Firearm Violence Surrounding the COVID-19 Pandemic: A Reopening Phenomenon

Introduction Past research has demonstrated a “reopening phenomenon” of increased firearm violence associated with the initial lifting of coronavirus disease 2019 (COVID-19) pandemic-related restrictions after the first wave. Now, with widespread societal reemergence from stay-at-home measures, we hypothesize another spike in firearm violence in the United States (US). Thus, the purpose of this study was to evaluate the trends in firearm violence before and after extensive community reopenings during the COVID-19 pandemic. Methods The Gun Violence Archive was utilized to collect data on daily firearm violence incidents, injuries, and deaths as well as on types of firearm violence. Mann–Whitney U-tests were performed for trends and types of firearm violence “before” (12/14/20-4/9/21) versus “after” (4/10/21-7/31/21) widespread societal reopening in the US. Additional analyses also sought to compare the after reopening time-period to historical data (2017-2020) of similar calendar dates, to better control for possible annual/seasonal variation. Results Median daily firearm violence incidents (153 versus 176, P < 0.001), injuries (89 versus 121, P < 0.001) and deaths (54 versus 58, P < 0.001) increased from before versus after reopening. Compared to all historical years, in the after reopening time-period there were consistent increases in total as well as mass shooting incidents/injuries/deaths (all P < 0.05). Conclusions Firearm violence incidents, injuries, and deaths increased after societal reemergence from the COVID-19 pandemic. In addition, there has been an increase in mass shootings despite a relative lull initially brought on by the pandemic. This suggests the “reopening phenomenon” has worsened an already substantial national firearm epidemic.


Introduction
The fears and uncertainty surrounding the emergence of the coronavirus disease 2019 (COVID- 19) were associated with a significant spike in firearm sales in the United States (US). [1][2][3] In the setting of ongoing socioeconomic upheaval brought on by the pandemic, many firearm purchasers obtained a weapon with the intention to protect themselves and their families. 4,5 Among these purchasers, a large proportion were first-time gun owners, 6 and approximately 40% of individuals admitted to storing a firearm unlocked. 4 In addition, pandemic-related stressors coupled with the use of alcohol 7 and/or drugs 8 were associated with increased firearm violence in the US in the wake of COVID-19. 1,[9][10][11] In a previous study published by our study group, a phenomenon of increased firearm violence associated with phased reopening was noted in California, Ohio, and the US as a whole. 12 Other studies have evidenced increased firearm violence incidents, 6 injuries, and deaths involving not only adults 12 but also children 13 associated with the COVID-19 pandemic. Although the intent of firearm purchase was for self-defense, the early COVID-19 era unfortunately saw an increase in accidental shooting deaths, with an increase in fatal and nonfatal child involved shootings as well as children killed by an adult with a firearm. 13 Conversely, there has been a decrease in defensive use, drug-involved, and home invasion shooting incidents. 12 However, there remains a paucity of literature regarding trends in firearm violence after the widespread societal reopening that occurred in 2021. 14 Understanding trends during this time-period is vital to the overall understanding of firearm violence-related trends before, during and after the pandemic. This information will hopefully help target future local interventions and/or legislation. In particular, the lifting of pandemic-related restrictions lends more opportunity to mass shootings and this is an area with a paucity of research in relation to the pandemic. Thus, the purpose of this study was to evaluate the trends in firearm violence before and after extensive community reopenings during the COVID-19 pandemic as well as compared to historical data for the years 2017 to 2020 to control for annual/seasonal variation. As a result of the increased number of Americans who now own a firearm paired with heightened emotions and potentially increased aggression related to traumatic pandemic experiences, 15 we hypothesize an increase in firearm violence following societal reemergence after COVID-19.

Materials and Methods
This study was deemed exempt by the University of California, Irvine Institutional Review Board. The manuscript was prepared using the guidelines outlined in the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement.
Retrospective firearm violence data (12/14/2020 to 7/31/ 2021) were obtained from the Gun Violence Archive (GVA). The GVA is an independent organization that seeks to publish accessible, comprehensive, national data on firearm violence and includes deidentified data only, and thus consent was not required for this study. This organization employs automated Internet queries to search and collect data from over 7500 sources, including police reports, media outlets, and more. 16 Data on total incidents, deaths, and injuries related to firearm violence, as well as on specific types of firearm violence, such as child involved shootings, home invasions, defensive use shootings, mass shootings, and officer involved shootings, were collected. Child involved shootings included children killed or injured by a child as well as children killed or injured by an adult regardless of accidental or intentional actions. Mass shootings were defined by the GVA and the Federal Bureau of Investigation as 4 persons shot or killed in a single incident, not including the shooter. Officer involved shootings were defined as any firearm violence during which a police officer was present. 16 To evaluate trends in firearm violence after societal reemergence following COVID-19, "before community reopening" was defined as 12/14/2020 to 4/9/2021 (117 d) and "after community reopening" was 4/10/2021 to 7/31/2021 (113 d). The date April 9, 2021 was selected as the cutoff for these time periods because according to the Centers for Disease Control and Prevention's COVID Data Tracker, this is the date when the US reached its daily vaccination peak with almost 4.5 million vaccines administered. 17 This date also closely coincided with many cities and states announcing plans to reopen given the decreasing COVID-19 burden. 18-21 The start date of 12/14/2020 was selected as this was the first date entered into the COVID Data Tracker, marking the beginning of vaccination rollout. 17 However, community reopening did not immediately follow these vaccination efforts given that COVID-19 cases in the winter months were surging. 22 Furthermore, to control for possible seasonal variation, GVA data for the dates 4/10 to 7/31 from the years 2017, 2018, 2019 and 2020 were compared to the "after community reopening" time-period.
The ManneWhitney U-test was performed to compare median daily firearm violence incidents, deaths, and injuries as well as types of firearm violence (i.e., child involved shootings, home invasions, defensive use shootings, mass shootings, and officer involved shootings) before versus after widespread community reopenings in the US. Similar statistics were run to compare the after reopening time-period with historical control data. A scatterplot was created to demonstrate the trends in firearm violence incidents and deaths from 12/14/2020 to 7/31/2021. Statistics were performed on IBM SPSS Statistics, Version 26 (IBM Corp., Armonk, NY). Statistical significance was set as P < 0.05.
Similarly, median daily child involved shooting incidents (before: two incidents versus after: three incidents, P < 0.001) and median daily child involved shooting injuries (before: two injuries versus after: three injuries, P < 0.001) increased after widespread community reopenings. With regards to trends in totals, there were increased total child-involved shooting incidents (before: 284 incidents versus after: 376 incidents) and child involved shooting injuries (before: 291 injuries versus after: 472 injuries) after reopening.
Contrarily, median daily home invasion incidents (before: three incidents versus after: two incidents, P ¼ 0.020) and deaths (before: one death versus after: 0 deaths, P ¼ 0.037) decreased after reopening compared to before reopening. Similarly, total home invasion incidents (before: 369 incidents versus after: 292 incidents) and home invasion deaths (before: 149 deaths versus after: 104 deaths) decreased after societal reopening.
The period after community reopening also had an increase in median daily mass shooting incidents (before: one incident versus after: two incidents, P < 0.001), mass shooting injuries (before: four versus after: 9, P < 0.001), and mass shooting deaths (before: 0 deaths versus two deaths, P ¼ 0.003). Moreover, total mass shooting incidents (before: 146 incidents versus after: 268 incidents), mass shooting injuries (before: 564 injuries versus after: 1150 injuries), and mass shooting deaths (before: 186 deaths versus after: 265 deaths) increased after reopening.
There were no changes in median daily defensive use and officer involved shooting incidents, injuries, and deaths before versus after widespread community reopening (all P > 0.05) ( Table 1).
After widespread community reopening compared to pandemic time-period historical data (2020 versus 2021) Median daily total firearm violence deaths (2020 historical: 55 deaths versus after: 58, P ¼ 0.033) increased significantly after    widespread community reopening compared to the 2020 pandemic time period control data. Moreover, median daily mass shooting deaths also increased during this time (2020 historical: one death versus after: two deaths, P ¼ 0.021) ( Table 5).

Discussion
Pandemic-related restrictions have been instrumental in curtailing the spread and severity of COVID-19. 23 Thus, given the stabilization of COVID-19 burden, the US has been enabled to resume daily societal activities. 24 However, this retrospective database study demonstrates increased overall firearm violence incidents, injuries, and deaths after widespread community reopenings in the US as well increases in the number of daily mass shooting incidents, injuries, and deaths. There was no difference in defensive firearm use and a decrease in home invasion incidents and deaths in the period after reopening. In all, these spikes in firearm violence following societal reopening point to a "reopening phenomenon" of significantly increased firearm-related incidents, injuries, and deaths associated with societal re-emergence. Firearm violence has continued to be a healthcare epidemic with around 40,000 firearm deaths annually in the US, 25 which is more than 11 times greater than other high-income countries. 26 Startlingly, the authors' current study suggests even further worsening of the increased firearm violence seen during the early phases of the pandemic 6,27-31 including domestic firearm violence 32 and pediatric firearm violence. 13 Interestingly, as described in the authors' previous study, the spikes in firearm violence were not seen with defensive shootings such as home invasion and defensive use. This is despite many Americans obtaining firearms for the purported reason of protection against socioeconomic uncertainties and fears of rising crime. 12 As such, given the decrease in COVID-19 related deaths, especially in urban cities, 33 a focus of resources must be given to curtail this public health crisis. On a community level, these efforts may include increased education on safe storage of firearms, 34,35 more resources dedicated to protecting adult and pediatric victims of domestic violence, 34 and improving access to mental health services, particularly for individuals who misuse alcohol 36 or other illicit substances. 37 On the state or national level, implementation of stronger gun laws may also help reduce the number of pediatric shooting incidents. 13 Mass shootings are an unfortunate "American problem". While Americans comprise approximately 4.4% of the global population, over 31% of the perpetrators of mass shootings historically have been American. 38 This retrospective study demonstrated an increase in mass shooting incidents, injuries, and deaths in the US following a relative lull brought on by the pandemic. This return of mass shootings has been predicted dating back to the beginning of COVID-19. 39 With increased congregation again at schools, places of worship, bars, restaurants, and concerts, there are more opportunities for this form of mass violence. 36 In addition, increased firearm purchases 6 as well as social, 40 economic, 41 and psychological stressors 42 brought on by the pandemic may be linked to this resurgence in mass shootings. 43 However, previous research has described potential interventions that may be effective at decreasing the incidence of mass shootings. 44,45 Examples include: preventing known perpetrators of domestic violence from purchasing a firearm, 44 utilizing extreme risk protection orders to reduce firearm access when the imminent risk of firearm violence is deemed high, 45 and increasing funding of  interdisciplinary research on mass shootings that will continue to inform future policy. As a retrospective database study, there are certain limitations inherent to the study design including classification/ coding errors and missing information. Additionally, as the GVA uses automated internet queries to search police and media reports, there may be an underrepresentation of parts of the country that receive less coverage. However, the GVA is nationally recognized and has been utilized for research previously. 13,46 Because there is no consensus definition of what constitutes a "mass shooting", 47 this study used the definition provided by the GVA, which aligns exactly with the definition provided by the Federal Bureau of Investigation. 48 Moreover, although these findings suggest that stronger gun laws may be appropriate to address the "reopening phenomenon" of increased firearm violence following widespread community reopening, the authors acknowledge that there are many other risk factors that must be addressed in order to curtail the surge of daily firearm injuries and deaths. Additionally, as the database does not allow for a breakdown of homicide, suicide, and accidental shootings, it is difficult to identify targeted public policy interventions at this time and a broader approach may be needed first to curtail the spike in firearm violence until more is known specifically about the types of shootings. And, finally, although the cutoff point for "before" versus "after" widespread community reopening was set based off vaccination data as well as reopening decisions from multiple major cities, the inability to set a true cutoff date that is applicable nationally is a significant limitation of this study.

Conclusions
This retrospective analysis demonstrates a worsening "reopening phenomenon" following widespread lifting of pandemic-related restrictions in the setting of increased firearm purchases and overall socioeconomic chaos brought on by the start of the COVID-19 pandemic. This study also provides evidence of an increase in mass shootings in the aftermath of the pandemic. As COVID-19 related deaths have begun to stabilize, increased efforts to curtail a worsening firearm related epidemic in the US are immediately needed.

Author Contributions
M.R.D. and J.N. were involved in conception and design, analysis and interpretation of the data, drafting the article, critically revising the article and providing final approval. C.K., B.S., L.S., C.D., A.G. and S.S. were involved in the design, interpretation of the data, critically revising the article and providing final approval.