Omitted “Resulted in Death” Enhancer Finding in Colorado Child Abuse Trials: Error, But Potentially Harmless Beyond a Reasonable Doubt; “Results in” Means But-For Causation

1. Introduction

People v. Ceus arises from the deaths of two young sisters, M.R. (about ten) and H.M. (about eight), who were exiled by a religious group’s leader, Madani Ceus, to a car on a Colorado farm and left there without food or water in intense summer heat. The People charged Ceus with two counts of first degree murder—victim under twelve and two counts of child abuse resulting in death, tried as lesser included offenses to the murder counts.

The appeal turned on a technical but constitutionally significant jury-instruction problem: the trial court did not require the jury to make a distinct finding—via special interrogatory or otherwise—that the child abuse “resulted in” the girls’ deaths, even though “death” elevates child abuse to a class 2 felony under section 18-6-401(7)(a)(I), C.R.S. (2025). The court of appeals found the omission reversible; the supreme court agreed there was error but held it harmless beyond a reasonable doubt and reinstated the convictions, remanding for consideration of Ceus’s remaining appellate claims.

2. Summary of the Opinion

• Instructional error: The trial court erred by failing to require distinct jury findings that the child abuse resulted in each child’s death (either through special interrogatories or elemental drafting that necessarily required the finding).
• Harmlessness: The error was constitutionally harmless because (i) the “resulted in death” point was not meaningfully disputed at trial, (ii) the instructions/verdict forms repeatedly framed the offense as “child abuse resulting in death,” and (iii) the evidence that the abuse resulted in death was overwhelming.
• Sufficiency: The evidence was sufficient to sustain the convictions; the statute does not require a formal caretaking relationship, and the proof supported at least reckless (and potentially knowing) child abuse.
• Disposition: The supreme court reversed the court of appeals’ judgment and remanded for consideration of remaining claims.

3. Analysis

3.1 Precedents Cited (and How They Shaped the Decision)

A. Who defines crimes; the need to instruct on each required fact

• Randolph v. People and Gorman v. People: Used for the foundational premise that the General Assembly defines criminal conduct and components of liability, and trial courts must properly instruct juries on what the prosecution must prove.
• Garcia v. People: Supplies the de novo review standard for whether instructions accurately state governing law.

B. Constitutional harmless error framework for instructional omissions

• Hagos v. People (quoting Chapman v. California): Establishes the “reasonable possibility the error might have contributed to the conviction” test and the prosecution’s burden on harmlessness.
• Delaware v. Van Arsdall: Supports considering the record “as a whole” when assessing harmlessness.
• Key v. People: Recognizes that instructions omitting correct element definitions are constitutionally deficient, yet harmlessness can be evaluated, including by reference to whether evidence on the issue was overwhelming and by reviewing instructions as a whole.
• Neder v. United States: The majority leans on Neder’s principle that omission of an element can be harmless beyond a reasonable doubt when the omitted matter was uncontested and supported by overwhelming evidence such that the verdict would be the same absent the error.
• People v. Grudznske (quoting People v. Draper): Colorado appellate authority applying a Neder-like analysis to omissions/definitions of elements, emphasizing harmlessness where evidence on the missing component is overwhelming.

C. Sentence enhancers, Apprendi, and child abuse “death” as an enhancer

• People v. Robinson and Armintrout v. People: Provide the key classification that the victim’s death in Colorado’s child abuse statute functions as a sentence enhancement factor rather than an element of the base child-abuse offense, even though it must be found by a jury when it increases the maximum penalty.
• Apprendi v. New Jersey, Erlinger v. United States, and Caswell v. People: Provide the constitutional rule that any fact increasing the statutory maximum penalty must be submitted to a jury and proved beyond a reasonable doubt—thereby requiring a jury finding on the “resulted in death” enhancer.
• People v. Ramirez: Cited for the proposition (as applied in Colorado) that sentence enhancers generally must be proved beyond a reasonable doubt to the jury.

D. What “results in” means, and how causation differs from proximate cause

• People v. Beverly (quoting Martinez v. People): Provides the majority’s doctrinal lever distinguishing statutes requiring “proximate cause” from those using “results in,” and reiterates the actual (“but for”) vs legal (“proximate”) causation divide when a statute demands a particular result.
• United States v. Jeffries and Burrage v. United States: Federal “results from” cases used to support the majority’s reading that “results” tracks an effect/outcome inquiry (but-for), not foreseeability/proximate cause.
• People v. Dunaway: Crucial comparator. The majority reads Dunaway as requiring a link between abuse and result but not importing proximate cause; the dissent reads Dunaway as explicitly requiring a “causal connection” and criticizes the majority for recharacterizing that concept as but-for only. Regardless, both sides accept Dunaway’s core point: proof of abusive conduct does not automatically prove the “result” enhancer.

E. Sufficiency standards and inferential reasoning

• Dempsey v. People and People v. Donald: Provide the sufficiency standard (substantial evidence viewed in the light most favorable to the People, reasonable inferences permitted).
• People v. Douglas: Cited for de novo review of sufficiency in the appellate posture (the opinion quotes it from the court of appeals).

F. Who can commit child abuse; no formal caretaking relationship required

• People v. Arevalo: States the expansive duty concept—every person must refrain from action endangering a child’s life/health—used to reject Ceus’s “no parental duty” sufficiency argument.
• People v. Mann: Supports the statute’s breadth to include not only direct harm but allowing harm.
• People v. Archer: A related case involving a codefendant, cited to reinforce that lack of a familial relationship does not foreclose liability under section 18-6-401.

G. Mens rea for child abuse

• People v. Deskins, People v. Noble, and Lybarger v. People: Supply the mental-state framework—knowing requires general awareness of abusive nature or circumstances; reckless requires conscious disregard of a substantial and unjustifiable risk in relation to child abuse.
• People v. Hall: Used to contextualize “substantial and unjustifiable risk,” emphasizing that because death is the gravest harm, even relatively slight risk of death can satisfy the standard.

H. “Fair trial, not perfect trial”

• People v. Mills: Invoked to frame the harmless error inquiry and to emphasize that identified instructional error does not always require reversal.

3.2 Legal Reasoning

A. The rule announced on instructions: distinct enhancer finding required

The court reaffirmed that Colorado’s child abuse statute separates (1) the abusive conduct (section 18-6-401(1)(a)) from (2) the “result” used to set offense level and punishment (section 18-6-401(7)). Although “death” is treated as a sentence enhancer (not a base element), Apprendi-type doctrine requires the jury to find it beyond a reasonable doubt when it increases the maximum penalty. Therefore, the jury must make a distinct “resulted in death” finding—either through special interrogatories (as contemplated by COLJI-Crim. 6-4:09) or by drafting elemental instructions so that guilt necessarily entails that finding.

B. The new limiting principle: omission can be harmless beyond a reasonable doubt

The opinion’s central operational holding is not that the “resulted in death” finding is optional—it is mandatory—but that its omission can be constitutionally harmless under a Neder/Hagos analysis when the record makes the missing finding effectively inevitable. The majority emphasized three cumulative considerations:

• Not meaningfully disputed: The defense theory focused on who should be held responsible, not on whether the deprivation and exposure caused the deaths; Ceus’s own statements (“It’s not rocket science”) were treated as reinforcing this.
• Instructions and verdict forms: The jury was repeatedly presented the offense label “child abuse resulting in death,” and it selected that option on the verdict forms.
• Overwhelming evidence: Evidence converged on starvation/dehydration/hyperthermia (or combination) as the likely causes, with no alternative cause suggested by the evidence.

Taken together, the court held there was no reasonable possibility the omission contributed to the verdict.

C. Clarifying “results in” vs proximate cause

A significant doctrinal move is the court’s reliance on People v. Beverly to treat “results in” as different from “proximate cause.” In the court’s framing, the statute asked whether death was an “effect, outcome, or consequence” of the abuse (a but-for style inquiry), not whether death was a foreseeable, proximate result. This clarification matters because it narrows the kind of causation dispute that could defeat harmlessness: the relevant question is whether the abusive deprivation/exposure produced death, not whether the death was a foreseeable consequence under a proximate-cause lens.

D. Sufficiency: expansive liability and proof of mental state

On sufficiency, the court rejected the argument that liability requires a formal caretaking relationship. It reiterated that the statute covers those who cause harm, threaten harm, or permit harm. The court also held evidence supported knowing or reckless mental state based on Ceus’s control over the group, directives restricting access to the children and to sustenance, and post-arrest statements indicating awareness and deliberate disregard (“shut it out”).

3.3 Impact

• Trial practice (instructions): The decision reinforces that prosecutors and trial courts must ensure a jury makes an explicit “resulted in death” (or other result) finding for section 18-6-401(7) classifications, typically by special interrogatory or by ensuring the elements as drafted necessarily require it.
• Appellate review (harmlessness): The case strengthens a Neder-like pathway for affirmance even when an Apprendi-required result finding is omitted—so long as the record shows the missing issue was effectively uncontested and overwhelmingly proven.
• Causation framing: By tying “results in” to an effect/outcome (but-for) inquiry and distinguishing it from proximate cause, the opinion may reduce the range of “causation” arguments that can defeat harmless error in child abuse result-enhancer cases.
• Scope of who can be liable: The sufficiency holding continues Colorado’s expansive approach: authority, control, and conduct endangering children can satisfy the statute even absent formal parent/guardian status.
• Future litigation: The dissent spotlights a likely fault line—whether Colorado should treat “results in” as requiring more than but-for causation (and whether Dunaway compels that). Parties may continue to litigate the degree of causation required and whether “responsibility” disputes should count as disputes over the enhancer.

4. Complex Concepts Simplified

• Sentence enhancer vs element: An “element” is part of the basic crime the jury must find to convict. A “sentence enhancer” is a fact that increases the punishment level. Under Apprendi v. New Jersey, an enhancer that increases the maximum penalty generally must still be found by a jury beyond a reasonable doubt.
• “Results in”: In this opinion’s usage, “results in” asks whether the death arose as an effect/outcome of the abuse (a but-for style question), not whether death was foreseeable under proximate cause.
• But-for vs proximate cause: “But-for” causation asks whether the harm would have happened without the conduct. “Proximate cause” adds a further limitation (often foreseeability and “natural and probable sequence”). The court treated “results in” as not importing proximate-cause limits.
• Constitutional harmless error: Even when the trial court makes a constitutional mistake (like omitting a required jury finding), the conviction can stand if the prosecution proves beyond a reasonable doubt that the error did not affect the verdict—i.e., there is no reasonable possibility it contributed.
• Special interrogatory: A targeted “verdict question” requiring jurors to answer a specific factual issue (here: “Did the child abuse result in death?”). It helps ensure the jury makes the exact finding required for an enhanced penalty.

5. Conclusion

People v. Ceus sets a practical appellate rule for Colorado child abuse prosecutions involving the “death” enhancer: a trial court’s failure to obtain a distinct jury finding that abuse “resulted in” death is instructional error, but it will not require reversal where the record—viewed as a whole—shows the missing finding was not meaningfully contested, was reinforced by how the case was tried and presented to the jury, and was supported by overwhelming evidence. The decision also reinforces an expansive view of who may be liable for child abuse (not limited to formal caretakers) and clarifies, through its reliance on People v. Beverly, that “results in” is treated as an effect/outcome (but-for) concept rather than proximate cause.