How to Prove Medical Negligence in Ireland: The Four Elements, the Dunne Test, and What Evidence You Actually Need

What is causation in Irish medical negligence law?

Causation is the legal requirement to prove that a healthcare provider's negligent act or omission directly caused, or materially contributed to, the patient's injury. A medical error alone does not create a valid claim. The error must have changed the outcome. Irish courts treat causation as distinct from breach of duty, and both must be proven independently to succeed.

The Supreme Court confirmed in Morrissey v HSE [2020] IESC 6 that the Dunne principles remain the governing test for medical negligence in Ireland ². Those principles address the standard of care (breach). Causation then asks a separate question: did that breach cause this specific harm?

A detail that catches many claimants off guard: you can prove clear negligence and still lose your case entirely if causation fails. In Quinn v Mid-Western Health Board [2005] IESC 19, the Supreme Court acknowledged negligent care during a birth but dismissed the claim because the infant's brain damage would have occurred regardless of the clinical error ⁵.

How do Irish courts apply the but-for test?

The but-for test asks one retrospective question: would the patient have suffered this specific injury but for the negligent act or omission? The claimant must prove the answer is "no" on the balance of probabilities, meaning 51% or more likely. The test was laid down alongside the Dunne principles in Dunne v National Maternity Hospital [1989] IR 91 and remains the primary standard in Irish clinical negligence.

Applying the test means separating what the disease or condition would have done naturally from what the clinical error added. In a delayed cancer diagnosis, for example, the plaintiff's expert must reconstruct the disease staging at the date diagnosis should have occurred, then compare that with staging at actual diagnosis. The gap between those two points is the "delay interval," and the worsened prognosis within that interval is what causation targets.

Unlike in England and Wales, where the Bolam-Bolitho test governs breach, Ireland's Dunne test applies a stricter standard: whether no competent practitioner of the same speciality would have acted the same way ¹. That strictness on breach also sharpens the causation question. The court asks not just "was there an error?" but "did this error, by a practitioner who acted as no peer would, change what happened to the patient?"

Causation in practice: two scenarios, opposite outcomes

The same type of claim, delayed bowel cancer diagnosis, can succeed or fail on causation depending on the staging evidence. These are composite illustrations based on common Irish claim patterns, not real cases.

The difference between these two outcomes is entirely in the medical evidence, not the legal test. The but-for question is identical in both cases. The answer depends on what the cancer was doing at the point when the GP should have acted. This is why causation expert evidence is the single most important element in a delayed diagnosis claim.

When the but-for test fails: material contribution in Ireland

Irish courts have not formally adopted the material contribution test as a standalone alternative to but-for, though practitioners increasingly invoke it in complex multi-factor cases. The Law Society Gazette (November 2024) [6] confirmed that Irish courts have "failed to consider an alternative test" to the standard but-for analysis.

The material contribution doctrine, developed in English case law through Bonnington Castings v Wardlaw [1956] and refined in Holmes v Poeton Holdings [2023] EWCA Civ 1377, allows a plaintiff to succeed by proving negligence made a contribution to the injury that was more than minimal (de minimis). The English Court of Appeal clarified in Holmes that this test applies to both divisible and indivisible conditions.

The Irish position remains cautious. In Quinn v Mid-Western Health Board [2005], Mr Justice Kearns rejected broader alternative causation tests, warning against compensating plaintiffs where a substantial connection to the defendant's conduct was absent ⁵. However, the High Court's 2025 approach in Tolan v Brindley Manor Federation of Nursing Homes [2025] IEHC 327 signalled growing flexibility. Mr Justice Barr refused to strike out a clinical negligence claim at the interlocutory stage, citing Fairchild and McGhee as authority for material contribution arguments where medical causation is scientifically unknowable.

From handling complex Irish clinical negligence files, the practical reality is that material contribution arguments surface most often in settlement negotiations rather than in formal judgments. The Civil Liability Act 1961 [3] already provides for apportionment among concurrent wrongdoers under sections 11, 12, and 34, which handles some scenarios that English courts address through material contribution doctrine. For most claimants, the more promising route is loss of chance.

Loss of chance: can you claim for a reduced probability of recovery?

The Irish Supreme Court recognises loss of chance as a compensable injury, even when the patient's survival probability was below 50%. In Philp v Ryan [2004] IESC 105, the plaintiff experienced an eight-month delay in prostate cancer diagnosis due to clear negligence ⁴. Medical experts could not conclusively determine whether the delay shortened life expectancy. The High Court awarded €45,000, which the Supreme Court increased to €100,000. The court held that the patient was wrongfully deprived of the opportunity to consider timely treatment and suffered severe mental distress from knowledge of the delay.

This differs from England and Wales. The House of Lords in Gregg v Scott [2005] established that loss of chance claims fail unless the initial survival probability exceeded 50%. A patient whose chance of survival dropped from 45% to 15% due to negligent delay would recover nothing under UK law. Under Irish law following Philp v Ryan, that same patient could recover compensation for the lost opportunity itself.

A tension exists between Philp v Ryan and Quinn v Mid-Western Health Board, decided just one year later. Quinn rejected loss-of-chance arguments for the physical brain injury, requiring strict but-for proof. The practical resolution in Irish litigation is to plead both: loss of chance for the deprivation of treatment opportunity and mental distress, while pursuing but-for or material contribution for the physical injury. This dual-track approach is central to how experienced practitioners handle delayed diagnosis causation in Ireland.

How causation analysis differs by claim type

We call this framework the Causation Complexity Matrix because the proof requirements and expert disciplines shift depending on which category the claim falls into. A surgical error case might need one expert report. A delayed cancer diagnosis case typically needs two or three. Getting the expert instruction wrong at the outset is the most common reason causation proof fails.

How the State Claims Agency defends on causation

The State Claims Agency (SCA) defends the majority of public hospital claims in Ireland, and its most common defence strategy is to admit the breach occurred but fight on whether it caused the injury. According to the NTMA Annual Report 2024 [7], the SCA manages estimated outstanding clinical negligence liabilities of €5.35 billion. It paid €210.5 million in clinical damages in 2024, with 43% of paid claims resolved through mediation.

In birth injury claims, for example, the SCA rarely disputes that a delay happened. Instead, its experts argue the brain injury had already occurred before the CTG became pathological, meaning the delay made no difference to the outcome. In delayed cancer diagnosis claims, the SCA's oncology experts typically argue the cancer had already progressed beyond the point where earlier intervention would have improved survival.

The Medical Protection Society's 2024 data [8] shows clinical negligence claims in Ireland take an average of 1,462 days to resolve (roughly four years), which is 56% longer than in the UK (939 days). A significant portion of that delay stems from causation disputes requiring multiple rounds of expert evidence. One aspect the official data doesn't capture: the SCA's willingness to settle often depends entirely on the strength of the plaintiff's causation evidence rather than the strength of the breach evidence.

Proving causation: why you typically need two expert reports

Irish courts require independent expert evidence confirming both negligence and causation before proceedings can issue. In most delayed diagnosis and complex treatment cases, these are separate expert reports from different specialists. The liability expert confirms the clinical error falls below the Dunne standard. The causation expert (often from a different medical discipline) confirms the error changed the patient's outcome.

The reason for this split is medical, not legal. A radiologist can say a scan was misread, but only an oncologist can say what would have happened to the cancer had it been caught at the earlier stage. A midwife expert can identify monitoring failures during labour, but only a neonatologist can determine when the hypoxic brain injury occurred relative to those failures. The Causation Complexity Matrix described above determines which expert combination your case requires.

Because Ireland's medical community is small and interconnected, both experts are typically sourced from the UK to avoid professional conflicts of interest. The High Court and Supreme Court have confirmed that it is unethical to issue medical negligence proceedings without a supportive expert report ². In practice, the most common causation failure point is vague expert instructions at the outset. The solicitor must frame precise questions for the causation expert, including: what was the natural progression of the condition, what difference would earlier intervention have made, and on the balance of probabilities, did the negligence change the outcome?

What questions must a causation expert report address?

The causation expert report must answer a specific set of questions that reconstruct what would have happened with proper care, then compare that to what actually happened. Vague instructions produce vague reports, and vague reports lose cases. The following questions form the core framework solicitors use when instructing a causation expert in Irish clinical negligence claims.

The causation expert must address:

1. What was the patient's condition at the point when the breach occurred? Establish the clinical baseline: disease stage, symptoms present, test results available.
2. What was the natural progression of that condition without any clinical intervention? The expert must describe what the disease or injury would have done on its own, untreated.
3. What would a competent clinician have done at that point? Identify the correct clinical pathway: the tests, referrals, or treatments that should have followed.
4. What would the outcome have been with that correct clinical pathway? This is the counterfactual reconstruction: the "but-for world" where proper care was given.
5. What actually happened because of the negligent act or omission? Map the real clinical timeline from breach to final outcome.
6. On the balance of probabilities, did the breach cause or materially contribute to the difference between those two outcomes? The expert must state a clear opinion, not a hedge.
7. Can the injury be divided between negligent and non-negligent causes? If the condition was progressing anyway, the expert should quantify (where possible) the additional harm attributable to the error versus the harm that was inevitable.

Between questions four and five lies the entire causation case. The gap between "what should have happened" and "what did happen" is the injury the court compensates. An expert who cannot articulate that gap clearly will not survive cross-examination. One detail that surprises clients: the causation expert may never examine you in person. Causation reports are often prepared entirely from medical records and imaging, because the question is retrospective, not about your current condition.

When causation experts disagree: how does the court decide?

When the plaintiff's causation expert says the error changed the outcome and the defendant's expert says it didn't, the judge must choose between two competing reconstructions of medical history. This is the most common trial scenario in Irish clinical negligence. The court does not average the two positions or split the difference. It evaluates which expert's reasoning is more internally consistent, better supported by the clinical records, and more methodologically sound.

Under Practice Direction HC132, the court can order expert meetings (sometimes called "hot-tubbing") where the plaintiff's and defendant's causation experts meet without lawyers present. They produce a joint statement identifying exactly what they agree on and exactly what they disagree on. This process often narrows the causation dispute to a single factual question: for example, what stage was the cancer at on a specific date, or when exactly during labour did the hypoxic event begin?

The Crumlish judgment illustrates how this plays out. Two experts presented competing positions on tumour doubling time. The High Court evaluated the underlying data each relied on, found the plaintiff's expert had been influenced by confirmation bias, and preferred the defendant's reconstruction. The Court of Appeal upheld this approach, confirming that trial courts are entitled to weigh expert evidence and prefer one opinion over another provided the reasoning is explained ¹².

Where experts disagree, the strength of the underlying records often determines which one the court believes. Where the medical records are detailed and contemporaneous, the court has raw material to test each expert's reconstruction against. Where records are sparse or incomplete (a common problem in Irish public hospital claims), the expert who builds a coherent narrative from limited data has the advantage. This is why obtaining complete medical records at the earliest possible stage is not just a procedural step. It is the foundation of the entire causation case.

Why causation evidence fails: the Goldilocks problem and confirmation bias

The two most common reasons causation evidence fails in Irish courts are the "Goldilocks problem" in delayed diagnosis and expert confirmation bias. Both emerged as critical factors in Crumlish v HSE [2024] IECA 244, where the Court of Appeal dismissed a breast cancer claim at "the first causation hurdle" ¹².

The Goldilocks problem in cancer causation

In delayed cancer diagnosis claims, the defence deploys what practitioners call the Goldilocks problem. The argument works in two directions, and either direction defeats causation. A slow tumour doubling time means the cancer was not growing aggressively, so the delay made little clinical difference to the outcome. A fast doubling time means the cancer was undetectable at the earlier date and grew rapidly between screening points, making it an "interval cancer" rather than a missed cancer. The plaintiff is caught between two positions: too slow and the delay doesn't matter, too fast and the tumour wasn't there yet.

In Crumlish, the HSE's expert (Prof John Crown) successfully argued this exact position. The plaintiff relied on a 45-day tumour doubling time derived from a 1993 academic paper. The High Court rejected the reliability of that calculation and found the tumour was likely undetectable in May 2017. The Court of Appeal upheld this finding. What many people don't realise is that even where both sides agree the cancer existed at the earlier date, the plaintiff must still prove it was detectable by the imaging or clinical methods available at that time.

Confirmation bias in expert reports

The High Court in Crumlish found that the plaintiff's expert "espoused the plaintiff's case too closely" and appeared to "marry the data to the plaintiff's account rather than considering doubling time as a science in isolation." The court applied the framework from Duffy v McGee [2022] IECA 254, which holds that an expert who loses independence will have the weight of their evidence reduced.

This is a practical warning for solicitors instructing causation experts. The expert must be given the clinical records and asked to form an independent view. Framing the instruction as "can you confirm the delay caused harm?" rather than "what is your opinion on whether the delay changed the outcome?" risks producing a report the court will discount. The difference between those two questions is often the difference between winning and losing on causation.

Does open disclosure affect causation evidence?

An open disclosure under the Patient Safety Act 2023 is not an admission of negligence and does not prove causation. The Act, commenced on 26 September 2024, requires hospitals to notify patients and families when a notifiable patient safety incident occurs. The hospital must hold a formal meeting, provide a written statement about what happened, and offer an apology where appropriate.

However, the information disclosed can inform the causation analysis. Open disclosure records may reveal the hospital's own contemporaneous understanding of what went wrong and when, which can help a causation expert reconstruct the timeline. The critical distinction: the disclosure itself is a factual notification, not a concession on whether the error changed the outcome. Independent expert evidence on causation remains mandatory regardless of what was disclosed.

What breaks the chain of causation?

An intervening act (novus actus interveniens) can break the causal chain between the original negligence and the final injury, relieving the first defendant of liability for downstream harm. Defendants in Irish clinical negligence claims argue this when a patient receives further treatment after the initial error, and that subsequent treatment itself caused additional harm.

The traditional threshold for breaking the chain required "gross negligence" by a subsequent treating doctor. The 2023 English High Court decision in Jenkinson v Hertfordshire CC dismantled this specific rule, holding that any subsequent medical intervention should be assessed under standard foreseeability and fairness principles rather than requiring an artificially high threshold of gross negligence.

Conversely, defendants cannot escape liability because a patient had pre-existing vulnerabilities. The "eggshell skull" rule requires the defendant to take their victim as they find them. The Irish High Court reaffirmed this in Higgins v Coleman [2025], extending it to psychiatric vulnerabilities: depression or adjustment disorder arising from negligent treatment is a continuation of the original harm, not an intervening act.

The limits of causation also restrict claims by family members. In Germaine v Day [2024], the High Court dismissed a nervous shock claim by the widow of a patient whose cancer was negligently diagnosed late. The court held that because the cancer was already incurable at the time of the error, the defendant's negligence did not cause the deterioration the widow witnessed.

Does the patient's own conduct affect the causal chain?

A patient who missed follow-up appointments, ignored symptoms, or didn't comply with prescribed treatment may face a contributory negligence argument, but this does not automatically break causation. Under section 34 of the Civil Liability Act 1961 ³, the court can reduce the compensation award to reflect the patient's own contribution to the harm, rather than dismissing the claim entirely. The defendant must prove the patient's conduct fell below a reasonable standard and that it contributed to the injury. A patient who was too unwell to attend appointments, or who was not clearly told the importance of follow-up, has a strong answer to this argument. The distinction matters: contributory negligence reduces the award. It does not eliminate the defendant's liability for the portion of harm they caused.

How do the 2025 Practice Directions change causation strategy?

Practice Directions HC131 and HC132, effective 28 April 2025, established a dedicated Clinical Negligence List in the High Court and imposed strict judicial control over expert evidence on causation. The directions were issued by the President of the High Court, Mr Justice David Barniville. The full text is published on courts.ie (Clinical Negligence List) [12] and analysed in detail by William Fry (2025) [10].

HC132 curtails the practice of "expert stacking," where parties would submit overlapping expert reports from multiple sub-specialties to overwhelm the opposition. The court now controls the number and scope of permitted experts and requires parties to identify the specific causation issues in dispute at an early stage. HC131 mandates a formal Certificate of Compliance before any case can receive a trial date, confirming that all causation issues have been fully pleaded, discovery is complete, and expert reports exchanged.

These reforms compress the causation timeline. Plaintiff solicitors must crystallise their causal arguments and achieve expert consensus far earlier in the litigation process than was historically necessary. Mediation must also be undertaken within specific timeframes once a trial date is fixed. The timing matters more than most guides suggest: if your causation expert report is not finalised before the compliance certificate deadline, you risk losing your trial date entirely.

How causation findings affect your compensation

The extent of proven causation directly determines the scope of compensation. Full causation means the defendant pays for all losses flowing from the negligence. Partial causation, where negligence worsened but did not solely cause the injury, typically results in an apportioned award reflecting only the additional harm caused by the error.

Under the Civil Liability Act 1961, section 34 ³, contributory negligence on the part of the patient (such as failing to attend follow-up appointments or not disclosing symptoms) can also reduce the award. Compensation amounts are assessed against the Judicial Council Personal Injuries Guidelines (2021) [11], with awards varying case by case based on the severity and permanence of injury.

Loss of chance damages (under Philp v Ryan) are calculated separately from physical injury damages. The €100,000 ultimately awarded by the Supreme Court in Philp reflected the mental distress, lost treatment opportunity, and aggravated damages, not the cancer itself. In practice, loss-of-chance arguments are most useful as a fallback where but-for causation for the physical injury remains uncertain.

What if your causation situation is more complex?

Most Irish medical negligence claims fall into the patterns described above. Some cases, however, involve complications that don't fit neatly: multiple defendants where each denies responsibility for the harm, concurrent wrongdoers under the Civil Liability Act 1961, or injuries that develop months after the original clinical error. The question then shifts from "did the error cause my injury?" to "which error, by which defendant, caused which part of the injury?"

Common questions about causation in medical negligence

What to consider next

References and sources

1. Dunne v National Maternity Hospital [1989] IR 91. Supreme Court of Ireland. Delivered 14 April 1989 by Finlay CJ. Established the six Dunne principles governing medical negligence in Ireland.
2. Morrissey v HSE [2020] IESC 6. Supreme Court of Ireland. Delivered 19 March 2020 by Clarke CJ. Reaffirmed Dunne principles as the governing test for clinical negligence.
3. Civil Liability Act 1961. Irish Statute Book. Sections 11, 12, and 34 govern apportionment and contributory negligence.
4. Philp v Ryan [2004] IESC 105. Supreme Court of Ireland. Delivered 16 December 2004 by Fennelly J. Established loss of chance as compensable in Ireland.
5. Quinn v Mid-Western Health Board [2005] IESC 19. Supreme Court of Ireland. Delivered 8 April 2005 by Kearns J. Confirmed but-for test as primary causation standard; rejected broader alternative tests.
6. Kate Ahern, "Untangling the Threads". Law Society Gazette, November 2024. Analysis of material contribution doctrine in Irish law.
7. NTMA Annual Report 2024: State Claims Agency data. Published July 2025. Clinical damages, outstanding liabilities, and mediation rates.
8. Medical Protection Society, "The Human and Financial Cost of Clinical Negligence Claims". Published January 2024. Data on resolution times and legal costs across jurisdictions.
9. Medical Council of Ireland: guidance on complaints about a doctor.
10. William Fry, "Clinical Negligence List in Irish High Court Established". Published April 2025. Analysis of Practice Directions HC131 and HC132.
11. Judicial Council Personal Injuries Guidelines. Adopted 2021. Governs general damages assessment bands in Ireland.
12. Crumlish v HSE [2024] IECA 244. Court of Appeal. Delivered October 2024 by Noonan J. Dismissed delayed breast cancer diagnosis claim at "the first causation hurdle." See Courts.ie.
13. Patient Safety (Notifiable Incidents and Open Disclosure) Act 2023. Commenced 26 September 2024.