In a recent study published in PLOS ONE, researchers evaluated to what extent prior infection and vaccination protected the residents and staff of the United States (US) California state prison system from breakthrough infection by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) Omicron sub-variant B.1.1.529.
Prisons and jails are risky settings where many coronavirus disease 2019 (COVID-19) outbreaks have occurred during the ongoing pandemic. The California Department of Corrections and Rehabilitation (CDCR) runs the second-largest US state prison system. They vaccinated their residents and staff such that 77.9% and 40.3% of their residents and staff had received a third (booster) messenger ribonucleic acid (mRNA) vaccine dose by late August 2021.
Yet, they identified an Omicron case within the CDCR system in December 2021. Soon after, there was an outbreak among their residents and staff, coinciding with the time of the worldwide Omicron wave. Examining these two high-risk populations ensured reliable ascertainment of infection during the study period owing to their high levels of testing. CDCR runs a multilayered, voluntary reverse-transcriptase–polymerase-chain-reaction (RT-PCR) testing program in response to detected COVID-19 outbreaks. Under this program, they mandatorily test all staff at least once weekly and staff that work in health care facilities at least twice weekly.
About the study
In the present study, researchers collected data from ~60,000 incarcerated persons and 17,000 prison staff in California during the Omicron outbreak period between December 24, 2021, and April 14, 2022, for the retrospective analysis of the risk of infection during the Omicron wave. They used weighted Cox models to compare the vaccine effectiveness (VE) and previous infection against the Omicron subvariant, B.1.1.529.
Notably, they considered previous infections that occurred in the study population before and during the period of Delta variant predominance. The models allowed the baseline hazard to vary according to prison, and inverse probability weighting reduced the confounding effects due to differences in baseline characteristics among cohort members.
The team computed covariate-balancing propensity scores per demographic, clinical, and carceral characteristics. They weighted residents per prison, COVID-19 risk score (0, 1, or ≥2), room type, gender, and age. Likewise, they weighted staff according to prison, position (custody or health care), age (18 to 55 years or more), and gender. The team estimated the extent to which mRNA vaccines protected against infection stratified by the number of doses that persons had received and whether they had had a previously documented infection before the start of the observation period.
Another arm of the study comprised a rolling matched-cohort design that evaluated the effectiveness of three vaccine doses in eligible prisoners relative to those who received only two doses. The researchers included only those residents in the study cohort who were imprisoned in a CDCR prison before January 1, 2021. Likewise, they covered correctional staff employed before January 1, 2021, and who had worked throughout 2021. Notably, by January 1, 2022, over 50% of the prisons had entered an outbreak phase, with the outbreak in the last prison commencing on January 18, 2022.
Among 59,794 residents who met the study inclusion criteria, 16.7% tested COVID-19-positive during the study period. Likewise, 30.3% of 16,572 staff members tested COVID-19-positive during the study period. The median interval between the start of their last infection and the prison outbreak was 393 days for residents; among staff, the same was 367 days.
The effectiveness of the previous infection against Omicron infection among unvaccinated persons ranged from 16.3% to 48.9%. Likewise, VE estimates ranged from 18.6% to 83.2% with two vaccine doses and from 40.9% to 87.9% with three vaccine doses. A booster dose increased VE estimates to 25% for those with no previous history of SARS-CoV-2 infection and 57.9% for those infected before the period of Delta predominance.
The secondary analysis used an alternative design that likely corrected for the biases that would have biased estimates of VE downward. Thus, it more effectively provided estimates among persons who became eligible for a third vaccine dose. Nevertheless, the conservative estimates from both analyses indicated additional protection from a booster mRNA dose against confirmed infection, irrespective of previous infection history.
Another notable observation of the study was that the resident population had lower VE estimates than the staff population for most combinations of exposure history. As testing was neither compulsory nor routinely done for residents, undiagnosed prior infections could have diluted the VE estimates among residents.
To conclude, although mRNA vaccines and previous infection provided less protection against Omicron breakthrough infection(s) relative to earlier SARS-CoV-2 variants, boosters continue to provide additional protection even among previously infected persons. Thus, booster vaccination remains significant for highly vulnerable populations bearing a disproportionate burden of COVID-19.
- Protection against Omicron from Vaccination and Previous Infection in a Prison System, Elizabeth T. Chin, David Leidner, Lauren Lamson, M.S., Kimberley Lucas, David M. Studdert, Jeremy D. Goldhaber-Fiebert, Jason R. Andrews, Joshua A. Salomon, NEJM 2022, doi: 10.1056/NEJMoa2207082