ED indicates emergency department. The time period is January 1, 2019 (week 1, 2019), to June 9, 2020 (week 23, 2020). The lockdown implementation occurred on March 23, 2020 (week 12, 2020). Trend lines show 6-week moving averages. To protect anonymity, weekly counts greater than 0 and less than 3 are not shown.
Customize your JAMA Network experience by selecting one or more topics from the list below.
Shepherd JP, Moore SC, Long A, Mercer Kollar LM, Sumner SA. Association Between COVID-19 Lockdown Measures and Emergency Department Visits for Violence-Related Injuries in Cardiff, Wales. JAMA. 2021;325(9):885–887. doi:10.1001/jama.2020.25511
Government policy responses to coronavirus disease 2019 (COVID-19), especially social distancing measures, have generated speculation on behavioral health effects, including interpersonal violence, domestic violence, and child abuse.1 We investigated the association between COVID-19 lockdown and emergency department (ED) visits for violence-related injuries in Cardiff, Wales, using detailed violence screening for all ED patients.2 We hypothesized that lockdown measures would decrease violence outside the home but would increase violence at home.
As one of a series of studies of violence trends, this study was approved by the Cardiff University Institutional Review Board and was deemed exempt from informed consent requirements.
Cardiff is served by a single ED at the University Hospital of Wales. Legally enforced social distancing restrictions were imposed across the UK beginning March 23, 2020 (week 12, 2020). Leaving home was allowed only for shopping for necessities, medical need, 1 exercise session per day, and travel to work if work from home was impossible. For all analyses, we examined ED attendances for violence-related injury, excluding self-injury, by week from January 1, 2019 (week 1, 2019), through June 9, 2020 (week 23, 2020). The prelockdown period was defined as week 1, 2019, through week 11, 2020 (a total of 63 weeks), and the postlockdown period as week 12, 2020, through week 23, 2020 (a total of 12 weeks).
A difference-in-differences regression model3 was used to evaluate for statistically significant changes before and after lockdown in ED visits for violence-related injury among 2 groups, patients injured outside the home and those injured at home. This was done through use of an interaction term between injury location (at home or outside the home) and time period (before lockdown or after lockdown). Time was included as a covariate, as was season (spring, summer, winter, and autumn). Statistical significance tests were 2-sided with a P < .05 cutoff.
The same difference-in-differences model was used to explore change in weekly mean number of violence-related ED visits after lockdown by age, sex, weapon involvement, and perpetrator type (acquaintance, family member, etc), with the threshold for significance Bonferroni adjusted to P < .0018 to account for multiple comparisons. Analyses were performed in Stata MP version 16.1 (StataCorp).
The mean number of total ED attendances per week decreased from 2889 (95% CI, 2847-2930) before lockdown to 1629 (95% CI, 1425-1834) after lockdown. The modeled mean number of violence-related ED attendances per week decreased from 28.4 (95% CI, 26.7-30.1) before lockdown to 16.5 (95% CI, 11.4-21.6) after lockdown. Mean weekly counts of injury at home were not significantly different before and after lockdown (before: 4.75 [95% CI, 4.10-5.40]; after: 6.00 [95% CI, 2.77-9.23]; difference, 1.19 [95% CI, −1.80 to 4.17]), but injury outside the home significantly declined (before: 22.62 [95% CI, 21.22-24.02]; after: 8.25 [95% CI, 5.68-10.82]; difference, −14.29 [95% CI, −17.27 to −11.31]) (Figure).
Analysis of characteristics of violent events before and after lockdown stratified by injury location (Table) revealed no significant changes among subgroups for injury at home. For injury outside the home, significant decreases were found in ED visits by female individuals younger than 18 years and by male individuals in all age groups, those injured with weapons, and those in which the perpetrator was a stranger, acquaintance, or security officer.
A significant decrease in ED visits in Cardiff by people injured by violence followed the UK lockdown, driven by a large reduction in ED visits due to violence outside the home. No significant increase in ED visits resulting from violence at home was noted.
Study limitations include lack of data on violent injury that did not result in ED treatment, limited data on injury severity (although a significant reduction in weapon use in violence outside the home after lockdown was found), and that the study took place in 1 city. Although the lockdown was uniform and UK-wide and Cardiff is similar to many UK cities,4 social and cultural factors not considered in this study together with different lockdown restrictions imposed in other countries may limit generalizability of these findings.5 It is possible but unlikely that reluctance to visit an ED because of fear of exposure to COVID-19 has materially affected estimates. Nonetheless, in Cardiff, these data suggest no associations between the UK lockdown and changes in violence at home, and large reductions in violence overall.
Corresponding Author: Jonathan P. Shepherd, PhD, Crime and Security Research Institute, Cardiff University, Friary House, Greyfriars Road, Cardiff CF10 3AE, Wales (email@example.com).
Accepted for Publication: December 8, 2020.
Author Contributions: Dr Moore had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Shepherd, Moore, Sumner.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Shepherd, Moore, Mercer Kollar, Sumner.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Moore.
Administrative, technical, or material support: Shepherd, Long, Mercer Kollar, Sumner.
Conflict of Interest Disclosures: None reported.
Disclaimer: The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the US Centers for Disease Control and Prevention.