ROGER JOHNSON v ANDREW WILLIAMS

In a civil claim involving competing medical explanations, the claimant bears the burden of proving causation on the balance of probabilities. The court is not required to choose between competing explanations where the evidence leaves it in doubt. An improbable explanation may be accepted only where the evidence makes the alternatives effectively untenable. The court must apply common sense and must not use the so-called Sherlock Holmes fallacy: eliminating some possibilities does not establish that the remaining improbable possibility is more likely than not. Expert evidence must be assessed by considering its logic, reliability and consistency with the evidence as a whole.

The claimant, a professional footballer, alleged that the defendant knee surgeon negligently caused a large rupture of the medial retinaculum during surgery to treat an infected knee. By closing submissions, breach was accepted if the defendant had caused the defect, leaving causation as the sole issue. The claimant relied principally on the subsequent swelling, repeated aspirations and MRI evidence of a substantial tear. The defendant relied on the absence of an observed defect during surgery and washout, and on possible post-operative mechanisms involving infection, tissue weakening and movement. The central question was whether the claimant had proved, on the balance of probabilities, that the large defect was caused during the surgery.

1. The claim was dismissed. The claimant had not proved on the balance of probabilities that the defendant caused the approximately 3cm defect during the synovectomy.
2. The court accepted that some rupture probably existed before the egg-shaped swelling became apparent on 23 or 24 March. It remained uncertain whether the large defect arose during surgery or developed afterwards.
3. The evidence strongly militated against a large defect having been caused and repeatedly missed. The defendant had good visualisation during the procedures. Fluid would probably have extravasated through a substantial defect during the pressured washouts on 17 and 19 March. Some swelling would also probably have appeared earlier.
4. The court rejected the submission that the competing explanations could be excluded merely because they were uncommon or unlikely. Applying the approach explained in Rhesa Shipping Co SA v Edmunds [1985] 1 WLR 948, the burden remained throughout on the claimant. The court could conclude that the evidence left causation in doubt and was not obliged to select the claimant’s improbable explanation.
5. The court considered that a multifactorial mechanism, or the widening of a smaller non-negligently caused defect through post-operative movement, could not be ruled out. Those possibilities were sufficient to prevent a finding that the alleged surgical error was more likely than not.
6. In evaluating the expert evidence, the court applied the principles identified in Bolitho v City and Hackney HA [1998] AC 232 and C v Cumbria University Hospitals NHS Trust [2014] EWHC 61. The evidence had to be assessed for logic and against the totality of the evidence, including contemporaneous clinical records. The claimant’s expert’s inflexible reasoning was rejected, while the defendant’s late change of expert position reduced the weight of parts of that evidence.