Image from CDC's report
COVID-19 is hard to contain in prisons and jails: people live close together, access to hygiene facilities and supplies varies, and there's little room to isolate the sick or quarantine contacts. Incarcerated people also have high rates of chronic disease, raising their risk of severe illness. And these facilities aren't closed: the virus moves in and out with staff, visitors, and people entering, transferring or leaving.
Early in the pandemic, testing in these settings was mostly limited to people with symptoms. How much did that miss?
The study
In May 2020, CDC asked 15 jurisdictions — the Federal Bureau of Prisons, 10 state prison systems and four city or county jails — for results from mass testing of incarcerated and detained people. Six responded, covering 16 adult facilities: 11 state prisons, three federal prisons and two county jails. Mass testing ran from April 11 to May 20, 2020, 2 to 41 days (median 25) after each facility's first symptomatic case.
What mass testing found
| People offered testing | 16,392 — a median of 54.9% of each facility's population |
| Known cases before (symptom-based testing) | 642 |
| Known cases after | 8,239 — 7,597 previously unrecognized infections |
| Increase per facility | a median of 12.1-fold |
| Prevalence found | 0%–86.8%, median 29.3% |
| Refusals | median 0%, up to 17.3% |
Known cases before and after mass testing, by facility. Image from CDC's report
Every facility that had found a case among incarcerated people by symptom-based testing found more by mass testing; the one that had found none found none again.
Dorms versus cells
Four jurisdictions reported results for 85 housing units in 12 facilities:
| Housing | Prevalence |
|---|---|
| Dormitories — open rooms holding 63 to 216 people | 0%–77.2%, median 42.6% |
| Cells — locked, one to eight people each | 1.8%–45.0%, median 14.6% |
Median prevalence was nearly three times higher in dormitories, suggesting housing layout may drive spread.
Testing again
In two federal prisons, quarantined close contacts who had tested negative were retested 7 days later: 20.5% (90 of 438) and 26.8% (84 of 314) were now positive. Controlling spread may take several rounds of testing, with other prevention measures; testing before releasing people from quarantine may also be warranted.
Why symptom checks fall short
- Asymptomatic and presymptomatic people — an estimated 40%–45% of those infected — can't be found by symptom checks.
- People may hide symptoms, fearing medical isolation and stigma. Refusals of up to 17.3% show the need for messages tailored to culture and language.
- In more than half the facilities, the first case was a staff member — so testing staff regularly, regardless of symptoms, may matter too.
How facilities used the results
- grouping people who tested positive for isolation, and their contacts for quarantine
- targeted testing elsewhere: at intake and release, before outside appointments, and for workers who move between areas, such as food or laundry service
But mass testing took heavy staff time and resources, and lack of space sometimes limited rearranging housing. Facilities asked for guidance on cheaper, faster options, such as point-of-care antigen or antibody tests.
Limits: the facilities were a convenience sample; results may depend on when in an outbreak testing happened and how much of the population was tested; symptom data weren't available; staff cases were likely undercounted; and a person's housing unit at testing may not be where they were infected.
Sources
Based on Liesl M. Hagan, Samantha P. Williams, Anne C. Spaulding and colleagues, "Mass Testing for SARS-CoV-2 in 16 Prisons and Jails — Six Jurisdictions, United States, April–May 2020," MMWR, volume 69, Centers for Disease Control and Prevention; a work of the United States government in the public domain.
Licence: CC0 1.0 (public domain) · Adapted from www.cdc.gov
1
0
0
0
Comments
