UNITED STATES DISTRICT COURT FOR THE DISTRICT OF NEW HAMPSHIRE

B.D., by their next friend, Christine Wellington

v. Case No. 21-cv-4-PB Opinion No.

Christopher Sununu, Governor of New Hampshire et al.

MEMORANDUM AND ORDER

The named plaintiff in this putative class action is seeking to represent

a class of adolescent foster children with mental disabilities who are in the

custody of the New Hampshire Division of Children, Youth, and Families

(DCYF) and either have been unnecessarily placed in a congregate care

setting or are at risk of being unnecessarily placed in congregate care. The

plaintiff’s primary claims are that these congregate care placements violate

the Americans with Disabilities Act and the Rehabilitation Act. The plaintiff

also asserts that the defendants are violating the Adoption Assistance and

Child Welfare Act by failing to provide class members with adequate case

plans. The matter is before me on the plaintiff’s motion for class certification. I. BACKGROUND

A. New Hampshire’s Foster Care System

DCYF, a division of New Hampshire’s Department of Health and

Human Services (DHHS), oversees the state’s child welfare services,

including its child protection services. Doc. 175-2 at 2. Children who have

been removed from their homes as a result of abuse or neglect may be placed

under the protective supervision or legal custody of DCYF. 1 Id. at 4-5. DCYF

is responsible for arranging placements and supportive services for children

under its care, as well as developing a case plan to guide each child’s time in

foster care. Id. at 5, 7-8.

When determining where to place a foster child, DCYF case workers

can choose between a variety of community-based placements or congregate

care placements. DCYF Policy 1600.2. 2 Community-based placements

1 A child is in “protective supervision” when the child has been placed in DCYF custody before a final finding of abuse or neglect. See

-C:3, XXIV. “Legal custody” refers to a child’s status after custody has been granted to DCYF following a final finding of abuse or neglect. See

-C:3, XVII. The parties do not distinguish between the two statuses for present purposes.

2 Some, but not all, of DCYF’s policies were submitted as exhibits. I take judicial notice of those policies that were not submitted, all of which are publicly available via the DHHS website. See DCYF Policy Manual, https://www.dhhs.nh.gov/programs-services/child-protection-juvenile- justice/dcyf-policy-manual (last visited Sept. 12, 2024); see also Fed. R. Evid. 201; United States v. Garcia,

(4th Cir. 2017); Daniels-Hall

2 include both kinship and foster home placements.

Kinship placements,

which include placements with relatives or other individuals who have an

existing relationship with a child, are the preferred type of placement.

-E:25, VIII (defining “kin”). Relatives may accept a child

placement with or without a license but unrelated kin must be licensed. Doc.

317-1 at 12.

A foster home placement is a placement with a licensed foster family.

Doc. 175-2 at 8. Foster home placements can be with either a general foster

home or an enhanced support foster home.

General foster homes provide

“day-to-day care in a family setting for children whose needs can be managed

in the community” without the need for specialized supervision or care.

DCYF Policy 1600.2. Enhanced support foster homes, on the other hand, are

foster homes that are trained and credentialed to offer a heightened level of

care for children with specific needs. Doc. 277-2 at 15 n.38.

Currently, the only enhanced support foster care offered in the state is

individual service option (ISO) foster care. Id. at 18. ISO foster homes provide

care to “children with chronic mental health, emotional, physical, or

behavioral needs” who require “intensive supervision and consistent

structure.” DCYF Policy 1600.2. Although not presently offered, the state’s

v. Nat’l Educ. Ass’n,

(9th Cir. 2010).

3 regulations and policies provide for two other forms of enhanced foster care:

therapeutic foster care and adolescent foster care. Doc. 277-2 at 15 n.38; see

also DCYF Policy 1603; DCYF Policy 1605. Therapeutic foster care provides

intensive clinical and therapeutic services to children with chronic mental or

behavioral health problems who require a higher level of care and

supervision than that offered by ISO foster care. DCYF Policy 1605.

Adolescent foster care supports foster children between 14 and 21 years of

age with “specialized needs[.]” N.H. Admin. R. He-C 6347.03(a).

Regardless of the placement type, children in community-based

placements are eligible to receive an array of community-based mental health

services provided through DHHS. Doc. 175-2 at 12-14. These services include

in-home therapeutic interventions, “wraparound” services, and crisis support,

as well as counseling and skills training.

Although community-based placements are preferred, DCYF may place

a child in a congregate care setting under certain circumstances. Congregate

care placements are placements with “qualified residential treatment

program[s],” which offer residential mental and behavioral health treatment

to children with “serious emotional or behavioral disorders or disturbances.”

(k)(4); see also

-C:19-f. These

placements differ significantly from community-based placements in that

they are in an institution, rather than a family setting, and typically require

4 children to comply with stringent rules and restrictions. Doc. 154-3 at 7. An

independent clinician must assess each child placed in residential care to

determine whether such a placement is warranted, and a state district court

judge must approve the placement.

-C:19-f.

Once a child has been placed, a DCYF case worker must develop a

written case plan for that child. DCYF Policy 1550. DCYF uses a standard

case plan template that collects information on the child’s background and

history, planned placement, and permanency goals (family reunification,

adoption, etc.). Doc. 207-11 at 1-7. The case worker must append to the case

plan a Youth Information Sheet, which provides detailed information

regarding the child’s family, medical, and educational history. DCYF Policy

1550; see also DCYF Form 1522. DCYF policy requires the case worker to

complete these forms within 60 days of the child’s removal from the home.

DCYF Policy 1550. Case plans must be updated every 6 months, as well as

following certain changes in circumstances, such as a change in the child’s

placement, permanency goal, or needs.

For foster children aged 14 or older, the case worker must also

complete a transition plan, which identifies the child’s achievements and

needs in various aspects of adult living, such as employment, budgeting, and

self-care. DCYF Policy 1695; see also DCYF Form 1695. This plan must be

5 completed within 60 days of the children turning 14, and must be updated at

least annually and following a change in placement. DCYF Policy 1695.

A child’s case plan and transition plan together provide a “critical

‘Roadmap’” that helps to “facilitat[e] the provision of timely, appropriate, and

coordinated services and supports that promote positive outcomes for youth

and families.” Doc. 154-1 at 5. Transition plans, in particular, ensure that

children who “age out” of the foster care system upon turning 18 are

adequately “prepared to live as independent young adults,” despite lacking

many of the ongoing support systems that benefit other children. Id. at 6.

B. Statutory Requirements

As a federally funded program, New Hampshire’s foster care program

must comply with a number of federal statutes, including Title II of the

Americans with Disabilities Act (ADA), section 504 of the Rehabilitation Act,

and the Adoption Assistance and Child Welfare Act (CWA). The ADA and the

Rehabilitation Act each prohibit public entities from discriminating on the

basis of disability in the administration of certain programs. 3

The ADA applies to public entities, including state agencies, whereas the Rehabilitation Act applies to programs receiving federal funds.

;

. Nonetheless, “[g]iven the textual similarities between [the two statutes], the same standards govern claims under both, and [courts] rely on cases construing [both statutes] interchangeably.” Ingram v. Kubik,

, 1256 (11th Cir. 2022) (cleaned up); accord Kiman v. N.H. Dep’t of Corrs.,

, 285 n.10 (1st Cir. 2006). Because

6 12132;

. One form of prohibited discrimination is the

“unjustified institutional isolation of persons with disabilities[.]” Olmstead v.

L.C. ex rel. Zimring,

(1999).

This prohibition is largely implemented through two regulations: the

integration mandate and the methods of administration regulation. The

integration mandate requires states to “administer services, programs, and

activities in the most integrated setting appropriate to the needs of qualified

individuals with disabilities.”

(d); see also

(d). “The most integrated setting is defined as a setting that enables

individuals with disabilities to interact with nondisabled persons to the

fullest extent possible.” Parent/Pro. Advoc. League v. City of Springfield,

(1st Cir. 2019) (hereinafter PPAL) (cleaned up). Nonetheless, the

state’s obligation to provide services in the most integrated setting “is not

boundless.” Olmstead,

. While a state is required to “make

reasonable modifications in policies, practices, or procedures” where

“necessary to avoid discrimination on the basis of disability,” it need not

make modifications that would “fundamentally alter the nature of the

service, program, or activity.”

(b)(7). Thus, the integration

the parties do not distinguish between the plaintiff’s ADA and Rehabilitation Act claims, I refer to both claims collectively as ADA claims for ease of reference.

7 mandate requires states to provide services in the community, rather than in

an institutionalized setting, if (1) “the State’s treatment professionals

determine that [community] placement is appropriate;” (2) “the affected

persons do not oppose such treatment;” and (3) community “placement can be

reasonably accommodated, taking into account the resources available to the

State and the needs of others with [disabilities].” Olmstead,

.

The methods of administration regulation, in turn, prohibits entities

from “utiliz[ing] criteria or methods of administration . . . [t]hat have the

effect of subjecting qualified individuals with disabilities to discrimination on

the basis of disability[.]”

(b)(3)(i); see also

(b)(3)(i). Under this regulation, states may not rely on methods of

administration that subject individuals to unjustified institutionalization.

Day v. District of Columbia,

(D.D.C. 2012).

The CWA is a federal statute embedded in the Social Security Act that

provides “federal funding for expenses associated with operating a foster care

system.” Connor B. ex rel. Vigurs v. Patrick,

(1st Cir. 2014).

The CWA amended Title IV-B of the Social Security Act, which provides

funding for state child welfare services, and created Title IV-E of the Act,

which provides reimbursement to states for foster care maintenance

payments made on behalf of certain eligible foster children. Vt. Dep’t of Soc.

& Rehab. Servs. v. U.S. Dep’t of Health & Hum. Servs.,

(2d

8 Cir. 1986). To receive federal funds under either Title IV-B or Title IV-E, a

state must submit a “State plan” to the Secretary of Health and Human

Services detailing its foster care system and compliance with various

statutory requirements. Connor B.,

.

State plans under both Title IV-B and Title IV-E must provide for the

development of a “case review system.”

(b)(8)(A)(ii) (Title IV-

B);

(a)(16) (Title IV-E). A case review system must, among

other things, ensure that “each child has a case plan designed to achieve

placement in a safe setting that is the least restrictive (most family like) and

most appropriate setting available and in close proximity to the parents’

home, consistent with the best interest and special needs of the child[.]”

(5). That case plan must be a written document that includes

certain information about the child, such as a description of the child’s

placement and an explanation of why that placement is appropriate, a plan

for ensuring that the child receives required care and services, up-to-date

health and education records, and, for children 14 years of age or older, a

description of the services needed to help the child transition to adulthood.

(1); see also 42 U.S.C. § 675a.

9 C. Factual and Procedural History 4

The named plaintiff, B.D., is a 15-year-old foster child who suffers from

post-traumatic stress disorder, attention-deficit/hyperactivity disorder, and

depression. 5 Doc. 310-1 at 5. B.D. was removed from their parent’s home at

the age of 13 and placed in the legal custody of DCYF after a finding of

parental neglect. Doc. 310-4 at 4. B.D. spent three days in an emergency

foster home immediately following their removal but was then moved to a

congregate care facility, where they have remained ever since. Id. B.D. has

expressed frustration with their placement in congregate care and would

prefer to live with a foster family. Doc. 310-1 at 7. Based on the opinion of

their child psychology expert witness, B.D. contends that they could remain

in the community if provided with an appropriate placement and

accompanying supports. Id. at 6.

Although B.D. was provided with a case plan shortly after their

removal from the home, it was lacking several required components,

including the Youth Information Sheet that details B.D.’s family, medical,

4 I focus here on the procedural history most relevant to the plaintiff’s motion for class certification but incorporate the more detailed account of this litigation’s history outlined in my prior orders. See Doc. 303; Doc. 315.

5 B.D. is pursuing this action pseudonymously. In an effort to further protect B.D.’s identity, the parties refer to B.D. using gender-neutral pronouns. I follow suit here.

10 and educational background. Doc. 310-4 at 7. B.D. received a transition plan

after turning 14, but the plan is “formulaic, lacking any meaningful

description of B.D.’s individual needs and goals” and fails to outline a

concrete plan for achieving the few goals that are described. Id. at 8. Neither

the case plan nor the transition plan have been updated over the course of

B.D.’s nearly two years in DCYF custody. Id. at 7-8.

B.D., acting through their next friend, filed a complaint on behalf of

themselves and a putative class of similarly situated individuals challenging

perceived deficiencies in New Hampshire’s operation of its foster care

program. 6 The putative class claims fall into two categories. 7 First, B.D.

alleges that the defendants are violating the ADA’s integration mandate and

methods of administration regulation by systematically placing and retaining

adolescent foster children (that is, foster children between 14 and 17 years of

age) with mental and behavioral disabilities in congregate care facilities,

even where those children could be appropriately placed in community-based

settings. Second, B.D. alleges that the defendants violate the CWA by failing

6 The complaint initially included five other named plaintiffs, all of whom have since exited DCYF custody. Doc. 324-1 at 2-3. For the reasons outlined in my order on the defendants’ motion to dismiss, those plaintiffs have been dismissed from the case.

7 The complaint also asserted a due process claim under the Fourteenth Amendment, which I dismissed in a prior order. See Doc. 49.

11 to ensure that adolescent foster children receive required case planning

services, including the provision of timely, up-to-date case and transition

plans that meet the substantive requirements outlined in the statute.

B.D. filed a motion for class certification pursuant to Federal Rule of

Civil Procedure 23(b)(2), seeking to certify a class of:

All children, ages 14 through 17, who:

(1) are, or will be, in the legal custody or under the protective supervision of DCYF under

-C:3 (XVII) and/or (XXV);

(2) have a mental impairment that substantially limits a major life activity, or have a record of such an impairment; and

(3) currently are, or are at serious risk of being, unnecessarily placed in congregate care settings.

Doc. 152 at 2. The defendants object, arguing that neither B.D.’s disability

discrimination claims nor their case planning claim are appropriate for class

resolution.

II. STANDARD OF REVIEW

A class action is “an exception to the usual rule that litigation is

conducted by and on behalf of the individual named parties only.” Califano v.

Yamasaki,

(1979). To warrant class action treatment,

the party seeking class certification must demonstrate that certification is

proper under Rule 23 of the Federal Rules of Civil Procedure. Smilow v. Sw.

Bell Mobile Sys.,

(1st Cir. 2003). The four prerequisites to the

12 certification of any class are numerosity, commonality, typicality, and

adequacy of representation.

A moving party must also demonstrate that

their claims fall within one or more of the circumstances listed in Rule 23(b).

Where, as here, the moving party seeks certification pursuant to Rule

23(b)(2), they must establish 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[.]”

“Rule 23 does not set forth a mere pleading standard.” Wal-Mart v.

Dukes,

(2011). Instead, parties seeking to certify a class

must prove by a preponderance of the evidence that the requirements of Rule

23 are satisfied. In re Nexium Antitrust Litig.,

(1st Cir. 2015).

“Once plaintiffs have made their initial showing, defendants have the burden

of producing sufficient evidence to rebut the plaintiff’s showing.”

Although the court may need to touch on the merits of a plaintiff’s

claims to determine whether the proposed class should be certified, “Rule 23

grants courts no license to engage in free-ranging merits inquiries at the

certification stage. 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 Inc. v. Conn. Ret.

Plans & Tr. Funds,

(2013).

13 III. ANALYSIS

B.D. argues that their claims are appropriate for class treatment

because they raise common questions regarding the legal sufficiency of the

defendants’ administration of its foster care system. The defendants disagree

and assert that B.D.’s claims turn on individualized considerations that

cannot be resolved or remediated on a class-wide basis. The defendants

further take issue with the proposed class definition and argue that B.D. has

failed to establish the required elements of numerosity, typicality, and

adequacy. I first consider the defendants’ arguments as to B.D.’s proposed

class definition before turning to the requirements of Rule 23.

A. Class Definition

The defendants argue that the proposed class definition is

inappropriate for two reasons. First, the defendants assert that whether an

individual is “unnecessarily” placed in congregate care or “at risk” of such an

unnecessary placement requires individualized fact finding and litigation. In

this way, the defendants contend, the proposed definition runs afoul of what

courts have referred to as the “implied requirement of ‘ascertainability[.]’”

See Shelton v. Bledsoe,

(3d Cir. 2015). Second, the

defendants argue that the class cannot be defined to include individuals who

are merely “at risk” of unnecessary placement in congregate care because an

ADA claim can only be premised on instances of actual institutionalization.

14 Although not an explicit prerequisite to class certification, the class

definition remains a relevant consideration insofar as its contours inform the

Rule 23 analysis. See Jackson v. Se. Pa. Transp. Auth.,

(E.D. Pa. 2009); O’Neill v. The Home Depot U.S.A., Inc.,

(S.D. Fla. 2006). Accordingly, I consider each of the defendants’ arguments

before proceeding to the elements of Rule 23.

1. Ascertainability

In addition to the enumerated requirements of Rule 23, some courts

have found that the rule contains an “implicit threshold requirement” that

the class be sufficiently ascertainable. See Sandusky Wellness Ctr., LLC v.

Medtox Sci., Inc.,

(8th Cir 2016) (collecting cases).

Ascertainability requires, at a minimum, that the class members be capable

of identification by reference to objective criteria without the need for

individualized fact finding and litigation. See William B. Rubenstein,

Newberg & Rubenstein on Class Actions § 3:3 (6th ed.) (hereinafter

“Newberg”) (noting that, despite “linguistic variations” among courts in

explaining the ascertainability requirement, “[a]ll courts essentially focus on

the question of whether the class can be ascertained by objective criteria”).

Among those courts that have recognized the implied requirement of

ascertainability, there is a circuit split as to whether classes proceeding

under Rule 23(b)(2) must meet the ascertainability requirement. See Cole v.

15 City of Memphis,

(6th Cir. 2016) (collecting cases). While

some circuits require all class actions to satisfy the ascertainability

requirement, the First Circuit is not among them. See Fitzmorris v. N.H.

Dep’t of Health & Hum. Servs.,

,

, at *7-8

(D.N.H. Nov. 27, 2023) (rejecting a similar argument).

The First Circuit has squarely held that ascertainability is a

“standard[] applicable to a subdivision (b)(3) class rather than to a

subdivision (b)(2) class.” Yaffe v. Powers,

(1st Cir. 1972).

Although the court subsequently recognized that ascertainability may be

required in (b)(2) classes where the class’s requested relief would require the

identification of class members, this is not such a case. See Crosby v. Soc.

Sec. Admin.,

, 580 (1st Cir. 1986). B.D. seeks changes to the

state’s operation of its foster system as a whole, which can be implemented

without the need to identify or provide notice to individual class members.

Accordingly, even if the defendants are correct that individualized inquiries

are required to determine which adolescent foster children are unnecessarily

placed in congregate care facilities or at risk thereof, it would provide no

basis for denying class certification.

2. “At Risk” Claims

The defendants next contend that the class cannot be defined to include

individuals who are at risk of unnecessary placement in a congregate care

16 facility because those individuals would fail to state a claim under Olmstead.

In the defendants’ view, placing an individual at risk of unnecessary

institutionalization, without actual institutionalization, does not violate the

ADA.

The defendants’ argument consists of only a single paragraph and fails

to grapple with the fact that the overwhelming majority of circuit courts to

have considered the issue have concluded that an ADA claim may be based

on a risk of unnecessary institutionalization. Compare Waskul v. Washtenaw

Cnty. Cmty. Mental Health,

, 460 (6th Cir. 2020), Steimel v.

Wernert,

(7th Cir. 2016), Davis v. Shah,

(2d Cir. 2016), Pashby v. Delia,

(4th Cir. 2013), M.R. v.

Dreyfus,

(9th Cir. 2012), and Fisher v. Okla. Health Care

Auth.,

(10th Cir. 2003), with United States v.

Mississippi,

, 392 (5th Cir. 2023). Moreover, the defendants do

not explain why the alleged legal defects in B.D.’s theory of the case would

preclude certification.

It would be premature to weigh in on the defendants’ arguments now,

which fundamentally go to the merits of the case and, in any event, are

underdeveloped. See Fed. R. Civ. P. 23 advisory committee notes (2003) (“an

evaluation of the probable outcome on the merits is not properly part of the

certification decision[.]”); Gooch v. Life Invs. Ins. Co. of Am.,

,

17 432 (6th Cir. 2012) (“the relative merits of the underlying dispute are to have

no impact upon the determination of the propriety of the class action.”)

(quoting Thompson v. Cnty. of Medina,

, 241 (6th Cir. 1994)); see

also Higgins v. New Balance Athletic Shoe, Inc.,

(1st Cir.

1999) (“The district court is free to disregard arguments that are not

adequately developed . . . . ”). B.D.’s proposed class definition tracks their

theory of the case, the viability of which can be determined at a later stage of

the proceedings following appropriate briefing and argument. Should the

defendants’ argument succeed on the merits, the class can be redefined

accordingly. See Fed. R. Civ. P. 23(c)(1)(C).

B. Numerosity

Rule 23(a)(1) requires the plaintiff to establish that the proposed class

is “so numerous that joinder of all members is impracticable[.]” Although

commonly referred to as the “numerosity” requirement, the inquiry under

Rule 23(a)(1) turns not on “the number of class members alone but the

practicability of joinder.” Anderson v. Weinert Enters. Inc.,

, 777

(7th Cir. 2021); accord Andrews v. Bechtel Power Corp.,

(1st Cir. 1985). Thus, while a “class of 40 or more members raises a

presumption of impracticability of joinder,” courts may consider a variety of

other factors as well. Newberg § 3:12; see also Garcia-Rubiera v. Calderon,

(1st Cir. 2009). Relevant factors include “judicial economy,

18 the claimants’ ability and motivation to litigate as joined plaintiffs, the

financial resources of class members, the geographic dispersion of class

members, the ability to identify future claimants, and whether the claims are

for injunctive relief or for damages.” In re Modafinil Antitrust Litig.,

(3d Cir. 2016).

To support their demonstration of numerosity, B.D. has provided

evidence that there are approximately 135 adolescent foster children in

DCYF custody who suffer from a “per se mental disability.” 8 Doc. 277-2 at 8.

Out of those 135 individuals, 104 experienced at least one congregate care

placement.

Tracey Feild, a child welfare expert with years of experience in

the field, opined that this rate of congregate care placements was “higher

than expected or appropriate” and that, in her experience, most adolescent

foster children with mental impairments can remain in the community with

appropriate supports. Doc. 154-3 at 9. Feild explained that it is widely

accepted that congregate care placements are appropriate only for “a small

subset of children with the most serious and acute mental impairments,” and

8 A “per se disability,” as used here, means an impairment that generally limits a major life activity and will therefore be regarded as a disability under the ADA “in virtually all cases.” See

(d)(2)(ii)-(iii). Such impairments include, for example, “[m]ajor depressive disorder, bipolar disorder, post-traumatic stress disorder, traumatic brain injury, obsessive compulsive disorder, and schizophrenia[.]” Id.; see also Doc. 277-2 at 8 n.13.

19 found it highly unlikely that such a significant percentage of adolescent foster

children in New Hampshire would fit within this “small subset.” Doc. 154-3

at 9; see also Doc. 178-5 at 24.

Feild’s view received further support from Dr. Theodore Cross, a child

psychologist and professor of social work, who explained that the “standard of

care for kids with mental impairment[s] is to provide care in a community-

based setting” and that this standard “can be applied to the vast majority of

youth.” Doc. 180-1 at 28-29. Cross explained that residential treatment

should be used only in “very rare” circumstances where the “child’s safety and

the safety of others are in question.” Doc. 322-8 at 13. Cross further explained

that children not currently in congregate care are at serious risk of

unnecessary placement in congregate care if:

(1) the child has clinically significant emotional or behavioral problems; (2) the child’s behavior was a factor for initial placement in out-of-home care; (3) the child’s behavior resulted in a placement disruption after 50 days in foster care; (4) the child returned to foster care after exit; or (5) the child was previously placed in congregate care or an inpatient psychiatric care facility.

Doc. 154-2 at 6.

This evidence, viewed collectively, demonstrates a sufficient likelihood

that there are at least 40 members of the proposed class. Although the

defendants are correct that B.D.’s evidence does not establish the precise

number of adolescent foster children who are unnecessarily placed in

20 congregate care or at risk of unnecessary placement in congregate care, no

such showing is required. See Newberg § 3:13 (collecting cases and noting

that “it is well settled that a plaintiff need not allege the exact number or

specific identity of proposed class members”); see also Gomes v. U.S. Dep’t of

Homeland Sec.,

, 99 (D.N.H. 2021). Rather, the court “may

draw reasonable inferences from the facts presented to find the requisite

numerosity.” McCuin v. Sec’y of Health & Hum. Servs.,

(1st Cir. 1987); see also Marcus v. BMW of North Am., LLC,

(3d Cir. 2012) (noting that, although “[m]ere speculation is

insufficient,” courts can find numerosity using “inferences” supported by

“circumstantial evidence”).

Considering B.D.’s data on the prevalence of congregate care

placements alongside the experts’ opinion that congregate care placements

are warranted only rarely gives rise to an inference that at least some subset

of adolescent foster children in congregate care were placed there

unnecessarily. Furthermore, the evidence indicates that a significant portion

of foster children have a “clinically significant emotional or behavioral

problem”—that is, a mental health problem that affects the child’s “quality of

life, well-being, and functioning”—and that a large number of foster children

have been placed in congregate care. Doc. 180-1 at 50, 57-58 (noting that the

“vast majority” of foster children with a mental health diagnosis are likely to

21 have a mental health diagnosis that impacts “a major life activity”); Doc. 216

at 2 (acknowledging that “[n]early 95 percent of Older Foster Youth had at

least one DSM diagnosis”); Doc. 153-16 at 2 (finding that 293 foster children

in New Hampshire were in congregate care placements on a given day in

2021). This evidence, viewed alongside Cross’ opinion that children with such

characteristics are at serious risk of unnecessary institutionalization,

supports an inference that an even greater number of adolescent foster

children are at risk of unnecessary placement in congregate care.

Aside from the sheer number of individuals in the putative class,

several characteristics of the proposed class members render joinder

impracticable. First, because children are constantly entering and exiting

DCYF custody, the class is inherently fluid such that it would be “not merely

impracticable but effectively impossible” to identify and join all class

members. Gomes, 561 F. Supp. 3d at 99 (cleaned up); see also Reid v.

Donelan,

, 189 (D. Mass. 2014). Second, the class is

geographically diffuse: it spans across the entire state and may even include

some New Hampshire children in out-of-state residential treatment

programs. See Risinger ex rel. Risinger v. Concannon,

(D.

Me. 2001). Finally, the class members are all children who suffer from

disabilities, many of whom are segregated from the greater community in

congregate care placements. Identifying, contacting, and bringing individual

22 litigation on behalf of each of these individuals would present significant

challenges that are only compounded by the fact that the putative class

members are under the defendants’ custody. See J.N. v. Or. Dep’t of Educ.,

, 264 (D. Or. 2021). These considerations, combined with the

likely number of putative class members, demonstrate that joinder would be

impracticable.

C. Commonality

Commonality under Rule 23(a)(2) requires the plaintiff to demonstrate

that there is at least one “question[] of law or fact common to the class.” To

satisfy this requirement, the plaintiff must prove that each of the class claims

“depend upon a common contention,” the “truth or falsity” of which “will

resolve an issue that is central to the validity of each one of the claims in one

stroke.” Wal-Mart,

. Thus, “[w]hat matters to class

certification is not the raising of common questions—even in droves—but

rather the capacity of class-wide proceedings to generate common answers

apt to drive the resolution of the litigation.”

(cleaned up). As the First

Circuit has recognized, such “common answers typically come in the form of a

particular and sufficiently well-defined set of allegedly illegal policies or

practices that work similar harm on the class plaintiffs.” PPAL,

(cleaned up).

23 “Mere allegations of systemic violations of the law” will not suffice. D.G.

ex rel. Stricklin v. Devaughn,

(10th Cir. 2010). Rather,

in cases challenging systemic deficiencies, commonality generally requires

proof of uniformly applicable policies or practices that allegedly drive the

class harm and give rise to common questions capable of yielding common

answers. See Parsons v. Ryan,

(9th Cir. 2014); see also

Fitzmorris v. N.H. Dep’t of Health & Hum. Servs.,

,

, at *5 (D.N.H. Apr. 17, 2023).

B.D. claims to have made such a showing here. B.D. alleges that the

defendants’ systematic failure to provide adolescent foster children with

legally sufficient case plans and appropriate community-based placements is

a result, not of discretionary decisions by individual case workers, but rather

of deficiencies in the defendants’ administration of the foster care program

writ large. Specifically, B.D. contends that the defendants fail to fund,

develop, and maintain an adequate array of community-based placements

and services and, furthermore, engage in inadequate case planning practices

by failing to regularly update case plans, inadequately training and

supervising their caseworkers, and utilizing an outdated IT system. 9

9 B.D. identifies other common practices that allegedly drive the class harm. Because I find that B.D. has proven the existence of several common practices, each of which gives rise to common questions that satisfy the

24 The defendants argue that B.D. has not demonstrated that these

practices are occurring but that, regardless, the allegedly common practices

do not give rise to common questions. In the defendants’ view, given the

inherently individualized and discretionary nature of foster care placements

and the plaintiff-specific inquiries that inform each of the class claims, there

is no common question that could resolve the claims of all class members in a

single stroke.

I begin by analyzing B.D.’s showing of commonality with respect to

their CWA claim before turning to their ADA claims.

1. CWA Claim

B.D.’s case planning claim is premised on the assertion that the

defendants systematically fail to provide adolescent foster children with case

plans that meet the substantive requirements of the CWA. Although there is

some evidence that the majority of foster children receive an initial case plan,

see Doc. 153-6 at 54; Doc. 326-30 at 3; the record indicates that the

defendants do not regularly update case or transition plans for adolescent

requirements of Rule 23(a)(2), I need not consider at the present time whether the other alleged practices are occurring in a way that is common to the class. See Immigr. & Naturalization Serv. v. Bagamasbad,

(1976) (“As a general rule courts . . . are not required to make findings on issues the decision of which is unnecessary to the results they reach.”); see also Wal-Mart,

(noting that “even a single common question” will satisfy commonality) (cleaned up).

25 foster children. Dr. Bryan Victor, an expert in data analysis, analyzed the

case files of approximately 184 adolescent foster children to determine how

many case and transition plans were provided to each child over the course of

a 15-month period. 10 Doc. 281-12 at 2. Victor found that over half of

adolescent foster children lacked a single case plan dated within the 15-

month period, and approximately three-quarters lacked a transition plan

dated within that same period. 11 Id. at 10. Out of those children who had an

initial case or transition plan developed, approximately 3% had updates

made to their case plan every seven months, and only 12% had updates made

to their transition plans every 13 months. Id. at 12-13. Furthermore, case and

transition plans were not regularly updated following a change in placement:

10 The 15-month period spanned from September 2021, when the defendants began using their current case planning template, to December 2022, when the defendants began production of the putative class members’ case files. Doc. 279 at 27 n.52.

11 The defendants submit evidence that Victor’s search of the case files failed to identify at least 12 case plans that were dated within the 15-month period under review. See Docs. 326-6 through 326-13; Doc. 326-22; Doc. 326- 23. Considering the additional 12 case plans would mean that 110 out of 184 foster children (59.7%) lacked a case plan dated during the period under review, calling into question Victor’s assertion that 122 out of 184 foster children (66.3%) lacked such a case plan. See Doc. 281-12 at 10. Regardless, even if Victor’s calculation was off by a few percentage points, the evidence demonstrates that more than half of adolescent foster youth lacked a case plan dated within the 15-month period under review.

26 Approximately 27% of children’s case plans and 8% of children’s transition

plans were updated within 60 days of a change in placement. Id. at 15.

The defendants argue that this evidence falls short of demonstrating

the existence of class-wide case planning deficiencies because, contrary to

B.D.’s contentions, the CWA does not require case or transition plans to be

updated on a regular basis. Rather, in the defendants’ view, all the law

requires is that foster children be provided with a case plan shortly after

removal and a transition plan sometime after turning 14. The defendants

contend that Victor’s analysis is therefore irrelevant because it speaks only to

whether “the youth’s case plan was updated during the [period under

review]” and “does not mean that the youth did not have a case plan or that

the case plan was not updated before or after the [period under review].” Doc.

322 at 30 (emphasis in original).

To be sure, Victor’s analysis does not demonstrate that the defendants

fail to provide adolescent foster children with initial case or transition plans.

But B.D.’s theory of the case is not so limited. B.D. contends that the

defendants violate the CWA, not only by failing to provide legally sufficient

initial plans, but also by failing to regularly update foster children’s case and

transition plans. See Doc. 316 at 34. Victor’s analysis is entirely relevant to

this theory and demonstrates that a substantial portion of the putative class

have suffered an alleged CWA violation. Whether B.D.’s theory of liability is

27 correct is an issue that can be determined at a later point in the proceedings.

What matters for class certification is that B.D. has demonstrated, through

Victor’s analysis, that the defendants do not regularly update case or

transition plans, either on a set basis or in response to changes in

placements. Accordingly, B.D.’s claims raise common questions regarding

whether the defendants’ failure to update case and transition plans violates

the CWA.

Beyond the failure to update case plans, B.D. has proven that the

defendants are engaging in a number of other common practices that are

allegedly insufficient to meet the defendants’ obligations under the CWA. 12

For example, although DCYF has some “adolescent case workers” that

receive specialized training on the needs of adolescents and the case planning

requirements that attach once children turn 14, not all adolescent foster

12 The defendants argue that B.D. previously framed these practices only as “potential remedies,” rather than “drivers of the alleged case planning harm,” and that they have had no opportunity to respond to this new theory. Doc. 340 at 1. Contrary to the defendants’ assertions, B.D. and their predecessors as named plaintiffs have consistently argued that the identified practices cause the defendants’ failure to provide adequate case plans. See, e.g., Doc. 154 at 16-18. Although B.D. has not explicitly used the term “drivers” when doing so, the defendants were clearly aware of the substance of B.D.’s theory because they responded to it. See, e.g., Doc. 195 at 27; Doc. 322 at 33. While it is true that B.D.’s expert witness stated at a deposition that the identified practices were simply “actions that DCYF could take . . . to enhance the case planning . . . process” and not necessarily violations of federal law, it is the plaintiff, and not their expert, who dictates the theory of the case. Doc. 322-15 at 30.

28 children are assigned to adolescent case workers, leaving many with case

workers who lack such qualifications. Doc. 280-2 at 2-3; Doc. 154-7 at 3. And,

although the CWA requires children over 14 and their families to be involved

in the case planning process, see

(1)(B);

(g)(1), DCYF does not require its case workers to engage in formal

discussions or hold meetings with a child and their family when developing a

case plan. Doc. 279 at 36. Rather, DCYF policy allows for the required

involvement to be achieved through “informal” discussions during “regular

interactions with the family leading up to the formalized plan development.”

DCYF Policy 1551.2; see also Doc. 279-14 at 2 (noting that “[c]ase plans are

not co-developed with parents” and that there is “[i]nconsistency statewide in

child’s involvement with case planning”). Furthermore, there is no formal

process for supervisors to monitor case worker’s case planning activities on

an ongoing basis. Doc. 280-7 at 4; Doc. 279 at 38. Finally, DCYF’s case

planning occurs through Bridges, an “aging legacy system which is over 23

years old” and in need of “replacement modernization.” Doc. 279-11 at 5; see

also Doc. 279-15 at 7. Bridges suffers from a number of deficiencies that can

make case planning more difficult, including usability issues, Doc. 279-11 at

5-6; Doc. 279-14 at 2, and lack of interoperability with other DHHS systems,

Doc. 153-7 at 39; Doc. 279-25 at 2. Although the defendants have plans to

29 replace Bridges with a new system, the system is not expected to be

implemented until the end of 2025. Doc. 279-24 at 9.

The defendants do not appear to contest that these practices are, in

fact, occurring, but nonetheless assert that they do not give rise to common

questions. In the defendants’ view, B.D.’s claims seek to challenge the

substantive sufficiency of adolescent foster children’s case plans, which can

only be determined by looking to each individual plaintiff’s case plan and

therefore cannot be litigated on a class-wide basis.

The defendants’ argument misconstrues B.D.’s theory of the case. B.D.’s

claim is premised on the assertion that the defendants’ practices fall short of

what is required to ensure that adolescent foster children are provided with

CWA-complaint case plans. See

(b)(8)(A)(ii); 675(5) (requiring

states to provide a “case review system” that “assur[es] that . . . each child

has a case plan” containing certain content). Thus, B.D. does not seek to

litigate the substantive sufficiency of any given class member’s case plan but

rather the capacity of the defendants’ case planning processes to deliver

substantively sufficient case plans. Resolving this claim will turn, not on the

contents of any individual case plan, but on the adequacy of the defendants’

processes for producing that case plan. “If [the defendants’ practices] are so

pervasively deficient as to be unlawful, there would not appear to be any

30 impediment to addressing [those] unlawful practice[s] in one stroke.” Elisa

W. v. City of New York,

, 125 (2d Cir. 2023).

All told, B.D. has demonstrated that the defendants are failing to

regularly update class members’ case and transition plans and, furthermore,

are engaging in several allegedly deficient case planning practices. B.D.’s

CWA claims therefore raise common questions regarding (1) whether the

alleged practices are, in fact, occurring; (2) whether those practices fail to

satisfy the state’s obligations under the CWA; and (3) whether the CWA

requires the defendants to update case plans on a set basis and/or following a

change in placement. Because each of these questions can be answered in

reference to common proof that will determine the viability of each class

member’s CWA claim in a single stroke, commonality is satisfied.

2. ADA Claims

B.D. asserts that the defendants are violating the ADA by

systematically and unnecessarily placing and retaining adolescent foster

children with mental impairments in congregate care. In support of their

claim, B.D. supplies evidence that New Hampshire’s use of congregate care

far exceeds the national average. A 2022 federal report found that 27% of all

foster children in New Hampshire are in congregate care placements,

compared to 9% of foster children nationally. Doc. 153-15 at 2; Doc. 153-16 at

2. The problem is even more pronounced for adolescent foster children with

31 mental impairments: According to Victor, B.D.’s data analysis expert, 77% of

adolescent foster children with mental impairments have experienced at least

one congregate care placement, with many experiencing two or more such

placements. 13 Doc. 277-2 at 8; see also Doc. 281-3 at 38.

Once adolescent foster children are placed in congregate care, they tend

to remain there for several months. The median length of congregate care

placements for adolescent foster children with mental impairments in New

Hampshire is 16 months. Doc. 277-2 at 9. By way of comparison, one study of

congregate care placements across 15 other states found that the median

length of a congregate care placement for adolescent foster children was

between 72 and 84 days.

And, because children are frequently returned to

congregate care after exiting, their cumulative time spent in congregate care

is even longer. On average, adolescent foster children with mental

impairments in New Hampshire spent 2.5 years in congregate care settings.

Id. at 8-9. B.D.’s child welfare expert opined that these statistics represent an

overreliance on congregate care—a concern that has been raised previously

13 Although the defendants criticize Victor’s data as stale and therefore not representative of the current state of affairs, they do not dispute that he utilized the most recent data available to him nor do they submit evidence that circumstances have changed.

32 by multiple third parties charged with analyzing New Hampshire’s foster

care system. Doc. 277-2 at 9-10; Doc. 153-1 at 19; Doc. 207-10 at 4.

B.D. contends that this overreliance on congregate care is the result of

the defendants’ deficient practices, which leave decisionmakers with virtually

no choice but to place and retain adolescent foster children in congregate

care. Specifically, B.D. contends that the defendants’ failure to fund and

maintain a sufficient array of community-based placements and services, as

well as their failure to ensure that foster children have adequate case plans,

systematically funnels adolescent foster children into congregate care. 14

The defendants deny that these common practices are occurring, but

they also contend that, even if they were, they fail to give rise to common

questions capable of yielding common answers. I begin by evaluating B.D.’s

evidence of the alleged practices before considering whether they satisfy the

requirements of commonality.

14 The defendants assert that B.D. and their predecessors first raised the theory that case planning failures are linked to the claimed ADA violation in their supplemental reply. Not so. Although B.D. did not previously label the defendants’ case planning failures as “drivers” of their ADA claim, they have asserted from the beginning that the lack of sufficient case plans results in unnecessary institutionalization. See, e.g., Doc. 154 at 5; Doc. 207 at 16. And, contrary to the defendants’ contentions, B.D.’s supplemental expert reports continued to press this claim. Doc. 310-4 at 2. The defendants therefore had adequate notice of this aspect of B.D.’s claims and the opportunity to respond accordingly.

33 a. Evidence of Practices

For the reasons I have explained, B.D. has adequately proved that the

defendants are engaging in certain allegedly deficient case planning

practices. B.D. has also demonstrated that the defendants are (1) failing to

take certain actions to develop the array of community-based placements, (2)

disproportionately investing in residential treatment facilities, and (3)

offering an insufficient array of community-based services.

i. Community-Based Placements

The evidence supports B.D.’s claim that the defendants are failing to

take certain steps to increase the supply of community-based placements. For

example, there is no dispute that the defendants do not currently offer

therapeutic foster care or adolescent foster care, despite having the

regulatory authority to do so. Doc. 277-29 at 3; Doc. 153-25 at 5. Although the

defendants are in the process of developing a therapeutic foster care program,

the project has been pending since at least 2019 yet remains in its early

stages. Doc. 277-29 at 3; Doc. 277-27 at 3.

Furthermore, there is no dispute that most kinship caregivers are paid

less than licensed foster families. Unlicensed kin are ineligible to receive the

foster care maintenance payments provided to licensed foster families and,

instead, are only eligible for a “TANF child-only grant.” Doc. 277-2 at 12. The

TANF grant pays $753 per month to a caregiver caring for one child, whereas

34 licensed families receive approximately $1,223 per month in foster care

maintenance payments per child. Doc. 317-1 at 12; Doc. 317-13 at 2.

Although kinship caregivers could receive foster care maintenance payments

by obtaining a foster family license, the vast majority—approximately 90%—

have not done so. Doc. 277-2 at 12; Doc. 277-9 at 2.

Feild, B.D.’s child welfare expert, opined that the TANF grant rates are

insufficient to cover the rising costs of raising a child and hinder kin’s ability

to serve as placements. Doc. 277-2 at 11-13; Doc. 322-1 at 32. Despite this,

the defendants have not increased payments to unlicensed kin by, for

example, supplementing the TANF payments, nor have they instituted a

process to ensure that all kin become licensed. Doc. 277-2 at 13; Doc 322-1 at

30; see also Doc. 277-13 at 2 (indicating that several states in the region

license all kinship caregivers); Doc. 317-14 at 8 (noting that 13 states provide

unlicensed kinship caregivers with foster care maintenance payments).

Additionally, the defendants have failed to recruit a sufficient number

of licensed foster homes. The defendants have regularly acknowledged that

they are in need of more foster homes to serve the state’s population of foster

children. See, e.g., Doc. 277-21 at 2; Doc. 277-22 at 2; Doc. 277-27 at 5. This

need is particularly urgent for adolescent foster children, as only a small

percentage of foster families are willing to serve as placements for teenagers.

Doc. 277-2 at 21; Doc. 278-31 at 40. Although the defendants have taken

35 some additional steps to recruit more foster families and, in particular, foster

families willing to serve teenagers, the total number of foster homes in the

state has fallen by 31% in recent years. Doc. 317-19 at 2.

A similar problem exists with regard to ISO foster care. There are

approximately 82 ISO foster homes in the state, which the defendants have

recognized is too few to meet the needs of all foster children. Doc. 277-2 at 17;

see also Doc. 153-5 at 8; Doc. 277-27 at 5. Despite recent efforts to increase

oversight of the child placing agencies contracted to recruit and train ISO

foster homes, there is no evidence that the defendants have meaningfully

increased the supply of ISO foster care.

Although the defendants do not dispute that they lack a sufficient

number of foster homes to serve the needs of adolescent foster children, they

contend that this shortage is due to factors outside their control and therefore

cannot constitute a uniform practice giving rise to common questions. The

defendants’ argument, however, goes to the merits rather than the existence

of a common practice. See Jonathan R. v. Justice,

, 308 (S.D.

W. Va. 2023). Because all class members are affected by what the defendants

do and do not do to increase the supply of available foster homes, the legal

sufficiency of the defendants’ efforts raises common questions capable of

class-wide resolution. Cf. Baby Neal v. Casey,

(3d Cir. 1994)

(finding commonality where a class of foster children challenged the

36 defendants’ failure to provide a sufficient number of appropriate placements);

Jonathan R., 344 F.R.D. at 305-306 (same); Wyatt B. v. Brown, No. 6:19-cv-

005556-AA,

, at *25 (D. Or. 2022) (same); Connor B. ex rel.

Vigurs v. Patrick,

(D. Mass. 2011) (same); see also S.R. v.

Pa. Dep’t of Hum. Servs.,

, 110 (M.D. Pa. 2018) (noting that

“systemic deficiencies in the availability of placements” are “exactly the type

of ‘common mode’ or practice predicating each alleged violation that was

noticeably absent from [Wal-Mart]”).

ii. Disproportionate Funding of Residential Treatment

B.D. asserts that the defendants invest a disproportionate amount of

funds into the development of residential treatment facilities, to the

detriment of congregate care placements. In 2022, DCYF spent 92% of its

placement funds on congregate care, expending only $1.5 million on

community-based placements for adolescent foster children compared to

$16.6 million on congregate care placements. Doc. 277-2 at 23-24. On

average, DCYF paid $7,317 per child in community-based placements,

compared to $119,763 per child in congregate care placements.

Feild

acknowledged that congregate care placements are generally more expensive

than community-based placements but nonetheless opined that the

defendants’ ratio of spending reflected a disproportionate investment in

congregate care. Doc. 322-1 at 22.

37 The defendants have also invested in the development of residential

treatment facilities. In 2021, DHHS contracted to add approximately 100 new

residential treatment beds and instituted a rate increase for residential

providers. Doc. 277-2 at 21; Doc. 278-17 at 3; Doc. 317-20 at 15-16. That same

year, DHHS requested approximately $232 million in funds to invest in

residential treatment programs over the course of three years. Doc. 317-20 at

15-16.

The defendants do not appear to dispute these underlying facts but

argue that their disproportionate investment in residential treatment

programs cannot constitute a common practice for at least two reasons. First,

the defendants note that investments in residential treatment programs are

overseen by the Bureau of Children’s Behavioral Health (BCBH), a separate

a division of DHHS, and are therefore unrelated to DCYF’s operation of its

foster care program. But BCBH and DCYF both fall under the purview of the

Commissioner of DHHS and, ultimately, the governor, both of whom are

named defendants in this case. Doc. 317-20 at 2. Even if the defendants are

correct that some of the funding was specifically earmarked for BCBH, the

decision to push other funding toward BCBH instead of DCYF is a common

practice attributable to the defendants.

Second, the defendants assert that the investments in residential care

were necessitated by a change in law that made residential treatment

38 available to all children, rather than just children in DCYF custody. But that

the investments were intended to benefit children outside of DCYF’s care

does not mean that it did not have an impact on children within DCYF’s care.

Feild explained that, in her experience, “if more congregate care beds become

available, they are likely to be filled” by adolescent foster children. Doc. 154-3

at 13. Furthermore, funds put toward residential treatment programs are

necessarily funds that did not go toward expanding community-based

placements. Thus, the defendants’ decision to invest in residential treatment

programs rather than community placements is a practice that is common to

the class, regardless of whether or not that decision was justified. Cf.

Kenneth R. ex rel. Tri-County CAP, Inc./GS v. Hassan,

(D.N.H. 2013) (finding commonality based, in part, on “practices relating to

the funding and provision of community-based services”).

iii. Community-Based Services

Finally, the evidence indicates that, although the state offers a broad

range of services to foster children in community placements, it lacks the

capacity to serve all foster children who might benefit from those services. As

early as 2018, a third-party consultant hired by the state observed that the

“current level of DCYF community based and residential services are not

sufficient to meet the needs of DCYF involved children, youth, and families.”

Doc. 153-7 at 36. These concerns were echoed in a 2018 federal audit of New

39 Hampshire’s foster care system, which noted that DCYF’s “ability to

generally meet the mental health needs of children and youth” was

negatively impacted by “a lack of quality mental health resources.” Doc. 153-6

at 42; see also Doc. 153-10 at 22 (noting that “[s]ervice availability was a

challenge across the state”). Then again in 2021, the New Hampshire Office

of the Child Advocate—an independent watchdog agency, see

-V:2—raised concerns about the state’s “limited community-based

services array for alternatives to congregate care” for children in DCYF

custody. Doc. 153-1 at 19.

Although the defendants criticize B.D.’s evidence as stale, they have

not provided any evidence of their own suggesting that circumstances have

meaningfully changed. Accordingly, the evidence supports B.D.’s contention

that the defendants have failed to maintain a sufficient array of community-

based services. See Baby Neal,

(certifying a class of foster

children challenging an insufficient array of services); S.R., 325 F.R.D. at 109

(same); Connor B.,

(same); see also Jonathan R., 344 F.R.D.

at 312 (collecting cases and noting that “several district courts have certified

classes of disabled individuals challenging a public entity’s failure to provide

community services to them”); cf. M.D., 675 F.3d at 847-848 (recognizing that

“a failure to correct a structural deficiency” could constitute a common

practice that gives rise to commonality).

40 b. Whether the alleged practices support commonality

The defendants argue that these common practices, even if adequately

proved, fail to establish commonality. The defendants contend that

commonality requires the plaintiff to prove not only the existence of common

practices but also that the common practices “cause the alleged classwide

harm or legal violation,” which B.D. has failed to do here. Doc. 195 at 23. The

defendants’ argument is principally based on PPAL, where the First Circuit

discussed the need for plaintiffs seeking class certification to identify a

“uniformly applied, official policy of the [defendant], or an unofficial yet well-

defined practice, that drives the alleged violation.”

. In the

defendants’ view, a practice cannot “drive[] the alleged violation” absent proof

of a causal link.

But the question of whether class action plaintiffs are required to

demonstrate a causal link between their challenged practices and alleged

harm was not at issue in PPAL, which turned entirely on the plaintiffs’

failure to prove the existence of a common practice. See

. Moreover,

as I recently explained in Fitzmorris, the language employed by both the

First Circuit in PPAL and the Supreme Court in Wal-Mart implies that it is

sufficient for plaintiffs to demonstrate that there are common practices that

allegedly cause their class harm.

, at *21. The question of

causation is one for the merits, and therefore need not be resolved at the class

41 certification stage. See Messner v. Northshore Univ. HealthSystem,

(7th Cir. 2012) (“the court should not turn the class certification

proceedings into a dress rehearsal for the trial on the merits.”); Sullivan v.

DB Inv., Inc.,

(3d Cir. 2011) (“the Rules and our case law

have consistently made clear that the plaintiffs need not actually establish

the validity of claims at the class certification stage.”).

The defendants next contend that B.D.’s claims turn on highly

individualized inquiries regarding the “legality of hundreds of placement

decisions” that defeat commonality. Doc. 195 at 25. This argument, however,

rests on a misunderstanding of B.D.’s claims. B.D. does not challenge any

particular placement decision, nor do they seek the sort of individualized

relief that would necessitate an inquiry into the propriety of any given class

member’s placement. See S.R., 325 F.R.D. at 109 (distinguishing between

claims that “seek review and determination of each individual class member’s

placement” and those that seek “system wide change”). Rather, B.D.’s claims

challenge specific acts and omissions in the defendants’ operation of its foster

care program that allegedly expose all class members to an ADA violation.

Accordingly, the focus of B.D.’s claims will be on the defendants’ actions, each

of which are common to the class, rather than any individual plaintiff’s

placement. These are precisely the sort of claims that have been broadly

recognized as appropriate for class treatment because they give rise to

42 common questions regarding the legal sufficiency of the defendants’ practices.

See PPAL,

n.14 (collecting cases that challenged “a definable

policy or practice imposed by a single entity or a small group of actors” which

“facilitated the formulation of questions apt for class resolution”); Elisa W.,

82 F.4th at 123 (“Where the same conduct or practice by the same defendant

give rise to the same kind of claims from all class members, there is a

common question.”) (quoting Johnson v. Nextel Commc’ns, Inc.,

(2d Cir. 2015)); see also Newberg § 3:20 (“When the party opposing the

class has engaged in some course of conduct that affects a group of persons

and gives rise to a cause of action, one or more of the elements of that cause

of action will be common to all of the persons affected.”).

The defendants nonetheless assert that Olmstead claims based on a

systemic failure to provide appropriate mental health care are fundamentally

incompatible with class treatment. In support of their contention, the

defendants rely on the Fifth Circuit’s recent decision in United States v.

Mississippi.

(5th Cir. 2023). In that case, the Department of

Justice (DOJ) brought suit against the state of Mississippi on the theory that

“systemic deficiencies in the state’s operation of mental health programs”

placed “every person in Mississippi suffering from a serious mental illness . . .

at risk of improper institutionalization.” Id. at 389 (emphasis omitted). The

Fifth Circuit reversed the district court’s finding in favor of the DOJ, noting

43 that a “claim of system-wide risk of institutionalizing some unspecified group

of patients” was incompatible with Olmstead, which required an inquiry into

“patient-specific” factors such as whether community placement is

appropriate and desired. Id. at 394. Although the court recognized that “a

number of circuits” had held that Olmstead claims could be litigated on a

systemwide basis, it nonetheless found those cases distinguishable because

they were brought by individual plaintiffs or classes challenging the failure to

provide certain “personal care services or medically necessary items[.]” Id. at

396. In the court’s view, although the consequences of failing to provide

medically necessary services or items were “susceptible of quantification and,

indeed, generalization,” the failure to provide “‘[a]pproriate’ treatment for

those with serious mental illness” could not be litigated on a system-wide

basis. Id.

As an initial matter, Mississippi is distinguishable from the present

case. The claims at issue in Mississippi were brought by the federal

government on the theory that a multifaceted, statewide system exposed a

diverse group of nearly 4,000 individuals to a uniform risk of

institutionalization. This case, in contrast, has been brought by a putative

class of less than 200 adolescents within the state’s care and custody against

a discrete set of state actors with authority over the children’s placements

and mental health care. The Fifth Circuit’s conclusions regarding the DOJ’s

44 ability to prove that Mississippi subjected thousands of unidentified

individuals across the state to an unnecessary risk of institutionalization

says little about this plaintiff’s ability to prove that the defendants subject a

narrow class of children within their care to an unnecessary risk of

institutionalization. Cf. D.G.,

(distinguishing a prior case

denying class certification on the basis that the “class of plaintiffs in [the

prior case] was far broader” and shared less in common than proposed class

at issue).

To the extent Mississippi stands for the proposition that Olmstead

claims based on the failure to provide appropriate mental health care can

never be litigated on a class-wide basis, it is unpersuasive. I am instead

persuaded by the long line of cases finding that such claims are appropriate

for class treatment precisely because they are susceptible to common proof

regarding the aggregate effect and legal sufficiency of identified state

practices. 15

15 See, e.g., Jonathan R., 344 F.R.D. at 313; Wyatt B.,

, at *27; J.N., 338 F.R.D. at 266; S.R., 325 F.R.D. at 109; N.B. v. Hamos,

(N.D. Ill. 2014); Kenneth R.,

; see also Lane v. Kitzhaber,

(D. Or. 2012) (noting that class certification has been granted “in almost every case” involving an alleged “violation of the integration mandate of the ADA”); Steven Schwartz & Kathryn Rucker, The Commonality of Difference: A Framework for Obtaining Class Certification in ADA Cases After Wal-Mart,

, 872 (2021) (noting that the theory that Olmstead cases cannot be litigated on

45 B.D.’s claims are no exception and raise a host of common questions

amenable to common answers. In order to succeed on their ADA claims, B.D.

will need to demonstrate that the defendants’ operation of its foster care

program subjects class members to unnecessary institutionalization or risk

thereof. M.R.,

; see also Shah,

. B.D. will also

need to demonstrate that the defendants’ practices could be reasonably

modified, considering the states resources and obligations to others with

disabilities. Olmstead,

. And the court will need to consider

the defendants’ argument that ADA claims cannot be based on a risk of

institutionalization, as well as its affirmative defenses that the integration

mandate is ultra vires and the proposed modifications would constitute a

fundamental alteration. Doc. 57 at 36. See In re Checking Account Overdraft

Litig.,

(S.D. Fla. 2015) (noting that affirmative

defenses may “raise common questions” where they can be addressed

“through the use of common evidence”); cf. Waste Mgmt. Holdings, Inc. v.

Mowbray,

(1st Cir. 2000) (“affirmative defenses should be

considered in making class certification decisions”). Accordingly, B.D.’s claims

give rise to common questions regarding (1) whether the challenged practices

a class-wide basis “has received little, if any, support, primarily because it runs contrary to, and would effectively upend, two decades of ADA class certification precedent”).

46 are, in fact, occurring; (2) whether those practices subject class members to

unnecessary institutionalization or a risk thereof; (3) whether those practices

can be reasonably modified; (4) whether instituting the requested

modifications would require the state to fundamentally alter its program; (5)

whether the integration mandate is ultra vires; and (6) whether ADA claims

can be based on a risk of unnecessary institutionalization. Because each of

these questions turn on common proof and give rise to common answers,

commonality is satisfied.

D. Typicality

Typicality under Rule 23(a)(3) requires that “the claims or defenses of

the representative parties are typical of the claims or defenses of the class[.]”

“The logic behind the typicality requirement ‘is that a class representative

will adequately pursue her own claims, and if those claims are typical of

those of the rest of the class, then her pursuit of her own interest will

necessarily benefit the class as well.’” Howard v. Cook Cnty. Sheriff’s Off.,

, 605 (7th Cir. 2021) (quoting Newberg § 3:28). Typicality is

satisfied where the named plaintiff’s claims “arise from the same event or

practice or course of conduct that gives rise to the claims of other class

members, and are based on the same legal theory.” Garcia-Rubiera,

(cleaned up). The named plaintiff’s claims “need not be identical” to

those of absent class members so long as they “share the same essential

47 characteristics.” Wright v. S. N.H. Univ.,

, 203 (D.N.H.

2021) (quoting Rapuano v. Tr. of Dartmouth Coll.,

, 648

(D.N.H. 2020)).

The defendants do not assert that anything about B.D.’s claims in

particular render them atypical. Instead, they argue that, because the ADA

and CWA claims turn on fundamentally individualized inquiries, litigating

B.D.’s claims will do little to advance the claims of absent class members.

This argument largely tracks the defendants’ arguments on commonality and

is unpersuasive for similar reasons. See Postawko v. Mo. Dep’t of Corrs.,

(8th Cir. 2018); see also Gen. Tel. Co. of Sw. v. Falcon,

, 157 n.13 (1982) (noting that the “commonality and typicality

requirements of Rule 23(a) tend to merge”). Like the other members of the

class, B.D. alleges that they were unnecessarily placed in congregate care and

given an inadequate case plan as a result of systemic deficiencies in the

defendants’ operation of its foster care program. Both B.D.’s claims and those

of absent class members challenge these systemic deficiencies under the same

legal theory which, for the reasons I explained, can be litigated through

common proof. Therefore, B.D.’s claims are typical of the class such that

advancing their claims will necessarily advance the interests of other class

members. See Jonathan R., 344 F.R.D. at 314-315; see also Rapuano, 334

F.R.D. at 648 (collecting cases and noting that “[c]ourts have found typicality

48 satisfied where a group of putative class members are exposed to the

systemic failures of an institution”)

E. Adequacy

Rule 23(a)(4), referred to as the “adequacy” requirement, requires class

plaintiffs to demonstrate that “the representative parties will fairly and

adequately protect the interests of the class.” To satisfy this requirement, the

plaintiff “must show first that the interests of the representative party will

not conflict with the interests of any of the class members, and second, that

counsel chosen by the representative party is qualified, experienced, and able

to vigorously conduct the proposed litigation.” Andrews,

.

The defendants do not argue that B.D.’s counsel and next friend lack

the capacity to represent the class. 16 The defendants argue, however, that

B.D. is nonetheless an inadequate class representative for two reasons.

First, the defendants assert that B.D.’s interest in enforcing their ADA

rights is in conflict with other members of the class who might prefer to be

placed in congregate care. As an initial matter, “[m]ere speculation as to

conflicts that may develop at the remedy stage is insufficient to support

16 The defendants filed a motion to disqualify three of the other named plaintiffs’ next friends, which I denied as moot after dismissing those plaintiffs from the action. The defendants have not sought to disqualify B.D.’s next friend, nor is there any indication that the concerns raised in the defendants’ motion to disqualify apply to B.D.’s next friend.

49 denial of initial class certification.” Cummings v. Connell,

(9th Cir. 2003) (quoting Soc. Servs. Union Local 535 v. Cnty. of Santa Clara,

, 948 (9th Cir. 1979)); see also Gunnells v. HealthPlan Servs.,

Inc.,

, 431 n.7 (4th Cir. 2003). Regardless, while it is certainly

possible that same members of the class may desire residential treatment

now or in the future, this is not the sort of discrepancy that “go[es] to the

heart of the litigation” and “renders a putative class action untenable.”

Matamoros v. Starbucks Corp.,

(1st Cir. 2012) (noting that

Rule 23 does not require “perfect symmetry of interests”). B.D.’s interest is in

preventing unnecessary institutionalization, not all residential treatment,

and there is no reason to think that remedies aimed at preventing

unnecessary institutionalization will negatively impact those who want or

need residential treatment. Moreover, because institutionalization is

necessarily temporary, developing the array of community services and

placements will benefit class members who desire residential treatment upon

their return to the community.

The defendants next assert that, because B.D. does not receive foster

care maintenance payments under Title IV-E, they cannot adequately

represent class members seeking to invoke the case planning requirements of

Title IV-E. Again, I disagree. The CWA’s case planning requirements appear

in both Title IV-B, which applies to all members of the proposed class, and

50 Title IV-E, which applies to foster children receiving foster care maintenance

payments. As I explained in my order on the defendants’ motion to dismiss,

there is no meaningful difference between the case planning requirements of

Title IV-B and those of Title IV-E. Therefore, even though B.D.’s case

planning claims are rooted in Title IV-B, they do not “implicate a

significantly different set of concerns” than those of absent class members

proceeding under Title IV-E. See Gratz v. Bollinger,

(2003).

Accordingly, B.D. is an adequate representative.

F. Rule 23(b)(2)

In addition to satisfying each of the prerequisites of Rule 23(a), parties

seeking class certification must also demonstrate that their class fits within

one or more of the circumstances outlined in Rule 23(b). B.D. seeks

certification pursuant to Rule 23(b)(2), which allows for class treatment

where the defendant “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[.]” This

requirement consists of two parts. “The class must sue over a uniform action

or inaction by the defendant, and it must request a uniform injunction or

declaratory judgment from the court.” Doster v. Kendall,

, 439

(6th Cir. 2022). A class may proceed under Rule 23(b)(2) “only when a single

51 injunction or declaratory judgment would provide relief to each member of

the class.” Wal-Mart,

.

The first requirement of Rule 23(b)(2) is satisfied for largely the same

reasons that commonality is satisfied. That is, by demonstrating that the

defendants engage in certain common practices, B.D. has established that the

defendants’ contested actions “apply generally” to the class. See Rodriguez v.

Hayes,

(9th Cir. 2010); M.D., 675 F.3d at 847-848.

As to the second requirement, B.D. has proposed injunctive relief that

they contend would provide relief to all class members. For example, B.D.

argues that the court could order the defendants to (1) undertake additional

recruiting efforts to “expand the pool” of qualified foster homes, (2) require

community-based service providers to “reserve a set number of slots” for

adolescent foster children, (3) implement a process for case workers to receive

“regular and timely feedback” on their case planning activities, or (4) ensure

that adolescent foster children are assigned to case workers with specialized

training and knowledge on the needs of those children. Doc. 154 at 32-33.

The defendants nevertheless argue that the proposed class-wide relief

does not satisfy the requirements of Rule 23(b)(2) for at least three reasons.

First, the defendants assert that, because placement needs vary from child to

child, requiring the defendants to increase the supply of any given placement

52 type would not benefit members of the class who do not qualify for that

placement type.

This argument, however, ignores the reality of foster care. Foster care

placements are a zero-sum game in that, when one child occupies a bed, it

necessarily denies another child the opportunity to occupy that bed. In this

way, all foster children are affected by the overall number of available beds,

regardless of their individual placement needs. Requiring the defendants to,

for example, increase the supply of enhanced support foster homes would

benefit not only those children who qualify for enhanced support foster homes

but also those children who can occupy the beds made vacant by moving

other children into enhanced support foster care. An injunction requiring the

defendants to take certain steps to increase the supply of available foster care

beds would therefore benefit all class members without regard their specific

needs. See S.R., 325 F.R.D. at 112 (certifying a (b)(2) class even though the

class members had different placement and service needs); Kenneth R.,

(similar).

The defendants next argue that B.D.’s claims cannot yield any

meaningful injunctive relief because the defendants “have already (of their

own accord) accomplished or begun undertaking nearly every step Plaintiffs

have requested[.]” Doc. 322 at 36 (emphasis omitted). The key word in the

defendants’ argument is “nearly.” To be sure, in the intervening years since

53 this case was first filed, the defendants have undertaken laudable efforts to

address many of the concerns raised in the complaint. But there is no

evidence that the defendants have abated or modified the common practices

identified in this order. These practices could provide the basis for

meaningful injunctive relief should B.D. succeed in demonstrating that they

are unlawful.

Finally, the defendants assert that B.D. has failed to demonstrate that

the relief they request would definitively resolve the class members’ injuries.

But, as I recently explained in rejecting a similar argument, Rule 23(b)(2)

does not require such a showing. See Fitzmorris,

, at *26

n.20. All that is required is that an appropriate injunction “is conceivable.”

Shook v. Cnty. of El Paso,

(10th Cir. 2008). Of course, an

injunction will only issue if B.D. succeeds on the merits by demonstrating

that the challenged practices violate class members’ rights under the ADA or

the CWA. Should this occur, it is certainly conceivable that the court could

craft an appropriate and effective injunction, similar to the one proposed by

B.D., ordering the defendants to modify those practices. Because that

injunction would modify the way in which the defendants administer the

foster care system writ large, it would benefit each member of the class

regardless of their individual needs or circumstances. See Rodriguez,

(noting that Rule 23(b)(2) “does not require [the court] to

54 examine the viability or bases of class members’ claims or declaratory

judgment, but only to look to whether class members seek uniform relief from

a practice applicable to all of them”). Accordingly, the requirements of Rule

23(b)(2) are satisfied.

IV. CONCLUSION

For the foregoing reasons, B.D.’s motion for class certification (Doc.

152) is granted. The court certifies a class of:

All children, ages 14 through 17, who:

(1) are, or will be, in the legal custody or under the protective supervision of DCYF under

-C:3 (XVII) and/or (XXV);

(2) have a mental impairment that substantially limits a major life activity, or have a record of such an impairment; and

(3) currently are, or are at serious risk of being, unnecessarily placed in congregate care settings.

B.D., by and through their next friend, Christine Wellington, is appointed as

class representative. B.D.’s counsel of record is appointed as class counsel

pursuant to Rule 23(g).

SO ORDERED. /s/ Paul J. Barbadoro Paul J. Barbadoro United States District Judge

September 18, 2024

cc: Counsel of record

55