Day 022 - 12 Sep 94 - Page 50
1 existing recommendations for substituting fish and chicken
2 for meats high in fat".
3
4 Without going into detail, what is your understanding of
5 the conclusions and the methodology of this study?
6 A. I would accept what he says; these prospective data
7 provide evidence for the hypothesis that a high intake of
8 animal fat increases the risk of colon cancer, but it does
9 not say that it definitely proves that a high intake of
10 fat is responsible for increasing the risk of colon
11 cancer.
12
13 Q. No cohort study would, would it?
14 A. No, but, as I say, what one is looking for is a
15 consistency of findings. Now, if we then go on and find
16 further studies which consistently reproduce this, then
17 one begins to think, well, there may be something in this.
18
19 Q. But this study, let me for the record say, involved 88,750
20 women, 34 to 59 years old, without a history of cancer
21 etc. They completed a dietary questionnaire in 1980 and
22 they have been followed up since, basically. This
23 original report is 1990. So, would you say this is one of
24 the most well-managed and significant cohort studies?
25 A. Well, it certainly is a very important cohort study,
26 yes.
27
28 Q. If you just have the results of the study: "After an
29 adjustment for total energy intake, animal fat was
30 positively associated with the risk of colon cancer."
31 I will not go into it in detail. We might as well get it
32 on the record: "P for trim O.01. The relative risk for
33 the highest as compared with the lowest quintile was 1.89
34 at a 95 per cent confidence interval of 1.13 to 3.15. No
35 association was found for vegetable fat."
36
37 Does that say something to you when there is one
38 particular kind of fat which shows an association and
39 another does not? Does that say something about total
40 fat?
41 A. One of the problems with all of these things is
42 actually accurately ascertaining what people are eating.
43 You cannot just draw total conclusions from one
44 study. For example, if you pass on to 19, 764,343 adults
45 were examined in a prospective study and they did not find
46 a relationship. So, I am only making the point that -----
47
48 Q. We are trying to look at specific studies because specific
49 studies have been referred to ---
50 A. Yes, but I have referred to this one as well.
51
52 Q. -- In examination.
53 A. In my reference to it I made the point that the
54 problem is that there not a consistency of reporting of
55 that finding.
56
57 Q. But there may be reasons for that?
58 A. There may be reasons for that.
59
60 MR. JUSTICE BELL: Can I understand the difference between tab
