United States v. SAFEHOUSE

Opinion

IN THE UNITED STATES DISTRICT COURT FOR THE EASTERN DISTRICT OF PENNSYLVANIA

UNITED STATES OF AMERICA, : Plaintiff, : CIVIL ACTION : No. 19-519 v. : : SAFEHOUSE, a Pennsylvania nonprofit : Corporation; JOSE BENITEZ, as President : and Treasurer of Safehouse, : Defendants. :

SAFEHOUSE, a Pennsylvania nonprofit : Corporation, : Counterclaim Plaintiff, : : v. : : UNITED STATES OF AMERICA, : Counterclaim Defendant, : : and : : U.S. DEPARTMENT OF JUSTICE; : WILLIAM P. BARR, in his official capacity : as Attorney General of the United States; : and WILLIAM M. McSWAIN, in his official : capacity as U.S. Attorney for the Eastern : District of Pennsylvania, : Third-Party Defendants. :

McHUGH, J. June 24, 2020 MEMORANDUM OPINION I. Introduction This is a declaratory judgment action in which I previously concluded that 21 U.S.C. § 856(a) cannot be read to bar supervised injection sites whose ultimate purpose is to reduce fatalities caused by opioid addiction. In accordance with the request of the parties I proceeded to enter a final appealable Order on February 25, 2020. ECF 142. Plaintiff Safehouse then promptly announced its intention to begin operation in South Philadelphia, leading the

Government to file an emergency motion for a stay of the Court’s February Order. Any urgency to address the motion quickly dissipated. Safehouse’s attempt to open without meaningful dialogue with the surrounding neighborhood was met with organized opposition supported by various elected officials, and the COVID-19 pandemic came to overshadow the opioid epidemic, understandably becoming the almost singular focus of local authorities. In view of those events, the Government’s motion to stay might almost be viewed as moot. But as Philadelphia begins to reopen, and with 963 opioid overdose deaths in Philadelphia in 2019, it is conceivable that Safehouse might yet seek to open before review by the Court of Appeals. In issuing a declaratory judgment, I consistently rebuffed the parties’ attempts to

transform an issue of statutory construction into a public policy debate. Ironically, the standard for evaluating a stay is grounded in equitable considerations, such as harm to the parties and “where the public interest lies,” so a broader discussion is necessarily now required. Under the traditional test, especially as to the issue of irreparable harm, the Government’s case for a stay is marginal. But these are not ordinary times. The combination of the pandemic and the momentous protests following the killing of Mr. George Floyd make this the wrong moment for another change in the status quo. Accordingly, the Government’s request for a stay will be granted. II. The Legal Standards Governing Requests to Stay Final Judicial Orders Safehouse raises a threshold issue—whether a declaratory judgment is properly the

subject of a request for a stay. Federal Rule of Civil Procedure 62, titled generally, “Stay of

2 Proceedings to Enforce a Judgment,” addresses money judgments and injunctions, but is silent as to declaratory judgments. Safehouse appears to argue that because my February 25 Order does

not mandate any specific action, there is no execution of the Order that can be stayed. ECF 149, at 3-4. There is little precedent addressing this issue. The Government invokes the All Writs Act, 28 U.S.C. § 1651, but without citation to any decisions that rely upon it as authority to stay a declaratory judgment. The Government also cites precedent holding that a court has inherent authority to stay orders apart from the Civil Rules. ECF 145, at 4-5 (citing United States v. Denver & Rio Grande W. R.R. Co., 223 F.2d 126, 127 (10th Cir. 1955)). The argument advanced by Safehouse is highly abstract. It appears that Safehouse seeks to draw a distinction between the enforcement consequences of a declaratory judgment and its preclusive effect. Intuitively, aside from the gap in Rule 62, it is not clear why a court cannot stay its own

declaratory judgment. The late Judge O’Neill of this Court sensibly held that where a declaratory judgment is requested to be stayed, the court should look to its practical effect. Robertshaw v. Pudles, 2014 WL 1976890, at *2 (E.D. Pa. May 15, 2014). Here, the effect of my ruling was that Safehouse was cleared to open, without fear of prosecution, a facility that the Government insists is illegal. In practical terms, the February 25 Order has significant consequences. I am persuaded that I retain the power to stay the judgment and will apply the traditional standard for considering a stay. The grant or denial of a stay focuses on four factors: “(1) whether the stay applicant has made a strong showing that he is likely to succeed on the merits; (2) whether the applicant will

be irreparably injured absent a stay; (3) whether issuance of the stay will substantially injure the other parties interested in the proceeding; and (4) where the public interest lies.” Nken v. Holder,

3 556 U.S. 418, 434 (2009) (quoting Hilton v. Braunskill, 481 U.S. 770, 776 (1987)). In deciding whether to grant the stay, courts balance all factors, with the first and second the “most critical.”

Id. “The party requesting a stay bears the burden of showing that the circumstances justify an exercise of [the Court’s] discretion.” Id. at 433-34. In the Third Circuit, In re Revel AC, Inc., 802 F.3d 558 (3d Cir. 2015), is the lead case that addresses application of the Nken factors. According to Revel, courts first assess whether the movant has satisfied the first and second Nken factors: Is the movant likely to succeed on the merits and will the movant suffer irreparable injury absent a stay? Id. at 571. To show a sufficient likelihood of success on the merits, the movant must demonstrate that it has “a reasonable chance, or probability, of winning.” Id. (quoting Singer Mgmt. Consultants, Inc. v. Milgram, 650 F.3d 223, 229 (3d Cir. 2011) (en banc)). A “reasonable chance of winning,” in the Court of Appeals’ prescription, “need not be more likely than not,” id. at 569, but it must be

“significantly better than negligible,” id. at 571. To demonstrate irreparable injury, the movant must evidence “harm that cannot be prevented or fully rectified by a successful appeal.” Id. at 568 (quoting Roland Mach. Co. v. Dresser Indus., 749 F.2d 380, 386 (7th Cir. 1984) (Posner, J.)). Further, the movant’s alleged injury absent a stay must be “likely, not merely possible.” Id. at 569 (cleaned up). It is important to note the different burden the moving party must carry as to each factor. While reference to “likelihood” under the first factor (likelihood of success on the merits) has been interpreted by our Court of Appeals as a standard less onerous than more likely than not, the use of “likely” for irreparable harm means “more apt to occur than not.” Id. Thus, the

showing required as to likelihood of success on the merits is significantly less than what is required for establishing irreparable harm.

4 Once beyond the first two factors, the court must balance the harm of the movant with the harm of the other parties interested in the litigation should a stay be issued. In doing so, courts in

this Circuit “weigh the likely harm to the movant (absent a stay) (factor two) against the likely irreparable harm to the stay opponent(s) if the stay is granted (factor three).” Id. Courts must also assess “where the public interest lies (factor four)—in effect, how a stay decision has consequences beyond the immediate parties.” Id. (internal quotation marks omitted).1 In balancing each of the four factors, “[k]eeping in mind that the first two factors are the most critical,” id. at 570, the Court of Appeals counsels that likelihood of success on the merits is “arguably the more important,” id. at 568. Accordingly, “even if a movant demonstrates irreparable harm that decidedly outweighs any potential harm to the stay opponent if a stay is granted, it is still required to show, at a minimum, serious questions going to the merits.” Id. (cleaned up). Correlatively, “depending on how strong a case the stay movant has on the merits,

1 The Government contends that it may “merge” its irreparable harm with the harm it believes the public will suffer absent a stay. ECF 145, at 10 (citing Nken, 556 U.S. at 435). According to the Government, this is so because “the government’s interest is the public interest.” ECF 145, at 10 (citing Pursuing America’s Greatness v. FEC, 831 F.3d 500, 511 (D.C. Cir. 2016)). However, in arguing for merger, the Government misreads the case law. In Nken, the Supreme Court concluded that the traditional stay factors, not a more demanding statutory standard, governed a court’s authority to stay an alien’s removal pending judicial review, and that, in applying the traditional factors, “[t]he first two factors of the traditional standard are the most critical.” Nken, 556 U.S. at 434. The Nken Court also suggested that courts should in particular circumstances merge the public’s interest with harms alleged by the Government. Id. at 435. But the Nken Court made clear that merger was appropriate only “when the Government is the opposing party” and after “an applicant satisfies the first two [stay] factors.” Id. (emphasis added); see also Pursuing America’s Greatness, 831 F.3d at 511 (citing Nken for the proposition “that, in the context of a stay, assessing the harm to the opposing party and weighing the public interest merge when the Government is the opposing party”). Here, unlike in Nken, the Government is not the party opposing a stay but the party applying for it, and it is the Government’s burden to “satisf[y] the first two factors,” not Safehouse’s. See Nken, 556 U.S. at 435. The status of the Government as the party requesting the stay may seem trivial, but permitting the Government to merge the public’s harms with its own only when opposing an application for a stay makes eminent sense. Generally, when the Government opposes a stay request, it has already secured a court judgment finding a violation of some federal law, and the applicant for the stay is the party found to have violated it. In such situations, granting the stay would prevent the Government from enforcing a law a court has found it has legal authority to enforce. Here, because my conclusion was that the Government lacks legal authority to apply Section 856(a) in the way that it seeks, the position it is advancing is not presumptively in the public interest.

5 a stay is permissible even if the balance of harms and public interest weigh against holding a ruling in abeyance pending appeal.” Id. at 571.

Revel could be read to imply that the movant must satisfy factors one and two before a court can balance the relative harms or consider the public’s interest. See id. (noting that stay factors one and two “both” are “necessary”). But Nken never imposed such a requirement, and Revel’s prescription is cast as a recommendation rather than a dictate. Revel observes that the stay analysis is fundamentally equitable, and that “all four stay factors are interconnected.” Id. Indeed, a district court’s granting of equitable relief is the paradigmatic “exercise of judicial discretion.” Virginian Ry. Co. v. United States, 272 U.S. 658, 672 (1926). Thus, while Revel’s step-by-step approach greatly assists in resolving the majority of stay requests, I understand Revel to nonetheless allow consideration of all four factors where circumstances, either particular to the parties, or forced upon them by world events, require it.

With that background in mind, I turn to the specifics of the Government’s motion. III. The Preliminary Burdens The Government has made the requisite showing of a more than negligible likelihood of success on the merits To carry its burden to show that it is likely to succeed on the merits, the Government need only demonstrate that it has “a reasonable chance, or probability, of winning.” Revel, 802 F.3d at 571. Those chances of prevailing on appeal must be “significantly better than negligible,” id., but “need not be more likely than not,” id. at 569. The Government has met that modest threshold here. The very nature of this case leaves room for reasonable minds to differ. I have previously observed that the question of whether Section 856(a) applies to the conduct planned by Safehouse is “complex,” “difficult,” and

6 altogether “novel.” See, e.g., Transcript of Oral Argument, at 4:19-5:13 (noting that “the questions [asked at oral argument] are questions that are important to [the Court] in grappling

with the complicated issues” presented by the case), ECF 131; Memorandum Opinion on Government’s Motion for Judgment on the Pleadings, at 7 (noting that the Court was tasked with “address[ing] a novel question of statutory construction”), ECF 133; id. at 56 (acknowledging that “[b]oth sides skillfully argue that Congress’s meaning in [Section] 856 is consistent with their own”); Memorandum Opinion on Safehouse’s Motion for Final Declaratory Relief, at 4 (noting that the reach of the statute was “a difficult and complex matter of first impression”), ECF 141. To illustrate, the Government will undoubtedly urge the Court of Appeals not to break from the other circuits that have concluded that the “purpose” requirement of Section 856(a)(2) can be satisfied by those using the premises rather than those making the place “available.” See

ECF 133, at 24-26. Maintaining uniformity among circuits is an important consideration for appellate courts, and though the Third Circuit is willing to chart its own path even where there is substantial precedent to the contrary, e.g., Cirko v. Commissioner of Social Security, 948 F.3d 148 (3d Cir. 2020), it could choose to align with those circuits. My conclusion that the Government has met its burden on this factor also depends on how I read the language of Section 856. Other courts saw a need to remedy a perceived redundancy between subparagraphs (a) and (b), whereas I did not. As noted in my previous opinion, ECF 133, at 8-9, even at the level of the Supreme Court, jurists ostensibly engaging in a “plain meaning” analysis can look at the same language and reach opposite results. See, e.g.,

Zuni Pub. Sch. Dist. No. 89 v. Dep’t of Educ., 550 U.S. 81 (2007).

7 Furthermore, the standard of review on appeal is broad. The parties have stipulated to all material facts, and I have not made any independent factual findings. ECF 137, Ex. A. Whether

Safehouse runs afoul of Section 856(a) is a pure question of law that the Court of Appeals will resolve de novo, with my opinion owed no more deference than any persuasive power it may contain. I stand by my analysis. I particularly stand by the proposition that words are tools and not talismans, and courts should hesitate to endow them with a meaning they could not possibly have, particularly in the context of criminal liability.2 But I have no hesitancy in concluding that reasonable minds may differ as to the correct result. That suffices to meet the Government’s burden under Nken’s first prong.

2 The parties agree that when a statutory provision is clear on its face, the text alone reveals what Congress meant by the words it employed. See ECF 47, at 5; ECF 48, at 20. But the Government is wrong to suppose that application of this principle inexorably leads to the conclusion it seeks. That is in part because the word “purpose,” which lies at the heart of the statute, operates on several levels. In enacting Section 856(a)(2), Congress generally prohibited any person who manages or controls any place from knowingly and intentionally making that place available for the purpose of unlawful drug use. In choosing those words, at one level, the drafters of the legislation had linguistic objectives. In my previous opinion, for instance, I observed that dictionaries (past and present) routinely define “purpose” as one’s ultimate objective. ECF 133, at 37-38. I also observed that common principles of grammar require that both Section 856(a)(1) and Section 856(a)(2) “have the same subject, identified in [Section] 856(b) as ‘any person,’” and that “the purpose requirement” in Section 856(a)(2) “applies to the person who acts knowingly.” ECF 133, at 20. But in enacting Section 856(a)(2), the drafters also had legal objectives, that is, objectives about what the “law” would be in light of their adopted text. In addition to generally prohibiting any person who manages or controls any place from knowingly and intentionally making that place available for the purpose of unlawful drug use, Congress certainly had a variety of more concrete convictions not only about the correct application of the statute to particular circumstances, but also convictions as to where it would not apply. The available legislative evidence supports the conclusion that Congress desired Section 856(a)(2) to outlaw “crack houses” but not to punish business owners who “knew” drug use occurred on their property but did not operate their property for that “purpose.” See, e.g., ECF 133, at 20-21, 29, 31. And various courts of appeals have concluded that Section 856(a)(2) does not automatically apply simply because there is some drug use on the premises, see ECF 133, at 39, even if the words of the statute could fit those facts. As noted in my prior opinion, the Government itself retreats both from its literalism and from its core reading of the statute when faced with challenging facts such as parents struggling to save an addicted child. ECF 133, at 41-42. The task for a court when interpreting a statute is to discern the objective Congress sought to achieve in the specific case before it. When language can be understood to mean different things, particularly when it comes to criminal liability, and it is indisputable that the drafters’ concrete convictions could not have encompassed the circumstances of a particular case, judges who do not test their conclusions by a variety of interpretive tools do so at their peril. And it is the combination of those tools that led me to the conclusion reached in my October decision.

8 The Government will not be irreparably injured by the denial of its request for a stay While the Government easily makes the requisite showing that it is likely to succeed on the merits, it does not carry its more onerous burden to show that it will be irreparably harmed absent a stay. To demonstrate irreparable injury, the Government must demonstrate “harm that cannot be prevented or fully rectified by a successful appeal.” Revel, 802 F.3d at 568 (internal citations omitted). Further, the possibility that “corrective relief will be available at a later date, in the ordinary course of litigation, weighs heavily against a claim of irreparable harm.” Id. The

injury must be “neither remote nor speculative, but actual and imminent.” Id. at 571 (citations omitted). And the Government must demonstrate that such injury “is likely,” that is, “more apt to occur than not.” Id. The Government makes several arguments to support its claim of irreparable harm. Except for its argument as to possible community impact, the harms it asserts are highly abstract and, even in the case of community impact, highly speculative. 1. The Government has not shown that the opening of Safehouse would frustrate any congressional determination concerning drug use. I take the Government’s first argument to be that allowing Safehouse to operate “threatens to overturn Congress’s determination that there is no legal use” for certain unlawful drugs. ECF 145, at 11. That contention lacks merit both legally and factually. The Government seizes upon a statement from a Safehouse witness at an

evidentiary hearing to the effect that a ruling in Safehouse’s favor would “decriminalize” drug use on its premises. See ECF 145, at 11-12 (citing ECF 127, at 237:2-12). My ruling did not reach such a conclusion. To the contrary, my opinion expressly noted that “opioid . . . possession is undisputedly criminal,” and that, “[b]y definition, a person cannot lawfully use or consume a substance that the person cannot even lawfully possess.” ECF 133, at 18, 40. At that

9 evidentiary hearing, I noted that “it’s self-evident that [users of unlawful drugs at Safehouse’s proposed supervised injection site would] be violating the law,” and that Jose Benitez, president

and treasurer of Safehouse, “at one point . . . conceded as much.” ECF 127, at 167:6-169:3. Mr. Benitez went on to concede that his staff would have no power to stop “the police or DEA [if they] tried to make an arrest at the site of a user,” even if staff might try to negotiate with the relevant authorities so that an arrest did not threaten any other constituency. ECF 127, at 167:24- 168:20. The Government remains free to arrest and charge users of narcotics if it chooses to exercise its prosecutorial discretion and employ law enforcement resources in that way. Nor does the fact that Safehouse provides a site where users of unlawful narcotics can inject under medical supervision as part of a public health initiative in any way deny the judgment reached by Congress. To the contrary, the need to inject under observation underscores the deadly potential of the banned substances. And though the Government scores a

semantic point when it observes that “safe injection” is an oxymoron when it comes to heroin, cocaine, and nonprescribed uses of Fentanyl, if one looks beyond wordplay, it cannot be disputed that Safehouse’s efforts are focused on the safety through supervision of the user, not the safety of the act of injection. In earlier correspondence from the Government to the Court, which Safehouse has attached to its response here, the United States Attorney referred to his office’s duty to enforce the law, which, if Safehouse were to open, would cause “chaos” and “deteriorate into a literal street fight.” ECF 137, Ex. B. As noted above, enforcing the law certainly would be the Government’s prerogative, but why such chaos needs to ensue is difficult to comprehend. The

crime that would occur on the premises of Safehouse is the same crime that already occurs daily

10 outside its doors and indeed in communities across the United States—and one that the federal government rarely prosecutes.

Indeed, in 2018 and 2019, there was not a single federal prosecution for drug possession in the Eastern District of Pennsylvania.3 Widespread prosecution of simple possession has only occurred in tandem with other law enforcement initiatives, such as border control strategies. In 2019, nationwide, there were only 560 federal cases brought for simple possession, and 243 of those emanated from just two districts, one in a border state.4 In 2018, there were 777 prosecutions for simple possession nationwide.5 When the federal government does choose to prosecute it, simple possession is a misdemeanor for a first-time offender, 21 U.S.C § 844, and, except for offenses involving unlawful entry, it results in probation or a fine more than seventy percent of the time.6 These numbers reflect the fact that simple possession and attendant use together represent

the prototypical example of a victimless crime, apart from the sad reality that users are victimizing themselves. Certainly, the cumulative consequences of such use have far broader negative effects, which account for societal efforts to prevent or stop it. But acts of individual use and possession have little impact in isolation, as reflected by the rarity of prosecution except to serve other law enforcement purposes. The Government has not explained its sudden

3 U.S. Sentencing Comm’n, 2019 Federal Sentencing Statistics for the Eastern District of Pennsylvania 3 (statistical information packet for fiscal year 2019 for the Eastern District of Pennsylvania accessed online); U.S. Sentencing Comm’n, 2018 Federal Sentencing Statistics for the Eastern District of Pennsylvania 3 (statistical information packet for fiscal year 2019 for the Eastern District of Pennsylvania accessed online). 4 U.S. Sentencing Comm’n, Fiscal Year 2019 Datafile 64 (accessed online). 5 U.S. Sentencing Comm’n, Fiscal Year 2018 Datafile 64 (accessed online). 6 U.S. Sentencing Comm’n, Weighing the Charges: Simple Possession of Drugs in the Federal Criminal Justice System 9 (Sept. 2016) (accessed online).

11 imperative to prosecute such cases, let alone one that would be catalyzed by Safehouse’s operation. And the Government can hardly claim irreparable harm from a confrontation of its

own creation that it could avoid with minimal impact on the rule of law. The Government’s position is therefore difficult to rationalize, except as a symbolic act. 2. The Government has not shown that the opening of Safehouse would cause more or riskier drug use. The Government also invokes a moral hazard surrounding supervised injection sites, relying on the theory that permitting such conduct normalizes it, and that reducing the “cost” of opioid use enables continued or riskier use. The Government further contends that in addition to an increase in destructive behavior, there likely will be an increase in crimes related to drug use. See ECF 145, at 11. There is an intuitive appeal to this argument, and in many ways, it exemplifies the fundamental premise of a law enforcement approach to opioid misuse as contrasted with a public health approach. As I previously opined—and as is reflected in my

conclusion as to the proper scope of Section 856—that debate is properly held in Congress. Given the procedural posture here, however, I must assess whether the Government has met its burden of establishing that the harm it invokes is more likely to occur than not. On that score, the Government relies exclusively on inference rather than facts, leaving it at a significant disadvantage.7 Safehouse, in opposition to the Government’s motion, points to a vast body of literature that evaluates the effects of supervised injection sites. As with most research, it is of varying quality, and the amount of available data makes it difficult to survey. But some researchers,

7 As noted in my second opinion entering final judgment, the Government strenuously resisted the development of a factual record. ECF 141, at 4.

12 publishing in peer-reviewed journals, have attempted to evaluate and summarize the data using various techniques of analysis, providing a useful way to access research in the field.

In 2014, French researchers selected 75 studies for review, including cohort studies, analytical studies, and evaluative studies.8 As to the specific issue of whether supervised injection sites increase drug use, they noted that, while “it was feared that [supervised injection sites] might foster the initiation of new users into intravenous drug use . . . no study found any increase in the total number of local [persons who inject drugs], irrespective of the [supervised injection sites] studied.”9 In 2017, Canadian researchers reviewed 341 potentially relevant studies conducted between 2003 and the date of publication,10 and screened them for quality using three evaluation methods. Based on the quality of the research, the authors chose for review 47 studies employing varying methodologies. Like the French researchers, the Canadian researchers did

not find that supervised injection sites resulted in normalization or increased usage: “While concerns persist that [supervised drug consumption facilities] may increase illicit drug use and

8 Chloé Potier, et al., Supervised injection services: What has been demonstrated? A systematic literature review, 145 Drug and Alcohol Dependence 48 (2014). The authors selected which research studies to review by applying two evaluation scales for quality, STROBE (for observational studies) or CHEERS (for medical economic studies). STROBE is an acronym for “Strengthening the reporting of observational studies and epidemiology,” a standard created by and International collaborative of researchers to improve the quality of research. CHEERS is an acronym for “Consolidated health economic evaluation reporting standards,” a similar standard established by the Professional Society for Health Economics and Outcomes Research. Id. 9 Potier, supra, at 64. 10 Mary Clare Kennedy, et al., Public Health and Public Order Outcomes Associated with Supervised Drug Consumption Facilities: a Systematic Review, 14 Current HIV/AIDS Report 161 (2017). The researchers screened the studies for quality using three evaluation methods, including the Quality Assessment Tool for Observational Cohort and Cross-sectional Studies. The Quality Assessment Tool was created by a branch of the National Health Institutes as a method for evaluating the quality of research. See Study Quality Assessment Tools, Nat’l Heart, Lung, and Blood Inst., available at https://www.nhlbi.nih.gov/health-topics/study-quality-assessment-tools.

13 discourage [people who use drugs] from seeking addiction treatment, such concerns are not supported by existing evidence.”11 The researchers also found that the frequent use of

supervised injection sites “has not been found to increase non-fatal overdose risk, which challenges the contention that these facilities promote riskier drug use practices (e.g. taking higher doses) associated with overdose.”12 With the case in its current posture, the question is not whether these researchers are ultimately correct. The question is whether, considered against the research to which Safehouse points, the Government’s argument based solely on inference suffices to meet its burden to establish irreparable harm. The research reviewed above, which summarizes much of the data that Safehouse has cited as the rationale for its operation, leads me to conclude that there is significant doubt as to whether the inference the Government would have me draw is supportable. Because the Government bears the burden, this aspect of its argument fails.

3. The Government has not shown that Safehouse opening without the full support of various governmental actors would cause tangible harm. The Government further contends that because Safehouse is proceeding on its own and without the guidance or supervision of various governmental and public health officials, there is a significant risk to the public’s safety. ECF 145, at 13. Specifically, the government argues that the “public will also be harmed because the ruling will radically alter the approach to opioid addiction treatment in the Commonwealth of Pennsylvania without providing an opportunity for the U.S. Secretary of Health and Human Services, the U.S. Surgeon General, the Pennsylvania General Assembly, the

11 Kennedy, supra, at 177. 12 Id.

14 Pennsylvania Board of Medicine, or the Philadelphia City Council to provide input or suggest safeguards or regulations for such unprecedented activity.” ECF 145 at 13. The Government

invokes an opinion from the United States Surgeon General that, based upon his review of the available data, the effect of “front-line strategies such as medication assisted treatment should be optimized before having conversations about more controversial interventions.” The Surgeon General adopted this position in part, he said, because “safer doesn’t mean safe.” Id. at 14. The Government accurately notes that in every other instance supervised injection sites have proceeded with the active participation of public officials. It also is correct when it observes that although the American Medical Association has endorsed the concept of supervised injection sites, its endorsement assumed the participation of governmental officials. ECF 127, at 52:8-59-19. In response to this and related arguments, the City of Philadelphia Health Commissioner, Dr. Thomas Farley, notes that Safehouse’s operational plan was created in

collaboration with his office, which consulted with experts in the field of opioid abuse. ECF 151, at 4-5. Commissioner Farley also emphasizes that “both the American Medical Association and the Pennsylvania Medical Association are in favor of overdose prevention sites.” Id. at 6. The Government is correct that it would be far better to have Safehouse commence operation with the direct involvement of public health officials above the level of municipal government. Whether Safehouse’s failure to do so constitutes irreparable harm is, however, a far closer question. The services Safehouse will provide—the administration of medications to combat overdose, defibrillation, and the administration of oxygen—are not medically complex. They are also the exact same services that Safehouse’s sister organization, Prevention Point,

already provides on the street, where the Government concedes their legality.

15 The Government also is correct that there are no established standards that would apply to how Safehouse operates. But as I observed in a previous opinion, that is typically the case as

new medical subspecialties evolve, even as to critically important areas such as specialty board qualifications for medical practitioners. Estate of Goldberg v. Nimoityn, 193 F. Supp. 3d 482, 490-91 (E.D. Pa. 2016). Standards emerge over time based upon the innovations of practitioners in the field, and the Philadelphia Health Commissioner represents that the Safehouse model is based upon the operation of similar facilities abroad, which are operating under the auspices of public health authorities. Further, there are well-established standards for the core medical interventions Safehouse would use. On balance, the Government has shown that how Safehouse is proceeding is less than optimal, but I cannot say that it rises to the level of irreparable harm. 4. The Government has not shown that the surrounding community will be adversely

affected absent a stay. The Government finally raises the possibility of harm to the community surrounding any location where Safehouse operates. Although I have previously held that such concerns are most properly addressed to local authorities, the equitable nature of the standard controlling the motion to stay necessarily requires me to consider community impact here. It should be emphasized that, however one gauges the true risk presented by public health initiatives such as supervised injection sites, the concerns of neighbors are understandable and worthy of consideration. While it is true that opioid abusers suffer from a medical condition, that condition frequently results not just in self-destructive behavior, but also behavior that causes some form of harm or disruption to others. Anyone with meaningful experience in recovery

work, and anyone who has ever had to ban a family member from the home because of persistent negative behaviors related to opioid addiction, understands that.

16 Once again, the Government presents mostly argument rather than evidence. The Government asks the Court to take judicial notice of “news media stories, reports of

governmental entities, and academic studies” that document the negative effects of supervised injection sites. ECF 155, at 5-6. However, neither the Government nor any amici cite any academic study. The media accounts to which the Government refers largely are set forth in the amicus brief filed by the Fraternal Order of Police and an assortment of local community groups. See ECF 148. As to reports from public authorities, the Government cites a single study from the Canadian province of Alberta, which, as discussed below, has proven controversial. The test for granting a stay requires that I look at the issue using a broad lens to determine whether the Government has carried its burden to show irreparable harm. Safehouse cites to a meaningful body of research dealing with the community safety aspects of supervised injection sites. For example, in 2019, a Canadian researcher conducted a systematic review of

data on supervised injection sites relating to crime, disorder, and community health issues such as discarded syringes.13 The author considered thirteen studies published in peer-reviewed journals that examined data from sites in Canada, Germany, Australia, Denmark, the United Kingdom, and the Netherlands.14 In all, the author concluded that “analysis of the relevant research literature on [supervised injection sites] and their effects on public crime, disorder and community health issues (discarded syringes) shows that the preponderance of evidence thus far

13 Laura Huey, What is Known About the Impacts of Supervised Injection Sites on Community Safety and Wellbeing? A Systematic Review, Sociology Publications (2019). 14 Like the researchers in the Potier and Kennedy studies, an attempt was made to evaluate the papers for the quality of their research, but such evaluation was deemed impractical by virtue of the fact that different researchers had each employed different study techniques. In any case, as I have mentioned previously, in resolving the Government’s motion to stay, the question is not whether these researchers are ultimately correct but whether the Government has met its burden to establish irreparable harm.

17 is tilted towards supporting the view of these sites as producing favorable outcomes not only for [individual drug users] but also for potentially enhancing the well-being of the local community.”15 These results were consistent with results from a team of French researchers,16

as well as those from a previous team of Canadian researchers.17 In contrast to this academic research, the Government through their amici cite multiple news stories reporting the perception that supervised injection sites cause deleterious community effects. ECF 148, at 6-8. I would expect the journalists reporting these experiences to accurately recount what they were being told and will further assume that those conveying the information were recounting their experiences as they perceived them to be. But it is difficult to rely on such anecdotal accounts as an evidentiary basis for finding irreparable harm in the face of substantial peer-reviewed research. Moreover, studies of methadone clinics suggest that a community’s perception of

supposed negative effects of drug treatment facilities does not necessarily conform to the sites’ actual effects. Patients attending methadone clinics are not violating the law but participating in medically assisted treatment under supervision. Their presence is nonetheless feared and

15 Huey, supra, at 16. 16 Potier, supra, at 64. In their review, the authors found no increase in crime beyond some increase in small-scale drug trafficking in the immediate vicinity of supervised injection sites. And, like the Canadian researchers, they concluded that the relevant data from Canada and Australia showed no increase in crime and reduced instances of public self-injection and discarded paraphernalia. For the sake of accuracy, it must be noted that not every study reviewed by these teams of researchers addressed crime or issues of public disorder, but of those that did the evidence did not reflect serious problems. 17 See Kennedy, supra, at 179.

18 shunned.18 This is consistent with research confirming that addiction is widely stigmatized.19 Such stigma or fear does not suffice to establish actual harm, let alone likely, irreparable harm.

The one substantial item of evidence cited by the Government is a recent report from the provincial government of Alberta, Canada.20 The report was compiled by a commission appointed after a freeze of funding for supervised injection sites enacted by a newly elected government opposed to their operation. The report’s principal conclusion was as follows: During the public consultations, the Review Committee was made aware of numerous concerns of residents living near the respective sites. Although stakeholder feedback on the socioeconomic impact was mixed, it was predominantly negative, except for the Edmonton town hall meeting. The issues raised ranged from increases in needle debris to increases in crime and increases in

18 See, e.g., C. Debra M. Furr-Holden, et al., Not in My Back Yard: A Comparative Analysis of Crime Around Publicly Funded Drug Treatment Centers, Liquor Stores, Convenience Stores, and Corner Stores in One Mid- Atlantic City, 77 Journal of Studies on Alcohol and Drugs 17 (2015). The community effect of methadone clinics provides a useful analogue more generally. Although there are no supervised injection sites in the United States, methadone clinics are lawful and recognized by public health experts as a significant tool for addressing opioid misuse. Nevertheless, many communities have opposed the establishment of such initiatives, advancing similar arguments to those now deployed by opponents of supervised injection sites. To combat that opposition, some researchers have evaluated whether methadone clinics are accompanied by the increase in crime and social disorder that their critics fear. In one study, which focused on the effects of the opening of nine outpatient methadone clinics in Philadelphia from 2007 to 2017, researchers at the University of Pennsylvania came to mixed conclusions. See Ruth A. Moyer & Greg Ridgeway, The effect of outpatient methadone maintenance treatment facilities on place- based crime, Journal of Experimental Criminology (2018). Near the facilities, they found a small increase in violent crime and a moderate increase in drug-related crime, but a reduction in property crimes and crime overall. These conclusions lend some support to the Government (outpatient drug treatment facilities can be magnets for at least some forms of criminal activity) but also to Safehouse (with appropriate police support, which the City has represented it will provide, ECF 151, at 7, risk to the surrounding community can be identified and mitigated). In the Furr-Holden study cited above, researchers at Johns Hopkins University evaluated crime statistics in the vicinity of 53 publicly funded outpatient drug treatment centers in Baltimore. Focusing on violent crime, the Johns Hopkins researchers found only a minimal difference in rates of a violent crime closer to such facilities. Furr-Holden, supra, at 22. A similar study conducted in Baltimore several years earlier drew the same basic conclusions, finding no statistically significant increase in crime within 100 meters of methadone treatment centers. Susan J. Boyd, et al., Use of a ‘microecological technique’ to study crime incidents around methadone maintenance treatment centers, 107 Addiction 1632 (2012). These conclusions, if anything, weigh against the Government’s contention that the surrounding community will suffer harm should a supervised injection site open. 19 See, e.g., Douglas L. Polcin, et al., Community context of sober living houses, 20 Addiction Research & Theory 480 (2012). 20 Alberta Health, Government of Alberta, Impact: A socio-economic review of supervised consumption sites in Alberta, Supervised Consumption Services Review Committee (March 2020) (accessed online).

19 overall social disorder since the sites opened. The Review Committee was also made aware of issues relating to the operations and oversight of various sites.21

The Report was first cited in the Government’s reply brief, and Safehouse has not filed any sur- reply. It is therefore difficult to assess the weight it should be assigned. But it was not published in a peer-reviewed journal or prepared in accordance with a specific research methodology, and it relied heavily upon town hall style meetings. Further, to the extent that the Government elsewhere urges me to take judicial notice of media accounts, it only seems fair to take note that the Alberta report was not well-received in a variety of circles.22

21 Id. 22 See, e.g., Benjamin Perrin, Alberta’s war against safe injection sites, Maclean’s (Mar. 12, 2020) (accessed online) (“[T]he review is dangerously biased, methodologically flawed and explicitly ignores over 100 peer-reviewed studies on the benefits of [supervised-consumption sites].”); Alana Smith, Academics question methodology of UCP- approved supervised consumption sites report, Calgary Herald (Mar. 8, 2020) (accessed online) (“The [] commissioned report into supervised drug-use sites is drawing ire from academics, some of whom are questioning its data collection, analysis and perceived findings.”); id. (“‘I would say that the only conclusion that a reasonable and fair-minded reader should draw is that the report is a political document and not an objective or scientifically credible evaluation of supervised consumption sites.’”) (quoting Cameron Wild, then professor at the University of Alberta School of Public Health); id. (“The report gives no indication of how participants were selected for the online survey and if there is any verification to prove people say and live where they said they do. . . . [or] whether someone could submit responses more than once.”); id. (“[T]he amount of participants from differing cities varied significantly, creating an imbalanced sample. . . . [and, thus, the] results of the report were then heavily driven by Lethbridge respondents, who greatly outnumbered participants from any other city.”) (paraphrasing in story of Jenny Godley, then professor in the University of Calgary’s sociology department); id. (“Let’s just say [this report] wouldn’t pass my undergrad stats class.”) (quoting Godley). The Alberta report is representative of continued political debate in Canada over supervised injection sites with periodic attempts to defund them or limit their operation. See Thomas Kerr et al., Supervised injection facilities in Canada: past, present, and future, 14 Harm Reduction Journal 28 (2017). That political debate has played out against the backdrop of a notable decision from the Supreme Court of Canada, Canada (Att’y Gen.) v. PHS Community Services Society, 2011 SCC 44 (Can.), which in itself has some degree of relevance as to the available evidence concerning the operation of supervised injection sites. The case arose when Canada’s Minister of Health refused to renew permission for the first facility in Canada (Insite in Vancouver) to continue operating. A trial was held, and evidence was submitted as to whether the site was effective in meeting its public health objectives and whether it had negative community impact. On appeal, in a unanimous opinion rendered on constitutional grounds, the Supreme Court permitted its continued operation and ordered the Minister to renew Insite’s operating charter. In doing so, the court construed the available evidence as demonstrating positive public health outcomes with no discernable community harm. Indeed, it noted that “Insite saves lives”; that Insite’s “benefits have been proven”; that “[t]here has been no discernable negative impact on the public safety and health objectives of Canada during its eight years of operation”; and that “[t]he effect of denying the services of Insite to the population it serves is grossly disproportionate to any benefit that Canada might derive from presenting a uniform stance on the possession of narcotics.” Id. ¶ 133. Such a conclusion from Canada’s highest court certainly lends support to Safehouse’s

20 On balance, in the face of substantial and seemingly credible social science research, the Government’s reliance upon media accounts and a single governmental report falls short of

meeting its burden as to the community impact aspect of irreparable harm. Considering the Government’s presentation as a whole, the irreparable harm it asserts is supported far more by argument than by evidence. In making such arguments, the Government has failed to carry its burden to show that it more likely than not will suffer irreparable harm absent a stay. Because of events separate from the proceedings in this case, Safehouse will not be irreparably harmed by a stay As to the balance of harms, Safehouse contends that a stay will cause it irreparable harm because it will be prevented from opening, depriving opioid users from potentially life-saving care, and undermining the Philadelphia Health Commissioner’s public health initiative. In support, it cites Third Circuit precedent dealing with alcohol treatment centers, Sullivan v. City of Pittsburgh, 811 F.2d 171, 183-84 (1987), and methadone treatment centers, New Directions Services v. City of Reading, 390 F.3d 293, 303-04 (3d Cir. 2007). ECF 149, at 17-18. It further argues that a stay would violate the religious beliefs of Safehouse’s founders, reviving their claims under the Religious Freedom Restoration Act previously dismissed without prejudice as moot.

The harm that Safehouse and its founders might otherwise have claimed has fallen prey to outside events. First, Safehouse’s naïve attempt to begin operation in the “City of Neighborhoods” without consulting the affected neighborhood generated substantial opposition,

characterization of the available evidence about the comparative risks and benefits of supervised injection sites and counsels against the Government’s argument that the opening of Safehouse would cause the public irreparable harm.

21 supported by elected officials,23 effectively preventing it from opening. Second, the opioid epidemic has been overshadowed by the COVID-19 pandemic, followed by momentous

demonstrations protesting excessive use of force by police, straining the City’s resources. As to the first issue, no state or local legislation has yet been enacted that would bar Safehouse from opening.24 Regardless, public reaction to Safehouse’s attempt to open without employing a transparent process assures that substantial outreach to both neighbors and elected officials alike will be required before operation is a realistic possibility. That was true even before Philadelphia was hit by the pandemic. To be sure, the threat imposed by COVID-19 does not diminish the threat posed by opioid addiction. But, by any measure, the challenges of coping with the pandemic and with recent community unrest understandably triggered by the death of George Floyd present strains on the City—government and citizens alike—that make the proposed opening of Safehouse unrealistic any time soon.

These same considerations guide my decision as to why a stay is now in the public interest. Recent events compel the conclusion that a stay is in the public interest In the first instance, the Government seeks to “merge” its alleged irreparable harm with the harm it believes the public will suffer absent a stay, arguing that it is entitled to a stay

23 Philadelphia City Council, Resolution No. 200195, adopted March 5, 2020. 24 Proposed state legislation, Senate Bill 933, Regular Session 2019-2020, currently pending in the Pennsylvania Senate Judiciary Committee, would impose requirements for the operation of supervised injection sites—including that there be three public hearings before an opening—but otherwise permit it to proceed if a medical director is on site and a public safety plan is put in place. Proposed municipal legislation, Philadelphia City Council, Bill No. 200189, introduced February 27, 2020, would not explicitly prohibit operation of supervised injection sites, but it would require any prospective operator of a supervised injection site to obtain support from the overwhelming majority of the surrounding community and, for practical purposes, operate as a ban. It remains striking that no legislation has been initiated federally. Eighteen months have elapsed since Safehouse stated its intention to begin operation. A one-sentence clarifying amendment to the statute would achieve what the Government seeks to achieve in the courts.

22 because “the government’s interest is the public interest.” ECF 145, at 10. As set forth in my discussion of the controlling legal standard, I reject that argument because it based upon a

misreading of Nken. Under Nken, such an argument only has merit where the Government is the party opposing a stay. See Nken, 556 U.S. at 435. The Government goes on to argue that the injunctive powers conferred by 21 U.S.C. § 856(e) reflect a congressional determination that irreparable harm will result if the law is not enforced, thereby compelling the issuance of a stay in this case. ECF 145, at 12 (citing Instant Air Freight Co. v. C.F. Air Freight, Inc., 882 F.2d 797, 799 (3d Cir. 1989)). Section 856(e) states that “[a]ny person who violates [Section 856(a)] shall be subject to declaratory and injunctive remedies” as set forth in a subsequent statutory provision. The Government’s argument would have merit if Congress by enacting Section 856(a) had in fact intended to criminalize supervised injection sites (and with it granted courts the power to enjoin their

operation), but here I have concluded that it did not. A stay is nonetheless in the public interest for reasons apart from those advanced by the Government. All change results in some level of disruption, and Philadelphia has already experienced two profound disruptions. It is confronted with a public health crisis even larger than the opioid epidemic, which has strained municipal government and individual citizens to an unprecedented degree, imposing extreme demands on the City’s resources and employees. At the same time revenues have dwindled because of the deep downturn in economic activity. The nerves of citizens are frayed by fear and uncertainty, and that was true before the death of Mr. Floyd and the widespread protests that arose in its aftermath.

The opening of Safehouse would require multiple public meetings, the time and attention of the City Health Department, and the allocation of police resources. Even if one assumes a

23 flawless opening process, the operation of Safehouse would represent a significant change in how the City responds to opioid abuse, and such change would necessarily be disruptive. As

committed as the founders and supporters of Safehouse may be, and without discounting the seriousness of the opioid epidemic, I am persuaded that a stay of the judgment is at this time manifestly in the public interest. IV. Conclusion I construe Revel as allowing a district court to balance all traditional stay factors, including the public’s interest, when circumstances warrant it. Although the Government has failed to show that it will suffer harms that cannot be prevented or fully rectified by a successful appeal, the ongoing challenges posed by COVID-19 and the deep social currents and introspection following the tragic killing of Mr. Floyd make this the wrong moment for another change in the status quo. Accordingly, the Government’s request for a stay will be granted. An

appropriate Order follows.

/s/ Gerald Austin McHugh Gerald Austin McHugh United States District Judge