Friday, 10 May 2019

Scope of Duty and Causation

There are few clinical negligence trials but it is surprising how, when they do come along, you see a cluster all dealing with the same issue. The currently fashionable issue is the relationship between causation and the scope of duty. Recent cases such as Duce (see here) and MNX (see here at case 21) have covered this issue and now we can add an interesting and detailed judgment from HHJ Cotter QC, the Designated Civil Judge in Bristol. Pomphrey v Secretary of State of Health and another is not yet on Bailii but is reported on Lawtel for those who subscribe. Judgment was given on 26 April 2019.

Mr Pomphrey criticised the defendants for failing to diagnose symptoms consistent with compression of the cauda equina nerve roots. The Judge found most of the allegations of breach not to be proved, but did find that there had been a negligent 10 day delay in operating on the Claimant to decompress the nerve roots of the cauda equina once the decision to operate had been made. The Claimant was unable to prove that the 10 day delay made any difference to the success of the operation - the delay by itself did not worsen his condition or cause him increased disability. However, during the operation the surgeon non-negligently caused a tear in the dura, leaving the Claimant with very severe and permanent neurological injury. 

The dura is the membrane that surrounds the spinal cord. The court heard that the tear probably occurred when the surgeon was using instrumentation to "bite" bone and ligament and, as the surgeon put it, "It's just pinched a little bit of the dural sac and that's opened a little window up." 

The risk of this non-negligent complication was of the order of 7%.

The Claimant contended that had the operation been performed, as it should, 10 days earlier, the chance of a dural tear would have been less than 50% . Therefore, on the balance of probabilities, but for the negligent delay, the complication would not have arisen and the Claimant would have been spared his serious injury.


The Claimant relied on a first instance judgment in Crossman v St George's [2016] EWHC 2878  (see case 33 here). In that case a plan for conservative management had negligently not been pursued, such that surgery - which would inevitably have been required in any event - was performed three months earlier than should have been the case. A non-negligent complication of radicular nerve damage arose and HHJ Peter Hughes QC, sitting in the High Court, held that on the balance of probabilities the complication would not have arisen had the surgery been performed on another date. It was a 0.5% to 1% risk only. Therefore the claimant should recover damages for the nerve damage caused because it would not have been caused but for the negligent delay. This was the finding even though the delay had not altered the extent of the risk of radicular nerve damage.

It is inexcusable to quote yourself but in September 2017 I gave the legal update to the annual PNBA Clinical Negligence conference and I wrote about Crossman in the handout, saying that it seemed to me to be a scope of duty case:

"Sometimes a strict “but for” test can result in spurious results – the mountain climber killed in the avalanche example. A GP negligently advises the mountaineer that he is fit for an expedition, when in fact he has a dodgy knee. But for the negligence the mountaineer would not have been on the mountain when the avalanche occurred. The conventional way of avoiding such results is to ask whether the avoidance of death by avalanche was within the scope of the duty of care of the GP. Likewise, it might be questioned whether avoidance of the risk of radicular nerve damage was within the scope of the duty of care of the doctor who changed the planned date of surgery. Would the Trust have been liable for any injury that occurred on the date of actual surgery? Suppose there had been an earthquake, the surgeon’s hand had slipped and severed an artery? On the reasoning in Crossman, wouldn’t the Trust have been liable?"

It would also follow from the Crossman approach, that a negligent delay in surgery of one hour, perhaps even of one minute, would be sufficient to establish causation.


In Pomphrey, HHJ Cotter QC found that causation was not established for two reasons.

First, he held that on the balance of probabilities the dural tear would have occurred even with earlier surgery. It would have been the same surgeon, he found, the same operation, the same instrumentation and the same anatomy (the Claimant's) as was the case when the tear occurred. There were no material differences between the circumstance of the operation as was, and the operation as it would have been but for the negligent delay:

"So I find as a fact that the delay made no difference to the outcome of the operation. Had there been no delay the operation would still have been carried out by Mr Patel and the same dural tear would have occurred." [278].

This was a finding on the particular facts of the case. A different judge might have come to a different conclusion on this issue. Naturally the fact that there is known to be a 7% risk of a complication arising does not necessarily mean that every patient comes to an operation with the same level of risk. The figure of 7% is a statistic that covers all patient-surgeon combinations. There may be certain patients who, because of their particular anatomy, and the surgeon's choice of instruments and technique, have a significantly higher risk of suffering the complication - but no-one could know that in advance. Other patients, for similar reasons, might in fact have no risk of suffering the complication unless the surgeon were negligent. In advance of the operation, the only advice that can be given is that statistics show that 7% of patients suffer this particular complication when undergoing this particular operation. However, it does not follow that just because the patient, surgeon, instrumentation and technique are "givens" that the same complication was going to arise whenever the operation was performed. On a different day the surgeon might well have approached the "biting" of bone and ligament slightly differently. "Non-negligence" covers a wide range of standards of care from the gold standard to just about acceptable. Just because the surgeon was not negligent when the dural tear was caused (or could not be proved to have been negligent) does not mean that he was on top form! There will be countless small variations in the performance of a standard operation. A painter using the same paints and brushes to create two portraits of the same subject, is unlikely to produce two identical paintings.   

The second reason why HHJ Cotter QC found against the Claimant on causation is more fundamental - it was not within the scope of duty of the Defendant to avoid a non-negligent complication (the dural tear), the risk of which was not affected by the breach of duty. As Defendant's Counsel, Andrew Kennedy put it to the Judge:

"the Defendant’s scope of duty did not extend to avoiding a risk inherent in the surgery that he was to
undergo. The fact that the Claimant sustained a dural tear was coincidental and not within the scope of the Defendant’s duty." [282]

The Judge had found that the dural tear would have occurred in any event, but:

"... given the scope of the relevant duty which was breached in this case (to avoid unreasonable delay) I would have declined Mr Samuel’s invitation to follow the reasoning/approach in Crossman and would have found that establishing simple “but for” causation ; based solely on the
operation taking place on a different day (or Mr Samuel suggested even at a different time on the same day) would not have been sufficient, without more,for the Claimant to establish causation. Indeed to do so would drive a coach and horses through well established causation principles." [297]

Note that Mr Samuel, Counsel for the Claimant, had recognised that the logic of the Crossman approach is that even a very short delay would establish causation.

Note that Chester v Afshar did not help the Claimant because this was not a consent case. Indeed in Chester, the court noted that on conventional principles causation would not have been established, and so it extended normal principle to allow for recovery of damages where a claimant had been deprived of their right to autonomy when making decisions about undergoing surgery that carried a risk of serious injury. Thus in a case that did not involve a loss of autonomy conventional principles remained. 

The Lesson from Pomphrey

The key lesson is that what the judge called an "unvarnished but for test" is  a necessary but insufficient test for establishing causation. The court should always also ask whether it was within the scope of the defendant's duty to prevent the injury in question. Defendants are not liable for coincidental injuries.

It would not be surprising if defendants try to run "scope of duty" more frequently in future clinical negligence litigation, but there is surely a risk of over-playing it. It is surely not always necessary for a claimant to establish that their precis injury should have been within the reasonable contemplation of the defendant at the time of breach. For example, a patient might have a constellation of signs and symptoms that are thought to be due to inflammatory bowel disease. Cancer is not within the differential diagnosis. There is a negligent failure to investigate the patient's condition. The investigations would have, surprisingly, revealed bowel cancer. Would it avail a defendant to say that because bowel cancer was, reasonably, not suspected before the investigations were performed, failure to diagnose and treat it were outwith its scope of duty? Sometimes medical investigations or procedures are carried out without knowing precisely what the outcome will be. 

One thing that is foreseeable, is that this run of cases on scope and causation is not over yet.