(PC) Coleman v. Newsom

Opinion

RALPH COLEMAN, et al., No. 2:90-cv-0520 KJM DB P Plaintiffs, v. ORDER GAVIN NEWSOM, et al., Defendants.

As set by court order, the court held a focused evidentiary hearing on October 15 and 16, 2019, to address unresolved issues the court identified after reviewing Dr. Golding’s whistleblower report and the court’s neutral expert’s investigation into Dr. Golding’s allegations. See ECF Nos. 6242, 6288. The court heard closing arguments from the parties on October 22, 2019. In addition, as authorized by the court, Dr. Golding filed a written closing argument. ECF No. 6362. On October 23, 2019, the court provided an oral pronouncement of its findings and conclusions in open court. Reporter’s Transcript of Proceedings (10/23/19 RT), ECF No. 6380. Those findings and conclusions, with record support, are memorialized in this order. In 1995, the court found “the California Department of Corrections . . . significantly and chronically understaffed in the area of mental health care services. . . . [It] does not have sufficient staff to treat large numbers of mentally ill inmates in its custody.” Coleman v. Wilson, 912 F.Supp. 1282, 1307 (E.D. Cal. 1995). In 2011, the United States Supreme Court observed that the record before that Court supported the conclusion “that the prison system remained chronically understaffed through trial [before a three-judge court] in 2008.” Brown v. Plata, 563 U.S. 493, 528 (2011).1 In October 2017, after more than two decades of remedial effort, this court issued an order requiring defendants to come into complete compliance with the staffing ratios in their 2009 Staffing Plan, ECF No. 3693, and the maximum ten percent staffing vacancy rate required by the court’s June 13, 2002 order, ECF No. 1383, with compliance to be achieved by October 2018. ECF No. 5711 at 30.2 In its October 2017 order, the court included a lengthy discussion of defendants’ request, made in a March 30, 2017 filing, see ECF No. 5591 at 4, for the court “to revisit the existing staffing ratios for psychiatrists.” Id. at 12-20.3 The court made clear defendants faced a

1 Brown v. Plata is a decision by the United States Supreme court “on appeal . . . from a three-judge District Court order . . . applicable to both” this action and to Plata v. Brown, Case No. C01-1351 JST (N.D.Cal.). Brown, 563 U.S. at 499-500. 2 With the exception of citations to page numbers in Reporter’s Transcripts of Proceedings and the deposition transcript of Dr. Kevin Kuich, references to page numbers in documents filed in the court’s Electronic Case Filing (ECF) system are to the page number assigned by the ECF system and located in the upper right hand corner of the page. 3 Defendants’ March 30, 2017 filing, ECF No. 5591, was in response to the Special Master’s Report on the Status of Mental Health Staffing and the Implementation of Defendants’ Staffing Plan (hereafter Special Master’s Staffing Report). As noted, the court discussed it extensively in its October 10, 2017 order, ECF No. 5711. It is one of the filings that has now, belatedly, been corrected by defendants as a result of the proceedings occasioned by the Golding Report. See ECF No. 6302. Although the Golding Report was issued in October 2018, and the Neutral Expert Report on issues raised by the Golding Report, ECF No. 6147, was issued in April 2019, defendants did not agree to correct this pleading until a meet and confer process required by court order and conducted by the parties in June and July 2019, see ECF No. 6187 at 2; the joint status report that followed that meet and confer process included defendants’ agreement to correct the pleading, see ECF No. 6302 at 2 (citing ECF No. 6226 at 30). Moreover, although the court does not in this order make specific findings about the adequacy of defendants’ corrections, it “heavy burden” in attempting to persuade the court those ratios should be revisited. Id. at 14, 18- 19. The court noted defendants’ request could “only be construed as a request to increase the existing caseload of prison psychiatrists” and that there was “scant evidence in the record to suggest this change would advance remediation of the Eighth Amendment violation in this case; rather there is strong evidence that such a change would slow progress toward the end of federal court oversight.” Id. at 19. Nonetheless, the court granted defendants limited leave to explore its request, deciding “not to preclude defendants from raising with the Special Master the issue of whether full implementation of the PMA [psychiatric medical assistant] program supports a change in the staffing ratios for psychiatrists.” Id. The court limited its permission because the record did not support a more extensive revisiting of the 2009 Staffing Plan and the time for defendants’ compliance with the Plan was past due. As of this writing, the record still does not support a more extensive review, and the time for compliance is even more seriously past due. For the year following the court’s October 2017 order, the parties, supervised by the Special Master, engaged in extensive negotiations over issues related to staffing compliance. Ultimately, defendants presented to plaintiffs and the Special Master a staffing proposal that would have cut by approximately twenty percent the total number of line psychiatry staff positions allocated throughout the prison system. See Reporter’s Transcript of Proceedings, October 15, 2019 (10/15/19 RT), ECF No. 6377, at 52:9-18.4 Plaintiffs were poised to accept the proposal. Before they did, however, on October 3, 2018, Dr. Michael Golding, Chief Psychiatrist does note that defendants have replaced a chart attached as Exhibit 2 to the Tebrock Declaration, ECF No. 5591-2, and have corrected two lines of ECF No. 5591. See ECF No. 6302 at 2-3. They have not, however, revised the statements in the Tebrock Declaration that describe Exhibit 2, see ECF No. 5591-2 ¶ 8, nor have they revisited the more general conclusion in the Tebrock Declaration that relied on the now-corrected chart, asserted in their response to the Special Master’s Staffing Report, that “CDCR clinicians, and particularly its psychiatrists, provide quality treatment at very high compliance rates despite the current staffing vacancies.” ECF No. 5591 at 14 (citing Tebrock Decl. ¶ 8). The court is still reviewing the adequacy of defendants’ corrections, generally. 4 At 52:10, the question as transcribed refers to a “2008 staffing proposal.” The reference is actually to a 2018 staffing proposal. See 10/15/19 RT at 52:15-17. for the California Department of Corrections and Rehabilitation (CDCR), sent a whistleblower report to the Plata5 Receiver. Id. at 54:23-55:6. The parties brought the report to this court’s attention on October 5, 2018, and it is that report that has led to the proceedings culminating in this order. As the findings in this order make clear, and contrary to defendants’ initial position -- maintained through the evidentiary hearing -- that no independent investigation of Dr. Golding’s allegations was necessary, those allegations in significant part justified the independent investigation and factfinding the court has undertaken. At this critical juncture, several key legal principles, articulated by the previously- assigned judge in this action, bear repeating: ‘Whatever rights one may lose at the prison gates, cf. Jones v. North Carolina Prisoners’ Union, 433 U.S. 119, 97 S.Ct. 2532, 53 L.Ed.2d 629 (1977) (prisoners have no right to unionize), . . . Eighth amendment protections are not forfeited by one’s prior acts. Mechanical deference to the findings of state prison officials in the context of the eighth amendment would reduce that provision to a nullity in precisely the context where it is most necessary. The ultimate duty of the federal court to order that conditions of state confinement be altered where necessary to eliminate cruel and unusual punishments is well established.’ Spain v. Procunier, 600 F.2d [189] at 193–94 [(9th Cir. 1979)] (emphasis added). Coleman v. Brown, 28 F.Supp.3d 1068, 1077-78 (E.D.Cal. 2014) (emphasis included in 2014 order). The same judge said not so very long ago, in his 2013 order denying defendants’ motion to terminate this action, “[t]he Eighth Amendment violation in this action is defendants’ ‘severe and unlawful mistreatment’ of prisoners with ‘serious mental disorders,’ through ‘grossly inadequate provision of ... mental health care.’” Coleman v. Brown, 938 F.Supp.2d 955, 969 (E.D.Cal. 2013) (quoting Brown v. Plata, 563 U.S. at 500, 502)). Just two years before that denial of termination, in its 2011 decision, the United States Supreme Court had observed that “[f]or years the ... mental health care provided by California’s prisons has fallen short of minimum constitutional requirements and has failed to meet prisoners’ basic health needs. Needless suffering and death have been the well-documented result.” Brown v. Plata, 563 U.S. at 501. 5 Plata v. Newsom, Case No. C01-1351-JST (N.D.Cal.). As the prior presiding judge also noted, once an Eighth Amendment violation is found and injunctive relief ordered, the focus shifts to remediation of the serious deprivations that formed the objective component of the identified Eighth Amendment violation. See Coleman v. Brown, 938 F.Supp.2d at 988. Remediation can be accomplished by compliance with targeted orders for relief or by establishing that the ‘violation has been remedied in another way.’ Id. Coleman v. Brown, 28 F.Supp.3d at 1077. Under no circumstances may remediation be accomplished by end runs and hiding the ball to create a false picture for the court, as has happened here. Given the constitutional deprivations underlying this case, and the court’s monitoring by way of a Special Master, defendants’ expenditure of so much time and effort to create records designed to advance litigation as the primary way to achieve a complete remedy or termination by other means is confounding. This court’s predecessor carefully constructed a process supervised by a Special Master that was intended to moderate court intrusion into defendants’ own remedial efforts. Such a process is arguably more respectful of defendants’ knowledge of their operations and their management prerogatives than a process whereby oversight is transferred to a receivership; it also is more hopeful that defendants can best determine how to meet their constitutional obligations to the seriously mentally ill inmates in their custody. At the same time, given the authority that here remains vested in defendants themselves, the importance of defendants’ transparent and accurate reporting is paramount: the court and the Special Master must be able to rely fully on defendants’ representations. As explained in this order, the court has concluded the reliability of those representations at multiple levels of the Coleman case structure is in serious doubt. If the approach of monitoring by a Special Master has contributed to play in the joints allowing for those misrepresentations, the court may need to revisit that structure in future proceedings. For now, that is a question for another day. Before detailing its findings and conclusions, the court sets forth in greater detail the background leading up to the evidentiary proceedings, which are now concluded. A. Matters Leading Up to Evidentiary Hearing As noted, on October 10, 2017, this court issued its order requiring defendants, within one year, to come into complete compliance with the staffing ratios in their 2009 Staffing Plan, ECF No. 3693, and also with the maximum ten percent staffing vacancy rate required by the court’s June 13, 2002 order, ECF No. 1383. ECF No. 5711 at 30. The court set a further status conference for October 11, 2018 and directed the parties to file a joint status report thirty days prior to the status conference; the status report was to address, as necessary, issues pertaining to enforcement of the order and the durability of the staffing remedy. Id. at 31. The court later continued the hearing to October 15, 2018, and expanded it to include an evidentiary hearing on the use of telepsychiatry to give defendants an opportunity to “prove that the changes they have effected, moving from limited use of telepsychiatry as a supplement to on-site psychiatry in the face of short-term staffing shortages, to the further expansion they appear to be implementing is consistent with the requirements of the Eighth Amendment.” ECF No. 5928 at 12; see also ECF No. 5933. On October 5, 2018, ten days before the scheduled hearing, the court received requests from both parties; plaintiffs requested a status conference, ECF No. 5936, and defendants requested a stay of proceedings, ECF No. 5938. The requests were based on the whistleblower report from Dr. Golding (hereafter Golding Report). Following a special status conference on October 10, 2018, the court vacated the original status conference set to consider enforcement of the October 10, 2017 order and the evidentiary hearing on the use of telepsychiatry. ECF No. 5949 at 5; ECF No. 5980. After hearing from the parties and Dr. Golding’s counsel both orally and in writing, ECF Nos. 5967, 5969, 5976-5978, on October 25, 2018, the court ordered the Golding Report filed on the public docket in redacted form.6 ECF

6 The redactions approved by the court were relatively minor, and made to protect “(1) current and former employees’ names; (2) current and former employees’ employment titles, but only where an employment title is expressly linked to a name or a specific prison so as to disclose a person’s identity; and (3) any other information that serves to identify an individual Nos. 5986-5988. A complete unredacted copy of the Golding Report is filed under seal. ECF No 5990. B. Appointment of Neutral Expert The court held a series of hearings, ECF Nos. 5964, 5980, 5995, issued an order to show cause why the court could not appoint its own neutral expert, ECF No. 6002, and considered the parties’ responses, ECF Nos. 6009-6012, 6015. On December 14, 2018, the court appointed Charles J. Stevens, Esq. of Gibson Dunn & Crutcher LLP as the court’s neutral expert under Federal Rule of Evidence 706, “to assist the court in investigating allegations raised in [the verified Golding Report) to determine whether defendants have committed any fraud on the court or the Special Master, or have intentionally provided false or misleading information to the court or the Special Master.” ECF No. 6033 at 1-2. The court filed an amended appointment order on January 8, 2019. ECF No. 6064. The amended order modified paragraphs A(2) and B(4) of the original appointment order, as requested by the neutral expert, and modified the first paragraph of that order to reflect events since the December 14, 2018 order was filed. Id. at 1 n.1. The court tasked the neutral expert with conducting an independent investigation to identify facts, if any, that raised a question whether defendants committed fraud on the court or intentionally misled the court or Special Master regarding seven issues the court specifically identified in its appointment order, as follows: a. Lengthening the intervals between psychiatric appointments beyond court-mandated timelines for inmate-patients at the Correctional Clinical Case Management System (CCCMS) and Enhanced Outpatient Program (EOP) levels of care who are transferred to new institutions by resetting the clock for such appointments from the time of transfer rather than from the last completed appointment, rescheduling such appointments at the maximum time allowed in the Program Guide, and reporting compliance with Program Guide requirements using the reset timelines. See Golding Report, ECF No. 5988-1 at 1, 14-23. b. Lengthening the interval between psychiatrist appointments for EOP inmate-patients and reporting compliance based on the extended intervals. See id. at 2, 23-26.

other than Dr. Golding, including government telephone numbers and email addresses.” ECF No. 5986 at 9. c. Combining CCCMS and EOP appointment compliance numbers into one reporting category. See id. at 26-27. d. Inflating compliance numbers by counting every encounter between a psychiatrist and an inmate-patient as an appointment for purposes of measuring Program Guide timeline compliance, without regard to whether the encounter was a psychiatry appointment or, e.g., a wellness check or a cell-front attempt to communicate with an inmate patient. See id. at 5-6, 54-57. e. The manner of reporting of scheduled appointments and missed appointments. See id. at 7-8, 35-47, 62-63. f. Failing to report that psychiatric supervisors were also performing some or all the functions of staff psychiatrists. See id. at 5, 56-57. g. The way in which medication non-compliance is measured. See id. at 8, 58-62. Id. at 2-3. The court did not delegate any ultimate fact-finding authority to the neutral expert, but reserved that critical role to itself. See ECF No. 6187 at 2. C. The Neutral Expert Report The neutral expert conducted a four-month investigation, and on April 22, 2019, submitted a report to the court. See ECF No. 6135 at 1. Without objection, on May 3, 2019, the court filed the unredacted Neutral Expert Report on the public docket. ECF Nos. 6146, 6147. The Neutral Expert Report points to substantial indications of defendants’ presenting misleading information to the court and/or the Special Master, including: 1. Defendants’ making the December 2016 business rule change to redefine “monthly” to lengthen the intervals between Enhanced Outpatient (EOP) appointments from 30 days to up to 45 days. While in effect, this business rule generated misleading data about defendants’ compliance with Program Guide requirements for routine EOP evaluations. ECF No. 6147 at 42, 48.

2. Defendants’ reporting of “Timely Psychiatry Contacts,” overstating compliance with Program Guide timeline requirements. Id. at 67. Specifically, California Department of Corrections and Rehabilitation (CDCR) data presented to the court and Special Master was inconsistent with Program Guide requirements and made defendants’ reports in this area appear more compliant with Program Guide timeline requirements than defendants actually were. Id. 3. Defendants’ reporting of psychiatric evaluations, erroneously skewed toward confidential evaluations. Id. at 68. “EHRS [Electronic Health Records System] data on compliance with Program Guide timelines for compliance with psychiatric evaluations is potentially misleading because it includes non-confidential encounters. . . .” Id. 4. Between 2016 and October 2018, defendants’ use of an incorrect and potentially misleading definition of the Appointments Seen as Scheduled performance indicator, resulting in data being provided to the court in June 2018 and to the Special Master as part of the Continuous Quality Improvement (CQI) evaluations in a “misleading manner.” Id. at 75-76.

5. Defendants’ submitting misleading data on “Timely Psychiatric Contacts” in support of their 2018 Staffing Proposal in that the data did not accurately reflect the extent to which appointments were seen by supervisors rather than line psychiatrists. Id. at 82.

6. Defendants’ submitting misleading data on the timeliness of mental health referrals, “because for medication noncompliant patients it only counted those patients for whom a psychiatrist ordered a medication noncompliance counseling appointment as a matter of discretion,” not all patients who require medication noncompliance appointments. Id. at 84. As a result, the performance indicator overstated compliance. Id. at 91.

7. Due to a bug in software, defendants counting cancelled noncompliance appointments as completed; however, “this inaccurate data was less favorable to CDCR than the corrected data.” Id. The court provided the parties an opportunity to file substantive responses to the Neutral Expert Report and then set the matter for a special status conference on June 10, 2019 to discuss issues raised in the briefing. ECF No. 6135 at 2. D. Review of Materials Defendants Claimed as Privileged During the course of the neutral expert’s investigation, defendants filed a motion for protective order, seeking to avoid producing to the neutral expert documents he requested but for which defendants asserted claims of attorney-client privilege and/or work product protection. ECF No. 6086. The court denied the motion for protective order, specifically noting the court had already provided those claims would not be waived by disclosure of “potentially privileged material . . . to the court’s neutral expert during the investigation and then to the court subject to the claims of privilege.” ECF No. 6096 at 6. Despite the court’s order, defendants still did not produce the documents to the neutral expert. See ECF No. 6147 at 14. The neutral expert declined to litigate the issues, concluding he “could make the findings requested by the Court without” doing so. Id. His findings were, therefore, “subject to the qualification that [he] did not review information claimed by Defendants to be protected by attorney-client or work product privileges.” Id. At the June 10, 2019 status conference, the court signaled it would require defendants to produce the privileged documents to the court for in camera review, and the court confirmed its tentative ruling by minute order the same day. ECF No. 6180. On June 14, 2019, defendants filed both a motion for reconsideration by this court, ECF No. 6188, and an emergency petition for writ of mandamus in the United States Court of Appeals for the Ninth Circuit. See Newsom v. USDC-SAC, CAD # 19-71493 (9th Cir. filed Jun. 14, 2019). On June 18, 2019, this court denied the request for reconsideration, ECF No. 6200, and the appellate court denied the mandamus request the next day. ECF No. 6202. Thereafter, defendants produced the documents, which the court itself reviewed in camera; during its review, the court discussed certain documents in camera on a few occasions with defendants. See, e.g., ECF Nos. 6270, 6323.7 As the procedural history reflects, defendants have resisted at every turn any reliance by the court on any portion of any document for which they have asserted a claim of privilege. The court’s general impression from its review of these documents is that defendants have overreached in a number of their privilege claims, although some claims of privilege would be sustained if not waived by reason of defense positions taken previously in these proceedings. Nonetheless, the court has determined that to venture further into the thicket of these privilege claims would waste valuable court time and resources and distract from the important, indeed imperative, tasks that remain to achieve delivery of constitutionally adequate mental health care to the plaintiff class. The court therefore has not prolonged these proceedings by issuing further

7 The court is maintaining as lodged documents those documents it reviewed in camera and has made a record of all in camera proceedings, which is being maintained under seal until further order of court. orders defendants are likely to appeal. Ultimately on the merits, after careful consideration, the court has determined it need not rely on any of the privileged documents.8 For purposes of these proceedings, it is enough to say that nothing in the privileged documents reviewed by the court supports a different conclusion than reached below. That is to say, nothing before the court indicates, directly or by way of inference, that anyone involved in presenting misleading information to the court committed intentional fraud. Rather, the picture that emerges from the documents reviewed in camera is consistent with the picture that emerges from the public record created over the course of the proceedings prompted by the Golding Report. It is this public record on which the court relies in making its findings and conclusions. A. Orders Narrowing Issues for Hearing In its June 14, 2019 order following the June 10, 2019 status conference, the court found that in five of the seven areas referred to the neutral expert for investigation, he had identified evidence that, “if confirmed through further proceedings and accepted by the court, could establish that misleading data has been presented to the court and/or the Special Master.” ECF No. 6187 at 2. These areas included Issues B, D, E, F and G as described in the court’s appointment orders. Id. The court set an evidentiary hearing to probe those five issues, “to take evidence, as necessary, to determine (a) whether misleading data was presented to the court and/or the Special Master; (b) if misleading data was presented, how and why that happened; and (c) what action is required to correct the record and avoid future submission of misleading data.” Id. The court also directed the parties to meet and confer in an effort to determine whether they could “stipulate to one or more of the underlying facts suggested by the results of the neutral expert’s investigation.” Id. On August 8, 2019, the court held a telephonic prehearing conference. ECF No. 6236. The court then filed an order on August 14, 2019, confirming and clarifying several 8 In open court on October 23, 2019, the court ordered defendants to file the non- privileged portion of the document identified as CDCR-PRIV 0000408-412. See 10/23/19 RT at 443:18-23. Defendants have complied with that order. ECF No. 6370. matters covered at that conference. ECF No. 6242. In particular, the court found the parties had stipulated to several key facts suggested by the Neutral Expert Report and defendants had admitted that misleading information was provided to the court. See id. at 5-11; see also 10/23/19 RT at 444:67. As a result, the court narrowed the scope of the evidentiary hearing, identifying the issues remaining for hearing as follows: • Why neither Dr. Leidner nor Dr. Ceballos consulted with Dr. Golding in connection with the decision to change the definition of “monthly” in the relevant business rule, and why no one from CDCR informed the Special Master or any member of his team about this change;

• How the Appointments Seen as Scheduled indicator was developed incorrectly and in the absence of consultation with Dr. Golding or other quality control measures, and what steps defendants plan to take to ensure indicators and definitions are developed with appropriate consultation and quality control in the future; and • Why defendants did not disclose in their 2018 Staffing Proposal whether, and to what extent, the reporting of data related to average frequency of patient contacts did not disclose the use of supervisory psychiatrists to complete caseload contacts with patients; and to what extent defendants knowingly relied on active participation of supervisory psychiatrists in performing the duties of line psychiatrists both in defendants’ 2018 Staffing Proposal and in supporting their representation that, if adopted, the 2018 Staffing Proposal would bring defendants into compliance with the October 2017 staffing order.9 ECF No. 6242 at 5, 9-10; see also ECF No. 6288 at 2. B. Evidentiary Hearing Schedule The court convened the evidentiary hearing commencing on October 15, 2019. ECF No. 6345. The court heard testimony from eight witnesses over two days, asking its questions first and then allowing the parties to ask questions; the court admitted several exhibits into evidence as moved by the parties. ECF Nos. 6345, 6350. In addition, the court received deposition testimony from Dr. Kevin Kuich, a psychiatrist, in lieu of his live testimony. See

9 The court accepted several other factual stipulations of the parties, drawn from evidence reported by the neutral expert, and referred issues related to those factual matters to the Special Master. See ECF No. 6242, passim. Annotated Deposition Transcript of Kevin Kuich, dated 9/19/2019 (Kuich Dep.), ECF No. 6406; see also ECF No. 6357 (resolving objections to parts of Kuich Testimony).10 Additionally, in accordance with the parties’ stipulation, the court accepted declarations from defendants’ attorneys Rae Onishi, Esq., Nicholas Weber, Esq. and Melissa Bentz, Esq. in lieu of their live testimony. See ECF No. 6337; see also 10/23/19 RT at 444:2-7. Following hearing, on October 22, 2019, the court heard closing argument from plaintiffs and defendants. ECF No. 6364; see also Reporter’s Transcript of Proceedings (10/22/19 RT), ECF No. 6379. The court also received a written statement regarding the evidence presented at hearing from Dr. Golding’s counsel. ECF No. 6362. On October 23, 2019, the court pronounced oral findings and conclusions in open court. ECF No. 6365. It is those findings and conclusions that are memorialized in this order. C. Witnesses As discussed above, the court heard live testimony from eight witnesses and received the deposition testimony of a ninth witness. The record memorializes the testimony of all the witnesses and ultimately their testimony speaks for itself. The court has considered all of the testimony and exhibits. The court’s assessment of certain witnesses in particular, including their credibility and the substance of their testimony, is central to the court’s findings and

10 The court reviewed the annotated Kuich deposition transcript and ruled on the parties’ objections prior to the time the annotated deposition transcript was filed on the docket. Compare ECF No. 6357 (filed 10/21/19) with ECF No. 6406 (filed 12/5/19). The texts of the two transcripts are identical. For technical reasons not clear to the court at this time, the annotated version used by the court to rule on the parties’ objections contained signals that suggested objections were one page later in the transcript than the actual pagination, which did not show in this version after page 7. For this reason, the page numbers cited in the court’s October 21, 2019 order ruling on objections, ECF No. 6357, do not match the page numbers in the annotated Kuich deposition transcript filed on the docket at ECF No. 6406. Accordingly, the court’s October 21, 2019 order is deemed amended to change each page number cited in that order by subtracting one from the cited page number. For example, the first objection cited in the order is at 24:25 to a question at 24:22-24, and the answer at 25:1 is disregarded. There is no suggestion in the record that this change has prejudiced or will prejudice any party. conclusions, and to the clarification, cleansing and purging necessary to move this case forward. The court therefore reviews its assessments of these selected witnesses below. 1. Dr. Michael Golding The court first heard from Dr. Golding. Based on the substance of his testimony and his demeanor on the witness stand, the court finds Dr. Golding credible. His observations and conclusions overall are well-founded. When he learned that the psychiatrist compliance indicator for timeliness of EOP appointments on the Mental Health Dashboard11 “had turned green,” he asked a Senior Psychiatrist Specialist member of his headquarters team, Dr. Melanie Gonzalez,12 to “look at the data to see what was going on.” 10/15/19 RT at 45:2-7; 76:6-11. Working with Dr. Kuich,13 see Kuich Dep. at 89:25-94:11, Dr. Gonzalez performed the requested analysis, and that analysis informs the conclusions in Dr. Golding’s Report as well as in her report. The defendants’ contentions articulated in their closing, that Dr. Golding “just about disagrees with everyone,” 10/22/09 RT at 426:4, and that he has a pro-psychiatry bias, are not well-founded and even if they contain a grain of truth do not undermine the doctor’s credibility. Whether or not Dr. Golding has a disagreeable side, which was not evident during his testimony, is irrelevant to whether he testified credibly and knowledgeably. And it would be understandable 11 In his report, Dr. Golding describes the Mental Health Dashboard as “CDCR’s self monitoring tool.” ECF No. 5988-1 at 4. The Mental Health Dashboard uses red, yellow, and green color coding to illustrate the degree of compliance with a wide variety of measures. See, e.g., Kuich Dep. at, e.g., 109:1-20 (“institutions were very much . . . into the dashboard, very much into the metrics, very much into how they’re performing. . . . Tremendous amount of pressure placed on the institutions by the regional staff to be able to get their metrics in the green. And not always did that effort to help the institution get in the green, look at the actual weeks and the details; it just involved work harder, work longer, get it done. I don’t care how you get it done. So if you need a cell side appointment, that’s what you need to do. If you need to do something, that’s what you need to do. We need to go from red to yellow or yellow to green.”). 12 Dr. Gonzalez sent a whistleblower report to the Plata Receiver on or about October 24, 2018, three weeks after Dr. Golding delivered his whistleblower report. See ECF No. 6363 at 1; 10/15/19 RT at, e.g., 54:23 (Golding Report issued October 3, 2018). 13 Dr. Kuich worked for CDCR from August 1, 2013 to mid-January 2019, first as a staff psychiatrist at California Health Care Facility (CHCF) in Stockton, then as a senior psychiatrist specialist at CDCR headquarters, and finally as the chief telepsychiatrist. Kuich Dep. at 8:7-9:5. He now works in Hawaii, in the same profession. Id. at 8:7. if he has a pro-psychiatry bias. He is, after all, a chief psychiatrist with CDCR, and it is hard to see how advocating for his professional counterparts and the integrity of the mental health care delivery system in CDCR displays a bias that undermines his credibility. Finally, contrary to defendants’ attempt to paint Dr. Golding as a solo outlier, his testimony and his report on the serious matters at issue here do not stand alone. Both Dr. Gonzalez’s report and Dr. Kuich’s deposition testimony corroborate Dr. Golding’s position in substantial and significant ways. Dr. Kuich’s deposition, in particular, provides a very helpful narrative, placing essential pieces of evidence into context in a way that brings the considerable bureaucratic dysfunction within defendants’ operations into clearer focus. Dr. Golding’s explanation for why he was not able to satisfactorily resolve the issues he raised internally and for why he provided his report to the Plata Receiver rather than the Coleman Special Master also are both credible and evidence of the dysfunction illuminated by these proceedings. In one respect, the court does not find other evidence in the record to support Dr. Golding’s strong belief that the Governor’s Office expressly directed the provision of misleading data to the court. Additionally, the court cannot on the present record and does not resolve whether then-Deputy Director Tebrock told Dr. Golding that by telling her about fraud he had “unburdened [him]self,” 10/15/19 RT at 26:19-27:1, because she is an officer of the court, a statement Ms. Tebrock denies making, see id. at 98:8-18. 2. Katherine Tebrock At times relevant to the events that gave rise to the Golding Report, Ms. Tebrock was Deputy Director of CDCR’s Statewide Mental Health Program. 10/15/19 RT at 82:13-20. She left that position voluntarily on July 12, 2019. Id. at 82:18-22. She is still employed by the State of California, though she does not work for CDCR and is not a gubernatorial appointee. Id. at 105:1-9. Ms. Tebrock is a person of obvious intelligence and significant abilities. While the court found her testimony credible, that testimony was also disappointing given the overall message it sent. Ms. Tebrock failed to fully accept responsibility for her own failures, including failures in the leadership she was required to exercise given her role as Deputy Director. Perhaps she too was a victim of the bureaucratic dysfunction so plainly evidenced by the record here, and not provided adequate leadership training and support to manage the demands of the complex environment in which she was working. That said, Ms. Tebrock signed at least one key declaration in this case during the relevant time frame and that declaration, which contained misleading information, was filed with the court. See 10/15/19 RT at 103:19-104:6 (discussing ECF No. 5591-2, March 30, 2017 declaration of Ms. Tebrock containing data based on changed business rule that extended timelines between EOP appointments from 30 to 45 days). And she did not ask anyone to correct the data in that declaration even after she became aware of Dr. Golding’s report. 10/15/19 RT at 103:19-104:26.14 She also signed at least five EOP Administrative Segregation Unit (ASU) Hub certification letters15 tendered to the Special Master between January 2017 and May 2017, which contained data created with the changed business rule. See 10/15/19 RT at 105:20-25 (Tebrock testimony that she signed EOP ASU Hub certification letters monthly); see also ECF No. 6330 at 4 (correcting EOP ASU Hub certification letters submitted to the Special Master between January 2017 and May 2017). Although defendants initially took the position they did not have to correct these letters, see ECF No. 6257 at 26-27, on October 10, 2019, defendants did send a letter to the Special Master containing corrected data for these letters.16 See ECF No. 6330. Although Ms. Tebrock testified at hearing that she would, if given the chance, correct any pleadings containing erroneous data, 10/15/19 RT at 123:22-124:2, this offer comes too late and rings hollow. Defendants have been given many

14 But see note 3 supra. 15 Since August 1, 2014, defendants have been required to “provide to the court and the Special Master monthly reports on whether each EOP ASU hub meets Program Guide requirements for an EOP ASU level of care.” ECF No. 5150 at 2-3 (revising ¶ 2c of ECF No. 5131 at 73:19-74:3). Defendants are prohibited from admitting any Coleman class member receiving mental health treatment at the EOP level of care “to any EOP ASU hub that has failed to meet or exceed Program Guide requirements for a period of more than two consecutive months.” Id. 16 Plaintiffs object that these corrections “do not go far enough.” ECF No. 6360 at 2. The court is in the process of reviewing corrections filed by defendants to date, as well as plaintiffs’ responses thereto and makes no findings at this time concerning the thoroughness or the accuracy of defendants’ corrections to the record. months since Dr. Golding filed his report to correct the record, including during the time Ms. Tebrock remained at CDCR. For reasons that are unclear to the court, Ms. Tebrock never availed herself of those opportunities. Ms. Tebrock’s handling of the defendants’ misguided 2018 staffing proposal was also inexplicably constrained, as if carefully curated to preclude meaningful input from psychiatry. Her explanation for her failure to give Dr. Golding a written draft of the staffing proposal before it was finalized – that the proposal was a court document to be wordsmithed by lawyers and therefore its substantive content tightly controlled by lawyers – is wholly unsatisfactory, given lawyers’ unbending obligation to ensure information submitted to the court is, among other requirements, “not being presented for any improper purpose” and “factual contentions have evidentiary support.” See 10/15/19 RT at 99:25-100:22; Fed. R. Civ. P. 11(b). When Dr. Toche later was asked what she made of Ms. Tebrock’s explanation in this respect, Dr. Toche declined to defend it, which speaks volumes. Reporter’s Transcript of Proceedings (10/16/19 RT), ECF No. 6378, at 351:19-352:10. 3. Dr. Laura Ceballos Dr. Ceballos is a psychologist who serves as Mental Health Administrator of Quality Management, Inpatient Facilities, for CDCR’s Statewide Mental Health Program and was appointed as Chief of Quality Management in 2009. ECF No. 6012-2 at 1-2. As plaintiffs elicited in their questioning during hearing and pointed out in their closing, Dr. Ceballos designed the Continuous Quality Improvement Tool (CQIT). 10/16/19 RT at 305:4-5; 10/22/19 RT at 402:17-24. The CQIT is the measurement tool identified in this case as key to demonstrating defendants’ progress toward and ultimate compliance with a durable remedy. See ECF No. 5477 at 3-4 (quoting ECF No. 4232 at 4-5). Despite her central role, Dr. Ceballos’s testimony betrayed little to no appreciation for the letter or the spirit of the court orders underlying the development of CQIT. Rather, for most of her testimony Dr. Ceballos maintained a false distinction between “internal” data she said defendants use without implicating any Coleman court orders, and “external” data formally reported to the court. See, e.g., 10/16/19 RT at 255:14-25, 260:19-261:3. In maintaining this distinction she consistently characterized her role as “a clinician . . . not an attorney.” See, e.g., 10/16/19 RT at 304:25. The Special Master has, in the past, considered Dr. Ceballos a key contact and relied on her to tell him or members of his team about significant developments related to this case. Yet Dr. Ceballos testified that she viewed the business rule change that lengthened the interval between EOP appointments from 30 to 45 days as insignificant such that she did not have to report that change to the Special Master or the court. She was consistent in saying she thought the change fell within the Program Guide requirement of “monthly” appointments for EOP patients and was made simply to help psychiatrists improve “continuity of care.”17 See, e.g., 10/16/19 RT at 278:23-280:10, 282:10-19. But her position does not withstand scrutiny in light of the record as a whole. Of particular significance, in late 2015 when the request to change the business rule was first raised, it was not approved by Dr. Golding and it went nowhere, yet when the request was made again a year later during a time when defendants were attempting to develop a staffing proposal to address the ongoing psychiatrist shortage the request was granted almost instantly, bypassing Dr. Golding altogether. See Section IV(B)(2)(a) infra. In sum, Dr. Ceballos’s testimony was in critical respects simply not credible. Whether the trust the Special Master previously placed in Dr. Ceballos can be maintained will be up to him, but in the court’s view it would be entirely reasonable for him to conclude that trust has been irreparably undermined. 4. Dr. David Leidner Despite plaintiff’s arguments to the contrary, the court found Dr. Leidner’s testimony credible. Fundamentally, Dr. Leidner distinguished himself from the witnesses other than Dr. Golding as someone who exhibited a conscience during his testimony. He understands he made mistakes, did not shrink from explaining that he had and worried about the consequences. Overall, the record suggests Dr. Leidner was working at the direction of others

17 In this context, “continuity of care” means allowing individual psychiatrists to follow their own patients on a continuous basis. See, e.g., 10/15/19 RT at 191:1-8. and working diligently in a dysfunctional system. See, e.g., 10/15/19 RT at 189:20-190:7, 195:15-16 (Leidner “took [his] marching orders from Dr. Ceballos and [his managers].”) While he should have been more aware of the results of his actions at the time, and in particular the change in the coding to effect the 30 to 45 day change in intervals between EOP appointments, he was not the key decision-maker on the issues now before the court. The court is not persuaded that Dr. Leidner’s appearance as a witness in proceedings in this case in 2013, see Reporter’s Transcript of Proceedings (12/4/13 RT), ECF No. 5013, at 2582-2643, undermines the credibility of his testimony during his second appearance at the October 2019 evidentiary hearing. Cf. 10/22/19 RT at 402:25-403:17. During the time relevant to these proceedings, Dr. Leidner worked at CDCR Headquarters. 10/15/19 RT at, e.g., 207:18-19. Relatively recently, he opted to return to a former position he held at California Men’s Colony. Id. at 207:18-21. It is evident that Dr. Leidner’s colleagues and managers at CDCR Headquarters view his departure as a significant loss. Their assessment makes sense to the court. In a properly designed system, with proper supervision, it appears Dr. Leidner’s significant skills could continue to play an important role both in these remedial proceedings and at CDCR Headquarters more generally. With these observations, the court turns to its broader findings and conclusions. A. Fraud on the Court As the court has reviewed in earlier orders, the current proceedings are grounded in this court’s “authority . . . [and] duty, to protect the integrity of the judicial process,” including the court’s “‘power to conduct an independent investigation in order to determine whether [the court] has been the victim of fraud.’” ECF No. 6002 at 4 (internal citations omitted); see also ECF No. 5786 at 2. The court has identified the standards applicable to a determination of whether there has been fraud on the court, as follows: “In determining whether fraud constitutes fraud on the court, the relevant inquiry is not whether fraudulent conduct ‘prejudiced the opposing party,’ but whether it ‘“harm[ed]” the integrity of the judicial process.’” United States v. Estate of Stonehill, 660 F.3d 415, 555 (9th Cir. 2011) (internal citations omitted). “Most fraud on the court cases involve a scheme by one party to hide a key fact from the court and the opposing party.” Id. Fraud on the court is shown only “by clear and convincing evidence” that a party tried “to prevent the judicial process from functioning ‘in the usual manner’”; it requires a showing of “more than perjury or nondisclosure of evidence, unless that perjury or nondisclosure was so fundamental that it undermined the workings of the adversary process itself.” Id. at 445. ECF No. 6002 at 4-5. The standard for a finding of fraud on the court is a high one and, as the court found from the bench, is one not met here. In particular, the court does not find the kind of “‘scheme . . . to hide a key fact from the court and the opposing party,’” with an emphasis on the word “scheme,” that is at the core of fraud on the court. B. Knowing Presentation of Misleading Information 1. Undisputed that Misleading Information Presented On the other hand, it is clear defendants have presented misleading information to the court. Defendants have admitted that several filings need to be corrected, see, e.g., ECF No. 6242 at, e.g., 5, 8, 9 (citing ECF No. 6226 at 3-4, 12-13, 16), and although they do not specifically concede that the necessary corrections are to misleading data, they have begun the process of correcting the record, which continues. See ECF Nos. 6302, 6330.18 As set forth in section IIC, supra, the neutral expert pointed to evidence suggesting defendants had presented misleading information to the court or the Special Master in several areas. The neutral expert’s key suggestions concerning the defendants’ presentation of misleading data are fully borne out, and then some, by the more developed record now before the court. Beyond the three issues set for hearing, discussed below, the record also shows, as the neutral expert’s report suggested, that defendants’ reporting on “Timely Psychiatry Contacts” resulted in reporting of misleading data by overstating Program Guide timeline requirements in two ways: (a) by counting all “non-confidential psychiatry contacts entered into EHRS toward Program Guide timeline requirements for EOP and CCCMS psychiatric evaluations” and (b) by

18 As noted, see note 3 supra, the court is in the process of review all of the pleadings relevant to a determination of the adequacy of defendants’ corrections to the record. defaulting appointments in EHRS to “confidential” without sufficient training and oversight to ensure proper use of this mechanism. ECF No. 6147 at 67-68. Defendants do not dispute the underlying facts, arguing instead that the Program Guide is insufficiently clear that these evaluations need to be confidential. See ECF No. 6242 at 7. But the court resolved that question in August 2019, concluding that an April 18, 2007 memorandum attached to the Program Guide “and a plain reading of the Program Guide support the conclusion these psychiatric evaluations must be confidential.” Id. Defendants stipulated that they have provided, to the court and/or the Special Master, data on Timely Psychiatry Contacts based on the interpretation the court found to be erroneous , and the court directed defendants to provide a date certain by which corrected data would be provided. Id. at 8 (citing ECF No. 6226 at 12-13, ¶¶ 9-10). Defendants’ responses to that order, ECF Nos. 6257 and 6330, and plaintiffs’ responses thereto, ECF No. 6301and 6360, are under submission. The neutral expert also pointed to evidence that defendants submitted misleading data on the timeliness of mental health referrals, “because for medication noncompliant patients it only counted those patients for whom a psychiatrist ordered a medication noncompliance counseling appointment as a matter of discretion,” not all patients who require medication noncompliance appointments. ECF No. 6147 at 84. As a result, the defendants’ performance indicator overstated compliance. Id. at 91. Based on representations of the parties, the court in its August 14, 2019 order found this issue attributable at least in part to a dispute in the “interpretation of how and why medication non-compliant patients are scheduled for follow-up under the CCHCS Medication Adherence Procedure policy” and, perhaps, to a dispute over “whether all medication non-compliance in fact must be captured.” ECF No. 6242 at 11. The parties’ stipulation and for approval of a memorandum clarifying this policy is pending before the court. See ECF No. 6393. The ultimate question for the court to decide, as it does below, is whether defendants’ presentation of misleading information was knowing and if so why. ///// ///// 2. Presentation of Misleading Information Was Knowing Plaintiffs urge the court to find defendants knowingly presented misleading information to the court and the Special Master. See, e.g., 10/22/19 RT at 390:21-391:6. Defendants respond by saying the court took no action on any misleading data, suggesting that it is only when a court takes action that reliance on misleading data creates a problem. Id. at 427:23-428:3. Given defendants’ argument with respect to the 2018 staffing proposal, it must be stressed there was no court action on that proposal because the then-impending agreement between the parties failed to materialize after Drs. Golding and Gonzalez issued their reports. Thus, the absence of court action on the proposal is because the whistleblower reports blocked the proposal’s presentation to the court in the first place, and not because defendants recognized that aspects of the proposal were based on flawed data and took action to correct the flaws. More broadly, as explained below, the weight of the evidence and the reasonable inferences to be drawn from the totality of the record before the court fully supports the finding that as to the first and third issues covered at hearing, defendants have engaged in knowing presentation of misleading information to the court and to the Special Master. With respect to the second issue, the record shows that the descriptor for “Appointments Seen As Scheduled” indicator did not accurately reflect the components of the indicator. While the court does not find defendants knowingly presented erroneous data created using this indicator, the failure to include an accurate descriptor should have been caught before any data was generated as part of a properly functioning quality management system. As noted above, there were four other areas in which the neutral expert identified evidence suggesting misleading information had been referred to the court and/or the Special Master. See Section IIC supra. In its August 14, 2019 order, ECF No. 6242, the court determined those four issues did not require an evidentiary hearing. As to the second and third items, defendants’ reporting of “Timely Psychiatry Contacts,” which overstates compliance with the timeliness requirements for psychiatry evaluations at the EOP and CCCMS levels of care, the court found a material dispute centered on conflicting interpretations of the Program Guide and whether psychiatry contacts must be confidential. See ECF No. 6242 at 6-7. The court resolved that issue, rejecting defendants’ interpretation of the Program Guide and concluding “that psychiatric evaluations must be ‘confidential’ to satisfy the 30 and 90 day Program Guide requirements.” Id. at 7. The court concluded it was “not necessary to examine the intent behind defendants’ erroneous interpretation” and has referred this matter to the All-Parties Workgroup for development of protocols consistent with the court’s clarifications. Id. at 8.As to the sixth and seventh items, involving submission of misleading data related to timeliness of referrals for medication non-compliant patients, the court determined no evidentiary hearing was required because it was not disputed that a software bug caused one of the errors and the other arose from a dispute over “interpretation of how and why medication non-compliant patients are scheduled for follow-up under the CCHCS Medication Adherence Procedure policy.” Id. at 11. That matter was referred to the All-Parties Workgroup, id., and a stipulation and proposed order with a proposed clarifying memorandum has been submitted to the court for review. See ECF No. 6393. Regarding the four matters before the court, without a full understanding of why defendants knowingly submitted misleading information to the court and Special Master, a proper solution cannot be identified. Therefore, the court has undertaken its own effort to determine why the misleading data was presented, as an essential step on the path to identifying appropriate remedies for this serious roadblock impeding meaningful progress toward full global remediation. a. Changing of EOP Business Rule from 30 to 45 Days i. Background The remedial plan for this action, the Mental Health Services Delivery System Program Guide (Program Guide) requires that a psychiatrist “evaluate each EOP inmate-patient at least monthly to address psychiatric medication issues.” Program Guide, ECF No. 5864-1 at 58. This requirement has been part of the remedial plan at least since 2006, when the court gave final approval to all but a limited number of provisions of the Program Guide that remained in dispute and then ordered the Guide’s immediate implementation. ECF No. 1753-4 at 8; ECF No. 1773. As the neutral expert reported, “[i]t is undisputed that the Program Guide does not define ‘monthly.’” ECF No. 6147 at 44. However, the Special Master19 informed the neutral expert that he “has consistently interpreted ‘monthly’ to be ‘30 days,’ since he began monitoring EOP routine appointments in the late 1990s . . .” ECF No. 6147 at 43. He provided the neutral expert with record support for his position, see id., which he has consistently articulated with the court. The neutral expert also noted documentary evidence showing that prior to the business rule change at issue defendants had also interpreted “monthly” in this context to mean 30 days. See, e.g., id. at 45 (“documents show that CDCR used 30 days to measure compliance with the Program Guide requirement that EOP patients have ‘monthly’ psychiatric evaluations. In 2016, for example, in connection with the production of data for the Special Master for the 27th Round of monitoring, CDCR used a definition of ‘every 30 calendar days after previous psychiatry contact’ to measure compliance with the ‘monthly’ requirement for EOP psychiatric appointments. ECF No. 6012-2 at 154 (Ceballos Decl. at Ex. 3).”). “A business rule is, at the most basic level, a specific directive that constrains or defines a business activity. . . Business rules can be applied to computing systems and are designed to help an organization achieve its goals. Software is used to automate business rules using business logic.” https://www.techopedia.com/definition/28018/business-rule. As relevant here, CDCR applies business rules to its computing systems to manage and report data. See, e.g., ECF No. 6012-3 at 2-4 (Leidner Decl.); 10/15/19 RT at 86:21-23 (testimony of Katherine Tebrock). As Dr. Leidner explains, “[f]rom 2010 to 2017, CDCR collected data regarding mental health treatment through the Mental Health Tracking System (MHTS), a web-based tool developed to track clinical contacts, referrals, and other data related to the provision of mental health services.” ECF No. 6012-3 at 2. Former Deputy Director Tebrock testified that MHTS “had been negotiated and discussed with the Special Master and his team over the course of many years” and that CDCR “had preserved or attempted to preserve, to the extent possible, all of those rules to maintain fidelity.” 10/15/19 RT at 87:3-7.

19 The Special Master has served in that capacity since November 1, 2007 and served as Deputy Special Master from February 15, 1996 until November 1, 2007. ECF Nos. 664, 2453. “Beginning in 2016, CDCR implemented an electronic health record called the Electronic Health Record System (EHRS) which replaced MHTS.” ECF No. 6012-3 at 2. Data collected in the EHRS is stored in the Health Care Data Warehouse, “a group of high-capacity computer servers into which data from numerous sources, including EHRS, CDCR’s Strategic Offender Management System (SOMS), and the Department of State Hospitals, are routed and stored. Historical data, such as MHTS data, are also stored in the Data Warehouse.” Id. at 2-3. Data in the system are used, among other things, to generate management reports, which “are designed to help staff provide timely patient care, manage resources, and give feedback on compliance with guidelines, including various requirements imposed by orders in this case.” Id. at 3. One such report is the Mental Health Performance Report, “which displays a set of indicators measuring performance in numerous areas, including compliance with mental health treatment required by the Program Guide, by health care regulations, and by CDCR health care policies and procedures.” Id. at 3. This report is comprised of “over 150 active indicators summarizing about 4,000,000 individual measurements each month. The logic underlying the indicators relies on over 230 business rules that stipulate a required action, the population it applies to, what triggers that requirement, and how much time is allowed to complete that requirement. All of these parameters are determined by the managers of the Statewide Mental Health Program.” Id. at 4. ii. Review of Evidence Until December 2016, the relevant business rule for the Program Guide requirement that a psychiatrist “evaluate each EOP inmate-patient at least monthly to address psychiatric medication issues” used a period of 30 days to measure compliance. See, e.g., 10/15/09 RT at 195:5-12 (testimony of David Leidner); 10/16/19 RT at 250:14-16 (testimony of Laura Ceballos); Neutral Expert Report, ECF No. 6147 at 45 (documents cited therein). Certain persons working within CDCR raised questions about changes to this business rule, initially in late 2015. The Neutral Expert Report points to evidence suggesting two psychiatrists who were working at CHCF,20 Drs. Jahangiri and Anand, raised the issue, which was then presented to CDCR headquarters in an email from Julie Kirkman, “a Medication Court Administrator and Pre- Release Coordinator at CHCF.” Id. At hearing, Dr. Leidner testified Dr. Jahangiri first raised the issue with him at a weekly state-wide webinar Dr. Leidner conducted in November 2015. 10/15/19 RT at 189:10-17. Dr. Leidner told Dr. Jahangiri it was “technically possible” to change the business rule and told him to raise the issue with Dr. Golding. Id. at 189:24-190:4. The neutral expert found evidence that Dr. Anand emailed Dr. Golding in February 2016, requesting to change the business rule from “30 days” to “monthly” “so as to help ease psychiatrists’ tracking issues, reduce staffing needs, help psychiatrists manage their own outpatient caseloads, and other issues.” ECF No. 6147 at 45-46. No action was taken on the request at that time, id. at 46, and it is undisputed that Dr. Golding did not approve the request. In December 2016, Ms. Kirkman raised the request again in a phone call with Dr. Leidner. 10/15/19 RT at 191:1-13. The same day, Dr. Leidner discussed the request in a telephone call with Dr. Ceballos. Id. at 192:6-193:5. During the call, Dr. Ceballos approved the request and told Dr. Leidner to make the change. Id. at 193:22-194:4. Dr. Ceballos also did not consult with Dr. Golding about the renewed request before she approved it. She testified that she “forgot” to consult him about the change even though she sent him an email on December 5, 2016 about other EOP-related matters. 10/16/19 RT at 283:15-284:5. Dr. Ceballos also did not discuss the request with then-Deputy Director Tebrock before approving it or, apparently, at any time. See 10/15/19 RT at 86:3-20. As noted above, Dr. Ceballos did not report the change to the Special Master or the court because she did not view it as “significant.” 10/16/19 RT at 249:19- 250:11. The Neutral Expert Report points to evidence that “[a]t least some of the psychiatry team was notified of the rule change in January 2017. See CDCR0016721-22. An email discussing the modified rule was forwarded to Dr. Golding by psychiatrists Dr. Mann and 20 CHCF is a medical and mental health care facility for inmate-patients in CDCR. It “provides medical care and mental health treatment to inmates who have the most severe and long term needs.” https://www.cdcr.ca.gov/facility-locator/chcf/. Dr. Lindgren, but it is unclear whether Dr. Golding read it. CDCR0016719-21.” ECF No. 6147 at 46. In his whistleblower report. Dr. Golding averred he first became aware of the change in March or April 2017, when the “headquarters psychiatry team noticed that all the 20 psychiatrists across the institutions seemed to be doing better in terms of seeing EOP patients in a more timely way.” ECF No. 5988-1 at 23. At hearing, he testified he was certain the business rule change would affect data provided to the court “because the dashboard for EOP had turned green, and people were contacting us. Psychiatrists were contacting us across the state and laughing about how much easier it was to be compliant at the EOP level of care, . . .” 10/15/19 RT at 44:12-45:6; see also id. at 76:6-11 (Golding testimony that he learned about the business rule change because Dr. Gonzalez told him “that the dashboard was looking greener than you expected it to be. . . .”). Dr. Golding first reported problems with the rule change to Drs. Ceballos and Leidner, who agreed to change it back. See id. at 16:20-21, 195:13-19. Specifically, Dr. Golding raised the issue with them on or about March 21, 2017, id. at 16:20-21, and Dr. Leidner changed the rule back on or about April 14, 2017, id. at 195:19. In her whistleblower complaint, Dr. Gonzalez avers that in December 2016, when she was working as a staff telepsychiatrist at CDCR headquarters, she noticed that her EOP patients “were suddenly due for follow-up appointments every 45 days or by the end of the next calendar month (whichever was sooner), rather than every 30 days.” ECF No. 6363 at 6. When she learned there had been no policy change regarding the required frequency of EOP psychiatry appointments, she reported the change in the Current Due Dates report to Dr. Golding and he “asked [her] to look into it further.” Id. She discovered that the change could actually permit a gap of 8 weeks between appointments to be reported as 75 percent compliant. Id. She reported her findings to Dr. Golding, who “followed up” with quality management staff regarding the rule change, which was “reverted back to 30 days shortly thereafter.” Id. At his deposition, Dr. Kuich testified that during the time frame the 30-day rule was changed “[t]here was a tremendous focus on EOP” to see if individuals at this level of care were properly placed. Kuich Dep. at 91:17-93:8. He testified that the push “was a systemwide push, not from psychiatry but from the system, to be able to look at EOP patients specifically and to find out” whether their needs were being met or could “be met at a lower level of care.” Id. at 91:21-24. He also noticed a “significant improvement” in compliance in this area, which he “liken[ed] to an improvement he noticed during a quality management meeting having to do with, . . . , use of nonformulary medications, that suddenly everyone who was noncompliant became compliant, and we couldn’t quite figure out what happened. And tracing it back, they had changed the formula for that, and that resulted in a better value, . . .” Id. at 93:5-13. Dr. Kuich believed that the change in the business rule from 30 to 45 days was not consistent with Program Guide requirements. Id. at 94:12-24. There is no dispute that defendants “submitted data using the modified [business] rule to the [c]ourt in” at least two court filings. See ECF No. 6242 at 5 (citing ECF No. 6226 at 3- 4, Issue B(A)(4), (11), (12)); see also ECF No. 6147 at 48 (citing ECF No. 5591 at 14 (Defendants’ Response to the Special Master’s Report on the Status of Mental Health Staffing and the Implementation of Defendants’ Staffing Plan); ECF No. 5591-2 at 4, 9 (Tebrock Decl. ¶ 10, Ex. 2); ECF No. 5601 at 8-9 (Defendants’ Reply to Plaintiffs’ Objections and Request for Additional Relief)). The Neutral Expert Report cites evidence that “CDCR also reported compliance figures from the “Timely Psychiatry Contacts” indicator during this time frame on at least one ASU EOP HUB certification. This report was not filed with the Court, but was submitted to the Special Master. See PLTF005299 (RJD).” ECF No. 6147 at 44. As noted, Dr. Golding made his request that the rule be changed back to 30 days on or about March 21, 2017. The Neutral Expert Report cites evidence that before the rule was changed back on April 14, 2017, “on March 28, 2017, Deputy Tebrock sent an email noting that the Governor’s office had asked ‘to explain in more detail what metrics can be used to show that the care by psychiatry is adequate.’ CDCR0016999,” and that “[o]n March 30, 2017, CDCR filed Defendants’ Response to the Special Master’s Report on the Status of Mental Health Staffing and the Implementation of Defendants’ Staffing Plan, ECF No. 5591, relying upon data under the 45- day rule. ECF No. 6012 at 13.” ECF No. 6147 at 48. ///// ///// iii. Findings The change in the business rule from 30 to 45 days was a material change that was inconsistent with implementation of the relevant Program Guide requirement as established through more than a decade of practice. The change should have been thoroughly vetted before it was implemented; thorough vetting would have included, at a minimum, consultation with Deputy Director Tebrock and Dr. Golding and reporting to the Special Master in advance of the change. Defendants’ argument that the change was prompted by a request from the field to improve continuity of care does not go far enough to explain why it was accomplished without the proper vetting. The request does appear to have come from a medication administrator at CHCF, and while implemented it apparently provided some relief to psychiatrists in the field by, for example, allowing them to go on vacation and still see the same patients in what Dr. Ceballos saw as a reasonable time frame. But the first time the request was made, in 2015, it went nowhere because Dr. Golding did not approve it. The second time the request was made, during a critical juncture when the Special Master was preparing a court-ordered “stand-alone report on the status of mental health staffing and implementation of defendants’ staffing plan,” see ECF No. 5564 at 6-7 (citing ECF No. 5477 at 8-9), no one checked with psychiatry and the change sailed through. Drs. Golding, Gonzalez and Kuich all learned of the change only after relevant “dashboards” designed to measure compliance with remedial requirements of this court changed from red to green overnight and then undertook an investigation to try to understand the reasons behind the change, which was not at all transparent. Given its timing, the rule change affected a period when the Special Master had been tasked with receiving monthly updates on the status of defendants’ implementation of their January 10, 2017 updated staffing plan and filing a stand- alone report on the status of mental health staffing and defendants’ implementation of that plan. See, e.g., ECF Nos. 5477, 5564. The context supports the clear inference of willful blindness at least, or reckless indifference to defendants’ obligations in this action. And, in specific answer to the questions posed for hearing on this issue in the August 14, 2019 order, this willful blindness or reckless indifference is attributable to Dr. Ceballos, the individual who authorized the change in December 2016. To a lesser extent, it is also attributable to systemic failures of CDCR management to ensure that all staff, including those tasked with developing key software codes, have a complete understanding of the remedial requirements of this action so that they can, to the extent possible, make sure the data reports produced to support defendants’ provision of constitutionally adequate mental health care will accurately measure the remedial requirements for this action. As noted above, defendants have undertaken to correct at least some of the misleading information in the record uncovered as a result of the Golding Report. See, e.g., ECF Nos. 6302, 6330. But they have undertaken that effort only after the court set this matter for evidentiary hearing and directed the parties to meet and confer in an effort to narrow the scope of issues for hearing. As noted, the court is currently in the process of reviewing the parties’ filings to determine whether in fact the record has been completely corrected and will address that question in a subsequent order. b. “Appointments Seen As Scheduled” Indicator i. Background The Plata Receiver initially developed the “Appointments Seen As Scheduled” indicator, 10/15/19 RT at 197:21-24, apparently for use as a metric to assess efficiency in scheduling and completing medical appointments. Annette Lambert, Deputy Director of Quality Management, Informatics and Improvement for California Correctional Health Care Services (CCCHS), run by the Receiver, told the neutral expert the “Appointments Seen as Scheduled” indicator was developed independently from the Program Guide by the medical unit, and adopted by Mental Health around 2016. She stated, “[F]rom the medical perspective we introduced seen as scheduled as an efficiency metric. And what we were primarily looking at is how much are we seeing cancellations of clinics based on factors that arguably are under our control.” Lambert Tr. at 82:16-20. ECF No. 6147 at 73. Similarly, Dr. Ceballos told the neutral expert that CDCR Mental Health updated its “Appointments Seen as Scheduled” indicator sometime in 2016 to match the CCHCS Health Care Dashboard indicator, and thereby include only those appointments that were missed due to a factor within CDCR’s control. Ceballos Tr. at 147:8-16, 148:9-10, 148:22-149:3. Id. ii. Review of Evidence At hearing, Dr. Leidner testified that “[a]t some point” he “was asked to replicate that” indicator on a performance report for mental health appointments. 10/15/19 RT at 197:25- 198:2. Witnesses told both the neutral expert and this court that the indicator was not directly connected to any Program Guide requirement or Coleman court order. See id. at 198:18-25 (Leidner testimony); see also ECF No. 6147 at 73 (“CDCR witnesses generally reported that the “Appointments Seen as Scheduled” . . . indicator[] [was] developed for internal use only for measuring the performance of the institutions. . . .”). Dr. Leidner testified that the criteria used in creating the indicator were the Plata Receiver’s criteria developed for medical appointments, and that while he had to recode the indicator to make it applicable to mental health appointments he did not make any substantive changes and tried “to just replicate their [the Plata] methodology.” 10/15/19 RT at 198:5-17. Dr. Rekart testified that after the Golding Report came out he heard the description of the indicator had not been updated to reflect certain changes that had been made to the indicator itself. 10/15/19 RT at 166:16-167:6. Dr. Leidner testified to the same understanding. Id. at 201:14-16. He also testified he thought “perhaps” the description matched the indicator initially, but he “failed to update” the description of the indicator when the Receiver’s staff changed some criteria in the indicator sometime before February 2016. Id. at 202:5-10, 208:7-10. Dr. Leidner testified that he learned of the error from Dr. Ceballos “sometime around October 9th, before [he] had any knowledge of the Golding report . . . and [he] basically changed it that day to match what it was really doing.” 10/15/19 RT at 224:20- 225:11. “At some point during the neutral expert’s investigation,” after the neutral expert’s investigation was over and before Dr. Leidner left CDCR headquarters he reported it to Dr. Ceballos and also discussed it during a telephone conference call. Id. at 206:13-207:9. Dr. Leidner left CDCR headquarters at the end of July 2019 and has not been involved in correcting the descriptor to match the indicator. Id. at 207:15-19.21 Most testimony suggested that, substantively, the indicator was developed as a means of determining how many appointments were cancelled for reasons within CDCR’s control. The day after Dr. Leidner’s testimony, in lieu of recalling him to the stand, defense counsel made an offer of proof in lieu of further testimony from Dr. Leidner, explaining that Dr. Leidner had, subsequent to testifying, clarified that the indicator is a ratio where the denominator is all mental health appointments that were either seen or canceled and not rescheduled for one of the four controllable cancellation reasons: Technical difficulties, modified program, lack of transport or provider unavailable. The numerator is all appointments from the denominator that were seen. 10/16/19 RT at 247:20-248:1.22 Despite this stated purpose, and although there seems to be general agreement that the “Appointments Seen As Scheduled” indicator was not directly related to any specific provision of the Program Guide, Dr. Golding testified he thinks “many aspects” of the Program Guide make measuring appointments seen as scheduled quite relevant. 10/15/19 RT at 68:9-69:7. It is undisputed that data generated using this indicator were provided to the court and to the Special Master. The neutral expert identified these specific ways in which these data were transmitted:  Defendants’ May 17, 2018 Staffing Proposal, ECF No. 5841- 2 at 4 n.5 (stating “[a]ppointments occurred as scheduled 98% to 100% of the time” over the prior 12 months to support

21 Dr. Leidner also testified that “at some point during . . . the neutral expert’s investigation” he learned that the criteria for the Plata Receiver’s Appointments Seen As Scheduled indicator and the criteria for the mental health Appointments Seen As Scheduled indicator did not match, even though the original instruction to him had been to replicate the Plata indicator for mental health. 10/15/19 RT at 205:13-206:20. He did not know “when they diverged . . . [or] if they ever exactly matched, but the intent was that they were supposed to.” Id. at 206:17-20. After the neutral expert’s investigation was over and before leaving CDCR headquarters he reported this to Dr. Ceballos and also discussed it during a telephone conference call. Id. at 206:13-207:9. Here as well, he does not know what has happened to the indicator since. Id. at 207:15-19. 22 Plaintiffs accepted this offer of proof as a correction of the record. Id. at 248:4-7. the assertion that “CDCR is meeting the needs of class members in the desert institutions”). • CQI data provided to the Special Master. See, e.g., PLTF000894, “CEN Mental Health Performance Report for 4/1/16 to 10/24/16” (reporting 100% of “Appointments Seen as Scheduled”); PLTF000896, “LAC Mental Health Performance Report for 3/1/16 to 9/26/16” (reporting 91% of “Appointments Seen as Scheduled”). • At least one CQI Report to the Special Master. See CDCR0019053, Regional Continuous Quality Improvement Review for RJD, October 10-14, 2016 at 12 (“ML EOP had 94% of appointments seen as scheduled” and “ML CCCMS had 94% (n=16,333) of their appointments seen as scheduled”). ECF No. 6147 at 73. iii. Findings Even accepting that the “Appointments Seen As Scheduled” indicator was created originally for “internal” measurement purposes and is not required by the Program Guide, Dr. Golding is correct that the Special Master relies on information generated using the indicator in conducting his monitoring. And even if the court credits Dr. Ceballos’ testimony that Dr. Golding does not understand the data, see 10/16/19 RT at 294:19-295:25, that testimony only highlights a more significant issue: defendants have not adopted processes necessary to make their data methodology fully transparent and understandable to all key stakeholders, including the chief psychiatrist. See also 10/15/19 RT at 36:1-21 (Golding testimony concerning psychiatrists’ lack of access to data for independent evaluation and that psychiatry leadership “can’t organize and program data to look across institutions to be able to see whether there are errors being made in medical care.”). As for the flawed descriptor, the error appears to have been unintentional, and Dr. Leidner brought it to the attention of Dr. Ceballos and others soon after he became aware of it. 10/15/19 RT at 207:2-5 At the same time, because the descriptor is the public-facing information about the code, it is critical to transparency that the descriptor be accurate. This is the type of error that a system needs to catch quickly, if not avoid altogether. Thus, Dr. Leidner’s testimony that he does not know whether the descriptor has been corrected since he left CDCR headquarters at the end of July 2019, id. at 207:13-19, is troubling. Moreover, defendants are not correct in believing that missed appointments are not relevant to patient care. Dr. Kuich’s deposition testimony is very instructive as to why properly tracking missed appointments is relevant. Such tracking contributes to qualitative analysis, including identification of trends and patterns at local institutions that is not available otherwise, absent a very detailed manual tracking. The court credits Dr. Toche’s testimony concerning remedial measures to be explored, which sounds promising assuming there is transparency and full communication with the Special Master and the court going forward. The court will address this issue further, as appropriate in its forthcoming remedial order. Finally, one of the specific questions posed in the court’s August 14, 2019 order with respect to this issue is how the Appointments Seen As Scheduled indicator was developed incorrectly and in the absence of consultation with Dr. Golding or other quality control measures, and what steps defendants plan to take to ensure indicators and definitions are developed with appropriate consultation and quality control in the future. The answer is found largely in one of the most significant issues surfaced through the hearing: the extent to which court-ordered coordination between this action and the Plata action, as well as defendants’ obligation to work with the Special Master in this case, appear to have gone off track. As the discussion in Section IV(C)(2) demonstrates, the boundaries between Plata and Coleman appear to have blurred in key respects. A return to robust and transparent coordination between the efforts in this action to remediate constitutionally inadequate mental health care and the efforts in Plata to remedy constitutionally inadequate medical care must be a key focus and priority going forward. c. Supervisors Acting As Line Staff i. Background In 2009, the court ordered defendants to “take all steps necessary to resolve all outstanding [mental health] staffing allocation issues” and “[t]o that end, . . . complete a staffing plan by the end of August 2009.” ECF No. 3613 at 2. After receiving an extension of time, defendants filed the required plan on September 30, 2009. ECF No. 3693. Defendants’ 2009 Staffing Plan is the controlling plan prescribing necessary levels of mental health care staffing in CDCR’s Mental Health Care Delivery System (MHSDS). “The plan provides staffing ratios for the programs at each level of defendants’ MHSDS and other ancillary programs. See ECF No. 3693 at 12-33. These ratios are expressed as one mental health staff person per x number of inmate patients. See id.” ECF No. 5711 at 3. The 2009 Staffing Plan contains ratios for staff psychiatrists at each level of the MHSDS, with the exception of intermediate and acute levels of inpatient care. See id. at 17-18 (citing ECF No. 3693 at 12-24). The 2009 Staffing Plan also provides ratios for supervising senior psychiatrists only in mental health crisis bed (MHCB) units and in mental health outpatient housing units (MH-OHUs). Id. at 18. Those ratios are extremely high relative to those for staff psychiatrists in the same units: in MHCB units, the ratio of staff psychiatrists to patients is 2.5 to 25, while the ratio of supervising senior psychiatrists to patients is 1:50, and in MH-OHUs, the ratio of staff psychiatrists to patients is 1:9 and the ratio for supervising senior psychiatrists to patients is 1:150. Id. at 18. The job description of Senior Psychiatrist, Supervisor in the 2009 Staffing Plan is as follows: Approximately one third of CDCR prisons will have a Senior Psychiatrist, Supervisor. Senior Psychiatrist, Supervisor positions will be allocated to prisons that do not have a Chief Psychiatrist but have a significant number of staff psychiatrists providing services to a largely stable inmate-patient population. Senior Psychiatrist, Supervisor positions will also be allocated to several prisons that have a Chief Psychiatrist and that have large and complex mental health services. Senior Psychiatrist, Supervisors will provide clinical supervision of staff psychiatrists, as well as various administrative tasks related to formulary, medication management, and continuity of psychiatric medications. Senior Psychiatrist, Supervisors are responsible for the generation and periodic reviews of LOPs pertaining to the practice of psychiatry and for ensuring that practices are consistent with the most recent departmental policies. Further, Senior Psychiatrist, Supervisors share responsibility for quality improvement activities including peer/professional review processes. ECF No. 3693 at 32. ii. Review of Evidence The neutral expert provided evidence that supervising psychiatrists see inmate patients for clinical appointments, that some level of participation by supervising psychiatrists in clinical patient care is appropriate, that the EHRS does not track the extent to which supervising psychiatrists see patients, and that “it does not appear that CDCR had ready access to a data set based on supervisor-only appointments.” See ECF No. 6147 at 80-82 and evidence cited therein. These facts were all confirmed in testimony at hearing. There is no dispute that CDCR supervising psychiatrists see inmate patients for clinical appointments, or that some level of participation in clinical care by supervising psychiatrists is appropriate. See, e.g., 10/15/19 RT at 74:9-14; Kuich Dep. at 206:21-207:5. At hearing, Angela Ponciano, Associate Director for CDCR’s Statewide Mental Health Program, testified that CDCR has not “tracked specifically” “how the number of patients seen by psychiatric supervisors providing direct patient care affects staffing ratios,” 10/15/19 RT at 134:19-22, that the EHRS does not include a performance report “that allows accurate reporting” of when supervisors act as line staff, 10/15/19 at 147:15-19, and that because of this the analysis of supervising psychiatrist patient contacts she conducted after the Golding Report came out was a multi-step process. Id. at 147:17-25. Dr. Golding contends this information can be tracked; at hearing he described two different methods for tracking the extent to which clinical services are provided by supervising psychiatrists. 10/15/19 RT at 50:3-51:21. Both methods involved running a caseload report using supervising psychiatrists’ names and generating a list of patients; these reports were then analyzed under two different methodologies, one controlling for caseload ratios at each level of care and the other by comparing the frequency of supervising psychiatrists’ patient visits with those of line psychiatrists. Id. Dr. Golding also testified that at least since 2018 his staff has called each institution monthly to determine the level of psychiatric coverage at the institution and “whether supervisors are providing coverage.” 10/15/19 RT at 48:17-20. He further testified that “[on numerous occasions” the reports generated by these calls “have been in the hands of Deputy Tebrock, Ms. Ponciano and Ms. Brizendine” and that “[at] one point they tried to get” Dr. Golding and his team to stop this monthly tracking, but the team continued to conduct the tracking. 10/15/19 RT at 48:17-24. ///// ///// In his report, Dr. Golding avers that [t]he staffing ratios CDCR reported to the court in the 2018 staffing report are incorrect. Sixty percent of psychiatric supervisors were seeing patients like line staff at least part time, and in some cases full time. The work was being done by a larger ratio of psychiatrists to patient that was reported, suggesting that fewer psychiatrists are needed per patient than is in fact the case. ECF No. 5988-1 at 5. At hearing, he testified that psychiatric supervisors spend about fifty percent of their time conducting line visits. 10/15/19 RT at 51:9-21. At his deposition, Dr. Kuich estimated that psychiatric supervisors have performed line duties more than a third but less than fifty percent of the time. Kuich Dep. at 30:14-31:1. After Dr. Golding issued his report, Ms. Ponciano conducted an analysis “on the number of CCCMS and EOP psychiatry appointments involving supervisors and chiefs at each institution.” ECF No. 6147 at 80; see also 10/15/19 RT at 135:1-15. She testified she “found that there were not 60 percent of supervisors carrying a full line staff caseload for those levels of care.” 10/15/19 RT at 136:3-5. She later performed a second analysis, which showed that if supervisors’ clinical contacts were removed from the “frequency of contacts” report on which defendants’ 2018 staffing proposal was based, the frequency of psychiatrist contacts for both CCCMS and EOP patients decreased. 10/15/19 RT at 138:13-139:6; see also ECF No. 6242 at 10 (citing ECF No. 6226 at 20 ¶ 6). The neutral expert conducted interviews in the field with psychiatrists who report variation in responsibilities from institution to institution, supporting the inference that Ms. Ponciano’s analysis underrepresents the actual amount of time psychiatrist supervisors spend performing the duties of line psychiatrists: Psychiatrists agreed that this issue varied significantly by institution, but all acknowledged that it was not uncommon for psychiatry supervisors to see patients. One psychiatrist noted that at some institutions, a single psychiatrist (sometimes a supervisor or chief) handles all IDTT appointments. Another psychiatrist noted that as a supervisor, she was assigned the case load of three line staff. One Chief Psychiatrist described that [s]he provides a lot of the care, but could not provide a specific volume. [S]he commented that if [s]he did not provide direct care, the institution would be out of compliance. Another Chief Psychiatrist reported that [s]he did not routinely see patients or have a set patient load. A senior supervising psychiatrist said it was expected that supervisors perform the same duties as line staff when there are staffing shortages—the culture of leadership was that psychiatrists should be utilized. That psychiatrist explained that when [s]he took on the senior supervisor position [s]he was doing line staff work at least 50% of the time, and [s]he currently still covers IDTTs and other line work when other psychiatrists are not available. The anecdotal evidence from multiple psychiatrists suggests that Ms. Ponciano’s data analysis undercounts the amount of patient care being provided by supervisors. It is clear, however, that the degree to which supervisors provide direct care varies widely by institution and over time, and so we were unable to more precisely quantify this activity. ECF No. 6147 at 81-82. As the court discussed in its August 14, 2019 order, the parties have stipulated “that the data on timely psychiatric contacts prepared in support of CDCR’s 2018 Staffing Proposal to reduce the number of psychiatrists was ‘potentially misleading’ because the proposal did not disclose that appointments with psychiatric supervisors were included in the data.” ECF No. 6242 at 10. The neutral expert also suggested other documents contain representations about staffing without accounting for “contributions from psychiatrist supervisors, including: • Monthly reports on staff psychiatrist vacancy rates. See, e.g., PLTF005201, PLTF005207. • Defendants’ Response to the Special Master’s Report. ECF No. 5591 at 14 (“Over the past year, inmates were seen timely . . . by their psychiatrist ninety percent of the time.”) (emphasis added), 15 (reporting 74% average fill rate for psychiatrists). • 27th Round Monitoring Data, Tab B: Staffing. See ECF No. 6012-2 at 69. • Other representations CDCR made to the Special Master and the Court regarding the adequacy of their current staff psychiatry staffing levels. See, e.g., Joint Status Report RE: October 11, 2018 Status Conference (Sept. 15, 2018), ECF No. 5922 at 4 (“CDCR expects that implementation of the proposed staffing plan will immediately lead to CDCR being at or above the staffing levels required in the Court’s June 13,

2002 Order (ECF No. 1383), and therefore immediately bring CDCR into compliance with the October 10, 2017 Order [ECF No. 5711 at 30].”).” ECF No. 6147 at 79-80. Again, the court is in the process of reviewing all relevant parts of the record to determine the scope and adequacy of necessary corrections. iii. Findings Taken together, all of the information in the record makes clear that psychiatry supervisors carry a significantly higher caseload than is contemplated by the 2009 Staffing Plan governing the remedy in this action. Defendants contend that the first analysis performed by Ms. Ponciano was not intended to show how much supervisor time was required, but, instead, to determine how much clinical time was necessary in the field, 10/22/19 RT at 422:14-25, but this argument misses the relevant point. Ms. Ponciano’s first analysis, conducted to support the ill- fated 2018 staffing proposal, masked the time supervisors spend providing line care and yielded numbers that overstated appointment timeliness compliance. At hearing, she testified to a second analysis she conducted subsequently, which showed that if supervisor contacts were removed from timely CCCMS contacts, the report would have showed CCCMS patients being seen on average .98 times every 90 days, rather than 1.07 times every 90 days. 10/15/19 RT at 137:21- 139:6. The latter analysis was not shared with the court or the Special Master before these proceedings. Ms. Ponciano also testified that discussions with the Special Master about CDCR’s use of psychiatric supervisors to perform line psychiatrist clinical duties have not begun but are planned for the “near future.” 10/15/09 RT at 139:14-23. Dr. Golding tried, apparently unsuccessfully, to surface the issue of supervisors serving as line staff in CDCR meetings he attended as a representative for psychiatry. See, e.g., 10/15/19 RT at 54:5-20. While Dr. Golding and Dr. Kuich disagree on the exact percentage of time supervisors have spent providing line care, with Dr. Golding estimating fifty percent and Dr. Kuich estimating from thirty-three to fifty percent, the difference is not material here. Regardless of which range the court credits, supervisors have clinical responsibilities far in excess of those contemplated in the 2009 Staffing Plan and defendants misleadingly withheld this information in an effort to make compliance numbers look better and to support significant reductions in psychiatrist staffing levels. Here the specific questions posed for hearing on this issue are why defendants did not disclose in their 2018 Staffing Proposal whether, and to what extent, the reporting of data related to average frequency of patient contacts did not disclose the use of supervisory psychiatrists to complete caseload contacts with patients; and to what extent defendants knowingly relied on active participation of supervisory psychiatrists in performing the duties of line psychiatrists, both in defendants’ 2018 Staffing Proposal and in supporting their representation that, if adopted, the 2018 Staffing Proposal would bring defendants into compliance with the October 2017 staffing order? The answers are found in the actions of critical players including Ms. Ponciano, in an environment of pervasive bureaucratic dysfunction as the hearing here revealed. Defendants failed either to consult key headquarters psychiatrists or to heed the clearly relevant information those psychiatrists attempted to provide concerning the overuse of supervising psychiatrists in performing clinical duties. Defendants thus failed to acknowledge the many negative consequences of the failure to either understand the extent of this overuse or its ramifications for the field. Dr. Golding testified that at one point he was even instructed to stop collecting institutional data on the extent to which supervising psychiatrists were performing line duties. This instruction appears to have come at a time in 2018, when defendants were under a court order to come into compliance with their Staffing Plan and the court-ordered vacancy rate and when they were engaged in their misguided attempt to artificially reduce the number of psychiatrists required to deliver constitutionally adequate mental health care to the plaintiff class. 3. Overall Summary All of the foregoing compels the conclusion that two of the issues discussed above, the change in the EOP timeline business rule and the failure to properly quantify and identify the extent to which supervising psychiatrists perform the duties of line staff psychiatrists, involved the knowing presentation of misleading information to the court. With respect to the Appointments Seen As Scheduled indicator, the court finds no knowing presentation of misleading information but, instead, a public-facing error in the descriptor that should have been caught sooner and that caused misunderstanding of the nature of the information presented. Taken together with defendants’ admissions prior to hearing, defendants have knowingly presented misleading information to the court in numerous areas critical to the remedy in this case and measuring compliance with that remedy. C. Other Concerns Identified Through Hearing 1. Marginalization of Psychiatry On a broader level, the record created through the evidentiary hearing demonstrates a marginalization of psychiatry that impedes defendants’ ability to achieve full compliance with the constitutional requirements embodied in the court-approved remedy. Dr. Kuich’s testimony explains the pressures and disincentives created by reliance on automation and electronic data: Psychiatrists are being made to practice in an environment that, among other things, “causes data to have to be massaged in certain ways to allow information to be more presentable to say we don’t need psychiatrists so we can get out of the lawsuit.” Kuich Dep. at 162:22-25. “And the more you automate this process to make sure that compliance happens, the more you take control out of the clinician to be able to determine what’s clinically relevant for the patient.” Id. at 167:14-18. In critical policy decisions affecting this case, psychiatry’s ability to provide substantive input also was severely constrained. As noted above, in preparing the 2018 staffing proposal, Deputy Director Tebrock declined to provide a copy of the proposal to Dr. Golding before it was finalized, and instead only read bullet points to him and then asked him to sign off. This process for incorporating the views of a psychiatric professional, given the stakes, is completely inadequate. Moreover, in addition to not being fully involved in discussions and development of the 2018 staffing proposal, Dr. Golding testified he was told to wait to talk about his staffing concerns until after that proposal was to have been filed with the court. 10/15/19 RT at 27:2-29:22. Non-psychiatrist members of the CDCR Mental Health management team also asked Dr. Golding and his team to stop their monthly tracking of staffing at various institutions, at a time when the team was tracking information CDCR officially said could not be tracked. Dr. Kuich’s access to data compiled by Ms. Ponciano and her team was extremely limited. See Kuich Dep. at 38:24-39:11. And Ms. Ponciano pulled Dr. Kuich out of one meeting for “about five minutes” to run some numbers by him for providing on-call services through telepsychiatry without giving him context, full information or time to evaluate the information she was putting together to support the staffing proposal. Id. at 65:3-21; 68:21-70:6. This exchange too is not the kind of meaningful discussion required under the circumstances, even if it enabled Ms. Ponciano to say honestly she had talked with D. Kuich. While Ms. Ponciano came across as credible in her testimony, she is an administrator who oversees operations and labor negotiations, who was tasked with coming up with the number of psychiatrists to hire and to cut. She did not have the full knowledge base to develop proposals on her own that would satisfy the requirements of this case. The headquarters environment described by Dr. Golding, led by former Deputy Director Tebrock, is also concerning. Dr. Golding had the impression, which he testified to credibly, that he was not able to speak to the Special Master. 10/15/19 RT at 21:14-23. Dr. Golding testified he now understands he can contact the Special Master directly as necessary. See 10/15/19 RT at 21:24-22:1. But his reasons for submitting his report to the Receiver instead of the Special Master were clearly articulated, consistent with his sense that he was not supposed to communicate with the Special Master directly; they also reflected his concern about reporting issues up the Coleman chain of command. 10/15/19 RT at 40:14-41:3. Dr. Kuich described multiple instances where he just gave up trying to report problems or make necessary changes, because his voice was never heard. See, Kuich Dep. at, e.g., 163:14-25. The actions reviewed above run counter to a key principle articulated in this case since the very beginning: “In order to provide inmates with access to constitutionally adequate mental health care, defendants must employ mental health staff in ‘sufficient numbers to identify and treat, in an individualized manner, those treatable inmates suffering from serious mental disorders.’” Coleman v. Wilson, 912 F.Supp. at 1306 (internal citation omitted). Psychiatrists are critical to appropriate mental health staffing, given that they are medical doctors bound by the Hippocratic Oath See Kuich Dep. at 33:8-9 (“Psychiatrists as physicians do have the Hippocratic Oath to do the best we can for our patients.”). This does not mean psychiatrists must always prevail in internal policy- and decision-making processes. But they must be meaningfully consulted; their professional views must be heard, considered and accounted for. Defendants’ marginalization of psychiatry and their clumsiness in the process reflects a significant lack of good judgment and bureaucratic dysfunction that, if allowed to continue, presents a major obstacle to successful remediation in this action. 2. Boundaries Between Plata and Coleman While there are areas of overlap between the Plata class action and this one, the two cases are distinct and it is clear defendants were not policing the boundaries between the two to protect and advance the remedies required in this case. For example, the Appointments Seen as Scheduled indicator, developed by the Plata Receiver, was simply adopted by CDCR Mental Health staff without any tailoring, as Dr. Leidner testified. More generally, the Receiver’s team is making changes to healthcare business rules that affect mental healthcare indicators, and it is the Receiver’s team running validation processes to check that code. See, e.g., 10/16/19 RT at 269:17-270:2. It appears defendants have tasked no one with a systemic review of changes to business rules to ensure compatibility with this Coleman case. Regardless of how the current practices have developed, all practices and procedures related to Coleman data collection and reporting must be made fully transparent immediately. Additionally, defendants’ Mental Health quality management team was heavily involved in the Plata Receiver’s development of the EHRS. As the record developed through the evidentiary hearing disclosed, many of psychiatry’s requests for a solution to critical scheduling problems linked to diagnosis and prescriptions have not been incorporated into EHRS. See Kuich Dep. at 174:15-175:7 Psychiatry’s requests for changes to EHRS “languished and were not addressed.” Id. at 146:7-10; see also id. at, e.g., 147:17-148:1. More broadly, EHRS is not tailored, as far as the court can tell, to take account of specific mental health issues implicated by this case. See, e.g., Golding Report, ECF No. 5988-1 at 71-75. A person entering data using EHRS can self-select his or her own title, see 10/15/19 RT at 148:10-20, a feature developed, according to Ms. Ponciano, to allow psychiatrists to change their title to supervisor in order to prescribe nonformulary medication. Id. at 161:15-19. It also appears that the training of psychiatrists, who defendants rely on for data entry, regarding the use of EHRS has been uneven and incomplete. See Kuich Dep. at 48:1-4 (regular training was “made available” but not mandatory); id. at 47:17-22 (training manual was “bare bones”, with refinements not transmitted to line staff level); id. at 23:17-20 (supervising psychiatrists’ acting as line staff impaired the needed “unobstructed time and undivided attention to train their psychiatrists in the ways of EHRS.”). Defendants acknowledge a need for more training; they must act on this acknowledgment and promptly initiate a robust training process to address all the deficiencies identified through the hearing. On a parallel track, this court will closely manage renewed coordination effort involving the Special Master in this case and the Plata Receiver, with the presiding judges at the table as appropriate. In the meantime, as the court cautioned from the bench, no one should rush into the breach to cement any new plan for improved data collection analysis and reporting, without obtaining this court’s advance approval. D. Reasons Defendants Knowingly Presented Misleading Information In the final analysis, inexplicably, it is apparent defendants lost complete sight of the reasons remediation is required here. Defendants adopted a laser focus in an effort to obtain termination of court supervision, which lead to a stark “ends justify the means” approach. Their litigation tactics have wholly missed the significance of the constitutional rights of the thousands of mentally ill persons defendants have in their custody. As the court said in its oral pronouncement following hearing, legal cases are not just words on a piece of paper and a series of jousting matches. Almost all legal cases have hearts and souls, as does this one in particular. This court’s predecessor, Judge Karlton, put his heart and his soul into this case, as reflected in the care he paid and his orders which stand today. This is a case that cries out for every single player to consult their hearts daily and keep their eyes hourly on those souls who are the members of the plaintiff class: the seriously mentally ill individuals housed behind bars in this state who have the absolute, undeniable right to constitutionally adequate treatment and care. The question of how and why defendants lost sight of this case’s touchstone is at once complicated and straightforward. Timing played a role, with awareness of the schedule of the Special Master’s monitoring rounds. The end of the prior Governor’s term does appear to have contributed to a pressure cooker environment. While no evidence has emerged of anyone in the Governor’s Office ever instructing any player to mislead the court, the Neutral Expert Report does capture a telling interaction. In March 2017, Ms. Tebrock sent an email noting that the Governor’s Office had asked for an explanation in more detail of what metrics could be used to show that the care by psychiatry is adequate. See ECF No. 6147 (citing CDCR0016999). This missive signaled an attention to data and a focus on compiling data to show that care was adequate, with the court as the audience. In the same general time frame, Ms. Tebrock’s explanation of the need for lawyers to wordsmith the court documents, as a reason for not showing the staffing proposal to Dr. Golding, also exposes that “ends justifying the means” approach, as opposed to one of engaging in responsible problem solving. The push to get dashboards from red to green is another marker. As Dr. Kuich, as someone who has left the department, explains: “[M]ental health felt that they were performing very well in many areas, that they could police themselves with data, that they were a structure. That they were sustainable. And the only piece that was the problem was that there weren't enough psychiatrists. And so if there was some way to show that with fewer psychiatrists we were meeting the metrics, that final block would tumble, and there would be no basis for the lawsuit.” Kuich Dep. at 122:15-23. In approving the change from 30 to 45 days, Dr. Ceballos facilitated dashboards turning to green overnight. A dashboard that’s green, makes it look as if a remedy is complete. In sum, litigation once again has trumped substantive compliance, a path defendants have taken repeatedly. It is not for this court to tell a party how to litigate its case, if it chooses litigation, believes it has that right and plays within bounds. The litigation efforts in this case, however, have exacted a steep, steep price at the expense of the plaintiff class. Despite defendants’ knowing presentation of misleading information to the court and the Special Master, and their having lost sight of the remedial purposes of this action, there are some hopeful signs. Dr. Toche has signaled that she has concrete plans going forward. She appears to have an ability to listen and to hear what others are saying. The court expects that she is thinking deeply about a proper response to plaintiffs’ counsel’s question about how she can work to make clear to those who work for and with her that “CDCR has been found to have violated the Constitution as to the mental health program and is under a remedial order supervised by this court” and that “Coleman is not just a word” but signifies a federal court order, “upheld by the United States Supreme Court[,] that governs a remedial process.” 10/16/19 RT at 359:12- 360:4. The court will direct that Dr. Toche provide a report on her answer to that question at its next status conference with the parties in early 2020. There also is a relatively new administration and with any new administration comes the chance to turn over a new leaf. The Deputy Legal Affairs Secretary for Criminal Justice in Governor Newsom’s office, Kelli Evans, has been present in the courtroom for the evidentiary proceedings. She also has attended settlement discussions convened by another judge of this court. Ms. Evans and her boss have an opportunity to step into the breach, to take the lessons from what has occurred and move forward in a way that can bring this case to a proper conclusion, if defendants can learn the lessons of their past mistakes, internalize the reasons behind those mistakes and identify meaningful solutions. The court will play its part by convening regular status conferences, resolving disputes as necessary and guiding the Special Master as appropriate. These proceedings have made clear the need for appropriate remedies. As noted above, defendants have been given an opportunity to fully cleanse and purge the record of misleading information. To some extent, they have availed themselves of that opportunity, and the court acknowledges those efforts. See, e.g., ECF Nos. 6302, 6330. Defendants have not, however, come forward to provide their own cogent, believable, supported and complete explanation as to why they presented misleading information. Rather, they have offered only partial excuses that do not fully acknowledge the totality of the record, pointing to inadvertent errors, absence of course correctors and vagueness in the Program Guide. See, e.g., 10/22/19 RT at 420:10-423:8. Moreover, this is the second time in less than ten years that defendants have embarked on a litigation strategy that delayed and frustrated compliance with staffing requirements, see, e.g., Coleman v. Brown, 938 F.Supp.2d at 984-989 (discussing ongoing significant staffing vacancies in order denying defendants’ motion to terminate this action), giving the court great pause. Against that backdrop, in the absence of defendants’ full acceptance of responsibility, the court has reached its own conclusions and any corresponding remedies must address the court’s findings. The parties are in apparent general agreement on the need for data certification. As required by the court’s bench order, they have now presented their individual views on proper approaches to such certification and that matter is submitted. See ECF Nos. 6383, 6384. Prior to hearing, plaintiffs identified a series of remedies, and they have now filed a brief setting forth their proposed remedies post-hearing, see ECF No. 6374, to which defendants have responded, ECF No. 6388. The court is prepared to seriously consider plaintiffs’ proposed remedies because fundamentally they are correct that this is the time to effect a sea change. This court must ensure no court is called upon again in the future to consider whether and how misleading information has been presented to it. The remediation called for by these proceedings will allow a long-delayed return to the big picture and the proper laser focus on quality of care for California’s seriously mentally ill prison inmates. Nothing has prevented work on that overarching goal during these proceedings, and it must be abundantly clear that focus on the quality of care for this class of prisoners is what will guide defendants’ true relief from court oversight. With respect to staffing in particular, ten years ago it was defendants themselves who submitted a staffing plan to this court followed by a budget change proposal to the California Legislature to “‘fully implement’” the staffing model described in that plan. Coleman v. Brown, 938 F.Supp.2d at 984 (quoting Ex. K to Kahn Decl., ECF No. 4325, at 93). The budget change proposal described the critical flaws in defendants’ prior staffing model and represented that the 2009 staffing plan identifies appropriate staffing levels to meet constitutional standards. Still today, however, psychiatrist staffing vacancies hover at the 30 percent mark. The court has heard many times the explanation of supply and demand, that there is insufficient supply, given the remote locations where psychiatrists are needed, to meet the needs of the plaintiff class. But these hearings have provided additional explanations and identified other contributors to the challenge in identifying psychiatrists, including an uninviting dysfunctional workplace that does not value the essential treatment perspectives that psychiatrists have to offer and creates an atmosphere where morale is low. While a change from 30 to 45 days, as one example, might provide a modicum of relief to an insufficient number of overburdened psychiatrists, here it was a misguided, unthinking fix, applying a very tiny bandage to a festering wound while the infection spreads throughout the body. Defendants simply must come to terms with the substance of the staffing plan and involve all key stakeholders in working with the proper focus to satisfy it. If, after addressing the problems these hearings have exposed, defendants honestly believe that their staffing plan, embodied in court orders, needs to be modified, they have the option, as they always have had, of seeking a modification from the court. Any such request would need to be properly justified and honestly supported, of course. In any event, defendants must acknowledge and account for the substantial findings in this court’s October 11, 2017 order, describing the heavy burden that must be met to support any increase in psychiatrists’ caseloads. In closing, the court repeats its observation from the bench, that nothing prevents the defendants coming forward with a more transformational option. In 2014, Judge Karlton observed, “California is not alone in ‘criminalizing mental illness,’” adopting the perspective of the sheriff of Cook County, Illinois quoted in a published article. Coleman v. Brown, 28 F.Supp.3d at 1073 n.5. “‘We’ve systematically shut down all of the mental health facilities, so the mentally ill have nowhere else to go. [The prison system has] become the de facto mental health hospital.’” Id. (internal citation omitted). Since the time this court assumed responsibility for this case, the mental health prison population numbers have risen. See ECF No. 5213 (August 29, 2014 Order reassigning case to undersigned). Given its experience with the case so for, the court agrees “that many of the problems giving rise to this suit and ongoing efforts at remediation arise from the inevitable tensions created by the distinct needs of custody supervision and the distinct need for mental healthcare.” Coleman v. Brown, 28 F.Supp.3d at 1073 n.5. If there is a transformational and pragmatic alternative to the prison as de facto mental health hospital, as a way to address a root contributor to the constitutional violation in this case, this court would entertain such a proposal. Unless or until such a constructive reform is possible, the staffing remedy in this case calls for defendants’ staffing plan in the context of the Program Guide to chart the way forward and be put front and center. Relatedly, Coleman data collection and reporting must be fixed, and it must be fixed to serve the policies and orders in this case, not the other way around. The policies and orders must not be drained of meaning in an effort to squeeze a square peg into a round hole. And the data must be fixed with all the key stakeholders at the table. It must be, as Dr. Toche appears to recognize, checked and double checked. All with an eye toward allowing the defendants ultimately, when they truly can, to accurately demonstrate to the court that the Constitution is finally satisfied. A remedial order will issue in the near future. A hearing will be set if the court needs to hear more from the parties before that order issues. DATED: December 17, 2019.