PHASE 2A CLASS CERTIFICATION OPINION

MYRON H. THOMPSON, District Judge.

The plaintiffs in this putative class-action lawsuit are dozens of state prisoners and the Alabama Disabilities Advocacy Program (ADAP). The defendants are officials of the Alabama Department of Corrections (ADOC): the Commissioner and the Associate Commissioner of Health Services.1 They are sued in their official capacities only.

In Phase 2A of this case, with which this opinion is concerned, ADAP and a subset of individual named plaintiffs assert the following claims: constitutionally inadequate mental-health treatment in Alabama prison facilities and involuntary medication without due process. They rely on the Eighth and Fourteenth Amendments, as enforced through 42 U.S.C. § 1983. Plaintiffs seek declaratory and injunctive relief. Jurisdiction is proper under 28 U.S.C. § 1331 (federal question) and § 1343 (civil rights).2 The case is proceeding on two parallel tracks consisting of ADAP's claims and the individual named plaintiffs' claims.

In August 2016, more than two years after this case was filed and after extensive discovery, the individual plaintiffs formally moved for certification of a Phase 2 class, while ADAP pursued its claims separately as an association whose constituents include the mentally ill prisoners in ADOC's custody.3

For reasons that follow, the motion for class certification filed by the individual plaintiffs (hereinafter referred to simply as plaintiffs) will be granted in part and denied in part, and two classes certified. The court has narrowed the scope of the mental-health care Eighth Amendment class, and winnowed down the involuntary-medication due-process claim to the portion suitable for class certification.

I. Rule 23(c)(1)(B)

Federal Rule of Civil Procedure 23(c)(1)(B) requires a certification order to "define the class and the class claims, issues, or defenses"; therefore, a court must consider the propriety of certification in relation to a specific putative class definition and specific putative class claims.

Plaintiffs seek certification of a Rule 23(b)(2) class consisting of "all persons with a serious mental health disorder or illness who are now, or will in the future be, subject to Defendants' mental health care policies and practices in ADOC facilities." Pls.' Reply Br. (doc. no. 890) at 57.4

The first claim plaintiffs seek to bring on behalf of this class is that defendants are violating the Eighth Amendment by failing to provide constitutionally adequate mental-health care to prisoners with serious mental illnesses. Specifically, they allege that defendants have been deliberately indifferent to a substantial risk of serious harm posed to the putative class members by a number of policies and practices (both separately and in combination). Although these policies and practices, according to both parties' experts, sometimes contribute to each other, it is helpful to parse them in order to ensure that the putative class can properly challenge them.

Plaintiffs take the position that they need not demonstrate the suitability for class-wide adjudication of any particular policy or practice, and that they need only demonstrate that the overarching claim of inadequate mental-health care is appropriate for certification. But, as in Parsons v. Ryan, a similar recent case, the court must dive somewhat deeper. 289 F.R.D. 513, 522-23 (D. Ariz. 2013) (Wake, J.) (certifying a state-wide class of prisoners challenging systemic deficiencies in both medical and mental-health care). Although the court agrees that plaintiffs have brought only one Eighth Amendment "claim," this claim— as it has been formulated by plaintiffs—turns on evaluating the risk of harm posed by particular policies and practices. The court will need to ensure, therefore, that these policies and practices are common to the class, and that representatives have been put at risk by them. Defendants, for their part, seek to atomize the claims of plaintiffs by focusing on the minutiae of unimportant distinctions, failing to see the forest for the trees (and leaves).

The court is thus left with the task of reaching an appropriate middle ground, and defining the policies and practices at issue in a way that recognizes themes articulated in the allegations and the voluminous quantity of record evidence the court has reviewed, while ensuring that defendants are not left to face a singular amorphous contention that the mental-health care they are providing is not good enough. Despite plaintiffs' protestations, the headings in their complaint do an admirable job in this respect, and the court has drawn heavily upon them in formulating the following list.

The eight policies and practices at issue in plaintiffs' Eighth Amendment claim are:

In addition, plaintiffs seek to raise on behalf of the class due-process claims related to involuntary medication. These, too, require some disaggregation. Plaintiffs seek to challenge three discrete policies or practices:

For the reasons set forth below, the court will certify two classes, as follows.

With respect to plaintiffs' Eighth Amendment challenge to the eight policies and practices enumerated above, the court will certify a class of "all persons with a serious mental-health disorder or illness who are now, or will in the future be, subject to defendants' mental-health care policies and practices in ADOC facilities, excluding the work release centers and Tutwiler Prison for Women." Plaintiffs (and defendants) have agreed that prisoners at the work release centers should be excluded from the class definition; the court will explain below, in the context of its discussion of commonality, why it has also decided to exclude the female prisoners at Tutwiler.

With respect to plaintiffs' challenge to the policy or practice of denying procedural due process to prisoners subject to involuntary medication orders by failing to provide them adequate notice of hearings and other protections provided for in the applicable regulation, the court will also certify a class of "all persons with a serious mental-health disorder or illness who are now, or will in the future be, subject to defendants' formal involuntary medication policies and practices." The court has decided, in an exercise of its discretion to manage this litigation, not to certify for class-wide litigation plaintiffs' substantive due-process challenge to the policy or practice of requiring involuntary medication without a recent finding of dangerousness—which turns at least in significant part on an inherently individualized determination—or their challenge to the policy or practice of obtaining consent through coercion. However, the individual plaintiffs who have brought these claims will proceed to trial; ADAP, through its associational standing, may represent the interests of unnamed prisoners who are subject to these policies and practices.

II. Evidentiary Burden

In Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338 (2011), the Supreme Court explained that "Rule 23 does not set forth a mere pleading standard. A party seeking class certification must affirmatively demonstrate his compliance with the Rule—that is, he must be prepared to prove that there are in fact sufficiently numerous parties, common questions of law or fact, etc." 564 U.S. at 350. A court should only certify a class if it "is satisfied, after a rigorous analysis that the prerequisites of Rule 23(a) have been satisfied." Comcast Corp. v. Behrend, 133 S.Ct. 1426, 1432 (2013).5

The Supreme Court has been clear that what the party seeking certification must "affirmatively demonstrate," and what the court must find, is that the requirements for class certification are met—not that the class will prevail on its claims. As the Court put it in Amgen Inc. v. Connecticut Ret. Plans & Trust Funds, 133 S.Ct. 1184, 1196 (2013), it "totally misapprehend[s] the essential point" of this case law to suggest that certification is improper unless plaintiffs are able to prove that the common question "will be answered in their favor." All they need to offer evidence to show—for purposes of certification— is that there exists a common question, the answer to which will be "apt to drive the resolution of the litigation." Wal-Mart, 564 U.S. at 350.

III. Daubert

The parties have raised Daubert objections to each other's experts. For purposes of the Phase 2A pretrial motions, however, the only objections that must be resolved are those to plaintiffs' mental-health expert, Dr. Kathryn Burns.6 Although the court has carefully considered all of the evidence in the record, it need not resolve plaintiffs' objections to defendants' experts until trial, because class certification is appropriate even assuming that defendants' expert evidence is admissible. The court therefore assumes— for purposes only of this opinion—that it is. Defendants do not raise Daubert objections to the reports of plaintiffs' other Phase 2A experts, Dr. Craig Haney7 and Eldon Vail.8

1. Standard

After the Supreme Court outlined a gatekeeping analysis for scientific expert testimony in Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993), and made that analysis applicable (with necessary modifications in the precise questions asked) to other technical expert testimony in Kumho Tire Co., Ltd. V. Carmichael, 526 U.S. 137 (1999), the Eleventh Circuit explained that the purpose of this analysis is "to ensure the reliability and relevancy of expert testimony . . . [and] to make certain that an expert, whether basing testimony upon professional studies or personal experience, employs in the courtroom the same level of intellectual rigor that characterizes the practice of an expert in the relevant field." United States v. Frazier, 387 F.3d 1244, 1260 (11th Cir. 2004) (en banc) (quoting Kumho Tire, 526 U.S. at 152). The appeals court went on to explain that under Federal Rule of Evidence 702, courts are to "engage in a rigorous three-part inquiry," considering "whether: (1) the expert is qualified to testify competently regarding the matters he intends to address; (2) the methodology by which the expert reaches his conclusions is sufficiently reliable as determined by the sort of inquiry mandated in Daubert; and (3) the testimony assists the trier of fact, through the application of scientific, technical, or specialized expertise, to understand the evidence or to determine a fact in issue." Id. (citation and internal quotation marks omitted). The proponents of the expert testimony (here, plaintiffs) bear the burden of establishing that these three requirements are met. See id.

Ordinarily, "[t]he safeguards outlined in Daubert are less essential in a bench trial"; a judge need not gatekeep for herself. M.D. v. Abbott, 152 F.Supp.3d 684, 709 (S.D. Tex. 2015) (Jack, J.) (citing Gibbs v. Gibbs, 210 F.3d 491, 500 (5th Cir. 2000)), appeal dismissed (Apr. 5, 2016). However, the Eleventh Circuit and others have held that when a court relies on expert testimony to find that a Rule 23 requirement has been met, the court must conduct a Daubert analysis and conclude that the expert's opinions satisfy its standard. Sher v. Raytheon Co., 419 F. App'x 887, 890-91 (11th Cir. 2011); see also In re Blood Reagents Antitrust Litig., 783 F.3d 183, 187 (3d Cir. 2015); Ellis v. Costco Wholesale Corp., 657 F.3d 970, 982 (9th Cir. 2011). Some courts have noted, however, that the Daubert analysis conducted for purposes of class certification may be narrower than that conducted for purposes of a trial on the merits, because "the inquiry is limited to whether or not the expert reports are admissible to establish the requirements of Rule 23." Fort Worth Employees' Ret. Fund v. J.P. Morgan Chase & Co., 301 F.R.D. 116, 126 (S.D.N.Y. 2014) (Oetken, J.).

Assessing an expert's qualifications is generally the most straightforward of these tasks. Defendants do not challenge Dr. Burns's qualifications, and as discussed below, the court finds that she is exceedingly well-qualified to opine on the matters she addresses in her report.

As for the court's assessment of an expert's methodology, the court is to determine "whether the reasoning or methodology underlying the testimony is. . . valid and . . . whether that reasoning or methodology properly can be applied to the facts in issue."9 Daubert, 609 U.S. at 592-93. The traditional factors a court may consider, where appropriate, are "(1) whether the expert's theory can be and has been tested; (2) whether the theory has been subjected to peer review and publication; (3) the known or potential rate of error of the particular [expert] technique; and (4) whether the technique is generally accepted in the [expert] community." Frazier, 387 F.3d at 1262. However, the court of appeals has emphasized that, particularly in cases like this one, involving non-scientific expert evidence, "[t]hese factors are illustrative, not exhaustive; not all of them will apply in every case, and in some cases other factors will be equally important. . . . Sometimes the specific Daubert factors will aid in determining reliability; sometimes other questions may be more useful. As a result, `the trial judge must have considerable leeway in deciding in a particular case how to go about determining whether particular expert testimony is reliable.' Kumho Tire, 526 U.S. at 152. Exactly how reliability is evaluated may vary from case to case. . . ." Frazier, 387 F.3d at 1262; see also United States v. Brown, 415 F.3d 1257, 1267-68 (11th Cir. 2005) (explaining that "the question of whether Daubert's specific factors are, or are not, reasonable measures of reliability in a particular case is a matter that the law grants the trial judge broad latitude to determine" (internal citations omitted)); Ruiz v. Johnson, 37 F.Supp.2d 855, 889-92 (S.D. Tex. 1999) (Justice, J.), rev'd on other grounds and remanded sub nom. Ruiz v. United States, 243 F.3d 941 (5th Cir. 2001) (explaining that an expert's evaluation of the quality of the medical care provided by a prison system is "not the type of testimony that necessarily implicates Daubert's requirement of scientific methodology"). To the extent that an expert witness relies "solely or primarily on experience, . . . the witness must explain how that experience leads to the conclusion reached, why that experience is a sufficient basis for the opinion, and how that experience is reliably applied to the facts." Morris v. Fla. Transformer, Inc., 455 F.Supp.2d 1328, 1331 (M.D. Ala. 2006) (Thompson, J.) (citing Fed. R. Evid. 702 advisory committee's note).10

As for helpfulness, "the court must ensure that the proposed expert testimony is relevant to the task at hand, . . . i.e., that it logically advances a material aspect of the proposing party's case." Allison v. McGhan Med. Corp., 184 F.3d 1300, 1312 (11th Cir. 1999) (citations and internal quotation marks omitted). Additionally, for expert evidence to be helpful to the finder of fact, it must offer insight "beyond the understanding and experience of the average citizen." United States v. Rouco, 765 F.2d 983, 995 (11th Cir. 1985). Although the relevance standard for expert evidence is higher than the liberal admissibility policy set forth in Rules 401 and 402, see Allison, 184 F.3d at 1309-10, Daubert itself makes clear that a court can deem expert evidence sufficiently helpful to be admissible but nonetheless conclude that it is insufficient to create a genuine dispute of material fact for purposes of summary judgment. 509 U.S. at 595-96; see also Hirsch v. CSX Transp., Inc., 656 F.3d 359, 362 (6th Cir. 2011) (same).

One final note bears mention. Here, because plaintiffs rely on the expert evidence of Dr. Burns with respect to their opposition to summary judgment and in support of their motion for class certification, the court is required first, to consider the admissibility of her opinions under Rule 702, and then to determine whether they, in conjunction with plaintiffs' other evidence, are sufficient both to create disputes of material fact regarding the merits of plaintiffs' claims for purposes of Rule 56, and to demonstrate the requirements of certification under Rule 23. As the court has explained previously, these analyses are "quite distinct" and "must be kept separate." Morris, 455 F. Supp. 2d at 1332 (quoting Rudd v. General Motors Corp., 127 F.Supp.2d 1330, 1336 (M.D. Ala. 2001) (Thompson, J.)). In this section of the opinion, the court will address only Rule 702 admissibility. Later in this opinion, it will address the role that this evidence plays in establishing compliance with Rule 23. Finally, in the court's simultaneously issued opinion on summary judgment, it has addressed at some length why this evidence creates material disputes of fact as to the merits of plaintiffs' claims.

2. Dr. Burns

As a preliminary matter, although defendants do not question Dr. Burns's qualifications, the court notes that they are quite impressive.11 Defendants' mental-health expert, Dr. Patterson, has himself recently recognized her to be a "top nationally recognized expert[] in the field of correctional mental health." Rep. on Suicides Completed in the California Dep't of Corr. and Rehab. at 1, Coleman v. Brown, No. 2:90-cv-520 (E.D. Cal. March 13, 2013), ECF No. 4376. She currently serves as the Chief Psychiatrist of the Ohio Department of Rehabilitation and Correction (as she did for a period of four years in the late 1990s). She has also long been familiar with the provision of mental-health care in Alabama's prisons, as she served as an expert in previous litigation that successfully obtained a remedial settlement, Bradley v. Harrelson. See Order Approving Settlement Agreement, Bradley v. Harrelson, No. 2:92-cv-70 (M.D. Ala. June 27, 2001) (Albritton, J.), ECF No. 412.

As to Dr. Burns's methodology, defendants essentially present two different arguments: that she did not see enough firsthand (of 15 major facilities, 102 prisoners), and that she (admittedly) did not randomly select prisoners to interview.12 In other prison cases, a number of courts have rejected Daubert attacks on experts (offered by both plaintiffs and defendants) on the grounds that their samples were non-random, and that they did not consider enough prisoners at enough facilities in order to extrapolate conclusions about the system as a whole. In Ruiz, 37 F. Supp. 2d at 890-92, the court rejected the defendants' arguments that "in the absence of statistical proof of violations . . ., the plaintiffs have not and cannot show system-wide violations," explaining that although "[s]tatistical procedures can increase the court's confidence in making inferences from a given set of data, . . . this is an issue of weight, rather than admissibility of a given methodology. Statistical models are simply not the only method for making general inferences from specific data."

As for non-random sampling, as the court put it in the context of the claims in Ruiz, "[t]he fact that 30 records show excessive use of force does not change because the records were selected non-randomly." Id. at 891. See also Coleman v. Wilson, 912 F.Supp. 1282, 1303 (E.D. Cal. 1995) (Karlton, J.) (rejecting a similar objection to expert declaration that the defendants contended were "unreliable insofar as they are based on medical files `pre-selected' by plaintiffs' counsel"). Another court recently discussed at some length the use of non-random sampling by prison experts. In Dockery v. Fisher, 2015 WL 5737608, at *5 (S.D. Miss. Sept. 29, 2015) (Barbour, J.), the court recognized that non-randomized qualitative research methods are both "accepted and mainstream in the scientific community," and, in the view of some experts, "more applicable to a proper evaluation of the delivery of health care at a prison." The court quoted plaintiffs' expert as explaining that "[w]hen sampling from people (patients, staff) and documents in qualitative research, random samples are to be avoided. Instead, the gold standard for sampling is `judgment sampling' or `purposeful sampling.' Instead of using random number generators to select samples, a judgment sample is chosen based on the expertise and judgment of a subject matter expert with knowledge of the system or process being assessed. The goal is to obtain a sample which is as broad, rich, and representative of the diversity of operational conditions as possible. . . . Judgment samples are appropriate because ensuring that all potential observational units in a population and sampling time frame have equal probability of selection is often not the most desired or beneficial strategy. Rather, we look to the subject matter experts to guide which areas, times of day, or segments of the population are most important to study and understand." Id. at *6 (citation and internal quotation marks omitted). The court also recognized the potential merit of defendants' arguments against this methodological approach, but concluded that this went to the weight to be given the expert evidence, rather than its admissibility. Id.; see United States v. Monteiro, 407 F.Supp.2d 351, 366 (D. Mass. 2006) (Saris, J.) ("It may well be that other methods . . . may prove to be the best method of analysis. However, Daubert and Kumho Tire do not make the perfect the enemy of the reliable; an expert need not use the best method of evaluation, only a reliable one.").

In another recent case about the adequacy of prison health care, Jama v. Esmor Correctional Services., Inc., 2007 WL 1847385, at *26-27 (D.N.J. June 25, 2007) (Debevoise, J.), the court rejected a Daubert challenge to an expert on the grounds that he had reviewed only the 78 medical request forms submitted by the nine plaintiffs, who were among 1600 detainees housed at the facility at issue. Plaintiffs argued that his use of "convenience" rather than random sampling was unreliable and that "his sample size [was] too small to generate reliable statistics," but the court agreed with defendants that these challenges went to the weight of his evidence rather than to its admissibility. The court and others recognized that this sort of sampling is particularly reasonable when it is part of a multifaceted review that considers not only the records and statements of individuals but also other sources such as deposition transcripts and other documents that allows an expert to "draw general conclusions." Id.; see also Parsons, 2014 WL 3721030, at *4.

Defendants cite a number of cases for the proposition that because Dr. Burns cited in her report only 29 examples of different individual prisoners receiving inadequate care or classification, the size of her sample was insufficient to allow her to draw conclusions about systemic inadequacies.13 These cases are, however, distinguishable from the case at bar. In Wal-Mart, the court concluded that 120 anecdotal affidavits describing instances of discrimination at 235 out of 3,400 stores were insufficient to "demonstrate that the entire company operates under a general policy of discrimination." 564 U.S. at 358 (citation and internal quotation marks omitted.) The individual prisoners whose care Dr. Burns discusses make up a significantly larger and more representative portion of the class than did the affiants in Wal-Mart, and, more important, the evidence about them is being used for an entirely different purpose. In Wal-Mart, the affidavits represented only one in approximately 12,500 class members; here, ADOC recognizes as mentally ill by placement on the caseload about 3,500 prisoners, meaning that the 43 prisoners Dr. Burns cites (even setting aside her representation that these are but illustrative examples drawn from notes that include others) represent in the vicinity of one in 80 class members. Moreover, while the Wal-Mart affiants related only to one in every 15 or so stores, Dr. Burns drew from about two-thirds of a much smaller number (15) of major correctional facilities.

Moreover, and critically, the Court in Wal-Mart had already rejected the plaintiffs' expert evidence, concluding that it did not establish that the company operated under a policy of discrimination; in the Court's discussion of the affidavits, it was simply explaining that they did not, on their own, suffice to support the inference that such a policy existed. With respect to all of Dr. Burns's opinions, by contrast, she cites other evidence to establish the existence of a policy or practice (such as statements by prison mental-health practitioners or internal documents). In some cases, the existence of the policy or practice is effectively undisputed; for example, defendants do not disagree that a certain number of practitioners with certain qualifications provide mental-health care to prisoners. What they dispute, and what Dr. Burns cites these prisoners to show, is that these policies and practices place prisoners at a substantial risk of serious harm.14

The court will address the remainder of defendants' case law—offered in support of their contention that Dr. Burns lacks a sufficient basis for her conclusions—more briefly. Although Lloyd Noland Hosp. & Clinic v. Heckler, 762 F.2d 1561, 1568 (11th Cir. 1985), recognized the obvious point that a "sample size [may be] too small" to draw certain conclusions from it, the issue in that case was worlds away: whether the Secretary of Health and Human Services had improperly relied on a study—which stated that "no broad conclusions can be drawn about the nature and extent of medical malpractice in this country" on the basis of the study—to reach a conclusion as to that very issue. Moreover, defendants suggest no basis for comparing the suitability of the sample size in the study at issue (or, indeed, what that sample size was) to that employed by Dr. Burns.

Dukes v. Georgia, 428 F.Supp.2d 1298, 1317 (N.D. Ga. 2006) (Forrester, J.), stands for the similarly unremarkable proposition that an expert may lack "sufficient data and information" to reach a conclusion. But the expert at issue in that case does not appear to have sampled anything; he simply speculated, without much basis, about whether a lab technician should have identified a strain of yeast. The case is hardly on point. Gilliam v. City of Prattville, 667 F.Supp.2d 1276, 1298 (M.D. Ala. 2009) (Fuller, J.), concluded that, even if an unqualified expert had been qualified, his testimony would not have been reliable for numerous reasons, including that he had not examined the autopsies of the individual about whose death he was testifying and had offered no data or explanation to support his theory as to the cause of death or to refute other theories. Again, not on point. Finally, in United States ex rel. Wall v. Vista Hospice Care, Inc., 2016 WL 3449833, at *12 (N.D. Tex. June 20, 2016) (Lynn, J.), the court was required to determine how many false claims had been submitted by physicians, which in turn required assessments of "individual physicians' judgment regarding individual patients." The court concluded (contrary to the view of at least one court in this circuit) that statistical sampling and extrapolation could not be used by experts to establish liability in such a case. Defendants apparently believe this case is relevant because mental-health care providers (obviously) exercise some judgment in caring for individual prisoners. This case is readily distinguishable, for two reasons: First, Dr. Burns has not engaged in statistical sampling or attempted to offer a precise quantitative assessment of whether specific instances of mental-health care were or were not adequate. Second, because it is the Commissioner and Associate Commissioner, not the mental-health practitioners who provide care, who are the defendants, this case does not turn on assessments of the case-specific "subjective clinical judgment" of mental-health staff, but rather on whether, in staffing, funding, and overseeing the operation of the mental-health care system, defendants have been deliberately indifferent to an objective risk of serious harm to mentally ill prisoners.

Defendants also cite a concurring opinion in EEOC v. Freeman, 778 F.3d 463, 469-70 (4th Cir. 2015) (Agee, J.), which expresses the judge's serious concern that the EEOC "continues to proffer expert testimony from a witness whose work has been roundly rejected in our sister circuits." This expert had attempted to offer statistical evidence in support of a disparate impact theory. His work "contained a plethora of analytical fallacies": although he had a complete data set, he opted to "ignor[e]" years of relevant data from more than half of the years and locations at issue, resulting in "an egregious example of scientific dishonesty." Even within the subset of available data he did consider, he had "cherry-picked" the data that supported his conclusions. Dr. Burns, by contrast, is not endeavoring to offer statistical evidence, and opted for a sampling method that other courts have recognized as appropriate for an expert in her field, and that was especially reasonable in light of the restrictions on her ability to gather data.15 She has considered all of the data she obtained in reaching her conclusions. Had Dr. Burns been provided access to, and had it been feasible to review, a dramatically larger sample, and had she then opted to base her assessments only on a small subset of those records and interviews, the concerns of defendants might be better founded.

With respect to defendants' contention that Dr. Burns did not tour all of defendants' facilities, the court agrees with another court recently to consider this very issue that "otherwise admissible testimony based on investigation of some facilities but not all is still probative to some extent and any limitation goes to the weight of the opinions." Parsons, 2014 WL 3721030, at *4. Moreover, as Judge Wake noted in that case, the number of tours plaintiffs' experts were allowed to complete, as well as the length of time they were allowed to spend in each facility and the number of prisoners whose records they were allowed to review and whose interviews they were permitted to conduct was circumscribed by defendants during the course of an extremely lengthy and contentious discovery mediation. See id. ("Defendants' strenuous objections to the burdensome nature of the multiple prison tours, [and] the Court's reduction of that burden by limiting the number of tours . . . weigh[s] against complete exclusion of the opinions.") The fact that Dr. Burns also reviewed policies and quality assurance documents relevant to other facilities, and that she reviewed the depositions of those responsible for providing or overseeing care at other facilities, further weighs in favor of considering her opinions even as to facilities she did not personally visit.16

As a last methodological point, defendants take issue with Dr. Burns's failure to review the entirety of the medical records of all of the prisoners she interviewed, and contend that this reflects a failure to confirm prisoners' self-reports. First of all, defendants concede that she did review the records of many of those she discusses specifically in her report. Furthermore, she may well have been able to reach reliable conclusions with respect to some issues without reviewing medical records. For example, to the extent that she disagreed with the classification of a severely mentally ill prisoner as an outpatient, she could well have both assessed the apparent severity of that prisoner's illness and determined that he was currently classified (and housed) as an outpatient without reviewing any documents. As another example, Dr. Burns described in her report observing prisoners who had been prescribed certain antipsychotics who exhibited movement disorders that are characteristic side effects of these medications. She is a psychiatrist who is capable of recognizing such a visible disorder without reference to records. Finally, although bald reliance on the totally uncorroborated statement of a single mentally ill individual might in some instances be questionable, Dr. Burns often describes statements by prisoners from her interviews as confirmatory of information she obtained from other sources, including directly from MHM. This is, therefore, not a basis to exclude her opinion, although defendants are free to cross-examine her on this point in order to undermine her conclusions at trial. See Allison, 184 F.3d at 1311 (explaining that Daubert analysis is "not intended to supplant the adversary system").

In sum, the court concludes that Dr. Burns's methodology is sufficiently reliable, and that her opinions are based on a sufficient quantity, and range of sources of, data, to be admissible as expert evidence under Daubert and Federal Rule of Evidence 702.

As for helpfulness, defendants insist that there is a "fit" problem between Dr. Burns's report and the claims in this case, because her reports states that various policies and practices subject prisoners to a "risk of harm"; they object to her failure to use the catchphrase from the case law, "substantial risk of serious harm." But as Dr. Burns herself pointed out in her deposition, she has offered evidence helpful to the court in making this ultimate legal determination. She has described in some detail the harms that she believes will, and in some instances did, result from defendants' policies and practices; it is the court's task to determine whether these harms are "serious," based on case law. Indeed, as Dr. Burns is not a legal expert and would not be applying this case law, any assessment she made as to the seriousness of these harms would not be very helpful.17 See Burkart v. Washington Metro. Area Transit Auth., 112 F.3d 1207, 1213 (D.C. Cir. 1997) ("Each courtroom comes equipped with a `legal expert,' called a judge. . . .").

Similarly, it is up to the court to determine whether the risk is sufficiently "substantial" to constitute an Eighth Amendment violation.18 Dr. Burns has offered evidence going to the incidence of the harms she has described; although her statements to this effect are not statistically precise, they are nonetheless very helpful. For example, she relies on chart reviews, interviews with prisoners, depositions of providers, and MHM quality assurance documents to conclude that for prisoners at a number of the facilities she toured, "virtually the only treatment being provided is psychotropic medication," that any individual psychotherapy being provided was infrequent and brief,19 and that very few or no outpatient therapy groups were being offered. Burns Report (doc. no. 868-2) at 38-39. In this instance, the incidence of harm is in her opinion extremely high. As another example, Burns opines that mental-health staff continue to prescribe certain long-acting antipsychotic injections even when they cause serious side effects such as severe restlessness, painful muscle spasms, and irreversible movement disorders; she reached this conclusion based on chart reviews and interviews with and observations of prisoners. Although her report does not reveal exactly what percentage of mentally ill prisoners are prescribed these medications, she does helpfully inform the court that "[m]any of the inmates interviewed displayed these types of movement disorders, but their prescriptions were continued rather than changed to medications less likely to cause these problems." Id. at 42. In addition, she provides six examples of prisoners who suffered in this particular manner. Id. at 42-43. As a third example, Dr. Burns opines that prisoners with mental illness are significantly underidentified; she believes that this underidentification stems from a deficient screening process that relies on unqualified and unsupervised staff to conduct evaluations. Although she does not state exactly what proportion of the prisoners with mental illness in ADOC custody she believes have not been identified, she does offer a number of statistics from other state prison systems; there is a range, but even the most conservative comparator statistics suggest substantial underidentification. Id. at 24-26.

As a final iteration of this argument, defendants contend that Dr. Burns is merely opining as to what care she thinks is good care or what care should be provided, and that this is not what the Constitution requires defendants to provide. Although defendants are quite correct that expert opinions as to desirable prison conditions are insufficient to establish constitutional minima, see Rhodes v. Chapman, 452 U.S. 337, 348 n.13 (1981), their argument badly misconstrues the evidence Dr. Burns is offering. Although she does express some opinions about how care should be provided (based, for example, on the fact that certain types of practitioners are "unqualified" to provide care without supervision, see Burns Report (doc. no. 868-2) at 16), she then goes on to explain how the failure to provide care in this way causes harm to prisoners. For example, she presents evidence, discussed above, to support her conclusion that these unqualified and unsupervised staff frequently fail to diagnose mental illness.

The court concludes that Dr. Burns's expert evidence will be helpful to the court in assessing whether the mental-health care provided to prisoners in Alabama falls below the constitutional floor, because she will bring to bear her experience, and the results of her investigation in this case, to help the court to understand the seriousness of the risk of harm posed by the challenged policies and practices. See Coleman, 912 F. Supp. at 1304 (rejecting a challenge that the magistrate judge had "improperly relied on expert testimony to establish constitutional minima").

Lastly, defendants attack Dr. Burns's qualifications as well as her methodology with respect to one fairly narrow issue: her ability to offer an opinion regarding the adequacy of correctional staffing. As for her qualifications, they note that she admitted that she has never been responsible for staffing a correctional facility with custody staff or received any training in doing so. As for her methodology, they note that she testified in her deposition that she did not perform (as part of her assignment in this case) any sort of staffing analysis or review for adequacy the numbers of correctional officers in any facility for any shift. The court is not convinced by the argument regarding methodology; it seems clear that an expert might reliably assess a staffing level to be too low and therefore likely to cause harm without determining what staffing level would be appropriate. However, the court is more troubled by the fact that although Dr. Burns is an esteemed expert in correctional mental health, plaintiffs have not demonstrated that her experience entitles her to opine on this separate, albeit related, issue. (Plaintiffs' response, that she has been involved in "evaluating and informing internal operations of correctional systems" through her role as a correctional mental-health administrator, is not convincing.) The court will therefore not rely on her opinion on this specific issue for purposes of the pretrial motions.20 That all said, plaintiffs have in fact offered another expert, Eldon Vail, who is an experienced correctional administrator; defendants have not raised a Daubert challenge to his qualifications to address this issue, and the court has already rejected their contention that he lacks an adequate basis for his conclusions. His opinion on the inadequacies of correctional staffing and their effects on delivery of mental-health care is more detailed and better supported than Dr. Burns's opinion on this point, though essentially in accord; the court will therefore consider his evidence instead of hers.

IV. Analysis

Federal Rule of Civil Procedure 23 "establishes the legal roadmap courts must follow when determining whether class certification is appropriate." Valley Drug Co. v. Geneva Pharm., Inc., 350 F.3d 1181, 1187 (11th Cir. 2003). The party seeking certification bears the burden of demonstrating that all of the requirements of Rule 23 have been met. Id.

In order for any certification motion to succeed, the putative class representatives must show that "(1) the class is so numerous that joinder of all members is impracticable; (2) there are questions of law or fact common to the class; (3) the claims or defenses of the representative parties are typical of the claims or defenses of the class; and (4) the representative parties will fairly and adequately protect the interests of the class." Fed. R. Civ. P. 23(a). In addition, a class must clear one of three additional hurdles: because the named plaintiffs in this case seek certification of a Rule 23(b)(2) class, they must also show that "the party opposing the class has acted or refused to act on grounds that apply generally to the class, so that final injunctive relief or corresponding declaratory relief is appropriate respecting the class as a whole." Fed. R. Civ. P. 23(b)(2).

The court's role at the class-certification stage is not to decide the underlying claims, but rather to determine whether the requirements for certification are met. See Eisen v. Carlisle & Jacquelin, 417 U.S. 156, 177-78 (1974). "Merits questions may be considered to the extent—but only to the extent—that they are relevant to determining whether the Rule 23 prerequisites for class certification are satisfied." Amgen, 133 S. Ct. at 1194-95 (citation omitted). A party seeking class certification must "affirmatively demonstrate his compliance with the Rule"; that is, plaintiffs must offer evidence sufficient to satisfy the court that the various requirements of Rule 23 have been met. Wal-Mart, 564 U.S. at 350.

Commonality, in particular, "cannot be determined without a precise understanding of the nature of the underlying claims," and a careful consideration of the evidence offered in support of the common thread, but "[o]f course, this does not mean that the plaintiffs must show at the class certification stage that they will prevail on the merits" of those claims. Parsons v. Ryan, 754 F.3d 657, 676 & n.19 (9th Cir. 2014) (citing, inter alia, Messner v. Northshore Univ. HealthSystem, 669 F.3d 802, 811 (7th Cir. 2012) ("[T]he court should not turn the class certification proceedings into a dress rehearsal for the trial on the merits.")).

Having articulated the framework for its analysis, the court will proceed to consider each requirement for certification in turn.

1. Standing

"[A]ny analysis of class certification must begin with the issue of standing"; only once the court finds that the named plaintiffs have standing may it consider whether they have "representative capacity, as defined by Rule 23(a), to assert the rights of others." Griffin v. Dugger, 823 F.2d 1476, 1482 (11th Cir. 1987). To show Article III standing, the named plaintiffs must show that they have been injured, that their injuries are fairly traceable to the defendants' conduct, and that a judgment in their favor would likely redress their injuries. See Mulhall v. UNITE HERE Local 355, 618 F.3d 1279, 1286 (11th Cir. 2010).

As defendants acknowledge, their arguments that the named plaintiffs lack standing to pursue the claims for which they seek certification "directly intersect[] and coincide[] with [their argument in their] pending Motion for Summary Judgment"—that is, defendants argue that they lack standing because their claims are moot, unexhausted, precluded or barred by the statute of limitations.21 Defs.' Opp. to Class Cert. (doc. no. 807) at 104. The court has engaged in an extensive discussion of all of these arguments in the course of its opinion denying defendants' motion for summary judgment. It rejected all of them except to the extent that it agreed that a few of the named plaintiffs' claims have become moot as a result of their releases from prison. As these plaintiffs have been dismissed from the case, they will not be certified as class representatives. Otherwise, defendants' arguments that the named plaintiffs lack standing to bring the claims they seek to have certified are rejected for the reasons set forth in the summary-judgment opinion.

2. Rule 23(a)

A. Numerosity

Rule 23(a)(1)'s requirement of numerosity is satisfied if joinder—the usual method of combining similar claims—would be impracticable. Although there is no strict threshold, classes containing more than 40 members are generally large enough to warrant certification. See, e.g., Cox v. Am. Cast Iron Pipe Co., 784 F.2d 1546, 1553 (11th Cir. 1986); see also Rubenstein, Newberg on Class Actions § 3.12 (5th ed.).

"Moreover, the fluid nature of a plaintiff class—as in the prison-litigation context—counsels in favor of certification of all present and future members." Henderson v. Thomas, 289 F.R.D. 506, 510 (M.D. Ala. 2012) (Thompson, J.) (citing Kilgo v. Bowman Transp., Inc., 789 F.2d 859, 878 (11th Cir. 1986) (affirming a certified class of 31 present members as well as future members who could not be identified); Green v. Johnson, 513 F.Supp. 965, 975 (D. Mass. 1981) (Freedman, J.) (finding numerosity after considering "the fact that the inmate population at these facilities is constantly revolving")); see also Pederson v. Louisiana State Univ., 213 F.3d 858, 868 n.11 (5th Cir. 2000) ("[D]istrict courts must not focus on sheer numbers alone. . . . [T]he fact that the class includes unknown, unnamed future members also weighs in favor of certification."); Reid v. Donelan, 297 F.R.D. 185, 189 (D. Mass. 2014) (Ponsor, J.) (certifying a class of detained plaintiffs and explaining that the numerosity threshold may be relaxed "when a party seeks only declaratory or injunctive relief, since the inclusion of future members increases the impracticability of joinder" (citing McCuin v. Sec'y of Health & Human Servs., 817 F.2d 161, 167 (1st Cir. 1987)); Rubenstein, Newberg on Class Actions § 3.15 (5th ed.) (explaining that the inclusion of future class members "may make class certification more, not less, likely" and citing two decisions certifying classes of prisoners, Hill v. Butterworth, 170 F.R.D. 509, 514 (N.D. Fla. 1997) (Paul, J.) ("This Circuit has held [that when] the alleged class includes future [members], necessarily unidentifiable[,] . . . the requirement of Rule 23(a)(1) is clearly met, for joinder of unknown individuals is clearly impracticable." (citation and internal quotation marks omitted)); Clarkson v. Coughlin, 145 F.R.D. 339, 346 (S.D.N.Y. 1993) (Sweet, J.) ("The class action device is particularly well-suited in actions brought by prisoners due to the fluid composition of the prison population. . . . Class actions therefore generally tend to be the norm in actions such as this." (citations and internal quotation marks omitted))).22

Plaintiffs have submitted evidence to show that in February 2016, there were 3,416 prisoners on the mental-health caseload (and argue that this figure significantly understates the number of seriously mentally ill prisoners in the State). This exceeds the customary numerosity threshold by a factor of 85. Defendants nonetheless have a bone to pick. They contend that "this number . . . says nothing about how many individuals can claim an actual injury traceable to the State's policies and procedures." Defs.' Opp. to Class Cert. (doc. no. 807) at 141. There are a number of problems with this argument.

First, courts look at numerosity in a fairly straightforward fashion: by assessing the practicability of joinder, in light of the number of people who fall within the definition of the class. Movants for class certification do not need to present evidence showing—one by one, many times over—that individual putative class members can proceed on the class claims; requiring as much would largely defeat the efficiency benefits of class-wide adjudication, and is unnecessary in light of the commonality requirement. See Rubenstein, Newberg on Class Actions § 3:11 ("[W]here joinder is impracticable, judicial economy weighs in favor of representative litigation of common issues for similarly situated plaintiffs."). Instead, they need to show that the class representatives can proceed on claims which are common to the class.

Plaintiffs' evidence regarding the size of the mental-health caseload goes directly to—indeed, conclusively resolves—the question whether more than 40 or so prisoners in ADOC custody have serious mental illnesses. And, again, the fact that the class contains as-yet-unknown future members also counsels strongly in favor of finding that Rule 23(a)(1) has been satisfied.

Second, even if the court were required to consider how many prisoners "can claim an actual injury traceable to the State's policies and procedures," rather than how many prisoners in ADOC custody have a serious mental illness, the court would reiterate that the actual injury being claimed in this case is exposure to the substantial risk of serious harm stemming from those policies and procedures. If, as plaintiffs claim, defendants provide inadequate numbers of mental-health care providers and thereby subject all prisoners with serious mental illnesses to a substantial risk of serious harm, then all class members can claim an actual injury—even those who have not yet suffered serious harm.

Finally, and for the sake of thoroughly responding to a thoroughly unconvincing argument, the court notes that plaintiffs have indeed presented evidence (some of it not subject to any direct rebuttal) to show that numerous individuals have in fact already suffered a range of serious harms. The reports of their experts are filled with examples of prisoners who the experts opine have suffered serious harm after receiving care that was inadequate due to the failure of defendants to provide enough funding, staffing, and oversight.23 A selection of these concrete harms includes: being denied necessary treatment other than psychotropic medication, being maintained on certain psychotropic medications despite suffering serious side effects and the availability of alternative medications, frequently missing doses of psychotropic medications or having them inappropriately or abruptly discontinued, being denied follow-up care after attempting to commit suicide, and being held in segregation without regard to the psychological deterioration it is causing.

Having addressed all of defendants' objections, the court finds that plaintiffs have met the numerosity requirement for the Eighth Amendment class.

The court also finds that the plaintiffs have met the numerosity requirement for the procedural dueprocess class. Plaintiffs have submitted undisputed evidence showing that about 70 to 80 individuals were subject to involuntary-medication orders as of March who have survived summary judgment would arguably be sufficient, on its own, to establish numerosity. 2016; therefore, this class—of prisoners subject to involuntary-medications orders—meets the presumptive numerosity threshold discussed above.

B. Commonality

i. Eighth Amendment Claim

Rule 23(a)(2) requires named plaintiffs seeking class certification to show that "there are questions of law or fact common to the class." In Wal-Mart, the Supreme Court explained that "this does not mean merely that they have all suffered a violation of the same provision of law. . . . [Rather,] [t]heir claims must depend upon a common contention . . . [which] must be of such a nature that it is capable of classwide resolution—which means that determination of its truth or falsity will resolve an issue that is central to the validity of each one of the claims in one stroke. What matters to class certification . . . is not the raising of common `questions'—even in droves—but, rather the capacity of a classwide proceeding to generate common answers apt to drive the resolution of the litigation." 564 U.S. at 350 (citation and internal quotation marks omitted). In short, commonality requires a showing that there is "some glue" holding the claims together. Id. at 352. However, plaintiffs seeking to demonstrate commonality under Rule 23(a)(2) need not show that common questions "predominate" over individual questions as required under Rule 23(b)(3); indeed, "even a single common question will do." Wal-Mart, 564 U.S. at 359 (citations and alterations omitted).

In Wal-Mart, the plaintiffs failed to satisfy the commonality requirement because they had not offered "significant proof" of a policy of discrimination that applied across the corporation's numerous stores.24 Id. at 353. Here, plaintiffs have identified eight different specific policies or practices, and offered significant proof that they are common to the class of prisoners in Alabama with serious mental illnesses.25 This is all plaintiffs need to do for purposes of class certification; they need to show that the policies and practices they challenge are common, not (yet) that the common policies and practices are unconstitutional. Put differently, they need to demonstrate the existence of common questions with common answers, not what those common answers are. See Amgen, 133 S. Ct. at 1194-95 (holding that merits questions are only to be considered to the extent they are relevant to Rule 23 determination).

The court will proceed as follows: First, it will discuss what must be shown to demonstrate that the policy or practice is common, generally and in the context of an Eighth Amendment substantial-risk-of-harm claim such as the one brought here. Second, it will discuss the evidence plaintiffs have offered to show that policies and practices they have challenged are indeed common to the class. Third and last, the court will address defendants' arguments that dissimilarities among class members defeat commonality.

As a general matter, a policy or practice need not affect every member of a class in order to be "common." See Rubenstein, Newberg on Class Actions § 3.23 (5th ed.) ("[T]he Fifth Circuit has held that questions affecting `most,' `a substantial number,' or `a significant number' of the class members will qualify as `common' questions. This question has rarely been addressed outside the Fifth Circuit though courts that have addressed it have adopted the Fifth Circuit's approach.").

Moreover, "class members can assert such a single common complaint even if they have not all suffered actual injury; demonstrating that all class members are subject to the same harm will suffice." Baby Neal ex rel. Kanter v. Casey, 43 F.3d 48, 56 (3d Cir. 1994) (emphasis in original) (citing Hassine v. Jeffes, 846 F.2d 169, 177-78 (reversing the denial of certification of a class of prisoners challenging the risks posed by unsafe prison conditions)). As another district court recently explained in a jail-conditions case, "the Fifth Circuit, like the Third, [has] expressly disagreed with the proposition that a policy must injure each class member to provide the foundation for class wide relief." Jones v. Gusman, 296 F.R.D. 416, 465-66 (E.D. La. 2013) (Africk, J.) (citing M.D. ex rel. Stukenberg v. Perry, 675 F.3d 832, 847-48 (5th Cir. 2012)). Instead, in M.D., the court of appeals explained that a valid claim for class-wide relief could be "based on an allegation that the [defendant] engages in a pattern or practice of agency action or inaction—including a failure to correct a structural deficiency within the agency, such as insufficient staffing—with respect to the class," even if plaintiffs do not contend that every member of the class has suffered an actual injury. 675 F.3d at 847-48 (internal citation and quotation marks omitted).

This case law supports a finding of commonality here, because being subjected to a substantial risk of serious harm is an actionable constitutional injury, even when a prisoner's physical or mental condition has not yet been detrimentally impacted. See Parsons, 754 F.3d at 681-83 (citing numerous cases, both within and outside the Eighth Amendment context, in which courts have found the commonality requirement satisfied, post-Wal-Mart, based on a systemic policy or practice that allegedly exposed the members of the class to an unconstitutional risk of harm); see also Dockery, 2015 WL 5737608, at *16-17 (recognizing as common questions "whether [the challenged] conditions and health care have either subjected prisoners to an unconstitutionally unreasonable risk of harm or, conversely, were sufficient to provide humane conditions of confinement"); Scott v. Clarke, 61 F.Supp.3d 569, 587 (W.D. Va. 2014) (Moon, J.) ("Plaintiffs allege that the policies and practices at FCCW—e.g., the defective sick call process; FCCW's refusal to refer or undue delay in referring prisoners for specialized care; the failure to maintain continuity in the provision of prescribed, potentially life-sustaining medications; etc.—reflect substandard medical care on the part of the Defendants. Whether these policies and practices place the Plaintiffs and other current and future FCCW prisoners at a substantial risk of serious harm to which the Defendants are deliberately indifferent implicates questions of fact and law common to the entire putative class."); Jones, 296 F.R.D. at 465-67 ("Whether certain conditions at OPP either by themselves, or through a mutually enforcing effect, put inmates at a substantial risk of harm is amenable to a common answer. . . . Similarly, whether OPP officials have been deliberately indifferent to any such risk can be demonstrated in a manner that is applicable to all class members. . . . Class Plaintiffs' Eighth and Fourteenth Amendment medical and mental-health care claims [likewise] warrant certification. . . . Class Plaintiffs have identified discrete and particularized practices including, for example, medication and suicide prevention practices, as well as staffing inadequacies, that are mutually enforcing causes of OPP's deficient conditions. . . . Accordingly, a class action is an appropriate vehicle for these claims." (citations and internal quotation marks omitted)); Hughes v. Judd, 2013 WL 1821077, at *9, *23-24 (M.D. Fla. Mar. 27, 2013) (Pizzo, M.J.) (finding commonality and certifying a class with respect to a Fourteenth Amendment risk-of-harm claim challenging a sheriff's failure to address a high level of violence among juvenile detainees and practice of using pepper spray in response to this violence), report and recommendation adopted as modified on other grounds, 2013 WL 1810806 (M.D. Fla. April 30, 2013) (Merryday, J.); M.D. v. Perry, 294 F.R.D. 7, 45 (S.D. Tex. 2013) (Jack, J.) (finding commonality and certifying a class with respect to a Fourteenth Amendment risk-of-harm claim challenging a policy or practice of allowing caseworkers in a foster care system to carry excessive caseloads that posed a risk of harm to children).

The fact that not every mentally ill prisoner is subjected to precisely the same level of risk also does not defeat commonality. Indeed, the last time a judge in this court heard a "systemic attack on the way that mental health care is provided to acutely and seriously mentally ill inmates in the Alabama prison system," it explained that "[t]hough there certainly may be some factual differences between the individual class members and the nature and severity of their illness, such individual differences do not defeat certification because there is no requirement that every class member be affected by the institutional practice or condition in the same way." Bradley v. Harrelson, 151 F.R.D. 422, 426 (M.D. Ala. 1993) (Albritton, J.); see also Parsons, 754 F.3d at 678-80 ("[A]lthough a presently existing risk may ultimately result in different future harm for different inmates—ranging from no harm at all to death—every inmate suffers exactly the same constitutional injury when he is exposed to a single statewide ADC policy or practice that creates a substantial risk of serious harm. . . . Even if some inmates are exposed to a greater or idiosyncratic risk of harm by the policy and practice of not hiring enough staff to provide adequate medical care to all inmates, that single policy and practice allegedly exposes every single inmate to a serious risk of the same basic kind of harm." (internal citations omitted)).

Plaintiffs have offered more than adequate evidence to establish the existence of the common policies and practices they challenge.26 This evidence, discussed in the court's opinion on summary judgment, could not be more different from the indefinite and speculative evidence offered by the expert in Wal-Mart, of a "culture" of discrimination manifested in millions of discretionary decisions. 564 U.S. at 353-54. Here, the experts point to very concrete policies and practices—such as using unsupervised and unqualified nurses to conduct intake screenings, or placing prisoners with serious mental illness in prolonged segregation—that undisputedly occur or that plaintiffs have offered considerable evidence (often in the form of admissions by MHM administrators) to show are widespread. See Parsons, 754 F.3d at 678 ("The putative class and subclass members thus all set forth numerous common contentions whose truth or falsity can be determined in one stroke: whether the specified statewide policies and practices to which they are all subjected by ADC expose them to a substantial risk of harm. . . . These policies and practices are the `glue' that holds together the putative class and the putative subclass; either each of the policies and practices is unlawful as to every inmate or it is not."). The only genuine debate in this case is as to the effects of these policies and practices—that is, whether or not they pose a substantial risk of serious harm. But this is not a question for the class-certification stage.

Instead of rehashing plaintiffs evidence in a second opinion, the court will simply note that plaintiffs' expert, Dr. Haney, reiterates at the conclusion of his report that the problems that he discussed in his reports were widespread and systemic. Unlike the expert in Wal-Mart, his conclusions are based not on untethered theorizing, but on a large quantity of concrete evidence. He explains as follows:

Haney Report (doc. no. 868-4) at 158-59.

As discussed at much greater length in the court's opinion denying summary judgment, Dr. Burns, plaintiffs' other mental-health expert, concurs, concluding that the "deficiencies in ADOC—inadequate staffing levels, qualifications, identification and classification, treatment and oversight of the mental health care—deny prisoners care for their serious mental illness leading to needless pain, suffering, self-injury, suicide and punishment for symptoms of untreated mental illness. These are systemic problems that can and should be addressed through changes to the mental health care delivery system." Burns Report (doc. no. 868-2) at 52.

Even Dr. Patterson, defendants' own expert, recognizes that a number of the policies and practices being challenged are pervasive, including that the "staffing of the facilities is insufficient and a significant number of the mental health staff are unlicensed practitioners" who had not received any documented supervision. Patterson Report (doc. no. 679-9) at 47. Moreover, he recognizes that other policies and practices, and concrete harms, stem from this source. For example, he observes "certain deficiencies in adequately identifying inmates during the reception and intake processes that are in need of mental health services, and therefore the numbers of inmates in need of mental health services are . . . underestimated and reflect the need for increased numbers of and properly trained and credentialed mental health staff, as well as the need for supervision by registered nurses during the intake process to identify those inmates in need of further evaluation for mental health services." Id. at 47.

Dr. Patterson is not merely suggesting that he found a few—or even a lot—of prisoners who were not identified; he is explaining that there is a consistent pattern of failure attributable to a very specific set of decisions about system-wide staffing. He agrees, then, that insufficiency of mental-health staffing is a pervasive problem, and that this causes prisoners not to be identified as in need of treatment, across the board. (Of course, not every mentally ill prisoner is missed; rather, every mentally ill prisoner is at risk of being missed.) It is thus undisputed—by the parties' experts if not by the parties—that at least one of the policies or practices challenged by plaintiffs is common to the class.27

Whether this policy or practice of understaffing subjects mentally ill prisoners to a substantial risk of serious harm is one question (of a number of them) apt to drive the resolution of this litigation. Unlike in Wal-Mart, the plaintiffs here have offered considerable evidence to show that this policy or practice is one of the reasons (indeed, the primary reason) that they are at serious risk of receiving inadequate care. This evidence is not merely of a "common pattern or practice," as in an oft-repeated violation, but rather of a "common answer to the crucial question why" prisoners have been put at a substantial risk of serious harm, Wal-Mart, 564 U.S. at 352: because ADOC provides inadequate numbers of and insufficiently qualified mental-health staff, resulting in, among other things, a failure to identify or recognize the acuity of mental illness in many prisoners. Plaintiffs here have done exactly what defendants accuse them of failing to do. Instead of attempting to distill "constellation of disparate but equally suspect practices" from varying experiences of class members, they have "identif[ied] the policy or custom they contend violated the [law] and then establish[ed] that the policy or custom is common to the class." Lightfoot, 273 F.R.D. at 326; see also Parsons, 754 F.3d at 673 (agreeing with the district court that "the problems identified in the provision of health care are not merely isolated instances but, rather, examples of systemic deficiencies that expose all inmates to a substantial risk of serious harm," because the evidence of the existence of a number of specific common policies and practices "suggests that the root cause of the injuries and threats of injuries suffered by Plaintiffs is the systemic failures in the provision of health care generally"); Jones, 296 F.R.D. at 466 ("The facts and law also demonstrate that Class Plaintiffs' Eighth and Fourteenth Amendment medical and mental health care claims warrant certification. These claims do not allege amorphous systemic deficiencies. Class Plaintiffs have identified discrete and particularized practices including, for example, medication and suicide prevention practices, as well as staffing inadequacies, that are mutually enforcing causes of OPP's deficient conditions. Accordingly, a class action is an appropriate vehicle for these claims." (citations and internal quotations marks omitted)).

Rather than discussing each of the remaining policies and practices at length, the court will merely discuss another example. With respect to the policy or practice of "placing prisoners in segregation without regard to its harmful effects on their mental health," the pertinent question is whether defendants treat mentally ill prisoners alike in segregating them, not whether this creates a substantial risk of serious harm to such prisoners. Although Dr. Morgan disagrees with Dr. Haney regarding the effects of this practice, and opines that the mental-health monitoring and treatment provided to prisoners in segregation is adequate, he at no point offers any evidence to suggest that the question—whether or not these policies and practices create a substantial risk of serious harm—cannot be adjudicated on a class-wide basis.28

This distinguishes the case at bar from Sher v. Raytheon Co., wherein the Eleventh Circuit reversed a grant of class certification and remanded for reconsideration because the district court had not "evaluat[ed] and weigh[ed] conflicting expert testimony," and "decline[d] to declare a proverbial, yet tentative winner" between these experts, at the class-certification stage. 419 F. App'x 887, 890 (11th Cir. 2011). In Sher, a case about release of toxins into groundwater, the plaintiffs attempted to show that they met the requirements for certification under Rule 23(b)(3) by presenting an expert who said that he could map the underground area affected and create a mathematical model to show how much the value of each surrounding property had diminished without requiring any individualized consideration; defendants' experts, by contrast, concluded that the identified area included properties with no contamination and that the mathematical model would not obviate the need for individualized determinations of value diminution. Id. at 889. Because the experts disagreed about whether the plaintiffs met the requirements of Rule 23, such that class-wide resolution was appropriate, it was necessary to "declare a . . . winner." Id. at 890; see also Ellis, 657 F.3d at 983-84 (holding that the district court should have resolved a dispute between the parties' experts as to whether gender disparities in promotion existed across the country or only in certain areas, because plaintiffs would be unable to show a common policy or practice unless they could show consistent nationwide discrimination). Here, Dr. Morgan does not suggest that some identifiable subset of mentally ill prisoners is not at risk of placement in segregation (if he did, the court would need to weigh his evidence to determine whether plaintiffs had shown commonality); instead, Dr. Morgan simply disagrees that the practice of placing mentally ill prisoners in segregation places them at a substantial risk of serious harm.

Before concluding this discussion of commonality, the court will address defendants' contention that commonality is defeated by variations between the conditions in different facilities and the differing mental-health needs of members of the class. Although it is true that there are some differences between the various men's facilities,29 in terms of the populations incarcerated there and the ways that certain forms of mental-health care are delivered (as an example, not all major facilities have residential treatment units), and in terms of the levels of staffing (some facilities have some psychiatrist coverage while others have none at all, and some facilities have especially severe shortages of custodial staff while others have somewhat more moderate shortages),30 this does not defeat commonality for two principal reasons.

First, plaintiffs' experts have cited evidence showing, and concluded, that the practices at issue are fairly consistent across the major facilities they inspected, including all of the major hubs for treatment of prisoners with mental illness; although there were slight variations, they did not identify any men's facilities as positive outliers where the system for providing mental-health care was appreciably different (or where adequate mental-health care was being provided).

Second, and perhaps more important, the named plaintiffs' records clearly reflect that they are transferred very frequently between a large number of different facilities; it is fair to presume that the other class members are too.31 In that sense, defining a subclass to include only those who are or will be housed at a subset of facilities would be a semantic distinction without much if any practical difference. The claim in this case revolves around a prospective risk of harm; even if some aspect of the mental-health care at a particular facility is idiosyncratically better or worse, it is clear that numerous class members not currently housed at that facility are likely to be exposed to it in the future.

Another comparison to Wal-Mart is instructive. In Wal-Mart, the fact that the class members worked at thousands of stores across the country made it impossible to show the existence of a common policy of discrimination based on discrimination at a subset of the stores. 564 U.S. at 356-58. But the female employees of Wal-Mart were not frequently moved to numerous different stores in different regions of the country at the whim of the defendant corporation. As the court pointed out, a woman working in Missoula, Montana, would be totally unaffected by discriminatory promotion practices at the stores in say, New Mexico. When a prisoner is in ADOC custody, however, he is subject to involuntary transfer at any time to another facility in the system, and therefore subject to the risks of harm that attend placement at those other facilities. Cf. Henderson, 913 F. Supp. 2d at 1285-86 (holding that the claims of a prisoner who was transferred to a different facility were not moot because there was "more than a reasonable basis to believe that she will be subjected to segregation in [the same facility] again").

Moreover, litigation of these system-wide issues would not preclude sensitivity to any salient differences between facilities in crafting a remedial order in the event that the court does find liability.32 The remedial order the Supreme Court affirmed in Brown v. Plata, 563 U.S. 493 (2011), for instance, ordered the California prison system to reduce overcrowding (which was resulting in unconstitutional medical care) to a certain percentage of design capacity across the State, while allowing it flexibility to adjust for the distinctive features and needs of different institutions by having some facilities be more crowded and some less. Coleman v. Schwarzenegger, 922 F.Supp.2d 882, 970 n.64 (E.D. Cal. 2009), aff'd sub nom. Brown v. Plata, 563 U.S. 493 (2011).

However, as noted above, the court has decided to define the class to exclude prisoners at Tutwiler Prison for Women. Although there is expert evidence regarding the mental-health care being provided at Tutwiler, and although many of the same policies and practices apply at that facility, there is no female class representative; unlike male prisoners who are frequently transferred between facilities, there are no transfers between Tutwiler and the rest of the men's facilities, and no named plaintiff will ever be incarcerated at Tutwiler. The court therefore concludes that it would be more appropriate for the adequacy of mental-health care at Tutwiler to be litigated by way of ADAP's associational standing.33

Defendants also argue that differences in the health care needs of class members preclude a finding of commonality. For this proposition, they rely on language from a Third Circuit decision, Rouse v. Plantier, 182 F.3d 192, 199 (3d Cir. 1999):

But Rouse was an appeal from a denial of qualified immunity (rather than an appeal from a class-certification decision)34 with respect to class claims for damages stemming from inadequate treatment in the past (rather than, as here, a purely prospective injunctive-relief class action alleging a current substantial risk of serious harm). In light of the posture of the case, the court's instruction to the district court, to focus on differences in the medical needs of different subsets of the plaintiff claims, is totally unremarkable: if the care provided in the past was adequate for, say, `stable' diabetics, they would not have suffered any retrospective constitutional violation, and would not be entitled to any damages. But here, such parsing is neither necessary nor appropriate: there are no damages to be apportioned, and, more important, even if a given class member admits that he has not in the past received care that violated the Constitution, he is not precluded from raising a prospective Eighth Amendment claim that defendants are currently subjecting him to a substantial risk of serious harm. Also, as a factual matter, the allegations in Rouse involved much more fine-grained aspects of care for a particular disease (for example, whether monitoring of blood sugar a certain number of times per day was sufficient), whereas the allegations in this case involve overarching inadequacies of staffing which, plaintiffs' experts have opined, contribute to a range of other systemic inadequacies that affect mentally ill prisoners irrespective of their particular diagnoses.

Finally, the court notes that it contemplated certifying subclasses, such as a subclass of those mentally ill prisoners receiving psychotropic medications, or a subclass of those held in segregation. However, the evidence in the record reflects that about two-thirds of mentally ill prisoners receive such medication, and that mentally ill prisoners who are not taking medication are frequently urged to do so (indeed, plaintiffs have offered evidence to suggest that prisoners who refused medication have sometimes been threatened with removal from the mental-health caseload). The record also reflects that prisoners with mental illness regularly move into and out of these restrictive housing units. As a result, certifying such subclasses would be an exercise in legal formalism; the court is satisfied that, to the extent that these medication and segregation practices create a substantial risk of serious harm to prisoners when they are taking medication or housed in segregation, this risk extends to all seriously mentally ill prisoners, who are all fairly likely to take medication and be placed in segregation at some point.

ii. Due Process Claim

The court also finds that the procedural due-process class satisfies the commonality requirement. As with the Eighth Amendment claims, plaintiffs presented evidence through their expert, Dr. Burns, showing that prisoners across the ADOC system have not received adequate process before being subjected to involuntary medication. While the parties agree that the ADOC's written policy is constitutionally adequate under Washington v. Harper, 494 U.S. 2010 (1990), Bui challenges whether the procedures are adequate in practice.35 This question applies to all class members, as they are all on involuntary-medication orders. The question can be answered in one stroke—namely, by determining whether ADOC's involuntary-medication practices adequately protect due-process rights. See Hightower ex rel. Dehler v. Olmstead, 959 F.Supp. 1549, 1557 (N.D. Ga. 1996) (considering procedural due-process claims, among others, brought by a certified class of patients in a state hospital), aff'd sub nom. Hightower v. Olmstead, 166 F.3d 351 (11th Cir. 1998).

C. Typicality

Although the commonality and typicality inquiries "tend to merge," the typicality requirement—which is "somewhat of a low hurdle"—focuses the court's attention on "whether a sufficient nexus exists between the claims of the named representatives and those of the class at large." Gen. Tel. Co. of Sw. v. Falcon, 457 U.S. 147, 158 n.13 (1982); Taylor v. Flagstar Bank, FSB, 181 F.R.D. 509, 517 (M.D. Ala. 1998) (Albritton, J.); Busby v. JRHBW Realty, Inc., 513 F.3d 1314, 1322 (11th Cir. 2008) (citation and internal quotations omitted). A class representative's claims are typical if they "arise from the same event or pattern or practice and are based on the same legal theory" as the class claims; they need not be identical. Williams v. Mohawk Indus., Inc., 568 F.3d 1350, 1357 (11th Cir. 2009) (citation and internal quotations omitted); see In re Healthsouth Corp. Sec. Litig., 257 F.R.D. 260, 275 (N.D. Ala. 2009) (Bowdre, J.). Defendants argue that plaintiffs have failed to identify class representatives able to challenge each of the policies and practices at issue in this case; although they frame this argument as going to adequacy, it actually sounds in typicality.

For purposes of typicality, plaintiffs need to show that there are named plaintiffs who have been exposed to the policies or practices that create the substantial risk of serious harm they challenge, not that they have actually suffered the harm in the past.36 So, for example, if lack of adequate medication management creates a substantial risk of a prisoner taking psychotropic medications suffering unnecessarily from serious side effects, there must be a named plaintiff who is taking psychotropic medication; he need not necessarily prove, at the class-certification stage (or, for that matter, at trial), that he has already suffered from side effects.

Hence, plaintiffs need to prove that there are named plaintiffs who (1) receive mental-health care from the mental-health care staff,37 (2) rely upon custodial staff to facilitate their receipt of mentalhealth treatment, (3) are assessed at intake, during classification reviews, and in response to referrals, (4) need or take psychotropic medications, (5) require psychotherapy, (6) engage or desire to engage in self-harm or to commit suicide, (7) are placed in segregation, and (8) receive disciplinary citations based on symptomatic behavior. There is no serious dispute,38 though, that all of the named plaintiffs fall within categories (1) and (3); that many of the named plaintiffs—including at least six who have been housed in segregation (Hartley, Jackson, Johnson, McCoy, Pruitt, and Williams)—fall within category (2); that at least eight (Braggs, Hartley, Johnson, McCoy, Pruitt, Wallace, and Williams) fall within category (4); and that at least two (in many cases, more) fall within each of the remaining categories.39

Although they need not do so, plaintiffs have in fact presented evidence that at least one of them has not only been exposed to but harmed by the risk created by each of the policies or practices at issue. Rather than engage in an exhaustive discussion that would largely rehash the extended section on individual harm in the court's opinion on summary judgment, the court will instead mention illustrative examples.40

As for defendants' policy or practice of failing to provide adequate numbers of and sufficiently qualified mental-health staff, plaintiffs have offered evidence to show that the inadequate staffing has adversely affected Hartley's treatment: his treatment plans repeat problem statements and goals without documenting changes in his mental state, and that his treatment team meetings are not attended by required staff members. In fact, defendants' own mental-health expert, Dr. Patterson, agrees with plaintiffs that Hartley's treatment plans, and his resulting mental-health care, are inadequate. He further opines that "Mental Health Treatment Planning" is "seriously deficient, at least in part because of the lack of adequate staffing." Patterson Report (doc. no. 679-9) at 52.

Plaintiffs have also offered evidence to show that named plaintiffs have been exposed to and harmed by defendants' failure to provide sufficient custodial staffing to avoid regular security-related interference with the provision of mental-health care. For example, Jackson's medical records indicate that a number of his mental-health appointments were canceled due to security issues arising from an insufficient number of ADOC officers.

With respect to the failure to identify or recognize the severity of mental illnesses, plaintiffs have offered evidence to show that although staff at a state mental hospital recognized, prior to Johnson's incarceration in 1996, that he suffered from depression and was possibly psychotic (and potentially incompetent to stand trial), he was not placed on the mental-health caseload for almost two decades, until he was placed on suicide watch; at that point, he was recognized to be suffering from paraonia and possible delusions. Furthermore, although he was at this point referred for mental-health treatment, there appears to be no evidence that the mental-health staff ever received or acted upon the referral. Moreover, Dr. Burns pointed to McCoy as an example of a prisoner who was severely ill and required residential care but was not being provided it due to his classification at an inappropriately low level.

Plaintiffs have offered evidence to show that named plaintiffs have been exposed to and harmed by defendants' practice of failing to appropriately prescribe and manage psychotropic medication and its side effects. For example, Hartley's medical records reflect that he has complained of a movement disorder caused by his antipsychotic medication, that treating this side effect has adversely affected his kidneys, and that he has requested and been denied an alternative antipsychotic medication which Dr. Burns explained would be less likely to cause a movement disorder.

With respect to defendants' failure to provide more than cursory psychotherapeutic care and counseling, plaintiffs have offered evidence to show that Braggs has not received the regular therapy—for anxiety, major depressive disorder, and post-traumatic stress disorder—contemplated in his treatment plans; that Pruitt's repeated requests to speak to counselors have been rejected; and that Wallace, despite his repeated suicide attempts and self-harm, and diagnoses of bi-polar disorder and intellectual disability, has not consistently received group therapy. Defendants' expert, Dr. Patterson, agrees with plaintiffs that Wallace's care has been inadequate, due in part to the lack of group therapy.

Plaintiffs have offered evidence to show that named plaintiffs have been exposed to and harmed by defendants' practice of inadequately monitoring and providing inadequate follow-up care to prisoners who are suicidal or engage in self-harm. Over the course of a six-month period, Pruitt was admitted to a crisis cell as a result of self-injurious behavior at least five times, for stretches sometimes lasting more than a week. As to inadequate monitoring of the crisis cells, Dr. Patterson recognized that there was no indication in Pruitt's records that he was seen by a psychiatrist or nurse practitioner during any of these admissions. He was attacked by other prisoners during two different admissions (disinfectant was thrown in his face and burning fabric was thrown onto his leg). Dr. Burns notes that he was not placed on the mental-health caseload after these admissions and opines that he was therefore denied adequate follow-up care.

As to defendants' placement of mentally ill prisoners in segregation without regard to its harmful effects on their mental health: McCoy has repeatedly been placed in segregation for extended periods of time, including one thirteen-month stretch, even though he was in a psychotic state and the mental-health staff had advised against prolonged isolation due to segregation's effect on his stability. Deterioration of McCoy's mental health during segregation was also documented in his records. Jackson was in segregation between 2007 and 2014; he was frequently transferred from one facility's segregation unit to another's. This extended stay in segregation has broken him down mentally, according to his own testimony and the opinion of one of plaintiffs' experts.

Plaintiffs have likewise presented evidence to show that Wallace and Pruitt have been harmed by defendants' policy or practice of disciplining prisoners for behavior that stems from mental illness. Wallace received nine disciplinary citations for instances of cutting his wrists, and Pruitt has also received disciplinary tickets for creating a hazard when he engaged in self-harm.

Finally, with respect to the procedural due-process class, plaintiffs have presented evidence showing that the class representative, Bui, was exposed to and harmed by defendants' involuntary-medication practices. He has been under an involuntary-medication order since November 2007, and the order has been renewed approximately every six months. Plaintiffs have shown that on some occasions, Bui was not present at his own involuntary-medication hearings; that many notices were given to him on the day of the hearing, or back-dated; and that on several occasions, he did not have a lay advisor during the hearing to explain the purpose of the hearing or to help him present objections to the hearing panel's findings.41

Rule 23(a)(3)'s typicality requirement has been satisfied with respect to both classes.

D. Adequacy

Rule 23(a)(4) requires that the "representative parties will fairly and adequately protect the interests of the class." This analysis "encompasses two separate inquiries: (1) whether any substantial conflicts of interest exist between the representatives and the class;42 and (2) whether the representatives will adequately prosecute the action." Valley Drug Co. v. Geneva Pharm., Inc., 350 F.3d 1181, 1189 (11th Cir. 2003) (citation omitted).43

"Adequate representation is usually presumed in the absence of contrary evidence," and generally exists for injunctive-relief classes, because there is no monetary pie to be sliced up. Access Now, Inc. v. Ambulatory Surgery Ctr. Grp., Ltd., 197 F.R.D. 522, 528 (S.D. Fla. 2000) (Seitz, J.).

As noted above, defendants' argument against the adequacy of the named plaintiffs as class representatives is that the named plaintiffs have not themselves raised the claims they now seek to bring on behalf of the class; this argument goes to typicality rather than adequacy. In any event, as the court has already explained, this argument fails. Defendants have also reiterated in the section of their brief on adequacy their contentions that the named plaintiffs do not have justiciable claims. This argument, too, has already been addressed in the section on standing.

Defendants have not suggested that the named plaintiffs have conflicts of interest with the class, and the court concludes that no such conflict exists. As for whether they will adequately prosecute this action, defendants challenged two of the named plaintiffs based on statements in their depositions arguably reflecting that they do not intend to represent other prisoners in this case or do not understand their roles as class representatives. However, these plaintiffs have been released from custody; because their claims are now moot, they cannot serve as class representatives. The remaining named plaintiffs are adequate representatives of the Eighth Amendment class, and Bui is an adequate representative of the procedural due-process class.

3. Rule 23(b)(2)

A class satisfies Rule 23(b)(2) in cases in which "the party opposing the class has acted or refused to act on grounds that apply generally to the class, so that final injunctive relief or corresponding declaratory relief is appropriate respecting to the class as a whole." Fed. R. Civ. P. 23(b)(2).

"Rule 23(b)(2) has been liberally applied in the area of civil rights, including suits challenging conditions and practices at various detention facilities." Bumgarner v. NCDOC, 276 F.R.D. 452, 457 (E.D.N.C. 2011) (Boyle, J.); see also Wright & Miller, 7AA Fed. Prac. & Proc. Civ. § 1776 (3d ed.) (discussing the range of civil-rights actions certified pursuant to Rule 23(b)(2), and explaining that "the class suit is a uniquely appropriate procedure in civil-rights cases"). Indeed, some courts have gone so far as to say that the rule's requirements are "almost automatically satisfied in actions primarily seeking injunctive relief." Baby Neal ex rel. Kanter, 43 F.3d at 59.

As plaintiffs have repeatedly explained and offered evidence to demonstrate throughout the litigation of this case, and as this court has found in addressing the commonality requirement, the problems of which they complain and the remedies they seek are systemic.44 The Supreme Court has approved of system-wide relief in prison cases involving "systemwide violation[s]" resulting from "systemwide deficiencies." Plata, 563 U.S. at 532.

Defendants are quite right that a single injunction would not be appropriate if plaintiffs sought an order requiring that each named plaintiff be provided the specific forms of treatment he has been denied. They cite Wal-Mart for the proposition that Rule 23(b)(2) "does not authorize class certification when each individual class member would be entitled to a different injunction or declaratory judgment against the defendant." 564 U.S. at 360. But Rule 23(b)(2) certainly does not preclude certification merely because the class members could bring separate claims regarding the adequacy of the particular treatment they have individually received—claims which, everyone agrees, could not be resolved with a single injunction or declaratory judgment—when those are not the claims they actually have brought in this lawsuit.

For some reason, defendants repeatedly misconstrue the claims in the case as demands for individual treatment of various kinds, and then proceed to argue that the individualized claims plaintiffs are not raising cannot be brought in the form of a class action.45 As the court explained at some length in its analysis of the applicable Eighth Amendment case law in its summary-judgment opinion, defendants' position is unwarranted; plaintiffs are not seeking adjudication of demands for particular individualized treatment, and the challenge they are actually bringing—to systemic deficiencies that create a substantial risk of serious harm—are well recognized.

Defendants declare that "an alleged delay in care at St. Clair bears no relationship of any kind to allegedly insufficient staffing at Hamilton, or allegedly insufficient intake procedures at Kilby." Defs.' Opp. to Class Cert. (doc. no. 807) at 167. This position is controverted by the evidence in this case. Plaintiffs' experts opine that all of these problems— and others—stem directly and ineluctably from a contractual arrangement that provides for inadequate numbers of and insufficiently qualified mental-health staff. Fundamentally, defendants err in forgetting that the language of Rule 23(b)(2) focuses on the nature of defendants' acts and omissions and the suitability of class-wide relief, and does not require that the class-wide relief benefit each class member in precisely the same way. Here, plaintiffs seek a remedy for defendants' decision to starve the mental-health care system in Alabama prisons of human and financial resources, and relief would be appropriate for everyone subjected to the substantial risk of serious harm plaintiffs claim this creates—that is, prisoners with serious mental illness.

It would, in fact, be impossible to grant any meaningful relief on this claim without granting it on a class-wide basis. See Wal-Mart, 564 U.S. at 360 ("The key to the (b)(2) class is the indivisible nature of the injunctive or declaratory remedy warranted." (citation and internal quotation marks omitted)). Although the court could order, for example, that a particular prisoner who currently requires psychiatric care be seen by a psychiatrist, it cannot order that adequate staffing be provided to the extent necessary to address any one particular prisoner's mental-health care needs as they develop in the future. Although deficiencies in the individual treatment now being provided could potentially (though much less efficiently) be addressed one by one, the risks defendants' policies and practices pose to mentally ill prisoners, whose needs change over time, could not. Moreover, any order remedying a systemic deficiency with respect to only one prisoner would rob Peter to pay Paul; the court would be playing a zero-sum game, shifting mental-health care resources from one prisoner to another.46 By the same token, there exists no individual remedy for the practice of filling residential treatment beds with segregation prisoners, thereby reducing the bed space available in those units and undermining the treatment provided there; the court obviously cannot order that a bed be reserved for each prisoner likely to require one, or order that a particular prisoner's therapy in the residential treatment unit—but only his—not be disturbed.

Nor could the court address inadequate intake screenings without also addressing inadequate staffing levels, when the deficiencies in screening are caused by the fact that there are not enough staff members qualified to conduct the screenings or supervise others in doing so. Moreover, as discussed above in the context of commonality, even ordering adequate staffing at a particular facility would not meaningfully redress the claims of prisoners currently housed at that facility, since the chance of a prisoner being moved to another one is very high.

The Rule 23(b)(2) cases to which defendants cite are all unavailing. In Lakeland Regional Medical Center, Inc. v. Astellas U.S., LLC, 763 F.3d 1280 (11th Cir. 2014), the court affirmed the denial of certification in a case where the plaintiff failed to identify the relief it sought other than "such declaratory and injunctive relief as appropriate in order to compel and ensure defendant[`s] future compliance with law." Id. at 1291 (citation omitted).47

Here, by contrast, plaintiffs have described in detail the specific policies and practices in which they ask the court to order defendants to stop engaging; to the extent that they seek an order requiring defendants to formulate a plan for remedying unconstitutional conditions, rather than setting out the details of that plan themselves, they do so only out of appropriate deference to the discretion of prison administrators. See Parsons, 754 F.3d at 689 n. 35 (explaining that requiring plaintiffs to articulate with great precision the terms of the injunction they seek is particularly inappropriate in "prison cases, given that an injunction in any such case must closely track the violations established by the evidence at trial, that any such relief must comply with the [Prison Litigation Reform Act]'s extensive requirements, that prison officials must play a role in shaping injunctions, that ultimate proof of some violations but not others might easily change the structure of a remedial plan, that conditions in prisons might change over the course of litigation, and that the class certification hearing is not a dress rehearsal of the trial on the merits (let alone a dress rehearsal of the remedy proceedings)"); Morrow v. Washington, 277 F.R.D. 172, 198 (E.D. Tex. 2011) (Ward, J.) ("Plaintiffs have set forth facts suggesting that Defendants' behavior was generally applicable to the class as a whole, making injunctive relief appropriate. The precise terms of the injunction need not be decided at this stage, only that the allegations are such that injunctive and declaratory relief are appropriate and that the class is sufficiently cohesive that an injunction can be crafted that meets the specificity requirements of Rule 65(d).").

Defendants' other cases fare no better. In Heffner v. Blue Cross and Blue Shield of Ala., Inc., 443 F.3d 1330, 1345 (11th Cir. 2005), certification of a Rule 23(b)(2) damages class was reversed because the court concluded that the class members were not entitled to relief (in the form of a refund of a deductible) absent an individualized showing that they had in fact relied on a summary plan description indicating that no deductible would be charged. Similarly, in Jamie S. v. Milwaukee Pub. Sch., 668 F.3d 481, 499 (7th Cir. 2012), the court reversed certification of a class in a case where "the intricate remedial scheme . . . require[d] thousands of individual determinations of class membership, liability, and appropriate remedies," explaining that certification under Rule 23(b)(2) is not appropriate when "the relief sought would merely initiate a process through which highly individualized determinations of liability and remedy are made." Here, by contrast, no individualized determinations are necessary or relevant. If the court grants relief, it will do so across the board, in one fell swoop. This relief will ensure that the statewide prison mental-health care system, with which all class members interact, is not operated pursuant to policies and practices that subject all prisoners to a substantial risk of serious harm.

This is exactly the kind of case for which Rule 23(b)(2) was intended.48

4. Rule 23(g)

In order to certify a class, the court must also appoint class counsel. In so doing, the court "must consider . . . (i) the work counsel has done in identifying or investigating potential claims in the action; (ii) counsel's experiences in handling class actions, other complex litigation, and the types of claims asserted in the action; (iii) counsel's knowledge of the applicable law; and (iv) the resources that counsel will commit to representing the class." Fed. R. Civ. P. 23(g)(1)(a). The court also "may consider any other matter pertinent to counsel's ability to fairly and adequately represent the interests of the class," Fed. R. Civ. P. 23(g)(1)(B).

Lawyers affiliated with the Southern Poverty Law Center, the Alabama Disabilities Advocacy Program, and the law firm of Baker, Donelson, Bearman, Caldwell & Berkowitz have represented named plaintiffs in litigating and negotiating the settlement of this case, and they now seek appointment as class counsel.49 As previously discussed for purposes of certification of the Phase 1 class in this case, the record reflects that these attorneys have substantial experience in litigating class actions and in the complex substantive areas of both prisoners'-rights and disability-rights law.

Notably, defendants take issue with the adequacy of only a subset of named plaintiffs' counsel. They argue that the Southern Poverty Law Center (SPLC) should not be appointed as class counsel, but they do not make the same argument with respect to Baker, Donelson or ADAP, both of which represent the individual named plaintiffs.

In any event, defendants' arguments with respect to SPLC are not convincing. Although they focus on one possible deficiency in SPLC's pre-filing investigation—namely, its failure to review the medical and mental-health records of all named plaintiffs prior to filing their complaint—they ignore the detailed and involved investigation that SPLC and plaintiffs' other counsel conducted prior to (and since) the filing of this case. Before filing suit, plaintiffs' counsel conducted a two-year investigation into health care in ADOC facilities, interviewed numerous prisoners, obtained some medical records, and obtained some records from defendants and their medical contractors through open records requests. They issued a lengthy report on their findings, which discussed the named plaintiffs in some detail. Especially in light of the difficulty of investigating secure institutions, this was plainly adequate.50

In addition, defendants argue that SPLC has taken positions in conflict with its fiduciary obligations to the putative class. Essentially, defendants complain that when, earlier this year, the Alabama legislature considered a proposal to invest hundreds of millions of dollars in building four large new prison facilities in the State, the president of SPLC sent a letter to a number of legislators opposing its passage. Whether or not this letter had any effect on legislative support for, or the eventual defeat of, the bill, nothing in it was in conflict with the interests of the putative class his organization seeks to represent.

As defendants themselves acknowledge, it was "the opinion of the Commissioner [that] inmates in the ADOC system would benefit from the [bill], because it would alleviate problems caused by overcrowding and understaffing." Defs.' Opp. to Class Cert. (doc. no. 807) at 171 (emphasis added). SPLC's president expressed a different opinion, favoring a legislative approach that prioritized "sentencing reform," and "demand[ed] that the prison system reform its management practices." Id.

Defendants' argument incorrectly presupposes either that the Commissioner's view was indisputably correct, or else that the only legislative alternatives were the passage of the bill as proposed, on the one hand, and no action at all on the other. This, too, is not the case; there is a panoply of possible legislative approaches to any issue as complex as prison reform. The fact that an SPLC official expressed a preference for options other than the one that happened to be on the table at a particular moment hardly reflects an abandonment of his fiduciary duty to his current or prospective clients.

Setting aside the fact that plaintiffs' counsel are free to take a different position than the Commissioner, it is not entirely clear how and why— even accepting his position—the prison construction project would benefit the plaintiff class. According to the statement of the Commissioner quoted by defendants, the salient impact of the bill would be on overcrowding and understaffing.51 Moreover, as with any "transformation" of the size contemplated here, the prison construction project would presumably impact the delivery of health care in numerous ways—some positive and some negative, many hard to predict. Plaintiffs note in replying to this argument that the record reflects that the Commissioner has also suggested that his proposal would allow ADOC to dramatically cut the cost of providing health care. The court certainly passes no judgment on the merits of SPLC's perspective, but it does seem perfectly reasonable for lawyers representing clients who seek more funding for health care to oppose a bill the proponent of which says would result in less funding for health care.

5. Ascertainability

Defendants argue that the Eighth Amendment class certified here is not ascertainable. However, there is serious reason to doubt that the judicially created ascertainability requirement applies to Rule 23(b)(2) classes. Even if it does, this class meets the requirement.

Although the Eleventh Circuit did hold in defendants' primary case, Little v. T-Mobile USA, Inc., 691 F.3d 1302, 1304 (11th Cir. 2012), that a class must be "clearly ascertainable"—that is, "its members can be ascertained by reference to objective criteria," Bussey v. Macon County Greyhound Park, Inc., 563 F. App'x 782, 787 (11th Cir. 2014) (citation and internal quotation marks omitted)—it did so in the context of a Rule 23(b)(3) damages class, not, as here, a Rule 23(b)(2) injunctive-relief class. Defendants have not cited, and the court is not aware, of any cases within this circuit applying the ascertainability requirement to a Rule 23(b)(2) class, much less any binding precedent doing so.

Moreover, the circuits that have squarely addressed the issue have generally concluded that the ascertainability requirement does not apply to Rule 23(b)(2) injunctive-relief classes. The recent discussion of this issue by the Third Circuit is detailed and persuasive.52

In Shelton v. Bledsoe, 775 F.3d 554 (3d Cir. 2015), the court distinguished Rule 23(b)(2) and (b)(3) classes by noting that the latter "`allows certification in a much wider set of circumstances,' including those `in which class-action treatment is not as clearly called for,'" and explaining that because Rule 23(b)(3) classes are an "`adventuresome innovation,' Congress included additional `procedural safeguards,'" such as individual notice and an opportunity to opt-out. Id. at 560 (quoting Wal-Mart, 131 S. Ct. at 2558, Comcast, 133 S. Ct. at 1432). In Rule 23(b)(2) classes, on the other hand, "`[t]he key. . . is the indivisible nature of the injunctive or declaratory remedy warranted—the notion that the conduct is such that it can be enjoined or declared unlawful only as to all of the class members or as to none of them.'" Id. at 561 (quoting Wal-Mart, 131 S. Ct. at 2557 (citation and internal quotation marks omitted). In sum, the court explained, "the identities of individual class members are less critical in a (b)(2) action than in a (b)(3) action." Id.

As the court noted in Shelton, an advisory committee note to Rule 23 describes cases which are specifically defined in terms of their unascertainability as "illustrative" examples of Rule 23(b)(2) classes: "various actions in the civil rights field where a party is charged with discriminating unlawfully against a class, usually one whose members are incapable of specific enumeration." Fed. R. Civ. P. 23 advisory committee's note (1966) (emphasis added). This court therefore agrees with the Third Circuit that a requirement—directly contrary to the express language of the advisory committee—"that the members of the class be capable of specific enumeration[] is inappropriate for (b)(2) classes."53 Shelton, 775 F.3d at 561.

As the Third Circuit recounted, two other circuits have reached the same conclusion.54 In Shook v. El Paso Cty., 386 F.3d 963, 972 (10th Cir. 2004), the court explained that "many courts have found Rule 23(b)(2) well suited for cases where the composition of the class is not readily ascertainable; for instance, in a case where the plaintiffs attempt to bring suit on behalf of a shifting prison population." The First Circuit has agreed. See Yaffe v. Powers, 454 F.2d 1362, 1366 (1st Cir. 1972) (holding that "the actual membership of [a Rule 23(b)(2)] class need not . . . be precisely delimited," because individual notice to class members is not required); see also Floyd v. City of New York, 283 F.R.D. 153, 171-72 & n.115 (S.D.N.Y. 2012) (Scheindlin, J.) (citing cases). Shelton also pointed out the "significant" fact that the Supreme Court's recent foray into Rule 23(b)(2), Wal-Mart, "lacks any inquiry into `ascertainability.'" 775 F.3d at 563.

For these reasons, the court concludes that the class proposed in this litigation need not be ascertainable. That said, the court concludes in the alternative that even if the requirement did apply, plaintiffs' proposed class definition would satisfy it.

Although defendants complain that the term "serious" mental illness is undefined, defendants themselves note in their motion for summary judgment that there is case law in this circuit articulating what counts as "serious" in the arena of Eighth Amendment mental-health claims: an illness "that has been diagnosed by a physician as mandating treatment or one that is so obvious that even a lay person would easily recognize the need for a doctor's attention." Hill v. Dekalb Reg'l Youth Det. Ctr., 40 F.3d 1176, 1187 (11th Cir. 1994) (explaining that "[t]his standard has been acknowledged and used by other federal circuit and district courts," and concluding that it was the "appropriate and guiding principle"). This is a rather straightforward, objective standard: a prisoner would be a member of this class if he has been diagnosed as needing mental-health treatment or if he is very obviously mentally ill.55 Of course, because this is not a Rule 23(b)(2) damages class action, and class members will not be entitled to individual notice or have the right to opt out, the court need not actually conduct this inquiry to determine whether each prisoner is or is not a member of the class.

To conclude, the court will briefly address the other two cases defendants cite, both of which are unreported decisions by out-of-circuit district courts: Johannes v. Washington, 2015 WL 5634446 (E.D. Mich. Sept. 25, 2015) (Michelson, J.), and Schilling v. Kenton Cty., 2011 WL 293759 (E.D. Ky. Jan. 27, 2011) (Bunning, J.). They are wholly consistent with the result reached here; indeed, Johannes actually recognizes that the ascertainability requirement applies "less stringently" in Rule 23(b)(2) cases. 2015 WL 5634446, at *10. These cases declined to certify or expressed concern about certifying what are known as "fail-safe" classes—classes that are defined in terms of the eventual merits adjudication. See Schilling, 2011 WL 293579, at *5 ("A class definition is . . . too general where it requires the Court to determine whether an individual's constitutional rights have been violated in order to ascertain membership in the class itself."). In this case, a fail-safe class would be one defined to consist of all prisoners who have received constitutionally inadequate mental-health care. Because plaintiffs have not proposed a fail-safe class, these cases suggest no reason not to certify.56

* * *

The motion for class certification filed by the remaining named plaintiffs is granted in part and denied in part. The court will certify a class of all persons with a serious mental-health disorder or illness who are now, or will in the future be, subject to defendants' mental-health care policies and practices in ADOC facilities, excluding the work release centers and Tutwiler Prison for Women. The court will also certify a class of all persons with a serious mental-health disorder or illness who are now, or will in the future be, subject to defendants' formal involuntary-medication policies and practices.

The court will decline to certify a class for the individual plaintiffs' due-process challenge regarding involuntary medication without a recent finding of dangerousness, or for their due-process challenge regarding coerced consent. However, the individual plaintiffs with these claims will proceed to trial, and ADAP, through its associational standing, may represent the interests of unnamed prisoners who are subject to these policies and practices.

An appropriate order will be entered.