Vance v. Aldrich (10th Cir. 2026): Generally Applicable “Red-Flag” Non-Enforcement Policies Are Not Affirmative State-Created Danger; Colorado CPLA Exclusively Governs On-Premises Mass-Shooting Injury Claims

1. Introduction

This consolidated appeal arises from the November 19, 2022 mass shooting at Club Q in Colorado Springs. Plaintiffs—survivors and certain estates—sued (1) El Paso County and its Sheriff (the “County Defendants”) and (2) entities and individuals affiliated with Club Q (the “Club Q Defendants”).

Against the County Defendants, Plaintiffs asserted substantive due process claims under 42 U.S.C. § 1983 using the Tenth Circuit’s “state-created-danger” doctrine, alleging County policies effectively prohibited law enforcement from seeking Extreme Risk Protection Orders (ERPOs) under Colorado’s “Red Flag Law,” thereby increasing vulnerability to private violence. Against the Club Q Defendants, Plaintiffs asserted Colorado claims under the Colorado Premises Liability Act (CPLA) and also pleaded negligence and wrongful-death theories based on asserted security and egress failures.

The appeal presented four key issues: (1) whether the County Defendants engaged in “affirmative conduct” sufficient for state-created-danger liability; (2) whether the district court had supplemental jurisdiction over the Club Q state-law claims; (3) whether the CPLA claims plausibly alleged causation after Colorado’s 2022 CPLA amendment responding to Rocky Mountain Planned Parenthood, Inc. v. Wagner, 467 P.3d 287 (Colo. 2020); and (4) whether the CPLA preempted negligence and wrongful-death claims.

2. Summary of the Opinion

3. Analysis

A. Precedents Cited

1) State-created danger and the “affirmative conduct” barrier

• DeShaney v. Winnebago Cnty. Dep't of Soc. Servs., 489 U.S. 189 (1989): Anchors the baseline rule that the Due Process Clause generally does not impose a duty to protect against private violence; state-created danger is an exception.
• Matthews v. Bergdorf, 889 F.3d 1136 (10th Cir. 2018): Supplies the Tenth Circuit’s two exceptions (special relationship; state-created danger) and enumerates the six-factor test used after the affirmative-conduct/private-violence preconditions.
• Est. of B.I.C. v. Gillen, 761 F.3d 1099 (10th Cir. 2014) and Est. of B.I.C. v. Gillen, 710 F.3d 1168 (10th Cir. 2013): Reiterate that actionable conduct must surpass “mere negligence or inaction.”
• Gray v. Univ. of Colo. Hosp. Auth., 672 F.3d 909 (10th Cir. 2012): Central to the holding. Establishing generally applicable policies/customs does not impose an “immediate threat of harm” with “limited range and duration,” and must be directed at a discrete plaintiff rather than the public at large.
• Ruiz v. McDonnell, 299 F.3d 1173 (10th Cir. 2002) and Robbins v. Oklahoma, 519 F.3d 1242 (10th Cir. 2008): Improper licensure / failure to ensure safety create an “indefinite” risk rather than an “immediate” threat; reinforce limits on policy-level causation theories.
• Graham v. Indep. Sch. Dist. No. I-89, 22 F.3d 991 (10th Cir. 1994): Knowledge of threats plus failure to act is not affirmative conduct.
• Armijo ex rel. Chavez v. Wagon Mound Public Schools, 159 F.3d 1253 (10th Cir. 1998): Illustrates affirmative conduct when officials leave a specific student alone with access to firearms shortly before harm—used to distinguish immediacy and plaintiff-specific targeting.
• Uhlrig v. Harder, 64 F.3d 567 (10th Cir. 1995): Considered risk issues in a workplace violence setting; here used to show it did not decide the “immediacy” point Plaintiffs sought.
• Currier v. Doran, 242 F.3d 905 (10th Cir. 2001): Plaintiffs relied on its “safety valve” language; the panel distinguished it because the state action there specifically and immediately placed identifiable children at risk and foreclosed aid.
• Out-of-circuit comparisons were addressed but treated as non-controlling in a doctrine without a uniform national test, including Dwares v. City of New York, 985 F.2d 94 (2d Cir. 1993), Pena v. Deprisco, 432 F.3d 98 (2d Cir. 2005), Kallstrom v. City of Columbus, 136 F.3d 1055 (6th Cir. 1998), Freeman v. Ferguson, 911 F.2d 52 (8th Cir. 1990), Okin v. Vill. of Cornwall-on-Hudson Police Dep't, 577 F.3d 415 (2d Cir. 2009), and Pauluk v. Savage, 836 F.3d 1117 (9th Cir. 2016).

2) Supplemental jurisdiction: confirming § 1367(a) before declining under § 1367(c)

• Carlsbad Tech., Inc. v. HIF Bio, Inc., 556 U.S. 635 (2009): Declining supplemental jurisdiction under § 1367(c) is discretionary, not jurisdictional—once § 1367(a) exists.
• Royal Canin U.S.A., Inc. v. Wullschleger, 604 U.S. 22 (2025): There is no discretion to decline supplemental jurisdiction when there is no supplemental jurisdiction to begin with—courts must determine § 1367(a) first.
• City of Albuquerque v. Soto Enters., Inc., 864 F.3d 1089 (10th Cir. 2017): In the Tenth Circuit, a court must have subject-matter jurisdiction and then choose to exercise it.
• United Mine Workers v. Gibbs, 383 U.S. 715 (1966) and City of Chicago v. Int'l Coll. of Surgeons, 522 U.S. 156 (1997): “Common nucleus of operative fact” is the constitutional and statutory touchstone for supplemental jurisdiction.
• Pettigrew v. Okla. ex rel. Okla. Dep't of Pub. Safety, 722 F.3d 1209 (10th Cir. 2013): Articulates the common nucleus test.
• Estate of Harshman v. Jackson Hole Mountain Resort Corp., 379 F.3d 1161 (10th Cir. 2004): Although ultimately about lack of original jurisdiction, it supported in dicta that claims arising from the same accident typically share a common nucleus.
• Additional “common nucleus” support: King Fisher Marine Serv., Inc. v. 21st Phx. Corp., 893 F.2d 1155 (10th Cir. 1990), Sinclair v. Soniform, Inc., 935 F.2d 599 (3d Cir. 1991), Feigler v. Tidex, Inc., 826 F.2d 1435 (5th Cir. 1987). Contrasts drawn with Joyner v. Morrison & Foerster LLP, 140 F.4th 523 (D.C. Cir. 2025) and discussion of Nanavati v. Burdette Tomlin Memorial Hospital, 857 F.2d 96 (3d Cir. 1998).

3) Colorado Premises Liability Act: causation after the 2022 amendment; exclusivity

• Jordan v. Panorama Orthopedics & Spine Ctr., PC, 346 P.3d 1035 (Colo. 2015): Defines the CPLA’s scope—liability in the defendant’s legal capacity as landowner for “conditions, activities, and circumstances” on the property.
• Rocky Mountain Planned Parenthood, Inc. v. Wagner, 467 P.3d 287 (Colo. 2020): Majority allowed a jury to find causation despite mass-shooter conduct; dissent argued shooter predominance defeats landowner causation as a matter of law. The 2022 amendment repudiates aspects of the majority’s analysis.
• City of Aurora v. Loveless, 639 P.2d 1061 (Colo. 1981): General proposition that causation is typically for a jury (but not immune from pleading standards).
• Taco Bell, Inc. v. Lannon, 744 P.2d 43 (Colo. 1987): Discussed security duties in a pre-CPLA context; the panel distinguished it as predating the CPLA and the 2022 amendment.
• Garcia v. Colorado Cab Co., 538 P.3d 328 (Colo. 2023): General proximate cause principles outside the CPLA; distinguished.
• Castaldo v. Stone, 192 F. Supp. 2d 1124 (D. Colo. 2001), Ireland v. Jefferson Cnty. Sheriff's Dep't, 193 F. Supp. 2d 1201 (D. Colo. 2002), Phillips v. Lucky Gunner, LLC, 84 F. Supp. 3d 1216 (D. Colo. 2015), Nowlan v. Cinemark Holdings, Inc., 2016 WL 4092468 (D. Colo. June 24, 2016): District court authorities reflecting skepticism that premises-related negligence can be a “substantial factor” relative to a mass shooter; acknowledged as not all directly on-point but consistent with the panel’s reading of the amendment.
• Macomber v. Nations Roof, LLC, 574 P.3d 757 (Colo. App. 2025), Vigil v. Franklin, 103 P.3d 322 (Colo. 2004), Larrieu v. Best Buy Stores, L.P., 303 P.3d 558 (Colo. 2013): Stand for the CPLA’s exclusivity when it applies, and that courts can decide applicability under ordinary dispositive standards.
• Tancrede v. Freund, 401 P.3d 132 (Colo. App. 2017), Tolle v. Steeland, LLC, 541 P.3d 24 (Colo. App. 2023), Grizzell v. Hartman Enters., 68 P.3d 551 (Colo. App. 2003): Confirm the CPLA’s broad reach, including safety measures and inadequate security as property conditions/circumstances.

B. Legal Reasoning

1) § 1983: why the ERPO “non-enforcement” policy did not satisfy state-created danger

The panel accepted that adoption/maintenance of the County’s ERPO policies could be “more than” pure inaction, but it held Plaintiffs still failed the affirmative-conduct precondition because the conduct must (i) impose an immediate threat of harm with limited range/duration and (ii) be directed at a discrete plaintiff rather than the public at large, per Gray v. Univ. of Colo. Hosp. Auth..

Two failures were dispositive:

• No immediate threat: The ERPO policies were generally applicable and created, at most, an “indefinite risk” of increased gun violence, with the shooting occurring years after the policies’ adoption (analogized to “improper licensure” in Ruiz v. McDonnell).
• Not plaintiff-directed: The policies affected the public broadly (anyone vulnerable to gun violence in the county). Plaintiffs were not “identifiable victims” at the time of policy adoption, echoing reasoning in Hernandez v. Ridley.

Currier v. Doran did not help Plaintiffs because that case involved state action that immediately and specifically increased risk to identifiable children and foreclosed aid. Here, the “safety valve” (ERPO mechanism) theory remained too generalized and temporally diffuse to satisfy the Tenth Circuit’s affirmative-conduct constraint.

2) Supplemental jurisdiction: “one constitutional case” despite different defendants and duties

On 28 U.S.C. § 1367, the panel clarified a sequencing principle: a court may decline under § 1367(c) only if supplemental jurisdiction exists under § 1367(a), relying on Royal Canin U.S.A., Inc. v. Wullschleger and City of Albuquerque v. Soto Enters., Inc..

Applying Gibbs, the panel found a “common nucleus of operative fact” because both the federal and state claims arose from the same mass shooting and its lead-up. Differences in defendants, duties, evidence, and timeframes did not defeat the nucleus where the operative event and injuries were shared—consistent with the panel’s reading of Estate of Harshman v. Jackson Hole Mountain Resort Corp. (dicta) and other multi-defendant “single event/project” cases.

3) CPLA causation after Colorado’s 2022 amendment responding to Wagner

The panel treated the 2022 amendment, Colo. Rev. Stat. § 13-21-115(2)(e), as a legislative repudiation of the Wagner majority’s approach to (a) foreseeability tied to “controversial” goods/services and (b) substantial-factor causation “without considering” whether third-party criminal acts were the predominant cause, “as noted by the dissenting justices.”

Critically, the panel read the amendment as aligning (at least strongly) with the Wagner dissent’s premise that mass shooters’ premeditated and intentional acts are ordinarily the predominant cause of victims’ injuries—often rendering a landowner’s alleged security deficiencies legally insignificant in the “substantial factor” calculus.

Although the panel did not announce an absolute rule that landowner causation can never be pleaded in a mass shooting, it held that—on these pleadings—Club Q’s alleged security, screening, training, staffing, and egress failures could not plausibly outweigh the shooter’s predominance, especially given the close factual parallels to the security allegations described in Rocky Mountain Planned Parenthood, Inc. v. Wagner.

4) CPLA exclusivity: why negligence and wrongful-death claims had to be dismissed

Independently of causation, the panel held the CPLA displaced Plaintiffs’ negligence and wrongful-death claims because the alleged injuries occurred on the property and “by reason of” conditions/circumstances/activities on the property—including security and safety measures—placing the case within the CPLA’s exclusivity clause, Colo. Rev. Stat. § 13-21-115(3).

Relying on Macomber v. Nations Roof, LLC, Vigil v. Franklin, and Larrieu v. Best Buy Stores, L.P., the panel rejected the district court’s view that preemption was too “fact-intensive” for the pleadings. The pleaded duty (“ensure their safety and well-being while on the premises”) matched the landowner-duty framework, confirming CPLA displacement at the Rule 12 stage.

C. Impact

1) Constitutional tort litigation over “red flag” enforcement choices

The decision reinforces (and arguably hardens) two limiting principles that make state-created-danger claims difficult in policy-driven gun-violence contexts: (1) generally applicable governmental policies rarely qualify as “affirmative conduct” because they create indefinite rather than immediate risk; and (2) plaintiffs must be a discrete, identifiable target, not part of the general public endangered by broader policy choices.

Practically, plaintiffs challenging systemic non-enforcement of public-safety tools (including ERPO regimes) will face dismissal unless they can plead conduct tightly linked in time and scope to specific victims (or a narrowly defined, identifiable group) and showing the state actor’s action—not omission—imposed an immediate, limited threat.

2) Colorado premises liability in mass-shooting cases post-2022 amendment

The panel’s reading of § 13-21-115(2)(e) signals that federal courts applying Colorado law may treat mass-shooter acts as presumptively predominant at the pleading stage, making CPLA causation difficult to allege absent unusually strong, specific facts showing the landowner’s conduct plausibly rivaled the shooter’s predominance.

Even more concretely, the opinion underscores that plaintiffs cannot avoid that problem by relabeling premises-security failures as common-law negligence or wrongful death: the CPLA remains the exclusive remedy when the injury is “by reason of” property conditions/circumstances/activities.

3) Federal jurisdiction practice

On § 1367, the opinion is a procedural corrective: district courts must confirm § 1367(a) before declining under § 1367(c). Substantively, it reflects a broad—but orthodox—view that a single catastrophic event can supply the “common nucleus” even where defendants and legal duties differ.

4. Complex Concepts Simplified

• State-created danger: A narrow due-process doctrine allowing liability when the state affirmatively creates or increases danger from private violence. In this circuit, it is not enough that officials failed to protect; plaintiffs must show affirmative conduct that created an immediate, plaintiff-specific risk.
• “Affirmative conduct” (vs. inaction): Not merely a decision or policy; it must functionally impose an immediate threat (limited in range/duration) aimed at a discrete plaintiff, not the public generally (Gray v. Univ. of Colo. Hosp. Auth.).
• Supplemental jurisdiction: Federal courts may hear related state-law claims if they are part of the same Article III “case or controversy” (share a “common nucleus of operative fact,” Gibbs). Only after that jurisdiction exists may a court decide whether to keep or decline the state claims.
• CPLA “predominant cause”: Colorado’s 2022 amendment instructs courts to consider whether a third-party criminal act was the predominant cause of harm, repudiating aspects of the Wagner majority and aligning with concerns expressed in the dissent that mass-shooter intent commonly dwarfs premises-security negligence in causation analysis.
• CPLA exclusivity / preemption: If the CPLA applies (injury on property because of property conditions/circumstances/activities), it displaces common-law negligence and similar claims against landowners.

5. Conclusion

Vance v. Aldrich delivers two particularly consequential rules. First, in the Tenth Circuit, policy-level decisions about public-safety tools (here, countywide ERPO non-use policies) generally do not satisfy the “affirmative conduct” requirement of the state-created-danger doctrine because they are not immediate, limited threats aimed at discrete plaintiffs. Second, under Colorado’s post-2022 CPLA framework, mass-shooter conduct will often be treated as the predominant cause of on-premises shooting injuries at the pleading stage, while the CPLA simultaneously forecloses attempts to plead around that barrier through negligence or wrongful-death theories due to its exclusivity.