Transmission potential of vaccinated and unvaccinated persons infected with the SARS-CoV-2 Delta variant in a federal prison, July-August 2021

Abstract

Background: The extent to which vaccinated persons who become infected with SARS-CoV-2 contribute to transmission is unclear. During a SARS-CoV-2 Delta variant outbreak among incarcerated persons with high vaccination rates in a federal prison, we assessed markers of viral shedding in vaccinated and unvaccinated persons.

Methods: Consenting incarcerated persons with confirmed SARS-CoV-2 infection provided mid-turbinate nasal specimens daily for 10 consecutive days and reported symptom data via questionnaire. Real-time reverse transcription-polymerase chain reaction (RT-PCR), viral whole genome sequencing, and viral culture was performed on these nasal specimens. Duration of RT-PCR positivity and viral culture positivity was assessed using survival analysis.

Results: A total of 957 specimens were provided by 93 participants, of whom 78 (84 %) were vaccinated and 17 (16 %) were unvaccinated. No significant differences were detected in duration of RT-PCR positivity among vaccinated participants (median: 13 days) versus those unvaccinated (median: 13 days; p = 0.50), or in duration of culture positivity (medians: 5 days and 5 days; p = 0.29). Among vaccinated participants, overall duration of culture positivity was shorter among Moderna vaccine recipients versus Pfizer (p = 0.048) or Janssen (p = 0.003) vaccine recipients. In post-hoc analyses, Moderna vaccine recipients demonstrated significantly shorter duration of culture positivity compared to unvaccinated participants (p = 0.02). When restricted to participants without reported prior infection, the difference between Moderna vaccine recipients and unvaccinated participants was more pronounced (medians: 3 days and 6 days, p = 0.002).

Conclusions: Infectious periods for vaccinated and unvaccinated persons who become infected with SARS-CoV-2 are similar and can be highly variable, though some vaccinated persons are likely infectious for shorter durations. These findings are critically important, especially in congregate settings where viral transmission can lead to large outbreaks. In such settings, clinicians and public health practitioners should consider vaccinated, infected persons to be no less infectious than unvaccinated, infected persons.

Keywords: COVID-19 transmission; Correctional facilities; Infectious disease outbreaks; Vaccination; Virus shedding.

Fig. 1

Timelines and results of nasal mid-turbinate specimens collected from enrolled participants, Federal prison, Texas, July 12—August 9, 2021. The timelines of specimen collection and laboratory results for 95 included participants are represented diagrammatically, indexed by the day of onset. Onset was determined to be either a) date of first onset of self-reported symptom(s) meeting the case definition of COVID-19 or b) date of first positive diagnostic SARS-CoV-2 test, whichever occurred first. Each participant is represented by a horizontal line corresponding to the investigation sampling period during their time-course of illness. Participants who were unvaccinated (including 2 participants who received only the first dose of a two-dose COVID-19 vaccine series) are depicted at the top of the figure, while vaccinated participants are depicted at the bottom. RT-PCR results are represented by solid circles (positive results) or open circles (negative results). For specimens with positive RT-PCR results for which viral culture was performed, culture results are indicated by overlaid blue boxes (positive culture results) or red boxes (negative culture results). Specimens with positive RT-PCR results with a cycle threshold (Ct) value>35 for which viral culture was not performed are indicated by overlaid orange boxes (indicated a presumptive negative viral culture result). Some participants provided specimens during case-finding testing while in quarantine and may have RT-PCR negative specimens collected prior to onset.

Fig. 2

SARS-CoV-2 RT-PCR test positivity survival curves for enrolled participants, Federal prison, Texas, July 12—August 9, 2021. Panels illustrate the results of Turnbull estimation survival functions with a primary endpoint of last positive reverse transcription-polymerase chain reaction (RT-PCR) test result. Solid lines indicate nonparametric maximum likelihood estimates and shaded regions correspond to 95% confidence intervals estimated through modified bootstrap. Survival functions are plotted by Turnbull interval midpoints. Onset was determined to be either a) date of first onset of self-reported symptom(s) meeting the case definition of COVID-19 or b) date of first positive diagnostic SARS-CoV-2 test, whichever occurred first. Panel A depicts RT-PCR positivity by vaccination status. Panel B depicts positivity by vaccine product among vaccinated participants. Panel C depicts positivity according to the time from completion of a COVID-19 vaccine/series to onset. Panel D depicts positivity according to history of known prior SARS-CoV-2 infection.

Fig. 3

RT-PCR Cycle Threshold distributions for enrolled participants with confirmed SARS-CoV-2 infection, Federal prison, Texas, July 12—August 9, 2021. Panels illustrate daily medians and interquartile ranges (IQRs) for reverse transcription-polymerase chain reaction (RT-PCR) cycle threshold (Ct) values among specimens with positive RT-PCR results. Solid lines indicate median Ct values and shaded regions indicate IQRs. Percentages at the top of each panel indicate the proportion of specimens with negative RT-PCR results each day. Onset was determined to be either a) date of first onset of self-reported symptom(s) meeting the case definition of COVID-19 or b) date of first positive diagnostic SARS-CoV-2 test, whichever occurred first. Panel A depicts RT-PCR positivity by vaccination status. Panel B depicts positivity by vaccine product among vaccinated participants. Panel C depicts positivity according to the time from completion of a COVID-19 vaccine/series to onset. Panel D depicts positivity according to history of known prior SARS-CoV-2 infection.

Fig. 4

SARS-CoV-2 viral culture test positivity survival curves for enrolled participants, Federal prison, Texas, July 12—August 9, 2021. Panels illustrate the results of Turnbull estimation survival functions with a primary endpoint of last positive viral culture test result. Specimens were included as presumptive negative results if no culture was performed but were accompanied by negative RT-PCR results or positive RT-PCR results with Ct > 35. Solid lines indicate nonparametric maximum likelihood estimates and shaded regions correspond to 95 % confidence intervals estimated through modified bootstrap. Survival functions are plotted by Turnbull interval midpoints. When Turnbull intervals are bounded by positive infinity (resulting from right-censoring in subgroups), survival functions are truncated by open points at the rightmost non-infinite intervals. Onset was determined to be either a) date of first onset of self-reported symptom(s) meeting the case definition of COVID-19 or b) date of first positive diagnostic SARS-CoV-2 test, whichever occurred first. Panel A depicts viral culture positivity by vaccination status. Panel B depicts positivity by vaccine product among vaccinated participants. Panel C depicts positivity according to the time from completion of a COVID-19 vaccine/series to onset. Panel D depicts positivity according to history of known prior SARS-CoV-2 infection.

Fig. 5

Subgroup analysis of SARS-CoV-2 viral culture test positivity survival curves for enrolled participants, Federal prison, Texas, July 12—August 9, 2021. Panels illustrate the results of Turnbull estimation survival functions with a primary endpoint of last positive viral culture test result. Specimens were included as presumptive negative results if no culture was performed but were accompanied by negative RT-PCR results or positive RT-PCR results with Ct > 35. Solid lines indicate nonparametric maximum likelihood estimates and shaded regions correspond to 95 % confidence intervals estimated through modified bootstrap. P-values of differences in survival functions (using the generalized Wilcoxon-Mann-Whitney method) are displayed at the bottom of each panel. Survival functions are plotted by Turnbull interval midpoints. When Turnbull intervals are bounded by positive infinity (resulting from right-censoring in subgroups), survival functions are truncated by open points at the rightmost non-infinite intervals. Onset was determined to be either a) date of first onset of self-reported symptom(s) meeting the case definition of COVID-19 or b) date of first positive diagnostic SARS-CoV-2 test, whichever occurred first. Panel A depicts viral culture positivity between Moderna vaccine recipients and unvaccinated participants. Panel B restricts this analysis to participants with no known prior infection, comparing positivity between Moderna vaccine recipients and unvaccinated participants Panel C restricts to unvaccinated participants and compares positivity between participants with and without known prior infection. Panel D compares positivity between vaccinated participants (any full primary series) without know prior infection versus unvaccinated participants with known prior infection.