New South Wales Court of Appeal CITATION : Forbes v Selleys Pty Limited [2004] NSWCA 149 HEARING DATE(S) : 27 October 2003 JUDGMENT DATE : 12 May 2004 JUDGMENT OF : Mason P at 1; Giles JA at 137; McColl JA at 138 DECISION : Appeal dismissed with costs. CATCHWORDS : NEGLIGENCE - product liability - causation - sudden illness after use of product - conflict among experts as to diagnosis and as to probable impact of particular chemical component (MDI) of product - possible cause not probable cause - treating doctor deferred to expert toxicologists as to critical scientific issue - trial judge's advantage - complex trial involving experts. (D) LEGISLATION CITED : Evidence Act 1995 (NSW), s 76 Trade Practices Act 1974 (Cth), Pt VA, s 75A, s 75AD, s 75AK Ahmedi v Ahmedi (1991) 23 NSWLR 288 Bennett v Minister of Community Welfare (1992) 176 CLR 408 Betts v Whittingslowe (1945) 71 CLR 637 Chappell v Hart (1998) 195 CLR 232 Commonwealth v Amann Aviation Pty Ltd (1991) 174 CLR 64 Forbes v Selleys Pty Ltd [2002] NSWSC 547 Fox v Percy [2003] HCA 22 CASES CITED : HG v The Queen (1999) 197 CLR 414 Jackamarra v Krakouer (1998) 195 CLR 516 Jones v Dunkel (1959) 101 CLR 298 Murphy v The Queen (1989) 167 CLR 94 Naxakis v Western General Hospital (1999) 197 CLR 269 Sellars v Adelaide Petroleum NL (1994) 179 CLR 332 Seltsam Pty Ltd v McGuinness (2000) 49 NSWLR 262 TC by his Tutor Sabatino v The State of New South Wales & Ors [2001] NSWCA 380 Warren v Coombes (1970) 142 CLR 531 PARTIES : Charles Barstow Wright Forbes (Appellant) Selleys Pty Limited (Respondent) FILE NUMBER(S) : CA 40640/02 COUNSEL : G. M. Watson SC/S.B. Loughnan (Appellant) M. T. McCulloch/D.F. Villa (Respondent) SOLICITORS : Matthews Folbigg (Appellant) Phillips Fox (Respondent) LOWER COURT Supreme Court JURISDICTION : LOWER COURT SC 20241/00 FILE NUMBER(S) : LOWER COURT Cripps AJ JUDICIAL OFFICER : IN THE SUPREME COURT OF NEW SOUTH WALES COURT OF APPEAL CA 40640/02 SC 20241/00 MASON P GILES JA McCOLL JA Wednesday, 12 May 2004 Charles Barstow Wright FORBES v SELLEYS PTY LIMITED JUDGMENT 1 MASON P: The appellant sued the respondent claiming damages for injuries arising out of a seizure he suffered on Monday 17 February 1997. It was alleged that it was due to toxic encephalopathy as a consequence of using a product known as "Selleys Space Invader" (SSI) on 15 February 1997. The respondent is the manufacturer of that product. 2 The claim, as pressed at trial, was based upon: • Part VA (s75A) of the Trade Practices Act 1974 (Cth) ; and • Negligence. 3 Each count depended upon the respondent's failure to warn of the risk of toxic encephalopathy caused by inhalation of SSI. Of course, more had to be established before liability ensued. The respondent also invoked statutory defences (cf s75AK) in relation to the trade practices count. 4 The claim failed because Cripps AJ was not satisfied that the appellant's illness was caused by the ingestion of fumes generated by the SSI (Forbes v Selleys Pty Ltd [2002] NSWSC 547). The other issues did not have to be addressed. 5 The conclusion was expressed in the following terms (J43): Conclusion I am not persuaded, on the balance of probabilities, that exposure to MDI (which I find to have been in the substance used by Mr Forbes on 15 February) caused or materially contributed to the encephalopathy he suffered. In law, I am required to be satisfied on the balance of probabilities that the plaintiff suffered toxic encephalopathy by reason of the use of SSI. At first blush it might be thought that the proximity between the use of SSI and the symptoms which ultimately lead to the plaintiff's coma establish a probable cause. But that finding depends on the capacity of MDI to have that consequence. Of course, none of the studies can prove conclusively that there could be no possible connection. But to my mind, the studies demonstrate that a relevant connection is rarely raised above a mere possibility. Thus it is possible, of course, that the plaintiff did suffer a toxic encephalopathy from the inhalation of MDI. However, I am not satisfied on the balance of probabilities that this was so. It is possible that he did not, notwithstanding that bacterial and viral causes (to the extent that they were capable of being identified) were eliminated. As I have said, Dr Darveniza said that had he understood the reference to "central nervous depression" as being a reference to the anaesthetic component (and not to a toxic component), he would have diagnosed the plaintiff's condition as an encephalopathy of unknown origin. 6 The appellant seeks a verdict in his favour, alternatively a new trial, contending that the judge erred in his approach to the causation issue and/or in his analysis of the facts. It is also submitted that the judge failed to give adequate reasons on key issues. Facts 7 Prior to Saturday 15 February 1997 the appellant purchased two 500g cans of SSI for the purpose of sealing part of a fireplace. He spent about 15-30 minutes that day applying the material. It was found that he followed the instructions on the can, in that he was careful to ensure that the area was well ventilated, he opened the windows and doors, and he placed two fans in appropriate positions. There was a breeze passing through the house. Nevertheless, from time to time his face was close to the area to which the SSI was directed. There was a strong draft coming down the chimney at the time. 8 The appellant's last memory was washing his hands after using the SSI. His next was of waking up some days later in hospital. What happened in the meantime was recounted in a history given by his wife which the trial judge accepted as correct. The appellant developed a headache in the afternoon of 15 February, which worsened on 16 February. He appeared unable to attend to domestic chores and there was a slight slurring of speech at lunchtime on 16 February. He went to the movies in the afternoon, but could not drive. In the evening, his condition progressively worsened and he spoke with "rambling incoherent words" and had difficulty with comprehension. He had a very restless night. On Monday 17 February he presented as ready for work dishevelled and incompletely dressed with his shirt out, trousers undone and without socks or shoes. He would not go to the doctor. He was driven to work, where shortly after arrival he collapsed and was taken to St Vincent's Hospital. He was admitted at about 9am in an unconscious condition. 9 He remained in a deep coma for at least two days, during which period he was intubated. The provisional diagnosis was encephalitis and it was thought he might not live. Subsequently, he made a rapid recovery and was discharged from hospital on 28 February 1997. He now has significant physical, psychosocial and cognitive loss. The illness has severely affected his working capacity and contributed to the breakdown of his marriage. 10 The appellant gave evidence that he read the warning label before purchasing the product. He would not have bought the cans if there had been a warning of a risk of headaches or dizziness or nausea on normal use, let alone a warning about the type of illness he suffered. 11 The history obtained from Mrs Forbes on the appellant's admission to hospital included reference that the appellant had an upper respiratory tract infection a week earlier; and that he had "donated blood as usual [on 14 February] but complained of feeling unusually tired". She said that he complained of headaches on the Saturday afternoon "not relieved by disprin but otherwise well". On the Sunday the headaches worsened, but the appellant was able to do some work at home and go to the movies in the afternoon. There was a progressive worsening of speech in the evening, with difficulty in comprehension. 12 The treating doctor was Dr Darveniza, a specialist neurologist. 13 Dr Darveniza's initial diagnosis was some sort of viral or bacterial encephalitis. (This diagnosis would not have implicated SSI.) The doctor arranged for several tests to be administered and commenced urgent treatment of what initially presented as a life-threatening illness. 14 The history given by Mrs Forbes at the hospital included reference to the use of SSI the previous Saturday. Dr Darveniza contacted the respondent, which immediately made available its MATERIAL SAFETY DATA SHEET (MSDS). This document provided the following information about the chemical makeup of SSI: CHEMICAL ENTITY PROPORTION Isocyanate terminated polymer V High 4,4' - Diphenylmethane di-isocyanate [MDI] Low 1,1,1,2 – Tetrafluoroethane [HFC 134a] Low Dimethylether Low Proportion (% weight per weight): V High >60%, High 30 -60%, Medium 10 - 30%, Low < 10%. 15 The case was fought on the basis that the particular cans of SSI used by the appellant contained quantities of chemicals as indicated in the label and the MSDS. 16 Under the heading "Health Hazard Information" there is a further subheading entitled "Inhaled". Under that was written the following: Vapour and spray mist is irritant to mucous membranes and respiratory tract. Inhalation of vapour or spray mist can result in headaches, dizziness, nausea and possible breathing difficulties due to respiratory sensitisation which may be delayed. Inhalation of high concentrations can produce central nervous depression, which can lead to loss of co-ordination, impaired judgment, and if exposure is prolonged, unconsciousness. 17 Dr Darveniza drew certain conclusions from this statement. They were later shown to be misplaced. 18 The SSI cans displayed information and/or warnings which became relevant to two issues: the nature of the warning to users given by the manufacturer and whether the manufacturer admitted the potential toxicity of its product in a relevant respect. The warnings included: SELLEYS SPACE INVADER is a polyurethane foam that expands on contact with moisture in the air. Use Selleys Space Invader inside and outside the house to fill large gaps, hollows and cavities, to block out draughts, dust, noise, vermin and birds. Selleys Space Invader has strong adhesion to most surfaces • can be cut, sanded, plastered and painted • is durable, and will not shrink • expands up to 30 times can content • uses an ozone friendly propellant • has a fire retardant additive. … DIRECTIONS Wear gloves and eye protection. Selleys Space Invader is difficult to remove from skin and clothing. • Shake can vigorously for 60 seconds. • Screw applicator nozzle onto the can valve (do not over tighten) and turn can upside down. • Gently apply pressure to the trigger to extrude Selleys Space Invader. (Vary pressure to control rate of flow.) • Fill approximately half the space. • Do not overfill as Selleys Space Invader will expand up to 2½ times the volume of the product extruded. • Continue to shake the can from time to time during use. • Selleys Space Invader will be tack free in about 20 minutes, and fully cured after approximately 5 hours. NOTES • Do not empty can in one continuous application. Confined or large spaces are best filled in 2 to 4 layers, allowing each to cure before applying the next. … CAUTION • Do not incinerate or puncture this can even when empty. • Keep upright in a cool place out of the sun. • Allow plenty of ventilation when applying. • Not suitable as a floation (sic) device. WARNING Intentional misuse by deliberately concentrating and inhaling contents can be harmful or fatal. SAFETY DIRECTIONS • Avoid contact of uncured product with skin and eyes. • Wear protective gloves and eye protection when applying…. Overview of trial on causation 19 This was a trial that mainly concentrated upon hard-fought scientific and medical issues. Considerable time and effort were spent in the investigation of hypotheses which were ultimately discarded completely. Some hypotheses remained, although very much in the realm of possibilities rather than probabilities. 20 The evidence on the causation issue included the account of the use of the product and the history of the development of symptoms to which reference has already been made. This included the hospital records. There was the evidence of the treating neurologist Dr Darveniza as to his observations, treatment and diagnoses of Mr Forbes. The appellant also relied upon the expert evidence of Dr Crank, a consultant toxicologist. He in turn cited studies from scientific literature. Most notably these included the "firemen studies" undertaken in 1976 to which reference will later be made. 21 The respondent relied principally upon the expert evidence of Dr Drew, who is a toxicologist. Dr Drew cited studies from scientific literature. There was also expert evidence from Dr Bisbey, a specialist in occupational medicine, and Dr Spira, a consultant neurologist. 22 Drs Darveniza, Crank and Drew were extensively cross-examined. These were the main witnesses upon whom the primary judge relied in forming his conclusions. 23 There were other witnesses whose evidence was tendered at trial. These included Dr Rawlinson, a specialist in the diagnosis and treatment of infections, whose report was tendered but not made the subject of cross-examination. The respondent relies upon this report in this Court. The appellant counters with the submission that, if it becomes necessary to bolster the judgment by reference to this further material, the judgment is deficient for exposure of essential reasoning process. The judgment makes no reference to Dr Rawlinson's report. 24 The critical issue was whether Mr Forbes' use of SSI on the Saturday caused or materially contributed to the medical condition he was suffering from when brought to St Vincent's Hospital on the following Monday. This involved determining the nature of his illness and whether a component of the SSI (the isocyanate called MDI) had caused it. 25 It is not suggested that the trial judge was equipped to determine either issue otherwise than according to expert evidence. His Honour had to resolve the conflict between the principal experts, Drs Drew and Crank, who "stood opposite each other at forty paces" (Mr Watson SC at CA Tr p20). 26 The evidence of the treating specialist, Dr Darveniza was obviously critical as to the nature of the disease, although it will be seen that his ultimate diagnosis was guarded as to aetiology. He nevertheless concluded that the patient had suffered "a toxic encephalopathy". 27 Dr Darveniza described encephalopathy and distinguished it from encephalitis in the following terms: Encephalopathy is some pathological process involving diffusely the whole brain of various causes, a wide variety of cause, whereas an encephalitis by definition is an inflammation of the brain, usually due to an infectious organism, but sometimes infectious organisms can cause encephalopathy themselves sometimes as well, so it is usually an inflammation due to a virus, bacteria, a fungi, and usually it is divided into acute encephalitis if the presentation is acute. Sometimes viruses can present as a chronic encephalitis over many many years, but they are very rare, and then occasionally there are other inflammatory conditions of the brain, encephalitis, which are not due to an infection, for instance, due to autoimmune diseases affecting the brain, for instance, and sometimes after an infection you get, you have the infection but then several days later you will come down with an encephalitic illness which is post infectious. It is not a direct viral infection, it is a post response to a previous virus. 28 The treatment administered to the appellant and the basis of the doctor's final diagnosis were described in the following terms: Well, obviously this man presented with an acute severe coma developing over a couple of days, and when he presented to the emergency department in fact he was in deep coma. So the management of such a patient really revolves around resuscitation was the first duty, to ensure an airway is obtained, that the person is breathing, airway, breathing and circulation to make sure they have got a circulation, because one cause for coma would be a cardiac arrest, for instance, if the heart stopped and whatever. And then your second thing is to resuscitate the patient and make sure their airway breathing and circulation are adequate and controlled, and then once you have done that often that diagnoses the coma on the way. Then the emergency department people would frequently be giving intravenous glucose on the speculative basis to make sure it wasn't a hypoglycaemic coma, and at St Vincents, where drug overdose is quite common, which in fact, when I think of it that way, the commonest cause of encephalopathy at our hospital is probably toxic due to drugs if you include all the drug overdoses that come into the hospital. Now, those encephalopathies, those toxic encephalopathies, are often reversible when the drug is metabolised. So the next thing would be an examination of the patient looking for causes of coma. So in addition to the blood pressure, the pulse, you would be doing a neurological examination looking for signs, raised intracranial pressure or abnormal reflexes or some sort of localising signs that tell you where the abnormality in the nervous system might be, and if there are no signs or symptoms which would point to encephalitis or a meningoencephalitis , meanwhile your investigations are underway - blood tests, make sure you know what their much (sic) blood count shows, what their renal function shows, all these metabolic causes of coma are looked for at the same time. Then if you don't find a metabolic cause as you are continuing on with your investigations and management of this patient, the next investigation would be probably a scan, a brain scan, either a CT or MRI if it was available, and then obviously a lumbar puncture. Now, quite honestly, if you do a lumbar puncture on the patient in this situation and it is normal, for all intents and purposes that's very much against an infective encephalitis due to a virus or a bacteria, particularly if the patient is in deep coma. So one would have grave doubts about the diagnosis of encephalitis in the absence of inflammatory cells in the spinal fluid. This gentleman had 2 lymphocytes. The number of normal lymphocytes in the spinal fluid is 5. The fact he had 2 provides no discrimination about the cause of the coma. That's within normal limits. This gentleman also had a mild rise in his spinal fluid protein. The normal range is up to 400 milligrams per litre and his was 480. Now, again that has no value, no discrimination value in terms of diagnosis, because it is not up enough. You could lie down for half an hour and get a protein level like that. In fact recumbency will put it up to twice normal, so the spinal fluid protein was of no value in discriminating. Now, on the other hand-- Q. Sorry, just to finish that, indiscriminating or determining that it was toxic? A. Determining whether it was toxic or due to a virus. Now, if anything, if it was due to a virus it is more likely to be elevated than if it is due to an encephalopathy. The fact it was near normal was more in keeping with an encephalopathic picture rather than an inflammatory infective picture. The other point in the CSP examination too was the spinal fluid pressure was normal, and again that would more point to an encephalopathic problem, an encephalopathy rather than an encephalitis. So now this man was in dangerous straits, he was in deep coma, in fact critically ill and there was some concern in the first 24 to 48 hours whether he might in fact die he was so deeply comatosed, and we did treat him, as I said in my report, as a meningoencephalitis with [cover] bacteria and viruses on a just case basis, you know, just in case. These are the very unusual form of an encephalitis we are seeing here. In a man that was just about to die you couldn't withhold that sort of treatment, anyway. But, as things turned out there was no substantiation for that diagnosis. Within two days he had made quite remarkable recovery, I was quite surprised, actually. Within 48 hours he seemed to be out of the woods almost entirely, and again that would be very much an encephalitis that was severe enough to produce coma. For someone with encephalitis to be in deep coma it is very unusual for them to make such a spontaneous and rapid recovery to the level that he did, whereas, as you might understand, that people with hypoglycaemia, low blood sugar, if you give them sugar then it gets better quickly. The same with toxic substances too. If the brain damage is not too severe then initially they often make quite a good recovery. Q. How relevant is your clinical assessment of him and the rate of recovery to a diagnosis? OBJECTION McCULLOCH: It may just be the form of the question. He is being asked how relevant it is. HIS HONOUR : What significance he attaches to it. WITNESS: the significance was it was against an infective form-- HIS HONOUR: Q. Is that because if it was encephalitis you would take a much longer time to come out of the coma? A. If you just have a bit of a headache you might recover quickly, if you are slightly drowsy you might recover quickly, but to be in the coma that puts you near death or close to death, in fact the commonest cause of infectious encephalitis is herpes simplex encephalitis due to the herpes virus which causes the cold sore, and if patients with herpes simplex encephalitis present in coma the prognosis for death, the mortality is 50%. So that is really a dividing point in encephalitis. If they present with coma, the mortality is 50%. So by the time a person has deep coma from an infective encephalitis their outcome is poor, so I would be very surprised with someone with an infective encephalitis to bounce back as he did to 90% of his potential, I suppose, within two or three days. HENNESSY : Q. Could you go on and then tell us the steps you took in ultimately coming to your conclusion? A. We did treat him as encephalitis without thinking that was the cause, and we did extensive investigations, as my report states, including MRI of the head, blood cultures, all sorts of things looking for other causes of coma and we didn't really come up with any specific results to diagnose any other cause of coma. For instance, we did his blood lead and Mercury levels, serology for arborviruses, urine drug screens - they are very important in coma, particularly around St Vincent's, urinary drug screens looking for common drugs which cause coma, and we couldn't find any of those. The only thing that became evident was this exposure to this particular substance, and in fact the family came with that history, that, you know, they were concerned about his exposure to the substance as well. OBJECTION . PRESSED . ALLOWED. Q. Go on, doctor. A. Well, when it transpired that the Space Invader, whatever you want to call it, is reported to have toxic effects, to me it seemed very likely that this was the most likely cause. Most household substances, whether it says it on the bottle or not, are toxic, I can tell you. …. HENNESSY: Q. Your diagnosis of toxic, that took into account the results of the various tests that you have outlined? A. Yes. Q. Together with your clinical assessment of the patient? A. Together with my clinical assessment of the patient, yes. 29 The evidence in this lengthy passage is at the heart of the appellant's case in the appeal. It demonstrates the processes whereby the treating specialist arrived at his diagnosis of toxic encephalopathy and his conclusion, reached during the treatment phase, that the respondent's product was the toxic culprit. The reasoning of the primary judge 30 Cripps AJ accepted the accuracy of the medical history provided by Mrs Forbes and recorded in the hospital notes. He also accepted the evidence of Dr Darveniza as to the treatment administered at hospital. 31 Cripps AJ also held that the appellant had followed the instructions on the SSI can (J12). 32 The tests and investigations supervised by Dr Darveniza eliminated the known causes of encephalitis, both bacterial and viral, that were capable of detection by such testing (J15). 33 Two matters became common ground at the trial (J20). First, it was agreed that, if a constituent part of SSI were capable of causing the problems experienced by the appellant, that part would be MDI. Second, it was agreed that the "health hazard information" in the MSDS relating to risks stemming from inhalation as disclosed in the MSDS stemmed from the HFC134a component, not the MDI. It followed that the MSDS did not contain an admission that the product was toxic in a relevant respect. 34 Most if not all matter is toxic in some sense and to some degree. For a time in this Court it appeared that senior counsel for the appellant was inviting the Court to revisit a case that included HFC 134a as the relevant toxin. Ultimately it was agreed that such a case had not been pressed at trial. It also became clear that the trial and appeal focussed on the possible and probably toxicity of SSI, not in the sense of its capacity to cause skin and airways reactions, but as a potential inducer of encephalopathy. 35 Against this background, Cripps AJ addressed the central issue of whether he was persuaded that MDI in the product used by Mr Forbes had caused the illness suffered. 36 As indicated, Dr Darveniza gave evidence that he diagnosed a toxic encephalopathy. He could not however totally exclude a viral cause, but thought that a virus was less likely than any other possible cause. Cripps AJ accepted this diagnosis and proceeded to examine the evidence on the basis that, more probably than not, the illness was toxic encephalopathy (J15, 20, 27, 28, 42). This did not however involve the judge excluding the possibility of viral or bacterial encephalitis (a possibility not excluded by Dr Darveniza himself) (see J27, 42, 43). 37 As Dr Darveniza's treatment of his patient progressed, he came to favour the diagnoses that (a) the illness was toxic encephalopathy and (b) SSI was the toxin. He did not abandon those diagnoses in his testimony, although he acknowledged the possibility of an alternative viral illness and of an alternative (unknown) toxin. 38 The primary judge found in effect that Dr Darveniza's views on the SSI being the toxin rested upon an assumption that Dr Darveniza did not seek to justify by reference to his personal expertise or experience (J21, 26, 32-33, 42, 43). This particular summary is disputed by the appellant, but I shall later seek to justify it. 39 The correctness of that assumption was rejected by his Honour on the probabilities, based particularly upon his assessment of the scientific evidence of Dr Drew. Diagnosis 40 Many of the appellant's submissions proceed from a diagnosis of toxic encephalopathy. From this springboard, the ultimate issue was posed as that of determining the probable toxin. 41 This does not accurately reflect the expert evidence and it skews the proper characterisation of the ultimate issue. That issue became one of determining whether it was proved that the MDI in the product had caused the illness actually suffered. 42 Since the evidence of all the experts recognised the possibility of an alternative diagnosis and alternative toxins, it was not incumbent on the judge to exclude a non-toxic cause merely because the probabilities pointed in the direction of toxic encephalopathy. 43 When a court is faced with determining on the balance of probabilities whether an event has occurred it treats the event as certain if the probability of it having occurred is greater than it not having occurred (Commonwealth v Amann Aviation Pty Ltd (1991) 174 CLR 64 at 122-3, Sellars v Adelaide Petroleum NL (1994) 179 CLR 332 at 365-7). This principle does not, however, enable the appellant to argue that Cripps AJ erred when he declined to treat as irrelevant the possibility that the appellant's illness was encephalitis unconnected with SSI. The evidence of Drs Darveniza, Drew and Rawlinson supported the possibility that the illness was not encephalopathy. The issue or event that had to be established on the balance of probabilities was whether SSI caused whatever illness eventuated. While Dr Darveniza thought it more probable than not that the illness was toxic encephalopathy (and this evidence was accepted by the primary judge), the causation issue remained one based on circumstantial evidence. 44 In Seltsam Pty Ltd v McGuinness (2000) 49 NSWLR 262 Spigelman CJ said (at 278): The courts must determine the existence of a causal relationship on the balance of probabilities. However, as is the case with all circumstantial evidence, an inference as to the probabilities may be drawn from a number of pieces of particular evidence, each piece of which does not itself rise above the level of possibility. Epidemiological studies and expert opinions based on such studies are able to form 'strands in a cable' of a circumstantial case. 45 The possibility that the disease was encephalitis was a "strand in the cable" to which the primary judge was entitled, indeed bound, to give some weight. 46 Dr Darveniza distinguished encephalopathy and encephalitis in the following terms: An encephalitis is a condition of the brain due to an inflammation of the brain and an encephalopathy is due, is a condition of the brain due to other pathological processes. 47 Dr Spira described encephalopathy as a collective noun referring to disorders of brain function. He said that it is "therefore not specific and is certainly not tantamount to a diagnosis". 48 The appellant's recovery meant that it was not possible for every known viral or bacterial cause to be tested and excluded. Dr Darveniza excluded as many known possible causes as he was able, but he recognised that a number of possibilities remained (untested for inclusion or exclusion). He pointed to tests that were done and observations made that firmly supported his preferred diagnosis. These are summarised in the appellant's written submissions. Having regard to the tests actually performed, Dr Darveniza thought that a viral cause was less likely than an alternative cause, but he could not and therefore did not totally exclude a viral cause (J27, 42, 43). These findings were supported by Dr Darveniza's evidence. 49 They are also supported by the opinion of Dr Rawlinson. 50 Senior counsel for the appellant submitted that this shows that the primary judge was approaching the ultimate issue as a search for a scientifically-proven case, contrary to the law's more pragmatic approach. I disagree. The judge was merely giving proper weight to Dr Darveniza's clinical opinion. Since that opinion did not exclude the possibility of viral or bacterial encephalitis the judge was not required to exclude it either, even when he accepted Dr Darveniza's conclusion that toxic encephalopathy was the probable diagnosis. Dr Darveniza's assumption as to the toxicity of SSI 51 Cripps AJ ultimately gave no weight to Dr Darveniza's opinion that SSI was the probable or possible toxin that caused the probable toxic encephalopathy that the doctor was well qualified to diagnose. His Honour effectively considered the doctor's view as to the toxin to be no better than that of a lay person who, being told of the history and the probable diagnosis of toxic encephalopathy, might readily conclude that there was a causal link to SSI. 52 There was debate before us as to whether this conclusion rested upon a concession by Dr Darveniza or a finding by the judge based upon his assessment of Dr Darveniza's testimony. There were also challenges as to the ambiguity and correctness of the finding. 53 Cripps AJ referred to this issue at several points in his reasoning. This indicated the importance of the matter in his disposition of the proceedings. He said: 21 Dr Darveniza did not know the chemical makeup of SSI and, as he has said, he is not a toxicologist. His opinion concerning the relationship between the inhalation of MDI and the plaintiff's encephalopathy was the statement in the MSDS together with the history given to him by Mrs Forbes. … 26 I record that I accept the opinion expressed by Dr Darveniza to be one that was honestly held. Nonetheless, and bearing in mind that had Dr Darveniza not been aware of the MSDS or, being aware of it he knew that the hazard information did not relate to a toxic substance in the SSI, he would have expressed the opinion that the plaintiff suffered an encephalopathy as a result of an unknown cause. … 32 Dr Darveniza agreed that if SSI was implicated that would not have affected his treatment. That statement does not mean, as submitted by the defendant, that unlike Dr Spira, Dr Darveniza was not concerned with the issue of cause. Dr Darveniza made it clear he was concerned with knowing the cause because that knowledge might have effected the treatment. However, in answer to the following question, " So to be clear about it, in the case of the diagnosis relating to Mr Forbes, you did not attribute the toxic event to any particular component individually; is that correct?" He said, "No, because it was beyond my expertise at the time". 33 As I have said, had Dr Darveniza understood that the hazard information in the MSDS was directed to a non-toxic substance, his diagnosis would have been encephalopathy of unknown cause. …. 42 … As I have said, had he not seen a reference [in the MSDS] to the "central nervous depression" (which, with respect to Dr Darveniza he misinterpreted) he would have simply recorded "encephalopathy cause unknown". …. 43 … As I have said, Dr Darveniza said that had he understood the reference to "central nervous depression" as being a reference to the anaesthetic component (and not to a toxic component), he would have diagnosed the plaintiff's condition as an encephalopathy of unknown origin. 54 These findings were consistent in their substance and were, in my view, justified on a fair reading of Dr Darveniza's evidence (see above). 55 The appellant contended otherwise. He also submitted that the conclusions gave no weight to the doctor's "clinical feel". I do not agree. That "clinical feel" enabled Dr Darveniza to eliminate the probability (but not the possibility) of viral or bacterial encephalitis. This left toxic encephalopathy as his preferred diagnosis. But nothing in Dr Darveniza's evidence embraced the probability that SSI contained the toxin except for (1) the doctor's initial but misplaced assumption that the warnings in the MSDS were directed at the MDI constituent and (2) (possibly) the doctor's assumption as a non-toxicologist that SSI contained the toxin having regard to the medical history and the hypotheses formed as a result of the treatment and rapid recovery of the patient. 56 Dr Darveniza did not claim expertise as an expert on isocyanate or as a toxicologist generally. By his own lights, he did not profess a capacity based on training, study or experience to express an admissible opinion on what had become the critical issue, ie the probability that the MDI component of SSI caused the appellant's illness. It followed that any explicit or implicit opinion from Dr Darveniza on this matter of specialist knowledge was inadmissible in light of s76 of the Evidence Act 1995 (NSW) (see HG v The Queen (1999) 197 CLR 414 at 427-8, 432, 442, 449, 454). On the matter at issue, Dr Darveniza did not claim knowledge or experience in relation to toxicology "outside the experience and knowledge of the judge and jury" (Murphy v The Queen (1989) 167 CLR 94 at 111, cited by Gaudron J in HG at 432[60]). Senior counsel for the appellant ultimately accepted this proposition. 57 Cripps AJ concluded from his interpretation of Dr Darveniza's evidence that, had the doctor known that the hazard information related to HFC 134a and not to a toxic substance in the SSI, he would have diagnosed encephalopathy of unknown origin (J26, 42, 43). The appellant challenges this finding, contending that it put words in the doctor's mouth. It is submitted that Darveniza's concession that HFC 134a was acquitted of causal responsibility should not have been restated (by the judge) as a diagnosis of encephalopathy resulting from "an unknown cause" (J26) or of "unknown origin" (J43). The appellant also submitted that the judge erred in equating an expert's concession that something was "possible" because it had not been excluded according to the rigours of scientific proof with a statement that the possibility had to be excluded altogether in the context of a civil trial. The particular vice is said to be that the judge put entirely aside the medical specialist's generally informed "clinical feel". 58 In my view, the judge did not fall into this error. As indicated, his final conclusion recognised the possibility that the MDI in the Selley's product was the cause of toxic encephalopathy, which was the appellant's probable illness. Furthermore, Dr Darveniza signalled his own agnosticism as to MDI, deferring to the toxicologists once his own conclusions based upon the warning in the label were found (and accepted by him) to be misplaced. Dr Darveniza nevertheless stuck to his guns to the extent that he favoured a diagnosis of toxic encephalopathy although he could not exclude alternate hypotheses. It was this that the judge was recognising when he summarised his understanding of Dr Darveniza's "bottom line" in the terms indicated in the previous paragraph. I therefore reject the submission that the judge misstated or misunderstood the substance of Dr Darveniza's ultimate diagnosis. I also reject the submission that the judge confused scientific and legal proof. 59 There was in fact a body of positive evidence suggesting more than purely hypothetical alternative possibilities. This was the evidence of Drs Spira and Rawlinson. It is unfortunate that Dr Rawlinson's evidence is not recorded or considered by the trial judge in his reasons. This counsels significant caution in any reliance upon it by the respondent to bolster the judgment in its favour. Nevertheless, the credibility of those two experts does not appear to have been challenged at trial and Dr Spira's views found general favour with the trial judge (J22, 41). Their evidence should therefore be given some weight in so far as it casts light on the ultimate issues. It can confidently be said that nothing in their evidence supports the appellant. 60 At one stage in his reasons the trial judge indicated that he placed "some significance" in the evidence (recorded in hospital progress notes) that the appellant "donated blood as usual but complained of feeling unusually tired". The appellant denied that he was "unusually" tired. He had given blood before and on the other occasions he had felt tired all day. He had also had an upper respiratory tract infection in the week before his collapse. The judge did not calibrate the significance of these factors. Dr Darveniza thought it was "unlikely" that the complaint of tiredness was an early symptom related to the appellant's eventual collapse. Ultimately he conceded and explained that it was "remotely possible" that there was a connection between the incident and the condition treated. These concessions show why the trial judge had no difficulty accepting Dr Darveniza's credibility. But his answers also demonstrate why it was open to the judge to conclude that a diagnosis of encephalitis had not been excluded. His Honour did not err in refusing to put encephalitis entirely out of the picture. The evidence as to the possible and probable impacts of MDI 61 The finding that SSI was used by Mr Forbes according to the instructions on the label was the basis upon which the scientific issues were joined. 62 Those issues addressed the capacity of MDI to cause neurological injury such as toxic encephalopathy. 63 Cripps AJ found that MDI was a possible cause of encephalopathy, but concluded that the MDI in the product used by the appellant was not the probable cause of his illness, even if that illness was encephalopathy (J28, 43). 64 MDI is described in the MSDS as: 4, 4 – Diphenylmethan di-isocyanate. It is referred to in the evidence as an isocyanate. 65 One way in which the appellant sought to prove the potential danger of isocyanates in general and MDI in particular was by reference to the warning on the SSI cans to the following effect: WARNING Intentional misuse by deliberately concentrating and inhaling contents can be harmful or fatal. 66 In this Court at least, the appellant sought to use this as a springboard for proving the causal link by admission. He also submitted that the respondent bore an evidentiary onus that it had failed to sustain as regards excluding SSI as the probably culprit. He joined this with a Jones v Dunkel (1959) 101 CLR 298 submission to the effect that the respondent must have had something in mind by the warning, whereas the respondent refrained from calling any evidence to explain it or to counter the admission implicit in it. The authors of the MSDS should have been called. 67 Apart from Dr Darveniza, the way in which the appellant sought to advance his toxicological case about MDI was by the evidence of Dr Crank, and the literature searches he conducted. 68 Dr Crank is a PhD with qualifications in organic chemistry. He is an independent consultant in chemistry and toxicology and has taught toxicology at the University of New South Wales for 26 years. He said that it is generally accepted by the medical and scientific community that the common isocyanates, such as TDI and MDI, have similar effects on the human body. He described isocyanates as well-known toxic and irritant compounds which can have an adverse effects on users, mostly by breathing fumes or dust or by skin contact. MDI irritates all mucous membranes including the eyes, nose, throat, mouth, the respiratory passages and the lungs. Once a person has become sensitised to an isocyanate the sensitisation is usually lifelong and it may be intense. 69 The difficulty for the appellant was that his illness was not preceded by respiratory problems of the type identified. Nor was Dr Crank's sensitisation hypothesis pressed by himself or the appellant. The case was fought on the basis of establishing that toxic encephalopathy stemming from MDI did not have to be preceded or accompanied by these symptoms. 70 Dr Crank referred to a study carried out on fireman exposed to the isocyanate TDI as a result of a factory fire in 1967. Some of the firemen were exposed while fighting the fire, but others were exposed while cleaning up the site and handling contaminated material. 15 cases of gastrointestinal distress were reported including four whose symptoms were delayed. 23 firemen complained of neurological symptoms. Euphoria, ataxia, limb tremors, dizziness, headache, difficulty in concentrating, poor memory and mental confusion were also experienced. Some of the symptoms cleared up after a few weeks but others persisted for years. Dr Crank said that it is generally accepted that MDI has similar toxic properties to TDI. He concluded that: the reported studies show that it is possible that one single exposure to isocyanates can lead to long-term mental damage, with adverse symptoms possibly continuing for years after the exposure. 71 The 1967 firemen study is documented in two papers published in 1976 in the British Journal of Industrial Medicine. The papers were tendered and differing evidence was led from Drs Crank (for the appellant) and Drew (for the respondent) as to the conclusions properly to be drawn from them. 72 Dr Crank pointed to material in the papers suggesting that there could be a delay in the onset of neurological symptoms, and that neurological symptoms appeared where they had not been preceded by skin irritation or respiratory problems. He made it plain that his opinion as to an association between exposure to the common isocyanates and neurological consequences was based solely on this report and the absence of any discrediting of it in later scientific literature. Dr Crank also disclaimed the expertise to express a medical opinion as to the consequences of exposure to isocyanates, whether through sensitisation or other means. 73 Dr Crank said that MDI has a distinctive pungent unpleasant smell. The smell increases as the amount of exposure increases. You can only smell MDI at a level which is considerably above the short-term exposure level according to WorkSafe Australia's literature. (The evidence and scientific dispute referable to the smell of SSI when it was applied by the appellant is addressed separately below.) 74 Under cross-examination, Dr Crank agreed that he had placed the most significant reliance on the two firemen papers. Those were the only papers derived from a literature research that provided any reference to neurological effects in relation to exposure to isocyanates. He also agreed that in each of the individual medical cases reported in the firemen study there was a complaint of respiratory distress occurring close enough to the fire to show an association in epidemiological terms. He was also driven to agree that the level of exposure to TDI recorded in the reports of the firemen study was that of a highly unusual exposure to a massive dose of TDI as well as the possibility that the toxin released by the fire was something other than TDI itself (see below). Dr Crank was not aware of a single case of anyone having suffered some form of neurological impairment without, at the same time, having complained of respiratory problems. The common response to isocyanates is respiratory. 75 The respondent's toxicologist was Dr Drew, currently a Diplomat of the American Board of Toxicology. He has extensive university and consultant experience. He has been on many key advisory boards and committees concerned with hazardous substances. 76 Dr Drew considered that there was no evidence that MDI was capable of causing neurological symptoms. He pointed to distinctions between TDI and MDI. He placed some reliance upon the conclusions relating to MDI drawn by the international group of experts who published Concise International Chemical Assessment Document 27: Diphenylmethane Diisocyanate (MDI), World Health Organisation, Geneva, 2001 (hereafter "CICAD"). That study had also been highly critical of findings by Reidy and Bolter in a Case Report "Neuropsychological toxicology of methylene diphenyl diisocyanate: a report of five cases" (1994) Brain Injury 285: see CICAD. The appellant had placed some reliance upon this Report. Cripps AJ addressed this issue at J39-40 in terms with which I would agree. 77 Dr Drew gave his evidence in terms of scientific proof. This is not said as a criticism. 78 He was extremely critical of Dr Crank's reliance upon the firemen study, stating that "its interpretation by Dr Crank in the context of Mr Forbes is quite misleading and comes close to junk science". Dr Drew explained in detail the distinctions that he saw between the causes and symptoms reported in the firemen study compared to those affecting the appellant. He adhered to this view in cross-examination, conceding only that the firemen study supported the hypothesis that TDI is capable of causing neurological symptoms in extreme contexts. Dr Drew pointed to other experiments conducted on TDI and the fact that in none of those experiments had neurological effects been observed. 79 There was evidence in the firemen study that chemicals other than TDI had been ignited. In Dr Drew's view, it was these other chemicals that "most probably" played a part in relation to the neurological conditions experienced by the firemen. Even though Dr Drew acknowledged that the firemen study provided some support for the view that neurological problems may occur as a result of exposure to TDI, he refused to extend this view to MDI. He recognised that you could have neurological effect without having respiratory effects, but he did not think that that was true for isocyanates. 80 Dr Spira is a consultant neurologist. He was retained by the respondent's solicitor, but he took a full history from the appellant. The only suggested error in the recorded history is one made in the appellant's favour, in that he assumed a 60% isocyanate level in SSI, whereas the level according to the label (the accuracy of which was not disputed) was 10%. 81 Dr Spira's initial conclusion was that viral encephalitis was a highly unlikely cause of the symptoms presented by the appellant. Dr Spira gave a second report after he had been given the opportunity of reading the reports of other experts, including Dr Crank. He adhered to his diagnosis of encephalopathy, but was at pains to state why he thought that its nature or cause remained a mystery. In his second report (dated 22 April 1999) Dr Spira concluded that a reaction to exposure of SSI was the most likely cause of the appellant's encephalopathy. However, like Dr Darveniza before him, he proceeded from the MSDS information, addressing the hypothesis (then accepted by him) that the appellant's encephalopathy was due to the toxic effects of SSI. He reported: Determining whether this is the case is of course difficult and depends critically on nature and degree of exposure, aspects which are all but impossible to determine at this stage. The extent to which he was sensitised to isocyanates by his previous exposure and whether he has a personal sensitivity to this chemical is of course unknown and is not testable. Nonetheless, contrary to Prof Bisby's comments to the contrary, it is apparent that Selleys Space Invader is capable of producing a toxic encephalopathy and this is reflected in the product information obtained from Selley's. I have a copy of the document which indicates the material safety data in relation to the agent. On p2 of that document, under the title Health Hazard Information, is a description of the effect of inhaling the vapour and in this it is said that "Inhalation of vapour or spray mist can result in headaches, dizziness, nausea and possible breathing difficulties due to respiratory sensitisation which may be delayed. Inhalation of high concentrations can produce central nervous depression, which can lead to loss of co-ordination, impaired judgement, and, if exposure is prolonged, unconsciousness." I believe that this is a good description of toxic encephalopathy. 82 Dr Spira gave a third report on 22 June 2001 and it was this opinion that he adhered to in his oral evidence. He said: Thank you for your letters of 1st and 20th June 2001 and for the document forwarded with the first letter containing information relating to the toxicity of isocyanates. My reading of those documents and subsequent search in medical databases on the topic have indicated that the isocyanates, the chief constituents in Selleys Space Invader (SSI), appear to be primarily respiratory toxins. This is particularly so in their vapour form, the exposure relevant in Mr Forbes' case. The reports describe asthma-like symptoms and it is said that the agents in high concentration may lead to severe bronchospasm, chemical pneumonitis and pulmonary oedema. In the world's worst example of thiocyanate intoxication in Bophal, India deaths resulted from cardiac arrest secondary to pulmonary oedema. In Mr Forbes' case there is no history of any respiratory involvement suggesting that exposure to the vapours of the SSI may have been minimal. This raises some doubt about its role in the events which saw his hospitalisation. In view of the lack of typical toxicity features in Mr Forbes, I searched for information relating to the neurological effect of thiocyanates to see if any reaction like that experienced by Mr Forbes has been described, I had previously read the leaflet supplied by Selleys title "Material Safety Data Sheet" which referred to the effects of exposure to inhalation of high concentrations in the terms "… central nervous system depression, which can lead to loss of coordination, impaired judgment, and, if exposure is prolonged, unconsciousness". Although this is not an apt description of Mr Forbes' exposure, on the basis of the fact that the constituents of SSI had central nervous system (CNS) effects at all raised the possibility that the encephalopathy demonstrated by Mr Forbes may have related to his exposure to SSI. Unfortunately the data sheet does not clarify whether the central effects described are independent of respiratory toxicity as all of the CNS features described are expected secondary effects of respiratory failure. Clearly Mr Forbes had no respiratory difficulties and in order for SSI to have been responsible for his encephalopathy and seizure it must have been via a direct CNS effect. The clinical features of Mr Forbes' syndrome included: headaches, joint pains, nausea, fever, confusion, tonic-clonic convulsion and coma. My search of the documents you forwarded and of the medical databases (Medline) revealed no similar case. Nonetheless in an article by Le Quesne et al, of 23 men who complained of neurological sequelae after single severe exposure to toluene di-isocyanate, five are said to have experienced euphoria, ataxia, and loss of consciousness. These men and nine others complained of headache, difficulty in concentration, poor memory, and confusion during the next three weeks. There is no description I have encountered in the literature of fever, arthralgia, seizure or of the slow development of confusion over two days. Overall I am left in a great deal of doubt about the role of SSI in Mr Forbes' presentation and the slow development of symptoms, the fever, joint pains and seizure experienced by him are strongly suggestive of a vial infection with an encephalitic component. This and the lack of a similar case to his own in the extensive literature search I conducted describing a large number of exposures to isocyanates makes SSI an unlikely cause. 83 Dr Spira's change of opinion between his second and third reports was further explained in his oral evidence. He summarised his third report in the following terms: I mean in essence, I could find no evidence of similar case in the literature to suggest that this is an expected outcome or even a rare outcome of Space Invader, and therefore I deferred to another possible explanation. 84 Dr Spira said that he had considered the "publications" and not just relied upon the views of others about them. 85 When asked what had happened to change his mind, he said: A. Well, first of all I had reviewed the whole picture and there were certain parts of it that don't follow the Selleys Space Invader model. One of them is the fact that he had a sort of pain syndrome, a febrile illness and arthralgia, joint pains. These are quite typical symptoms of a viral illness, and I can think of no way in which thiocyanates can be implicated in producing that, but I think the most important piece of information that I did not know and that I do now is that there was a preceding illness, that he had been unwell in some way prior to using the Selleys Space Invader. There had been an upper respiratory tract infection, which is pretty much a classical preamble to viral encephalitis. It usually follows a trivial upper respiratory tract infection which occurs over the preceding 14 days, and I felt that this was an important part of the history that I didn't have when I first saw this gentleman, and that also leaned me towards sort of a typical, nonetheless, encephalitis, rather than – I guess I was balancing the data and I came down to the because of the associated symptoms of arthralgia and fever. 86 The appellant is critical of this evidence. In his written submissions he submits that this opinion failed to grapple with the weight of the primary diagnosis of Dr Darveniza. 87 At the end of the day, Dr Spira's view seems to have carried comparatively little weight with the trial judge, beyond his evidence that there was nothing in the literature to support the probability that exposure to isocyanates caused neurological problems in the absence of respiratory problems (J41). The appellant failed because the treating neurologist (Dr Darveniza) essentially deferred to the evidence of the toxicologists as regards the relevant toxicity of the TDI. It may be that comparatively little attention was paid to Dr Spira's evidence because he too professed no toxicological expertise. In any event, it was for the appellant's counsel to cross-examine Dr Spira to lay any evidentiary ground work for challenging his conclusions. I do not think that those conclusions are undermined because they depended upon assumptions whose existence and weight was not put to Dr Spira by the party seeking to challenge in this Court whatever weight was and/or should be placed upon Dr Spira's views. I have already indicated that nothing in Dr Spira's third report and oral evidence provided any real support for the appellant's case. 88 Associate Professor Rawlinson is a medical virologist who directs the Division of Virology in the Department of Microbiology and the Department of Infectious Diseases at Prince of Wales Hospital. He specialises in the diagnosis and treatment of infections. Professor Rawlinson considered that it was more than likely that there was an alternative explanation for the appellant's injury apart from exposure to the chemicals in the SSI product. Several possible infective agents were identified. 89 Professor Rawlinson tended to exclude a diagnosis of toxic encephalopathy stemming from SSI, but in so doing based himself upon Dr Drew's toxicological expertise. 90 Professor Rawlinson's conclusions were: The diagnosis of Mr Forbes' encephalitis was approached appropriately during his admission to St Vincent's Hospital 17.2.97. The key diagnoses sought were of treatable illness that required urgent therapy. The diagnosis of toxic encephalopathy appears to have been made after the exclusion of viral encephalilitides, and the diagnostic tests in clinical use then (and now) do not exclude all viral encephalilitides. In the absence of a toxic cause, the most likely diagnosis is of a viral encephalitis of an unknown origin. An opinion should be sought as to the relevance of the EEG findings, but the CSF findings that are not typical for viral encephalitis, are by no means inconsistent with viral encephalitis. Overall, Mr Forbes' history fits best with a diagnosis of viral encephalitis, although the limited tests performed did not diagnose viral encephalitis. 91 Professor Rawlinson was not cross-examined. 92 The respondent also relied upon reports from Professor Bisby an expert in occupational medicine and toxicology. He identified the likely possible causes of the appellant's condition as either viral encephalitis or a toxic encephalopathy due to some inhaled chemical exposure, drugs or poison or unusual food poisoning. The trial judge made no reference to Dr Bisby's evidence in his reasons. Neither party placed any significant reliance upon Professor Bisby's evidence in the appeal. The trial judge's methodology in addressing the causation issue 93 Senior counsel for the appellant in this Court (Mr Watson SC) commenced his oral submissions with an issue that had not been raised in the written submissions signed by a different senior counsel (Mr A J Meagher SC); and which had not been flagged in the grounds of appeal. 94 The submission was that the respondent bore an evidentiary onus to rebut a finding that its product was the probable cause of the illness. It was submitted that this principle applied because the respondent was a manufacturer that knew or ought to have known about its product and because of the findings that MDI could possibly have caused a toxic encephalopathy. 95 Senior counsel relied upon Spigelman CJ's discussion in Seltsam in relation to the use of epidemiological evidence suggesting an increase in risk as a circumstantial fact. Reliance was also placed on passages in Chappell v Hart (1998) 195 CLR 232 at 238-9, 247-8, 257, especially the statement by McHugh J (at 247-8) that: However, once the plaintiff proves that the defendant breached a duty to warn of a risk and that the risk eventuated and caused harm to the plaintiff, the plaintiff has made out a prima facie case of causal connection. An evidentiary onus then rests on the defendant to point to other evidence suggesting that no causal connection exists. Examples of such evidence are: evidence which indicates that the plaintiff would not have acted on the warning because of lack of choice or personal inclination; evidence that no alternative course of action would have eliminated or reduced the risk of injury. Once the defendant points to such evidence, the onus lies on the plaintiff to prove that in all the circumstances a causal connection existed between the failure to warn and the injury suffered by the plaintiff. 96 In TC by his Tutor Sabatino v The State of New South Wales & Ors [2001] NSWCA 380 I discussed the issue of onus and causation at length (at [53]-[101]). I cited lengthy passages from Chappell, Naxakis v Western General Hospital (1999) 197 CLR 269 and Seltsam. I concluded (at [59]) that Australian law has not adopted a formal reversal of onus of proof of causation in negligence, although a robust and pragmatic approach to proof permits, but does not compel, a favourable finding in particular circumstances. I also indicated (at [70]) my respectful agreement with what Spigelman CJ wrote in Seltsam. 97 A plaintiff may however be assisted by a shifting in the evidentiary burden of proof in relation to causation. This concept originated in a dictum of Dixon J in Betts v Whittingslowe (1945) 71 CLR 637 at 649 that has been cited with approval in Bennett v Minister of Community Welfare (1992) 176 CLR 408 at 420-1 (Gaudron J), in Chappell at 238-9 (Gaudron J) and 273-4 (Kirby J). See also Naxakis at 278-9 (Gaudron J) and 296 (Kirby J). 98 In the seminal passage in Betts, Dixon J said at 649 that: … The breach of duty coupled with an accident of the kind that might thereby be caused is enough to justify an inference, in the absence of any sufficient reason to the contrary, that in fact the accident did occur owing to the act or omission amounting to the breach of statutory duty. 99 Betts was itself a case of breach of an imperative statutory duty to secure fence dangerous machinery. The plaintiff was injured when unfenced dangerous machinery injured him, but in circumstances where there was no evidence accepted by the trial judge as to how the accident happened. The particular accident was obviously one that fell within the scope of Dixon J's principle, quoted above. In TC, I said (at [96]): Betts and the cases which follow it suggest an approach to reasoning that enables a robust trier of fact to proceed from breach and specific harm to causation. The decision-maker may infer causation if the breach was such that in the ordinary course of events (as perceived by the decision-maker) that type of harm is a consequence of the breach. Kirby J expressed the same notion in Chappel and Naxakis … when he cited Betts as a case where breach was closely followed by damage. In such cases, unless the defendant can point to some particular reason why the instant case is outside the norm, causation may (but not must) be inferred. Simply because a risk is "not far-fetched or fanciful" (cf Wyong Shire Council v Shirt (1980) 146 CLR 40 at 48) does not mean that the trier of fact is required to infer that it came home or that the defendant's negligent conduct caused it to happen or materially contributed to its happening. 100 In my view, the primary judge did not err in his approach to the causation issue. There is nothing to which our attention was drawn to indicate that he was asked to apply reasoning stemming from Dixon J's dictum. In any event, the detailed scientific enquiry embarked upon at the trial showed that every piece of light that could be shone upon the causation issue was directed at it. The possibility that the relevant toxin was the subject of the warning on the label was excluded by the evidence, including the evidence about the MSDS that led to it being common ground that the only possibly relevant toxin in SSI was the MDI. 101 In his favour, the appellant could point to the short period that elapsed between his use of the product and the early symptoms of his sudden illness. There was also the scientific evidence, accepted by Cripps AJ, that the MDI component could possibly have been the trigger for toxic encephalopathy, which in turn was the most likely diagnosis of the appellant's illness. 102 But there was no suggestion that the respondent held back any information relevant to the causation issue. Unlike many negligence cases, that issue did not turn upon hypothesising what the plaintiff might have done if the defendant had done something expected of it. In the present case, the causation issue turned upon the state of scientific knowledge about something that had never happened before in similar conditions, according to the evidence and the literature. At the end of the day it was agreed that the only possible exception in the literature related to the firemen study discussed above. Issue was well and truly joined between the toxicologists who alone professed expertise to extrapolate conclusions from that study, which addressed circumstances markedly different from the present case. 103 Cripps AJ correctly ruled that the plaintiff bore the ultimate onus of proof on causation. The appellant/plaintiff did not fail because of some flaw in the legal methodology adopted by his Honour to address the scientific causation issue. Did the trial judge err in his conclusion on causation? 104 Eight of the eleven grounds of appeal contend that the primary judge erred in failing to hold that the MDI was the probable cause of encephalopathy. The appellant builds his case upon the trial judge's findings of primary fact, submitting that this Court is in as good a position as the trial judge to address the ultimate issue. The appeal is to be addressed according to the principles enunciated in Warren v Coombes (1970) 142 CLR 531 at 551. 105 I have already addressed the issue of the trial judge's methodology as to causation. The evidence of the experts has also been set out above. 106 Once the trial focussed upon the relevant toxicity of MDI, the essential conflict was between Drs Crank and Drew and the conclusions they derived from the scientific literature. The judge was also able to consider that literature in light of the expert evidence. It was relevant both as to its general silence as to any neurological impact of MDI (see esp CICAD) and as to the guarded, qualified and criticised conclusions of the firemen study and the Reidy and Bolter Case Report. It was well open to the judge to prefer the conclusions of Dr Drew. In light of them, the finding that the case on causation failed on the probabilities was well-nigh inevitable. 107 The appellant was critical of the trial judge for not taking a "robust and pragmatic" attitude to the scientific evidence. But this criticism is misplaced. His Honour was fully aware of his capacity to infer probable cause from the epidemiological possibility evidence. But he was not bound to do so. Robust and pragmatic decisions are not always made in the plaintiff's favour. The evidence relating to the odour of SSI 108 At the hearing of the appeal senior counsel for the appellant raised a further issue that had not been flagged in the grounds of appeal or the written submissions. 109 The appellant sought in effect to argue that the SSI he had used contained a dangerous quantity of MDI because of the smell emitted during application. 110 It was not suggested that the particular can used by the appellant contained a quantity of MDI in excess of the "low <10%" referred to in MSDS. The thrust of the submission was that the quality of the MDI could nevertheless have been closer to the dangerous "exposure standard" identified in a WorkSafe Australia publication. 111 In his evidence in chief, the appellant said that he had been able to smell the SSI during application. He first described the smell as "strong". He next referred to it as being "a very strong smell coming out of the fireplace, coming out of the hole in the fireplace there". In cross-examination he said "initially I could smell the product…. It's like I got used to it. I didn't smell it anymore but it initially smelled quite strong". 112 Dr Crank interviewed the appellant before completing his report. He does not record any history of the appellant having smelt the SSI. He simply noted that, according to the appellant, "the chimney was a source of drafts and that it was possible that the fumes from the can may have blown back from the chimney towards him". 113 According to the history given to Dr Spira, the appellant "curiously was unable to [smell] SSI on the second occasion [15 February 1997] despite the fact that his wife in another part of the house complained of the powerful odour". The appellant told Dr Bisby "I didn't smell anything, but my wife did – she complained at the time – actually I did smell it at first but very briefly". 114 Histories recorded by Dr Shores and Dr Roberts are also silent or negative as to the appellant having smelt anything, but record statements that his wife commented on a ("lousy") smell. 115 Mrs Forbes said that "there was a smell which is why I took our daughter out of the room, yeah, that was it". 116 Cripps AJ adverted to the issue of smell as it had been presented at trial in the following terms (J38): The plaintiff has asked me to accept estimates made by Dr Crank concerning the dose to which the plaintiff may have been exposed in the course of applying SSI. It is to be recalled that the firemen in the "firemen study" were subjected to massive doses of TDI. Dr Crank made a number of assumptions, which I think are questionable at best. He posed a theoretical model about the amount of MDI that could be released in an enclosed space at a particular equilibrium point. He had regard to the fact that there was, probably, ten grams of MDI in the two cans and that the plaintiff may have been exposed to it. I have already referred to the fact that the plaintiff gave evidence that the area was well ventilated, a breeze was blowing and indeed he had turned on fans. The product was not used in an enclosed space. It would seem to me that it is unrealistic for me to assume that all the MDI would have been in the atmosphere at any one time bearing in mind that large quantities of it were hardening in the material that had been applied. Moreover, I gain little assistance by the inference said, by Dr Crank, to arise from the odour referred to by Mrs Forbes. It is argued that if exposure had risen to a level where it could be detected by smell, then a dangerous amount of MDI had escaped. So much may be accepted, but the danger is the risk of sensitisation (not claimed) or respiratory problems. It does not assist in determining whether it causes brain damage. Moreover, the smell test is unhelpful bearing in mind that the evidence appears to be that an odour threshold (if it is met and even if people are capable of recording the odour of MDI) generally cannot be reached unless the substance is heated (and this substance was not). 117 These findings have survived the appellant's belated attack on them in the appeal. 118 There was evidence at trial that isocyanates are not normally detected by odour until the concentration reaches 10 – 20 times the exposure standard stipulated by WorkSafe Australia in a particular publication. 119 Dr Crank sought to quantify the extent of the appellant's exposure to MDI by assuming that the appellant had reported being able to smell the product, and relating the odour threshold adopted by the National Occupational Health and Safety Commission (NOHSC) to the exposure limits to isocyanates. The odour threshold to NOHSC is 0.4 parts per million. According to Dr Crank, this exposure threshold is 220 times the Threshold Limit Value (ie the level of exposure that the typical worker can experience without an unreasonable risk of disease or injury (and 63 times the Short Term Exposure Limit)). 120 The inference that was sought from this evidence was that because the appellant could smell the MDI, he must have been exposed to hazardous quantities of that substance. 121 As the respondent points out in its supplementary submissions, one flaw in this reasoning was exposed during Dr Crank's cross-examination. Assuming it was MDI that was smelt (see further below), the odour threshold at 0.4 ppm is greater than the vapour saturation point of MDI which is 0.13 ppm. Dr Crank therefore conceded that the only way in which the concentration of MDI could have exceeded the odour threshold was if it were heated or if it were present as a particulate. Yet there is no evidence of either situation in this case. 122 It follows that, whatever it was that the appellant and/or his wife could smell, it was physically impossible for it to have been MDI. 123 As indicated, the case was run on the basis that SSI had been applied in accordance with the manufacturer's specifications. These did not involve heating of the product. 124 There was another difficulty for the appellant as regards this alternative hypothesis. The appellant's evidence about odour was necessarily subjective. But, taken at its highest, it did not approach the descriptions of MDI's smell. Dr Crank described MDI as having a "pungent, unpleasant smell", or a "pungent irritant odour". Did the judge fail to find material facts or to state his reasons? 125 During the hearing of the appeal the appellant sought and obtained leave to amend his grounds of appeal by adding the following: 10. The trial judge erred in law in failing to give reasons: (a) as to whether or not the appellant's condition was caused by an infective encephalitis; (b) to permit the identification of any cause alternative to toxic encephalopathy caused by inhaling Selleys Space Invader 11. In the event the trial judge found that MDI was incapable of causing toxic encephalopathy, the trial judge erred in failing: (a) to make a specific finding to that effect; and (b) to give sufficient reasons for the finding. 126 In my view, these grounds were not established. 127 None of the doctors or scientists was able to identify a finite list of infective causes of encephalitis. But they were able to establish a range of "possibles", and issue was joined as to the likelihood of them having led to the appellant's symptoms. The possibility of viral encephalitis (of unidentified origin) was accepted by the witnesses and the judge (J43). The duty to find facts and give reasons did not, in my view, rise so high as to require the judge to have gone beyond the level of medical and scientific certainty. 128 As to ground 11, Cripps AJ did not find that MDI was incapable of causing toxic encephalopathy. On my reading of J28, 39 and 43 his Honour held that it was indeed a possibility, if only because epidemiological evidence did not exclude it. That was in accordance with the scientific evidence to which reference has already been made. Such evidence was fairly recounted in the primary judge's reasons. Conclusion 129 Since the appellant has failed to persuade me that the trial judge erred as regards his conclusion on causation, it is unnecessary to consider the remaining issues of foreseeability, negligence and the application of ss75AD and 75AK of the Trade Practices Act. 130 I confess to having found this a troubling case. The lay instinct is to draw a conclusion of cause and effect from the raw data about the onset of symptoms shortly after the use of the product. The appellant urges this Court to pay due regard to this matter. But it is not suggested that the judge failed to do so. Nor is it suggested that this is a case where the tribunal of fact may disregard the scientific evidence. 131 At the end of the day, Dr Darveniza was not in a position to take the critical issue any further than that of an informed lay person. Naturally, he knew better than anyone what were the symptoms presented by the appellant and what were the range of possible and/or probable causes for his disease. But Dr Darveniza did not profess expertise as a toxicologist. And the opinion about the impact of SSI drawn by him at one stage in the appellant's treatment was triggered by an inference he drew from the MSDS. By the conclusion of the trial no one suggested that that inference was properly drawn. 132 In this context, the issue became a dispute as to probabilities based upon strongly divergent scientific evidence. The judge did not fall into the error of requiring the appellant to prove his case at the level of scientific proof, nor did he approach the fact-finding task in any way indicative of legal error. 133 The parties did not invite the Court to read the full transcript of the scientific evidence. I have nevertheless done so, albeit rather cursorily, paying particular attention to the passages to which reference was made in submissions. 134 In my view, the primary judge was clearly entitled to prefer the evidence of Dr Drew over that of Dr Crank. At the end of the day, the latter placed the great weight of his argument upon conclusions drawn from the firemen study. Dr Crank's cross-examination and Dr Drew's evidence illustrated the difficulties of doing this at several levels. 135 From time to time Kirby J has raised suggestions that the principles governing appellate review of technical disputes may differ somewhat from those governing "ordinary" factual disputes (see Ahmedi v Ahmedi (1991) 23 NSWLR 288 at 291-2, Jackamarra v Krakouer (1998) 195 CLR 516 at 543). The present view of this Court is generally to the contrary (see Ahmedi). 136 One principle is however, constant. That is the notion that the trial judge has an advantage stemming from the very fact that he or she heard the totality of the evidence in the context of the trial (see generally Fox v Percy [2003] HCA 22 at [237]). Sometimes that advantage is squandered. Sometimes a small portion of strong evidence is overlooked at first instance. But there is nothing in the present appeal that takes the case outside the general principle. I am unpersuaded that Cripps AJ erred in law or fact. Accordingly the appeal should be dismissed with costs. 137 GILES JA: I agree with Mason P. 138 McCOLL JA: I agree with Mason P. ********** Last Modified: 05/13/2004 DISCLAIMER - Every effort has been made to comply with suppression orders or statutory provisions prohibiting publication that may apply to this judgment or decision. The onus remains on any person using material in the judgment or decision to ensure that the intended use of that material does not breach any such order or provision. Further enquiries may be directed to the Registry of the Court or Tribunal in which it was generated.