When 10% is material: breast screening case appealed

Sydney South West Area Health Service v Stamoulis

The New South Wales Court of Appeal recently delivered its decision on the liability of BreastScreen NSW, Sydney South West (BreastScreen), for a delay in diagnosing breast cancer in a patient who had been part of its screening programme for five years. The patient could not allege BreastScreen caused her breast cancer - it did not. Her case arose from a complaint that BreastScreen, by failing to recall her for further testing in February 2006, had caused a 10 month delay in the diagnosis of her cancer and an increased risk that the cancer would metastasise.

The facts of the case

Ms O’Gorman attended BreastScreen for a screening mammogram of her breasts in 2002, 2004 and February 2006. Following the 2006 mammogram, two radiologists employed by BreastScreen separately and independently reviewed the mammogram and found that it was normal.

In January 2007, Ms O’Gorman detected a lump in her left breast. A mammogram and ultrasound revealed that she had breast cancer. Ms O’Gorman underwent chemotherapy, followed by a mastectomy in August 2007 and radiation therapy. In May 2008 metastatic tumours were found in her lungs and brain.

Ms O’Gorman brought proceedings against the operators of BreastScreen, alleging that BreastScreen was negligent in failing to inform her of a suspicious mass in her breast following the 2006 mammogram, failing to recall her for further testing and failing to refer the mammogram for further opinion.

First instance decision

At first instance, Justice Hoeben found that BreastScreen was negligent in failing to recall the plaintiff for further investigation following the 2006 mammogram, and that there was a 90% chance that the cancer would have been detected if she had been recalled. His Honour also accepted the expert evidence of both parties that the delay in diagnosis increased the risk of the cancer metastasising by approximately 10% (from 38% to 42%).

Despite arguments by BreastScreen that Ms O’Gorman’s loss should be assessed on the basis of having lost the chance of a better outcome, Justice Hoeben found that the metastasis of the cancer was a separate injury, the risk of which had been increased by the delay in diagnosis, and therefore causation was established.

BreastScreen appealed the decision, challenging His Honour’s findings on negligence and causation.

Decision of the New South Wales Court of Appeal

In assessing causation - that is, did the delay in diagnosis cause or contribute to the injury (metastatic cancer) - the Court stated that a material increase in the risk of injury followed by the occurrence of that risk does not equate to a material contribution for the purpose of causation. In order to prove causation, a plaintiff must establish that it was probable that the risk in fact came home.

Justice Giles described the phrase “risk coming home” as requiring a finding that it was probable that the injury was suffered because of the increased risk. Although it could not be said that the cancer would not have metastasised anyway, it could be said that statistically the risk of the cancer metastasising was increased and the plaintiff was brought materially closer to a point where there was a substantial and real chance that tumours would develop. On this basis, His Honour found that on the balance of probabilities the increased risk had materially contributed to the metastases.

Justice Ipp, with whom the other judges concurred, subject to the additional comments of Justice Giles, cited a number of cases recognising the dangers of applying evidence which tends to quantify possibilities mathematically and in applying epidemiological or statistical evidence in a mechanical way. His Honour noted that the fact that experts cannot infer causation on the balance of probabilities does not mean that a Court cannot - a causal connection may be found even when evidence does not go beyond the possible.

On appeal, BreastScreen argued that the case was on all fours with the decision in Gett v Tabet [2009] NSWCA 76 in which the Court found that the plaintiff had failed to prove, on the balance of probabilities, that she would have had a better outcome.

In Gett v Tabet the plaintiff argued that if a CT scan had been performed on 13 January 1991, it would have detected a brain tumour, allowing for earlier treatment to relieve intracranial pressure, thereby avoiding or minimising the effects of a seizure she had the following day.

In rejecting this argument, Justice Ipp noted that in Gett v Tabet the Court found that the plaintiff’s lost chance ranged between “speculative and having some effect.” In no respect did the evidence lend itself to a degree of precision.

However, in Ms O’Gorman’s case, there was statistical evidence that if the cancer had been detected in March 2006, there was a 62% chance it would not have metastasised. Whilst this alone did not establish causation, it showed that there was a strong possibility that if the cancer had been detected in March 2006 it would not have metastasised.

His Honour went onto find that evidence that the risk of metastasis had been increased by 10% by January 2007, a risk which in itself was not negligible and was sufficient to tip the scales from a strong possibility that the cancer would not have metastasised to a finding, on the balance of probabilities, that the delay caused the metastasis.

Whilst there was some argument on appeal as to whether it should be accepted that the 10% increased risk was a risk to Ms O’Gorman individually, as opposed to a general statistical risk, the Court found that either way the 10% increase was sufficient to support a finding of causation.

Expert evidence

During the first instance proceedings, in addition to the evidence of independent experts, BreastScreen sought to rely upon written statements of the two radiologists who had examined Ms O’Gorman’s 2006 mammogram and determined that it was “normal”.

These statements covered the radiologists’ interpretation of the 2006 films and the basis for their findings that the mammogram was “normal” and did not warrant recall for further investigation. Their opinions had in large been dismissed by the trial judge because it was felt, as employees of BreastScreen, they had an interest in the proceedings.

The Court of Appeal held that such evidence constituted expert evidence and was not inadmissible merely because the expert, in this case the radiologists, had an interest in the proceedings. The Court held that such evidence was admissible and that any issues regarding the reliability of the evidence, given the expert’s material interest in the proceedings, should go to the weight given to the evidence by the Court.

Implications

The Court of Appeal made it clear that mathematical possibilities and statistical and epidemiological evidence cannot be applied mechanically. However, the Court’s finding that the respondent’s failure to diagnose the cancer, and resultant increase in the chance of metastasis from 38%-42%, was in fact causative of the metastasis, is difficult to reconcile with its own determination - it is an application of the statistics.

It is difficult to see how a small decrease in the chance that metastasis would not have occurred, from 62% to 58%, was sufficient to justify a finding on the balance of probabilities that the delay materially contributed to the metastasis.

The decision is also difficult to reconcile with the findings made in Gett v Tabet where the Court determined that the plaintiff had lost a 15% chance, but went onto find that this loss had not, on the balance of probabilities materially contributed to the plaintiff’s brain damage.

Aside from the epidemiological evidence supporting the 10% chance in the present proceedings, the only difference between the cases appears to be that in O’Gorman, in the absence of any negligence, there was a greater likelihood of the cancer not metastasising, than there was of the plaintiff having an improved outcome if her tumour had been picked up one day earlier in Gett v Tabet. Notably, Justice Ipp commented that this in itself was a further basis distinguishing Gett v Tabet.

Rather than clarify the circumstances in which it is likely that an increased risk will be found to have materially contributed to a plaintiff’s injury, the case arguably muddies the waters as to the extent to which a risk must be increased in order to be found to have materially contributed to an injury.

 

Written by Kerrie Chambers, Partner and Ashleigh Lester, Solicitor.