Braggs v. Dunn

257 F. Supp. 3d 1171

• Court: District Court, M.D. Alabama
• Decided: June 27, 2017
• Docket: CIVIL ACTION NO. 2:14cv601-MHT (WO)
• Status: Published

Opinion

LIABILITY OPINION AND ORDER AS TO PHASE 2A EIGHTH AMENDMENT CLAIM

Myron H. Thompson, UNITED STATES DISTRICT JUDGE

I. INTRODUCTION 1180

II. PROCEDURAL BACKGROUND 1180

III. FACTUAL BACKGROUND 1181

A. ADOC Facilities and Organizational Structure 1181

B. MHM Organizational Structure 1183

C. Summary of Factual Findings 1184

1. Fact Witnesses 1184

2. Expert Witnesses 1186

IV. EIGHTH AMENDMENT LEGAL STANDARD 1188

V.FINDINGS OF FACT AND CONCLUSIONS OF LAW 1189

A. Serious Mental-Health Needs 1189

B. Actual Harm and Substantial Risks of Serious Harm Posed by Inadequate Care 1191

1. Contributing Conditions 1193

a. Overcrowding 1193

b. Mental-Health Understaffing 1194

c. Correctional Understaffing 1197

2. Identification and Classification of Prisoners’ Mental-Health Needs 1201

a. Inadequate Intake Process 1201

b. Inadequate Referral Process 1203

c. Inadequate Classification of Mental-Health Needs 1204

d. Inadequate Utilization of Mental-Health Units 1205

3. Inadequate Treatment Planning 1206

4. Inadequate Psychotherapy 1208

5. Inadequate Inpatient Care 1212

a. Improper Use of Mental-Health Units 1212

b. Inadequate Out-of-Cell Time and Programming 1213

c. Lack of Hospital-Level Care 1217

6. Inadequate Suicide Prevention and Crisis Care 1218

a. Failure to Provide Crisis Care, to Those Who Need It 1221

b. Placement of Prisoners in Crisis in Dangerous and Harmful Settings 1224

c. Inadequate Treatment in Crisis Care 1225

d. Unsafe Crisis Cells 1226

e. Inadequate Monitoring of Suicidal Prisoners 1229

f. Inappropriate Release from Suicide Watch and Inadequate Follow-up 1230

7. Inappropriate Use of Disciplinary Actions 1231

8. Inappropriate Placement and Inadequate Treatment in Segregation 1235

a. Background on Segregation 1236

i. Consensus among Correctional and Mental-Health Professionals on Segregation ' 1236

ii. ADOC’s Segregation Units 1238

b. ADOC’s Segregation of Mentally Ill Prisoners 1240

i. ADOC’s Segregation-Placement Practices 1240

ii. Treatment and Monitoring in Segregation Units 1242

c. Segregation of Prisoners with Serious Mental Illness 1245

9. Tutwiler 1247

10. Other Issues 1249

C. Deliberate Indifference 1250

1. ADOC’s Knowledge of Harm and Risk of Harm 1252

2. ADOC’s Disregard of Harm and Risk of Harm 1255

a. ADOC’s Failure to Exercise Oversight of the Provision of Mental-Health Care 1257

b. ADOC’s Unreasonable Responses to Identified Deficiencies 1260

D. Ongoing Violation 1262 .

E. Ex parte Young Defenses 1265

VI. CONCLUSION 1267

I. INTRODUCTION

The plaintiffs in this phase of this class-action lawsuit are a group of seriously mentally ill state prisoners and the Alabama Disabilities Advocacy Program (ADAP), which represents mentally ill prisoners in Alabama. The defendants are the Commissioner of the Alabama Department of Corrections (ADOC), Jefferson Dunn, and the Associate Commissioner of Health Services, Ruth Naglich, who are sued only in their official capacities. The plaintiffs assert that the State of Alabama provides constitutionally inadequate mental-health care in prison facilities and seek injunctive and declaratory relief. They rely on the Eighth Amendment, made applicable to the States by the Fourteenth Amendment and as enforced through 42 U.S.C. § 1983. Jurisdiction is proper under 28 U.S.C. § 1331 (federal question) and § 1343(a)(3) (civil rights).

After a lengthy trial, this claim is now before the court for resolution on the merits. Upon consideration of the evidence and arguments, the court finds for the plaintiffs in substantial part. Surprisingly, the evidence from both sides (including testimony from Commissioner Dunn and Associate Commissioner Naglich as well as that of all experts) extensively and materially supported the plaintiffs’ claim.

II.PROCEDURAL BACKGROUND

This extremely complex case has been split into three phases: Phase 1 involved claims under Title II of the Americans with Disabilities Act (ADA), codified at 42 U.S.C. § 12131 et seq., and § 504 of the Rehabilitation Act, codified at 29 U.S.C. § 794, claiming discrimination on the basis of physical disabilities and failure to accommodate those disabilities. The parties settled Phase 1. See Dunn v. Dunn, 318 F.R.D. 652 (M.D. Ala. 2016) (Thompson, J.). Phase 2A involves Eighth Amendment, ADA, Rehabilitation Act, and due-process claims regarding mental-health care. The parties settled the Phase 2A ADA and Rehabilitation Act claim. The due-process claims are pending before the court for settlement approval.¹ Phase 2B will focus on medical-care and dental-care claims under the Eighth Amendment,

This opinion resolves only the Phase 2A Eighth Amendment claim of inadequate mental-health care.² The court has certified a Phase 2A plaintiff class consisting of all persons with a serious mental illness who are, or will be, confined within ADOC’s facilities, excluding Tutwiler Prison for Women and the work-release centers. See Braggs v. Dunn, 317 F.R.D. 634 (M.D. Ala. 2016) (Thompson, J.). While mentally ill prisoners at Tutwiler are not part of the class, ADAP, as Alabama’s designated protection and advocacy organization for the mentally ill, brought claims on their behalf A seven-week trial followed.

III. FACTUAL BACKGROUND

Mental-health care in this opinion refers to screening, treatment, and monitoring of mental illnesses, as well as ADOC’s policies and practices regarding mentally ill prisoners, including decisions on disciplinary sanctions and housing placements.³ Before diving in to the details of weeks’ worth of testimony and thousands of pages of documentary evidence regarding mental-health care within ADOC, the court pauses to provide some background information on ADOC and its mental-health contractor, as well as a summary of the factual findings.

A. ADOC Facilities and Organizational Structure

ADOC runs 15 major facilities (14 for men and the Tutwiler Prison for Women) and houses around 19,500 prisoners in its major facilities.⁴ Approximately 3,400 prisoners are on the mental-health caseload, meaning that they receive some type of mental-health treatment, such as counseling or psychotropic medications.

MAJOR ADOC FACILITIES⁵

Facility Location ¡Population

Bibb Brent 1847

Bullock Union Springs 1522

Donaldson Bessemer 1474

Draper Elmore 1144

Easterling. Clio 1457

Elmore Elmore 1186

Fountain Atmore 1242

Hamilton Hamilton 275

Holman Atmore • 941

Kilby Mt. Meigs 1126

Limestone Harvest 2214

St. Clair Springville 975

Staton Elmore 1382

Ventress Clayton 1254

Tutwiler Wetumpka 880

Three of the major facilities, Bullock, Donaldson, and Tutwiler, serve as ‘treatment hubs’ for mental-health services, containing a residential treatment unit (RTU) and/or a stabilization unit (SU). These two types of units, together referred to as ‘mental-health units’ or ‘inpatient-care units,’ house and treat the most severely mentally ill prisoners. The rest of those on the mental-health caseload receive their care through outpatient services: they live in a unit that is not focused on treatment and ordinarily must go to a different part of the prison to see a mental-health provider.

Under the administrative regulations governing ADOC’s mental-health care, RTUs are for mental-health patients who suffer from “moderate impairment in mental health functioning” that puts them at risk in a general-population setting. Joint Ex. 107, Admin. Reg. § 613-2 (doc. no. 1038-130). RTUs are intended to provide a therapeutic environment to mentally ill patients and to help them develop coping skills necessary for placement in general population. RTUs can be ‘closed,’ meaning that each patient lives in an individual cell with little time spent outside the cell; ‘semi-closed,’ meaning that the patient still stays in an individual cell but is let out of the cell more often; or ‘open,’ meaning that the patient lives in an open dormitory with other RTU patients. ■

SUs are for patients who are suffering from acute mental-health problems — such as acute psychosis or other conditions causing an ■ acute risk of self-harm — and have not been stabilized through other interventions. SUs are intended to stabilize the patient as. quickly as possible so that the patient can return to a less restrictive environment. All SU patients are housed in individual cells.

Altogether, the two male treatment hubs have 346 RTU beds and 30 SU beds: Bullock has 250 RTU beds and a 30-bed SU for male prisoners, and Donaldson has an additional 96-bed RTU. Tutwiler has 30 RTU beds and eight SU beds for women. These units provide services to about 2 % of ADOC’s overall population.

ADOC is headed by Commissioner Dunn. Associate Commissioner for Health Services Naglich heads the Office of Health Services (OHS), which is responsible for overseeing the provision of medical and mental-health care to prisoners. ADOC uses private contractors to deliver medical and mental-health care services to prisoners. Under the mental-health contract with a third-party vendor, OHS has access to the contractor’s internal documents and records, and the contractor is required to send certain reports, such as monthly operating reports and annual contract-compliance reports, to OHS. The only OHS staff member with mental-health expertise is Dr. David Tytell, the chief clinical psychologist. Dr, Tytell serves as the main liaison between the mental-health contractor and ADOC, and communicates with the contractor’s program director at least weekly. ADOC also directly employs ‘psychological associates,’ who are counselors responsible for conducting certain psychological tests at intake and for providing group sessions and classes for non-mentally ill prisoners. They report to their respective facilities’ wardens, rather than OHS or the mental-health contractor.

B. MHM Organizational Structure

MHM Correctional Services, Inc. is ADOC’s contractor for mental-health care. MHM is a for-profit corporation that provides medical and mental-health services to correctional facilities across the country.

MHM’s regional office in Alabama is headed by its program director Teresa Houser. She serves as the main liaison between ADOC and MHM- Dr. Robert Hunter, a psychiatrist who serves as the medical director for the Alabama regional office, is charged with supervising psychiatrists and certified registered nurse practitioners (CRNP) stationed at various ADOC facilities. Both Houser and Hunter communicate frequently with ADOC officials, including Associate Commissioner Naglich and Dr. Tytell.

MHM employs a variety of administrative and clinical personnel to fulfill its contract with ADOC. In its regional office, Houser supervises various administrators and managers, such as the continuous quality improvement (CQI) manager, who conducts informal audits of MHM’s performance, and the chief psychologist, who supervises psychologists and conducts training for MHM employees. At the facility level, MHM employs site administrators to provide administrative oversight; these administrators are counselors by training. MHM also employs approximately 45 full-time ‘mental-health professionals’ (MHPs), who are masters-level mental-health counselors, at prisons across the State. As of December 2016, MHM employed four psychiatrists and eight CRNPs in Alabama; these providers are qualified to diagnose mental illnesses, prescribe psychotropic medication, and provide psychotherapy across multiple facilities. MHM also employs three psychologists and three.registered nurses (RNs) for the entire State. The RNs are stationed at the three treatment hubs, Bullock, Donaldson, and Tut-wiler; they administer medication, provide crisis intervention, and supervise the licensed practical nurses (LPNs) at their facilities. MHM employs approximately 40 LPNs, individuals with 12 to Í5 months of health-care training. The LPNs are responsible for conducting mental-health intake at Kilby and Tutwiler, monitoring medication compliance, maintaining medication records, and conducting side-effects monitoring tests for psychotropic medications. While the LPNs stationed in the mental-health treatment units are supervised by the on-site RN, at all other places, including at intake screening, LPNs have no on-site supervision. Lastly, MHM employs six to eight activity technicians, who organize or assist in therapeutic, social, and recreational activities for patients in mental-health units.

C.- Summary of Factual Findings

1. Fact Witnesses

Over the course of seven weeks, the court heard testimony as to whether ADOC’s mental-health care violates mentally ill prisoners’ constitutional rights. The trial opened with the testimony of prisoner Jamie Wallace, who suffered from severe mental illnesses, intellectual disability, and substantial physical disabilities. Wallace stated that he had tried to kill himself many times, showed the court the scars on arms’ where he made repeated attempts, and complained that he had not received sufficient treatment for his illness. Because of his mental illness, he became so agitated during his testimony that the court had to recess and reconvene to hear his testimony in the quiet of the chambers library and then coax him into completing his testimony as if he were a fearful child. The court was extremely concerned, by what it had seen and heard from this plaintiff, about the fragility of his mental health. At the end of Wallace’s testimony and out of his presence, the court informed the attorneys for both sides that it wanted a full .report on his mental condition and the steps that were being taken to address that condition. Unfortunately, and most tragically, ten days after Wallace testified, he killed himself by hanging. Because it appeared that adequate measures may not have been put in place to prevent Wallace’s suicide, the court put the parties into mediation to attempt to come up with immediate, interim procedures to prevent future prisoner suicides. The parties eventually came up with such procedures. Without question, Wallace’s testimony and the tragic event that followed darkly draped all the subsequent testimony like a pall.

The plaintiffs’ case then proceeded with testimony from Commissioner Dunn, who aptly described the prison system as wrestling with a “two-headed monster”: overcrowding and understaffing. Dunn Testimony at 26. The court also heard from Associate Commissioner Naglich and MHM’s program director Houser, for whom overcrowding and understaffing (both as to correctional staff, as noted by Dunn, and mental-health staff) were a mantra. They, with admirable candor, as with many other fact witnesses and the experts from both sides, essentially agreed that the staffing shortages, combined with persistent and significant overcrowding, contribute to serious systemic deficiencies in the delivery of mental-health care.

The inadequacies in the mental-health care system start at the door, with intake screening for prisoners who need mental-health care. ADOC boasts one of the lowest mental-illness prevalence rates among correctional systems in the country. But this is not because Alabama has fewer mentally ill prisoners than the rest of the country or the best mental-health care system for its prisoners; rather, according to experts from both sides, this is because a substantial number — likely thousands — of prisoners with mental illness are missed at intake and referrals for evaluation and treatment are neglected. As a result, many ADOC prisoners who need mental-health care go untreated.

Even when identified, mentally ill prisoners receive significantly inadequate care. Mental-health and correctional staffing shortages drive inadequate treatment. Individual and group counseling sessions are delayed or canceled due to shortages of counselors and correctional officers to escort prisoners to the sessions and to provide security. As a result, mental-health staff often have to resort to cell-side contacts, which cannot be considered substitutes for meaningful, confidential, out-of-cell appointments. Treatment planning is often pro forma and not individualized and fails to provide a meaningful and consistent course of treatment. Mental-health units intended as a therapeutic environment for the most severely ill prisoners operate like segregation units, with little counseling, therapeutic programming, or out-of-cell time. ADOC does not provide hospital-level care for those who need it.

ADOC also fails to provide adequate care to prisoners expressing suicidality and undergoing mental-health crises. Mental-health staff fail to use appropriate risk-assessment tools to determine suicide risk. ADOC has an insufficient number of crisis, or ‘suicide-watch,’ cells — special cells for the protection of suicidal prisoners. Because they have a limited number of cells to work with, they gamble on which prisoners to put in them and frequently discount prisoners’ threats of self-harm and suicide. The insufficient number of crisis cells also results in the use of unsafe rooms such as shift offices to house suicidal prisoners. The suicide-watch cells that do exist are dangerous: visibility into many of the cells is poor, making it difficult to monitor; many cells have tie-off points for ligatures that can be used for suicide attempts; dangerous items used for inflicting self-injury are often found. Prisoners in these cells receive less contact with and less monitoring by providers than the acuity of their condition demands. When they are released to general population or segregation, prisoners receive inadequate follow-up.

ADOC’s segregation practices inflict further harm on prisoners suffering from inadequate mental-health care. Due to the effects of isolation, placement in segregation endangers mentally ill prisoners, and the risk of harm increases with the length of isolation and the severity of their mental illness. This danger is compounded by the limited access to mental-health care and monitoring available within ADOC’s segregation units and dangerous conditions inside the cells. Despite these dangers, ADOC does not have a meaningful mechanism that prevents mentally ill prisoners from being placed in segregation for lengthy periods of time. Moreover, many mentally ill prisoners land in segregation due to symptoms of mental illness. This combination of conditions is often deadly: most suicides in ADOC occur in segregation.

For years, ADOC has failed to respond reasonably to these problems. Despite knowledge of serious and widespread deficiencies, it has failed to remedy known problems and exercised very little oversight of its mental-health care contractor. Associate Commissioner Naglich, who is in charge of contract monitoring, admitted that she has been aware of the contractor’s deficient performance and inadequate quality-control process; however, she does not monitor the contractor to ensure that it provides minimally adequate care. Moreover, ADOC officials admitted on the stand that they have done little to nothing to fix problems on the ground, despite their knowledge that those problems may be putting lives at risk.

The psychological and sometimes physical harm arising from these systemic deficiencies is palpable. Unidentified and. under-treated mental illness causes needless pain and suffering in the form of persistent or worsening symptoms, decompensation,⁶ self-injurious behavior, and suicide. The skyrocketing suicide rate within ADOC in the last two years is a testament to the concrete harm that inadequate mental-health care has already inflicted on mentally ill prisoners.

In fact, as explained earlier, the court had a close encounter with one of the tragic consequences of inadequate mental-health care during the trial. Over the course of the trial, two prisoners committed suicide, one of whom was named plaintiff Jamie Wallace. Prior to his suicide, defendants’ expert, Dr. Patterson, concluded based on a review of Wallace’s medical records'that the care he had received was inadequate. Dr. Haney, a correctional mental-health care expert, met Wallace months before his death, while he was housed in a residential treatment unit, and in his report expressed serious concerns about the care he was receiving,⁷ Wallace’s case was emblematic of multiple systemic deficiencies. Wallace testified, and his records reflected, that mental-health staff did not proyide much in the way of consistent psychotherapeutic treatment, which is distinct from medications administered by nurses and cursory ‘check-ins’ with staff. MHM clinicians recommended that he be transferred to a mental-health hospital, but ADOC failed to do so. His psychiatrist at the time of his death testified that the medically appropriate combination of supervised out-of-cell time and close monitoring when he was in his cell was unavailable due to a shortage of correctional officers. As a result, Wallace was left alone for days in an isolated cell in a treatment unit, where he- had enough time to tie a sheet unnoticed; because his cell, was not suicide-proof, he was able. to find a tie-off point from which to hang himself.

The case of Jamie Wallace is powerful evidence of the real, concrete, and terribly permanent harms that woefully inadequate mental-health care inflicts on mentally ill prisoners in Alabama. Without systemic changes that address these pervasive and grave deficiencies, mentally ill prisoners in ADOC, whose symptoms are no less real than Wallace’s, will continue to suffer.

2. Expert Witnesses.

Plaintiffs and defendants presented five experts in the correctional mental health and correctional administration fields.⁸ By and large, experts from both sides agreed that ADOC facilities are suffering from severe systemic deficiencies that are affecting the delivery of mental-health care. For example, experts from both sides agreed that ADOC suffers' from severe overcrowding; correctional understaffing; mental-health staff shortages; deficient treatment planning; inadequate psychotherapy; inadequate use of mental-health units; inappropriate placement of segreg'ation inmates; and inappropriate use of segregation for mentally ill prisoners.

Defendants’ correctional mental-health care expert, Dr. Raymond Patterson, is a forensic psychiatrist who has worked for various state and federal correctional institutions as a provider and as a consultant. In preparation for his testimony, he reviewed the individual plaintiffs’ medical records and deposition transcripts, visited and conducted audits of six facilities, and reviewed ADOC regulations, MHM policies and procedures, MHM monthly reports, and other expert reports. His conclusions regarding systemic deficiencies in ADOC’s mental-health care system-largely tracked those of Dr. Kathryn Burns, one of the plaintiffs’ experts: he credibly concluded that ADOC needs more mental-health staff; ADOC’s identification and classification of mental illness are inadequate; MHM’s unlicensed practitioners should be supervised; treatment planning is, deficient; too few patients are getting inpatient care; ADOC should provide hospitalization as an option for the most severely ill patients; and suicide prevention measures are inadequate.⁹

Defense expert Robert Ayers is a correctional administration expert who has been involved in the California prison system for over 40 years. In preparation for giving his opinion, Ayers reviewed plaintiffs’ expert reports, visited six facilities, and talked with ADOC and MHM staff during those visits. He agreed with plaintiffs’ experts that ADOC facilities are understaffed and overcrowded. He opined that ADOC’s written policies related to mental-health care seemed to be adequate. However, he credibly explained that, mainly due to the severe understaffing and the lack of documentation, he had reasons to doubt that correctional officers and mental-health staff were actually complying with ADOC policies and procedures. He also concluded that ADOC was not providing an adequate level of care to all prisoners with mental-health needs.

Dr. Kathryn Burns, the chief psychiatrist for the Ohio Department of Rehabilitation and Correction, is a correctional mental-health expert for the plaintiffs. To prepare for her testimony, Dr. Burns visited nine major ADOC facilities, touring housing units, mental-health treatment areas, and crisis cells; she held formal interviews with 77 prisoners and spoke to an additional 25 prisoners at cell-front; she also reviewed documents such as medical records, ADOC regulations, MHM’s quality-improvement (or ‘continuous quality improvement’ or ‘CQI’) and multidisciplinary-team meeting minutes, suicide tracking sheets, and audit results. Based on her review of this evidence, she identified a wide range of problems in the delivery of mental-health care, including: insufficient mental-health staffing and correctional staffing; inadequate identification and classification of mental illness; inadequate treatment, including cursory counseling appointments, inadequate treatment plans, dearth of group counseling, and inadequate use of mental-health units; and inadequate response to self-injurious behavior and mental-health crises. Dr. Burns credibly opined that these inadequacies, separately and taken together, subject mentally ill prisoners to a substantial risk of harm from untreated symptoms, continued pain and suffering, decompensation, self-injurious behavior, and suicide.

Dr. Craig Haney, a professor of psychology at the University of California Santa Cruz, is an expert for the plaintiffs in the psychological effects on prisoners of incarceration and particularly of segregation. His testimony focused on the state of segregation units and their impact on prisoners’ mental health, based on his visits to seven facilities, interviews with numerous prisoners, and review of documents such as deposition transcripts of ADOC and MHM personnel, medical records, monthly statistical reports, and quality-assurance documents, among others. He testified that segregation units he saw were “degraded, dilapidated, deplorable,” and that these units and conditions have a significant negative psychological impact on prisoners. Haney Testimony at vol. 1, 79. Furthermore, he explained how ADOC’s segregation practices harm mental health of all prisoners, and especially that of prisoners who are already mentally ill.

Lastly, plaintiffs’ expert Eldon Vail is a correctional administration expert who has worked in corrections for over 30 years. Vail toured seven prisons, spending a day at each, and conducted confidential interviews with 42 prisoners. He also reviewed ADOC policies and procedures, meeting minutes, reports and logs generated by ADOC, deposition testimony of ADOC and MHM personnel, and other documentary evidence. His testimony focused on matters of prison administration, including security, staffing, and behavior management, and the impact of these factors on the provision of mental-health care and on prisoners’ mental health. He credibly testified that the level of correctional under-staffing at ADOC was so low as to be “shocking,” and that it has cascading effects on mental-health care: inadequate staff to transport prisoners to appointments and supervise treatment activities; inadequate staff to monitor segregation inmates, who have higher suicide risks; and overcrowded crisis cells filled with prisoners who feel unsafe due to violence in general-population dorms. Vail Testimony at vol. 1, 34.

IV. EIGHTH AMENDMENT LEGAL STANDARD

The Eighth Amendment’s prohibition on “cruel and unusual punishments” extends to a State’s failure to provide minimally adequate medical care that “may result in pain and suffering which no one suggests would serve any penological purpose.” Estelle v. Gamble, 429 U.S. 97, 103, 97 S.Ct. 285, 50 L.Ed.2d 251 (1976); Harris v. Thigpen, 941 F.2d 1495, 1504 (11th Cir. 1991) (“Federal and state governments ... have a constitutional obligation to provide minimally adequate medical care to those whom they are punishing by incarceration.”). The State’s obligation to provide medical care to prisoners includes psychiatric and mental-health care. Rogers v. Evans, 792 F.2d 1052, 1058 (11th Cir. 1986) (“Failure to provide basic psychiatric and mental-health care states a claim of deliberate indifference to the serious medical needs of prisoners.”). The ‘basic’ mental-health care that States must provide if needed by a prisoner includes not only medication but also psychotherapeutic treatment. See Greason v. Kemp, 891 F.2d 829, 834 (11th Cir. 1990) (“Even if this case involved failure to provide psychotherapy or psychological counselling alone, the court would still conclude that the psychiatric care was sufficiently similar to medical treatment to bring it within the embrace of Estelle.”). The State’s obligation remains even if it has contracted with private parties to provide medical care. West v. Atkins, 487 U.S. 42, 56, 108 S.Ct. 2250, 101 L.Ed.2d 40 (1988). That is, the State is liable for the contractor’s unconstitutional policies and practices if the contractor is allowed to determine policy either “expressly or by default.” Ancata v. Prison Health Servs., Inc., 769 F.2d 700, 706 n.11 (11th Cir. 1985).

To prevail on an Eighth Amendment challenge, plaintiffs must prove that prison officials acted with deliberate indifference to serious medical needs. Estelle, 429 U.S. at 105-06, 97 S.Ct. 285. This inquiry consists of both objective and subjective tests. The objective test requires showing that the prisoner has “serious medical needs,” Estelle, 429 U.S. at 104, 97 S.Ct. 285, and either has already been harmed or been “incarcerated under conditions posing a substantial risk of serious harm.” Farmer v. Brennan, 511 U.S. 825, 834, 114 S.Ct. 1970, 128 L.Ed.2d 811 (1994). Subjectively, a prisoner must show that a prison official acted with deliberate indifference to that harm or risk of harm: that is, the official must have “known[ ] of and disregarded[ ] an excessive risk to inmate health or safety.” Id. at 837, 114 S.Ct. 1970; see also Farrow v. West, 320 F.3d 1235, 1245 (11th Cir. 2003).

V. FINDINGS OF FACT AND CONCLUSIONS OF LAW

In this section, the court first discusses the basis for its finding that the plaintiffs have serious mental-health needs that require mental-health treatment. The court then lays out the common factors contributing to .the substantial risks of harm in ADOC: shortages of mental-health staff, understaffing of correctional officers, and overcrowding. After that, the court proceeds through seven different ways in which ADOC’s mental-health care system has caused actual harm and a substantial risk of serious harm; the treatment of mentally ill prisoners at Tutwiler; issues on which the court does not, at this time, find for the plaintiffs; and the defendants’ knowledge of such harm and risks, and their failure, to act in- a. reasonable manner to mitigate those risks. The section concludes with a discussion of the defendants’ legal defenses .based on Ex parte Young, 209 U.S. 123, 28 S.Ct. 441, 52 L.Ed. 714 (1908).

A.. ¡Serious Mental-Health Needs

To prove an Eighth Amendment claim based on inadequate mental-health care, plaintiffs must show that they have serious mental-health care needs. A serious need is “one that has been diagnosed by a physician as mandating treatment or one that is so obvious that even a lay person would easily recognize the necessity for a doctor’s attention.” Farrow v. West, 320 F.3d 1235, 1243 (11th Cir. 2003). Thus, courts may find the existence of serious needs even when prison staff have failed, to recognize an inmate’s need for treatment. Danley v. Allen, 540 F.3d 1298, 1310-11 (11th Cir. 2008) (finding that plaintiff, whose requests to see a nurse had been rebuffed, demonstrated a serious medical need in that he had difficulty breathing and swollen, burning eyes, and a fellow inmate brought his condition to the attention of correctional officers), overruled on other grounds, Randall v. Scott, 610 F.3d 701, 709 (11th Cir. 2010). A serious mental-health care need was found where a doctor, nurse, and correctional officials recognized that a prisoner “engaged in self harm” and “showed outward signs of mania and depression.” Jacoby v. Baldwin Cty., 596 Fed.Appx. 757, 763 (11th Cir. 2014).

One of the factors that courts consider in finding a serious medical need is “whether a delay in treating the need worsens it.” Danley, 540 F.3d at 1310. “The tolerable length of delay in providing medical attention depends on the nature of the medical need and the reason for the delay.” Hill v. Dekalb Reg’l Youth Det. Ctr., 40 F.3d 1176, 1188 (11th Cir. 1994) (citation omitted). Factors relevant to determining the tolerable length of delay include the “seriousness of the medical need,” “whether the delay worsened the medical condition,” and “the reason for delay.” Id. at 1189.

Because this is a Rule 23(b)(2) class action lawsuit challenging defendants’ actions “on [a] ground[ ] that applies] generally to the class” — that is, defendants’ provision of inadequate mental-health care— the plaintiffs must show that serious mental-health needs exist on a system-wide basis, rather than on an individual basis.¹⁰ Fed. R. Civ. P. 23(b)(2). As explained in the class-certification opinion, the plaintiffs’ claim and the remedies they seek are systemic. Braggs v. Dunn, 318 F.R.D. 652, 667 (M.D. Ala. 2016). In other words, “plaintiffs are not seeking adjudication of demands for particular individualized treatment,” and any relief the court grants “would be appropriate for everyone subjected to the substantial risk of serious harm plaintiffs claim [ADOC’s inadequate mental-health care system] creates — that is, prisoners with serious mental illness.” Id. at 668.

It is clear that a number of prisoners in ADOC’s custody have serious mental-health needs, and the issue is undisputed. As a preliminary matter, MHM places prisoners on the caseload only if they have been diagnosed with a condition that requires treatment. Therefore, all prisoners on the caseload meet the legal requirement for having a serious mental-health need. Prisoners on the mental-health caseload have wide-ranging illnesses, such as bipolar disorder, schizophrenia, schizoaffective disorder, major depressive disorder, mood disorders, borderline personality disorder, anxiety, and PTSD.¹¹

Furthermore, the court heard testimony from multiple prisoners, both named plaintiffs and class members, who clearly exhibited serious mental-health needs. For example, plaintiff R.M. has been diagnosed with paranoid schizophrenia and admitted that he is out of touch with reality; he testified to what were obviously his delusions regarding his blood relationships to three different well-known terrorist figures and his owing billions of dollars to the United States treasury. Similarly, medical records made clear that plaintiff Q.B. has suffered from years of delusion and hallucination; he was on involuntary psychiatric medication orders for years while in ADOC custody. Lastly, as explained earlier, plaintiff Jamie Wallace¹² had been diagnosed with bipolar disorder and schizophrenia, among other mental-health conditions, and he testified that he heard voices of his deceased mother telling him to cut himself. In sum, plaintiffs presented more than sufficient evidence establishing their serious mental-health needs.

Because only prisoners "with serious mental-health needs have a cognizable Eighth Amendment claim, when the court refers to ‘mentally ill prisoners’ in this opinion, it is referring to only those with serious mental-health needs.

B. Serious Harm and Substantial Risks of Serious Harm Posed by Inadequate Care

In addition to showing a serious medical need, plaintiffs must establish that they have been subjected to serious harm, or a substantial risk of serious harm — the second part of the ‘objective’ test under the Eighth Amendment jurisprudence — as a result of inadequate mental-health care. Put another way, plaintiffs must show that their serious medical need, “if left unattended, ‘poses a substantial risk of serious harm.’” Farrow v. West, 320 F.3d 1235, 1243 n.13 (11th Cir. 2003)(quoting Farmer v. Brennan, 511 U.S. 825, 834, 114 S.Ct. 1970, 128 L.Ed.2d 811 (1994)). Defendants may be held liable for “incarcerating prisoners under conditions posing a substantial risk of serious harm.” Farmer, 511 U.S. at 834, 114 S.Ct. 1970.¹³

The “serious harm” requirement “is concerned with both the ‘severity’ and the ‘duration’ of the prisoner’s exposure” to the harm, such that an exposure to harm “which might not ordinarily violate the Eighth Amendment may nonetheless do so if it persists over an extended period of time.” Chandler v. Crosby, 379 F.3d 1278, 1295 (11th Cir. 2004) (citation omitted). While mere discomfort is insufficient to support liability, id., “unnecessary pain or suffering” qualifies as serious harm. LaMarca v. Turner, 995 F.2d 1526, 1535 (11th Cir. 1993).

Plaintiffs may bring an Eighth Amendment challenge to a condition that is already inflicting serious harm on them at the time of the complaint or to prevent serious harm which is substantially likely to occur in the future — a substantial risk of serious harm. As the Supreme Court explained in Helling v. McKinney, 509 , 113 S.Ct. 2475, 125 L.Ed.2d 22 (1993), a case in which a prisoner challenged his prolonged exposure to secondhand smoke, “a remedy for unsafe conditions need not await a tragic event,” because “the Eighth Amendment protects against future harms to inmates,” even when the harm “might not affect all of those exposed” to the risk and even when the harm would not manifest itself immediately. Id. at 33-34, 113 S.Ct. 2475. In other words, plaintiffs must show “that they have been subjected to the harmful policies and practices at issue, not (necessarily) that they have already been harmed by these policies and practices.” Dunn v. Dunn, 219 F.Supp.3d 1100, 1123 (M.D. Ala. 2016)(Thompson, J.). In the class-action context, the plaintiff class must show that it, as a whole, has been subjected to policies and practices that create a substantial risk of serious harm. Braggs v. Dunn, 317 F.R.D. 634, 654 (M.D. Ala. 2016)(Thompson, J.).

Moreover, multiple policies or practices that combine to deprive a prisoner of a “single, identifiable human need,” such ás mental-health care, can support a finding of Eighth Amendment liability. Gates v. Cook, 376 F.3d 323, 333 (5th Cir. 2004) (“Conditions of confinement may establish an Eighth Amendment violation ‘in combination’ when each would not do so alone, but only when they have a mutually enforcing effect that produces the deprivation of a single, identifiable human need such as food, warmth, or exercise — for example, a low cell temperature at night combined with a failure to issue blankets.”)(citing Wilson v. Seiter, 501 U.S. 294, 304, 111 S.Ct. 2321, 115 L.Ed.2d 271 (1991)). The Eleventh Circuit Court of Appeals has recognized this ‘totality of conditions’ approach in prison-conditions cases. See, e.g., Hamm v. DeKalb Cty., 774 F.2d 1567, 1575-76 (11th Cir. 1985).

Mentally ill ADOC prisoners, defined here as prisoners with serious mental-health needs, have suffered harm and are subject to a substantial risk of serious harm due to ADOC’s inadequate mental-health care. Based on the trial testimony, the court finds seven interrelated areas of inadequacy: (1) identification and classification of prisoners with mental illness; (2) treatment planning; (3) psychotherapy; (4) inpatient mental-health care units; (5) crisis care and suicide prevention; (6) use of disciplinary actions for symptoms of mental illness; and (7) use of segregation for mentally ill prisoners. In all seven areas, experts from both sides by and large agreed about significant flaws affecting mentally ill prisoners.¹⁴ MHM and ADOC staff also recognized and corroborated the existence and severity of these issues. Even Associate Commissioner Naglich essentially agreed that some of these were problems so significant that they must be fixed as soon as possible, because lives are at risk.¹⁵ These inadequacies, alone and in combination, subject mentally ill prisoners to actual harm and a substantial risk of serious harm — including worsening of symptoms, increased isolation, continued pain and suffering, self-harm and suicide.

1. Contributing Conditions

Three conditions contribute to all of the deficiencies in ADOC’s treatment of mentally ill prisoners: understaffing of mental-health care providers, understaffing of correctional officers, and overcrowding.¹⁶ Associate Commissioner Naglieh and defendants’ expert witnesses largely agreed with plaintiffs that these conditions present significant challenges to the system today. Correctional and mental-health un-derstaffing, both alone and in combination, impose substantial risks of serious harm to mentally ill prisoners, and overcrowding compounds these risks.

a. Overcrowding

ADOC facilities are significantly and chronically overcrowded. Publically available information on ADOC’s inmate population and capacity plainly lays out the magnitude of overcrowding: ADOC’s September 2016 monthly statistical report states that ADOC held 23,328 prisoners in facilities that are designed to hold only 13,318; this brings the' occupancy rate to over 175 %. PI. Ex. 1260, September 2016 Monthly Statistical Report (doc. no. 1097-19) at 2, 4.¹⁷ Plaintiffs’ expert Vail testified that the magnitude of overcrowding in ADOC is the worst he has seen in his career in corrections and consulting for other correctional systems across the country. According to Vail, California, whose overcrowded correctional system was found to be unconstitutional, approached an occupancy rate of 170 %; a three-judge court subsequently ordered the State to lower the occupancy rate to 137.5 %, a target rate that was affirmed by the Supreme Court. Brown v. Plata, 563 U.S. 493, 539-42, 131 S.Ct. 1910, 179 L.Ed.2d 969 (2011). The sheer magnitude of overcrowding within ADOC has meant that some ADOC facilities, including Kilby, Bibb, Staton, and Easterling, house more than double the number of prisoners they are designed to hold. PI. Ex. 1260, September 2016 Monthly Statistical Report (doc. no. 1097-19) at 4. Even maximum-security facilities use open-bay dormitories filled with wall-to-wall rows of double bunk beds, holding up to 240 prisoners in a single room, where officers do not have a line of sight on" most of the prisoners they are assigned to supervise.

b. Mental-Health Understaffing

ADOC has maintained mental-health staffing levels that are chronically insufficient across disciplines and facilities. Witness after witness identified significant mental-health staffing shortages as one of the major reasons for ADOC’s inability to meet the rising mental-health care needs of prisoners. Most significantly, Associate Commissioner for Health Services Naglich admitted that MHM has been understaffed since 2013 and remains understaffed today. MHM’s program director Houser stated bluntly that MHM staffing shortages make it difficult to “do the work required under the contract,” and that the current caseload for MHM' staff does not meet an “acceptable standard.” Houser Testimony at vol. 2, 24-25.

Over the course of the trial, evidence showed that the mental-health caseload per MHM provider has been increasing since ,2008; largely due to three reasons: (1) an increasing number of prisoners with mental-health needs across ADOC; (2) multiple budget cuts over the years; and (3) ADOC’s long-time refusal to increase the, authorized number of mental-health staff positions despite repeated requests from MHM, even when an initiative to transfer some of the caseload to ADOC staff-so-called ‘blending of services’ — was not implemented as planned.¹⁸

ADOC’s prisoner population has had increasing needs for mental-health services over the last decade. As múltiple MHM providers and expert witnesses from both sides testified, ADOC’s prisoner population has become more mentally ill over the last decade, both in terms of the number of individuals who need mental-health care and in terms of the acuity of mental-health care needs. MHM’s medical director, Dr. Hunter, testified that, the number of prisoners receiving regular mental-health services within ADOC (also known as being ‘on the caseload’) has been’ increasing since 2003, which has been “concerning” and “tax[ing his] ability to adequately do” what he is required to. do under the contract. Hunter Testimony at_. (For transcripts that are not yet finalized, the court leaves the page numbers blank.) He also explained that, since 2003, the number of prisoners coming into the system with severe mental illness has been increasing. MHM’s own documents showed that between 2008 and 2016, the mental-health caseload increased by 25 % across all facilities. PI. Dem. Ex. 25, Pricing, Caseload and Staffing Comparison Over Time (doc, no. 1071-5).

As the need for mental-health services has been increasing substantially, MHM and ADOC have been hiring fewer and fewer providers over the years, exacerbating the staffing shortage. In 2009, ADOC reduced MHM’s compensation under the contract and the number of authorized positions to be hired by MHM. In 2013, the state legislature further reduced ADOC’s mental-health care budget by 10 %. ADOC and MHM then re-negotiated their 2013 contract to reduce the previously agreed-upon “minimum required staffing,” cutting close to 20 full-time equivalent positions. Naglich Testimony 2-211; PI. Dem. -Ex. .140, MHM Staffing Increase Chart (doc. no. 1148-59); see also PI, Dem. Ex. 25, Pricing, Caseload, and Staffing Comparison Over Time (doc. no. 1071-5). During that same contract renewal period, ADOC and MHM also reduced the number of positions that are covered by the contractual ’staffing rebate’ provision, under which MHM must pay back ADOC if it does not fill all authorized positions. In other words, the revision allowed MHM to leave clinical staff positions unfilled without being penalized, even though the overall number of authorized positions had already been reduced. Houser described this latter modification as a way to make the reduction in payment and staffing under the contract “more palatable for MHM.” Houser Testimony at vol. 1, 49.

Another driving ■ force behind MHM’s mental-health understaffing is ADOC’s failure to implement the ‘blending of services’ initiative - successfully. Houser explained that this initiative was established in 2009 in response to ADOC’s reduction in both the amount it would pay to MHM under the contract and in the staffing provided for in the contract: MHM’s caseload would be reduced 'by transferring treatment of prisoners with lower-acuity mental-health issues to ADOC’s psychological associates; the initiative was an “attempt to make sure that the inmates received mental health services” despite the staffing reduction and increasing caseloads. Houser Testimony at vol. 1, 14. However, ADOC failed to implement the initiative across its facilities: MHM’s staffing was reduced, but at many facilities, psychological associates did not take over any caseload from MHM. Naglich explained that, because some wardens were resistant to letting psychological associates carry significant caseloads, MHM staff remained responsible for most of the patients, even though there were now fewer MHM providers than before. Houser testified that blending of services is not currently happening anywhere in ADOC in the way it was designed to happen, despite MHM’s reduced staffing levels. 'ADOC’s chief clinical psychologist Dr. David Tytell admitted that the initiative, has failed to work. However, ADOC has not restored MHM’s staffing to the pre-2009 level.¹⁹

The result of ADOC’s refusal to increase MHM’s staffing level or even to restore staffing to the pre-2009 level has been chronic shortages of mental-health care providers. Dr. Hunter testified that the staffing shortage has had a significant impact on scheduling of psychiatric visits and medication management. Several mental-health counselors testified that their caseloads have soared; Houser testified that MHP caseloads at some facilities have been twice what they should be, which is “never an acceptable standard.” Houser Testimony at vol. 2, 25. Increasing caseloads due to understaffing have also led to a high turnover rate among staff: according to Houser, staff resign because of their frustration with increasing caseloads, leaving the rest of the staff with even higher caseloads; recruiting also suffers because of the overwhelming caseloads that mental-health staff are expected to manage. MHM’s monthly operating report submitted to ADOC for May 2016 described the problem in stark terms: “Mental health caseloads are running high at many of the facilities. Staff has attempted to accommodate the increased numbers, however quality cannot be maintained at current staffing levels.” Joint Ex. 343 (doc. no. 1038-702) at 19. As explained in more detail in the following sections, this understaffing also has prevented MHM from providing care that complies with ADOC’s administrative regulations, the contract, and professional standards for minimally adequate care in a prison system.²⁰

Not surprisingly, experts from both sides opined that ADOC does not have a sufficient number of mental-health staff for a system of its size. Dr. Patterson, the defense expert, concluded based on his review of medical records and site visits that ADOC’s mental-health care system is significantly understaffed. Plaintiffs’ expert Dr. Burns agreed with this assessment based on her review of medical records and MHM internal records, which revealed that caseloads for psychiatric providers and counselors were too large to allow for sufficient counselling or therapeutic group activities. Dr. Burns concluded that ADOC needs more psychiatric staff, psychologists, registered nurses, and activity technicians.²¹

MHM’s corporate office — which exercises contract-compliance oversight but does not directly provide care in Alabama — has repeatedly raised mental-health under-staffing in the annual clinical contract-compliance review reports (hereafter ‘contract-compliance reports’) sent to Associate Commissioner Naglich’s Office of Health Services. Starting in 2011, each annual contract-compliance report included information on multiple facilities that were suffering from staffing shortages, “compromising [MHM’s] ability to provide monthly follow-up for all caseload inmates.” PL Ex. 1190, 2011 Contract-Compliance Report (doc. no. 1070-8) at 15. The 2013 report also noted the impact of the staffing reduction that year, stating that “[d]espite the increase in the size of the caseload across ADOC, MHM’s contract has been compressed to include significant staffing cuts at all sites.” PI. Ex. 114, 2013 Contract-Compliance Report (doc. no; 1070^4) at 1. The report also warned that, at Donaldson, where one of the two male residential treatment units is located, “[c]urrent staffing pattern does not support the delivery of adequate services to inmates and that they have been reduced to providing minimal and ‘triage-based’ services rather than effective and thoughtfully planned treatment.” Id. at 5. In 2016, MHM reported significant backlogs in treatment and staffing shortages at Donaldson and Bullock, the two male facilities that house ADOC’s most seriously ill mental-health patients. PI. Ex. 115, 2016 Contrach-Compliance Report (doe. no. 1070-5). Even after the partial staffing increase in September 2016, Houser stated that MHM remains understaffed and pointed to mental-health un-derstaffing as a cause for a plethora of issues, including insufficient identification of mental illness at intake and referrals; missed counseling appointments and group sessions; and inadequate monitoring of prisoners in mental-health crises.

Based on Associate Commissioner Nag-lich’s testimony and other evidence, the court finds that MHM has been consistently and significantly understaffed at least since 2013, and that it is still understaffed even after ADOC approved a small staffing increase in September 2016 as part of its one-year contract extension.

c. Correctional Understaffing

In addition to mental-health understaff-ing and overcrowding, a significant shortage of correctional officers also hinders the delivery of mental-health care and poses a substantial risk of harm to prisoners who need mental-health care. As with mental-health staffing shortages, witness after witness, including both defendants, testified that a significant shortage of correctional officers has been one of the biggest obstacles to providing mental-health care in ADOC. In Associate' Commissioner Naglich’s words, the problem of insufficient mental-health staffing is “compounded by” the lack of sufficient correctional staffing at ADOC. Naglich Testimony at vol. 2, 208.

ADOC has reported an ever-increasing shortage of correctional officers in its annual reports arid monthly operating reports since 2006. In 2010, ADOC summarized that “[c]orreetional staffing continues to fall- short of required levels — impacting the inmate to officer ratio and overtime necessary to cover essential posts,” and reported that the shortage rate was 12.2 % at close-custody (highest security) facilities and 21.2 % at medium-security facilities. Joint Ex. 463, Vail Expert Report (doc. no. 1038-1048) at 39 (quoting ADOC Annual Report FY 2010). Essentially the same statement regarding the officer shortage appeared in every annual report until 2013, when the shortage rate across facilities shot up to 43.3 %. The report in 2015 showed officer shortage rates of over 25 % at 13 of the 15 major prisons and over 50 % at six of those; the highest was 68 % at Bibb. Donaldson was barely undpr 25 %; only one prison, Hamilton, the facility for the elderly and the' infirmed, was below 25 %. Id. at 39-40 (citing ADOC Annual Report FY 2013, 2015). As of September 2016, ADOC reported having filled only about half of the authorized positions for correctional officers. PI. Ex. 1260, September 2016 Monthly Statistical Report (doc. no. 1097-19) at 16 (showing 51.1 % overall staffing level).²² The staffing level continued to drop throughout 2016, according to Associate Commissioner of Operations Grantt Culliver.

Understaffing has been a persistent, systemic problem that leaves many ADOC facilities incredibly dangerous and out of control. Defendants’ correctional administration expert Robert Ayers observed multiple high-security units not being monitored at all and an entire unit at Bibb overseen by a single control booth officer and a single officer on the floor; he opined that such understaffing was “not acceptable.” Ayers Testimony -at_Plaintiffs’ correctional administration expert Vail agreed with this conclusion and elaborated that many facilities are struggling to have sufficient numbers of correctional officers to station at. least one officer per dorm— including the highest-security facilities, such as Holman and Kilby.. Not surprisingly, a severe .shortage of officers leads to dangerous and violent conditions, especially in high-security facilities, with overcrowded dormitories.²³ In these conditions, prisoners and correctional officers alike are justifiably afraid for their safety — a jarring image that many prisoner-witnesses and experts, painted in their testimony. For example, class member M.P., w;ho is now housed in Ventress, stated repeatedly how dangerous it was to be in a general-population dorm at St. Clair; he was enormously relieved to be transferred to another prison.²⁴ Multiple experts also testified that during their site visits, prison officials did not allow them to enter certain parts of the prison, such as the second and third tiers of the Holman segregation unit and a whole half of Bibb, because the officials could not guarantee their safety.²⁵

As a result of the officer shortage, ADOC has an exceedingly high overtime rate. Overtime rate refers to the, proportion of the number of hours worked by correctional officers as overtime compared to the total number of hours worked. A high overtime rate undermines security and officer morale, which in turn has negative implications for mental-health care. ADOC’s chief of staff Steve Brown admitted that the current overtime rate of over 20 % is not sustainable in the long run, because it decreases retention of officers and increases the number of disciplinary actions against officers. Multiple vulnerability analyses — ADOC’s internal critical assessments of each facility’s security risks — also found that mandatory overtime and overuse of overtime have affected staff morale and contributed to high turnover rates. PI. Ex. 146, Bullock Vulnerability Analysis (doc. no. 1087-3); PI. Ex. 185, Donaldson Vulnerability Analysis (doc. no. 1087-6); PI. Ex. 204 Elmore Vulnerability Analysis (doc. no. 1087-8).²⁶

This chronic and severe correctional un-derstaffing has compromised mental-health care in many ways. Most significantly, as discussed in more detail in Part V.B.4, correctional officers are needed to provide security for mental-health programming and escort prisoners from their cells to appointments if they are not in general population. Due to insufficient correctional staffing, appointments and group activities are frequently canceled- and delayed, significantly impairing MHM staffs ability to provide treatment. See, e.g., PI. Ex. 115,2016-Contract-Compliance Report (doc. no. 1070-5) at 3 ,(MHM staff not being able to access patients:at Bullock, Donaldson, Holman, St. Clair,, and Staton due to correctional staffing shortages, and expressing concern about their own safety at five facilities); PI. Ex. 105, 2014 MHM Implementation Review Report (doc. no. 1070-3)- at 3 (20 to 70 % of mental-health appointments were canceled due to correctional officer shortages at the Donaldson residential treatment unit in 2014). Based on the testimony of Ayers, one of the defense experts, and almost all MHM providers and managers who testified, the court is convinced that the correctional staffing level falls intolerably short of providing adequate care to prisoners who need to be escorted to their mental-health appointments.

Second, understaffing impacts correctional officers’ ability to supervise mentally ill prisoners effectively. According to plaintiffs’ expert Vail, understaffing compromises overworked correctional officers’ alertness and ability to respond to incidents, crises, and emergencies, and to exercise the patience and restraint necessary to supervise mentally ill prisoners. This effect is even more pronounced in segregation and crisis cells. Without sufficient correctional staff, officers are unable to check on prisoners isolated from the rest of the population as frequently as they must in order to guarantee their safety. As a result, decompensating prisoners go unnoticed, leading to extended suffering without access to treatment, and more frequent crisis situations.

Correctional understaffing, combined with overcrowding, also has a more direct impact on prisoners’ mental health. The combination of overcrowding and under-staffing leads to an increased level of violence, both because of the difficulty of diffusing tension and violence in an overcrowded open-dormitory setting, and because of the lack of supervision by correctional officers. See PI. Ex. 1260, ADOC September 2016 Monthly Statistical Report (doc. no. 1108-37) at 12 (reporting nearly 200 assaults with serious injuries and seven homicides in the fiscal year ending in September 2016). According to Dr. Haney, plaintiffs’ expert on correctional mental health and solitary confinement, prisoners’ legitimate fear of violence is a common source of anxiety and mental instability: for prisoners who already suffer from mental illnesses, this environment increases their likelihood of decompensation. The level of danger and lack of control arising from overcrowding and insufficient staffing also contributes to a punitive culture, in which officers prioritize security concerns over mental-health treatment and are quick to treat mental-health symptoms as behavioral problems; dealing with violence and emergencies also diverts correctional resources away from regular mental-health programming and treatment. Untreated or undertreated mental illness in turn creates a greater need for mental-health services, provision of which is limited by the very shortage of officers that created the increased need in the first instance. Furthermore, mental-health problems are much more likely to go unnoticed in overcrowded and understaffed prisons, because correctional officers who are spread too thin are less likely to notice any unusual behavior by a particular prisoner. These observations made by Dr. Haney all rang true in the evidence before the court. Lastly, as Dr. John Wilson, a psychologist who serves as one of the directors of MHM’s national Clinical Operations Department, explained to MHM’s program director Houser, “experience and research” confirm that suicides tend to increase with overcrowding, and “basic unrest at a systems level” can cause a spike in suicides. PI. Ex. 1224, October 1, 2015 Email from Wilson to Houser (doc. no. 1117-24) at 2. In fact, the suicide rate within ADOC has more than doubled in the last two years, as ‘unrest at a systems level’ continues to plague ADOC facilities. Taken together, ADOC’s low correctional-staffing level, in the context of its severely overcrowded prisons, creates a substantial risk of serious harm to mentally ill prisoners, including continued pain and suffering, decompensation, self-injury, and suicide.

2. Identification and Classification of Prisoners’ Mental-Health Needs

As one expert put it, ADOC’s mental-health care system “falls apart at the door”: the system fails to identify and classify appropriately those with mental illnesses, and the effect of this under-identification cascades through the system. Haney Testimony at vol. 1, 30. Because of inadequate identification and classification, seriously mentally ill prisoners languish and decompensate in ADOC without treatment, ending up in crisis care and engaging in destructive — sometimes fatal — self-harm.

Timely identification and appropriate classification of prisoners with mental illness are essential to a functioning mental-health care system. As experts explained, and as common sense would dictate, mental-health treatment cannot begin unless providers are aware of who needs treatment and for what. Failure to identify those who need mental-health services denies them access to necessary treatment, creating a substantial risk of harm to those who remain unidentified. See LaMarca v. Turner, 995 F.2d 1526, 1544 (11th Cir. 1993) (affirming conclusion that systematic denial of access to treatment constitutes deliberate indifference to a serious medical need).

a. Inadequate Intake Process

ADOC’s system for identifying prisoners with mental illness is significantly inadequate. According to three experts — defense expert Patterson and plaintiffs’ experts Burns and Haney — the percentage of prisoners within ADOC with mental illness (referred to as the ‘prevalence rate’) is substantially lower than the national average: the average rate of mental illness for men in correctional systems ranges between 20 % and 30 %; ADOC’s prevalence rate is between 14 % and 15 %. See Joint Ex. 346, June 2016 MHM Monthly Statistical Report (doc. no. 1038-708) at 1.

As experts from both sides testified, ADOC’s prevalence rate is abnormally low and reflects that the system is under-identifying prisoners with mental illness. Defense expert Dr. Patterson explained that experts do not expect to see much variation in actual prevalence rates across correctional systems, and that he has not seen anything that suggests that ADOC would have a lower prevalence rate than other correctional systems for any reason other than under-identification. Dr. Bums agreed and explained that it is highly likely that the abnormally low prevalence rate is due to under-identification, rather than because Alabama prisoners have fewer mental-health issues compared to those in other States. She added that she does not know of any States that have lower prevalence rates than Alabama. Assuming that ADOC’s actual prevalence rate for mental illness actually tracks the national figure of between 20 % and 30 %, somewhere between 1,200 and 3,600 prisoners should be receiving mental-health care but are not, because between 5 % and 15 % of ADOC’s 24,000 prisoners have not been identified as having a mental illness.

A closer examination of the two main processes of identifying prisoners with mental-health care needs — intake and referral — sheds light on why ADOC’s prevalence rate is so low. First, ADOC’s mental-health screening process at intake fails to identify a substantial number of prisoners with mental-health issues. Licensed practical nurses, who have very limited training, are responsible for conducting mental-health screening for prisoners at intake at Kilby (for all male prisoners) and Tutwiler (for all female prisoners). No higher-level provider supervises the LPNs during the intake process. The intake LPN fills out forms and questionnaires and decides whether to refer a prisoner for further examination by a psychiatrist or a nurse practitioner. If the LPN determines that a prisoner does not need to be referred tq a psychiatrist or nurse practitioner, a mental-health code of MH-0, denoting no need for mental-health care, is entered into the system. Prisoners who are designated as MH-0 by an LPN do not receive any further evaluation or any mental-heálth treatment unless reférred to mental-health services later by a staff member or the prisoners themselves. On the other hand, if the LPN refers the prisoner for evaluation, a psychiatric provider completes' an evaluation, gives a diagnosis if appropriate, and assigns a mental-health code, which determines the level of care the prisoner subsequently receives and ranges from MH-0 (no mental-health need) to MH-6 (in need of hospitalization).²⁷

Experts from both sides agreed, and the court finds, that the intake screening process conducted by an LPN without any on-site supervision by a higher-level provider contributes to under-identification of prisoners with mental illness. This is because LPNs, who only have 12 to 15 months of general medical training — very little of which may be related to mental health— are not qualified to assess the presence or acuity of mental illness symptoms based on information obtained during the intake process. Intake forms that LPNs fill out include questions, that require clinical assessments, rather than simple yes-or-no questions based on physical observations. See Joint Ex. 85, Admin. Reg. § 601 Mental Health Forms and Disposition (doc. no. 1038-106);, Burns'Testimony at vól. 1, 44-45. According to the experts, LPNs are not qualified to make such clinical assessments. Moreover, although LPNs may make referrals- -based on self-reported symptoms of mental illness, a proper intake system cannot solely rely on self-reporting to identify mental-health needs. As Dr. Burns testified, the use of unsupervised LPNs for. intake mental-health screening presents an “obvious” risk of under-identification. Burns Testimony at vol.l, 61-62.²⁸.

The use of inadequately supervised LPNs for intake is compounded by insufficient mental-health staffing. Houser testified that MHM does not have sufficient staffing or space to conduct mental-health screenings at Kilby (where all male prisoners are screened), and her staff have had to send prisoners to other facilities without conducting the initial intake screening. This in turn has increased the workload for mental-health staff at the receiving facilities and has created delays in the provision of mental-health care to those who need treatment. Dr. Patterson, the defense expert, agreed that insufficient staff at intake has led to insufficient identification of prisoners with mental illness, and that this failure to identify increases the risk of continued pain and suffering and potential suicides among those who are not receiving the mental-health care they need.

b. Inadequate Referral Process

The other mechanism for identifying and classifying prisoners with mental illness, the referral process, is riddled with delays and inadequacies. The purpose of the referral process is to identify prisoners whose mental illnesses develop during their incarceration and prisoners whose mental-health needs were not identified during the intake process. Furthermore, the referral process enables the system to respond to the changing mental-health needs of prisoners as they arise, regardless of their initial mental-health assessment results. In a functioning system, referrals from prisoners or staff would be triaged based on the urgency of the articulated needs: some may warrant immediate action, such as placement in a suicide-watch cell or an immediate evaluation by a psychiatrist, while others may be addressed over a longer period of time. According to Dr. Patterson, the defense expert, triaging is important because the assessment process enables clinicians to determine appropriate next steps, and delays in doing so pose a risk of untreated symptoms, including a risk of death from critical yet unmet treatment needs.

As with the intake screening procedure, experts from both sides concluded that ADOC’s referral process suffers from serious deficiencies. First, ADOC does not have a system to triage and identify the urgency of each request, and to make referrals according to the level of urgency. MHM’s contract-compliance reports have identified this issue year after year, starting in 2011: the reports stated that processed referral slips did not reflect acuity levels, and the logs of referrals did not record the relevant date and time information, making it impossible to ensure timely processing and referrals. Despite perennial indications that referral requests were being processed in a haphazard manner, ADOC still does not have any system of tracking and processing referrals to ensure that urgent requests are actually referred to providers, or that providers are able to handle requests in a timely fashion: an audit performed- by defense experts in May 2016 revealed that referral forms still do not note urgency levels that would enable triaging.²⁹

Second, the referral process is inadequate because correctional officers are ill-positioned to notice behavioral changes. As ■plaintiffs’ expert Vail testified, severe overcrowding and understaffing make it difficult for correctional officers to notice behavioral changes. It is simply unrealistic to rely on ADOC’s overburdened correctional officers to identify and refer prisoners who may need mental-health treatment, except perhaps for those prisoners with the most obvious symptoms of mental illness.

In addition to delaying treatment or leaving mental-health symptoms untreated, ADOC’s broken referral process has contributed to the phenomenon of prisoners engaging in self-harm or other destructive behavior in order to get attention of mental-health staff. Experts described examples of “increasingly desperate acts” to get the attention of MHM and necessary services, such as self-injury, fire setting, and suicide attempts. Joint Ex. 460, Burns Expert Report (doc. no. 1038-1044) at 29; Haney Testimony at vol. 1, 72 (describing frequent fires in segregation units as desperate attempts to get attention for their needs, including mental-health needs). The court also heard from class member J.A., who has repeatedly engaged in self-harm and expressed suicidal ideation. After summarizing his various attempts to obtain mental-health services while in segregation, including starting fires, J.A. observed, “[Gjetting help in prison is harder than getting out of prison.” J.A. Testimony at _These are snapshots of unnecessary pain and suffering that could be avoided or at least minimized if prisoner requests for mental-health services were being addressed on a timely basis.

c. Inadequate Classification of Mental-Health Needs

ADOC also fails to classify the severity of mental illnesses accurately. The mental-health coding system is intended to reflect the level of functioning a mental-health patient has and correspond to his or her treatment needs and housing requirements. Through multiple revisions, the coding system now includes 13 different codes, ranging from MH-0 tó MH-9, with sub-codes for some levels, such as MH-2d. In broad strokes, a higher numbered MH code reflects more intensive care needs: MH-0 refers to no mental-health care need; MH-1 and MH-2 refer to mild impairment or stable enough to receive only outpatient care; MH-3 through MH-5 refer to those who need inpatient care, in either the residential treatment unit (RTU) or intensive stabilization unit (SU); MH-6 refers to those who need to be hospitalized. See Joint Ex. 105, Admin. Reg. § 613 (doc. no. 1038-127).³⁰

Testimony from multiple witnesses and experts made clear that ADOC’s mental-health coding system often fails to accurately reflect prisoners’ mental-health needs.³¹ For example, plaintiff R.M. has been coded MH-2 and housed in general population for most of his incarceration since 1994, despite his severe paranoid schizophrenia and resulting delusions. He was eventually given a higher code and transferred to the Bullock RTU, but Dr. Burns testified that he may need an even higher level of care, and that he suffered from inadequate care while housed for years in an outpatient facility. Likewise, a prisoner identified as # 12 in Dr. Burns’s report was clearly delusional and believed that televisions and radios were speaking to him; he was in an outpatient facility at the time of his interview with Dr. Burns, but needed to be in a long-term, inpatient facility due to the severity of his schizophrenic symptoms. An email from Associate Commissioner Naglieh to Dr. Hunter in December 2015 discussed a schizophrenic prisoner who was clearly delusional and eventually killed another prisoner and threatened to kill a correctional officer; he had been coded as MH-1, which is intended to denote someone who is stabilized with a ‘mild’ impairment. Lastly, Dr. Haney gave examples of patients who have been repeatedly placed on suicide watch for engaging in self-harm and suicide attempts but were designated as MH-0— that is, not having any mental-health treatment needs — including plaintiffs L.P and R.M.W., and former plaintiff J.D. Haney Testimony at vol. 2, 113-20; see PI. Dem. Ex. 131, Movement History of Exemplar Plaintiffs (doc. no. 1126-10).

d. Inadequate Utilization of Mental-Health Units

As experts from both sides concluded, ADOC does not adequately utilize residential treatment unit beds and fails to provide residential-level care to those who need it, leading to persistent or worsening symptoms. Defendants’ expert Dr. Patterson opined that roughly 15 % of prisoners on the mental-health caseload should be housed in RTU or intensive stabilization unit settings; in -other words, approximately 515 ADOC prisoners should be housed in the RTU or the SU.³² However, only 310 of the 376 RTU and SU beds were being used to house prisoners with mental-health needs as of S¿ptember 2016. Joint Ex. 344, September 2016 Monthly Operating Report (doc. no. 1038-703). This practice of not filling even existing mental-health unit beds has persisted for years, as reflected in MHM’s monthly operating reports. See, e.g.. Joint Ex. 321, December 2015 Monthly Operating Report (doc. no. 1038-666) at ADOC0319118-19; Joint Ex. 320, December 2014 Monthly Operating Report (doc. ho. 1038-665) at ADOC0319016-17 (showing 299 beds occupied in December 2015 and 177 beds occupied in December 2014).³³ Dr. Patterson credibly opined that this significant shortfall suggests ADOC has been under-identifying those who need residential treatment — a problem that starts with the inadequate intake screening, process. He also observed another flaw in RTU admission management: he explained that those who are repeatedly sent to the SU should be admitted, to the RTU to receive more long-term, intensive treatment, rather than being released back to general population after a stay, in the SU. He also npted that prisoners who are admitted to RTUs often stay only for a short period, despite their pronounced needs for long-term treatment. Because there is little programming available in the RTU, the utility of an RTU placement is quickly exhausted, according to Patterson.

Dr. Burns agreed with Dr.. Patterson’s assessment that ADOC needs to house more patients in the RTU, especially when RTUs have available beds. She also observed during her facility visits multiple prisoners who needed residential treatment but were in general population.

In sum, ADOC’s significantly inadequate identification and classification practices create a substantial risk of serious harm to prisoners with mental illness. These practices result in a failure to treat or under-treatment of prisoners’ serious mental-health needs. As will be discussed later, these practices also have a downward-spiral effect on the rest of the system: those who do not get needed treatment often end up in crisis cells, frequently receive disciplinary sanctions, and. may be placed in segregation, where they have . even less access to treatment and monitoring.

3. Inadequate Treatment Planning

Correctional systems have a duty to provide minimally adequate mental-health care to prisoners with serious mental-health needs. Estelle v. Gamble, 429 U.S. 97, 104, 97 S.Ct. 285, 50 L.Ed.2d 251 (1976) (deliberate indifference to serious medical needs of prisoners constitutes “unnecessary and wanton infliction of pain proscribed by the Eighth Amendment”) (internal quotation and citation omitted); Greason v. Kemp, 891 F.2d 829, 834 (11th Cir. 1990) (holding that prisoners have a constitutional right to psychiatric care under Estelle v. Gamble). Expert testimony from both sides established that such minimally adequate care requires treatment planning. Treatment planning is the foundation of all forms of health care; through the process, providers involved in the treatment identify the patient’s target symptoms, treatment goals, and - next steps, and coordinate long-term care as necessary. When staff from multiple disciplines — for example, psychiatric, psychological, nursing, and even correctional— are involved in a patient’s treatment, treatment planning should involve key people from each discipline in order to ensure consistent and informed treatment. Treatment planning is particularly important in the prison context, where prisoners have almost no ability to ensure the consistency of them own treatment; it is even more crucial in the context of, ADOC, where prisoners are frequently transferred across correctional facilities and the staff turnover rate is high. As experts described, without coordinated long-term planning, treatment is often ineffective and runs a substantial risk of prolonging pain and suffering of those who have treatable mental illnesses. Failure to provide meaningful .treatment planning constitutes a substantial deviation from acceptable standards of prison health care; such deviations can pose a substantial risk of serious harm to those who have serious psychiatric needs. Steele v. Shah, 87 F.3d 1266, 1269 (11th Cir. 1996) (noting that providing care where the quality is “so substantial a deviation from accepted standards” can constitute an Eighth Amendment violation).³⁴

ADOC fails to provide adequate treatment planning. First, experts for both sides found that ADOC’s treatment plans are not individualized to each prisoner’s symptoms and needs, résulting in ‘cookie-cutter’ plans that remain the same even though there may have been changes in that prisoner’s mental-health state. As defense expert Dr. Patterson explained, a patient’s lack of progress in treatment ■does not justify the use of a cookie-cutter treatment plan: providers should try different interventions that could be effective, rather than sticking to the same intervention when the patient is not responding to it. Likewise, treatment plans should reflect the changes in the treatment environment, such as an admission to the SU or placement on suicide watch. However, ADOC treatment plans often have general patient goal statements such as “identify triggers” or “identify coping mechanism” repeated in subsequent plans, without showing any progress or change in the mental state of the patient; they also often fail to reflect the fact that the patient has been placed in a different environment' that would impact his or her mental health and treatment mode. Whether the rote repetition results from a lack of follow-through on the plans or mere sloppiness in filling' out the plans, both present hazards to prisoners with mental illness.

ADOC’s treatment-team meetings are also inadequate. Treatment-team meetings are an essential part of the treatment planning process, where providers from various disciplines involved in the patient’s treatment discuss developments and next steps to ensure coordinated care. However, the meetings at ADOC happen haphazardly, with members of the treatment team missing from the meetings and signing new treatment plans on different days. This haphazard attendance creates a risk of different providers having an inconsistent approach or course of treatment for the same patient because some of the treatment .team are unaware that a new treatment plan has been put into effect. Furthermore, the meetings frequently occur without any participant, with prescription privileges, especially at some outpatient facilities where the only provider with prescription privileges is a nurse practitioner who visits the facility as infrequently as pnce per month. As a result, treatment plans are often developed without the input of a provider with expertise in psychotropic medication. Experts from both sides agreed that ADOC’s treatment planning without all necessary participants is problematic and falls below the standard of care because it deprives patients of a coherent treatment plan and- continuity of care.

Inadequate treatment planning subjects mentally ill prisoners to the risk of exacerbating symptoms, prolonged pain and suffering, serious injury from self-harm, and even death. As Dr. Burns explained, treatment plans serve an essential function of making sure that all’providers’ treatment is consistent. Dr. Burns credibly opined that not having a consistent approach to a prisoner’s treatment poses a. risk of exacerbating or neglecting problems that may arise from mental illness, such ás self-injury. Specifically, according to Dr. Burns, failing to address the issue of repeated self-injury due to a lack of coordinated treatment and inconsistent approaches by different providers creates a substantial risk that patients will continue to -engage in self-harm; these patients can eventually end up disabled or dead as a result of continued self-harm. Defense expert Patterson agreed with Burns’s emphasis on the critical importance of coordinated'treatment and identified inadequate treatment planning as one of the most significant deficiencies in ADOC’s mental-health care system.

In the context of ADOC, where transfers of prisoners and changes in providers are frequent, the impact of inadequate treatment planning is exacerbated. Because written treatment plans are generic, counselors and patients often have to start from scratch when patients are moved from counselor to counselor. A former mental-health professional testified that prisoners who are transferred to a new counselor are often adversely affected, not only because the counselor has to start anew the process of building rapport with the prisoner, but also because treatment plans and progress notes often contain insufficient information to enable a different provider to learn about the patient or continue a consistent course of treatment. Plaintiff C.J. also testified to the difficulties in having to start over with a new counselor after each transfer. In sum, without the continuity of care and consistent treatment approaches provided through proper treatment planning, providers are substantially hindered from addressing symptoms of mental illness, exposing patients to continued pain and suffering, worsening self-injurious behavior, serious bodily injury, or even death.

4. Inadequate Psychotherapy

Constitutionally adequate mental-health care in prisons requires more than simply providing psychotropic medications to mentally ill prisoners. Prison systems must provide not only psychotropic medication but also psychotherapy or counseling to prisoners who need it to treat their serious mental-health needs. See Greason v. Kemp, 891 F.2d 829, 834 (11th Cir. 1990) (adopting district court’s conclusion that “[e]ven if this case involved failure to provide psychotherapy or psychological coun-selling alone, the court would still conclude that the psychiatric care was sufficiently similar to medical treatment to bring it within the embrace of Estelle”). As Dr. Burns explained, from a clinical perspective, having both modalities of treatment— medication and counseling — is important because one particular modality does not work for everyone. According to Dr. Burns, research indicates that seriously mentally ill patients need counseling and medication, along with non-structured or recreational activities, and that psychotherapy is an effective and essential mode of treatment for mental illness. She credibly opined, and the court finds, that not providing individual or group therapy poses a substantial risk of serious harm, including continued symptoms, pain, and suffering, as well as self-harm and suicide attempts.

Insufficient mental-health and correctional staffing at ADOC undermines the availability and quality of individual and group counseling sessions. First, as explained earlier, inadequate mental-health staffing combined with the increasing number of prisoners on the mental-health caseload has driven up the number of prisoners on each counselor’s caseload. As a result, both the frequency and quality of counseling sessions have suffered over time, according to both experts and MHM providers. MHM’s medical director Dr. Hunter testified that the caseload has increased in recent years to the point of taxing his staffs ability to carry out MHM’s contractual obligation: MHM counselors’ caseloads have increased from 60 patients to between 80 and 90; some facilities have only one counselor, who treats more than 100 patients; nurse practitioners’ caseloads have increased from 10-15 patients per day to 20-25 patients per day.

MHM’s program director Houser also testified that caseloads for counselors were sometimes twice as much as they should be; as a result, she said, counselors are “continually getting behind.” Houser Testimony at vol. 2, 25. In addition to seeing patients, counselors also have to attend meetings, document their treatment actions, design treatment plans, go on rounds in segregation units, and respond to crises as they arise.³⁵ Due to counselors’ increasing caseloads and mounting job responsibilities, individual counseling appointments are frequently canceled or delayed. For example, during a spot audit of the caseload at Bibb, 212 out of 213 cases had overdue counseling appointments. PI. Ex. 576, December 2, 2015 Email from Davis-Walker to Houser (doc. no. 1112— 26). Defense expert Patterson also observed that counseling appointments are frequently delayed due to staffing shortages and opined that “these delays contribute to a failure to provide necessary mental health services”; the potential harm in such delayed appointments includes “continued pain and suffering of mental health symptoms including suicide and disciplinary actions due to inadequate treatment.” See Joint Ex. 461, Patterson Expert Report (doc. no. 1038-1046) at 64.

Caseloads that are — as MHM’s Houser put it — much higher than an “acceptable standard” may explain why so many prisoners testified that ‘counseling sessions’ do not amount to much. Dr. Patterson’s review of the medical records within ADOC revealed that most progress notes from counseling sessions only contained short descriptions of symptoms, instead of reflecting clinical judgments and overall assessments of the patient’s progress. Similarly, Dr. Burns noted that the overwhelming majority of progress notes she reviewed indicated that the patient was ‘fine,’ had ‘no complaints,’ or had nothing to talk about. She explained that a short, vague statement like “I’m alright” is not a sufficient indicator of a stable mental-health state: instead of moving on to the next patient simply because the patient’s initial self-reporting does not expressly indicate distress, the clinician should probe deeper; notes on asking follow-up questions about medications, mood, job assignments, or disciplinary sanctions would reflect a proper counseling session. Based on the prisoners’ descriptions and the experts’ observations, the court finds that counseling sessions are often inadequate.

The chronic lack of sufficient correctional staffing has also contributed to frequent disruptions in the provision of psychotherapy. Dr. Burns credibly opined that insufficient correctional staff has interfered with access to treatment, as evinced by frequently canceled or delayed individual counseling sessions and group sessions. In particular, as she noted, the frequency of counseling sessions for those in segregation is especially low due to officer shortages: since segregation inmates must be escorted from their cells by correctional officers, mental-health appointments are frequently canceled or delayed when there are not enough officers to cover both the essential security posts and mental-health appointments. Ayers, defendants’ correctional expert, also credibly opined that ADOC was failing to respond to the needs of mentally ill prisoners due to the correctional staffing shortage. Likewise, a nurse practitioner at Donaldson credibly testified that she has experienced a persistent problem of not being able to see patients due to a lack of correctional staffing, and that the problem has been getting worse over the years. She and other providers testified that when insufficient correctional staffing does not allow prisoners to be escorted to the mental-health offices, the mental-health providers may go' to the cells themselves and attempt to talk-tó their patients at the cell-front. However, as agreed by MHM’s medical director Hunter and experts Burns and Haney, these cell-front check-ins are insufficient as counseling -and do not constitute actual mental-health treatment; Haney explained that these contacts serve solely a monitoring purpose — that is, to ensure that the patient is responsive and not decompensat-ing, rather than to treat the underlying mental illness. Indeed, while visiting five different facilities and their segregation units, the court observed the difficulty of standing outside a closed cell door to speak to a prisoner about mental-health needs: most cell doors are solid with small, perhaps 12-by-6-inch windows, some of which were completely fogged over- and others shielded by wire mesh or obfuscated by paper pasted on the window, either by the prisoner or from outside; and most of these segregation or high-security cells áre in' large, auditorium-like spaces, where sounds echo throughout the units, resulting in a panoply of unintelligible yet very loud noises. Conducting a counseling session across the door in these loud spaces seemed nearly impossible: the court had a hard time imagining having a meaningful conversation'in such an environment, let alone a conversation for the purposes of mental-health treatment.

As with these 1 cell-front sessions, ADOC’s provision of psychotherapy often lacks confidentiality. Experts and other clinician witnesses explained that confidentiality between providers and patients is a hallmark of and a necessary condition for mental-health treatment, yet some ADOC facilities lack a confidential setting for counseling sessions. Obviously, cell-front interactions between mental-health staff and prisoners are not confidential, as many staff witnesses testified, and as the court observed 'firsthand. Moreover, many facilities lack mental-health offices with windows and doors that would ensure the visibility of the counseling session to the correctional officer who is providing security without sacrificing sound confidentiality. For example, as the court saw on its tour of St. Clair Correctional Facility, 'the walls in the mental-health offices do not extend from floor to ceiling, and they lack doors; in other words, the offices resemble tall cubicles. Anyone nearby, including other prisoners and the correctional officer who escorted the -prisoner there, could hear the content of a counseling session. Moreover, correctional officers often stand by the door of counseling offices' with the door ajar for safety purposes, and counseling sessions are sometimes held in lieutenant’s offices where other correctional officers are present arid holding disciplinary hearings. As Dr. Haney explained, prisoners often do not feel safe sharing their mental-health issues in the presence of correctional officers or other prisoners because what they share with the mental-health staff may make it easier for others to exploit 'them; as a result, the lack of confidentiality undermines the effectiveness and quality of counseling sessions.³⁶

The quality of psychotherapy also suffers due to use of unsupervised, unlicensed counselors, referred to as ‘mental health professionals’ in ADOC. The court finds, based on expert testimony from both sides, that the lack of supervision for unlicensed MHPs is a significant, system-wide problem affecting the delivery of mental-health care within ADOC. ADOC’s own contract for mental-health care specifies that all MHPs must be licensed. However, only four out of 47 MHPs employed at. ADOC were licensed as of February 2016, and this problem has persisted for years.³⁷ The standard of care and state regulations mandate that an unlicensed counselor be supervised by a licensed psychologist, who is required to co-sign the counselor’s notes and review the treatment provided. Because MHM employs only three psychologists, most MHPs work at prisons without a psychologist, and the chief psychologist of MHM, Dr. Woodley, provides.no actual supervision to unlicensed MHPs. In fact, most MHPs’ clinical work is supervised by their respective site administrators, who are also mostly unlicensed counselors with their own caseloads — in other words, the supervisors generally have the same level of credentialing and education as the MHPs they are supervising. If the site administrators have any problems, they consult with Dr. Woodley. Dr. Patterson, a defense expert, credibly opined that it is unacceptable for an unlicensed counselor, rather than a licensed psychologist, to supervise another unlicensed counselor. He identified the lack of supervision of unlicensed providers as a systemic deficiency.

ADOC’s provision of group therapy is also inadequate. Dr. Burns testified that infrequent and. inadequate individual counseling can pose a substantial risk of serious harm to prisoners with mental illness, if the same patients do not have access to group therapy. Burns further explained that group therapy is especially important in a correctional system, which often does not have enough resources to provide individual counseling to all of the prisoners who need psychotherapy. Group sessions, like individual therapy, help prisoners with mental illness manage their symptoms, so that they do not deteriorate to the point of needing residential treatment; outpatient group therapy also enables mental-health staff to identify those who need more intensive treatment. Burns opined that therapy groups on depression, post-traumatic stress disorder, and medication management issues should always be offered to those on’ the' mental-health caseload, and that not offering such group treatment in the context of an under-resourced correctional mental-health system creates a substantial risk of harm to prisoners suffering from those illnesses. Despite the importance of group therapy for those who receive inadequate individual therapy, many seriously mentally ill ADOC prisoners with little access to individual therapy also have little access to group therapy. MHM’s program director Houser admitted that groups have been not happening at many facilities, including RTUs and SUs, due to the correctional staffing shortage.-

In sum, mental-health understaffing, correctional understaffing, the use of unsupervised, unlicensed counselors, and lack of confidentiality all undermine the efficacy and frequency of psychotherapy for mentally ill prisoners within ADOC. These conditions have created a substantial risk of serious harm for those who need counseling services, leaving them at a greater risk for continued pain and suffering, self-injurious behavior, suicidal ideation, and, as discussed later, disciplinary actions in response to symptoms of mental illness.

5. Inadequate Inpatient Care

Problems of inadequate psychotherapy and treatment planning become even more pronounced for prisoners in mental-health units, where ADOC houses the most severely mentally ill prisoners in its custody. Mental-health units (also referred to as inpatient units) include residential treatment units and intensive stabilization units. These units, which are located at Donaldson, Bullock, and Tut-wiler, house about 2 % of prisoners within ADOC’s custody.³⁸ Given that prisoners housed in mental-health units have already been identified as having the most severe mental-health needs within ADOC, these patients are at higher risk of de-compensation than other mentally ill prisoners if treatment is insufficient or if their housing environment is not therapeutic. And yet, despite ADOC and MHM’s awareness of these prisoners’ acute needs, the most severely mentally ill have been receiving grossly inadequate care; in fact, one of the experts described ADOC’s mental-health units as operating “almost exactly the same way” as segregation, as illustrated by the placement of segregation inmates without mental-health needs in the same unit and the inadequate out-of-cell time and treatment. Haney Testimony at vol. 2, 104.

a. Improper Use of Mental-Health Units

ADOC has had a persistent and longstanding practice of placing segregation inmates without mental-health needs in mental-health units. This practice allows prisoners without mental-health needs to occupy beds that should be reserved for prisoners who have heightened mental-health care needs and seriously undermines the therapeutic purpose of the mental-health units. Starting in 2012 and continuing through 2016, in its yearly contract-compliance reports, quarterly continuous quality improvement (CQI) meetings, and monthly operating reports, MHM repeatedly discussed ADOC’s problematic placement of segregation inmates in the RTU and SU. ADOC’s own audit of the Donaldson RTU in 2013 also identified the presence of segregation inmates without mental-health needs as a problem. While Associate Commissioner Naglich testified that segregation inmates were moved out of the Bullock SU by the end of 2013, evidence showed that the problem continued through 2016. For example, Brenda Fields, a clinical operations associate from MHM’s corporate office, testified that the presence of segregation inmates in the RTUs and SUs was noted as a problem in early 2016. Most recently, in December 2016, the list of prisoners in the Donaldson RTU included 13 segregation prisoners who did not have a mental-health code appropriate for mental-health units. PI. Ex. 1264, December 2016 Donaldson Segregation List (doc. no. 1099-8) at 14; Cul-liver Testimony at_

Dr. Tytell, ADOC’s chief clinical psychologist, explained that wardens place segregation inmates in the RTU or the SU when they do not have space for them elsewhere. He explained that MHM currently is expected to contact him or Nag-lich whenever this happens, but did not confirm whether this was always the case. Nevertheless, according to Tytell, the problem has been recurring. He conceded that it is ultimately the wardens, rather than the mental-health staff, who decide how cells in the mental-health units are used.

As all experts, MHM providers, and Dr. Tytell agreed, placing segregation inmates in a mental-health treatment unit is highly problematic. The reasons are multifold. First, having segregation inmates in the same unit as mental-health patients creates a security risk for mental-health patients: the segregation inmates’ presence prevents programming from taking place and diverts correctional officers’ attention away from mental-health patients and their needs. MHM’s medical director Dr. Hunter testified that'housing segregation prisoners in mental-health units compromises mental-health treatment, and that he has made this clear to ADOC. Dr. Woodley, MHM’s chief psychologist, informed ADOC that the presence of segregation inmates in the Bullock SU “undermine[s] the utility of this unit making it nearly impossible to operate it for its intended purposes.” Joint Ex. 323, February 2016 MHM Monthly Operating Report (doc. no. 1038-668) at 23.

Second, as Dr. Tytell and other experts explained, because mental-health inmates are particularly vulnerable, and those placed in segregation generally have behavioral problems, the presence of segregation inmates increases mental-health patients’ risk of being victimized through manipulation or violence. MHM and Dr. Tytell were aware of this risk, as one of the MHM staff members explained during a CQI meeting that using the Bullock RTU as a “disciplinary dorm” is “putting our vulnerable [inmates] at risk.” PI. Ex, 717, July 2015 Quarterly CQI Meeting Minutes (doc. no. 1044-11) at MHM029600. Associate Commissioner Naglich also agreed that segregation inmates in mental-health units can cause tension within the unit and anxiety to mental-health patients.

The housing of segregation inmates in mental-health units also contributes to the shortage of SÚ cells for those who actually need urgent mental-health treatment. Associate Commissioner Naglich acknowledged that patients awaiting SU admission could be in an “emergency” situation, as these patients require the highest level of care available within ADOC. Naglich Testimony at vol. 1, 208. However, since 2011, the Bullock SU has had a backlog of patients awaiting admission. While Naglich maintained that after the 2013 audit, ADOC actually moved all segregation inmates out of the Bullock SU in order to alleviate the backlog, she was unable to produce any documentation supporting her testimony. (During her testimony Naglich reassured the court that she could produce documents showing that she did move segregation inmates out of the Bullock SU in 2013. However, when she did bring in documents purportedly showing such transfers, none of them actually shoyred that any segregation inmates 'were moved out of the SU.) Moreover, in 2016, MHM continued to report that segregation inmates were still present in the SU, and that SU cell shortages were causing delays for patients who need SU-level care. Clearly, the placement of segregation inmates in SU beds continues to affect the most severely ill.

b. Inadequate Out-of-Cell Time and Programming

ADOC’s mental-health units often fail to serve their' therapeutic purpose due to insufficient out-of-cell time and scarce programming for their patients. One of the plaintiffs’ experts, Dr. Haney, who for multiple decades has studied isolation and segregation in correctional facilities, noted that ADOC’s ‘celled’ mental-health units³⁹ resemble and operate like segregation units.⁴⁰ Experts on both sides pointed to specific traits of ADOC’s mental-health units that contribute to this segregation-like atmosphere and the lack of a therapeutic milieu: the presence of segregation inmates within the mental-health units, as explained above; a severe lack of out-of-cell time; and a lack of meaningful treatment activities.

Out-of-cell time is crucial for patients housed ,iri mental-health units. Without bringing patients out of their cells for counselling sessions, treatment team meetings, group sessions, and activities, placement in a ‘mental-health unit’ does no good for patients who need the highest level of care; careful observation and treatment cannot happen when confined in a small cell all day. In fact, without out-of-cell time and effective treatment, housing severely mentally ill prisoners in a mental-health unit is tantamount to “warehousing” the mentally ill. See Wyatt v. Aderholt, 503 F.2d 1305, 1309 n.4 (5th Cir. 1974) (affirming the district court’s finding that a state mental hospital was functioning as a “ ‘warehousing institution ,.. wholly incapable of furnishing treatment to the mentally [ill] and ... conducive only to the deterioration and debilitation of the residents’ ”)(quoting Wyatt v. Stickney, 344 F.Supp. 387, 391 (M.D. Ala. 1972) (Johnson, C.J.)). Furthermore, as Dr. Haney explained, out-of-cell time is especially important for mentally ill, prisoners for two reasons. First, mentally ill prisoners experience more pressure and stress from a confined environment, and they have a more acute need to relieve that type of stress due to their vulnerable mental state; in other words, isolation makes it mor¿ likely that their conditions will deteriorate. In that sense, out-of-cell time is in and of itself therapeutic. Second, out-of-cell .time ensures that mental-health patients’ socialization skills do not atrophy to the point that they become uncomfortable with human interaction altogether.

Patients housed in ADOC’s- mental-health units receive very little out-of-cell time. This puts them at a substantial risk of continued pain and suffering, decompen-sation, and self-harm. As Dr.. Haney observed, at the Donaldson RTU, patients with serious mental illnesses are left inside their cells virtually all day, with no daily activities; this is similar to ADOC’s treatment of segregation inmates, whose out-of-cell time at ADOC does not exceed five hours per week. Dr. Burns concluded, and the court agrees, that the-RTUs and SUs offer “little treatment except for psychotropic medication due to staffing level shortages of both treatment and custody staff.” Burns Testimony at vol. 1, 26. Dr. Haney also noted that an unduly harsh and punitive practice limiting property makes mental-health units far from therapeutic and exacerbates prisoners’ idleness. He observed that mental-health unit inmates are allowed very little property, which means that they do not have books to read or other things to keep them engaged while spending the vast majority of their time in their cells. The court also observed firsthand the-idleness of seriously mentally ill prisoners during its visits to Bullock and Donaldson’s mental-health units: the majority of prisoners in those units were lying in their cells, often in a fetal position and facing the wall; there appealed to be np way to engage in .any remotely meaningful activity in the cell.

Dr. Patterson, the defense mental-health expert, testified that, in prisons around the country, the standard out-of-cell time for those in mental-health units is ten hours of structured therapeutic activity and - ten hours of unstructured activity per week. While a standard practice within the industry does not necessarily set the constitutional floor, a substantial deviation from the acceptable professional standard could support a finding of an Eighth Amendment violation. Steele, 87 F.3d at 1269.

Patients in ADOC’s RTUs and SUs get a vanishingly small amount of time outside their cells compared to the standard practice. In 2013, MHM acknowledged that the lack of programming was problematic for the Bullock SU, telling ADOC that “[i]n-creased programming will assist in staffs ability to stabilize inmates sooner and address the waiting list problem thus easing the bottleneck.” PI. Ex. 689, MHM Corrective Action — Donaldson May 2013 (doc. no. 1069-6) at 13-14. As of June 2016, three years after the 2013 audit, patients in the SU at Bullock were still getting about 30 minutes of individual therapeutic contact per week and about 2.5 hours of non-therapeutic group contacts per week. Joint Ex. 346, June 2016 Monthly Operating Report (doc. no. 1038-708) at 4,⁴¹

Prisoners in RTUs do not fare much better than in the SUs: Dr. Patterson found that RTU programming — which provides prisoners’ main opportunity to leave their cells — is inadequate.⁴² MHM and ADOC’s internal documents also recognized this lack of out-of-cell time for RTU inmates in the 2013 Donaldson audit: the audit results revealed, that no groups were being held -for Donaldson RTU patients, and that providers were having difficulties keeping appointments due to correctional staffing shortages. MHM’s corrective-action plan following the audit stated that “ADOC not enforcing-the out of cell time and not supporting MHM' with the process” is a challenge in ensuring that RTU patients are let out of their cells daily. PI. Ex. 689, MHM Corrective Action — Donaldson May 2013 (doc. no. 1069-5) at 12. The problem of inadequate out-of-cell time at the Donaldson RTU has' continued in spite' of the corrective-action plan: in early 2016; MHM’s corporate office recommended “continued advocacy for RTU patients to receive outdoor recreation,” PI. Ex. 115, 2016 Contract-Compliance Report (doc. no. 1070-5) at 15. As of September 2016, Donaldson RTU patients were getting fewer than two group contacts per week on average. Joint Ex. 344, September 2016 Monthly Operating Report (doc. no. 1038-703) at 3.

In addition to the lack of general out-of-cell time, mental-health units also fail to provide an adequate amount of treatment to these severely mentally ill prisoners because of shortages of mental-health staff. MHM’s program director Houser testified that groups have not been taking place at many facilities, including RTUs and SUs; indeed, an alarmed site administrator at Donaldson informed Houser in August 2015 that staffing losses at the facility have made it all but impossible to meet the needs of patients at the RTU. In December 2015, Houser asked Dr. Hunter to have one of the psychiatric providers at Bullock, Dr. Edward Kern, provide more services in the RTU in addition to his work in the SU. Dr. Hunter responded that, because the SU was so short-staffed and needed to be prioritized, shifting resources to the RTU would be difficult; he also noted that Dr. Kern had returned after a week of vacation to “what was essentially a zoo on [the SU].” PI. Ex. 382, Email from Houser to Hunter (doc. no. 1112-6). The 2013 Donaldson audit also found that the psychiatric coverage was insufficient and the logs for RTU rounds by providers were not being kept, making it impossible to tell whether RTU patients were getting any check-ins or treatment or whether their progress was being monitored. PI. Ex. 689, MHM Corrective Action — Donaldson May 2013 (doc. no. 1069-5) at 1-2.

The correctional staffing shortage also affects the amount of therapeutic care that patients at Donaldson and Bullock receive. Houser admitted that a lack of officers for the RTUs and SUs often cause the cancellation of group activities. The impact of the officer shortage was also consistently documented by ADOC and in reports that were sent to Associate Commissioner Naglich and OHS for their review. For example, during OHS’s 2013 audit of Donaldson, the auditors noted numerous deficiencies caused by the correctional staffing shortage. First, mental-health staff were manning laundry and showers instead of providing mental-health care, because there were not enough correctional officers to perform those basic duties. Scheduled activities and out-of-cell time were not being provided due to the correctional officer shortage, and MHM’s corrective-action plan stated that the “[RTU] has to be conceptualized as an RTU and not as segregation.”, PI. Ex. 689, MHM Corrective Action — Donaldson May 2013 (doc. no. 1069-5) at 9. The same was true at the Bullock SU: the problem of ‘access to patients’ — meaning that mental-health staff were unable to provide treatment to patients due to correctional officer shortage — was first identified in 2013, and then again in 2014 and 2016 contract-compliance reports. See PI. Ex. 114, 2013 Contract-Compliance Report (doc. no. 1070-4); PI. Ex. 105, 2014 Contract-Compliance Report (doc. no. 1070-3); PI. Ex. 115, 2016 Contract-Compliance Report (doc. no. 1070-5). As Naglich testified, because there are simply not enough correctional officers, the problem of accessing patients in RTUs and SUs recurs on a regular basis, even when it has been temporarily alleviated through reassigning officers to particularly problematic areas. As a result, patients in the SU often receive their individual psychiatric contact via cell-front check-ins. As explained earlier, this utter lack of confidentiality negates the therapeutic utility of these contacts. Such cursory contacts with the most severely ill patients are gravely inadequate.

The severe effects of warehousing, rather than treating, seriously mentally ill prisoners was crystalized in two incidents at the Donaldson RTU, where two different patients set their cells on fire out of frustration about not getting let out of their cells. The internal email reporting one of the incidents explained that the problem of not letting patients out of their cells was due to correctional staffing shortages. PI. Ex. 518, January 22, 2016 Email from Wynn-Scott to Houser (doc. no. 1112-18).

Jamie Wallace’s last 10 days in the Bullock stabilization unit further exemplify the inadequate treatment provided to the most severely ill patients: his medical records for his final 10 days reflected no group activities, one cell-side treatment plan note, and two psychiatric progress notes.

The lack of out-of-cell treatment in mental-health units adds the risk of harm posed by the harsh effects of isolation to that posed by inadequate treatment in general. As Associate Commissioner Nag-lich admitted, inadequate treatment of patients in inpatient units can lead to “additional exacerbation of their mental health symptoms,” including further hallucinations and delusions, and suicide. Naglich Testimony at vol. 3, 144-45. In addition, as experts testified, mentally ill prisoners are at a substantial risk of decompensating and being subject to prolonged pain and suffering when placed in an isolated environment. In other words, ADOC’s failure to provide adequate treatment and out-of-cell time in mental-health units forces the most severely mentally ill patients to face yet another risk factor for decompensation, even though their placement was for the specific purpose of alleviating the symptoms of their mental illness. Inadequate out-of-cell time and treatment in this context therefore compounds the risk of harm that is already inherent in a nonfunc-tioning mental-health care system,

c. Lack of Hospital-Level Care

ADOC also creates a substantial risk of serious harm to prisoners at the most severe end of the mental-health spectrum, because it does not provide hospital-level care or a hospitalization option for prisoners housed there. According to experts from both sides, hospital-level care or hospitalization should be available when patients pose a danger to self or others and interventions in the SU do not improve their condition: due to the harmful effect of isolation in an SU cell, staying in the SUs cannot be a long-term solution for patients who experience repeated episodes of deterioration.

Although many ADOC prisoners require hospital-level care, very few actually receive it. Virtually all psychiatric providers who testified agreed that they knew or noticed ADOC prisoners who needed to be transferred to a hospital. ADOC’s administrative .regulations dictate that those who are kept in the SU for over 30 days without stabilizing should be considered for hospitalization; the same provision also mandates that the treating psychiatrist recommend a transfer to a state psychiatric hospital if the treatment team determines that all mental-health interventions possible within ADOC have been exhausted, and that the inmate has not responded to those interventions. Joint Ex. 138, Admin. Reg, § 634, Transfer to State Psychiatric Hospital (doc no. 1038-168). However, ADOC virtually never transfers patients to hospitals, except in the case of prisoners nearing the end of their sentence. Dr. Hunter and Associate Commissioner Naglich corroborated this point, and Dr. Kern could recall only four prisoners in the last six years who were transferred to a hospital before the end of their sentences. Dr. Kern explained that MHM tries to deal with acutely ill patients’ symptoms within ADOC even though ADOC cannot provide hospital-level care, instead of pursuing hospitalization as required by the administrative regulation, because the waiting list for a bed in a hospital can be six months long or longer.

Several factors differentiate hospital-level care from what is provided in ADOC, as defense expert Dr. Patterson explained. Hospitals are able to offer a high level, of monitoring for suicidal and decompensat-ing patients while not isolating them in a cell: hospitals, or hospital-like environments are better at treating severely mentally ill patients because patients can leave their rooms to request help from staff, instead of having to wait until correctional officers or mental-health staff check on them; most of the patients’ interactions in a hospital are based on doctor-patient or nurse-patient relationship's, rather than guard-prisoner relationships; and the goal of the staff is to treat the patients, rather than to incarcerate them. Dr. Kern' also admitted that dealing with patients who need hospital-level care within an SU or RTU is challenging because in those units, providers have a very limited ability to give patients out-of-cell time. He also added that, if he could, he would like to- have SU patients “four to six, possibly eight hours out of their cell every day,”, but that this is impossible because “there are not enough security staff.” Kern Testimony at 21-, In other words, without a hospitalization option or another method of providing hospital-level care, the providers are forced to choose between the benefits of close, monitoring and restriction of activity and the harmful effects of isolation and losing socialization opportunities.

Both Dr. Patterson and Dr. Burns expressed strong disapproval of ADOC’s failure to provide hospital-level care. As Dr. Burns put it, waiting for an unstable patient’s end of sentence to transfer him or her to a hospital is akin to “someone with chest pain who has to wait until they’re released from prison to get taken to a hospital to have the chest pain treated. We wouldn’t do-that in the case of chest pain. I’m not. sure why we do it in the case of inmates with serious mental illness.” Burns Testimony at vol. 1, 168-69. Dr. Patterson opined that there, should be a hospital-like setting or actual hospitalization of patients with the most severe cases of mental illness; he did not see any hospital-like environment within the ten facilities he toured. He also explained .that placement in the stabilization unit, the highest level of care available within ADOC, should be a time-limited treatment intervention, because the SU is a highly isolated setting and likely to exacerbate conditions of those prisoners experiencing acute symptoms; and that if a patient is not stabilized in the SU, the patient should be moved to a hospital. Patterson emphatically stated, without any qualification, that refusing to transfer patients to mental health hospitals until the end of their sentences is simply “wrong,” 'and that it puts the most severely ill patients at a substantial risk of harm, Patterson Testimony at vol. 1, 174. In other words, for the most severe cases of acute mental illness, there is no alternative to a hospital setting, due to these stark differences in treatment options and milieu.

The grave risk of serious harm in failing to provide hospital-level care to severely ill prisoners was quite obvious in the case of Jamie Wallace. Less than two months before he testified in court, clinicians recommended that Wallace be transferred to a hospital. Despite the clinical recommendation, ADOC chose not to pursue hospital admission. In court, Wallace testified that voices in his head told him to kill himself; and indeed, he had attempted suicide multiple times. After testifying in court, highly agitated and destabilized, Wallace languished in a crisis cell and an SU cell before ending his life. Less than two weeks had passed since his testimony, regarding inadequate mental-health care in ADOC.

6. Inadequate Suicide Prevention and Crisis Care

Like its inpatient care, ADOC’s suicide-prevention procedures and crisis care suffer from serious deficiencies. Identification, treatment, and monitoring of those who have heightened suicide risks are important because they provide the last safety net before the worst possible outcome in mental-health care: suicide. Reflecting its importance, courts have held that a minimally adequate mental-health care system must have a functioning suicide-prevention program. See, e.g., Ruiz v. Estelle, 503 F.Supp. 1265, 1339 (S.D. Tex. 1980) (Justice, J.) (“[I]dentification, treatment, and supervision of inmates with suicidal tendencies is a necessary component of any mental health treatment program.”), aff'd in part, rev’d in part on other grounds, 679 F.2d 1115 (5th Cir. 1982), opinion amended in part and vacated in part, 688 F.2d 266 (5th Cir. 1982), cert. denied, 460 U.S. 1042, 103 S.Ct. 1438, 75 L.Ed.2d 795 (1983); Madrid v. Gomez, 889 F.Supp. 1146, 1258 (N.D. Cal. 1995) (Henderson, C.J.) (adopting the suicide-prevention program standard from Ruiz as part of “constitutional minima”); see also Greason v. Kemp, 891 F.2d 829, 835-36 (11th Cir. 1990) (‘Where prison personnel directly responsible for inmate care have knowledge that an inmate has attempted, or even threatened, suicide, their failure to take steps to prevent that inmate from committing suicide can amount to deliberate indifference.”); Waldrop v. Evans, 871 F.2d 1030, 1036 (11th Cir. 1989) (finding that failure of a prison staff member to notify competent authorities regarding the inmate’s dangerous psychiatric state and self-harm may constitute deliberate indifference).

Prisoners are at an elevated risk of suicide due to the conditions prevalent in ADOC facilities. As Dr. John Wilsbn, the psychologist from MHM’s national Clinical Operations Department, explained to MHM’s program director Houser, “[experience and research confirm” the following: “Suicides increase with crowding, drugs, assaults, low staffing rates, lack of meaningful programming, and significant changes in facility mission/population such that inmates are moving between facilities more frequently or are uncertain about whether they will be housed or ... when there is basic unrest at a systems level, it can cause a spike.” PI. Ex. 1224, October 1, 2015 Email from Wilson to Houser (doc. no. 1117-24) at ‘2. Given the widespread presence of these factors in ADOC, the need for effective suicide prevention and crisis care cannot be overstated.

Suicide prevention consists of assessing and managing suicide risk: "assessing the risk entails using a suicide risk-assessment tool to identify those who are at heightened risk and the level of that risk; managing the risk involves both short- and long-term care that provides meaningful therapeutic contact to alleviate suicide risk. Suicide prevention also involves physically restricting suicidal prisoners’ ability to harm themselves. The short-term care provided to prisoners who are undergoing acute mental-health crises is called ‘crisis care.’

The standard of care in correctional mental-health care and ADOC regulations require that a suicidal'prisoner be placed in a special ■ cell that, minimizes risks of self-harm and suicide. These ‘crisis cells’ or ‘suicide-watch cells’ must be free of structural designs that would facilitate self-harm or suicide attempts, such as tie-off points where prisoners can tie a ligature to hang themselves; they also must be free of items that prisoners can use to harm themselves, such as sharp items and ropes. Patients on.suicide watch are stripped of personal belongings and regular clothes and given a suicide-proof blanket and a suicide smock.

Both correctional officers and mental-health staff have the ability to place any prisoner on suicide watch, after which a mid- or high-level provider is required to conduct a thorough mental-health assessment that includes the use of a suicide risk-assessment tool.⁴³ While on suicide watch, patients are to receive a high level of care in order to resolve the crisis and return to a less isolated and restrictive setting as soon as possible; such care includes close monitoring, daily re-evaluation of treatment plans, and frequent contacts with mid- or high-level providers such as psychiatrists and psychologists. If a patient on suicide watch is not stabilized within 72 hours, mental-health staff is required to evaluate the patient for admission to the stabilization unit. Discontinuing suicide-watch procedures requires an order from a psychiatrist, psychologist, or a nurse practitioner after an in-person evaluation.

ADOC prisoners in crisis may alternatively be placed in a crisis cell on mental health observation (MHO). MHO refers to a similar, short-term monitoring status for patients whose conditions are not as acute as those on suicide watch but still merit an observational status, or who have been recently released from suicide watch. Patients on MHO may be released only by a psychiatrist, psychologist, or nurse practitioner, but — unlike in suicide watch — the patients are allowed regular clothes and limited property.

ADOC’s suicide-prevention efforts and crisis care suffer from multiple inadequacies. First, ADOC and MHM’s use of a suicide risk-assessment tool is too limited to adequately identify those at high risk. Moreover, many prisoners at heightened risk of suicide or self-harm do not receive crisis care because of a severe shortage of crisis cells and staffing, and due to a culture of skepticism towards threats of suicide. Second, suicidal prisoners are often placed in unsafe environments both because of the shortage of crisis cells and because many crisis cells contain unsafe physical structures, such as tie-off points, and dangerous items that can be used for self-harm. Third, prisoners who are identified as suicidal receive inadequate monitoring and treatment. Lastly, inappropriate releases from suicide watch and a lack of follow-up care often push suicidal prisoners back into crises again and again, driving up the demand for crisis cells and diverting resources away from day-to-day, long-term treatment.

ADOC’s inadequate crisis care and long-term suicide-prevention measures have created a substantial risk of serious harm, including self-harm, suicide, and continued pain and suffering. ADOC has experienced a dramatic increase in suicide rates in the last two years. Alabama’s reported suicide rate was five per 100,000 between 2000 and 2013; by fiscal year 2015-2016, the rate had shot up to over 37 per 100,000. This is more than double the national average of 16 suicides per 100,000 prisoners in state and federal correctional systems. Patterson Testimony at vol. 2, 27; see also U.S. Department of Justice, Bureau of Justice Statistics, Mortality in Local Jails and State Prisons, 2000-2013 — Statistical Tables (2015) at Table 28. In the fiscal year starting in October 2016, the rate is projected to be over 60 per 100,000, based on the first three months of the year. See PL Ex. 1267, 2015-2016 Chart of ADOC Suicides (doc. no. 1108-38). Defense expert Dr. Patterson testified that he does not know of any prison system that has a suicide rate over 25 or 30 per 100,000. It is in the context of the magnitude of the suicide rate at ADOC that the court now considers ADOC’s failure to provide a functioning suicide-prevention system,

a. Failure to Provide Crisis Care to Those Who Need It

ADOC fails to provide suicide-prevention services and crisis care to many prisoners who need it. This failure stems from inadequate identification of those who are at heightened risk of suicide, combined with a culture of cynicism toward prisoners’ threats of suicide and self-harm and a severe shortage of crisis cells. The majority of suicides in ADOC are committed by prisoners who are not on the mental-health caseload, which means that many of the prisoners’ needs were never identified through the intake or referral process, and no intervention happened before their deaths. See PL Ex. 1267, 2015-2016 chart of ADOC suicides (doc. no. 1108-38) (showing eight out of 11 suicides between September 2015 and December 2016 committed by those who were not on the mental-health caseload).

According to correctional mental-health experts on both sides, the administration of a suicide risk-assessment and management tool by a qualified provider is widely recognized to be an essential part of mental-health care: it should be used as a part of the intake screening process and whenever a prisoner threatens or attempts to harm himself or actually does so. The purpose of a suicide risk-assessment tool is to assess whether a prisoner presents an increased risk of suicidal behavior in order to manage that risk through early intervention. As defense expert Dr. Patterson explained, the suicide risk-assessment tool must be completed in a face-to-face encounter by a high-level provider or a mid-level provider with high-level supervision, because the tool comes with clinical guidelines and requires clinical judgment.

ADOC and MHM did not use a suicide risk-assessment tool for many years and only recently began using one only at intake. While examining ADOC’s mental-health care in connection with this case, defense expert Patterson noticed that no suicide risk-assessment tool was being used, even though MHM had such a tool. Dr. Patterson recommended that a risk-assessment tool be used throughout the system, including at intake, upon placement in a crisis cell, and any other time a prisoner is deemed to have a heightened suicide risk. He also specifically indicated that a casual assessment without using a form is not acceptable — a form with appropriate clinical guidelines should be used in each instance.

As a result of this exchange, which took place in the summer of 2016 — more than two years after this lawsuit was filed— MHM began using a suicide risk-assessment tool at intake for new prisoners entering ADOC.⁴⁴ However, contrary to Patterson’s recommendations, MHM is not using the assessment tool when prisoners threaten or engage in self-harm or are placed in crisis cells. For example, Dr. Patterson found no suicide risk assessment had been completed for plaintiff Jamie Wallace in December 2016 despite his repeated threats of self-harm and suicide and his stay in a crisis cell shortly before he killed himself. Prisoners who threaten suicide frequently do not receive any kind of face-to-face assessment by high-level providers, let alone one involving a risk-assessment tool. The failure to perform proper suicide risk assessments to identify prisoners with a heightened risk of suicidal behavior places seriously mentally ill prisoners at an “obvious,” substantial risk of serious harm. Burns Testimony at vol. 1, 63.

A chronic shortage of crisis cells also contributes to ADOC’s failure to provide crisis care to those who need it. While the exact number of crisis cells sufficient for any given prison system depends on'the needs of the population and the treatment options available, it is clear that the number of crisis cells in ADOC is grossly inadequate. Witness after witness — including Associate Commissioner Naglich and MHM managers — agreed that having two crisis cells for 3,800 prisoners at the Sta-ton-Draper-Elmore complex, two crisis cells for 1,900 prisoners at Bibb, and two crisis cells for 2,700 prisoners at Fountain, is insufficient.⁴⁵ MHM’s medical director Hunter testified that the number of crisis cells in each of the 15 major facilities within ADOC is insufficient. In addition to the low-number of crisis cells across the system, the backlog of placements at the Bullock stabilization unit has contributed to the shortage: when the SU does not have a bed for the most acutely ill prisoners, often due to the presence of segregation inmates, mentally ill prisoners end up staying in crisis cells for much longer than 72 hours, though the explicit purpose of crisis cells is to serve as a short-term placement while the prisoner stabilizes.⁴⁶

Because ADOC has a limited number of cells to work with, ADOC and MHM staff gamble on-which prisoners to put in them and-frequently-discount prisoners’ threats of self-harm and suicide, instead of properly evaluating suicide threats by having a qualified provider administer a suicide risk-assessment tool. For example, in CQI meetings and multidisciplinary staff meetings, MHM staff discussed “cabling] their bluff” and “tak[ing] thé gamble” on prisoners who threatened to commit suicide or severely injure themselves. PI. Ex. 720, February 2014 Quarterly CQI Meeting Minutes (doc. no. 1044-14) at MHM029579; see also PI. Ex. 718, April 2015 Quarterly CQI meeting minutes (doc. no. 1044-12) at MHM029570 (discussing concerns about feigning suicidality to avoid being sent to segregation and that it is ADOC’s responsibility to find a safe place for “genuinely suicidal inmates” (emphasis in original)). Discussions during these meetings included statements such as “99 % often do not act on their threats.” PI. Ex. 721, January 2015 Quarterly CQI meeting minutes (doc. no. 1044-15) at MHM029614. Staff meeting minutes and medical records of patients also included conclusory statements suggesting that prisoners who are claiming suicidality and self-harm tendencies are in fact malingering or seeking ‘secondary gains’ — such as getting out of a segregation cell, or getting away from an enemy, or debt problems. In response to MHM staffs use of this type of language in medical records, MHM’s Chief Psychologist, Dr. Woodley, instructed staff to not use “malingering” and “secondary gain” in written documentation because one “cannot know [a prisoner’s] motivations for certain.” PI. Ex. 721, January 2015 Quarterly CQI meeting minutes (doc. no. 1044-15) at MHM029614. Contrary to this instruction, MHM staff continued to write off prisoners’ threats of self-harm as motivated by inmate-to-inmate debt or secondary gains, rather than conducting a proper assessment. In the March 2015 monthly operations report, MHM reported to ADOC that there have been “occasions where the inmate would not be placed on watch despite claiming to be suicidal, especially if the inmate is well known to the treatment staff as having a history of bluffing and/or no actual attempts.” Joint Ex. 328, March 2015 Monthly Operations Report (doe. no. 1038-673) at 14. A progress note from Jamie Wallace’s medical records dated five days before he committed suicide was representative of this culture: it noted that Wallace was “using crisis cell/threats to get what he wants.” Joint Ex. 496, Jamie Wallace Medical Records (doc. no. 1037-1062) at ADOC0399861. In sum, MHM staff frequently treat threats of self-harm as behavioral rather than mental-health issues, writing off threats instead of delving deeper to address underlying mental-health needs through a mental-health evaluation and suicide risk assessment.⁴⁷

This skeptical approach towards threats of self-harm poses substantial and obvious risks. First, those who should be on suicide watch may not receive the crisis care that they need and may kill or harm themselves. Gambling with threats of self-harm is dangerous: obviously, as experts and MHM staff agreed, not all prisoners who express suicidality are feigning it, and a number of prisoners do in fact become suicidal and engage in self-harm. Furthermore, the risk of misinterpreting a prisoner’s motivation is heightened by MHM’s failure to use a suicide risk-assessment tool after an instance or threat of self-harm. Second, hostile attitudes towards prisoners in mental health crises can “cause inmates to become more aggravated and agitated,” making it more difficult to treat the inmate. Houser Testimony at vol. 2, 160. Third, prisoners who make threats or engage in self-harm but are not actively suicidal may nevertheless suffer from underlying mental-health issues that need to be addressed.

As experts from both sides agreed, no bright line distinguishes ‘behavioral problems’ from ‘mental-health problems’: even if someone is engaging in self-harm for ‘secondary gains,’ a high-level clinician should evaluate the underlying mental-health issues, for four reasons. First, the presence of suicidality is not a yes-or-no question; according to the experts, it is well established that suicide risk is on a continuum, and a meaningful suicide-prevention program requires monitoring for an increased risk of suicide. Second, even in the absence of genuine suicidal ideation, engaging in self-harm is a mental-health issue because it indicates suffering from psychological distress and a lack of proper coping mechanisms to resolve problems. Therefore, insíéad of ignoring those who resort to self-harm to seek attention, staff should provide assistance.⁴⁸ Third, as Dr. Burns cautioned, chalking up instances of self-harm to behavioral problems not deserving of treatment may actually encourage such behavior: research in behavioral management shows that negative reinforcement of self-harm is more likely to prompt the prisoner to engage in more dramatic and even lethal self-harm. Finally, people who engage in self-harm can also accidentally kill or severely injure themselves without having a specific intent to do so; therefore, monitoring and assessment are necessary even if a prisoner’s suicidality is deemed not genuine.

To emphasize, the court does not mean to suggest that a prisoner must always be kept on suicide watch upon a threat of suicide; as experts noted, some threats of suicide or self-harm are not genuine. However, as the experts explained, these threats should not be written off without the use of an appropriate suicide risk-assessment tool by a qualified provider in a face-to-face evaluation. Based on the overall assessment of the evidence, the court finds that ADOC’s current practice was devoid of any system to ensure that suicidal prisoners are appropriately evaluated.

b. Placement of Prisoners in Crisis in Dangerous and Harmful Settings

Due to the chronic shortage of crisis cells, ADOC frequently places those on suicide watch in inappropriate environments, such as offices for correctional staff (also called ‘shift offices’), libraries, and segregation cells. These inappropriate placements put suicidal prisoners at a grave risk of self-harm and suicide.

ADOC and MHM have repeatedly placed suicidal prisoners in dangerous environments due to a lack of available crisis cells. MHM’s Dr. Hunter complained to ADOC in his March 2015 monthly operating report that ADOC officers at some facilities were placing prisoners on suicide watch in cells that are not crisis cells to avoid having to travel. He was also aware of at least a dozen times when prisoners in crisis at Bibb Correctional Facility were placed in shift offices over the weekend while waiting for transportation to another facility. ADOC’s Dr. Tytell recalled multiple instances in 2015 in which prisoners in crisis were being housed in shift offices for multiple days; he admitted that this practice was inappropriate but commented, “[Y]ou have to work with what you got.” Tytell Testimony at_During a 2016 ADOC tour, Dr. Haney found a prisoner who was housed in a mental-health office;⁴⁹ the prisoner had been there for over a day without receiving any treatment, even though he was deemed to be suicidal. Lastly, at least one documented case of suicide in the last three years occurred while a prisoner awaiting crisis-cell placement was housed in a room behind a shift office.⁵⁰ Houser Testimony at vol. 3, 55.

This practice of placing suicidal prisoners in unsafe environments increases the risk that prisoners will engage in self-harm, including suicide attempts. The consensus among the experts from both sides, as well as MHM and ADOC staff, was that housing a suicidal inmate in a space like a shift office is quite dangerous: not only are these places full of items that can be used for self-harm, but, depending on where the prisoner is placed, such placements can also cut off suicidal prisoners from the treatment that they desperately need.

Placing suicidal prisoners in cells that are either in or adjacent to death row or segregation also poses a number of problems. Holman Correctional Facility’s suicide-watch cells are located on death row. As plaintiffs’ expert Dr. Haney explained, the “juxtaposition of prisoners who are potentially suicidal with prisoners who are under a sentence of death” is “extremely problematic” for those in the throes of a mental health crisis. Haney Testimony at vol. 1, 101-02; Joint Ex. 459, Haney Expert Report (doc. no. 1038-1043) at 35. Defense expert Dr. Patterson agreed: bringing prisoners in crisis from general population into a death-row unit would make them more likely to decompensate, because death-row units are not designed to be therapeutic; moreover, death-row units are largely self-contained and are subject to their own regulations that are likely harsher and more punitive than the regulations in an ordinary unit. Dr. Patterson also expressed concern that death-row inmates would retaliate against inmates in crisis cells, creating even more stress for these vulnerable prisoners.⁵¹

c. Inadequate Treatment in Crisis Care

The care provided to prisoners on suicide watch is also grossly inadequate. ADOC and MHM fail to provide adequate treatment to patients in crisis cells and, to make matters worse, frequently keep them in crisis cells for much longer than appropriate or necessary.

As Dr. Burns credibly opined, out-of-cell counseling sessions for prisoners on suicide watch are important both because they can help eliminate- suicidal thoughts and because they assist providers in meaningfully modifying treatment plans to address the causes of a crisis. However, prisoners on suicide watch -and mental-health observation are not consistently receiving out-of-cell appointments with counselors.

•Prisoners are frequently kept for extended periods of time in crisis cells, instead of being transferred to an-RTU' or SU for intensive, longer-term treatment. According to ADOC’s administrative regulations, anyone who is on suicide watch for more than 72 hours should be considered for placement in a mental-health unit. As experts on both sides agreed," crisis-cell placement is meant to be temporary and should not last longer than 72 hours, because the harsh effects of prolonged isolation in a crisis cell can harm patients’ mental health. However, since as far back as 2011, MHM has, by its own report, considered transferring prisoners- in crisis to treatment units only in a small fraction of the crisis • placements that last longer than 72 hours. See PI. Ex. 1190, 2011 Contract-Compliance Report (doc. no. 1070-8) at 22 (in 2011, only 20 % of those housed in crisis cells for over 72 hour's were considered for transfer); PI. Ex. 105, 2014 Contract-Compliance Report (doc. ho. 1070-105) at 11 (in 2014, 29 %); PI. Ex. 115, 2016 Contract-Compliance Report (doc. no. 1070-3) at 11 (in 2016, 13 %). MHM’s CQI manager testified that extended stays in crisis cells are “sometimes” necessary because there is a “full house” in the appropriate treatment unit. Davis-Walker Testimony at vol. 2, 102. See also PI, Éx. 1219, September 2014 Emails between MHM and ADOC (doc. no. 1047-10)(dis-cussing a prisoner who was on suicide watch for 25 days at Bibb, waiting for, a transfer to Bullock SU).

Contrary to the CQI manager’s characterization, documentary evidence showed that prisoners are in fact frequently kept in crisis cells for much longer than 72 hours. See PI. Ex. 721, January 2015 Quarterly CQI Meeting Minutes (doe. no. 1044-15) at 4- (showing examples of long crisis-cell stays, such 'as 240 hours at Limestone, 429 hours at Staton, and 620 hours at Ventress, and suggesting that weekend hours were not being counted); PI. Dem, Ex. 141, 2016 Crisis Cell Placements (doc. no. 1156-2) (showing that a-majority of facilities have multiple prisoners being housed in crisis cells for longer than 144 hours, some of them exceeding 200 hours, in 2016). At St. Clair, Dr. Haney confirmed that one of the prisoners he interviewed had been housed in a barren suicide-watch cell in the infirmary for five months — well beyond the intended duration of crisis-cell stays. These extremely lengthy stays in crisis cells contribute, in turn, to a shortage of crisis cells throughout the system. They also illustrate that prisoners are not getting the treatment they need to stabilize and be moved out of crisis cells, or that ADOC and MHM are leaving these mentally ill prisoners in extremely-isolated environments for longer than appropriate.

d. Unsafe Crisis Cells

Despite their purpose of preventing self-harm and suicide, crisis cells in ADOC facilities are unsafe. First, crisis cells are ridden with physical structures that provide easy opportunities to commit suicide. Experts from both sides agreed that having crisis cells free of tie-off points is a critically important feature of suicide prevention in prisons. The National Commission on Correctional Health Care (NCCHC), a professional organization that promulgates standards for correctional health care and provides accreditation to facilities that follow those standards, requires that crisis cells be free from tie-off points that can be used for self-injurious behavior.⁵² ADOC’s history makes clear the critical importance of this issue: all but one suicide within ADOC in the last two years happened by hanging. However, many of ADOC’s crisis cells have easily accessible tie-off points, such as sprinkler heads, hinges, fixtures, and vents, making them incredibly dangerous for suicidal prisoners. In fact, defense expert Dr. Patterson stated that making crisis cells suicide-proof is the “number-one issue” to be addressed. Patterson Testimony at vol. 1, 296. •

Examples of unsafe crisis cells abound. As Dr. Haney noted and the court saw firsthand during prison visits in February 2017, in the Bullock SU, where some prisoners on suicide watch are kept, sprinkler heads are located directly above the sink and the toilet, making it easy for suicidal prisoners to climb up to tie a ligature on the sprinkler head. In fact, that is how Jamie Wallace committed suicide while housed in an SU cell at Bullock. As plaintiffs’ experts observed, crisis cells in St. Clair, Kilby, and Holman all have tie-off points; MHM’s Houser also admitted that many crisis cells across ADOC facilities are out of compliance with NCCHC standards for suicide cells because they have tie-off points.

Unsurprisingly, MHM staff have repeatedly expressed concerns about the safety of crisis cells in multiple facilities, as reflected in contract-compliance reports and CQI meeting minutes: in 2011, staff expressed concern about the unsafe features of crisis cells at Fountain; in 2012, staff reported concerns about the safety of Yentress crisis cells; in 2016, MHM’s contract-compliance report stated that crisis cells in Holman are unsafe because of the open bars on the doors. ■

Another dangerous aspect of many ADOC crisis cells is the difficulty of monitoring the prisoner inside. The design of the cell doors and windows and the layout of the facilities often prevent a direct line of sight into the cell. For example, Dr. Haney testified that suicide-watch cells at Donaldson, located in the infirmary and known as Z-Cells, had grates over the windows that made it very difficult to see into a cell even when standing directly in front of a door and peering in. At St. Clair, Dr. Haney noted that suicide-watch cells were located in a hallway in the infirmary; they, too, were hard to see into and easy to ignore. PI. Dem. Ex. 107, St. Clair Suicide Watch Cell (doc. no, 1125-62). Associate Commissioner of Operations Culliver noted that even though Holman crisis cells have barred fronts, it is nonetheless impossible to see into these cells from the officers’ cube located closest to them. Culliver also acknowledged that the solid crisis-cells doors at many facilities, including Bullock, Donaldson, Fountain, Kilby, and St. Clair, make it impossible for an officer or mental-health provider on the unit to see into the cells and check on the prisoners housed within them without walking up to the door and looking -through the small glass window.

ADOC’s practice of allowing prisoners in cells to cover the windows with paper or other material exacerbates the visibility problem. Dr. Haney noticed this practice in Donáldson, Holman, St. Clair, and Bibb, describing it as incredibly problematic because it blocks any type of monitoring entirely. Dr. Haney witnessed a particularly disturbing incident while touring Bibb. He entered the infirmary and went to speak with the prisoners housed in the crisis cells. As he was speaking to one, a lawyer touring the facility with him discovered that a prisoner in another crisis cell was, at that very moment, attempting to hang himself — the prisoner had somehow procured a cord to wrap around his neck and had attempted to cover the window with a blanket. Allowing prisoners to cover the windows of their cells is dangerous in any context, but it is particularly unacceptable for prisoners known to be suicidal. Due to the visibility problems with many ADOC suicide-watch cells, defense expert Patterson opined that suicidal prisoners should be under direct, constant observation while in those cells. He also explained that camera observation by an officer at the control station may not be sufficient, because by the time that officer notices a suicide attempt, it might be too late; moreover, the officer likely has other responsibilities that would preclude careful monitoring of any single cell.

The dangerousness of crisis cells and the significant risk of harm caused by such conditions are compounded by ADOC’s rampant failure to prevent introduction of dangerous items into crisis cells. Admittedly, the parties in 2014 reached a settlement that prohibits ADOC officers from providing disposable razor blades to prisoners on suicide watch and in segregation. See January 16, 2015 Order Denying Motion for Preliminary Injunction (doc. no. 84).⁵³ However, the problem of dangerous items in crisis cells has continued, according to a number of ADOC officials and MHM staff. Suicidal prisoners have access to inappropriate items — such as sharp implements — either because they bring the items with them when placed on suicide watch and correctional officers do not search them, or because correctional officers or inmate ‘runners’ who perform various housekeeping tasks around the unit bring the items to the crisis cells. MHM’s Houser stated that prisoners have access to improper items in safe cells at a number of facilities, including specifically Donaldson, St. Clair, Staton, and Holman; she was not sure whether this problem had been addressed at any of these facilities. Dr. Hunter of MHM and Associate Commissioner Culliver both testified that finding sharp objects in a suicide-watch cell has been a problem at Bibb, despite the installation of flaps on cell doors that were intended to stem the flow of contraband. Lastly, Holman’s crisis cells are particularly problematic, as Associate Commissioners Naglich and Culliver admitted: although passing prisoners are able to slip items to those housed in crisis cells at a number of facilities, this sort of exchange is particularly easy at Holman, where the crisis cell doors have open bars. Yet, when asked what MHM had done to address this issue, Houser responded that after each incident, MHM staff would “ask [ADOC] to please do a better job.” Houser Testimony at vol. 3, 16. She could not identify any other efforts either by MHM or ADOC to address this issue.⁵⁴

e. Inadequate Monitoring of Suicidal Prisoners

The unsafe features of crisis cells heighten the importance of monitoring prisoners for signs of decompensation or suicide attempts. However, ADOC’s •monitoring practices are woefully inadequate.

According to ADOC’s administrative regulations and the standard of care for mental-health care in prisons, suicide-watch checks should take place at staggered, or random, intervals of approximately every 15 minutes, rather than exactly every 15 minutes. For prisoners on mental-health observation, these staggered checks should occur approximately every 30 minutes. Staggered intervals prevent prisoners from timing their suicide attempts, because otherwise they can predict exactly when checks will occur. Such monitoring procedures are all the more crucial when suicidal inmates are housed in cells that have little visibility: as plaintiffs’ expert Vail bluntly stated, without regular checks, “[Y]ou have no idea if they’re alive or dead.” Vail Testimony at vol. 1, 96.

Dr. Burns and Dr. Haney both testified that many of the monitoring logs they had seen during their site visits and document review had pre-printed times or had handwritten pre-filled times at exact intervals. This practice reflects prison staffs lack of understanding that checks should be performed at staggered intervals, and makes it impossible to ensure that staggered checks are actually happening. Associate Commissioner Naglich admitted that staff are not permitted to use monitoring logs with pre-printed times, but that some continue to use them. She also testified that officers and staff are not permitted to handwrite times and signatures in advance of, or in lieu of, their actual checks. However, during the post-trial prison tours, the court came across multiple logs where times at 15- or 30-minute intervals had been pre-filled; even though the parties had agreed during the trial to correct this practice, and the court had ordered compliance with the agreement several weeks before the tours. This evidence of noncompliance greatly troubled the court, as it showed that policy changes are not being implemented on the ground eveq when a court order is involved.

For the most acutely suicidal, constant — rather than staggered-interval— watch is necessary. As Dr. Burns opined, correctional systems must have a constant-watch procedure for individuals whose risk of suicide is the highest, due to their engagement in self-injurious behavior or threat of suicide with specific plans: if a prisoner is waiting for an opportunity to kill himself, it is too dangerous to walk away, and he must be constantly observed. For this reason, the NCCHC standards classify constant-watch procedures as an “essential” standard, and MHM is contractually obligated to follow all NCCHC standards.⁵⁵

ADOC and MHM had not provided constant watch for acutely suicidal inmates prior to Jamie Wallace’s death. During the trial, in the wake of Jamie Wallace’s suicide, the court urged the parties to propose interim measures tó prevent more suicides. Plaintiffs then filed a motion for temporary restraining order seeking to institute constant wátch and other suicide-prevention measures. Plaintiffs’ Emergency Motion for Temporary Restraining Order (doc. no. 1075). The parties reached an interim agreement in early January. Phase 2A Interim Relief Order Regarding Suicide Prevention Measures (doc. no. 1102). The agreement mandated a constant-watch procedure for those deemed acutely suicidal and forbade using pre-printed or pre-filled forms for other types of suicide watch. While defense counsel represented to the court that it was Commissioner Dunn’s intent to keep the constant-watch procedure until told otherwise by the court or experts, the court also heard testimony that the current implementation of suicide-prevention measures and constant watch is not sustainable.⁵⁶ The parties defined ‘constant watch’ as a “procedure that ensures one-on-one visual contact at all times, except to the extent that the physical design allows an observer to maintain an' unobstructed line of sight with no more than two people on suicide watch at once.” Interim Agreement Regarding Suicide Prevention Measures (doc. " ño. 1102-1). MHM’s Houser testified that the implementation has been difficult because some facilities do not have a layout conducive to constant watch, due to the location of the windows on cell doors and structures that obstruct a direct line of sight into crisis cells. As a result, MHM has had to transfer some prisoners to other facilities. Another obstacle in the implementation stems from a lack of sufficient correctional staffing: for example, the Holman crisis cells, located on death row, are unsafe for mental-health staff, because without sufficient correctional staffing on duty, .prisoners often throw objects from second and-third tiers at the mental-health staff conducting constant watch on the first tier. Finally, according to Houser, the annual budget fpr a permanent constant-watch procedure is projected to be-over $4 million, but MHM was initially provided only $200,000 to meet the immediate needs of the interim agreement mandating constant watch.⁵⁷

f. Inappropriate Release from Suicide Watch and Inadequate Follow-up

Prisoners are routinely released from suicide watch improperly and receive inadequate follow-up care after their release from suicide watch. These practices create a substantial risk of recurring self-injurious behavior and suicide.

As experts from both sides explained, suicidal prisoners should be released only with the approval of a psychiatric provider (psychiatrist or ' nurse practitioner) who has made a face-to-face assessment that their condition was sufficiently stabilized to warrant it. In 20Í6, MHM reported to ADOC that it was discharging patients from suicide watch without,a face-to-face assessment; the decisions were based instead on whatever information lower-level mental-health staff communicated over the phone to on-call doctors and nurse practitioners. A-nurse practitioner at'Donaldson and St. Clair testified that generally she will not authorize the release of a prisoner from suicide watch at St, Clair without seeing him in person; however, when she is not at St. Clair (a significant majority of the hours in the week), staff call Dr. Hunter to authorize the release remotely. Associate Commissioner Naglich admitted that this practice of authorizing' suicide-watch release without a face-to-face evaluation was not specific to any particular facilities, but that it reflected a general shortage of psychiatrists; she further agreed that it put the prisoners at risk of premature release. Evidence also showed that prisoners have, on occasion, been released from suicide watch by correctional staff without any mental health assessment at all; this is even more unacceptable. See, e.g., PI. Ex. 436, September 19, 2014 Email between Houser and ADOC (doc. no. 1074-26) (notifying Naglich about a death-row inmate who was released from a crisis cell by ADOC officer, without notice or approval by mental-health staff).

According to experts on both sides, follow-up care is necessary upon release from suicide watch both for prisoners on the mental-health caseload and for those who are not. For those who are already on the mental-health caseload, follow-up care entails incorporating what providers learned from the most recent crisis into the prisoner’s treatment plans and modifying interventions in order to address the factors that contributed to. the self-injurious behavior or suicidal ideation. For those who were not on the caseload, follow-up care allows providers to assess whether the prisoner’s risk of self-injury remains, low, and to determine whether the prisoner should be added to the mental-health caseload to address underlying mental-health issues. As Dr. Burns credibly opined, the failure to provide follow-up care that addresses the root of self-injurious behavior creates a substantial risk that the self-injurious behavior will continue and result in serious injury or death.

The follow-up care provided to many prisoners upon their release from suicide watch at ADOC is woefully inadequate. Both Dr. Haney and Dr. Burns observed multiple instances of prisoners who were released directly from crisis cells back into segregation, with little or no follow-up treatment in subsequent weeks. For example, experts observed that plaintiffs L.P., R.M.W., and C.J. and prisoner J.D. all had a pattern of cycling between crisis cells and segregation with little follow-up treatment after crisis-cell release. As explained further, later, prisoners in ■ segregation— even those on the mental-health caseload — have little access to meaningful treatment, due to severe staffing shortages that prevent prisoners from being brought out of their cells and a lack of group activities. ■ ‘ ■

Once again, Jamie Wallace provides a concrete example of the lack of follow-up care and the resulting harm. During his testimony, he repeatedly insisted that he rarely received therapeutic care when not on suicide watch. Dr. Burns corroborated his testimony, noting that despite his very acute mental illness, Wallace had only one individual counseling session in the two-month period following a suicide watch placement in 2015, and that his treatment plan did not change or reflect the fact that he came off of suicide watch in late August 2016. The same lack of follow-up care was repeated in 2016: he was discharged from suicide watch two days before he commit ted suicide; in those two days, he received no follow-up care.

In sum, the combination of inadequate identification of needs for crisis care, unsafe cells, inadequate monitoring, and inadequate treatment has created a substantial and. grave risk of serious harm for ADOC’s prisoners who have a high risk of engaging, in self-injurious behavior and suicide attempts.

7. Inappropriate Use of Disciplinary Actions

ADOC has an unacceptable practice of disciplining mentally ill prisoners for behavior that stems from their mental illnesses and doing so without adequate regard for the disciplinary sanctions’ impact on mental health. Mentally ill prisoners are routinely- disciplined for harming themselves or attempting to do so. These punitive practices in turn subject mentally ill prisoners to a substantial risk of decom-pensation and increased suffering. Cf.Coleman v. Wilson, 912 F.Supp. 1282, 1320 (E.D. Cal. 1995) (Karlton, J.) (“[Bjeing treated with punitive measures by the custody staff to control the inmates’ behavior without regard to the cause of the behavior, the efficacy of such measures, or the impact of those measures on the inmates’ mental illnesses” violated -seriously mentally ill prisoners’ Eighth Amendment rights); Casey v. Lewis, 834 F.Supp. 1477, 1548-49 (D. Ariz. 1993) (Muecke, J.) (finding that using lockdowns to punish seriously mentally ill prisoners’ behavior stemming from their illness constitutes an Eighth Amendment violation).

Imposing disciplinary sanctions on prisoners for engaging in self-injury creates an additional risk of harm beyond that stemming from inadequate treatment. As plaintiffs’ expert Burns explained, because ADOC’s practice treats self-injury solely as a behavioral problem rather than a mental-health problem, it fails to address the underlying mental-health issues through treatment; responding to self-harm in this manner is likely to escalate the self-injurious behavior, potentially resulting in serious physical injury or even death. Furthermore, if a disciplinary action results in segregation, mentally ill prisoners are at an even higher risk of harm — as will be discussed in detail later — because of the detrimental effects of isolation and of the limited access to treatment, both of which can in turn worsen underlying mental illness.

The practice of punishing prisoners for engaging in self-harm is common and system-wide at ADOC, despite a written policy purporting to prohibit it. ADOC’s administrative regulation states that, although they are not exempt from compliance with rules and regulations, inmates “will not be punished for symptoms of a mental illness.”⁵⁸ ■ Joint Ex. 128, Admin. Reg. § 626 (doc. no. 1038-151). ADOC has engaged in a practice of automatically disciplining prisoners who engage in self-injurious behaviors. In fact, Naglich’s Office of Health Services deemed this practice problematic as early as 2013, when it conducted an audit of services provided in Donaldson. As a result, MHM’s post-audit corrective-action plan stated that ADOC is to stop “automatically applying] disciplinary sanctions to male inmates who engage in self-injurious behavior.” PI. Ex. 689, MHM Corrective Action — Donaldson May 2013 (doc. no. 1069-5) at ADOC045459.⁵⁹ The person responsible for implementing this change was Dr. Ron Cavanaugh of OHS, who was to review files of prisoners who may have been sanctioned for symptoms of mental illness and send instructions on how to deal with self-injurious behavior to ADOC officials in charge of supervising the disciplinary process.

Although the 2013 corrective-action plan required follow-up action to address this issue, ADOC did not take meaningful action to change this practice, and prisoners continue to face sanctions for self-injurious behavior. Associate Commissioner Nag-lich’s staff could find no documentation of any file reviews conducted by Dr. Cava-naugh or instructions sent to Associate Commissioner Culliver or the regional coordinators, who according to Naglich were the officials responsible for enforcing this policy change. Associate Commissioner Culliver was likewise not aware of any policy change or new instructions regarding self-harm and disciplinary sanctions. Dr. Tytell, who replaced Dr. Cavanaugh after his death and was aware of this issue at Donaldson, testified that he and Associate Commissioner Naglich have discussed that imposing disciplinary sanctions for self-injury continued to be a problem, including in the RTU and SU. However, Dr. Tytell has done nothing to monitor, let alone address, this issue. When asked what, if anything, she personally has done to implement this policy change, Associate Commissioner Naglich admitted that she had reviewed only one single prisoner’s disciplinary record; in that case, she intervened to recommend that convictions be removed based on the indications in his medical records that he was decompensat-ing at the time of the infraction.

Not surprisingly, in 2016, plaintiffs’ expert Dr. Bums credibly concluded that “desperate acts to get the attention of MHM staff and necessary services,” including self-injury and suicide attempts, “often result in disciplinary action and placement in segregation where mental health treatment is even more difficult to access.” Joint Ex. 460, Burns Expert Report (doc. no. 1038-1044) at 29. She also saw evidence of prisoners with untreated serious mental illness being “essentially punished for symptoms of their psychiatric illness,” such as prisoners with bipolar disorder being placed in disciplinary segregation for untreated manic behaviors. Burns Testimony at vol. 1,27-28.

A related problem is ADOC’s inadequate mental-health evaluation process for prisoners facing disciplinary charges. Not taking mental health into consideration when determining appropriate sanctions is dangerous because certain sanctions, such as placement in segregation, expose mentally ill prisoners to a substantial risk of worsening symptoms and significantly reduced access to monitoring and treatment.

Under ADOC’s administrative regulations, disciplinary actions against prisoners whose mental-health code is MH-1 or above require consultation with mental-health staff: once a prisoner on the caseload is charged with a disciplinary infraction, MHM’s mental-health counselors are required to conduct a mental-health evaluation and complete a computerized module. Ostensibly, this system allows the counsel- or to have input into the disciplinary process and to communicate in writing to the disciplinary hearing officer: (1) whether “mental health issues affected the inmate’s behavior at the time of the charge”; (2) whether there are “mental health issues to be considered in disposition if found guilty”; and (3) whether mental-health staff would be present at the hearing. Joint Ex. 467, Mental Health Consultation to the Disciplinary Process, Inmate File of Jamie Wallace (doe. no. 1038-1052) at ADOC031346.

However, the system falls far short in practice: these mental-health evaluations are often brief and perfunctory, and the counselors conducting them understand their role to be limited to an assessment of capacity or knowledge of their infraction, rather than providing input on the mental-health implications of any punishment. For example, Sharon Trimble, an MHM counselor at Kilby, testified that her evaluation process entails informing the prisoner of the charge against him, describing the incident at issue, letting him explain what happened, and making sure that he understands the reasons for a disciplinary hearing, This, in her view, amounts to an "assessment of the prisoner’s competency; her evaluation concludes when the prisoner “say[s] that [he] did it.” Trimble Testimony at_She does not otherwise assess whether the prisoner’s behavior is related to his mental illness, and she has never made'recommendations as to the appropriateness of possible sanctions, including whether placement in segregation was contraindicated by the prisoner’s mental illness:

Strikingly, Associate Commissioner Naglich herself did not havé a clear understanding of the purpose of the consultation process. While she understood that the consultation process should address whether “the mental health issues eontribute[d] to the conduct,” she was unsure about whether it involved anything else. Naglich Testimony at vol. 2,15. She understood the second question in the module — whether mental illness should be considered in determining the punishment — to relate not to the appropriateness of various sanctions in light of the prisoner’s mental illness but rather to be largely duplicative of the first question, regarding culpability. She believed it .to be asking “how cognizant or responsible was the inmate at the time of the charge and should that be considered in the disposition if he’s found guilty.” Id.

■ As explained in the next section, ami as agreed .by experts on both sides, it is critical that mental illness be considered in determining punishment for infractions, because placing mentally ill prisoners in segregation significantly increases the risk of decompensation. ADOC’s failure to ensure that mental-health staff can and in fact do express their views as to whether particular prisoners will be harmed by placement in segregation (or some other disciplinary sanction) creates a substantial risk of serious harm.

Moreover, the disciplinary consultation process consistently fails to perform even the limited functions Trimble and Naglich ascribed to it. ADOC’s 2013 audit of Donaldson and a quality-improvement study conducted by MHM around the same time recognized that mental-health consultations were often acting as little more than a rubber' stamp. The Donaldson audit found that “answers provided by [mental health] appeared to conflict with patients’ clinically documented mental health status” — in other words, the consultation documentation from mental-health staff did hot reflect the diagnoses in the medical record of the prisoner who was being disciplined. PL Ex. 689, MHM Corrective Action — Donaldson May 2013 (doc. no. 1069— 5) at ADOC045459. MHM found that “95 % were declared competent to stand hearing with no qualifiers for MH factors,” and, relatedly, “that [MHM’s] staff did not understand how to fill out form.” PI. Ex. 715, July 2013 Quarterly CQI Meeting Minutes (doc. no. 1044-9) at 4. Not surprisingly, MHM counselor Trimble was aware of only one instance in the course of five years in which a prisoner was not sanctioned because his behavior was considered a result of his mental illness.

The consequences of ADOC’s policy of disciplining prisoners for engaging in self-harm combined with the dysfunctional consultation process are frequently egregious: when they attempt to hurt or kill themselves, mentally ill prisoners are routinely found guilty of and punished for “intentionally creating a. security, safety, or health hazard,” and -often are placed in segregation. For example, Jamie Wallace was given disciplinary sanctions and sent to segregation for self-injury and suicide attempts multiple times between 2013 and his suicide in 2016. See Joint Ex. 467, Inmate File of Jamie Wallace (doc. no. 1038-1062) at ADOC031352 (Jan. 8, 2013, for cutting his neck with a metal top of a smokeless tobacco can); ADOC031661 (Feb. 3, 2013, attempting to hang himself); ADOC031341 (Nov. 12, 2013, penetrating his ears and bottom lip with a metal object); ADOC031528 (May 25, 2014, intentionally cutting his left wrist); see also PI. Dem. Ex. 2, Summary of J.W. Suicide Attempts (doc. no. 1058-16) (showing six occasions of being sent to segregation for inflicting self-harm, and 12 disciplinary actions for self-harm in total). Records of plaintiff L.P. also reflect that he has received disciplinary segregation for self-harm incidents; plaintiff R.M.W. and class member M.P. testified that they have received multiple disciplinary actions for intentionally creating a security, safety, or health hazard when they had cut themselves. These instances of punitive response can also lead to even graver harm: Dr. Hunter, the medical director of MHM, acknowledged that the combination of a recent disciplinary action and the prospect of a segregation placement was a common factor among prisoners who committed suicide. The trend in suicides since October 2015 corroborated this testimony. In sum, ADOC’s disciplinary process has inflicted actual harm and created a substantial risk of serious harm for mentally ill prisoners.

8. Inappropriate Placement and Inadequate Treatment in Segregation ■

Segregation — also known as restrictive housing or solitary confinement — generally refers to the correctional practice of keeping a prisoner in a cell for 22.5 hours .or more a day, usually in a single-person cell, only letting the prisoner out for brief ‘yard’ time and showers.⁶⁰ In ADOC, segregation takes two different forms: disciplinary and administrative. Disciplinary segregation is a type of punishment whereby prisoners are allowed to have extremely limited personal property in their cells and lose privileges such as telephone use and family visits. Administrative segregation is used to separate prisoners.from the general population, generally for safety reasons; prisoners in administrative segregation do not formally lose privileges, but are still subject to some property restrictions and receive little out-of-cell time.

Trial testimony revealed that segregation has a profound impact on prisoners’ mental health due to the harmful effects of isolation; this impact is worse for those who are already mentally ill. According to the experts, the risk 'of decompensation increases with the duration of isolation and the severity of the prisoner’s mental illness.

Plaintiffs ask the court to declare that, due to the risk of harm, mentally ill prisoners as a general matter should never be placed in segregation. However, the court sees no need to reach that broad conclusion, for here, the evidence is 'over-whelming that the ADOC’s current segregation practices pose an unacceptably high risk of serious harm to prisoners with serious mental-health needs. As the testimony of experts and defense witnesses made abundantly clear, ADOC lacks a functioning process for screening out prisoners who should not be placed in segregation due to mental illness or ensuring that they are not sent there for dangerously long periods, and mentally ill prisoners in segregation receive inadequate treatment and monitoring. It is simply undeniable that these practices pose a grave danger to many mentally ill prisoners placed in segregation.

This section discusses the ways in which ADOC’s segregation practices place these prisoners at a substantial risk of serious harm. After explaining the consensus developed in recent years regarding the harmful psychological effects of segregation in general and on mentally ill prisoners in particular, the discussion turns to the specific risks of harm posed by ADOC’s segregation practices. Finally, the court discusses the heightened level of danger segregation poses to those prisoners with the most serious mental-health needs — that is, those who have conditions classified as serious mental illnesses, a. Background on Segregation

i. Consensus among Correctional and Mental-Health Professionals on Segregation

Mental-health and correctional professionals have recognized that long-term isolation resulting from segregation, or solitary confinement, has crippling consequences for mental health. Dr. Craig Haney, who has studied the psychological effects of solitary confinement for more than 30 years, explained that isolation of the type experienced by prisoners in segregation has harmful psychological effects even on those who are not mentally ill, and even mentally healthy prisoners can develop mental illness such as depression, psychosis, and anxiety disorder during a prolonged period of isolation. Summarizing years of research in his field, Dr. Haney explained: “[T]he nature and magnitude of the negative psychological reactions ... underscore the stressfulness and painfulness of this kind of confinement, the lengths to which prisoners must go to adapt and adjust to it, and the risk of harm that it creates. The potentially devastating effects of these conditions are reflected in the characteristically high numbers of suicide deaths, and incidents of self-harm and self-mutilation that occur in many of these units.... These effects are not only painful but can do real harm and inflict real damage that is sometimes severe and can be irreversible.They can persist beyond the time that prisoners are housed in isolation and lead to long-term disability and dysfunction.” Joint Ex. 459, Haney Expert Report (doc. no. 1038-1043) at 130-31; see also Davis v. Ayala, — U.S.-, 135 S.Ct. 2187, 2210, 192 L.Ed.2d 323 (2015) (Kennedy, J., concurring) (summarizing case law and historical texts that “understood!; ] and questioned” the “human toll wrought by extended terms of isolation” and observing that “research still confirms what this Court suggested over a century ago: Years on end of near-total isolation exact a terrible price.”) The psychological harm from segregation can also lead to symptoms like hallucinations, chest pain, palpitations, anxiety attacks, and self-harm, even among previously healthy people. Burns Testimony at vol. 1, 209; see also Palakovic v. Wetzel, 854 F.3d 209, 225-26 (3d Cir. 2017) (summarizing the “robust body of legal and scientific authority recognizing the devastating mental health consequences caused by long-term isolation in solitary confinement,” including “anxiety, panic, paranoia, depression, post-traumatic stress disorder, psychosis, and even a disintegration of the basic sense of self-identity,” as well as physical harm). The depth of the psychological impact of such isolated confinement conditions on human beings was also reflected in Senator John McCain’s observation about his prisoner-of-war experience in Vietnam: '“[Solitary confinement] crushes your spirit and weakens your resistance more effectively than any other form of mistreatment. Having, no one else to rely on, to share confidences with, to seek counsel from, you begin to doubt your judgment and your courage.” PI. Ex. 1272, 2012 Congressional Testimony of Dr. Craig Haney (doc. no. 1126-3) at 9 (quoting from Richard Kozar, John McCain: Overcoming Adversity (2001) at 53).

The serious psychological harm stemming from segregation is even more devastating for those with mental illness. As Dr. Haney explained, mentally ill prisoners are highly likely to decompensate in such an isolated environment, and it is more difficult to deliver treatment to those in segregation units. In other words, mentally ill prisoners in segregation are hit with a double-whammy: they are exposed to a heightened risk of worsening symptoms, while having less access to treatment they need. As , a result of the growing body of evidence on the destructive effects of segregation, a general consensus among correctional and psychiatric professionals, while not necessarily establishing a constitutional floor, has developed in the last ten years: placement and duration of segregation should be strictly limited for mentally ill prisoners. For example, as the experts explained, the National Commission on Correctional Health Care has issued a position statement declaring that mentally ill prisoners should not be placed in segregation absent extenuating circumstapces, and even in those circumstances, the stay should be shorter than 30 days.⁶¹

Defense witnesses agreed that mentally ill prisoners should rarely be placed in segregation for,prolonged periods of time. Dr. Hunter, MHM’s medical director, testified that it is “generally recognized” in the profession, including within ADOC, that prolonged segregation is deleterious to mental health, because of the combination of sensory deprivation and sensory overload: a severe lack of stimulation arises when confined to one space for over 23 hours a day without any meaningful social interactions; sensory overload comes from the chaotic environment of segregation units, filled with loud noises and malodors. Hunter Testimony at_ ADOC’s chief psychologist Dr. Tytell and MHM. psychiatrist Dr. Kern also agreed that overwhelming research shows that prolonged isolation has gravely detrimental effects on mental health, especially for those with pre-existing mental illness. Lastly, Ayers, a defense expert, opined that based on his experience as a correctional administrator, mentally ill prisoners should generally not be placed in segregation; if they are, it should only occur with the explicit approval and hands-on involvement of mental-health staff, and such prisoners should be placed on a fast-track to be moved into more therapeutic settings,

ii. ■ ADOC’s Segregation Units

The court heard overwhelming evidence, including from experts on both sides, that the conditions in ADOC’s segregation units pose serious risks for mentally ill prisoners — beyond the inherent psychological risks of segregation. ADOC prisoners receive very little out-of-cell time; they are left idle for almost all hours of the day with very little property allowed in the cell; the physical conditions of the segregation cells are often deplorable; and the design of the cells often makes it difficult to monitor the well-being of the prisoners. Associate Commissioner Culliver testified that they “try to give them five hours a week” of out-of-cell time, which means that even when ADOC officers are able to meet their goal, prisoners spend • on average over 23 hours per day inside of a cell. Culliver Testimony at_As for idleness, not only do segregation prisoners lack access to programming, but they are allowed very few items in their cells to occupy themselves: only a Bible and their current legal paperwork. As Dr. Haney credibly testified based on his extensive experience, it is quite unusual for. segregation inmates to be denied access to any other books or a radio. Furthermore, segregation units within ADOC are in significant disrepair, exacerbating the inherent stress of being confined to a small cell and wprsening its impact on mental health. As reflected by photographs admitted into the record and as the court witnessed' firsthand during facility visits, segregation cells are often poorly lit, with little natural light and only small grated windows, if any. The court observed that they are often-filled with the smell of burning paper and urine; some are extremely dirty with-what appears to be dried excrement smeared on the walls and floors; and loud noises travel through the segregation units, some of which house from anywhere between 20 to 50 people on multiple levels.-⁶² The court witnessed an overpowering sense of abandonment and despair, with a prolonged stay crushing all hope.⁶³.

The combination of the lack of any meaningful activity or social contact and the stressors of living in’ a dilapidated, filthy, and loud housing unit for almost 24 hours per day results in a heightened risk of decompensation for mentally ill prisoners and a heightened risk of developing serious mental-health needs for those who were initially healthy. In addition, as Dr. Haney credibly testified, it is much more difficult for staff to detect, decompensation of prisoners while they are housed in segregation: when prisoners remain in their cells around the clock, mental-health staff have a harder time observing the patient and diagnosing illnesses effectively, and correctional officers and fellow prisoners also lack sufficient regular contact with the prisoner to notice the onset of symptoms of mental illness. This difficulty adds to the danger.⁶⁴

The design of ADOC’s cells and units in which they exist poses additional obstacles for effective monitoring in segregation units. ADOC segregation units often lack visibility into cells, both because of small windows on the doors, which are often grated or difficult to see through, and because of-the layout of the cells and the units. Unfortunately, as experts from both sides testified, because of understaffing, officers cannot constantly . walk near the cells and are generally unable to monitor what is going on inside. This means that mentally ill prisoners in segregation — including those identified as mentally ill, those with undiagnosed mental illnesses, and those who develop mental illness while in segregation — are at a heightened risk for decompensation without anyone noticing.

These problems exist throughout ADOC facilities. For example, Easterling’s unit has tiny windows on doors that do not allow correctional officers to observe inside without being directly in front of the door; as Dr. Haney credibly testified, correctional officers often do not feel safe standing very close to the door because they risk having bodily fluids or food thrown at them through the food-tray slot or the cracks between the door and the wall. (Indeed, the court was repeatedly warned not to walk too close to the doors for that reason during facility tours.) As the court saw firsthand, Donaldson and St. Clair facilities have the same problem of very little visibility into the cells from the officers’ station, due to small windows and dim lighting. Lastly, Bibb’s segregation units might be the most egregious in terms of visibility: each housing unit has its own segregation unit of a few cells shut off from the rest of the unit, down a long hallway and through a door, with no line of sight from the central officer station and officers entering the space to check on the prisoners only periodically. Dr. Haney was surprised that such units were maintained, because prisoners in these cells have no way of alerting officers if anything was going wrong; they are completely dependent for their safety upon periodic trips that officers make from the central officer station. In fact, Dr. Haney recommended that Bibb’s segregation units be closed immediately: he explained that he has never recommended any unit to be closed immediately in his four decades of doing this work, but he thought the risk of harm was too great at Bibb because so little monitoring is available. Defense correctional expert Ayers’s testimony also - raised concerns: he credibly testified to his suspicion that, because of understaffing and safety concerns, correctional officers were not walking down the hallway away from the central cube at Bibb as frequently as they claimed.

b. ADOC’s Segregation of Mentally Ill Prisoners

The evidence clearly establishes that placements of mentally ill prisoners in segregation endangers those prisoners, and that the risk of serious harm to those prisoners increases based on the seriousness of the prisoner’s illness, the length of the stay in segregation, and the dangerous conditions discussed above. Against this backdrop, the court explains the ways in which ADOC’s placement practices and treatment of mentally ill prisoners in segregation create a substantial risk of serious harm.

i. ADOC’s Segregation-Placement Practices

Due to the risks of decompensation created by segregation in general and by ADOC’s segregation units in particular, it is critically important that ADOC consider a prisoner’s mental health condition when deciding whether to place the prisoner in segregation, and if so, for how long. But here, overwhelming evidence makes clear that ADOC does not ensure that those with a heightened risk of serious harm from mental illness are not placed in segregation or that they are not sent there for dangerously long periods.⁶⁵ In particular, as discussed earlier, ADOC does not have a functioning system for evaluating mental-health risks when deciding whether to place prisoners in segregation; it also fails to evaluate these risks when determining the length of any segregation placement. The result is that prisoners whose mental illness makes them likely to be harmed by segregation are placed there anyway.

ADOC’s current process for placing prisoners in segregation does not adequately consider the impact of segregation on mental health. As explained in the section on disciplinary sanctions, ADOC’s administrative regulations mandate that during disciplinary proceedings, mental-health staff provide input to ADOC regarding the impact of mental illness on the prisoner’s competency at the time of the offense and at the time of the hearing and give recommendations for the disposition of the offense and the type of sanctions that should be imposed. However, as discussed earlier, MHM staff and ADOC officials expressed confusion as to what role, if any, mental-health staff should play in the disciplinary process, and mental-health staff largely have rubber-stamped ADOC’s decisions to send mentally ill prisoners to segregation.

Even when MHM has recommended against placing a particular prisoner or a group of mentally ill prisoners in segregation, there is evidence that ADOC has ignored such input. As MHM’s program director Houser testified, ADOC has overridden MHM’s recommendations that prisoners whose mental-health code is above MH-3 (which requires residential treatment in a mental-health unit) should not be placed in segregation; she also gave an example of a prisoner who was put in segregation despite MHM’s recommendation. She further explained that because MHM is not authorized to move any prisoners, ADOC can override MHM’s clinical judgment and house RTU patients in segregation. Indeed, ADOC correctional staff are not required to follow the recommendations of the mental-health staff in disciplinary proceedings. Likewise, while regulations require that prisoners in segregation undergo periodic mental-health evaluations, ADOC is not required to move the prisoner if the mental-health evaluation reveals that continued placement in segregation would be detrimental to the prisoner’s mental health. Joint Ex. 127, Admin. Reg. § 625 (doc. no. 1038-150) (“The ADOC psychologist or psychological associate will consult with the Warden or designee when their [segregation] mental health assessment indicates that continued placement in [segregation] is contraindicated by changes in the inmate’s mental status and functioning. Alternative strategies to facilitate the inmate’s mental stabilization will be offered.”).⁶⁶

For their part, MHM staff have been hesitant to oppose ADOC on the placement of mentally ill prisoners in segregation. MHM staff discussed ADOC’s use of segregation on mentally ill prisoners during a staff meeting in 2013, expressing frustration that ADOC was over-using segregation on mentally ill prisoners: the meeting summary read, “DOC is over using segregation on MH inmates. They want to punish them. We must be diligent in calling it from a treatment perspective in disciplinary consult. Put MH as factor in the bad behavior. Long term segregation can be detrimental mental well-being. ... Do not recommend a disciplinary action. Say MH is a major factor. We are reluctant to do it because of influence of DOC.” PI. Ex. 715, July 2013 Quarterly CQI Meeting Minutes (doc. no. 1044-9) at 4.

ADOC also fails to ensure that prisoners with serious mental-health needs are not subjected to extremely lengthy periods of segregation. Dr. Haney described examples of several plaintiffs and one former class member who have bounced between segregation units and suicide-watch cells over lengthy periods of time; three were never put on the mental-health caseload despite repeated instances of self-harm. See PI. Dem. Ex: 131, Movement History of Exemplar Plaintiffs (doc. no. 1126-10). In particular, plaintiff C.J.’s eight-yearlong movement history shows that he has been in segregation or suicide-watch cells for all of those eight years; his mental-health code was eventually elevated to MH-2, but his treatment plan did not change despite his clear deterioration over the years.⁶⁷ See Joint Ex. 459, Haney Expert Report (doc. no. 1038-1043) at A39.⁶⁸

Not surprisingly given ADOC’s disregard for segregation’s impact on mental health, mentally ill prisoners are overrepresented in ADOC segregation. While only 14 % of the ADOC population is on the mental-health caseload, mentally ill prisoners make up 21 % of those in -segregation. Looking at individual facilities year by year, most facilities’ segregation units have a far higher rate of mentally ill prisoners compared to the general population: throughout 2014, 2015, and 2016, Bibb, Easterling, Kilby, St;-Clair, Staton, and Ventress each had a disproportionately high number of mental-health patients in segregation; ■ Holman and Limestone’s segregation population also had a disproportionately high number of mental-health patients more than half of the time period. Only four of the 12 major male facilities— Bullock, Donaldson, Fountain, and Hamilton — did not have disproportionate numbers of mental-health patients in segregation for most of the three years.⁶⁹ See-Pl. Dem. Ex. 127, Overrepresentation of the Mentally Ill in .Segregation, 2014-2016 (doc. no. 1126-8).

Experts on both sides were alarmed by ADOC’s systematic overuse of segregation for mentally ill prisoners, who are most vulnerable to the risk of deterioration in such an isolated environment. Ayers, a defense expert for correctional administration who reviewed ADOC records, was troubled by forms he saw for administrative segregation in which the reason for segregation placement was ‘psychiatric.’ Dr. Haney and Dr. Burns were also troubled by the number of prisoners with unaddressed mental illnesses they encountered in segregation units. In sum, ADOC lacks a functioning process for screening out prisoners who should not be placed in segregation due to mental illness or ensuring that they are not sent there for dangerously long periods.

ii. Treatment and Monitoring in Segregation Units

ADOC prisoners with serious mental-health needs must contend not only with dangerous and unhealthy conditions^ in segregation units but also with significantly less access to mental-health treatment. Mental-health patients’ needs are considerably greater in segregation due to the harsh effects of isolation, yet instead of receiving more treatment to mitigate these effects, prisoners in segregation have less access to , care than in general population and are not adequately monitored for signs of decompensation. The court heard extensive evidence that, due to staffing shortages, mental-health treatment and monitoring .in segregation are gravely more limited than in general population, and nonexistent at some facilities. This denial of minimal medical care contributes to the substantial risk that prisoners in segregation with serious mental-health needs will decompensate, experience increased pain and suffering, or worse, harm or kill themselves.

As Houser, MHM’s program director, credibly testified, even though mental-health patients’ needs are considerably greater in segregation due to the harsh effects of isolation, prisoners in segregation are not allowed to leave.their cells for mental-health groups or therapeutic activities. As a result, mental-health patients in segregation ^receive less treatment than they otherwise would outside segregation, despite their heightened need.⁷⁰

On top of the lack of access to group therapy or other programming, ADOC’s segregation prisoners have very little access to individual treatment. For example, in the month of June 2016, the number of ‘seg interventions’ — that is, out-of-cell treatment encounters with mental-health staff — at seven facilities with mentally ill prisoners in segregation was zero, despite having many, sometimes dozens of, mental-health patients in those units; three facilities had more than zero but fewer than five seg interventions. See Joint Ex. 346, June 2016 MHM Monthly Operations Report (doc. no. 1038-708) at 2.

The dearth of individual treatment in segregation is mainly due to correctional understaffing. Houser observed that mental-health patients in segregation were not getting the services they required, “not by [MHM’s] choice,” but because of ADOC’s failure to bring inmates out of their segregation cells for treatment. Houser Testimony at vol. 2, 100. MHM staff, have consistently complained of the difficulties of reaching patients in segregation due to the chronic correctional staffing shortage. See, e.cj,, PI. Ex. 950, July 2014 Holman Multidisciplinary Meeting Minutes (doc. no. 1,097-4) (reporting issues with psychiatric providers seeing patients. in segregation due to “walks, feeding, and DOC shortage, etc.”); PI. Ex, 1191, 2012 Contract-Compliance Report (doc. no. 1070-9) (noting the lack of documentation or notes for treatment of mentally ill prisoners in segregation).

In the absence of correctional officers to provide security and escort for segregation prisoners who need mental-health treatment, mental-health staff have to conduct cell-front check-ins, instead of actual treatment sessions. - But because segregation units are not hospitable environments for a personal conversation — let alone confidential conversations — these- interactions are brief and cannot replace individual counseling sessions.⁷¹

‘Segregation rounds,-’ whereby mental-health counselors go arotmd the segregation unit to check on the well-being of prisoners, also are of limited utility due to understaffing and visibility issues. ADOC regulations require that these rounds happen at least twice per week. As with other cell-front encounters, segregation rounds are not meant to replace individual psychotherapy.⁷² However, within ADOC, segregation rounds do not adequately serve even the limited purpose they are intended to serve. Dr. Hunter described them as ‘drive-bys,’ sometimes even without verbal exchanges. The cursory nature of the monitoring was further crystalized by the testimony of staff who conduct these rounds. Dr. Tytell, who served as an ADOC psychologist at Donaldson before taking his current position, testified that segregation rounds for over 120 prisoners at Donaldson took between 1.5 hours and 2 hours, including the time to walk between cell blocks — meaning no more than one minute per prisoner on average. A former counsel- or at Bibb testified that it would take her 35 minutes to an hour to complete the rounds at all six housing units with 18 double-celled cells, meaning one to two minutes per prisoner, including the time to walk between six housing units. A lack of visibility into many of these cells — due to small, sometimes covered windows, blocked views, and safety concerns associated with standing too close to the door— makes it even more difficult to provide effective monitoring.

Even these cursory rounds by MHM staff do not actually happen as often as they should, or at all at some facilities. The lack of documentation of segregation rounds combined with the acute staffing shortages led defense expert Ayers to doubt that ADOC was able to conduct segregation rounds as often as required. The site administrator for Holman confirmed Ayers’s belief, by credibly testifying that insufficient segregation rounds have been a problem at Holman since 2008 due to staffing shortages, and that the problem has only worsened since then. According to her, at Holman, instead of a separate mental-health segregation round, a counselor accompanies the warden and other security officers during a weekly segregation review board, where the warden and other officials walk from cell to cell to review each segregation prisoner’s status and potentially change the prisoner’s segregation sentence based on their conduct. Sometimes, she is able to visit only one prisoner in segregation per week due to the correctional staffing shortage.

Monitoring by ADOC staff in segregation is also ineffective. Correctional expert Vail credibly opined that ADOC lacked enough correctional staff to conduct monitoring rounds in segregation every 30 minutes — the level of monitoring in segregation units necessary to keep prisoners safe from self-harm and suicide. Indeed, he saw logs at ADOC that suggested that no segregation checks were done for multiple hours. Even defense expert Ayers, while not explicitly concluding that monitoring was inadequate, implied so by saying that better monitoring of segregation inmates would address the high suicide rates within ADOC.

This lack of monitoring is even more troubling given that ADOC segregation cells are not suicide-proof. Many segregation cells have grates, sprinkler heads, and other structures that could be used as tie-off points. Furthermore, during the facility tour, the court saw many segregation prisoners with ropes hanging across their cells as clothes lines, which can be easily used to commit suicide. Allowing prisoners to cover their cell door windows with papers further heightens the risk of suicide.

The dearth of individual encounters outside the cell, haphazard cell-front encounters, and inadequate monitoring in ADOC all show that ADOC fails to provide adequate treatment and monitoring.

In sum, the evidence is clear that ADOC’s segregation practices — inadequate screening for the impact of segregation on mental health, and inadequate treatment and monitoring — pose a substantial risk of serious harm to prisoners with serious mental-health needs. This serious inadequacy also has effects on other areas of mental-health care. According to Dr. Haney, this is because “[i]t’s very difficult to deliver adequate mental-health care in isolation units, and mentally ill prisoners deteriorate in isolated units. So the inadequacies of the mental health system actually are exacerbated by the use of isolation for mentally ill prisoners.” Haney Testimony at vol. 1, 29. In other words, ADOC’s segregation practices perpetuate a vicious cycle of isolation, inadequate treatment, and decompensation.

The skyrocketing number of suicides within ADOC, the majority of which occurred in segregation, reflects the combined effect of the lack of screening, monitoring, and treatment in segregation units and the dangerous conditions in segregation cells. Because prisoners often remain in segregation for weeks, months, or even years at a time, their decompensation may not become evident until it is too late— after an actual or attempted suicide.⁷³ Since September 2015, seven of eleven suicides within ADOC facilities happened in segregation units; of the four that have occurred since October 2016 (the current fiscal year), all but one involved a prisoner in segregation.⁷⁴ As explained above, these suicide numbers are astounding compared to the national average across state prison systems. By subjecting mentally ill prisoners to its segregation practices, ADOC has placed prisoners with serious mental-health needs at a substantial risk of continued pain and suffering, decompensation, self-injurious behavior, ■■ and even death, and the court cannot close its eyes to this overwhelming evidence.

c. Segregation of Prisoners with Serious Mental Illness

The court heard significant evidence that extended-1 segregation — even absent consideration of the conditions at ADOC— poses a substantial risk of harm to all mentally ill prisoners, and- plaintiffs asked the court to so conclude. However, as mentioned before, because ADOC’s segregation practices1 fall so far short of protecting prisoners with serious mental-health needs from a grave risk of decompensation and other harms, the court need not, at this time, decide whether segregation poses an unacceptably high risk of harm to all mentally ill prisoners as a general matter. That said, the testimony of the experts, clinicians who work for ADOC, and e^en Associate Commissioner Naglich herself overwhelmingly established that one particular subset of prisoners with serious mental-health needs should never be placed in segregation in the absence of extenuating circumstances: those who suffer from a ‘serious mental illness.’

As discussed earlier, ‘serious mental illness’ is a term of art in the field of psychiatry that refers to a certain subset of particularly disabling conditions. Serious mental illness is defined by the diagnosis, duration, and severity of the symptoms. Certain diagnoses, such as schizophrenia and disorders accompanied by psychosis, are by definition serious mental illnesses, because they last a lifetime and are accompanied by debilitating symptoms; others, such as major depression and anxiety disorder, may be considered serious mental illnesses depending on the severity of the individual’s symptoms.

As Dr. Bums credibly opined based on the literature in the field, those who suffer from serious mental illness should not be put in segregation as a general matter because prisoners with serious mental illness experience worsening symptoms in such an isolated environment, and because they are likely to have reduced access to treatment in segregation units. Burns added that, even when extenuating circumstances exist, segregation placements for such prisoners should still be short term, and access to necessary treatment must be provided. Indeed, as Dr. Burns pointed out, the American Correctional Association and the American Psychiatric Association take the position that seriously mentally ill people should not be placed in segregation unless absolutely necessary, and if so, they should only remain for the shortest duration possible — no longer than three to foul-weeks. American Correctional Association, Restrictive Housing Performance Based Standards, August 2016; American Psychiatric Association, Position Statement on Segregation of Prisoners with Mental Illness (2012).

Associate Commissioner Naglich candidly agreed with Dr. Burns that placing seriously mentally ill prisoners in segregation is “categorically inappropriate,” and'that such placement is tantamount, to “denial of minimal medical care.” Naglich Testimony at vol. 5, 73. She described .a, new mental-health coding system in development at ADOC that would prevent all prisoners with serious mental illness from being placed in segregation. While she could not tell the court when the “rollout” of the new system would be complete, she assured the court that once completed, “no seriously mentally ill inmate would be housed in a segregation setting.” Naglich Testimony at vol. 5, 67. MHM’s program director Houser agreed with the bright-line rulé against placing prisoners with serious mental illness in segregation: she explained that prisoners classified as MH-3 or above, which are designated for RTU or SU placements and considered to have a serious mental illness, should never be in segregation because “their mental health capacity would not allow them to be able to be maintained in such an environment.” Houser Testimony at vol. 2,1Ó9.

While there was no dispute between the parties that placing seriously mentally ill prisoners in segregation amounts to denial of minimal care, a question was -raised- as to whether the new system that Associate Commissioner Naglich described has been implemented. Associate . Commissioner Culliver, -who has , the primary responsibility for inmate placements, transfers, and correctional staffing levels, testified after Naglich that there had not been any recent official policy change on the placement of mentally ill prisoners in segregation, and that he did not know about any changes that would prohibit officers from placing certain prisoners in segregation or would limit the duration of segregation placements. Naglich’s subordinate, Dr. Ty-tell, later testified that an effort to change the coding system began only after Nag-lich testified that the policy change was already being rolled out, and that no new official coding system existed. He further explained that she instructed him to email the wardens at Donaldson to move ten individuals whose mental-health code was MH-2 or higher out of segregation and into the RTU, only after her testimony in court. She did not instruct him to do so with any other facility, and Tytell was not aware of any other facilities moving mentally ill prisoners out of segregation units at the tune of his testimony in January 2017. Based on the evidence presented— especially given Associate Commissioner Gulliver’s lack of knowledge or involvement in a major change to segregation policy — the court cannot conclude ' that ADOC has implemented this policy change of not placing prisoners with serious mental illness in segregation.⁷⁵ Given the consensus on the substantial risk of harm of decompensation for these mostly severely mentally ill prisoners, the court concludes that it is categorically inappropriate to place prisoners with serious mental illness in segregation absent extenuating circumstances; even in extenuating circumstances, decisions regarding the placement should be with the involvement and approval of appropriate mental-health staff, and the prisoners should be moved out of segregation as soon as possible and have access to treatment and monitoring in the meantime.

9. Tutwiler

As ADOC’s only major facility for women, Tutwiler Prison for Women serves as the treatment hub for all female prisoners in Alabama: While the same' factors contributing to inadequate mental-health care; — mental-health understaffing, correctional understaffing, and overcrowding— apply to Tutwiler, the provision of mental-health care at Tutwiler differs in some ways. This is because Tutwiler'administra-tors, as a result of other litigation, have revised policies to make them more ‘gender-responsive’ and ‘trauma-informed’— that is, responsive to female prisoners’ experience of past traumatic events.⁷⁶ Some of these revisibns involve regulations governing mental-health care.⁷⁷,

Yet, despite these policy changes, the care provided to mentally ill prisoners at Tutwiler suffers from' some of the same inadequacies that affect mental-health care for men. Tutwiler lacks adequate mental-health and correctional staffing. As in the facilities for men, a significant portion of mentally ill patients are not being identified or appropriately classified; no suicide risk-assessment tool is used outside of intake; and the provision of counseling sessions is seriously inadequate. The court has sufficient evidence before it to conclude that these problems pose a substantial risk of serious harm to Tutwiler prisoners with serious mental-health needs.

Tutwiler suffers from the same serious deficiencies in identification and classification of prisoners’ serious mental-health needs. The mental-health identification and classification processes at Tutwiler function the same way as at male correctional facilities: an LPN conducts the initial intake screening, without any on-site supervision by an RN or any other higher-level provider. Tutwiler also lacks a triage system for referral requests, and therefore requests to see a mental-health provider do not get classified or tracked to ensure that they are processed. The resulting under-identification is apparent in the number of prisoners on the mental-health caseload. Experts from both sides testified that women in prison have a significantly higher incidence rate of mental illness compared to their male counterparts: the estimated rate ranges between 75 to 80 %, according to Dr. Burns. At Tutwiler, only 54 % of prisoners are on the mental-health caseload. Joint Ex. 846, June 2016 Monthly Operating Report (doc. no. 1038-708). As with the rest of the system, experts from both sides testified that the low rate stems from ADOC’s inadequate intake arid referral processes. Experts from both sides also testified that an insufficient number of prisoners are getting care in mental-health units at Tutwiler despite the severity of their illnesses. As explained above, such inadequate identification and classification of serious mental-health needs create a substantial risk of serious harm by failing to treat mental illness.

Expert testimony also showed that no suicide risk-assessment tool is being used at Tutwiler, except at intake, as is the case in male facilities. As explained earlier, failing to assess suicide risks of prisoners who threaten or attempt self-harm or suicide places those prisoners at a substantial risk of harm.

As at the male prisons, individual counseling sessions at Tutwiler are frequently delayed and canceled due to shortages of mental-health staff and correctional officers. An ADOC psychologist at Tutwiler testified that the correctional staffing shortage that causes such delays and cancellations of counseling sessions is a topic of discussion at almost every multidisciplinary meeting. Furthermore, MHM contract-compliance reports and the minutes from CQI meetings consistently reported that Tutwiler’s caseload is “bursting at [the] seams,” and that MHM had difficulty meeting outpatient needs for counseling. Ex. 670, April 2015 Quarterly CQI Meeting Minutes (doc. no. 1056-7) at MHM031224; see also PI. Ex. 532, 2015 Contract-Compliance Report (doc. no. 1070-7) at 4, 13 (“At Tutwiler, staff are attempting to mánage extremely large caseloads, which at times can be very challenging”; “significant staffing shortages in psychiatry” reported at Tutwiler); PI. Ex. 114, 2013 Contract-Compliance Report (doc. no. 1070-^4) at 1-2 (discussing decrease in treatment availability at Tutwiler due to staffing cuts and increasing size of caseload across all facilities).

In sum, inadequate identification and classification of mental-health needs, inadequate screening for suicide risk, and inadequate psychotherapy create a substantial risk of serious harm to mentally ill prisoners at Tutwiler. On the other hand, while also concerned about the number of crisis cells, suicide-watch placements, segregation placements, and treatment and monitoring available in segregation and in crisis care at Tutwiler, the court does not have sufficient evidence to find that those areas pose a substantial risk of serious harm to Tutwiler’s prisoners.⁷⁸

10. Other Issues

This section discusses several issues on which the court does not at this time find for the plaintiffs. First, there is substantial evidence that periodic mental-health evaluations for all prisoners in segregation are inadequate, but the court, out of an abundance of caution and exercising its discretion, leaves this issue to be further addressed by the parties. Second, evidence was insufficient to establish a substantial risk of serious harm arising from ADOC’s medication management practices or the supervision of certified registered nurse practitioners.

On the first issue, substantial evidence suggested that ADOC is not conducting adequate periodic mental-health assessments of prisoners in segregation to identify those who become mentally ill while in segregation. Dr. Haney credibly opined that periodic out-of-cell assessments are necessary not only to monitor for decom-pensation among those identified as mentally ill, but also to identify prisoners not on the mental-health caseload who may develop mental illness while in segregation. Just as identification and classification of mental-health needs at intake are essential in a functioning mental-health care system, it is also essential to identify those who need mental-health treatment in segregation. ADOC’s own administrative regulation requires periodic mental-health assessments of prisoners in segregation, even for those who are not on the caseload, though it does not appear to require out-of-cell assessments. Joint Ex. 127, Admin. Reg. § 625 (doc. no. 1038-150). However, evidence suggested that such assessments at ADOC are cursory at best. For example, as discussed above, plaintiff R.M.W.’s segregation mental-health evaluation form completed in the same month when she was sent to suicide watch twice and had multiple incidents of self-injury simply had some check marks and stated “inmate appropriate for placement” and “segregation placement not impacting inmate’s mental health.” Joint Ex. 404, March 28, 2014 Review of Segregation Inmates R.M.W. (doc. no. 1088-859) at MR017081. No mention of her suicide-watch placements or self-injury episodes was included, and no suicide risk-assessment tool was completed. Ample evidence of correctional and mental-health understaffing — and the fact that staff are often unable to conduct segregation rounds consisting of much shorter, cursory cell-front interactions — also suggests that ADOC is unable to provide meaningful mental-health assessments of prisoners in segregation. However, the court believes that it should solicit more input from the parties before determining whether ADOC is conducting adequate periodic mental-health assessments of prisoners in segregation. Therefore, the Eighth Amendment finding remains open as to this discrete issue, and the court will take it up with the parties after this opinion is issued.

Second, the court is able to conclude on the record before it that plaintiffs did not present sufficient evidence to establish that prisoners in ADOC custody face a substantial risk of serious harm in two areas: medication management and supervision of certified registered nurse practitioners. Plaintiffs did not present sufficient evidence to establish that ADOC’s medication management practices are inadequate based on ADOC allegedly letting cost concerns override clinical needs and not being responsive to patients’ concerns about side effects. While plaintiffs presented anecdotal evidence of providers’ refusal to continue previously prescribed medications or to switch medications despite continuing side effects, the court did not see any independent clinical assessments of these patients’ medication needs. Absent any contrary clinical assess-merits, credibility findings, or more direct evidence of ADOC’s failure to prioritize patients’ clinical needs over medication costs, a constitutional determination about the adequacy of these kinds of medication decisions would invade the province of psychiatric providers’ medical judgment. Estelle v. Gamble, 429 U.S. 97, 107, 97 S.Ct. 285, 50 L.Ed.2d 251 (1976) (holding that matters for “medical judgment” do not raise an Eighth Amendment concern), The testimony established only that clinicians talk about the cost of medications during meetings, and that managers commend providers for keeping prices down as a team; further, some prisoners were discontinued on medications they were originally prescribed, but there is no documentation about the reasons those medications were discontinued. However, these unconnected dots are not sufficient to. find that ADOC prioritizes cost concerns over clinical needs when making prescription décisions, because the court is ill-equipped to discern whether the decisions were clinically inappropriate. Furthermore, 'even plaintiffs’ expert Burns found that keeping the cost of medications in mind when making prescribing decisions was not on its own inappropriate or unusual, especially because MHM clinicians’ requests for medications that are not pre-approved for use are almost always granted. In other words, absent contrary clinical findings, there is not enough evidence to find that ADOC systematically overrides clinical needs due to cost concerns such that its medication management practices are constitutionally inadequate.

In addition, plaintiffs argued that ADOC’s certified registered nurse practitioners were not properly supervised by psychiatrists. Evidence suggested that some of the CRNPs employed by MHM could not meet the state regulatory requirement that they collaborate with .an on-site psychiatrist at least 10 % of the hours they work. However, evidence also showed that psychiatrists do supervise and collaborate with CRNPs through other, more informal channels. Therefore, there is insufficient evidence to establish that inadequate supervision has created a substantial risk of serious harm for mentally ill prisoners. -

C. Deliberate Indifference

Having found that ADOC’s mental-health care system creates substantial risks of serious harm to mentally ill prisoners (defined in this opinion as those with serious mental-health needs), the court now turns to the deliberate-indifference prong of the Eighth Amendment inquiry. In order to prove an Eighth Amendment violation, plaintiffs must show not only that state officials subjected mentally ill prisoners to a substantial risk of serious harm, but also that defendants acted with deliberate indifference to that risk. As discussed below, despite being repeatedly informed that significant deficiencies existed, ADOC has disregarded and failed to respond reasonably to the actual harm and substantial risks of serious harm posed by its deficient mental-health care system.⁷⁹

To establish deliberate indifference, plaintiffs must show that defendants had subjective knowledge of the harm or risk of harm, and disregarded it or failed to act reasonably to alleviate it. Thomas v. Bryant, 614 F.3d 1288, 1312 (11th Cir. 2010). Officials must “be aware of facts from which the inference could be drawn that a substantial risk of serious harm exists,” and “draw the inference.” Farmer v. Brennan, 511 U.S. 825, 837, 114 S.Ct. 1970, 128 L.Ed.2d 811 (1994). The defendant’s subjective awareness of a risk of harm can be determined based on circumstantial evidence, including “the very fact that the risk was .obvious.” Id. at 842, 114 S.Ct. 1970. In other words, if a particular risk was “longstanding, pervasive, well-documented, or expressly noted by prison officials in the past, and the circumstances suggest that the defendant-official being sued had been exposed to information concerning the risk and thus ‘must have known’ about it,” such evidence permits a trier of fact to conclude that the officials had actual knowledge of the risk. Id. at 842-43, 114 S.Ct. 1970 (internal citation omitted).

The disregard prong can be proven in many ways. In the area of medical care, disregard of a risk of harm may consist of “failing to provide care, delaying care, or providing grossly inadequate care,” when doing so causes a prisoner to needlessly suffer the pain resulting from his or her illness. McElligott v. Foley, 182 F.3d 1248, 1257 (11th Cir. 1999). Put differently, Eighth Amendment liability may be found if a defendant with subjective awareness of a serious need provides “an objectively insufficient response to that need.” Taylor v. Adams, 221 F.3d 1254, 1258 (11th Cir. 2000). Although considered part of the subjective component, the requirement that the defendant disregard a risk of harm actually evaluates her response (or lack thereof) by an objective ‘reasonableness’ standard. Farmer, 511 U.S. at 847, 114 S.Ct. 1970.

In some circumstances, a defendant’s disregard of a known risk is quite obvious. For example, the defendant might “simply refuse[ ] to provide” medical care known to' be necessary. Ancata v. Prison Health Servs., Inc., 769 F.2d 700, 704 (11th Cir. 1985) (allegations that prisoner required a-psychiatric evaluation that defendants refused to provide satisfies disregard requirement). If a defendant provides some medical care, the Constitution does not require that the care be “perfect” or the “best obtainable.” Harris v. Thigpen, 941 F.2d 1495, 1510 (11th Cir. 1991). Nonetheless, a defendant’s disregard of the risk can still be found through “delaying the treatment,” providing “grossly inadequate care,” making “a decision to take an easier but less efficacious course of treatment,” or providing “medical care Which is so cursory as to amount to no treatment at all.” McElligott, 182 F.3d at 1255 (collecting cases). In other words, a choice to provide care known to be less effective because it is easier or cheaper can constitute deliberate indifference. In the context of mental-health care, “the quality of psychiatric care can be so substantial a deviation from accepted standards as to evidence deliberate indifference to those- serious psychiatric needs.” Steele v. Shah, 87 F.3d 1266, 1269 (11th Cir. 1996) (citing Greason v. Kemp, 891 F.2d 829, 835 (11th Cir. 1990)). Deliberate indifference can also be found when “[a] prison official persists in a particular course of treatment in the face of resultant pain and risk of permanent injury” to the prisoner. Rouse v. Plantier, 182 F.3d 192, 197 (3d Cir. 1999).

In challenges to a correctional institution’s provision of medical care, evidence of systemic deficiencies can also establish the ‘disregard’ element of delibérate indifference. Harris, 941 F.2d at 1505. For example, this element may be met “by proving that there are ‘such systemic and gross deficiencies in staffing, facilities, equipment, or procedures that the inmate population is effectively denied access to adequate medical care.’ ” Id. (quoting Ramos v. Lamm, 639 F.2d 559, 575 (10th Cir. 1980), cert. denied, 450 U.S. 1041, 101 S.Ct. 1759, 68 L.Ed.2d 239 (1981)). As. an eviden-tiary matter, these systemic deficiencies may be identified by a “series of incidents closely related in time” or “[r]epeated examples of delayed or denied medical care.” Rogers v. Evans, 792 F.2d 1052, 1058-59 (11th Cir. 1986). Further, prison officials’ efforts to correct systemic deficiencies that “simply do not go far enough” when weighed against the risk of harm also support a finding of deliberate indifference, Laube v. Haley, 234 F.Supp.2d 1227, 1251 (M.D. Ala. 2002) (Thompson, J.), because such efforts are not “reasonable measures to abate” the identified substantial risk of serious harm. Farmer, 511 U.S. at 847, 114 S.Ct. 1970.

Finally, the defendant institution’s response to a known risk must be more blameworthy than “mere negligence.” Farmer, 511 U.S. at 835, 114 S.Ct. 1970 (citing Estelle v. Gamble, 429 U.S. 97, 106, 97 S.Ct. 285, 50 L.Ed.2d 251 (1976)). In other words, the defendant must have disregarded the risk with “more than ordinary lack of due care for the prisoner’s interests or safety.” Id. (quoting Whitley v. Albers, 475 U.S. 312, 319, 106 S.Ct 1078, 89 L.Ed.2d 251 (1986)). However, while an “inadvertent failure” to provide adequate medical care does not satisfy the deliberate-indifference standard, Estelle, 429 U.S. at 105-06, 97 S.Ct. 285, in challenges to health-care systems, repeated examples of negligent conduct support an inference of systemic disregard for the risk of harm facing mentally ill prisoners. See Ramos, 639 F.2d at 575 (“In class actions challenging the entire system of health care, deliberate indifference to inmates’ health needs may be shown by proving repeated examples of negligent acts which disclose a pattern of conduct by the prison medical staff.”).

In an official-capacity suit, the suit is not “against the official personally, for the real party in interest is the [governmental] entity”; therefore, the deliberate-indifference inquiry focuses on the institution’s “historical indifference” to the identified risk of harm, rather than the named defendant official’s personal indifference. LaMarca v. Turner, 995 F.2d 1526, 1542 (11th Cir. 1993) (finding that substitution of a new defendant, “a dedicated public servant who is trying very hard to make [the prison] an efficient and effective correctional institution” does not preclude a deliberate-indifference finding); see also Laube, 234 F.Supp.2d at 1249 (“[T]he real parties in interest-are the responsible entities: the Department of Corrections and, ultimately, the State of Alabama.”).

This case is likely sui generis in the extent to which the top ADOC officials had personal knowledge of the substantial risks of serious harm posed by its deficient care and has not responded reasonably to those risks. Much of the evidence came from ADOC officials’ own mouths: defendants — particularly Associate Commissioner Naglich — and other officials readily admitted to the existence of serious deficiencies, the risk of harm arising from them, and ADOC’s failure to respond. As a result, although plaintiffs do not have to prove personal deliberate indifference by the named defendants in order to establish institutional deliberate indifference, the court’s finding of deliberate indifference is well supported by defendants’ own admissions of knowledge and failure to act, in addition to the other circumstantial evidence more typically seen in official-capacity suits.

1. ADOC’s Knowledge of Harm and Risk of Harm

The inadequacies plaguing ADOC’s mental-health care system were pervasive and well-documented in multiple ways: ADOC received monthly statistical reports and annual contract-compliance reports from MHM; ADOC communicated with senior MHM managers through emails and quarterly CQI meetings; ADOC received corrective-action plans from MHM after compliance reviews and audits; ADOC also performed two audits of MHM’s performance since 2011. As a result, ADOC has been well aware of the risks presented by the deficiencies in its mental-health care.

ADOC has been well aware of the significant and adverse impact of overcrowding, mental-health understaffing, and correctional understaffing on the provision of mental-health care. Associate Commissioner Naglich admitted that, since 2010, MHM has been struggling to meet contractual requirements due to staffing cuts and increasing caseloads. In addition, MHM’s program director Houser repeatedly raised concerns about inadequate mental-health staffing with Naglich, requesting for over a year to amend the contract to increase staffing across facilities; she also told Naglich repeatedly that MHM needed more counselors in order to meet the rising demand, because ADOC’s psychological associates were not taking counseling caseloads from MHM providers as anticipated.

Both Dunn and Naglich have been aware that persistent correctional under-staffing has interfered with MHM’s ability to provide mental-health care. According to Naglich, in the years since 2010, MHM has repeatedly informed ADOC that the lack of sufficient correctional staffing has been seriously impacting its ability to provide care. ADOC’s own audit of the Donaldson RTU in 2013 also revealed that check-in rounds, individual appointments, and regularly scheduled activities had to be delayed or canceled due to the limited number of officers assigned to the mental-health unit'. At least since 2013, Naglich ' has repeatedly complained to- ADOC’s Commissioner, former Commissioner, and Associate Commissioner of Operations about the chronic shortage of correctional officers interfering with mental-health care. She characterized correctional under-staffing as “probably , one of the most serious problems facing the department,” Naglich Testimony atvol. 2,174-75.

Ample evidence ■ also demonstrates ADOC’s knowledge of the risks of harm arising out of the specific deficiencies in the treatment of mentally ill prispners discussed earlier. First, MHM managers repeatedly informed ADOC in their reports and emails that the deficiencies arising out of staffing shortages — including difficulties in providing timely counseling sessions and activities — were seriously undermining their ability to provide care. Second, Nag-lich admitted thát that the failure to meet the mental-health needs of prisoners with serious mental illness — in other words, the risk of harm arising from failing to identify prisoners in need of mental-health care and providing them'with the appropriate level of care — puts them at risk of decom-pensating.

ADOC was also well aware of the specific deficiencies. To begin, ADOC was aware that its processes for identifying and classifying mentally ill prisoners were inadequate. ADOC has had a persistently low prevalence rate of mental illnpss, and ADOC - officials have known that LPNs with extremely limited training are responsible for identifying prisoners’ needs for mental-health services. Moreover, Associate Commissioner Naglich was informed of the persistent’ pattern of self-injury, attempted suicides, and suicides involving prisoners who had not been identified as mentally ill; 'MHM’s corporate office had repeatedly informed her in contract-compliance reports that requests for mental-health services were not being processed appropriately according to their urgency level. In sum, the circumstances make clear that she had been exposed to information concerning the problems and thus ‘must have known’ about them. Farmer, 511 U.S. at 842-43, 114 S.Ct. 1970.

Deficiencies in treatment planning have been longstanding, persistent, and well documented, including in reports directly delivered to Associate Commissioner Nag-lich. MHM notified ADOC of the lack of individualization of treatment plans for years in audits and quarterly CQI meetings. MHM’s annual contract-compliance reports to ADOC between 2011 and 2016 also noted that treatment plans were inadequate across all levels of care, from outpatient to crisis care. ADOC’s own 2013 audit of Donaldson identified as a problem that treatment team meetings — where treatment planning occurs — frequently were held without all necessary participants.

The problem of insufficient counseling services has also been longstanding and well known. First, Naglich admittéd her knowledge of a persistent shortage of counselors and increasing caseloads, as well as a chronic shortage of correctional officers for escorting prisoners to appointments. Second, multiple sources informed her and other ADOC officials of serious problems in the provision of group counseling services; she also admitted that the shortage of correctional officers hindered MHM’s ability to provide group therapy sessions. Contract-compliance reports given to Naglich repeatedly informed her that multiple facilities were not getting enough group counseling sessions over the years. MHM’s monthly operations reports to ADOC, which contain statistics on the number of individual treatment encounters and group sessions each month, also made clear that little group counseling was occurring at multiple prisons. For example, the monthly operations report for April 2016 showed that no outpatient group therapy was offered at Donaldson, Easter-ling, Kilby, or St. Clair. Moreover, MHM has repeatedly discussed the problem of increasing caseloads for counselors and the unavailability of group treatment at many facilities during- quarterly CQI meetings, which ADOC Chief Psychologist Tytell attends on behalf of the agency.

ADOC officials have also been aware of the array of well-documented problems plaguing inpatient-level care. MHM has been reporting low utilization rates for RTU and SU beds - to Naglich and her office every month; Naglich admitted that she has been aware of the presence of prisoners in segregation without any mental-health needs in mental-health units, and that this disrupts the therapeutic environment; ADOC’s audit of Donaldson revealed that patients were not getting sufficient out-of-cell time and counseling; and Naglich has known that ADOC does not provide hospital-level care to patients who need it.

ADOC officials have been well aware of the -inadequacies in suicide prevention and crisis- care. Commissioner Dunn personally reviews suicide-incident reports and has been aware of the precipitous increase in the suicide rate in the last two years; he has been also aware that most of the suicides were committed by hanging and in segregation. For her part, Naglich has known even of the specific, system-wide conditions that create substantial risks of suicide: she was notified of the chronic crisis-cell shortage⁸⁰ and the backlog at the Bullock SU that has been driving the shortage; MHM complained to her about unsafe crisis cells with tie-off points and low visibility, and her ' office’s own audit included the same findings; and MHM repeatedly reported to Naglich that sharp items were found in crisis cells. Naglich also admitted that not having a constant-watch procedure for the most acutely suicidal inmates is a serious problem that poses a risk of harm in such a way that “someone could die.” Naglich Testimony at vol. 3, 228.

Perhaps most dramatically, ADOC has been aware of the actual harm and the substantial risk of serious harm that ADOC’s segregation practices pose to mentally ill prisoners. Commissioner Dunn has been aware of the fact that mentally ill prisoners resided in segregation, and that segregation could exacerbate their mental illness. Naglich has been receiving monthly reports that showed overrepresentation of mentally ill prisoners in segregation. MHM staff repeatedly communicated to ADOC officials — both orally and in writing — their concern about ADOC’s placement of mentally ill prisoners in segregation. MHM’s annual contract-compliance reports between.2012 and 2016 reported that multiple facilities had disproportionate numbers of prisoners on the mental-health caseload in segregation and recommended further review of the mental-health consultation process and monitoring. See PI. Ex. 1191, 2012 Contract-Compliance Report (doc. no. 1070-9); PL Ex. 114, 2013 Contract-Compliance Report (doc. no. 1070-4); PI. Ex. 115,2016 Contract-Compliance Report (doc. no. 1070-5). Moreover, MHM leadership has communicated the grave and potentially lethal risks of such segregation practices to ADOC officials, including Naglich. For example, over the last four to five years on multiple occasions, Dr. Hunter, MHM’s medical director, has had discussions with ADOC leadership regarding mentally ill prisoners’ potential to deteriorate while in segregation. MHM’s program director Houser has repeatedly informed ADOC officials that placement of mentally ill prisoners in segregation should be avoided because of the potential harm to those prisoners. Naglich herself admitted that housing mentally ill prisoners in segregation is “categorically- inappropriate.” Naglich Testimony at vol. 5,73.

While aware of the substantial risk of serious harm posed by segregation, ADOC has also known that certain ADOC disciplinary practices result in frequent placement of mentally, ill prisoners in segregation. Associate Commissioner Naglich admitted that ADOC has had a practice of disciplining prisoners for engaging in self-injurious behaviors. Furthermore, both MHM’s and ADOC’s own audits revealed that the mental-health consultation component of the disciplinary process was not properly functioning to keep mentally ill prisoners out of segregation.

Lastly, ADOC has also been aware of the inadequate monitoring and access to treatment for prisoners in segregation. ADOC’s chief psychologist Tytell informed Naglich that segregation rounds by mental-health staff were not being done properly and - that mental-health patients in segregation were not receiving treatment. Furthermore, ADOC officials have been well aware that segregation placement has been a common factor among suicides. Indeed, the great danger to mentally ill prisoners in segregation is obvious: prisoners are locked away for weeks at a time in cells with little monitoring and easy access to the means to kill themselves. In other words, ADOC has been aware of the actual harm that has resulted from segregation practices, in addition to the substantial risk of serious harm that ADOC’s segregation practices have imposed on mentally ill prisoners.

In sum, evidence established that ADOC has been aware of the gross deficiencies found in its treatment of mentally ill prisoners.

2. ADOC’s Disregard of Harm and Risk of Harm

Despite its knowledge ' of actual harm and substantial risks of serious harm to mentally ill prisoners, ADOC has failed to respond reasonably to identified issues in the delivery of mental-health care. On a global level, the state of the mental-health care system is itself evidence of ADOC’s disregard of harm and risk of harm: in spite of countless reports, emails, and internal documents putting ADOC on notice of the actual harm and substantial risks of serious harm posed by the identified inadequacies in mental-health care, those inadequacies have persisted for years and years. Suicide risk-assessment tools are still not being used outside of intake; referral requests are still not being triaged according to them urgency levels; records from late 2016 indicated a continued lack of individualized treatment plans and inadequate frequency of individual and group counseling; segregation prisoners without mental-health needs are still found in mental-health units; no hospitalization option or hospital-level care for the most severely ill exists; suicidal prisoners continue to be housed in unsafe cells without adequate monitoring; and mentally ill prisoners still are placed in segregation without a meaningful mental-health consultation process and have even less access to treatment.⁸¹

In addition to its failure to respond reasonably to these deficiencies, ADOC’s disregard for the substantial risk of serious harm to mentally ill prisoners manifested itself in two additional ways: its persistent refusal to exercise any meaningful oversight of MHM’s delivery of care; and its unreasonable responses to critical incidents and discrete issues brought to their attention over the years.

a. ADOC’s Failure to Exercise Oversight of the Provision of Mental-Health Care

ADOC’s Office of Health Services, run by Associate Commissioner Naglich, has done vanishingly little to exercise oversight of the provision of care to mentally ill prisoners. This failure exemplifies ADOC’s disregard of the substantial risk of serious harm to mentally ill prisoners within ADOC. Two facts provide important context for understanding this failure: first, ADOC has been well aware of the inadequacies in the treatment of mentally ill prisoners discussed above; second, as explained in this section, ADOC has known that MHM’s own quality-control process is hopelessly inadequate in implementing corrective actions. Despite clear indications that the same inadequacies persisted year after year, and that its contractor has been failing to implement corrective actions, ADOC chose to exercise close to no oversight, abdicating its constitutional obligation to ensure that the provision of mental-health care is minimally adequate.⁸² Such inaction is clearly unreasonable and therefore amounts to deliberate indifference.

As an initial matter, a brief overview of MHM’s quality-control processes illustrates the unreasonable nature of ADOC’s response. Though designed for ‘continuous quality improvement,’ MHM’s quality-control processes do not ensure that the identified deficiencies are corrected, mainly because many of the necessary corrective actions require cooperation and action by ADOC. MHM’s corporate office’s annual contract-compliance audit is the only system-wide review of MHM’s performance that either MHM or ADOC conducts. Once the review is complete, MHM sends a contract-compliance report, as well as a corrective-action plan, to Naglich’s Office of Health Services.⁸³ However, it is unclear whether anyone within MHM monitors the implementation of corrective actions. Moreover, for many of the identified deficiencies, MHM cannot address them effectively without ADOC’s help: corrective actions — such as obtaining adequate staff to facilitate therapy appointments and group activities — often require action by ADOC officers and, crucially, more staffing.⁸⁴ As a result, without action on ADOC’s part, contract-compliance reports often note the same problems recurring year after year: for example, multiple annual reports found that treatment plans were not updated consistently; that crisis cells in various facilities were unsafe, for suicidal prisoners; and that prisoners in segregation and mental-health units were not getting regular treatment due to the shortage of correctional officers.⁸⁵

The regional-level quality-improveinent exercises — which includes quarterly audits and ‘spot audits’ by MHM’s CQI-manager — also do not seem to result in corrective actions. In fact, the CQI manager admitted that no one is responsible for ensuring that site administrators address the issues identified through her spot checks: she is only responsible for reporting the findings, not addressing the problems; no documentation of site-level follow-up is required. In her own words, “the buck doesn’t stop with anyone.” Davis-Walker Testimony at vol. 2,152.⁸⁶

'Naglich was well aware of MHM’s' inability to address identified problems: In fact, Naglich blamed MHM for most of the deficiencies in mental-health care at ADOC and expressed particular dissatisfaction with MHM’s CQI process: she complained in court that MHM identifies problems, but does not help ADOC solve those problems. But this was the proverbial pot calling the kettle black:, in spite of her concerns about MHM’s internal oversight and her knowledge of deficiencies in.care, Nag-lich and OHS — the only ADOC department with responsibility for monitoring mental-health care — have done almost nothing that resembles ‘quality-improvement’ or even bare-bones contract monitoring in response.

First, Associate Commissioner Naglich admitted that she does not review. the contract-compliance reports in full or take actions based on their findings. She asserted that Dr. Tytell, the only staff member at OHS with mental-health expertise, is responsible for reviewing the reports. However,. Tytell denied ever receiving the reports or being responsible for reviewing them.⁸⁷ Not surprisingly, Naglich’s testimony also revealed that neither she nor anyone else in her office has taken any corrective measures in response to the numerous inadequacies identified in the reports. '

Second, ADOC has failed to monitor MHM’s provision of mental-health care, despite' having the tools to do so. ADOC’s contract with MHM grants it access to MHM’s files and the right to conduct scheduled and unannounced performance reviews. The contract also authorizes ADOC to assess fines for noncompliance found during formal audits. However, ADOC has not made use of these provisions. Since 2011, Naglich’s office has conducted only one informal audit — in response to a specific concern raised by a medical provider about a mentally ill inmate — and one ‘pilot audit,’ both of which were limited to the Donaldson facility.⁸⁸ (Only the first, informal audit produced a written report.) Despite Naglich’s own assessment that MHM was “not measuring up,” Naglich Testimony at vol. 4, 121, ADOC has not audited, even informally, mental-health care at any prison other than Donaldson, and has not conducted formal audits at any prisons. Because it has not conducted any formal audits, ADOC has not been able to assess MHM any fines for contractual noncompliance.

Even when ADOC conducted the informal audits at Donaldson, it did nothing to address the identified problems. The 2013 informal audit of Donaldson revealed that the care provided at the Donaldson RTU was deficient in many ways — so much so that Associate Commissioner Naglich described it as a “failed audit.” Naglich Testimony at vol. 2, 55. As discussed earlier in more detail in Section V.B.5, the audit revealed that providers had difficulties accessing patients because of the correctional staffing shortage; group programming was inadequate; bed space in the treatment units was used to house segregation inmates; mental-health staffing was inadequate; security for mental-health staff was inadequate; and patients were not getting sufficient out-of-cell time. MHM’s corrective-action plan identified tasks for both ADOC and MHM. However, Naglich was unable to identify a single follow-up action taken by her office or MHM to address any of these issues. ADOC’s lead auditor, Brendan Kinard, admitted that OHS did not do anything to resolve problems identified in the Donaldson audit.

Associate Commissioner Naglich offered no reasonable explanation when pressed about the reason for the lack of follow-up after the dismal results of the 2013 Donaldson audit. She blamed the death of Dr. Cavanaugh, the chief psychologist of OHS and Dr. Tytell’s predecessor, who unexpectedly passed away in March 2014. According to Naglich, Cavanaugh had been responsible for contract monitoring, including conducting formal and informal audits of MHM’s delivery of care, and for ensuring that the quality of mental-health care is adequate. However, Naglich’s excuse did not hold water: when asked to produce any documentation of audits or follow-ups done by Dr. Cavanaugh before he passed away, she was unable to do so; he apparently produced no written reports or emails about his findings or audits. According to MHM’s program director Houser, Cavanaugh conducted no system-wide or even facility-wide audits; he simply performed ‘reviews’ that did not result in corrective-action plans or written reports.

Moreover, the testimony of Dr. Tytell, who took Dr. Cavanaugh’s place later in 2014, made clear that ADOC still does little to ensure that MHM is meeting contractual requirements. Tytell admitted that he does not conduct any system-wide or facility-wide audits, and that he only examines patient records when he is trying to learn something about a specific patient. He attends MHM’s quarterly CQI meetings, which last a whole day, but he leaves around lunch time; he has missed one or two of the four quarterly meetings in the last year. He does not look into issues raised at CQI meetings, unless specifically told to do so by Naglich.⁸⁹ Likewise, although he receives programming logs and monthly operations reports from MHM, Tytell does nothing with them because the information is “already old data” that is “a couple months behind.” Tytell Testimony at_

The 2015 Donaldson ‘pilot audit’ also exemplified ADOC’s inadequate response to identified problems in the provision of mental-health care. While conducting the audit, Tytell became concerned that many of the medical records he was examining were not meeting the benchmarks, and called Associate Commissioner Naglich in the middle of the audit to report the “dismal” results. Tytell Testimony at_ Naglich simply told him to finish the audit. On the last day of the audit, Tytell informally discussed his preliminary findings in an exit interview with the site administrator and two people from MHM’s regional office. However, no one at OHS formally communicated with MHM regarding the problems found in the audit or gave written feedback, even though many of the same inadequacies from the 2013 audit were identified again, and Houser specifically asked for feedback. Because there was no written report, MHM did not develop any corrective-action plans.

After he revised the audit tools based on the ‘pilot audit’ results, Tytell asked Nag-lich whether OHS should re-audit Donaldson using the new tools. Naglich told him to not worry about it. Naglich also told him not to conduct any more audits of any other facilities. Tytell disagreed with the decision not to re-audit but did as he was told: as he explained, he “learned to stay in [his] lane,” that is, “to do as I am ordered.” Tytell Testimony at_OHS has not conducted any audit using the revised audit tools since then.

In sum, in failing to exercise adequate oversight of MHM’s performance and to address deficiencies identified in the “failed” results of the 2013 Donaldson audit and the “dismal results” of the 2015 Donaldson audit, ADOC’s response to its knowledge of harm and risk of harm in the mental-health care system has been objectively unreasonable,

b. ADOC’s Unreasonable Responses to Identified Deficiencies

ADOC has also failed to respond reasonably to discrete issues that come to its attention, even when lives may be at stake. In response to many of the deficiencies identified above, ADOC officials admitted to doing nothing in response to being informed. ADOC officials also repeatedly testified that they simply told someone else about the risk of harm being created by deficient treatment of mentally ill prisoners, and took no other action, even though informing someone else within ADOC previously had failed to result in any change. Insisting upon a course of action that has already proven futile is not an objectively reasonable response under the deliberate-indifference standard.

Examples of such unreasonable responses abound. First, as multiple ADOC and MHM staff admitted, sharp items in crisis cells have been a recurring problem in multiple facilities. When Dr. Tytell was asked about this problem, he simply stated: “I’m always told that things will be taken care of and things will be done. How to check'up on it and follow up on it, I don’t know how unless I’m told that it happens again.” Tytell Testimony at_ Tytell’s statement epitomized ADOC’s inadequate response to problems that pose serious risks to prisoners: the sole ADOC official with mental-health expertise insists on passing the buck even when the issue involves' self-harm by suicidal prisoners, and even when his past experience • has clearly shown him that simply bringing problems to the attention of others does not fix those problems.

Associate Commissioner Naglich likewise shirked responsibility when asked about the issue of sharp items found in crisis cells: even though she knew that correctional officers were not following protocol by failing to search crisis cells for sharp items that could be used for self-harm, she maintained that she does not have the authority to tell correctional officers to follow the protocols “because it’s a security concern, so all that we can do is relay that concern to security.” Naglich Testimony at vol. 4, 115. However, Associate Commissioner Culliver credibly testified that as an associate commissioner herself, Naglich has the authority to tell correctional officers to comply with administrative regulations and protocols.

Associate Commissioner Naglich’s testimony was also full of admissions that, despite knowledge of risks' of harm, ADOC took no action at all. She admitted that shé had known about problems regarding visibility into crisis cells at least since ADOC’s 2013 audit of Donaldson, but she did not know what, if anything, had been done to correct these problems in the years since. When- asked why she has not done anything personally to- address this issue that she acknowledged as “critical,” she stated that she does not have enough staff to do so. Naglich Testimony at vol. 1, 173-74. Naglich also admitted that ADOC officials did nothing in response to their own audit finding that ADOC had a practice of automatically applying disciplinary sanctions for self-injury.-Likewise, she took no action in response-, to MHM’s repeatedly-expressed concern — which she shared— that mentally ill prisoners are overrepresented in segregation, until after she told the court that mentally ill prisoners should not be in segregation. She unconvincingly testified that if she had been notified that the mentally ill were disproportionately being housed in segregation, shp “would have looked at each one- of those facilities.” Naglich Testimony at vol. 5, 138. Naglich, in fact,- had been informed for years that mentally ill prisoners have been overrepresented in segregation.⁹⁰ Yet, she admitted that she never inquired into the facilities reported to have disproportionate numbers of mentally ill prisoners in segregation.

ADOC’s response to the skyrocketing suicidb rate also demonstrates a frankly shocking level of disregard for a known substantial risk of serious harm. At the highest level, Commissioner Dunn testified that he personally tracks suicide rates and has looked at incident reports; he is, of course, aware of the sharp increase in suicide rates in' the last two yeaps within ADOC. However, more than a month after this trial began, hé testified that he has not ordered his staff to take any concrete measures other than asking his chief of staff, Steve Brown, to “look into it.” Dunn Testimony at vol. 1,- 45. He has never attended any meetings regarding suicides, or asked for a written report or follow-up after suicide-related meetings that took place in October 2015 and, October 2016.

Associate Commissioner Naglich was not only aware of the increase in the suicide rate, but «Iso the risk factors for suicides. Yet, she and other ADOC officials made almost no effort to address the problem. In Naglich’s view, suicide is a risk for anyone with untreated mental illness — in other words, a lack of treatment, as well as a lack of acute care and suicide-prevention measures, places all mentally ill prisoners at risk of the most serious bodily harm possible. She attended an October 2015 meeting focused on the increase in suicide rates, where she learned that segregation placement was a common factor among suicides. However, neither she nor anyone else at ADOC took any action to change ADOC’s housing of mentally ill prisoners in segregation, and no follow-up meeting was scheduled until October 2016. During that 12-month period, six prisoners committed suicide, doubling the annual rate from 2015. After the second meeting, ADOC again took no action. Appallingly, ADOC officials directly responsible for mental health — Naglich and Tytell — and prisoner placement — Culliver—all admitted that they were aware of the sharp rise in suicides, participated in these two meetings on the suicide rate, and took no action.

Associate . Commissioner Naglich attempted to explain this lack of response by stating that a new mental-health coding system prohibiting placement of seriously mentally ill prisoners in segregation was in the middle of a roll-out at the time of her testimony in December 2016. However, as explained earlier, her representation was disputed by the testimony of two • of her colleagues, who explained that OHS moved ten mentally ill prisoners out of segregation into the Donaldson RTU only after her testimony, and that there was no official policy change.⁹¹ In a way, Naglich’s belated transfer of the prisoners is all the more damning: the fact that she and Dr. Tytell could move ten RTU-level prisoners out of segregation and into the RTU over the course of a few weeks suggests that it was well within their ability to prevent seriously mentally ill prisoners from being housed in segregation. They could have taken this action in 2015, after the first meeting on suicides, or in 2016, after the second meeting, rather than waiting until January 2017. By that time, twelve more people, including a plaintiff in this lawsuit, had committed suicide.

D. Ongoing Violation

Before granting injunctive relief against a state official for an Eighth Amendment violation, the court must find that the violation is ongoing-and-continuous in order to fall under the Ex parte Young exception of the Eleventh Amendment bar. 209 U.S. 123, 28 S.Ct. 441, 52 L.Ed. 714 (1908); see Idaho v. Coeur d’Alene Tribe of Idaho, 521 U.S. 261, 281, 117 S.Ct. 2028, 138 L.Ed.2d 438 (1997) (“An allegation of an ongoing violation of federal law where the requested relief is prospective is ordinarily sufficient to invoke the Young fiction.”). In interpreting this requirement, the Eleventh Circuit has held that “the ongoing and continuous requirement merely distinguishes between cases where the relief sought is prospective in nature, i.e., designed to prevent injury that will occur in the future, and cases where relief is retrospective.” Summit Med. Assocs., P.C. v. Pryor, 180 F.3d 1326, 1338 (11th Cir. 1999). In. this case, plaintiffs are seeking prospective injunctive relief to remedy serious inadequacies in the mental-health care system that will continue to put mentally ill prisoners at a substantial risk of serious harm if not corrected. However, during the trial, defendants suggested that in three different areas of mental-health care at issue here, ADOC has started remedying the inadequacies, rendering plaintiffs’ .claims as to those areas moot and not suitable for resolution by the court. The court disagrees, and addresses each area in turn.⁹²

First, during the trial, defendants repeatedly argued that the 2014 partial settlement between the parties regarding the distribution of razor blades to prisoners in crisis cells and segregation units has rendered the issue of dangerous items in crisis cells moot. However, the settlement deals solely with the policy of distributing razor blades for shaving to prisoners in those units, rather than the distinct issue of keeping dangerous items — including but not limited to razor blades — from being introduced into crisis cells by other means. Multiple employees of ADOC and MHM testified that the presence of dangerous items in crisis cells has been an ongoing problem. Accordingly, the 2014 settlement of the razor-distribution issue does not moot this inquiry or prevent the court from finding an ongoing .violation.

Second, defendants argued that the January 2017 interim agreement ‘revamping1suicide • prevention 1 protocols moots the issue of suicide prevention and crisis care in general. However, as discussed earlier, suicide prevention encompasses much more than requiring constant watch for the most acutely- suicidal prisoners and ensuring staggered-interval checks of others. In fact, various suicide prevention measures discussed in this case are not covered by the interim agreement, and defendants have not implemented them, despite their knowledge of the risk of harm posed by the current conditions. For example, as this court saw firsthand during its facility tours after the- trial, segregation cells and some crisis-cells continue to have easily accessible tie-off. points, despite the fact that most suicides happen in segregation cells.⁹³ Likewise, despite the recommendation of defendants’ own expert that a suicide risk-assessment tool be used for all prisoners at a heightened risk of suicide, not just prisoners coming through the intake process for the first time, ADOC has failed to assess prisoners for suicide risks outside of the intake process.⁹⁴ When asked by the court why this part of Dr. Patterson’s recommendation is not being followed, Associate Commissioner Naglich answered that she is “not sure where all it’s being used” and “it would be a question better asked of MHM.”⁹⁵ Nag-lich Testimony at vol. 3,231.

Furthermore, evidence suggests that even the limited remedial actions covered by the interim agreement have not been fully implemented. Allegations of noncompliance with the constant-watch procedure resulted in a modification of the interim agreement in order to allow plaintiffs’ counsel frequent monitoring visits to crisis cells. The court also witnessed - firsthand during the post-trial site visits that essential parts of suicide-watch procedures were still not being followed: many forms for 15-minute and 30-minute staggered-interval checks of prisoners on suicide watch and mental-health observation were pre-filled and at exact intervals. ADOC’s inability to carry out the terms of the interim agreement even in anticipation of this court’s announced visit illustrates a severe, ongoing dysfunction in the system, a striking indifference by ADOC to a substantial risk of serious harm, or both. Needless to say, the court finds that the inadequacies in ADOC’s suicide-prevention measures are ongoing.

Partly for this reason, the court declines to rely on Commissioner Dunn’s testimony that he intends to abide by the interim agreement’s constant-watch procedures until an expert or the court tells him otherwise. Dunn’s statement regarding his intent to enforce it indefinitely is not reliable given the evidence of noncompliance already shown and Houser’s testimony that the budget and the layout of crisis cells make constant watch unsustainable. In addition, Dunn’s statement of intention is not enforceable in court, especially given that the order approving the interim agreement specifically states that it does not resolve any of the issues raised in trial. See Interim Agreement on Suicide Prevention Measures (doc. no. 1102). In other words, defendants have not satisfied the requirements for making a claim moot by voluntary cessation: the Commissioner’s statement cannot be said to have “completely and irrevocably eradicated the effects of the alleged violation,” and there is a reasonable expectation that the alleged violation may recur, due to the risk and evidence of non-compliance and the unen-forceability of the defendant’s statement in court. See Reich v. Occupational Safety and Health Review Comm’n, 102 F.3d 1200, 1202 (11th Cir. 1997) (holding that a request for injunctive relief may become moot if: (1) “it can be said with assurance that there is no reasonable expectation that the alleged violation will recur and (2) interim relief or events have completely and irrevocably eradicated the effects of the alleged violation.”).

Third, defendants have also failed to show that the inquiry into their segregation practices has become moot or that they have stopped placing seriously mentally ill prisoners in segregation.⁹⁶ As discussed above, evidence suggests that the new coding system as described by Associate Commissioner Naglich has not yet been implemented. ADOC’s failure, to address such an obvious risk of harm despite their knowledge of the issue for over two years vividly illustrates that the violation is ongoing and will continue if the defendants are left to their own devices.

E. Ex parte Young Defenses

Defendants advance two arguments regarding the Ex parte Young doctrine: first, that the defendants, sued in their official capacities, lack the authority to implement the remedy, and therefore cannot be proper defendants; second, that the remedy would require the State to expend money, and therefore is barred by the Eleventh Amendment — an argument that this court already rejected in the summary judgment opinion. Neither argument is viable under the Eleventh Amendment case law and Ex parte Young, 209 U.S. 123, 28 S.Ct. 441, 52 L.Ed. 714 (1908).

The case law does not support the argument that the Commissioner and the Associate Commissioner were not the proper defendants to sue due to their alleged lack of authority to implement the remedy. The Supreme Court rejected this line of argument in Papasan v. Allain, 478 U.S. 265, 106 S.Ct. 2932, 92 L.Ed.2d 209 (1986), where the State of Mississippi contended that plaintiffs had not sued officials who could grant the relief requested, which was to remedy the State’s unequal distribution of the benefits from the State’s school land. The Court' held that one of the named defendants, the Secretary of State, was a proper defendant because he was responsible under a state statute for “general supervision” of the local school officials’ administration of the lands in question; because of those responsibilities, he could be properly enjoined under Ex parte Young. Id. at 282 & n.14, 106 S.Ct. 2932. The Court’s holding ensured that if &■ state official violates the Constitution while carrying out a responsibility created by virtue of the defendant’s office, that defendant may be enjoined under Ex parte Young. See also Ex parte Young, 209 U.S. at 157, 28 S.Ct. 441 (explaining that “the fact that the state officer, by virtue of his office, has some connection with the enforcement of the act, is the important and material fact”). The defendant’s authority to implement the remedy was not relevant to the Ex Parte Young analysis.

This circuit has repeatedly held that defendants simply must have “‘have some connection with’ the unconstitutional act or conduct complained of’ in order to be proper defendants for an injunctive-relief suit under Ex parte Young. Luckey v. Harris, 860 F.2d 1012, 1015-16 (11th Cir. 1988) (quoting and citing Ex parte Young, 209 U.S. at 157, 28 S.Ct. 441 (internal alterations omitted)). For "example, in Grizzle v. Kemp, 634 F.3d 1314 (11th Cir. 2011), the Eleventh Circuit held that the Secretary of State is a proper defendant in a suit challenging the legality of a state election law — even though that official cannot implement the relief of changing the law — since he has “both the power and the duty to ensure that [local boards of elections] comply with Georgia’s election code,” which “ ‘sufficiently connects] him with the duty of. enforcement”’ for the potentially unconstitutional law. Id. at 1319 (quoting Ex Parte Young, 209 U.S. at 161, 28 S.Ct. 441). Conversely, in Summit Med. Assocs., PC v. Pryor, 180 F.3d 1326 (11th Cir. 1999), the court found that a state prosecutor is not a proper defendant in a lawsuit challenging a -private civil-enforcement statute creating, a private cause, of action, because a prosecutor has .-no connection with the enforcement of a civil statute that enables an affected private individual to sue. The application of Ex parte Young in the Eleventh Circuit as well .as other circuits is palpably distinct from the defendants’ formulation, which elides the distinction between having “ ‘some connection’ ... with the conduct complained of,” Luckey, 860 F.2d at 1015-16 (quoting Ex Parte Young, 209 U.S. at 157, 28 S.Ct. 441), and the “authority to remedy the alleged wrongs.” Defs.’ Ex parte Young Trial Br. (doc. no. 1098) at 12,

Applying the proper formulation of Ex parte Young, the Commissioner and the Associate Commissioner have the constitutional duty to provide minimally adequate mental-health care as the officials responsible for running the Alabama Department of Corrections • and its Office of Health Services;. therefore, they have “the ability to commit the unconstitutional act” of failing to provide minimally adequate mental-health care, and the Ex parte Young doctrine applies. Okpalobi v. Foster, 244 F.3d 405, 421 (5th Cir. 2001) (en banc).

Defendants also seem to argue that any time a state official requires someone else’s cooperation in order to remedy a constitutional violation, that state official’s unconstitutional act is immune from suit. This cannot be. The Ex parte Young case law is replete with examples where a court finds the conduct of a state agency unconstitutional, even when the named defendants in their official capacities cannot remedy the violation alone. For example, the Eleventh Circuit found that the Ex parte Young-doctrine applied to a'lawsuit challenging the adequacy of counsel provided to indigent criminal defendants in the State of Georgia in Luckey v. Harris, 860 F.2d 1012 (11th Cir. 1988), quoting the “some connection” language from Ex parte Young. Remedying some of the allegations in that case, such as inadequate supervision ofi court-appointed criminal defense counsel,'would have required third parties’ cooperation, including hiring new personnel'to supervise defense attorneys and related budget appropriations — -just as potential remedies proposed by the parties in this case might require an additional budget appropriation and the recruitment of new -personnel. In other words, the fact that the named defendants in their official capacities may need third parties’ cooperation to cany out some of the potential remedies does not bar Ex parte Young’s applicability, because the doctrine only requires “some connection” between the alleged'wro'ngdoing and the officials’ responsibility.

Defendants’ second argument — that the remedy would require state expenditures in violation'of the Eleventh Amendment — is equally unavailing. Defendants argue that because the Commissioner and the Associate Commissioner do not have the authority to appropriate more money to their own budget, they are immune from this lawsuit under the Eleventh Amendment. The Supreme Court has clearly held that the Eleventh Amendment does not bar an order requiring expenditure of state funds if it is ancillary to injunctive relief for an ongoing violation. In Edelman v. Jordan, 415 U.S. 651, 668, 94 S.Ct. 1347, 39 L.Ed.2d 662 (1974), the Supreme Court noted that the Eleventh Amendment did not bar suits that had “fiscal consequences to state treasuries” that “were the necessary result of compliance with decrees which by their terms were prospective in nature.” The Court in Edelman also observed that having to spend more money from the state treasury because the State needs to conform its conduct to the court order is an “ancillary effect” that is “a permissible and often an inevitable consequence of the principle announced in Ex 'parte Young.” Id. at 668, 94 S.Ct, 1347. The Court reiterated this principle in Milliken v. Bradley, 433 U.S. 267, 289, 97 S.Ct. 2749, 53 L.Ed.2d 745 (1977), upholding a district court’s order requiring the State defendants to pay one-half of the additional costs attributable to a remedial education scheme to support school desegregation. In both of these cases, the Supreme Court recognized that the State must pay for ancillary costs of prospective, injunctive relief, regardless of whether the named defendants in their official capacities — who were standing in for the State based on the Ex parte Young fiction — had the ability to appropriate more money to their own budget. See also Lane v. Cent. Ala. Cmty. Coll., 772 F.3d 1349, 1351 (11th Cir. 2014) (“The Supreme Court has recognized that compliance with the terms of prospective injunctive relief will often necessitate the expenditure of state funds.”) (citing Edelman v. Jordan, 415 U.S. 651, 94 S.Ct. 1347, 39 L.Ed.2d 662 (1974)). In fact, rather than precluding relief, courts have found inadequate funding to be a basis for finding of deliberate indifference. See, e.g., Wellman v. Faulkner, 715 F.2d 269, 273 (7th Cir. 1983).

In sum, defendants are not immunized from liability arising from ongoing constitutional violations simply because they lack financial resources or the authority to mandate certain specific measures that might remedy the violation. On the contrary, the Ex parte Young doctrine allows this court to find liability and ensure that the prison system provides minimally adequate mental-health care.

VI. CONCLUSION

For the reasons above, the court holds that the Commissioner of the Alabama Department of Corrections and the Associate Commissioner of Health Services, in their official capacities, are violating the Eighth Amendment rights of the plaintiff class and of plaintiff Alabama Disabilities Advocacy Program’s constituents with serious mental-health needs who are in ADOC custody. Simply put, ADOC’s mental-health care is horrendously inadequate. Based on the abundant evidence presented in support of the Eighth Amendment claim, the court summarizes its factual findings in the following roadmap, identifying the contributing factors to the inadequacies found in ADOC’s mental-health care system:

(1) Failing to identify prisoners with serious mental-health needs and to classify their needs properly;

(2) Failing to provide individualized treatment plans to prisoners with serious mental-health needs;

(3) Failing to provide psychotherapy by qualified and properly supervised mental-health staff and with adequate frequency and sound confidentiality;

(4) Providing insufficient out-of-cell time and treatment to those who need residential treatment; and failing to provide hospital-level care to those who need it;

(5) Failing to identify suicide risks adequately and providing inadequate treatment and monitoring to those who are suicidal, engaging in self-harm, or otherwise undergoing a mental-health crisis;

(6) Imposing disciplinary sanctions on mentally ill prisoners for symptoms of their mental illness, and imposing disciplinary sanctions without regard for the impact of sanctions on prisoners’ mental health;

(7) Placing seriously mentally ill prisoners in segregation without extenuating circumstances and for prolonged periods of time;⁹⁷ placing prisoners with serious mental-health needs in segregation without adequate consideration of the impact of segregation on mental health; and providing inadequate treatment and monitoring in segregation.

The court further finds that persistent and severe shortages of mental-health staff and correctional staff, combined with chronic and significant overcrowding, are the overarching issues that permeate each of the above-identified contributing factors of inadequate mental-health care.

[[Image here]]

Accordingly, it is ORDERED that the court and the parties will meet to discuss a remedy. The court emphasizes that given the severity and urgency of the need for mental-health care explained in this opinion, the proposed relief must be both immediate and long term. No partial final judgment shall issue at this time as to the claim resolved in this entry.

DONE, this the 27th day of June, 2017.

1 . Earlier in the litigation, the parties also reached a settlement regarding the distribution of razor blades to mentally ill prisoners.

2 . The defendants did not raise or re-argue exhaustion of administrative remedies during or after the trial, and did not argue exhaustion in their post-trial filings as a reason they should prevail. See Defendants' Post-Trial Brief (doc. no. 1282); see also Dunn v. Dunn, 219 F.Supp.3d 1100 (M.D. Ala. 2016).

3 . The provision of mental-health care to Alabama’s prisoners has been litigated at least three times before. See Laube v. Campbell, 333 F.Supp.2d 1234 (M.D. Ala. 2004) (Thompson, J.) (approving settlement agreement that provides for inpatient care, suicide prevention and treatment, crisis intervention, and counseling services in a class-action lawsuit brought on behalf of women incarcerated in Alabama); Bradley v. Harrelson, 151 F.R.D. 422 (M.D. Ala. 1993) (Albritton, J.) (certifying a class of severely mentally ill male prisoners); Pugh v. Locke, 406 F.Supp. 318 (M.D. Ala. 1976) (Johnson, J.) (ordering the State to provide minimally adequate mental-health care, including identification of mentally ill prisoners and provision of care by qualified mental-health professionals), aff'd and remanded sub nom. Newman v. Alabama, 559 F.2d 283 (5th Cir. 1977), cert. granad in part, judgment rev'd in part on other grounds, and remanded sub nom. Alabama v. Pugh, 438 U.S. 781, 98 S.Ct. 3057, 57 L.Ed.2d 1114 (1978).

4 . ADOC also houses an additional 4,500 prisoners in work centers and work-release centers, bringing the total population in custody to around 24,000.

5 . See Pl. Ex. 1260, September 2016 Monthly Statistical Report (doc. no. 1097-19).

6 . Decompensation refers to exacerbation of symptoms of mental illness and impaired mental functioning; it calls for a "more structured or sheltered setting for more intensive treatment interventions." Burns Testimony at vol. 1, 173.

7 . During their meeting, Wallace began to cry, leaned over the interview table, and told Dr. Haney, with tragic prescience, "[T]his place is killing me.” Joint Ex. 459, Haney Expert Report (doc, no. 1038-1043) at 40.

8 .In a separate order with an opinion to follow, the court finds that four of the experts' methodologies survive Daubert challenges, No objection was raised against plaintiffs’ expert Dr. Craig Haney.

9 . Based on his review of medical records and deposition testimony, Dr. Patterson also offered his opinions about whether individual plaintiffs’ care was adequate. However, because this is a case alleging systemic inadequacies in the delivery of mental-health care, the court need not determine the adequacy of care for any particular individual. Furthermore, because Dr. Patterson did not meet with any of the plaintiffs, and deposition transcripts, by Dr. Patterson’s own admission, are not a reliable source for determining credibility or making clinical diagnoses of an individual, the court gives little weight to his opinions as to whether the care provided to the individual plaintiffs was adequate.

10 . Earlier in the litigation, this court certified a class consisting of “persons with a serious mental-health disorder or illness who are now, or will in the future be, subject to defendants' mental-health care policies' and practices in ADOC facilities, excluding work-release centers and Tutwiler Prison for Women.” Braggs v. Dunn, 317 F.R.D. 634, 640 (M.D. Ala. 2016) (Thompson, J.).

11 . The concept of ‘serious mental-health need’ in the Eighth Amendment context . should not be confused with ‘serious mental illness,’ a term of art in the mental-health care field. As plaintiffs’ psychiatric expert Dr, Burns testified, ‘serious mental illness’ can be defined by three components: the diagnosis, the degree of disability, and the duration of the diagnosis or disability. Certain diagnoses are by definition serious mental illnesses, because they last a lifetime and are accompanied by debilitating symptoms; these diagnoses include bipolar disorder, schizophrenia, schizoaffective disorder,, major depressive disorder with psychotic features, and any other diagnoses with psychosis. Dr, Hunter, MHM’s medical director, agreed with this assessment, testifying that a person with well-controlled schizophrenia still has a serious mental illness, because it requires continued treatment, even if he or she is only mildly impaired at the moment. Other diagnoses, like anxiety and PTSD, may reflect a serious mental illness depending on the' degree and duration of the impairment. Dr. Burns testified that ADOC’s administrative definition of serious mental illness tracks this understanding of serious mental illness. See Joint Ex. 88, Admin. Reg. § 602 (doc. no. 1038-1039) at 11 (defining "serious mental illness” as "[a] substantial disorder of thought, mood, perception, orientation, or memory such as those that meet the DSM- IV criteria for Axis I disorders .... [and] persistent and disabling Axis II personality disorders.”). According to experts on both sides, treatment of serious mental illnesses requires, at a minimum, multidisciplinary efforts to coordinate and implement interventions, including psychotherapy or counseling, psychotropic medications, and monitoring for signs of decompensation or progress. It also requires careful treatment planning and maintaining medical records in order to ensure continuity of care.

12 . When the trial began, the court used full names of prisoner-witnesses, but the parties agreed to use initials after Jamie Wallace’s testimony.

13 . While courts have sometimes used the "serious need” and "substantial risk of serious harm” tests interchangeably, they appear to be somewhat distinct: the "serious need” requirement examines whether a prisoner has a medical problem requiring attention; the “substantial risk of serious harm” test examines whether the defendant’s inattention to or mistreatment of the medical need threatens serious harm to the prisoner. Of course, a plaintiff may face a serious medical need because defendants’ inattention has caused or exacerbated a medical condition, see, e.g., Helling v. McKinney, 509 U.S. 25, 113 S.Ct. 2475, 125 L.Ed.2d 22 (1993) (concluding that prisoner’s claim based on potential future effects of exposure to tobacco smoke could be a viable Eighth Amendment claim), but this does not change the fact that the focus of the "serious need” inquiry is the prisoner’s condition, while the "substantial risk of serious harm” inquiry focuses on the effects of inadequate health care.

14 . The 'stacked Swiss cheese' analogy, well known in the healthcare and risk-management contexts, may be useful here. In this analogy, a layer of Swiss cheese represents a mechanism to prevent harm, and an error is a hole in that layer. Ideally, each layer is sufficiently redundant to catch or ameliorate errors and to prevent holes from lining up. However, if each hole is too big, errors from each layer compound and result in an inadequate system. See James Reason, Human Error: Models and Management, 320 Brit. Med. J. 768 (2000). Applied to this context, each layer of mental-health care within ADOC— identification of symptoms at intake and referral; treatment planning; provision of psychotherapy; inpatient care; crisis care; and consideration of mental health in prisoner placement decisions — is riddled with too many holes to prevent mentally ill prisoners from falling through the cracks. Moreover, each layer's error is compounded by latent errors in inter-related layers of care: for example, delinquent counseling appointments fail to address a sudden deterioration in a prisoner's condition, which is worsened by the lack of a properly functioning referral system and a suicide-watch protocol.

15 .As discussed later, some of the policies and practices affecting mentally ill prisoners are determined by ADOC, others by MHM: for example, ADOC is responsible for staffing decisions and placement of prisoners in mental-health units and segregation; MHM is responsible for policies and practices in intake screening, the referral system, treatment planning, and psychotherapy. However, ADOC is still liable for policies and practices determined by MHM, for three reasons. First, ADOC’s decisions regarding mental-health staffing, correctional staffing, and overcrowding have directly impacted MHM’s policies and practices, such as frequently delayed and cancelled counseling sessions and the use of LPNs to conduct intake screening. Second, for some of the practices, ADOC has expressly authorized MHM to determine them on its behalf by contracting out its constitutional obligation to provide mental-health care. Third, even when ADOC has not expressly authorized MHM to make these policies — that is, when MHM’s policies and practices contravene ADOC’s administrative regulations or contractual requirements — ADOC through its lack of oversight has de facto delegated its decision-making authority to MHM. See Ancata v. Prison Health Servs., Inc., 769 F.2d 700, 706 n.11 (11th Cir. 1985) (holding that ”whe[n] a governmental entity delegates the final authority to make decisions," either expressly or by default, then "those decisions necessarily represent official policy” in the context of contracting out medical care for prisoners). Therefore, the court finds that ADOC is liable for the policies and practices described here, despite the fact that MHM is the entity providing mental-health care and determining some of the policies and practices related to mental-health care.

16 . Defendants advanced a few versions of the argument that variability across different facilities negates ADOC’s liability: defendants argued that experts visiting seven, eight, or nine facilities instead of visiting all 15 facilities renders their opinions irrelevant or not reliable; that certain facilities are not as overcrowded as others; and that plaintiffs did not prove that every single facility suffers from a shortage of crisis cells. As explained in the commonality and typicality analyses in the class certification opinion, Braggs v. Dunn, 317 F.R.D. 634, 655-66 (M.D. Ala. 2016), evidence of systemic practices that may have differing levels of impact at different facilities may establish liability against ADOC: mentally ill prisoners are subject to a substantial risk of serious harm from practices that are common in ADOC facilities no matter where they are housed currently, because they may be housed in any of these facilities in the future due to ADOC’s frequent and unpredictable transfers of prisoners across facilities.

17 . Parties have put forth evidence regarding the Alabama Prison Transformation Initiative, a proposal by the now-former Governor to build new prisons. At this point, the court does not see any need to determine the effects of the proposal, because the case at hand asks the court to evaluate whether the current state of mental-health care in existing ADOC facilities is constitutionally inadequate, rather than whether a hypothetical system of mental-health care in new prisons would be adequate.

18 . After years of refusing to increase staffing, ADOC approved a small staffing increase in September 2016, shortly before the trial in this case, when it extended the contract with MHM for another year. However, both Associate Commissioner Naglich and MHM’s program director Houser testified that understaffing has persisted despite the recent increase,

19 . Chronic mental-health understaffing is also compounded by vacancies that are left unfilled for many months.

20 . Examples of inadequate care caused by mental-health shortages include: lack of timely provision of counseling services; inadequate treatment planning; and inadequate monitoring of suicidal patients as well as those housed in mental-health units and segregation units.

21 , Dr. Burns also testified that the mental-health staffing requirements in a 2001 settlement agreement between ADOC and a class of male prisoners provide a helpful benchmark for adequate staffing levels. See Order Approving Settlement Agreement, Bradley v. Harrelson, No. 2:92-cv-70 (M.D. Ala. June 27, 2001) (Albritton, J.), ECF No. 412. Dr. Bums explained that while the number of ADOC prisoners in need of mental-health services has increased since the Bradley settlement, ADOC has entered into mental-health contracts that provide significantly fewer high-level practitioners, as well as more practitioners with lower levels of qualification, compared to the Bradley requirements. For example, under Bradley, ADOC was required to provide eight psychiatrists for approximately 20,600 prisoners; today, it employs five psychiatrists for close to 24,000 prisoners. While the staffing requirements derived from an out-of-court settlement do not set a constitutional floor for adequate mental-health care, the comparison with the Bradley settlement is relevant, though not dispositive, for determining whether the current staffing levels are adequate.

22 .Throughout the trial, there was confusion as to how ADOC defined 'authorized positions’ for the purpose of deriving shortage rates published in their annual reports. During the defendants’ case, ADOC’s chief of staff Steve Brown finally clarified that the number of authorized positions was determined based on a staffing ratio of 1:6 or 1:7, which were ratios that ADOC considered close enough to the "ideal” ratio of one correctional officer for every five inmates. However, as plaintiffs’ expert Eldon Vail and ADOC officials explained, adequate staffing numbers cannot be calculated by simply dividing the inmate population by the staffing ratio that is deemed to be ideal; rather, it requires a facility-by-facility determination that" considers numerous variables, such as the layout and design of the facilities, level of security, level of programs and activities, and state and local standards and statutes, Vail also explained that 1:5 is not an "ideal" ratio but likely' the average of staffing ratios from state correctional systems that responded to a survey conducted by the Association of State' Correctional Administrators. The ratios also do not take multiple shifts and leave time into account. Therefore, while ADOC relied on the authorized position numbers derived from such calculations in its annual reports, the shortage rates in those reports are not reliable indicators of under-staffing, except as a metric to measure change in staffing over time. However, as shown later, there is ample evidence, both from expert testimony and ADOC staff's testimony, that ADOC suffers from a serioüs correctional staffing shortage.

It is alarming that ADOC has not conducted any staffing analysis in the last decade to determine, exactly how many officers are needed to keep officers and prisoners safe within its facilities. It is, also alarming that ADOC's own reports have been relying on authorized-position numbers based on rudimentary ratios that do not take into consideration the actual layouts .of facilities. This failure to conduct any staffing analysis is all the more troubling because at least one ADOC official, Associate Commissioner Grantt Cul-liver, has the expertise to conduct staffing analyses and has been training other state correctional officials on hqw to conduct staffing analyses. Vail also' testified that it is not resource-intensive to obtain a staffing analysis from the National Institute of Corrections, since the Institute provides grants and other resources to state prison systems that host training for correctional officials in their own facilities, as ADOC has done.

23 . Vail explained that ADOC’s use of open dormitories in maximum-security facilities is almost unheard of in corrections,

24 . - The witness’s fear is well-warranted: St. Clair is the most violent facility in ADOC, accounting for a quarter of assaults with serious injuries within the system, while housing only 4 % of ADOC prisoners. PI, Ex. 1260, September 2016 Monthly Statistical • Report (doc. no,1108-37) at 4, 12.

25 . In fact, although the court' has visited a number of prisons over the years, the United States Marshals Service, in consultation with defense counsel, advised against the court’s visit to Holman Correctional Facility in this case due to safety concerns.

26 . A related issue is the new set of staffing ratios that ADOC Chief of Staff Brown presented during the trial, which counted overtime hours performed by existing correctional officers as additional officers. These ratios are also misleading. First, according to plaintiffs' expert Vail, counting overtime hours as additional full-time correctional officers is not the standard practice to determine whether correctional staffing is adequate. Second, these ratios do not take into consideration that officers working overtime are less effective than officers working standalone shifts, or that the overtime rate in ADOC is extremely high compared to other correctional systems, especially in facilities such as Donaldson, Kilby, St, Clair, Tutwiler, Draper, Holman, Bullock, and Easterling, Def. Dem. Ex, 19 (doc.-no. 1148— 60) (showing the eight facilities with 15 % or higher overtime rate), Furthermore, as with the authorized-position calculations discussed above, these ratios do not account for the fact that many ADOC facilities are designed with little direct line of sight from officer stations into prisoner living areas, and have dorms with rows and rows of bunk beds obstructing officers' views; both factors' require higher officer-to-inmate ratios than facilities with better line of sight or fewer bunked dorms.

Lastly, even if the court were to accept the current staffing ratios , calculated by Brown’s staff as accurate, only,two of the 14 facilities meet the 7:1 (for medium custody) or 6:1 (for close custody) thresholds. In other words, even using this overly inclusive metric to measure staffing sufficiency, ADOC is significantly understaffed. .

27 . Associate Commissioner Naglich testified that psychological associates, who have master's degrees in counseling and are employed by ADOC, also have the ability to refer prisoners to psychiatric providers at intake. However, other evidence suggested that this rarely, if ever, happens. Dr, Hunter explained that ADOC’s intake process, which involves psychological tests, is a parallel track to MHM’s screening process, and that they do not overlap; the court interpreted this to mean that ADOC’s psychologicál associates do' not interact with psychiatric providers on the MHM side for further evaluation of prisoners. In addition to Dr. Hunter’s testimony, no documentary evidence could be found to support Naglich’s assertion that psychological associates do refer prisoners for further examinations during the intake process. See also Joint Ex, 100, Admin. Reg, § 610 (doc, no. 1038-122) (detailing the mental-health screening process to be conducted by the contractor staff), Given Dr, Hunter’s familiarity with the ■ intake process and the lack of any documem tation of psychological associates’ referrals, the court finds that the initial intake process is primarily or entirely done by an LPN.

28 . Experts from both sides also observed that the intake process does not include an assess- : ment for-suicide risk, a serious systemic-issue that may have contributed to the recent dramatic increase in the suicide rate. See Joint Ex. 461, Patterson Expert Report (doc, no. 1038-1Ó46) at 69 (concluding that ADOC’s lack of suicide risk evaluation and management is an area of substantial concern); Burns Testimony at vol. 1, 63; PL Ex. 1267, 2015-2016 Chart of ADOC Suicides (doc. no. 1108-38) (showing 12 suicides between September 2015 and December 2016). ADOC has now implemented suicide risk assessments as part of their regular intake procedure based upon Dr. Patterson’s recommendation. However, as discussed in more detail later, ADOC has not incorporated suicide risk-assessment tools into other parts of the mental-health care system, despite. Dr. Patterson’s recom: mendation to do so.

29 . Plaintiffs have objected to the use of the audit results on Daubert grounds, contending that the methodology used to conduct the audit was not reliable and has not been accepted in the field of correctional mental-health care as a way of evaluating adequacy of care, Based on Dr. Patterson's testimony on the methodology, the audit results are admitted. However, as will be explained more extensively in a separate Daubert opinion, limitations in the methodology and implementation of the audit have been taken into consideration in evaluating their weight.

30 . ADOC’s mental health coding system was amended twice in 2016. According to the latest version, a new level (MH-9) refers to those who cannot be transferred to any facility and must be held at the current housing facility. However, the description of the code does not give any specifics about the patient’s symptoms and only specifies who may revise such a code. Joint Ex. 107, Admin. Reg. § 613-2 (doc. no. 1038-130). There is no MH-7 or MH-8 in the system.

31 . Dr. Burns explained that inappropriate classification of mentally ill patients partially stems from a lack of proper documentation in treatment plans and progress notes. Combined with a high turnover rate of staff and frequent transfers between facilities, inadequate documentation means that information about a patient’s symptoms and treatment is not well preserved. As a result, symptoms are evaluated without the context and history of each patient, leading to a higher risk of under-classifying and underestimating the acuity of mental illnesses.

32 . These numbers are based on the number of patients currently on the mental-health caseload. Because ADOC misses a significant portion — at least 5 % of the inmate population, or a third of those who are already on the caseload — of those who need mental-health care during its intake screening and referral processes, it is likely that even more prisoners need residential mental-health care than calculated here.

33 . As explained in more detail later, many of the cells in the mental-health unit are being used to house segregation prisoners without anyméntal-health needs.

34 . As an aside, the court notes that treatment planning can be viewed as serving similar purposes as medical recordkeeping, which also ensures continuity of care and coordination between different providers. Courts have held that maintaining accurate and complete records of mental-health treatment is an essential component of a minimally adequate mental-health care system. See, e.g., Ruiz v. Estelle, 503 F.Supp. 1265, 1339 (S.D. Tex. 1980) (Justice, J.), aff'd in part, rev'd in part on other grounds, 679 F.2d 1115 (5th Cir. 1982), opinion amended in part and vacated in part, 688 F.2d 266 (5th Cir. 1982), cert. denied, 460 U.S. 1042, 103 S.Ct. 1438, 75 L.Ed.2d 795 (1983); Balla v. Idaho State Bd. of Corrections, 595 F.Supp. 1558, 1577 (D. Idaho 1984) (Ryan, J.) (adopting the Ruiz standard of - six essential components of a minimally adequate mental health treatment program, including complete and accurate records); Coleman v. Wilson, 912 F.Supp. (E.D. Cal. 1995) (Karlton, J.) (same).

35 . Much testimony centered around how long a typical counseling session lasts. The court heard conflicting testimony: some counselors testified that most sessions do not last longer than 30 minutes; some refused to give a more concrete estimate altogether, other than saying their sessions might range from ten minutes to two hours. Moreover, there is no documentary or otherwise reliable evidence establishing such numbers. Given the lack of documentation, the number of patients on each clinician’s caseload at any given moment is a more reliable proxy for the quality and. frequency of therapy: even if some patients receive sufficiently long counseling sessions, the context of overwhelming caseloads means that the clinicians are not able to give such counseling sessions to most patients, by virtue of not having enough time in the day and having other duties. Therefore, the court finds that the overwhelming caseloads of psychiatric providers and counselors as relayed .by Dr. Hunter and Houser are more indicative of the quality and frequency of counseling sessions for the vast majority of prisoners on the mental-health caseload.

36 . By the same token, Dr. Haney- found it problematic that at some facilities ‘inmate newsletters’ identify exactly who is on the mental-health caseload, thereby increasing the stigma of mental-health care and discouraging prisoners from seeking mental-health treatment due to fear of exploitation by others.

37 . Associate Commissioner Naglich testified that unlicensed counselors can work in ADOC facilities only if they obtain a license within six months of starting employment. However, this testimony was contradicted by the employment data, as well as Houser’s testimony that MHM cannot hire licensed counselors within the contract- budget. Indeed, of the four current and former MHM counselors who testified at . trial, none had become licensed despite having worked in ADOC for multiple years.

38 . Practices within Tutwiler’s mental-health units are discussed separately in Section V.B.9.

39 . A part of each RTU is an ‘open RTU,' consisting of dormitories with rows of beds, rather than individual cells.

40 . Dr. Haney was not the only one who thought ADOCs celled mental-health units were indistinguishable from segregation units. In the corrective-action plan provided to ADOC in 2013, MHM stated that “concep-tualiz[ing] the [Bullock] SU !as a treatment unit, not as segregation” was necessary to further the goal of stabilizing patients. Ironically, or perhaps not surprisingly. Associate Commissioner Naglich also had a hard time distinguishing between segregation units and stabilization units during her testimony, frequently referring to stabilization units as ¡segregation units.

41 . The presence of segregation inmates in the mental-health units contributes to the dearth of out-of-cell time afforded to mentally ill prisoners in those units. Dr. Haney testified that it is very difficult to operate a unit that has mixed populations, and that it is not surprising that a unit that contains both segregation inmates and mental-health patients would be treated like a segregation unit. This is partially because correctional officers get confused as to how to operate a unit with two conflicting purposes — discipline and treatment. Relatedly, having segregation inmates in the unit means that mental-health patients cannot be let out of their cells as easily, especially when correctional staffing is minimal or inadequate.

42 . Tellingly, Dr. Patterson stated that while Tutwiler’s RTU programming is much better than RTUs at malé facilities — Bullock arid Donaldson — it is ■ still only "close to adequate,” but not adequate. Patterson Testimony at vol. 1, 92.

43 . Suicide risk-assessment tools include a checklist of risk factors and protective factors, such as age, gender, length of sentence, contact with family, and engagement in treatment, just to name a few. These tools also require the clinician to make a holistic assessment based on the answers and the appropriate weight of each factor to estimate the overall risk of suicide.

44 . No efforts have been made to administer the risk-assessment tool to the vast majority of prisoners in the system, who went through intake before that date.

45 . Plaintiffs’ expert Vail testified that inadequate outpatient and routine care would increase the need for crisis care, and inadequate crisis care would also increase the number of those who are placed on suicide watch over and over again, reinforcing the need for more crisis cells. Documentary evidence of prisoners being repeatedly placed on suicide watch supports this conclusion.

46 . The O dorm at Kilby, a small cell block with 13 cells, has been used as overflow crisis cells for the rest of the system. Transferring suicidal prisoners from their home institution to crisis cells at a different institution, while preferable to housing them in a place that is not suicide-proof, poses a host of problems. First,, it takes multiple correctional officers to transport a prisoner, which exacerbates the problem of correctional-officer shortages and may not even be possible depending on the , staffing level at the time the crisis is happening. Second, it jeopardizes the prisoner's mental state even further, since changing the environment of someone who is already in crisis can add to the distress the prisoner is experiencing. Dr. Tytell referred to the transfer experience as potentially "traumatic." Ty-tell Testimony at_Third, it interferes with continuity of care, where the team that was familiar with the patient’s symptoms and treatment is no longer in charge of treatment, and a new team of providers must get up to ■speed to treat á new patient, all in the context of time-sensitive care.

A related issue that arose during the trial is how to characterize the O dorm' at Kilby. Trial testimony showed that the O dorm is sometimes also used as a segregation overflow for Kilby. PI. Ex. 1257, Duty Post Log for O Dorm (doc. no. 1097-20) (showing "seg walk” and "seg shower” as part of tasks completed by correctional officers in O dorm). This practice makes it more likely that Kilby will run out of crisis cells. It also raises the same concerns that experts and MHM providers expressed regarding housing suicidal prisoners in a segregation unit or housing mental-health patients with 'heightened treatment needs and increased vulnerability near segregation inmates in general, as explained in more detail in the section on segregation practices.

47 . Some correctional officers also fail to take threats of self-harm seriously, and even worse, respond in dangerous ways. As Dr. Hunter testified, ADOC officers have responded to prisoners’ threats of self-harm with sarcasm or cracked jokes about suicidality, and even challenged inmates to follow through with suicide threats; on several occasions, ADOC officers essentially called a prisoner’s bluff and then that person attempted suicide.

48 . Interestingly, on the topic of 'secondary gains,’ plaintiffs’ expert Vail posited that with a functioning protective-custody system, in which prisoners who feel unsafe can be moved away from their enemies, the problem of trying to determine who is genuinely suicidal would be alleviated. While all prison systems have conflicts among prisoners and a risk of inmate-on-inmate violence, he explained, prisoners in other correctional systems who feel unsafe generally pursue other avenues to protect themselves, such as requesting protective custody or transfer, rather than requesting to be placed on suicide watch. According to Vail, prisoners generally do not request suicide watch solely for protection because suicide cells are not, generally speaking, a desirable environment. This testimony suggests that to the extent non-suicidal ADOC prisoners actually are electing such an undesirable cell environment for protection, either the protective custody system is inadequate, or general population dorms are ridden with so much violence that it is rational for prisoners to choose suicide-watch cells over the dangerous environment of general population. Vail’s observations also indicate that improving the protective-custody system and addressing the underlying problem of prisoner safety could be a safe and effective way of ensuring that only those who need crisis cells are placed there; it would also be a solution that does not involve placing suicidal prisoners at a substantial risk of serious harm by taking a gamble on whether prisoners are actually suicidal.

49 . Dr. Haney also noted that this prisoner was locked in die office with no access to a bathroom.

50 . A related problem is the inadequate mental-health staffing at prisons that provide only outpatient mental-health care: those who are confined in crisis cells, or even more inappropriate settings, such as shift offices or libraries, do not have access to mental-health care over the weekend, because most outpatient-only prisons do not have mental-health staff on weekends. In other words, despite their 'crisis' condition, suicidal prisoners in these facilities are often left in a crisis cell or a non-suicide-proof environment for multiple days waiting to see a mental-health staff member.

51 .ADOC also sometimes places crisis-care inmates in stabilization units. This is problematic not so much for the suicidal inmates but for the others: according to Dr. Hunter of MHM, housing suicidal prisoners in crisis cells within an SU negatively impacts MHM’s ability to care for the severely mentally ill already in the unit.

52 . While professional standards like those promulgated by NCCHC do not necessarily set the constitutional floor for minimally adequate mental-health care under the Eighth Amendment, substantial deviations from accepted standards can indicate an Eighth Amendment violation. See Steele v. Shah, 87 F.3d 1266, 1269 (11th Cir. 1996) (holding that providing care where the quality is "so substantial a deviation from accepted standards” can constitute deliberate indifference). Moreover, ADOC's contract with MHM requires MHM to comply with those standards.

53 . Defendants argued that inappropriate items found in crisis cells can no longer be part of the case because of this settlement. However, the problem is broader in scope: the settlement agreement to discontinue providing razor blades by no means discharges ADOC’s responsibility to ensure that objects with which suicidal prisoners can engage in self-harm are not found in crisis cells.

54 . The risk of allowing suicidal prisoners access to sharp implements is obvious. However, as Dr. Burns explained, sharp items pose a serious risk even to prisoners who do not have any intention of killing themselves but engage in cutting; it is easy to cut too deep by accident and cause potentially fatal bleeding.

55 . NCCHC promulgates two types of standards: essential and important. As the terms would indicate, the distinction between the two denotes the relative importance of each standard. The essential standards are mandatory conditions for accreditation by NCCHC; only 85 % compliance with important standards is required.

56 . In addition, there were allegations of noncompliance with the constant watch procedures at Kilby. See Plaintiffs’ Motion to Renew the Temporary Restraining Order Regarding Suicide Prevention Procedures (doc. no. 1171). This allegation of non-compliance will be discussed in infra Part V.D.

A separate issue is whether Commissioner Dunn’s representation that he will enforce the interim agreement indefinitely is binding on ADOC or his successors. This issue is taken up in Part V.D.

57 . Houser explained that prior to the interim agreement, MHM could not staff constant watch under the current contract amount and was not expected to do so, even though NCCHC standards mandate, constant watch.

58 . Defendants elicited testimony from various practitioners and prisoners that it is sometimes appropriate to discipline a prisoner for a violation of administrative rules despite the fact that he suffers from a mental illness. But plaintiffs have not disputed this point. Instead, they have offered evidence to show that many mentally ill prisoners are punished as a direct result of their mental illness, which the experts credibly testified is harmful. For example, defense counsel asked multiple prisoners, including plaintiff Jamie Wallace and class member M.P., whether it was ‘appropriate’ to be disciplined for having a contraband — a razor blade, for example — in the cell; however, these prisoners actually had received disciplinary sanctions for engaging in self-injurious behavior, not for having contraband.

59 . While the corrective-action plan for Donaldson specifies “male inmates,” Associate Commissioner Naglich testified that she understood the policy change — to cease automatic disciplinary sanctions for engaging in self-harm — applied to both male and female prisoners. Naglich Testimony at vol. 2, 135.

60 . As Dr. Haney explained in his testimony before Congress, exercise time for segregation prisoners hardly involves a ‘yard.’ PI. Ex. 1272, 2012 Congressional Testimony of Dr. Craig Haney (doc. no. 1126-3) at 5-6. Rather than an open space with greenery, the exercise yards that the court observed at ADOC facilities for segregation prisoners were often small and fenced in with concrete surfaces. Some of the facilities allow only one inmate at a time in a 'cage,' a subdivided section of the yard that is fenced in and hardly bigger than the segregation cell itself. Some of the yards, such as the one in Kilby, also had fences totally enclosing the yard, including a fenced ceiling, truly evoking the feeling of a cage.

61 . See National Commission on Correctional Health Care, Solitary Confinement Position Statement on Solitary Confinement, 2016; Burns Testimony at vol. 1, 204.

As Dr. Haney explained, prison systems around the country are also moving away from using solitary confinement in general— even for healthy people — unless it is absolutely necessary. See, e.g., Joint Ex. 459, Haney Expert Report (doc. no. 1038-1043) at 133 (referencing Rick Raemisch, My Night in Solitary, N.Y. Times (Feb. 20, 2014), available at http ://www.nytimes. com/2014/02/21/opinion/ my-night-in-solitaiy.html (describing the experience of the head of the Colorado Department of Corrections spending 20 hours in a segregation cell and the efforts to bring down the number of mentally ill prisoners in administrative segregation to 'single digits among 500 prisoners in segregation); Teriy ICupers, et al., Beyond Supermax Administrative Segregation: Mississippi's Experience Rethinking Prison Classification and Creating Alternative Mental Health Programs, 36 Crim. Just. & Behav.' 1037 (2009) (describing the reforms in the Mississippi Department of Corrections significantly reducing the population in administrative segregation and its effept on misconduct, violence, and use of force)).

62 . Dr. Haney .also observed that Bullock’s segregation unit has a practice of removing mattresses from cells so. that prisoners cannot rest on them during the day, which he described as ''extraordinarily draconian.” Haney Testimony at vol. 1, 117. PI. Dem. Ex. 60, Bullock Main Camp, B Dorm (doc, no, 1125— 20). The court also observed that Kilby's large segregation unit (also known as 'big seg’) has extremely small cells that are only a foot or two longer than the length of a single-sized mattress and only a narrow strip of space that barely fits a toilet, in a stifling unit of fifty cells stacked on top of each other without any ventilation or transparent windows facing outside. PL Dem. Ex. 80 & 81, Kilby C Dorm (docs. no. 1125-38, 1125-39).

63 . The coúrt notés that the worst thing that could happen in this context is for the correctional officers and ADOC officials to get accustomed to such conditions.

64 . Admittedly, ADOC uses double-celling in some segregation units, which means putting two prisoners into a single segregation cell. At first blush,, this practice might seem to mitigate the harmful effects of solitary confinement. However, double-celled segregation has an even more severe impact on the mental health of prisoners. Dr, Haney credibly explained that double-celled prisoners "in some ways ... have the worst of both worlds: they are 'crowded' in and confined with another person; inside a small cell but — and this is the crux of their ‘isolation’ — simultaneously isolated from-the rest of the mainstream prisoner population, deprived of even minimal freedom of movement, prohibited from access to meaningful prison programs, and denied op■portunities for any semblance of 'normal' social interaction.” Joint Ex. 459, Haney Expert Report (doc. no. 1038-1043) at. 109.

65 . Experts from both sides explained that alternatives to placing mentally ill prisoners in segregation exist. Prison systems across the country, ranging from Maine to Mississippi, have reduced the number of prisoners in segregation generally, and significantly reduced the mentally ill population in segregation. Joint Ex. 459, Haney Expert Report (doc. no. 1038-1043) at 133. For example, California operates a separate housing unit that is devoted to mentally ill prisoners who have committed disciplinary infractions. These units provide 20 hours of out-of-cell time per week, as well as structured and unstructured therapeutic activities. Arizona has begun similar reforms, providing more programming and out-of-cell time to mentally ill prisoners who committed disciplinary infractions. Haney Testimony at vol. 2, 154-55; Ayers Testimony at

66 . There is sufficient evidence that these mental-health evaluations in segregation are inadequate, which will be discussed later in section V.B.10.

67 . Plaintiff C J. is also an example of prisoners who experience whát was referred to during the trial as 'segregation rotation,’ whereby a prisoner is sent from one segregation unit at a facility to another segregation unit at another facility every few months. C.J.’s movement history indicated that he has been rotating among three different segregation units in the last eight years, averaging eight months at each facility at a time. PI. Dem. Ex. 131, Movement History of Exemplar Plaintiffs (doc. no. 1126-10). This practice, according to Associate Commissioner Culliver, is used to "give staff a break” and "give the inmate an opportunity to restart.” Culliver testimony at _Culliver did not know how many people were on segregation rotation currently, or how many mentally ill prisoners are on segregation rotation.

This practice adds an additional set of risk factors to the already debilitating and harmful practice of housing mentally ill patients in segregation for prolonged periods of time. Dr. Haney testified that moving mentally ill prisoners from one environment to another disrupts treatment, because of lack of continuity of care and providers: a new set of staff must get to know the patient, and the usefulness of the information that staff have already gathered on the person gets lost when the prisoner is transferred. C.J. testified that he often has to start anew with new counselors at each facility, and when he goes back to the.old facility after a year or two of absence, the former counselor is often no longer working there because of the high turnover rate. Furthermore, as .Dr. Haney testified, frequent transfers of mentally ill prisoners have an adverse impact on their mental health because they have a more difficult time adjusting to new environments than those who are not mentally ill.

68 . Dr, Haney also stated that cycling between .segregation and general population may also indicate that those prisoners are likely suffering from the after-effects of prolonged stays in segregation, which are leading to more disciplinary infractions.

69 . The plaintiffs' summary chart and MHM’s monthly operations reports count Draper and Elmore as part of Staton, because the three facilities are in the same complex.

70 . According to Houser, MHM and ADOC discussed a pilot project for long-term treatment programming in the segregation unit at St. Clair, but the project never got off the ground because of the lack of support from ADOC.

71 . As discussed in the section regarding sound confidentiality and psychotherapy, most ADOC segregation units are not conducive to having a cell-front conversation, due to heavy solid doors and very loud units with dozens of cells in a single unit. As the court saw during its tours of five prisons, none of the units — even the ones at Bibb, where only three cells are in a unit — were conducive to confidential conversations, because of the proximity to other cells and prisoners,

72 .Furthermore, as Dr. Haney testified, while segregation rounds by mental-health staff are crucial for checking for signs of decompensation or crisis, they cannot replace periodic out-of-cell clinical assessments of prisoners’ mental-health status, because it is difficult to observe someone’s behavior and accurately assess the prisoner’s mental health through " cell-front encounters.

One vivid example of ADOC's failure to monitor segregation prisoners’ fnental-health status concerned plaintiff R.M.W. After a month of segregation placement during which she was twice sent to a crisis cell and had multiple episodes of self-injury, the segregation mental-health evaluation form indicated that the inmate was “appropriate for placement” and the recommendation was “segregation placement not impacting inmate’s mental health.” Joint Ex. 404, March 28, 2014 Review of Segregation Inmates — • R.M.W. (doc. no. 1038-859) at MR017081. Nothing in her medical records suggests that a suicide-risk assessment was done after any of the episodes or before this review to ascertain the impact of segregation and likelihood of recurring self-harm.

73 . While no aggregate data on the average or typical lengths of segregation stays were presented, the court, during its visits to six facilities, was able to view forms on the front of segregation cells showing how long the prisoner had been there: most were there for at least several weeks, some for months or even over a year. As discussed earlier, some inmates, like plaintiff C J., are placed on ‘segregation rotation,’ which can keep prisoners in segregation units for years on end. Experts on both sides unequivocally denounced ADOC’s practice of prolonged segregation stays.

74 . The only one that did not take place in segregation was plaintiff Wallace, who was in the Bullock stabilization unit. See PL Ex. 1267, 2015-2016 Chart of ADOC Suicides (doc. no. 1108-38).

75 . The court further notes that the system that Associate Commissioner Naglich described would prevent the placement of seriously mentally ill prisoners in segregation only if the mental-health coding system were accurately classifying prisoners’ mental-health needs.

76 . Defense counsel suggested that the approval of certain policies at Tutwiler by monitors hired by the U.S. Department of Justice signifies that those policies are constitutionally adequate. However, there are two flaws with this argument. First, the DOJ monitor was not necessarily evaluating policies to ensure that mental-health care was adequate under the Eighth Amendment: the lawsuit that resulted in the monitoring was not about mental-health care, nor was the monitor’s job to set the constitutional floor of mental-health care. Second, the monitors' approval of certain policies, such as segregation placement, does, not mean that ADOC’s actual practices are constitutionally adequate.

77 .For example, newly implemented practices include limiting pre-disciplinary hearing segregation to 72 hours, submitting monitoring logs for segregation cells to an independent reviewer, and having a compliance visit to the stabilization unit every six months to ensure 15-minute interval checks.

78 . The court also notes that the experts from both sides presented affirmative evidence that the care being provided in the Tutwiler RTU is adequate, or close to adequate.

79 . Defendants also asserted that because the named ADOC officials were not involved in the direct provision of mental-health care to prisoners, they could not have been deliberately indifferent to the plaintiffs' serious mental-health needs. This court has already rejected this argument. See Dunn v. Dunn, 219 F.Supp.3d 1100, 1157-60 (M.D. Ala. 2016).

80 . ADOC Associate Commissioner Culliver and the regional • coordinator for medical care, Brendan- Kinard, also have been aware of crisis-cell shortage and the resulting placement of suicidal prisoners in' non-crisis cells for years.

81 . Likewise, the current levels of' mental-health and correctional understaffing and overcrowding also illustrates ADOC’s disregard of risk of harm. First, ADOC’s response to the shortages of mental-health and correctional staff have been objectively insufficient, because systemic and gross deficiencies arising from understaffing have persisted and effectively denied prisoners access to adequate mental-health care. Taylor, 221 F.3d at 1258 (holding that the ‘disregard’ prong under Estelle and Farmer can be satisfied through an "objectively insufficient response” by prison officials); Harris, 941 F.2d at 1505 (deliberate indifference can be established through "systemic and gross deficiencies in staffing” that effectively deny prisoners access to adequate medical care); see also Coleman v. Wilson, 912 F.Supp. 1282, 1319 (E.D. Cal. 1995)(Karlton, J.)(“[G]iven the nature and extent of the crisis and its duration,” defendants’ purported efforts to remedy the acute shortage of mental-health staff in the prison system were not sufficient to defeat a deliberate-indifference finding). Furthermore, difficulties in recruiting do not negate the fact that understaffing has caused this serious systemic deficiency. See Wellman v. Faulkner, 715 F.2d 269, 273 (7th Cir. 1983) (failure of a prison to fill authorized position weighs "more heavily against the state than for it,” partly because the authorized salary was woefully inadequate and the prison’s effort was insufficient); Madrid v. Gomez, 889 F.Supp. 1146, 1227 (N.D. Cal. 1995) (Henderson, J.) (finding "recruitment difficulties do not excuse compliance with constitutional mandates.”). In other words, ADOC’s failure to provide mental-health and correctional staffing sufficient to operate a minimally adequate mental-health care system is in itself an unreasonable response under the deliberate-indifference standard.

The same logic applies to overcrowding. While it is true that ADOC does not have the authority to release prisoners or stem the inflow of prisoners from the state’s criminal justice system, ADOC’s response to overcrowding has been objectively insufficient. This is because the court does not consider the overcrowding problem in a vacuum. ADOC has been well aware of the magnitude and impact of overcrowding on every facet of its operations for years. ADOC’s efforts — belatedly pushing for construction of new prisons in 2016, for example — to alleviate the problem have been too little and too late, as reflected in the current 170 % occupancy rate. Considering the institution’s historical deliberate indifference to the problem of overcrowding, rather than what ADOC has done under the current leadership only, the court finds that ADOC has disregarded to the harm and risk of harm caused by overcrowding and understaffing.

82 . In fact, instead of penalizing MHM for its known inadequacies, ADOC extended the contract with MHM for one more year in September 2016. Associate Commissioner Naglich credibly testified that, as a result of her negative view of MHM's performance, she recommended awarding the mental-health contract in 2013 to another contractor, rather than renewing the contract with MHM; She likewise stated that before the department extended its contract with MHM in 2016, she told Commissioner Dunn that MHM was not "measuring up,” Naglich Testimony at vol. 4, 121, adding that Dunn was also dissatisfied with all the issues that ADOC has had with MHM. And yet, ADOC renewed the contract with MHM regardless.

83 . One former MHM employee testified that these audit results are not reliable, because MHM staff on site pull medical records to be audited ahead of time and get them up to par before the corporate auditors review them. Plaintiffs’ expert Dr. Bums also observed that the facility staff select the files for review, rather than the corporate office randomly selecting the files. While this testimony raises a concern that the reports may have minimized negative findings, the court relies on them to the extent the reports still found serious deficiencies in the provision of mental-health care.

84 .MHM's corrective-action plans reflect this conundrum: while MHM is required to send a corrective-action plan in response, much of what is required to fix the deficiencies identified in contract-compliance reports involves ADOC actions. For example, follow-up findings in the corrective-action plan for 2016 included statements such as: "This is a work in progress. Due to the staffing issues currently are not being completed during the required time frame”; "Still working with the MHPs on ensuring that the treatment plan is completed during this time frame of admission.” PL. Ex, 1247, -July 2016 Bullock IP Corrective Action Plan (doc. no. 1,099 — 10).

85 . To be clear, the court notes that many identified problems could be fixed by MHM, but are not.

86 . Asked how she knows that MHM is meet- ■ ing.contract compliance goals if all follow ups are done at the site level and she does not see any - of those results, the CQI manager responded, "[O]bviously, you do not understand quality.” Davis Walker Testimony at vol. 2, 237. She testified that the purpose of CQI is “refining [] processes],” which she defined as determining how to collect data and reflect it in a database. Davis-Walker Testimony at vol. 1, 83. This singular focus on process rather than substance on her part led to one of the more bizarre exchanges of this trial: she insisted that all spot-audit results showing failures to meet contract or regulatory standards were exclusively attributable to data-entry problems, and never to any actual failure to provide appropriate care. For example, she insisted that noncompliance reported in the audit, such as treatment plans that were "outdated or requiring review,” reflected database entry problems, even though finding the date of the latest treatment plan did not involve looking in the database. Davis-Walker Testimony at vol. -2, 130-32. Needless to say, the court did not find credible her testimony that all identified problems are attributable to mere data entry errors.

87 ,Lynn Brown, the only other person within OHS who interacts with MHM regularly, also denied ever seeing the reports or being responsible for reviewing them.

88 , MHM’s program manager Houser explained that Naglich told her the 'pilot audit’ would not "count” and that the results would not be used as an "I gotcha.” Houser Testimony at vol. 2,. 176,

89 . Lynn Brown, the other ADOC employee who attends the MHM CQI meetings but does not have any mental-health training, testified that she is not responsible for reporting from the meetings unless specifically told to do so. This office-wide lack of involvement in the CQI process further supports the finding that ADOC has chosen to abdicate its duty of ensuring that MHM's delivery of mental-health care is minimally adequate.

90 . As explained in the knowledge section, MHM managers, including Dr. Hunter and Houser, have both discussed this issue with her on multiple occasions. In addition, MHM has been sending monthly operations reports and annual contract compliance reports stating the same.

91 . Tytell also tried to evade responsibility by saying that he was not responsible for the actual transfer or monitoring of the transfer: he first contended that whether mentally ill prisoners are actually being transferred out of segregation was up to the wardens and site administrators, because they have the mental-health codes of prisoners in segregation; he insinuated that he did not have any way of monitoring the movement of mentally ill prisoners in and out of segregation, However, he then admitted'that both himself and Associate Commissioner Naglich do have access to the mental-health codes of prisoners in segregation, At the time of his testimony, Dr. Tytell had never run a report to ascertain how many mentally ill prisoners remain to be moved out of segregation, and Naglich had never requested to see such a report. .

92 . The interplay between the mootness inquiry and the ongoing-violation requirement under Ex parte Young is somewhat unsettled. However, the Eleventh Circuit, along with the Fifth Circuit and the Sixth Circuit, has suggested that a threat of recurrence sufficient to render a claim not moot should also be sufficient for the ongoing-violation requirement. See Nat’l Ass’n of Bds. of Pharm. v. Bd. of Regents of the Univ. Sys. of Ga., 633 F.3d 1297, 1308-09 (11th Cir. 2011) (treating a dispute regarding whether the plaintiff alleged an ongoing violation as a mootness inquiry); K.P. v. LeBlanc, 729 F.3d 427, 439 (5th Cir. 2013) (rejecting the contention that a non-moot claim did not meet the ongoing-violation requirement, because "[that] theory, if accepted, would work an end-run around the voluntary-cessation exception to mootness where a state actor is involved”); Russell v. Lundergan-Grimes, 784 F.3d 1037, 1047 (6th Cir. 2015) (“[A]t the point that a threatened injury becomes sufficiently imminent and particularized to confer Article III standing, that threat of enforcement also becomes sufficient to satisfy ... Ex.parte Young.")); see also Muhammad v. Crews, No. 4:14CV379-MW/GRJ, 2016 WL 3360501, at *6 n.5 (N.D. Fla. June 15, 2016) (Walker, J.) (summarizing the case law). Here, the court addresses both the mootness argument and the ongoing-violation argument, though the analyses overlap.

93 . MHM’s program director Houser explained that ADOC started looking into fixing the doors on the Holman suicide watch cells (which have bars that can provide a tie-off point) during the last week of December 2016. This was close to a month after the trial had begun, and years after MHM started reporting to ADOC that Holman crisis cells are not safe. These belated actions illustrate that without a court order, ADOC will continue to look the other way despite the glaring deficiencies that put mentally ill prisoners at a substantial risk of serious harm, including death.

94 . As discussed earlier, many prisoners who commit suicide while in ADOC custody are not actually on suicide watch at the time; in fact, many reside in general population units without receiving any mental-health treatment, This suggests that meaningful remedial suicide prevention efforts cannot be confined to those already identified as high risk, but also must include identifying those at high risk among the general population.

95 , The court attributes this lack of knowledge to the Associate Commissioner being overwhelmed due to understaffing.

96 . Defendants did not make this argument explicitly during the trial, but the court addresses it since Associate Commissioner Nag-lich’s contention regarding the new coding system could be construed as arguing that plaintiffs’ claim regarding ADOC’s segregation practice is now moot.

97 . The court recognizes that ‘extenuating circumstances' and 'prolonged periods of time’ are somewhat ambiguous terms but leaves them to be defined during the remedy phase with the parties’ input.