Thus, on remand, the Estate must prove that one or more officers acted with deliberate indifference to Thao’s serious medical needs and that GCCJA’s training failures were a moving force behind that violation.

IV. Detailed Legal Analysis

A. Municipal liability under § 1983: the Monell framework

Under Monell v. Dep’t of Soc. Servs., 436 U.S. 658 (1978), a municipality (or applicable governmental entity such as a public trust) cannot be held liable under § 1983 solely on a respondeat superior theory; it is liable only for its own policies, customs, or decisions that cause a constitutional deprivation. As the Tenth Circuit summarized:

• A plaintiff must show:
  1. execution of a policy or custom, and
  2. a direct causal link between that policy or custom and a constitutional injury.
• Municipal “policy or custom” can take many forms, including:
  • a formal regulation or policy statement,
  • a widespread informal custom,
  • decisions by officials with final policymaking authority,
  • ratification by such policymakers, or
  • a failure to train or supervise employees, where that failure reflects deliberate indifference.

The court’s opinion is especially useful in clearly distinguishing two different types of Monell theories:

1. Facially unconstitutional policy – When an official municipal policy “itself violates federal law,” proving the existence of that policy—and that it was applied to the plaintiff—“puts an end to the question” of municipal culpability. There is no need to separately prove “deliberate indifference.”
2. Policy of inaction / failure‑to‑train, supervise, or hire – When a municipality is alleged to have failed to provide adequate training, supervision, or hiring practices, the plaintiff must establish a “rigorous standard of culpability”: deliberate indifference to a known or obvious risk of constitutional harm, typically shown by a pattern of similar violations or, in rare cases, by a “patently obvious” single‑incident scenario.

This distinction, grounded in Supreme Court cases such as Board of the County Comm’rs of Bryan County v. Brown, 520 U.S. 397 (1997), and City of Canton v. Harris, 489 U.S. 378 (1989), is central to the court’s analysis of both claims.

B. Excessive force and facially constitutional policies

1. Eighth Amendment excessive force standard

Because Thao was a convicted prisoner (as opposed to a pretrial detainee), his excessive‑force claim arises under the Eighth Amendment. The court applied the classic standard from Whitley v. Albers, 475 U.S. 312 (1986), and Hudson v. McMillian, 503 U.S. 1 (1992), as summarized in Giron v. Corr. Corp. of Am., 191 F.3d 1281 (10th Cir. 1999) and Redmond v. Crowther, 882 F.3d 927 (10th Cir. 2018):

• Objective prong: Was the force objectively harmful enough to amount to a constitutional violation?
• Subjective prong: Did the officer act “maliciously and sadistically for the very purpose of causing harm,” or in a good‑faith effort to maintain or restore discipline?

The Tenth Circuit explicitly declined to decide whether Officer Henneman’s tasing Thao while handcuffed in the elevator violated this standard, because the only claim on appeal against GCCJA was predicated on an alleged municipal policy, not on any pending claim against the individual officer.

2. The Monell question: Is GCCJA’s Taser policy facially unconstitutional?

The Estate’s theory was that GCCJA’s written policy “authorized” the tasing of a small, handcuffed inmate surrounded by six officers, and thus the policy itself was unconstitutional. The district court incorrectly analyzed this as a failure‑to‑train/deliberate‑indifference issue. The Tenth Circuit corrected that framing:

• For an affirmative policy claim, the question is whether the policy, on its face, directs or authorizes conduct that violates constitutional standards. If so, no further inquiry into the municipality’s mental state is required.
• Deliberate indifference comes into play only where the alleged policy is one of inaction (e.g., failure to train, supervise, or discipline), not where the policy is a straightforward rule governing officer conduct.

GCCJA’s policy provided that:

Detention officers [are authorized] to use only the force which is reasonably necessary to achieve lawful objectives, defend themselves or others from physical harm, to prevent escapes, and then only as a last resort, and overcome resistance. In no event is physical force used as punishment. … [A Taser] will not be deployed on a handcuffed individual without articulable extenuating circumstances.

The court reasoned:

• The policy on its face limits use of force. It calls for reasonable, non‑punitive force and restricts Taser deployment on handcuffed individuals to situations with “articulable extenuating circumstances.”
• Nothing in the policy instructs officers to use force “maliciously and sadistically for the very purpose of causing harm.”
• Thus, in the sense used in Christensen v. Park City Municipal Corp., 554 F.3d 1271 (10th Cir. 2009), the policy’s “straightforward enforcement” does not violate the Constitution; rather, it appears designed to comply with constitutional standards.

Crucially, even if GCCJA admitted in discovery that staff “acted consistent with official written policy or practices” in tasing Thao, that admission:

• Was made in the context of GCCJA’s view that extenuating circumstances justified the Taser use; and
• Could, at most, support a custom or failure‑to‑train theory about how the policy is applied—not override the unambiguous text of the formal written policy.

The court noted that an admission in discovery can be powerful evidence for other Monell theories (for example, a widespread practice undercutting written rules), but it does not transform a facially constitutional policy into a facially unconstitutional one.

Because the Estate chose to premise its municipal liability claim only on the formal policy’s alleged facial invalidity, and did not pursue an informal custom or failure‑to‑train theory concerning the use of Tasers, the court held that GCCJA was entitled to summary judgment on the excessive‑force claim.

C. Failure‑to‑train and suicide prevention: the single‑incident doctrine

1. Eighth Amendment deliberate‑indifference standard for medical care and suicide

Prison officials have a constitutional duty to “take reasonable steps to protect prisoners’ safety and bodily integrity,” including providing necessary medical and mental‑health care. In suicide cases, courts treat the claim as a failure to provide adequate medical care. Under Estelle v. Gamble, 429 U.S. 97 (1976), and later refinements such as Cox v. Glanz, 800 F.3d 1231 (10th Cir. 2015), the plaintiff must show:

1. Objective component: A serious medical need—in suicide cases, a sufficiently apparent risk of self‑harm.
2. Subjective component: That the official knew of and disregarded an excessive risk to inmate health or safety—that is, was deliberately indifferent.

Importantly, a municipal failure‑to‑train claim cannot succeed unless there is at least one underlying constitutional violation by an officer (or officers) whose training was deficient. See Crowson v. Washington County, Utah, 983 F.3d 1166, 1187 (10th Cir. 2020).

2. The single‑incident failure‑to‑train framework (Lance and Valdez)

Ordinarily, deliberate indifference in a failure‑to‑train case is demonstrated by a pattern of similar constitutional violations by untrained employees. See Connick v. Thompson, 563 U.S. 51, 62 (2011). However, in rare cases, a plaintiff may prevail without such a pattern under a “single‑incident” theory if the failure to train is so obvious that constitutional violations are highly predictable.

In Lance v. Morris, 985 F.3d 787 (10th Cir. 2021), and later in Valdez v. Macdonald, 66 F.4th 796 (10th Cir. 2023), the Tenth Circuit adopted the Second Circuit’s test from Walker v. City of New York, 974 F.2d 293 (2d Cir. 1992), stating that a plaintiff bringing a single‑incident failure‑to‑train claim must show:

1. Deficient training policy or custom – The municipality maintained a policy or custom of inadequate training in a relevant respect.
2. Obvious, closely related injury – The injury suffered was a plainly obvious consequence of the deficient training and closely related to the deficiency.
3. Deliberate indifference – Policymakers:
   1. knew to a moral certainty that employees would confront a given situation,
   2. knew that the situation presented employees with a difficult choice that training could make less difficult, and
   3. knew that the wrong choice would frequently cause constitutional deprivations.

The Tenth Circuit in Thao focused on the first requirement—whether GCCJA had a policy or custom of deficient training regarding suicide risk identification.

3. Conflicting evidence about training: why summary judgment was improper

The district court concluded that GCCJA had shown its officers were trained on supervising inmates, identifying suicide risks, and handling such situations, and that the Estate “proffered no evidence to the contrary.” The Tenth Circuit found this characterization incorrect and held that genuine factual disputes exist.

a. Formal policies and state jail standards

• GCCJA’s written policies and the Oklahoma State Jail Standards were the only formal materials given to new hires.
• The state standards require:
  • Triage screening upon admission to identify significant psychiatric problems or suicide risks before an inmate is placed in housing, and, when such risks are present, prompt transport to a medical facility and housing in an observable location.
  • Facilities to implement a “medical/mental health screening” by trained facility personnel using an approved questionnaire, or screening by licensed medical staff.
• Crucially, these standards say little about how line officers should identify suicide risks in already housed inmates beyond the initial intake screening. They do not, for example, instruct that repeated statements like “kill me” or “I’m going to commit suicide” require immediate intervention.
• For “turnaround” inmates, GCCJA did not actually perform the required triage screening. Instead, it relied solely on USMS Form 553, which contained only one mental‑health checkbox (suicide watch/psychiatric decompensation in the last month). This practice departed from the letter of the state standards.

From this, a reasonable jury could find that GCCJA’s “policy” was to not provide any formal, written training on identifying and responding to suicide risks in already housed inmates—particularly rapid‑turnaround federal inmates like Thao—and instead rely on USMS paperwork and informal “experience”‑based learning.

b. “Experience‑based” training and inconsistent officer testimony

Warden Gerlach’s testimony was pivotal. He indicated:

• As of November 2017, the “only” training on identifying mental‑health problems came from the written policies and state standards.
• He believed (incorrectly) that the state standards themselves satisfied the facility’s contractual duty to provide a “comprehensive suicide prevention program.”
• Any additional skill in recognizing mental‑health crises or suicide risk in housed inmates was acquired “through experience, not through a book or anything else,” and he could not say what training, if any, officers on Thao’s shift had actually received.

In contrast, some officers reported training, but in nonspecific terms. The record showed:

• One supervisor (Farley) testified that he understood any “threat of harm to oneself” including “I want to die” required taking the inmate to the nurse.
• Another officer (Duncan) testified that he had some instruction on warning signs like changes in behavior and appearance.
• The officer most directly involved (Henneman) first denied receiving training on identifying mental‑health episodes, and later stated that if an inmate said something about suicide, they were to go “straight to medical.”

The Tenth Circuit concluded that a jury could reasonably interpret this record in more than one way:

• One view (favorable to GCCJA) is that all officers received a consistent message: any suggestion of self‑harm must be taken seriously, and offenders should be brought immediately to medical staff.
• An alternative view (favorable to the Estate) is that:
  • GCCJA had no uniform, formalized training in suicide risk identification; and
  • Officers’ understanding depended heavily on their individual experience, leading to serious gaps—as evidenced by:
    • Henneman’s inability to recognize Thao’s repeated pleas (“come kill me,” “I’m gonna fucking commit suicide”) as suicidal, and
    • His statement to paramedics that Thao had been “yelling” but was “fine.”

The court already had evidence that:

• Thao’s statements were clear and repeated for nearly an hour.
• Officers heard them (as indicated by their responses), yet no one removed him from the camera‑less Cell 126 or called medical staff.
• Facility practice forbade housing inmates on suicide watch in Cell 126, suggesting an appreciation that the cell was unsuitable for suicidal inmates.

Given this, the Tenth Circuit held that a reasonable jury could find that:

• GCCJA’s training regimen on suicide risk identification was deficient (first Lance element), and
• The inconsistent and experience‑based approach to mental‑health training may have been substantially certain to lead to constitutional violations, particularly in a jail handling large volumes of transient inmates.

The court explicitly declined to resolve the remaining Lance/Valdez elements (obviousness, causal link, and deliberate indifference), leaving them for the district court to assess on remand.

4. Underlying constitutional violation and “systemic failure”

GCCJA argued that even if the training was deficient, summary judgment could be affirmed because no officer actually violated Thao’s constitutional rights. The Tenth Circuit did not decide that issue but reminded that:

• A failure‑to‑train claim “may not be maintained without a showing of a constitutional violation by the allegedly un-, under-, or improperly‑trained officer.” Crowson, 983 F.3d at 1187.
• Thus, on remand, the district court must determine whether any officer was deliberately indifferent to Thao’s serious medical needs.

The Estate attempted to sidestep this requirement by invoking a “systemic failure” theory, arguing that, in the context of inadequate medical care, a municipality can be liable without proof of an individual officer’s violation. The Tenth Circuit clarified:

• “Systemic failure” is a distinct Monell theory, recognized in Crowson, where the constitutional violation can arise from the aggregate actions of multiple officers acting pursuant to a policy.
• But the Estate did not plead such a theory; it pleaded a traditional failure‑to‑train claim, which requires proof of at least one underlying violation.
• The court declined to recharacterize the claim on appeal. Evidence of systemic problems may be highly relevant to deliberate indifference and causation, but does not eliminate the need for an underlying violation in a failure‑to‑train case.

On remand, therefore, the Estate must:

1. Show that one or more officers were deliberately indifferent to an obvious risk that Thao would harm himself (Eighth Amendment violation); and
2. Connect that violation to GCCJA’s failure to train on suicide risk identification under the Lance/Valdez framework.

V. Precedents Cited and Their Influence

A. Supreme Court and Tenth Circuit Monell jurisprudence

• Monell v. Dept. of Soc. Servs., 436 U.S. 658 (1978)
  Established that municipalities are “persons” under § 1983 but liable only for their own policies/customs, not under respondeat superior. The court applies this baseline to require a “policy or custom” plus causation.
• Board of Cty. Comm’rs v. Brown, 520 U.S. 397 (1997)
  Emphasized that when a formal policy is itself unconstitutional, issues of culpability and causation are straightforward; but where the alleged policy is one of inaction (e.g., failure to train), a “rigorous” deliberate‑indifference standard applies. The Tenth Circuit relies on this to separate the excessive‑force policy claim from the training claim.
• City of Canton v. Harris, 489 U.S. 378 (1989)
  First articulated the failure‑to‑train doctrine: a municipality can be liable when its failure to train reflects deliberate indifference to rights and is closely related to the injury. Thao applies this via Lance and Valdez.
• Connick v. Thompson, 563 U.S. 51 (2011)
  Held that a pattern of similar constitutional violations is “ordinarily necessary” to show deliberate indifference in failure‑to‑train claims, though limited “single‑incident” exceptions exist. Thao proceeds under this exception.
• Christensen v. Park City Mun. Corp., 554 F.3d 1271 (10th Cir. 2009)
  Describes “facial unconstitutionality” in terms of a policy’s “straightforward enforcement.” Thao borrows this language to evaluate GCCJA’s use‑of‑force policy.
• Crowson v. Washington Cty., Utah, 983 F.3d 1166 (10th Cir. 2020)
  Clarified that failure‑to‑train claims require an underlying individual violation, and distinguished “systemic failure” theories where the violation arises from the aggregate acts of multiple officers. Thao relies on Crowson both to require an underlying violation and to limit the Estate’s attempt to rebrand its claim as “systemic failure” on appeal.
• Lance v. Morris, 985 F.3d 787 (10th Cir. 2021) and Valdez v. Macdonald, 66 F.4th 796 (10th Cir. 2023)
  Adopt and refine the single‑incident failure‑to‑train test (using Walker). Thao directly applies their three‑part framework and focuses on the first part—existence of a policy or custom of deficient training—as the basis for reversing summary judgment.

B. Eighth Amendment excessive force and medical care cases

• Whitley v. Albers, 475 U.S. 312 (1986); Hudson v. McMillian, 503 U.S. 1 (1992)
  Provide the two‑prong excessive‑force standard applied in prisoner cases. Thao cites these to frame the Eighth Amendment context.
• Cox v. Glanz, 800 F.3d 1231 (10th Cir. 2015)
  Treats jail suicide claims as inadequate medical care cases under the deliberate‑indifference test. Thao explicitly invokes this line of authority in labelling Thao’s claim as an Eighth Amendment medical‑care issue.
• Est. of Burgaz v. Bd. of Cty. Comm’rs for Jefferson Cty., Colo., 30 F.4th 1181 (10th Cir. 2022)
  Reiterates that jailers cannot guarantee safety but must take reasonable steps to protect inmate safety; used in Thao to frame the duty owed to Thao.

C. Walker’s three‑factor deliberate‑indifference test

In Walker v. City of New York, 974 F.2d 293 (2d Cir. 1992), the Second Circuit articulated a three‑factor test for deliberate indifference in failure‑to‑train cases. Lance and Valdez, and now Thao, formally embed this test into Tenth Circuit law, requiring plaintiffs to show:

1. Predictable recurrence of the situation;
2. A difficult choice for employees that training could alleviate; and
3. A high likelihood that the wrong choice will cause constitutional deprivations.

In the jail‑suicide context, it is easy to see how these factors may be satisfied:

• Jail officers know to a moral certainty that they will encounter inmates expressing despair or suicidal ideation.
• Deciding whether to treat a given statement as a genuine suicide risk or as mere “acting out” is a “difficult choice” that targeted training could make less difficult.
• Choosing not to treat explicit pleas like “I’m going to commit suicide” or “just kill me” as red flags will “frequently cause” grave deprivations—the inmate’s death.

Although Thao does not complete the deliberate‑indifference analysis, it sets up the path for how the Walker/Lance/Valdez framework may be applied by the district court and by future litigants.

VI. Complex Concepts Simplified

1. “Facially unconstitutional” policy vs. unconstitutional application

• A policy is facially unconstitutional if, by its own terms, it requires or authorizes conduct that is unconstitutional in all or nearly all applications (e.g., “jailers may beat inmates whenever displeased”).
• A policy is facially constitutional but may be unconstitutionally applied when officers violate either the policy’s terms or apply it in an unreasonable way (e.g., abusing a Taser despite a policy allowing only reasonable, necessary force).
• Thao holds GCCJA’s Taser policy is facially constitutional; if Henneman’s tasing was unconstitutional, it was because he violated or misapplied a facially valid policy, not because the policy itself was defective.

2. Failure‑to‑train vs. systemic failure

• Failure‑to‑train: The municipality’s training program is inadequate in a specific respect (e.g., no training on suicide risk), and that deficiency causes an officer to violate someone’s rights. Requires:
  • a specific training deficiency,
  • deliberate indifference,
  • a causal link to an individual officer’s constitutional violation.
• Systemic failure: A distinct theory where the constitutional violation is produced by the aggregate of many officers’ actions operating under a flawed system (e.g., chronic understaffing, broken communication procedures, conflicting directives), such that the municipality is effectively violating rights “through twenty hands rather than two.”
• Thao underscores that one cannot simply label a failure‑to‑train claim as “systemic failure” to avoid proving an underlying individual violation; they are doctrinally separate theories.

3. “Single‑incident” failure‑to‑train

• Normally, plaintiffs show deliberate indifference in a training claim by pointing to a pattern of similar past incidents (e.g., multiple prior suicides in the same jail under similar circumstances).
• In a “single‑incident” theory, the plaintiff instead argues that even without a prior pattern, the need for training was so obvious, and the risks so grave, that the municipality was deliberately indifferent in not training officers. Classic example: no training on deadly‑force rules for armed police officers.
• Thao proceeds under this doctrine: jail authorities know they will face suicidal inmates, but the record permits an inference that they failed to provide even basic, uniform training on how to recognize and respond to such crises.

4. Suicide risk as a “serious medical need”

• Under Eighth Amendment medical‑care doctrine, a medical or mental condition is “serious” if a reasonable person would recognize it as needing prompt medical attention.
• Suicidal ideation and threats of self‑harm clearly qualify; ignoring such risk can constitute deliberate indifference if officials know of and disregard the risk.
• In Thao, the Estate will need to prove on remand that officers actually perceived Thao’s statements (“I’m gonna fucking commit suicide,” “come kill me”) as indicating serious risk, or that the risk was so obvious that they should be deemed aware of it.

VII. Likely Impact and Practical Implications

A. For jail and prison administrators

1. Suicide‑prevention training cannot be left to “experience.”
   The court’s willingness to send this case to a jury based on evidence of ad hoc, experience‑based learning sends a strong signal: facilities must implement formal, uniform, and documented training on:
   • recognizing signs of suicide risk, including verbal threats, expressions of hopelessness, and drastic behavioral changes;
   • mandatory steps when such signs appear (e.g., immediate clinician notification, suicide watch, removal from hazardous cells).
2. Turnaround/transit inmates need genuine mental‑health screening.
   GCCJA’s practice of treating transient federal inmates as exempt from state jail triage requirements, and relying solely on USMS Form 553, is now spotlighted as problematic. Even if the Tenth Circuit did not issue a formal holding on the point, its analysis makes clear that:
   • State screening standards apply irrespective of the length of stay.
   • Relying on a single check box on an external form is not equivalent to meaningful mental‑health screening by trained personnel.
3. Risky physical environments (like Cell 126) demand heightened policy controls.
   The opinion notes that:
   • Cell 126 lacked a camera and had a solid hatch blocking the window.
   • Officers knew suicidal inmates should not be housed there.
   • Yet Thao was left there, unsupervised for over an hour, while voicing clear suicidal intent.
   Facilities should revisit the use of such “backup” or “shower” cells for inmates in distress and ensure policies strictly control when and how they may be used.
4. Documentation of training matters.
   The presence of a mental‑health PowerPoint was not enough; the lack of attendance records and mandatory participation meant the court could not credit it as proof that Thao’s supervising officers were actually trained. Administrators should:
   • Make critical trainings mandatory,
   • Maintain clear attendance and testing records, and
   • Update training materials to match evolving case law and standards.

B. For § 1983 litigants and counsel

1. Choose the Monell theory carefully.
   Thao illustrates the strategic consequences of limiting an excessive‑force municipal claim to a facial policy theory. The Estate likely could have framed additional theories based on:
   • a widespread practice of allowing Tasers to be used on restrained inmates, or
   • a failure to train or supervise on Taser limitations.
   Instead, by arguing only that the policy itself was unconstitutional, the Estate invited an analysis focused on textual facial validity, and lost that claim even though the underlying force may have been excessive.
2. Failure‑to‑train claims require granular evidence of what training actually occurred.
   Thao shows the importance of:
   • deposing policymakers on training content, frequency, and documentation;
   • contrasting formal policies with actual practices, particularly with high‑risk groups like suicidal inmates;
   • juxtaposing different officers’ understandings to show inconsistency or gaps.
3. Systemic‑failure theories must be pleaded explicitly.
   Where evidence suggests a facility‑wide breakdown (e.g., chronic understaffing, pervasive disregard of screening rules, repeated prior suicides), counsel should consider pleading a distinct systemic failure Monell theory consistent with Crowson, rather than relying solely on failure‑to‑train. That approach:
   • acknowledges the possibility that no single officer’s conduct might fully capture the violation, and
   • shifts focus to how the institution as a whole operates.
   Thao makes clear that courts will not retrofit such a theory on appeal.
4. Suicide cases are fertile ground for single‑incident failure‑to‑train claims.
   Because the risk of inmate suicide is well‑known and recurring, and the consequences of inadequate training are extreme, many jail suicide cases can plausibly fit within the “patently obvious” single‑incident framework, especially when:
   • there is little or no formal training on suicide risk, and
   • officers’ responses to clear cries for help appear indifferent or mocking.
   Thao provides a concrete appellate endorsement that such a claim is triable to a jury under the right evidentiary showing.

VIII. Conclusion

Thao v. Grady County Criminal Justice Authority sharpens two important lines in § 1983 municipal‑liability doctrine.

First, it clarifies that when a plaintiff challenges a formal use‑of‑force policy as unconstitutional, the focus is on the policy’s text and facial operation, not the municipality’s “deliberate indifference.” GCCJA’s Taser policy, by limiting force to reasonable, non‑punitive uses and restricting Tasers against handcuffed inmates absent “articulable extenuating circumstances,” was held facially constitutional. Any excessive force by an officer in contravention of that policy may create individual liability but does not, without more, support Monell liability under a facial‑policy theory.

Second, the decision demonstrates how the single‑incident failure‑to‑train doctrine can be deployed in jail suicide cases. The court found that the Estate introduced sufficient evidence to permit a jury to conclude that GCCJA provided no consistent, formal training on recognizing suicide risks in already housed inmates, instead relying on general state standards and unsystematic, “experience‑based” learning. In the context of a jail that regularly encounters suicidal inmates, and in light of Thao’s protracted pleas for death, that alleged training deficiency is serious enough to warrant trial.

By reversing summary judgment on the failure‑to‑train claim, the Tenth Circuit underscores that suicide‑prevention training in correctional settings is not optional institutional housekeeping, but a constitutional imperative. At the same time, by affirming on the excessive‑force policy claim, the court cautions plaintiffs to distinguish carefully between claims attacking the content of official policies and those attacking how officers are trained to implement them.

In sum, Thao is likely to become a significant reference point in the law of jail suicides and Monell liability, guiding both correctional administrators in structuring suicide‑prevention protocols and litigants in framing and proving municipal‑liability claims in the Tenth Circuit and beyond.