Day 254 - 22 May 96 - Page 27
1 in energy adjusted nutrients intake.
2 A. They actually do not. What they say is, look at the
3 bottom of page 358, overall relative risks, in table 2 we
4 show the pooled quintile specific relative risk of breast
5 cancer as compared with the lowest quintile. So quintile
6 No. 1, which is the reference, is the lowest quintile of
7 intake.
8
9 Q. Yes, but we do not know what the intake was.
10 A. No, obviously they divided the intake according to a
11 range, and quintile No 5 is the quintile with the highest
12 fat intake, quintile No. 1 is the quintile with the lowest
13 fat intake.
14
15 Q. Yes, but we do not know what the range is or the -- or what
16 the range within the quintiles either.
17 A. What they have said is, quintile No. 1 is where the
18 energy is less than 20 per cent of total intake.
19
20 Q. Where do you get that?
21 A. That is in the summary. I am just trying to find the
22 data for you.
23
24 MR. JUSTICE BELL: What I suggest is we take the 5-minute break
25 there. It occurs to me that there might be an explanation
26 in that the figures might vary quite considerably, not
27 necessarily within the United States but between the
28 different countries, or might vary to some extent. They
29 must have had that information, it is a question of how
30 much you actually put into your paper at the end of the
31 day, I suppose. But if you do not mind, while everyone
32 else has a 5-minute break, having a look through and see if
33 we can find the answer, Dr. Arnott.
34
35 (Short adjournment)
36
37 MR. JUSTICE BELL: Can you find an answer to it?
38 A. Yes, my Lord. Basically, what they appear to have done
39 they have not specified exactly what these quintiles
40 represent, except that quintile 1 is the quintile with the
41 lowest intake and they have taken that as the reference
42 level. But they have used a variable energy intake
43 relating the intake to 1,600 calories, and I think this is
44 obviously a statistical analysis which they have not fully
45 explained in how they have divided the quintiles up. But
46 they have essentially based it on a number of factors, like
47 nutrients density they quote, which is the ratio of energy
48 from fat, to obtain the total intake since this is often
49 used, they quote, to make dietary recommendations.
50
51 That is as much information as they actually give in the
52 paper regarding how they have divided the groups of
53 patients up into these five categories.
54
55 Q. Might I suggest that you put to Dr. Arnott the conclusion
56 you are aiming at because he may agree with me if it is
57 related to whether, for instance, there might have been a
58 very great or a very small difference or we just cannot
59 tell from the report itself.
60
