Through serial testing of detained persons from quarantined dormitories at a Louisiana detention facility, we identified rapid and widespread SARS-CoV-2 transmission, a large number of asymptomatic infections, and shedding of replication-competent virus in persons with asymptomatic and presymptomatic infections. Despite early adoption of certain prevention and mitigation measures, the cumulative incidence among affected dormitories in facility X was 78%. Of persons who tested positive for SARS-CoV-2, 47% (52/111) were asymptomatic, of which 12 had positive viral culture results with replication-competent virus, indicating infectiousness. In this relatively young population, Ct values were similar regardless of symptom status; the lowest Ct values were among persons with presymptomatic infection, indicating high viral load. These findings add to the evidence that presymptomatic and asymptomatic persons can transmit SARS-CoV-2.
This investigation demonstrated the usefulness of testing shortly after SARS-CoV-2 introduction and at multiple time points to comprehensively identify infections and mitigate transmission. Serial testing identified 52% (58/111) of the COVID-19 cases identified during the investigation. In dormitories A–E, 2 of 53 positive samples from day 0 testing had replication-competent virus, suggesting many persons in these dormitories were convalescent. In dormitory F, 89% (40/45) of residents tested positive for SARS-CoV-2 18 days after all testing negative on day 0; 59% had replication-competent virus. The timing of initial testing in dormitories A–E (2–4 weeks after the first case) and the long testing interval (18 days) in dormitory F limited the usefulness of serial testing to provide data needed to mitigate transmission. Once SARS-CoV-2 introduction into a correctional or detention facility is suspected or confirmed, widespread testing of detained persons and staff at short intervals could quickly identify infections and inform cohorting by infection status to prevent further transmission. In nursing homes, facilitywide testing closer in time to the identification of a COVID-19 case was associated with fewer cases within the facility. Facilities with resource constraints for which widespread testing is not feasible should work with the local health department to determine the most effective testing strategy for their facility.
To complement symptom screening and address the challenges of early detection of SARS-CoV-2, correctional and detention facilities might consider both periodic testing at regular intervals (e.g., 7–14 days) and serial testing of close contacts at short intervals (e.g., 3–4 days) to identify newly acquired infections, infections missed in previous rounds of testing, and new introductions.[8,12,20] Increased dormitory density might also be a risk factor for viral transmission; the lowest cumulative incidence occurred in dormitory E, which had lowest occupancy. Some facilities have reduced occupancy as a mitigation strategy. Novel testing approaches (e.g., pooled testing), point-of-care rapid antigen assays, and less intrusive specimen collection methods are urgently needed to enable efficient SARS-CoV-2 testing. This investigation found no differences in handwashing and mask use between persons who tested positive or negative for SARS-CoV-2. A small proportion overall (13%) reported always using a mask which, along with close living quarters, might have limited the effectiveness of these personal mitigation measures.
During follow-up, 22 persons tested positive ≥14 days after their first positive result and 1 person tested positive 48 days after symptom onset. Four persons had positive rRT-PCR results ≈3 weeks after the first positive result, which was longer than that seen in previous investigations of patients with mild illness.[22,23] However, replication-competent virus was not isolated from these specimens or any specimens collected >9 days after symptom onset. This finding lends support to facilities using symptom-based criteria for release after 10 days of isolation, with resolution of fever and improvement of other symptoms, instead of test-based criteria.
Phylogenetic analysis identified 3 distinct clusters of SARS-CoV-2 infection from 41 specimens collected within the same month from detained persons in dormitories A, D, E, and F. Given the genetic distance between the groups within a short time period and the overall diversity of sequences from the COVID-19 outbreak, there was likely >1 introduction of SARS-CoV-2 into the facility before May 29. In addition to mitigation measures to prevent SARS-CoV-2 spread within a facility, measures should be taken to limit introductions into the facility, including routine symptom screening and testing at entry, use of face masks, and systematic assignment of staff to specific dormitories.
We identified 4 primary limitations to this investigation. First, serial testing was initiated 2–4 weeks after the first case was identified in dormitories A–E, which limited our ability to assess the impact of testing and cohorting on preventing transmission if most detained persons had been infected before the investigation. In addition, persons who tested negative for SARS-CoV-2, including 53% who reported COVID-19 symptoms, might have had COVID-19 and cleared their infections by the time of testing, leading to an underestimation of the prevalence of SARS-CoV-2 infection. No antibody testing was performed; thus, the extent of prior infection cannot be estimated. Second, detained persons might have limited recall of mild symptoms and symptom timing, particularly symptoms occurring >2 weeks before testing, potentially resulting in an overestimation of the prevalence of asymptomatic infection. Also, follow-up symptom assessments were not conducted among persons with positive test results from dormitory F, thus potential presymptomatic detained persons remained classified as asymptomatic. Third, given our inclusion of symptoms reported up to 6 weeks before testing, misclassification of symptoms caused by other pathogens or allergies could have occurred. Finally, no systematic testing of facility staff or detained persons in other dormitories was part of this investigation.
In correctional and detention facilities, prevention and mitigation of SARS-CoV-2 transmission requires a combination of measures. Testing is necessary to identify asymptomatic and presymptomatic persons who can silently transmit the infection. Although symptom screening alone was not sufficient to identify SARS-CoV-2 infections, it could serve as a signal for SARS-CoV-2 introduction and initiation of widespread testing. To increase sensitivity of symptom screening, screenings should use an expanded COVID-19 symptom list based on the latest evidence and guidance, and barriers to symptom reporting, such as medical care costs or concerns over medical isolation, should be minimized.[18,25,26] Multiple rounds of widespread testing for detained persons and staff might be necessary for early detection of virus introduction, particularly when there are high rates of transmission in the surrounding community and ongoing risk for reintroduction. When initiated early in an outbreak, results from serial testing 3–4 days after an exposed person first tests negative for SARS-CoV-2, paired with mitigation strategies, might help limit transmission among detained persons. SARS-CoV-2 testing in these congregate settings will likely be most effective when timed soon after viral introduction, inclusive of all potentially exposed staff and detained persons, and combined with infection control mitigation strategies such as medical isolation and quarantine.
We thank persons incarcerated and detained at the detention facility, detention facility staff members, Louisiana Department of Health officials, Louisiana Office of Public Health Laboratory officials, Lauren Franco, Julian Grass, Jennifer Huang, Hannah Kirking, Eric Manders, Claire Midgely, Erin Moritz, Amy Schumacher, Margaret Williams, the Public Health Institute, and the CDC COVID-19 Epidemiology Task Force for participating in this study.
Emerging Infectious Diseases. 2021;27(2):421-429. © 2021 Centers for Disease Control and Prevention (CDC)