Full opinion text
FINDINGS AND RULINGS YOUNG, District Judge. I. INTRODUCTION Anonymized minors acting on behalf of a class of approximately 8500 children (collectively, the “Plaintiffs”) who, after being removed from their family homes due to abuse or neglect, have suffered harm or are exposed to harm in the custody of the Massachusetts Department of Children and Families (“DCF” or the “Department”), bring this suit under 42 U.S.C. section 1988 against high-level officials in the Commonwealth’s administrative bureaucracy (collectively, the “Defendants”) for allegedly circumscribing the myriad constitutional and statutory rights of the class members. Specifically, the Plaintiffs contend that the Defendants violated (1) class members’ right to substantive due process; (2) class members’ constitutionally guaranteed liberty, privacy, and associational interests, particularly the right to familial association; (3) certain provisions of the Adoption Assistance and Child Welfare Act of 1980 (“AACWA”), 42 U.S.C. §§ 670-676, specifically those pertaining to foster care maintenance payments and individualized case plans; and (4) class members’ right to procedural due process. The sheer number and breadth of allegations raised by the Plaintiffs in this case effectively amount to an all-out assault on the Massachusetts foster care system, in which the Plaintiffs request all manner of declaratory and injunctive relief. The Defendants, for their part, have moved for judgment on the record. A. Procedural Posture On April 15, 2010, lead plaintiffs Connor B., Adam S., Camila R., Andre S., Seth T., and Rakeem D. (collectively, the “Named Plaintiffs”), by their next friends and on behalf of all others similarly situated, filed a complaint in this district against Massachusetts Governor Deval Patrick and the heads of the Massachusetts Executive Office of Health & Human Services and DCF. Compl., ECF No. 1. That same day, the Plaintiffs filed a motion to certify a class and appoint class counsel. Pis.’ Mot. Class Certification & Appointment Class Counsel, ECF No. 2. The Defendants moved to dismiss the complaint on August 20, 2010, Mot. Hon. Deval L. Patrick Dismiss Compl. Against Him Pursuant Fed. R.Civ.P. 12(b)(1) & 12(b)(6), ECF No. 17; Defs.’ Mot. Dismiss Compl. Pursuant Fed.R.Civ.P. 12(b)(1) & 12(b)(6), ECF No. 18, and subsequently filed an opposition to the Plaintiffs’ motion for class certification, Defs.’ Opp’n Pis.’ Mot. Class Certification, ECF No. 32. Following additional filings by both parties, Judge Michael Ponsor issued a memorandum and order on January 4, 2011, denying the Defendants’ motions to dismiss and leaving for later resolution the Plaintiffs’ motion for class certification. Connor B. ex rel. Vigurs v. Patrick, 771 F.Supp.2d 142 (D.Mass.2011) (Ponsor, J.). Nearly two months later, Judge Ponsor revisited the remaining motion, granting the Plaintiffs class certification and referring the case to Magistrate Judge Kenneth Neiman for further adjudication. Mem. & Order Regarding Pis.’ Mot. Certify Class & Appoint Class Counsel, ECF No. 49; Elec. Order, Feb. 28, 2011. The case was eventually reassigned to this Court on November 19, 2012. Elec. Notice, Nov. 19, 2012, ECF No. 203. On December 3, 2012, the Defendants moved for partial summary judgment on the Plaintiffs’ substantive due process count and for full summary judgment on all of the remaining counts in the Plaintiffs’ complaint. Defs.’ Mot. Partial Summ. J., ECF No. 209. At a motion hearing held on January 10, 2013, -the Court denied the Defendants’ motion as matter of judicial economy. Elec. Clerk’s Notes, Jan. 10, 2013, ECF No. 272. The case proceeded to trial on January 22, 2013. Elec. Clerk’s Notes, Jan. 22, 2013, ECF No. 291. On April 30, 2013, following the close of the Plaintiffs’ case-in-chief, the Defendants filed a motion for judgment on the record and appended to their motion a memorandum of law in support. Defs.’ Mot. J. R., ECF No. 316; Mem. Law Supp. Defs.’ Mot. J.R. (“Defs.’ Mem.”), ECF No. 317. The Plaintiffs submitted a brief in opposition to the Defendants’ motion on May 16, 2013. Pis.’ Mem. Law Opp’n Defs.’ Mot. J.R. (“Pis.’ Opp’n”), ECF No. 356. The motion was heard on May 21, 2013, but the Court declined to rule on it at the hearing, deciding instead to take the matter under advisement and adjourn the case without day. Mot. Hr’g Tr. 37:17-18, May 21, 2013, ECF No. 364. B. Child Welfare Regulatory Framework Under Title IV-E of the Social Security Act, 42 U.S.C. §§ 670-676, the Children’s Bureau of the Administration for Children and Families (“ACF”), which sits within the U.S. Department of Health & Human Services (“HHS”), allots federal funds to states to assist in their provision of foster care services. See Title IV-E Foster Care, Children’s Bureau (May 16, 2012), http://www.acf.hhs.gov/programs/cb/ resource/titleive-foster-care. In order to qualify for Title VI-E funds, state foster care agencies must meet a long list of federal requirements. See 42 U.S.C. § 671(a). The receipt of said funds is additionally conditioned upon participation in and the successful completion of Child and Family Services Reviews (“CFSRs”). Trial Ex. 808, Children’s Bureau Child & Family Servs. Reviews Fact Sheet (“CFSR Overview”) 1. HHS has established seven outcome measures spread across three categories to grade a state agency’s performance: (1) A title TV-E agency’s substantial conformity will be determined by its ability to substantially achieve the following child and family service outcomes: (i) In the area of child safety: (A) Children are, first and foremost, protected from abuse and neglect; and, (B) Children are safely maintained in their own homes whenever possible and appropriate; (ii) In the area of permanency for children: (A) Children have permanency and stability in their living situations; and (B) The continuity of family relationships and connections is preserved for children; and (in) In the area of child and family well-being: (A) Families have enhanced capacity to provide for their children’s needs; (B) Children receive appropriate services to meet their educational needs; and (C)Children receive adequate services to meet their physical and mental health needs. 45 C.F.R. § 1355.34(b)(1). In addition to these seven outcome measures, HHS has adopted six statewide data indicators to assist in determinations of whether a state is in substantial conformity with Title IVE: (1) the recurrence of maltreatment; (2) the incidence of child abuse and/or neglect in foster care; (3)the number of re-entries into the foster care system; (4) the length of time to achieve reunification; (5) the length of time to achieve adoption; and (6) the stability of foster care placement. See Title TV-E Foster Care Eligibility Reviews and Child and Family Services State Plan Reviews, 65 Fed.Reg. 4020, 4024 (Jan. 25, 2000) (codified at 45 C.F.R. pts. 1355-1357); see also Admin. Children & Families, U.S. Dep’t Health & Human Servs., Background Paper: Child and Family Services Reviews National Standards 1 (2012), available at http://www.acf. hhs.gov/sites/defaul1/files/cb/cfsr backgroundpaper.pdf. States that fail to meet the requirements of a CFSR must draft and implement a Program Improvement Plan (“PIP”) that establishes performance improvement goals in the areas in which the states are not in substantial conformity with federal standards. See CFSR Overview 2. There is great incentive to keep pace with the goals set forth in the PIPs, as states that fail to do so are assessed penalties as punishment for their noncompliance. See id. C. Child Welfare Standards National child welfare standards established by the Council on Accreditation (“COA”) and the Child Welfare League of America (“CWLA”) provide the normative backdrop against which DCF’s challenged practices and policies must be cast. COA is an accrediting organization that performs research on best practices in child welfare and sets professional standards for child welfare agencies. See Trial Tr. vol. 11, 33:5-9, 33:23-34:1, Feb. 28, 2013, ECF No. 335. COA standards are derived both from the opinions of independent panels (comprised of experts drawn from across the social services spectrum) and from the relevant academic literature. Rule 30(b)(6) Dep. Council Accreditation, Richard Klarberg 37:19-20, 41:20-42:4, Aug. 9, 2012, ECF No. 226-1. Standards that have been transcribed in draft form are then disseminated for comment to individuals who have expertise in the applicable field. Id. at 41:14-19, 46:17-22. The comments received are customarily incorporated into existing standard statements and reviewed again by the independent panelists, id. at 47:7-12, making the process of COA standard-setting rather iterative and dynamic. CWLA is a public, non-profit agency that, like COA, conducts best-practices research and sets industry standards for child welfare services. Trial Tr. vol. 11, 33:15-21. CWLA standards are set by a process similar to the one used for the development of COA standards. Rule 30(b)(6) Dep. Child Welfare League Am., Linda Spears 77:12-78-24, Aug. 14, 2012, ECF No. 226-1. D. Expert Witness Testimony Over the course of the bench trial, seven expert witnesses (all for the plaintiff) were called to the stand to testify. The Court takes the time here to provide a brief biography of the witnesses and explain their particular relevance or contribution to the matter under review. 1. Named Plaintiff Case File Review Dr. Lenette Azzi-Lessing (“Dr. Azzi-Lessing”) is a tenured associate professor of social work at Wheelock College, located in Boston, Massachusetts. Trial Tr. vol. 3, 103:3-18, Jan. 25, 2013, ECF No. 327. Dr. Azzi-Lessing conducted a review of five of the Named Plaintiffs’ DCF case files, which date from the children’s entry into the foster care system through early 2012. Trial Tr. vol. 6, 76:5-7, 79:3-10, Feb. 4, 2013, ECF No. 330. The aim of Dr. Azzi-Lessing’s case file review was to assess DCF’s effectiveness in providing the five Named Plaintiffs with “safety, permanency and well-being.” Id. at 71:14. In reaching her conclusions, Dr. Azzi-Lessing drew upon COA and CWLA standards, academic literature, federal welfare regulations, internal DCF policies, and her own teaching experiences. See id. at 71:9-14, 73:14-74:2. 2. Children’s Research Center Case File Study The Children’s Research Center (“CRC”), a division within the National Council on Crime & Delinquency, was retained by the Plaintiffs to perform a study of DCF case files to determine whether and the extent to which DCF met prevailing standards of case practice in foster care. See Trial Tr. vol. 4, 4:8-16, 20:15-20, Jan. 30, 2013, ECF No. 328. Dr. Raelene Freitag (“Dr. Freitag”), the director of CRC, managed the study and was a coauthor of the report that contained CRC’s findings. Id. at 4:12-13, 30:7-15. Dr. Kristen Johnson (“Dr. Johnson”), a senior researcher at NCCD, assisted Dr. Freitag at all stages of the study, performing study and research design, establishing data collection protocols, training case readers, conducting data cleaning and analysis, and co-authoring the CRC report. See id. at 22:25-23:2; Trial Tr. vol. 5, 67:22-68:3, 78:1-15, Jan. 31, 2013, ECF No. 329. Dr. Erik Nordheim (“Dr. Nordheim”), a statistics professor at the University of Wisconsin-Madison in Madison, Wisconsin, served as Dr. Johnson’s consultant and advised CRC on its study’s sample design. See Trial Tr. vol. 4, 31:14-32:2; Trial Tr. vol. 6, 33:19-21, 36:5-14. CRC conducted a longitudinal study that involved an examination of a random and representative sample of 484 DCF case files, equally divided into two cohorts of foster children who were followed for a period of thirty months. See Trial Ex. 1066, Compliance Foster Care Case Practice Standards Mass. Dep’t Children & Families: Longitudinal Study Two Cohorts (“CRC Study”); Trial Ex. 1065, Compliance Foster Care Case Practice Standards Mass. Dep’t Children & Families: Longitudinal Study Two Cohorts app. C (“CRC Study Appendix C”); see also Trial Tr. vol. 4, 24:24-25:2, 27:2-4, 31:6-13, 32:12-22; Trial Tr. vol. 5, 84:6-9. The first cohort, labeled the “entry cohort,” was comprised of children who had entered the Massachusetts foster care system during the 'twelve-month window between July 1, 2009, and June 30,2010. See Trial Tr. vol. 4, 25:6-9. This particular group of foster children was selected for the purpose of assessing current DCF practice for children just entering foster care. Id. at 25:9-13. The second cohort, labeled the “two-year cohort,” gave CRC a picture of long-term DCF practice, as the cohort consisted of children who had been in foster care for two or more years as of July 1, 2009. See id. at 25:14-20, 55:11-12. The CRC study covered a wide range of “key points” in foster care practice, touching upon reunification, permanency, placement stability, maltreatment in care, and family visitation, among other subjects. Trial Tr. vol. 5, 77:13-22. 3. Psychotropic Medication Review Dr. Christopher Bellonci (“Dr. Bellonci”) is a board-certified adult and child psychiatrist who presently works as an assistant professor at Tufts University School of Medicine and as an attending psychiatrist at Tufts Medical Center, both of which are located in Boston, Massachusetts. See Trial Tr. vol. 1, 111:24-112:3, 119:24-120:3, Jan. 22, 2013, ECF No. 325. The Plaintiffs retained Dr. Bellonci in March 2012 to review three of the Named Plaintiffs’ case files and produce a report speaking to the degree to which DCF met the standards governing child welfare practices concerning the administration of psychotropic medication and mental health services. Trial Tr. vol. 2, 36:2-6, 36:21-37:10,114:13-19, Jan. 24, 2013, ECF No. 326. He paid particular attention to whether DCF obtained informed consent from the relevant parties to whom they were responsible, provided adequate oversight of the psychotropic drug administration process, and had in place an adequate monitoring system. See id. at 116:21-117:2. To support his findings, Dr. Bellonci relied upon his personal experience working with the foster care population and knowledge of other states’ foster care systems, academic literature, federal guidelines, and standards set forth by the American Academy of Child and Adolescent Psychiatry (“AACAP”). See id. at 38:3-10, 43:11-16, 43:25-44:10, 77:4-13. 4. Management Reviews Catherine Crabtree (“Crabtree”) is a senior project leader at the Center for Government and Public Affairs at Auburn University at Montgomery in Montgomery, Alabama, where she consults with public agencies and non-profit institutions on questions concerning organizational reform and performance management. Trial Tr. vol. 11, 7:19-22, 8:5-12. Crabtree also has an extensive background in child welfare, mental health, and other social services work in Tennessee and Alabama. See id. at 10:6-13:1, 14:25-16:7, 16:19-18:15. In the spring of 2012, the Plaintiffs contacted Crabtree to do a management review of the Massachusetts foster care system. See id. at 24:3-14. The management review focused on “four critical building blocks” of child welfare agency practice, id. at 29:18-19: (1) the number of skilled and trained DCF personnel; (2) the balance of foster care placements and services; (3) the presence of quality assurance systems; and (4) the existence of accountable and stable leadership. See id. at 29:15-31:2. In the course of her review, Crabtree turned to a variety of sources, including state and federal regulations, national professional standards, DCF’s internal policies and communications, and deposition testimony. See id. at 25:14-26:3. Arburta Jones (“Jones”) boasts a long work history in various positions in the child and social welfare systems in New Jersey. See Trial Tr. vol. 8, 115:6-135:1, Feb. 6, 2013, ECF No. 332. The Plaintiffs retained Jones in May 2012 to study issues related to the safety of children in DCF custody who are situated in out-of-home placements. Id. at 135:4-14. Jones’s review focused on the quality of four “hinge pins” of child welfare, id. at 141:16: (1) family visitation; (2) foster home licensing; (3) foster care investigations; and (4) internal and external accountability systems. Id. at 141:13-142:1. Jones’s analysis depended largely upon the depositions of DCF staff and executives; DCF’s internal data, reports, and communications; CWLA and COA standards; data produced by the Massachusetts Office of the Child Advocate (the “OCA”); and state and federal regulations. Id. at 135:24-136:11,138:11-20. II. FINDINGS OF FACT The amount of material that this Court has been called upon to review to inform its findings is, quite frankly, voluminous: over the'course of the last nine months, the Court has scrutinized numerous depositions and party filings, heard twenty-four days of trial testimony, read nearly 3000 pages of trial transcript, and studied just under 1200 trial exhibits (which themselves collectively comprised tens of thousands of pages of documents). Obviously, it would be near impossible to take into consideration every last shred of evidence on every topic broached in this case. Naturally, then, the Court necessarily is selective in its reproduction of the record, condensing the most salient kernels of information and deemphasizing (or altogether omitting) those facts that are not as germane to this Court’s conclusions. A. Maltreatment in Foster Care HHS has conducted two rounds of CFSRs, the first spanning the years from 2000 to 2004 and the second spanning the years from 2007 to 2010. See Nat’l Conference of State Legislatures, State Progress Toward Child Welfare Improvement 2 (2010), available at http://www.ncsl.org/ documents/cyf/progress-cw-improvement. pdf. For the first-round CFSRs, HHS set the national standard for the acceptable incidence of child abuse or neglect in foster care at 0.57%, meaning that a 0.57% rate of substantiated maltreatment would constitute the seventy-fifth percentile of all states’ performance on this statewide data indicator. See Trial Ex. 488, Background Paper: Child & Family Servs. Reviews Nat’l Standards 2-3. The first-round CFSR for Massachusetts, completed in 2001, found a 0.85% incidence of child abuse or neglect in 1997 and a 0.94% incidence of child abuse or neglect in 1999, rates that were 0.28 and 0.37 percentage points above the national standard, respectively. Trial Ex. 55, Mass. Statewide Assessment: Mass. Child & Family Servs. Review (“FirsNRound CFSR Statewide Assessment”) CSF-000000425. For the second-round CFSRs, HHS refrained this statewide data indicator as the absence of maltreatment of children in foster care by foster parents or facility staff and pegged the national standard for this new metric at 99.68% (effectively tightening the old standard from 0.57% to 0.32%). See The Data Measures, Data Composites, and National Standards to be Used in the Child and Family Services Reviews, Notice, 71 Fed.Reg. 32,969, 32,973 (June 7, 2006); The Data Measures, Data Composites, and National Standards to be Used in the Child and Family Services Reviews; Corrections, Notice, 72 Fed.Reg. 2881, 2886 tbl. A (Jan. 23, 2007). In other words, if 99.68% or more of foster children statewide were not victims of a substantiated or indicated maltreatment, that state would be deemed to have met the national standard. The second-round CFSR for Massachusetts, completed in 2007 and relying in part upon state child welfare data from 2004,'2005, and the first three months of 2006, found a 98.72% absence of child abuse or neglect in the Commonwealth during the time period, 0.37 percentage points below the national standard and marking a decline in performance from the first-round CFSR. See Trial Ex. 58, Final Report: Mass. Child & Family Servs. Review (“Second-Round CFSR Final Report”) CSF-000000747, CSF-000000750. Due to its poor performance in the second-round CFSR, Massachusetts submitted a PIP for approval by HHS. See generally Trial Ex. 33, Mass. Child & Family Servs. Review Program Improvement Plan. The Department agreed to achieve an improvement goal of 99.03%, See id. at 54, but later negotiated a lower improvement goal of 98.8%, Trial Ex. 587, DCF PIP Quarterly Report: Quarter One DCF000063281. DCF met the latter negotiated improvement goal. 'Trial Ex. 588, DCF PIP Quarterly Report: Quarter Two DCF000554885. A third-round CFSR, though widely expected, has not yet taken place. See Children’s Bureau, U.S. Dep’t of Health and Human Servs., Child and Family Services Review: Technical Bulletin # 6 (Feb. 4, 2013), available at http://www.acf.hhs.gov/ sites/default/files/cb/cfsr- — tb6.pdf. That said, HHS’s annual publication reporting child maltreatment data serves to fill the gaps for more recent years. This publication provides state-by-state data with respect to the absence of maltreatment in foster care for a given federal fiscal year. See, e.g., Trial Ex. 1088, Child Maltreatment 2010; Trial Ex. 1161, Child Maltreatment 2011. From 2006 to 2011, Massachusetts reported absence of maltreatment rates of 99.05%, 99.14%, 98.93%, 99.16%, 99.22%, and 99.30%, respectively. See Child Maltreatment 2010, at 57 tbl. 3-20; Child Maltreatment 2011, at 55 tbl. 3-15. These results, when compared to those of other states, placed the Commonwealth fourth worst out of forty-six reporting states in 2006, the seventh worst out of forty-six reporting states in 2007, the fourth worst out of forty-eight states in 2008, the seventh worst out of forty-nine states in 2009, the eighth worst out of forty-seven states in 2010, and the seventh worst out of forty-nine states in 2011. See Child Maltreatment 2010, at 57 tbl. 3-20; Child Maltreatment 2011, at 55 tbl. 3-15. The CRC longitudinal study also documented rates of alleged abuse or neglect committed against children in the study’s entry and two-year cohorts. Among 242 children in the entry cohort, forty-four allegations of abuse or neglect during the DCF observation period were reported. See CRC Study tbl. 18. Twelve of these allegations were substantiated, causing children to be removed from their environments in eleven cases. Id. This means that roughly 5% of the children in the entry cohort sample population were victims of substantiated abuse or neglect, and that approximately 91.6% of these children were removed from their environments. The CRC study reported comparable incidence rates for the two-year cohort. Among 242 children in the two-year cohort, fifty-six allegations of abuse or neglect during the DCF observation period were reported. Id. tbl. 45. Ten allegations were substantiated, and children were removed from their environment in eight of these cases. Id. This means that roughly 4.1% of the children in the two-year cohort sample population were victims of substantiated abuse or neglect during the observation period, and that approximately 80% of these children were removed from their environments. Statistics on the two-year cohort prior to the observation period are striking. Of 240 children in the two-year cohort, forty-three children were victims of substantiated maltreatment prior to July 2009. See CRC Study Appendix C tbl. C59. This means that at the start of the observation period, 17.9% of children in the two-year cohort had already experienced substantiated maltreatment while in DCF custody. In roughly 37.2% of these cases, the children had been substantiated victims between two and nine times. See id. The five Named Plaintiff ease files reviewed by Dr. Azzi-Lessing documented egregious instances of maltreatment. For example, at the age of six, Connor B. was placed in a foster home for four to six weeks with a teenager known to be at risk for sexually abusing younger children. See Trial Tr. vol. 6, 83:12-22, 84:10-23. He reportedly raped Connor B. repeatedly during the course of his stay. See id. at 85:15-23. The teenager was subsequently removed from the foster home, and DCF revoked the foster home’s license to host foster children. Id. at 115:24-116:13. In the case of Adam S., DCF initially screened out complaints of abuse or neglect that allegedly took place in Adam S.’s adoptive home, which included instances of corporal punishment and force-feeding. Id. at 117:9-19. Eventually, a DCF investigator initiated a review of the foster home, which resulted in the removal of certain foster children from the home. Id. at 118:18-22. But because the investigator did not provide the remaining children with protective services, Adam S. and his sisters suffered brutal beatings by their adoptive parents until they were ultimately removed from the adoptive home as well. See id. at 118:18-120:6. Andre S. and his sister, both under the age of four, resided in an overcrowded foster home with seven other children for over two years. See Trial Tr. vol. 7, 28:1-4, 28:13-21, Feb. 5, 2013, ECF No. 331. At a later preadoptive foster placement at their cousin’s home, Andre S. and his sister were reportedly prompted to engage in sexual acts with one another and to watch the cousin and her boyfriend have sex and take drugs together. See id. at 35:14-19, 38:14-22. Andre S.’s sister also reported being raped by the cousin’s boyfriend on numerous occasions. Id. at 38:23-39:1. The children were eventually removed, and DCF later investigated and substantiated the majority of the reported events. See id. at 39:2-9. Sworn trial testimony provided by Lauren James (“James”), a former ward of DCF custody, also speaks to maltreatment in care. See Trial Tr. vol. 1, 28:24-25, 29:25:30-2. James attested to a wide range of negative experiences in foster care, including, among other things, sharing beds with other foster children, id. at 32:24-33:1, performing excessive amounts of housework, id. at 35:11-36:12, having inadequate food, id. at 36:13-37:15, having strained contact with family members, id. at 39:17-24, and taking prescribed psychotropic medications beginning at age six or seven, id. at 62:8-24. B. Family Visits and Placements The maintenance of family relationships is an issue of key concern for foster care agencies. According to DCF policy, foster children must be given the opportunity to receive their parents and siblings on visits at least once per month. Trial Ex. 1, DCF Case Practice Policy & Procedures Manual DCF POL (7/08) 140. Children taken into foster care custody are also expected to be relocated to a foster home or other placement with or in close proximity to their siblings and other family members, unless the placement would endanger a child’s safety. See 42 U.S.C. § 671(a)(31)(A); Mass. Gen. Laws ch. 119, § 23(c); 110 Mass.Code Regs. 7.101(l)(b), (e). By natural extension, then, Massachusetts regulations express an unambiguous preference for placements with relatives, known as “kinship placements,” and placements of children into homes in which there are no other foster children, known as “child-specific placements,” over other potential arrangements. See 110 Mass.Code Regs. 7.101(2)(a)-(b) (listing “placement with a kinship family” and “placement with a child-specific family” atop the hierarchy of possible placement resources). DCF must screen kinship placements and child-specific placements before a foster child is relocated, however, and the Department is statutorily obliged to reassess these placements on an annual basis. 110 Mass.Code Regs. 7.108(l)-(2), 7.113(1). In its first-round CFSR, Massachusetts’s performance in achieving continuity in family and area relationships was found to be in substantial conformity with federal law. Trial Ex. 56, Child & Family Servs. Review: Final Assessment (“FirsNRound CFSR Final Assessment”) CSF-000000331; see also id. at CSF-000000331-34 (dubbing as “strengths” Massachusetts’s efforts to ensure that children in out-of-home placements remained in close proximity to their former communities, maintained relationships with their families, and were placed in safe homes with relatives). The Commonwealth did not achieve these results in the second round, however. See Second-Round CFSR Final Report CSF-000000780-91 (labeling Massachusetts’s performance with respect to placements with siblings, kinship placements, and preserving family connections as areas needing improvement). If one were to extrapolate from the results of the CRC study, it would appear that child-family visits are a relatively rare occurrence: only 20.9% and 37.6% of children in the entry cohort received consistent monthly visits from siblings and parents, respectively, for the entirety of the thirty-month review period. CRC Study tbl. 16. Nor does Massachusetts boast a sterling record with respect to the suitability of placements. Roughly 31.9% of children in DCF custody have been placed outside of their local home area, and some 18.6% of children were placed altogether outside their region of origin, see Trial Ex. 418, Proximity Placement Tables 12, which makes visits with family members and caseworkers, commuting to and from school, and attending doctor’s appointments more difficult, see Trial Tr. vol. 11, 118:2-20. The CRC study found that with respect to sibling placement, 69% of children in the entry cohort with siblings also in foster care were placed with at least one sibling for at least part of their time in care, and 49.5% of children in the entry cohort were placed with all of their siblings for at least part of their time in foster care. See CRC Study tbl. 8. The rate of sibling placement for children in the two-year cohort was markedly worse: only 43.9% of children were placed with at least one sibling, and a mere 18.7% of children were placed with all of their siblings. See id. tbl. 37. Reasons for the lapses in sibling placements were documented in only 53.8% of cases in the entry cohort and 38.4% of cases in the two-year cohort. CRC Study Appendix C tbl. C28. In addition, children are sometimes removed to kinship and child-specific placements that have not yet received formal authority to operate. See Trial Ex. 512, Unapproved Homes Active Placements (reporting the list of unapproved kinship and child-specific homes in the Boston, Northern, and Southern regions with active placements). C. Placement Stability Placement stability is another goal that often proves elusive in Massachusetts. DCF regularly makes use of a variety of short-term placements. See, e.g., Dep. Joy E. Cochran (“Cochran Dep.”) 159:22-160:23, Apr. 13, 2012 (describing DCF’s use of night-to-night placements, where foster children are placed for periods of about a week or less and then moved to other locations); Dep. Raymond W. Pillidge (“Pillidge Dep.”) 184:17-185:15, Mar. 7, 2012 (describing DCF’s use of hotline homes, where foster children may be placed for a few days or a weekend in the event that there is a lack of available foster homes). These short-term placements disrupt the lives of children in care and are often used for purposes other than those for which they were designed. See, e.g., Pis.’ Designations Fact Dep. Mary Gambon (“Gambon May Dep.”) 165:24-167:17, May 14, 2012, ECF No. 226-1 (confirming former DCF Commissioner Angelo McClain’s (“McClain”) belief that hotline homes and night-to-night placements are “not good for kids,” id. at 166:6-7, and agreeing that multiple placements are harmful to foster children, id. at 167:13-15); Pillidge Dep. 184:17-185:1 (agreeing that hotline homes are to be used only for emergency placements after the close of regular business hours); see also 110 Mass.Code Regs. 7.101(2)(e) (positioning “placement in a shelter/short term program or group home” on a lower rung of a prioritized list of placement options). Additional problems are posed by DCF’s reliance on Stabilization, Assessment and Rapid Reintegration (“STARR”) facilities and Intensive Foster Care (“IFC”) placements. STARR facilities are intended to be used as “up-to^!5-day placements] for children or youth who may be coming into the department’s care or custody ... [and who may warrant] a period of more intensive assessment and stabilization before determining the next appropriate level of care.” Pis.’ Designations Rule 30(b)(6) Dep. Robert E. Wentworth, Jr. Re: Purchased Servs. Licensing, Delivery, & Oversight 13:2-8, Dec. 29, 2011, ECF No. 226-1. Likewise, IFC programs “provide therapeutic services and supports in a family-based placement setting to children and youth [who] ... are transitioning from a residential/group home level of care ... or discharging from a hospital setting.” Trial Ex. 385, Intensive Foster Care Scope Serv. DCF000465967. Due to a lack of available foster homes, however, foster children are frequently moved into STARR facilities and IFC placements even when they do not meet the eligibility criteria for entering such placements. See Trial Ex. 407, Email Frances Carbone to Perry Trilling & Amy Kershaw DCF003355182 (conveying a DCF’s employee stated reasons for using STARR facilities, including “providing] quick family treatment sometimes to diffuse the situation so that the child can return home”); Gambon May Dep. 213:9-17 (describing DCF’s practice of “hoteling,” which refers to the assignment of a foster child to an IFC placement even if the child does not exhibit the requisite behavior or medical conditions that would qualify her for the placement). In certain instances, children remain in these facilities for longer than is recommended. See, e.g., Trial Ex. 673, N. Regional Office — Apr. 2011, at 1 (revealing that twenty-one children in DCF’s Northern Region remained in a STARR facility for longer than the prescribed forty-five-day period). These persistent placement problems can primarily be traced to a single root cause: there is a severe shortage in the number of foster homes in Massachusetts. DCF’s general goal is to maintain a pool of around 4000 unrestricted homes for foster care placements, see Pis.’ Designations Rule 30(b)(6) Dep. Mary Gambon Re: Recruitment & Retention Foster Homes (“Gambon Oct. Dep.”) 94:12-95:12, Oct. 18, 2011, ECF No. 226-1, but the Department has fallen short of that target since the late 1990s, Trial Ex. 955, DCF Quarterly Reports, 2008-2012, Tab Q3 2012, at 62 fig. 28. Staffing shortfalls contribute substantially to the drag in foster home recruitment. See, e.g., Gambon Oct. Dep. 50:9-51:7 (explaining that four additional statewide recruiters would be necessary in order to fulfill DCF’s recruitment objectives); cf., e.g., id. at 17:14-18 (certifying that DCF’s central office in Boston has only four employees dedicated, and only in part, to the recruitment of foster homes). What’s more, neither bolstering the administrative ranks nor obtaining the requisite number of foster homes will resolve the ongoing placement challenges related to ensuring a child’s unique fit with a prospective placement, a consideration which rightly figures prominently in placement decisions. See Gambon Oct. Dep. 96:20-97:14 (“At any point in time, when you look at the number of homes that [DCF] ha[s] ... you’ll have sometimes upwards of 1,000, 1,500 homes without placements .... [I]t’s all about matching the right home.... [W]e’re looking at are we able to bring in homes that can take sibling groups, are we able to bring in homes that can take certain types of behavior? Are we able to bring in homes that can work more closely with families? Are we able to bring in homes that can commit to a family over time as opposed to taking a child for three to five days until we can find another home? So a lot of it is not just based on numbers but also the type of homes you look at.”); cf. Trial Ex. 671, Findings Initial Child & Family Service Reviews: 2001-2004, at 22 (including “[mismatching placements to children’s needs” among common concerns regarding placement stability). D. Permanency Because substitute care is a less-than-optimal outcome for children, achieving permanency for children in foster care is among DCF’s highest-priority objectives. See 110 Mass.Code Regs. 1.02(3)-(4). To this end, DCF is called upon to “direct [its] efforts toward reunification of child(ren) and parent(s),” and “[a]s soon as it is determined that reunification is not feasible, the Department [is instructed to] take swift action to implement another permanent plan, such as adoption or guardianship.” 110 Mass.Code Regs. 1.02(4). Massachusetts regulations mandate that DCF provide to every family receiving foster care services a service plan that lays out the conditions necessary to achieve one of three goals, the two most relevant being (1) the reunification of a foster child with her family, or (2) the provision of an alternative permanent home for a foster child. 110 Mass.Code Regs. 6.01(l)(b)-(c), 6.02. One of the principal ways to gauge success in this area is to appraise the rate at which foster children who have exited the foster care system reenter it. See Trial Tr. vol. 19, 16:4-17:1, May 7, 2013, ECF No. 347. In the first-round CFSR, Massachusetts’s rate of reentry was 22.3%, more than two-and-a-half times greater than the national standard of 8.6% during the 2000-2001 review period. FirsWRound CFSR Final Assessment CSF-000000306. By the second-round CFSR, only 15.7% of children leaving the Massachusetts foster care system reentered it, a figure still above the national median performance of 15% during the 2006-2007 review period. Second-Round CFSR Final Report CSF-000000765. In federal fiscal year 2010, 15.3% of Massachusetts foster children who were discharged on the basis of reunification reentered foster care within twelve months from the date of discharge, Children’s Bureau, U.S. Dep’t of Health & Human Servs., Child Welfare Outcomes 2008-2011: Report to Congress 178 (2012), available at http://www.acf.hhs.gov/sites/ default/files/cb/cwo08 — ll.pdf [hereinafter Child Welfare Outcomes 2008-2011 ], which placed the Commonwealth forty-first among fifty-two reporting jurisdictions, Trial Ex. 1084, Table Summaries Child Welfare Outcomes Data 2010 (“Child Welfare Outcomes Tables 2010”) 6. The next year, the rate of reentry crept up to 15.6%, Child Welfare Outcomes 2008-2011, at 178, and Massachusetts’s national rank fell to forty-third, Trial Ex. 1085, Table Summaries Child Welfare Outcomes 2008-2011: Report Congress (“Child Welfare Outcomes Tables 2008-2011”) 7. Although the reported rates of reentry from federal fiscal years 2010 and 2011 marked improvements over those seen in the prior decade, they nevertheless deviated fairly significantly from the national medians at that time. See Child Welfare Outcomes 2008-2011, at viii tbl. 2 (showing median rates of 12.6% and 11.8% for 2010 and 2011, respectively). Where reunification is impossible or impracticable, the timeliness of adoptions serves instead as a bellwether of progress in permanency. See Trial Tr. vol. 19, 17:13-18:6. According to results from the first-round CFSR for federal fiscal year 1999, the median length of time for foster children in Massachusetts to achieve adoption was 49.28 months. First-Round CFSR Statewide Assessment CSF-000000428-30. During federal fiscal year 2010, Massachusetts achieved a composite score of 83.7 in timeliness of adoptions, Child Welfare Outcomes 2008-2011, at 178, which represents forty-seventh place among fifty-two reporting jurisdictions, Child Welfare Outcomes Tables 2010, at 7. For fiscal year 2011, Massachusetts obtained a composite score of 76.2, Child Welfare Outcomes 2008-2011, at 178, which caused it to drop two places to forty-ninth place, Child Welfare Outcomes Tables 2008-2011, at 8. On a micro level, Dr. Azzi-Lessing’s case file review squared with the national results. She found that DCF had failed to achieve permanent placements for all five Named Plaintiffs. See Trial Tr. vol. 7, 73:3-77:15. What’s more, DCF’s efforts in this regard were plagued by inconsistency and “inertia,” Trial Tr. vol. 7, 75:21, causing the Named Plaintiffs to “languish” in foster care for years, id. at 73:8, without clear prospects for permanency. E. Case Worker Visitation Federal law requires that caseworkers visit the children in foster care to whom they are assigned on a monthly basis. 42 U.S.C. § 624(f)(1)(A). Research shows that a correlation exists between the frequency of caseworker visits and favorable foster care outcomes. Trial Tr. vol. 17, 107:20-21, May 3, 2013, ECF No. 342. From 2008 to 2011, between 43% and 50% of children received monthly visits from the caseworkers, Child Welfare Outcomes 2008-2011, at 173, which fell far below the 90% benchmark set for this metric during that time period, Trial Ex. 153, Monthly Caseworker Visits Data Fiscal Year (FY) 2007, at 3. CRC’s review indicated that only 12.9% of children in the two-year cohort received consistent monthly contact from their caseworker throughout the two-year review period. CRC Study tbl. 41. The study also found that 16.1% of children in the entry cohort received no contact at all from their caseworker during their first month in DCF’s care. CRC Study tbl. 12. F. Services 1. Preparing Foster Children for Adulthood The provision of life skills to foster children is a core responsibility of child welfare agencies. See, e.g., Trial Ex. 2, Case Practice Policy & Procedures Manual DCF POL 201 (“It is critical that youth served by [DCF] be systematically and comprehensively prepared for independent living to enable them to function as productive members of society.”). To this end, 42 U.S.C. section 675(5)(H) provides, in relevant part: [D]uring the 90-day period immediately prior to the date on which [a foster] child will attain 18 years of age, ... a caseworker on the staff of the State agency, and, as appropriate, other representatives of the child [must] provide the child with assistance and support in developing a transition plan that is personalized at the direction of the child, [which] includes specific options on housing, health insurance, education, local opportunities for mentors and continuing support services, and work force supports and employment services, ... information about the importance of designating another individual to make health care treatment decisions on behalf of the child if the child becomes unable to participate in such decisions and the child does not have, or does not want, a relative who would otherwise be authorized under State law to make such decisions, and provides the child with the option to execute a health care power of attorney, health care proxy, or other similar document recognized under State law, and is as detailed as the child may elect.... 42 U.S.C. § 675(5)(H). Many children, however, “age out” of foster care without ever having learned the necessary life skills to succeed outside the walls of a foster placement. In both the first- and second-round CFSRs, the provision of independent living services was deemed an area needing improvement. See First-Round CFSR Final Assessment CSF-000000327; Second-Round CFSR Final Report CSF-000000777. Indeed, in the third quarter of state fiscal year 2012, nearly as many children left their placements at the age of eighteen as were adopted. DCF Quarterly Reports, 2008-2012, Tab Q3 2012, at 60 tbl. 21. 2. Medical Services Every child, upon entry into the Massachusetts foster care system, must be “screened and evaluated under the early and periodic screening, diagnostic and treatment [ (“EPSDT”) ] standards established by Title XIX of the Social Security Act.” Mass. Gen. Laws ch. 119, § 32. EPSDT services are manifold: among others, they include regular pediatric preventive healthcare visits, physical and nutritional assessments, and developmental and behavioral screening. See Trial Ex. 621, Early & Periodic Screening, Diagnosis, & Treatment (EPSDT) Med. Protocol & Periodicity Schedule (Med. Schedule) & EPSDT Dental Protocol & Periodicity Schedule (Dental Schedule) DCF006427603-10. “Medical passports” are documents prescribed by Massachusetts regulations for use in the foster care system to keep track of a child’s medical, dental, mental health, and developmental history for the length of the child’s stay in foster care. 110 Mass.Code Regs. 7.124. DCF personnel are required to maintain medical passports for each child in its care, id., and are expected regularly to update the passports with relevant information regarding each child’s responses to medication and treatment interventions, see Trial Tr. vol. 2, 60:13-61:3, 63:7-15. Evidence in the record suggests that foster children in Massachusetts commonly receive medical screenings in an untimely fashion, if at all. In 2011, just 12.1% of foster children received on-time, seven-day medical visits; 7.1% received on-time, thirty-day medical visits; and 18.2% completed a medical visit. Trial Ex. 610, Monthly Compliance Medical Screenings Due 2011, at 1. It is also rare to have medical information that is fully complete. The CRC researchers found that only 52.1% and 73.7% of children in the entry cohort and children in the two-year cohort were provided medical passports at their scheduled time. CRC Study Appendix C tbl. C37. Dr. Bellonci opined that he rarely sees medical passports when meeting with patients, which reflects something akin to the rule rather than the exception. Trial Tr. vol. 2, 62:1-17 (“In my 20 years of working with this population, I saw medical passport approximately three or four times. And that’s not an unusual experience.” Id. at 62:5-7). 3. Psychotropic Medication Children in foster care often enter the system with a history of biological and psychological problems, due in no small part to abuse, neglect and frequent environmental changes. See Trial Ex. 17, AACAP Position Statement Oversight Psychotropic Medication Use Children State Custody: Best Principles Guideline (“AACAP Guidelines”) 1. As a result, foster children are often prescribed psychotropic drugs to treat persistent behavioral and emotional issues. See id. National data suggests that foster children in Massachusetts are prescribed psychotropic drugs at rates exceeding those in other states. A report published by the U.S. Government Accountability Office (the “GAO”) in 2011, reviewing psychotropic drug prescription rates and oversight practices in five states (including Massachusetts) in 2008, found that Massachusetts had the highest percentage of children prescribed psychotropic medication among the states considered in the report. Compare Trial Ex. 16, Foster Children: HHS Guidance Could Help States Improve Oversight Psychotropic Prescriptions (“GAO Report”) 102 app. XVII (noting that 39.1% of foster children in Massachusetts are prescribed psychotropic drugs), with id. at 101 app. XVII (noting that 22.0% of foster children in Florida are prescribed psychotropic drugs), id. at 103 app. XVII (noting that 21.0% of foster children in Michigan are prescribed psychotropic drugs), id. at 104 app. XVII (noting that 19.7% of foster children in Oregon are prescribed psychotropic drugs), and id. at 105 app. XVII (noting that 32.2% of foster children in Texas are prescribed psychotropic drugs). More particularly, 4.9% of children between the ages of zero and five, 44.8% of children between the ages of six and twelve, and 53.4% of children between the ages of thirteen and seventeen had been prescribed psychotropic drugs, which exceeded the prescription percentages for nonfoster children in each demographic band by between two and four times. Id. at 102 app. XVII. Overall, foster children in Massachusetts were also 3.8 times more likely to be prescribed psychotropic drugs than children outside of foster care in the Commonwealth, see id., which sat squarely in the middle range among the five states, See id. at 101 app. XVII, 103-05 app. XVII. There are numerous side effects associated with taking psychotropic medications, including hallucinations, mania, paranoia, headaches, nausea, sexual and menstrual problems, rashes, suicidal thoughts, and pancreatic and liver damage, See id. at 41-44 app. II, and additional indicia point to the potential for other health risks. Research suggests that the concomitant use of five or more psychotropic drugs by children holds no medical benefits, that higher doses of psychotropic drugs may prove less effective than the recommended dose, and that no medical evidence supports the use of psychotropic drugs in children under the age of one. See id. at 14-15. The GAO report found that 1.33% of foster children in Massachusetts are prescribed five or more psychotropic medications concomitantly, id. at 107 app. XVIII, which exceeds the percentage of children prescribed five or more medications in all four of the other surveyed states, See id. at 106 app. XVIII, 108-10 app. XVIII. Moreover, 2.21% of foster children in Massachusetts were given dosages of psychotropic medications exceeding the maximum amounts cited on FDA-approved drug labels, id. at 107 app. XVIII, a percentage that dwarfed those found in three of the four other states, See id. at 106 app. XVIII, 108-10 app. XVIII. The administration of psychotropic drugs is only one piece of the puzzle, however: informed consent, clinical oversight, and monitoring systems are required to have a fully functioning medical apparatus in the foster care system. Psychotropic drugs cannot be prescribed without the informed consent of either a child’s parent or guardian or the state, standing in loco parentis, see Trial Tr. vol. 2, 9:20-10:3; AACAP Guidelines 2, and according to AACAP guidelines, the assent of the child herself also ought be obtained, AAC-AP Guidelines 2. Clinical oversight and documentation — manifested in the form of medical passports — are particularly helpful, given the frequent movement of foster children from one placement to the next. Cf. Trial Tr. vol. 2, 64:24-65:12 (explaining that medical treatment of children who experience multiple placement moves is challenging because of the multiplicity of sources a medical provider must canvas to collect a patient’s complete medical history). Further, federal law calls for the deployment of comprehensive monitoring systems, 42 U.S.C. § 622(b)(15)(A) (mandating that child welfare services plans contain, inter alia, “a plan for the ongoing oversight and coordination of health care services for any child in a foster care placement”), which aim to moderate the risks associated with psychotropic medications, see AACAP Guidelines 2 (noting that child welfare agencies should, at a minimum, “[e]stablish guidelines for the use of psychotropic medications for youth in state custody”). Proof of unauthorized or excessive psychotropic prescriptions were rampant in the three Named Plaintiffs’ case files reviewed by Dr. Bellonci. See Trial Tr. vol. 2, 80:5-82:19. Similarly, medical files reviewed in the CRC study rarely included critical information such as when a child had first been prescribed a medication or even whether a child was being treated with medication at all. See Trial Tr. vol. 4, 50:1-9. Massachusetts has a mechanism for overseeing the administration of certain psychotropic drugs, however. Pursuant to the Rogers process, courts are formally required to authorize the prescription of antipsychotic medications — a subclass of drugs within the family of psychotropic medications, Trial Tr. vol. 1, 126:20-25 — to children in DCF custody. Trial Tr. vol. 10, 116:1-5, Feb. 15, 2013, ECF No. 334. In 2009, a working group comprised of representatives from Massachusetts health and welfare agencies and university researchers, the Massachusetts Department of Mental Health, the OCA, and DCF was convened to address issues related to the process of obtaining consent for the prescription of psychotropic medications to children. See Trial Tr. vol. 2, 31:16-17; Trial Tr. vol. 10, 116:6-9, 116:13-20; Trial Ex. 630, Rogers Process Working Grp. Minutes DCF004956552. A study commissioned by the OCA in service of the Rogers working group identified a host of benefits of the Rogers process — namely, secondary review of antipsychotic prescriptions prior to treatment, the appointment of guardians ad litem, and dedicated resource allocation. Trial Ex. 1090, Examination Rogers Process' Youth Custody Mass. Dep’t Children & Families (“Rogers Report”) 8. A number of challenges were also brought to the fore: the Rogers process was without adequate standardization, coordination, and quality assurances; was slow-going; failed to take proper account of medical expertise; and featured inadequate systemic oversight of antipsychotic medication administration. Id. at 9-10. Perhaps most troubling, drugs that are not listed as Rogers-process eligible — namely, nonantipsychotic psychotropic medications, of which there are numerous kinds — are prescribed without the consent of a court or of DCF. Pis.’ Designations Fact Dep. Jan Nisenbaum 194:1-8, May 18, 2012, ECF No. 226-1. The study offered a number of recommendations that might resolve these issues, see Rogers Report 13-16, but it is unclear whether any of the recommendations that emerged out of the Rogers working group were ever implemented, see Trial Tr. vol. 2, 35:23-36:1. G. Caseloads Massachusetts General Laws chapter 18B, section 7 empowers the Commissioner of DCF with the authority to “establish reasonable caseload rates.” Mass. Gen. Laws eh. 18B, § 7(a). The term “caseload” refers to the number of children for which an individual caseworker is responsible at a given time. Trial Tr. vol. 11, 31:25-32:1. Because the mere number of children per caseworker fails fully to capture the actual burden that caseworkers must shoulder, however, “workload” is the more favored metric to use to analyze the reasonability of caseloads. See id. at 32:9-12. A caseworker’s workload comprises a “multitude of tasks related to the actual caseload of one child,” id. at 32:22-23, including traveling to and from the child’s home to facilitate sibling visits, scheduling appointments, composing permanency requirements, and performing data entry, See id. at 32:12-21. For a workload to be deemed manageable, national standards dictate that caseworkers “simply have to have enough time to meet their practice requirements based on how their state is organized, or how their agency is organized,” id. at 35:12-18, taking into consideration caseworkers’ qualifications and subject-matter competencies, id. at 35:18-20. Generally speaking, a workload study must be undertaken to gauge the appropriateness of caseworker caseloads. Id. at 36:20-22. Nevertheless, national standards suggest that, even in the absence of a workload study, caseloads of between twelve and eighteen children represent the maximum that a caseworker can carry while still satisfying her job requirements. See id. at 36:22-25 (noting that CWLA recommends caseload ratios of between twelve and fifteen children per caseworker); id. at 37:16-18 (noting that COA recommends caseload ratios of between twelve and eighteen children per caseworker). DCF employs individuals known in internal agency parlance as “ongoing” caseworkers who provide continuous services to children in the foster care system. See id. at 38:8-17. Unlike their equivalents in most other states’ child welfare agencies, DCF’s ongoing caseworkers carry “mixed” caseloads, meaning that they are responsible for providing services to children placed both in their family home or in a foster home. See id. at 40:23-41:3; Trial Ex. 642, Dep’t Children & Families Proposal Reducing Ongoing Caseloads (“15 Families Initiative Presentation”) DCF010275221. DCF has not conducted or commissioned its own workload study, nor has it adapted those performed by other child welfare agencies. See Trial Tr. vol. 11, 45:19-24; Trial Tr. vol. 16, 92:11-15. Instead, DCF’s current caseload standards — which derive from a collective bargaining agreement originally struck in the mid-1980s between the Commonwealth and the union representing the interests of DCF caseworkers, see Trial Ex. 641, Dep’t Children & Families Rationale for Reducing Ongoing Caseloads DCF010278533 — prescribe that ongoing caseworkers be assigned eighteen cases per month. Trial Ex. 21, Collective Bargaining Agreement 121. In 2005, the two contracting parties agreed in a memorandum of understanding to assign weights to caseloads to account more accurately for caseworkers’ actual workload. Trial Ex. 647, Mem. Understanding Between Dep’t Soc. Servs. & DSS Chapter SEIU Local 509 Concerning Family Res. Workers DCF003540781 (weighting, for example, 0.6 points per active foster home, 0.9 points per foster home undergoing licensure, 0.75 points per licensed foster home on probation, and 0.4 points per out-of-state foster home). Since at least 2008, scores of DCF caseworkers have carried caseloads in excess of the agreed-upon eighteen cases. See, e.g., Trial Ex. 957, Workers Weighted Caseloads Greater 18 Reports, 2008-2012 (“Caseloads Reports”), Tab June 2008, at 57 (reporting that as of June 30, 2008,1034 social workers across six regional offices and a special investigations unit had weighted workloads of over eighteen cases); id. Tab June 2009, at 48 (reporting that as of June 30, 2009, 836 social workers across six regional offices and a special investigations unit had weighted workloads of over eighteen cases); id. Tab June 2010, at 24 (reporting that, as of June 30, 2010, 777 social workers across six regional offices had weighted workloads of over eighteen cases); id. Tab June 2011, at 15 (reporting that, as of June 30, 2011, 492 social workers across four regional offices had weighted workloads of over eighteen cases). Indeed, as of August 2012 (the last month from which the parties could gather facts in this case), 390 caseworkers had weighted workloads of greater than eighteen cases. Id. Tab August 2012, at 11. Observing that the 18:1 caseload ratio established in the 1980s collective bargaining agreement was “[r]apidly [b]ecoming [ojbsolete,” 15 Families Initiative Presentation DCF010275219, DCF launched a campaign known as the “15 Families Initiative,” with the aim of reducing caseload ratios to 15:1, see Pis.’ Designations Fact Dep. Angelo McClain DCF’s Chief Staff (“McClain Dep.”) 132:1-23, May 25, 2012, ECF No. 226-1. DCF acknowledged that such a reduction would bring the Commonwealth in line with national trends and standards. 15 Families Initiative Presentation DCF010275220-21. In order to achieve this goal, however, DCF estimated that it would need nearly two hundred additional ongoing caseworkers at an approximate annual cost of about $10,100,000. Id. at DCF010275224. H. Qualifications and Training Optimally, child welfare agencies should implement social worker training development programs that feature, among other elements, new-hire curricula, on-the-job and supervised training, performance evaluations, continuing in-service training, channels for feedback, and plans for routine reviews and updates of training practices. See Trial Tr. vol. 11, 93:6-21. See generally Trial Ex. 1092, Building Effective Training Sys. Child Welfare Agencies. DCF regularly hires ongoing caseworkers through a competitive, Department-sponsored internship program. Pis.’ Designations Rule 30(b)(6) Dep. Olga I. Roche Re: Staffing, Caseloads, & Training 146:1-147:1, Jan. 20, 2012, ECF No. 226-1. Candidates are often recent college graduates holding a bachelor’s degree or an associate’s degree paired with relevant work experience. Id. In either case, caseworkers typically arrive at DCF full-time with a fair degree of prior experience. See id. New recruits must participate in a competency training program, held monthly, where incoming caseworkers are trained in areas ranging from child development and diversity to deescalation and safety techniques. Trial Ex. 374, Annual Progress & Servs. Report: Fed. FY2012 (“DCF FY2012 Progress Report”) DCF007413805. Funding cuts, however, have led to a reduction in the number of training programs, See id. at DCF007413695 (recountin