Characterizing predictors of COVID-19 vaccine refusal in an urban southern California jail population

Background: Correctional populations have been disproportionately affected by COVID-19, and many large outbreaks have occurred in jails and prisons. Vaccination is a key strategy to reduce the SARS-CoV-2 transmission in carceral settings. Although implementation can be challenging due to vaccine hesitancy and medical mistrust, correctional settings provide largely equitable healthcare access and present a unique opportunity to identify potential predictors of vaccine hesitancy independent of access issues.

Methods: We retrospectively analyzed electronic health record data for individuals offered COVID-19 vaccination at the Los Angeles County Jail between January 19, 2021, and January 31, 2023, and used multivariable logistic regression to determine predictors of COVID-19 vaccine refusal.

Results: Of the 21,424 individuals offered COVID-19 vaccination, 2,060 (9.6 %) refused. Refusal was associated with male sex ([aOR] = 2.3, 95 % CI (1.9, 2.8)), age 18-34 ([aOR] = 1.2, 95 % CI (1.1, 1.4), referent group: age 45-54), Black race ([aOR] = 1.2, 95 % CI (1.1, 1.4)), reporting ever being houseless ([aOR] = 1.2, 95 % CI (1.1, 1.3)), and having a history of not receiving influenza vaccination while incarcerated ([aOR] = 2.4, 95 % CI (2.0, 2.8)). When analyzing male and female populations separately, male-specific trends reflected those seen in the overall population, whereas the only significant predictor of vaccine refusal in the female population was not receiving influenza vaccination while in custody ([aOR] = 6.5, 95 % CI (2.4, 17.6)).

Conclusion: Identifying predictors of vaccine refusal in correctional populations is an essential first step in the development and implementation of targeted interventions to mitigate vaccine hesitancy.