Braggs v. Hamm

IN THE DISTRICT COURT OF THE UNITED STATES FOR THE

MIDDLE DISTRICT OF ALABAMA, NORTHERN DIVISION

EDWARD BRAGGS, et al., )

)

Plaintiffs, )

) CIVIL ACTION NO.

v. ) 2:14cv601-MHT

) (WO)

JEFFERSON S. DUNN, in his )

official capacity as )

Commissioner of )

the Alabama Department of )

Corrections, et al., )

)

Defendants. )

PHASE 2A OPINION AND ORDER

ON MONITORING OF EIGHTH AMENDMENT REMEDY

Previously, this court found that the State of

Alabama provides inadequate mental-health care in its

prisons in violation of the Eighth Amendment’s

prohibition against cruel and unusual punishment. See

Braggs v. Dunn, 367 F. Supp. 3d 1340 (M.D. Ala. 2019)

(Thompson, J.); Braggs v. Dunn, 257 F. Supp. 3d 1171

(M.D. Ala. 2017) (Thompson, J.). The issue now before

the court is the development of a plan to monitor

compliance with the court’s orders to remedy that

constitutional violation. The court will adopt in large

part the defendants’ plan--substantial portions to which

the plaintiffs have agreed--with some alterations. Most

significantly, the court will adopt the defendants’

overarching proposal that, in light of their own

admission that they lack the capacity to self-monitor,

outside experts will initially monitor compliance and

will draw on their expertise to develop many of the

details of the monitoring plan. See Defs.’ Response

(doc. no. 2295) at 14. Those outside experts will train

and eventually hand control over to an internal

monitoring team, building the capacity of the Alabama

Department of Corrections (ADOC) to regulate itself. The

court hopes that this monitoring scheme will help the

ADOC attain timely, meaningful, and sustainable

compliance with the court’s remedial orders on

mental-health care and bring this litigation to an end

as soon as is reasonably possible.

I. PROCEDURAL BACKGROUND

The plaintiffs in this class-action lawsuit are ADOC

inmates who have mental illness and the Alabama

Disabilities Advocacy Program, which represents mentally

ill inmates in Alabama. The defendants are the ADOC

Commissioner and the ADOC Associate Commissioner of

Health Services, who are both sued in only their official

capacities. In a liability opinion, this court found

that ADOC’s mental-health care was, “[s]imply put, ...

horrendously inadequate.” Braggs, 257 F. Supp. 3d at

1267. The court laid out seven factors contributing to

the Eighth Amendment violation, in addition to the

“overarching” problems of understaffing and

overcrowding. Id. at 1267-68. After two months of

mediation to develop a comprehensive remedial plan, it

became apparent that the remedy was too large and complex

to be addressed all at once. The court therefore severed

the remedy into the various contributing factors, to be

addressed seriatim. See Phase 2A Revised Scheduling

Order (doc. no. 1357).

The court has now issued remedial opinions and orders

regarding, among other things, understaffing, see Braggs

v. Dunn, No. 2:14cv601-MHT, 2018 WL 985759 (M.D. Ala.

Feb. 20, 2018) (Thompson, J.), and inpatient treatment,

see Braggs v. Dunn, No. 2:14cv601-MHT, 2020 WL 2789880

(M.D. Ala. May 29, 2020) (Thompson, J.). The court has

also issued several remedial orders temporarily adopting

the parties’ stipulations regarding other contributing

factors, see, e.g., Braggs, v. Dunn, No. 2:14cv601-MHT,

2018 WL 2168705 (M.D. Ala. Apr. 25, 2018). In March

2020, based on the parties’ agreement that their

stipulations temporarily satisfy the Prison Litigation

Reform Act (PLRA), 18 U.S.C. § 3626(a)(1)(A), the court

issued an interim injunction extending these orders

until, at the latest, December 30, 2020. See Interim

Injunction (doc. no. 2793). The issue of whether the

stipulations satisfy the requirements of the PLRA beyond

that date is set for a hearing in September, and the

court will defer judgment as to whether the measures are

warranted until that hearing has occurred.

Throughout the process of resolving each remedial

issue, the question of monitoring compliance with the

court’s orders has repeatedly arisen. The issue of

monitoring “raises important questions regarding, on the

one hand, the duty of courts to avoid overly intruding

into the executive matter of prison administration, and

on the other hand, the duty of courts to ensure that the

constitutional violations they find are effectively

remedied in a timely fashion.” Phase 2A Order on

Monitoring (doc. no. 1927) at 2-3. In pursuit of the

proper balance of these important interests, the court

opted to resolve the issue of monitoring separately from

all substantive remedial orders and on a global scale,

rather than as to each individual order.1 See id.

1. As part of its order for immediate relief for

suicide prevention, the court established an interim

external monitoring scheme and required ADOC to establish

a formal internal monitoring scheme. See Braggs v. Dunn,

383 F. Supp. 3d 1218, 1282 (M.D. Ala. 2019) (Thompson,

J.). Both forms of monitoring were narrowly focused on

the immediate suicide-prevention relief. See id.

However, at the request of the parties, the court stayed

that order and substituted the parties’ voluntary

agreement. See Order (doc. no. 2569) at 2. At this

point, the order remains stayed pending the court’s

The court also suggested to the parties that the

scheme for court monitoring should include not only

‘external’ monitoring by experts (an

‘external-monitoring team’ or EMT) but eventual

‘internal’ monitoring by ADOC itself (an

‘internal-monitoring team’ or IMT). In other words, for

part of the period of monitoring, the court substituted

internal monitoring for external monitoring. As

explained in more detail later, the model would divide

the traditional period of monitoring into three parts,

bringing ADOC into the process earlier and in a more

substantive role than usual. In the first phase, the EMT

will assess and monitor ADOC's compliance with the

court's remedial orders; next, the EMT, as part of its

monitoring, will train the ADOC, through its IMT, how to

monitor itself; and, finally, ADOC, through its IMT, will

monitor itself.

determination of whether the parties’ agreement complies

with the PLRA. See Phase 2A Revised Scheduling Order

(doc. no. 2784) at 5.

The court adopted this model in the hope that it

would facilitate a more effective, less intrusive process

and avoid an indeterminate period of external monitoring.

External monitoring and internal monitoring complement

each other: external monitors offer an outside

perspective on ongoing issues, while internal monitors

have more familiarity with and investment in the

remediation efforts. External monitoring will also

provide valuable information for ADOC, allowing it to

more effectively implement its own system of internal

monitoring.

The court further believed that self-monitoring

would help ADOC develop internal buy-in, resulting in

more active cooperation and timely compliance. This

method invites ADOC to be directly involved in the

monitoring effort, encouraging collaboration and

investment in reform rather than an adversarial posture.

The internal monitoring team will work with and learn

from the external monitoring team, building ADOC’s

capacity and making the eventual termination of court

oversight more seamless.

Finally, in light of the fact that the ultimate goal

of this litigation is not just monitoring of ADOC but

adequate monitoring by ADOC, this model should build

ADOC’s internal capacity and help it sustain compliance

over the long term. Ultimately, the court hopes that

this hybrid monitoring process will finally bring to an

end the history of repeated litigation ADOC has

confronted over its mental-health care since the 1970s,

as described in the final section of this opinion. Both

the plaintiffs and the defendants have agreed to this

model, should the court order external monitoring.

When the court reached the monitoring issue, it first

gave the defendants an opportunity to propose an overall

plan and allowed the plaintiffs to respond. See Defs.’

Proposed Monitoring Plan (doc. no. 2115); Pls.’ Response

(doc. no. 2133). The court then held a hearing on the

defendants’ proposed monitoring plan in which it heard

testimony from plaintiffs’ correctional psychiatry

expert Dr. Kathryn Burns; plaintiffs’ correctional

administration expert Eldon Vail; Executive Director of

the Alabama Disabilities Advocacy Program James Tucker;

ADOC Associate Commissioner for Health Services Ruth

Naglich; ADOC Commissioner Jefferson Dunn; and four

individuals the defendants proposed as the external

monitoring team: Larry Linton, MargaRita Pauley,

psychiatrist Dr. Robert Stern, and psychologist Dr. David

Clayman.

In April 2019, following the hearing and the parties’

subsequent briefing on the monitoring issue, the United

States Department of Justice (DOJ) issued a findings

letter regarding unsafe conditions in ADOC facilities,

including due to understaffing, overcrowding, and

violence. See DOJ Findings Letter (Pls.’ Ex. 2739); SPLC

Letter to Governor Ivey and Commissioner Dunn (doc. no.

2472). The parties subsequently filed a joint

motion--which the court granted--to stay all matters

under submission in this litigation, including the

monitoring issue, for 90 days to allow the parties to

pursue a global resolution, via mediation, between the

parties as well as with DOJ. See Joint Notice and Mot.

to Stay (doc. no. 2560); Order (doc. no. 2569). This

stay was twice extended upon joint motions of the parties

to allow for further mediation. See Order (doc. no.

2608); Phase 2A Revised Scheduling Order (doc. no. 2720).

On March 25, 2020, the parties informed the court during

an on-the-record conference call that they had been

unable to reach an agreement on the monitoring issue in

their negotiations and that the issue was thus again

submitted to the court for resolution. See Phase 2A

Revised Scheduling Order (doc. no. 2784). Today’s

opinion fully resolves the remedial monitoring issue.

Though the remedies for all seven factors

contributing to the constitutional violation have not yet

been reduced to final orders with PLRA findings, and

though there remain some additional remedial issues for

resolution (for example, segregation and inpatient

treatment), the court need not wait to issue those orders

prior to resolving the monitoring issue. This is because

the court’s order today is not specific to any particular

remedial measures in that it does not set up the means

of measuring compliance; rather, the court’s order

establishes an overarching monitoring structure and

scheme, the details of which the court leaves to be filled

in by the experts, as both sides agree is appropriate.

II. THE MONITORING SCHEME

The parties are to be commended for reaching

significant areas of agreement on the issue of

monitoring. They agree that the monitoring scheme here

has two fundamental goals: (1) to oversee compliance with

the court’s remedial orders and (2) to build ADOC’s

capacity to exercise sustainable internal oversight of

mental-health care--that is, to identify and correct

problems. They also agree on the overarching structure

of this monitoring. Specifically, they agree that the

scheme should: (1) include the EMT; (2) consist of three

phases, with the EMT teaching--and then ultimately

handing the reins over to--the IMT; (3) empower the EMT

to determine many of the details of how to carry out

monitoring, including fashioning performance measures

and audit tools; and (4) consist of a number of essential

components of monitoring, including document review,

observation, feedback, consultation, and handoff to ADOC

to monitor itself going forward.

At the same time, there are some disagreements

regarding specifics under each broader point of

agreement. These areas of dispute are outlined below.

The court will first summarize each area of dispute and

then explain the court’s resolution.

Overall, the defendants assert that the court should

approve their plan without entering any order on

monitoring because ADOC should be allowed to “voluntarily

undertake culture change.” Defs.’ Response (doc. no.

2295) at 9. But, as described in detail in the final

section of this opinion, ADOC’s record as set forth in

the liability opinion--and inadequate implementation of

remedial orders in this case--shows that Alabama’s prison

officials are unable to change their system without

monitoring. While the court is encouraged by

Commissioner Dunn’s own admission that monitoring is

needed, see Dunn Nov. 26, 2018, Trial Tr. (doc. no. 2250)

at 16 (“The Court desires to see certain outcomes in the

department. I desire to see certain outcomes in the

department. And there's a way in which we can create a

monitoring structure, if you will, that will enable us

to do that.”), it shares the plaintiffs’ concerns about

allowing the defendants to implement their plan without

a court order; “[m]id-litigation assurances are all too

easy to make and all too hard to enforce, which probably

explains why the Supreme Court has refused to accept

them.” W. Alabama Women's Ctr. v. Williamson, 900 F.3d

1310, 1328 (11th Cir. 2018) (affirming an injunction

despite a non-binding clarification from the State),

cert. denied sub nom. Harris v. W. Alabama Women's Ctr.,

––– U.S. ––––, 139 S. Ct. 2606 (2019); see also Stenberg

v. Carhart, 530 U.S. 914, 940 (2000) (cautioning against

accepting an Attorney General's non-binding

interpretation of a state law).

The court agrees with the defendants that it is

critical for ADOC to “buy in” to the process of attaining

compliance with the court’s remedial orders. However,

this does not require that the defendants be given

unfettered discretion to drive the monitoring

process--buy-in can be achieved by implementing much of

the defendants’ plan as proposed, involving ADOC staff

in monitoring efforts from the beginning, and appointing

monitors in which all parties have faith.2 These

priorities are reflected in the monitoring plan described

below.

But before turning to these specific areas of

dispute, the court must turn to the question whether the

court’s monitoring order is governed by the PLRA. This

statute provides that a “court shall not grant or approve

any prospective relief unless the court finds that such

2. Significantly, the court’s order today does not

involve a receivership and does not infringe on the

autonomy of ADOC any more than the Constitution requires.

Ultimately, it is ADOC that is responsible for carrying

out its obligations under the Constitution, and the

court’s order does not shift that responsibility.

relief is narrowly drawn, extends no further than

necessary to correct the violation of the Federal right,

and is the least intrusive means necessary to correct the

violation of the Federal right.”

18 U.S.C. § 3626(a)(1)(A). The plaintiffs say that, with

an exception or two, the PLRA does not apply, while the

defendants say that it applies fully. As explained later

in this opinion, however, the court need not resolve this

issue, for, whether the PLRA is applicable or not, the

court finds that the monitoring scheme it has fashioned

fully complies with the statute.

In discussing why each part of the court’s order

satisfies the PLRA's need-narrowness-intrusiveness

requirement, 18 U.S.C. § 3626(a)(1)(A), the court in its

analysis will largely focus on a single inquiry: Is the

monitoring provision necessary to correct the

constitutional violation found? As the court has

discussed in prior opinions, this single inquiry is

distilled from the three requirements of the PLRA--that

the relief is (1) “narrowly drawn,” (2) “extends no

further than necessary,” and (3) “is the least intrusive

means necessary to correct the violation,” 18 U.S.C.

§ 3626(a)(1)(A)--and allows for a more streamlined

analysis. See Braggs v. Dunn, 383 F. Supp. 3d 1218, 1252

(M.D. Ala. 2019) (Thompson, J.). “[I]f the ordered

relief is necessary to correct the violation, then--by

definition--no other form of relief would be sufficient

to correct it. And if no other form of relief is

sufficient [to] correct the violation, then the ordered

relief is--by definition--‘narrowly drawn’ and the ‘least

intrusive means necessary’ to correct it; any narrower

or less intrusive relief would not be sufficient.” Id.

at 1252 n.19 (citing 18 U.S.C. § 3626(a)(1)(A)).

Similarly, if the ordered relief is necessary, then it

‘extends no further than necessary,’ because any part of

the relief extending further than what is necessary would

render it unnecessary.” Id.

A. External Monitoring Team

The parties agree that the monitoring scheme should

involve, from the outset, an EMT, but disagree about the

composition of the EMT, the process for selecting its

members, the status of its members, and the rules

regarding conflicts of interest.

1. Composition

a. Dispute

Both the plaintiffs and the defendants propose that

the EMT consist of at least one psychiatrist, one

psychologist, and one nurse to represent the professional

perspectives of the members of ADOC’s treatment teams.

The plaintiffs also assert that the EMT should have an

expert in correctional administration (whom they call a

“security monitor”); a counselor or social worker, as

would also be included in treatment teams; and a data

monitor with expertise in collection and processing. The

defendants contend that a correctional administration

expert is unnecessary because the court already receives

quarterly correctional staffing reports. See Defs.’

Response (doc. no. 2295) at 38-39.

b. The Court’s Resolution

The court will adopt the parties’ common proposal

that the EMT must include at least one psychiatrist, one

psychologist, and one nurse. The court will also require

the inclusion of a correctional administration expert,

in accordance with the opinions of both of the

plaintiffs’ experts.

As plaintiffs’ correctional expert Eldon Vail

credibly testified, improving ADOC’s mental-health care

requires the cooperation and compliance of not just

mental-health staff, but of correctional staff as well.

See Vail Nov. 29, 2018, Trial Tr. (doc. no. 2340) at 7.

For example, Vail testified that: “Access to treatment

is dependent oftentimes on the ability of correctional

staff to escort people, make sure that they get there.

Once they get there, there's issues of confidentiality

that need to be understood by the correctional staff and

honored in the context of them doing their security job.”

Id. at 8. Correctional officers also play an important

role in ensuring that self-referrals by inmates are

communicated to mental-health staff. See id.

A correctional administration expert is also

necessary to evaluate, for example, whether “extenuating

circumstances” exist to require the placement of an

inmate with a serious mental illness (SMI) in

segregation--the only scenario in which such placement

is permissible. See Braggs v. Dunn, 257 F. Supp. 3d

1171, 1245-46 (M.D. Ala. 2017) (Thompson, J.). The

judgments of correctional staff regarding whether

circumstances are “extenuating” cannot be fully evaluated

by mental-health experts on the EMT, for they may lack

expertise in correctional management and would be unable

to ascertain whether such security-related judgments are

reasonable. By contrast, as Vail testified, a

corrections expert would be able to ask staff: “What is

exceptional about this situation, and were there other

ways that you could handle it? And by the way, here’s a

way that you might have been able to handle it. Would

that have worked?” Vail Nov. 29, 2018, Trial Tr. (doc.

no. 2340) at 23. Commissioner Dunn also agreed that the

remedial orders contain security-related matters that

require the participation of correctional staff. See

Dunn Nov. 26, 2018, Trial Tr. (doc. no. 2250) at 201-202.

Because of the intricate involvement of correctional

staff in provision of care, some security-related

matters, which are part of the court’s remedial orders

must be monitored as part and parcel of mental-health

monitoring. The need for a corrections expert on the EMT

clearly follows.

Plaintiffs’ experts Vail and Dr. Burns both testified

that it is particularly necessary to include a

corrections expert on the EMT in light of the ongoing

challenges posed by correctional understaffing at ADOC.

In the liability opinion, the court found understaffing

problematic not in isolation, but as an issue that

“permeate[s] each of the ... identified contributing

factors of inadequate mental-health care.” Braggs, 257

F. Supp. 3d at 1268. While the quarterly staffing reports

submitted to the court are critical to understanding

ADOC’s progress on the understaffing issue, the reports

alone do not provide any information as to whether

custody staff are supporting and implementing the

required improvements to mental-health care; in other

words, they do not reflect the critical impact

understaffing (which impact could be significantly and

adversely changing if there is a significant adverse

change in overcrowding) might be having on the delivery

of mental-health care. A correctional administration

expert is necessary for the EMT to be able to evaluate

whether, based on the availability of officers in

overcrowded conditions, correctional staff are being

deployed properly to both “keep[] people safe” and

“mak[e] sure that prisoners get mental health

treatment”--a compliance issue that cannot be ascertained

by simply reviewing the total number of staff reported.

Vail Nov. 29, 2018, Trial Tr. (doc. no. 2340) at 18; see

also Burns Dec. 6, 2018, Trial Tr. (doc. no. 2254) at

112-13.

At the same time, the court will not initially

require the inclusion of the other members proposed by

the plaintiffs on the EMT--a counselor or social worker,

and a data expert--but will instead leave it to the EMT,

which would be more knowledgeable than the court about

such matters, to determine whether additional members or

support staff are necessary. If it so determines, the

EMT may ask the court to appoint additional members,

either on an ad hoc or permanent basis. Accordingly, the

court adds only one additional member to the defendants’

proposed monitoring team structure.

The composition of the EMT, largely proposed by the

defendants, meets the PLRA's

need-narrowness-intrusiveness requirement. See 18

U.S.C. § 3626(a)(1)(A). As described, the addition of a

correctional administration expert is necessary to ensure

that officers and clinical staff alike are complying with

the court’s remedial orders. Particularly in light of

ADOC’s continued difficulties with understaffing, the

court must be sure that ADOC is doing all that it can to

comply with the court’s orders by properly training and

deploying the limited staff that it has.

The court believes there should also be a head of

the EMT with the administrative abilities to coordinate

monitoring efforts and to serve as a liaison with the

court. Before deciding who this person should be, the

court will solicit input from the parties. This

additional, administrative provision meets the PLRA's

need-narrowness-intrusiveness requirement because the

team requires a leader to manage its operations

effectively and to coordinate with the court. See 18

U.S.C. § 3626(a)(1)(A).

2. Selection Process

a. Dispute

While the defendants contend that the EMT members

should be selected from a pool of candidates that was

exclusively proposed by them at the monitoring hearing,

the plaintiffs argue that there should be a collaborative

selection process between the plaintiffs and the

defendants. The defendants propose two alternative teams

of external monitors: one led by individuals previously

affiliated with Corizon Correctional Healthcare, the

former medical care provider for ADOC, and one comprised

of individuals employed by PSIMED, Inc., a company that

subcontracts with Wexford Health Sources, Inc. to provide

mental-health care in West Virginia prisons. See Pls.’

Proposed Opinion (doc. no. 2260) at 13. The defendants

further assert that members of the EMT must be licensed

in Alabama if they are to make recommendations based on

clinical judgment that would require a license. Defs.’

Proposed Monitoring Plan (doc. no. 2115) at 28. They

contend that the plaintiffs essentially waived the

opportunity to participate in the selection of the EMT

members by failing to propose candidates during the

remedial hearing. See Defs.’ Response (doc. no. 2295)

at 27-28.

By contrast, the plaintiffs argue that there should

be a collaborative selection process between the

plaintiffs and the defendants, under which the parties

first attempt to reach an agreement on the EMT members,

and then, if they cannot, submit proposed candidates to

the court for selection. The EMT members, the plaintiffs

contend, should not be limited to those licensed in

Alabama.

b. The Court’s Resolution

The court will adopt the plaintiffs’ proposal for a

joint selection process, with the condition that the

court will have ultimate authority to approve the experts

agreed upon by the parties. The court will not limit the

pool of candidates for the EMT to those licensed in

Alabama.

The court finds it both unwise and unfair for the

plaintiffs to have no role in selecting the EMT members;

joint selection is necessary to ensure impartiality of

the EMT and trust by all parties and the court in their

monitoring work. The court believes that such

collaboration is necessary to ensure that both parties,

not just the defendants, buy into the process. The court

was also not impressed by the monitors proposed by the

defendants, some of whom had no court monitoring

experience and some of whom presented serious questions

about possible conflicts of interest.3 Meanwhile, the

court finds it unnecessarily restrictive to require that

the experts be licensed in Alabama, though, as described

in the section below on qualitative review, the court

3. At the time of the monitoring trial, PSIMED was

employed in the West Virginia Department of Corrections

as a subcontractor of Wexford Health Sources, Inc.,

ADOC’s current provider of mental-health care. This

business relationship between the proposed EMT members

and ADOC’s vendor could create, at a minimum, a perceived

conflict of interest, which the defendants themselves

sought to avoid under their proposed monitoring plan.

See Defs.’ Proposed Monitoring Plan (doc. no. 2115) at

24. Even more concerningly, the owner of PSIMED was also

involved in a corruption charge brought by the West

Virginia Ethics Commission against the former

Commissioner of the West Virginia Department of

Corrections. See Stern Nov. 27, 2018, Trial Tr. (doc.

no. 2415) at 68-73. Other individuals proposed by the

defendants had previously worked for ADOC’s prior

healthcare contractor, Corizon Correctional Healthcare,

another possible conflict of interest.

will authorize the EMT to evaluate clinical judgments.

Such an eligibility requirement could unnecessarily limit

the pool of available candidates. Finally, members

should be chosen based on their background and

expertise--with a preference for those with monitoring

experience--which should make clear their qualification

for the role.

The court will give the parties a reasonable period

of time to reach an agreement about the composition of

the EMT. If the parties are unable to agree on all EMT

members, the court will select any missing members of the

EMT or will come up with a means for selection. This

process will also extend to the replacement of an EMT

member who is removed or resigns. Removal of an EMT

member for good cause may be proposed by either party and

will be subject to the court’s approval.

The requirements for selection and qualifications of

the EMT meet the PLRA's need-narrowness-intrusiveness

requirement. See 18 U.S.C. § 3626(a)(1)(A). The

cornerstone of the defendants’ proposed monitoring plan,

which the court will largely adopt, is that the EMT will

be responsible for filling in key details based on its

members’ expertise. The choice of experts, therefore,

is a critical issue to ensure the success of the

monitoring phase of this case, and thus ADOC’s ability

to provide constitutionally adequate mental-health care

in the long-term. Still, the court’s resolution again

gives the defendants an opportunity to suggest potential

experts, who will be appointed if they are satisfactory

to both the plaintiffs and the court. Under this

collaborative process, the defendants can work to ensure

that staff members will not “come to resent” each expert

chosen “as an outside critic who is not invested in the

department’s mission.” See Defs.’ Response (doc. no.

2295) at 7.

3. Status of EMT Members

a. Dispute

The defendants propose that the EMT members serve as

independent contractors of ADOC but that ADOC shall not

“possess any supervisory authority over the [EMT’s]

activities, reports, findings, or recommendations.”

Defs.’ Proposed Monitoring Plan (doc. no. 2115) at 11.

Despite this, the defendants assert that ADOC should have

the power to negotiate each EMT member’s fees, expenses,

and budget for monitoring. See id. at 10. By contrast,

the plaintiffs contend that, for the monitors to remain

“neutral and independent, they should be court-appointed

under Federal Rule of Evidence 706 and paid by through

the court’s registry rather than directly by ADOC.”

Pls.’ Proposed Opinion (doc. no. 2260) at 23.

b. The Court’s Resolution

The court will not order the EMT to be paid by the

defendants through the court’s registry pursuant to

Federal Rule of Evidence 706. The court believes that

the most efficient and least intrusive arrangement is for

ADOC to pay the EMT directly, rather than through the

court, which would be complex and likely more expensive.

Therefore, the court will allow the defendants to

negotiate the fees, expenses, and budgets of the EMT,

subject to the court’s approval in the event a dispute

arises. (The court has found, from past experience, that

the State has been fair in such negotiations, and hopes

that it will continue to be the same here. See, e.g.,

Consent Decree (doc. no. 11) at 114, United States v.

Alabama, No. 2:15cv368-MHT (M.D. Ala. June 18, 2015)

(Thompson, J.)). And though ADOC will pay the EMT

directly, EMT members should be introduced to ADOC

employees and any other parties as neutral and totally

independent. The EMT members will not be under ADOC’s

supervision, but under the direction of the court. (The

court finds, again from past experience, that monitors

of state institutions can be fair and neutral despite

being paid by the state. See, e.g., id.) The court

finds this resolution satisfies the PLRA’s

need-narrowness-intrusiveness requirement, as it is

substantially in line with the defendants’ proposal. See

18 U.S.C. § 3626(a)(1)(A).

4. Conflict of Interest Rules

a. Dispute

The defendants’ proposal includes a restriction that

EMT members may not “testify in any other civil action

or proceeding concerning or relating to any act or

omission of ADOC or its employees, contractors, or

agents, or testify regarding any subject or matter that

any [EMT] member learned, or might have learned, as a

result of his or her performance as a member of the [EMT],

or serve as a non-testifying expert regarding any subject

or matter that any [EMT] member learned, or might have

learned, as a result of his or her performance as a member

of the [EMT].” Defs.’ Proposed Monitoring Plan (doc. no.

2115) at 29. The defendants contend that such a rule is

necessary to ensure ADOC employees and contractors feel

the EMT members are “on their side” and will not be

gathering evidence against them. Defs.’ Response (doc.

no. 2295) at 36. The plaintiffs assert that this strict

limitation would deter otherwise willing potential

experts and that instead, possible conflicts of interests

of EMT members should be reviewed on a case-by-case

basis.

b. The Court’s Resolution

The court agrees with the defendants that to ensure

cooperation between the EMT and those implementing the

remedial measures, some limitation on participation in

future actions against the State and its contractors is

necessary. However, the rules proposed by the defendants

are overly restrictive, and there is a serious question

as to whether they are even enforceable by this court--in

particular, with regard to limitations on participation

in another court proceeding, state or federal. The court

will restrict EMT members from providing paid testimony

or serving as paid non-testifying experts in any action

against ADOC or its employees or against an ADOC

contractor in an action the subject matter of which

pertains to ADOC facilities specifically. As many prison

healthcare providers are active in multiple states, the

court will not issue a blanket ban on EMT members

testifying or serving as experts in actions involving

ADOC contractors outside of Alabama. The parties may

raise concerns about conflicts of interest in such cases

on a case-by-case basis.

The defendants have not provided any legal basis to

restrict or prevent the EMT experts from otherwise

testifying in other cases if subpoenaed to do so. The

court will therefore not impose that limitation.

The court finds the adoption of some restrictions on

the ability of EMT members to testify, largely proposed

by the defendants, meets the

need-narrowness-intrusiveness requirement. See 18

U.S.C. § 3626(a)(1)(A). The court’s resolution loosens

the defendants’ proposed restriction only to the extent

necessary to ensure that the parties can jointly recruit

qualified experts who may otherwise be deterred by their

inability to find other work if they accept appointment

in this case.

B. Measuring Compliance

1. Establishing Performance Measures

a. Dispute

The parties agree that the monitoring scheme should

empower the EMT members, who are the experts here, to

determine many of the details of how to carry out

monitoring, including determining what performance

measures to use. Performance measures are the metrics

by which the monitors are to evaluate whether the

defendants are complying with the court’s remedial

orders. See Defs.’ Proposed Monitoring Plan (doc. no.

2115) at 2-3. In their initial proposed monitoring plan,

the defendants proposed 259 potential performance

measures to evaluate compliance with then-existing

remedial orders. Id. For example, proposed performance

measure number 37 states that “[i]ntake RN [registered

nurses] shall conduct intake MH [mental-health] screening

in a confidential location,” Defs.’ Proposed Monitoring

Plan (doc. no. 2115-1) at 3, and is meant to measure

compliance with one of the court’s orders regarding

identification of inmates in need of mental-health

treatment. See Identification Order (doc. no. 1794-1).

Despite having initially proposed these performance

measures, it is clear from testimony that Commissioner

Dunn and Associate Commissioner Naglich understand the

defendants’ proposal to allow for the EMT to modify the

259 measures proposed as it deems appropriate. See Dunn

Nov. 26, 2018, Trial Tr. (doc. no. 2250) at 148

(testifying that if the EMT determines the performance

measures fail to adequately address the remedial orders,

the EMT members “have the ability to create measures that

do”); Naglich Nov. 7, 2018, Trial Tr. (doc. no. 2249) at

180-82 (testifying that the EMT should be allowed to

develop and change performance measures and that there

is nothing wrong with the EMT coming up with the

performance measures in the first place, “as long as

they’re reflective of” the remedial orders).

This point of agreement is emblematic of the overall

theme of the defendants’ monitoring plan, in which they

acknowledge that the EMT “must drive the process of

filling in the open components” of the plan. Defs.’

Response (doc. no. 2295) at 14. The defendants

“intentionally left open certain aspects” of their

proposed plan so that the EMT could “guide ADOC in filling

in these remaining details.” Id. The defendants seek

to empower the EMT to fill out the details of the plan

because of the EMT’s “expertise.” Id. at 15. The

plaintiffs agree it is appropriate to give the EMT

significant authority to determine how to conduct

monitoring, including as to the performance measures

used.

b. The Court’s Resolution

The court will adopt the defendants’ plan to give

the EMT authority to modify the 259 initially proposed

performance measures, including by removing them,

changing their language, or creating entirely new

performance measures. The EMT, in exercise of its unique

expertise, recognized by the defendants, is to create the

performance measures necessary to evaluate the

defendants’ compliance with the court’s remedial orders.

Once the performance measures are established but before

monitoring begins, the parties will be given an

opportunity to raise objections to any of the proposed

measures through a standard dispute resolution process:

the objection must be raised first with the EMT, then via

mediation, and finally with the court if still

unresolved.

The court finds that allowing the experts, rather

than the court, to establish the performance measures

necessary to monitor compliance meets the

need-narrowness-intrusiveness requirement. See 18

U.S.C. § 3626(a)(1)(A). Because the performance measures

will be limited to the court’s remedial orders, which

must also satisfy the PLRA, they will be narrowly

tailored to evaluate only ADOC’s progress with regard to

the remedies ordered in this case.

2. Self-Correction

a. Dispute

The plaintiffs seek a provision requiring ADOC to

create corrective action plans when monitoring reveals

noncompliance. Pls.’ Response (doc. no. 2133) at 75-78.

The plaintiffs concede that the PLRA applies to this

proposed order because it does not have the goal of

informing the court of ongoing violations but is intended

to improve ADOC’s internal capacity to comply with all

remedial orders. Id. at 54. They assert, however, that

creation and auditing of corrective action plans is

absolutely necessary to ensure compliance. Id. The

plaintiffs also urge the court to require ADOC to

designate a “Compliance Coordinator” to oversee the

creation and implementation of corrective action plans

and other monitoring activities, and to designate

additional clinical staff who are dedicated solely to

internal monitoring. Id. at 52-53, 60-61.

The defendants state that they expect corrective

action planning to be part of ADOC’s response to findings

of noncompliance. See Defs.’ Response (doc. no. 2295)

at 38 n.4. However, they assert that, “[w]hile

corrective action is constitutionally required in the

face of a constitutional violation, a specific corrective

action plan is not.” Id. at 38. They argue that a

requirement that ADOC create corrective action plans

would therefore not meet the

need-narrowness-intrusiveness requirement of the PLRA.

See 18 U.S.C. § 3626(a)(1)(A).

b. The Court’s Resolution

The court will not order ADOC to create corrective

action plans, though the court agrees with both parties

that doing so will likely be a necessary step toward

achieving compliance. In the event of finding a

deficiency, the EMT should provide instruction to ADOC

as to what corrective action should be taken, suggest how

to plan for that action, and monitor whether that action

has been taken. Nonetheless the court will accept, for

now, Commissioner Dunn’s assertion that “common sense”

will lead ADOC to self-correct when presented with

evidence of its noncompliance. Dunn Nov. 26, 2018, Trial

Tr. (doc. no. 2250) at 153-54. Common sense will also

dictate that the court will be very concerned to learn

through the monitoring process that ADOC has failed,

absent a valid reason, to take corrective actions

suggested by the EMT. As the defendants have repeatedly

acknowledged, “self-identification and self-correction

of systemwide problems is the best long-term solution.”

Defs.’ Proposed Monitoring Plan (doc. no. 2115) at 6.

Plaintiffs’ expert Dr. Burns credibly testified that

self-correction “is kind of the definition of continuous

quality improvement: That you self-monitor, identify

deficiencies, study the problem, prepare a solution,

apply the solution, restudy the issue to see that it has

been resolved.” Burns Dec. 6, 2018, Trial Tr. (doc. no.

2254) at 89. The capacity to self-monitor necessarily

includes an understanding of “how a system needs to

change in response to internal experience.” Id. at 90.

The court will not be convinced that external monitoring

is no longer needed until ADOC has demonstrated that it

has the capacity to engage in both of these twin elements

of self-monitoring. The experts on the EMT are best

positioned to help ADOC develop this capacity and may do

so through the “feedback” and “consultation” processes

described below. The court finds that this resolution

is narrowly tailored and meets the

need-narrowness-intrusiveness requirement of the PLRA.

See 18 U.S.C. § 3626(a)(1)(A).

The court will also not order ADOC to designate a

“Compliance Coordinator” or otherwise dictate how ADOC

organizes and manages its internal monitoring staff. For

now, the court will defer to ADOC to establish an IMT as

it deems appropriate. The defendants have informed the

court that they plan to appoint to the IMT one

psychiatrist, one psychologist, two registered nurses,

and any other staff deemed necessary to complete

quarterly reports. See Defs.’ Proposed Monitoring Plan

(doc. no. 2115) at 12. While the court understands the

plaintiffs’ concern that appointing members of the IMT

who already have clinical duties may mean they are

stretched too thin, the provisions plaintiffs seek are

overly intrusive. The issue of whether the IMT

established by ADOC has sufficient expertise and capacity

to monitor compliance will be properly before the court

when the court considers whether to terminate external

monitoring, as described in the “Handoff” section of this

opinion.

3. Qualitative Review

a. Dispute

Though initially a point of concern for the

plaintiffs, it now appears the parties agree that in

assessing whether ADOC is complying with the performance

measures, and thus remedial orders, the EMT can make

“qualitative assessments relating to clinical

decision-making.” Defs.’ Response (doc. no. 2295) at 30;

see also Pls.’ Proposed Opinion (doc. no. 2260) at 26.

The defendants urge that such assessments may not

constitute second-guessing of the reasonable judgment of

clinicians, but acknowledge, based on the testimony of

Dr. Robert Stern, one of their own proposed external

monitors, that assessment of many performance measures

will “require[] a clinical perspective.” Stern Nov. 27,

2018, Trial Tr. (doc. no. 2415) at 116. For example, Dr.

Stern testified that the EMT should be able to determine

whether a counseling progress note contains merely

“gibberish,” rather than clinical observations, or

whether a referral decision is within the range of

appropriate actions. Id. at 150. According to Dr. Stern,

there would be “something wrong” if a monitoring plan did

not allow for this type of qualitative review. Id. at

123; see also Burns Dec. 6, 2018, Trial Tr. (doc. no.

2254) at 92-93 (testifying that that monitors must be

able to assess clinical judgment).

b. The Court’s Resolution

Based on the parties’ agreement, the EMT will have

authority to make qualitative assessments of whether

clinicians are making reasonable clinical judgments.

This assessment will often entail evaluating whether the

clinicians based their judgment on clinically relevant

information, as well as evaluating the documented reasons

for making the judgment. As described by plaintiffs’

expert Dr. Burns, the EMT’s role is not to second-guess

clinicians' clinical judgments, but to see “if there’s a

rationale involved for the choices that are made and the

interventions that are provided.” Burns Dec. 6, 2018,

Trial Tr. (doc. no. 2254) at 93. As such, the EMT cannot

find that the defendants did not satisfy a performance

measure because a clinician made a decision that was

reasonable, but different from the decision the EMT would

have made.

The court finds this resolution satisfies the

need-narrowness-intrusiveness requirement of the PLRA.

See 18 U.S.C. § 3626(a)(1)(A). Reviewing clinical

judgment is necessary to assess compliance with remedial

orders because practitioners cannot comply with remedial

orders while engaging in clinically unreasonable

judgment. For example, if clinicians systematically fail

to diagnose people with SMIs despite clear indicators,

ADOC will be unable to attain compliance with the court’s

remedial orders regarding identification of people in

need of mental-health care. The parties’ agreement

regarding qualitative evaluations and the testimony of

witnesses from both sides support the court’s finding

that this resolution satisfies the PLRA. See 18 U.S.C.

§ 3626(a)(1)(A); see also Braggs v. Dunn, 383 F. Supp.

3d 1218, 1253 (M.D. Ala. 2019) (Thompson, J.) (finding

the defendants’ agreement to a remedial provision is

strong evidence of PLRA compliance).

4. Audit Tools

a. Dispute

In order to perform qualitative review, the parties

agree that the EMT shall have the authority to develop

“audit tools.” See Pls.’ Proposed Opinion (doc. no.

2260) at 27; Defs.’ Response (doc. no. 2295) at 17-18.

While the term “audit tool” has not been precisely

defined by either party, it essentially refers to the

method or procedure by which the EMT members assess

compliance with the performance measures. See Defs.’

Proposed Monitoring Plan (doc. no. 2115) at 15-16; Burns

Dec. 7, 2018, Trial Tr. (doc. no. 2256) at 242-43. For

example, the defendants’ proposed performance measure 45

is that “any referral by the intake [registered nurse]

must be designated as emergent, urgent, or routine.”

Defs.’ Proposed Monitoring Plan (doc. no. 2115-1) at 4.

The proposed audit tool for that performance measure--

referred to as a “Method of Measurement” in the

defendants’ performance measures chart--is to “review a

minimum of 10 charts for inmates referred for evaluation

by the intake RN [registered nurse] during the applicable

quarter for each identified facility.” Id.

Just as both parties propose empowering the EMT to

develop performance measures, they also propose

empowering the EMT to develop audit tools for evaluating

compliance with their performance measures. It is not

clear, however, whether the defendants agree that the

EMT’s authority to create the audit tools should be

complete or whether their proposal would require final

approval from ADOC.

Overall, the defendants’ plan also limits the EMT to

evaluating no more than three facilities per quarter.

This would seem to mean that only 12 of the 14 major

facilities would be evaluated even once in a year.

b. The Court’s Resolution

The court will order that the EMT will have the

authority and discretion to develop and adjust audit

tools as it deems appropriate to evaluate compliance with

the performance measures and ADOC’s developing ability

to self-correct. Since the audit tools are another

important element of the monitoring plan to be filled in,

the defendants acknowledge that the EMT should devise

them, in light of their expertise. See Defs.’ Proposed

Monitoring Plan (doc. no. 2115) at 18. As with the

performance measures, the audit tools will be limited to

assessing ADOC’s progress as to the court’s limited

remedial orders and its capacity to self-monitor. And,

as with the performance measures, the parties will be

given an opportunity to raise objections to any of the

proposed audit tools before monitoring begins, first with

the EMT, then via mediation, and finally with the court

if still unresolved.

The court will also leave it to the experts to decide

how many and which facilities to evaluate on a quarterly

basis within the constraints of the rest of today’s

order. It is in the interest of all parties and the

court for ADOC to achieve compliance and attain the

capacity to self-monitor as swiftly and effectively as

possible. Though seemingly aimed to minimize the

intrusion to ADOC facilities, arbitrarily limiting the

number of facilities to be evaluated each quarter could

have the opposite effect, slowing ADOC’s progress toward

termination of the court’s remedial orders and extending

the duration of external monitoring.

The court finds that this resolution satisfies the

PLRA's need-narrowness-intrusiveness requirement. See

18 U.S.C. § 3626(a)(1)(A). Allowing the EMT to create

the tools and schedule it deems necessary to evaluate

compliance is essential to ensure the EMT’s expertise

shapes and drives the overall monitoring scheme.

5. Contempt and Dispute Resolution

a. Dispute

The defendants’ plan includes a significant

limitation on the ability of the plaintiffs to initiate

contempt proceedings, allowing for the plaintiffs to do

so only where the monitoring team has issued three

consecutive quarterly reports finding noncompliance with

respect to the same performance measurement and the

plaintiffs have exhausted “all reasonable efforts to

resolve any sustained noncompliance through the dispute

resolution process.” Defs.’ Proposed Monitoring Plan

(doc. no. 2115) at 21. Under the defendants’ plan, the

plaintiffs must first raise concerns about noncompliance

with the EMT and the defendants, then engage in mediation

to resolve any disputes prior to filing a contempt

motion. The plaintiffs assert that such a limitation

unduly limits their ability to timely raise concerns

about noncompliance with remedial orders, including those

that are a matter of life or death.

b. The Court’s Resolution

The court will not adopt in full the defendants’

proposed limitation on the plaintiffs’ ability to

initiate contempt or other proceedings, which the court

finds an unwarranted restriction. The court understands

the plaintiffs’ concern that three quarters is simply too

long for some forms of noncompliance to continue

unremedied. For example, because the risk of suicide is

“so severe and imminent,” see Braggs v. Dunn, 383 F.

Supp. 3d 1218, 1227 (M.D. Ala. 2019), the remedial

measures on suicide prevention must be complied with

100 % of the time, see Burns Dec. 7, 2018, Trial Tr.

(doc. no. 2256) at 122-23.

However, following the court’s prior practice in this

case, see, e.g., Order (doc. no. 1926) (directing parties

to mediate before the defendants would be ordered to

respond to the plaintiffs’ contempt motion), the court

will order that, absent an extraordinary urgency, the

plaintiffs must initially mediate their concerns before

initiating contempt proceedings. This restriction

balances the defendants’ interest in having the

opportunity to correct noncompliance without litigation

with the plaintiffs’ concern about the need to ensure

complete and immediate compliance with some remedial

orders. The court finds this provision satisfies the

need-narrowness-intrusiveness requirement of the PLRA.

See 18 U.S.C. § 3626(a)(1)(A).

Logically, even in the absence of this restriction,

it would be more difficult for the plaintiffs to succeed

on a non-urgent contempt motion brought prematurely in

the monitoring process. And the court expects that other

elements of this order will obviate the need perceived

by the plaintiffs to initiate contempt proceedings early

on. By ensuring the plaintiffs have a role in selecting

the EMT members, who will fill in many of the details

within the monitoring structure, the plaintiffs will have

greater trust in the monitoring plan to address ADOC’s

deficiencies. Meanwhile, improvements to ADOC’s efforts

and ability to self-correct will help to assure the

plaintiffs that ADOC is taking meaningful steps toward

compliance.

C. Components of Monitoring

The parties are in agreement as to the overall

structure and phases of monitoring, and the court will

order that the monitoring scheme comport with this

agreement. This includes that the monitoring scheme will

consist of three phases, with an external monitoring team

teaching--and then ultimately handing the reins over

to--an internal monitoring team; that the IMT should

consist of professionals with clinical expertise who are

exclusively chosen and employed by ADOC; and that the

monitoring scheme should consist of five essential

components of monitoring outlined above and described in

detail below.

The parties disagree, however, about the timeline

for the phases of monitoring, as well as various

components within this broader structure. The court will

describe and resolve each disagreement in turn.

1. Document Review

a. Dispute

The parties agree that ADOC must produce

documentation to the EMT, and that the EMT shall have

ultimate authority to identify which documents to review.

However, the parties disagree as to the sample size that

must be reviewed. The defendants propose requiring a

minimum sample size of documents to be reviewed for each

performance measure (generally 10 documents), while the

plaintiffs contend that the sample size should be left

to the EMT’s discretion. Once the responsibility for

monitoring has been transferred to the IMT, the

plaintiffs propose that the IMT cannot reduce or change

the amount of documentation identified for review by the

EMT unless ADOC has been found in compliance with respect

to a particular order.

The plaintiffs also propose that documents produced

to the EMT or IMT be produced to plaintiffs’ counsel,

while the defendants argue that the plaintiffs should

receive only “all of the documentation that the

Compliance Teams actually rely upon in evaluating

compliance.” Defs.’ Response (doc. no. 2295) at 29.

Finally, the plaintiffs propose a list of types of

documents to serve as a starting point for review, in

addition to any others requested by the EMT. See Pls.’

Proposed Opinion (doc. no. 2260) at 67.

b. The Court’s Resolution

With regard to document review, the court will again

leave much to be decided by the experts. The parties

appear to agree that the EMT may review an unlimited

number of documents. Rather than set a minimum number

of documents, as the defendants propose, the court will

leave it to the EMT to decide both how many and which

documents to review. While the court acknowledges the

defendants’ concern that a review of too few documents

may result in an unrepresentative and unfair sample, this

is an issue that the qualified experts on the EMT will

be capable of recognizing and addressing as appropriate.

The court will not adopt the plaintiffs’ proposed

requirement that the IMT follow precisely the choices for

document review made by the EMT. The issue of how to

review an adequate number or scope of documents is one

on which the court expects the EMT will train the IMT,

as discussed below. Evidence that the IMT is failing to

follow that training may support a finding that external

monitoring continues to be necessary.

The court declines to adopt the plaintiffs’ proposal

that all documents produced to the EMT or IMT also be

produced to plaintiffs’ counsel. The court is unsure

whether, in practice, there will be a meaningful

difference between the parties’ proposals on this point.

Therefore, though willing to revisit the issue if this

resolution proves unworkable, the court will first adopt

the defendants’ narrower position that plaintiffs’

counsel receive “all of the documentation that the

Compliance Teams actually rely upon in evaluating

compliance.” See Defs.’ Response (doc. no. 2295) at 29.

Because the court has difficulty defining in the abstract

the exact meaning of “actual reliance,” the court will

initially leave it to the defendants to interpret this

language and will simply address any issues as they

arise. In accordance with the defendants’ proposed EMT

member Dr. Stern’s testimony, the EMT must also retain

all documents it reviews in the event its findings are

contested by either party. See Stern Nov. 27, 2018,

Trial Tr. (doc. no. 2415) at 111.

The court finds that this requirement regarding

document review is inherently narrow, as it is adopted

in large part from the defendants’ proposal and the

parties’ agreements and again gives discretion to the

experts to determine what is necessary. Thus, the court

finds it meets the need-narrowness-intrusiveness

requirement of the PLRA. See 18 U.S.C. § 3626(a)(1)(A).

2. Observation

a. Dispute

The parties agree that site visits should be part of

monitoring. In broad strokes, the key difference is

that, while the defendants propose that the EMT conduct

site visits upon its request, the plaintiffs want to

require at least yearly site visits to each facility at

the start of monitoring.

The defendants propose that, within 60 days of the

entry of this order, any EMT member who has not visited

ADOC facilities should visit as many facilities as he or

she deems necessary to become familiar with them.

Subsequently, under the defendants’ plan, EMT members may

request a site visit to any facility, but if any party

disputes such a request or deems it unnecessary or

inappropriate, such request shall be resolved through

mediation.

By contrast, the plaintiffs propose requiring that,

in the first phase of monitoring, each major ADOC

facility be visited by the EMT members at least once, but

no more than twice, per year, unless extraordinary

circumstances require more visits. See Pls.’ Proposed

Opinion (doc. no. 2260) at 41-42. As facilities come

into compliance, the plaintiffs propose, the members may

reduce the frequency of visits. With regard to

scheduling site visits, the plaintiffs propose that the

EMT determine the schedule of visits in consultation with

the defendants, and that prior to each quarter, the EMT

notify the defendants of the facilities to be visited

during that quarter with at least two weeks’ notice

before any site visit.

The plaintiffs also propose limiting each site visit

to no more than five full business days at any facility

with inpatient treatment units or intake processing, and

no more than three full business days at any other

facility, unless explicitly agreed to by the parties or

permitted by the court upon the EMT's request. The

defendants do not have a proposal regarding the duration

of site visits but have repeatedly raised concerns that

site visits may be distracting and intrusive.

The defendants similarly do not have a proposal

regarding the scope of observation of the EMT members

allowed during site visits. The plaintiffs propose that

monitors be allowed to (1) observe processes, including

treatment, with the permission of patients; (2) inspect

areas where inmates are housed and where mental-health

services and programming are performed; (3) conduct

confidential interviews with inmates who receive or

request mental-health services; (4) speak with

correctional and mental-health staff as needed, in a way

that limits as much as possible the impact on the

provision of care; (5) have access to documents they

request either on site or within 10 days after a visit;

and (6) have, to the extent possible, a confidential

setting in which to conduct interviews. Finally, the

plaintiffs propose that each party be allowed to send a

single attorney each day for site visits. The defendants

do not appear to have a position on this issue.

b. The Court’s Resolution

The court will order regular site visits by the EMT

throughout the monitoring process, rather than just at

the outset. However, the court will not dictate the

precise number required and will leave that decision to

the EMT. The court credits the testimony of plaintiffs’

experts Dr. Burns and Vail as well as the defendants’ own

witness Dr. Stern, who all agreed that site visits are

necessary to help understand and verify ADOC’s

documentation, to catch issues that are undetectable

through document review, to foster collaboration between

the EMT and ADOC staff, and to allow for the teaching and

cultural shift required for ADOC to eventually

self-monitor. See Stern Nov. 27, 2018, Trial Tr. (doc.

no. 2415) at 152-53; Vail Nov. 29, 2018, Trial Tr. (doc.

no. 2340) at 21; Burns Dec. 6, 2018, Trial Tr. (doc. no.

2254) at 64.

As proposed by the defendants, it will be up the EMT

to determine the exact frequency of visiting each

facility, including prior to the start of monitoring. In

recognition of the defendants’ concern that overly

frequent site visits may be disruptive to the facilities’

operation, however, the court will limit the number of

site visits by the EMT to no more than two per facility

per year, unless explicitly agreed to by the parties or

permitted by the court upon the monitors’ request. This

requirement is based on the testimony of both Dr. Burns

and Vail, who stated that they would expect the EMT to

conduct two site visits per facility per year, at least

to start, with an emphasis on those facilities “that have

the most involvement in carrying out all of the remedial

plans,” such as the mental-health treatment hubs. Burns

Dec. 6, 2018, Trial Tr. (doc. no. 2254) at 64; see also

Vail Nov. 29, 2018, Trial Tr. (doc. no. 2340) at 25-28.

Because it is up to the EMT which facilities to visit and

how often, it is possible, and perhaps likely, that fewer

than two visits to some facilities will occur in a year.

The court will require the EMT to determine the

schedule for site visits in consultation with the

defendants and to inform the defendants of a proposed

visiting schedule on a quarterly basis, giving no less

than two weeks of notice prior to any visit. The court

finds the defendants’ proposal that ADOC may dispute any

site visit as “unnecessary or inappropriate” overly

broad. Instead, there will be a presumption that a site

visit will occur as deemed appropriate by the EMT--a

presumption that may be overcome based on extraordinary

circumstances, such as an extreme security risk. The

court expects that scheduling disputes will be resolved

between the EMT and the defendants; however, any disputes

that cannot be resolved are to be submitted for mediation

and then to the court if the parties are still unable to

resolve the matter.

Though the court will leave the issue of duration of

site visits to the EMT, the court will do so with the

restriction that site visits may last no more than four

full business days at any facility with inpatient

treatment units or intake processing, and no more than

three full business days at any other facility, unless

explicitly agreed to by the parties or permitted by the

court upon the monitors’ request. The court adopts this

proposal as an assurance to the defendants and based on

testimony by plaintiffs’ expert Dr. Burns that site

visits could be disruptive “if they lasted 30

days ... but the plan would be that you would design this

to last probably three or four days at a given site.”

Burns Dec. 7, 2018, Trial Tr. (doc. no. 2256) at 242.

The court will also adopt the plaintiffs’ proposal

regarding the extent of access by the EMT during the site

visits. The court finds this level of access minimally

required for monitors to ensure a meaningful and accurate

review of ADOC’s compliance. Dr. Burns credibly

testified that site visits provide insight that cannot

be gathered from reviewing the paperwork, particularly

regarding how various processes work together in the

context of each facility. See Burns Dec. 6, 2018, Trial

Tr. (doc. no. 2254) at 57.

Observing processes in person is also useful to

determine whether they are being done “as intended to be

done in terms of all the various components.” Id. at 56.

For example, Dr. Burns testified that compliance with the

remedial order requiring that suicide-watch checks be

made at staggered intervals cannot be fully evaluated

without a site visit. See id. at 165-67. The available

documents regarding this remedial measure are the

observation logs kept by staff. Upon review of these

logs from the months prior to the monitoring trial, Dr.

Burns testified that the logs noted checks at staggered

times, but repeated at precisely the same times over

multiple logs in different months. See id. The

explanation Dr. Burns suggested was that the logs were

likely filled in, in advance. See id. However, without

a site visit “to observe the observers and their logs

contemporaneous with the observation” and to watch the

actual practice of staff, monitors would be left unsure

of when the checks actually happened and whether the logs

are trustworthy. Id. at 167. Plaintiffs’ expert Vail

testified that allowing the monitors to interview ADOC

staff will also increase the monitors’ understanding of

staff actions and decisions before they reach a

conclusion regarding compliance. See Vail Nov. 29, 2018,

Trial Tr. (doc. no. 2340) at 15-16.

Both Vail and Dr. Burns agreed that interviewing

inmates is another important way to detect patterns of

compliance or noncompliance with remedial measures. For

example, Dr. Burns testified that interviewing inmates

is sometimes necessary to investigate issues such as how

quickly patients are being evaluated in response to their

requests for mental-health treatment and whether they are

seen by mental-health staff in a confidential space. See

Burns Dec. 6, 2018, Trial Tr. (doc. no. 2254) at 55.

The court will leave the issue of including attorneys

during site visits up to the EMT. While Dr. Burns

credibly testified that the presence of multiple

attorneys during visits can be disruptive, she also

stated that the presence of counsel during exit

interviews could be appropriate. See id. at 48. The

court believes the EMT members are best positioned to

evaluate when and whether to invite a limited number of

attorneys from both parties to be present.

The court finds that these provisions meet the

need-narrowness-intrusiveness requirement of the PLRA.

See 18 U.S.C. § 3626(a)(1)(A). The requirement regarding

site visits balances the defendants’ concerns about

disruptions caused by in-person monitoring with the need

for monitors to achieve a complete understanding of

ADOC’s operations and its progress toward constitutional

compliance. As plaintiffs’ experts and defendants’ own

witness Dr. Stern testified, some site visits,

particularly initial visits, “are absolutely necessary.”

Stern Nov. 27, 2018, Trial Tr. (doc. no. 2415) at 102;

see also Vail Nov. 29, 2018, Trial Tr. (doc. no. 2340)

at 25 (agreeing that site visits are “necessary to

establish the working relationships with the people in

the system, and it’s also necessary to be able to see and

feel the progress that's occurring or lack of progress”).

However, the court agrees with plaintiffs’ expert Vail

that, “beyond a required ... visit [to] each facility, I

think [the issue of site visits] should be left up to the

judgment of the monitors.” Vail Nov. 29, 2018, Trial Tr.

(doc. no. 2340), at 59.

The court’s relief embraces the defendants’ logic

that site visits should occur as deemed necessary by the

experts. It also limits the intrusion into ADOC’s

operations by capping the duration of each visit and

allowing a total of only two visits per facility per

year, absent the consent of all parties or the court’s

approval. As facilities demonstrate their compliance

with remedial orders, logically the EMT should find that

less monitoring is required and thus the number of site

visits will diminish.

3. Feedback

Citing Dr. Burns, the plaintiffs and defendants both

list “feedback” and “consultation” as two of the five

components of monitoring. Given the similarity between

these two concepts, it is important to define the

difference between them. The court refers to “feedback”

as activity by the monitors to inform the court and

parties of the defendants’ compliance with remedial

orders and the Constitution. As part of this process,

they may also advise defendants on best practices,

policy, and more.4 “Consultation,” by contrast, refers

to actions taken to teach ADOC how to self-monitor.

a. Dispute

In broad strokes, whereas the defendants propose that

feedback essentially be limited to the quarterly reports

produced by the EMT, the plaintiffs contend that feedback

should also include face-to-face meetings.

4. Dr. Burns testified that as part of

“consultation,” the parties can advise the defendants on

policy development, such as suggesting that ADOC look at

model policies or practices from other States or

suggesting certain ADOC facilities consult with one

another. See Burns Dec. 6, 2018, Trial Tr. (doc. no.

2254) at 68-69. However, for purposes of the analysis

here, the court will treat this type of advising as

“feedback,” and “consultation” as exclusively related to

teaching about monitoring. This is because advising on

policies and practices is more closely related to giving

feedback about compliance with orders and the

Constitution than it is with consulting about monitoring.

With regard to the written reports, the parties agree

that quarterly monitoring reports should be submitted to

the court, and that both parties should have the

opportunity to contest the reports or to provide feedback

to the EMT. However, the parties’ proposals include

three minor discrepancies: (1) the plaintiffs propose

that, during the second phase, the IMT, in addition the

EMT, produce quarterly reports to learn how to do so; (2)

the defendants propose that all quarterly monitoring

reports as well as supporting documentation be kept under

seal and treated as confidential; and (3) the defendants

propose that “[n]o Quarterly Evaluation Report, or

finding or recommendation of the [EMT] and/or [IMT]

within a Quarterly Evaluation Report, is admissible

against ADOC, its employees, contractors, or contractor’s

employees in any other civil action or other proceeding.”

Defs.’ Proposed Monitoring Plan (doc. no. 2115) at 20-21.

With regard to the face-to-face feedback, the

plaintiffs propose that for each site visit, there be an

entrance and exit interview between the monitoring team

and the institutional staff as well as any party

representatives that are present.

Finally, the plaintiffs also propose that the EMT be

able to communicate with the plaintiffs and the

defendants (and presumably their representatives),

separately or together, at its discretion, outside of the

context of a site visit. By contrast, the defendants’

proposal provides for regular communication (outside of

quarterly reports) only between the EMT and ADOC and

prohibits the EMT from informing one party “of any

decision, recommendation, or report in advance of the

time it is given to all parties.” Defs.’ Proposed

Monitoring Plan (doc. no. 2115) at 21. Specifically, the

defendants’ proposal provides that the EMT “may convene

conference calls on a monthly or otherwise regular basis

with ADOC to discuss implementation of the Remedial

Orders, to obtain updates, and to address questions or

concerns.” Id.

b. The Court’s Resolution

The court will largely adopt the parties’ agreement

that the EMT must issue its quarterly reports to the

parties within 30 days of the conclusion of each quarter

to provide an opportunity to object or provide feedback

prior to submission to the court. See Defs.’ Response

(doc. no. 2295) at 10. In its reports, the EMT shall

identify the basis for its findings (documentation,

interviews, etc.). As to when the reports are to be

filed, the court will make that decision after it has

input from the EMT.

Once filed, in light of the strong presumption in

favor of public access to judicial records, there will

not be a blanket seal on the reports; rather, the either

party or the EMT may move to seal prior to the EMT’s

filing of the reports on a case-by-case basis. See Braggs

v. Dunn, 382 F. Supp. 3d 1267, 1270 (M.D. Ala. 2019)

(Thompson, J.) (finding that “[t]he public has a

common-law right to inspect and copy judicial records and

documents” (citing Nixon v. Warner Commc’ns, Inc., 435

U.S. 589, 597 (1978) & Newman v. Graddick, 696 F.2d 796,

802-04 (11th Cir. 1983)). As this court has stated,

“Alabamians indisputably have a powerful interest in

overseeing ADOC’s performance.” Id. at 1272. The

defendants’ proposal that the reports be inadmissible in

other proceedings is due to be rejected, as the

defendants do not identify any legal basis for such a

requirement. The court will address in the

“Consultation” section below the parties’ disagreement

regarding the IMT’s participation in producing quarterly

reports.

With regard to the face-to-face feedback, the EMT

has authority, upon its request, to conduct entrance or

exit interviews with ADOC personnel. ADOC shall comply

with these requests, absent extenuating circumstances,

such as an extreme security risk. As described, if the

EMT deems participation of the parties’ lawyers

appropriate during these interviews, the court will

permit such participation. Plaintiffs’ expert Dr. Burns

testified that these in-person meetings are important to

provide immediate feedback to ADOC regarding the

monitor’s findings and for the monitors to make

suggestions of improvements where appropriate. See Burns

Dec. 6, 2018, Trial Tr. (doc. no. 2254) at 66-67. Vail

testified that on-the-ground feedback from ADOC to the

monitors is also important to ensure that the monitors

have an accurate impression of ADOC’s activities and to

correct any misunderstandings prior to the writing of the

monitoring reports. See Vail Nov. 29, 2018, Trial Tr.

(doc. no. 2340) at 15. This two-way feedback, he

testified, helps the monitoring process because “the

people on the ground in any state know the details of

their system better than an outsider ever will, no matter

how good the monitor is.” Id.

Finally, with regard to communication between the

parties and the EMT, the court will allow the EMT to meet

with and otherwise communicate with the parties,

including ADOC staff, at its discretion, separately or

together. Fluid communication will aid the two-way

delivery of feedback. The defendants’ own proposed

monitor Dr. Stern also testified that communication

between the EMT and counsel for all parties is important

during the monitoring process. See Stern Nov. 27, 2018,

Trial Tr. (doc. no. 2415) at 54. However, if the monitors

rely on information from these communications in the

quarterly reports, they must cite the communication in

the report. Consistent with the defendants’ proposal,

the court will order that the EMT may not provide the

quarterly report to one party ahead of the other.

The court finds that these provisions regarding

feedback meet the need-narrowness-intrusiveness

requirement of the PLRA. See 18 U.S.C. § 3626(a)(1)(A).

Once again, the parties’ agreement regarding the filing

of quarterly reports is strong evidence of PLRA

compliance. See Braggs v. Dunn, 383 F. Supp. 3d 1218,

1253 (M.D. Ala. 2019) (Thompson, J.). Allowing the EMT

discretion to request face-to-face feedback when they

believe it is necessary will help to ensure swift and

productive collaboration between the EMT and ADOC--a

critical feature of this monitoring scheme. Such

meetings will be limited to what is necessary, as

determined by the experts on the EMT, to limit the

intrusion into ADOC’s operations.

4. Consultation

a. Dispute

As used here, “consultation” refers to the EMT’s

actions to teach the defendants how to self-monitor.

Consultation differs from “feedback” in that it is not

focused on informing the parties and the court about

compliance. In broad strokes, the parties agree that

teaching should be a big part of the monitoring scheme.

The parties’ disagreement lies in how this teaching and

capacity-building will take place. The distinctions in

the proposed division of labor between the EMT and

IMT--including the EMT’s supervisory role over the

IMT--reflect the parties’ different proposed processes

by which the EMT will teach the IMT how to monitor.

In the first phase, the parties agree that the EMT

will exclusively conduct monitoring. However, the

parties’ proposals differ slightly in that the plaintiffs

propose that the IMT observe the EMT during the first

phase, see Pls.’ Proposed Opinion (doc. no. 2260) at 72,

whereas the defendants simply propose that the EMT make

the results of its quarterly evaluations available to the

IMT, see Defs.’ Proposed Monitoring Plan (doc. no. 2115)

at 13.

In the second phase, the plaintiffs propose that the

IMT observe the EMT and participate only “as

appropriate,” Pls.’ Proposed Opinion (doc. no. 2260) at

72, whereas the defendants propose that the EMT and IMT

jointly conduct the evaluations, see Defs.’ Proposed

Monitoring Plan (doc. no. 2115) at 13-14. Unlike the

defendants, the plaintiffs also propose that the IMT

produce separate draft quarterly evaluations during the

second phase (in addition to the evaluations produced by

the EMT), which the EMT would review for accuracy,

thoroughness, and efficacy. This way, according to

plaintiffs, the EMT “will be able to teach the Internal

Monitoring Team the skills they need to take the lead in

monitoring and will be able to evaluate the progress of

the Internal Monitoring Team in developing the capacity

to take over the monitoring process.” See Pls.’ Proposed

Opinion (doc. no. 2260) at 74. Another crucial

difference, discussed in the “Handoff” section below, is

that defendants propose that the second phase

automatically end after two years of monitoring.

In the third phase, the plaintiffs propose that the

EMT continue to observe the IMT and retain the authority

to return monitoring to the second phase. By contrast,

the defendants want monitoring in the third phase to be

exclusively conducted by the IMT, without EMT

supervision.

b. The Court’s Resolution

The court will order that the IMT’s responsibilities

and involvement will be largely driven by what the EMT

members, who are the experts, believe is necessary.

During the first and second phases, the EMT shall be

allowed to invite the IMT to observe the EMT in its

monitoring activities, unless there is an activity that

the EMT requests to conduct without being observed, such

as a confidential interview. During the second phase,

the EMT shall retain primary responsibility for

monitoring and producing the quarterly reports; however,

the IMT shall participate in those activities to the

extent the EMT deems appropriate. The court will not

adopt the plaintiffs’ proposal that, during the second

phase, in addition to the EMT-produced report, the IMT

produce a separate draft report for the sole purpose of

training. Instead, the court will leave it to the EMT

to determine how best to involve the IMT in drafting

reports and other monitoring activities as it deems

necessary for training and capacity-building. Finally,

the court will order that, during the third phase, the

EMT shall remain available for consultation with the IMT,

the parties, and the court, should any issues or disputes

arise.

The court finds these steps necessary to ensure that

the IMT not only receives the necessary training, but

also that it is capable of executing the work of

monitoring in accordance with that training. Allowing

for collaboration between the EMT and IMT in every phase

of monitoring is essential to ensure that, one day, ADOC

will have the capacity to self-monitor. The court also

expects that involving the IMT from the very beginning

of monitoring will help to improve the working

relationship between the EMT and ADOC and help develop

the IMT into a “cultural force” for institutional change.

Vail Nov. 29, 2018, Trial Tr. (doc. no. 2340) at 69.

This provision satisfies the

need-narrowness-intrusiveness requirement of the PLRA.

See 18 U.S.C. § 3626(a)(1)(A). The consultation element

of monitoring is particularly critical in light of the

admission of the defendants that ADOC lacks the

“training” to self-monitor. Defs.’ Proposed Monitoring

Plan (doc. no. 2115) at 2. Commissioner Dunn himself

testified that he instructed his staff to develop a

monitoring plan that focused “on training, education, and

building internal capacity.” Dunn Nov. 26, 2018, Trial

Tr. (doc. no. 2250) at 7. Consultation is “one of the

ways in which the monitor helps the agency develop the

internal knowledge and skill set to be able to assume

responsibility for internal monitoring, ongoing,

continuous quality improvement process.” Burns Dec. 6,

2018, Trial Tr. (doc. no. 2254) at 68. The court’s order

limits the consultation element of monitoring to only

what is necessary to ensure that ADOC develops the

ability to self-monitor, rather than requiring the

redundant reporting by the IMT in the second phase or the

intrusive supervision over the IMT in the third phase

sought by the plaintiffs. Consultation, like all

elements of the court’s monitoring order, is also limited

in scope to the requirements of this court’s remedial

orders and is thus narrowly tailored to redress the

violations found by this court.

5. Handoff to ADOC

The “handoff” to ADOC involves two central issues.

First, when does the EMT transfer primary monitoring

responsibilities to the IMT, that is, when does

monitoring move from phase two to phase three? Second,

when does court monitoring terminate? The parties

disagree on both issues.

a. Timing of Monitoring Phases

i. Dispute

As to the first issue of transition from phase two

to phase three of monitoring, the defendants propose

automatic termination of the EMT and transition to phase

three after two years of monitoring--four quarters in

phase one and four quarters in phase two. By contrast,

the plaintiffs contend that the EMT “cannot automatically

disband after two years.” Pls.’ Proposed Opinion (doc.

no. 2260) at 52. The plaintiffs propose that the EMT

conduct monitoring until the EMT determines that the IMT

“has developed the competence to lead the monitoring

efforts,” id. at 72, or until the defendants demonstrate

to the court “that the remedies being monitored can be

terminated because they are no longer necessary to remedy

ongoing constitutional violations,” id. at 52. The

plaintiffs further assert that, in the third phase of

monitoring, the EMT should continue to exercise oversight

of the IMT’s monitoring, including directing which

facilities are to be audited.

The defendants take issue with the fact that the

plaintiffs “do not propose any clear point in time for

transitioning from the [EMT] to the [IMT].” Defs.’

Response (doc. no. 2295) at 33. Under the plaintiffs’

plan, they complain, the determination of when the IMT

is prepared to assume primary monitoring responsibilities

would be left “entirely to the subjective opinion of the

[EMT] without holding the [EMT] accountable for training

and mentoring the [IMT].” Id. The defendants argue that

the EMT would have no incentive to conclude its work and

allow ADOC to take over monitoring.

ii. The Court’s Resolution

The court will adopt the defendants’ proposal that

after one year, monitoring will automatically shift from

phase one to phase two. However, the court will order

that monitoring will move from phase two to phase

three--that is, the IMT will assume responsibility for

monitoring--when the court determines, after a hearing,

that the IMT is sufficiently competent that monitoring

by the EMT is no longer necessary. Of course, in phase

three, monitoring by the court would continue, albeit by

way of the IMT, until monitoring is no longer needed.

The defendants’ critique and proposal suffer from

two key problems. First, the defendants’ plan would in

no way incentivize the IMT or ADOC to build its internal

monitoring capacity. The fixed, two-year cut off sends

the message that, regardless of the IMT’s monitoring

capacity after two years, the IMT would take over and

external monitoring will end at that time. Second, and

perhaps more importantly, a fixed two-year cut-off would

mean that external monitoring would end regardless of the

status of compliance by ADOC. As plaintiffs’ expert Dr.

Burns stated, under the defendants’ proposal, the

transition would occur “based on the passage of time, not

necessarily improvements in the system.” Burns Dec. 6,

2018, Trial Tr. (doc. no. 2254) at 51. As explained

above, external monitoring is necessary to address

ongoing constitutional violations. This is because ADOC

has failed to self-identify and self-correct problems

with its provision of mental-health services to inmates,

see Braggs v. Dunn, 257 F. Supp. 3d 1171, 1257 (M.D. Ala.

2017) (Thompson, J.), and because this failure has

continued since the liability opinion, demonstrated by

ADOC’s ongoing failure to self-monitor compliance with

remedial orders, as described in the final section of

this opinion. Therefore, external monitoring will

continue to be necessary until the defendants have the

capacity to self-monitor; that is, until the IMT has

built and demonstrated its competency both to identify

and correct deficiencies, or rather, ADOC itself has

built and demonstrated such. The court has no reason to

believe that ADOC, via the IMT, will automatically obtain

competency after two years of monitoring, that is, that

it will have a documented, substantial track record of

identifying and successfully correcting deficiencies.

Cf. Burns Dec. 6, 2018, Trial Tr. (doc. no. 2254) at 116

(estimating that in the best-case scenario, ADOC will

reach constitutional compliance in five years).

The court shares the defendants’ concern that

monitoring may extend indefinitely. Therefore, the court

will rely on the process set forth by the PLRA for making

the determination that the IMT is ready to assume

responsibility for monitoring--an approach that ensures

external monitoring continues only as long as it is truly

necessary. Accordingly, at any point at least two years

after the entry of this order, any party may move to

terminate the monitoring by the EMT. See 18 U.S.C.

§ 3626(b)(1). (After all, ADOC itself may conclude,

after two years of experience with the EMT and in light

of then-current conditions, that it then realizes that

it is not quite ready for self-monitoring.) Pursuant to

that motion, the court will terminate the EMT’s

monitoring and hand it over to the IMT unless the

plaintiffs demonstrate, at an evidentiary hearing, that

external monitoring remains necessary to correct a

current and ongoing constitutional violation, and that

the monitoring order continues to meet the

need-narrowness-intrusiveness test. See 18 U.S.C.

§ 3626(b)(3); see also Cason v. Seckinger, 231 F.3d 777,

782-83 (11th Cir. 2000). In the event of such a hearing,

the court will consider, among other evidence, the

opinions of the members of the EMT as to whether the IMT

has developed the capacity to assume monitoring

responsibilities and engage successfully in

self-correction. The court expects that, through the

various mechanisms in this process designed to build the

IMT’s capacity, once monitoring shifts into phase three,

the IMT will take over monitoring for the remaining

duration of the court’s oversight. However, the court

will reserve the authority to return monitoring to phase

two and re-engage the EMT experts as primary monitors if

the court determines, upon the motion of either party and

after an evidentiary hearing, that the IMT is no longer

fulfilling its obligations.

The court understands the defendants’ concern that

the experts on the EMT, who are paid for their work, may

have competing interests with regard to handing over the

reins to ADOC. However, this concern is insufficient to

support automatic and arbitrary termination of external

monitoring after two years. And regardless of the EMT’s

motivations, the court’s resolution will provide ADOC the

opportunity to show the court that it is ready to take

over monitoring.

All that said, the court agrees with the defendants

that the IMT would benefit from having clear goals and

criteria for determining whether it has the ability to

assume responsibility for monitoring. Students should

know what their teachers require for a passing grade.

Accordingly, in phase one, the EMT will develop and

provide the IMT with criteria--much like performance

measures--for eventually assessing whether it is able to

assume the reins of monitoring. The IMT’s performance

in meeting these criteria would serve as evidence, in a

future hearing to terminate phase two, of whether the IMT

has the capacity to take over monitoring. To continue

with the teaching metaphor, members of the EMT can

testify about whether the IMT has a passing grade; the

defendants and IMT can always disagree and argue for a

better grade. Ultimately, it will be the court that

determines whether the IMT has the passing grade to

graduate and take over monitoring. As stated, the court

will not adopt the plaintiffs’ proposal that the EMT

continue to exercise a supervisory role over the IMT once

the IMT takes over monitoring. Instead, the court will

require that the EMT remain available for consultation

with the court and the parties as needed.

The court’s resolution satisfies the

need-narrowness-intrusiveness requirement of the PLRA.

See 18 U.S.C. § 3626(a)(1)(A). The entire exercise of

monitoring would be just that, an exercise, if the EMT’s

oversight and consultation were to end before ADOC has

gained the capacity to self-monitor. However, under this

scheme, the EMT will exist only as long as it is necessary

to help ADOC achieve that end.

Moreover, the court would emphasize that this case

may be unique in that the monitoring scheme adopted here

does not consist exclusively of external monitoring.

Instead, the court has substituted internal monitoring

for part of the external monitoring that would otherwise

be imposed. To this extent, the monitoring scheme is less

intrusive than it otherwise would be.

b. Termination of Court Monitoring

i. Dispute

The defendants assert that court monitoring--as well

as the underlying remedial orders--should terminate when

there is “sustained substantial compliance” with

performance measures. See Defs.’ Proposed Monitoring

Plan (doc. no. 2115) at 30. By contrast, the plaintiffs

propose that court monitoring continue until the

underlying remedial orders are terminated based on a

finding, under the PLRA, that they are no longer

necessary to correct a current and ongoing violation.

See Pls.’ Proposed Opinion (doc. no. 2260) at 28, 75.

The defendants further assert that when ADOC achieves

“sustained substantial compliance” with respect to a

single performance measure, all court monitoring of that

particular performance measure should cease, and the

portion of the court’s remedial order related to that

performance measure should also terminate. Defs.’

Proposed Monitoring Plan (doc. no. 2115) at 21. They

propose that, “[a]s a general matter, the Compliance

Teams should use eighty-five percent (85 %) as the

threshold for substantial compliance,” and many of their

proposed performance measures specify 85 % as the

threshold.5 Id. at 18. Crucially, however, the

5. For example, the defendants’ proposed performance

measure number 139 is that “[a] member of the MH staff

will complete a MH Progress Note for each inmate

participating in a group activity or therapy and place

the note in the inmate’s record.” Defs.’ Proposed

Monitoring Plan (doc. no. 2115-1) at 15. Their proposed

audit tool is to “[r]eview a minimum of 10 charts for

inmates on the MH caseload who participated in a group

activity or therapy during the applicable quarter for

each identified facility.” Id. Their proposal states

that “substantial compliance” shall be 85 % or greater.

See id. Accordingly, to achieve substantial compliance

with this performance measure, at least 85 % of the

reviewed charts must show that a member of the

defendants maintain that the EMT has the authority under

their plan to modify the proposed compliance percentage

for a particular performance measure and make it higher.

See Defs.’ Response (doc. no. 2295) at 21-23.

According to the defendants, “sustained substantial

compliance” occurs with respect to a particular

performance measure when the EMT determines that ADOC has

substantially complied with that performance measure for

three quarters at any ADOC facility or facilities.

Defs.’ Proposed Monitoring Plan (doc. no. 2115) at 9-10.

The substantial compliance need not be for three

consecutive quarters, as long as there is no intervening

quarter with a finding of noncompliance (generally less

than 65 % compliance) as to that performance measure.

Id. Consequently, under the defendants’ plan, it would

appear that termination of monitoring of a performance

measure at all ADOC facilities--as well as the

termination of the corresponding portion of the remedial

mental-health staff completed a mental-health progress

note.

order--can be based on a finding of substantial

compliance in just a few facilities, or even a single

one, in three, not necessarily consecutive, quarters.

The defendants propose that, even though court

monitoring of a performance measure will end once there

is sustained substantial compliance, ADOC will continue

to monitor compliance with the performance measure at

least on an annual basis as part of its ongoing continuous

quality improvement activities (CQI). See id. at 14. Of

course, the results of this CQI self-oversight would not

be reported to the court or the plaintiffs. Under the

defendants’ plan, all court monitoring and the court’s

remedial orders would end as soon as sustained

substantial compliance is achieved for every performance

measure.

The plaintiffs disagree with the defendants’

definition of substantial compliance. Their main

objections are that the defendants’ definition is purely

numerical, rather than also qualitative, and that, as

their expert Dr. Burns and the defendants’ proposed

monitor Dr. Stern testified, some remedial measures must

be complied with 100 % of the time, rather than 85 % of

the time, to be effective. See Burns Dec. 7, 2018, Trial

Tr. (doc. no. 2256) at 123; Stern Nov. 27, 2018, Trial

Tr. (doc. no. 2415) at 156. Instead, the plaintiffs

propose that the court adopt the definition of

substantial compliance utilized in the opinion adopting

the United States v. Alabama settlement agreement

regarding conditions at Tutwiler prison. See No.

2:15cv368-MHT, 2015 WL 3796526 (M.D. Ala. June 18, 2015)

(Thompson, J.) (opinion adopting settlement agreement).6

They thus propose that the EMT “may identify a numerical

threshold for any remedial requirement they deem

appropriate, but the monitors must also consider whether

6. Specifically, plaintiffs propose that

“‘[s]ubstantial [c]ompliance’ indicates that defendants

have achieved material compliance with most or all

components of the relevant remedial requirement.” Pls.’

Proposed Opinion (doc. no. 2260) at 64. “Material

[c]ompliance,” in turn, “requires that, for each remedial

requirement, defendants have developed and implemented a

policy incorporating the requirement, trained relevant

personnel on the policy, and relevant personnel are

complying with the requirement in actual practice as

measured both quantitatively and qualitatively.” Id.

policies are developed, staff are trained, and the

policies are followed.” Pls.’ Proposed Opinion (doc. no.

2260) at 53. Another area of disagreement is that the

plaintiffs insist that substantial compliance as well as

sustained substantial compliance be measured on a

facility-by-facility basis. This differs from the

defendants’ proposal that sustained substantial

compliance for all facilities can be found based on

substantial compliance in just one or a few facilities.

Even though the plaintiffs propose a different

definition of substantial compliance, their proposal does

not make clear the immediate relevance of substantial

compliance. Unlike the defendants, they do not tie the

end of monitoring of each performance measure to the

achievement of substantial compliance. Instead of

connecting the termination of court monitoring to

substantial compliance, the plaintiffs propose that,

under the PLRA, “the defendants may move to terminate the

court’s remedial orders (and thus any concomitant

monitoring) after two years, and the court will order

such if the evidence shows that the remedial orders are

no longer necessary to correct current and ongoing

violations.” Id. at 75.

ii. The Court’s Resolution

The court will adopt the same approach as above with

regard to the second “handoff” issue of termination of

court monitoring. That is, the court will simply rely

on the process set forth by the PLRA for determining when

to terminate relief. Accordingly, at any point at least

two years after the entry of this order, any party or

intervenor may move to terminate court monitoring

altogether. See 18 U.S.C. § 3626(b)(1). The court will

terminate court monitoring unless the plaintiffs

demonstrate, at an evidentiary hearing, that court

monitoring remains necessary to correct a current and

ongoing constitutional violation, and that court

monitoring continues to meet the

need-narrowness-intrusiveness test. Id. § 3626(b)(3);

see also Cason v. Seckinger, 231 F.3d 777, 782-83 (11th

Cir. 2000). In the event of such a hearing, the inquiries

will be whether ADOC is complying with the remedial

orders and whether the IMT has the ability to identify

and successfully correct problems without court

oversight.

Accordingly, crucial evidence for determining

whether court monitoring remains necessary will be

whether ADOC has achieved substantial compliance with the

performance measures. As both parties agree, the EMT

will have the authority to define substantial compliance

for each performance measure. This could include, if the

EMT so determines, the ability to take into account more

qualitative criteria, such as “whether policies are

developed, staff are trained, and the policies are

followed.” Pls.’ Proposed Opinion (doc. no. 2260) at 53.

As proposed by the defendants, even after monitoring is

transferred from the EMT to the IMT, the IMT must continue

to use the EMT’s definitions of substantial compliance,

unless a change is approved by the plaintiffs or the

court.7

There is merit to the defendants’ argument that

“loose and subjective standards” risk having ADOC and its

mental-health vendor “feel that they [are] aiming at a

moving target.” Defs.’ Response (doc. no. 2295) at 13-14.

Dr. Burns agreed that the expectations for the

mental-health staff’s performance should be very clear.

See Burns Dec. 7, 2018, Trial Tr. (doc. no. 2256) at 220,

251. These concerns are adequately taken into account

by allowing the EMT to create a definition of substantial

compliance that includes both quantitative percentages,

as well as consideration of whether policies are

developed, staff are trained, and the policies are

followed. These assessments can be objectively and

empirically determined. And, by clarifying from the

7. The defendants proposed that, “[u]nlike the

[EMT], the [IMT] shall not deviate from the substantial

compliance and partial compliance percentages utilized

by the [EMT] during the Phase I and Phase II evaluations,

unless such deviation is approved by (a) Plaintiffs or

(b) the Court.” Defs.’ Proposed Monitoring Plan (doc. no.

2115) at 18.

outset that substantial compliance will be crucial

evidence for showing that court monitoring is no longer

necessary, the defendants receive “fair notice of what

standard they must reach” to terminate monitoring.

Defs.’ Response (doc. no. 2295) at 13.

The court need not resolve whether, as proposed by

the plaintiffs, substantial compliance must be shown on

a facility-by-facility basis for each performance

measure, or whether sustained substantial compliance at

a few facilities is sufficient to terminate monitoring

of the performance measure for all facilities, as

proposed by the defendants. This is because the question

presented at the hearing to terminate court monitoring

will be whether court monitoring remains necessary.

Obviously, if the defendants present evidence of

sustained substantial compliance in just one or two

facilities for a performance measure, that evidence may

be less compelling than if they present evidence of

sustained substantial compliance throughout the system.

The EMT experts shall have the authority, without a

hearing, to stop evaluating a particular performance

measure at a particular facility, or stop evaluating a

facility altogether, based on their own determination of

sustained substantial compliance. As with the

defendants’ proposal that the IMT must continue to use

the definitions of substantial compliance developed by

the EMT, the IMT must also continue to evaluate the

performance measures and facilities that the EMT is still

assessing at time of transition to the IMT. The IMT may

only stop assessing them upon a determination, after a

hearing, that they are no longer necessary, or by consent

of the parties. Because the duration of court monitoring

is tied to the status of compliance and capacity of the

IMT, rather than an arbitrary schedule, monitoring will

only “intrude” as long as necessary to report and correct

the constitutional violation.

Finally, the underlying remedial orders do not

terminate just because monitoring of them ends; the

underlying remedial orders will terminate separately

based on the same procedure set forth in the PLRA--that

is, when, after a hearing, they are determined to be no

longer necessary. 18 U.S.C. § 3626(b)(3); see also Cason

v. Seckinger, 231 F.3d 777, 782-83 (11th Cir. 2000). As

Dr. Burns testified, the inquiries for terminating

monitoring and terminating the remedial orders are

different; even if ADOC is not yet in adequate compliance

with all of the court’s remedial orders, if “there’s a

plan in place and the internal team and the department

ha[ve] demonstrated that they know how to take that

information, develop a plan of correction, implement it,

and change things,” then external monitoring may no

longer be necessary. Burns Dec. 6, 2018, Trial Tr. (doc.

no. 2254) at 74. Of course, the evidence of substantial

compliance to show that monitoring is no longer necessary

will also be directly relevant to proving that the

underlying remedial order is no longer necessary. Thus,

there may be overlap between when court monitoring and

underlying remedial orders end.

III. MONITORING AND THE PLRA

It is a complicated question whether the PLRA’s

requirements apply to the court’s order regarding

monitoring. With regard to some of the court’s

monitoring provisions adopted today, the parties appear

to agree that the PLRA governs. The plaintiffs admit

that, “[w]here the remedy includes setting up internal

processes for self-monitoring and self-correction,

courts must determine whether the [PLRA’s] requirement

is met.” Pls.’ Response Regarding PLRA (doc. no. 2213)

at 3. As stated, the parties disagree, however, about

whether external monitoring is “prospective relief” and

therefore subject to the PLRA’s

need-narrowness-intrusiveness requirement. 18 U.S.C. §

3626(a)(1)(A). The defendants contend that it is. By

contrast, the plaintiffs argue that, to the extent

monitoring is limited to informing the court whether the

defendants comply with court orders, the requirement does

not apply because such monitoring constitutes a means to

relief, as opposed to “prospective relief” within the

meaning of the PLRA. Id.

The caselaw is unclear as to whether the

need-narrowness-intrusiveness requirement applies to

court monitoring. Some district courts agree with the

plaintiffs that monitoring is a means to relief, rather

than “prospective relief,” and therefore is not subject

to the requirement. See, e.g., Carruthers v. Jenne, 209

F. Supp. 2d 1294, 1300 (S.D. Fla. 2002) (Hoeveler, J.)

(“Clearly monitoring is not an ‘ultimate remedy’ and only

aids the prisoners in obtaining relief.”); Benjamin v.

Fraser, 156 F. Supp. 2d 333, 342-43 (S.D.N.Y. 2001)

(Baer, J.) (holding that monitoring “cannot be relief”

and “[t]o find otherwise would conflate relief with the

means to guarantee its provision”), aff’d in part,

vacated in part on other grounds, 343 F.3d 35 (2d Cir.

2003). On the other hand, the Second Circuit stated in

dictum that it was “somewhat problematic” for the

district court in the case before it to conclude that

monitoring is not relief within the meaning of the PLRA.

Benjamin v. Fraser, 343 F.3d 35, 48-49 (2d Cir. 2003),

overruled on other grounds by Caiozzo v. Koreman, 581

F.3d 63 (2d Cir. 2009). The appellate court reasoned

that placing the monitoring body beyond the reach of the

PLRA would “frustrat[e] one of the Act’s broad goals of

limiting ‘the micromanag[ing] [of] State and local prison

systems.’” Id. at 49 (citing 146 Cong. Rec. S14611,

14626 (1995) (statement of Sen. Dole)). Additionally,

because the monitoring body at issue had “substantial

responsibilities,” there was “no easy distinction between

relief itself and the monitoring of relief.” Id. After

making these observations, the Second Circuit refrained

from resolving whether monitoring constituted

prospective relief, because it held that the district

court had made the appropriate

need-narrowness-intrusiveness findings. See id.

Ultimately, this court need not resolve whether

external monitoring is “prospective relief” subject to

the PLRA’s need-narrowness-intrusiveness requirement,

because, as elaborated above, the monitoring ordered here

satisfies the requirement. So, to the extent monitoring

must meet the requirement, it does.

Several courts have indicated that in applying the

need-narrowness-intrusiveness requirement, a history of

non-compliance helps justify an intrusive remedy. See

Benjamin, 343 F.3d at 49 (finding that an independent

monitoring body with substantial responsibilities

satisfied the need-narrowness-intrusiveness requirements

in part because of the district court’s finding that “the

nearly twenty year history of incomplete compliance with

the consent decrees amply attests to the need for

external monitoring” (citation omitted)); see also

Benjamin v. Schriro, 370 Fed. App'x 168, 171 (2d Cir.

2010) (unpublished) (“The needs-narrowness-intrusiveness

requirement of the PLRA notwithstanding, we find that

nearly a half-decade of untruthfulness, non-compliance

and inaction constitutes sufficient justification for the

intrusiveness of a subsequent order to compel compliance

with an original order entered pursuant to the PLRA that

has been ignored.”); Clark v. California, 739 F. Supp.

2d 1168, 1233-35 (N.D. Cal. 2010) (Breyer, J.) (stating

that, in fashioning relief, courts “may take into account

a history of noncompliance with prior orders,” and

finding that further orders satisfied the

need-narrowness-intrusiveness requirements in part

because of evidence of non-compliance with prior orders

and because “defendants have demonstrated an inability

to take remedial steps absent court intervention”).

Thus, the evidence of ADOC’s failure to comply with the

remedial orders in this case, described in detail in the

final section of this opinion, would likely justify even

a far more intrusive monitoring order than the court

enters today.

Nonetheless, the monitoring scheme that the court

orders today is the least intrusive possible as it is

largely drawn from the defendants’ own plan. And the

fact that the defendants agree that some degree of both

external and internal monitoring is necessary, see, e.g.,

Dunn Nov. 26, 2018, Trial Tr. (doc. no. 2250) at 35

(Commissioner Dunn testifying: “Q. Have you ever taken

the position that the state should not be subject to

monitoring or oversight on these issues? A. No.”),

further supports the court’s finding that the overall

monitoring scheme satisfies the PLRA, see 18 U.S.C.

§ 3626(a)(1)(A); Braggs v. Dunn, 383 F. Supp. 3d 1218,

1253 (M.D. Ala. 2019) (Thompson, J.) (finding the

defendants’ agreement to a remedial provision is strong

evidence of PLRA compliance).

IV. WHY COURT-ORDERED MONITORING IS NEEDED

The defendants argue that ADOC should be allowed to

“voluntarily undertake culture change” without a court

order.8 Defs.’ Response (doc. no. 2295) at 9. But ADOC

8. Overall, the court is not persuaded by the

defendants’ central critique of what they call the

“historical,” “coercive” model of court-ordered

monitoring, which they claim has failed. Defs.’ Response

(doc. no. 2295) at 7. Though the defendants never

explicitly define what they consider the “historical”

approach, in their critiques they refer to monitoring

schemes that involve court-appointed external monitors

and termination of monitoring based on progress rather

than a set schedule. See Dunn Nov. 26, 2018, Trial Tr.

(doc. no. 2250) at 9-14. According to the defendants,

this approach “‘is basically based on compulsion [with]

someone looking over your shoulder all the time and

has had ample opportunity, in this litigation and for

decades prior, to correct voluntarily its failings

regarding the mental-health care in its prisons, and its

conduct during the course of this case has been

coming in and, in effect, taking over a core function of

the department...,’ i.e. the continuous quality

improvement (“CQI”) process of improving the delivery of

mental-health care.” Defs.’ Response (doc. no. 2295) at

7 (quoting Dunn Nov. 26, 2018, Trial Tr. (doc. no. 2250)

at 9). The court is unpersuaded by this argument because

the defendants presented absolutely no evidence of the

failure of this “historical” model. They highlight that,

as described below, there has been repeated litigation

over prison conditions in Alabama. But that is not

evidence that it was the monitoring that led to the

recurrence of unconstitutional prison conditions. It is

an equally plausible explanation--and, indeed, the court

finds it is more likely--that the lack of effort or

resources invested by ADOC to make s

This text is long and has been trimmed here. Open the source document for the complete record.