Judge Rotenberg Educational Center, Inc. v. Commissioner of the Department of Developmental Services

NOTICE: All slip opinions and orders are subject to formal
revision and are superseded by the advance sheets and bound
volumes of the Official Reports. If you find a typographical
error or other formal error, please notify the Reporter of
Decisions, Supreme Judicial Court, John Adams Courthouse, 1
Pemberton Square, Suite 2500, Boston, MA, 02108-1750; (617) 557-
1030; SJCReporter@sjc.state.ma.us

SJC-13298

JUDGE ROTENBERG EDUCATIONAL CENTER, INC.,1 & others2 vs.
COMMISSIONER OF THE DEPARTMENT OF DEVELOPMENTAL SERVICES
& another.3

Bristol. May 3, 2023. - September 7, 2023.

Present: Budd, C.J., Gaziano, Lowy, Cypher, Kafker,
& Wendlandt, JJ.

Developmentally Disabled Person. Department of Developmental
Services. Probate Court, Revocation of decree, Judicial
discretion. Judgment, Relief from judgment. Practice,
Civil, Relief from judgment. Regulation. Administrative
Law, Regulations. Constitutional Law, Separation of
powers.

1 Formerly known as Behavior Research Institute, Inc.

2 Leo Soucy, individually and as parent and next friend of
Brendon Soucy; Peter Biscardi, individually and as parent and
next friend of P.J. Biscardi; and both as representatives of the
class of all patients at Judge Rotenberg Educational Center,
Inc. (JRC), their parents, and their guardians. The former
executive director of JRC was originally listed as a plaintiff
as well.

3 Commissioner of the Department of Early Education and
Care. The director of the Office for Children, the predecessor
in interest to the defendants, was originally listed as a
defendant in her ex officio capacity.
 2

Civil action commenced in the Bristol Division of the
Probate and Family Court Department on February 28, 1986.

A motion to terminate a consent decree, filed on February
14, 2013, was heard by Katherine A. Field, J.

The Supreme Judicial Court granted an application for
direct appellate review.

Timothy J. Casey, Assistant Attorney General (Christine
Fimognari, Assistant Attorney General, also present) for the
defendants.
Max D. Stern (Joseph M. Cacace, Alexandra H. Deal, & C.
Michele Dorsey also present) for Leo Soucy & others.
Michael P. Flammia (Christian B.W. Stephens, Matthew D.
Rodgers, & Trevin C. Schmidt also present) for Judge Rotenberg
Educational Center, Inc.
Kathryn Rucker, Mona Igram, Steven J. Schwartz, & Richard
M. Glassman, for The Arc of Massachusetts & others, amici
curiae, submitted a brief.
Felicia H. Ellsworth & Charles C. Kelsh, for American
Academy of Pediatrics & others, amici curiae, submitted a brief.

KAFKER, J. The instant appeal concerns a long-standing

controversy over the treatment and welfare of a particularly

vulnerable population living within our Commonwealth. These

individuals suffer from severe developmental and intellectual

disabilities that, left untreated, cause them to engage in

grievous self-harm, maiming, and other life-threatening

behaviors. They reside in small group homes under the care of

Judge Rotenberg Educational Center, Inc. (JRC), a facility that

employs the use of aversive interventions -- most notably,

electric skin shock -- as part of its treatment approach to

severe behavioral issues. JRC, which stands as the sole
 3

facility in the country to use electric skin shock on the

developmentally disabled, currently operates under the

protection of a thirty-six year old consent decree. That decree

was entered, and has remained in place, after State agencies

resorted to pretextual and bad faith regulatory practices to

disrupt JRC's operations in the 1980s and 1990s. The State

agencies that remain bound by the decree have since moved for

its termination. That motion was denied by a judge in the

Probate and Family Court (Probate Court), and the matter now

comes before us on appeal.

For many mental health advocates, the controversial

interventions used by JRC sound reminiscent of the

institutionalization and abuse inflicted on the developmentally

disabled in decades past. Yet the families of these clients

claim that JRC has been singularly effective in preventing their

children from engaging in severely self-injurious and

destructive behaviors, such as gouging their own eyes,

puncturing their own bodily orifices, and violently attacking

others. These families characterize JRC's methods as a

treatment of last resort -- one sought after alternative

treatments either failed to protect their children from self-

harm or left them continually sedated and restrained. This case

thus involves a heart-wrenching issue: continue to protect a

controversial practice that has widely been criticized, or pave
 4

the way for its prohibition at the risk of subjecting these

vulnerable patients to a life of sedation and restraint, or

extreme self-injury.

The propriety of this controversial treatment does not

reach us in a vacuum, however. The record before us contains

extensive findings of fact made by the judge below, based on a

forty-four day evidentiary hearing that closed in 2016, with 788

exhibits and nearly thirty witnesses. Among those findings was

the judge's conclusion that the Commonwealth had yet again

resorted to bad faith regulation of JRC in 2010, and that, as of

2016, the medical community remained divided as to whether JRC's

treatment approach fell outside the professional standard of

care for the most severely disabled patients.

It is particularly troubling that the case is before us on

an evidentiary record that closed seven years ago, especially

given the fact-intensive nature of the issues at stake.

Nonetheless, because the parties have urged us to decide this

appeal without remanding for additional findings, we assess the

parties' arguments on the record we have been given. That

record compels us to conclude that the defendants have failed to

demonstrate that the judge's findings were clearly erroneous,

based on the evidence before her in 2016.

We stress, however, that our conclusion does not foreclose

the possibility that new developments will occur, or have
 5

occurred, bearing on these factual issues. Moreover, nothing in

our decision or the consent decree prevents the Department of

Developmental Services (department) from exercising its existing

authority to contest the use of electric skin shock on

individual JRC patients at their yearly substituted judgment

hearings in the Probate Court. Nor is the department precluded

from enforcing the consent decree's requirement that electric

skin shock be used only where it is the least intrusive, most

appropriate treatment. The fact that the department has largely

chosen not to do so informs the context within which we rule on

this issue. That being said, today we decide only the narrow

question of whether the judge below abused her discretion in

concluding that the department failed to establish that the

consent decree should be terminated based on the evidentiary

record before the Probate Court in 2016. We conclude that she

did not, and thus affirm the denial of the defendants' motion.4

4 We acknowledge the amicus brief submitted by the American
Academy of Pediatrics, the American Association on Intellectual
and Developmental Disabilities, the National Association of
State Directors of Developmental Disabilities Services, the
National Association of State Directors of Special Education,
the International Association for the Scientific Study of
Intellectual and Developmental Disabilities, the National
Association for the Dually Diagnosed, and the Massachusetts
Chapter of the American Academy of Pediatrics, as well as the
amicus brief submitted by The Arc of Massachusetts, the
Disability Policy Consortium, the Massachusetts Developmental
Disability Council, the Federation for Children with Special
Needs, and MassFamilies. In addition, we allow the plaintiffs'
joint motion for leave to respond to the briefs of the amici
 6

1. Background. We summarize the relevant factual findings

of the judge below, supplemented where necessary by undisputed

evidence in the record. See Connor v. Benedict, 481 Mass. 567,

568 (2019). Because the record before the Probate Court closed

in 2016, any references to "current" practices, procedures, or

statistics is only current as to that date, unless otherwise

noted.

a. JRC and its treatment methods. Since 1975, JRC5 has

operated a residential program in the Commonwealth to provide

treatment and educational services for individuals with

intellectual disabilities, developmental disabilities, and

behavior disorders. At present, JRC operates forty-four houses

in the Commonwealth, where clients live in a residential

setting. The judge below credited testimony that patients are

housed in a very humane environment and the staff is well

trained. As of March 2015, the total number of clients enrolled

at JRC was 244.

Many of the developmentally disabled patients at JRC suffer

from severely problematic behaviors, including aggressive,

curiae. See Mass. R. A. P. 17 (b), as appearing in 481 Mass.
1635 (2019).

5 At the time of its founding, JRC was known as Behavior
Research Institute, Inc. See note 1, supra. For the sake of
consistency, we shall refer to the facility as JRC throughout
this decision.
 7

destructive, and self-injurious behaviors. Indeed, a number of

patients have come to JRC after being expelled from other

facilities unable to address the severity of their behavioral

issues. For some of these patients, prior programs had resorted

to restraint or heavy sedation in an attempt to manage their

harmful behaviors. Because JRC generally does not turn patients

away, for numerous families, JRC was the only program willing to

accept their son or daughter.

Unlike other facilities, JRC has a policy of avoiding or

minimizing the use of psychotropic medication to treat its

patients. Instead, JRC has long relied upon applied behavior

analysis (ABA) to treat patients. This involves conducting a

"functional behavior assessment," i.e., studying the

relationship between problematic behaviors and the conditions

that precede them. JRC then uses positive reinforcement, e.g.,

rewards, to encourage desirable behaviors and negative

consequences, or "aversives," to discourage undesirable

behaviors. Typically, JRC relies in the first instance on

positive reinforcement procedures alone. However, if JRC's

positive programming fails to accomplish a patient's treatment

goals, or if it does not effectively treat the patient's

problematic behavior, the family is given the option of

including aversives as part of their son or daughter's treatment

plan. At the evidentiary hearing, several former JRC patients
 8

and parents credited JRC's aversive treatments with

significantly improving these patients' problematic behaviors.

At the time this litigation first began in 1986, JRC

employed a variety of physical aversives in a hierarchical

fashion with increasing levels of intrusiveness. See Natrona

County Sch. Dist. No. 1 v. McKnight, 764 P.2d 1039, 1045 n.4

(Wyo. 1988) (listing hierarchy of aversives utilized at JRC in

1985). In the years after the consent decree was entered, JRC

shifted away from this existing hierarchy of aversives in favor

of using the "Graduated Electronic Decelerator" (GED), a device

that administers a two-second electric shock to the surface of

the skin, usually on the arm or the leg. At present, the GED is

the primary physical aversive used at JRC. JRC utilizes two

versions of the device: the GED-3A and the GED-4. The former

delivers a current of 15.25 milliamps, and the latter delivers a

current of 41 milliamps.

JRC administers the GED to discourage specific problematic

behaviors. When a JRC staff member observes one of these

behaviors, a second staff member verifies that the behavior is

one for which use of the GED has been authorized pursuant to a

substituted judgment action, see note 6, infra, and the first

staff member then activates the GED. Normal application of the

device results in transient pain.
 9

Before a patient may be treated with the GED, a JRC

clinician must first develop a treatment plan. Each plan is

reviewed and approved by JRC clinicians, as well as a human

rights committee and a peer review committee. After JRC

develops the treatment plan, it must petition the Probate Court

for substituted judgment6 authorization to use the GED on that

patient. Once the Probate Court has approved the plan, JRC must

petition for reauthorization on an annual basis to continue

treating the patient with the GED. In the treatment plan

provided to the Probate Court, JRC is required to identify the

behaviors that it intends to target with the GED, and a JRC

clinician must aver that the GED is the least intrusive, most

effective treatment for the patient in question. The department

has the ability to oppose the use of the GED on a particular

patient at these yearly substituted judgment proceedings, but,

in practice, the department rarely does so. In one 2014

proceeding where the department did choose to participate, the

6 Substituted judgment proceedings are used as a "means by
which incompetents may exercise their right to refuse or
terminate treatment. . . . The judge, after hearing, must try
to identify the choice which would be made by the incompetent
person, if that person were competent, taking into account the
present and future incompetency of the individual as one of the
factors which would necessarily enter into the decision-making
process of the competent person" (quotation and citation
omitted). Guardianship of Doe, 411 Mass. 512, 518 (1992).
 10

Probate Court ultimately sided with the department and declined

to authorize the use of the GED on that patient.

As of 2014, thirty percent of JRC's patients had treatment

plans that included the use of court-authorized aversives. The

remaining seventy percent were treated using positive

programming alone. As of the close of evidence in 2016, few JRC

patients treated with the GED were minors. Counsel for the

defendants has since represented to this court that there are

currently no children receiving the GED as part of their

treatment plan.

b. History of current litigation. The procedural history

of this litigation began almost forty years ago, and the matter

last came before this court in 1997. See Judge Rotenberg Educ.

Ctr., Inc. v. Commissioner of the Dep't of Mental Retardation

(No. 1), 424 Mass. 430 (JRC I), S.C., 424 Mass. 471, 424 Mass.

473, and 424 Mass. 476 (1997). We need not repeat the entire

history of this case, much of which is covered in our prior

decision. See id. at 433-442. In short, this litigation began

after the Office for Children (OFC) issued a set of emergency

orders in 1985 requiring JRC to immediately cease the use of

physical aversive treatments and to halt the intake of new

patients.7 A judge in the Probate Court would later find that

7 Because JRC provided treatment to children with special
needs, in a full-time residential setting, it was at that time
 11

OFC had issued these orders "based upon no medical foundation,"

and that OFC attempted to hide this fact by retroactively

altering a report that had been "laudatory to [JRC] in all

substantial respects."

In response to the emergency orders, JRC and a class

consisting of all JRC patients and their parents and guardians

filed suit, alleging various constitutional and civil rights

violations. Thereafter, a judge in the Probate Court entered a

preliminary injunction enjoining OFC from enforcing its orders

and found that OFC had engaged in bad faith regulation of JRC.

The parties subsequently reached a settlement agreement, and on

January 7, 1987, the Probate Court approved and incorporated the

agreement as an order of the Probate Court (consent decree). As

part of the settlement agreement, licensing responsibility for

JRC was transferred from OFC to the Department of Mental Health;

later, that responsibility was transferred to the Department of

Developmental Services.8

required to obtain a license from the Office for Children (OFC).
See G. L. c. 28A, §§ 9, 11, as amended through St. 1981, c. 726,
§ 1. OFC is now known as the Department of Early Education and
Care (DEEC). See Commonwealth v. Power, 76 Mass. App. Ct. 398,
400 n.2 (2010). DEEC is the other named defendant in this
appeal.

8 At the time that the Department of Developmental Services
(department) became a party to the case, it was known as the
Department of Mental Retardation. See G. L. c. 19B, § 1, as
amended through St. 2008, c. 182, § 9.
 12

The consent decree contained a number of provisions, which

are discussed at length in JRC I, 424 Mass. at 433 n.5, 443-445,

448, and included a requirement that both parties act in good

faith. Another provision required JRC to obtain authorization

from the Probate Court, by way of substituted judgment

proceedings, before it could employ physical aversives in the

individual treatment plan of a client unable to give consent.

This was the only provision that was explicitly designed to

survive the termination of the consent decree. The decree

called for compliance reviews to occur at six-month intervals,

with the decree to terminate automatically after the second such

review "unless the [Probate] Court, for good cause shown related

to the terms or substance of [the settlement] agreement, orders

otherwise." The Probate Court subsequently issued an order on

July 7, 1988, extending the settlement agreement indefinitely.

No party objected to this extension.

Shortly after the consent decree was entered in 1987,

regulations were promulgated to govern the appropriate use of

physical aversives. The regulations classified aversive

interventions into one of three "levels," depending on severity.

See 104 Code Mass. Regs. § 20.15(3) (1987). Level three was

comprised of the most severe aversive treatments, including any

intervention that "involve[d] the contingent application of

physical contact aversive stimuli" or "pose[d] a significant
 13

risk of physical or psychological harm to the individual." See

104 Code Mass. Regs. § 20.15(3)(d). Any program seeking to use

level three aversives was required to apply for a special

certification from the department.9 See 104 Code Mass. Regs.

§ 20.15(4)(f) (1987). After conducting a review and inspection

of the program, the department would grant, grant with

conditions, or deny the program a level three certification for

a period not to exceed two years. See 104 Code Mass. Regs.

§ 20.15(4)(f)(7),(9).

The regulations further specified that a program would only

be eligible to receive a certification for the use of level

three aversives if, "prior to the effective date of this

regulation, . . . the program had been using one or more level

III interventions pursuant to a Behavior Modification plan for

one or more clients of the program." See 104 Code Mass. Regs.

§ 20.15(4)(f)(11). Additionally, and in accord with the consent

decree, level three interventions could not be used on a patient

unable to provide consent, absent authorization from the Probate

Court, obtained by way of a substituted judgment proceeding.

See 104 Code Mass. Regs. § 20.15(4)(e) (1987). Moreover, these

9 At the time the regulations were first promulgated, the
Department of Mental Health was the agency responsible for
certifying programs for use of level three aversives. See 104
Code Mass. Regs. §§ 2.02(2), 20.15(4)(f) (1987). This
responsibility was subsequently transferred to the Department of
Developmental Services.
 14

interventions could only be used "to address extraordinarily

difficult or dangerous behavioral problems that significantly

interfere with appropriate behavior and or the learning of

appropriate and useful skills and that have seriously harmed or

are likely to seriously harm the individual or others." See 104

Code Mass. Regs. § 20.15(4)(b)(5) (1987). JRC is the only

program in the Commonwealth certified to use level three

aversives.10

In 1993, six years after the consent decree was entered,

the department launched a campaign to "disrupt the operations of

JRC by every conceivable means," with the intent of putting JRC

out of business. See JRC I, 424 Mass. at 454. This included

"interfering with JRC's relationships with funding agencies and

JRC's fiscal operations," as well as imposing "a severe and

essentially constant burden on the JRC staff by having to

respond to an unrelenting stream of bad faith regulatory

10It should be noted, however, that there are references in
the record to other facilities that have, at various times, used
a level three aversive because they employed time-outs beyond
fifteen minutes. See 104 Code Mass. Regs. § 20.15(3)(d)(2)
(1987) (defining level three aversives to include time-outs in
excess of fifteen minutes). See also Judge Rotenberg Educ.
Ctr., Inc. v. Commissioner of the Dep't of Mental Retardation
(No. 1), 424 Mass. 430, 447 n.20 (JRC I), S.C., 424 Mass. 471,
424 Mass. 473, and 424 Mass. 476 (1997) (noting that department
had permitted use of aversive therapies on individual patients
at various facilities, even though department official conceded
that "there is no authority in the regulations for approval of
Level III procedures 'in the absence of a certification as a
program'").
 15

demands" made without justification. Id. at 456-457. At one

point, the department ordered JRC to discontinue level three

aversives for six patients, and later, it decertified JRC as a

provider of level three aversives altogether. The department's

actions led JRC to file a complaint alleging that the department

was in contempt of the consent decree.

After a thirteen-day trial, a judge in the Probate Court

found that the department had engaged in bad faith regulation of

JRC, held the department in contempt of the consent decree, and

placed it in receivership to oversee and manage its interactions

with JRC. This court affirmed the finding of contempt on

appeal, but narrowed the scope of the receivership. See id. at

463, 466-467. In 2003, the parties agreed to a winding down of

the receivership. By order of the Probate Court, the

receivership came to an end in 2006, thereby restoring the

department's regulatory oversight of JRC. The order did not,

however, terminate the consent decree.

c. Department's regulatory conduct postreceivership. In

August 2007, a former JRC resident called the facility,

impersonated a staff member over the telephone, and ordered JRC

employees to administer dozens of electric shocks to two

patients in the middle of the night. The caller also ordered

the employees to place a third patient on a four-point restraint
 16

board, despite the fact that this was not authorized by the

patient's treatment plan.

In the wake of the incident, JRC was investigated by the

department, along with a number of other entities, including the

Department of Social Services, the Department of Early Education

and Care, the Disabled Persons Protection Commission, and an

independent monitor. These investigations identified a number

of issues that contributed to this horrible incident, and the

department issued an action plan to JRC in early 2008 requiring

the facility to, inter alia, ensure that staff personally

witness the targeted behavior before using the GED, and minimize

the time between observing the behavior and administering the

GED.

In January 2008, the Secretary of the Executive Office of

Health and Human Services (EOHHS), JudyAnn Bigby, sent a

memorandum to Governor Deval Patrick with recommendations for

ways in which the administration could "change the State's

policy toward JRC without running afoul of the [consent

decree]." Bigby made clear that she was "personally outraged by

the continued practice of electric skin shock therapy" and

believed it to be outside the current standard of care. She

tasked EOHHS Assistant Secretary Jean McGuire with forming and

leading a clinical advisory group on the use of aversives. The

initiative resulted in a memorandum authored by one of its
 17

members, Dr. Charles Hamad (Hamad memo or memo), a psychologist

at University of Massachusetts Medical School (UMass Medical).

Upon receiving Hamad's draft of the memo, McGuire suggested

a number of edits, including the minimization of one expert's

opinion that she felt "looked like a rationale for keeping one

place [that uses electric skin shock] open in the country (which

would be the one we already have)." Hamad approved McGuire's

edits, which included a new sentence stating, "In brief, our

conclusion is that neither the professional literature nor the

practice arena supports the use of aversive contingent

interventions for behavior management of people with

intellectual or other disabilities that may involve serious

behavioral problems." The final version of the Hamad memo was

attached to a subsequent policy review memorandum that McGuire

drafted and sent to Bigby in December 2008, which listed various

political and regulatory options for changing the

administration's policy toward the use of electric skin shock.

One year later, Bigby sent a memorandum to the Governor

with an update on the status of JRC since the August 2007

incident. She noted that there had been "considerable

improvement in Executive agency collaboration and oversight of

JRC, which in turn has led to noteworthy progress in JRC's

performance." She also stated that the department's level three

"certification team has recently completed a monitoring review
 18

and found JRC to be in substantial compliance with previously

imposed conditions." Although Bigby reiterated her belief that

the use of aversives "does not reflect the community standard of

care or best practices," she concluded that, "at least for now,"

the administration should "continue the current close monitoring

and regulation of JRC and . . . not pursue any other options at

this time."

In May 2010, the Governor's chief legal counsel met with

representatives of advocacy groups opposed to aversive

treatments. McGuire subsequently informed the department of the

issues discussed at the meeting, including the advocates'

recommendation that the department "make every use of the

upcoming certification to assure that we are tough on /

responsive to those areas where he [sic] continues to be non-

compliant or has slipped."

The following month, the department's 2010 certification

team, headed by Dr. Philip Levendusky, completed its report on

JRC's most recent application for level three certification.

The team concluded that JRC was in "substantial compliance" with

prior conditions imposed by the department and recommended that

JRC be given a one-year recertification to use level three

aversives with certain conditions of compliance. However, in a

subsequent series of communications between Levendusky, the

department's general counsel, and the department's commissioner,
 19

the 2010 report was further revised without consulting the

remaining team members. In the course of these revisions, the

department's general counsel removed the "substantial

compliance" language from the report as well as the

recommendation that JRC be issued a one-year certification. The

final, revised version of the report instead extended JRC's

existing certification by fourteen working days. Within that

time, JRC was required to submit a corrective action plan for

obtaining compliance with all of the report's conditions, and to

submit additional progress reports at forty-five-day intervals.

Following the issuance of this report, and in the course of

complying with the deadlines contained therein, JRC exchanged a

number of reports and correspondences with the department in

which JRC challenged the department's ability to impose certain

conditions, and the department rejected various of JRC's

assertions of compliance. Eventually, in the summer of 2011,

JRC and the department agreed to mediation before the former

receiver. The parties ultimately reached an agreement in July

2012 resolving their dispute.

While the mediation was still ongoing, the department

amended its behavior modification regulations to prohibit the

use of level three aversives, except for "individuals who, as of

September 1, 2011, [had] an existing court-approved treatment

plan" authorizing their use. See 115 Code Mass. Regs.
 20

§ 5.14(4)(b)(4) (2011). In effect, these regulations imposed a

prospective ban on the use of level three aversives for new JRC

patients. After the regulations went into effect, the

department convened a group of experts to serve on an advisory

subcommittee charged with promulgating new guidelines for the

Statewide implementation of the department's favored treatment

approach, Positive Behavior Supports (PBS).11 In advance of the

subcommittee's discussions of the ABA literature concerning the

efficacy and acceptability of using specific procedures to

decelerate problematic behaviors, a representative of the

department informed the subcommittee co-chair that "it won't

matter whats [sic] in the literature if [the commissioner] does

not like it." Some members of the subcommittee nonetheless went

on to voice support for the use of level three aversives in

certain limited circumstances, and the group expressed

discomfort with draft guidelines that would ban specific

interventions. The department subsequently instructed the

11Positive Behavior Supports (PBS) focuses on the
conditions that precede problem behaviors and the environmental
changes that can be made to improve a client's quality of life.
The judge below found that PBS was more accurately described as
a philosophy or general approach to treatment, rather than a
subdiscipline within the field of psychology.
 21

subcommittee not to address the issue of level three aversives

as part of their work.12

d. Procedural history leading to instant appeal. In 2013,

while the advisory subcommittee's discussions remained ongoing,

the defendants filed a motion in the Probate Court to terminate13

the consent decree, pursuant to Rule 60 of the Rules of the

Probate Court (2013) and Mass. R. Civ. P. 60 (b) (5), 365 Mass.

828 (1974). The department argued that termination of the

decree was warranted because the department had long since

abandoned its history of bad faith regulation, and because the

12The department went on to amend its regulations that year
to remove certain level two aversives, including procedures
requiring significant physical exercise, unpleasant sensory
stimuli like loud noises or bad tastes, and meal delays. See
115 Code Mass. Regs. § 5.14(3)(c)(1) (2013). However, it was
not until 2020 that new regulations went into effect replacing
the existing regulations governing behavior modification with a
PBS framework. See 115 Code Mass. Regs. §§ 5.14, 5.14A (2020).

13The parties have characterized the defendants' pleading
as a motion to "vacate" the consent decree. However, the
defendants' motion "did not challenge the grounds on which [the
consent decree] was earlier entered," but "sought only to
prevent its prospective application." MacDonald v. Caruso, 467
Mass. 382, 384 n.4 (2014). Accordingly, the motion is most
appropriately understood as a request to terminate, rather than
vacate, the decree. See id. See also Inmates of Suffolk County
Jail v. Rouse, 129 F.3d 649, 662 (1st Cir. 1997), cert. denied,
524 U.S. 951 (1998) ("While terminating a consent decree strips
it of future potency, the decree's past puissance is preserved
and certain of its collateral effects may endure. Vacating a
consent decree, however, wipes the slate clean, not only
rendering the decree sterile for future purposes, but also
eviscerating any collateral effects and, indeed, casting a
shadow on past actions taken under the decree's imprimatur").
 22

primary physical aversive used by the facility, electric skin

shock, was outside the professional standard of care.

A judge in the Probate Court held an evidentiary hearing on

the motion, which took place over the course of forty-four days

between October 2015 and October 2016. On June 20, 2018, the

judge issued a written memorandum of decision denying the

motion. The judge found that the department had engaged in bad

faith regulation of JRC in 2010, just as it had in prior

decades. On this basis, the judge concluded that, as of 2018,

the consent decree remained necessary to protect JRC from bad

faith conduct such as had occurred eight years prior. The

judge's 2018 decision also concluded that the department had

failed to demonstrate a significant change in circumstances that

would warrant termination of the consent decree. In reaching

this conclusion, the judge found that the department had failed

to show that, as of the close of evidence in 2016, there was a

professional consensus that level three aversives were outside

the standard of care. The defendants timely filed a notice of

appeal, and the case was entered in the Appeals Court in August
 23

2021.14 Thereafter, this court granted the parties' joint

application for direct appellate review.15

2. Standard of review. Rule 60 (b) (5) permits the court

to grant relief from a judgment with prospective effect where

"it is no longer equitable" for the judgment to remain in place.

This requires the moving party to demonstrate a significant

change in circumstances since the entry of the judgment that

would warrant its modification or termination. See MacDonald v.

Caruso, 467 Mass. 382, 388 (2014), and sources cited. This

standard is a flexible one, and its application depends upon the

individual facts of the case and the nature of the judgment at

issue. See Rufo v. Inmates of Suffolk County Jail, 502 U.S.

14The three-year delay between the department's notice of
appeal and the entry of this case in the Appeals Court appears
to have been due to the size of the record, compounded by delays
in receiving searchable copies of the electronic transcript
files, as well as lapses in communication between the clerk's
office of the Probate and Family Court (Probate Court), the
parties, and the stenographers.

15JRC has argued that DEEC's appeal should be dismissed
because DEEC failed to file a brief after the case had been
entered in this court. DEEC, which joined in the notice of
appeal and docketing statement, has since moved to join the
department's appellate brief. DEEC has argued that it failed
join the department's brief at the time of filing because
counsel for the defendants "mistaken[ly]" believed that DEEC had
no further obligations under the decree, given that no children
enrolled at JRC are approved for use of level three aversives.
JRC has failed to articulate any prejudice that would stem from
allowing DEEC's motion to join the department's briefing.
Accordingly, we allow DEEC's motion to join the department's
brief and decline JRC's invitation to dismiss DEEC's appeal.
 24

367, 380-381 (1992). See also Alexis Lichine & Cie. v. Sacha A.

Lichine Estate Selections, Ltd, 45 F.3d 582, 586 (1st Cir.

1995). Thus, consent decrees that implicate "the supervision of

changing conduct or conditions," which "are thus provisional and

tentative," are more likely to warrant modification than consent

decrees that "give protection to rights fully accrued upon facts

so nearly permanent as to be substantially impervious to

change." Rufo, supra at 379, quoting from Justice Cardozo's

often-cited articulation of the standard in United States v.

Swift & Co., 286 U.S. 106, 114-115 (1932).

The decision whether to grant relief from judgment under

rule 60 (b) rests within the sound discretion of the trial

judge. See Atlanticare Med. Ctr. v. Division of Med.

Assistance, 485 Mass. 233, 247 (2020) (Atlanticare).

"Accordingly, the denial of a motion under Rule 60 (b) will be

set aside only on a clear showing of an abuse of discretion"

(quotation and citation omitted). Id. In effect, this means

that the decision will be affirmed unless the judge below "made

a clear error of judgment in weighing the factors relevant to

the decision . . . such that the decision falls outside the

range of reasonable alternatives" (citation omitted). Dacey v.

Burgess, 491 Mass. 311, 317 (2023). Here, the department's

argument that the judge abused her discretion is focused

primarily upon contesting two factual findings derived from the
 25

evidentiary record before the Probate Court in 2016: (1) the

department's continued bad faith regulatory conduct toward JRC;

and (2) the absence of a professional consensus whether level

three aversives fall outside the accepted standard of care.

"To prevail on appeal on the basis of an assault on a

judge's factual findings is no easy matter, for we accept the

judge's findings of fact as true unless they are 'clearly

erroneous'" (citation omitted). Millennium Equity Holdings, LLC

v. Mahlowitz, 456 Mass. 627, 636 (2010). Under this "clearly

erroneous" standard, "the judge's findings come here well armed

with the buckler and shield" (alteration, quotation, and

citation omitted). JRC I, 424 Mass. at 452. That is, any

finding based partly or wholly on oral testimony will be upheld,

unless there is no evidence to support it or the reviewing court

"is left with the definite and firm conviction that a mistake

has been committed" (citation omitted). Kendall v. Selvaggio,

413 Mass. 619, 620-621 (1992). See Demoulas v. Demoulas Super

Mkts., Inc., 424 Mass. 501, 510 (1997) ("So long as the judge's

account is plausible in light of the entire record, an appellate

court should decline to reverse it"). It is not enough that

other evidence exists to support a different finding, or even

that this court might have weighed the evidence differently in

the first instance. See Brandao v. DoCanto, 80 Mass. App. Ct.

151, 154 (2011).
 26

3. Timeliness of department's motion. JRC argues, as a

threshold matter, that we need not reach the merits of this

appeal because the defendants' motion to terminate the consent

decree was untimely. Although the judge below did not deny the

motion on this basis, she observed, in accord with the

plaintiffs' argument, that the motion had been filed "long after

the existence of both reasons that Defendants proffer" as

necessitating termination of the consent decree.

Motions under rule 60 (b) (5) must be filed "within a

reasonable time," determined in light of all the circumstances

of the case. Atlanticare, 485 Mass. at 247-248, quoting Mass.

R. Civ. P. 60 (b). In making this determination, "a judge may

consider the reasons for delay; the ability of the movant to

learn of the grounds earlier; prejudice to the parties, if any;

and the important interest of finality" (citation omitted).

Atlanticare, supra at 248. Where, as here, the judgment at

issue binds public officials, the court also considers the

governmental and public interests at stake. See id. See also

Shakman v. Chicago, 426 F.3d 925, 934 (7th Cir. 2005) (under

Federal analog, "any consideration of a 'reasonable time' for

filing a [Fed. R. Civ. P.] 60(b) motion with respect to the

. . . Consent Decree must take into account the nature of that

litigation as well as the resulting prejudice, if any, to the

present elected officials and the public they represent"). At
 27

bottom, however, "[t]here is no set formula" for determining

reasonableness in this context. Atlanticare, supra. Compare

id. at 247-249 (seven-year delay did not render motion untimely

in "highly unusual circumstances" of case, including conflicting

decisions between United States Court of Appeals for First

Circuit and this court that would otherwise "lead to confusion

and administrative deadlock"), with Owens v. Mukendi, 448 Mass.

66, 76-77 (2006) (listing cases where delays of two or three

years rendered motion untimely).

Applying these principles, we find that the department's

motion to terminate the decree was timely. The governmental

interests are significant, as denial on the grounds of

untimeliness "would effectively 'bind all future [regulatory

officials]' . . . to the decree's proscriptions," solely because

their predecessors failed to bring the motion at the earliest

available opportunity. Doe v. Briley, 562 F.3d 777, 781 (6th

Cir. 2009), quoting Rufo, 502 U.S. at 392. Additionally, the

prejudice to the plaintiffs is comparatively limited. Indeed,

any delay inures to the advantage of JRC. As long as the decree

remains undisturbed, JRC continues to benefit from the decree's

limitation on the regulatory authority that the department may

exercise over the facility. Cf. Doe, supra (rejecting argument

that motion to terminate decades-old consent decree was untimely

where, inter alia, "the only apparent consequence of the delay,
 28

so far as [the nonmovant was] concerned, [was] that the decree

remained in place for some [thirty] years longer than it

probably should have").

Further, while we recognize that some of the grounds for

relief raised in the motion date back to the 1990s, the

department's primary arguments -- the department's record of

good faith compliance and a new medical consensus -- concern

gradual developments. Moreover, given that the department

sought to argue that it had a long-standing record of acting in

good faith, any delay in raising the argument was a reasonable

response to the decree itself; the delay allowed the department

time to demonstrate that it had learned from its mistakes and

had made the necessary institutional reforms. Cf. Associated

Bldrs. & Contrs. v. Michigan Dep't of Labor & Economic Growth,

543 F.3d 275, 279 (6th Cir. 2008), cert. denied, 556 U.S. 1127

(2009) ("An unduly strict reading of the reasonable-time

requirement, moreover, would tend to force premature [Fed. R.

Civ. P.] 60(b)(5) motions due to a State's fear of losing

forever the opportunity to correct an injunction or consent

decree"). Accordingly, we turn to the merits of the defendants'

motion.

4. Satisfaction of purpose underlying consent decree.

Changed circumstances exist to warrant termination of a consent

decree, as opposed to its mere modification, where the moving
 29

party demonstrates that the purpose of the decree has been

achieved. See 12 Moore's Federal Practice § 60.47[2][c], at 60-

178 & n.22 (3d ed. 2023). Although we have not provided

specific guidance on how to determine whether the purpose of a

consent decree binding public officials has been satisfied,

Federal courts have looked to whether the State has demonstrated

that it is currently in "substantial, good-faith compliance"

with the fundamental purpose of the consent decree and "unlikely

. . . [to] return to its former ways." Peery v. Miami, 977 F.3d

1061, 1075 (11th Cir. 2020), quoting Board of Educ. of Okla.

City Pub. Sch. v. Dowell, 498 U.S. 237, 247 (1991). See Shakman

v. Pritzker, 43 F.4th 723, 728 (7th Cir. 2022) ("A party

claiming to have satisfied the terms of a consent decree must

show that it has achieved the objectives of that decree . . .

and implemented a durable remedy"). To assess whether

termination is warranted on that basis, we look first to the

underlying purpose of the decree.

a. Purpose of consent decree. When this case was last

before this court on appeal, we explained the context and over-

all function of the consent decree as follows:

"The action that resulted in the settlement agreement was
brought because the parents and guardians of JRC patients
alleged that OFC was denying individual patients their
constitutional rights to certain treatments and was not
regulating JRC in good faith. The settlement agreement
sought to remedy this situation while allowing the
 30

department to continue to fulfil its statutory duties to
regulate mental health facilities."

JRC I, 424 Mass. at 450. In describing the decree, we did not

go so far as to state that its purpose was to guarantee the

right of access to aversives, as the plaintiffs' claims to that

effect were never actually adjudicated and determined by the

court. That being said, the terms of the consent decree, along

with the underlying proceedings, do reveal two main purposes.

First, the consent decree was intended to ensure that the

department's predecessor in interest, OFC (and later, the

department itself), would regulate JRC in good faith and avoid

engaging in unauthorized, "unilateral interference" with

individual treatment plans. Id. at 445-447. Second, the

consent decree was intended to permit JRC to continue using

aversives on individual patients, but only subject to judicial

supervision, by way of substituted judgment proceedings. See

id. at 444. The decree contemplated that the department would

be allowed to participate in these proceedings, and that JRC

would only receive authorization where the proposed treatment

was the least intrusive and most appropriate to the client's

needs. See id. at 444 n.15. The consent decree otherwise

preserved the department's regulatory authority.16 See id. at

16We recognize that the consent decree contained a
provision calling for a court monitor to evaluate JRC's
compliance with department regulations that did not concern
 31

445 ("Indeed, there is no provision in the agreement that

provides the department gave up any regulatory authority").

Whether these purposes have been fulfilled remains hotly

disputed. We address each one in turn.17

b. Findings of bad faith regulation. Here, the judge

below found that the purpose of the decree had not been

fulfilled because the department engaged in bad faith regulation

level three aversives. We previously declined to address the
permissibility of this provision, stating:

"We do not consider whether the portion of the agreement
providing that it was the court monitor, not the
department, that was to oversee compliance with all other
applicable State regulations except those related to Level
III aversives and undertake general monitoring of JRC's
treatment and educational program constituted an
impermissible delegation of regulatory authority. The
findings of the judge with respect to this portion of the
settlement agreement are not necessary for our decision
here; we note, moreover, that neither side disputes that
JRC was required to be certified according to the
department's regulations, and it is that certification
process and its relationship to the settlement agreement
that is before us."

JRC I, 424 Mass. at 445 n.19. Regardless, the winding down and
eventual termination of the receivership resulted in these other
regulatory functions being returned to the department.

17 Because the second purpose -- concerning JRC's ability to
use aversives pursuant to court authorization -- implicates the
interplay between the consent decree and the department's
residual regulatory authority, we address it as part of our
discussion of the separation of powers argument raised by the
defendants.
 32

in 2010.18 In support of this finding, the judge relied upon

three subsidiary findings: (1) the alterations made by McGuire

to the 2008 Hamad memo concerning the use of aversives; (2) the

alterations made by the department's commissioner and general

counsel to the 2010 certification report; and (3) the

department's decision to accept certain conditions proposed by

the 2010 certification team concerning the acceptable use of the

GED that the judge concluded were "impermissibl[e] . . .

treatment decisions."

On appeal, the department asserts that the Hamad memo did

not affect the department's regulation of JRC because the

department was not involved in the creation of the memo and did

not rely on it in any way. The department further argues that

the changes made to the 2010 certification report were largely

nonsubstantive, and that the alterations made were "reasonable

exercises of the [c]ommissioner's ultimate authority to approve,

approve with conditions, or disapprove a Level III program,"

citing 115 Code. Mass. Regs. § 5.14(4)(f)(7) (2011). Finally,

the department claims that the conditions in the 2010

certification report concerning the acceptable use of the GED

18We note that this bad faith finding is based on conduct
that occurred thirteen years ago and an evidentiary hearing that
concluded in 2016. We stress again that our analysis does not
foreclose the possibility that new developments have occurred
since the record closed here bearing on these factual issues.
 33

were properly within the purview of the department's regulatory

authority. We conclude that the department's alterations to the

2010 certification report, particularly the removal of the

team's substantial compliance finding and the dramatic reduction

in certification length, support the judge's finding of bad

faith. This finding of bad faith is further supported by the

department's unilateral decision, without first assessing the

scientific evidence, to impose a regulatory change that would

prohibit JRC from using level three aversives on new patients.

"Bad faith is a 'general and somewhat indefinite term' that

goes beyond 'bad judgment' or 'negligence,' suggesting 'a

dishonest purpose or some moral obliquity,' a 'conscious doing

of wrong,' or a 'breach of a known duty through some motive of

interest or ill will'" (citation omitted). Buffalo-Water 1, LLC

v. Fidelity Real Estate Co., 481 Mass. 13, 25-26 (2018). See

JRC I, 424 Mass. at 454. In the context of State action, this

includes the use of an otherwise lawful power for an improper

purpose. See Pheasant Ridge Assocs. Ltd. Partnership v.

Burlington, 399 Mass. 771, 776 (1987). In effect, bad faith

requires an inquiry into the subjective intent behind a party's

actions, in addition to the actions themselves. See Bank of

Am., N.A. v. Prestige Imports, Inc., 75 Mass. App. Ct. 741, 754-

755 (2009), and cases cited (discussing "foundational
 34

definition" of bad faith, which involves "subjective focus" on

"knowing and conscious wrongdoing").

i. Hamad memo. The judge below found that the Hamad memo,

in its final form, "was shaped significantly by [EOHHS]

Assistant Secretary McGuire herself and did not represent an

independent, objective review." The judge then cited the Hamad

memo in her discussion of bad faith, describing the document as

the primary source for a memorandum from McGuire that was

"intended to create a justification for [the department] to

pursue a path that would eventually end with the elimination of

contingent aversive treatment at JRC." Although we do not

discount the Hamad memo, we do not consider it as significant as

the judge for the reasons discussed infra. It does, however,

provide further, albeit limited, support for the more compelling

evidence of bad faith relating to the department's manipulation

of the 2010 certification report.

We recognize, as did the judge below, that McGuire made

numerous alterations in the Hamad memo. This included

downplaying one expert's opinion that "contingent electric shock

might conceivably be needed . . . for a very, very small number

of exceptional cases where the individual's behavior was so

extreme as to be life threatening" and adding a statement that

"neither the professional literature nor the practice arena

supports the use of aversive contingent interventions for
 35

behavior management of people with intellectual or other

disabilities that may involve serious behavioral problems."

However, McGuire's revisions are largely in accord with the

thrust of Hamad's original draft. The original memo contained a

list of policy recommendations for consideration by EOHHS,

including a recommendation to file legislation banning

aversives, which the original memo described as "reflect[ing] a

consensus view reached after completion of various review

activities conducted under your direction over that [sic] last

[six] months." Moreover, both versions of the memo effectively

contain the same conclusion that "alternatives to contingent

aversive techniques are not only the preferred methods to treat

extreme behavior disorders but have clearly become the practice

standard in the field of developmental disabilities."

It is nonetheless apparent from the record that the Hamad

memo did not provide an independent, objective review of

aversives. Notably, Hamad did not seek to interview either of

the independent psychologists who evaluate and prepare reports

on the patients for whom JRC seeks use of the GED. Nor did

Hamad follow up on information he received about clinicians at

Johns Hopkins University and the University of Florida, as well

as psychologists in Boston, who supported considering aversives

where alternative treatments had failed. These shortcomings

support the judge's finding that the Hamad memo did not
 36

constitute an independent, objective review of the standard of

care.

However, as the department emphasizes, the record does not

contain evidence to support the finding that the Hamad memo

played a role in the department's subsequent regulatory actions

toward JRC. Although the advisory group in which Hamad

participated included several department clinicians, it does not

appear that department officials were involved in the drafting

of the Hamad memo, let alone McGuire's subsequent revisions.

Nor was any evidence presented indicating that the department

was influenced by, or even aware of, the contents of the Hamad

memo at the time of the 2010 certification process. That said,

the memo's origin, revisions, and methodology suggest a result-

oriented approach that lends some contextual support for the

more significant basis for the judge's finding of bad faith:

the department's revisions to the 2010 certification team

report.

ii. Revisions to 2010 certification team report. Firmer

support for the judge's finding of bad faith can be found in the

revisions to the 2010 certification team report. The judge

found that "many parts of the final report . . . were entirely

rewritten" by the department's general counsel and the

commissioner, including "significant substantive changes" made

without the approval or knowledge of team members other than
 37

Levendusky. The judge further concluded that even though

Levendusky approved the changes, he was not the "driving force"

behind them, and that the involvement and influence of the

commissioner was akin to the bad faith regulation of the 1980s

and 1990s.

Although we agree with the judge's over-all conclusion that

these changes support a finding of bad faith, we do not agree

that "many" parts of the report were "entirely rewritten."19 The

important substantive changes to the report, which totaled more

than thirty pages, consisted of (1) the deletion of the

"substantial compliance" language and the team's recommendation

for a one-year recertification; (2) the revised recommendation

to extend JRC's existing certification by only fourteen days;

and (3) the addition of burdensome documentation requirements,

with short turn-around times, contained within the summary of

conditions.20

It is readily apparent that these three changes were both

significant and improper. The impetus for removing the

19While language was removed from the "Safety Review of GED
and GED-4 Device" section, as well as the "Peer Review" section,
and small revisions were made to the "Level II Interventions in
Use" section, these edits appear to be more stylistic than
substantive.

20We also note the alteration of condition (2) (g) from
requiring JRC to engage a "multidisciplinary" team to instead
requiring an "external" one comprised of at least three
clinicians with ABA expertise.
 38

"substantial compliance" language and the one-year certification

recommendation both originated from the commissioner and her

general counsel, not Levendusky. While Levendusky was the first

to suggest adding deadlines, he was not considering changing the

one-year certification recommendation at the time the report was

drafted, and he did, in fact, find JRC to be in substantial

compliance with prior certification conditions. Further, the

removal of the "substantial compliance" language was far from

mere semantics. The commissioner admitted at the hearing that

under the department's own policies, a finding of substantial

compliance would have resulted in a one-year certification

recommendation, and thus, removal of that language was necessary

to justify the department's decision to grant a shorter

certification length. Accordingly, the commissioner's decision

to extend JRC's existing certification by only fourteen days was

improper under the department's own policies, given the 2010

certification team's actual finding of substantial compliance.

And by limiting JRC's certification extension to only fourteen

days, the department put JRC under significant undue and

unjustified pressure, placing all of its patients' aversive

treatment plans in jeopardy. Further compounding this pressure

was the additional requirement that JRC provide substantial

documentation reflecting compliance within relatively tight

deadlines.
 39

Considering these improper revisions within their

surrounding context, the judge's finding of bad faith in 2010

was not clearly erroneous. Prior to the events in question,

Bigby had sent a memorandum to the Governor indicating that the

certification team had "recently completed a monitoring review

and found JRC to be in substantial compliance with previously

imposed conditions," noting that "JRC staff [had] been very

cooperative and improvement in the program [was] evident" and

that "[b]y all accounts, the situation at JRC [was] as good as

it [had] ever been." Things appear to have changed when, four

months later, the Governor's chief legal counsel met with

disability advocates who recommended "mak[ing] every use of the

upcoming certification to assure that [the administration is]

tough on / responsive to those areas where [JRC] continues to be

non-compliant or has slipped." McGuire relayed this message to

the department's commissioner and general counsel, indicating

that the Governor's chief legal counsel would expect "an update

on this certification process, once the team's work is done but

before we issue the decision." McGuire would later remark in an

e-mail message that she also told the commissioner that McGuire

"did not think [the administration] would support another six

month certification." And when the department finally sent the

revised certification report to EOHHS, the department's general

counsel made a point of highlighting to McGuire that, with the
 40

limited fourteen-day extension, JRC's certification "could be

pulled at day [fourteen] or day [forty-five] if [JRC's] response

isn't sufficient." The department's general counsel did so

despite acknowledging that, for some of the report's findings of

noncompliance, "these are really professional judgment issues."

On the whole, this evidence supports the judge's inference

that the removal of the substantial compliance language, the

dramatic reduction in certification length from one year to

fourteen days, and the imposition of burdensome and time-

sensitive follow-up requirements did not amount to a good faith

assessment of JRC's regulatory compliance, but an attempt to

appease advocates opposed to JRC and maximize the

administration's ability to justify a revocation of JRC's

certification. See Lynch v. Crawford, 483 Mass. 631, 644

(2019), quoting Commonwealth v. Casale, 381 Mass. 167, 173

(1980) ("intent is a matter of fact, which is often not

susceptible of proof by direct evidence, so resort is frequently

made to proof by inference from all the facts and circumstances

developed at the trial"). This improper motive supports a

finding of bad faith.21

21In light of our conclusion that the department's
revisions to the 2010 certification report supported the judge's
finding of bad faith, we need not address the third basis for
the judge's finding of bad faith -- namely, her determination
that "by accepting certain recommendations of the 2010 Level III
Certification Team, [the department] impermissibly made
 41

c. Whether purpose of consent decree was satisfied given

passage of time. We next consider the judge's holding that this

bad faith conduct demonstrated that the purpose of the consent

decree had not been fulfilled as of 2018, and whether that

ruling was an abuse of discretion given the passage of time.

For the reasons discussed infra, we conclude that it was not.

In reaching this conclusion, we recognize that the basis for the

judge's finding of continued bad faith relies heavily on conduct

that occurred in 2010, eight years prior to the denial of the

motion in 2018. Further, it is apparent that after the parties

mediated their dispute concerning the 2010 recertification

process, the department went on to issue a new one-year

certification, with conditions, to JRC in 2013. As of the close

of evidence in this case, JRC's 2014 application for

recertification was still outstanding, but as far as we are

aware, there have been no additional allegations of bad faith by

the department in the course of performing its regulatory

oversight duties between 2010 and the commencement of the

hearing in the instant case. Moreover, it has now been over ten

years since the department's motion to terminate was filed.

treatment decisions for JRC clients." We further note that the
complex interplay between the consent decree and the
department's residual regulatory authority is an issue we
address separately in our discussion of the department's
separation of powers argument.
 42

To be sure, the passage of time, combined with the turnover

of administrations and leadership in an agency, as well as the

cessation of bad faith regulatory misconduct, can provide

support for the eventual termination of a consent decree that

binds public officials. See Frew v. Hawkins, 540 U.S. 431, 441-

442 (2004); Inmates of Suffolk County Jail v. Rouse, 129 F.3d

649, 656-657 (1st Cir. 1997), cert. denied, 524 U.S. 951 (1998).

The Legislature delegates power to an executive agency to make

and enforce rules in accordance with that agency's expertise in

light of changing conditions. See Borden, Inc. v. Commissioner

of Pub. Health, 388 Mass. 707, 723-724, cert. denied sub nom.

Formaldehyde Inst., Inc. v. Frechette, 464 U.S. 936 (1983);

Mostyn v. Department of Envtl. Protection, 83 Mass. App. Ct.

788, 797 (2013). Consent decrees enmesh the judiciary in

ongoing oversight of such policy-making decisions, and may serve

to "improperly deprive future officials of their designated

legislative and executive powers." Frew, supra at 441. These

risks are compounded for decrees that last decades, requiring

ongoing judicial supervision over subsequent actors who are far

removed from the original actors' bad faith misconduct. See

Rufo, 502 U.S. at 392 ("To refuse modification of a decree is to

bind all future officers of the State, regardless of their view

of the necessity of relief from one or more provisions of a

decree that might not have been entered had the matter been
 43

litigated to its conclusion"). Thus, to the extent that a

consent decree is based on agency misconduct, evidence

establishing that the improper conduct of the past has been

abandoned, and that the agency has been acting in good faith,

would support termination of the consent decree. See Peery, 977

F.3d at 1075. Contrast MacDonald, 467 Mass. at 388-389 (where

court order at issue binds private parties, neither passage of

time nor movant's ongoing compliance are normally sufficient,

without more, to justify termination).

Here, however, the issue of bad faith regulation as of the

judge's ruling in 2018 arises not only from the manipulation of

documents in 2010 or expert opinion in 2008, but also from the

department's continued insistence on using the regulatory

process to achieve a predetermined outcome regarding level three

aversives -- namely, to eliminate a treatment protocol that the

Legislature has repeatedly declined to ban,22 that judges in the

Probate Court have regularly authorized through substituted

22Although there has been no shortage of legislative
proposals to ban aversive treatments, none has passed. See,
e.g., 2023 House Doc. No. 180; 2021 House Doc. No. 225; 2019
House Doc. No. 123; 2017 House Doc. No. 93; 2015 House Doc.
No. 89; 2015 Senate Doc. No. 80; 2013 House Doc. No. 106; 2013
Senate Doc. No. 30; 2011 Senate Doc. No. 51; 2011 House Doc.
No. 77; 2009 House Doc. No. 154. Other proposals to restrict or
study aversive treatments have similarly failed. See 2023 House
Doc. No. 170; 2022 House Doc. No. 4956; 2015 Senate Doc. No. 79;
2013 Senate Doc. No. 28; 2011 Senate Doc. No. 49; 2009 House
Doc. No. 183; 2009 Senate Doc. No. 45.
 44

judgment, and that the department itself had agreed to permit

when it chose to bind itself to the consent decree -- without an

objective consideration of the evidence concerning the use of

the aversives, and without adhering to the legal requirements

imposed upon the department by the courts. In the instant case,

we conclude that the judge could reasonably find that the

consent decree remained necessary in 2018 to prevent bad faith

regulation because the regulations promulgated by the department

in 2011 again demonstrated its intention to reach this

predetermined outcome without first objectively evaluating the

medical evidence or moving to terminate the consent decree.

The record indicates that, in 2010, after recent

legislative efforts to ban electric skin shock had failed, Bigby

sent a memorandum to the Governor with other policy options to

restrict or eliminate electric skin shock. At that time, she

cautioned that a regulatory ban could be construed as bad faith

regulation, and recommended tabling any policy proposals until

the Attorney General completed a criminal investigation into the

August 2007 incident. After the completion of that

investigation, Bigby authored a memorandum in April 2011 with

EOHHS's "recommendations for next steps in our regulatory

relationship with JRC." The first recommendation was to move

for termination of the consent decree. The second

recommendation, made "alternatively, or concurrently" to the
 45

first, was for the department to promulgate regulations to

prospectively ban level three aversives.

Two months later, the department proposed the 2011

regulations. These regulations were proposed only a year after

the 2010 certification report, and while the dispute about JRC's

compliance with the conditions contained in that report remained

ongoing. From the record, it also appears that there was no

effort by the department to undertake an independent objective

review of level three aversives prior to the passage of these

regulations. The department apparently did not convene experts

who considered the issue until after the regulations had already

gone into effect.

Most importantly, by choosing to pass the 2011 regulations

before moving to terminate the consent decree -- which, as

explained infra, the department was required to do -- the

department effectively sought to use its regulatory power as an

"end run" around the consent decree. In so doing, the

department again demonstrated that it was determined to alter

its policy toward aversives, regardless of the existence of the

consent decree and the legal constraints contained therein. It

was only later, nearly one and one-half years after those

regulations were promulgated, that the department chose to come

before the Probate Court to seek termination of the consent

decree. All of this supports the judge's conclusion that the
 46

consent decree remained necessary in 2018, despite the passage

of time, to preclude bad faith regulation by the department. We

therefore turn to the issue of changed circumstances of fact and

the judge's finding that no such change had occurred to warrant

termination of the decree.

5. Existence of changed circumstances of fact. In

entering the consent decree in 1987, the Probate Court found

that JRC's use of physical aversives was safe, effective, and

professionally acceptable. At that time, the Probate Court also

referenced earlier findings in which the Probate Court had

determined that JRC's use of physical aversives was "consistent

with professional practice" and was employed "in lieu of

antipsychotic medication and other more restrictive procedures,

such as seclusion and painful electric-shock." The department

contends that this is no longer the case, both because JRC now

employs electric skin shock and because the use of electric skin

shock is not within the professional standard of care. The

department also asserts that, regardless of whether electric

skin shock falls within the general standard of care, its

practical implementation at JRC does not. We address each

contention in turn.

a. Invention of GED. Although the judge did not

explicitly address whether the invention of the GED constituted

a change in circumstances, her failure to do so was not an abuse
 47

of discretion. While the consent decree predates the use of

electric skin shock treatment at JRC, the decree concerns the

use of "all aversive procedures which are presently used or

which may be proposed for use at [JRC]," apart from exceptions

not relevant here. The consent decree also explicitly states

that "[n]othing in this agreement shall preclude [JRC] from

developing new . . . aversive procedures." Given that the

consent decree patently contemplated the development of new

aversives, the fact that the GED was not in use at that time is

clearly insufficient, without more, to warrant termination of

the decree. See Rufo, 502 U.S. at 385 ("modification should not

be granted where a party relies upon events that actually were

anticipated at the time it entered into a decree").

In reaching this conclusion, we recognize that the findings

of fact in support of the consent decree referenced earlier

findings in which the Probate Court had determined that JRC's

aversive techniques were less restrictive than "painful

electric-shock." That finding was derived from uncontroverted

testimony offered at the preliminary injunction hearing in 1986,

wherein one of JRC's expert witnesses testified about

"contingent electroshock." Despite JRC's assertion to the

contrary, this does appear to be a reference to electric skin

shock akin to the GED. At the 1986 hearing, the expert

described "electroshock" as consisting of shocks that "would be
 48

administered for a very, very brief period[,] sometimes, merely

seconds," and explicitly distinguished it from electroconvulsive

therapy.

Nonetheless, this does not alter our analysis. The same

expert -- whose testimony was credited by the Probate Court in

1986 -- stated that contingent electric shock remained "less

aversive than . . . large dosages of drugs, [or] . . .

electroconvulsive shock therapy." The expert further offered

that he would consider using contingent electroshock if a

patient was "likely going to kill [him- or herself]" and nothing

else had "proved to be effective." Another expert, quoting from

professional literature, offered testimony at one of the six-

month review hearings in 1987 that "very intense punishment such

as shock . . . should be considered for immediate inclusion in

treatment" where there is "imminent and extreme physical danger

or when the self-injurious behavior is so intrusive as to

prevent participation in habilitative and humanizing

activities," or when other interventions have not reduced the

self-injurious behavior. Accordingly, the invention of an

electric skin shock device by JRC does not constitute an

unforeseen change in circumstances that would warrant

termination of the consent decree.

This is not to say that JRC's turn toward electric skin

shock as a physical aversive does not require specific
 49

consideration. As stated, in entering the consent decree, the

Probate Court found that, as of 1987, JRC's use of physical

aversives was safe, effective, and professionally acceptable. A

change in the safety, efficacy, or professional acceptability of

the physical aversives used by JRC would constitute a

significant and unforeseen change in circumstances. Thus,

although the mere invention of the GED, and its use by JRC, is

not a change in circumstances, a finding that its usage is not

safe or professionally acceptable would be. With these

principles in mind, we turn to the judge's findings as to the

standard of care and assess whether those findings were clearly

erroneous based on the evidentiary record before the Probate

Court in 2016.

b. Electric shock and standard of care. In denying the

defendants' motion, the judge below found that, as of the close

of evidence in 2016, there was still no professional consensus

that the use of level three aversives fell outside the standard

of care to treat severely self-injurious and violent behavior.

The department argues that the judge improperly conflated

evidence as to the acceptability of aversive treatments in

general with evidence as to the acceptability of electric skin

shock in particular. On the latter subject, the department

asserts that the evidence is clear: there is "no serious

dispute" as to the professional consensus that electric skin
 50

shock is outside the standard of care for individuals with

developmental disabilities.

We recognize, of course, that a professional consensus does

not require unanimity. In any profession, on the most difficult

issues, unanimity of opinion is often nearly impossible to

achieve. See Planned Parenthood Fed'n of Am., Inc. v. Gonzales,

435 F.3d 1163, 1172 (9th Cir. 2006), rev'd sub nom. Gonzales v.

Carhart, 550 U.S. 124 (2007) ("By medical consensus, we do not

mean unanimity or that no single doctor disagrees, but rather

that there is no significant disagreement within the medical

community"). That said, our inquiry is limited to whether,

based on the evidence before the Probate Court in 2016, the

judge's finding that no professional consensus existed at that

time as to JRC's use of physical aversives was clearly

erroneous.

Our review of the record indicates that there was support

for the judge's finding as of the close of evidence in 2016. In

fact, it appears that when the department filed its motion to

terminate the decree in early 2013, now a decade ago, there was

an ongoing debate about the potential necessity of level three

aversives among the very experts that the department elected to

consult in formulating practitioner guidelines. Separate and

apart from any clinicians tasked with reviewing JRC's regulatory
 51

compliance or treatment plans,23 experts that the department

selected to serve on its PBS advisory subcommittee expressed

ambivalence in 2012 and 2013 about whether electric skin shock

was outside the acceptable standard of care. Indeed, the

subcommittee was nearly unanimous24 in its rejection of draft PBS

guidelines on the use of procedures to "decelerate challenging

behavior," which included language prohibiting electric skin

shock and other level two and three aversives, because members

were "uncomfortable with banning specific procedures." The co-

chair of the subcommittee, Dr. Christopher Fox, suggested an

alternative set of guidelines that would call for

individualized, evidence-based treatments, with rigorous

training and monitoring requirements.25 Another member of the

23There was also testimony, which the judge below
referenced in her findings, to indicate that the independent
clinicians who monitor JRC's regulatory compliance and treatment
plans believed that the GED remained within the professional
standard of care.

24Although the subcommittee co-chair described the
subcommittee's opinion as "unanimous" on this issue, he also
noted that three members were absent from the portion of the
meeting in which the issue was discussed.

25In a later e-mail message, Fox went on to acknowledge
that the 2011 regulations, which predated the formation of the
subcommittee, had already served to limit the use of electric
shock to those patients with existing GED treatment plans; he
nonetheless opined that, "[i]n an ideal world I would like all
interventions to be available," even though "in the world as it
exists currently that is not the case."
 52

subcommittee, Dr. Steve Woolf, expressed a similar sentiment,

writing:

"[Level three interventions] should be implemented based on
three ethical considerations: 1) client's right to safe
and humane treatment, 2) the behavior analyst's
responsibility to use the least restrictive procedure, and
3) the client's right to effective treatment. In my
experiences, [there] is a very small minority of clients
that may require . . . a level three intervention. Banning
these evidenced-based [sic] positive punishment treatments
raises very important ethical concerns when serving clients
with chronic life-threatening problem behaviors. Failing
to use these procedures that research has shown to be
effective in suppressing self-destructive behavior that
have [sic] not responded to positive reinforcement,
extinction, or less intrusive intervention is unethical
because doing so withholds potentially effective treatment
and risks maintaining a dangerous state. . . .

"I would agree to stronger regulation, oversight, and
quality assurance monitoring of these punishment based
procedures. However, the outright prohibition of level
three [interventions] requires more time to study."26

Other members of the subcommittee similarly expressed

concerns that more work was necessary on this issue, with one

member stating that "practices regarding the most severely

behaviorally challenged individuals requires a much greater

degree of collaboration, specification, research and consensus

than has been achieved thus far."

26 The department points out that this e-mail message was
subject to an evidentiary objection, and the judge admitted it
for a limited purpose. However, the judge later admitted the
same e-mail message as a separate exhibit, without limitation,
and the department did not object.
 53

The department seeks to downplay these discussions by

highlighting the fact that the experts did not explicitly

identify electric skin shock in their comments, and argues that

they were instead referencing other level three aversives.

However, the theme that emerges from all of these communications

is a discomfort with banning any specific procedures in that

category, which would include electric skin shock, without

additional evidence and research. And importantly, when the

concerns of these experts were relayed to the department, the

department responded by silencing any further debate among the

subcommittee as to level three aversives. Indeed, from the

outset of the subcommittee's consideration of this topic, the

department bluntly informed the co-chair that "it [wouldn't]

matter" if the ABA literature supported the efficacy and

professional acceptability of specific decelerative procedures

when it came to procedures that the commissioner "[did] not

like."

There was also evidence that this debate was not isolated

to experts consulted by the department. The 2016 edition of the

ABA textbook "Contemporary Behavior Therapy (Sixth Edition),"

excerpts of which were admitted at trial, states that "mild

electric shock often is an effective and efficient means of

significantly reducing self-injurious behaviors." Additionally,

Dr. Richard Foxx, a national expert in this area, believed that
 54

the use of electric skin shock may be necessary to treat a

"very, very small number of exceptional cases where the

individual's behavior was so extreme as to be life

threatening."27

Testimony provided by the plaintiffs about the efficacy of

JRC's treatment methods formed another source of evidence that

the judge could reasonably consider in assessing this issue.

Although the department dismisses this evidence as "anecdotal,"

the testimony credited by the judge reflects that, for many

families with children at JRC, its treatment methods were not

only effective, but also considered more humane than the course

of restraint and pharmacological sedation to which their

children had previously been subjected. One mother testified

that, prior to JRC, her daughter had a long history of school

expulsions and hospitalizations due to her severe aggressive

behaviors. The daughter had previously been prescribed Abilify

27While not necessary to our analysis, we also note that
two separate Federal court cases involving JRC from 2010 and
2012 reference the existence of such a debate within the context
of addressing claims brought under the Individuals with
Disabilities Education Act, 20 U.S.C. §§ 1400 et seq. See
Bryant v. New York State Educ. Dep't, 692 F.3d 202, 215 (2d Cir.
2012), cert. denied, 569 U.S. 958 (2013) (referencing "ongoing
debate among the experts regarding the advantages and
disadvantages of aversive interventions and positive-only
methods of behavioral modification"); Alleyne v. New York State
Educ. Dep't, 691 F. Supp. 2d 322, 332 (N.D.N.Y. 2010) ("It is
readily apparent that the use and benefits of aversives in an
educational setting is a divisive issue among educational
professionals").
 55

and Risperdal, among upwards of twenty other medications, and

had been subject to long periods of seclusion and restraint at

prior placements. All were unsuccessful in treating her violent

behaviors. By the time she enrolled at JRC, it was the only

facility in the Commonwealth that was willing to take her. And

in contrast to the prior treatment inventions she had received,

JRC's treatment protocol was effective in minimizing her

behavioral problems, allowing her to go on field trips and other

outings. As her mother testified, "[My daughter] says her whole

world opened up. . . . She has gone from a person that is

isolated and medicated and injured and unhappy to a young person

that is happy and able to live in a world and experience what

other people experience." A father testified that his son came

to JRC with incredibly harmful behavioral issues, including

rectum and throat gouging, eye picking, and self-induced

vomiting. After being placed at JRC and treated with the GED,

and in contrast to prior pharmacological treatments, the

dangerous behaviors substantially decreased. The father

testified that his son is "happier now than he's ever been" and

engages in hobbies and field trips.

A former JRC patient who testified at trial described

experiencing a similar journey. Prior to JRC, she had

repeatedly been expelled from residential placements, and had

been rejected from as many as thirty-seven programs, due to
 56

extremely violent behaviors that she exhibited toward herself

and others. During this time, she was treated with numerous

medications, which she testified had the effect of making her

feel like a "zombie," and was repeatedly placed in physical

restraints, including straightjackets. When she finally came to

JRC and began treatment with the GED, her self-injurious

behaviors drastically decreased, until they went away

completely. She eventually went on to receive her high school

diploma, obtained gainful employment, and now has children of

her own. These testimonials are also echoed in a description

offered by one of the independent clinicians tasked with

evaluating JRC treatment plans, in an e-mail message sent to the

department's general counsel:

"Having visited institutions and programs all over the
country, and in some foreign countries, I have rarely, if
ever, seen clients with the degree of disability seen at
JRC dressed in shirts and ties, living in community housing
and earning weekends at community recreation, shopping, and
dining activities."

To be sure, despite these examples, and as the judge below

appropriately recognized, the use of level three aversives

remains bitterly contested and controversial, even when it is

limited to a class of patients for whom other treatment

protocols have failed, and authorized only through substituted

judgment proceedings. As the judge acknowledged, JRC stands

alone in using electric skin shock to treat such patients, when
 57

other facilities would decline to do so. And as the department

highlights, the National Association of State Directors of

Developmental Disabilities Services has rejected the use of

electric skin shock, many clinicians regard electric skin shock

as a treatment that does not fall within the standard of care,

and as the judge found, approximately one-half of States have

banned its use on the developmentally disabled. Nonetheless, we

cannot conclude that the judge's finding regarding the use of

aversives was clearly erroneous based on the evidentiary record

before the Probate Court in 2016. See Demoulas, 424 Mass. at

510 ("Where there are two permissible views of the evidence, the

factfinder's choice between them cannot be clearly erroneous"

[citation omitted]).

In reaching this conclusion, however, we remain troubled

that we do so based on a record that is nearly a decade old.

The correspondence between members of the PBS subcommittee in

2012 and 2013 reflects a concern that additional evidence,

research, and dialogue would be necessary to achieve a

consensus. Yet, in response to those concerns, the department

decided that "it was not appropriate" for the subcommittee to

consider the issue further. We also do not know whether these

experts later changed their mind based on additional

information, or whether other significant research and treatment

developments have taken place since the close of evidence in
 58

2016. And when asked at oral argument whether this case should

be remanded for further findings in this regard, the department

was adamant that it not be. Thus, we do not reach the propriety

of electric skin shock treatment in 2023, as we do not have the

record to do so, and we therefore do not foreclose the

possibility that new scientific developments or a more recent

evidentiary record would suffice to demonstrate a change in the

standard of care. See MacDonald, 467 Mass. at 394 ("Although we

conclude that the judge here, on this record, did not abuse her

discretion in denying the defendant's motion to terminate the

abuse prevention order, we leave open the possibility that the

defendant might be able to meet his burden if he were to renew

his motion with a stronger evidentiary foundation").

c. JRC's implementation of GED. The department contends

that, regardless of whether the use of electric skin shock is

acceptable as a general matter, its use at JRC is improper

because it is not employed solely as the least restrictive

method of treatment. The department points to expert testimony

and video footage admitted at trial, which shows eleven specific

instances in which the GED was applied to seemingly minor

behaviors, as proof that "JRC regularly misuses GED."

Importantly, the department does not appear to be arguing

that JRC is violating or subverting the authorization provided

by its court-approved treatment plans. Rather, the department
 59

principally takes issue with some of the behaviors for which JRC

has been granted court approval to use the GED. Yet the

department retains the authority to participate in the annual

substituted judgment proceedings in which those individual

treatment plans are approved. And as we have previously stated,

if the department's monitoring of JRC "reveals any problems [in

an individual treatment plan], that information should be

brought to the judge who has authorized the use of aversive

treatments." JRC I, 424 Mass. at 447 n.20. However, as the

judge below found, the department regularly declines to do so,

despite being given the opportunity to weigh in on a yearly

basis, and despite having access to the materials that JRC uses

in support of its substituted judgment petitions. See 115 Code

Mass. Regs. § 5.14(4)(d)(6) (2011). Given the department's

failure to utilize these existing means of preventing any

unjustified application of the GED in particular circumstances,

we cannot discern why those existing corrective measures are

inadequate and why elimination of the consent decree in total is

an appropriate remedy. The department can and should raise

these specific concerns in the yearly substituted judgment

proceedings before the Probate Court.

6. Whether continued enforcement of consent decree

violates separation of powers. The department further argues

that the decree interferes with the department's regulatory
 60

authority, in violation of the separation of powers expressed in

art. 30 of the Massachusetts Declaration of Rights. We

disagree.

This is not the first time that we have considered the

relationship between the consent decree and the department's

constitutional regulatory authority. In response to a similar

argument raised by the department in JRC I, 424 Mass. at 445, we

indicated that "to read the [consent decree] as a delegation of

all regulatory authority" would raise constitutional concerns.

However, the consent decree contained no such provision to this

effect, and we concluded that it was reconcilable with art. 30.

See id. In so doing, we distinguished those regulatory powers

that the department retains from those actions that must give

way to the consent decree and judicial enforcement. In

explaining that distinction, we stated that the department

retained "authority regarding certification requirements [and]

compliance with applicable regulations," but that the consent

decree reserved "the ultimate decision on an individual's

treatment" to the judiciary, via substituted judgment. See id.

at 445-446. We also explained more specifically that the

department was precluded from using "bad faith regulatory

practices . . . [to] ensure that no individual . . . receive[s]

aversive therapies at JRC." Id. at 449.
 61

We address this "bad faith" regulatory constraint first,

and its relevance to the evidentiary record before the court as

of 2016.28 Given the department's history of using its

regulatory power in bad faith to halt the use of physical

aversives and interfere with JRC operations, it was

constitutionally permissible to impose certain restrictions on

regulatory changes by the department that would limit the use of

level three aversives. This is not a separation of powers

problem. Rather, the department's own bad faith regulatory

practices (and those of its predecessor) justified imposing

limitations on its regulatory authority, by way of a consent

decree, as a form of remedial action. See JRC I, 424 Mass. at

461; Matter of McKnight, 406 Mass. 787, 807 (1990) (Liacos,

C.J., dissenting) (general practice of judicial deference to

agency expertise "is not absolute; it gives way in the face of

agency misbehavior"). By agreeing to be bound by the decree,

the department agreed to additional restrictions on its own

ability to regulate level three aversives in any manner that

would exceed the constraints imposed by the consent decree. The

department also bound itself to the requirement of demonstrating

28In so doing, we note that we have not been presented with
any allegations or evidence of bad faith since that date and do
not purport to address whether any bad faith conduct has
occurred in the seven years that have elapsed since the close of
evidence.
 62

a change in circumstances before it could escape the constraints

contained within the decree.

These constitutionally permissible constraints precluded

the regulatory change proposed by the department in the 2011

regulations. The use of level three aversives was authorized by

the Probate Court, pursuant to the substituted judgment process,

when it was found to be the least intrusive and most appropriate

means of preventing significant harm for an individual patient.

The 2011 regulations took that power away from the Probate

Court, and thus constituted an impermissible end run around

substituted judgment proceedings. Further, the department was

well aware of the existing consent decree at the time it chose

to promulgate the 2011 regulations, and yet made no attempt to

terminate the decree prior to doing so. It is not a separation

of powers problem to enforce the consent decree and its

constraints in this context or to consider the 2011 regulations

as another example of bad faith regulatory misconduct.

Nor do we find persuasive the department's contention that

a prospective regulatory ban on level three aversives is

permissible because it does not interfere with any existing

patient's treatment plan or the substituted judgment process

overseen by the judiciary. This is far too narrow a reading of

our prior decision in JRC I. The consent decree's limitation on

the regulatory powers of the department, which came about as a
 63

result of the bad faith conduct of the department's predecessor,

was not limited to existing JRC patients but extended to the

department's supervision over JRC's operations more generally.

Nor were these constraints limited to interference with the

substituted judgment process in an individual patient's

treatment plan. See JRC I, 424 Mass. at 449 ("it would be

absurd to conclude that, although the agreement was intended to

settle claims that the department's predecessor was improperly

denying the patients needed aversive therapy, the department

could, through bad faith regulatory practices, ensure that no

individual could receive aversive therapies at JRC").

Thus, the department may not prospectively ban the use of

level three aversives for all new patients, in the absence of

changed circumstances, without running afoul of the consent

decree. The existence of such a change in circumstances

requires a judicial determination to that effect, not a

unilateral decision by the department. If the department could

simply pass a new regulation at any point to prospectively ban

the use of level three aversives, the consent decree would be a

pointless paper tiger, ignoring the department's past misconduct

and the resulting consequences.

This does not mean that the department is powerless to

prevent the improper use of the GED. The judge below found that

"physical aversive treatment has not been effective for all JRC
 64

students and may not be the least restrictive procedure

available to treat every student receiving physical aversive

treatment." To the extent that the department agrees that this

is the case for any particular patient, it can and should

register those objections with the Probate Court. We believe

this division of authority is in keeping with JRC I and

separation of powers principles.

Finally, we address the department's argument that failing

to terminate the consent decree violates the department's

statutory mandate. The department is charged with "mak[ing]

regulations for the operation" of providers of residential

services like JRC, see G. L. c. 19B, § 15 (a), as well as

"adopt[ing] regulations . . . which establish procedures and the

highest practicable professional standards for the reception,

examination, treatment, restraint, transfer and discharge of

persons with an intellectual disability in departmental

facilities," see G. L. c. 123B, § 2. The statutory scheme

requires that this latter type of regulation "be adaptable to

changing conditions and to advances in methods of care and

treatment and in programs and services for persons with an

intellectual disability." Id.

Such a mandate must certainly be respected. Further, we

note that the department's ability to pass regulations unrelated

to level three aversives is totally unaffected by the consent
 65

decree. The only issue is whether the department can change

regulations related to level three aversives. In this regard,

evidence of changing conditions and advances in methods of care

and treatment are critical considerations in assessing whether

changed circumstances justify termination of the consent decree

and its limitation on the department's regulatory authority.

The judge's fact findings, however, reject the conclusion that

advances in methods of care and treatment as of the close of

evidence in 2016 supported the elimination of level three

aversives for these deeply troubled patients. Rather, the

expert testimony from 2015 and 2016, or at least the judge's

fact finding regarding that testimony, supported preservation of

level three aversives as an option of last resort for this

particular group at that time. We express no opinion whether

further medical advances since the hearing, or a better

evidentiary record regarding such advances, would justify

lifting the consent decree now or in the future.

7. Existence of changed circumstances of law. Finally, we

address the department's remaining arguments as to changes of

law that would warrant termination of the consent decree. For

the reasons discussed infra, the judge did not abuse her

discretion in declining to grant relief on this basis.

a. Change in Federal reimbursement policy for JRC

services. The department highlights that the Centers for
 66

Medicare & Medicaid Services, a division of the Department of

Health and Human Services that oversees the Federal

administration of Medicaid and Medicare, indicated in 2012 that

it would no longer deem JRC's services eligible for

reimbursement from its Home and Community-Based Services waiver

program.29 As a result, the Commonwealth has expended additional

funds to make up for the shortfall in Federal reimbursement.

From 2012 to 2015, this amounted to $7.7 million.

Although the judge did not address this change in Federal

policy, her failure to do so was not an abuse of discretion.

Even though financial constraints "are a legitimate concern of

government defendants," they are normally assessed within the

context of "tailoring a consent decree modification," rather

than its wholesale termination. Rufo, 502 U.S. at 392-393.

This is not to say that financial constraints could not warrant

termination, but only that the department has not sought to

explain the impact of this funding burden or what strain it has

placed on State resources. Without any such information, we are

29The Home and Community-Based Services (HCBS) waiver is a
program that enables States to receive Federal funding for
community-based services provided to individuals who would
otherwise be institutionalized.
 67

unable to conclude that this is evidence per se to warrant

termination of the decree.30

b. 2011 regulations. The 2011 regulations, through which

the department prospectively sought to ban the use of level

three aversives on new patients, do not constitute a change in

circumstances either. These regulations were promulgated by the

department, a party bound by the decree, and cannot form the

basis for permitting the department to escape, extrajudicially,

the obligations it voluntarily agreed to assume, for the reasons

discussed supra. That much should have been clear from our

prior opinion. See JRC I, 424 Mass. at 449 (observing that it

would be "absurd" to conclude that department could sidestep

obligations under consent decree by resorting to "bad faith

regulatory practices" for purpose of "do[ing] indirectly what

[the] order makes clear [it] cannot do directly"). See also

Delaware Valley Citizens' Council for Clean Air v. Pennsylvania,

533 F. Supp. 869, 876 (E.D. Pa.), aff'd, 678 F.2d 470 (3d Cir.

1982) ("A party should not be permitted, however, to obtain a

30As the department appears to acknowledge in its reply
brief, the decision by the Centers for Medicare & Medicaid
Services (CMS) to stop reimbursements for JRC services was not
competent evidence of a change in the professional standard of
care, as the department presented evidence of CMS's decision
only for the limited purpose of showing that Federal funding for
HCBS waiver participants at JRC had been revoked.
 68

modification of a consent decree because of changed

circumstances of its own creation").

By contrast, a legislative ban on the use of electric skin

shock would constitute a change in circumstances.31 See Rufo,

502 U.S. at 388. And indeed, it is apparent that, during the

relevant period at issue in this case, EOHHS's preferred

strategy for changing the Commonwealth's policy toward electric

skin shock was a legislative ban. It was only in 2010, after no

legislative solution materialized, that Bigby provided the

Governor with other policy options to restrict or eliminate

31The department and the amici also make reference to a
rule promulgated by the Food and Drug Administration (FDA) in
2020 that banned the use of electric shock devices for treatment
of severe self-injurious or aggressive behavior. See 85 Fed.
Reg. 13,312 (2020). This rule -- which was promulgated after
the judge issued her decision below -- was later vacated by the
United States Court of Appeals for the District of Columbia
Circuit as exceeding the FDA's authority. See Judge Rotenberg
Educ. Ctr., Inc. v. United States Food & Drug Admin., 3 F.4th
390, 393 (D.C. Cir. 2021). We note, however, that in December
2022, Congress amended the statutory language that formed the
basis for the District of Columbia Circuit's decision to vacate
the rule. See Pub. L. No. 117-328, § 3306, 136 Stat. 4459, 5834
(2022). In a letter filed pursuant to Mass. R. A. P. 16 (l), as
appearing in 481 Mass. 1628 (2019), the department indicates
that the FDA has recently announced its intent to issue a
proposed rule that would again ban the use of devices like the
GED. If the FDA does, in fact, promulgate the same rule again,
that may well warrant termination of the decree. See
Atlanticare Med. Ctr. v. Division of Med. Assistance, 485 Mass.
233, 247 (2020). See also Rufo v. Inmates of Suffolk County
Jail, 502 U.S. 367, 388 (1992) ("A consent decree must of course
be modified if, as it later turns out, one or more of the
obligations placed upon the parties has become impermissible
under federal law").
 69

aversives. However, Bigby's first instinct was correct -- any

change in circumstances cannot be manufactured by way of

regulatory changes promulgated by the very agency bound by the

decree.

If the department seeks to get out from under the decree,

it must either wait for a legislative solution, provide more

robust evidence that electric skin shock is outside the standard

of care than the record it relied upon in 2016, or establish an

ongoing record of good faith regulatory conduct toward JRC. In

the interim, of course, the department is always free to

intervene in any individual substituted judgment proceeding

where it objects to the use of the GED for a particular patient.

Indeed, in the one recent case where the department chose to do

so, it prevailed. The wisdom of the department's decision not

to avail itself of this option for any other patient is not

before us.

Judgment affirmed.