Arksey v Cambridge University Hospitals NHS Foundation Trust

In a clinical negligence claim, the claimant bears the burden of proving factual causation. An increase in risk does not, by itself, establish that the breach made a material contribution to the injury. The material-contribution approach is confined to cases involving contributory causes and operates as an evidential route to proving causation. It does not replace the ‘but for’ test in an ordinary single-cause case.

A court may draw reasonable inferences from the evidence, but cannot invent past or hypothetical facts to overcome evidential gaps. In assessing delayed treatment, the court must evaluate the clinical decision at the time it was made and avoid reasoning retrospectively from a later successful outcome.

The claimant alleged that the defendant hospital negligently failed to admit her after a sentinel bleed from a cerebral aneurysm on 3 November 2012. She suffered a major subarachnoid haemorrhage the following day and sustained significant disability. The defendant admitted the failure to admit and scan, but disputed causation.

The claimant also alleged negligent delay in inserting a ventricular-peritoneal shunt for hydrocephalus. The defendant admitted that the shunt should have been inserted between 31 January and 14 May 2014, but denied any earlier breach. The court was required to determine causation in relation to the aneurysm claim and the proper timing and consequences of the shunt treatment.

1. Causation. The claimant had to prove that the admitted failure to admit and scan caused the relevant damage. Following Gregg v Scott [2005] 2 AC 76, evidential difficulty caused by negligence did not remove that burden. The court could draw reasonable inferences, but could not invent past or hypothetical facts.
2. The evidence did not establish that admission, bed rest, monitoring or advice would have delayed or prevented the re-bleed. The claimant’s proposed comparison between hospital and out-of-hospital re-bleeding lacked reliable comparative data and did not establish that any difference in blood pressure was causative. An increase in risk could not be translated into causation without a proper evidential basis.
3. The material-contribution doctrine, developed through Bonnington Castings v Wardlaw [1956] AC 613 and McGhee v National Coal Board [1973] 1 WLR 1, is confined to cases involving contributory causes. It provides an evidential means of proving material contribution where the breach increased the risk, but it does not generally displace the ‘but for’ test. The approach in Chester v Afshar [2004] 3 WLR 297 remained an exceptional departure from the usual rule.
4. Shunt treatment. The decision to offer a shunt after the claimant’s recovery had plateaued and convincing ventriculomegaly had been identified was reasonable. The treatment was offered in the hope, rather than expectation, of improvement. Its later success could not establish that it should have been offered earlier. The appropriate assessment was the information available when the clinical decisions were made.
5. Judgment was entered for the claimant only for the admitted delay in inserting the shunt between 31 January and 14 May 2014. The aneurysm-related claim and the allegation of any earlier shunt breach failed. The longer-term outcome was not materially altered by the admitted delay; its causative effect was limited to postponing recovery, with damages to be assessed or agreed later.

First-instance judgment. No appellate history was stated in the judgment.