding in part to the increased outpatient management of many injuries as part of an overall attempt to conserve hospital resources needed for patients with SARS-CoV-2.
Our study identified an increase in the number of injured patients with histories of alcohol use disorder, dementia, and functional dependencies, without significant change in those with cancer, psychiatric comorbidity, or substance use disorder. These findings suggest that these patients with alcohol use disorder, dementia, and functional dependencies may have less support during times of isolation and therefore be more likely to suffer from trauma. Chiba et al. found an increase in patients presenting with substance use disorder and no change in patients with a history of alcohol use disorder, while we found the opposite.
Our study did not evaluate substance and alcohol positivity at presentation but focused on history in patients.
We noted an increased rate of penetrating injury rather than blunt injury, in concordance with several recent studies. Sheets et al. published a multicenter study of 88 hospitals and 169,892 patients, which demonstrated significantly more penetrating injuries and burn injuries compared with prepandemic levels, as did smaller studies performed in other areas of the United States.
We observed increases in both stab and gunshot wound injuries. Specifically, stab wound injuries increased to 26% of assaults compared with 17% of assaults previously. An increased rate of stab wounds was also seen by Chiba et al. and may be attributed to an increased proximity of individuals for prolonged periods of time during the EO.
Though some studies have found an increase in assault and domestic abuse, our study did not.
Like Chiba et al., we observed no major difference in motor vehicle collisions and agree with their conclusion that empty roads may have led to more aggressive and experimental driving habits. We also found no change in workplace-related injuries. This could be because essential workers and those who could not work from home during the pandemic were still working during the EO and remained vulnerable to injury in the workplace.
Interestingly, and in contrast to other studies, ISS was not found to be significantly different from prepandemic levels in our study (ISS (5 [IQR 1-9] versus 5 [IQR 1-9], P = 0.73). Chiba et al. evaluated 1202 injury-related admissions during the same time period in Los Angeles County, and found that patients suffered less severe trauma during the lockdown as well as less severe head and chest trauma.
A possible explanation for this finding could be the lower number of interhospital transfers and the possibility that less severely injured patients were triaged to stay at their presenting institutions or be discharged for follow-up with a subspecialist. This was possibly unique to the institutions studied in this paper as the NJ governor gave these three centers additional responsibility in triaging patients in their respective regions during the pandemic.
Similar to Chiba et al., we observed a lower rate of head and neck trauma, perhaps due to the increase in penetrating injury and decrease in blunt injury. We also observed a reduction in facial trauma, which may be explained by an increased number of patients being triaged to follow-up in the office instead of transferred to our centers.
HLOS and ICULOS were similar in the prepandemic and pandemic period. There was a trend toward more days on mechanical ventilation (3.18 d, 0 [IQR 0-2] versus 1.9 d, 0 [IQR 0-1]) although this did not achieve statistical significance (P = 0.08). The number of procedures performed, blood transfusions administered, and mortality rate were not statistically significantly different between the groups, suggesting overall similar resource utilization. Despite initial concerns about airborne and droplet spread of COVID-19, the rates of tracheostomies remained similar. A greater percentage of patients were discharged home (rather than to rehabilitation or long-term acute care hospitals) during the pandemic (80% versus 78%, P = 0.02). This finding is not unexpected given that these facilities faced staffing issues and lack of bed availability during this timeframe; in addition, patients were likely unwilling to pursue rehabilitation due to the increased spread of COVID-19 at rehabilitation and nursing facilities.
Our study had several notable limitations. As a registry-based study, it is retrospective in nature and only includes patient data recorded in the respective databases. Variables relating to preexisting conditions may not be accurate and preexisting conditions may be underreported. Variables such as “abuse” as listed in the trauma database lack specificity (domestic, pediatric, and elder) and may not illustrate a clear etiology between centers. Abuse is suggested by a clinician and therefore subjective and potentially underreported. Additionally, due to differences in reporting, the number of patients requiring blood transfusions, rather than the total quantity of blood products transfused, were obtainable. Further investigation is required to quantify the volume of blood transfused, as blood availably is strained during times of crisis due to fewer donors.
We chose to compare 2020 merely to the year prior, and further investigation is welcome to greater durations and larger number of previous years. This study only evaluated level one trauma centers in NJ; level two and three trauma centers were not evaluated. It is possible that these centers had varying experiences. Regardless, we believe this adds important insight into trends in the care of the injured patient, and resource utilization for said patients under exceptional circumstances such as a global pandemic. Evaluation of these trends will allow for improved resource utilization, planning, and policy creation in the event of similar crises.
During the COVID-19 EO, NJ observed an overall decrease in the volume of injured patients, with significant locoregional variation in volume across the state, associated with surges in COVID-19 infection itself. Increased rates of penetrating injury during the EO were noted, however, overall injury severity was similar, and resources utilized for injured patients were unchanged from prepandemic levels. Those with a history of alcohol use disorder, dementia, and functional dependence may be at higher risk of traumatic injury during lockdown. These findings suggest that additional legislation may be required for injury prevention, or dedicated resources must be maintained to care for injured patients, despite ongoing infectious pandemics.