HOTBEDS OF
INFECTION
How ICE Detention Contributed to
the Spread of COVID-19 in the United States

Accnowledgements
Authors
This report was a collaborative effort of Detention Watch Network (DWN) and the primary
authors, Gregory Hooks and Bob Libal, with contributions from Setareh Ghandehari of
DWN. Additional review and edits were provided by Aimee Nichols, Carly Pérez Fernández,
and Silky Shah.

Gregory Hooks

Bob Libal

Gregory Hooks is Professor of

Bob Libal is a nationally recognized

Sociology at McMaster University.

organizer, writer, and strategist who

In prior research, he has challenged

has worked for more than 15 years on

the widespread belief that prisons

issues related to prison privatization,

bring jobs and other economic

immigration enforcement, and mass

benefits to host communities. More

incarceration. Bob is author or co-author

broadly, his research examines spatial

of numerous reports and articles and

inequality, inequality and social policy,

is regularly interviewed by national,

environmental inequality (from both

regional, and local press on issues

military and civilian sources), and peace,

related to prison privatization, criminal

war and development.

justice reform, immigration enforcement,
and the business of prisons.

About Detention Watch Network
Detention Watch Network is a

strategy and build partnerships on a

national coalition of organizations

local and national level.

and individuals building power
through collective advocacy,
grassroots organizing, and strategic
communications to abolish immigration
detention in the United States. Founded
in 1997 by immigrant rights groups,
DWN brings together advocates to unify

Design: Martyn Andrés Bonaventura
Cover Image: Fernando Lopez
© December 2020
Detention Watch Network
Visit www.detentionwatchnetwork.org
— Page 2 —

Introduction
Since the World Health Organization

immigration detention system, operated

(WHO) declared COVID-19 a global

by ICE, has a well-documented history

pandemic on March 11, 2020, the virus

of medically negligent and abhorrent

has devastated communities around the

conditions.3 Experts have long

world. The United States has seen more

condemned the agency for violating

than twelve million confirmed cases

international norms and placing the

at the time of writing this report, the

health and welfare of detained people at

highest in the world.

risk. Despite the overwhelming evidence,

The U.S. response to the deadly virus
continues to be an example of what
not to do, largely ignoring the advice of
public health professionals and other
experts, doing little to stop the spread
of infection, and displaying a callous
disregard for the safety and health of
its residents. The results have been

the restrictive and punitive detention
system has continued to grow over
the last several decades. While these
defects and steady growth of the system
predate 2016, the Trump administration
further expanded the detention system,
promoted punitiveness, and degraded
health and safety conditions.4

both tragic and predictable. Federal and
state policies have been inconsistent
and ineffective, and medical supplies
(from personal protective equipment
[PPE] to testing supplies) have been
unavailable.1 In short, the U.S. has failed
— and continues to fail — spectacularly.
With roughly five percent of the world’s
population, the U.S. accounts for 20
percent of COVID-19 cases and deaths

ICE’s failure to release
people from detention
during the pandemic

worldwide.2

added over 245,000 cases

The response of Immigration and

to the total U.S. caseload.

Customs Enforcement (ICE) to the
pandemic also played a role in
the spread of COVID-19. The U.S.

— Page 3 —

Recommendations
The immigration detention system

In this context, it is not surprising

is cruel and unnecessary. Its defects

that ICE detention facilities have

are only heightened during a global

been uniquely vulnerable to the

health crisis. People navigating their

novel coronavirus, with an infection

immigration cases should be able to

rate that far outweighs the infection

do so with their loved ones and in

rate among the population.

community, not behind bars.

Immigration advocates and public
health officials warned of these risks

The only just and long-term
solution is to free all people
from detention.

in March and made clear the only

In the meantime,

distance. People in detention fearing

way forward was to immediately
release people from detention
so that they could safely socially
for their lives have spoken out and

1. ICE must immediately heed the

protested, asking to be released in

advice of public health experts

one of the few ways that they can:

by significantly and quickly
reducing the number of people in
detention.

5

2. ICE must halt enforcement
activities.
3. ICE must halt all transfers within
the immigration detention
system as well as all transfers
from state and local jails and
prisons.
4. ICE must adopt a moratorium on
deportations in conjunction with
the above recommendations.

refusing meals. From March to July
2020, nearly 2,500 people joined in
COVID-19-related hunger strikes in
detention centers nationwide.6
ICE refused to heed the warnings
and even evaded court orders
requiring them to reduce numbers.7
As expected, the virus swept
through the ICE detention system,
impacting detained people and
those working in detention facilities
at disproportionate rates, as well as
their families.8
The impact of ICE’s failure to
adequately respond to the pandemic

— Page 4 —

was far reaching and multilayered.

Based on the findings of this report, ICE’s

People working at detention centers

failure to release people from detention

travel to and from their homes and

during the pandemic added over

communities, potentially introducing

245,000 cases to the total U.S. caseload.

the virus both to people detained and
to their communities. Multiple reports
revealed that those working in ICE
detention centers were not regularly
wearing PPE. In addition, detained

ICE’s deadly detention
system

people were not given adequate access

ICE operates a sprawling network of 221

to soap or PPE.9 Further, ICE’s continued

dedicated and non-dedicated detention

and irresponsible transfer of people

facilities throughout the U.S.12 This

throughout the detention system also

system has proven deadly to those it

facilitated the spread of the virus.10 As

detains. More than 200 people have

community transmission surged out

died in ICE custody since the creation of

of control in the spring and summer of

the Department of Homeland Security

2020, counties with detention facilities

(DHS) in 2004. In Fiscal Year (FY) 2020,

and surrounding counties endured

21 people died in ICE detention.13 Even

higher rates of infection. Even as

excluding deaths due to COVID-19, FY

COVID-19 cases have surged across the

2020 was the deadliest year for people

country, ICE has ramped up enforcement

detained by ICE since 2005.14

activities,11 creating a recipe for disaster
for those in detention and surrounding
communities.
This report adds to the body of research
that points to ICE’s abuse and medical
neglect of people in detention, and its
failure to adequately respond to the
ongoing COVID-19 pandemic. ICE’s
failure to release people led to higher
numbers of COVID-19 cases in counties
where detention centers are located and
the economic areas that surround them.

— Page 5 —

People navigating their
immigration cases should be
able to do so with their loved
ones and in community, not
behind bars.

(t
were exposed while in custody.17

exposure to mumps and chicken pox.

Prevention (CDC), 84 percent of patients

10 detained people at the time — for

to the Center for Disease Control and

in quarantine — or about one in every

mumps cases in 57 facilities. According

In 2019, ICE had to place 5,200 people

August 2019, there were 898 reports of

spreads rapidly through the system.

between two ICE detention centers. By

have long warned infectious disease

mumps among immigrants transferred

medical neglect.15 Public health officials

Services reported five confirmed cases of

products, inadequate food, abuse, and

the Texas Department of State Health

including lack of access to basic hygienic

contagious diseases. In October 2018,

that typify immigration detention,

appropriately respond to outbreaks of

documented the egregious conditions

ICE has repeatedly failed to

Detention Watch Network has previously

Advocates attributed the outbreaks to
inadequate medical care in a lawsuit
against the agency.16

ADELANTO DETENTION CENTER | PHOTO CREDIT: ALONSO YÁÑEZ, LA OPINION

— Page 6 —

COVID-19 in ICE detention
facilities

were detained in facilities with active

Against this backdrop, it is no surprise

tested positive shortly after their

that COVID-19 spread rapidly in the

deportations.20

ICE detention system. Throughout the
pandemic, ICE failed to provide adequate
supplies of soap and PPE to people in
detention and to detention center staff.
Testing was inadequate and irregular.
As of November 19, 2020, ICE has
reported 7,339 positive cases among
detained people out of a total of 62,080
individuals who had been tested. To
date, eight people in ICE detention have
died of complications from COVID-19.18
The number of those who have died due
to COVID-19 contracted at ICE facilities
may, in fact, be higher. It is possible that
a number of people contracted the virus
in detention, were not tested, and then
passed away after being released from
detention or deported.

COVID-19 cases without testing them.
People deported to countries including
India, Haiti, Guatemala, and El Salvador

At the Stewart Detention Center in
Columbus, Georgia, three detained
people have died and at least 379 have
tested positive for the virus. Santiago
Baten-Oxlaj, a 34-year-old Guatemalan
immigrant, died in May. Mr. Baten-Oxlaj
was detained for six weeks at Stewart
and was infected with the virus during
his detention.21 Weeks before his death,
people detained at Stewart went on
hunger strike demanding the most basic
health precautions and advocacy groups
demanded their release.22
In May, officials in Pearsall, Texas
raised the alarm after every local case
of COVID-19 could be traced back to
ICE’s negligence at the South Texas

In at least one detention facility — the
Mesa Verde ICE Processing Center

ICE Processing Center. Local officials
expressed concern that GEO Group, the
company that operates the facility, failed

in Bakersfield, California — ICE
purposefully rejected universal testing
because it would be too difficult to

to respond to emails or properly keep
the community appraised as the virus
quickly spread in the facility.23

quarantine all detained people who
may test positive.19 ICE has even

In August, a federal judge ordered

been blamed for spreading the virus
internationally by deporting people who

ICE to stop transferring people to the
detention center in Farmville, Virginia

— Page 7 —

after 339 detained people tested positive

1. Counties with ICE detention centers

for COVID-19. The judge commented

were more likely to report COVID-19

that social distancing was not enforced

cases earlier in the pandemic than

and that many staff at the facility did

counties without a detention center.

not wear proper protective equipment,

Not only were counties with ICE

including masks.24

facilities more likely to see an initial
case in the spring of 2020, these

In October, the conditions devolved so

counties were also more likely to

badly at the nearly 2,000-bed Adelanto

confront a serious outbreak (at

Detention Center in California that a

least 15 cases), a major outbreak

federal judge ordered the administration

(more than 250 cases) and a health

to immediately begin releasing people

care emergency (more than 2,500

from detention. More than 160 people

confirmed cases).

in detention and 30 staff members were
infected at the facility, even as ICE and

2. The heightened risk was not limited

GEO Group attempted to expand the

to the county where an ICE detention

facility.

facility was located. Nearby counties

25

were also more likely to confront a

ICE’s failed response
contributed to the spread
of COVID-19 throughout the
country

serious COVID-19 outbreak in the
spring of 2020.
3. As the pandemic raged in the
summer of 2020 (May – August),
COVID-19 spread more rapidly in
economic areas with ICE detention
facilities.

Our analysis explores how ICE’s failures
contributed to the spread of COVID-19
across the country, adding to the

4. Larger ICE detention facilities
contributed to accelerated growth in

body of research documenting ICE’s

COVID-19 cases in nearby counties.

mismanagement, grievous medical
negligence, and lack of transparency.

Counties in economic areas with the
largest immigrant detention centers
added an estimated 150 COVID-19

Key findings of the report include:

cases per 100,000 residents to
baseline estimates.

— Page 8 —

August, our analyses reveal that ICE

Studying community
transmission of COVID-19:
Counties and multicounty
economic areas

detention facilities were responsible

Because people commute, shop, and

for over 245,000 COVID-19 cases

socialize across counties, for this

throughout the country. These cases

study we examined the impacts of ICE

were concentrated in multicounty

detention for the county in which an

economic areas where ICE facilities

ICE facility was located. Those who

are located.

work in one county but live in another

5. Taken as a whole, the spread of
COVID-19 due to ICE’s negligence was
dramatic. Across the United States,
the COVID-19 caseload surged over
the summer of 2020. ICE exacerbated
the pandemic. Between May and

6. If a country had reported 245,000
cases on August 1st to the World
Health Organization, that country
would have ranked 16th in the
world, meaning that the number of
COVID-19 cases attributed to spread
caused by ICE detention in the
United States would have outranked

county are exposed to COVID-19 in more
than one county. And, if they become
infected, they can infect people in more
than one county. For this reason, we also
examined community transmission of
COIVD-19 to nearby counties, specifically
counties co-located in Bureau of
Economic Analysis economic areas.26

countries including Germany, France,

Consider the Farmville Detention Center

and Canada.

(FDC), located in Farmville, Virginia

7. California, Texas, and Arizona had
the most net COVID-19 cases due
to the presence of ICE detention
facilities. Arizona had by far the most
net additional COVID-19 cases per
100,000 residents.

(Prince Edward County), approximately
65 miles from Richmond. Including
Prince Edward County, 39 counties are
in the Richmond economic area. FDC
experienced a severe outbreak during
the summer of 2020, with more than 75
percent of people detained there testing
positive for COVID-19.27 In addition to
raising concerns about the management
of the facility, this high rate of infection

— Page 9 —

– Table 1 –
Number of people detained by county and economic area

LITTLE OR NO
ICE DETENTION

HIGHER RATE OF
ICE DETENTION

157 counties with 2 or

COUNTY

2,983 counties
With 0 or 1 person detained

more people detained
Median number of people
detained among these counties:
60 people

1,345 counties in BEA area with
more than 25 people in ICE

MULTI-COUNTY
BEA ECONOMIC
AREA
(179 AREAS)

detention.
Median number of immigrants
detained in the BEA economic
area among these counties: 245
people

— Page 10 —

at FDC elevated the risk to residents of

1 and 2 below), a simple contrast is

Farmville, residents of Prince Edward

presented – counties in the United

County, and those residing in nearby

States with one or zero immigrants

counties. In the following analyses, we

in ICE detention versus counties with

include a measure to assess exposure to

at least two people in ICE detention.

the county in which the facility is located,

Parallel to the county-specific measure,

and we include a second measure that

the multicounty measure contrasts

taps into exposure to counties in the

counties in economic areas with 25 or

larger economic area.

fewer people in ICE detention to those in
multicounty economic areas with more

In analyses focused on the arrival of
COVID-19 in the spring of 2020 (Figures

than 25 people in ICE detention.28

ADELANTO DETENTION CENTER | PHOTO CREDIT: ALONSO YÁÑEZ, LA OPINION

— Page 11 —

In 2020, active ICE detention facilities
were located in a minority of counties
across the country. Among the 3,143
counties included in these analyses,
the vast majority (2,983 counties – 95
percent) have no detention facility (0 or
1 person detained). In the remaining 157
counties (about 5 percent of all counties)
ICE detained at least two people in Fiscal
Year 2020. Among these 157 counties,
the median number of people detained
was 60. In other words, half of these 157
counties had fewer than 60 people in
detention; the other half had more than
60 people in detention.

Counties and multicounty
economic areas with
ICE facilities were more
likely to confront
serious COVID-19
outbreaks
When considering the role that ICE
played in the spread of COVID-19, it
is not enough to simply look at the
presence of ICE facilities. A myriad
of factors (other than ICE facilities)
impacted the timing and severity of

However, when we zoomed out to
consider economic areas, rather than
individual counties alone, we found that
more than two-thirds of counties in
the U.S. (2,211 counties or 70 percent)
are located in an economic area where
ICE detention centers are present.
Only 30 percent of counties (929) are
in economic areas with zero people in
ICE detention. A significant number of
counties are in economic areas in which
ICE detention is more prominent. Fortythree percent of counties (or 1,345) had
more than 25 people in ICE detention in
their economic area. The median
for these counties is 245 people in ICE
detention across the economic area.

COVID-19 outbreaks.29 Our analyses
consider the presence of an ICE facility in
a county and in a multicounty economic
area (25 or more people detained in the
larger economic area).30
Using this approach, it is possible to
hold constant a host of factors that
might influence the timing of COVID-19
arriving in a county and zero in on
factors of concern. Logistic regression31
was employed to evaluate the possibility
that counties (Figure 1, next page) were
at heightened risk of (a) COVID-19 being
present in the county by April 1st (at
least one case), (b) COVID-19 being
present in the county by May 1st (more
than 2 cases), (c) significant outbreak (15

— Page 12 —

– Figure 1 –
Impact of ICE Detention Facility in the County on Confirmed
Cases of COVID-19 (various measures)

Percent of counties

Between March 1 and May 1, 2020 (3,071 counties)
89%

90%

80%

70%

80%

81%

69%

60%
60%

53%
50%

40%

30%

18%

20%

13%
10%

2%

4%

0%
0 -1 person
Detained by ICE

More than 1 person
Detained by ICE

Figure 1 is focused on five unwanted COVID-19 milestones:
Presence:
More than 2 cases (May 1)

Serious outbreak:
More than 250 cases (May 1)

Presence:
At least one case (April 1)

Major outbreak:
More than 2,500 cases (May 1)

Significant outbreak:
More than 15 cases (May 1)

cases) by May 1st, (d) serious outbreak

confirmed more than 15 cases, counties

(more than 250 cases by May 1st), and

with a facility were 7 points more likely

(e) a major outbreak (more than 2,500

to have done so (60%).

cases by May 1st).

ICE facilities also heightened the risk of

After holding all other variables constant
(i.e., all other variables in the logistic
regression model held at their respective

more serious outbreaks:
•

counties with 0 or 1 person detained,

means), presence of an ICE facility (more

compared to 18% of counties with 2

than one person detained) significantly

or more people detained.

increased a county’s risk of a COVID-19
event. Whereas 69% of counties
were dealing with at least one case of

More than 250 cases (May 1): 13% of

•

More than 2,500 cases (May 1):
Few counties were dealing with an

COVID-19 by April 1st, an ICE detention

outbreak of this magnitude. But

facility made this 11% more likely (80%).

counties with ICE facilities (3.5%)

A month later (May 1st), more than 2

were at significantly greater risk than

cases had been confirmed in 80% of

counties without an ICE facility (2.3%).

counties without an ICE facility and in
89% of counties with a facility. Similarly,

Figure 1 reveals that the novel

whereas slightly more than half (53%)

coronavirus does not respect the walls

of counties without an ICE facility had

and fences surrounding ICE detention
facilities. ICE employees, vendors,
contractors, and visitors bring the virus
with them as they travel to and from

The presence of an ICE
facility significantly increased
a county’s risk of a
COVID-19 event.

the facility where it can spread rapidly
in congregate settings. Especially in
the initial spread of the virus, residents
of counties in which ICE facilities were
located were more likely to report
COVID-19 cases early in the pandemic
and were at heightened risk of a serious
outbreak. Nor does the virus respect
county boundaries. As reported in
Figure 2 (on page 16), the heightened

— Page 14 —

risk attributable to ICE facilities extended

Presence (as of May 1st):

to counties across multicounty economic

•

areas. Figure 2 is also focused on

More than 15 cases: The risk of
having 15 or more cases increases

unwanted COVID-19 events. Notably,

by 5% -- 51% of counties with 25

our analyses provide evidence that that

or fewer persons detained in the

presence of an ICE facility did not make it

entire BEA area compared to 56% of

more likely that a multicounty economic

counties in BEA area in which more

area reported COVID-19 cases early in

people are detained.

the pandemic, only that the presence of
an ICE facility made it more likely that

Larger outbreaks (as of May 1st):

the multicounty economic area faced a

•

serious outbreak:

More than 100 cases: 21% of
counties in BEA areas with fewer

Presence:
More than 15 cases (May 1)

people detained reported 100 cases,

•

Significant outbreak:
More than 100 cases (May 1)

counties in a BEA area with more

•

Serious outbreak:
More than 250 cases (May 1)

•

•

compared to one-fourth (25%) of
than 25 persons detained.
•

More than 250 cases: For counties
in a BEA area with more than 25

Major outbreak:
More than 2,500 cases (May 1)

persons detained, the risk of a major
outbreak was over 14%. The risk was

The contribution of ICE detention in

approximately 2% lower in BEA areas

multicounty economic areas to several

with fewer persons detained by ICE.

COVID-19 events are displayed in Figure
2 (next page).

•

More than 2,500 cases: Few counties
were dealing with an outbreak of

Figure 2 shifts the focus from the

this magnitude. But counties in a

impact of ICE detention facilities in a

BEA area with more than 25 persons

county (Figure 1) to the impact of ICE

detained were twice as likely to

detention facilities in nearby counties:

confront a major outbreak (3.0%)

more than 25 people detained across

when compared to counties few

the multicounty BEA economic area.

persons detained by ICE (1.5%).

The focus continues to be on unwanted
COVID-19 milestones:

When compared to Figure 2, the
increased risk is higher for each

— Page 15 —

– Figure 2 –
Impact of ICE Detention Facility across the Multicounty BEA Economic Area
on Confirmed Cases of COVID-19 (various measures)

Percent of counties

Between March 1 and May 1, 2020 (3,071 counties)

60.0%

55.8%
50.8%
50.0%

40.0%

30.0%

24.7%
21.4%
20.0%

14.2%
12.3%
10.0%

3.0%

1.5%
0.0%

0 -25 people
Detained by ICE

More than 25 people
Detained by ICE

Figure 2 is focused on four unwanted COVID-19 milestones as of May 1st:
Presence:
More than 15 case

Serious Outbreak:
More than 250 cases

Significant Outbreak:
More than 100 cases

Mayor Outbreak:
More than 2,500 cases
— Page 16 —

COVID-19 event in Figure 1. This should

Building on these analyses, the emphasis

not be surprising. Figure 1 is comparing

now shifts to the impact of ICE facilities

the 157 counties in which ICE detains

as the COVID-19 pandemic spiraled out

2 or more people to nearly 3,000

of control over the summer of 2020.

counties in which 0 or 1 person has been
detained in 2020. In Figure 2, however,

detention. The increased prevalence of

COVID-19 spread more
rapidly in multicounty
economic areas with ICE
facilities

COVID-19 displayed in Figure 2 highlights

Building on the preceding analyses, we

the comparison is between 1,345
counties in multicounty economic areas
with more than 25 people detained
by ICE and counties in multicounty
economic areas with lower levels of ICE

the risks posed by ICE detention facilities
across many more counties.

next considered the magnitude of the
impact, i.e. the number of additional

Figure 1 provides evidence that counties
with ICE facilities were more likely to
report cases of the novel coronavirus
early on – and heightened risk of serious
outbreaks as well. Figure 2 highlights the
increased risk of serious outbreaks for
counties near those with ICE facilities.

cases that could be attributed to the
presence of an ICE facility. As is common
in health research, the dependent
variable is not the absolute number
of cases. Instead, it is the number of
COVID-19 cases per 100,000 residents.
Poisson regression was employed to
estimate impacts on COVID-19 caseloads
per 100,000 residents.32
We used the average daily population

Counties near ICE facilities

in ICE detention in a county and in the

experienced an increased

the number of cases (per 100,000

risk of a serious outbreak.

to the presence of an ICE detention

surrounding economic area to estimate
residents) that could be attributed
center in a county and the surrounding
economic area. In order to focus on the
relationship between the number of

— Page 17 —

– Table 2 –
Impact of ICE Detention Facilities across BEA Economic Areas
Additional Cases of COVID-19 per 100,000 Residents
between May 1 and August 1, 2020 (3,114 counties)

PERCENTILE RANKING
(Persons detained by ICE
in BEA Economic Area)

CONFIRMED CASES
(per 100,000 residents)

INCREASE OVER
BASELINE

791

-

791

0.0%

798

0.8%

830

5.0%

861

8.9%

950

20.2%

BASELINE:
0 People Detained

50TH PERCENTILE:
5 People Detained

75TH PERCENTILE:
131 People Detained

90TH PERCENTILE:
785 People Detained

95TH PERCENTILE:
1,376 People Detained

99TH PERCENTILE:
2,959 People Detained

— Page 18 —

people detained by ICE and the spread

detention do not diverge significantly

of COVID-19, we included a range of

from the baseline estimate. In fact, at

control variables.

the 50th percentile (5 persons detained

33

The Poisson regression estimates did
not provide evidence that ICE facilities
contributed to growing caseloads of
COVID-19 in the county in which they
were located. However, providing

in the BEA area), a county might
expect a negligible increase (less than
1 additional case per 100,000 on top
of the baseline estimate). However,
as the detained population increases
so does the severity of the COVID-19

additional evidence that the callous

outbreak. Counties in a BEA economic

mismanagement of ICE facilities

areas at the 75th percentile (131 people

contributed to community spread,

detained) were expected to confirm

these analyses did provide evidence
that as the size of ICE detention in the
multicounty economic area increased, so
did the incidence of COVID-19 (see

approximately 7 additional cases per
100,000, and the caseload increased by
39 per 100,000 residents for counties at
the 90th percentile (5% increase). The

Table 2).

situation was worse still for counties in
BEA areas at the 95th percentile (1,376

Table 2 displays additional cases
attributable to ICE facilities in the BEA
economic area. The chart focuses on
an “average” county, i.e., the mean

people detained): 70 additional cases
per 100,000 residents. At the extreme
(99th percentile, 2,959 or more people
detained in the economic area), it is

was assumed for all variables in the

estimated that the number of additional

Poisson regression model, except

cases was more than 150 cases (per

people detained by ICE in the BEA
area. The “baseline” assumes no one
(0) is detained by ICE in the BEA area:
it is estimated that this “average”
county confirmed 791 COVID-19 cases
(per 100,000 residents) between May
1st and August 1st. Table 2 reports
additional cases on top of this baseline
as the number of people detained by
ICE increases. Counties in economic
areas with relatively few people in ICE

100,000 residents) higher – i.e., a 20%
increase in the COVID-19 caseload.
Whereas Table 2 is concerned with
cases per 100,000 residents, these
results can also be used to estimate
the net additional cases attributable to
ICE detention for each county.34 Table
3 (next page) summarizes impacts
across the 25 states where ICE exerted

— Page 19 —

– Table 3 –
Impact of ICE Detention Facilities on 25 States Experiencing Highest Impact
Net Additional Cases of COVID-19 Confirmed
(May 1 - August 1)

RANK

STATE

NET ADDITIONAL
CASES

TOTAL
POPULATION

NET ADDITIONAL
CASES
PER 100,000 RESIDENTS

1

California

111,415.9

39,148,760

284.6

2

Texas

35,564.4

27,885,196

127.5

3

Arizona

28,793.7

6,946,685

414.5

4

Florida

19,906.5

20,598,140

96.6

5

New York

11,429.9

19,618,452

58.3

6

Illinois

10,840.3

12,821,497

84.5

7

New Jersey

5,305.8

8,881,845

59.7

8

Louisina

4,866.7

4,663,616

104.4

9

Mississippi

3,006.8

2,988,762

100.6

10

Washington

1,673.5

7,294,336

22.9

11

Massachusetts

1,601.1

6,830,193

23.4

12

Connecticut

1,501.3

3,581,504

41.9

13

Georgia

1,232.3

10,297,484

12.0

14

Colorado

1,129.4

5,531,141

20.4

15

Minnesota

990.9

5,527,358

17.9

16

Virginia

983.2

8,413,774

11.7

17

Pennsylvania

793.4

12,791,181

6.2

18

Alabama

792.0

4,864,680

16.3

19

New Mexico

442.9

2,092,434

21.1

20

Tennessee

406.7

6,651,089

6.1

21

Indiana

370.4

6,637,426

5.6

22

Michigan

333.6

9,957,488

3.3

23

Maryland

326.2

6,003,435

5.4

24

Nevada

290.0

2,922,849

9.9

25

Rhode Island

268.9

1,056,611

25.4

the strongest impact in the spread of

the rank in Table 3 was based on net

COVID-19.

additional cases per 100,000, Arizona
would be first – and by a wide margin:

Recall that the dependent measure

its estimated 414.5 additional cases per

in our estimation (as summarized in

100,000 is approximately 130 additional

Table 2) is the number of additional
cases per 100,000 residents (rightmost
column in Table 3). Calculating net
additional cases is based on the state’s

cases higher than the second highest
(California with 284.6 additional cases
per 100,000 residents).

total population and additional cases

Adopting a similar approach, we also

per 100,000. For each of the three top

ranked economic areas (Table 4).

ranked states (California, Texas and
Florida), net additional cases per 100,000
residents exceeded 100. However, if

Comparable to states, calculating net
additional cases in a BEA area is based

ELOY DETENTION CENTER | PHOTO CREDIT: STEVE PAVEY
— Page 21 —

– Table 4 –
Impact of ICE Detention Facilities on 25 Economic Area Experiencing Highest
Impact Net Additional Cases of COVID-19 Confirmed
(May 1 - August 1)

RANK

BEA
ECONOMIC AREA

NET
ADDITIONAL
CASES

TOTAL
POPULATION

ADDITIONAL
CASES
PER 100,000
RESIDENTS

1

Los Angeles-Long Beach-Riverside, CA

112,563

20,678,296

544

2

Phoenix-Mesa-Scottsdale, AZ

27,549

5,260,048

523

3

New York-Newark-Bridgeport, NY-NJ-CT-PA

18,524

23,602,788

78

4

Miami-Fort Lauderdale-Miami Beach, FL

18,165

6,855,487

265

5

Houston-Baytown-Huntsville, TX

13,187

7,809,735

168

6

Chicago-Naperville-Mich. City, IL-IN-WI

11,137

10,457,692

106

7

San Antonio, TX

8,871

2,736,961

324

8

Dallas-Fort Worth, TX

6,931

8,892,231

78

9

McAllen-Edinburg-Pharr, TX

4,391

1,356,787

323

10

Lafayette-Acadiana, LA

3,503

867,513

403

11

Jackson-Yazoo City, MS

3,055

1,661,397

183

12

Boston-Worcester-Manchester, MA-NH

1,969

8,594,883

22

13

Jacksonville, FL

1,871

1,884,231

99

14

Seattle-Tacoma-Olympia, WA

1,671

5,168,694

32

15

El Paso, TX

1,257

1,208,018

104

16

Denver-Aurora-Boulder, CO

1,155

4,558,349

25

17

Minneapolis-St. Paul-St. Cloud, MN-WI

1,047

5,533,996

18

18

Richmond, VA

825

1,745,675

47

19

Shreveport-Bossier City-Minden, LA

814

557,323

146

20

Columbus-Auburn-Opelika, GA-AL

771

494,720

156

21

Washington-Baltimore-Northern Virginia, DC-MD-VA-WV

575

10,040,033

5

22

Austin-Round Rock, TX

543

2,181,797

24

23

Albany, GA

523

607,225

86

24

Memphis, TN-MS-AR

518

2,047,494

25

25

Monroe-Bastrop, LA

427

337,021

126

on the total population and additional

in states (Table 3) and multicounty BEA

cases per 100,000. The Los Angeles area

economic areas (Table 4), we calculated

is both a major population center and

the net effect of ICE detention for the

experienced the highest impact from ICE

entire United States. We estimate that

detention activities (544 additional cases

ICE detention activities were linked to

per 100,000 residents). Consequently,

an additional 245,581 cases from May

the number of additional cases in the

1st to August 1st. Because the United

Los Angeles area exceeds 100,000 while

States’ management of COVID-19 is

the second highest impact (Phoenix)

and has been a spectacular failure, this

had an additional 27,549 cases due to

additional caseload may seem modest

ICE detention facilities. Each of the top

in comparison to the overall number

six areas had an increased caseload

of COVID-19 cases in the U.S. After all,

that exceeded 10,000 cases, and the 17

by August 1st, the United States had

highest ranked areas had more than

confirmed nearly 4.5 million cases

1,000 additional cases by August 1st.

(and this is likely an undercount given

Shifting the focus to additional cases
per 100,000 residents (rightmost

the ongoing problems with testing).
However, if ICE’s contribution to the
COVID-19 pandemic is compared to

column in Table 4), only Los Angeles
and Phoenix areas were hit by more
than 500 additional cases. However,
the comparable measure for Lafayette

national caseloads around the world, the
impact of ICE’s poor management of the
pandemic becomes readily apparent.

(Louisiana) exceeded 400 cases, and
Miami, San Antonio, and McAllen (Texas)
were left to cope with more than 250
additional cases per 100,000 residents.
Most of the areas listed in Table 4 had
to come to terms with more than 100
additional cases per 100,000 residents.

A national tragedy

ICE detention activities
were linked to an
additional 245,581 cases
from May 1st to August 1st.

Adopting the same approach that was
used to estimate net additional cases

— Page 23 —

– Table 5 –
Net Additional Cases of COVID-19 Due to ICE Detention (May 1- August 1)
Compared to Countries with 100,000 or More Confirmed Cases

RANK

(as of August 1, 2020)37

COUNTRY

1

USA

2

CONFIRMED
CASES

CASES PER
100,000 RESIDENTS

4,456,389

1,588

Brazil

2,610,102

1,541

3

India

1,695,988

187

4

Russia

845,443

632

5

South Africa

493,183

984

6

Mexico

416,179

396

7

Peru

407,492

1,565

8

Chile

355,667

2,008

9

Iran

304,204

406

10

United Kingdom

303,185

466

11

Spain

288,522

733

12

Colombia

286,020

874

13

Pakistan

278,305

130

14

Saudi Arabia

275,905

854

15

Italy

247,537

419

16

Additional cases attributable to ICE facilities

245,581

17

Bangladesh

237,661

166

18

Turkey

230,873

294

19

Germany

209,653

266

20

Argentina

185,373

624

21

France

175,920

419

22

Iraq

124,609

429

23

Canada

115,799

322

24

Qatar

110,695

3,985

25

Indonesia

108,376

50

Table 5 (previous page) lists all countries

by ICE was more than 13 times higher

with at least 100,000 confirmed cases

than that of the general population. If a

as of August 1st – and inserts the cases

nation reported an infection rate of this

attributable to ICE in this ranking. If the

magnitude to WHO, it would have the

cases linked to ICE were the reported

highest rate of infection in the world –

caseload of a country, that country

and by a wide margin.

would have ranked 16th in the world –
nearly tied with Italy (the site of an early
and severe outbreak). Equally disturbing
is the monthly case rate among people
detained by ICE. In a research letter
published by the Journal of the American
Medical Association (JAMA), Erfani,
Uppal, and Lee35 calculated a monthly
case rate of 6,683 as of August 2020. The
rate of infection among people detained

The broader mismanagement of the
pandemic helps explain the large
number of cases linked to ICE facilities.
Table 5 reports the cases per 100,000
as of August 2020 (rightmost column).
With nearly 1,588.7 cases per 100,000,
the United States had one of the highest
infection rates in the world (only Qatar
[3,985] and Chile [2,008] were higher).

FREE THEM ALL ACTION | PHOTO CREDIT: MARCELA HERNANDEZ
— Page 25 —

Among high-income countries, the

15,000 and 55,000 respectively). This

infection rate in Spain is roughly half

translates into fewer than 50 cases per

of the US rate; for the United Kingdom

100,000 residents. Had the United States

and Italy, the infection rate was less

managed the pandemic comparably to

than a third. For Canada and Germany,

these countries, there might have been

it was lower still. If the infection rate

fewer than 10,000 COVID-19 cases linked

was comparable to Germany or Canada

to ICE detention facilities. Instead, as of

(roughly 20% of that found in the

August 1st, due to the perverse synergy

United States), it is quite possible that

between these two policy failures, over

cases linked to ICE detention facilities

245,000 cases can be traced back to ICE

would have been below 50,000. Neither

detention.

Korea nor Japan are listed in Table 4
because their total caseloads were well
below 100,000 as of August (roughly

While ICE detention contributed to over
245,000 COVID-19 cases in the US, and
the presence of an ICE facility made
a serious outbreak more likely, it is
important to note that the people being

A research letter published by

detained by ICE bear no responsibility

the Journal of the American

the result of (uncontrolled) community

Medical Association found

the walls and fences of ICE facilities.

that the rate of infection

detention, people who are detained are

among people detained

other people in the U.S. Rather detention

by ICE was more than 13

provide the perfect storm for the spread

times higher than that of the

about the consequences. Their failure

general population.

the infection rate in ICE facilities and in

for this result. These 245,000 cases are
transmission of COVID-19 – beyond
Outside of the risk factors posed by their
no more or less likely to get sick than
centers, like other congregate settings,
of a virus. ICE knew this and was warned
to release people resulted in a spike in
surrounding communities.

— Page 26 —

Conclusion and Recommendations
For years, the United States has ignored

spread quickly due to congregate

the advice of experts on immigration

settings. ICE refused to listen to these

detention and has displayed a disregard

warnings. The failure to release people

for the dignity, safety, and health of

from custody in the spring and summer

people in detention. Time and again

of 2020 – despite recommendations

these chronic failures have been

advanced by advocates and public health

exemplified by a culture of abuse and

experts – proved to be catastrophic for

medical neglect. These failures are

people detained, for those working in

endemic in the current system, but they

detention centers, and for those living in

are avoidable. Human rights abuses

surrounding communities.

and medical neglect could be avoided if
those navigating immigration cases were
able to do so at home with their families
and in their communities, not behind
bars. Now during the ongoing global
pandemic, these failures have created
conditions for infection rates to soar.
Medical professionals, advocates, and,
most notably, detained immigrants
themselves called on ICE to release
people from detention as the COVID-19
pandemic grew in early spring 2020,
noting the unique vulnerability people
face in detention. Meanwhile ICE
continued enforcement operations
and transfers between facilities, while
people working in detention centers
went back and forth from work and
home in nearby communities, creating
conditions for exposing the virus to
people in detention, where it could

— Page 27 —

The failure to release
people from custody
proved to be catastrophic
for people detained, for
those working in detention
centers, and for those
living in surrounding
communities.

Our analysis demonstrates that
counties with ICE facilities were more
likely to report cases of COVID-19
early on. The impact then rippled
into the surrounding communities.
Counties with ICE facilities and
their surrounding multicounty
economic areas were more likely to
face a serious outbreak than those
without ICE facilities. The spectacular
mismanagement of COVID-19
resulted in spiraling community
transmission of the disease. Once
introduced into ICE detention,
COVID-19 transmitted quickly
within facilities and the surrounding

The immigration detention
system is cruel and
unnecessary. Its defects are
only heightened during a global
health crisis. People navigating
their immigration cases should
be able to do so with their loved
ones and in community, not
behind bars.
The only just and long-term
solution is to free all people
from detention.

communities and counties (Table

In the meantime,

2). The consequences of this

1. ICE must immediately heed the

transmission were magnified many
times over by the uncontrolled
spread of COVID-19 in communities
across the country.36
As COVID-19 cases are rising sharply
and will continue to climb in the
coming months, now is the time
for immediate action to mitigate
the spread of COVID-19 inside and
outside the detention system.

advice of public health experts by
significantly and quickly reducing the
number of people in detention.38
2. ICE must halt enforcement activities.
3. ICE must halt all transfers within the
immigration detention system as
well as all transfers from state and
local jails and prisons.
4. ICE must adopt a moratorium on
deportations in conjunction with
above recommendations.

Endnotes
1

Altman Drew, “Understanding the US failure
on Coronavirus—an Essay by Drew Altman,”
The BMJ, September 14, 2020, https://www.
bmj.com/content/370/bmj.m3417; David
Crow and Hannah Kuchler, “US Coronavirus
Surge: ‘It’s a Failure of National Leadership’,”
Financial Times, July 17 2020, https://www.
ft.com/content/787125ba-5707-4718-858b1e912fee0a38.

2

Johns Hopkins University, “Coronavirus
Resource Center,” 2020, https://coronavirus.
jhu.edu/map.html

3

Human Rights Watch, Systemic Indifference:
Dangerous & Substandard Medical Care in
US Immigration. Detention, May 8, 2017,
https://www.hrw.org/report/2017/05/08/
systemic-indifference/dangeroussubstandard-medical-care-us-immigrationdetention; J. Rachel Reyes, “Immigration
Detention: Recent Trends and Scholarship,”
The Center for Migration Studies of New
York (CMS), https://cmsny.org/publications/
virtualbrief-detention/

4

Franklin Foer, “How Trump Radicalized ICE,”
The Atlantic, September 2020, https://www.
theatlantic.com/magazine/archive/2018/09/
trump-ice/565772/

5

Human Rights First, “Physicians for Human
Rights and Human Rights First, Public Health
Experts, Medical Doctors, Prison Experts,
and Former ICE Officials Urge Releases from
Immigration Detention Facilities to Control
the Spread of COVID-19.” April 17, 2020,
https://www.humanrightsfirst.org/resource/
public-health-experts-medical-doctorsprisonexperts-and-former-ice-officials-urgereleases.

6

Bárbara Suarez Galeano, “Listen to the
Brave Hunger Strikers in ICE Detention.”
Common Dreams. July 14 2020, https://www.
commondreams.org/views/2020/07/14/
listen-brave-hunger-strikers-ice-detention.

7

City Service News, “ICE Ordered to Reduce
Population at Adelanto Immigrant Detention
Center Hit by COVID-19,” Los Angeles
Times, September 30, 2020, https://www.
latimes.com/california/story/2020-09-30/
ice-detention-center-covid-19-outbreak.
Hailey Konnath, “Judge Slams ICE for Spotty
Compliance With Virus Safety,” 360 Law,
October 8, 2020, https://www.law360.com/
articles/1318310/judge-slams-ice-for-spottycompliance-with-virus-safety

8

Tanvi Misra. “About 900 private ICE detention
staff positive for COVID-19.” Roll Call, July 14
2020. https://www.rollcall.com/2020/07/14/
about-900-private-ice-detention-staffpositive-for-covid-19/

9

“Detained Immigrants Seek Federal Court
Order to Gain Access to Soap, Sanitizer,
PPE Amid COVID-19 Outbreak at Privately
Run Detention Facilities.” Yahoo News,
April 7, 2020, https://finance.yahoo.com/
news/detained-immigrants-seek-federalcourt-150600107.html

10 Hamed Aleaziz, “Federal Officials Now
Say That Transferring Detainees Between
Jails Holding Immigrants Contributed To
Coronavirus Outbreaks,” BuzzFeed News,
October 6, 2020, https://www.buzzfeednews.
com/article/hamedaleaziz/dhs-reportdetainee-transfers-covid-spread
11 Hamed Aleaziz. “ICE Arrested More Than
150 Immigrants In A Nationwide Sweep,”

— Page 29 —

Buzzfeed News, November 19, 2020.
https://www.buzzfeednews.com/article/
hamedaleaziz/ice-arrests-immigrantsnationwide-sweep.

https://www.ice.gov/coronavirus, accessed
September 8, 2020.

12 U.S. Immigration and Customs Enforcement,
“Dedicated and Non-dedicated Facilities
List as of April 6, 2020,” Spreadsheet
available online at: https://www.
ice.gov/doclib/facilityInspections/
dedicatedNonDedicatedFacilityList.xlsx

19 Andrea Castillo, “ICE Deliberately Limited
Testing at Bakersfield Immigration Facility
with COVID-19 Outbreak,” Los Angeles
Times, August 6, 2020, https://www.
latimes.com/california/story/2020-08-06/
amid-coronavirus-outbreak-at-bakersfieldimmigration-facility-emails-show-icedeliberately-limited-testing

13 Catherine Shoichet, “The Death Toll in
ICE Custody Is the Highest it’s Been in 15
Years,” CNN, September 30, 2020, https://
www.cnn.com/2020/09/30/us/ice-deathsdetention-2020/index.html.

20 Emily Kassie and Barbara Marcolini,
“How ICE Exported the Coronavirus,” The
Marshall Project, July 10, 2020, https://www.
themarshallproject.org/2020/07/10/how-iceexported-the-coronavirus.

14 Alex Nowrasteh, “8 People Died in
Immigration Detention in 2019, 193 Since
2004,” Cato Institute, January 8, 2020,
https://www.cato.org/blog/8-people-diedimmigration-detention-2019-193-2004.

21 Hamed Aleaziz, “An Immigrant Man In
ICE Custody Died After Contracting The
Coronavirus,” Buzzfeed News, May 25, 2020,
https://www.buzzfeednews.com/article/
hamedaleaziz/immigrant-ice-coronavirus-death.

15 American Civil Liberties Union, Detention
Watch Network, and National Immigrant
Justice Center, “Fatal Neglect: How ICE
Ignores Deaths in Detention, ” February
2016, https://www.detentionwatchnetwork.
org/sites/default/files/reports/Fatal%20
Neglect%20ACLU-DWN-NIJC.pdf

22 Mijente, “Immigrants in Stewart Detention
Center on Hunger Strike to Demand
COVID19 Precautions, Releases as Georgia
Becomes Southeast Epicenter of Pandemic,”
Latino Rebels, March 27 2020, https://
www.latinorebels.com/2020/03/27/
immigrantsstewartdetention/

16 Border/Lines, “Week 27: ICE Detention and
Infectious Diseases.” Borderlines Newsletter,
April 10, 2020, https://borderlines.substack.
com/p/week-27-ice-detention-and-infectious.

23 Perla Trevizo, “COVID-19 Cases at a Texas
Immigration Detention Center Soared. Now,
Town Leaders Want Answer,” Texas Tribune,
May 11, 2020, https://www.texastribune.
org/2020/05/11/covid-19-cases-soar-texasimmigrant-detention-center-town-wants-answer/

17 Setareh Ghandehari and Gabriela Viera,
Courting Catastrophe: How ICE is Gambling
with Immigrant Lives Amid a Global
Pandemic, Detention Watch Network, March
2020, https://www.detentionwatchnetwork.
org/pressroom/reports.
18 U.S. Immigration and Customs Enforcement,
“COVID-19 ICE Detainee Statistics by Facility,”

24 Associated Press, “Judge: Outside Experts
Can Visit Immigrant Detention Center,” ABC
News, August 17, 2020, https://abcnews.
go.com/Health/wireStory/judge-experts-visitimmigrant-detention-center-72422641

— Page 30 —

25 Rebecca Plevin, “Judge Orders ICE to
Reduce Population at Adelanto Detention
Center amid COVID-19 Outbreak,” The
Desert Sun, October 15, 2020, https://www.
desertsun.com/story/news/2020/10/15/
judge-orders-ice-reduce-populationadelanto-detention-center-amid-covid-19outbreak/3667578001/
26 Kenneth P. Johnson and John R. Kort, “2004
Redefinition of the BEA Economic Areas,”
Survey of Current Business (November 2004),
pages 68-75. The BEA examined linkages
among counties, primarily on the basis of
commuting patterns and media markets.
Through this analysis, more than 3,100
counties were sorted into 179 economic
areas. The emphasis that BEA economic
areas place on commuting is well suited for
the purposes of this report.
27 Priscilla Alvarez, “Nearly 75% of detainees
at US immigration facility in Virginia have
coronavirus,” CNN, July 23, 2020, https://
www.cnn.com/2020/07/23/politics/
immigration-ice-detention-coronavirusfarmville/index.html; Keyris Manzanares,
“Nearly 90 percent of ICE detainees in
Farmville tested positive for COVID-19 after
transfers from Florida, Arizona,” WRIC
(Channel 8), July 16, 2020, https://www.wric.
com/news/taking-action/covid-19-casesspike-at-farmville-ice-detention-center-aftertransfers-from-florida-arizona/
28 U.S. Immigration and Customs Enforcement
2020, op. cit.
29 Data on COVID-19 cases are provided
by: New York Times. 2020. “Coronavirus
(Covid-19) Data in the United State.” Website:
https://github.com/nytimes/covid-19-data.
This website is updated daily (providing data
for each county in the United States).

30 In addition, these models included a number
of control variables: black population, Native
American / Alaskan Native population,
Hispanic population, population living in
group quarters, incarcerated persons (in
county and in BEA area), metropolitan versus
nonmetropolitan county, adults without
health insurance, diabetes prevalence, and
population density. For details on data
sources, model specification and related
details, please see: Hotbeds of Infection: How
ICE Detention Contributed to the Spread of
Covid-19 in the United States (Methodological
Appendix).
31 Statwing, “A User-Friendly Guide to Logistic
Regression,” http://docs.statwing.com/a-userfriendly-guide-to-logistic-regression/, 2020.
32 For additional information on Poisson
regression and its use in health research,
see: Frome, Edward, and Harvey Checkoway.
1985. “Use of Poisson Regression Models
in Estimating Incidence Rates and Ratios.”
American Journal of Epidemiology 121:309–
323, https://doi.org/10.1093/oxfordjournals.
aje.a114001; Osborne, Jason. 2017.
“Poisson Models: Low-Frequency Count
Data as Dependent Variables’, pp. 283304 in Regression & Linear Modeling: Best
Practices and Modern Methods. Thousand
Oaks, CA: Sage [Accessed 5 November 2020],
doi: 10.4135/9781071802724; Population
Health Methods (Columbia University). 2020.
“Confounders in Time-Series Regression.”
Website: https://www.publichealth.columbia.
edu/research/population-health-methods/
confounders-time-series-regression,
accessed October 4, 2020. Frome and
Checkoway 1985; Osborne 2017, Population
Health Methods [Columbia University] 2020).
33 Control variables included: confirmed
COVID-19 cases (per 100,000) on May 1st,

— Page 31 —

black population, Native American / Alaskan
Native population, Hispanic population,
incarcerated persons (in county and in
BEA area), population density, adults with
less than 9 years of education, non-citizen
population, residents of nursing homes,
residents of group quarters (other than
prisons and nursing homes), meatpacking
plants (in county and BEA area), and dummy
variable for each state. For details on data
sources, model specification and related
details, please see: Hotbeds of Infection:
How ICE Detention Contributed to the
Spread of Covid-19 in the United States
(Methodological Appendix).

Epidemiological Update 1 (August 17).
https://www.who.int/emergencies/diseases/
novel-coronavirus-2019/situation-reports.
38 Human Rights First, “Physicians for Human
Rights and Human Rights First, Public Health
Experts, Medical Doctors, Prison Experts,
and Former ICE Officials Urge Releases from
Immigration Detention Facilities to Control
the Spread of COVID-19.”

34 For details on Poisson regression and the
approach to calculating increased COVID-19
cases per 100,000 residents (Table 1) and
total cases (Table 2, 3 and 4), please see:
Hotbeds of Infection: How ICE Detention
Contributed to the Spread of Covid-19 in the
United States (Methodological Appendix).
35 Parsa Erfani, Nishant Uppal, Caroline H. Lee,
“COVID-19 Testing and Cases in Immigration
Detention Centers, April-August 2020,” JAMA,
October 29, 2020, https://jamanetwork.com/
journals/jama/fullarticle/2772627.
36 For details on statistical analyses supporting
these conclusions, see: Hotbeds of Infection:
How ICE Detention Contributed to the
Spread of COVID-19 in the United States
(Methodological Appendix).
37 Sources: World Health Organization. 2020.
Coronavirus Disease (COVID-19). Situation
Report – 194. Data as received by WHO from
national authorities by 10:00 CEST, 1 August
2020. Additional information available
online at: https://www.who.int/emergencies/
diseases/novel-coronavirus-2019/situationreports; World Health Organization. 2020.
Coronavirus Disease (COVID-19). Weekly

— Page 32 —