Inquest Scope and “Coronial Causation”: Police Omission Cannot Be Excluded as “Speculative” Without Key Criminal-Justice Materials

1. Introduction

Linda O'Brien died on 9 May 2020 after falling from a window of her flat in the early hours. Alan McMahon was in the flat and later called emergency services. The surrounding evidence included a history of domestic abuse, the existence of restraining orders against Mr McMahon, and concerns that Ms O'Brien had been assaulted shortly before the fall and may have been attempting to escape.

The appellant, Sharon O'Brien (next of kin), challenged the assistant coroner’s decision to restrict the scope of the inquest largely to events of 8–9 May 2020. She argued that the inquest also needed to investigate earlier police acts/omissions (notably on 7 April 2020, when police attended a domestic incident but did not arrest Mr McMahon for breach of a restraining order) because those omissions might have contributed to him being present on 9 May 2020.

The High Court dismissed the judicial review. The Court of Appeal allowed the appeal, holding that both the coroner’s approach to causation and the High Court’s approach to review were legally flawed, and that the scope decision had been taken prematurely and irrationally.

2. Summary of the Judgment

• The Court of Appeal quashed the coroner’s decision notice and remitted the matter for reconsideration by another coroner.
• The coroner’s conclusion that there was “no coronial causation” between police involvement on 7 April 2020 and the death on 9 May 2020 was inadequately reasoned and reached in ignorance of key materials (notably Mr McMahon’s antecedent history).
• It was premature and irrational to treat the criminal justice outcomes following a hypothetical arrest (charge, bail, remand, sentence) as unknowable “speculation” without first obtaining straightforward, relevant information and applying established frameworks (sentencing guidelines; CPS decision-making).
• The High Court erred by effectively making its own factual findings on causation instead of reviewing whether the coroner’s decision was premature/irrational.
• Because jury and Article 2 decisions were premised on the coroner’s causation conclusion, they too required reconsideration.

3. Analysis

3.1 Precedents Cited

R (Tainton) v HM Coroner [2016] EWHC 1396 (Admin)

This was the central authority on the meaning of “causation” in the coronial context. The Court of Appeal accepted that R (Tainton) v HM Coroner correctly distinguishes:

• Standard of proof: balance of probabilities; and
• Threshold of contribution: the act/omission must have “more than minimally, negligibly or trivially contributed” to death.

The Court of Appeal made an important clarification: the High Court had relied on R (Tainton) v HM Coroner to suggest (in effect) that where outcomes are uncertain the issue becomes inherently speculative. The Court of Appeal emphasised that the “context” of Tainton (terminal illness and loss of a chance) does not re-write the general test. In a case like this—where causation depends on assessing how the criminal justice system would probably have responded—Tainton does not justify foreclosing inquiry simply because counterfactual assessment is required.

R (Dawson) v HM Coroner for East Riding and Kingston upon Hull Coroners District [2001] EWHC Admin 352; [2001] Inquest LR 233

Cited via Tainton, R (Dawson) v HM Coroner for East Riding and Kingston upon Hull Coroners District supports the “more than minimal” contribution threshold. It underpins the idea that coronial causation is not satisfied by merely trivial contributions but equally does not require the act/omission to be the sole or dominant cause.

R v Turner [1975] QB 834 (and Folkes v. Chadd (1782) 3 Doug.K.B. 157)

The High Court rejected the notion of opinion evidence about the likely progress of prosecution/bail/sentence as “purely speculative and of no probative value.” The Court of Appeal held that this was an impermissible blanket approach.

Relying on R v Turner (which in turn referenced Folkes v. Chadd (1782) 3 Doug.K.B. 157), the Court confirmed the orthodox principle: expert opinion is admissible where it provides information likely to be outside ordinary judicial experience; it is unnecessary only where judge/jury can safely decide without it.

In practical terms, the Court’s message is that coroners (and reviewing courts) must not exclude potentially relevant expert assistance by labelling it “speculative” in the abstract; the admissibility and weight of such evidence depend on whether it genuinely fills a knowledge gap and is appropriately constrained.

3.2 Legal Reasoning

(a) The correct question: probability, not “certainty”

The coroner’s post-decision reasoning included the assertion that “It simply cannot be known” whether an arrest would have led to incarceration at the relevant time. The Court of Appeal held that this applied the wrong lens: the legal test is balance of probabilities, not knowledge or certainty. The relevant inquiry was whether, more likely than not, arrest on 7 April 2020 would have prevented Mr McMahon being present on 9 May 2020 (e.g., by remand or sentence), and—if his presence materially contributed to the fall—whether the police omission thereby materially contributed to death.

(b) Prematurity and irrationality: deciding without key materials

A major plank of the Court of Appeal’s decision is that the coroner drew a causation line too early, without obtaining “the key document in the case”: Mr McMahon’s antecedents. The Court highlighted that:

• the police (an interested party) would readily have this information;
• antecedents and prior history bear directly on likely bail/remand outcomes and on likely sentence; and
• without them, the coroner could not rationally conclude that custody on 9 May was beyond proof on the balance of probabilities.

(c) Counterfactual assessment is not “random”: structured legal frameworks exist

The Court rejected the notion that criminal justice outcomes are effectively unknowable. It pointed to structured decision-making frameworks that allow reasoned prediction:

• Sentencing Guideline for breach of a restraining order (effective 1 October 2018): categories of culpability/harm, starting points and ranges, and listed aggravating factors (including recent convictions and history of disobedience to court orders).
• Domestic Abuse: overarching principles (effective October 2018), emphasising the seriousness of domestic abuse, escalation risk, and that victim withdrawal/attitudes do not diminish seriousness.
• Code for Crown Prosecutors: realistic prospect of conviction + public interest test; the Court considered it “simply inconceivable” that the public interest would not be met on these facts.
• Bail Act 1976: despite the presumption in favour of bail, remand is routine where statutory exceptions apply—particularly substantial grounds to believe the defendant would commit further offences on bail.

On the evidence described (including an earlier swift custody outcome in August 2019), the Court considered it “highly likely” that, if arrested and sentenced prior to 9 May 2020 (especially on a guilty plea), Mr McMahon would have been in custody on that date.

(d) High Court’s error: substituting its own view rather than reviewing the coroner’s approach

The Court of Appeal held the High Court wrongly embarked on deciding causation itself—reasoning through what “we do not know” and treating the chain as “entirely speculative”—instead of answering the judicial review question: whether the coroner’s decision to exclude the police conduct from scope was premature and irrational. This was not merely a difference of evaluative judgment; it was a misdirection as to the court’s function on review.

(e) Consequential effects: jury and Article 2

The coroner’s causation conclusion underpinned:

• the decision that the inquest should not be with a jury (linked to section 7(2)(b) of the Coroners and Justice Act 2009); and
• the decision that Article 2 was not engaged.

Because the causation analysis was flawed, these decisions must be reconsidered as well. The Court also indicated it did not accept the High Court’s concern that including police conduct would necessarily turn the inquest into an impermissibly wide “surrogate public inquiry.”

3.3 Impact

(1) A practical discipline on coroners’ “scope” discretion

The judgment reinforces that coroners have wide discretion to set scope, but that discretion must be exercised on a properly informed and legally correct foundation. Where a proposed line of inquiry turns on a counterfactual criminal justice pathway (arrest → charge → bail/remand → sentence), a coroner must not dismiss it as “speculation” without:

• obtaining obvious key material (e.g., antecedents, charging history, relevant records);
• considering structured legal frameworks (sentencing guidelines, bail law, CPS Code); and
• considering whether tightly controlled expert evidence is needed.

(2) Lower threshold for jury consideration and interaction with police omissions

The Court’s approach makes it harder for coroners to avoid jury inquests (where potentially required) by prematurely deciding there is no causation from police acts/omissions. If there is “reason to suspect” a police omission may have resulted in death, scope and jury questions should not be closed down by under-investigated causation conclusions.

(3) Domestic abuse context: institutional decision-making must reflect reality

A notable theme is the Court’s insistence that police, CPS, and courts are expected to act with awareness of domestic abuse dynamics (including coercive control and victim accommodation of abusers). The Court explicitly rejected treating the victim’s tolerance of the abuser’s presence as mitigating or as making enforcement outcomes less likely.

(4) Admissibility of expert evidence in coronial causation disputes

By invoking R v Turner, the Court signalled that expert evidence about criminal justice process (properly framed) may be admissible to assist coroners/juries in evaluating probabilities. The impact is not that experts will routinely opine on “what would have happened,” but that a categorical refusal is wrong; relevance and utility must be assessed case by case.

4. Complex Concepts Simplified

• Scope of an inquest: the boundaries of what the inquest will investigate (which events, time periods, systems, and actors).
• Interested person (IP): someone with a formal right to participate in the inquest because of their connection to the death (e.g., family, potentially responsible individuals, public bodies).
• Coronial causation: whether an act/omission made an “actual and material” contribution to death. It need not be the sole cause, but must be more than minimal.
• Standard of proof vs threshold:
  • Standard = how sure the fact-finder must be (here, balance of probabilities).
  • Threshold = how significant the contribution must be (more than minimal/trivial).
• Article 2 (ECHR): the right to life. When engaged, it can require a more searching investigation into state involvement/failures (including policing) surrounding a death.
• “Speculation” vs “probability”: the Court drew a line between ungrounded guesswork and a structured assessment of likely outcomes using legal rules, guidelines, and available evidence (including antecedents).

5. Conclusion

The Court of Appeal established a clear and practical principle for coronial decision-making: a coroner cannot lawfully narrow the scope of an inquest by treating a plausible police-omission causation chain as mere “speculation” where key, readily obtainable materials (especially antecedents) and structured legal frameworks (bail law, CPS Code, sentencing guidelines) have not been properly considered. The judgment also rejects a blanket exclusion of expert evidence on such matters, and it requires reconsideration of consequential decisions on jury inquests and Article 2 where they rest on an under-informed causation conclusion.