key: cord-0855050-x3zdsc0c
authors: Maleitzke, Tazio; Pumberger, Matthias; Gerlach, Undine A.; Herrmann, Carolin; Slagman, Anna; Henriksen, Louise S.; von Mauchenheim, Frederic; Hüttermann, Nils; Santos, Anabel N.; Fleckenstein, Florian N.; Rauch, Geraldine; Märdian, Sven; Perka, Carsten; Stöckle, Ulrich; Möckel, Martin; Lindner, Tobias; Winkler, Tobias
title: Impact of the COVID-19 shutdown on orthopedic trauma numbers and patterns in an academic Level I Trauma Center in Berlin, Germany
date: 2021-02-16
journal: PLoS One
DOI: 10.1371/journal.pone.0246956
sha: ffaa001e6b708bceb9bc8b606a9331279e2463f2
doc_id: 855050
cord_uid: x3zdsc0c

BACKGROUND: The COVID-19 pandemic led to the implementation of drastic shutdown measures worldwide. While quarantine, self-isolation and shutdown laws helped to effectively contain and control the spread of SARS-CoV-2, the impact of COVID-19 shutdowns on trauma care in emergency departments (EDs) remains elusive. METHODS: All ED patient records from the 35-day COVID-19 shutdown (SHUTDOWN) period were retrospectively compared to a calendar-matched control period in 2019 (CTRL) as well as to a pre (PRE)- and post (POST)-shutdown period in an academic Level I Trauma Center in Berlin, Germany. Total patient and orthopedic trauma cases and contacts as well as trauma causes and injury patterns were evaluated during respective periods regarding absolute numbers, incidence rate ratios (IRRs) and risk ratios (RRs). FINDINGS: Daily total patient cases (SHUTDOWN vs. CTRL, 106.94 vs. 167.54) and orthopedic trauma cases (SHUTDOWN vs. CTRL, 30.91 vs. 52.06) decreased during the SHUTDOWN compared to the CTRL period with IRRs of 0.64 and 0.59. While absolute numbers decreased for most trauma causes during the SHUTDOWN period, we observed increased incidence proportions of household injuries and bicycle accidents with RRs of 1.31 and 1.68 respectively. An RR of 2.41 was observed for injuries due to domestic violence. We further recorded increased incidence proportions of acute and regular substance abuse during the SHUTDOWN period with RRs of 1.63 and 3.22, respectively. CONCLUSIONS: While we observed a relevant decrease in total patient cases, relative proportions of specific trauma causes and injury patterns increased during the COVID-19 shutdown in Berlin, Germany. As government programs offered prompt financial aid during the pandemic to individuals and businesses, additional social support may be considered for vulnerable domestic environments.

a1111111111 a1111111111 a1111111111 a1111111111 a1111111111

Since first reports of respiratory tract infections due to a novel coronavirus in late 2019 the rapid spread of SARS-CoV-2 has led to a global health pandemic with vast impacts on society, economy and healthcare systems [1, 2] . As a result, governments implemented drastic shutdown measures to slow-down infection rates and contain the disease outbreak. While some communities are slowly returning to normality, increasing infection numbers in the United States of America and India during the summer and fall of 2020 are proof that the fight against COVID- 19 has not yet been won [3, 4] .

On March 16 th 2020, the German government and the federal states announced a temporary shutdown of major social institutions, including schools, universities, restaurants, bars, theaters, non-essential businesses and nightclubs. Social contacts were restricted, following a one-household-plus-one rule, and the shutdown was maintained for a little over a month. On April 20 th 2020, Berlin officials started gradually re-opening small businesses, restaurants, schools and other institutions with new regulations including the use of face masks in most indoor public spaces.

In combination with nationwide shutdowns, quarantine and isolation led to fundamental shifts in daily routines of individuals and businesses by an unforeseen magnitude. Comparable deprivations of social and material resources have previously only been observed after natural disasters. After Hurricane Katrina in 2005 [5] and the tsunami disaster in Sri Lanka in 2004 [6] , intimate partner violence increased significantly and was reported to prevail for up to a year after the events [7] . Reasons for the rise in violence included increased emotional stress, unemployment, reduced income and increased substance abuse [8, 9] .

Scientific evaluations of trauma and injury patterns in emergency departments (EDs) during the COVID-19 pandemic are still scarce and only slowly emerging [10] [11] [12] [13] , yet credible news outlets and United Nations reports have pointed out a rise in domestic violence [14, 15] . While potential psychological trauma due to the COVID-19 pandemic has recently been summarized [16] , data on physical trauma is urgently needed. Testing facilities at hospitals proved to be highly frequented by patients and healthcare staff during the SARS-CoV-2 pandemic [17] . Yet, due to fear of uncontrolled in-hospital exposure to the virus, subsequent self-treatment of minor injuries and limited practice of harmful behavior (e.g. sports, driving), overall ED admissions and trauma patterns may have been altered during the COVID-19 pandemic.

In this study we evaluated total and daily numbers of patient cases and contacts treated in an academic Level I Trauma Center ED in Berlin, Germany, during the 35-day COVID-19 shutdown (SHUTDOWN) and a calendar-matched control period in 2019 (CTRL). Trauma causes and injury patterns of orthopedic trauma patients during the SHUTDOWN were compared to the CTRL period. Additionally, the number of daily cases and contacts were compared to a shutdown-preceding 35-day transition (PRE) and shutdown-succeeding transition period (POST) in 2020.

All medical records of patients admitted to our central ED during the following periods were retrospectively evaluated: March 16 th until April 19 th 2019 and February 10 th until May 24 th 2020. We included all patients who sought medical care at the ED regardless of whether they were admitted, transferred or directly discharged following acute treatment. To compare periods with different social distancing restrictions, according to the German Infection Protection Act, we defined the following groups: iv. POST period: April 20 th -May 24 th 2020 (35 days) (Fig 1) .

Our Level I Trauma Center is one of five in Berlin, Germany, serving a community of approximately 3.7 million people. 63,394 patients were treated in our ED in 2019, and 18,038 of these were categorized as orthopedic trauma cases.

Absolute and daily average values were calculated for ED patient cases and contacts. Patient cases were defined as the number of patients who presented to the ED during aforementioned time periods. Patient contacts were defined as the number of medical specialties that were consulted per case. Example: A patient presenting with a distal radius fracture and shortness of breath was seen by both, an orthopedic trauma surgeon and an internal medicine physician, resulting in two patient contacts for this case. Ethical approval was obtained from the local hospital ethics committee (Ethikkommission der Charité -Universitätsmedizin Berlin: EA1/ 082/20). Trauma calls included trauma team activation for suspected multiple trauma, coma and aortic dissection. All orthopedic trauma patients self-identified as female or male, therefore no additional genders were reported.

Daily total cases, daily total contacts and daily orthopedic trauma cases and their standard deviations (SDs) were calculated. Daily numbers were plotted as a calendar time function for the CTRL (2019) as well as for the PRE, SHUTDOWN and POST (2020) periods (Fig 2A-2C ). Boxplots were created for the same periods (Fig 2D-2F ). Upper and lower whiskers represent the respective minimum and maximal values.

Average daily numbers of ED patient cases and contacts were compared between different time periods by means of incidence rate ratios (IRRs). Incidence rates were determined as the cumulative case numbers per period divided by the number of days in the corresponding time periods. IRRs and corresponding 95% confidence intervals (CIs) for comparison of different periods were determined by negative binomial regression analysis. For the explorative design of the study, the 95% CIs were not adjusted for multiple testing and p-values were not reported for that reason.

Furthermore, for orthopedic trauma cases we reviewed different pre-selected variables (S1 File) from ED charts and analyzed changes between the SHUTDOWN and CTRL period. We focused on changes of incidence proportions of the pre-selected variables. Incidence proportions were defined as cumulative numbers of cases of specific pre-selected variables including trauma causes and injury patterns divided by the cumulative number of orthopedic trauma cases within respective periods. To compare incidence proportions, we calculated risk ratios (RR) and related 95% CIs. Again, confidence levels were not adjusted for multiple testing due to the explorative design of the study. All statistical analyses were conducted in R version 4.0.0 and R Studio version 1.2.5001 [18] . RRs and 95% CIs were calculated using the "riskratio" function in the R package "epitools" [19] .

The total number of ED patient cases during the 35-day SHUTDOWN period was 3743 and 5864 during the calendar-matched CTRL period. The total number of patient contacts decreased from 6762 during the CTRL to 4255 during the SHUTDOWN period and a similar pattern was seen for daily orthopedic trauma cases during the CTRL (n = 1822) and SHUT-DOWN period (n = 1082). Daily numbers decreased for all three parameters during the PRE compared to the CTRL period and during the SHUTDOWN compared to the PRE period. Finally, an increase in daily numbers was observed during the POST compared to the SHUT-DOWN period, again for all three parameters. Total and daily numbers as well as IRRs are summarized in Fig 2 and Table 1 .

To assess potential changes in demographics, trauma causes and injury patterns of orthopedic trauma patients during the COVID-19 shutdown, the development of absolute numbers and incidence proportions during the 35-day SHUTDOWN were evaluated and compared to a calendar-matched 35-day CTRL period. For further relative assessments of data, RRs were calculated with respect to overall orthopedic trauma patient cases during respective periods. Absolute numbers, incidence proportions and RRs are displayed in detail in Tables 2 and 3 . 

Absolute numbers of female and male patients declined during the SHUTDOWN compared to the CTRL period without any difference in gender distribution on a relative scale. Trauma calls and trauma related deaths in the ED <24 hours and >24 hours showed no relative differences during the SHUTDOWN compared to the CTRL period.

Incidence No relevant changes were observed for incidence proportions of nightlife-related accidents, injuries resulting from self-harm, suicide attempts, assault-related injuries and robbery-related injuries comparing both periods. Increased incidence proportions were, however, observed for domestic violence-related injuries during the SHUTDOWN compared to the CTRL period with an RR of 2.41 (SHUTDOWN vs. CTRL, 95% CI: [1.22; 4 .74]) ( Fig 3B) .

Incidence proportions of overall traffic-related accidents remained almost constant in the two periods assessed. Similarly, incidence proportions of accidents involving motor vehicles, pedestrians and public transport did not differ between study periods. In contrast, incidence All RRs with a CI not including 1 are highlighted in bold. 95% CI = 95% confidence interval; CTRL = control; ED = emergency department; IRR = incidence rate ratio; PRE = pre shutdown transition; POST = post shutdown transition, SD = standard deviation; SHUTDOWN = shutdown. (Fig 3C) . orthopedic presentations, including low back pain, local infections, check-up visits and cases with an internal medicine focus and a subsequent referral to internal medicine, remained without relative differences during respective periods (Fig 4A) . (Fig 4B and 4C ). Numbers are summarized in Table 2 .

Overall fracture incidence proportions increased during the SHUTDOWN period, most pronouncedly for humerus fractures where the RR was 3.09 (SHUTDOWN vs. CTRL, 95% CI: [1.53; 6.21]) (Fig 5) . Incidence proportions and RRs of skull, facial, cervical spine, radius/ulna, hand, clavicle, thoracic spine, rib, pelvic/sacral, lumbar, femoral, tibia/fibula, patella, foot and Table 3 . open fractures remained similar during the SHUTDOWN and CTRL period. Absolute numbers, incidence proportions and RRs for fractures and patients with intracranial hemorrhages are shown in Fig 5 and Table 3 .

During the PRE period 402 patients were tested for COVID-19 in our ED. Of these, five test results were positive (1.2%). During the SHUTDOWN period 905 patients were tested for COVID-19 in the ED. Of these, 83 test results were positive (9.2%). During the POST period our ED tested 1042 patients for COVID-19. Of these, 14 test results were positive (1.3%). Of note, testing criteria changed between February and May 2020 and especially in the POST period also asymptomatic patients were tested before ward admission.

In the current study we describe the impact of the COVID-19 shutdown on patient numbers in an academic Level I Trauma Center ED in Berlin, Germany, by comparison to three other periods before (CTRL, PRE) and after (POST) the shutdown (SHUTDOWN). For the CTRL and SHUTDOWN periods, we analyzed trauma causes and injury patterns in detail.

Compared to the CTRL period in 2019, daily numbers of orthopedic trauma cases declined by nearly 15% during the PRE and by over 40% during the SHUTDOWN period. Correspondingly, an increase of 17% was observed during the POST compared to the SHUTDOWN period. The u-shaped development of ED and orthopedic trauma cases over time displays a gradual decrease and slow increase in patient numbers during the COVID-19 shutdown period. U-shaped recoveries were also described for other sectors, like aviation and trade during the COVID-19 pandemic [20, 21] . In case of further COVID-19 related shutdowns in the future, the longitudinal development of ED and orthopedic trauma cases could turn into a wshape [21] .

As previously reported, trauma admissions show seasonal differences with a mid-year peak and the "trauma season" spanning from April to November [22] . Although falls and motor vehicle accidents were reported to be more frequent in the winter season [23] , trauma numbers in March/April were comparable to those in February/March in ordinary years without a pandemic [24] . This allows the assumption that lower orthopedic trauma cases during the PRE compared to the CTRL period, may have been caused by people already being more cautious during the early months of the COVID-19 pandemic, although a shutdown was not yet in place.

Our findings go in accordance with a previous report from the Midland trauma registry in New Zealand [10] . The group found a 43% decrease of injury-related admission numbers during the national COVID-19 lockdown compared to a control period in 2019 in a Level I Trauma Center. In their study, no statistical analyses could be performed on injury patterns due to a low number of included patients (n = 195) [10] . In contrast, we compared 1082 orthopedic trauma cases during the SHUTDOWN in 2020 with 1822 cases during the CTRL period in 2019 regarding trauma causes and injury patterns. The observation of a relevant increase in domestic violence confirms concerns that were previously raised by other authors regarding intimate partner violence during the COVID-19 pandemic and concomitant shutdown environments [25] [26] [27] . A suspected high number of unreported cases of intimate partner violence further underlines the relevance of this finding [28] . Concrete social support plans are warranted during quarantine and shutdown scenarios to protect vulnerable individuals and families in potential future pandemics and related shutdowns [29] . Recent data from the United Kingdom indicate that inaccessibility to social support during the COVID-19 pandemic led to a reduction in well-being and increased anxiety in the elderly [30] . The importance of a timely response to mental health needs has been described for previous pandemics [31] and can also be seen in the increased workplace violence-related injuries observed during the SHUT-DOWN period in this study, which might have been caused by psychological stress due to fear of going into work as essential workers despite a global pandemic and a concomitant risk of infection. Effective psychological support can also be provided by non-psychiatric support groups, if mental health professionals are scarce and training is available [31] .

Our data disclose higher rates of acute and regular substance abuse in orthopedic trauma patients during the SHUTDOWN compared to the CTRL period. This is in accordance with observations from the Hubei province in China during the COVID-19 lockdown, where an increase in hazardous and harmful alcohol use during the COVID-19 lockdown was observed [32] . Similarly, data from the 2003 SARS epidemic in China pointed toward a risk of alcohol dependence and abuse during infectious disease outbreaks [33] .

We demonstrated that injuries resulting from self-harm and suicide attempts did not differ between the two periods assessed. This was expected as suicide numbers seem to rise only after a certain delay following catastrophic events [34] . Increased suicide rates after economic crises [35, 36] and natural disasters [34] have been well-documented, and a rise in suicide numbers in the months and years following the COVID-19 pandemic is anticipated [37] . Tele-counselling alongside 24/7 crisis response services for emotional, mental and behavioral support have been suggested and implemented as tools to tackle COVID-19-related suicide intentions [38, 39] . Further, continuous and transparent communication between healthcare officials, governments and society may decrease anxiety and create sustainable information structures in times shaped by uncertainty and fear [40, 41] .

We observed that not only the total number of ED cases decreased, but also the number of ED visits, that were likely unrelated to the COVID-19 shutdown itself, including low back pain and unspecific pain. This finding may be explained by a tendency to seek less medical aid during pandemic-related shutdowns as recently demonstrated in a retrospective analysis from 15 Italian cardiovascular centers [42] . In their study, the authors described a reduction in acute coronary syndrome-related hospital admissions during the COVID-19 lockdown in Italy. They observed markedly elevated mortality rates not fully explained by SARS-CoV-2 infections during the same period. The group hypothesized that a relevant number of deaths from acute coronary syndrome may have occurred unnoticed during the COVID-19 lockdown in Italy, as patients did not seek medical assistance [42] . In this study we found an absolute and relative decrease in the number of patients presenting with household-related injuries during the SHUTDOWN period. As people spent more time at home during shutdowns, it is unlikely that the number of household-related injuries decreased. Yet, if people with life-threatening conditions like acute coronary syndrome were less likely to seek medical care when experiencing symptoms during the COVID-19 pandemic, we speculate that this could also be the case for patients with less severe conditions. Whether these patients were in fact not in need of medical help provided by ED services or will require medical support at a later stage (e.g. for infections, missed fractures) remains speculative. This finding however raises questions about the potential prevention of ED visits due to non-urgent and chronic conditions and could help to educate people on alternatives to ED care in those cases [43] . Data from a tertiary trauma center in Spain reported no differences in the number of osteoporotic hip fractures between the period of March 14 th to April 2 nd 2020 (n = 36) and a control period in 2019 (n = 43) [11] . In opposition to this, two trauma institutions in Italy observed a relevant decrease in femoral fracture numbers during a two-month period from February 22 nd to April 18 th 2020 (n = 121) compared with the previous year (n = 169) [44] . Our data showed an absolute but not a relative decrease in femoral fractures during the SHUTDOWN (n = 17) compared to the CTRL (n = 31) period. Reasons for such a decline in fractures, which are mainly sustained at home or in nursing homes, remain hypothetical and although our hospital is considered a central treatment facility for COVID-19 patients, a shift in patient distribution with femoral fractures to other hospitals by emergency and first responder services was not confirmed by the services. Social distancing and contact restrictions were also implemented in nursing homes, which led to reduced community activities and by that a potentially reduced risk of falling.

Although the number of overall fractures decreased, we observed an absolute and relative increase of humerus fractures. Data from Italy showed an overall decrease of 65% for shoulder and elbow trauma during March and April 2020 compared to the same period in 2019, yet the prevalence of proximal humerus fractures did not differ between the two periods. The vast majority of cases was caused by an accidental fall at home [45] . As people spent more time at home during the COVID-19 shutdown, this may explain why the number of this specific injury was found to be increased in our study.

Our data show that with an overall decline in orthopedic trauma patients, the absolute number of patients requiring surgical care decreased as well. These findings are particularly relevant for resource planning and logistics during potential future pandemics and shutdowns. In our academic Level I Trauma Center ED in Berlin, Germany, the reallocation of orthopedic trauma healthcare providers and operating theatre staff to high maintenance COVID-19 wards was largely discussed and well-considered prior to the start of the pandemic in Berlin, Germany. According to our data, these strategies proved useful and justifiable with regards to the numbers of orthopedic trauma cases.

Limitations of this study are its retrospective and monocentric character as well as the lack of adjustments for potential confounding factors. While we were able to analyze a great number of orthopedic trauma cases in one of Europe's largest university hospitals, differences in absolute numbers and incidence proportions should always be compared to results from other large trauma institutions to validate the findings. We hope these data will aid future national and international pandemic plans regarding ED service assessments and the distribution of supporting social services.

Supporting information S1 File. List of variables for orthopedic trauma patients. (DOCX) S1 Data. (XLSX) 

World Health Organization declares global emergency: A review of the 2019 novel coronavirus (COVID-19)

A pneumonia outbreak associated with a new coronavirus of probable bat origin

Beware of the second wave of COVID-19

An interactive web-based dashboard to track COVID-19 in real time

Intimate partner violence and Hurricane Katrina: predictors and associated mental health outcomes

Violence against women and natural disasters: findings from post-tsunami Sri Lanka

Violence against women in disasters: A study of domestic violence programs in the United States and Canada

Investigating the Increase in Domestic Violence Post Disaster: An Australian Case Study

Natural disasters and social order: Modeling crime outcomes in Florida

Variation in volumes and characteristics of trauma patients admitted to a level one trauma centre during national level 4 lockdown for COVID-19 in New Zealand

Impact of the COVID-19 Pandemic on an Emergency Traumatology Service: Experience at a Tertiary Trauma Centre in Spain

Where Have All the Fractures Gone? The Epidemiology of Pediatric Fractures During the COVID-19 Pandemic

Medical emergencies during the COVID-19 pandemic-an analysis of emergency department data in Germany

Domestic abuse surge in coronavirus lockdown could have lasting impact, MPs say. The Guardian

UN chief calls for domestic violence 'ceasefire' amid 'horrifying global surge

The psychological impact of quarantine and how to reduce it: rapid review of the evidence

COVID-19: the case for health-care worker screening to prevent hospital transmission

R: A language and environment for statistical computing. R Foundation for Statistical Computing

epitools: Epidemiology Tools. R package version 0.5-10

The future of airport post COVID-19

Simulating the trade effects of the COVID-19 pandemic: Scenario analysis based on quantitative trade modelling

Identifying temporal patterns in trauma admissions: Informing resource allocation

Seasonal Variation of Trauma in Western Massachusetts: Fact or Folklore? Trauma Surg Acute Care Open

Relationship between weather and seasonal factors and trauma admission volume at a Level I trauma center

An increasing risk of family violence during the Covid-19 pandemic: Strengthening community collaborations to save lives

Pandemics and violence against women and children

COVID-19: Reducing the risk of infection might increase the risk of intimate partner violence

Missed opportunities: emergency department visits by police-identified victims of intimate partner violence

The pandemic paradox: The consequences of COVID-19 on domestic violence

A UK survey of COVID-19 related social support closures and their effects on older people, people with dementia, and carers

Mental health interventions and supports during COVID-19 and other medical pandemics: A rapid systematic review of the evidence

Epidemic of COVID-19 in China and associated Psychological Problems

Alcohol abuse/dependence symptoms among hospital employees exposed to a SARS outbreak

Suicide after natural disasters

Modelling suicide and unemployment: a longitudinal analysis covering 63 countries

Impact of 2008 global economic crisis on suicide: time trend study in 54 countries

COVID-19, unemployment, and suicide

COVID 2019-suicides: A global psychological pandemic

Psychological support in times of COVID-19: the Essen community-based CoPE concept

Effective health communication-a key factor in fighting the COVID-19 pandemic

COVID-19: fighting panic with information

Reduced Rate of Hospital Admissions for ACS during Covid-19 Outbreak in Northern Italy

Emergency department visits for nonurgent conditions: systematic literature review

Proximal femur fractures in COVID-19 emergency: the experience of two Orthopedics and Traumatology Departments in the first eight weeks of the Italian epidemic

The impact of COVID-19 on shoulder and elbow trauma: an Italian survey

Authors T.M. and F. N. F. are fellows of the BIH Charité Clinician Scientist Program. The program was initiated and led by Prof. Dr. Duska Dragun to enable resident physicians to pursue careers in academic medicine and research. With great sadness we have received the news that Prof. Dr. Dragun passed away on December 28th of 2020. This publication is dedicated to her as a mentor, friend, role model and stellar scientist.