Mario Schembri v Ian Marshall [2020] EWCA Civ 358

The Court of Appeal addressed the appropriate approach to causation in a medical negligence case where a particular mechanism for a positive outcome could not be proven.

Full judgment, dated 10 March 2020


The court was presented with a case where, had the Appellant not failed to refer the deceased to hospital, she would not have died. The judge at first instance found that the Respondent could not prove that the deceased would have survived by any particular mechanism, but nevertheless the tenor of the statistics and the expert evidence was such that, taking a holistic view, she probably would have done.

The court found that the judge’s approach was clear and was consistent with the previous authorities and dismissed the appeal, giving guidance on the approach to causation and, in particular, statistical evidence in such cases.


This appeal related to a failure by the Appellant to refer the Respondent’s late wife (“the deceased”) to hospital. She had a history of pulmonary embolism and attended at his surgery complaining of chest pain and breathlessness. A admitted that he should have referred D to hospital that afternoon. He didn’t, and the following morning D suffered a cardiac arrest the following morning from which she did not survive.

The parties agreed that had D been referred to hospital by A she would have been diagnosed with pulmonary embolism and treated with either anticoagulants or thrombolysis drugs. They further agreed that she would have been prescribed heparin (the anticoagulant of choice) by at the latest 9pm of the day she saw A. The difference between the parties was to what would have happened after that.

McCombe LJ gave the judgment of the court. At paragraph 10 onwards he set out the disputed matters. R alleged in his particulars that D would both (i) probably have avoided the pulmonary embolism in the first place and (ii) have survived the same had it occurred if she had been in hospital. A denied the same, stating that the clot that embolised on the morning of D’s death was likely there the day before and the heparin would not have dispersed or dissolved it in the less than 12 hours available, only stopped it growing. Whilst the fibrinolytic enzymes in the blood gradually dissolve the clot and reduce the risk of part of it detaching and causing PE, that takes more time than 12 hours. Accordingly, D would still have died regardless of earlier provision of heparin.

By the end of trial, the parties’ positions were:

  • for R, that though heparin alone would not have prevented the death, it would have had a beneficial effect and was relevant to causation as a whole; and
  • for A, that heparin would have stopped the size of the clot in the leg from increasing, but that it would have remained at 95% of its eventual size (the judge accepted this evidence).

From paragraph 13 onwards, Stewart J’s statement of the issues is reproduced. The questions he posed himself were whether R had proven that there were progressive emboli overnight before D’s death, if so, whether these would have been picked up in hospital, if so, what thrombolysis would have been prescribed and with what effect. If the former were negative, would thrombolysis given on arrest have saved her? If not, is it, looking at the evidence holistically, nevertheless more likely than not that D would have survived had she been in hospital?

The judge found that there were no progressive emboli during the night. He found that had there been they would have been picked up, but thrombolysis would not have been prescribed. 

He considered whether thrombolysis given on arrest would have saved her with relation to the medical literature. The review is at 16-19. The papers give figures for survival that are far higher than 50%, but the judge commented about the differences between some of the papers and D’s situation. The judge found that had D been given alteplase (the thrombolysis drug of choice) three hours before her arrest, she probably would have survived. However, he had already found that she would not have been prescribed it at any point before her arrest. 

He then reviewed the literature on provision of thrombolysis in cardiac shock and arrest. It suggested that more than 50% of people in shock survived, but less than 50% of those in arrest did. However, the figures included (i) those already in shock or arrest on presentation (which would not have been the case) and (ii) people who were over 75, which was not the case. 

Interestingly, though D had only been in cardiogenic shock for a couple of minutes, the length of the same had she been in hospital receiving treatment and with a crash team on call may have been very different. 

Looking at that issue in isolation, the judge found that R couldn’t prove that D would have survived, but her chances would have been significantly increased. The meaning of that finding was disputed between the parties.

Moving to his final question, whether D would have survived, taking a holistic approach, the judge referred to Drake v Harbour [2008] EWCA Civ 25, in which Toulson LJ said at 28:

In the absence of any positive evidence of breach of duty, merely to show that a claimant’s loss was consistent with breach of duty by the defendant would not prove breach of duty if it would also be consistent with a credible non-negligent explanation. But where a claimant proves both that a defendant was negligent and that loss ensued which was of a kind likely to have resulted from such negligence, this will ordinarily be enough to enable a court to infer that it was probably so caused, even if the claimant is unable to prove positively the precise mechanism. That is not a principle of law nor does it involve an alteration in the burden of proof; rather, it is a matter of applying common sense. The court must consider any alternative theories of causation advanced by the defendant before reaching its conclusion about where the probability lies. If it concludes that the only alternative suggestions put forward by the defendant are on balance improbable, that is likely to fortify the court’s conclusion that it is legitimate to infer that the loss was caused by the proven negligence.

The judge went on to say that R did not have to identify the precise mechanism by which D would have survived, he only had to show she would have. He then said that he had made a number of “close call” findings based on hypothetical situations. Essentially, he made the point that finely balanced decisions on a significant number of highly technical, wholly hypothetical situations, had to be weighed against “important overall evidence”. These factors were that overall most people do not die of PE, D was not elderly and was in good health. And that notwithstanding that she would have been diagnosed with PE on arrival, the arrest she suffered would have been so sudden and massive that it could not be effectively treated (29-34).

Ultimately the judge found that it was inappropriate to infer what would have happened in hospital solely from what happened outside of hospital, particularly by way of overly detailed or forensic examination of particular issues. He felt that the clear balance of the literature and of the expert evidence was in favour of the conclusion that D would have survived (35).

A, unsurprisingly, appealed on the basis that the judge, having found R had not proved his case on the detailed analysis, should not have posed himself the separate, holistic question. Further, he should not have based his finding on generalities and he failed to give sufficient weight to the A’s experts.

R counter-appealed on the basis that even if the judgment was not upheld on the primary basis, the judge should have found that C would have survived if she had been given thrombolysis only on going into cardiogenic shock.

McCombe LJ addressed the competing arguments from paragraph 38 onwards. He firstly set out R’s case concerning not having to set out a precise cause and A’s acknowledgment of the same, while stating that where the judge had specifically found that the R’s case was not proven on the specific issues he should not have gone on to find that it was on some more general basis. Then he recited A’s argument in reliance on Wardlaw v Farrar [2003] EWCA Civ 1719, in which (in a similar situation to this case) the court of appeal had rejected the use of general statistical evidence to challenge specific findings on causation (41-44). Particular reliance was placed on Brooke LJ’s dicta that “[w]hile judges are of course entitled to place such weight on statistical evidence as is appropriate, they must not blind themselves to the effect of other evidence which might put a particular patient in a particular category, regardless of the general probabilities.”

In response, R placed reliance on a passage from Clerk & Lindsell at para 2-30, which reads:

On the other hand, care should be taken not to take the logic of this reasoning too far in the opposite direction. If the evidence is that, say, 80 per cent of patients survive with prompt treatment, but 20 per cent die even with prompt treatment, the fact that the patient died following delayed treatment does not establish that he probably fell into the 20 per cent category at the outset and therefore the delay did not contribute to the death. The assessment of causation would turn upon the detailed medical evidence, both as to the overall statistical chances of survival and the particular condition and circumstances of the patient.

Finally, McCombe LJ referred to the judgment of Lord Nicholls in Gregg v Scott [2005] 2 AC 176 at paras 27, 28 and 32. At 27 and 28 Lord Nicholls acknowledged that in clinical negligence cases statistics may be the main evidential aid to what would have happened in hypothetical situations, but they are still only trends, which have to be applied to a particular Claimant. At 32 he stated that the quality of statistics must also be assessed. Nevertheless:

statistical evidence of a diminution in perceived prospects will often be the nearest one can get to evidence of diminution of actual prospects in a particular caseThis conclusion is the more compelling when it is recalled that the reason why the actual outcome for the claimant patient if treated promptly is not known is that the defendant by his negligence prevented that outcome becoming known.”

At para 50 McCombe LJ found that a key issue was that the judge’s finding on whether or not D would have survived if treated with thrombolysis on going into shock was that, although R couldn’t prove that she would have, she may or may not have. The judge legitimately paused for thought, after asking the first three questions, and considered statistical evidence that was highly favourable to R. That was particularly legitimate because of the large number of unknowns in this case. The judge’s setting out of the questions was clear and amounted to primarily asking whether a particular mechanism for avoiding the death could be established, but that secondarily involved asking, in the absence of an identifiable particular mechanism, whether or not it was nevertheless likely that D would have survived (52-57). The judge’s approach was correct and the appeal should be dismissed.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s