Day 030 - 03 Oct 94 - Page 49
1 works.
2
3 I will not go into too much detail on this, but what they
4 did was a sophisticated epidemiological exercise which
5 started off quite simply. What they did was to compare
6 the rates of death in cancer registries where the rates of
7 death from specific cancers, the lowest to the reverse,
8 those cancers registries where the rates of death were
9 highest and, in common with other research workers whose
10 conclusions chose are regarded as reliable, they said that
11 it is fair to say that environmental factors account for
12 80 to 90 per cent of all cancer deaths, that is to say,
13 the pure genetic connection is not very substantial in the
14 case of cancer.
15
16 What then becomes more sophisticated is how do you
17 calculate within the environmental causes the dietary
18 factor? That is really the subject of this large book
19 which you may be happy to know is not in court today, but
20 the conclusion they came to was that alcohol reliably
21 could be said to cause something like three or four per
22 cent of all cancer death. That was a pretty hard figure.
23 Again, I think that has been referred to in evidence
24 earlier in this case; whereas the diet as a whole they
25 proposed an extremely broad range of between ten and 70
26 per cent, but had some reason to believe that roughly the
27 mean figure of 35 per cent or roughly over a third of all
28 deaths from cancer could be attributed to diet.
29
30 That is a negative way of putting it. You can put it
31 positively of course and, putting it crudely, what they
32 were suggesting is that if you eat the wrong kind of diet,
33 you are going to increase your chance -- your risk of many
34 cancers. If you eat the right kind of diet, you thereby
35 decrease your risk.
36
37 The extent of the risk depends on the cancer. In the case
38 of lung cancer, for example, there is good evidence (and
39 was then) that vegetables and fruit or, rather, particular
40 chemical constituents of vegetables and fruit are powerful
41 protectors against all cancers, including lung cancer, but
42 it is not reckoned that amongst smokers you are going to
43 reduce your chances of lung cancer very much if you eat
44 lots of fruit and veg or if you drink orange juice or
45 whatever; whereas stomach cancer -- not mentioned so far
46 today -- and colon cancer and, to a lesser extent, breast
47 cancer and also prostate cancer are regarded as very
48 heavily influenced by diet of the ones I have mentioned,
49 stomach cancer and colon cancer most of all.
50
51 Q. So this approach of identifying percentages of cancers
52 that could be reduced with dietary changes, has that
53 approach continued -- is there a consensus emerging about
54 that?
55 A. There is not consensus -- it depends what you mean by
56 "consensus". Yes, there is consensus if you mean that
57 Doll and Peto's estimate or, as they would call it, I
58 think, guesstimate of this very wide range is constantly
59 cited; it is cited by government, it has been referred to
60 by the chief medical officer in addresses he has given,
