Bowers and Repatriation Commission
[2000] AATA 845
•22 September 2000
DECISION AND REASONS FOR DECISION [2000] AATA 845
ADMINISTRATIVE APPEALS TRIBUNAL )
) No V1997/84
VETERANS' APPEALS DIVISION )
Re NORMA JEAN BOWERS
Applicant
And REPATRIATION COMMISSION
Respondent
DECISION
Tribunal Mr B. H. Pascoe, Senior Member Mr A. Argent, Member Associate Professor J. Maynard, Member
Date22 September 2000
PlaceMelbourne
Decision The Tribunal affirms the decision under review.
.....…(Sgd) B. H. Pascoe............
Senior Member
CATCHWORDS
VETERANS' AFFAIRS – whether veteran's death from non-Hodgkin's lymphoma war-caused – whether reasonable hypothesis – whether smoking cause of non-Hodgkin's lymphoma – whether hypothesis supported by evidence
Veterans' Entitlements Act 1986
Repatriation Commission v Webb (1998) 51 ALD 575
Bushell v Repatriation Commission (1992) 175 CLR 408
Byrnes v Repatriation Commission (1993) 177 CLR 564
REASONS FOR DECISION
22 September 2000 Mr B. H. Pascoe, Senior Member Mr A. Argent, Member Associate Professor J. Maynard, Member
This is an application to review a decision of the Veterans' Review Board ("VRB") of 11 November 1996 which affirmed a decision of the respondent dated 24 May 1993 that the death of the applicant's late husband, Mr George Bowers, was not war-caused.
At the hearing the applicant, Mrs Norma Bowers, was represented by Mr A. Larkin of counsel and the respondent by Ms J. McCulloch, an advocate of the respondent. Evidence was given by Professor J. Zalcberg, a haematologist and medical oncologist; Professor H. Peach, an epidemiologist; and Professor R. Fox, a haematologist and medical oncologist.
The late Mr Bowers died on 14 June 1981, aged 57, and the cause of death was certified to be poorly differentiated lymphoma, duration one year, and central nervous system infiltration with lymphoma, duration two months. Mr Bowers' original claim was refused by a Repatriation Board in 1982 and by the Repatriation Commission in 1984. A new claim was lodged in April 1993 and it is the decision on this claim which is before the Tribunal. Mrs Bowers did not appear at the hearing of the VRB, had not stated the basis of her claim and had not provided any medical evidence in support of her case. The VRB had been advised that an attempt had been made to obtain a medical report from a Dr D. Bernshaw, oncologist, without success. It was the understanding of the VRB that Mrs Bowers contended that there was a link between the late veteran's death and exposure to radiation in Japan in 1946/1947.
The veteran served in the Australian Army from 6 January 1942 to 14 February 1947. His service constituted eligible war served and operational service under the Veterans' Entitlements Act 1986 ("the Act"). Before this Tribunal, the applicant had abandoned the contention of a causal link between the death and exposure to radiation. The causal link argued was that the veteran's smoking had caused the poorly differentiated lymphoma, generally known as non-Hodgkin's lymphoma ("NHL"). The respondent conceded that the veteran's smoking habit was related to war service so that the dispute related to whether a reasonable hypothesis connecting such smoking with his death had been raised.
Under section 120(1) of the Act the Tribunal, standing in the shoes of the respondent, shall determine that the death of a veteran was war-caused unless it is satisfied, beyond reasonable doubt, that there is no sufficient ground for making that determination. Under subsection (3) of section 120, the Tribunal shall be satisfied, beyond reasonable doubt, that there is no sufficient ground for determining that the death was war-caused if, after consideration of the whole of the material before it, it is of the opinion that such material does not raise a reasonable hypothesis connecting the death with the circumstances of the particular service rendered by the veteran. As the claim was made prior to 1 June 1994, section 120A and any relevant Statement of Principles has no application. Consequently, the question in issue here is whether the material before the Tribunal raises a reasonable hypothesis connecting the death from poorly differentiated lymphoma with smoking.
Professor Zalcberg is the Director of haematology and medical oncology at Peter McCallum Cancer Institute. He provided a written report, dated 28 July 1997, which stated:
"According to the statement made by Mrs Bowers (dated 18/3/1997), her husband started smoking during his enlistment and continued to smoke until his terminal illness, a period of 34-39 years (the exact date of his commencement of smoking during his enlistment was not specified). I have previously reviewed the evidence which supports the aetiological association between tobacco consumption and non-Hodgkins lymphoma and continue to stand by this report as there has been no recent literature to deny the probable association between long-term tobacco use and the subsequent development of non-Hodgkins lymphoma.
Although tobacco consumption was not thought to be important in the aetiology of haematopoietic malignancies prior to 1986, such a relationship has now become more widely accepted. (A functional classification of the types of tumours that make up the haemopoietic malignancies is appended). Increasingly, a relationship between tobacco consumption and the development of various subtypes of the haemopoietic malignancies has been demonstrated. A prospective study in Seventh Day Adventists revealed a doubling in the risk of acute leukemia and three times the risk of myeloma. In each case, these risks were related to the dose of tobacco consumed. Similarly, a prospective study in US Veterans as well as studies performed by the American Cancer Society have revealed an increase in the risk of leukemia, proportional to the number of cigarettes smoked.
With respect to non-Hodgkin's lymphoma, a 20% increase in the incidence of this group of disease was seen amongst 3.7 million male US Veterans when compared to risks in the general population obtained from the SEER (Surveillance, Epidemiology and End-Results) data-base. Finally, a population based case-control study performed in Iowa (USA) revealed a 40% increase in the risk of non-Hodgkin's lymphoma in smokers. These risks were greatest in those who had smoked for the longest duration.
The hypothesis that cigarette smoking is important in the aetiology of haematopoietic malignancies is corroborated by the fact that this concept is biologically plausible. Tobacco smoke contains benzene and plutonium amongst hundreds of other chemicals which individually are known to be carcinogenic.
In view of the fact that Mr Bowers started smoking as a result of his war-service and the fact that the risk of developing non-Hodgkin's lymphoma is increased in smokers, I believe that the development of his disease and the subsequent death of the patient from complications of this process should be considered as war-caused as defined under The Act."
In his oral evidence, Professor Zalcberg accepted that the increase in non-Hodgkin's lymphoma among US veterans was not linked with tobacco. He agreed also, that there was no reference to lymphoma in the Seventh Day Adventist study and that myeloma and acute myeloid leukemia have different aetiological causes compared to NHL. Whilst Professor Zalcberg recognised that most studies referred to either haemopoietic cancers generally or emphasised myeloma rather than lymphoma, he believed that they supported a possible connection between smoking and lymphoma. He believed that there was no proof either way and the connection was a "plausible hypothesis" given that carcinogens in tobacco were known to cause some blood cancers.
Professor Peach is Head of the Department of Public Health and Community Medicine at the University of Melbourne and had provided two reports dated 20 April 1998 and 18 May 2000. In his first report he reviewed some 22 studies between 1976 and 1994. He commented on their reliability and whether they had excluded chance, bias and confounding as explanations for any apparent association between smoking and lymphoma. In that report he concluded that: "The material available to me as a whole has failed to raise the hypothesis that smoking 36 pack years is causally associated with non-Hodgkin's lymphoma". On 18 May 2000, Professor Peach provided an updated report on the epidemiological research into such an association. He analysed some 20 studies and concluded:
"1. Of the five prospective studies, only one found an association between ever having smoked and dying from non-Hodgkin's lymphoma and another found an association between being a former smoker and follicular lymphoma. The former had a small number of deaths, was possibly biased in favour of finding an association between smoking and non-Hodgkin's lymphoma, lost 23% of the cohort to follow-up and did not take account of any confounding factors other than age. The latter performed multiple comparisons thereby increasing the chance of finding a positive association between non-Hodgkin's lymphoma and smoking, was possibly biased in favour of finding an association between non-Hodgkin's lymphoma and smoking, did not take account of confounding factors and failed to find a trend between increasing risk of follicular lymphoma and increasing duration and intensity of smoking to support causation. Although all but one of the larger prospective studies may have been biased against finding an association between current smoking and non-Hodgkin's lymphoma due to their recording smoking status only on entry into the cohort, they also found no association between non-Hodgkin's lymphoma and being an ex-smoker.
2. Of the eight population-based case-control studies only three found a significant association between smoking and non-Hodgkin's lymphoma and one between past smoking and non-Hodgkin's lymphoma. Of the former, one study did not exclude chance and confounding and the association disappeared when the effect of a bias in obtaining smoking data was taken into account. Another did not include men, involved multiple comparisons which produced only a weak association which was not tested for a dose-response relationship. The third found an association only among heavy smokers under 45 years of age. The latter of the four studies involved multiple comparisons and the authors themselves cast doubt on whether the association they found between past smoking and non-Hodgkin's lymphoma was a causal one.
3. Of the four hospital-based case-control studies, only two found an association between smoking and non-Hodgkin's lymphoma. One found an association between ever smoking and non-Hodgkin's lymphoma when cases were compared to cancer controls, whose smoking habits may not have been similar to those of the general population, but not when cases were compared with population controls. The study was conducted in Canada whose tobacco is not comparable to that of other countries. The other study was conducted in Uruguay and the multiple comparisons yielded associations only with smoking black tobacco or hand-rolled cigarettes. The odds of a smoker of black tobacco or hand-rolled cigarettes had wide confidence limits and were not adjusted for confounding factors. A dose-response relationship was not explored. The relevant of the effect of smoking black tobacco and hand-rolled cigarettes in Uruguay to the effects of smoking in other countries is unclear.
4. An analysis of the combined data from three population-based case-control studies found no association between smoking and non-Hodgkin's lymphoma among men in any age group.
5. The hypothesis that smoking is causally associated with non-Hodgkin's lymphoma has been tested in prospective, population-based case-control studies and hospital-based case-controlled studies, the weight of evidence is against the hypothesis. The strongest evidence in favour of an association between smoking and non-Hodgkin's lymphoma is indicative of an effect of heavy smoking among men under 50 years of age.
6. The above material does not point to smoking being a cause of the late Veteran's condition whether it is non-Hodgkin's lymphoma."
In his oral evidence, Professor Peach said that it was appropriate to recognise that smokers differ frequently in other ways such as education, socio-economic factors, diet, etc. from non-smokers and these other facts have to be considered in assessing the reliability of any study. Further, it would be expected that a study proposing a link between smoking and lymphoma would demonstrate a "dose response" or that the greater the consumption of tobacco products the greater the risk of lymphoma and no study had shown this. He said that laboratory studies had not produced any reliable evidence of such a link and had not been able to produce lymphoma in laboratory animals with smoke. Professor Peach said that this was in contrast to studies into lung cancer with showed strong consistent evidence between studies of the link between smoking and that disease. The studies relating to lymphoma had been few, inconsistent and failed to produce evidence of cause and effect.
Professor Fox, who is the Director, Department of Clinical Haematology and Medical Oncology at the Royal Melbourne Hospital, provided two reports dated 29 October 1997 and 10 May 2000. In his report of 29 October 1997 he considered the studies referred to by Professor Zalcberg and other studies between 1993 and 1997. He concluded that "the vast number of papers of epidemiological studies looking at non-Hodgkin's lymphoma and smoking have not shown any link". He believed that Professor Zalcberg had attempted to suggest that a link with acute myeloid leukaemia could be transferred to NHL without any specific direct data. He noted that NHL is much more common that acute myeloid leukaemia yet the vast majority of studies have found no link with smoking. In his report of 10 May 2000 Professor Fox reviewed four further studies published in 1997 and 1998 and concluded that, given that information "it does not appear possible to link smoking to the development of non-Hodgkin's lymphoma". In his oral evidence, Professor Fox said that the studies supported a possible risk of smoking and myeloma but not lymphoma. He maintained that the myeloproliferative and lymphoproliferative areas have different accepted aetiological factors and it was speculative to say there was a link between the two.
It was submitted for the applicant that the evidence of Professor Zalcberg had raised a reasonable hypothesis connecting the death of the veteran with his smoking, which had been accepted as being war-caused. It was said that this was the view of an eminent medical practitioner supported by some studies and it was irrelevant that other practitioners disagreed with that hypothesis. It was submitted that it was unnecessary under section 120 of the Act for the causation to be proved, provided the hypothesis was not fanciful or tenuous and properly held by a practitioner eminent in the relevant field of medicine.
For the respondent it was submitted that the hypothesis relied upon was no more than a possibility, not supported by studies or other practitioners and was too remote and tenuous to be accepted as reasonable for the purposes of section 120(3).
This application had been stayed pending a decision of the Federal Court in Repatriation Commission v Webb (1998) 58 ALD 575. In that case, the hypothesis raised was that the veteran's war service caused him to smoke, the smoking caused him to contract NHL and NHL was the cause of the veteran's death. In a decision of this Tribunal it was concluded that the evidence did not point to a diagnosis of NHL and that the hypothesis was not reasonable. On appeal, Finkelstein J set aside that decision and remitted the matter to be heard and determined again. The reason, which is relevant to this present case, was that the Tribunal had been in error in finding that the hypothesis connecting smoking with NHL "lacks support in the medical studies" and that the Tribunal had been in error in identifying its task as evaluating or weighing the evidence of a medical expert against the medical studies. On appeal to the Full Court of the Federal Court, Tamberlin, Finn and Marshall JJ, dismissed the appeal and remitted the matter to the Tribunal. A major part of the decision related to the Tribunal's non acceptance of that part of the hypothesis which related to the cause of death being NHL. In this present case there is no dispute as to the diagnosis of cause of death. In respect to the association between smoking and NHL, their Honours, in a joint judgement said (at pages 580 and 581):
"His Honour found that the tribunal erred in not having regard to the evidence of Professor Fox as to the existence of studies which supported an association between smoking and NHL. His Honour said that the issue was not the correctness of Professor Fox's view but whether the tribunal was wrong when it said that there were no medical studies to support the connection between smoking and NHL.
The evidence indicates that the relevant report, which is that of Williams and Horm was raised in oral evidence and that Dr Fox agreed it was impossible to imagine at a 'biological level' that data in the relevant report was meaningful. This conclusion was accepted by the tribunal.
However, Professor Fox referred to another report, relating to a religious group of non-smokers, which showed a significantly lesser incidence of NHL when compared with the general incidence of NHL in the community at large. Professor Fox, however, conceded that it was not possible to conclude anything from this study.
In the light of this evidence, we do not consider that the tribunal erred in stating that the hypothesis connecting smoking with NHL lacked support in the medical studies in the material before the tribunal. However, as the High Court indicated in Bushell's case lack of support in the profession is not decisive on the question of whether a reasonable hypothesis existed."
In respect to the evaluation of the evidence, their Honours said (at page 581):
"In our view, the tribunal reasons indicate that it did not simply confine its task to an evaluation or weighing of medical opinions as against each other in order to select the preferable expert or make any determination as to the correct version of competing views. The task was approached on the basis of its consideration of the material before it. Any examination of the material before the tribunal necessarily involves having regard to conflicting medical opinions and, as pointed out in Bushell's case, this is a proper exercise to be undertaken by the tribunal in the performance of its task. This must involve some degree of evaluating or weighing medical opinions in the light of all the material presented to the tribunal. We are not satisfied that any error of law has been shown as a consequence of referring to evaluating or weighing the evidence."
The two studies referred to which had been noted by Professor Fox in his earlier oral evidence before the Tribunal in that case were so mentioned by him as having been seen by him since writing his formal report. In this case, both studies were covered by Professor Fox in his report.
In discussing the operation of section 120(3) of the Act, their Honours in Webb's case (supra) referred to two decisions of the High Court and said (at pages 578 and 579):
"In considering the operation of s 120(3), it is necessary to bear in mind the principles set out by the High Court in Bushell v Repatriation Commission (1992) 175 CLR 408. In that case at 414-416, it is said in the joint judgment of Mason CJ, Deane and McHugh JJ that:
"The material will raise a reasonable hypothesis within the meaning of s 120(3) if the material points to some fact or facts (`the raised facts') which support the hypothesis and if the hypothesis can be regarded as reasonable if the raised facts are true. ... So, in determining whether a hypothesis is reasonable ... it is not decisive that a connexion has not been proved between the kind of injury which occurred and circumstances of the kind which constitute the relevant incidents of the veteran's service. Nor is it decisive that the medical or scientific opinion which supports the hypothesis has little support in the medical profession or among scientists.
...
However, a hypothesis cannot be reasonable if it is 'contrary to proved scientific facts or to the known phenomena of nature'. Nor can it be reasonable if it is 'obviously fanciful, impossible, incredible or not tenable or too remote or too tenuous'.
... the case must be rare where it can be said that a hypothesis, based on the raised facts, is unreasonable when it is put forward by a medical practitioner who is eminent in the relevant field of knowledge. Conflict with other medical opinions is not sufficient to reject a hypothesis as unreasonable. As we have earlier pointed out, it is not the function of s 120(3) to require the commission to choose between competing hypotheses or to determine whether one medical or scientific opinion is to be preferred to another. This does not mean, however, that ...the commission cannot have regard to the medical or scientific material which is opposed to the material which supports the veteran's claim. Indeed, the commission is bound to have regard to the opposing material for the purpose of examining the validity of the reasoning which supports the claim that there is a connexion between the incapacity or death and the service of a veteran. But it is vital that the commission keep in mind that that hypothesis may still be reasonable although it is unproved and opposed to the weight of informed opinion.
If the material does raise a reasonable hypothesis ... the commission must determine that the injury, disease or death was war caused 'unless it is satisfied, beyond reasonable doubt, that there is no sufficient ground for making that determination'.
...
The commission will be satisfied beyond reasonable doubt 'that there is no sufficient reason for making [the] determination' if it is satisfied beyond reasonable doubt that it cannot accept the raised facts or so many of them as are necessary to support the hypothesis." (emphasis added)The High Court again considered the operation of s 120(3) of the Act in Byrnes v Repatriation Commission (1993) 177 CLR 564 at 571. In a joint judgment, Mason CJ, Gaudron and McHugh JJ summarised the operation of the provision as follows:
"(1) First sub-s.(3) of s 120 is applied: do all or some of the facts raised by the material before the Commission give rise to a reasonable hypothesis connecting the veteran's injury with war service? The hypothesis will not be reasonable if it is contrary to known scientific facts or is obviously fanciful or untenable. If the hypothesis is not reasonable, the claim fails. Proof of facts is not in issue at this point. (2) If a reasonable hypothesis is established, sub-s.(1) of s 120 is applied. The claim will succeed unless: (a) one or more of the facts necessary to support the hypothesis are disproved beyond reasonable doubt; or (b) the truth of another fact in the material, which is inconsistent with the hypothesis, is proved beyond reasonable doubt, thus disproving, beyond reasonable doubt, the hypothesis.
…"
In this case we do have a medical practitioner who is eminent in the relevant field of knowledge, Professor Zalcberg, who has put forward a hypothesis linking smoking and NHL. However, we had two equally eminent medical practitioners who were of the opinion that the proposed hypothesis was not reasonable. Their respective evidence may be summarised as follows:
Professor Peach, an expert in epidemiology, made comments about the nature of the studies and looked at each in detail:
(a)of the 5 prospective studies only one found an association but this, in Professor Peach's opinion, could have been subject to bias;
(b)of the 8 population based studies only 3 found a significant association between smoking and NHL and on analysis these did not exclude chance and confounding and association disappeared when the effect of bias was taken into account.
(c)of the 4 hospital based case control studies only 2 found an association but are not relevant when compared to the type of tobacco in the countries (Canada and Uruguay). No dose response relationship was present.
Professor Peach was of the opinion that the weight of evidence does not point in published studies to smoking being a cause of NHL.
Professor Zalcberg cited articles which indicated a relationship exists between cigarette smoking and acute leukemia and with multiple myeloma (both myeloproliferative cancers);
(a)Professor Zalcberg aggregated the haemopoietic cancers together bringing in the lymphoproliferative part of the classification together with the myeloproliferative portion;
(b)There were no specific arguments presented by Professor Zalcberg for this linking between myeloproliferative and lymphoproliferative disease apart from his chart of haematopoietic cancer (unreferenced) and comments about the possibility of a single stem cell.
(c)In his oral evidence Professor Zalcberg accepted that his hypothesis was "plausible" with no proof either way.
Professor Fox stated that it was not appropriate to extrapolate between the 2 diseases and that the myeloproliferative and lymphoproliferative areas are separate in that:
(a)they have different accepted aetiological factors;
(b)no common aetiological factor was described;
(c)it was speculative to say there was a link between the two.
Professor Fox was of the opinion that the evidence presented does not provide a reasonable hypothesis to link cigarette smoking as a cause of NHL.
In our view the evidence as to the link between smoking and NHL has not been established to make the hypothesis a feasible one. There is a relationship established in the literature between myeloproliferative cancer and cigarette smoking but not a relationship between NHL and cigarette smoking. The hypothesis advanced is, in our view, no more than tenuous and too remote to be regarded as reasonable.
It follows that, pursuant to section 120(3), we are of the opinion, after consideration of all of the material before us, that the material does not raise a reasonable hypothesis connecting the death of the veteran from NHL with his smoking. As a result, we are satisfied beyond reasonable doubt, that there is no sufficient ground for determining that the death of the veteran was war-caused.
The decision under review is affirmed.
I certify that the seventeen (17) preceding paragraphs are a true copy of the reasons for the decision herein of
Mr B. H. Pascoe, Senior Member
Mr A. Argent, Member
Associate Professor J. Maynard, MemberSigned: .....................................................................................
Personal AssistantDate/s of Hearing 26 June 2000
Date of Decision 22 September 2000
Counsel for the Applicant Mr A. Larkin
Solicitor for the Applicant Williams Winter & Higgs
Solicitor for the Respondent Ms J. McCulloch, Departmental advocate
0
2
0