Investigators hypothesized that the COVID-19 'Stay At Home" order contributed to an
increase in the incidence of violent traumatic injuries, particularly in areas with lower
socioeconomic status in Kern County, as measured by census data, which was previously
shown to be true in other areas of America. Investigators conducted a retrospective
analysis of injuries sustained before and during the pandemic and further explored the
median income of the area of injury to determine if socioeconomic status was a
determinant of violent crime incidence rates. This study aims to better understand the
causality of the increase in crime, further analyze the results, and identify trends in
future public health crises.

This retrospective cross-sectional minimal risk study utilized the Kern Medical's trauma
registry to identify patients in Kern County who presented to the emergency department
with a gunshot wound (GSW), stab wound, or assault between March 2019 and February 2021.
All reported blunt injuries, motor vehicle incident-related injuries, suicide attempts,
or unintentional injuries were excluded. Each case was categorized as pre-COVID (March
2019 - February 2020) or COVID (March 2020 - February 2021) and further subclassified
based on the incident zip code. Kern County zip codes were combined into urban regions as
NW (zip codes:93312, 93314), NE (93301, 93305, 93306, 93308), SW (93311, 93313), and SE
(93304, 93307, 93309, 93241), with a separate category defined as rural (outside
Bakersfield city limits). These regions were also defined by median income as lower
income (≤$50k: NW and SW) and higher income (>$50k: NE, SE, Rural) based on California
census data. This study was approved by the Institutional Review board of Kern Medical
Hospital.

Tabulated data were analyzed using odds ratios, chi-square tests of Independence, Woolf's
test for homogeneity of Odds Ratios, Fisher's exact test, and the Generalized
Cochran-Mantel-Haenszel test. Regions were identified with median incomes of ≤$50k and
>$50K then were analyzed using a Cochran-Mantel-Haenszel test with an ad hoc analysis
looking at the occurrence of GSWs and stab wounds pre-COVID and during COVID. A
Bonferroni p-value adjustment was performed for all multiple comparison p-values.
Additional demographic variables were collected, including patient demographics,
follow-up, and death, but they were not included in these analyses.