Jacob Pfaller v. Mark Amonette

55 F.4th 436

Opinion

PUBLISHED

UNITED STATES COURT OF APPEALS FOR THE FOURTH CIRCUIT

No. 21-1555

JACOB PFALLER, Administrator of the Estate of Danny Harold Pfaller, Plaintiff - Appellee,

v. DR. MARK AMONETTE, in his individual capacity, Defendant - Appellant,

and DR. LAURENCE SHU-CHUNG WANG, in his individual capacity, Defendant.

------------------------------ RIGHTS BEHIND BARS, Amicus Supporting Appellee.

No. 21-1612

JACOB PFALLER, Administrator of the Estate of Danny Harold Pfaller, Plaintiff - Appellee,

v. DR. LAURENCE SHU-CHUNG WANG, in his individual capacity,

2

Defendant - Appellant,

and DR. MARK AMONETTE, in his individual capacity, Defendant.

------------------------------ RIGHTS BEHIND BARS, Amicus Supporting Appellee.

Appeal from the United States District Court for the Eastern District of Virginia, at Richmond. Robert E. Payne, Senior District Judge. (3:19-cv-00728-REP)

Argued: September 16, 2022 Decided: December 15, 2022

Before WILKINSON, WYNN, and DIAZ, Circuit Judges.

Affirmed in part, reversed in part, and remanded by published opinion. Judge Wynn wrote the opinion, in which Judge Diaz joined. Judge Wilkinson wrote a separate opinion concurring in part and dissenting in part.

ARGUED: Andrew Nathan Ferguson, Erika L. Maley, OFFICE OF THE ATTORNEY GENERAL OF VIRGINIA, Richmond, Virginia, for Appellants. John Michael Shoreman, MCFADDEN & SHOREMAN, Washington, D.C., for Appellee. ON BRIEF: Mark R. Herring, Attorney General, K. Scott Miles, Deputy Attorney General, Laura Maughan, Assistant Attorney General, Michelle S. Kallen, Acting Solicitor General, Brittany M. Jones, Deputy Solicitor General, Laura H. Cahill, Assistant Attorney General, Rohiniyurie Tashima, John Marshall Fellow, OFFICE OF THE ATTORNEY GENERAL OF VIRGINIA, Richmond, Virginia, for Appellant Dr. Amonette. Erin B. Ashwell, Chief Deputy Attorney General, A. Anne Lloyd, Assistant Attorney General, OFFICE OF THE ATTORNEY GENERAL OF VIRGINIA, Richmond, Virginia; Lynne Jones Blain, M. Scott Fisher, Jr., HARMAN CLAYTOR CORRIGAN WELLMAN, Glen Allen, Virginia, for Appellant Dr. Wang. Mario B. Williams, Dallas S. LePierre, HDR LLC, Atlanta,

3

Georgia, for Appellee. Oren Nimmi, RIGHTS BEHIND BARS, Washington, D.C., for Amicus Rights Behind Bars.

4

WYNN, Circuit Judge:

Danny Pfaller died from liver cancer while he was a prisoner with the Virginia Department of Corrections (“Department”). His estate sued several prison officials under 42 U.S.C. § 1983 and Virginia law, alleging that they violated the Eighth Amendment and state law by failing to provide Pfaller treatment for his chronic hepatitis C until it was too late.

Defendants in this appeal are Dr. Mark Amonette and Dr. Laurence Shu-Chung Wang. Plaintiff alleges that Dr. Amonette designed treatment guidelines for inmates with hepatitis C that unconstitutionally excluded Pfaller from receiving treatment. Plaintiff also alleges that Dr. Wang failed to follow those guidelines and committed both medical malpractice and Eighth Amendment violations in denying him appropriate treatment. Defendants unsuccessfully moved for summary judgment, alleging that they were protected by qualified immunity and, on Dr. Wang’s part, derivative sovereign immunity.

For the reasons that follow, we reverse the district court’s denial of sovereign immunity to Dr. Wang and denial of qualified immunity to Dr. Amonette but affirm its denial of qualified immunity to Dr. Wang.

I.

Because this case is before us on interlocutory appeal, the following facts are recounted as the district court viewed them, and in the light most favorable to Pfaller. See Hicks v. Ferreyra, 965 F.3d 302, 305 (4th Cir. 2020).

5

A.

Hepatitis C is a disease caused by a viral infection of the liver. In certain individuals, hepatitis C can persist as an asymptomatic infection for years. In others, the virus can lead to liver inflammation, fibrosis (liver scarring), cirrhosis (liver tissue death), and even terminal liver cancer.

For many years, the only curative treatment for hepatitis C was a course of interferon-based drugs. However, these drugs offered a low cure rate (40 to 50%) and caused major side effects, including life-threatening neuropsychiatric and autoimmune disorders. In 2014, the Food and Drug Administration began approving a suite of new drugs called direct-acting antivirals for treatment of hepatitis C patients. These drugs offered great promise. Not only were they less likely to cause serious side effects, but they also boasted cure rates of 90 to 100%. By 2015, direct-acting antivirals were available for treating hepatitis C patients.

In response to these medical advances, Dr. Amonette, the Department’s chief physician, developed new hepatitis C treatment guidelines (“Guidelines”) for the Department. Under the Guidelines, the Department agreed to refer inmates with hepatitis C to a clinic at Virginia Commonwealth University (“VCU”) based on certain criteria. These criteria sorted inmates into three groups based on APRI and FIB-4 scores that assessed their level of fibrosis (if any). 1 Inmates who scored at the high end of the scale

1

“APRI” stands for “aspartate aminotransferase to platelet ratio.” J.A. 116. It is a noninvasive way to estimate fibrosis, or liver scarring. J.A. 397. “FIB-4” is short for “Fibrosis-4 index,” and is another formula used to assess fibrosis based on a patient’s age,

6

were to be “automatically referred to VCU for evaluation without any additional testing.” J.A. 116. Inmates who scored in the middle tier were to receive “additional testing to determine whether [they] should be referred for evaluation.” Id. And those who scored at the low end were not to be referred for treatment and instead were to “receive periodic laboratory blood testing and chronic care appointments with a medical provider.” Id. Outside of these criteria, a physician could also refer an inmate to the VCU clinic “if there [were] other findings suggestive of advanced liver disease.” J.A. 303. Once referred, the inmate would receive an antiviral prescription unless there was some other medical reason not to treat them.

Dr. Amonette explains that these Guidelines were designed to ensure that those with the greatest need were treated first. Plaintiff’s expert disputes this, stating that the Guidelines were actually a tool for excluding patients from treatment. The parties agree that when resources are limited, prioritization of patients with the most advanced disease can be a reasonable strategy. But whether the Department’s resources were actually limited is disputed.

B.

Danny Pfaller was an inmate with the Department from 1999 to 2018. As early as 2007, Pfaller tested positive for hepatitis C. Beginning in 2015, Pfaller had his blood drawn

platelet count, and other factors. J.A. 116, 397. Citations to the “J.A.” refer to the parties’ Joint Appendix filed in this appeal.

7

every six to twelve months to assess his APRI and FIB-4 scores, consistent with the Guidelines. During this time, Dr. Wang served as Pfaller’s primary physician.

Between 2015 and 2018, Dr. Wang twice failed to follow the Guidelines in treating Pfaller. On October 16, 2015, Pfaller tested into the middle tier of the Guidelines criteria and therefore qualified for fibroscan testing to determine the extent of any fibrosis. But Dr. Wang did not refer him. Two years later, following several tests that fell into the lowest tier, Pfaller tested into the middle tier again. But once again, Dr. Wang did not refer him for more testing. Dr. Wang claims both failures were a mistake. He states that he did not refer Pfaller for a fibroscan because he thought the inclusion number for the middle tier was 1.5 on the FIB-4 test, when it was in fact 1.45.

On May 7, 2018, Pfaller tested for a third time into the middle criteria, and this time, Dr. Wang acted. He submitted a request for a fibroscan on May 14. Meanwhile, Dr. Wang began to see Pfaller for numerous physical examinations showing increasingly severe signs of liver disease. In early June, Pfaller visited Dr. Wang for abdominal pain and swelling, and Dr. Wang ordered hepatitis C genotype testing and prescribed medication to alleviate the symptoms. Later that month, Pfaller visited Dr. Wang for weight gain and bowel obstruction. Dr. Wang found that Pfaller’s abdomen was distended, and he was retaining fluid—symptoms of liver disease—so Dr. Wang prescribed diuretics. During this time, Dr. Wang knew Pfaller “was on the list to be scheduled for a Fibroscan at VCU,” but took no further measures to ensure Pfaller was actually tested. J.A. 1958. In early July, Pfaller visited the clinic three more times, complaining of shortness of breath, a cough, and fluid retention.

8

On July 11, 2018, Pfaller visited Dr. Wang again for fullness in his abdomen. Dr.

Wang examined him and concluded his hepatitis C was causing fluid retention and cirrhosis. Two months after he originally put in a referral, he now ordered a fibroscan “asap,” which was performed six days later. J.A. 1959. A follow-up CT scan revealed a mass on Pfaller’s liver. After further testing, Dr. Wang referred Pfaller to VCU’s oncology department. Pfaller was diagnosed with untreatable liver cancer in early September. He died only a month later, on October 3, 2018.

C.

Plaintiff Jacob Pfaller, who is Pfaller’s son and the administrator of Pfaller’s estate, filed suit on behalf of the estate in October 2019. Three claims are relevant to this appeal. To start, Plaintiff filed Eighth Amendment claims against Dr. Wang and Dr. Amonette in their individual capacities, alleging they were each deliberately indifferent to Pfaller’s serious medical needs. Additionally, Plaintiff filed a state-law medical-malpractice claim against Dr. Wang in his individual capacity.

At the close of discovery, Dr. Wang and Dr. Amonette each moved for summary judgment. The district court rejected Defendants’ arguments in two separate decisions. First, the court found that Plaintiff had pointed to several genuine disputes of material fact as to whether Dr. Wang and Dr. Amonette were deliberately indifferent. See Pfaller v. Clarke, 630 B.R. 197, 207 (E.D. Va. 2021) (Dr. Amonette); Pfaller v. Clarke, No. 3:19CV728, 2021 WL 1776189, at *6 (E.D. Va. May 4, 2021) (Dr. Wang). The court also concluded that “Pfaller’s Eighth Amendment right to receive adequate medical care and to be free from officials’ deliberate indifference to his known medical needs” was clearly

9

established at the time in question, so Defendants were not entitled to qualified immunity. Pfaller, 630 B.R. at 215; Pfaller, 2021 WL 1776189, at *10. Finally, the court rejected Dr. Wang’s sovereign-immunity defense to the malpractice claim. Pfaller, 2021 WL 1776189, at *14. Both Defendants timely appealed.

II.

We first examine the district court’s denial of qualified immunity to both defendants on the constitutional claim. Our review is limited by the posture of this case on interlocutory appeal. Hicks, 965 F.3d at 308.

Although denials of summary judgment are generally not subject to appellate review, denials of qualified immunity are an exception and may be appealed immediately. Id. Nonetheless, our review is limited only “to the extent [a denial of qualified immunity] turns on an issue of law.” Id. (emphasis omitted) (quoting Gould v. Davis, 165 F.3d 265, 268 (4th Cir. 1998)).

In conducting such a review, we answer only a “narrow legal question: if we take the facts as the district court gives them to us, and we view those facts in the light most favorable to the plaintiff, is the defendant still entitled to qualified immunity?” Williams v. Strickland, 917 F.3d 763, 768 (4th Cir. 2019) (footnote omitted); see also Iko v. Shreve, 535 F.3d 225, 234 (4th Cir. 2008) (“[W]e possess no jurisdiction over a claim that a plaintiff has not presented enough evidence to prove that the plaintiff’s version of the facts actually occurred, but we have jurisdiction over a claim that there was no violation of clearly established law accepting the facts as the district court viewed them.” (quoting Winfield v. Bass, 106 F.3d 525, 530 (4th Cir. 1997) (en banc))).

10

A.

Our qualified-immunity inquiry proceeds in two steps, which we may address in whichever sequence “will best facilitate the fair and efficient disposition of [the] case.” Halcomb v. Ravenell, 992 F.3d 316, 319 (4th Cir. 2021) (citation omitted). First, a plaintiff must show a violation of a constitutional right. Thompson v. Virginia, 878 F.3d 89, 97 (4th Cir. 2017). And second, “the right at issue must have been ‘clearly established’ at the time of the defendant’s alleged misconduct.” Id. (quoting Pearson v. Callahan, 555 U.S. 223, 232 (2009)).

The source of the right at issue here is the Eighth Amendment, which prohibits the infliction of “cruel and unusual punishments.” U.S. Const. amend. VIII. This prohibition extends to “the treatment a prisoner receives in prison and the conditions under which he is confined.” Helling v. McKinney, 509 U.S. 25, 31 (1993). Because “adequate . . . medical care” is a basic condition of humane confinement, Farmer v. Brennan, 511 U.S. 825, 832 (1994), a prison official’s “deliberate indifference to serious medical needs of prisoners constitutes the unnecessary and wanton infliction of pain proscribed by the Eighth Amendment,” Estelle v. Gamble, 429 U.S. 97, 104 (1976) (citation and internal quotation marks omitted).

To establish a claim for deliberate indifference to serious medical needs, a prisoner “must satisfy the Supreme Court’s two-pronged test set forth in Farmer v. Brennan.” Scinto v. Stansberry, 841 F.3d 219, 225 (4th Cir. 2016). Farmer’s first, “objective” prong requires a plaintiff to prove that the alleged deprivation was “sufficiently serious.” Farmer, 511

11

U.S. at 834 (citation omitted). Farmer’s second, “subjective” prong requires a plaintiff to show that prison officials acted with “deliberate indifference.” Scinto, 841 F.3d at 225.

This second prong itself has two subparts: a plaintiff must show the prison official (1) had “actual knowledge of the risk of harm to the inmate” and (2) “‘recognized that his actions were insufficient’ to mitigate the risk of harm to the inmate arising from his medical needs.” Iko, 535 F.3d at 241 (emphasis omitted) (quoting Parrish ex rel. Lee v. Cleveland, 372 F.3d 294, 303 (4th Cir. 2004)). “[M]ere negligence” won’t meet this standard, De’lonta v. Angelone, 330 F.3d 630, 634 (4th Cir. 2003), but neither is “actual purposive intent” required, De’lonta v. Johnson, 708 F.3d 520, 525 (4th Cir. 2013). Instead, deliberate indifference is most akin to criminal-law recklessness. Campbell v. Florian, 972 F.3d 385, 395 (4th Cir. 2020), as amended (Aug. 28, 2020). Further, so long as the official who knew of a substantial risk to inmate health or safety “responded reasonably to the risk,” they cannot be found liable under the Eighth Amendment, “even if the harm ultimately was not averted.” Farmer, 511 U.S. at 844.

Even if a plaintiff proves an Eighth Amendment right was violated, our inquiry isn’t over. Under the second prong of the qualified-immunity analysis, an official is nonetheless entitled to immunity “if the right was not so ‘clearly established’ that ‘a reasonable official would understand that what he is doing violates that right.’” Thompson, 878 F.3d at 98 (quoting Hope v. Pelzer, 536 U.S. 730, 739 (2002)). The state of the law at the time must have given an official “fair warning” that his treatment of the prisoner was unconstitutional. Id. (quoting Hope, 536 U.S. at 741).

12

In performing this analysis, a court must pinpoint the precise constitutional right at issue in order to determine whether it was clearly established. See Halcomb, 992 F.3d at 319–20. In doing so, a court must be careful not to define the right “at a high level of generality ‘[b]ecause the dispositive question is whether the violative nature of particular conduct is clearly established.’” Id. at 320 (quoting Estate of Armstrong ex rel. Armstrong v. Village of Pinehurst, 810 F.3d 892, 907 (4th Cir. 2016)). Nonetheless, Eighth Amendment claims don’t require the same level of specificity that is needed in, for example, the Fourth Amendment context. See Thorpe v. Clarke, 37 F.4th 926, 940 (4th Cir. 2022). And “[t]here is no requirement that the ‘very action in question [must have] previously been held unlawful’ for a reasonable official to have notice that his conduct violated that right.” Scinto, 841 F.3d at 236 (quoting Hope, 536 U.S. at 739).

The extent of the analysis required for this second prong, however, varies considerably across cases. This Court recently held in Thorpe v. Clarke that “when ‘plaintiffs have made a showing sufficient to’ demonstrate an intentional violation of the Eighth Amendment, ‘they have also made a showing sufficient to overcome any claim to qualified immunity.’” 37 F.4th at 934 (quoting Beers-Capitol v. Whetzel, 256 F.3d 120, 142 n.15 (3d Cir. 2001)). That’s because, as Thorpe opined, “qualified immunity does not shield ‘those who knowingly violate the law.’” Id. (quoting Ashcroft v. al-Kidd, 563 U.S. 731, 743 (2011)). And because the Eighth Amendment’s deliberate-indifference standard requires knowing conduct, an official who was deliberately indifferent could not also believe “that [their] actions comported with clearly established law.” Id. at 939 (quoting Beers-Capitol, 256 F.3d at 142 n.15). Thus, in Thorpe, we “effectively collapse[d]”

13

qualified immunity’s two inquiries into one, holding that dismissal on qualified-immunity grounds “remains improper so long as the officers’ mental state remains genuinely in issue.” Id. at 934 (citation omitted).

We don’t believe, however, that Thorpe directs this Court to eliminate the clearly established law prong in every case where a dispute of fact related to the defendant’s mental state remains. 2 Critically, the Supreme Court has not so limited itself.

In Taylor v. Barkes, 575 U.S. 822 (2015) (per curiam), the Supreme Court reversed the lower courts’ denial of qualified immunity. The lower courts had held there were genuine issues of material fact as to whether there was an Eighth Amendment violation. Yet the Court concluded that, even if that were so, qualified immunity nevertheless attached because the right at issue was not clearly established. Id. at 827. Thus, we know from Taylor that not all disputes of material fact as to deliberate indifference will freeze our application of qualified immunity where warranted.

Instead, Eighth Amendment cases exist on a spectrum of intent and harm. Thorpe most neatly applies to a prison doctor, who doesn’t need case law to tell him his patient

2

Defendants argue that Thorpe only applies to cases at the motion-to-dismiss stage.

As discussed below, the lack of factual development at that stage does appear to have been relevant to the outcome in Thorpe. But Thorpe’s reasoning doesn’t hinge on its procedural posture. Moreover, this Court has already relied on Thorpe to vacate a grant of summary judgment where the district court found that the defendants may have intentionally discriminated against the plaintiff based on his religion yet nonetheless granted immunity because the defendants did not violate a clearly established right. Coleman v. Jones, No. 20-7382, 2022 WL 2188402, at *6 (4th Cir. June 17, 2022) (unpublished but orally argued) (“[W]hen . . . there is a triable issue as to whether defendants engaged in a constitutional violation that incorporates intent as an element, qualified immunity may be inappropriate.”).

14

deserves fair treatment. Or a prison guard, who doesn’t need case law to tell him he can’t abuse an inmate. Thorpe collapses qualified immunity’s two-pronged inquiry in these cases because each prong merely duplicates the other’s work; these Eighth Amendment violations inherently include knowing disregard for the law. See Thorpe, 37 F.4th at 938– 40.

But in other cases, there may be more attenuation between the risk of harm and the defendant’s knowledge that his conduct is constitutionally deficient. And in these cases, a defendant is less able to “use his own ‘state of mind’ as ‘a reference point’ to ‘assess conformity to the law,’” as Thorpe envisions. Id. at 939 (quoting Thompson, 878 F.3d at 106). Thorpe itself recognized this: “What Defendants actually demur is that they did not know, until [a later] decision . . . , that the . . . conditions they promulgated posed a substantial risk of serious harm in violation of the Eighth Amendment. They may well end up on the winning side of that argument after the evidence comes in, but for now, these fact[ual] issues compel us to move this case forward unless Defendants’ entitlement to qualified immunity appears on the face of the complaint.” Id. at 935 (emphasis added) (citations and internal quotation marks omitted).

This principle is illustrated by the very cases at hand—Thorpe on one end of the spectrum, Taylor on the other. In Thorpe, we found (and the defendants didn’t even challenge) that the plaintiffs adequately alleged that prison officials knew the harm that long-term solitary confinement caused, yet disregarded it. Id. at 933. We emphasized the officials’ consciousness of a serious risk of harm: that the plaintiffs suffered a range of severe physical and mental harms, such as schizoaffective disorder, psychosis, bouts of

15

disorientation, and severe weight loss; that for years, the defendants “had daily contact with [the p]laintiffs”; that the defendants had been pressured to abandon similar solitary- confinement systems several times before; and that numerous studies and judicial precedents had documented “the severe and often permanent damage caused by” such systems. Id. at 935–96. In short, the harm was “obvious,” id. at 934, and “the pre-existing law [was] not in controversy,” id. at 935 (quoting Ortiz v. Jordan, 562 U.S. 180, 190 (2011)).

Under those facts, which sufficiently alleged “an intentional violation of the Eighth Amendment,” we concluded that the plaintiffs “‘also made a showing sufficient to overcome any claim to qualified immunity.’” Id. at 934 (emphasis added) (quoting Beers- Capitol, 256 F.3d at 142 n.15). We relied on the Supreme Court’s decision in Ortiz v. Jordan, where the Court likewise declined to grant qualified immunity to officers who were deliberately indifferent to an inmate’s safety. See id. at 934–35. The facts in Ortiz, like those in Thorpe, demonstrated an obvious harm, where prison officials failed to distance the plaintiff from her assailant after they were “adequately informed” of a sexual assault. Ortiz, 562 U.S. at 191. Ortiz indicated that those facts plainly fell under the clearly established law that the Eighth Amendment protects prisoners from situations where an official knows they face a substantial risk of serious harm in the form of violence, but “disregards that risk by failing to take reasonable measures to abate it.” Id. at 190 (quoting Farmer, 511 U.S. at 847).

In Taylor, by contrast, the Supreme Court examined whether a prison implemented adequate suicide prevention protocols and concluded that the defendants were entitled to

16

qualified immunity because the right at issue was not clearly established. Unlike the defendants in Thorpe and Ortiz, the Taylor defendants had not “personally interacted with [the prisoner] or knew of his condition before his death.” Taylor, 575 U.S. at 824. A constitutional violation could arise, then, only if the prison’s suicide-screening measures were inadequate. Id. at 824–25. Yet as the Court opined, no case law actually identified what minimum procedures a prison had to use. Id. at 826–27. Thus, the Court held that “even if the [prison’s] suicide screening and prevention measures contained the shortcomings that [the plaintiffs] allege[d], no precedent . . . would have made clear to [the defendants] that they were overseeing a system that violated the Constitution.” Id. at 827.

Put another way, even if the defendants’ suicide-prevention protocols were wanting, they weren’t clearly unconstitutional. The defendants’ states of mind in crafting those policies, then, could not be a guiding compass to ensuring those policies met constitutional muster, unlike in Thorpe. See Thorpe, 37 F.4th at 939 (stating that precedent isn’t needed when an official can “use his own state of mind as a reference point to assess conformity to the law” (citation and internal quotation marks omitted)). In such a case, the “clearly established” prong of the qualified-immunity analysis continues to perform work independent of the “constitutional-violation” prong—again, unlike in Thorpe. And so, in cases like Taylor, even where they may be a dispute of material fact as to the underlying constitutional violation, a court may still inquire into whether a right was clearly established to determine if the defendants are entitled to qualified immunity.

To review, in order to defeat a defendant’s claim for qualified immunity, a plaintiff must show: (1) a violation of a constitutional right, which in the case of alleged deliberate

17

indifference to serious medical needs requires a showing that (A) the alleged deprivation was sufficiently serious (the objective prong) and (B) prison officials acted with deliberate indifference (the subjective prong), which in turn requires a showing that the prison official (i) had actual knowledge of the risk of harm and (ii) recognized that his actions were insufficient to mitigate that risk of harm; and (2) that the constitutional right was clearly established at the time of the defendant’s actions. If a plaintiff fails to satisfy either prong of the qualified-immunity test, the defendant is entitled to qualified immunity. But in some cases, like Thorpe, the court need not separately determine whether the constitutional right was clearly established if there remains a genuine issue of material fact as to an official’s deliberate indifference, because that potential deliberate indifference would, if established, necessarily include an awareness of the illegality of the defendant’s actions.

B.

We turn first to whether Dr. Wang, Pfaller’s primary physician, is entitled to qualified immunity under this two-step framework. We conclude that (1) Plaintiff has presented sufficient facts for a reasonable jury to find that Dr. Wang was deliberately indifferent to Pfaller’s serious medical needs and (2) Pfaller’s right to treatment was clearly established by 2015, the year of Dr. Wang’s first alleged violation. Thus, we affirm the district court’s denial of qualified immunity to Dr. Wang.

1.

Dr. Wang doesn’t contest Farmer’s first, objective prong that the alleged deprivation here was sufficiently serious. Nor could he. We have previously found that this

18

prong is satisfied by failure to treat an “inmate’s serious medical needs,” including, specifically, hepatitis C. Gordon v. Schilling, 937 F.3d 348, 356–57 (4th Cir. 2019).

Dr. Wang’s dispute instead centers on the subjective prong and whether he was deliberately indifferent to Pfaller’s hepatitis C. On this issue, the district court found that Plaintiff had established at least two genuine disputes of material fact. The first dispute concerns Dr. Wang’s failure to refer Pfaller for additional testing on two occasions. As the court recognized, it is undisputed that Pfaller’s bloodwork qualified him for additional testing three times: 1) on October 16, 2015, Pfaller’s FIB-4 score was 1.48, which exceeded the 1.45 cutoff for referral; (2) on July 12, 2017, Pfaller had a FIB-4 score of 1.46, which again exceeded the 1.45 cutoff; and (3) on May 7, 2018, Pfaller had an APRI of 0.55 and a FIB-4 of 2.18, both of which exceeded the criteria for additional testing. But Dr. Wang only referred Pfaller for a fibroscan after the last encounter. By then, it was much too late.

Drawing all reasonable inferences in Plaintiff’s favor, as we must at summary judgment, Knibbs v. Momphard, 30 F.4th 200, 213 (4th Cir. 2022), Plaintiff raises a genuine issue of material fact as to whether Dr. Wang was deliberately indifferent. Specifically, a reasonable jury could conclude that Dr. Wang knew Pfaller qualified for additional testing but declined to do anything about it. As Dr. Wang acknowledges, he understood that the Guidelines set the standard of care for the treatment of patients with chronic hepatitis C. The Guidelines explicitly and unambiguously indicated that Pfaller qualified on each occasion. It is a reasonable inference that Dr. Wang, as a staff physician, was aware of those Guidelines.

19

This inference is further bolstered by the fact that, although the relevant portions of the Guidelines did not change between June 2015 and May 2018, they were revised as a whole at least six times in that time period. Presumably, Dr. Wang would have received a copy of these revised guidelines each time; after all, he acknowledged the Guidelines set the standard of care for treating prisoners with hepatitis C. If we infer that Dr. Wang did his job and reviewed each copy carefully to ensure nothing important had changed, that means he had at least six opportunities to confirm the FIB-4 cutoff for additional testing was indeed 1.45. A reasonable jury could therefore infer that Dr. Wang knew the true APRI and FIB-4 cutoffs but nevertheless declined to refer Pfaller for testing for two and a half years.

Dr. Wang counters that he did not refer Pfaller because he made a simple mistake.

Specifically, he claims in his declaration that he “believed the inclusion number for the [FIB-4] was 1.50 when in fact it was 1.45.” J.A. 1954. For support, Dr. Wang points to the fact that once Pfaller’s FIB-4 score exceeded 1.5 on May 7, 2018, he referred him for a fibroscan. Dr. Wang argues that without any direct evidence to contradict his account, there is no genuine issue of material fact. At most, he contends, the facts show only negligence, which is not enough to state an Eighth Amendment violation.

Like the district court, we decline at this stage to accept Dr. Wang’s self-serving assertion as fact, for several reasons. First, Dr. Wang tries to bolster his claim of mistake by noting that as soon as Pfaller’s FIB-4 score exceeded 1.5 on May 7, 2018—the number he allegedly believed was the threshold—Dr. Wang referred him for testing. But as the district court observed, Pfaller’s May 7 APRI score also exceeded the cutoff for additional

20

testing. So, critically, we have no way of knowing which score actually influenced Dr. Wang’s referral, other than his own testimony. Moreover, the May 7, 2018, FIB-4 score was 2.18—far higher than the earlier scores of 1.48 and 1.46. A reasonable jury could infer that Dr. Wang opted to ignore the first two scores, despite knowing about the Guideline cutoff, but felt the third score was too high to reasonably ignore.

Next is Dr. Wang’s testimony itself. To accept that Dr. Wang’s repeated errors were at most mere negligence would require crediting his own self-serving assertion. Of course, a jury may ultimately choose to credit Dr. Wang’s assertion over Plaintiff’s circumstantial evidence that Dr. Wang was aware Pfaller qualified for further testing. But that decision rests on an evaluation of Dr. Wang’s credibility, which we are not permitted to weigh at the summary-judgment stage. See Hensley ex rel. North Carolina v. Price, 876 F.3d 573, 584 n.6 (4th Cir. 2017). And while self-serving affidavits offered by the non-movant can sometimes defeat summary judgment, e.g., Lovett v. Cracker Barrel Old Country Store, Inc., 700 F. App’x 209, 212 (4th Cir. 2017) (unpublished but orally argued), here it is the movant—Dr. Wang—who offers his own statements as the key evidence in support of summary judgment. That is insufficient. E.g., Knibbs, 30 F.4th at 222 (“Because [the defendant] is the moving party, we are constrained to assume that the jury will not credit his evidence and will instead accept the [plaintiff]’s proffered evidence on disputed fact questions. . . . [We would] contravene[] Rule 56 [if we were to] accept[] [the defendant]’s self-serving statements and read[] the evidence in the light most favorable to him.”).

21

Moreover, even if we did credit Dr. Wang’s self-serving assertion, it is not the panacea he believes it is. To accept Dr. Wang’s claim of mistake, we would need to assume that he did not verify the Guidelines criteria either time he reviewed Pfaller’s bloodwork, since the Guidelines explicitly set forth the 1.45 threshold for FIB-4 scores. But this, too, raises an issue of deliberate indifference, because “prison officials may not simply bury their heads in the sand and thereby skirt liability.” Makdessi v. Fields, 789 F.3d 126, 133 (4th Cir. 2015). For example, we have held that a prison guard may not avoid liability if he “refused to verify ‘underlying facts that he strongly suspected to be true’ [or] ‘declined to confirm inferences of risk that he strongly suspected to exist.’” Id. at 133–34 (quoting Brice v. Va. Beach Corr. Ctr., 58 F.3d 101, 105 (4th Cir. 1995)).

Here, it is undisputed that Dr. Wang knew Pfaller’s blood results fell right near the line of warranting further screening. Yet, by his own account, he failed to check his guess about their meaning against the Guidelines, despite knowing the increasing risk of liver disease that higher scores denoted.

Finally, Dr. Wang is incorrect when he argues that Plaintiff has failed to produce any evidence contradicting Dr. Wang’s account. To be sure, Plaintiff relies on circumstantial evidence to establish an issue of material fact as to Dr. Wang’s deliberate indifference. But that isn’t a problem. Our Court has consistently recognized that a plaintiff can show subjective knowledge through circumstantial evidence. See, e.g., id. at 133. After all, rarely will courts see a case where a defendant readily agrees that he was deliberately indifferent. And here, the fact alone that Dr. Wang failed to refer Pfaller for further testing when he demonstrated signs of an increasingly serious medical condition “raises an

22

inference [of] deliberate indifference.” Scinto, 841 F.3d at 232 (quoting Miltier v. Beorn, 896 F.2d 848, 853 (4th Cir. 1990)).

Our colleague in dissent chafes against the inferences for Pfaller that we must make at this stage. Instead, the dissent accepts one conclusion and one alone: that Dr. Wang made a mistake. To be sure, not just one mistake, but two mistakes, in two different years. Because Dr. Wang uttered it, the dissent concludes that it must be. And apparently it need not matter that we can only believe this conclusion if we believe Dr. Wang himself— undoubtedly a question weighing on his credibility and therefore the province of a jury alone. But the law here is simple: we cannot contravene Rule 56 by crediting Dr. Wang’s self-serving affidavit. See Knibbs, 30 F.4th at 222.

Beyond that most fundamental error, the dissent’s view comes with other hitches.

Take one: the dissent suggests that there’s no evidence Dr. Wang even read or knew the Guidelines’ cutoffs at all. See Dissent at 44 (arguing that the Guidelines’ existence is not evidence that Dr. Wang knew the correct thresholds for further testing or treatment). But even accepting that Dr. Wang had “access” to the Guidelines yet did not “actually use[] that information,” id. (quoting Danser v. Stansberry, 772 F.3d 340, 348 n.10 (4th Cir. 2014)), that too demonstrates an issue of deliberate indifference for the reasons stated earlier. That is, of course, if reasonable inferences are actually drawn as they should be at summary judgment. And surely Dr. Wang himself would not go so far as to argue cavalierly that he never even used the very Guidelines he testified were the Department’s standard of

23

care for treating hepatitis C. 3 See Farmer, 511 U.S. at 842 (explaining that the Eighth Amendment doesn’t allow prison officials “to take refuge in the zone between ignorance of obvious risks and actual knowledge of risks.” (citation and internal quotation marks omitted)).

The district court also found that Plaintiff established a genuine dispute of material fact as to whether Dr. Wang responded appropriately once he did finally request a fibroscan for Pfaller. As noted earlier, Dr. Wang requested a fibroscan about a week after Pfaller’s May 7 bloodwork came back. However, Dr. Wang did not definitively order a fibroscan until July 11—around two months later. In the interim, Pfaller visited Dr. Wang several times complaining of abdominal pain, swelling, a distended abdomen, and fluid retention. All of these conditions are symptoms of advancing liver disease. Even though Dr. Wang knew Pfaller “was on the list to be scheduled for a Fibroscan at VCU” and that no fibroscan had taken place, J.A. 1958, Dr. Wang “did not press to have the previously ordered test performed.” Pfaller, 2021 WL 1776189, at *7. By the time he did, it was too late. Based on these facts, we believe a reasonable jury could conclude that Dr. Wang knew about the risk of Pfaller’s advancing liver disease yet took no concrete actions between May and July 2018 to diagnose or treat the disease.

3

The dissent also blurs the deliberate-indifference standard with an intent-to-harm standard. But it wasn’t Plaintiff’s burden to show that Dr. Wang acted with hope that Pfaller actually be harmed. “[I]t is enough that the official acted or failed to act despite his knowledge of a substantial risk of serious harm.” Farmer, 511 U.S. at 842.

24

Dr. Wang’s arguments don’t persuade us otherwise. He argues that he took some steps to follow up with Pfaller during this time, such as ordering hepatitis C genotype testing and prescribing medication to alleviate Pfaller’s symptoms. While that is more than nothing, a jury might conclude that Dr. Wang, as a medical doctor, must have known these half measures “were insufficient to mitigate the [true] risk of harm” from liver cirrhosis and cancer. Iko, 535 F.3d at 241 (citation, internal quotation marks, and emphasis omitted); see also Heyer v. U.S. Bureau of Prisons, 849 F.3d 202, 211 (4th Cir. 2017) (“[T]he mere fact that prison officials provide some treatment does not mean they have provided ‘constitutionally adequate treatment.’” (quoting De’lonta, 708 F.3d at 526)).

Indeed, we rejected a similar set of facts in an analogy in De’lonta. There, we imagined a scenario where prison officials prescribed a painkiller to a prisoner after a fall, but the prisoner continued showing symptoms that “by all objective measure, [showed that he] require[d] evaluation for surgery.” De’lonta, 708 F.3d at 526. In that scenario, we wondered, “[w]ould prison officials then be free to deny him consideration for surgery, immunized from constitutional suit by the fact they were giving him a painkiller?” Id. The answer was obvious: we thought not. See id.

Likewise, here. Dr. Wang wasn’t off the hook because he prescribed some medication. It’s not a question of disagreement over Pfaller’s preferred care, as Dr. Wang unpersuasively argues. See Hixson v. Moran, 1 F.4th 297, 303 (4th Cir. 2021). Rather, it’s a question of whether Pfaller failed to receive adequate treatment at all.

The dissent repeatedly chastises this Court for interfering in a matter of “prison administration.” Dissent at 39, 47–48. But that’s not the issue in this appeal; instead, the

25

issue here concerns a prisoner’s constitutional right to adequate medical care. And while the dissent makes much of Virginia’s policy priorities, see id. at 49–51, it places considerably less emphasis on this constitutional guarantee.

Lest we forget, prisoners are entirely “dependent on the State for food, clothing, and necessary medical care.” Brown v. Plata, 563 U.S. 493, 510 (2011). And a prison that deprives prisoners of those needs “is incompatible with the concept of human dignity.” Id. at 511. We cannot turn a blind eye to such constitutional infirmities. Cf. id. (“Courts may not allow constitutional violations to continue simply because a remedy would involve intrusion into the realm of prison administration.”). In its rush to raise the alarm on a host of policy concerns, the dissent ignores this Court’s most basic role—and the limits of that role at summary judgment.

Based on the disputed issues of material fact here, we believe the appropriateness of Dr. Wang’s actions should be resolved by a jury.

2.

Having concluded that Plaintiff raises a genuine issue of material fact as to an Eighth Amendment violation, we turn to the other prong of the qualified-immunity analysis— whether Pfaller’s right to treatment was clearly established. Of course, if Dr. Wang’s conduct sufficiently resembles that of the defendants in Thorpe, the analysis of the two prongs collapses into one, and we must find that Pfaller has made a showing sufficient to overcome qualified immunity at this stage. See Thorpe, 37 F.4th at 934. But, while we believe the facts here more closely resemble Thorpe than Taylor, we need not expressly decide whether to collapse our qualified immunity inquiry as to Dr. Wang. That’s because,

26

either way, we reach the same outcome: Dr. Wang was on notice that he was violating Pfaller’s constitutional rights.

First, we must define the right. The district court held that the right at issue is “Pfaller’s Eighth Amendment right to receive adequate medical care and to be free from officials’ deliberate indifference to his known medical needs.” Pfaller, 2021 WL 1776189, at *10. This, Dr. Wang argues, is far too general. Instead, he proffers that the proper inquiry is “whether it would have been clear to Dr. Wang that he was providing inadequate medical care in violation of the Eighth Amendment by failing to schedule a specific type of follow- up test (fibroscan) for Pfaller until July 2018.” Wang’s Opening Br. at 24.

We disagree with Dr. Wang’s overly narrow articulation. First, his argument is premised almost entirely on direction from the Supreme Court regarding Fourth Amendment claims. But as we recognized in Thorpe, the Supreme Court has not demanded the same level of specificity in Eighth Amendment cases. Thorpe, 37 F.4th at 940; see also Hope, 536 U.S. at 738–39 (declining to inquire whether “the very action in question has previously been held” to violate the Eighth Amendment and rejecting qualified immunity because “the risk of harm is obvious”); Taylor v. Riojas, 141 S. Ct. 52, 54 (2020) (per curiam) (“Confronted with the particularly egregious facts of this case, any reasonable officer should have realized that Taylor’s conditions of confinement offended the Constitution.”).

As we explained in Scinto v. Stansberry, “[a] prisoner’s right to adequate medical care and freedom from deliberate indifference to medical needs has been clearly established by the Supreme Court and this Circuit since at least 1976.” Scinto, 841 F.3d at

27

236. As Pfaller’s treating physician, Dr. Wang was on notice at all relevant times—late 2015 through 2018—that he could not choose to delay or withhold treatment from Pfaller when he knew it was medically necessary and was, in fact, mandated by the very Guidelines he operated under. See id. (holding that it was clearly established in 2005 that denying a prisoner his prescribed insulin can violate the Eighth Amendment); Jehovah v. Clarke, 798 F.3d 169, 181–82 (4th Cir. 2015) (emphasizing that refusal to treat serious medical needs can constitute deliberate indifference); Smith v. Smith, 589 F.3d 736, 739 (4th Cir. 2009) (explaining that delay in treatment can violate Eighth Amendment).

Take Smith as one case that put Dr. Wang on notice. There, an inmate alleged that he was prescribed treatment by his doctor for a foot infection but was unable to receive any treatment because a nurse later ripped up the order slip that authorized it. Smith, 589 F.3d at 737. The district court held there was no constitutional violation, but we reversed. We explained that “mere delay or interference [with medical treatment] can be sufficient to constitute a violation of the Eighth Amendment.” Id. at 739. By destroying the means for the inmate to access treatment, the nurse ultimately deprived him of necessary care. See id.

After Smith, Dr. Wang was on notice that he could not refuse necessary medical care to Pfaller. In this case, his supervisor—the Department’s chief physician—had ordered through the Guidelines that Pfaller receive further testing when his FIB-4 score rose above a specific level. That level was high enough to indicate progression of his hepatitis C and worrying signs of liver disease. Yet Dr. Wang allegedly ignored this directive and delayed the necessary treatment—for two and a half years. Smith, along with our other Eighth Amendment cases, put him on notice that this denial of care was constitutionally deficient.

28

Dr. Wang counters that Smith “does not establish, much less clearly establish, that monitoring a patient’s chronic [hepatitis C virus] is unconstitutional.” Wang’s Reply Br. at 6–7. But clearly established law doesn’t require identical facts. None of the above cases, of course, deal with the precise issue of hepatitis C treatment. But requiring such specificity in the case of a treatment provider’s decisions would allow a doctor limitless opportunity to deny medical care unless the precise required treatment for a specific underlying illness had been addressed by our Court. We decline to distort qualified immunity into such an absolute immunity.

Likewise, Dr. Wang was on notice that providing some care—even if unreasonable or deficient—does not clear the constitutional bar. See Jehovah, 798 F.3d at 181–82 (establishing that providing “some treatment” but ignoring or failing to treat other serious symptoms can violate the Eighth Amendment); De’lonta, 330 F.3d at 635. Thus, he was on notice that failing to adequately respond to Pfaller’s signs of severe liver disease in 2018 could also be constitutionally deficient. The district court was correct to deny qualified immunity to Dr. Wang at this stage.

C.

Next, we address whether Dr. Amonette, the Department’s chief physician who designed the Guidelines, is entitled to qualified immunity. Given the difference in the roles that Dr. Amonette and Dr. Wang played, it is necessary to analyze them separately. See Thompson, 878 F.3d at 107. And ultimately, the two officials go on separate tracks.

We will assume without deciding that Dr. Amonette’s actions amounted to deliberate indifference and turn directly to the question of how to consider the clearly

29

established prong of the qualified-immunity analysis. We conclude that Dr. Amonette’s conduct appears closer to the defendants’ conduct in Taylor than in Thorpe, so we must proceed to our usual “clearly established” inquiry. And, after evaluating the case law, we conclude that Dr. Amonette was not on sufficient notice that he was violating a clearly established right.

To start, it bears emphasizing that “the lodestar for whether a right was clearly established is whether the law ‘gave the officials “fair warning” that their conduct was unconstitutional.’” Iko, 535 F.3d at 238 (quoting Ridpath v. Bd. of Governors Marshall Univ., 447 F.3d 292, 313 (4th Cir. 2006)). The district court defined the constitutional right at issue for Dr. Amonette in the same way it defined it as to Dr. Wang: “Pfaller’s Eighth Amendment right to receive adequate medical care and to be free from officials’ deliberate indifference to his known medical needs.” Pfaller, 630 B.R. at 215. But while the case law establishing that right sufficed to give Dr. Wang fair warning that he could not deny necessary medical treatment to Pfaller as his primary care provider, it was not enough to give Dr. Amonette fair warning that his system-wide treatment Guidelines for a new drug were constitutionally deficient. Those contexts differ.

Like the defendants in Taylor, Dr. Amonette never “personally interacted with”

Pfaller. Taylor, 575 U.S. at 824. He only learned of Pfaller’s medical condition shortly

30

before Pfaller’s cancer diagnosis. So Dr. Amonette’s conduct could have violated the Constitution only if the Guidelines themselves were inadequate. 4 On this question, we must remember qualified immunity’s purpose: it “gives government officials breathing room to make reasonable but mistaken judgments about open legal questions.” Ashcroft, 563 U.S. at 743. And here, there was—and remains—an open question as to what kind of treatment protocol for administering direct-acting antivirals is constitutionally sufficient in a prison system.

Dr. Amonette, for his part, created a system of prioritization where the sickest inmates received treatment first. Those that did not qualify for treatment nonetheless received continuous monitoring. At the time, the medical community and the Federal Bureau of Prisons agreed. They recognized that while immediate treatment of all hepatitis C patients was recommended, prioritization was reasonable where resources were limited.

The district court, of course, found there was a genuine issue of material fact as to whether the Department’s resources were actually limited. The limitation at issue here was the number of providers at the VCU clinic who could provide treatment to inmates. Dr. Amonette relied on VCU’s specialty providers because he believed it was not “clinically

4

To be sure, we do not hold that an official must have personally interacted with a prisoner to have violated the Eighth Amendment. In fact, we have expressly rejected that view in a case involving Dr. Amonette himself. See Gordon, 937 F.3d at 362 (“Put simply, Amonette may not escape liability by claiming that he did not know the identities of the inmates who would suffer under his policies.”). But the attenuation between Dr. Amonette’s conduct and the risk here is relevant to why we find this case more similar to Taylor than Thorpe and therefore why it is necessary to delve into the clearly established law analysis.

31

appropriate” to have the Department’s own primary care providers prescribe the new direct-acting antivirals “without specialty input.” J.A. 117.

Plaintiff’s expert, however, stated that the Department’s providers could have felt comfortable treating inmates with direct-acting antivirals. Even assuming that is true, we find it notable that this dispute over resources arises directly from the novelty of the drug and the open question of how to appropriately prescribe it. Under these facts, we decline to hold that Dr. Amonette was clearly on notice that he should have ordered the Department’s primary care providers to prescribe this novel treatment rather than referring patients to specialists for treatment.

Meanwhile, various Courts of Appeals opinions have cut different ways regarding whether similar treatment guidelines pass constitutional muster or violate clearly established law. Compare Atkins v. Parker, 972 F.3d 734, 739–40 (6th Cir. 2020) (holding that prison medical director was not deliberately indifferent for failing to provide direct- acting antivirals to all inmates), Hoffer v. Sec’y, Fla. Dep’t of Corr., 973 F.3d 1263, 1277 (11th Cir. 2020) (same), and Bernier v. Allen, 38 F.4th 1145, 1157 (D.C. Cir. 2022) (finding there was no clearly established right in 2015 that inmate should immediately receive new antiviral treatment), with Abu-Jamal v. Kerestes, 779 F. App’x 893, 900 (3d Cir. 2019) (holding that denying direct-acting antivirals to inmate could constitute an Eighth Amendment violation).

This varying case law shows the inherent gray area that Dr. Amonette was operating in. And “qualified immunity protects public officials from bad guesses in gray areas.” Durham v. Horner, 690 F.3d 183, 190 (4th Cir. 2012) (citation and internal quotation marks

32

omitted); see also Taylor, 575 U.S. at 825 (“We do not require a case directly on point, but existing precedent must have placed the statutory or constitutional question beyond debate.” (quoting Ashcroft, 563 U.S. at 741)). Tasked with designing a treatment regimen for a novel drug, Dr. Amonette wasn’t given notice by our case law that his ultimate choice—a system that prioritized treatment with direct-acting antivirals for the sickest patients while offering monitoring for others—was (as we have assumed) deficient. And the record, as the district court interpreted it, doesn’t sufficiently show that Dr. Amonette knew that the Guidelines, while wanting, violated the Eighth Amendment. “In short, even if the [Guidelines] contained the shortcomings that [Plaintiff] allege[d], no precedent on the books in [2015] would have made clear to [Dr. Amonette] that [he was] [designing] a system that violated the Constitution.” Taylor, 575 U.S. at 827.

That, of course, doesn’t mean that the novelty of a treatment regimen will always immunize an official’s bad guess. There are certainly situations where the risk of medical harm is obvious and a right to a specific treatment was clearly established. See Hope, 536 at 741 (“[O]fficials can still be on notice that their conduct violates established law even in novel factual circumstances.”). But we can’t say that’s the case here. A prisoner’s purported right not to be subjected to a treatment regimen that prioritized antiviral treatment to prisoners with the most advanced levels of fibrosis was not clearly established when Dr. Amonette designed the Guidelines in 2015.

III.

The final question posed in this appeal is whether Dr. Wang is entitled to sovereign immunity on Pfaller’s medical-malpractice claim. On appeal, Dr. Wang claims that he is

33

entitled to summary judgment on Plaintiff’s medical-malpractice claim because, as a Virginia state employee, he is shielded by derivative sovereign immunity. We agree.

In Virginia, to determine whether derivative sovereign immunity applies to a state employee, a court must “focus on four, non-exclusive factors.” Patterson v. City of Danville, 875 S.E.2d 65, 70 (Va. 2022). These factors are (1) “the function th[e] employee was performing”; (2) “the extent of the state’s interest and involvement in that function”; (3) “[w]hether the act performed involves the use of judgment and discretion”; and (4) “the degree of control and direction exercised by the state over the employee.” James v. Jane, 282 S.E.2d 864, 869 (Va. 1980). “Because this multi-factor test is broadly worded and capable of disparate applications,” courts should conduct a “fine-grained analysis that looks to prior applications involving fact patterns that most closely parallel the case before [them].” Patterson, 875 S.E.2d at 70 (citation and internal quotation marks omitted). And here, we are aided by a recent Supreme Court of Virginia decision that involves another medical-malpractice claim brought against none other than Dr. Wang himself. That decision, Patterson v. City of Danville, which arrived while this appeal was pending, largely controls the outcome here.

The first two factors—the nature of the function performed by the employee and the extent of the Government’s interest and involvement in that function—are usually considered together. See, e.g., Lohr v. Larsen, 431 S.E.2d 642, 644 (Va. 1993). In general, “if the function that a government employee was negligently performing was essential to a governmental objective and the government had a great interest and involvement in that function, those factors would weigh in favor of the employee’s claim of sovereign

34

immunity.” Id. In contrast, “if that function has only a marginal influence upon a governmental objective, and the government’s interest and involvement in that function are ‘slight,’” sovereign immunity is disfavored. Id. (quoting James, 282 S.E.2d at 870).

The district court found that both factors weighed against sovereign immunity in the present case. Specifically, the court noted that while Virginia “undoubtedly has an important interest [in] the provision of medical care to inmates, . . . the state’s ‘paramount’ interest at [the prison] is incarceration.” Pfaller, 2021 WL 1776189, at *12. Based on its reading of Virginia case law, the court concluded that “if the state’s paramount interest is not the provision of medical care, these factors weigh against sovereign immunity.” Id.

That reading was squarely rejected by the Supreme Court of Virginia’s subsequent decision in Patterson. In that case, the court considered whether derivative sovereign immunity extended to Dr. Wang as an employee at another, minimum-security facility. Patterson, 875 S.E.2d at 67. The court noted that Dr. Wang’s function—providing medical care to incarcerated patients—“was not simply a benevolent act of governmental grace.” Id. at 72. Rather, the state “had a constitutional and statutory duty to provide medical care to incarcerated patients” through physicians like Dr. Wang. Id. “In such circumstances, the governmental ‘interest and involvement’ is at its apogee.” Id. (citation omitted). So, the court concluded that “[t]here can be little doubt” that the first two factors of the test favored sovereign immunity. Id.

This case fares no different. Dr. Wang’s function at Pfaller’s prison was to provide medical care to incarcerated patients, just as it was in Patterson. Under federal and state law, Virginia is required to provide such care. See Va. Code Ann. § 53.1-32(A); Williams

35

v. Commonwealth, 810 S.E.2d 885, 888 (Va. 2017). In those circumstances, the state’s interest and involvement in that function “is at its apogee.” Patterson, 875 S.E.2d at 72. The district court erred by discounting that interest and focusing too heavily on the state’s “paramount” interest in incarceration. Under Patterson, the first two factors clearly weigh in favor of finding sovereign immunity.

The third factor—whether the act performed involves the use of judgment and discretion—is a closer call. If the challenged act involves “broad discretion,” that tends to “weigh heavily in favor of a government employee’s claim of immunity.” Lohr, 431 S.E.2d at 645. Conversely, if the employee is merely “performing a ministerial act,” they will likely be liable in negligence. Id.

The district court found this factor weighed against sovereign immunity because “Wang’s discretion was severely curtailed by the [Department] Guidelines.” Pfaller, 2021 WL 1776189, at *13. Specifically, it noted the Department “prescribed the course of monitoring and treatment for inmates with Hepatitis C” via the Guidelines, “and Wang’s role was to follow that prescription.” Id. So, it concluded that Dr. Wang was performing a primarily ministerial act, which is not shielded by sovereign immunity.

This analysis undercounts Dr. Wang’s discretion. Although his discretion was constrained by the Guidelines, it was not eliminated. Even if patients like Pfaller did not meet the APRI and FIB-4 cutoffs, Dr. Wang could still request a referral if he concluded that those patients were displaying other symptoms “suggestive of advanced liver disease.” J.A. 240. And as the district court recognized, in certain “emergency circumstances” Dr. Wang could circumvent the Guidelines criteria entirely. Pfaller, 2021 WL 1776189, at *13.

36

Decisions like these necessarily involve “the use of judgment and discretion.” Lohr, 431 S.E.2d at 645; see also Pike v. Hagaman, 787 S.E.2d 89, 93 (Va. 2016) (nurse “was exercising discretion in caring for” plaintiff even if discretion was cabined by doctor’s orders); cf. Patterson, 875 S.E.2d at 72 (noting that the Supreme Court of Virginia’s “cases uniformly emphasize the highly discretionary character of professional medical care”). Thus, while this factor tends to weigh against immunity, the district court overstated by how much.

That leaves the fourth factor: the degree of government control over the employee.

In general, “[a] high level of control weighs in favor of immunity; a low level of such control weighs against immunity.” Lohr, 431 S.E.2d at 646. At first glance, this fourth factor (government control) appears to be at odds with the third factor (employee discretion). However, the Supreme Court of Virginia has clarified that government control has “diminished relevance . . . in the context of a medical professional.” Patterson, 875 S.E.2d at 71. After all, “when a government employee is specially trained to make discretionary decisions”—like a medical professional is—“the government’s control must necessarily be limited in order to make maximum use of the employee’s special training and subsequent experience.” Lohr, 431 S.E.2d at 646. Therefore, when examining government control of state physicians, we are less concerned with the government’s control of the disputed action at issue (which necessarily involves the use of medical judgment), and more concerned with the government’s control of the physician’s work environment in general. See id. (examining whether the clinic controlled the patients seen by the provider, the equipment he used, and the procedures he could perform).

37

For example, in Patterson, the Supreme Court of Virginia found this factor weighed in favor of sovereign immunity because the evidence showed, among other things, that Dr. Wang (1) “had no control over the patients that he was obligated to treat”; (2) was required to use “City-owned medical equipment and supplies”; (3) “did not possess or control any of the medical records of his patients”; (4) “was governed by medical policies and procedures promulgated by the Virginia Board of Corrections”; and (5) was subject to supervision by the prison director. Patterson, 875 S.E.2d at 73.

In the instant case, the district court concluded that this factor weighed in favor of sovereign immunity for many of the same reasons: (1) Dr. Wang “cannot choose his own patients or refuse to see certain patients”; (2) “[a]ll the equipment he uses is provided by” the Department; and (3) he was required to follow the Department’s Guidelines when treating hepatitis C patients. Pfaller, 2021 WL 1776189, at *14. Given the similarities of these facts to those that the Supreme Court of Virginia found determinative in Patterson, we agree with the district court that this factor favors sovereign immunity.

Overall, three of the four factors strongly weigh in favor of sovereign immunity, and one only moderately weighs against it. Therefore, we conclude that the district court erred by rejecting Dr. Wang’s sovereign-immunity defense.

IV.

For the foregoing reasons, we affirm the district court’s denial of qualified immunity to Dr. Wang on the Eighth Amendment claim. But we must reverse the denial of qualified immunity to Dr. Amonette. Additionally, we reverse the denial of sovereign immunity to

38

Dr. Wang on the medical-malpractice claim. We remand for further proceedings consistent with this opinion.

AFFIRMED IN PART,

REVERSED IN PART,

AND REMANDED

39

WILKINSON, Circuit Judge, concurring in part and dissenting in part:

The plaintiff’s Eighth Amendment claim against Dr. Wang fails as a matter of law.

The undisputed facts show that Dr. Wang was not deliberately indifferent to Pfaller’s serious medical needs. Pfaller proffered no evidence that Dr. Wang “had actual knowledge” that not ordering fibroscan testing when Pfaller’s FIB-4 scores were 1.46 and 1.48 created “an excessive risk” to his health. Danser v. Stansberry, 772 F.3d 340, 347 (4th Cir. 2014). Nor was the risk “so obvious” that Dr. Wang must have known of it. Rish v. Johnson, 131 F.3d 1092, 1099 (4th Cir. 1997). Rather, all the evidence suggests that Dr. Wang acted in good faith and was, at worst, negligent. Because there is no evidence to the contrary, the “undisputed facts demonstrate” no violation of Pfaller’s constitutional rights. Winfield v. Bass, 106 F.3d 525, 533 (4th Cir. 1997) (en banc).

The Supreme Court has clearly instructed that Eighth Amendment deliberate-

indifference claims are the narrow exception to the general rule that federal courts should not second-guess the medical judgments of doctors in state prisons. The Court has stated in no uncertain terms that “an inadvertent failure to provide adequate medical care” or a physician’s “negligent” diagnosis or treatment does not violate the Eighth Amendment. Estelle v. Gamble, 429 U.S. 97, 105–06 (1976). The majority contravenes Estelle and collapses the distinction between deliberate indifference and negligence. In so doing, it further thrusts the federal courts into the minutiae of state prison administration to the detriment of the Constitution’s reservation of powers to the states and sound state policymaking.

40

When deliberate indifference bleeds into negligence, the losers will ultimately be inmates themselves. Inmates deserve decent medical care. Yet prison practice is not a desirable option for many physicians. Many doctors prefer to treat patients who have not been convicted of serious crimes. It would seem elemental that law should counter, not contribute to, these disincentives. Yet while doctors outside prison gates enjoy robust protections against negligent medical-malpractice suits, those treating the incarcerated face, as here, Eighth Amendment negligence claims in contravention of the deliberate- indifference standard. Virginia law, as the panel unanimously concludes, would not permit this suit to proceed. See Patterson v. City of Danville, 875 S.E.2d 65 (Va. 2022). Yet Virginia’s red light is superseded by the majority’s green.

Privileging prisoner suits in this manner will chase more willing or altruistic doctors away. It is sad indeed that Pfaller died. It would also be sad, and very wrong, if we were to lay his death at this doctor’s door. Diluting the deliberate-indifference standard, and running roughshod over Virginia law, not only fractures our federal system. It will consign inmates over time to ever more substandard medical care.

I.

We must journey to the headwaters of prisoner Eighth Amendment medical-care claims in order to appreciate how undesirably far from the Supreme Court’s teachings we have come. See Estelle v. Gamble, 429 U.S. 97 (1976). The language of limitation flows throughout the Estelle opinion. Estelle concluded that “deliberate indifference to serious medical needs of prisoners constitutes the unnecessary and wanton infliction of pain proscribed by the Eighth Amendment.” 429 U.S. at 104 (quotation marks and citation

41

omitted). But this does not mean “that every claim by a prisoner that he has not received adequate medical treatment states a violation of the Eighth Amendment.” Id. at 105. Rather, “an inadvertent failure to provide adequate medical care cannot be said to constitute an unnecessary and wanton infliction of pain or to be repugnant to the conscience of mankind.” Id. at 105–06 (quotation marks omitted). So, “a complaint that a physician has been negligent in diagnosing or treating a medical condition does not state a valid claim of medical mistreatment under the Eighth Amendment.” Id. at 106. “Medical malpractice” is not a “constitutional violation.” Id.

The facts of Estelle are instructive. “Even applying the[] liberal standards”

applicable to pro se complaints, the Court held that the plaintiff failed to state an Eighth Amendment claim against his treating physician. Id. at 107. The plaintiff’s doctors had diagnosed his injury and prescribed medication, but the plaintiff complained that he should have received additional diagnostic testing and treatment. Id. The Court disagreed, noting that “the question whether an X-ray or additional diagnostic techniques or forms of treatment is indicated is a classic example of a matter for medical judgment.” Id. The “medical decision not to order an X-ray, or like measures, does not represent cruel and unusual punishment” and is “[a]t most . . . medical malpractice.” Id. The Court held that the plaintiff’s complaint against his doctor did not survive a motion to dismiss, id. at 108— a conclusion that applies with special force here given that Pfaller needed to meet an even higher burden at summary judgment, see infra Part II. Thus, Estelle created the Eighth Amendment deliberate-indifference claim but, from the very beginning, limited such

42

claims to conduct that is far more egregious than mere negligence and indeed far more egregious than anything that took place in this case.

Estelle controls this decision.

II.

Our circuit has tried to give effect to Estelle and its progeny by establishing a “very high standard” that deliberate-indifference plaintiffs must meet to survive a defendant’s motion for summary judgment. Danser, 772 F.3d at 347 (quoting Grayson v. Peed, 195 F.3d 692, 695 (4th Cir. 1999)). The majority today ignores that standard, strays from Estelle and our caselaw, and collapses the distinction between deliberate indifference and negligence. In truth, the undisputed facts refute the suggestion that Dr. Wang was deliberately indifferent to Pfaller’s medical needs.

A.

Let’s give Estelle its due. To establish deliberate indifference, a plaintiff must “show[] that the prison official ‘kn[ew] of and disregard[ed] an excessive risk to inmate health or safety.’” Danser, 772 F.3d at 347 (quoting Farmer v. Brennan, 511 U.S. 825, 837 (1994)). This is a “very high standard,” Grayson, 195 F.3d at 695, “which requires that a plaintiff introduce evidence suggesting that the prison official had actual knowledge of an excessive risk” to the plaintiff’s health or safety, Danser, 772 F.3d at 347 (emphasis added). The official “must both be aware of facts from which the inference could be drawn that a substantial risk of serious harm exists, and he must also draw the inference.” Id. (quotation marks omitted). Crucially, a plaintiff’s “unsupported speculation” that the defendant was aware of the risk “is insufficient to create a disputed issue of material fact

43

for purposes of summary judgment.” Id. at 348 n.10; see Elliott v. Leavitt, 99 F.3d 640, 644 (4th Cir. 1996).

Rather, a plaintiff can show actual knowledge in either of two ways: The plaintiff can produce “direct evidence supporting a conclusion that the prison officials actually knew” of a risk of harm—for instance, “deposition testimony or affidavit[s].” Rish, 131 F.3d at 1099. Short of such direct evidence, the plaintiff can demonstrate knowledge through “circumstantial evidence” by showing that the “risk was so obvious that it can be inferred that the prison officials knew of it.” Id. (citing Farmer, 511 U.S. at 842); see Brice v. Va. Beach Corr. Ctr., 58 F.3d 101, 105 (4th Cir. 1995) (noting that the risk must be “so obvious . . . that the [officer] did know of it because he could not have failed to know of it.”). If the plaintiff cannot make at least one of these showings, his deliberate-indifference claim fails as a matter of law. See Rish, 131 F.3d at 1100–01; Danser, 772 F.3d at 348–49.

B.

The undisputed facts here negate any possibility of deliberate indifference and instead show Dr. Wang’s good faith. The plaintiff failed to produce even an iota of evidence that Dr. Wang was deliberately indifferent to Pfaller’s serious medical needs. He produced no direct evidence that Dr. Wang had actual knowledge of an excessive risk to his health. Nor was the risk so obvious that Dr. Wang must have known of it. All Pfaller can muster is “unsupported speculation” that Dr. Wang was aware of the risk, which “is insufficient to create a disputed issue of material fact for purposes of summary judgment.” Danser, 772 F.3d at 348 n.10.

44

Pfaller first alleges that Dr. Wang was deliberately indifferent in failing to order a fibroscan when Pfaller’s FIB-4 scores were 1.48 in October 2015 and 1.46 in July 2017. See J.A. 1954–57; Br. of Appellee at 10–11. He argues that Dr. Wang was deliberately indifferent because he ignored the VDOC guidelines recommending fibroscans for FIB-4 scores greater than 1.45. Br. of Appellee at 10–11; see J.A. 1954–55. But the VDOC guidelines are not evidence that Dr. Wang knew the correct thresholds and ignored them: “[D]efendants’ mere access to information [is] insufficient to show on summary judgment that defendants actually used that information.” Danser, 772 F.3d at 348 n.10 (citing Othentec Ltd. v. Phelan, 526 F.3d 135, 142 (4th Cir. 2008)). Pfaller produced no evidence that Dr. Wang “had actual knowledge” that not ordering a fibroscan when Pfaller’s FIB-4 scores were 1.48 and 1.46 would create an “excessive risk” to his health. Id. at 347.

Nor was this risk “so obvious” that Dr. Wang must have known about it. Rish, 131 F.3d at 1099. The lack of obvious risk here is apparent given the undisputed facts: Pfaller showed no symptoms of liver decompensation when these FIB-4 scores were calculated. See J.A. 1713–14. Dr. Wang was a family medicine doctor, not a specialist in HCV. Id. at 1496–97. And the actual VDOC FIB-4 cutoff (1.45) was very close to and easily confused with what Dr. Wang believed to be the cutoff (1.5), and a different VDOC HCV policy used 1.5 as its cutoff. See id. at 1954–55. This is scarcely the sort of mistake that was so obvious that Dr. Wang must have been aware of it.

Pfaller also argues that Dr. Wang exhibited deliberate indifference between May and July 2018, when he ordered a fibroscan but did not immediately follow up. Br. of Appellee at 10–11. But Pfaller has introduced no evidence that in so doing, Dr. Wang

45

“knew of and disregarded an excessive risk” to Pfaller’s health. Danser, 772 F.3d at 347 (quotation marks omitted and alterations adopted).

The following facts are undisputed: Dr. Wang calculated Pfaller’s FIB-4 score to be 2.18 on May 9, 2018 and ordered a fibroscan less than a week later, on May 14. J.A. 1958. There was no way for Dr. Wang to conduct a fibroscan himself; he had to request that a third party, VCU, schedule Pfaller for a fibroscan. Id. Dr. Wang saw Pfaller twice in June for symptoms of liver disease and prescribed him seven different medications to relieve those symptoms. Id. at 1958–59. At this time, Dr. Wang knew that Pfaller was already “on the list to be scheduled for a Fibroscan at VCU.” Id. at 1958. When Pfaller appeared in Dr. Wang’s clinic on July 11, 2018 “with complaints of fullness in his abdomen,” Dr. Wang followed up with VCU, ordering a fibroscan “asap.” Id. at 1959.

As a matter of law, these undisputed facts do not establish deliberate indifference.

There is no evidence that Dr. Wang actually knew that his course of action—continuing to treat Pfaller with medication while waiting for VCU to complete his request for a fibroscan—created an excessive risk to Pfaller’s health. Nor was this risk so obvious that Dr. Wang must have been aware of it: There is surely some reasonable period that a doctor might wait after ordering testing from a third party. That decision, like the question whether to order “additional diagnostic techniques” in the first place, “is a classic example of a matter for medical judgment.” Estelle, 429 U.S. at 107. Treating Pfaller with medication while waiting a month or two for VCU to conduct a fibroscan was a “medical judgment” that “does not represent cruel and unusual punishment.” Id.

46

C.

Even assuming, purely arguendo, that Dr. Wang had actual knowledge of a substantial risk to Pfaller’s health, no case can be made that Dr. Wang deliberately disregarded that risk. See Estelle, 429 U.S. at 104; Danser, 772 F.3d at 347. There is not a single shred of evidence that Dr. Wang had any animus or other motive to deliberately ignore Pfaller’s medical needs. Why would he? Did he possess some malevolent temperament? Did he bear some grudge against Pfaller? The majority does not and cannot suggest any such thing. Given the dearth of evidence of deliberateness in the record, no reasonable person could conclude that this element of the deliberate-indifference standard was met.

Far from deliberately ignoring Pfaller’s needs, the undisputed facts show that Dr.

Wang tried to treat Pfaller’s HCV and made, at worst, honest mistakes about the FIB-4 threshold for fibroscan testing. Dr. Wang did not order a fibroscan when Pfaller’s FIB-4 scores were below what he believed to be the threshold, but he did order one as soon as Pfaller received a score above that value. See J.A. 1958. Once again, when Pfaller presented with symptoms of liver disease in June 2018, Dr. Wang prescribed him seven different types of medication. Id. at 1958–59. And when VCU had not scheduled Pfaller’s fibroscan after two months, Dr. Wang demanded that he receive the test “asap.” Id. at 1959. A deliberately indifferent doctor would not have done any of this.

The majority repeatedly concedes that Dr. Wang provided Pfaller with medical care for his HCV but contends that this fact is irrelevant to whether he was deliberately indifferent. See Maj. Op. at 23–24. The majority’s position raises the unknowable question

47

of how much medical care is enough and abandons this whole area of law to indeterminacy. Often the best medical care lies in monitoring and observation, not in disturbing patient stasis. But given the majority’s rule, physicians will be tempted to overprescribe and overtest to avoid Eighth Amendment claims—the easiest path for minimizing liability. The majority ultimately substitutes judicial judgment for medical judgment. But our chambers are not a doctor’s office. We cannot don our white coats and apply our own preferred diagnostic and treatment regimens in place of physicians’ medical judgment. By forsaking Estelle’s high standard for deliberate-indifference claims, the majority encourages courts to venture into medical judgments that we are woefully unqualified to make.

Indeed, the majority’s opinion reads like a routine negligence case and of course, in such a case, the inferences at the summary-judgment stage belong to the non-moving party. What the majority neglects, however, is the fact that the value of any immunity and heightened standard of proof lies in resolving claims at an earlier rather than later stage of litigation. This whole dimension of the problem is ignored by my good colleagues. In addition to being wrong on the outcome, the majority is wrong on the process. By embracing routine full-dress litigation even in the face of so many undisputed material facts, it has doubled down on the difficulties of its position and further undermined the value that the deliberate-indifference standard was intended to have.

III.

A.

When federal courts gratuitously interfere with state prison administration, we damage the Constitution’s differentiation between the enumerated powers of the federal

48

government and the residual powers of the states. It is true that Article I, section 8 is a great charter of federal enumerated power, including the national economic engine that is the Commerce Clause. See U.S. Const., art. I, § 8. But at the same time, the “powers not delegated to the United States by the Constitution, nor prohibited by it to the States, are reserved to the States respectively, or to the people.” U.S. Const. amend. X. These residual powers must include something of meaning if our system of dual sovereignty is to be preserved and the states not relegated to the position of lesser sovereigns or inferior partners. As Justice Iredell noted long ago, “[e]ach State in the Union is sovereign as to all the powers reserved. . . . [T]he United States have no claim to any authority but such as the States have surrendered to them.” Chisholm v. Georgia, 2 U.S. (2 Dall.) 419, 435 (1793).

The often salutary growth of federal power since must not obscure the democratic virtues that state decision-making still retains. If the residual power of the states is to mean anything, it would seem to preserve a good amount of state autonomy over state institutions like schools and prisons. The administration of schools, for example, implicates highly local, policy-driven judgments about how the state “provides for the education of its children”—“an area in which [the Court] has traditionally deferred to state legislatures.” San Antonio Indep. Sch. Dist. v. Rodriguez, 411 U.S. 1, 40 (1973). State prisons and their medical facilities implicate the same basic local policy concerns. “[C]ourts are ill equipped to deal with . . . problems of prison administration,” which “require expertise, comprehensive planning, and the commitment of resources.” Procunier v. Martinez, 416 U.S. 396, 405 (1974). For this reason, “[t]raditionally, federal courts have adopted a broad hands-off attitude toward problems of prison administration.” Id. at 404. Operating prisons

49

and providing medical care to inmates are among the residual powers most clearly reserved to the states. The majority’s granular intrusion is, yet once again, exactly what Estelle v. Gamble sought to avoid.

Those who control the levers of federal power—be they legislators or judges—risk arrogating to themselves powers that our Constitution never meant them to have. I am disappointed that my esteemed colleagues in the majority have joined the company of federal overlords who view the states as little more than junior varsity sovereigns.

B.

This case is a perfect example of how a federal court can undermine a state’s policy decisions by allowing a suit for a constitutional violation to go forward where no such violation exists. The plaintiff in this case brought, in addition to his federal deliberate- indifference claim, a state medical-malpractice claim based on Dr. Wang’s alleged negligence in treating Pfaller’s HCV. J.A. 80–81. All members of this panel rightly agree that in light of the Virginia Supreme Court’s decision in Patterson v. City of Danville, 875 S.E.2d 65 (Va. 2022), Pfaller’s malpractice claim is barred by state sovereign immunity. I readily join in the majority’s fine opinion on this point. See Maj. Op. Part III.

Patterson is important because it demonstrates how Virginia conceives of its own interest in determining how to provide medical care to its prisoners. Virginia applies a four- factor test to determine whether a government employee is protected by sovereign immunity. Id. at 70. These factors are “(i) the nature of the function performed by the employee; (ii) the extent of the [state’s] interest and involvement in the function; (iii) the degree of control and direction exercised by the [government employer] over the employee;

50

and (iv) whether the act complained of involved the use of judgment and discretion.” Id. (quotation marks omitted). Crucially, the Patterson Court held that the state’s “governmental interest and involvement” in “provid[ing] medical care to incarcerated patients . . . is at its apogee.” Id. at 72 (quotation marks omitted). Given this fact, along with the discretionary nature of medical care and the degree of governmental control over the doctor, the Court held that the medical-malpractice suit was barred by state sovereign immunity. Id. at 73.

The Virginia Supreme Court has thus determined that medical-malpractice suits alleging a doctor’s negligence implicate the state’s weighty interest in providing medical care to prisoners and are therefore barred by sovereign immunity. One might disagree with Virginia’s invocation of sovereign immunity in these circumstances, but mere disagreement with a state policy does not license a federal court to cavalierly supersede it in the name of enforcing a federal right.

That is precisely what the majority does today. In one breath, the majority rejects Pfaller’s state medical-negligence claim on the basis of sovereign immunity. With the next, the majority allows a federal claim predicated only on negligent acts to proceed in the guise of a deliberate-indifference claim. The majority effectively supersedes the state’s judgment that sovereign immunity ought to bar medical-negligence suits. It accomplishes the precise end that the state sought to avoid: subjecting state prison doctors to suit for what are essentially negligence allegations. This case, like all too many others, is a classic example of how federal constitutional activism has wrested from the states control over their own institutions.

51

The consequences will be severe. Virginia’s interest in claiming sovereign immunity from medical-malpractice suits is grounded in legitimate concerns about its prison doctors being inundated with such claims. The majority invites a big increase in lawsuits that threaten to lodge in federal judges the power to second-guess state prison doctors’ exercise of “medical judgment.” Estelle, 429 U.S. at 107. Under the majority’s rule, a plaintiff offering nothing more than mere speculation that a doctor was deliberately indifferent can overcome the doctor’s immunity and force him to stand trial.

It does not take an economist to predict what will happen when a state’s prison doctors are hauled into federal court for any perceived deficiency in medical care and branded with the stigma of deliberate indifference for honest mistakes: Good doctors will be strongly discouraged from providing their services in state prisons. Thus, unwarranted federal interference with state prisons works not only to the detriment of our federal system, but to the well-being of prisoners themselves.

IV.

The moral of this case is simple. In the words of the great Scottish poet Robert Burns, “The best-laid schemes o’ Mice an’ Men gang aft agley.” Robert Burns, To a Mouse (1785). The Supreme Court, with the best of intentions, sought to crack the door a little bit to a limited category of prisoner medical-mistreatment claims where the high standard of deliberate indifference was met. But those best intentions have gone badly awry. Over time,

52

what the Supreme Court intended as a small crack of the door has been flung wide by the lower federal courts. This case and Estelle v. Gamble bear scant resemblance.

Not only that. The “best-laid” plans of the majority will “gang aft agley” as well, as its admirable desire for adequate medical care for inmates will likely occasion just the opposite.

This did not need to be. That the majority bumps so squarely against both constitutional law and the law of unintended consequences should have given it pause. I do concur in Parts II.C and III of its opinion. I respectfully dissent in part and would reverse the judgment below in its entirety and remand with instructions to dismiss the complaint.