Valentine v. Collier

Full title: LADDY CURTIS VALENTINE, et al., Plaintiffs, v. BRYAN COLLIER, et al.…

Court: UNITED STATES DISTRICT COURT SOUTHERN DISTRICT OF TEXAS HOUSTON DIVISION

Date published: Sep 29, 2020

490 F. Supp. 3d 1121 (S.D. Tex. 2020)

Opinion

CIVIL ACTION NO. 4:20-CV-1115

2020-09-29

Laddy Curtis VALENTINE, et al., Plaintiffs, v. Bryan COLLIER, et al., Defendants.

Denise U. Scofield, Benjamin David Williams, Brandon W. Duke, Corinne Stone, Michael Terrell Murphy, Robert Lawrence Green, III, John R. Keville, Winston & Strawn LLP, Houston, TX, David Anthony James, Michael C. Singley, Scott Charles Medlock, Jeffrey S. Edwards, The Edwards Law Firm, Austin, TX, for Plaintiff Laddy Curtis Valentine. Brandon W. Duke, John R. Keville, Winston Strawn LLP, Houston, TX, Jeffrey S. Edwards, The Edwards Law Firm, Austin, TX, for Plaintiff Richard Elvin King. Christin Cobe Vasquez, Leah Jean O'Leary, Office of the Attorney General, Ralph M. Molina, Shawn E. Cowles, Pro Hac Vice, Texas Office of the Attorney General, Austin, TX, for Defendants.

Keith P. Ellison, U.S. District Judge

Denise U. Scofield, Benjamin David Williams, Brandon W. Duke, Corinne Stone, Michael Terrell Murphy, Robert Lawrence Green, III, John R. Keville, Winston & Strawn LLP, Houston, TX, David Anthony James, Michael C. Singley, Scott Charles Medlock, Jeffrey S. Edwards, The Edwards Law Firm, Austin, TX, for Plaintiff Laddy Curtis Valentine.

Brandon W. Duke, John R. Keville, Winston Strawn LLP, Houston, TX, Jeffrey S. Edwards, The Edwards Law Firm, Austin, TX, for Plaintiff Richard Elvin King.

Christin Cobe Vasquez, Leah Jean O'Leary, Office of the Attorney General, Ralph M. Molina, Shawn E. Cowles, Pro Hac Vice, Texas Office of the Attorney General, Austin, TX, for Defendants.

Keith P. Ellison, U.S. District Judge

FINDINGS OF FACT AND CONCLUSIONS OF LAW

I. BACKGROUND ...1127 II. FINDINGS OF FACT ...1128

A. The Parties...1128

B. The COVID-19 Pandemic...1129

C. The Wallace Pack Unit...1131

D. Credibility of Defendants...1131

E. COVID-19 Spread in the Pack Unit...1134

F. Initial Measures Taken by TDCJ...1136

G. The CMHC B-14.52 Policy... 1136

H. Implementation of Preventative Measures at the Pack Unit...1138

i. Social Distancing in Common Spaces...1138

ii. Social Distancing Within the Dorms...1139

iii. Hand Hygiene...1141

iv. Cleaning and Sanitation...1143

v. Laundry Exchange...1146

vi. Masks and Personal Protective Equipment ("PPE")...1147

vii. Testing...1148

viii. Quarantining and Isolating...1151

ix. Contact Tracing...1152

x. Education...1153

I. Compliance with Policy B-14.52...1154

J. Grievance Process...1154

III. CONCLUSIONS OF LAW ...1156

A. Plaintiffs’ Pending Motion for Class Certification...1156

i. Legal Standard...1157

ii. The Mobility-Impaired Subclass...1157

iii. The Disability Subclass...1159

B. Legal Standard for a Permanent Injunction...1160

C. Administrative Exhaustion Under the PLRA...1160

D. Permanent Injunction Analysis...1163

i. Actual Success on the Merits...1163

a. Eighth Amendment...1163

b. Americans with Disabilities Act and Rehabilitation Act...1170

ii. Irreparable Harm...1172

iii. Balance of Harm to Parties and the Public Interest...1173

IV. PERMANENT INJUNCTION ...1174

* * * * *

The Court submits the following Findings of Fact and Conclusions of Law pursuant to Rule 52(a)(1) of the Federal Rules of Civil Procedure.¹

1 Any Finding of Fact that should be a Conclusion of Law shall be deemed such, and any Conclusion of Law that should be a Finding of Fact shall be deemed such.

I. BACKGROUND

Plaintiffs Laddy Curtis Valentine and Richard Elvin King, individuals incarcerated at the Wallace Pack Unit ("Pack Unit"), filed this case as a putative class action on March 30, 2020. (Doc. 1.) Plaintiffs allege that Defendants have failed to properly protect them and other similarly situated Pack Unit inmates from the COVID-19 pandemic. Plaintiffs claim that Defendants are violating the Eighth Amendment by acting with deliberate indifference to the health of the Pack Unit population, which is more vulnerable than the general public is to serious illness or death from COVID-19. (Doc. 1 ¶¶ 74–79.) Plaintiffs also claim that Defendants are violating the Americans with Disabilities Act ("ADA") and the Rehabilitation Act by refusing to accommodate individuals with disabilities by, e.g., providing measures that protect against the spread of COVID-19 in the Pack Unit. (Doc. 1 ¶ 81–89.) Plaintiffs seek injunctive and declaratory relief only. (Doc. 1 ¶¶ 96–100.) After an evidentiary hearing, the Court entered a preliminary injunction order on April 16, 2020. (Doc. 40.) On April 22, 2020, the Fifth Circuit stayed the Court's preliminary injunction order pending appeal. Valentine v. Collier , 956 F.3d 797, 801–05 (5th Cir. 2020) (" Valentine I "). On June 5, 2020, the Fifth Circuit vacated the Court's preliminary injunction order on the basis that Defendants "ha[d] substantially complied with the measures ordered by the district court" and remanded the case to this Court for further proceedings on the permanent injunction. Valentine v. Collier , 960 F.3d 707, 707 (5th Cir. 2020) (" Valentine II ").

After the case was remanded, this Court certified a general class of inmates at the Pack Unit and a high-risk subclass of inmates who have medical conditions or who are of older age. Valentine v. Collier , No. 20-cv-1115, 2020 WL 3491999 (S.D. Tex. June 27, 2020) (Doc. 160) (" Class Cert. Order"). The Court also denied Defendants’ Motion to Dismiss. (Doc. 172). As explained infra Part III(A), the Court has also certified a mobility-impaired subclass of inmates who require the use of a walker, cane, crutches, or wheelchair to ambulate.

On July 13, 2020, the case proceeded to a bench trial before this Court. Over the course of the eighteen-day trial, the Court received exhibits and heard sworn testimony from fact and expert witnesses. Following trial, the Court granted Plaintiffs’ motion to admit newly discovered evidence revealed during and after trial. (Minute Entry Aug. 25, 2020). Having considered the evidence, testimony, oral arguments presented during the trial, post-trial filings, and all applicable law, the Court sets forth the following Findings of Fact and Conclusions of Law.

II. FINDINGS OF FACT

A. The Parties

1. Named Plaintiff Laddy Curtis Valentine is a 69-year-old man who is currently incarcerated at the Pack Unit. Mr. Valentine suffers from chronic medical conditions including drop foot, atrophy and weakness in the upper-left extremity, high blood pressure, hypertension, nerve damage from a stroke, and limited ability to grip with his left hand. (Tr. 1-148:20–149:20, 150:4–17 (Valentine).)²

2. Named Plaintiff Richard Elvin King is a 73-year-old man who is currently incarcerated at the Pack Unit. Mr. King has diabetes, high blood pressure, and chronic kidney failure. (Tr. 2-10:6–11:13 (King).) As a diabetic, Mr. King has difficulty regulating his blood sugar. (Id. ) Mr. King also suffers from hyperlipidemia, diabetes, and kidney disease. (PTX 184 at 11, 14.)³

3. Mr. Valentine and Mr. King (collectively, "Plaintiffs") represent three certified classes of inmates at the Pack Unit (collectively, the "Class"):

a. The General Class includes:

All current and future inmates incarcerated in the Texas Department of Criminal Justice Wallace Pack Unit who are subjected to TDCJ and the Texas Correctional Managed Health

2 The transcript for the eighteen-day bench trial before this Court is located at docket entries 304–336. The Court will cite to the transcripts as "(Tr. [day]-[page]:[line]–[page]:[line] (witness).)."

3 The Court will cite to trial exhibits for Plaintiffs and Defendants as "(PTX ## at [page].)" and "(DTX ## at [page].)," respectively.

Care Committee's policy and practice of failing to provide protection from exposure to COVID-19 during the class period.

b. The High-Risk Subclass includes:

All current and future inmates incarcerated in the Texas Department of Criminal Justice Wallace Pack Unit who are subjected to TDCJ's policy and practice of failing to provide protection from exposure to COVID-19 during the class period and who are, according to the CDC, most at risk for severe illness, injury, or death from COVID-19 due to their age or their health conditions, including the following individuals:

• People aged 65 years or older;

• People with chronic lung disease or moderate to severe asthma ;

• People who have serious heart conditions;

• People who are immunocompromised including patients undergoing cancer treatment;

• People with other underlying medical conditions, particularly if not well controlled, including, but not limited to, those with diabetes, renal failure, or liver disease; and

• People of any age with severe obesity (body mass index [BMI] = 40).

(Doc. 160 at 3.)

c. The Mobility-Impaired Subclass includes:

All current and future inmates incarcerated in the Texas Department of Criminal Justice Wallace Pack Unit who suffer from a disability that substantially limits one or more of their major life activities, are subjected to TDCJ and the Texas Correctional Managed Health Care Committee's policy and practice of failing to provide protection from exposure to COVID-19 during the class period, and who require the use of a walker, cane, [crutches], or wheelchair to ambulate.

(Doc. 248 at 11.)

4. Defendant Texas Department of Criminal Justice ("TDCJ") is the prison system for the state of Texas and is an agency of the state. TDCJ operates 104 prisons around the state, including the Pack Unit, the unit at issue here. (Tr. 1-41:10–22 (Collier).)

5. Defendant Robert Herrera is the senior warden of the Pack Unit. (Doc. 205 ¶ 6.) As the warden of the Pack Unit, Mr. Herrera oversees the staff and inmates housed at the Pack Unit. (Id. )

6. Defendant Bryan Collier is the Executive Director of TDCJ who oversees the activities of TDCJ. (Doc. 205 ¶ 5; Tr. 1-41:10–22 (Collier).) Mr. Collier is responsible for Mr. Herrera's actions or inactions at the Pack Unit. (Tr. 10-98:23–99:2 (Collier).)

B. The COVID-19 Pandemic ⁴

4 The facts in this subsection are not in dispute and subject to judicial notice. (See Doc. 354 ¶¶ 5–11; Doc. 355 ¶¶ 1–9.)

7. Coronavirus disease 2019 ("COVID-19") is an infectious disease caused by severe acute respiratory syndrome coronavirus 2 ("SARS-CoV-2"). This novel coronavirus has spread around the world since the end of 2019, resulting in a global pandemic.

8. The COVID-19 pandemic has sparked a public health emergency across the United States, including

in Texas. The first case of COVID-19 in the United States was confirmed on January 21, 2020. On January 30, 2020, the World Health Organization declared the COVID-19 outbreak a "Public Health Emergency of International Concern." Statement on the Second Meeting of the International Health Regulations (2005) Emergency Committee Regarding the Outbreak of Novel Coronavirus (2019-nCoV) , WHO (Jan. 30, 2020), https://www.who.int/news-room/detail/30-01-2020-statement-on-the-second-meeting-of-the-international-health-regulations-(2005)-emergency-committee-regarding-the-outbreak-of-novel-coronavirus-(2019-ncov). The next day, the U.S. Secretary of Health and Human Services declared that COVID-19 presented a public health emergency under section 319 of the Public Health Service Act, 42 U.S.C. § 247d. Proclamation, Proclamation on Declaring National Emergency Concerning the Novel Coronavirus Disease (COVID-19) Outbreak , White House (Mar. 13, 2020). On March 13, 2020, the President of the United States proclaimed that "the COVID-19 outbreak in the United States constitutes a national emergency, beginning March 1, 2020." Id. That same day, Texas Governor Greg Abbott announced that COVID-19 "poses an imminent threat of disaster" and declared "a state of disaster for all counties in Texas." (PTX 119.) Governor Abbott continued to extend the disaster declaration throughout trial and on August 8, 2020, issued yet another extension, stating that "[e]veryone must do their part to slow the spread of COVID-19 by wearing a mask, practicing social distancing, and washing your hands frequently and thoroughly." Press Release, Governor Abbott Extends State Disaster Declaration for COVID-19 , Office of the Tex. Governor (Aug. 8, 2020), https://gov.texas.gov/news/post/governor-abbott-extends-state-disaster-declaration-for-covid-19. On March 25, 2020, President Trump also declared that a "major disaster" exists in the State of Texas due to COVID-19. Statement & Release, President Donald J. Trump Approves Texas Disaster , White House (Mar. 25, 2020), https://www.whitehouse.gov/briefings-statements/president-donald-j-trump-approves-texas-disaster-declaration-6/.

9. Common symptoms of COVID-19 include, but are not limited to, fever, cough, fatigue, shortness of breath, and loss of smell and taste. While most cases result in mild symptoms, other cases may progress to acute respiratory distress syndrome ("ARDS"), multi-organ failure, septic shock, blood clots, other serious illnesses, and death.

10. COVID-19 is highly communicable and primarily spreads between people through close proximity or contact. It can also be transmitted by touching contaminated surfaces. COVID-19 is also particularly transmissible because of its long incubation period of up to two weeks, as well as its asymptomatic or presymptomatic presentation in many individuals, which allows infected individuals to unknowingly spread the virus

while exhibiting only mild or even no symptoms.

11. Individuals over the age of 65, individuals who are severely obese (body mass index greater than or equal to 40), and individuals with comorbid conditions—such as individuals who are immunocompromised, have chronic lung disease, moderate to severe asthma, serious heart conditions, or other underlying medical conditions—are at higher risk for serious illness or death if they contract COVID-19.

12. There is no vaccine or cure for COVID-19.

C. The Wallace Pack Unit

13. The Pack Unit is a Type-1 Geriatric prison in the Texas Department of Criminal Justice ("TDCJ") prison system. (PTX 1.) It is located in Grimes County, Texas. (Id. ) The facility has a maximum capacity of 1,478 people. (Id. )

14. As of June 29, 2020, 1,132 people are incarcerated at the Pack Unit. (PTX 7C at 23.)

15. As a Type-1 Geriatric unit, the Pack Unit primarily houses individuals who are elderly and/or have health problems. Approximately 800 of those inmates are over the age of 65. (Tr. 3-52:25–53:6 (Herrera).)

16. The Pack Unit consists of 20 dorms inside the perimeter fence and a trusty camp outside the perimeter. There are 16 dorms in the main building and 4 dorms in the expansion dorm. (PTX 2.)

17. Each dorm in the main building can house approximately 54 inmates, except for two wheelchair dorms, which can house 30 inmates each. Two of the dorms in the expansion dorm have 48 beds each; the other two dorms in the expansion dorm have 93 beds each. Each of the trusty camp dorms can house up to 107 inmates. (PTX 2.)

18. When trial began, the Pack Unit had 49 wheelchair-bound inmates and 87 inmates who used walkers. (PTX 337 at 2–3.)

19. Most Pack Unit inmates live in small cubicles in a dormitory setting. Each cubicle is made of a waist-high wall; each cubicle contains a bunk. The cubicles are set next to each other, such that each cubicle shares walls with adjacent cubicles. Most inmates are within six feet of another inmate when they are sleeping in their bunks. (Tr. 11-81:4–6 (Wilder), 2-91:14–92:13 (Rudloff).)

20. The Pack Unit has communal bathrooms. Communal showers are shared by multiple dorms. (DTX 14.)

D. Credibility of Defendants

21. At the outset, the Court is concerned about the credibility of Defendants’ representations and experts. These concerns are as follows.

22. Before, during, and after trial, counsel for Defendants provided status updates of the number of Pack Unit inmates who had tested positive or negative for COVID-19 or who had recovered.

a. The reliability of these numbers, however, is undermined by Defendants’ repeated failure to disclose the underlying test results to Plaintiffs or the Court despite Plaintiffs’ repeated requests and the Court's repeated orders. Prior

to trial, the Court ordered Defendants to produce records of Pack Unit inmate COVID-19 related hospitalizations, contract tracing forms or logs, and COVID-19 test results for Pack Unit inmates and employees. (Minute Entry July 8, 2020) (granting Plaintiffs’ Motion to Compel Document Production and Supplemental Discovery Responses (Doc. 199)). By the second week of trial, Defendants still had not produced the underlying test results and failed to produce the underlying test results even after agreeing to do so. (Tr. 10-5:1–9 (Defendants’ counsel).)

b. Internal communications between TDCJ officials further raise concerns about the credibility of the COVID-19 numbers reported to the Court. For example, on May 29, 2020, Lorie Davis texted Billy Hirsch: "Get the list of names from pack that back the numbers we provided." (PTX 338.) Two days later, she stated: "Strike team numbers were wrong for pendings on pack." (Id. ) Another text message exchange on June 15, 2020 between Ms. Davis and Oscar Mendoza demonstrates that TDCJ officials did not have a ready or reliable system for tracking or determining the number of inmates who had been hospitalized due to COVID-19. (PTX 344.) Following a lengthy exchange regarding the number of hospitalizations, Mr. Mendoza texted, "Lorie .... I need [an] absolute number ... Please .... what's the final number ... [I'm] going with 17 Lorie. Your number." (Id. at 3–4.) The Court understands that Defendants were working under difficult and time-pressed circumstances, but at no point during trial did Defendants represent to the Court uncertainty about these numbers.

c. The credibility of the reported number of inmates who tested positive for COVID-19 and who had recovered is further undermined by the way TDCJ conducted mass testing in Pack Unit. As will be discussed in greater detail infra , only those inmates who had previously tested negative for COVID-19 were tested in subsequent rounds of testing. Inmates who had previously tested positive were not retested, even if they remained symptomatic.

d. Similarly, the credibility of the reported number of inmates who tested positive for COVID-19 and who had recovered is undermined by how TDCJ interpreted the test results. First, TDCJ appears to have adopted an opaque system for determining which inmates were recovered. On July 14, 2020, Ms. Davis texted Mr. Mendoza that there were "13 active positive on Pack today," but that "[a]s I understood you said Robert said 14 this morning." (PTX 337 at 3.) She reconciled the discrepancy as follows: "Offender Mayfield, a previous recovered offender, returned to Pack last night from HG, who tested positive. We have been counting them only once as positive and then stay recovered. [S]o it is 13 this morning too. That is part of reconciling with Robert. Which we are doing as we speak." (Id. (emphasis added).) The definition of "recovery" used by TDCJ in practice, however, conflicts with how TDCJ officially defines recovered inmates, according to Dr. Stephanie Zepeda, one of Defendants’

experts who helped develop TDCJ's response to COVID-19. Dr. Zepeda testified that TDCJ defines recovered as "[n]o longer symptomatic. They've been released from medical isolation." (Tr. 12-136:1–9 (Zepeda).) Moreover, internal communications suggest that, even if a Pack Unit inmate tests positive, they are not counted as positive in the reported numbers unless they are physically at Pack Unit. On July 15, 2020, Ms. Davis texted Mr. Hirsch asking, "What is pack active positive this morning." (PTX 338 at 1.) Mr. Hirsch responded, "22 when I reconciled yesterday afternoon. JC doesn't think it's changed. Jessica is communicating with Mary to check movement overnight that may have changed it this morning." (Id. ). He subsequently texted: "There are now 21 active positives at Pack. O/f Hargrow ... went to a FWH last night, and is now ert to HG." (Id. ). Without the underlying test results, the Court is concerned about how Defendants arrived at the COVID-19 counts they reported to the Court and whether they are accurate.

23. Defendants’ credibility is further undermined by actions and modifications to TDCJ's practices that were made right before, and in apparent response to, hearings before this Court or the trial itself.

a. As will be discussed in greater detail infra , while TDCJ officials testified that they are conducting mass testing on a weekly basis at Pack Unit, that plan was not devised until two weeks before trial. (Tr. 7-53:13–18 (Davis); Tr. 1-50:1–10, 1-51:8–10 (Collier)). Indeed, by July 17, 2020, three days after trial commenced, the plan was not yet finalized. (PTX 335 at 1 (text from Dr. Linthicum dated July 17, 2020, stating that "there is not a written plan other than the paper I faxed to you on Pack. We were supposed to meet today to finalize the plan at 1:00PM").) TDCJ officials declined to document the long-term testing plan in writing. (Tr. 1-51:11–52:25 (Collier).)

b. TDCJ's COVID-19 response policy was established on March 20, 2020 and was most recently revised on June 11, 2020. (PTX 115.) However, TDCJ did not begin discussions to create compliance teams to oversee and ensure that TDCJ was following its policy until June 24, 2020, two weeks before trial. (Tr. 10-90:4–10 (Collier).) Moreover, while a compliance team visited Pack Unit, TDCJ did not produce any report or documentation of the compliance team's findings. (Tr. 10-90:22–91:13 (Collier).)

c. Mr. Collier sent Mr. Monroe, the regional director, to conduct a site visit at Pack Unit on July 15 or 16, 2020, based on testimony from inmate witnesses at trial. (Tr. 10-97:17–98:11, 10-160:12–14 (Collier)). Mr. Monroe was removed from Defendants’ trial witness list several days later on July 20, 2020, and therefore Mr. Monroe never testified to his observations during this site visit. (Tr. 10-160:12–19 (Collier).)

d. In a text conversation between Lorie Davis and Dr. Linthicum on May 19, 2020, the two were discussing whether to move positive inmates at Pack into the SHU. When Ms. Davis said that "OM [presumably, Oscar Mendoza] hasn't been letting me move anybody

much off pack," Dr. Linthicum responded, "We can ask him. I think it would look more favorable in the Court's eyes to move them." (PTX 337 at 5 (emphasis added).) The Court held a status conference three days later on May 22, 2020. These statements undermine the Court's confidence that TDCJ officials were following a deliberate strategy to mitigate the spread of COVID-19, rather than taking ad hoc steps in response to litigation proceedings.

e. Shortly before trial, TDCJ placed red tape on the floors in Pack Unit to demarcate social distance spacing, a measure which Defendants acknowledge could easily have been implemented months prior, in the early stages of the pandemic. (Tr. 4-108:2–109:22). A photograph revealed in post-trial evidence production dated July 12 further raises the possibility that the tape had subsequently been removed. (PTX 327.)

24. Defendants have misrepresented certain facts. In a status update to this Court, TDCJ claimed to have moved positive inmates to Dormitory 2 in an effort to further social distancing. (Doc. 126-1 at 10.) When at trial an inmate testified that he had been housed with positive inmates in Dormitory 2 after he tested negative for COVID-19, Mr. Herrera denied that TDCJ moved positive inmates into Mr. Beal's dorm. (Tr. 4-183:15–184:12 (Herrera).) Additionally, on May 11, 2020, TDCJ's website announced that Pack Unit had 10 positive tests. (PTX 65 at 3.) Yet, TDCJ reported to the Fifth Circuit in a letter that 22 inmates at Pack Unit had tested positive. (PTX 64 at 1.)

25. Finally, Defendants’ experts Ms. Follenweider and Dr. Beard each relied on a single, pre-arranged visit to the Pack Unit. Their credibility as neutral and accurate observers of conditions on the ground is thereby reduced. (Tr. 13-33:11–24 (Follenweider); Tr. 14-107:16–21 (Beard).) These visits, which Defendants expressly stated they needed to "plan" for, do not have the same indicia of reliability as testimony by TDCJ inmates who observed conditions in the Pack Unit on a daily basis. (PTX 343 at 1.) Even Defendants’ other expert, Dr. Zawitz, emphasized the importance of unstaged visits for an expert to be able to form a credible and reliable opinion about the conditions and public health practices at the Pack Unit. (Tr. 9-29:16–30:13 (Zawitz).)

E. COVID-19 Spread in the Pack Unit

26. On March 30, 2020, when this suit was filed, there were no known cases of COVID-19 in the Pack Unit. (See generally Doc. 1.)

27. Leonard Clerkly was the first Pack Unit inmate who tested positive for COVID-19. He died on April 11, 2020. His positive COVID-19 test was confirmed via autopsy on April 13. The next day, April 14, TDCJ announced that Mr. Clerkly's death was due to "viral pneumonia due to COVID-19." (PTX 41 at 1.)

28. The second positive case of COVID-19 in the Pack Unit was discovered on April 15, 2020. (PTX 42 at 3.)

29. On May 9, 2020, TDCJ announced 8 positive test results at the Pack

Unit. (PTX 63 at 3.) On May 11, 2020, TDCJ announced on its website that there were 10 positive tests. (PTX 65 at 3.) By May 16, 2020, the number of positive tests reported by TDCJ had jumped to 32. (PTX 74 at 3.) And by May 27, 2020, TDCJ reported that 5 Pack Unit inmates who had tested positive for COVID-19 had died, 12 were hospitalized, and 191 others had tested positive. (PTX 76 at 6–7.)

30. As of July 24, 2020, over 497 Pack Unit inmates had tested positive for COVID-19, 74 inmates had been hospitalized, and 19 people had died. (Doc. 356-18 at 22 (collecting data from various admitted trial exhibits).) Since trial, at least one other COVID-19 positive inmate at the Pack Unit has died, bringing the total number of deaths to 20. (Tr. 13-42:5–7 (Follenweider).) By contrast, there were 11 natural deaths at the Pack Unit in all of 2019. (PTX 335 at 2.) In total, at least 505 Pack Unit inmates have tested positive. (Tr. 13-42:7–11 (Follenweider); Tr. 12-132:19–24 (Zepeda).)

31. The risk to Pack Unit inmates increases as the number of COVID-19 cases in Texas climbs, because employees and staff come in and out of the prison. Mr. Collier recognized this increased risk when he testified that "we have a bigger threat of COVID now than we did in March, a significantly higher threat because more people have COVID. And that risk of our employees being in the community around more people with it I think is higher than it was even a few months ago." (Tr. 10-7:16–21 (Collier).)

32. Post-trial, the Court requested an affidavit from Defendants summarizing the history of strike team testing at the Pack Unit. Defendants provided the following information to the Court (Docs. 379, 381):

Date	Total Tests	# Negative	# Positive	Invalid / Not Returned
May 12-14	1,179	1,030	144	5
June 23-25	878	690	172	16
July 9-10	697	676	15	5
July 21-22	677	665	2	10
July 28-30	666	636	13	17
Aug 5-6	642	636	4	2
Aug 10-12	638	635	1	2
Aug 18-19	633	Not yet reported	Not yet reported	Not yet reported

As the above chart demonstrates, during the August 10-12 round of testing, TDCJ reported one positive test among inmates. By August 31, more individuals had tested positive, as the TDCJ website reported 14 active cases of COVID-19. As of September 28, the TDCJ website listed 4 active cases among inmates at the Pack Unit indicating that COVID-19 has not been fully contained.⁵

33. Defendants also summarized the history of testing of employees at

5 Texas Department of Criminal Justice, COVID-19 Update (Sept. 28, 2020), https://txdps.maps.arcgis.com/apps/opsdashboard/index.html#/dce4d7da662945178ad5fbf3981fa35c.

the Pack Unit as follows (Docs. 379, 381):

Date	Total Tests	# Negative	# Positive	Invalid / Not Returned
May 12-14	288	30	2	256
June 3-6	262	249	12	1
June 22-25	271	258	13	0
July 9-10	245	241	3	1
July 21-22	250	242	6	2
July 28-30	246	238	1	7
Aug. 2-5	248	243	1	4
Aug. 10-12	248	240	1	7
Aug. 18-19	250	Not reported	Not reported	Not reported

As of September 28, the TDCJ website listed 62 employees who have been infected and "recovered," and 1 active employee case.⁶

6 Id.

F. Initial Measures Taken by TDCJ

34. TDCJ Executive Director Bryan Collier first began discussing TDCJ's response to COVID-19 with Dr. Lanette Linthicum, Director of the Health Services Division for TDCJ, and Ms. Lorie Davis, Director of the Correctional Institutions Division of TDCJ, at the end of February 2020. (Tr. 9-185:9–16 (Collier).)

35. On March 4, 2020, Mr. Collier and Dr. Linthicum first met with representatives from the University of Texas Medical Branch (UTMB) and Texas Tech Health Sciences Center. (Tr. 9-188:17–189:16 (Collier).)

36. A follow-up meeting with the UTMB and Texas Tech representatives took place on March 11, 2020. (Tr. 9-189:14–16 (Collier).)

37. On March 13, 2020, the Office of the Governor granted TDCJ's request to suspend all in-person inmate visitation at all TDCJ units. (Tr. 9-195:5–17 (Collier).)

38. On March 16, 2020, TDCJ activated the "Command Center" at the TDCJ Administrative Headquarters Building in Huntsville and began having daily conference calls to discuss TDCJ's COVID-19 response. (DTX 38 at 1; Tr. 9-227:19–21 (Collier).)

39. On March 20, 2020, the Office of the Governor granted TDCJ's request for a statutory waiver of all inmate medical copays. (PTX 21 at 1.)

40. On March 24, 2020, TDCJ began manufacturing masks for its staff and inmate population and distributed them first to individuals who were 65 years or older. (Tr. 9-202:20–25, 9-203:18–24 (Collier).)

G. The CMHC B-14.52 Policy

41. The Correctional Managed Health Care Committee ("CMHCC") had its first meeting to discuss planning and response to COVID-19 at TDCJ on February 26, 2020. CMHCC is a statutorily created body responsible for developing the inmate healthcare plan in TDCJ. Its membership is comprised of representatives from TDCJ, Texas Tech, and UTMB, as well as several governor-appointed

positions. (Tr. 12-56:19–23 (Zepeda).) It is a separate entity from TDCJ. Dr. Lannette Linthicum, Director of the Health Services Division for TDCJ, is TDCJ's representative on CMHCC. (Tr. 12-55:4–8 (Zepeda).)

42. The Correctional Managed Health Care ("CMHC") B-14.52 Policy ("Policy B-14.52" or "B-14.52") is CMHCC's policy on how to manage COVID-19 in TDCJ facilities. (PTX 108 at 1.) It was drafted by a subcommittee of the Joint Infection Control Committee ("JICC")⁷ and approved by the three CMHCC joint medical directors: Dr. Owen Murray for UTMB, Dr. Denise DeShields for Texas Tech, and Dr. Lannette Linthicum for TDCJ. (Tr. 12-55:4–8 (Zepeda).)

43. Although B-14.52 was designed to manage COVID-19 in TDCJ facilities, CMHCC did not seek any input from Ms. Davis nor apparently from any other TDCJ official in designing the policy. Ms. Davis testified that she was not at all involved in drafting B-14.52. (Tr. 7-41:3–25 (Davis).) She testified that she did not provide any information to CMHCC regarding TDCJ's individual facilities’ physical space, staffing population, inmate population, or operations or other resources. (Tr. 7-42:10–43:9 (Davis).) Mr. Sullivan testified that he was not authorized to change B-14.52 and that his job was just to follow the policy. (Tr. 6-230:23–231:14 (Sullivan).) Mr. Herrera also testified that he did not have any authority to make any changes to B-14.52. (Tr. 3-95:15–18 (Herrera).)

44. Dr. Stephanie Zepeda testified about her involvement in developing B-14.52 as a member of JICC. (Tr. 12-53:14–19 (Zepeda).) Dr. Zepeda admitted, however, that she had not been to the Pack Unit in many years, and she did not have personal knowledge of how B-14.52 was implemented at the Pack Unit or any other TDCJ unit. (Tr. 12-118:15–119:12 (Zepeda).)

45. The first version of Policy B-14.52 was approved and adopted by the joint medical directors on March 20, 2020. (PTX 108 at 1.)

46. Policy B-14.52 applies to all 104 TDCJ institutions and healthcare facilities, including the Pack Unit. (Tr. 1-68:24–69:18 (Collier).) There are no COVID-related policies or supplements that are specific to the Pack Unit. (Tr. 1-69:14–21 (Collier).) Mr. Mendoza attested in his video deposition that "[t]here would be no policies specifically developed for the Pack Unit. The policies that are being developed ... are mitigating measures for TDCJ and all of its facilities." (Tr. 7-30:2–5 (Mendoza).)

47. There have been six updates made to Policy B-14.52 since the first version was adopted on March 20, 2020. (DTX 2–7.)

48. On March 23, 2020, three days after Policy B-14.52 was first adopted, the Centers for Disease Control (CDC) posted its Interim Guidance on Management of Coronavirus Disease 2019 (COVID-19)

7 The Joint Infection Control Committee is a subcommittee of CMHCC responsible for overseeing and coordinating statewide healthcare services. (Tr. 12-54:3–6 (Zepeda).)

in Correctional and Detention Facilities. (PTX 88.)

49. The second version of Policy B-14.52 went into effect on March 27, 2020. This second version incorporated recommendations contained in the CDC's interim guidance on correctional and detention facilities. (DTX 2.)

50. Policy B-14.52 and all its revisions were disseminated to the wardens by Lorie Davis, Director of the Correctional Institutions Division. The wardens of the prisons are responsible for making sure that the measures in Policy B-14.52 are implemented in their prisons. (Tr. 7-10:17–23 (Davis).)

H. Implementation of Preventative Measures at the Pack Unit

i. Social Distancing in Common Spaces

51. Social distancing is the practice of maintaining a sufficient distance from another person to lower the risk of contracting a contagious disease like COVID-19. (Tr. 4-243:1–5 (Young).) Social distancing of six feet is an important measure in preventing the spread of COVID-19 at Pack Unit. (Tr. 4-243:6–18 (Young), Tr. 9-95:22–96:3 (Zawitz).)

52. Although Policy B-14.52 requires use of cloth face masks at all times, the wearing of masks does not replace the need for social distancing under the policy. (PTX 115 at 5.)

53. According to the Pack Unit social distancing plan, created by Warden Herrera, some social distancing measures were put into place at Pack Unit before the unit was placed on precautionary lockdown. Dayroom access was limited, based on a rotating schedule, and inmates would be kept six feet apart at the pill window. (DTX 35 at 1–2.)

54. Although Warden Herrera's social distancing plan stated that one dorm would be fed at a time in the dining hall, allowing each inmate to have his own table, (DTX 35 at 1), undisputed testimony stated that two dorms were fed at a time, necessitating at least two inmates to be seated at each table, (Tr. 2-40:17–41:10 (King), Tr. 3-33:4–12 (Reynolds), Tr. 2-179:10–16 (Beal), Tr. 3-249:5–10 (Herrera).) With two inmates seated per table, there was no way for inmates to maintain six feet of distance from each other. (Tr. 2-67:8–10 (King), Tr. 2-179:10–16 (Beal)), (Tr. 11-160:20–163:13 (Wilder).) One inmate, Mr. Pennington, testified that, when feeding inmates started taking too long, TDCJ started sitting three or four inmates per table again. (Tr. 2-256:13–17 (Pennington).)

55. After Mr. Clerkly tested positive for COVID-19, the Pack Unit went on precautionary lockdown. (Tr. 3-248:1–7 (Herrera).) During precautionary lockdown, inmates are primarily confined to their dorms. Common areas like the dining hall, recreation yard, law library, education room, and the dayrooms are no longer in use. Inmates receive their meals and their medication in their dorms. Inmates still leave their dorms to shower if their dorm does not have adjoining showers, and if they need to go to the medical wing. (Tr. 3-248.12-249:25 (Herrera).)

56. Inmates living in dorms without adjoining showers continue to shower in showers that are shared by multiple dorms. (Tr. 1-157:6–22 (Valentine), Tr. 2-202:2–11 (Butaud).) A full dorm goes down to shower together at the same time. (Tr. 1-157:14–16 (Valentine), Tr. 2-245:10–15 (Pennington).) With an entire dorm at the showers at once, there is no possibility of social distancing, because the showers are right next to each other. (Tr. 1-157:6–22 (Valentine), 2-202:12–15 (Butaud).) This is especially a problem for handicapped inmates, who have to wait on handicap benches until a handicap shower becomes free for use. (Tr. 1-157:14–18 (Valentine).) The benches become full, so individuals are sitting right next to each other without any protection. (Tr. 1-157:14–18 (Valentine).) Mr. Pennington testified that, although the showers are only to be used by one dorm at a time, another dorm will begin showering while he is still in the showers. (Tr. 2-245:23–247:2 (Pennington).)

57. Warden Herrera testified that the policy at Pack Unit is to keep inmates six feet apart while they are in the hallways. (Tr. 4-17–22, 4-56:1–4 (Herrera)). Officers sometimes enforce social distancing in the hallways, but not consistently. (Tr. 2-179:22–180:2 (Beal), Tr. 3-32:7–9 (Reynolds).) The Pack Unit Social Distance Plan does not include any policies on social distancing in the hallways. (DTX 35.)

58. Around the beginning of July, the Pack Unit placed for the first time red tape on the floor marking six feet of distance in the hallways and on the floor leading to the shower area. (Tr. 4-108:2–17 (Herrera).)

ii. Social Distancing Within the Dorms

59. Multiple inmates testified that they cannot socially distance while they are in their cubicles. (Tr. 1-157:23–158:8 (Valentine); Tr. 2.1-17:23–18:5 (King).) If inmates occupy neighboring cubicles, they are only a few feet apart at most, and may even be within a foot of each other, depending on where the inmates are in their cubicles. (Tr. 1-158:2–8 (Valentine); Tr. 2-18:3–12 (King); Tr. 2-91:14–22 (Jones); Tr. 2.2-201:19–21 (Butaud).) The cubicle walls are 43 inches high, (Tr. 11-81:4–6 (Wilder)), and do not provide a barrier between inmates that would protect against COVID-19 transmission when they are either seated or standing. (Tr. 1-158:2–8 (Valentine); Tr. 2-18:18–20 (King); Tr. 2.1-92:6–13 (Jones); Tr. 2-200:23–201:5 (Butaud).)

60. No one at TDCJ has ever instructed Pack Unit inmates to sleep head-to-foot in order to increase social distancing in the dorms. (Tr. 1-105:6–14 (Collier); Tr. 1-158:9–12 (Valentine); Tr. 2-21:16–20 (King).) Mr. Collier testified that he had a discussion with Ms. Davis in May about sleeping head-to-foot, and Ms. Davis and her staff "had significant concerns" and felt it would "create potential issues ... that may outweigh the benefits," such as compliance issues. (Tr. 10-50:9––51:8 (Collier).) Mr. King testified that he felt uncomfortable sleeping with his head facing the opening of his cubicle. (Tr. 2-51:15–52: 12 (King).) Warden Wilder cited similar

concerns among inmates as a reason for not advising inmates to sleep head-to-foot. (Tr. 11-80:20–81:3 (Wilder).)

61. Multiple inmates testified that there were empty cubicles in their dorms during the pandemic, and yet, no TDCJ staff members ever rearranged inmates within their dorms to spread them out and facilitate social distancing. (Tr. 1-160:2–11 (Valentine); Tr. 2-201:6–14 (Butaud).)

62. Two dorms in E-wing, 17 Dorm and 19 Dorm, were under construction when the pandemic began, and so, did not hold any inmates. (Tr. 3-64:1–5 (Herrera).) Construction on the dorms was completed April 5, 2020. (Tr. 3-64:14–15 (Herrera).) Together, 17 Dorm and 19 Dorm had a total capacity of 167 inmates. (PTX 2 at 3.)

63. TDCJ held the dorms empty while it considered moving inmates from a geriatric dorm in Estelle Unit into the Pack Unit dorms. (Tr. 9-193:7–25 (Collier).) The plan to move inmates from Estelle Unit into the E-wing dorms was abandoned at some point in mid-April. (Tr. 7-16:25–17:25 (Mendoza).) Instead, TDCJ decided to "allow Warden Herrera" to use both dorms for Pack Unit inmates. (Tr. 7-17:9–12 (Mendoza).) No prisoners were moved into 17 Dorm or 19 Dorm until May 4, 2020. (Tr. 3-70:12–15 (Herrera).)

64. The education building was repurposed in mid-to-late April to house inmates, in order to facilitate separation of inmates with pending or positive test results from those with negative test results. (Tr. 3-219:17–220:8 (Herrera); Tr. 11-56:18–21 (Wilder).)

65. Non-dorm spaces within Pack Unit other than the education building have not been used to house inmates. TDCJ notes that spaces like the gymnasium are not air-conditioned, so inmates who are class members in the air-conditioning settlement cannot be housed there. (Tr. 4-58:4–10 (Herrera); Tr. 11-67:23–3 (Wilder).) However, there have been no plans or discussions to consider using the gymnasium for inmate housing after summer ends, when TDCJ has no obligation to air-condition any housing units. (Tr. 1.1-82:2–15 (Collier).)

66. Mr. Collier testified that he has never had discussions with anyone about potentially using early release of inmates as a method of increasing social distancing within Pack Unit. (Tr. 1-115:16–116:11 (Collier).)

67. Mr. Collier testified that TDCJ has not delayed release of individuals scheduled for release on parole, unless the individual is on medical restriction or isolation. (Tr. 1-115:22–116:11; Tr. 10-61:10–17, Tr. 10-194:21–23 (Collier).) However, one inmate testified that his release has been delayed because no one has approved his intended housing. (Tr. 3-15:13–16, Tr. 3-29:7–10 (Reynolds).) It is unclear why his approval has been delayed, and whether the delay is caused by an employee of TDCJ or an employee of the Texas Board for Pardons and Paroles, a separate agency. Mr. Reynolds knows of other Pack inmates who are in similar positions on the parole

issue. (Tr. 3-24:13–24 (Reynolds).)

iii. Hand Hygiene

68. Before the pandemic, inmates received five small bars of soap per week. Inmates continue to receive an allotment of five bars of soap a week. (Tr. 11-60:15 (Wilder).) However, they are given access to additional soap as necessary. (Tr. 1-53:22–24 (Collier); Tr. 2.2-255:13–19 (Pennington); Tr. 11-61:6–10 (Wilder).) Currently, boxes of soap are available to inmates without restriction at officers’ podiums in the hallway and at the laundry exchange, near the showers. (Tr. 2-254:15–255:15 (Pennington); Tr. 11-61:6–10 (Wilder).) Inmates have testified that they have previously attempted to obtain extra soap when the boxes were empty and had been denied by officers. (Tr. 1-166:15–167:4 (Valentine); Tr. 2-136:6–14 (Jones).) Other inmates testified that when the boxes of soap are empty, officers will bring more soap within a few hours. (Tr. 2-49:5–7 (King).)

69. Prior to the pandemic, inmates received one roll of toilet paper per week. Inmates continue to receive one roll once a week, on Fridays. (Tr. 11-61:20–24 (Wilder); Tr. 1-178:25–179:7 (Valentine).) However, they may now ask officers for an extra roll of toilet paper when they run out. (Tr. 11-62:10–20 (Wilder).) But inmates have testified that they have been consistently denied additional toilet paper when they ask for more. (Tr. 1-179:16–180:16, 181:18–22 (Valentine); Tr. 2-101:4–15 (Jones); Tr. 2-203:22–204:7 (Butaud).) These inmates were told by officers that they did not have extra toilet paper, and that they would have to wait until their usual allotment on Friday. (Tr. 1-180:4–16 (Valentine); Tr. 2-101:9–11 (Jones); Tr. 2-204:1–10 (Butaud).) Mr. Jones testified that his dorm has never been notified by TDCJ that they may receive additional toilet paper upon request. (Tr. 2-101:12–15 (Jones).)

70. TDCJ has never provided tissues to the inmates. (Tr. 1-179:6–7 (Valentine); Tr. 2-101:1–3 (Jones); Tr. 2-203:20–21 (Butaud).)

71. Without access to tissues or enough toilet paper, inmates have been sneezing and coughing into their hands. (Tr. 1-179:6–15 (Valentine).)

72. Inmates have one face towel, which they can exchange every day. (Tr. 3-186:5–7 (Herrera).)

73. Multiple dorms had broken sinks for a significant portion of this pandemic. Mr. Valentine's current dorm, C Wing 11 Dorm, had at least one broken sink for about a month. (Tr. 1-167:11–16, 168:4–7 (Valentine).) The entire time that Mr. Valentine was in 14 Dorm, only five of the nine sinks worked; a fifth sink broke right before he was transferred. (Tr. 1.2-168:16–21 (Valentine).) 12 Dorm, only four out of the nine sinks were working until about a week-and-a-half before trial started. (Tr. 2-204:11–205:1 (Butaud).) The same dorm still has no hot water and only one toilet for the entire dorm of 54 people. (Tr. 2-204:16–19 (Butaud).)

74. At trial, TDCJ officials represented for the first time that they are in the process of installing temporary

handwashing stations around Pack Unit. (Tr. 3-186:8–10 (Herrera).) Warden Herrera testified that Mr. Mendoza showed him a picture of the temporary handwashing stations for the first time a few days before trial began. (Tr. 3-221:15–18 (Herrera).) Warden Herrera agreed that they would be useful for Pack Unit. (Tr. 3-222:4–6 (Herrera).) The temporary handwashing stations arrived at Pack Unit a few days into trial and were placed in the hallways and near the gates at Pack Unit while trial was underway. (Tr. 3-222:17–223:1 (Herrera).) Some, but not all units received temporary handwashing stations. (Tr. 10-66:16–24 (Collier).)

75. TDCJ has never provided hand sanitizer to Pack Unit inmates. (Tr. 1-64:4–5 (Collier).)

76. Providing hand sanitizer in addition to soap and water, particularly where hand washing at a sink is not readily available, is an important measure in stopping the spread of COVID-19. (Tr. 4-227:21–229:15 (Young); Tr. 9-95:9–17 (Zawitz).) Hand sanitizer increases ready access to cleanliness, and medical studies and journal articles show that hand sanitizer is effective at killing COVID-19. (Tr. 4-229:8–15, 230:10–231:14 (Young).)

77. Access to hand sanitizer is particularly critical to mobility-impaired inmates in wheelchairs who have difficulty keeping their hands clean. In traveling to and from the sink to wash their hands with soap and water, wheelchair-bound inmates must touch the tires of their wheelchairs, which may track dirt, urine, fecal matter, or germs from the floor. (Tr. 2-95:4–96:16 (Jones); Tr. 2-193:15–22 (Butaud) ("[I]f I wash my hands, I've got to turn around and roll myself with the wheelchair away from the sink.... So they get dirty again."); Tr. 2-160:4–11 (Beal) (explaining that he was unable to return to his cubicle after washing his hands with his hands still clean because he had to touch the wheels of his wheelchair); Tr. 2-193:15–22 (Dove).) TDCJ did not perform an analysis to determine whether hand sanitizer could be provided to wheelchair-bound inmates at Pack Unit. (Tr. 1-75:21–76:16 (Collier).)

78. Access to hand sanitizer is also important for Pack Unit inmates whose medical conditions impair their mobility. (Tr. 1-169:13–170:3 (Valentine) (explaining that, due to his stroke, it is difficult for him to walk to the sink to wash his hands in preparation for a meal or after coughing or sneezing into his hands); Tr. 2-247:11–17 (Pennington) (explaining that it is difficult for him to walk from his cubicle to the sink due to a large tumor and that the distance prevents him from cleaning his hands as frequently as he would like).)

79. Jails and prisons across the country have provided inmates with hand sanitizer to combat the spread of COVID-19. Examples of such facilities include Cook County Jail in Chicago, Illinois, where Defendants’ experts Dr. Zawitz and Ms. Follenweider work or worked, (Tr. 9-87:14–88:1 (Zawitz); Tr. 12-228:6–21 (Follenweider)), and the California prison system, which Defendants’ expert Dr. Beard used to run, (Tr. 14-187:6–20

(Beard).) California is one of approximately 30 states that are providing hand sanitizer to inmates. (Tr. 14-187:12–20 (Beard).) At TDCJ's Roach facility, inmates are entrusted with handling hand sanitizer, including rebottling hand sanitizer bought in bulk for TDCJ staff use. (Tr. 1-76:23–77:1 (Collier).)

80. As with these facilities, it is possible for Pack Unit to devise a plan to safely distribute hand sanitizer to inmates. (Tr. 14-188:6–12 (Beard); PTX 340 at 1–2 (Collier text message stating in a text that "[i]f we need to [provide hand sanitizer to inmates] we know we can figure it out so it is your call").)

iv. Cleaning and Sanitation

81. At the Pack Unit, inmates work as dorm janitors, also called service support inmates ("SSI" or "janitor"). Janitors are responsible for cleaning everything in their assigned dorm other than individual cubicles, including the adjoining bathrooms and dayrooms. (Tr. 2-23:1–24:1 (King).) Janitors work in pairs, in 12-hour shifts. (Tr. 2-88:13–17 (Jones).)

82. TDCJ did not increase the number of janitors at Pack Unit in response to COVID-19. (Tr. 3-137:19–24 (Herrera).)

83. At the beginning of each 12-hour shift, each pair of dorm janitors receives cleaning supplies from TDCJ staff members. (Tr. 2-24:19–24 (King).) The janitors did not receive any training from TDCJ. (Tr. 2-24:14–16 (King).) Each pair of janitors receives two ounces of a liquid disinfectant that is manufactured by TDCJ, called Double D. (Tr. 2-24:17–24 (King).) They also receive two ounces of a scouring powder, called Bippy, which is similar to Comet cleaner. (Id. ) The TDCJ staff member also pours about a quarter-cup of powder bleach into a standard mop bucket. (Id. ; Tr. 2-150:1–7 (Jones).) Mr. King testified that two dorms share a small, gallon-and-a-half plastic pump-up bleach sprayer, which TDCJ officials place some powder bleach in at the beginning of the shift and janitors fill with water. (Tr. 2-25:22–26:7 (King).) He testified that the bleach sprayer is empty halfway through the shift, and so to conserve the bleach solution janitors sometimes dilute it with water—even so, the solution is depleted by mid-afternoon. (Tr. 2-26:8–12 (King).)

84. These are the same cleaning supplies that were provided to janitors before COVID-19. (Tr. 2-87:8–19 (Jones).)

85. Prior to the preliminary injunction hearing in this case in April, each pair of janitors was only given one pair of gloves to share. (Tr. 2-66:6–14 (King).) Each janitor is now provided a pair of clean vinyl gloves at the beginning of the shift. (Tr. 2-59:16–28 (King).) Additional gloves are often not available, even upon request. (Tr. 2-89:18–22 (Jones).)

86. Multiple inmates who currently serve as inmate janitors testified as to their experiences as janitors. The janitors testified that the cleaning supplies provided at the beginning of a 12-hour shift only last until halfway through the shift. (Tr. 2-25:4–15 (King);

Tr. 2-87:20–25 (Jones).) Mr. Jones testified that janitors may run out of cleaning chemicals even more quickly depending on how many inmates have accidents in the shower or in the bathroom. (Tr. 2-88:1–8, Tr. 2-88:13–17 (Jones).) He also testified that, after half the dorm is mopped, the bleach in the mop bucket is already dirty and thus often must be thrown out and certainly does not last all day. (Tr. 2-88:9–12 (Jones).) No new supplies are provided until the start of the next shift. (Tr. 2-25:16–18 (King); Tr. 2-88:18–89:25 (Jones) ("[W]hen it's out, it's out. Everything is out until the second shift.").) Mr. King testified that the amount of cleaning supplies has not increased during the pandemic, and as of July 14, the date of his testimony, he was still receiving the same amount of cleaning chemicals. (Tr. 2-24:11–21, 67:21–68:24 (King).) The inmates also testified that requests for more cleaning products by inmate janitors have been denied. (Tr. 2-26:13–27:4 (King) (testifying that he was told by Ms. Maxie, a TDCJ staff member who distributes the chemicals, that "we get what we're going to get"); Tr. 2-89:2–8 (Jones).) The Court finds the testimony of individual janitors with direct knowledge of their day-to-day activities more credible than statements by high-level TDCJ officials about the same.

87. While pump sprayers containing bleach solutions are available for inmates to use to clean their individual cubicles, multiple inmates testified that multiple dorms share each sprayer, such that the sprayers run out of solution halfway through the day. (Tr. 2-25:4–18 (King).)

88. While some inmate janitors have helped inmates spray down their cubicles, there are rules that forbid any inmate, including those working as janitors, from entering another inmate's cubicle. (Tr. 2-69:3–11 (King).)

89. Inmate janitors are assigned to clean the dorms in which they live. In wheelchair dorms, all inmates are in wheelchairs, so the inmate janitors are also in wheelchairs. (Tr. 2-83:1–5 (Jones).) Multiple janitors testified that, due to their physical and medical conditions, they had great difficulty cleaning the dorms. Marvin Jones, a wheelchair-bound janitor, testified that it is very difficult to mop using the TDCJ-issued mop and bucket while using a wheelchair. (Tr. 2-83:6–16 (Jones).) He described the difficulty as follows: "Being in a wheelchair, trying to sweep and mop, you have to maneuver the wheelchair and grab the wheels and maneuver, move forward, move backwards, traverse, all the while trying to guide the mop or the broom ... and also I'm trying to push the mop bucket, as well." (Tr. 2–83:8–13 (Jones).) Mr. Jones also testified that it is particularly difficult for him to effectively fulfill his cleaning responsibility due to a condition in his right hand that prevents him from being able to hold anything, including a mop. (Tr. 2-83:17–24 (Jones).) When Mr. Wilder was asked whether he was concerned about disabled inmates serving as janitors, Mr. Wilder responded that he was not because inmates "could put a broom against his neck and push

it with a wheelchair." (Tr. 11-239:13–19 (Wilder).)

90. Harold Dove is also a wheelchair-bound janitor, who is legally blind and is paralyzed on the right side of his body. (Tr. 2-85:18–22 (Jones); Tr. 2-188:17–25, 189:2–6 (Dove).) Mr. Dove testified that he is physically unable to work as a janitor because he is unable to mop or sweep. (Tr. 2-190:24–191:4 (Dove).) Both Mr. Dove and his wife wrote to the warden, major, captain, and sergeant in April 2020 and complained that Mr. Dove was not capable of working as a janitor. (Tr. 2-191:20–192:9, 192:10–24 (Dove).) Neither of them ever received a response. (Tr. 2-192:8–9, Tr. 2-192:16–20 (Dove).)

91. Pack Unit was on notice of Mr. Dove's condition at least as of April 20, 2020, when it received a grievance complaining that the inmate's dorm did not have adequate dorm janitors because Mr. Dove was legally blind and another dorm janitor was an above-the-knee amputee with only one good arm. (PTX 131 at 45.) Mr. Wilder signed off on a response to the grievance on April 23, 2020. (Tr. 11-229:3–9 (Wilder).) As of June 22, 2020, Mr. Dove was still assigned as a janitor. (PTX 7A at 11008; Tr. 2-194:16–20 (Dove) (testifying on July 14 that "[t]he last I looked ... I was assigned to 4 Dorm as a janitor; so if it's been changed, it's been since I've been in quarantine.").)

92. Warden Herrera and Warden Wilder testified that Mr. Jones and Mr. Dove were assigned to be janitors based on the medical restrictions listed in their health summaries; their health summaries did not list medical restrictions that would have prohibited assignment as janitors. (Tr. 3-139:17–19, 141:5–8 (Herrera); Tr. 11-231:9–15 (Wilder).) In choosing janitors among the wheelchair-bound inmates, no one ever went to personally speak with inmates to assess whether they were actually able to perform the job duties. (Tr. 3-144:6–10 (Herrera).) Despite a grievance filed by an inmate on April 23, 2020 raising this concern, the issue of a blind janitor was not addressed in Warden Wilder's response to the grievance. (Tr. 11-233:17–23 (Wilder).) Mr. Dove was eventually taken off his janitorial assignment. (Tr. 3-141:9–15 (Herrera); Tr. 11-232:10–13 (Wilder).) It is unclear when he was reassigned, although the reassignment occurred after mid-June. (Tr. 2-194:16–22 (Dove).)

93. Beginning less than two weeks before trial commenced, around July 4, Pack Unit began using an electrostatic sprayer to aid in disinfecting the facility. (Tr. 3-35:16–21, 40:18–41:1 (Reynolds); Tr. 3-186:15–187:1 (Herrera).) The sprayer sprays Vital Oxide, a disinfectant, in a mist. (Tr. 3-216:23–217:18 (Herrera).) A staff member has used the sprayer on at least two occasions to spray all parts of the Pack Unit with disinfectant. (Tr. 3-218:8–17, 219:13–16 (Herrera).) Mr. Collier testified that TDCJ tried to acquire Vital Oxide in late March or early April, but was unable to purchase it anywhere. (Tr. 10-51:18–25 (Collier).) However, Mr. Collier states there are no documents or emails documenting

such attempts. (Tr. 10-198:5–15 (Collier).)

94. Multiple inmates have testified that the shared showers, which are used by multiple dorms, are not cleaned between different dorms. (Tr. 1-164:17–165:6 (Valentine); Tr. 2-202:16–203:13 (Butaud); Tr. 2-160:12–15 (Beal); Tr. 2-247:3–5 (Pennington).) Inmates state that, upon entering the shower with their dorm, they find dirty razors, bloody bandages, dirty diapers, soap, towels, and discarded clothing, left behind by previous dorms. (Tr. 1-164:25–165:6 (Valentine); Tr. 2-202:16–203:13 (Butaud).) Mr. Butaud testified that he has seen the same bloody gauze in the shower for months. (Tr. 2-203:5 (Butaud).) Thomas Ray Pennington testified that he can see the shower area and the inmates being brought into the shower area while he is working in the laundry issue room. (Tr. 2-245:25–246:20, (Pennington)). Mr. Pennington testified that, most of the time, no one cleans the showers between dorms. (Tr. 2-246:16–20 (Pennington).)

95. As discussed infra , multiple inmates have testified that 16 dorms share the same laundry exchange but that the surfaces of the laundry exchange are not cleaned in between use.

v. Laundry Exchange

96. Inmates exchange their soiled clothing for clean clothing right before they enter the shower area through a laundry exchange window. (Tr. 1-162:4–24 (Valentine); Tr. 2-243:12–14 (Pennington).) Inmates work behind the window, passing out the clean clothes. (Tr. 1-162:6–13 (Valentine); Tr. 2-243:6–16 (Pennington).)

97. Inmates place their dirty clothing on a ledge and push the clothing through the window. (Tr. 1-162:4–163:7 (Valentine); Tr. 2-98:5–10 (Jones); Tr. 2-168:24–169:9 (Beal).) Inmates must touch the ledge with their hands when they push their clothes through the window. (Tr. 2-98:15–23 (Jones).) All 16 dorms in the main building use the same laundry window for clothing drop-off. (Tr. 2-100:11–18 (Jones).) Multiple inmates testified that they had never seen the window ledge being cleaned, even between dormitories using it. (Tr. 1-165:13–16 (Valentine)); (Tr. 2-99:7–13, 100:20–23 (Jones); Tr. 2-169:10–11 (Beal).)

98. The inmate workers handing out clean clothing wear masks, but the inmates entering the showers have already taken their masks off to be exchanged when they interact with the inmate workers behind the window. (Tr. 2-243:17–21 (Pennington).) The inmate workers place the clean clothing in each inmate's hands, but the workers are not wearing clean gloves. (Tr. 2-244:1–7 (Pennington).) Instead, the inmate workers all share "the same pair of long plastic gloves." (Tr. 2-244:8–11 (Pennington).) Those inmate workers who are not provided gloves will sometimes touch their bare hand on another inmate's bare hand while passing out clothing. (Tr. 2-244:12–15 (Pennington).) There is no opportunity for the inmate workers to wash their hands immediately after touching someone else's hands, although they have the opportunity at a

later point. (Tr. 2-244:16–20 (Pennington).)

vi. Masks and Personal Protective Equipment ("PPE")

99. Under Policy B-14.52, masks "should be worn at all times unless it restricts [sic] breathing or interferes with activities of daily living." (PTX 115 at 5.)

100. Initially, TDCJ provided only those inmates at Pack Unit who were 65 and older with masks. (Tr. 10-109:9–15 (Collier); PTX 299 at 1 (email dated April 13, 2020, stating: "Scratch giving the entire Pack Unit two (2) masks. Only the offenders that are 65 and older at pack will get a mask").) Inmates are provided cloth, not surgical, masks. (Tr. 1-208:6–9 (Valentine).)

101. Multiple inmates have observed officers and staff entering negative dormitories in full PPE that they had already worn into other dormitories. (Tr. 1-182:7–183:12 (Valentine); Tr. 2-33:23–34:10 (King); Tr. 2-103:14–24, 105:17–19 (Jones).)

102. Multiple inmates have testified that officers will frequently "shakedown" cubicles without wearing gloves or without changing their gloves between cubicles. (Tr. 1-184:5–12 (Valentine); Tr. 2-34:23–25 (King).) A shakedown is a procedure pursuant to which an officer enters an inmate's cubicle in search of contraband; the process involves touching an inmate's belongings, lifting the mattress and blankets, and inspecting the inside of an inmate's locker and desk. (Tr. 1-183:14–21 (Valentine).) Mr. King testified that he had never seen any guards change gloves between cubicles during shakedowns. (Tr. 2-35:2–5 (King).) Mr. Jones testified that he has seen officers leave a quarantine dorm and enter his dorm to conduct a shakedown without changing any PPE or gloves. (Tr. 2-103:19–104:6 (Jones); Tr. 2-206:14–25 (Butaud) (describing this as an "everyday occurrence").)

103. Multiple inmates have testified that officers will regularly pass out sack meals to inmates in their dorms with no gloves on. (Tr. 1-187:7–21 (Valentine); Tr. 2-206:14–25, 207:18–21 (Butaud).) Mr. Valentine testified that, when one inmate pointed out that the officers were not wearing gloves to hand out lunch, the officer responded: "If you don't like it, put it on paper." (Tr. 1-187:18–23 (Valentine).)

104. Multiple inmates have testified that they regularly observe officers in the unit who are not properly wearing face masks. (Tr. 1-187:7–17 (Valentine); Tr. 2-101:16–102:21 ("More than once a day.") (Jones); Tr. 2-206:1–25 (Butaud); Tr. 3-13:23–14:15 (Reynolds); Tr. 2-32:20–33:1 (King); Tr. 2-262:17–22 (Pennington).) Inmates have also observed officers not wearing masks and not social distancing at the same time. (Tr. 2-102:13–21 (Jones); Tr. 2-206:6–12 (Butaud); Tr. 3-14:1–3 (Reynolds).)

105. TDCJ plays a video educating inmates about COVID-19 and protective measures to combat the spread of the virus three times a day. That video depicts numerous instances of correctional officers not wearing masks or practicing

social distancing. (Tr. 10-100:19–101:20, Tr. 10-104:12–17 (Collier).)

106. One inmate listed 13 TDCJ officers, including Mr. Wilder and Mr. Herrera, who frequently do not practice social distancing or wear face masks. (Tr. 2-207:1–14 (Butaud); Tr. 2-167:5–25 (Beal) (naming five officers, including Mr. Wilder, who frequently did not wear their masks).) Multiple inmates testified that Lieutenant Brown frequently did not wear a mask inside and outside the dorms, (Tr. 1-185:7–15 (Valentine); Tr. 2-31:11–25 (King); Tr. 2-167:23–25 (Beal); Tr. 3-13:17–22 (Reynolds)), and in one instance when an inmate notified Lieutenant Brown that other officers were not wearing masks in the dorms, he was told to "shut up and go back to [his] bunk," (Tr. 3-13:10–16 (Reynolds)). Lieutenant Brown later tested positive for COVID-19. (Tr. 3-170:21–171:1 (Herrera).)

vii. Testing

107. Prior to mid-April, the only inmates at Pack Unit who received testing were those who showed symptoms of COVID-19 and were taken off-site to hospitals for testing. (Tr. 10-22:12–21 (Collier).) Until Mr. Clerkly's positive test on April 13, 2020, no Pack Unit inmates who had been tested at hospitals had tested positive.

108. After Mr. Clerkly died and subsequently tested positive for COVID-19 during his autopsy, TDCJ tested the 54 inmates who lived in Mr. Clerkly's dorm. (Tr. 10-24:4–17 (Collier).) All of the inmates in Mr. Clerkly's dorm tested negative. (Tr. 10-24:16–17 (Collier); Tr. 3-189:18–21 (Herrera).) No one else was tested by TDCJ based on previous contact with Mr. Clerkly before his death.

109. Approximately one month following Mr. Clerkly's death, on May 12, 2020, TDCJ conducted its first "strike team testing." Strike team testing consists of teams of TDCJ employees deployed to different units to test inmates and staff who have not previously tested positive. (Tr. 3-190:5–10 (Herrera); Tr. 10-26:6–14 (Collier); Tr. 10-24:18–20 (Collier).)

110. The strike teams administer a viral test manufactured by Curative Medical Inc. called the Curative-Korva test. (PTX 76 at 7.) This test is a type of real time polymerase chain reaction ("RT-PCR") that relies on detecting COVID-19 through oral, rather than nasal, swabs. (PTX 76 at 6; PTX 285 at 1–2.) The Food and Drug Administration ("FDA") granted use under an Emergency Use Authorization ("EUA"). (PTX 285 at 1.) The EUA, however, is limited to "patients with symptoms of COVID-19" and has not been approved for testing of asymptomatic individuals. (PTX 285 at 1, 3 ("Testing of nasal swabs and oral fluid specimens ... is limited to patients with symptoms of COVID-19."); Tr. 5-19:13–18 (Young)). The EUA also advises that negative results from specimens taken from oral swabbing "should be confirmed by testing of an alternative specimen type." (PTX 285 at 1; Tr. 5-16:18–23 (Young) ("[I]f you have a negative test with [the Curative-Korva] test ... you should retest with a different test.").)

111. The members of the strike team were trained on how to administer the Curative-Korva oral swab test, which is self-administered. (Tr. 10-27:15–23 (Collier); Doc. 379-1 at 3.)⁸ Strike team members observe while inmates cough into their sleeves three to five times to improve the quality of the oral test sample, swab the inside of their mouths, and place the swab into a collection tube. (PTX 76 at 7.) After the samples are collected, they are shipped overnight to California, where the Curative labs re located. (PTX 76 at 8); (Tr. 10-131:8–11 (Collier).)

112. In the first round of strike team testing, 1,179 inmates and 313 employees at Pack Unit were tested. (PTX 76 at 5.) Not all employees were tested. (Tr. 1-156:22–157:1 (Valentine).) It took 7 to 14 days for Pack Unit to receive test results for the inmates. (Tr. 10-131:1–4 (Collier).) As of May 26, 2020, TDCJ had received results from only 876 tests; of those results, 136 inmates tested positive with 307 tests pending, and 2 employees tested positive with 309 tests pending. (PTX 76 at 5.)

113. Such prolonged turnaround times for test results impede effective containment of the spread of COVID-19. This is because inmates who have not contracted the virus continue to cohabit and mingle with inmates who are positive pending the results, thereby increasing the possibility of transmission. (Tr. 5-15:18–16:23 (Young); Tr. 9-158:17–23 (Zawitz) (explaining that, when test results take 5 to 14 days, some of the people who test negative are necessarily exposed to those that tested positive); Tr. 12-267:5–15 (Follenweider) (testifying that the long turnaround period for test results is "a frustration" because it impedes the ability to effectively protect those who have not yet been exposed to the virus).) Dr. Young testified that the standard practice for an outbreak in a confined space is to test individuals with a rapid turnaround time for results. (Tr. 5-129:21–130:8 (Young).)

114. Curative Labs's website states that it can receive and return thousands of test results in 24 hours. (Tr. 10-132:22–133:2 (Collier).) However, Mr. Collier testified that he did not know if TDCJ staff members ever discussed with Curative Labs a 24–48-hour turnaround window for test results. (Tr. 10-133:24–134:9 (Collier).) Mr. Collier never instructed his staff to try to get results from Curative in any particular time frame, nor did he ever express that 7 to 14 days was too long a timeframe. (Tr. 10-134:18–136:4 (Collier).)

115. When TDCJ received the test results, they did not immediately retest those inmates who had been exposed to inmates who tested positive. (Tr. 11-260:16–21 (Wilder) ("There was no protocol to do that.").)

116. TDCJ did not have any plans to continue strike team testing following the first round of testing. (Tr. 7-32:10–35:25 (Mendoza).) As

8 Post-trial, the Court requested Defendants produce an affidavit outlining the history of mass testing and test results at the Pack Unit. (Doc. 368.) While not formally admitted into evidence, the Court relies on the affidavit as it would other fact filings in the case.

of June 4, 2020, TDCJ still did not have any plans, formal or informal, to conduct retesting at Pack Unit. (Tr. 1-49:23–25 (Collier).)

117. Six weeks after the first round of strike team testing, TDCJ conducted a second round of strike team testing at Pack Unit from June 23 to 25. (PTX 288 at 1); (Tr. 3-200:4–8 (Herrera).) Only those inmates who had previously tested negative were tested, for a total of 878 inmates tests. (Doc. 379-1 at 3.) In the second round of strike team testing, 172 new inmates tested positive for COVID-19. (Doc. 379-1 at 3.)

118. At some point shortly before trial, Ms. Davis and Dr. Linthicum devised a testing plan at Pack Unit referred to as the "long-term care testing" plan. (Tr. 7-53:12–18 (Davis); Tr. 1-50:1–10 (Collier).) The plan was authorized by Mr. Collier approximately two to three weeks before trial. (Tr. 1-51:8–10 (Collier).) However, it appears that by July 17, 2020, three days after trial commenced, the plan was not yet finalized. (PTX 335 at 1 (text from Dr. Linthicum dated July 17, 2020, stating that "there is not a written plan other than the paper I faxed to you on Pack. We were supposed to meet today to finalize the plan at 1:00PM").)

119. The long-term care testing plan is based on the CDC's guidance for mass testing of COVID-19 in nursing homes. (See PTX 98, 99; Tr. 7-53:12–18 (Davis).) Ms. Davis and Dr. Linthicum chose to apply the long-term care testing plan to four units—Pack, Duncan, Terrell, and Jester III—because the populations at those units were very similar to nursing home populations. (Tr. 7-53:19–24, 55:6–22 (Davis).) Ms. Davis testified that she viewed these four units as "long-term care facilities," a term used in the CDC's guidance on testing in nursing homes. (PTX 99 at 2; Tr. 7-58:4–8 (Davis).) The CDC's guidance on testing in nursing homes expressly applies to long-term care facilities. (PTX 99 at 2.)

120. The plan itself is to test the population, identify individuals who have negative results, and retest those with negative results, as well as individuals who were never tested. (Tr. 7-54:1–21 (Davis).) Retesting of individuals with negative results or who have never been tested continues every seven days until there are no more positive results returned. (Tr. 1-50:20–51:3 (Collier)); (Tr. 7-54:1–21 (Davis).) Once there are no more positive results coming from a unit, TDCJ will conduct two more tests, seven days apart, then create a schedule to continue testing employees at the unit at least every fourteen days. (Tr. 1-51:4–7 (Collier)); (Tr. 7-54:13–19 (Davis).) In implementing the long-term care testing plan, TDCJ will continue to use the same strike team and same Curative-Korva test that it has been using for its previous mass or strike team tests. (Tr. 7-54:1–7 (Davis).)

121. The long-term care testing plan is nowhere documented in writing. (Tr. 1-51:11–12 (Collier); Tr. 7-59:15–60:18 (Davis); PTX 335 at 1; PTX 337 at 2 (text message from Mr. Mendoza asking: "Are there any written plans ... to retest the 4 units. Pack. Duncan. And the two others." Ms. Davis responded:

"No sir. Nothing written.")). There are no plans to put it in writing. (Tr. 7-59:15–60:21 (Davis).)

122. Since TDCJ implemented the long-term care testing plan in early July, it has deviated from the seven-day testing cadence it set forth. For instance, the third round of strike team testing did not occur until July 9-10, 2020—more than two weeks after the previous round of strike team testing. (Doc. 379-1 at 3.) The fourth round of strike team testing did not occur until eleven days after that. (Doc. 379-1 at 3; Tr. 8-33:16–18.)

123. As of July 24, 2020, TDCJ receives results from the Curative-Korva test in about four-and-a-half-days. (Tr. 10-32:9–13 (Collier)). The results from one round of testing continue to be returned in batches, in a staggered manner. (Tr. 10-45:14–22 (Collier).)

124. As of July 23, 2020, Warden Herrera had not been notified about the long-term care testing plan. (Tr. 4-7:14–19 (Herrera).)

125. TDCJ has never had a written plan for its testing protocol. (Tr. 7-31:25–32:9 (Mendoza).)

viii. Quarantining and Isolating

126. Policy B-14.52 describes the protocol for isolating and quarantining inmates with confirmed or suspected COVID-19, or who were potentially exposed to COVID-19. (PTX 115 at 6–11; Tr. 3-197:17–22 (Herrera).)

127. Under Policy B-14.52, when inmates test positive for COVID-19, they are placed into medical isolation, in which they are housed in a dorm only with individuals who have also tested positive. (PTX 115 at 7; Tr. 3-194:10–14 (Herrera).) Symptomatic inmates are housed separately from asymptomatic inmates. (PTX 115 at 7.) Medical staff from UTMB monitor these inmates by taking temperatures and talking to inmates at least once a day. (PTX 115 at 7; Tr. 3-196:20–197:3 (Herrera).) Under the policy, inmates with symptoms should be observed at least twice per day. (PTX 115 at 7.)

128. Individuals suspected of having COVID-19 must also be placed in medical isolation; positive test results are not required for isolation. (PTX 115 at 18.)

129. Symptomatic inmates are kept in medical isolation until at least 72 hours after they no longer experience fever without the use of fever-reducing medication and improvement in their respiratory symptoms, and at least 14 days have passed since their symptoms first appeared. (PTX 115 at 7.) Asymptomatic inmates may be released from medical isolation 10 days after the collection date of the positive test as long as they have not subsequently developed symptoms. (PTX 115 at 7.)

130. Individuals in a dorm where an inmate has tested positive are placed in medical restriction, because of their possible exposure to COVID-19. (PTX 115 at 10; Tr. 3-195:9–13 (Herrera).) Inmates in medical restriction can be released 14 days after the last exposure of a case for everyone in the cohort. (PTX 115 at 10.) Those in medical restriction must be checked every day for fever and other symptoms.

(PTX 115 at 11; Tr. 3-197:4–8 (Herrera).)

131. A room in the education department is now being used as a quarantine dorm, for inmates who are brought back to Pack Unit from off-site hospitals. (Tr. 3-195:22–196:2 (Herrera).) All inmates who are brought back to Pack are subject to a 14-day quarantine before they are returned to the general population. (Tr. 3-196:2–3 (Herrera).) Individuals in quarantine are monitored for symptoms. (Tr. 3-196:7–11 (Herrera).)

132. Inmate testimony at trial shows that these isolation measures are not always followed at Pack Unit. Several inmates testified that inmates who had tested negative remained in the same dorm as those who tested positive for hours or days. Mr. King testified that two inmates in his dorm remained in the dorm for a month after testing positive. (Tr. 2-14:14–15:9 (King).) Mr. Jones testified that an inmate in his dorm was not removed for two or three days after he first began exhibiting symptoms and was tested. (Tr. 2-81:23–25 (Jones).) Mr. Jones also testified individuals who tested positive during mass testing were not removed from the dorm for over two hours after receiving their positive test results, and no one in the dorm was provided additional protective gear. (Tr. 2-82:1–14 (Jones).)

133. Inmates testified that after testing positive and being quarantined they are neither medically examined by a doctor or retested for COVID-19 before returning to negative dorms. (See, e.g. , Tr. 1-155:7–11 (Valentine) (testifying that he neither saw a doctor nor was retested for COVID-19 prior to being moved back to a non-segregated dorm); Tr. 2-200:11–18 (Butaud) (same).) Mr. Beal testified that he was transferred from a quarantine dorm back to a negative dorm even though he was still symptomatic. (Tr. 2-162:25–165:3 (Beal).) Mr. Beal spoke to medical and security about his continuing symptoms and filed a grievance stating his concern that his transfer would put other inmates at risk, but was told that he would nonetheless be transferred because he had completed the minimum quarantine period. (Tr. 2-162:25–165:3 (Beal).) Mr. Beal was harassed by inmates in the negative dorm because of his symptoms. (Tr. 2-165:4–9 (Beal).)

134. Mr. Collier was aware that inmates could still be positive for COVID-19 past the 14-day quarantine period. On April 26, 2020, Mr. Collier received a text informing him that "FYI One of our first positives is still testing positive and shedding virus at 21 days. The states 14 day isolation with no retesting is questionable at best." (PTX 346 at 1.)

ix. Contact Tracing

135. There is no written policy, plan, or instructions for contact tracing among inmates or staff at TDCJ. (Tr. 1-105:15–106:5 (Collier).) In a text message dated July 4, 2020, Mr. Mendoza asked Ms. Davis: "Question ... when it comes to Contact Tracing ... do we have written procedures on how they should be done???" (PTX 337 at 2.) Ms. Davis responded: "Not that I know of," and explained

that staff should fill out a blank form. (PTX 337 at 2.) Mr. Mendoza also asked whether there "was any training done on how to use the Tracing Forms or was it an informal process of ‘here they are,’ " (PTX at 2), but there is no response in the record. TDCJ appears to use a 48-hour window for contact tracing. (Tr. 1-61:22–62:3, 106:16–20 (Collier).)

136. Even when the contact tracing forms were used, they were not always used properly. After Mr. Clerkly's death, officials did not fill out much of the contact tracing form and only listed those individuals who Mr. Clerkly came into contact with on the day he died. (PTX 186 at 3); (Tr. 1-108:21–23 ("I see nothing filled out, yes.") (Collier).) However, inmates testified that Mr. Clerkly used to sit in a common area outside of his dorm before visiting the infirmary, and not long before he died he had contact with Mr. Valentine and several other inmates. (Tr. 1-189:5–10 (Valentine).) None of the inmates was contacted related to contact tracing following Mr. Clerkly's death, nor were any inmates outside of Mr. Clerkly's dorm. (Tr. 1-190:20–24 (Valentine).)

137. Multiple inmates testified that, after they tested positive for COVID-19, no one from TDCJ reached out to them about contact tracing. (Tr. 1-188:22–24 (Valentine); 2-157:23–158:14 (Beal); 2-242:14–23 (Pennington).) The inmates were not asked about where they had been or who they had close contact with in the days before they were tested or became symptomatic. (Tr. 2-158:15–23 (Beal).)

x. Education

138. Educational posters describing the symptoms of COVID-19, how to prevent its spread, and what to do if inmates begin feeling ill are placed around the facility. (Tr. 1-192:13–17 (Valentine); Tr. 11-69:1–4 (Wilder); DTX 10.) The posters appear to be in English only. (Tr. 1-192:18–25 (Valentine).)

139. Pamphlets entitled "COVID-19: What You Need to Know About the Coronavirus Disease 2019" have also been distributed to inmates. (Tr. 1-193:18–21 (Valentine).)

140. Mr. Dove, a legally blind inmate, testified that he could not read any of the posters or pamphlets provided by TDCJ. (Tr. 2-192:25–193:6, 193:11–14 (Dove).) No one has read Mr. Dove the information from the COVID-19 pamphlets or posters. (Tr. 2-193:7–10 (Dove).)

141. A video produced by TDCJ about preventing the spread of COVID-19 has been played on the dayroom TVs three times a day since mid-April. (Tr. 3-243:4–8 (Herrera); Tr. 11-69:5–10 (Wilder); DTX 9.) The dayroom TVs have been turned to face the cubicles, such that inmates can generally see the TV screens from their cubicles. (Tr. 3-244:12–17 (Herrera); Tr. 11-69:7–10 (Wilder).) Warden Herrera testified that no inmate had ever complained to him that he could not see or hear the video. (Tr. 3-245:8–12 (Herrera).) However, Mr. Valentine testified that the video is inaudible from the center of the dorm, and especially those who are hearing impaired have difficulty hearing the video.

(Tr. 1-177:23–178:13 (Valentine).) The video is also only in English, even though there are inmates who speak only Spanish. (Tr. 1-177:23–178:7 (Valentine).)

142. Two to three times a week, there are announcements in English and Spanish informing inmates that the medical co-pay has been suspended. (Tr. 11-71:8–11 (Wilder).)

I. Compliance with Policy B-14.52

143. Of TDCJ's 104 units across Texas, Pack Unit had the most complaints and grievances related to COVID-19. (Tr. 10-72:25–73:4, 10-187:18–25 (Collier); Tr. 4-114:18–115:16 (Herrera).) In an email from Marvin Dunbar to Mr. Collier on May 26, 2020, Mr. Dunbar reported that these complaints related to failure to wear masks and gloves appropriately, lack of cleaning supplies issued to inmates, inability to social distance, and exposure to COVID-19 by staff and inmates who have tested positive. (PTX 301 at 1.)

144. Despite these grievances and although Policy B-14.52 was established on March 20, 2020, and was most recently revised on June 11, 2020, (PTX 115 at 1), TDCJ did not begin discussions to create compliance teams until June 24, 2020. (Tr. 10-90:4–10 (Collier).) The compliance teams are to oversee and ensure that TDCJ is following Policy B-14.52 by, for example, providing sufficient hand soap, disinfecting, and requiring social distancing and mask wearing. (Tr. 10-89:15–90:2 (Collier).)

145. As of July 24, 2020, a compliance team had visited Pack Unit only once, and TDCJ has not received any official report or documentation that TDCJ is in fact following COVID-19 protocol. (Tr. 10-90:22–91:13 (Collier).)

146. On July 27, 2020, in a text message following up on the compliance assessment team, Mr. Hirsch wrote that "[Mr. Collier] said after talking to Gene last week he felt we were not doing everything we should have been and was wondering if that might be a factor. Thinks [sic] like restricting, isolating, PPE access, cleaning supplies." (PTX 338 at 1.)

147. Mr. Collier sent Mr. Monroe, the regional director, to conduct a site visit at Pack Unit on July 15 or 16, 2020, based on testimony from inmate witnesses at trial. (Tr. 10-97:17–98:11, 10-160:12–14 (Collier).) Mr. Monroe was removed from Defendants’ trial witness list several days later on July 20, 2020, and therefore Mr. Monroe never testified to his observations during this site visit. (Tr. 10-160:15–19 (Collier).)

J. Grievance Process

148. Inmates are instructed by their TDCJ-issued orientation handbooks to attempt informal resolution of conflicts before filing a Step 1 grievance. (Tr. 10-201:4–13 (Collier); PTX 9 at 7 ("Offenders shall pursue an informal resolution with staff, when possible, prior to filing a formal grievance."); PTX 10 at 86–87 ("An attempt to informally resolve your problem shall be made before filing a grievance.").) Grievances that were filed before attempting informal resolution can be rejected on that basis. (Tr. 10-201:14–24 (Collier); PTX 9 at 7 ("When an attempt

to informally resolve an issue is not documented on the grievance form, the unit grievance investigator may return the grievance to the offender without an investigation.").)

149. Inmates have 15 days after the alleged incident or problem to submit a Step 1 grievance. (PTX 9 at 8; PTX 10 at 87.) Unless a grievance is coded as an emergency, TDCJ grievance staff may take up to 40 days to respond to the Step 1 grievance. (PTX 9 at 8); (PTX 10 at 87.) The grievance staff can extend its own deadline by an additional 40 days, for a maximum response time of 80 days after the Step 1 grievance was filed. (PTX 9 at 10.)

150. Inmates may appeal a Step 1 decision by filing a Step 2 grievance within 15 days of receipt of the Step 1 response. (PTX 9 at 8; PTX 10 at 87.) TDCJ grievance staff may take up to 40 days to respond to the Step 2 grievance. (PTX 9 at 8–9; PTX 10 at 87.) The grievance staff can extend its own deadline by an additional 40 days, for a maximum response time of 80 days after the Step 2 grievance was filed. (PTX 9 at 10.)

151. For his first pandemic-related grievance, Mr. Valentine attempted informal resolution through an I-60 form on March 27, 2020. (DTX 11 at 2.) Mr. Valentine submitted the corresponding Step 1 grievance on March 31, 2020. (DTX 11 at 3.) In his Step 1 grievance, Mr. Valentine requested hand sanitizer, gloves, masks, and additional cleaning supplies for individual cubicles, in order to prevent the spread of COVID-19. (DTX 11 at 2.) At the top of the first page, Mr. Valentine wrote "(EMERGENCY GRIEVANCE)." (DTX 11 at 2.) TDCJ responded to Mr. Valentine's Step 1 grievance on April 22, 2020. (DTX 11 at 3.) As of June 5, 2020, Mr. Valentine had not received a response to his Step 2 grievance (Doc. 137-1 at 3.)

152. On April 2, 2020, Mr. King filed a Step 1 grievance without attempting informal resolution. (DTX 12 at 2–3; Tr. 2-61:2–4 (King).) The grievance involved inmate transfers to the Pack Unit. (DTX 12 at 2; Tr. 2-60:14–19 (King).) TDCJ responded to the grievance on May 4, 2020, even though there was no attempt at informal resolution. (DTX 12 at 2; Tr. 2-61:2–7 (King).) However, Mr. King clarified that there was no means for attempting informal resolution, since his complaint was lodged against the State Classification Department, located in Huntsville, Texas. (Tr. 2-61:15–18 (King).)

153. Prior to May 26, 2020, COVID-related grievances were not treated differently from other types of grievances. Even if inmates tried to designate COVID-related grievances as emergencies, there was no way for these grievances to be processed as emergencies unless they satisfied the criteria of a normal emergency classification. (Tr. 11-204:12–16, 11-213:13–17 (Wilder).) Inmates’ attempts to classify their grievances as emergencies were not considered in deciding whether a grievance raised an emergency. (Tr. 11-215:3–13 (Wilder).) Warden Wilder testified that he still does not consider COVID-related grievances that did not fulfill another emergency classification

to be an emergency grievance that requires a shortened time frame for response. (Tr. 11-213:15–23 (Wilder).)

154. Given the rapid spread of COVID-19, multiple inmates contracted the virus while they awaited Pack Unit's response. Mr. Valentine, for example, began his grievance process by attempting informal resolution on March 27, 2020, (Tr. 1-210:7–211:22 (Valentine)), and filed a COVID-related grievance on March 31, 2020, (Doc. 137-1 at 2, 5–6). Mr. Valentine tested positive for COVID-19 as did three other Pack Unit inmates before the Pack Unit fully responded to his grievance. (PTX 43, 158, 160, 186, 189, 196.) Mr. Butaud filed a Step-1 grievance on April 3, 2020, related to COVID-19. (PTX 131 at 17–18.) TDCJ granted itself an extension and did not respond for 81 days. (Tr. 3-177:4–18 (Herrera).) Like Mr. Valentine, Mr. Butaud also tested positive for COVID-19 while his grievance was pending. (PTX 131 at 17–18; Tr. 2-196:17–197:1 (Butaud).)

155. On May 26, 2020, TDCJ implemented a new process for handling COVID-related grievances. (PTX 157 at 2.) The new COVID-19 Grievance Process created a new code for grievances that are related in any way to COVID-19 or TDCJ's measures responding to COVID-19. (PTX 157 at 3.) All grievances related to COVID-19 are exempt from almost all screening criteria. (PTX 157 at 3.) Additionally, all COVID-related grievances must be processed along an expedited time frame—a maximum of 15 days per step, for a total of 30 days—with no possibility for extensions. (PTX 157 at 3–4.) The new process became effective on May 26, 2020. (PTX 157 at 2; Tr. 11-83:4–11 (Wilder).)

III. CONCLUSIONS OF LAW

Plaintiffs contend that Defendants’ conduct is violating (1) the Eighth Amendment, and (2) the Americans with Disabilities Act ("ADA") and Rehabilitation Act ("RA"). They seek a permanent injunction requiring Defendants to carry out a number of measures to adequately protect class members from COVID-19, as well as the appointment of a special master to enforce the injunction. (Doc. 354 ¶¶ 414-15.)

A. Plaintiffs’ Pending Motion for Class Certification

Before addressing the merits of Plaintiffs’ claims, the Court addresses Plaintiffs’ Emergency Supplemental Motion for Certification of Two Additional Subclasses, filed July 21, 2020. (Doc. 248). As noted supra , this Court has already certified the General Class and the High-Risk Subclass. Class Cert. Order, 2020 WL 3491999, at *1. After additional discovery, Plaintiffs moved to certify two additional sub-classes under Rule 23(b)(2): the "Disability Subclass" and the "Mobility-Impaired Subclass," defined as follows:

The Disability Subclass : All current and future inmates incarcerated in the Texas Department of Criminal Justice Wallace Pack Unit who suffer from a disability that substantially limits one or more of their major life activities, are subjected to TDCJ and the Texas Correctional Managed Health Care Committee's policy and practice of failing to provide protection from exposure to Covid-19 during the class period, and who are at increased risk of serious illness, injury, or death if exposed to Covid-19 due to

their disability and any medical treatment necessary to treat their disability.

The Mobility-Impaired Subclass : All current and future inmates incarcerated in the Texas Department of Criminal Justice Wallace Pack Unit who suffer from a disability that substantially limits one or more of their major life activities, are subjected to TDCJ and the Texas Correctional Managed Health Care Committee's policy and practice of failing to provide protection from exposure to Covid-19 during the class period, and who require the use of a walker, cane, or wheelchair to ambulate.

(Doc. 248 at 11.) To Plaintiffs’ proposed definition of the Mobility-Impaired Subclass, the Court adds individuals who require crutches to walk. Plaintiffs Valentine and King are members of the purported Disability Subclass, and Mr. Valentine is a member of the purported Mobility-Impaired Subclass. Id.

The Court applies the same legal standard and framework that it did in its previous order granting certification of the General Class and the High-Risk Subclass. See Class Cert. Order, 2020 WL 3491999, at *2–3. For the reasons that follow, the Court certifies the Mobility-Impaired Subclass but not the Disability Subclass.

i. Legal Standard

The requirements for class certification under Rule 23(a) are numerosity, commonality, typicality, and adequacy. In addition, Plaintiffs seek to certify these subclasses under Rule 23(b)(2), which is the class action mechanism for plaintiffs seeking injunctive or declaratory relief. Ordinarily, a court may not perform "free-ranging merits inquiries at the certification stage," Amgen Inc. v. Conn. Ret. Plans & Tr. Funds , 568 U.S. 455, 466, 133 S.Ct. 1184, 185 L.Ed.2d 308 (2013) ; however, the certification questions often "overlap" with the merits of the underlying claims. Wal-Mart Stores, Inc. v. Dukes , 564 U.S. 338, 351, 131 S.Ct. 2541, 180 L.Ed.2d 374 (2011).

ii. The Mobility-Impaired Subclass

The Court finds that Plaintiffs’ proposed Mobility-Impaired Subclass meets the requirements of Rule 23(b)(2).

Numerosity. Rule 23(a)(1) requires that the class be "so numerous that joinder of all members is impracticable." Additionally, where there are subclasses, "each subclass must independently meet the numerosity requirement." Leal v. Paramount Restaurants Grp., Inc. , No. 12-CY-038-J, 2013 WL 1363616, at *3 (N.D. Tex. Apr. 4, 2013). There are "no mechanical rules" to determine whether a class is too numerous; rather, that inquiry turns on "practicability." Watson v. Shell Oil Co. , 979 F.2d 1014, 1022 (5th Cir. 1992). At the same time, the "rule of thumb adopted by most courts is that proposed classes in excess of 40 generally satisfy the numerosity requirement." 1 McLaughlin on Class Actions § 4:5 (16th ed.) (collecting cases).

When trial began, the Pack Unit had 49 wheelchair-bound inmates and 87 inmates who used walkers. (PTX 337 at 2.) Thus, based on this Court's findings of fact, the proposed Mobility-Impaired Subclass has more than 40 members and presumptively meets the numerosity requirement. This number may also change over time, if new prisoners are transferred to or from the Pack Unit. Defendants dispute numerosity, arguing that there is no "geographical dispersion" among class members and, at the time, that Plaintiffs’ estimates of the size of the class were unsupported by evidence. (Doc. 275 at 14–18.) However, geographical dispersion is not a requirement but rather one factor among many that courts may consider when determining whether joinder is impractical. For example, the Court has already certified the General Class and the High-Risk Subclass, all of whose members are located at the Pack Unit. The Court determines that joinder of over 100 mobility-impaired incarcerated individuals is impractical given the late stage of this litigation. Moreover, such joinder would have been impractical at any stage of this litigation, given that the purported class members are disabled, incarcerated, and seek emergency injunctive relief as protection against imminent health risks. The class is therefore numerous.

Commonality. Rule 23(a)(2) requires that there be "questions of law or fact common to the class." In practice, this means that the legal claims of the class "depend upon a common contention" that is "capable of classwide resolution." Wal-Mart , 564 U.S. at 350, 131 S.Ct. 2541. In other words, the answer to that contention must "resolve an issue that is central to the validity of each one of the claims in one stroke." Id. The Mobility-Impaired Subclass raises at least one common question, the answer to which will resolve particular claims with regard to that class "in one stroke." That is: does a lack of access to hand sanitizer and other cleaning materials for hand hygiene violate the ADA with respect to mobility-impaired individuals? Either the answer to this question is "yes," in which case TDCJ is required to provide that specific accommodation for the entire subclass, or the answer is "no," in which case class-wide relief is not warranted. But either way, the answer to that question can be resolved with respect to every mobility-impaired class member at once. This is all Plaintiffs need, as Wal-Mart teaches that "even a single common question" may satisfy commonality as long as it admits a common answer that drives the litigation. 564 U.S. at 359, 131 S.Ct. 2541 (citation and internal modifications omitted); see also Yates v. Collier , 868 F.3d 354, 366–67 (5th Cir. 2017) ("Rule 23(b)(2) requires common behavior by the defendant toward the class." (quoting In re Rodriguez , 695 F.3d 360, 365 (5th Cir. 2012) )).

Defendants attempt to analogize the Mobility-Impaired Subclass to the proposed class in Wal-Mart , arguing that resolution of Plaintiffs’ ADA claims "would require highly individualized inquiries." (Doc. 275 at 7.) While this may be true of the Disability Subclass, as different disabilities might require different accommodations, it is not true of the Mobility-Impaired Subclass, through which Plaintiffs allege a unique harm tied to a particular disability that would apply to all subclass members and which can be remedied with a single injunction. Unlike in Wal-Mart , where the class members’ claims were based on "millions of employment decisions" within the "discretionary decisionmaking of ... thousands of managers," 564 U.S. at 345, 352, 131 S.Ct. 2541, the harm alleged here is the result of decisions made by a single centralized body—TDCJ—and the officials in charge of the Pack Unit. And unlike in Wal-Mart , where the answer to the common question—"why was I disfavored" in employment decisions—necessitated an individualized inquiry, 564 U.S. at 352, 131 S.Ct. 2541, the analysis of whether Defendants’ conduct violates the ADA with respect to individuals with mobility impairments is common. That analysis would require Plaintiffs to demonstrate: "(1) that they are qualified individuals within the meaning of the Act; (2) that they are being excluded from participation in, or being denied benefits of, services, programs, or activities for which the Secretary is responsible, or are otherwise being discriminated against by the Secretary; and (3) that such exclusion, denial of benefits, or discrimination is by reason of their disability." Lightbourn v. Cty. of El Paso , 118 F.3d 421, 428 (5th Cir. 1997). The answers to the Lightbourn factors are the same for all subclass members, and in fact the Fifth Circuit has found commonality for proposed classes bringing ADA claims based on specific disabilities. See, e.g. , Yates , 868 F.3d at 366 (affirming certification of class of prison inmates "suffering from disabilities that may impact ... their ability to withstand extreme heat"); Lightbourn , 118 F.3d at 425 (affirming certification of class of voting-age citizens in Texas "who are blind or mobility-impaired"). Given that the Mobility-Impaired Subclass likewise requests common relief for the same alleged harm stemming from the same kind of disability, the subclass satisfies commonality. And to be sure, "no two individuals have the exact same risk" of contracting COVID-19 due to Defendants’ alleged discriminatory conduct, but this "obvious fact does not destroy commonality." See Yates , 868 F.3d at 363 (citation and internal modifications omitted).

Typicality and Adequacy. Rule 23(a)(3) requires that the Plaintiffs’ claims be "typical" of the rest of the class, and Rule 23(a)(4) requires that "the representative parties will fairly and adequately protect the interests of the class." These inquiries "tend to merge" with commonality. Wal-Mart , 564 U.S. at 349 n.5, 131 S.Ct. 2541. Mr. Valentine has required a walker to ambulate since 1993 due to a drop foot and nerve damage in his legs. (Tr. 1-149:1–150:17 (Valentine).) His disability is therefore typical of those of the other Mobility-Impaired Subclass members, and the harm he alleges—difficulty maintaining a medically necessary level of hand hygiene as a result—is also typical of the subclass. Defendants argue that typicality is not satisfied due to the "unique circumstances attendant to each individual inmate's medical needs," but as discussed with respect to commonality, no such individualized inquiry is needed with respect to inmates who allege the same harm stemming from the same disability. (See Doc. 275 at 23.) Finally, the Court finds that the class representatives as well as class counsel satisfy adequacy for the same reasons as in its previous order granting class certification, and Defendants raise no new arguments to the contrary. See Class Cert. Order, 2020 WL 3491999, at *11–13 ; (Doc. 275 at 17–18).

Rule 23(b)(2) Requirements. Finally, Plaintiffs seek to certify this subclass under Rule 23(b)(2), which requires that "the party opposing the class has acted or refused to act on grounds that apply generally to the class, so that final injunctive relief or corresponding declaratory relief is appropriate respecting the class as a whole." The Supreme Court has stated that the "key to the (b)(2) class is the indivisible nature of the injunctive or declaratory remedy warranted." Wal-Mart , 564 U.S. at 360, 131 S.Ct. 2541 (quotation omitted). The Fifth Circuit has previously held that Rule 23(b)(2) certification requires that "(1) class members must have been harmed in essentially the same way; (2) injunctive relief must predominate over monetary damage claims; and (3) the injunctive relief sought must be specific." Yates , 868 F.3d at 366 (internal quotations omitted) (quoting Maldonado v. Ochsner Clinic Found. , 493 F.3d 521, 524 (5th Cir. 2007) ). The Mobility-Impaired Subclass easily satisfies these requirements. The class members request specific relief—access to hand sanitizer and other hand hygiene materials—that would remedy the alleged harms. They also request no monetary damages whatsoever, unlike the class members in Wal-Mart , 564 U.S. at 360, 131 S.Ct. 2541, who sought individualized amounts of backpay. Accordingly, the Mobility-Impaired Subclass ought to be certified.

iii. The Disability Subclass

However, the Court agrees with Defendants that the Disability Subclass does not satisfy commonality. As Plaintiffs note, the class raises a number of common questions such as whether the class members fall within the scope of the ADA, whether TDCJ has provided reasonable accommodations to them, and whether TDCJ has denied them access to the prison's services, programs or activities by making it more difficult for individuals to wash their hands. (Doc. 248 at 14–15.) But the Supreme Court has emphasized that what matters "is not the raising of common questions," but whether those questions are capable of being answered with a single remedy. Wal-Mart , 564 U.S. at 350, 360, 131 S.Ct. 2541 (citation and internal modifications omitted). For the Disability Subclass, these questions cannot be answered as to the class as a whole, given the wide range of disabilities that are protected by the ADA. In the cases Plaintiffs cite, the purported subclasses challenging conduct as violative of the ADA are defined more specifically to allege a common harm which can be remedied by a single "reasonable accommodation" as required by the ADA, often tailored to specific disabilities. See, e.g. , Yates , 868 F.3d at 366 (class of individuals with disabilities which made them "particularly susceptible to heat"); Lightbourn , 118 F.3d at 425 (class of individuals "who are blind or mobility-impaired"). Put another way, Plaintiffs’ Mobility-Impaired Subclass satisfies commonality for the same reason the general Disability Subclass fails—the former identifies a common harm which can be remedied in a single stroke, whereas the alleged harms to the Disability Class cannot be remedied by a single set of reasonable accommodations.

The Court therefore certifies Plaintiffs’ proposed Mobility-Impaired Subclass but not the Disability Subclass. The Court now turns to the merits of Plaintiffs’ claims.

B. Legal Standard for a Permanent Injunction

The standard for a permanent injunction is "essentially the same" as that for a preliminary injunction, except the plaintiff must show "actual success" rather than "a likelihood of success" on the merits of their claims. Winter v. Nat. Res. Def. Council, Inc. , 555 U.S. 7, 22, 129 S.Ct. 365, 172 L.Ed.2d 249 (2008) (quoting Amoco Prod. Co. v. Gambell , 480 U.S. 531, 546 n.12, 107 S.Ct. 1396, 94 L.Ed.2d 542 (1987) ); see also eBay Inc. v. MercExchange, LLC , 547 U.S. 388, 391, 126 S.Ct. 1837, 164 L.Ed.2d 641 (2006). The Court therefore applies the following familiar four-factor test, with that caveat in mind. Plaintiffs must demonstrate: (1) "[actual success] on the merits," (2) "that he is likely to suffer irreparable harm in the absence of ... relief," (3) "that the balance of equities tips in his favor," and (4) "that an injunction is in the public interest." Winter , 555 U.S. at 20, 129 S.Ct. 365. Importantly, a "permanent injunction is appropriate only if a ‘defendant's past conduct gives rise to an inference that, in light of present circumstances, there is a " ‘reasonable likelihood’ of future transgressions." ’ " SEC v. Life Partners Holdings, Inc. , 854 F.3d 765, 784 (5th Cir. 2017) (quoting SEC v. Gann , 565 F.3d 932, 940 (5th Cir. 2009) ).

C. Administrative Exhaustion Under the PLRA

As a threshold matter, Defendants have renewed their argument that Plaintiffs failed to exhaust administrative remedies as required by the PLRA, and their claims are therefore barred. (Doc. 355 at 56–57.) In response, Plaintiffs argue that they were not required to exhaust TDCJ's grievance procedure because it was unavailable to them, within the meaning of the PLRA. (Doc. 354 ¶¶ 394–407.) Ordinarily, the PLRA requires inmates to exhaust "available" administrative remedies before bringing suit in federal court to challenge prison conditions. 42 U.S.C. § 1997e(a). This requirement is "mandatory" and there is no exception even for "special circumstances." Ross v. Blake , ––– U.S. ––––, 136 S. Ct. 1850, 1856, 195 L.Ed.2d 117 (2016). However, the statutory text requires the remedy actually to be "available"—in other words, "capable of use" in order to obtain "some relief." Id. at 1859. The Supreme Court has articulated "three kinds of circumstances in which an administrative remedy, although officially on the books, is not capable of use to obtain relief" and is thus unavailable within the meaning of 42 U.S.C. § 1997e(a). Id. One such scenario is where the remedy "operates as a simple dead end." Id. However, these exemplary scenarios are non-exhaustive. See Class Cert. Order, 2020 WL 3491999 at *6 (collecting cases).

The issue of whether Plaintiffs have exhausted any "available" remedies has been extensively briefed and argued before the Court several times in this litigation. Most recently, this Court determined that "TDCJ's grievance procedure was not ‘capable of use to obtain some relief,’ ... under two theories of unavailability." Id. First, the Court found that TDCJ's grievance process "did not fit the problem Plaintiffs were facing," as it was "not designed with a worldwide pandemic in mind" and was thus unable to provide relief to inmates facing an imminent risk of serious illness or death. Id. at *7. Second and relatedly, the Court determined that "TDCJ's grievance process was a ‘simple dead end’ " because it was unable to provide even the minimal amount of relief that Plaintiffs would need "in order to be protected from the pandemic to the degree required by the Eighth Amendment." Id.

In considering exhaustion once again, the Court re-iterates that the issue of availability under 42 U.S.C. § 1997e(a) is a "fact-based analysis," and "further factual development could change [a Court's] preliminary evaluations of the issue of exhaustion and availability."⁹ Id. at *5 (citing Ross , 136 S. Ct. at 1859 ); accord Valentine I , 956 F.3d 797, 806 (5th Cir. 2020) (Higginson, J., concurring) (noting that further factual development may change a court's determination as to the availability of administrative remedies). The Court also notes that the relevant factual circumstances for this analysis are those that existed at the time Plaintiffs filed their complaint. See, e.g. , Capozzi v. Pigos , 640 F. App'x 142, 145 (3d Cir. 2016). This commonsense principle flows naturally from the requirement that an inmate is generally required to exhaust any administrative remedies before filing suit in federal court. Otherwise, an inmate faced with a grievance process incapable of use would be placed in the impossible and untenable position of either (a) filing suit potentially subject to dismissal if the grievance process was later changed, or (b) waiting indefinitely to vindicate his or her constitutional rights. See id. Thus, the Court does not consider any changes to TDCJ's grievance processes made after Mr. Valentine and Mr. King filed the original complaint. What matters is the process that existed at the time of filing.

9 Defendants argue that the Fifth Circuit's holding in Valentine I , 956 F.3d at 804, "is binding on this Court" as to the question of exhaustion. (Doc. 355 at 57.) But as explained by this Court in its class certification opinion, that "panel made its ruling on a different procedural posture, and on a much narrower record." Class Cert. Order, 2020 WL 3491999, at *5. Given that exhaustion is a fact-based analysis, that opinion's conclusion does not bind this Court's renewed determination of the issue on a substantially different factual record, now after a full trial.

With these background principles in mind, the Court once again concludes that TDCJ's grievance process was incapable of use at the time Plaintiffs filed their lawsuit. That analysis was based primarily on four factors, given the record at the time. These were: (1) the fact that Mr. Valentine had begun the informal administrative process, as he was required to by TDCJ policy, before filing this lawsuit; (2) the fact that over two months elapsed from the time Mr. Valentine and Mr. King filed grievances, and they remained unresolved; (3) the rapid deterioration of circumstances at the Pack Unit while these grievances were pending, including 267 inmates testing positive and eighteen inmates having died as of June 27; and (4) Defendants’ subsequent acknowledgment that the existing grievance process was inadequate in light of COVID-19 and the implementation of a new set of procedures. Class Cert. Order, 2020 WL 3491999, at *5–6. The evidence presented at trial confirms or bolsters each of these findings.

Defendants insist that, contrary to this Court's earlier findings, Plaintiffs "Valentine and King have successfully navigated TDCJ's grievance process and have largely obtained the relief they have requested in their grievances." (Doc. 355 at 44.) In support of this assertion, Defendants point to various examples of TDCJ policy changes supposedly made in response to grievances filed by Plaintiffs Valentine and King. (Doc. 355 at 42–43.) These include changes in access to soap and cleaning supplies; testing; and a request to stop transfers into the Pack Unit.¹⁰ (Doc. 355 at 44–45.)

10 Defendants additionally contend that requests for increased social distancing and sanitizing the shower area between use had already been granted at the time Mr. Valentine filed his grievances. (Doc. 355 at 43.)

The trial record tells a slightly different story, however. In some of these instances, TDCJ changed its policies prior to a grievance being filed, such as with regard to Mr. Valentine's request for testing. (Compare DTX 11 at 2384–85 (Mr. Valentine's Step 1 grievance for testing, filed May 10, 2020) with Tr. 10-26:4–27:23 (Collier, testifying that strike teams were created May 9, 2020).) During cross-examination, Mr. Collier repeatedly stated that he was "not sure" whether remedies would be available under the TDCJ grievance policy as constituted prior to any changes made in light of COVID-19. (Tr. 10-208:22–209:12 (Collier).) Thus, post-trial, there is insufficient evidence to find that TDCJ was in fact responsive to grievances filed in the systematic way that would be required to render the process available to provide relief during the pandemic.

The grievance process also operated too slowly, given the risk to human life posed by COVID-19. For instance, Mr. Valentine attempted informal resolution beginning March 27, 2020. He requested hand sanitizer, gloves, masks, and additional cleaning supplies in order to prevent the spread of COVID-19. Defendants are correct to point out that TDCJ responded to Mr. Valentine's Step 1 grievance in under a month. However, inmates are required by the grievance process to subsequently filed a Step 2 grievance, to which he hadn't received a response as of June 5, well after he had tested positive for COVID-19 and multiple inmates had died. An administrative process is not available if there is a likelihood the inmate will die or suffer severe illness while waiting for a response.

Indeed, for a number of other inmates, there is no doubt that TDCJ was either "unable" or "unwilling to provide any relief to aggrieved inmates." Ross , 136 S. Ct. at 1859. One need look no further than the tragic case of Mr. Norris, who filed a grievance on April 27 complaining that a member of the medical staff was "not wearing gloves or masks." (PTX 152 at 6 (e-mail list of COVID-19 related grievances filed Apr. 27).) According to Mr. Valentine's testimony, Defendants were also on notice that Mr. Norris was grappling with severe medical problems, and a sergeant told Mr. Valentine that he "didn't have time for this." (Tr. 15-18:10–19:12 (Valentine).) Less than a week later, Mr. Norris died, with COVID-19, while his grievance was pending. (PTX 196 (Mr. Norris's preliminary autopsy).) Defendants provided no evidence that any steps were taken to address Mr. Norris's grievance in the interim. Mr. Norris's death demonstrates unequivocally that the TDCJ grievance process was unavailable and incapable of providing even "some relief," as Ross requires. Similarly, TDCJ deemed other emergency grievances filed by inmates relating to COVID-19 as presenting issues that were "not grievable." (PTX 131 at 61–62 (Mr. Reynolds's Grievance Form).)

At trial, counsel for Defendants insisted that "this is the law"—"[e]ven if a hundred people have died, the plaintiffs would have had to wait until the grievance process was exhausted before filing suit." (Tr. 18-159:20–160:11). But the Court declines what ultimately amounts to an invitation to interpret the PLRA's exhaustion requirement in a way that raises serious constitutional doubts. See Class Cert. Order, 2020 WL 3491999, at *8 (citing Nielsen v. Preap , ––– U.S. ––––, 139 S. Ct. 954, 972, 203 L.Ed.2d 333 (2019) ); accord Malam v. Adducci , 452 F.Supp.3d 643, 662 (E.D. Mich. 2020) (holding that mandatory immigration detention statute must admit an exception where "continued detention is in violation of the United States Constitution"). Though the Court need not decide the issue, such an interpretation would almost certainly run afoul of the state's duty to provide basic medical care to individuals from whom it has "take[n] ... the means to provide for their own needs." See Brown v. Plata , 563 U.S. 493, 510, 131 S.Ct. 1910, 179 L.Ed.2d 969 (2011). The Court thus avoids this difficult constitutional question by adopting the "fairly possible" construction of the PLRA that, in light of these extraordinary times, the regular TDCJ grievance process was simply incapable of use by inmates whose lives were threatened by—and in some tragic instances, claimed by—the COVID-19 pandemic. See Nielsen , 139 S. Ct. at 972.

Simply put, Plaintiffs have demonstrated that the existing grievance process, designed for run-of-the-mill requests in ordinary times, was "utterly incapable of responding to a rapidly spreading pandemic like Covid-19, ... much in the way they would be if prison officials ignored the grievances entirely." Valentine v. Collier , ––– U.S. ––––, 140 S. Ct. 1598, 1600–01, 206 L.Ed.2d 930 (mem) (Sotomayor, J., respecting denial of application to vacate stay). Thus, the PLRA does not bar Plaintiffs’ claims.

D. Permanent Injunction Analysis

The Court now turns to the merits of Plaintiffs’ claims, beginning with the Eighth Amendment.

i. Actual Success on the Merits

a. Eighth Amendment

The state has a constitutional duty to protect the people it incarcerates from "a substantial risk of serious harm" and to "take reasonable measures to guarantee the[ir] safety." Farmer v. Brennan , 511 U.S. 825, 828, 832, 114 S.Ct. 1970, 128 L.Ed.2d 811 (1994) (quotation omitted). This includes the responsibility to provide basic human needs such as "adequate ... medical care." Id. at 832, 114 S.Ct. 1970. At the same time, an individual seeking relief from conditions of confinement under the Eighth Amendment must demonstrate prison officials’ " ‘deliberate indifference’ to a substantial risk of serious harm." Id. at 828, 114 S.Ct. 1970. Deliberate indifference exists where a prison official "knows that inmates face a substantial risk of serious harm and disregards that risk by failing to take reasonable measures to abate it." Id. at 847, 114 S.Ct. 1970. This is an "extremely high standard to meet." Cadena v. El Paso Cty. , 946 F.3d 717, 728 (5th Cir. 2020).

"[D]eliberate indifference is a question of fact often made out by ‘inference from circumstantial evidence.’ " Valentine , 140 S. Ct. at 1598 (Sotomayor, J., respecting the denial of application to vacate stay) (quoting Farmer , 511 U.S. at 842, 114 S.Ct. 1970 ). Such an inference may follow "from the very fact that the risk was obvious." Farmer , 511 U.S. at 842, 114 S.Ct. 1970. At the same time, that inference cannot stem "merely from a negligent or even a grossly negligent response to a substantial risk of serious harm." Thompson v. Upshur Cty. , 245 F.3d 447, 459 (5th Cir. 2001). But if a defendant-official "had been exposed to information concerning the risk and thus ‘must have known’ about it," that is "sufficient" to reason that he or she "had actual knowledge" of it as well. Farmer , 511 U.S. at 842–43, 114 S.Ct. 1970. In other words, a defendant-official need not admit the requisite mental state in order for a plaintiff to prove deliberate indifference.

At the preliminary injunction stage, this Court found that Plaintiffs, on a much narrower record, were "likely" to establish that conditions at the Pack Unit violated their Eighth Amendment rights. Valentine v. Collier , 455 F.Supp.3d 308, 321-27 (S.D. Tex. 2020). The Fifth Circuit disagreed and granted a stay of that injunction pending appeal. Valentine I , 956 F.3d at 806. The stay panel determined, based on "Defendants’ written evidence and Plaintiffs’ live witness testimony" at the time, that TDCJ officials had not been deliberately indifferent to Plaintiffs’ medical needs. Id. at 799–801. But the deliberate indifference inquiry is "intensely fact-based." Id. at 807 (Higginson, J., concurring). The Court now once again takes up the issue of whether Plaintiffs have proven deliberate indifference with the benefit of a fully developed record from an 18-day trial, including extensive live testimony from both sides.

The facts before this Court post-trial present a wholly different picture of the risks faced by individuals at the Pack Unit in the preceding months as well as in the months ahead than did the facts before this Court in April. Most saliently, at the time of the Fifth Circuit's decision, one individual had died—Mr. Clerkly—and mass testing had not begun, so the extent of the risk posed to inmates at the Pack Unit was wholly unknown. In fact, Mr. Clerkly was the only individual known to have been infected at all. As such, neither this Court at that time nor the Fifth Circuit panel that stayed the injunction had an indication of the magnitude of the impending outbreak at the Pack Unit. In stark contrast, the Court is now confronted with the "dramatically changed" and sobering reality that 20 men have died and over 40% of the inmates held at the Pack Unit have tested positive—undoubtedly, a "human tragedy." See Valentine II , 960 F.3d at 707 (Davis, J., concurring in judgment). This makes the infection rate and the overall death rate at the Pack Unit significantly higher than that of Texas or the United States as a whole. The Court's analysis is grounded in these grim statistics. To be sure, the deliberate indifference inquiry is subjective, not objective, so these statistics in and of themselves do not control the analysis. But at the same time, the scale of death that has struck the Pack Unit is not something this Court dismisses lightly, and it ultimately frames these conclusions of law.

Defendants do not contest that COVID-19 poses a substantial risk to individuals incarcerated at the Pack Unit, a fact that is indisputable given the number of inmates who have died and that there continue to be active cases at the unit. The Court also finds that Defendants were and are aware of that risk, which knowledge can be inferred when "the risk [is] obvious." Farmer , 511 U.S. at 842, 114 S.Ct. 1970. Rather, the dispute centers around Defendants’ mental state—whether they "disregard[ed]" that risk—here, "an excessive risk to inmate health." Id. at 837, 114 S.Ct. 1970.

At the conclusion of trial, this Court is faced with two competing narratives of whether the actions by TDCJ evince deliberate indifference. Defendants argue that a number of actions taken by TDCJ officials since the pandemic began defeat Plaintiffs’ claims. These include the creation and implementation of Policy B-14.52; substantial compliance with CDC guidelines; Mr. Collier's early participation in meetings related to the pandemic; the implementation of asymptomatic testing; Mr. Herrera's implementation of precautionary measures, Policy B-14.52, and other policies; and Defendants’ voluntary compliance with measures ordered by this Court in its preliminary injunction. (Doc. 355 at 58–60.) Defendants additionally argue that any "inept, ineffective, or negligent" actions and "occasional lapses" in enforcement do not rise to the level of recklessness that the Eighth Amendment demands. (Doc. 355 at 60.)

In response, Plaintiffs argue that Defendants’ conduct rises above the level of the occasional misstep and constitute systemic policy failures as well as a pervasive pattern of deficient compliance with what policies do exist. (Doc. 354 ¶¶ 380-81.) This is based on Defendants’ personal knowledge that, at various times, the Pack Unit had the highest number of deaths of any unit within TDCJ (in addition to, at one point in time, the highest number of COVID-19 related grievances of any unit) together with the failure to adopt Policy B-14.52 to the Pack Unit despite its uniquely vulnerable population; the absence of certain "critical steps" to slow the spread COVID-19; and the failure to consistently implement, follow, and ensure compliance with the policies that were in place. Id.

The Court acknowledges that Defendants have taken a number of steps to address the spread of COVID-19, including initial consultations with medical experts and the adoption and implementation of Policy B-14.52 at the Pack Unit. But the Court views these measures as the most basic steps that TDCJ could have taken to prevent mass death within the prison walls on an unimaginable scale. Designing a policy and implementing some of the measures therein does not automatically satisfy Defendants’ constitutional obligations, especially in the face of an unprecedented public health crisis. Moreover, for the reasons explained in its Findings of Fact, the Court, albeit reluctantly, only partially credits Defendants for these efforts. Text messages demonstrating confusion about the methodology behind reporting statistics including a lack of transparency regarding the methodology behind test results; modifications to the Pack Unit made just in time for trial in order to "look more favorable" to this Court; visits to the Pack Unit that were seemingly staged for Defendants’ experts; the fact that grievances were not dealt with promptly even while the death toll mounted; and the fact that the overall guidance was not modified for the Pack Unit whatsoever, even as it continued to be the unit hit the hardest by the pandemic across TDCJ and in spite of its vulnerable population, have all contributed to the Court's skepticism that Defendants are in fact consistently implementing many of the procedures and policies that they claim to be.

After considering the relevant facts, the Court ultimately finds that Defendants’ conduct has demonstrated deliberate indifference to Plaintiffs and the class members’ medical needs by recklessly disregarding obvious and known risks to inmate health and safety. Broadly, this conclusion is based on: (1) Defendants’ lack of a systematic and sustainable approach to slow the spread of COVID-19; (2) a failure to abide by basic public health guidance including but not limited to the steps outlined in Policy B-14.52; and (3) the ongoing risk to inmates coupled with uncertainty that the existing measures will continue absent a permanent injunction.

1. Lack of a systematic approach

The Court first concludes that TDCJ's approach in confronting the pandemic was not systematic and lacked indicia of effecting long-term changes that will be consistently carried out until the pandemic is under control within the state of Texas and the country, which may be months from now. First, the process of designing Policy B-14.52 indicates that it was not responsive to the needs of individual units or TDCJ facilities in general. As the Centers for Disease Control has indicated, prisons "present[ ] unique challenges for control of SARS-CoV-2 transmission."¹¹ Yet TDCJ officials were not involved in designing any part of the policy, and Ms. Davis testified that she did not provide any information to the committees in charge of drafting the policy regarding the characteristics of TDCJ facilities such as the physical space, staffing, the inmate population, or operations. Mr. Herrera testified that he had no authority to make changes to how B-14.52 was implemented at the Pack Unit. The lack of communication between those in charge of drafting TDCJ's public health response and those with knowledge of the specific needs as well as constraints of TDCJ facilities as a whole, let alone the Pack Unit, further indicates TDCJ's failure to take the most basic steps in response to COVID-19.

11 Interim Guidance on Management of Coronavirus Disease 2019 (COVID-19) in Correctional and Detention Facilities , Centers for Disease Control (last updated July 14, 2020), https://www.cdc.gov/coronavirus/2019-ncov/community/correction-detention/guidance-correctional-detention.html.

Second, even at the time of trial—months after the first known infection at the Pack Unit, Defendants had failed to document the long-term testing plan in writing. As Dr. Young testified, this is unusual in medicine and makes it substantially more difficult for TDCJ employees to follow and implement the testing plan as well as for TDCJ officials to know if the plan is being followed. High-level TDCJ officials acknowledged that there was no testing plan put in writing, nor could Mr. Collier commit to putting one in writing in the future. The same problem plagued TDCJ's plan, to the extent it existed, to implement contact tracing. As of July 4, Ms. Davis told Mr. Mendoza that there were no written procedures at all regarding how to carry out contact tracing. (PTX 337 at 2.) And at trial, Mr. Collier confirmed that he was also not aware of any written policies for how TDCJ employees were to carry out contact tracing.

Third, in addition to failing to put essential policies in writing, TDCJ's response to COVID-19 has lacked any kind of consistent audit or compliance regime. As Mr. Collier testified at trial, the compliance team had visited the Pack Unit only once as of the end of trial, and Warden Herrera stated that he was unaware of compliance audits ever being performed at the Pack Unit. As of July 27, Ms. Davis stated in a text message that TDCJ needed a better plan to "start holding wardens accountable," implying that there was little in the way of accountability for compliance before. (PTX 338.) At the same time, Ms. Davis never visited the Pack Unit, despite knowing that the Pack Unit was experiencing among the largest outbreaks within TDCJ. In other words, in addition to failing to put certain key policies in writing, TDCJ officials themselves had no systematic means of measuring or understanding whether policies—written or unwritten—were being followed.

This lack of a compliance regime likely manifested in the fact that Plaintiffs and other witnesses observed multiple lapses in implementing both written and unwritten policies at the Pack Unit. Defendants characterize such "occasional lapses" as possibly indicating negligence, but not deliberate indifference. But at trial, Mr. Collier and Mr. Herrera both acknowledged that the Pack Unit had the highest number of COVID-19-related grievances filed of any of TDCJ's 104 facilities. Moreover, TDCJ officials never investigated why this number (as well as the correspondingly higher number of infections and deaths) was so much higher than at similar facilities such as the Luther Unit. These complaints related to staff members’ failure to wear PPE, a lack of cleaning supplies, the inability to social distance, and exposure to COVID-19 by staff and inmates who had tested positive or were awaiting test results.

The lapses identified in testimony by Pack Unit inmates cannot be described as "occasional" or merely negligent. As Mr. Valentine, Mr. Butaud, Mr. Reynolds, Mr. King, and Mr. Pennington all testified, staff non-compliance with regard to wearing PPE and social distancing were regular, daily features of life in the Pack Unit. This included while correctional staff were cleaning cubicles, passing out meals or simply walking around generally. The inmates were able to recall the details of non-compliance with startling specificity and testified to being spoken to harshly when they pointed out such non-compliance, even by staff members who would later tragically test positive for the virus. (See, e.g. , Tr. 3-13:10–16 (Mr. Reynolds told by Lieutenant Brown to "shut up and go back to [his] bunk" after notifying Mr. Brown that correctional officers were not wearing PPE in the dorms).) These anecdotes are but some of many in which Plaintiffs have shown, as required by the Eighth Amendment, that prison officials "ignored ... complaints" in a way that "clearly evince[d] a wanton disregard for any serious medical needs." See, e.g. , Domino v. Tex. Dep't of Criminal Justice , 239 F.3d 752, 756 (5th Cir. 2001) (quoting Johnson v. Treen , 759 F.2d 1236, 1238 (5th Cir. 1985) ).

Thus, taken together, the lack of policies tailored to the Pack Unit, the lack of written policies, the lack of a compliance regime to ensure compliance with policies, and the predictable result of these forces—consistent non-compliance with basic public health protocols—rises above the level of mere negligence and demonstrates deliberate indifference. As revealed at trial, Defendants and other TDCJ officials were well aware of the shortcomings listed above, and nevertheless chose to stay the course, even after a number of inmates died. Importantly, most of these deficiencies had not been remedied by the time of trial and represent long-term defects in TDCJ's response to COVID-19 that pose an ongoing risk to the Pack Unit. Defendants still could not commit to putting the most basic public health measures such as a plan for testing in writing at any point in the future. Put simply, Defendants were well aware of these obvious policy failures and chose to ignore rather than address them, and their conduct therefore rises to the level of recklessness in violation of the Eighth Amendment.

2. Failure to abide by basic public health guidance

The Court additionally finds that Defendants demonstrated deliberate indifference by failing to take obvious precautionary public health measures upon which all medical professionals would agree. At a very basic level, TDCJ did not commence mass testing—one of the core mechanisms by which COVID-19 may be contained due to the prevalence of asymptomatic and presymptomatic individuals—until May 12, two months after President Trump and Governor Abbott declared COVID-19 to be public health disasters. The testing plan itself was plagued by a number of issues, in addition to never being documented in writing. For example, and perhaps more troubling than the initial delay in implementing testing, there were no plans to retest individuals following the first round of mass testing. This remained the case as of June 4, over four months after Mr. Collier states that he first began discussing TDCJ's response to COVID-19. Defendants explained the delay at trial by arguing that mass testing in the volume needed by TDCJ was simply impossible due to the difficulty of obtaining test kits. Yet TDCJ had the testing capacity, through UTMB with whom it already had a working relationship, to carry out at least some mass testing in mid-April, based on messages exchanged between UTMB and Dr. Linthicum. (Tr. 15-28:13-32:9 (Young)). Further, perhaps as a result of the plan not being documented, testing was not consistently carried out according to the seven-day schedule that was purportedly set, especially in the latter weeks of June and early weeks of July, until trial began.

TDCJ also used defective tests. Both Plaintiffs’ and Defendants’ experts agreed that the long turnaround time for the Curative tests—between one and two weeks—was simply too long to effectively contain the spread of the virus. TDCJ's policy of continuing to house individuals together whose tests were pending rather than maintaining them in medical isolation almost certainly contributed to the spread of disease, as individuals who would subsequently test positive were interacting with the general Pack Unit in the meantime. Defendants did not explore the possibility of using tests with shorter turnaround times, in the range of 24 to 48 hours (as advertised on Curative's website), despite knowing of this possibility. Given the widespread, general knowledge that asymptomatic transmission is one of the ways COVID-19 spreads so quickly, the failure to explore other testing options and continuing—through and after trial—to use tests that TDCJ knows have limited to know effectiveness in containing disease spread, demonstrates an obvious disregard of a known risk. And in addition to having a turnaround time that was too slow to be effective, the tests themselves were only approved under the FDA's Emergency Use Authorization and had not been approved for testing of asymptomatic individuals. Both the FDA and Dr. Young testified at trial that negative results from the Curative tests should have been retested, yet TDCJ nevertheless chose to use Curative tests for mass asymptomatic testing—precisely the use case they were not designed for.

Apart from deficiencies in its testing plan, TDCJ failed to consider or implement other basic public health measures, particularly with regard to disinfecting the dorms and other common spaces, the issuance of PPE, and social distancing. Inmate testimony about the laundry exchange routine highlight many of these flaws. All dorms in the main building use the same laundry window, but inmates testified that they had never seen the window ledge being cleaned. Moreover, inmates entering the showers had already taken off their masks when they received clean clothes from other, masked inmates. Inmate testimony likewise revealed that the showers were not cleaned between use by different dorm units. Tellingly, despite presumably needing to disinfect the Pack Unit more regularly, TDCJ did not increase the number of janitors in response to COVID-19, and the inmate who worked as janitors were never trained on any new cleaning regimen after the pandemic began. Despite TDCJ's representations, multiple inmates testified to not receiving a sufficient amount of cleaning supplies (and no more than prior to the pandemic), even after having requested the same.

The practice of requiring mobility-impaired individuals to be janitors additionally calls into question whether Defendants were and are keeping TDCJ sanitized to the minimum extent required to avoid the spread of COVID-19. As both Mr. Jones and Mr. Dove testified, TDCJ required individuals with serious physical disabilities to perform janitorial work throughout the pandemic. Mr. Jones is in a wheelchair and has a condition in his right hand that prevents him from being able to hold anything, including a mop. Mr. Dove similarly is wheelchair bound and paralyzed on one side of his body. In addition to these disabilities, Mr. Dove is blind. Nevertheless, Mr. Dove was assigned to be a janitor, and two months after his grievance regarding his job was supposedly addressed and well over three months into the pandemic, Mr. Dove had not been reassigned. When asked at trial whether it was appropriate for such individuals to be made to work as janitors, Mr. Wilder responded that he was unconcerned because Mr. Jones or Mr. Dove "could put a broom against his neck and push it in a wheelchair." (Tr. 11-239:13–19 (Wilder).) Such a statement is a textbook example of actual knowledge coupled with deliberate indifference or a reckless disregard for the basic health and safety of inmates, which depends on surfaces being regularly cleaned. To be sure, occasionally lapses in the cleanliness of prisons likely would not rise to the level of an Eighth Amendment violation in normal times. But employing disabled individuals who patently cannot carry out janitorial duties, after being put on notice in the midst of a global pandemic, surely does.

Next, the CDC has described social distancing as the "cornerstone of reducing transmission of respiratory diseases such as COVID-19."¹² It is undisputed that strict social distancing did not occur within the Pack Unit, either in common spaces or in the dorms. Instead, Defendants argue that "TDCJ implemented social distancing as much as operationally possible in a correctional environment." (Doc. 355 at 9.) Yet TDCJ ignored the most basic steps to increase social distancing within the communal-living setting of the Pack Unit. For example, at no point did TDCJ ever even potentially consider using authorized early release as a means to increase social distancing. TDCJ left empty for a month two dorms in E-Wing, which were under construction when the pandemic began but complete by early April. TDCJ also did not implement head-to-foot sleeping, a policy recommended by the CDC. While it is true that at least one inmate expressed discomfort at the idea, TDCJ did not discuss the possibility of implementing the policy generally with exceptions for individuals who had legitimate concerns about the policy. In common spaces, dorms were fed two at a time, two inmates were seated at each table, and this increased to three and four inmates per table at various points. The Pack Unit took steps to visually demarcate social distancing in common spaces with red tape markings on the floors only in July, months after the pandemic began. Finally, the social distancing policies that were in place were continually violated by TDCJ staff, as observed by inmates on a daily basis.

12 Interim Guidance on Management of Coronavirus Disease 2019 (COVID-19) in Correctional and Detention Facilities , Centers for Disease Control (last updated July 14, 2020), https://www.cdc.gov/coronavirus/2019-ncov/community/correction-detention/guidance-correctional-detention.html.

TDCJ's response to the need for additional handwashing capacity—one of the key ways to contain the spread of COVID-19—has also been inadequate to inmates’ needs. Multiple inmates testified that sinks in the Pack Unit, at times more than half of existing sinks, were broken for much of the pandemic. TDCJ installed temporary handwashing stations only in the days leading up to and the early days of trial. As late as July 15, the third day of trial, Ms. Davis was texting TDCJ employees to get "hand washing stations prepped and in place" and to send her pictures. (PTX 337 at 1.) Nowhere did TDCJ provide evidence that installing handwashing stations on that particular schedule was in response to any kind of written plan, rather than in response to the ongoing trial. In fact, communications between Mr. Mendoza and TDCJ employees demonstrate that at least eight other facilities had obtained multiple handwashing stations as of July 9, 2020. (PTX 342 at 2.) Yet TDCJ had not done so at the Pack Unit, which houses one of TDCJ's most vulnerable populations and had already experienced a massive outbreak by that time.

Finally, as of trial, TDCJ had no plan to carry out contact tracing, which is another measure identified by the CDC and other public officials in mitigating the spread of COVID-19 once positive cases are found. And the unwritten plan that did exist was not carried out consistently, as indicated by the barely-filled out forms introduced into evidence at trial as well as communications between TDCJ officials asking for clarifying information regarding the contact tracing plan, the response to which was that no written plan existed.

In sum, Plaintiffs have identified a pattern of policy failures coupled with implementation and enforcement failures in response to COVID-19 that constitute deliberate indifference to the medical needs of inmates.

b. Americans with Disabilities Act and Rehabilitation Act

The Americans with Disabilities Act ("ADA") and the Rehabilitation Act ("RA") have the "same legal standard[ ]" and allow for the "same remedies." Kemp v. Holder , 610 F.3d 231, 234 (5th Cir. 2010). In order to prevail on a claim under either, a plaintiff must show: "(1) that he has a qualifying disability; (2) that he is being denied the benefits, services, programs, or activities for which the public entity is responsible, or is otherwise discriminated against by the public entity; and (3) that such discrimination is by reason of his disability." Hale v. King , 642 F.3d 492, 499 (5th Cir. 2011). The ADA places an "affirmative obligation" on the state "to make reasonable accommodations" for persons with disabilities in the provision of public services. Smith v. Harris Cty. , 956 F.3d 311, 317 (5th Cir. 2020) (quoting Bennett-Nelson v. La. Bd. of Regents , 431 F.3d 448, 454 (5th Cir. 2005) ); accord Tennessee v. Lane , 541 U.S. 509, 531, 124 S.Ct. 1978, 158 L.Ed.2d 820 (2004) ("[F]ailure to accommodate persons with disabilities will often have the same practical effect as outright exclusion."). The ADA protects people incarcerated in state prisons. Pa. Dep't of Corr. v. Yeskey , 524 U.S. 206, 213, 118 S.Ct. 1952, 141 L.Ed.2d 215 (1998).

Defendants argue that the ADA does not apply here because ADA claims are not available under "exigent circumstances." Hainze v. Richards , 207 F.3d 795, 801 (5th Cir. 2000) ; (see Doc. 355 at 55). The authority Defendants cite is inapposite. Hainze dealt with a situation in which a police officer responding to a reported disturbance had to "instantaneously identify, assess, and react to [a] potentially life-threatening situation[ ]" as he arrested the plaintiff, who was holding a knife. Id. at 801. That court emphasized the narrowness of its decision by "simply hold[ing] that [an ADA] claim is not available under Title II under circumstances such as presented herein." Id. Of course, the situation at hand is "life-threatening" (to both inmates and correctional officers), but it is not an "exigent circumstance" that relieves the government of its duty to abide by the ADA within the narrow bounds of the Hainze exception, let alone for months on end or indefinitely. If anything, the pandemic presents a situation in which the state must be more sensitive, not less, to what are ultimately the heightened public health needs of inmates.

Plaintiffs argue that TDCJ's decision to not distribute hand sanitizer to inmates—in particular, the Mobility-Impaired Subclass—effectively denies those individuals the services of medical treatment, proper hygiene, and safe conditions of confinement. (Doc. 354 ¶ 386–87.) Defendants contend that its decision to do so was based on security concerns—the possibility that inmates would either ingest the sanitizer or that the sanitizer would be used as an accelerant. But hand sanitizer could surely be provided to members of the Mobility-Impaired Subclass in daily, incremental quantities too small for misuse. The Court finds TDCJ's decision to deny members of the Mobility-Impaired Subclass access to hand sanitizer and failure to reasonably accommodate their disability, given the testimony at trial regarding their inability to practice appropriate hand hygiene—a medical necessity in light of the pandemic—using only soap-and-water handwashing stations. Indeed, not only would such an accommodation be reasonable; it would likely be life-saving in certain instances. In fact, the decision to deny hand sanitizer to mobility-impaired individuals also contravenes basic public health guidance, as explained by Dr. Young and Dr. Zawitz at trial. Whereas Defendants read public health guidance to recommend issuing hand sanitizer only where handwashing with soap and water is impossible, the Court agrees with testimony at trial explaining that the proper approach is that using both in conjunction is much more effective especially where, as here, individuals cannot readily access regular handwashing where needed.

The Court also finds Defendants’ security rationale unpersuasive. Jails and prisons across the country, including large jail systems such as the Cook County Jail in Chicago Illinois and the California prison system, have managed to safely do so. In fact, units within TDCJ such as the Roach Unit allow inmates to handle hand sanitizer for staff use. TDCJ officials even acknowledged it would be feasible to issue hand sanitizer to inmates and nevertheless made the deliberate choice not to. Balanced against potential security risks was the very real risk that mobility-impaired individuals who could not easily access sinks or the temporary handwashing stations would contract COVID-19, as indeed many did.

Defendants further argue that the decision not to issue hand sanitizer to disabled individuals cannot be discriminatory because Policy B-14.52 applies to all Texas prisons. (Doc. 355 at 61.) But this argument misunderstands TDCJ's obligations under the ADA. As explained above, the ADA imposes an "affirmative obligation" to ensure that disabled individuals have equal access to services; it does not merely require the state to treat disabled individuals the same as it treats non-disabled individuals. By virtue of their disability, individuals who must use wheelchairs or walkers to ambulate are being denied access to the basic service of being able to keep their hands clean. Thus, TDCJ's decision to not issue hand sanitizer to members of this subclass constitutes a failure to reasonable accommodate their disabilities in violation of the ADA and RA.

Plaintiffs have demonstrated actual success on the merits for both their Eighth Amendment as well as their ADA and RA claims for the Mobility-Impaired Subclass. The Court now turns to the remaining factors that govern whether it should grant a permanent injunction.

ii. Irreparable Harm

Next, a plaintiff must prove that he or she will be irreparably harmed by Defendants’ conduct. There is no doubt that Plaintiffs have been irreparably harmed in the past, given the number of infections and deaths at the Pack Unit. But a "permanent injunction is appropriate only if a ‘defendant's past conduct gives rise to an inference that, in light of present circumstances, there is a " ‘reasonable likelihood’ of future transgressions." ’ " Life Partners Holdings, Inc. , 854 F.3d at 784 (quoting Gann , 565 F.3d at 940 ). In other words, injunctive relief must remedy a "real and immediate threat of future or continuing injury apart from any past injury." The Ark. Project v. Shaw , 775 F.3d 641, 648 (5th Cir. 2014). Still, while "insufficient alone," "past wrongs may help establish the threat of future injury." Id. (citing O'Shea v. Littleton , 414 U.S. 488, 495–96, 94 S.Ct. 669, 38 L.Ed.2d 674 (1974) ).

The Court finds that the risk to inmates at the Pack Unit is continuing and imminent, based on Defendants’ past conduct as well as representations made during trial about future conduct. Many of the issues in TDCJ's response to COVID-19 identified in this Court's analysis of Plaintiffs’ Eighth Amendment and ADA claims not only posed risks to inmate health in the past but do so on an ongoing basis, for four reasons. First, the lack of written policies or a compliance or audit regime to ensure enforcement of policies means there is a high likelihood of non-compliance with basic public health procedures in the future, just as there has been prior to and during trial. Second, many of the policies that TDCJ implemented were done so either on the eve of or during trial. The Court does not have confidence that, without an injunction in place, TDCJ will continue to carry out policies that are appropriate to safeguard inmates’ health and safety. Third, the existing policies are inadequate to protect inmates going forward. The track record at the Pack Unit speaks for itself—twenty deaths and a 40% infection rate is more than sufficient for the court to infer "future injury" from "past wrongs," if TDCJ represents, as it did multiple times during trial, that it has no major plans to change or improve its current approach to COVID-19. Id. And fourth, although the Court comes to this conclusion with the utmost regret, the credibility of the representations made by certain TDCJ officials and witnesses has been placed in doubt.

In closing at trial, counsel for Defendant argued that the TDCJ policies were working, given the relatively lower number of active cases in the unit now than in May, June, or July. But even setting aside the issue of whether the test results are accurate, whether enough tests are being performed, and whether TDCJ's policies are sufficient or are being strictly followed, the Court fears that Defendants have lulled themselves into a false sense of security. As reported on TDCJ's website, weeks after trial, another 14 active cases were found at the Pack Unit. The website now reports 4 active cases among inmates, indicating that the pandemic continues to spread, even if not with the speed it did initially. And as this relentless pandemic has demonstrated repeatedly, the time it may take for the Pack Unit to suffer another outbreak could be a matter of just a few days or a week of failing to follow health precautions. As Mr. Collier acknowledged at trial, due to the unrelenting spread of COVID-19 in the state of Texas since the pandemic began, "we have a bigger threat of COVID now than we did in March" and in particular, there is a very real risk that employees might contract COVID-19 and bring it back to the prison. (Tr. 10-7:16–21 (Collier).) Yet TDCJ plans to discontinue strike team testing if they have three weeks with zero positive cases. Given that COVID-19 was necessarily first introduced into the Pack Unit from outside the prison, such a plan fails to account for the possibility, indeed likelihood, that COVID-19 will be introduced into the prison once again with no means of detecting it. Accordingly, the Court finds that the inmates at the Pack Unit continue to be at risk of irreparable harm in the form of serious illness or death, and a permanent injunction is warranted.

iii. Balance of Harm to Parties and the Public Interest

Finally, the Court considers the balance of harm to the parties and whether the injunction is in the public interest. These prongs of the injunction analysis "merge when the Government is the opposing party," "because the government's interest is the public interest." Pursuing America's Greatness v. FEC , 831 F.3d 500, 511 (D.C. Cir. 2016) (quoting Nken v. Holder , 556 U.S. 418, 435, 129 S.Ct. 1749, 173 L.Ed.2d 550 (2009) ) (emphasis in original).

Even in ordinary times, "public health" is a "significant public interest[ ]." Grand River Enters. Six Nations v. Pryor , 425 F.3d 158, 169 (2d Cir. 2005). District courts have consistently recognized that this is a particularly acute consideration when considering issues relating to COVID-19 in prisons or detention centers, due to the fact that prisons and jails have been among the largest sources of outbreaks in the country. See, e.g. , Malam , 452 F.Supp.3d at 662 ("Protecting public health ... is in the public interest."); Hartman v. Acton , No 2:20-CV-1952, ––– F.Supp.3d ––––, ––––, 2020 WL 1932896, at *11 (S.D. Oh. Apr. 21, 2020) ("In considering ... the public interest, ‘a court must at the very least weigh the potential injury to the public health.’ " (citation omitted)); Thakker v. Doll , 451 F.Supp.3d 358, 372 (M.D. Pa. 2020) ("Efforts to stop the spread of COVID-19 and promote public health are clearly in the public's best interest."); cf. Basank v. Decker , 449 F. Supp. 3d 205, 216 (S.D.N.Y. 2020) ("In the highly unusual circumstances posed by the COVID-19 crisis, the continued detention of aging or ill civil detainees does not serve the public's interest."). Thus, curbing the spread of COVID-19 within the Pack Unit is of critical importance not only to the inmates who live there but to the public more generally.

Courts have also recognized that "it is always in the public interest to prevent the violation of a party's constitutional rights." Melendres v. Arpaio , 695 F.3d 990, 1002 (9th Cir. 2012) (quoting G & V Lounge, Inc. v. Mich. Liquor Control Comm'n , 23 F.3d 1071, 1079 (6th Cir. 1994) ); see also Gannett Co. v. DePasquale , 443 U.S. 368, 383, 99 S.Ct. 2898, 61 L.Ed.2d 608 (1979). Given that the Court has determined that Defendants’ conduct is violative of the Constitution, it is in the public interest for the Court to enjoin Defendants’ conduct.

The harm to Plaintiffs if an injunction is not put in place has been described in detail above, and it outweighs the harm to TDCJ. In the past, Defendants have argued that an injunction would be "unduly burdensome, ... waste resources, and set a precedent for courts to micro-manage the operations of prisons during a pandemic." (Doc. 36 at 33.) But at trial, Defendants did not provide evidence of budgetary or financial concerns with the relief requested by Plaintiffs. Even if they had, the Fifth Circuit has made clear that a state agency's "fiscal interests" do not outweigh the "public interest" of "safeguarding public health." Planned Parenthood of Gulf Coast, Inc. v. Gee , 862 F.3d 445, 471–72 (5th Cir. 2017) ; accord Laube v. Haley , 234 F. Supp. 2d 1227, 1252 (M.D. Ala. 2002) ("The threat of harm to the plaintiffs cannot be outweighed by the risk of financial burden or administrative inconvenience to the defendants."). The Court additionally fails to see how the injunction would be unduly burdensome to TDCJ. Several of the measures in the injunction are measures that TDCJ itself has acknowledged constitute medically necessary responses to COVID-19 and some are measures that TDCJ is already purporting to carry out. The Court merely seeks to ensure that TDCJ is in fact implementing many of these measures, at least some of which the Court remains concerned were done in preparation for trial but for which there is no written plan to continue in the future. And balanced against the "burden" of complying with the injunction is the obvious fact that taking public health precautions to keep inmates safe also keeps TDCJ employees safe, many of whom have already fallen sick and continue to be at high risk of being exposed to the virus. The injunction thus benefits not only Plaintiffs but, to a large extent, Defendants as well and the communities in which TDCJ's employees live.

Finally, the Court does not seek to micro-manage TDCJ nor become "enmeshed in the minutiae of prison operations." Bell v. Wolfish , 441 U.S. 520, 562, 99 S.Ct. 1861, 60 L.Ed.2d 447 (1979). It recognizes that courts should be hesitant before rushing to regulate prison administration, where the state is traditionally owed a great measure of deference. At the same time, the Court reiterates that it must balance legitimate penological concerns with the fundamental principle that the "[p]rison walls do not form a barrier separating prison inmates from the protections of the Constitution." Turner v. Safley , 482 U.S. 78, 84, 107 S.Ct. 2254, 96 L.Ed.2d 64 (1987). At bottom, federal courts have a "duty to protect constitutional rights." Id. Accordingly, the injunction contains relief that is tailored to the matter at hand, based on extensive trial testimony including from public health experts, yet allows for latitude in implementation while aligning Defendants’ conduct with their constitutional obligations.

* * * * *

In imposing this injunction, the Court is fully sensitive to the potential objections to it.¹³ The injunction may be seen as micro-management of the state's conduct, or a burden to the government's budget, or as assuming a responsibility that should be left for the legislature. Against all that is the simple proposition that we must not treat with deliberate indifference those whom we have chosen to imprison. The Supreme Court has clearly established that obligation and has done so in circumstances far less severe than the circumstances to which the Pack Unit inmates are now exposed. Asking less of prison authorities—so as to avoid micro-management, or a budgetary impact, or an intrusion on legislative terrain—could easily translate into more lives lost and more inmates sickened. The difficulties that may be endured by prison authorities are modest relative to the harm to be averted.

13 As set forth previously, the PLRA, interpreted correctly with the Constitution, cannot be understood as prohibiting judicial relief while inmates are dying. Defendants’ arguments otherwise are unpersuasive. (Tr. 18-159:20–160:11.)

IV. PERMANENT INJUNCTION

For the above reasons, the Court finds that Plaintiffs have met the burden of demonstrating that permanent injunctive relief is warranted in this case. The Court accordingly GRANTS Plaintiffs’ request for a permanent injunction, and it is ORDERED that Defendants, their agents, representatives, and all persons or entities acting in concert with them are enjoined as follows:

• Provide unrestricted access to hand soap and clean (regularly washed) or disposable hand towels to facilitate frequent handwashing;

• Provide members of the Mobility-Impaired Subclass access to hand sanitizer that contains at least 60% alcohol;

• Provide sufficient cleaning supplies for each housing area, including bleach-based cleaning agents and CDC-recommended disinfectants; provide additional cleaning supplies as requested by inmate janitors; train janitors on additional cleaning practices to be carried out in light of COVID-19;

• Provide new (either disposable or washed) gloves and masks each time inmates perform new tasks, such as beginning a janitorial shift or working in the laundry exchange;

• Create a plan to allow for regular cleaning of common surfaces with bleach-based cleaning agents;

• Create a plan to allow for regular cleaning of the cubicles of inmates who are physically unable to do so themselves;

• Enforce social distancing and the wearing of PPE among TDCJ staff;

• Mark common spaces with red tape to denote safe social distancing practices;

• Create a plan for inmates to sleep head-to-foot with exceptions for legitimate concerns by individual inmates;

• Use common spaces for temporary housing of inmates without disabilities;

• Limit transportation of inmates in and out of the Pack Unit other than for medical appointments or release from custody;

• Create a comprehensive weekly testing program using tests that are approved by the FDA for asymptomatic testing and with a turnaround time for results of 48 hours or less, and document that plan in writing;

• Continue weekly testing until the pandemic is brought under control within the state of Texas, even if multiple weeks pass with zero positive cases;

• Quarantine inmates who are awaiting test results from individuals who are known to have tested negative;

• Create a written plan to implement contact tracing when an inmate or staff member tests positive;

• Document in writing all TDCJ policies in response to COVID-19; and

• Institute a regular audit and compliance program to ensure compliance with the measure in this injunction and other written policies in response to COVID-19.

This injunction shall go into effect 15 days from today, on October 14, 2020.

IT IS SO ORDERED. Signed at Houston, TX on September 29, 2020.