Skip to main content

Identification of presymptomatic and asymptomatic cases using cohort-based testing approaches at a large correctional facility – Chicago, Illinois, USA, May 2020

Our take —

Among 224 quarantined inmates at the Cook County Jail in Chicago, Illinois, more SARS-CoV-2 infections were identified among inmates receiving three tests during a 14-day quarantine compared to a single test at the end of the 14-day quarantine period (17 cases vs. 2 cases). Most new cases were identified within 1-3 days after the start of quarantine, and many were asymptomatic. SARS-CoV-2 attack rates were likely underestimated by high testing refusal rates, diagnostic test characteristics, and potential COVID-19 recoveries in the study population. These results highlight the importance of early and frequent testing among close contacts of cases in congregate settings.

Study design

Prospective Cohort

Study population and setting

Between May 1 and 19, 2020, incarcerated individuals in the Cook County Jail (Chicago, Illinois, United States) were placed into a 14-day quarantine after at least 1 laboratory-confirmed SARS-CoV-2 infection was identified in a housing unit. Two testing strategies were used to detect SARS-CoV-2 transmission in quarantined housing units following potential exposure: 1) single-time testing (at day 14 of the quarantine period) or 2) serial testing (3 time points during the quarantine period: day 1, days 3-5, days 13-14). All individuals in a quarantined housing unit received the same type of testing strategy. The investigators compared COVID-19 positivity (confirmed by RT-PCR) between housing units receiving single-time and serial testing, respectively.

Summary of Main Findings

Among 137 individuals in the serial testing cohort, 17 of the 96 individuals who received at least one test (18%) had a positive test result for SARS-CoV-2: 16 on day 1, and 1 on days 3-5. Of the 87 individuals in the single-time period cohort, only 2 out of 76 individuals who were tested on the final day of quarantine (day 14), had a positive test result. Among newly identified COVID-19 cases (N = 19), 63% were pre-symptomatic or asymptomatic at the time of diagnosis.

Study Strengths

Investigators compared COVID-19 detection rates in incarcerated populations receiving two distinct testing strategies using individual-level data, which is infrequently available in epidemiologic studies of COVID-19 transmission.


Assignment of the testing protocol to housing units was not random, and there may have been differences in underlying transmission risks between the two groups that were unmeasured or unaccounted for. The estimated attack rate was substantially lower (11%) than what has been observed in other enclosed/incarcerated settings. Possible explanations for this include the high testing refusal rates (13%) in the source population and possible previous SARS-CoV-2 infections that were not ascertained (either by self-report or through antibody testing). Given that one-fourth of quarantined inmates with negative SARS-CoV-2 test results reported illness symptoms consistent with COVID-19 (e.g., loss of taste smell, headache), COVID-19 diagnoses may have been misclassified due to suboptimal RT-PCR test sensitivity.

Value added

This is the first study to compare effectiveness of differenting COVID-19 testing strategies in an incarceration setting, where COVID-19 transmission remains unacceptably high.

This review was posted on: 14 January 2021