Smokingissues
Lobacco was first brought into modem use by the
Frenchman
iea_,- Nicot in the mid-sixteenth century. Since
then, the pleasures of
=6 ~ng have attracted a strong following
worldwide, major industries
have zrown from it and governments have come to
rely heavily on
toba=o as a source of revenue.
But from as far back as James I the practice of
smoking has had its
critics- Today a body of medical opinion maintains
that smoking can
caus;e or can contribute to, various diseases.
Despite the fact that massive attention has been
directed to the
subject - with perhaps unprecedented media
interest - I believe that
very. *;Mle is in fact known about either the
underlying statistics ofthe
cl-i---J risks, or about the way in which present
day cigarettes differ in
desizz and smoking characteristics from the mainly
non-filter cigarettes
that Z-40minated the scene until the early 1960s.
My aim is to present a summary ofsome ofthe key
components of
the ='es, not to make judgements or to present
opposing arguments ir
the c:i-- ' on smoking and health.
going controversy
Thus, I wish to stress that the findings and
interpretations I shall
pre=r are those of independent medical and
scientific researchers.
Whi:e BAT does nor necessarily adopt these views,
they are interesting
in the Lighr of existing controversy about
smoking.
1 :ecognise char no summary can do justice to such
a highly
com::7ex issue, but I trust that what I will say
will be of interest to this
aud!=ce.
Qq
CD
CC)
(A
rIJ
U1
BATCo document for Province of British Columbia 29
October 1999


B-MA20- ars[ady
W
Of UK ctors
do
It is perhaps fitting to start by making reference
to the world
famous BIMA prospective epidemiological studv of
British doctors
started in 1951 by Dr R Doll and Dr A B Hill - and
subsequently
reported by Professor Richard Doll and Dr R Pt:o
after 20 years
of observation in the British Medical journal of
December 1976.
From this pioneering study there has developed a
body of
opinion that:
(a) smoking is statistically associated as a
risk factor with cert ain diseases
(b) the risk is related to "dose", ie to the
daily consumption ofcigarettes.
By far Elie major area ofinterest has been the
statistical
association ofsmoking with the highly emotive
subject ofcancer
- in particular of lung cancer. In recent times,
however, the
incidence of lung cancer has started to decline
strongly in the UK
and esperi2fly in the younger age groups. This is
shown in the
figure below, where statistics for four age groups
relating to the
years 1971, 1976 and 1979 are compared with those
for the base
year 1966.
120
110
100
90
80
70
60
50 -
40 -
30 -
20 -
10 -
0
1966
2
1971 1976
Age
65-74
55-64
45-54
35-44
1979
C)
00
NJ
CT\
BATCo document for Province of British Columbia 29
October 1999


Perhaps as a result ofthis decline in lung cancer
Ehere is now
an increasing shift in comment towards the
association ofsmoking
with heart disease. I think, however, it is
instructive to put the
computed statistical mortality risks from lung and
heart disease
and indeed from some other common causes ofdearh -
into
relative perspective; and to do this I shall draw
directly an the
previously mentioned 1976 publication ofProfessor
l3oll and
Dr Pero that derailed the causes ofdeath of 10,072
doctors out of
the 34,440 doctors who took part in the BMA
20-year study.
In this presentation, I merely reproduce the
computed age-
standardised mortaLity rates for Non-Smokers and
for the general
group that were designated 'Current or Ex-smokers'
for the major
causes ofdeath - totalling some 6,000 out of the
total of 10,072.
It is not my purpose to go into fine derail, and
indeed I
acknowledge J~t I have ignored the fact that in
several cases the
computed statistical risks increased with the
level of smoking.
But I believe that the table does draw attention
to some points that
I believe are nor generally known:
1. Death from heart and cerebrovascular was very
much greater than that of
diseases out-weighed lung cancer by a lung-cancer
for Current/Ex- smokers.
factor of some 10 to 1. 3. The actual number
ofdeaths, and
2. The computed age-standardised computed
statistics, for some other
mortality rate ofhear- and common causes
ofmortalirv make
cerebrovascular disease for Non-smokers
interestinq reading. In mos'z cases the
association with smoking was marginal.
BATCO document for Province of BritiSh Columbia 29
October 1999
icauses ~,SJA67.j:ijJ,4 3


0
Further ical
epjdemiolog
studies interpretation
and
I shall now turn to the evidence that is being
cited by
independent medical and sciendfic researchers that
the modern
cigarette, with its substantially lower smoke
'deliveries', is
associated with a lower risk to health.
It should be noted that most ofthe acknowledged
authorities I
shall refer to charaaerise themselves as
'anti-smokers'.
First, let me surnmarise the findings ofa number
ofmajor
worldwide epidemiological studies that have been
interpreted in
terms of cigarette design:
PERCENTAGI~ REDUCTION IN INCIODENCE OF
_-SMOKI .NO ASSOCIATED DISEASES IN M .&N
Disease ource'.
ompanson
C
Reduction
Lung'cancer -',-..Bross& Gibson (19.68) .. .Filter
v Plain 41
600)' F terv
11 plain
.45
hTiN
LowV ig 18
Filter v Plain 46
1978) Filter v Plain 16
Plain 24
Heartdise'ase `- Low v High TIN 10
Pl.
Ijean(1977)-- ,'Filterv ain
.'Ha ome(1978) -' Filter v Plain
16
Lary=cancer -Tynderetal.(1979) FiltervPlain 39
Stroke :-Dean (1977) Filter v Plain 32
Chronic brqnchitis Dea '
(1977) Filter v Plain
34
Of all the epidemiological studies to date,
probably the most
comprehensive, and well known, is the so-called
"One IMillion"
prospective study sponsored by the American Cancer
Society -
started in 1959 by Dr E C Hammond, and now
continued by
Dr Lawrence Garfinkel, Vice President ofthe
American
Cancer Socim.
4
CD
co
Co
BATCo document for Province of BritiSh Columbia 29
October 1999


The Followine is arectnt direct quotation from a
paper by
Dr Garfinkel to the 1979 World Smoking and Health
Conference:
Analvsis made ofthe mortality in cigarette smokers
in
the American Cancer Society study showed that
there was
a small but consistent decrease in both men and
women in
those who smoked cigarettes with low tar and
nicotine
compared to those smoking high tar and nicotine.
Over a 12 year period, the lung cancer rate
averaged
26% lower.
But perh2ps the most interesting recent
publication is that of
the US pathologist Dr 0 Auerbach, ofthe Veterans
AdministrationlMedical Cenrre, New Jersey, who in
two periods,
1955-60 and 1970-77, examined 20,425 sections
taken at autopsy
from the bronchial tubes of445 non-lung, cancer
deaths.
The results were expressed in two forms.
First, the percentage ofslide sections that
exhibited pre-
cancerous changes that might have developed with
further time
Cr
into a lung cancer.
_4% Sections w#hPre-cancerous Changes
-4-
--.~.,(A) 1955-60
(B) 1970-77
on oier' 0
im S
N
.-Per y
..20-39*day-'.-*i~-;--'"".-...'-~-;-13~2 8
y 5
-Z p
'Group A (1955-6 1211
death'&-'-'~t'i54i;giilir'si~okiis
Croup B (1970-77) 234 deaths-'. 181 regular
sm"okers
Second, the percentage ofslides that exhibited
7007o or more
atypical" cells, again classifiied by smoking
habit.
80
TO 66.6
60
50
40
30
20 12.2
10
0 0 0 0.1 0 0.1
A B A B A B A B
NEVER SMOKED SMOKED SMOKED
S.140KED I PACK 1-2PACKS 2 + PACKS
REGULARLY A DAY A DAY A DAY
BATCo document for Province of BritiSh Columbia 29
October 1999


Ln a joint presentation of the findings a~ 'he
1979 Cold Spring
Har'l-our Conference, which was entitled'A Safe
Cigarette-% the
authors concluded With the statement that "The
tenfold decrease
in carcinoma-in-siru obsmed in this studv, we
believe, has
trem,ndous significance in the future of lung
cancer in the US."
In a discussion ofthe paper Dr Gio Gori, a Deputy
director of
the US National Cancer Institute, commenced: "And
perhaps
what we see here is the forerunner of what we will
be seeing in
epidemiology in a few years from now."
The paper was published in the New England journal
of
Medicine (22.8.79) jointly by Dr Auerbach and by
Drs H:-mond
and Garfinkel of the American Cancer Society - but
only after
the latter co-authors had satisfied themselves
about the accuracy of
the findings through repeat checking and by an
independent
r-nination ofa sample ofthe slides by another
person.
A ffirther coyr-ent on the Auerbach findings, and
on the
general subject ofthe reported decreasing
incidence ofsmoking-
associated diseases, was given by Dr Richard Peto,
Reader in
Cancer Studies at Oxford University, in an article
published in
"Nature" on 27 March 1980:
Smokers of 'Less Dangerous' cigarettes have
already
been found in various epidemiological studies to
have
disease rates which are materially lower th--
smokers of
other cig:-ettes.
At autopsy they have far fewer 'Pre-Malign-nt'
histological -h-nges in their bronchi a-i, perhaps
due to
the changes in cigarette composition 10 or 20
years ago,
male lung cancer death rates in early middle age
are now
decreasing in North America, in Brit-;,, and in
Finland.
A more recent comment an lower delivery products
was
given by Professor Doll, Dr Peto and Mr Copeland
in a paper in
the British Medical journal (7th March 198 1):
"The trends in tar
yield may well explain the reduction in lung r-er
in the U.K.
better rh- has been suspected hitherto."
lklany other independent scientific authorities
have made
si-il-r suggestions that the reduced deliveries
oftar, nicotine and
oth,-r components oftobacco smoke are primarily
responsible for a
reduced incidence ofsmoking associated diseases.
It has been
pointed out, however, by Dr E L Wynder and Dr D
Hoffman of
the American Heart Foundation, that because ofthe
very long
induction period believed to be associated w-ith
lung disease
(20-30-40 years) any correlation with the clizagge
in cigarette
tA
CD
(:D
BATCo document for Province of BritiSh Columbia 29
October 1999


deliveries could onlv relate to the earlv 1960
trend to Flitered
cigarettes with reduced car and nicotine
deliveries - not to the
current move started in the mid-'70s towards even
lower deliveries
by means ofvenEilated cippings, porous paper and
modified
tobacco blends. Thus, they stated in a paper in
the .4inerican
journal ofPublic Health, November 1980:
46 Presently available epidemiological data relate
only to
those who began to smoke cigarettes with high
'tar' and
nicotine yields.
Only future studies will demonstrate the risks for
those who initiated and continued their smoking
habit with
low yield cigarettes.
One wav that has been suggested to rest the basic
hypothesis
that low delivery cigarettes are associated with a
low health risk is
to carry out a major comprehensive prospective
epidemiological
survey - in which groups of people who had chosen
to smoke (a)
low delivery cigarettes, and (b) higher delivery
cigarettes
throughout most ofthe 1970s were documented, and
then their
smoking habits and their health followed for at
least 10 years.
However, there are serious questions about
conducting such a
test, eg the large number ofsubjects that would be
necessary,
difficulty ofge=.g reliable data, the
self-selection of subjects, the
greatly differing environments, the organisation
and the cost.
Before leaving the subject of epidemiological
evidence
relating to smoking and health, it is pertinent to
note the recent
study of c. 16,000 men aged 35-64 years who
attended the BUPA
Centre, London, for comprehensive health-screening
examination
between March 1975 and December 1978. In a paper
published
by Dr A Bailey in "The Lancet" on 18 October 1980
- in
collaboration with Dr Nicholas Wald of the Cancer
Epidemiology
and Clinical Trials Unit, University of Oxford -
Vitamin A
(retinol) levels in 86 subjects who had developed
cancer were
compared with 172 controls who had not developed
cancer.
The following was the striking finding:
Low retinol levels were associated with an
increased
risk of cancer. The association was independent of
Age
Smoking habits
Serum-cholesterol level
and was greatest for men who developed lung
cancer.
c0
BATCo document for Province of BritiSh Columbia 29
October 1999


. Chaxigqs in
cigarette design
The main features ofa modern low delivm cigarette,
with
ventilated tipping, is illustrated below:
Permeable Paper Tobacco Air Vents Filter
Air Air
.4 Smoke
,
filtered
and
diluted
with air
I think it may be ofinterest at this stage to give
some facts
regarding lower delivery cigarettes. The
present-day cigarette
looks physically almost identical to the cigarette
ofthe past.
But the deliveries ofthe components considered
harmfal in early
cigarettes are very different from those in the
average modem
cig
.,a rette - and vastly different from those of the
advanced
cigarettes which are available in many countries.
First, some statistics about how the
"sales-weighted" average tar
and nicotine deliveries have changed significantly
since the
mid-1960s:
SALES -GHTED TAR
"S.,
S--
S~MERLNND
E
% %
% %
%
%
%
',I % 7; 74 .1 1.
13-
%
%
GEK-I'l
% %
%
%
%%
C:)
T:NE
71 72 *1 71
N)
BATCO document for Province of BritiSh Columbia 29
October 1999
S-1. -.-TED' Q-T,1Z


It will be noted that while tar levels tend to be
relatively
similar in the countries cited, there is a much
greater difference in
the levels ofnicorine - which presumably is more
sensitive to
'national character'.
The next table indicates the current position
regarding
commercial products in typical Virginia and US
blend styles:
LOW DELIVERY CIGARETTES - 1960 V 1980
Smoke component Delivery levels per cigarette (mg)
"Typicar' "Avenge" Commercial products
before 1960 in 1980 available 1980
1 ~irjinla- iyle
cigarettes,
e.g. U.K., Canada
Tar condensave 33 :15 1
cotine-
2
1.2
, 0.1
Carbon monoxide 7~.:-..20 li
s
V.S. bl ty e
cigarqtes
Tar co~ensate
43
15
Nicotine 70.2-7-
Carbon monoxide `23 17 2
CZ)
co
9
BATCo document for Province of BritiSh Columbia 29
October 1999


While on the subject ofdelivery statistics, it is
perhaps
:-!-tinent at -.his stage to note that despite the
strong ami-smoking
-.,.,ssures in the pas[ years, the general
worldwide rrend of
imoking is upwards - even in many developed
countries:
WORLD CIGARETTE CONSL;.MPTION TRENDS
(Cigarette production: billion)
(Log: x 10') 5,OOC de Zoete & Bevan 16.1.1981
Total
4,50C
4,000
3,50C
3,000
2'5oc
2,OOC - Developed
countries
Communist
1,50C - bloc
0 Developing
countries
50-3
1973 74 75 76 77 78 79 1980E 1985E
It will be seen that the world consumption in 1980
of4500
billion cigarettes is estimated to increase bv
1985 to 5000 billion
cigarettes.
If we note the assumption that the average
deliveries in 1985
will be 15 me tar and I mg nicotine, then the
cigarettes will be
only about one-third as strong as the typical 1960
product.
Thus, in terms OfEhe deliveries, the 5000 billion
in 1985 would be
-quivalent to only about 1500 billion of the
1960-cype product.
10
CD
c0
~A
BATCo document for Province of BritiSh Columbia 29
October 1999


Thus, by cnanzing the design ofthe cigarette, the
tobacco
industry has already reduced the tar and nicotine
deliveries (and
those oforher chemical components in the gas and
particulate
phases ofsmoke) by the equivalent oftwice the
projected total
world consumption of the assumed typical cigarette
in 1985.
And there will continue to be a trend downwards
commensurate
with consumer choice.
I leave it to you to decide on the relative merits
- or
effirectivcness - ofmaintaining the current 'ban
everything'
approach ofmany of the anti-smoking pressure
groups, compared
with the more selective approach recommended by
others of
encouraging smokers to switch to low or ultra-low
delivery
ci2arettes.
Critical level .4;
Of daily intake
Having summarised the reports from various
epidemiological
studies, and described the way in which both the
design and the
delivery level ofcigarettes have changed markedly
since the 1960s,
I should now like to give a briefaccount ofthe
work, and beliefs,
of Dr Gio God - who in 1976 caused such a stir in
the US anti-
smoking scene that he was fired from his position
as Deputy
Director, Division of Cancer and Prevention,
National Cancer
Institute, by the then US Secretary of Health,
Education and
Welfare, Mr Joseph Califarno.
The basic thesis that Dr Gori developed was that
there must
exist maximum daily intakes ofvarious smoke
components below
which the risk ofsmoking-associated disease is
minimal. Based on
a number of epidemiological surveys carried out in
1950-60,
Dr Gori deduced that a person in those days who
smoked 2
cigarettes a day in 1950-60 had a risk of
mortality (from all diseases)
equal to that ofa non-smoker.
CD
c0
U-1
BATCo document for Province of BritiSh Columbia 29
October 1999


From the average deliveries of cre-1960 US
cigarertes Dr Gio
Gori computed average "critical leveis"
ofdailv;nrake for the
various components - none of which levels were to
be exceeded
when smoking present day cigarettes lest the
smoker should
exceed the equivalent oftwo pre- 1960 Cigarettes
in so far as the
most commonly-cited suspect components were
concerned.
These average critical levels were as Lndicated
below:
-Smoke component Average
pre-1960 US -daily critical
cigarette 7v2lues,
'Tar (mg) .43 86
(mg) 3 6
Co (mi)
-23
`_46
.~NOx(ug) 270 540
HCN (ug) 410 820
Acrolein (,ug) 130 .-260
A brief illustration ofthe working ofthe
hypothesis is as
follows: considering a hypothetical cigarette
giving the three main
deliveries as below, the maximum number of
cigarettes that could
be smoked with minimal risk would be 12, ic the
lowest ofthe
three critical values (which in this case would
relate to nicotine):
Hypothetical Daily Number OfCig2rertes
Cigarette giving "critical leveW'
Tar (mg) 4 22
Nicotine (mg) 0.5 12
Carbon Monoxide (mg) 3 15
Although not popular among the medical world, nor
to date
adopted in the US, it is still possible that the
critical level
approach will have an influence on product
development and an
thinking in the area of smoking and health.
Certainly Dr Gori himselfis still strongly
propounding his
theory ofcritical values. In a recent 1980 paper
on "Observed no-
effect thresholds and the definition ofless
hazardous cigarettes"
published in the journal of Environmental
Pathology and
Toxicology, Vol 3, 1980, Dr Gori concluded with
the followin!z
thought-provoking paragraph:
2
L.,4
BATCo document for Province of British Columbia 29
October 1999


The technology for producing cigarettes that, at a
use
level of 10 to 20 per day, deliver total smoke
components
within the critical values has been developed and
can be
applied on a mass scale by the skilled cigarette
manufacturer. This is the single most important
and
potentially successful disease-prevention
opportunity in
contemporary society. It can be seized by means of
responsible marketing decisions in the cigarette
industry,
by a major PubUc education drive leading smokers
to new
patterns ?facceptance, and by the promulgation of
judicious legislitiviincentive's.'
Cubo n-- Monoxide
mdNicotine
I should next like to make brief co=ents on two
much
discussed components in cigarette smoke - carbon
monoxide and
nicotine.
Carbon monoxide is a natural by-product present in
the
human body whether one smokes or not, and it can
be absorbed
from the environment, eg from exhaust combustion
gases.
It is toxic in large amounts because it affects
the oxygen-carrying
capacity ofthe blood. Studies on animals exposed
to carbon
monoxide, at levels much above that of cigarette
smoke, have
presented a confusing picture. But the general
view is that healthy
individuals are well able to tolerate the blood
levels arising from
either the environment or from inhaling cig3rette
smoke.
Current expert opinion may be cited:
Letter to UK Secretaries ofState
bv Lord Hunter, 5 December 1979
Although the epidemiological evidence suggests
that
tobacco smoke is associated with a number of types
of
cardiovascular disease, we have found the evidence
to
indicate that carbon monoxide is the principal
factor to be
less than convincing.
C)
CO
(_14
U-4
-14
BATCo document for Province of BritiSh Columbia 29
October 1999


E L Wynder and D Hoffman
New Eng
gland journal ofMedicine - 19.4-79
49 A reduction in lung-cancer risk among smokers
of
filter cigarettes has been shown in a prospective
study by
Hi-ond and in our studies.
In addition, H----nd has noted that the risk of
heart
attacks is reduced ;-I Koch has reported a
reduction in
periplic.-I vascular disease.
From this evidence one would infer that carbon
monoxide does not have a major role since carbon
monoxide was not substantially reduced in filter
cigarettes
until products with perforated- filter tips were
introduced.
Report ofthe US Surgeon General 1981
"The ('11mging Cigarette"
Carbon monoxide has been impugned as a harmful
constituent of cip mrte smoke. There is no
evidence
av-;I-ble, however, that permits a deter-;nation
of
ch-Ses in the risk ofdiseases due to variations in
carbon
monoxide levels.
One may note from an earlier table that carbon
monoxide
delivezies are in any case reduced to very low
levels in modem
ventilated cigarettes.
The second substance, nicotine, has equally been
very much
discussed:
- as a stimulant
- as a rranqi,illker
- as a poison
- as a pharmocological habit-forming factor in
smoking
- as a (-hort-term) accelerator of heart rate
- as a constrictor ofarteries
- as a carcinogen or co-carcinogen
14
CD
c0
co
BATCo document for Province of British Columbia 29
October 1999


But, irrespective afthe fact that nicotine levels
in smoke are
very low in absolute terms, and have fallen
significantly in the past
30 years from above 3 mg/cigarette to around I mg
or below, it is
interesung to note the recent comments ofDr E L
Wvnder in the
discussion on a paper on cardiovascular disease
given at the 1979
Cold Spring Harbour Conference:
66 If you measure cotinine-nicotine levels in the
blood
system of cigar and pipe smokers, you find they
are as high
as those among cigarette smokers.
* ' ' *We know there is no increased risk for
coronary
diseases among cigar and pipe smokers, even with
their
clcvated cotinine-nicotine levels.
Ambientsmoke
Finally, I should now like to turn to the emotive
question of
ambient smoke - also known principally by the
anti-smoking
movement as 'Passive Smoking' or'Involuntary
Smoking. I refer,
of course, to the inhaling of tobacco smoke by
non-smokers.
It is not my place to comment on the social
aspects, but I wish
to refer to the alleged health risk of inhaling
ambient smoke as it is
frequently used by the anti-smoking groups to
initiate their
campaigns to make smoking socially unacceptable.
The liter-2ture on the subject is voluminous - one
review I
have just seen contains some 100 references - and
the view
generally expressed until very recently is that
there is no risk of
disease. Thus, it was noted in the December 1978
draft status
report ofthe US National Cancer Institute that
"The risk of
cancer of the respiratory tract, emphysema, or
cardio-vascular
disease does not appear to be increased by
passively inhaling
smoke generated by others".
CO
15 (-N
UN
BATCo document for Province of BritiSh Columbia 29
October 1999


Other more recent statements are as follows:
197 9 US Suz.,eon General's Report
66 Healthy non-smokers exposed to cigarette smoke
have
little or no physiological response to the smoke,
and what
response does occur may be due to psychological
factors.
1979 Chairman ofthe American Heart Association
Task
Force on the Environment and Cardiovascular Disime
66 Studies indicate that non-smokers have
negligible
levels of carboxyhemoglobin under good conditions
of
ventilation, and with no ventilation have
acceptably low
levels.
1980 C Hugod, K Hawkins and P Astrup
69 It is pointed out that in spite of an often
considerable
subjective discomfort, exposing non-smokers to
tobacco
smoke under realistic conditions will not
cause;-h-lation
ofsuch amounts ofthe components oftobacco smoke
traditionally considered harmful, that a lasting,
adverse
health effect in otherwise healthy, grown up
individuals
seems probable.
nese conclusions seem entirely in keeping with the
measured very heavy dilution of'sidestream'
ande-xhaled smoke
in a t~-pical room, office, train, etc under
realistic conditions:
CHARACTERISTICS
MEASURED TOBACCOSMOKE
,
(W C I-Ends and M W Fmt)
.
-i Z Nc~v En glai ni'd J o urnai I
i
edicif
-;L-: 17.4 75
sure
venge ca
quivalent F ter_
1 codne (
~~ -Ci tte 1~~9ked/per dav
n
Commureitrain
4.9 0.004
Commuter bus 6.3 0.005
Bus waiting room 1.0 0.001
Adfline waiting room 3.1
0.003
Restaurant 5.2 0.004
Cocktail lounge 10.3 0.009
Station loun" 2.8 0.002
CD
c0
BATCo document for Province of BritiSh Columbia 29
October 1999


T-his generally accepted situation was
challenz.:a', however,
last year by a paper published by Dr White and Dr
Fmeb - the
forr.-I-er well known as a highly active
campaigner a-2inst smoking
(eg the California Propositions: Number 5 in 1978
and Number
10 in 1980).
The re;ort oftheir work in the New England journal
of
NlecEcine, which was based on measurement of
Forztd Expiratory
Volu.me ofnon-smokers - who were claimed eithe.-:o
have been
exposed or not-exposed in their working
environment to cigarette
smoke - has received strong criticism from other
research
workers on the grounds:
(a) the design ofthe study lacks credibility
(b) the findings lack credibility (non-smokers
said to show
effects equivalent to smoking up to 10 cigarerLts
a day)
(c) several ofthe lung-function tests used are
con:-oversial as
regards their importance in lung disease.
Much more recently, however, in January this year,
the
British Medical journal carried a paper by Dr
Hiray-arna on a study
on the incidence oflung cancer in Japanese
non-smoking wives of
smoking husbands - and reported an enhanced risk
when
husbands smoked. The study was based on a 14-year
prospective
epidemiology study involving some 9 1,000
non-smoking wives
out ofa total study group of265,000 adults.
Perhaps at this stage, one might merely note th~-r
the Japanese
as a race are very different from the Western
world in terms of
1 i%, ing envircrunent and social cultures - and
also i= their health
&position, eg stomach cancer is three times more
common than
lung cancer, whereas the reverse is true in the
UK.
At about the same time as the appearance of the
Japanese
srudy, details were 21S0 published in the
Internarioral Journal of
Cancer of a small retrospective case-control study
o3 women in
Greek hospitals. This claimed to show a greater
risk of lung cancer
among non-smoking wives ofsmokers, and that the
relative risk
increased with the number ofcigarettes smoked by
the husbands.
However, the numbers ofcases were small, and the
authors
themselves stated: "This study has ob,, ious
limitations and is
offiered principally to suggest that ffinher
investiption ofthis
issue should be pressed."
CD
co
17 41~.
BATCo document for Province of British Columbia 29
October 1999


These two reports f~orri Japan and Greece at-. in
complete
contrast, however, to [he Findings from a srudv
just carried out in
the US using the very much larger bank ofdaia from
the
American Cancer Sociey 'One Million' prospective
epidemiological study plus the (quarter million)
US Veremns
study. It is understood that a report of the work
is to be published
shortly in a leading US medical journal, and that
the report will
contain two important conclusions:
1. Over the period 1960-1972, there was no
material trend in lung
cancer mortality among middle-aged non-smokers
ofeither sex.
2. Non-smoking wives married to smoking husbands
do not have a
significantly higher risk of lung cancer compared
with non-
smoking wives married to non-smoking husbands.
Both findings are statistically inconsistent with
those reported
for Japan and Greece.
The subject is clearly far too complex to afford a
simplistic
interpretation of the reported negative findings
without a great
deal more research. But suffice it to say, the
evidence offiered. thus
far that ambient tobacco smoke causes disease in
healthy non-
smokers is highly suspc-.-.
Research
Despite the continuing controversy, I want to
assure you that
our company is very concerned about the issue of
smoking and
health, and that we are committed to furthering
scientific research
which will hopefully provide answers to some ofthe
questions
raised in this talk.
To that end, for many years we have been, and are
presently,
sponsoring high-quality research by independent
scientists on
various aspects ofthe smoldng issue. We also
undertake agpeaE
deal of related research ourselves, both at the
Central R&D
laboratorv at Southampton and in several ofthe
overseas
laboratories ofBAT Associate Companies.
And I wish to emphasise that BAT's involvement in
this area
is only part ofan industr-.--wide effort - which
must approach
L100 million per annurn throughout the world. I
feel that this
indusrr,,, commitment and effort is not widely
known or
recognised.
CD
18
CXD
BATCo document for Province of British Columbia 29
October 1999


Conclusion
For obvious reasons, it has not been possible for
me to cover
all the facets ofthe smoking and health
controversy. I hope,
however, that by covering the significant issues,
I have been able
to provide some insight into the opinions
ofcertain independent
scientists and researchers on the subject.
I have also tried to give some indication ofthe
way the tobacco
industry shares the concerns ofthose who are
working in this area,
and has tried to respond to those concerns (a) by
undertaking and
supporting research, and (b) by introduc' r g new
products in line
with medical and sciendfic opinions and with the
changing
demand ofconsumers.
Finally, I hope that I have succeeded in
illustrating the
complexity ofthe smoking and health issue. Complex
issues
seldom, ifever, have easv solutions, and the
smoking and health
issue is certainly no exception. However, the
tobacco indusm
appreciates these complexities, and is making
every responsiiie
effort to meet the concerns ofthose who work in
this area. We feel
that progress can only come from continued
research - and from
reasonable and open discussions about the issues.
Lim
CD
00
BATCo document for Province of British Columbia 29
October 1999