Causation: A negligent defendant's best friend!

Posted on July 1, 2010 by Julian Johnson

 I admit that I am partisan, when it comes to comment upon cases that I have been involved in.  Read the following bearing this in mind.  As you will gather, they are not 'rose-coloured' glasses I see this through..

I have previously posted in relation to the case of Wright v WA Country Health Services (effectively Broome Hospital), in which I have acted for Billy Wright.

The final (now sad) chapter in this case came with the Court of Appeal's decision yesterday, upholding the hospital's appeal and dismissing Billy's claim.  

The Court's conclusion was that the evidence at trial did not justify the conclusion that Billy would have been better off, had his infection which lead to his initial attendance at the hospital, been treated at such hospital, rather than his being at home.

The result left by this decision is as follows, in terms of the facts of the case:

1) Billy was negligently treated by hospital staff when he attended Broome Hospital's A+E department, following the onset of severe abdominal pain.  He should not (as he was) have been discharged after being kept for observation for an hour or so, when he continued to experience significant pain, the cause for which had not been identified.

The hospital did not appeal against this conclusion by the trial judge (it could not).

2) Had Billy been kept at the hospital (as he should), within an hour or so, it would have been identified that his condition was deteriorating.  He developed rigors and probably at such point had a temperature, indicative of infection.  Presumably, had this been identified, tests would have been done to identify the cause for this infection and (initially broad spectrum) antibiotics would have been commenced.

Again the hospital did not (and could) appeal against this conclusion.

3) Billy's condition deteriorated over the following 30+ hours, such that when his family took him back to Broome Hospital, he was in critical condition and at risk of dying.

The hospital did not appeal this, because Billy's condition when he re-attended the  hospital is self-evident from the records.

All of the above seems reasonable. The 'catch' comes at the next step....

4) The bug responsible for Billy's illness was never identified.  Blood cultures and x-rays taken after he returned to the hospital confirmed his critical illness but not its cause.  This is not remarkable.  Evidence at trial was that this can commonly occur.  Further, perfectly understandably given they were trying to save his life, staff at Broome Hospital had given Billy large doses of IV antibiotics when he returned to Broome Hospital, which it was accepted could explain why the bugs responsible for his infection were not identifiable.

Here is the killer step....

5) Because such bug could not be identified, the hospital's insurer argued (and this was accepted by the Court of Appeal) that it could not be proven that Billy would necessarily have been better off had he been kept at the hospital under observation, rather than being discharged, at his initial visit.  In other words, despite the fact they ought to have seen his progressing symptoms, if they had done the right thing and kept him at hospital, the view taken is that it could not be proven that any treatment at the hospital would have been effective and would have made any difference to his condition, compared with its development while he lay in bed at home.

Quite apart from the difficulty of this outcome as a matter of common sense (in my humble opinion) such conclusion also ignores the fact that uncertainty as to the identity of the responsible bug was at least contributed to by the hospital's negligence in the first place.  Had they done what they should have, kept Billy for observation and identified his deterioration within an hour or so, it seems to me that the relevant bug must have been a good chance of being identified by blood test etc at that time or at least, the further information as to Billy's progressive symptoms that would have been available would have assisted in identifying the most likely cause.  The problem of proving the type of bug present was directly a consequence of the hospital's negligence.

Despite my putting this last point to the Court of Appeal during argument - that this link between the hospital's negligence and any uncertainty as the bug's identity should tend towards a conclusion in BIlly's favour, the reasons for the Court's decision are silent on the point.

All in all a bitter outcome for my client and I.  My client, understandably in my view, has difficulty accepting the Law's reasoning as achieving a just result....

Tags: , , , , , , ,

Amaca -v- Ellis - An Anticlimax?

Posted on March 9, 2010 by Julian Johnson

 The High Court delivered its much anticipated decision in this case on 3 March 2010.  I have to say my impression is a sense of anticlimax...

As I have said in an earlier post, this case was seen amongst some (including me!) as providing an opportunity for the High Court to provide useful guidance as to the appropriate approach for deciding whether negligence or other breach of duty should be found to have caused an adverse outcome, in complex cases.  Cases involving 'evidentiary gaps' and were identifying whether or not a poor outcome was a consequence of the lack of care/negligence was difficult, because of the complexity of the science behind such conclusion etc, are some such cases.

The case, as is well known, was an asbestos exposure case.  Mr Ellis had been exposed to asbestos dust in the course of his work and by the time of this appeal there was no dispute that Amaca were at fault in allowing such exposure to occur.

The complexity to the case came from the fact that Mr Ellis's medical condition about which the claim arose (and from which he had died by the time of this appeal), was lung cancer.  

Apart from exposure to asbestos, on the scientific evidence, it was clear this could have resulted from his heavy tobacco smoking habit (15 to 20/day for 26 years).  In fact the unanimous scientific evidence was that such smoking was likely to have contributed to the occurrence of his cancer.  The key question was whether his asbestos exposure had also probably contributed.

The medical evidence confirmed (at least on Mr Ellis' case) that it was quite possible that both his smoking and asbestos exposure could have contributed to his cancer, either through a collective effect or as independent (but cumulative) causes.

The critical legal question was whether, in such circumstances Mr Ellis' exposure to asbestos could be said to have made "a material contribution" to the occurrence of his cancer (the accepted legal test).

Surprisingly, Bret Walker SC, the fabulous Counsel appearing on Mr Ellis' behalf, accepted for the purposes of this appeal, that the asbestos exposure should only be accepted as a cause of his lung cancer, if it were accepted on the evidence that it was more likely than not that had such exposure not occurred, his cancer would have been avoided (a lot of negatives in this proposition I know - don't worry, if you say it 20x you'll get the gist!).  

This was an acceptance of what is known as the 'but for' test, a traditional, non-exclusive test used by Courts to determine whether negligent conduct causes a particular bad outcome (or 'injury").

On the expert evidence (which was primarily epidemiological), the High Court found that such test was not met.  It was not  more likely than not that had Mr Ellis not been exposed to the asbestos he was, that his lung cancer would not have occurred.  This was therefore the end of his claim (given the concession by Mr Walker SC) and the appeal by Amaca was upheld.

In fact the majority of what was, by contemporary standards, a relatively short decision, dealt with this review/consideration of the complex scientific evidence on this topic.  No statements of broad application or modifications to the Law concerning the approach to deciding so called 'causation' questions were made.

The disappointment of all this (for me at least) is that it seems to me there was a question the case raised, which the court could have (but were not asked to) discussed and explained.  This was the fact it is clearly accepted  that in some circumstances negligent action can be accepted as causing harm even where it cannot be shown that had such negligent not occcurred, the relevant outcome would have been avoided.  In other words, where the 'but for' test is not met.  The most obvious category of such case is where 2 causes for such injury, each sufficient to cause the injury occur simultaneously.  Another is where multiple causes each contribute collectively to the occurrence of harm and the negligence is only one such cause.  This latter category is one which is confronted fairly often in medical negligence claims.

In this case, it seems to me that even if Mr Ellis' asbestos exposure could not be shown to have been necessary for the lung cancer to occur, this did not automatically mean it was not sufficient to be found to have in fact contributed to its occurrence, particularly where it was possible such exposure acted in conjunction with Mr Ellis' smoking.  It is a pity the Court were not asked and so did not comment on the appropriate approach to considering whether such asbestos exposure should have been found to have been a cause of Mr Ellis cancer, on this basis, if, as they did, they were not satisfied that it could be concluded that it would not have occurred but for such exposure.

It is also a pity the court were not asked to consider and discuss the role policy should play in deciding whether, in Law, negligent conduct should be found to have caused harm.  It is well recognised by previous case law that policy has a role to play, one would have thought most commonly in situations where science struggles to provide guidance as to whether an 'injury' is caused by negligence or not.  Because of the constraints on the way Mr Ellis' argument on appeal was run (no doubt for good reasons, though I am not privy to them), this issue was also not explored by the Court.

Tags: , , , , ,

Ellis Argued - now we wait!

Posted on November 9, 2009 by Julian Johnson

The High Court has now heard argument concerning this important (potentially ground breaking) decision and we are now in the Court's hands, for their decision.

The transcript of argument can be found here.  Although reading the transcript is not the same as hearing 1st hand, it seems Bret Walker SC had a torrid time, at least from some members of the Court!  

Interestingly, and this is close to a common concern in medical negligence claims, emphasis was placed by him on the capacity of the negligent party to know the information from which firmer conclusions about whether "A caused B" could be drawn.  It is often the case in medical negligence claims that when it is difficult to prove whether negligent medical care caused a particular bad outcome, that frustration is felt that the answer to such question would be known with certainty - if only the negligent care hadn't occurred (eg if the negligent failure to diagnose the breast cancer hadn't occur, we would have known whether there was or was not lymph node spread etc).

It seems to me that at least the more outspoken members of the Court during argument, expressed considerable scepticism at the approach of our Court of Appeal, in accepting that the Plaintiff's exposure to asbestos had contributed to his lung cancer, even though at least some evidence suggested it was 100 times more likely his cancer was a consequence of his smoking.

It also looks to me that the Court may be contemplating narrowing the traditional degree of contribution from negligent (or other tortious) conduct to an injury, for liability to arise.  

Traditionally, enough was shown if negligent action made a "material" contribution, which has often been accepted as met, when the contribution was more than negligible.  

in other words, if the Court accepted there were multiple causes for injury or illness, damages would be awarded if one of such causes was due to negligence (even if it was a minor or secondary cause - and even if it was likely the injury/illness may have occurred even if no such negligence occurred).  Further (and this has always been the difficult part to this, for me conceptually!), liability would then be for the entire consequences of the illness.  There would often not be a substantial reduction in the damages awarded, for the chances the same outcome would have occurred, even if the negligence had not occurred.

From my reading, several members of the Court were grappling, during argument, as to whether for a contribution to be 'material,' and so liability and damages follow, a greater extent of contribution should be required.

It will be interesting to see how the dynamics of argument translate into the Court's decision.

As a footnote, have to say I liked Justice Gummow's comment, when the submission was put that deciding whether a factor was or was not a cause of disease, was a matter of 'logic,' that:

"Logic itself is a house of many mansions"

Tags: , , , , , ,

A Balance of Probabilities does NOT mean >50% (at least in NSW)

Posted on September 3, 2009 by Julian Johnson

Further to an earlier post, the New South Wales Court of Appeal has recently delivered its decision concerning the appeal in relation to Ms O’Gorman’s delay in diagnosis of breast cancer claim.

Tragically in the period between the initial decision in her favour and this appeal, Ms O’Gorman died from her metastatic breast cancer.   As a consequence the appeal was contested by her Estate.
The Area Health Service responsible for BreastScreen New South Wales, the Sydney South-West Area Health Service succeeded in its appeal and the Court of Appeal have ordered a re-trial on the issue of negligence.

The primary reason for the Appeal being upheld and a re-trial being ordered was the Court of Appeal’s view that the trial judge had been wrong to refuse to allow the two radiologists involved in assessing Ms O’Gorman’s original breast images to give evidence as to whether they had taken reasonable care in examining her breast images.

This is a not uncommon situation. A defendant seeks to bolster its case by reinforcing its “independent” expert witnesses’ opinions by the views of the particular doctors concerned. I have had first hand experience of this in several cases at and approaching trial in the last two or three years.

Consistent with the New South Wales Court of Appeal’s decision, the generally accepted position is that such evidence cannot be put forward unless its substance has been disclosed an appropriate period prior to trial (as with any other expert’s opinion).  By this means the parties are given an appropriate opportunity to consider such opinion, its foundation and prepare cross examination etc.

Subject to this requirement, such expert evidence can generally be put forward, albeit (and this point was acknowledged by the New South Wales Court of Appeal), there will always be issues as to the weight to be attached to such opinion given the witnesses' lack of impartiality.

More interestingly in a general sense, the Court of Appeal in this case made some interesting observations in relation to causation questions.

The lead decision was provided by Ipp JA. Justice Ipp reiterated that in New South Wales at least, showing that negligent medical care has increased a risk of injury and such injury has occurred, is not enough to show that such injury occurred because of negligence.

The position in New South Wales at least, following Seltsam v McGuiness [2000] NSWCA 29 is that the Court must still be satisfied on a balance of probabilities that the risk of injury created by the lack of care is indeed the cause and the injury has happened because of the negligence medical care.

Perhaps I am missing something, but this does not explain to me how the facts in Ms O’Gorman’s case satisfied this test. Hopefully someone reading this can enlighten me!

As indicated in my earlier post, the findings by the trial judge (accepted on appeal) where that because of the delay in diagnosis and treatment of Ms O’Gorman’s cancer, her chances of metastases occurring had increased by 10% (though from what % is not clear to me).

Because this had occurred by the time of trial, I would have thought it could be looked at as a case in which Ms O’Gorman had lost, because of the alleged negligence, the 10% (or other figure, depending on her overall risk of this) chance that such metastasis would not have occurred. This was actually how the defendant tried to argue the case at trial, though by appeal time Gett v Tabbet [2009] NSWCA 76 had been decided and so any assessment of the loss as “a chance” would have had disastrous consequences for Ms O’Gorman's claim (her estate).

It could be said that the strong likelihood on such evidence was that the negligence had not caused her metastases and death. Certainly on a scientific balance of probabilities, it appears it was > 50% probability that such metastases and death would have occurred in any case.

The trial judge’s finding was that rather than a lost chance, what BreastScreen’s negligence had caused was the metastasis itself (ie in effect, somehow she was in the % who otherwise would not have developed such spread). This was accepted as the right approach by Justice Ipp and other members of the Court.

Although I confess I have difficulty in understanding the Court’s reasons (no doubt, due to limitations on my part), my impression is that there is a “likelihood threshold” at play.  In other words, where the increase in risk of a particular injury by negligent care is very small (or uncertain), the “jump” to accept that when such form of injury has occurred, this is because of the negligence, will not be made.

Put another way, though I hasten to add the Court did not say this (!), for an increase in risk to be found to have caused injury, the increase in risk may need to be more than just material. This seemed Ipp JA’s explanation (not in so many words) why Gett v Tabbet did not exclude success here. At [131] Ipp JA suggested that Gett’s case failed because the patient had been unable to establish "there would, as distinct from may have been a better medical outcome." If this is the test, I would have thought Ms O’Gorman should have failed as well.

Ipp JA suggested (I think unfairly) that the evidence in Gett as to the prospects for the patient with better care being improved was “at most, speculative” [132].

Ipp JA’s decision goes on to make interesting reading in terms of the difference between “probabilities” in maths or science and on the other hand, in Law. He refers to statements in Jones v Sutherland Shire Council [1979] 2 NSWLR 206 as to the meaning of the critical legal phrase “a balance of probabilities”. This emphasises an impressionistic, subjective concept. The quote reads:

“The first step, in a sense, simply involves an assessment of what is human experience: it is, or is not, a fact that human experience is so. But such an assessment is not made as a mathematical calculation. It involves other kinds of reasoning and judgements, the correctness of which cannot be demonstrated by mathematics or ordinary logic. Therefore, the subjective confidence which a person or a Court will have in the correctness of the assessment may vary.”

As can be seen therefore, it is absolutely wrong, though very common (in this jurisdiction at least!), to suggest in a medical negligence case that whether proven negligence can be shown to have caused damage is a matter decided by the expert evidence and expert opinions, as to whether as a matter of science, this is likely.

It is true that expert evidence will inform the Judge concerning the scientific issues relevant to a mathematical assessment of the probability of a link between negligent care and the outcome.

As Justice Ipp and the Court of Appeal’s decision however illustrates, this is only a first step in the process. The Judge’s “valued” judgement as to whether in law a link should be accepted must then be performed and may quite legitimately reach a contrary conclusion to science (as perhaps in Ms O’Gorman’s case).

Ipp JA relies upon this meaning of “a balance of probabilities” in concluding that the scientifically evident “strong possibility” (which as stated, was in fact 10% or so) that had the radiologist detected the tumour it would not have metastasised, would not require a “great deal of evidence… to tip the scales from the strong possibility to a finding that, on a balance of probabilities, the failure to detect the tumour lead to metastases” [148].

Obviously these issues of the role of science in proving causation in complex litigation and what should happen where science is unable to explain or assist a Court as to a potential link between potentially negligent acts and adverse outcomes (so called “evidentiary gap” cases) is something the High Court is likely to clarify when Amaca and Ellis are delivered.  Watch this space!
Tags: , , , , , , , , , ,

Awaiting Amaca and Ellis

Posted on August 4, 2009 by Julian Johnson

 One rumour I have heard suggests that the High Court will be asked in Amaca and Ellis to review the degree of causation needed for negligent action to be taken to have caused loss.  These critical cases are due to be heard by the Court, later this year, probably in October.

At present (and for as long as I have been in practice), it is sufficient if the negligence made a "not-insignificant" (or material) contribution to the injury.  In other words, it doesn't matter that the bad outcome was caused by a range of factors, provided one of them was the doctor/hospital/other health provider's negligent care.

I noted with interest similar reform efforts reported in NY in the US, where a move is being considered to bring the Law in line with other States, which require, apparently, a predominant (>50%) contribution to the injury or other bad outcome, from the negligent care.

Lets hope this trend isn't picked up in Australia.  Unlike NY, there is no evidence to suggest a need to curb litigation, particularly in the health services sector.  What is needed is clarity in the approach to complex cases, where science struggles with answering whether negligence has contributed to loss.

Tags: , , , ,

Failure to Warn - another Warning (about how difficult these cases are)

Posted on June 14, 2009 by Julian Johnson

Reading the recent West Australian Court of Appeal decision in Gingin -v- Coomb [2009] WASCA 92, handed down last month.  This was a case concerning a catastrophically injured young man who suffered injury when he lost control of his trail-bike when riding it in a designated off-road recreational area, near Lancelin, a beach side town an hour or so North of Perth.

Unfortunately for Mr Coomb, the Court of Appeal (2:1) reversed the trial judge's finding that the Shire were in breach of the duty they owed him, in failing to more adequately warn him of the potential for such injury when riding in the dune area.

I have to say I am very surprised at the evidence at trial (which it does not seem was contradicted) that because the dunes in the designated area were "ever changing, depending on wind strength and direction" (Martin CJ @ R26), it was not feasible for the Shire to inspect the area and cordon off obviously dangerous areas, despite it being found the Shire encouraged use of such area (R20) and that a designated area was set aside for this activity (and was very popular).

It surprises me that it seems the dune face Mr Coombs fell down, which was 10 - 15m high at an angle of 80 degrees, with a hard rock bottom, would not have been an obvious hazard, had it been inspected even a month or 2 before the accident.  In fact very shortly prior to Mr Coombs' accident (on the same day) another rider was fatally injured at the precise same location. This suggests the area was commonly crossed and again, in my view, indicates that inspection/identification of the particular and unusual hazard of the particular location ought to have been possible. The danger (or near inevitability of injuries) for someone riding at anything other than very low speed, over such a dune, and confronting such a face, seems absolutely clear as a matter of common sense.

The case is another example of the difficulty of demonstrating claims, based on an alleged failure to warn. See the contrasting factual conclusions by Martin CJ (in the majority) -v- McLure JA on this.

Once again, the case faced a fundamental evidentiary gap, because Mr Coombs was not asked (as he should have been) and so did not say, how a better sign warning of the particular risk of this sort of dune, would have changed the way he approached the relevant dune on his trail bike (appreciating the difficulty of this, because he had no recollection of the accident).

Although this is not a medical negligence case, it is relevant, given a 'failure to warn' is a very common type of medical negligence or malpractice case investigated.  The key question of whether the person making claim can show that if warned, they would have acted differently (in a medical context, would not have had the surgery etc), remains the most substantial hurdle for claims.

image: GrahamKing 

Tags: , , , , , ,