Opinion
No. 3:20-cv-00569 (MPS)
01-19-2021
JAMES WHITTED, individually, and on behalf of all others similarly situated, Petitioner, v. DIANE EASTER, Warden of Federal Correctional Institution at Danbury in her official capacity, Respondent.
ORDER OF SPECIFIC PERFORMANCE OF CLASS SETTLEMENT AGREEMENT
In September 2020, I approved a settlement of this class action brought by inmates of Federal Correctional Institution Danbury ("FCI Danbury") whose medical conditions make them vulnerable to serious illness from COVID-19. Under the settlement, the Warden of FCI Danbury and the U.S. Bureau of Prisons (the "BOP") agree to review class members to determine whether to transfer them to home confinement due to the risks associated with COVID-19 using a process outlined in the Settlement Agreement. The Settlement Agreement provides that I retain jurisdiction to resolve disputes that arise between the parties by issuing orders of specific performance. This ruling resolves a series of disputes that have arisen concerning the standards the BOP is using to review inmates for home confinement, the adequacy of the explanations it is providing to support decisions to deny transfers to home confinement, and other matters. In particular, the class has filed a motion for enforcement of the settlement agreement that asks me, among other things, to bar the BOP from relying on certain factors that the class contends are not relevant to the issues of inmate and public safety on which the BOP should be focusing in the home confinement review process mandated by the Settlement Agreement. ECF No. 296. These factors include things like non-violent disciplinary infractions and the percentage of time an inmate has served. After considering the parties' memoranda and associated exhibits, and the January 8 oral argument, I grant in part and deny in part the class's motion for the reasons and to the extent set forth below.
I assume familiarity with the factual and procedural background of this case as set forth in the Ruling on Motion for Temporary Restraining Order and Motion to Dismiss, ECF No. 30, the Settlement Agreement, ECF No. 238-1, the first Order of Specific Performance and associated memorandum, ECF Nos. 289, 304, and the submissions relating to the class's second motion to enforce the Settlement Agreement, ECF Nos. 296-305, 307-08, 310-13, 319-23. I also incorporate by reference the legal standards set forth in the memorandum supporting the first Order of Specific Performance. ECF No. 304 at 3 n.1 (jurisdiction); id. at 13-15 (settlement agreements, specific performance). I address the specific issues raised in the class's motion to enforce the Settlement Agreement below.
I. DISCUSSION
A. The BOP's Home Confinement Decisions Often Do Not Reflect Application of the Correct Standard
The class seeks a series of orders about the factors the BOP may rely on in making home confinement decisions, among which are orders that would direct the BOP to "[c]onsider all CDC [r]isk factors as valid risk factors that render a home confinement applicant medically vulnerable . . . and [afford] all CDC [r]isk factors 'substantial weight' in any home confinement decision"; "[e]xplicitly consider an individual's age as a CDC risk factor"; "[n]ot rely in home confinement decisions on a 'Low' PATTERN Score as a basis to deny home confinement to class members"; "[n]ot rely in home confinement decisions on factors that are only tangentially related to public safety, including percentage of time served or recent disciplinary history that is both non-violent and non-sexual". ECF No. 296 at 33-34. More generally, the class argues that the "statement of reasons on the BOP's worksheets [generated during the home confinement review process] often fail to evidence that the BOP has engaged in any balancing between public safety and inmate safety . . . ." Id. at 1. The class has submitted a large volume of worksheets reflecting recent denials of home confinement and generated in the home confinement review process contemplated by the Settlement Agreement. By contrast, the Warden and BOP (collectively, the "BOP") assert that the Settlement Agreement does not limit the BOP's discretion as to either its substantive judgment in individual home confinement decisions or its judgment as to which considerations bear on public safety or an inmate's suitability for home confinement. ECF No. 310 at 1. As to several of these issues, I agree with the class that the BOP's decisions as reflected in the worksheets do not comply with the standards required by the Settlement Agreement. I begin with a general discussion of the standards for the home confinement review process required by the Settlement Agreement and my role in enforcing those standards, and then proceed to the specific issues in dispute.
As discussed below, PATTERN is a risk assessment tool used by the BOP to estimate the risk of recidivism.
1. Standards for Home Confinement Review and Judicial Enforcement Contemplated by the Settlement Agreement
The Settlement Agreement makes clear in several places that "the provisions of the May 12, 2020 TRO" ("TRO") govern the BOP's home confinement reviews. See ECF No. 238-1 ¶ 3 (requiring all List Two inmates to be reviewed for home confinement "pursuant to" the Attorney General's memoranda, current BOP guidance, "and the standards set forth in the May 12, 2020 TRO", but stating that "to the extent" the latter standards differ from the Attorney General's memos or the current BOP guidance, "the provisions of the May 12, 2020 TRO shall control"; and "BOP agrees to have [its Home Confinement Committee ("HCC")]] make a home confinement determination based upon the standards set forth in the May 12, 2020 TRO."); id. ¶ 6 (same for non-list inmates who are added to class). The May 12, 2020 TRO was issued as preliminary relief in a habeas petition brought by the class claiming, among other things, that the Warden at FCI Danbury was failing to protect inmates from the risk of severe illness or death posed by COVID-19 in violation of the Eighth Amendment, in particular, by failing to use her expanded authority under the March 2020 CARES Act to place inmates on home confinement.
The TRO requires the Warden at FCI Danbury to "assign[] substantial weight in [the home confinement review] to the inmate's risk factors for COVD-19 based on CDC guidance", and to eliminate "all requirements" that (1) "the inmate have served some portion of his or her sentence . . .", (2) "a 'primary or prior offense' not be a violent offense . . .", and (3) "the inmate be 'without incident reports in the past 12 months (regardless of severity level)'". ECF No. 30 at 71. A footnote elaborates on the elimination of these "requirements," stating "I do not mean to suggest that these considerations are irrelevant to the Warden's assessment of the appropriateness of home confinement for a particular inmate[,]" and noting, as an example, that an inmate's violent history "is obviously relevant to the assessment of the danger his or her release might pose to the public and thus should be accorded weight in the determination." Id. n.33. The footnote then states: "The purpose of eliminating these and other categorical exclusions is to the make the process more individualized and more amenable to a proper and reasonable balancing of inmate safety and public safety." Id. (emphasis added).
The concept of balancing public safety against inmate safety in the home confinement review process is a recurring theme in the opinion accompanying the TRO. The introduction to the opinion states: "I grant in part the inmates' motion for a temporary restraining order and issue an order that requires the Warden . . . to adopt a process that is . . . more clearly focused on the critical issues of inmate and public safety than the current process." ECF No. 30 at 2; see also id. at 57 ("Petitioners have made a preliminary showing that, in these circumstances, the Eighth Amendment requires that the [home confinement determination] be made promptly and that it focus on the critical issues of inmate and public safety."); id. at 68 ("The temporary restraining order that follows is aimed at . . . focusing that process [for evaluating inmates for home confinement] on achieving a 'reasonable' balance between the risks to inmate safety and the risks to public safety.").
It is thus clear that, in adopting the standards of the May 12, 2020 TRO, the Settlement Agreement requires the BOP to balance inmate safety against public safety and, in that balancing process, to "assign substantial weight" to the inmate's COVID-19 risk factors when making decisions on whether to transfer an inmate to home confinement. Contrary to the BOP's argument, this requirement is not a mandate (or a license) for the BOP to consider the "totality of the circumstances" - a term that appears nowhere in the TRO, the underlying opinion, or the Settlement Agreement. Instead, it directs the BOP to weigh certain factors - those related to inmate safety and public safety - and not others, and directs it to accord a specified albeit imprecise quantity of weight - "substantial weight" - to a set of inmate-safety-related factors, i.e., COVID-19 risk factors.
The term does appear in former Attorney General Barr's March 26 memorandum. See ECF No. 281 at 33 (directing that "home confinement [should be granted] only when BOP has determined—based on the totality of the circumstances—that transfer to home confinement is likely not to increase the inmate's risk of contracting COVID-19."). But as noted, the Settlement Agreement requires the BOP to follow the TRO's standards to the extent there is a conflict with the Attorney General's March 26 memorandum.
The BOP argues that it has broad discretion to define what "public safety" means in this context and can thus rely on a wide range of considerations in making home confinement decisions. It points out that the Settlement Agreement provides that the BOP's "substantive determination of whether to place any particular member of the Medically Vulnerable Class on home confinement shall not be subject to judicial review, except that the Court may order the BOP to have [it] reconsider a home confinement denial that is based upon erroneous factual underpinnings . . . or a failure to comply with the terms of the Agreement, and may compel specific performance of the Agreement." ECF No. 134-1 at 17. The Settlement Agreement's assignment to the BOP of the task of reading the scale, however, is not a grant of authority to determine the ingredients of the items to be weighed. In other words, "public safety" does not mean whatever the BOP says it means. To the contrary, as used in the TRO, the underlying opinion, the Settlement Agreement, and former Attorney General Barr's memoranda, "public safety" has a specific meaning, i.e., an immediate danger to the public from the release of an inmate into the community.
Neither is 18 U.S.C. § 3624, the statute that governs the BOP's exercise of its authority to transfer inmates to home confinement during the final portion of their sentence. After the CARES Act expanded that authority to allow the BOP to make such transfers during any portion of an inmate's sentence, I found, on a preliminary basis in the TRO ruling, that the BOP was not exercising that authority in compliance with the Eighth Amendment. Thereafter, the parties entered into the Settlement Agreement, which both released the BOP from my preliminary finding and bound it to new, more specific obligations pertaining to home confinement transfers. It is the Settlement Agreement, not Section 3624, that defines the parameters of the home confinement review process for members of the class.
The term "public safety" appears in the TRO in the footnote quoted above. See ECF No. 30 at 71 n.33. The example immediately preceding the sentence in which the term appears discusses an inmate with a "violent history[,]" id.—obviously, an inmate with an active propensity for violence presents an immediate danger to the public if he is released. The underlying opinion confirms that "public safety" is focused on immediate danger to the public. The opinion describes the requirement of serving a specified percentage of a sentence and the presence of an incident report "no matter the seriousness" as "at best, only tangentially related to public safety," ECF No. 30 at 48, because neither of those circumstances presents an immediate threat to the public. The Settlement Agreement uses "public safety" in the same sense of immediate danger: it requires the BOP to "endeavor" to transfer individuals approved for home confinement to home confinement within 14 days of approval "unless public safety or the absence of any home in which to place the inmate would make it unsafe to move the inmate to home confinement within the 14 day period." ECF No. 238-1 ¶ 11. "Public safety" in that context plainly is not referring to things like the portion of the sentence served or minor disciplinary infractions. Finally, the Attorney General's April 3 memorandum uses "public safety" to describe a danger that might be mitigated by the electronic monitoring of an inmate. In that document, the Attorney General authorized the BOP to transfer inmates to home confinement "even if electronic monitoring is not available, so long as BOP determines in every such instances that doing so is appropriate and consistent with our obligation to protect public safety." The memorandum elaborates that protecting public safety means protecting against the "risk[] to the public from released prisoners engaging in additional criminal activity, potentially including violence or heinous sex offenses." In short, the consistent use of the term "public safety" in the Settlement Agreement, the TRO, and related documents shows that it has a fixed, clear meaning that is not amenable to expansion by the BOP.
Other inmates whose profiles would present obvious risks to the public safety include those with recent conduct demonstrating an active propensity for distributing dangerous drugs, illegally possessing firearms, or engaging in sex-related offenses. Another obvious example would be the risk that an inmate infected with COVID-19 might infect members of the public. By listing these examples, I do not intend to be exclusive. But the farther removed an inmate's offense of conviction, criminal or disciplinary history, or other circumstances are from presenting an immediate danger to the public, the more the BOP will need to explain its conclusion that those circumstances are related to public safety.
Federal Bureau of Prisons, Memorandum for Director of Bureau of Prisons from Attorney General William Barr: Increasing Use of Home Confinement at Institutions Most Affected by COVID-19 (April 3, 2020), https://www.bop.gov/coronavirus/docs/bop_memo_home_confinement_april3.pdf.
Id.
Given all this, I conclude that the Settlement Agreement permits a finding by the Court that the BOP is "fail[ing] to comply with the terms of the Agreement" in cases in which the worksheet documenting the home confinement review, when read as a whole, does not make clear that the BOP has made an effort to (1) balance the inmate's safety against public safety, properly understood as meaning immediate danger to the public from release of the inmate, and (2) assign "substantial weight" to the "inmate's risk factors for COVD-19 based on CDC guidance". On the other hand, the Settlement Agreement does not permit such a finding when the Court (or class counsel) disagrees with the product of the BOP's effort, for example, because, in the Court's or class counsel's view, the BOP has struck the balance incorrectly or given too much weight to public-safety-related factors.
I discuss in greater detail below the Settlement Agreement's requirements regarding the length and quality of the explanation.
In its motion to enforce, the class seeks several specific orders from the Court. See ECF No. 296 at 33-35. After oral argument and through the parties' filings, it is apparent, based on agreements reached by the parties, that some of the issues raised are now moot. I address the remaining issues below.
Specifically, the parties agreed: (1) that the Warden and BOP are required to list and consider all of a class member's CDC risk factors on that individual's home confinement review sheet; (2) that the Settlement Agreement requires a re-review for home confinement where an inmate's worksheet does not list one or more that individual's Tier I or Tier II risk factors, except in limited circumstances (i.e., where an error or mistake of fact is not material, see ECF No. 134-1 at 12); (3) that the Warden and BOP are required to re-review inmates who were approved for placement in an RRC but were denied home confinement without any explanation; and (4) to work collaboratively to identify the facts underlying individual home confinement denials, where such facts are not apparent from the face of the worksheet and not otherwise obtainable from the individual or his or her criminal docket. Thus, the requests for relief in sections (a)(iii), (b)(i), and (c) are DENIED as moot. See ECF No. 296 at 33-35. In addition, the Government agreed, "when producing worksheets to class counsel going forward, [that the Government's] counsel will provide the date of the Home Confinement Committee's decision." ECF No. 324 at 2. I therefore also deny as moot the request for relief in section (g) of the supplemental memorandum. See ECF No. 307 at 29.
2. CDC Risk Factors
The parties disagree about whether the BOP is properly considering all CDC risk factors and giving "substantial weight" to these factors as required by the Settlement Agreement. Specifically, the class seeks an order directing the BOP to "[c]onsider all CDC Risk factors as valid risk factors that render a home confinement applicant medically vulnerable, including Tier II conditions, and that all CDC Risk factors are to be afforded 'substantial weight' in any home confinement decision". ECF No. 296 at 33 ¶ (a)(i). "Tier II conditions," or Tier II risk factors, refer to medical conditions that, according to the CDC, might place a person at a higher risk of developing severe illness from COVID-19. By comparison, Tier I risk factors are conditions that, according to the CDC, do place a person a higher risk of developing severe illness as a result of COVID-19.
People with Certain Medical Conditions, CDC, https://www.cdc.gov/coronavirus/2019-ncov/need-extra-precautions/people-with-medical-conditions.html.
The core of this dispute is focused on the following language that appears on many of the worksheets in which the BOP has denied home confinement to an inmate - usually an inmate who has only Tier II risk factors: the inmate "does not have any risk factors that would put him/her at a higher risk of developing severe illness as a result of COVID-19 based on CDC guidelines." On its face, this language appears to contradict other portions of the worksheet listing the CDC risk factors. More generally, it is inconsistent with the idea that all members of the class - and thus all those for whom such worksheets are being generated - are "medically vulnerable" on account of their having one or more CDC risk factors. The BOP argues that, though the above-quoted language is inartful, it is intended to capture the notion that an inmate with only Tier II risk factors does not face a fully proven risk from COVID-19, because Tier II "conditions 'might,' but are not proven to, place an inmate at increased risk of severe COVID-19." ECF No. 310 at 8 (emphasis in original). The BOP further asserts that Tier I and Tier II conditions present different risks according to the CDC and, therefore, should be afforded different weights in individual home confinement decisions. I agree with the BOP that the Settlement Agreement allows it to accord less weight to Tier II factors. " Substantial weight" in this context must be understood to define a spectrum, and Tier I factors, which are supported by more robust scientific evidence of an association with COVID-19, belong at the weightier end of that spectrum. But I agree with the class that the BOP's language stating that an inmate "does not have any risk factors that would put him/her at a higher risk of developing severe illness as a result of COVID-19" is, at best, ambiguous and fails to demonstrate that the BOP is complying with its obligation to assign "substantial weight" to an inmate's COVID-19 risk factors.
See, e.g., ECF No. 298 at 60 (using this phrase to describe O.F., who has two Tier II conditions: 28 percent BMI and hypertension); id. at 64 (same for G.G., who has one potential Tier II condition: asthma); see also People with Certain Medical Conditions, Centers for Disease Control and Prevention ("CDC") (last visited January 12, 2021), available at https://www.cdc.gov/coronavirus/2019-ncov/need-extra-precautions/people-withmedical-conditions.html.
Under the Settlement Agreement, each worksheet must list all an inmate's risk factors, including both Tier I and Tier II factors. ECF No. 134-1 at 11.
The TRO issued before the CDC revised its guidance to separate the conditions associated with severe illness due to Covid-19 into those that "do" create a higher risk of severe illness and those that "might" create such a risk. When the TRO was drafted, all the conditions listed in the class definition were deemed by the CDC to place the individual at higher risk of severe illness from COVID-19.
The Settlement Agreement requires the BOP to "assign[] substantial weight" to the inmate's COVID risk factors, but it qualifies this statement with the phrase "based on CDC guidance." ECF No. 30 at 71. The CDC explains that the distinction between Tier I factors and Tier II factors is based on the strength of the scientific evidence supporting the conclusion that there is an association between each risk factor and severe illness. Tier I conditions are supported by the "strongest and most consistent evidence", which is "[d]efined as consistent evidence from multiple small studies or a strong association from a large study." Tier II conditions are supported by either "mixed evidence"—"[d]efined as multiple studies that reached different conclusions about risk associated with a condition"—or "limited evidence"—"[d]efined as consistent evidence from a small number of studies." Given these gradations in the quality of the evidence supporting particular risk factors and given that the Settlement Agreement requires the BOP to keep the "National Health Technology Administrator—or a substitute medical clinician . . . [as] a member of the [HCC] to assist in the Committee's home confinement reviews," ECF No. 134-1 at 10-11, I find that the Settlement Agreement's command to assign "substantial weight" to an inmate's risk factors based on CDC guidance must be understood to leave some discretion to the BOP in determining each inmate's risk profile. The presence of a medical professional on the HCC means that the BOP is in a better position than the class or the Court to make that determination. As a result, although the BOP must give "substantial weight" to each risk factor, the agency retains some discretion in the precise amount of weight to assign to each factor based on CDC guidance, a specific inmate's medical history, and the input of the medical professional on the HCC. And this discretion may frequently, and properly, result in a finding that an inmate's Tier II conditions are outweighed by public safety considerations, even when the same public safety considerations would not outweigh Tier I medical conditions.
Scientific Evidence for Conditions that Increase Risk of Severe Illness, Centers for Disease Control and Prevention (last visited January 12, 2021), available at https://www.cdc.gov/coronavirus/2019-ncov/need-extra-precautions/evidence-table.html.
Id.
The above-quoted language the BOP is apparently using when an inmate has only Tier II risk factors does not unambiguously show, however, that the BOP is properly acting within its discretion under the Settlement Agreement. The BOP undoubtedly knows that all the inmates it is reviewing suffer from some condition that makes them "medically vulnerable" enough to be in the class, and, as I explain below, the Settlement Agreement does not demand that the BOP explain itself in detailed, precise terms. Nonetheless, saying of every inmate with only Tier II conditions that he or she "does not have risk factors that would put [him or her] at higher risk of developing severe illness as a result of COVID-19" leaves open the possibility that the BOP believes that an inmate with Tier II conditions faces zero increased risk of severe illness from COVID-19 or that the BOP is simply overlooking the facts—the inmate's Tier II conditions. Either possibility would violate the requirement that the BOP assign "substantial weight" to the inmate's COVID-19 risk factors. As a result, because the BOP's "does not have risk factors" language makes it impossible to determine whether it is complying with the requirement to accord substantial weight to an inmate's CDC risk factors, I find that the worksheets using that phrase fail to demonstrate that the BOP has complied with the Settlement Agreement.
Ordinarily, in this situation, I would require the BOP to re-review specific inmates for home confinement, as the Settlement Agreement authorizes me to do. See ECF No. 134-1 at 17. But a court's authority to award specific performance is an equitable power, Restatement (Second) of Contracts, § 358 cmt. a., and I do not believe it would be equitable to issue such sweeping relief here. The BOP has limited resources, and it should not be required to divert its staff to filling out worksheets more carefully just to demonstrate compliance with the Settlement Agreement, even when the underlying litigation is prompted by a public health emergency. It is not at all clear that ordering re-reviews will make a difference for many inmates. For example, for inmates whose conditions barely qualify them for Tier II status (such as someone whose BMI barely exceeds 25) or for inmates who, though they have serious co-morbidities, also have criminal records, offenses of conviction, or disciplinary histories that raise serious public safety concerns, there would be little point in requiring a re-review. As noted, under the Settlement Agreement, as long as it is weighing the correct considerations and assigning some level of "substantial weight" to COVID-19 risk factors, the BOP has plenary discretion to read the scale and determine whether a particular public safety consideration outweighs a particular COVID-19 risk factor. It thus would not be equitable to require a re-review in cases where the BOP violated the "substantial weight" requirement by invoking the formula quoted above but other facts apparent from the face of the worksheet make it likely the violation was harmless. As a result, the parties shall meet and confer to identify specific inmates whose individual cases (1) contain the "does not have risk factors" language, and (2) otherwise are strong candidates for re-review. Then, after the parties have agreed upon a list of inmates whose cases contain this error and who otherwise warrant re-review, the BOP shall re-review that list of inmates for home confinement in accordance with the terms of the Settlement Agreement.
See, e.g., ECF No. 298 at 60 (O.F., who has two Tier II conditions (28 percent BMI and hypertension), was denied home confinement in part because O.F. "does not have any risk factors that would put [him/her] at a higher risk of developing severe illness as a result of COVID-19 based on CDC guidelines." But O.F.'s worksheet also lists "Past Violence- Assault and Battery with a dangerous weapon, 5/23/00", a "Medium Risk of Recidivism-based on Pattern Score", and "no viable release plan . . . ." Id. It is not difficult to imagine that, even on re-review, the BOP would be well within its discretion to deny O.F. home confinement under the terms of the Settlement Agreement, assuming that the worksheet does not contain material factual errors or omissions.).
The class also seeks an order directing the BOP to "[e]xplicitly consider an individual's age as a CDC risk factor to be afforded 'substantial weight' for any class member over the age of 50, in conjunction with all other CDC medical risk factors . . . ." ECF No. 296 at 33 ¶ (a)(ii). But the class makes no allegation that the BOP fails to list an inmate's age on the worksheet, or that the BOP has refused to re-review an inmate for home confinement based on a factual error related to age. Further, the class does not identify anything in the Settlement Agreement, the TRO, the CDC guidelines or elsewhere that supports its cut-off at the age of 50. The CDC views age as a sliding scale of increasing risk from COVID-19 as one gets older, but it does not set clear age cut-offs. Older Adults, Centers for Disease Control and Prevention, https://www.cdc.gov/coronavirus/2019-ncov/need-extra-precautions/older-adults.html. As a result, I find that a worksheet is adequate under the Settlement Agreement as to an inmate's age as long as it correctly lists the inmate's age and otherwise appears to provide substantial weight to that inmate's CDC risk factors.
3. Percentage of Time Served and Disciplinary History
The parties disagree as to whether the BOP may rely on certain factors, including percentage or amount of time served or an inmate's disciplinary history that is both non-violent and non-sexual, in denying a class member home confinement. Thus, the class seeks an order directing the BOP to "[n]ot rely in home confinement decisions on factors that are only tangentially related to public safety, including percentage or amount of time served or recent disciplinary history that is both non-violent and non-sexual[.]" ECF No. 296 at 34 ¶ (a)(vi). I agree with the class that these factors bear no clear or necessary relationship to public safety.
The percentage or amount of time an inmate has served bears no necessary relationship to public safety, properly understood. The BOP argues in its brief that the earlier in an inmate's sentence he or she is released, the greater the "flight" risk he or she presents. But this is speculative; the BOP has presented no data suggesting that any inmates released to home confinement have fled the jurisdiction to which they have been released. In any event, the HCC does not invoke the "flight" argument in any of its references to percentage of time served in the worksheets, and thus the argument appears to be a post-hoc rationalization offered by counsel, rather than the HCC's own rationale for denying transfer to home confinement. Rather, it appears that when the HCC cites percentage or amount of time served in the worksheets, it does so as a measure of the degree to which the portion of the sentence served has reflected the "seriousness of the offense." This is a relevant consideration under the statute governing federal sentencing, 18 U.S.C. § 3553(a), but it is not under the Settlement Agreement, which focuses on balancing public safety and inmate safety. Thus, the BOP may not rely on this consideration in denying transfers to home confinement or, if it does so, it must offer a more detailed explanation demonstrating a connection between percentage of time served and public safety, properly understood.
See, e.g., ECF No. 298 at 20 (M.B., who had served 12.0 percent of the statutory time, was denied home confinement in part because the "current percentage of time served diminishes the seriousness of the offense.").
Disciplinary incidents are more naturally relevant to public safety, especially when they involve violence, threats of violence (including making weapons), or sexual transgressions. In addition, a pattern of breaking even minor or technical rules can demonstrate an inability to conform one's conduct to the law that could pose a risk to the public safety. But it is hard to see how one or two minor and non-violent disciplinary infractions, or even a greater number of very old infractions, could establish such a pattern. So if the BOP is going to rely on disciplinary incidents, the worksheet must make clear why they are related to public safety. In some cases—as when the worksheet indicates that the disciplinary incident involved an assault—this will be obvious. But when it is not obvious, the BOP must provide more explanation to show that it has fulfilled its balancing duties under the Settlement Agreement.
See, e.g., ECF No. 298 at 236-37 (B.W., whose worksheet provided a disciplinary history as follows: "Yes. 06-19-2020[;] Code 316 Being in Unauthorized Area[;] Code 332 Smoking in Unauthorized Area."—absent a more detailed explanation, smoking and being in an unauthorized area once in the past year is not a relevant public safety consideration under the Settlement Agreement).
As a result, I find that the BOP has violated the terms of the Settlement Agreement when it has relied on percentage or amount of time served or a disciplinary history that is non-violent, non-sexual, and not part of a pattern of misconduct demonstrating an inability to follow the law.
Again, although the Settlement Agreement authorizes me to order a comprehensive re-review in these cases, I find that a more focused remedy would be more efficient and practical, and thus more equitable. Thus, I again order the parties to meet and confer to identify specific inmates whose individual home confinement cases (1) were denied at least in part because of the percentage or amount of time served or because of a disciplinary history that is non-violent, non-sexual, and does not shown a clear pattern of misconduct suggesting an inability to follow the law, and (2) who are otherwise strong candidates for re-review, i.e., those whose worksheets do not otherwise raise significant concerns about public safety, properly understood. Then, after the parties have agreed upon a list of inmates whose cases contain one or more of these errors and otherwise merit re-review, the BOP shall re-review that list of inmates for home confinement in accordance with the terms of the Settlement Agreement. It is my hope that the parties will focus the re-review process on the inmates who face the greatest dangers from COVID-19 and are least likely to present a danger to public safety.
4. Low PATTERN Scores
The parties also disagree as to the relevance to public safety of a "Low" score on the BOP's "PATTERN" risk assessment tool, which is aimed at predicting whether an inmate will reoffend based on several characteristics or "risk factors". The tool assigns each inmate to one of four categories for risk of recidivism: minimum, low, medium, or high. A "Low" PATTERN score is supposed to indicate a higher risk of recidivism than a "minimum" score. The class seeks an order directing the BOP to "[n]ot rely in home confinement decisions on a 'Low' PATTERN Score as a basis to deny home confinement to class members[.]" ECF No. 296 at 34 ¶ (a)(v). As the Government points out, however, the risk of recidivism is plainly related to public safety, and the BOP uses an inmate's PATTERN score to estimate that risk. The class has criticized the reliability of the PATTERN tool, and others have done so as well. See, e.g., https://www.propublica.org/article/bill-barr-promised-to-release-prisoners-threatened-by-coronavirus-even-as-the-feds-secretly-made-it-harder-for-them-to-get-out. But while the tool may be imperfect, the BOP is using it to gauge a factor relevant to public safety, i.e., risk of recidivism, and the Settlement Agreement leaves to the BOP the determination of how much weight to accord a particular PATTERN score, as this is part of the substantive determination of whether it is appropriate to transfer an inmate to home confinement. While the BOP may not categorically deny inmates home confinement on the basis of a particular PATTERN score, because doing so would not reflect an individualized balancing of public safety against inmate safety, it is free to rely on an inmate's PATTERN score, including a "Low" score, in its weighing analysis. As a result, I deny the class's request for an order prohibiting the BOP from relying in part on an inmate's Low PATTERN score in denying a class member home confinement.
As discussed above, not all types of recidivism pose identical risks to public safety, properly understood. See note 4, supra.
That said, there may be individual cases where the Settlement Agreement requires more explanation when a denial rests primarily on a Low PATTERN score. Again, the BOP's task is to determine which weighs more heavily in a particular case - the risk to the inmate or the risk to the public safety. To show that it has performed this task, the BOP needs to make some showing on the worksheet that it has decided that one risk outweighs the other. Thus, for example, where a "low PATTERN score" is the only factor being relied upon (or the only factor other than percentage of time served or some other non-public-safety-related consideration), and the inmate has multiple Tier I risk factors, the BOP will likely need to provide more detail about why it has read the scale to tilt against transfer to home confinement. At the end of the day, it is within BOP's discretion to make that determination, but, as discussed below, the Settlement Agreement requires some explanation of the factual basis for its decisions. So the higher the risk to the inmate or the fewer (or lighter) the public safety considerations, the more the BOP will be required to explain why it denied the transfer to home confinement to demonstrate that it complied with its balancing obligation.
See https://www.cdc.gov/media/releases/2020/p0625-update-expands-covid-19.html ("The more underlying medical conditions people have, the higher their risk."); see also, e.g., ECF No. 298 at 47 (K.D., 37 years old, whose worksheet shows one Tier I risk factor (BMI of 44.8 percent) and two Tier II risk factors (hypertension, asthma), no history of violence or sexual offenses, a minimum security level, no history of discipline or escape, and an apparently valid release plan, was denied home confinement "due to her low recidivism risk level, as well as the percentage of time served for the offense."). Based on the worksheet, it is not obvious why these two factors present a danger to public safety that outweighs K.D.'s risk of severe illness from COVID-19, or that the BOP complied with its obligations under the Settlement Agreement to accord substantial weight to an inmate's CDC risk factors and provide a factual basis, relevant to public safety, that outweighs the risk to inmate safety.
In its supplemental memorandum, the class also seeks relief limiting the BOP's consideration of the lack of a viable release plan and certain supervision-related requirements. ECF No. 307 at 29. For the reasons set forth in the Government's brief, ECF No. 324 at 2-7, I deny this request. I encourage the parties to meet and confer to discuss any specific cases related to this request for relief the class believes warrant a reexamination. --------
B. The Settlement Agreement Does Not Require Lengthy "Case Narratives"
The parties also disagree about the quality of the explanations the BOP has provided on the worksheets when denying home confinement. Specifically, the class seeks an order directing the BOP to: (1) reflect its consideration of an inmate's CDC risk factors "in the case narrative or reasons for denial provided by the BOP"; (2) reflect its consideration of an inmate's age as a CDC risk factor "in the case narrative or reasons for denial provided by the BOP"; (3) "[r]eflect in any home confinement denial that 'substantial weight' was actually given to the individual's CDC risk factors and include an explanation of why these risk factors are overcome by substantial evidence of countervailing public safety considerations;" (4) "provide an individualized explanation of the reasons why the public safety concerns specific to the individual class member's circumstances are sufficient to overcome their serious medical vulnerability to COVID-19"; and (5) "[n]ot rely in home confinement decisions on any exclusion previously lifted by this Court's TRO as the sole basis for denying home confinement, such as disciplinary history or offense of conviction, unless the BOP details the nature of the underlying factual circumstances and why they raise substantial evidence of countervailing public safety considerations sufficient to overcome the individual's serious medical vulnerability to COVID-19". See ECF No. 296 at 33-34 ¶¶ (a)(i)-(ii), (a)(iv), (a)(vii). I find that the Settlement Agreement does not, in general, require a lengthy explanation of a home confinement denial as long as the worksheet, taken as a whole, presents a sufficient factual basis to infer that the BOP determined that public safety factors, properly understood and supported by the facts, outweighed the risk of severe illness to the inmate from COVID-19.
The Settlement Agreement states that if the HCC denies home confinement, "it will comply with the May 12, 2020 TRO regarding a written explanation of the factual basis for any factors relied upon for the denial." ECF No. 238-1 ¶ 11. The only mention in the TRO of a written explanation states that the Warden must provide the Court with "an individualized explanation for each denial of home confinement as to [a medically vulnerable inmate], including at least a brief description of the factual basis for any factors deemed to outweigh the danger to the inmate from COVID 19." ECF No. 30 at 73; see also Settlement Agreement ¶ 10, ECF No. 134-1 at 11 ("Respondent agrees that if she determines that [a class member] is not eligible for home confinement, she will comply with the May 12, 2020 TRO and provide a written explanation of the factual basis for any factors relied upon for the denial."). The Settlement Agreement also states that the BOP will provide class counsel with "copies of the documents generated" in the home confinement review process, including the worksheets and "the reason for denial," and that "the form of the documents produced under this Section will be consistent with the worksheets that have been disclosed in this litigation (see, e.g., ECF 59)." Id. ¶ 11. The worksheets "disclosed in this litigation" are one-page documents consisting primarily of a table listing the inmate's COVID risk factors, age, current or past offenses, percentage of time served, pending charges or detainers, security level, recent discipline, PATTERN score, and other related information. Below the table there is a "case narrative," a "home confinement committee review," or both, each consisting of two or three sentences indicating reasons the BOP deemed the individual appropriate or inappropriate for home confinement.
When the Settlement Agreement is read as a whole, I find that each worksheet generated by the BOP to document its decision must be sufficient to identify the facts supporting its conclusion that, in the case of a denial, public safety factors outweighed the danger to the inmate from COVID-19. This is not an exacting standard. It does not require the BOP to explain its decision in anything approaching the level of detail one would expect from a legal opinion. To the contrary, the reference to the form of the documents being consistent with the worksheets disclosed in the litigation makes clear that what is expected is, as described above, a table format listing information in various categories and then an explanation of a couple of sentences at the bottom providing further explanation of the factors the HCC found most weighty. I agree with the Government that the proper way to evaluate whether these worksheets comply with the Settlement Agreement is to read them as a whole and not to interpret the explanation at the bottom as a comprehensive summary of the HCC's reasons for denial. At the same time, to be compliant with the Agreement, there needs to be enough information in the table and the narrative at the bottom, taken together, to enable the reader to glean the facts upon which the BOP relied in deciding that public safety factors outweighed the danger to the inmate. Thus, neither the "case narrative" nor the "home confinement committee review" at the bottom of the worksheet need list every CDC risk factor or every public safety factor, invoke the words "substantial weight" or "outweighs," or incant other similar formulas. The critical point is that either the facts shown in the table or the narrative explanations below the table, or both, must identify the facts weighed on each side of the scale. Again, rather than order a re -review of all the worksheets that do not satisfy this modest standard, I direct the parties to meet and confer and identify the non-compliant worksheets for the strongest candidates for home confinement, based on the public safety/inmate safety standards discussed in this opinion, and arrange for a re-review of the cases of those inmates.
C. The Settlement Agreement Requires the BOP to Review for Home Confinement Class Members Who Are Transferred to Other Prisons.
Lastly, the parties disagree as to whether inmates who qualify as class members but are subsequently transferred to a facility other than FCI Danbury are eligible for home confinement review under the Settlement Agreement. As a result of this disagreement, the class seeks an "[o]rder [directing] the BOP to not refuse to conduct a home confinement review or a re-review for any recognized class member it transfers out of FCI Danbury[.]" ECF No. 296 at 35 ¶ (d). The Government opposes this order because, it contends, the intent of the Settlement Agreement was to provide home confinement only to inmates currently at FCI Danbury. I agree with the class.
The Settlement Agreement defines the class as "consisting of any person incarcerated at FCI Danbury anytime from the Effective Date until the termination date of this Agreement, October 31, 2021 . . ." who is either a List 1 or List 2 inmate, or who possesses one or more Tier 1 or Tier 2 COVID-19 risk factors as determined by the CDC. ECF No. 134-1 at 5-6. This definition plainly applies to any inmate who was at FCI Danbury and who otherwise qualified as a class member, regardless of the length of time the inmate spent at FCI Danbury or whether the inmate remains at FCI Danbury. No other provision in the Settlement Agreement addresses inmates transferred out of FCI Danbury to another prison. As a result, the Settlement Agreement requires the BOP to consider for home confinement inmates transferred out of FCI Danbury to another prison, as long as such inmates otherwise qualified as class members at some point prior to the transfer.
The fact that the Warden of FCI Danbury no longer has authority over inmates transferred to another facility does not change the analysis. The BOP itself is a party to the Settlement Agreement and is bound by its terms. It is not excused from complying with the Agreement simply because a class member has been transferred to a prison that was not the subject of the underlying lawsuit and is outside the District of Connecticut, and the BOP has pointed to nothing that would prevent it from effectuating the transfer to home confinement of such an inmate. It is true, as the BOP noted at oral argument, that the inmate's overall risk profile may have changed due to, for example, a higher or lower infection rate of COVID-19 at the inmate's new facility; but I know of no reason the Home Confinement Committee should not be able to take such changed circumstances into account in making its decisions.
II. CONCLUSION
For the reasons and to the extent stated above, the class's motion to enforce the settlement agreement (ECF No. 296) is GRANTED IN PART and DENIED IN PART, as follows:
(1) Within 14 days of this Order, the parties shall meet and confer to identify specific class members whose cases warrant re-review under the terms of this Order and the Settlement Agreement; and
(2) Within 14 days of such identification, the BOP shall conduct a re-review of inmates identified under paragraph (1) for home confinement under the terms of the Settlement Agreement, except that, if for some good reason the BOP cannot meet that deadline, it shall inform class counsel of that good reason before expiration of the 14 days.
(3) To the extent the parties are unable to agree on the class members whose cases warrant re-review under paragraph (1), they shall utilize the mechanisms specified in the Settlement Agreement for resolving such disagreements.IT IS SO ORDERED.
/s/_________
Michael P. Shea, U.S.D.J. Dated: Hartford, Connecticut
January 19, 2021