Causation and the 'but for' test: explaining medical causation in an expert report

Most of the attention in a clinical negligence case goes to breach of duty: was the care acceptable, and if not, where did it fall short. But a finding of breach decides nothing on its own. The claim still has to clear causation, the question of whether that breach actually made a difference to the outcome, and it is on causation that otherwise strong cases quietly fail. An expert report that treats causation as an afterthought, or folds it into the criticism of the care, leaves the most contestable part of the case underbuilt.

What is causation, and why does it decide cases?

Causation is the bridge between the failing and the harm. It is not enough to establish that the treatment fell below the standard a responsible body of practitioners would have met. The claimant must also show that the substandard care caused, or materially contributed to, the injury complained of, and must do so on the balance of probabilities: more likely than not.

This is where cases turn because breach and consequence are genuinely separable. A clinician can manage a patient poorly and the patient can still come to the same harm they would have suffered with flawless care, because the underlying condition was always going to take that course. When that is so, the claim fails on causation no matter how clear the breach. The expert who understands this builds the causation analysis with the same care as the breach analysis, rather than assuming one follows from the other.

What is the 'but for' test?

The standard test for causation is the 'but for' test, and it asks one disciplined question: but for the breach, would the claimant have avoided the harm? If the answer is that the injury would probably have happened anyway, the breach did not cause it in the eyes of the law, however poor the care was.

The test forces the expert to confront a counterfactual: what would probably have happened if the patient had been managed properly. That is a clinical judgement, and it is the heart of a causation opinion. It is not answered by restating that the care was substandard. It is answered by setting out, on the evidence, the course the condition would most likely have taken with appropriate treatment, and comparing it with the course it actually took.

How should an expert structure a causation opinion?

The most common failing is to treat causation as a continuation of the criticism rather than as a separate question. A sound opinion keeps the two analyses visibly apart and works the causation question through in its own right.

• State the actual outcome: what in fact happened to the patient, tied to the records that show it.
• State the counterfactual: what would probably have happened with non-negligent care, and on what clinical basis.
• Identify the difference between the two, which is the injury attributable to the breach rather than to the underlying condition.
• Address timing where it matters: not only whether harm would have occurred, but when, since a delay that changes the outcome is itself the causative link.
• Say plainly which test is being applied, and if it is material contribution rather than 'but for', explain why.

Set out this way, the opinion lets the reader follow the reasoning from the records to the conclusion and test it at each step. Collapsed into a single paragraph that asserts the breach caused the harm, it invites exactly the cross-examination it should be built to withstand.

When does material contribution apply?

The 'but for' test does not fit every case. Where an injury is indivisible and several factors combined to produce it, it can be impossible to say that the breach alone, on its own, caused the harm. In that situation the law allows a claimant to succeed by showing that the breach made a material contribution to the injury, meaning a contribution that was more than negligible.

This is a different question, and an expert who drifts into it without saying so creates confusion. If the opinion rests on material contribution, it should name the route, explain why the ordinary 'but for' analysis does not work on these facts, and address the contribution the breach made rather than implying it was the sole cause. Used precisely, it is a legitimate and important tool. Used loosely, as a way of avoiding a hard counterfactual, it weakens the report.

Why does the factual basis matter so much in causation?

A causation opinion is only as good as the chronology underneath it, because the counterfactual is built entirely on what the records show and when. The timing of a presentation, the point at which a test result became available, the gap between a referral and a clinic letter: these are the facts that decide whether earlier action would have changed the outcome, and each of them has to be traceable to its source page.

A causation opinion is a clinical argument about timing and sequence. If the dates underneath it cannot be located in the records, neither can the opinion.

This is where a sound factual foundation does its quiet work. When every key date is anchored to the document it came from, the expert can build the counterfactual on solid ground, and can defend it when the other side disputes the sequence. When the chronology is loose, the causation analysis inherits that looseness, and a single misremembered date can undo an otherwise careful opinion.

What are the common mistakes in a causation analysis?

The recurring errors are predictable. The first is assuming causation follows from breach, so the opinion proves the care was poor and stops there. The second is failing to address the counterfactual, criticising what was done without saying what proper care would have achieved. The third is confusing the tests, leaning on material contribution without acknowledging it or applying it where the ordinary test would do.

The most damaging mistake is the unsupported timing claim: the assertion that earlier treatment would have changed the outcome, made without grounding it in the dates the records actually show. It may well be right, but if the sequence cannot be demonstrated, the opinion cannot be defended, and causation is precisely where the other side will press hardest.

How does a clear causation analysis support the expert's duty to the court?

Under Part 35 the expert's duty is to help the court on matters within their expertise, and causation is often the matter that most needs that help. A judge has to decide, on competing accounts, what would probably have happened with proper care. The expert who lays out the actual outcome, the counterfactual, and the difference between them, each tied to the evidence, gives the court something it can actually use. The expert who asserts a conclusion does not.

It is also what holds up under cross-examination and in the experts' discussion. A causation opinion built on a sourced sequence keeps the argument about clinical judgement rather than about whose recollection of the bundle is correct. The counterfactual stays anchored to the records, and the expert is never reduced to defending a date they can no longer find.

Why it matters

Causation is where a well-evidenced case is either carried home or quietly lost, and it rests on getting the timing and sequence right. That is the principle ALLDOQ is built on. It reads the whole record, assembles the chronology in one auditable file, and attaches the source page to every entry, so the expert builds the counterfactual on dates that can be opened and checked rather than on memory. The sequence is in order, the timing is defensible, and every fact the causation opinion turns on can be traced back to the page it came from, which is exactly what a causation analysis needs to stand up.