COVID-19 in Louisiana's Correctional and Detention Facilities, March–April 2020

This report describes the first weeks of the COVID-19 pandemic, in spring 2020, and the CDC guidance in force at the time. Quarantine and isolation guidance has changed since; see CDC's current recommendations.

Jails, prisons and detention centres make infectious disease hard to control. Space for distancing, isolation and quarantine is limited, and what happens inside does not stay inside: of the more than 10 million admissions to U.S. jails each year, roughly 55% of detainees go back to their communities each week, and staff live in the surrounding towns. High rates of COVID-19 in these facilities can therefore feed wider community spread.

Tracking cases

The first COVID-19 case in an incarcerated person in Louisiana was reported on March 25, 2020, about 2 weeks after the state's first case. Epidemiologists at the Louisiana Department of Health (LDH) enrolled correctional and detention facilities in active surveillance: a daily email asked each for its count of confirmed and suspected cases among incarcerated or detained people and among staff, and for the day's census of people held. The total number of staff was not requested.

On April 4, after a first look at the data, LDH asked CDC for help on site — to measure the outbreaks, find out how far CDC's COVID-19 guidance was known and followed, and train regional epidemiologists to advise facilities. A CDC team arrived on April 6.

From March 25 to April 22, surveillance recorded 489 laboratory-confirmed cases among incarcerated or detained people and 253 among staff, in 46 (32%) of Louisiana's 144 correctional and detention facilities.

Of the 46 facilities, 17 reported cases in both groups, 15 in staff only and 14 in incarcerated or detained people only. They were spread across all nine of the state's health regions and held anywhere from 12 to more than 5,000 people, juveniles and adults: 31 local jails, 11 state facilities, one federal and three private. In the 31 facilities with cases among the people held, the median share of the population confirmed infected over the period was 3%, ranging from under 0.01% to 50%.

A telephone assessment tool

With outbreaks in many places at once, LDH and CDC built the COVID-19 Management Assessment and Response (CMAR) tool, modelled on a CDC infection-control assessment for health care facilities. It walked a facility's health administrators, and leaders such as the sheriff or warden, through CDC's interim guidance for correctional and detention facilities, which recommended:

• suspending transfers and visits;
• hand hygiene supplies, including running water, for everyone;
• symptom screening and a 14-day quarantine for new arrivals before they joined the general population;
• symptom screening for staff at the start of every shift;
• dedicated space for medical isolation and quarantine;
• symptom screening, and coordination with local public health, before anyone was released;
• personal protective equipment for staff and incarcerated people whose duties could expose them to a case;
• assigning staff to particular housing units.

The tool also prompted discussion of problems and lessons learned, and after each section gave a short description of the recommended approach, so that LDH epidemiologists learned the guidance as they used it.

From April 8 to 22, CDC and LDH held 24 CMAR calls: 13 with facilities among the 31 that had cases in the people held, and 11 with facilities among the 113 that did not. These facilities housed men and women, adults and juveniles, from 14 to more than 1,500 people. Dormitory-style housing was reported in 92% of the facilities with cases and 64% of those without.

What the facilities said

All 24 had suspended visits, provided hand hygiene supplies and were screening new arrivals for symptoms; all but one screened staff at shift change. Administrators knew and broadly understood the guidance. What they struggled with was space: there was nowhere to quarantine close contacts one by one, and in dormitory housing people could not keep apart.

• Isolation. Of the 23 facilities able to isolate, most (eight of 12 with cases, nine of 11 without) could isolate suspected and confirmed cases individually; the rest isolated confirmed cases together in groups.
• Quarantine. Of the 23 able to quarantine, 10 of 12 with cases and six of 11 without had little room to quarantine close contacts individually and grouped them instead.
• Transfers. Eleven of the 13 facilities with cases had stopped transfers, against five of 11 without. Those still transferring had cut how often.
• Screening before release. Six of 13 with cases, and two of 11 without.
• Masks or cloth face coverings for everyone held. Nine of 13 with cases; three of 11 without.
• Staff assigned to one unit. Seven of 13 with cases and three of 11 without; staff often had to work across several units.

Facilities said some people held did not report illness because they did not want to be isolated, and in some places a medical visit cost money. At two facilities, daily screening turned up people with fever who had not noticed or had not mentioned their symptoms.

Some facilities went beyond the guidance: releasing people early or lowering bail to reduce crowding, requiring a negative RT-PCR test before ending someone's quarantine, and moving patients to facilities with more space to isolate them. Staff worried about spread from people without symptoms and about people shedding virus after isolation ended.

Their worries had some basis. At two facilities, testing close contacts without symptoms at the end of their 14-day quarantine found six of 10 positive at one and nine of 19 at the other. Two facilities had patients still testing positive when their symptom-based isolation would have ended; they moved people who tested positive into "step-down" units, grouped together for another 7 days after isolation.

What it means

Louisiana has the second-highest incarceration rate in the country, with 144 correctional and detention facilities and an estimated 45,400 people held on any day. The facilities understood the guidance, but their physical, logistical and security constraints made it hard to carry out; where contacts could not be quarantined individually, the virus could spread inside the quarantine units themselves.

Some people without symptoms tested positive at the end of quarantine, though whether they carried live virus is unknown, and more research was needed on spread from people without symptoms in these settings. Facilities that tested people in quarantine, or used the test-based route out of isolation, might need more isolation space — and positive results had been reported up to 36 days after symptoms began, without its being known whether those people were still infectious.

The authors note limits: staff case counts were missing for some facilities and are likely an undercount; testing and case finding varied between facilities; taking part in CMAR was voluntary; the tool was revised as the work went on; and a telephone call cannot confirm what a facility actually does.

No single approach suits every facility, since they differ in size, population, layout and operations, and the people held depend on the system for their protection. CMAR gave public health officials a systematic, easy way to advise facilities and to build local capacity; LDH kept using it, and other local, state and federal agencies can too. Protecting people in these facilities protects the communities around them.

Sources

Based on "Public Health Response to COVID-19 Cases in Correctional and Detention Facilities — Louisiana, March–April 2020," by Megan Wallace and colleagues (CDC; Louisiana Department of Health), Morbidity and Mortality Weekly Report 69(19), posted as an MMWR Early Release on May 8, 2020, published by the Centers for Disease Control and Prevention; rewritten in hubnx's own words.

• The CMAR tool is published as the report's supplementary material: https://stacks.cdc.gov/view/cdc/87561
• The report's social media graphic is not reproduced.