[New search]
[Printable RTF version]
[Help]
Case No: QBENI 1999/1156/A2
IN THE SUPREME COURT OF JUDICATURE
COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM MR JUSTICE THOMAS
Royal Courts of Justice
Strand, London, WC2A 2LL
Wednesday 19th April 2000
B e f o r e :
LORD JUSTICE KENNEDY
LORD JUSTICE BUXTON
and
LADY JUSTICE HALE
- - - - - - - - - - - - - - - - - - - - -
|
|
ELVICTA
WOOD ENGINEERING LTD and JAMES NEAL SERVICES LTD
|
Appellants
|
|
-
and -
|
|
|
HUXLEY
|
Respondent
|
- - - - - - - - - - - - - - - - - - - - -
(Transcript of the Handed Down Judgment of
Smith Bernal Reporting Limited, 180 Fleet Street
London EC4A 2HD
Tel No: 0171 421 4040, Fax No: 0171 831 8838
Official Shorthand Writers to the Court)
- - - - - - - - - - - - - - - - - - - - -
Michael Spencer QC & Colin Nixon (instructed by Wansborough Willey
Hargrave for the appellant)
Wyn Williams QC & Bernard Powell (instructed by Behr & Co.for
the Respondent)
- - - - - - - - - - - - - - - - - - - - -
Judgment
As Approved by the Court
Crown Copyright ©
LORD JUSTICE KENNEDY:
1. This is a defendant's appeal from a decision of Thomas J in relation
to the only live issue in the case, namely the issue of causation.
2. Agreed facts and Issue.
In 1954 the claimant, then 21 years of age, began to work for the first
defendant as a machine operative. In due course the business was taken over by
the second defendant. In 1980 the claimant was made redundant. He then worked
in business on his own account, and in May 1992 he was found to be suffering
from sino-nasal cancer. It was not adenocarcenoma (hereinafter AC) which has
been shown to be caused by exposure to hardwood dust, but squamous cell
carcinoma (SCC). During the many years of his employment with the defendants
the claimant had been heavily exposed to wood dust, mainly hardwood but some
softwood, and in this action such exposure has been admitted to have
constituted a breach of statutory duty and negligence. At trial the claimant
contended that, at least on the balance of probabilities, it was that wrongful
exposure which caused his SCC. That was the issue in the case which the judge
resolved in favour of the claimant, and so awarded the damages which, subject
to liability, had been agreed.
3. Nature of the case
Three expert witnesses were called, Professor Seaton for the claimant and Mr
Baer and Doctor Gallimore for the defendants. Professor Seaton after many
years as a consultant chest physician has been specialising in recent years in
the field of occupational health. Mr Baer has since 1993 been consultant ENT
surgeon at the Conquest Hospital, St Leonard's on Sea, East Sussex, and Dr
Gallimore has since 1992 been Honorary Consultant histopathologist at the Royal
Free Hospital in London. Dr Gallimore examined a biopsy of 18th May 1997, and
provided evidence as to the nature of the claimant's condition which has not
been disputed. Professor Seaton and Mr Baer gave expert evidence which was
intended to assist the judge to decide whether, on the balance of
probabilities, the claimant's SCC was shown to have been caused by his exposure
to wood dust. Professor Seaton did not himself examine the claimant, but he
and Mr Baer examined a number of studies reported in medical journals, and
brought to bear their skill in interpreting those publications. Professor
Seaton's opinion, as expressed in the final paragraph of his report of 5th
August 1998, was that -
"While the evidence associating squamous carcinoma of the nasal sinuses and
work with wood is less strong than that for adenocarcinoma, there is in my
opinion sufficient to conclude on the balance of probabilities that such an
exposure is causative in the case of an individual who (1) has had prolonged
and heavy exposure, especially if that exposure has been to hardwood dust, and
(2) has developed the tumour at a time consistent with the known natural
history of carcinogenesis."
Mr Baer's opinion was to the opposite effect, and he accepted in
cross-examination that his conclusion hung primarily on a study published by
Demers and others in 1995. It will be observed that save for Dr Gallimore's
examination of the biopsy the skill which each expert brought to bear was his
skill in sifting and evaluation of medical studies carried out by others, a
skill which over the years Professor Seaton working in the filed of
occupational health must have had to exercise even more than his two
colleagues. Furthermore his experience was longer, he having qualified in 1962
as against 1981 (Mr Baer) and 1983 (Dr Gallimore). I mention those matters
only because there has been criticism of the judge for preferring the evidence
of Professor Seaton to that of Mr Baer. To a large extent that criticism
focused on the way in which the experts, and in particular Professor Seaton and
Mr Baer, evaluated and interpreted the studies which constituted the common
data available to both sides, and that is something to which I will turn at
later stage in this judgment, but in an important paragraph towards the end of
his judgment the judge said -
"In reaching my conclusion I have taken into account my assessment of
Professor Seaton and Mr Baer. Although Mr Baer had researched the medical
literature thoroughly and, as an ENT consultant surgeon, had extensive of
experience of sino-nasal cancer, he was very much less experienced than
Professor Seaton, particularly in the study of industrial disease. Professor
Seaton was a most impressive witness of great experience and authority; he was
candid in the views he expressed and gave his evidence in a detached and
independent manner. He squarely faced up to the views raised by Mr Nixon in
his very skilful and searching cross-examination. Although a detailed analysis
of his reports were shown to point to some imprecision of language and he had
not mentioned all the literature, his oral evidence to me was very clear, fair
and objective; it had taken into account all the literature and other evidence
available by the time of the trial. I preferred his evidence to that of Mr
Baer who was far more heavily reliant on the published papers than his own
experience. I accept Professor Seaton's conclusions on the evidence."
Mr Michael Spencer QC, for the appellants, is critical of that paragraph. In
paragraph 6(a) of his skeleton argument he submits that the judge "did not have
sufficient regard for the relative experience and expertise of the parties'
expert witnesses." In paragraph 7 he points out rightly that both experts were
reliant on the literature, but then goes on to submit that Mr Baer "had great
experience of sino-nasal cancer and its origins". As I have already indicated
Mr Baer's clinical experience was not crucial in this case, and he made clear
his reliance on the Demers study. Before the judge Professor Seaton was
cross-examined by Mr Nixon on the basis that in order to mislead he only
referred to studies which favoured his hypothesis, and in somewhat more
restrained terms a similar line of cross-examination was adopted in relation to
Mr Baer. In this court Mr Spencer submitted that Professor Seaton's treatment
of the material available showed that he was deliberately misleading the court.
Having now had an opportunity to read the transcripts of evidence I see no
reason to conclude that either expert was falling short of his duty to the
court in any way, and I would emphatically reject the submission made by Mr
Spencer in this court, which was not put to Professor Seaton when he gave
evidence, and which does not even feature in the grounds of appeal. As to Dr
Gallimore the judge said -
"In his report and his evidence in chief, his evidence was that there was no
convincing evidence of an association between wood dust and squamous cell
carcinoma. He accepted that this part of his evidence was based on a reading
of the papers and it was not part of his expertise to consider the causal
connection between a disease and a carcinogen. I have therefore attached less
weight to his opinion on the issue of causation as it was outside the area of
his expertise; I have treated the principal experts on the issue of causation
as being Professor Seaton and Mr Baer."
That approach has not been seriously challenged before us nor, in my judgment,
could it have been. As to those who were "principal experts on the issue of
causation" the judge was right to consider not only what they said, but also
what he regarded as their relevant expertise, and the impression they created
when giving evidence. Of course if the experts misinterpreted the material
with which they had to work, and in particular if they misinterpreted it for
partisan reasons, then that would be bound to affect the judge's assessment of
them overall, but if they were each doing what they could then, as it seems to
me, it is quite impossible for this court to go behind the judge's evaluation
of them and of the impact of their evidence, and it should not attempt to do
so.
4. Medical Frontiers
When he began his submissions in this court Mr Spencer pointed out that
sino-nasal cancers of all types are relatively rare, and although it is
accepted that hardwood dust causes AC "conventional medical learning" does not,
he submitted, regard exposure to wood dust as causative of SCC. The claimant
is therefore, Mr Spencer submitted, "seeking to advance medical frontiers". I
doubt if the claimant has any such ambitions, and certainly our concern in this
court is very different. We merely have to decide whether, on the material
presented to him, as interpreted by the witnesses, the judge was entitled to
conclude as he did, namely that on the balance of probabilities the cause or a
significant cause of this claimant's SCC was his wrongful exposure to large
amounts of wood dust over many years. That does not mean that we are entitled
to ignore known limitations on the value of statistical material or anything of
that sort, but what it does mean is that we do not have to search for medical
certainty, and even if we do uphold the judge's decision we can contemplate
with equanimity the possibility that at some time in the future it may be
shown, on a balance of probabilities or perhaps to an even higher standard,
that the judge was wrong.
5. Studies
I turn now to the studies that suggested a link between inhalation of wood
dust and SCC for two reasons - first because they constituted the raw material
which was interpreted by the experts, and secondly, because, as the judge said,
his preference for Professor Seaton's evidence was reinforced by his own
consideration and analysis of the evidence.
The link between wood dust and AC was first suggested in 1965 as a result of a
study involving wood workers in High Wycombe, but it was not until 1972 that
there was any suggestion of a link between wood dust and SCC. Acheson and
others in a nation-wide study (Adenocarcinoma of the nasal cavity and sinuses
in England and Wales -Brit J Industr. Med. 1972 p.21) then said -
"In addition in providing abundant further evidence of the association between
nasal adenocarcinoma and work in the furniture industry the evidence suggests
that a smaller but nevertheless material risk of developing other histological
types of nasal cancer may exist for workers in the industry ........
It is worth noting .... that there is also a significant excess of nasal
cancers other than adenocarcinomas among woodworkers .... such an excess is not
surprising since a proportion of the anaplastic carcinomas maybe presumed to be
of glandular origin, and squamous metaplasia in the nose is a common phenomenon
.....
The strong suggestion from table 4 that there is also an increased risk of
histological types of tumour other than adenocarcinoma in furniture workers
will come as no surprise to pathologists. A reappraisal of the published
material from the original survey from Oxford showed that four nasal tumours
other than adenocarcinomas (two transitional cell, one squamous, and one
anaplastic) had arisen in Buckinghamshire furniture workers up to the end of
1967, two in males and two in females. It is extremely difficult to calculate
a realistic `expected' figure but using generous assumptions, namely that the
average annual risk of nasal cancer in the general population of both sexes is
10 per million, and that the population of furniture workers at risk was five
thousand, 0.5 cases would be expected over a decade."
The focus of the study was on AC, and as the study envisaged, more work needed
to be done, but meanwhile other studies raised the question of a link between
wood dust and SCC. In 1977 a Danish study by Andersen and others recorded in
its "discussion" section -
"Adenocarcinoma of the nose is a rare disease with annual incidence of 0.9 per
million in the total population. More than two thirds of those affected were
wood workers in the furniture industry, making it abundantly clear that this
disease is a associated with work in the furniture industry. ....
Looking at ectodermal cancers in the nose other than adenocarcinoma it appears
that 10 out of 99 tumours in men occurred in woodworkers ... this is
significantly higher than expected in the number of wood workers in relation to
the total male population in 1935 or 1969. Occupational exposure to wood dust
thus may be of importance in the development of other nasal cancers than
adenocarcinoma. The predominance of ectodermal tumours in males
(males/females: 114/43) in comparison with the sex ratio in the other tumour
types (males/females:15/14) also supports this hypothesis."
In 1982 a Scandinavian study by Boysen and Solberg drew on some monographs
published by the International Agency for Research on Cancer in 1981 when
asserting that workers in the furniture industry were at risk of AC, and also
faced "a smaller but significant risk of SCC".
In 1991 Vaughan and Davis published their paper on wood dust exposure and SCC
of the upper respiratory tract. They used data from two population based case
control studies in Washington State, and although the numbers were small and
the research method imperfect the findings did suggest that exposure to
softwood dust increases the risk of both sino nasal and nasopharyngeal SCC.
In 1995 Demers and others considered and re-worked twelve earlier studies from
7 countries). After dealing with AC they say -
"The squamous cell results present a more ambiguous picture. Women who had
been employed in wood dust-exposed jobs appeared to have a doubling of risk.
Although based on small numbers, some evidence for a dose-response with
duration of employment, especially in moderate and high wood dust-exposed jobs
was also observed. However, among men the risk of squamous cell cancer did not
appear to be related to employment in wood-related jobs or to level of wood
dust exposure. Although men employed for 30 or more years in wood dust-exposed
jobs appeared to have a slightly increased risk, it diminished when only
moderate or highly exposed jobs were considered. The relative risk was higher
when only long-term employment in moderate and high wood dust-exposed jobs in
logging, sawmills, and wood pulp preparation were considered. Possible
explanations for the differences in the relative risk of squamous cell cancer
between men and women include differences in exposure to chemical agents in the
workplace or sex -related differences in susceptibility or wood dust
exposure."
6. Caution
We were rightly warned by Mr Spencer not to give too much weight to what maybe
only hypothesis derived from a slim statistical base. Epidemiologists
recognise the principal risks - chance, bias and confounding. They try to
allow for those risks by various means, but, as Dr MacRae says in his chapter
on Epidemiology in Powers & Harris on Medical Negligence, to which Mr
Spencer invited us to have regard -
"It is a considerable understatement to say that no agreed criteria for coming
to a judgement regarding causation from statistical associations in
observational epidemiology exists. Indeed, there is no consensus whether or
not causation can be addressed at all by applying criteria to epidemiological
evidence."
Mr Spencer submitted that we should only attached weight to the views
expressed in the studies to which I have referred where the surveys carried out
by the observers used results which after the application of somewhat arbitrary
epidemiological criteria (which inevitably rely upon somewhat arbitrarily
chosen measures of significance) yielded results which an epidemiologist would
regard as being of statistical significance. I see no reason whatsoever to
adopt that approach in this case. I, like the judge, prefer to look first to
the expert witnesses to see what they made of the studies, bearing in mind that
Professor Seaton actually lectures in epidemiology to post-graduate students,
and Mr Baer also, as he indicated, keeps abreast of developments in that
field.
7. Experts' Conclusions
(A) Professor Seaton.
Professor Seaton's first report was dated 15th March 1996. In that report he
asked himself the question: "Does exposure to hardwood dust cause sinus
carcinoma other than AC?" He then referred to some of the literature and
concluded that "on the strong balance of probabilities exposure to hardwood
dust is an established cause of both squamous and AC of nose and nasal
sinuses". As Mr Spencer pointed out, Professor Seaton did not refer to the
Demers study (which had only just been published in the United States), and his
reference to Boysen was inappropriate, but those were points fully explored
before the judge at the trial, as were the various criticisms which could be
made of other studies which were referred to.
After seeing the reports of Dr Gallimore and Mr Baer, Professor Seaton, in a
letter dated 28th November 1997, did comment on the Demers study, which he
described as a "very thorough and careful statistical review". After
summarising the effect of that study Professor Seaton continued -
"The fact that the tumour is now known to be a squamous carcinoma weakens your
case slightly. However, it should be pointed out that the paper by Demers and
colleagues is simply a summary of other previous studies, several of which have
shown evidence of an association. The question is whether this is due to
chance or due to variations in exposure between different industries and
different parts of the world. This difference in exposure could be due to
intensity of exposure or type of wood. Thus, it is not possible to comment
with certainty that there is an association and the argument would have to
proceed along the lines of balance of probability. Considering the individual
case of Mr Huxley. The argument is thus:
(1) He developed an unusual tumour of his sinus.
(2) Exposure to wood dust is known to cause certain tumours of the sinus.
(3) He was exposed heavily and carelessly to wood dust at a time when it was
known that such exposures had the potential to cause cancer.
(4) A number of studies have shown such an association though it is fair to
say others have not.
(5) Some of the studies that have not have not even been able to show the
well-established association and should be discounted.
(6) Other studies that have not may not have done so because the specific
wood or trades in their area were not capable of causing this particular
tumour.
(7) In Mr Huxley's case, on the balance of probabilities, the presence of a
rare tumour in someone who worked with a known carcinogen, even though the
tumour was not of the typical histological type, was in fact due to exposure to
that carcinogen.
If the matter came to court this is the argument that I would use. My opinion
is that it is more likely than not that his tumour was caused by his exposure
if you are able to demonstrate, as I think you are, that he was indeed exposed
to high concentrations of wood dust at a time appropriate to cause later
development of cancer. Ultimately you must decide whether my argument is
likely to convince the judge."
Mr Spencer was critical of all of that. As I understood his argument it was
that the professor was shown to be partisan, and that in the end his approach
amounted to no more than "post hoc propter hoc". In my judgment the criticism
is unfounded. The professor was properly taking into account the Demers study,
and demonstrated his approach to legal causation in the light of that study.
He developed that approach further in his report of 5th August 1998 which set
out the nine Bradford Hill criteria which are used by epidemiologists to assist
when considering evidence of causation. He then looked at the other possible
causes of the claimant's condition, and applied the Bradford Hill criteria,
going on to say -
"In summary this evidence suggests, but does not prove to the level of medical
certainty, that woodworking in certain circumstances may be associated with an
increased risk of squamous carcinoma of the nasal sinuses. Taking the
well-known precautionary principle, it would be sufficient, even in the absence
of the very strong association with adenocarcinoma, for an occupational
physician to advise strongly that concentrations in the workplace should be
kept as low as practicable. In my opinion this constitutes sufficient evidence
of causation `on the balance of probabilities'."
Mr Spencer has really no quarrel with that save for the last sentence, and
Professor Seaton concluded with the opinion cited at the start of this
judgment.
As the judge said, Professor Seaton was extensively cross-examined in relation
to all of the material, and at times made concessions, but essentially his
evidence reflected the contents of his reports.
(B) Mr Baer
Mr Baer's first report was prepared as a result of his examination of the
claimant on 19th April 1996. At that stage there was still some uncertainty as
to the type of cancer from which the claimant was suffering, and Mr Baer said
that if it was shown to be SCC "current medical knowledge would suggest that
the tumour is not due to wood dust exposure." He cited the Acheson and Demers'
studies, to which I have referred.
Mr Baer's second report was prepared in the light of Dr Gallimore's
confirmation that the claimant was suffering from SCC, and having seen
Professor Seaton's first report Mr Baer commented "a thorough search of the
literature suggested that any link between wood dust exposure and squamous cell
carcinoma is weak."
In the witness box Mr Baer maintained his position, making clear his reliance
on the Demers study, but in cross-examination he was invited to consider the
other studies to which I have referred, and he accepted that, for example, the
International Agency for Research of Cancer statement cited by Boysen, was on
the face of it authoritative.
(C) Dr Gallimore
Dr Gallimore's position was as indicated by the judge.
8. Bradford Hill criteria
In the course of his judgment the judge referred to the evidence given by
Professor Seaton and Mr Baer in relation to the criteria suggested by Sir
Austin Bradford Hill as a useful means of evaluating epidemiological evidence
of causation. In this court Mr Spencer submitted that at times the judge
misused or misinterpreted those criteria. There are nine of them but not all
can be invoked in this case, nor is it considered necessary to have positive
responses in relation to all nine criteria in order to reach a conclusion.
They are-
(1) Strength of association between the disease and the alleged cause.
As Professor Seaton put it in his report "a strong association, one that is
highly unlikely to have occurred by chance, suggests a cause and effect
relationship." He found that whereas the figures in relation to AC were so
strong that they almost proved causation on their own, in the case of SCC the
figures could be considered no more than a starting point, though highly
suggestive of an association. Mr Baer considered the strength of the
association to be weak. Mr Spencer says that the judge did not express a view,
but he did. He accepted the evidence of Professor Seaton.
(2) Consistency of the association. As Professor Seaton put it in his
report, "do all, or most of the properly conducted studies of this relationship
point in the same direction?"
Again the judge set out the positions adopted by the two experts, and went on
to say of the apparently inconsistent Demers study that its level of detail
about the other studies which it considered does not enable one to undertake
an analysis of the reasons for that inconsistency. Mr Spencer submits that the
judge misunderstood the weight of the evidence to the opposite effect,
particularly in his reference to Boysen and Solberg, but as I have attempted to
demonstrate, the meta-analysis in Demers apart, the substantive studies brought
to the attention of the judge do show as he said "an appreciable measure of
consistency".
(3) Specificity of the association. As the judge said this criterion
is concerned with whether the disease is confined to a particular group exposed
to a particular agent, and, as Professor Seaton accepted, AC is caused by
hardwood dust but SCC has several causes. As Mr Spencer submits, this
criterion in this case, takes matters no further.
(4) Relationship in time. The judge, rightly on the evidence, adopted
Professor Seaton's question "does the disease follow the exposure after an
appropriate time interval?" He answered the question in the affirmative. Mr
Spencer now submits that the judge misunderstood the criterion, which is
related to general causation. Dr MacRae says of this criterion -
"This means that the exposure must precede the disease. This is sometimes
difficult to determine, especially if the disease could have been present in a
latent form prior to exposure."
In his third report Professor Seaton says -
"With respect to time relationships, the evidence mentioned above, and
especially that in Mr Huxley's case, strongly supports a causal association.
cancer usually requires at least fifteen years of moderate level exposure and a
lag of a further 10 to 20 years before the tumour develops."
I am far from satisfied that there was any relevant misunderstanding
by the judge.
(5) Biological gradient. It is common ground that this factor involved
an examination of the relationship of the exposure its intensity or duration to
the alleged cause. As Professor Seaton put it "can the risk of disease be
related to the exposure to the suspected cause." The judge rightly considered
the Demers paper under this head, which suggested a weak relationship between
SCC and duration of employment. The judge then accepted Professor Seaton's
reservations in relation to the Demers study (related to lack of information as
to its underlying studies) but he did not, as Mr Spencer submits, rely upon the
Demers study to demonstrate a biological gradient.
(6) Biological Plausibility. As Professor Seaton said, the
relationship postulated in this case makes scientific sense. Wood dust causes
AC and respiratory carcinomas derive from a common stem cell. The working of
wood generates predominately particles of a size preferentially deposited in
the nose which suggests an appropriate biological mechanism for their causation
of SCC of the nasal sinus. That was not really disputed, but the judge rightly
reminded himself of Dr Gallimore's "uncertainties in evaluating causation from
a pathological standpoint."
(7) Coherence of the evidence. In Professor Seaton's words "do the
reserved relationships conflict with what else is known of the natural history
and biology of the disease?" As Mr Spencer accepts that question had to be
answered, as Professor Seaton answered it, in the negative. Mr Spencer submits
that the question is only significant if the question has to be answered in the
affirmative.
(8) Experimental evidence. It was common ground that there was no such
evidence.
(9) Analogy. Whether the observations accord with previously described
associations in other fields. In his third report Professor Seaton said in
relation to this criterion -
"There is little to suggest that industrial or other environmental
carcinogens cause single cell types of tumour response. Certainly, of the two
most important respiratory carcinogens cigarette smoking is associated with the
entire range of lung cancer cell types with a tendency to a small deficit of
adenocarcinomas, and asbestos is also associated with the whole range but with
an excess of adenocarcicnomas. In other words, environmental carcinogens alter
the relative proportions of cell types among the tumours they cause rather than
causing cancer of one exclusive type."
The judge in his judgment summarised that by saying -
"Professor Seaton considered the evidence that cigarette smoke and asbestos
caused cancers of different types and were not causative of cancer of a single
cell type was a useful analogy."
That was perhaps too brief a summary, but there is no reason to believe that
the evidence was not fully understood.
In my judgment such criticisms as can be made of the judge's review of the
Bradford Hill criteria are of no weight. The criteria were understood and used
by the experts, and understood by the judge, and he rightly looked to the
experts to assist him as to their proper use.
Professor Seaton said that using the Bradford Hill criteria he could not say
there was a clear cut answer. He considered that the epidemiological evidence
proved on a balance of probability but not to a level of medical certainty that
exposure to wood dust was associated with an increased risk of SCC. Mr Baer
disagreed, but the judge, as he was entitled to, adopted Proffessor Seaton's
view.
9. Grounds of Appeal
I turn finally to the grounds of appeal. They attack the judge's preference
for the evidence of Professor Seaton, his use of the Bradford Hill criteria,
and his ultimate conclusion in relation to causation. For the reasons I have
set out I conclude that the attack fails. In my judgment there was ample
evidence in this case to enable the judge to decide as he did, and I would
dismiss this appeal.
Lord Justice Buxton:
I gratefully adopt the statement of the issues and the account of the evidence
that has been set out by Kennedy LJ. I also would dismiss this appeal.
It was the cornerstone of the appellants' criticism that it was necessary
first to be satisfied on the basis of epidemiological studies that there was
generally a more than random or chance possibility of SCC being caused by
exposure to hardwood dust; and only if the Judge were properly so satisfied was
it permissible to go on and consider whether the instant individual case is one
that fits into the overall pattern of causation. Mr Spencer said that the
Judge, and Professor Seaton, had taken these steps in the wrong order; or, at
least, had not followed the proper scheme with sufficient rigour. They had had
regard, in particular, to the plausibility of the claimant's illness being
caused by exposure to dust, and in so doing had given insufficient weight to,
or had overlooked, the difficulties as to the statistical analysis of the
epidemiological studies.
The Judge did, however, address the epidemiological studies, and in my view at
the appropriate stage of his judgment. But it was contended that he had been
wrong in his approach because, in the terms of paragraph 8 of the Notice of
Appeal, he
placed excessive weight upon the scientific studies by Acheson et al (1972),
Andersen et al (1977), Boysen et al (1982) and Vaughan and Davies (1991) and
insufficient weight upon the Bradford-Hill criteria and the study by Dermers et
al (1995).
I am afraid that I found this concentration upon the statistical implications
of the evidence, and the Judge's handling of it, extremely unsatisfactory. No
statistician was called at the trial, and we were therefore sought to be drawn
into areas of technical expertise without any specific guidance upon them. I
am, however, entirely clear that in these circumstances the Judge was fully
entitled, indeed bound, to act on such evidence as was made available to him,
and not to seek to enter upon his own analysis or critique of the implications
of the scientific studies. In so doing he was further plainly entitled, and
for my part I would say also bound, given the view that he formed of the
witness, to follow the guidance that he received from Professor Seaton. As my
Lord has pointed out, although Professor Seaton was not a professional
statistician he was highly distinguished in a field, occupational medicine,
which inevitably engages the techniques of population studies that are found in
the material relied on in this case. The Judge was right to compare his
qualifications in that respect favourably with those of Dr Baer, who agreed in
cross-examination that his evidence and opinion was based almost entirely on
the Demers study. It was therefore both second-hand, and self-supporting: in
that it assumed the paramountcy of the Demers study. If the appellants wished
to contend as a matter of statistical expertise that, as it were, the Demers
study trumped all other work in the field, then they had to call a witness
skilled in that expertise to say so: which Dr Baer was not.
The Judge is therefore not to be criticised for the respective weight that he
gave to the various scientific studies. He did that because what he rightly
considered to be the most reliable evidence available to him pointed him in
that direction. The premise of the appellants' criticism, that the Judge
should have accepted that the evidence established no general pattern of
causality, thus disabling him from considering any further the case of this
particular plaintiff, therefore falls away.
Had the appellants been right on their main point, it would not have been
appropriate for the Judge to consider the Bradford-Hill criteria at all,
because their only function is to assess whether a general pattern of causality
can be applied to establish causation in the instant case. However, the Judge
having concluded as he did on the epidemiological studies, the question does
arise as to whether the present case falls within that causal pattern. My
Lord has set out how, in performing that taks, the Judge approached the
Bradford-Hill criteria and, subject to some expansion on one point which I deal
with below, I respectfully entirely agree with his account and observations.
I would also add, more generally, that there was some disposition on the part
of the appellants in their argument before us, and certainly in the Notice of
Appeal, to treat the Bradford-Hill criteria as if they were some sort of series
of threshold tests, each of which the case had to pass before it could be
treated as a reliable case of causality. We were not shown Sir Austen
Bradford-Hill's original paper, but from the way in which his criteria have
been handled in other material it seems clear that this approach, if it was
that of the appellants, is incorrect. The criteria are no more than reminders
of the factors that should be reviewed before assuming that a connexion that is
statistically significant over the whole population is established in the
instant case. It was Professor Seaton who introduced the Bradford-Hill
assessment into the case, but he said in his report, rightly in my
estimation:
Not all of these criteria need to be satisfied and indeed some if strong are
sufficient by themselves.
The "criteria" are in truth no more than guidelines for the skilled man: nor
indeed could at least some of them, expressed in wholly unspecific terms such
as "biological plausibility" or "analogy", be anything else. Looking at the
criteria in this way, the relevant skilled man, Professor Seaton, was satisfied
that the causal relationship was established with sufficient certainty in this
particular case. The Judge was plainly entitled to act on that evidence.
I wish to add something in relation to one of the criteria reviewed by the
Judge, biological plausibility. This concerns the medical explanation for the
presence of the malign condition in the particular patient. If a doctor can
give an explanation of how the actual symptoms seen in the patient might have
been caused by the exposure complained of, then a case based on statistical
association is materially strengthened. That issue can be illustrated by the
early disputes about studies linking smoking with lung cancer, which were
sought to be devalued because they showed a "merely" statistical association,
without identifying the cancer-inducing agent, or explaining how it worked on
the human body. Professor Seaton placed importance in his report on the fact
that the tumour arose in a nasal mucus membrane that showed evidence of chronic
irritation; and that Dr Gallimore's histological examination had shown Mr
Huxley to have a severely dysplastic squamous epithelium. Professor Seaton
said, and the Judge accepted, at page 7 of the judgment, that
"squamous dysplasia was considered an important precursor to the development
of carcinoma. It was strong evidence of prolonged pre-cancerous irritation
such as would have been caused by wood dust which would then have developed
into cancer."
The Judge returned to this point when dealing with biological plausibility
under the Bradford-Hill criteria, but as my Lord has pointed out he did not
regard Professor Seaton's conclusion as at all dispositive. The issue took on
a somewhat larger role in the appeal, because Mr Williams came close to arguing
that this medical explanation of Mr Huxley's condition, once it was properly
accepted by the Judge, largely concluded the case in Mr Huxley's favour. That
contention caused Mr Spencer in reply to mount a frontal attack on this part of
Professor Seaton's evidence. He said that Professor Seaton had misunderstood
the findings of Dr Gallimore on which the evidence was based. The inflammation
had been found in the post-operative, not pre-operative, examination, and might
have been caused by the cancer itself; and the dysplasia referred to by Dr
Gallimore was not pre-cancerous dysplasia, but a cautious description by Dr
Gallimore of what he thought was almost certainly in itself a cancerous
condition.
I have gone into this matter in some little detail because it played a
significant role in the argument before us. That significance was, however,
forensic only, and evoked by the course taken by the respondent's argument.
The point was not seen as sufficiently significant to be put to Professor
Seaton at the trial, and it did not feature in the Notice of Appeal: there the
complaint was that the Judge placed excessive weight on Professor Seaton's
theory as to causation, rather than, as it was before us, that the theory was
one on which the Judge could not act at all. That original course was taken
because (as indeed I understood Mr Spencer to agree), even if Professor Seaton
was mistaken as Mr Spencer contended, it could not be said to be medically
implausible that Mr Huxley could have contracted his cancer from wood-dust: the
only point was that the medical evidence did not in itself establish a causal
connexion between the wood-dust and the cancer. That connexion was, rather, to
be established on the balance of probabilities on the basis of all the
evidence: an enquiry that the Judge properly carried out.
LADY JUSTICE HALE:
I have read and agree with both judgments.
Order: Appeal dismissed with costs; legal aid assessment.
(Order does not form part of the approved judgment)
BAILII:
Copyright Policy |
Disclaimers |
Privacy Policy |
Feedback |
Donate to BAILII
URL: http://www.bailii.org/ew/cases/EWCA/Civ/2000/139.html
