Day 017 - 25 Jul 94 - Page 51
1 A. I think this is a problem with all of these studies.
We have talked about this on several occasions. There is
2 the problem of trying to devise the ideal questionnaire to
ascertain what people are really eating and its relevance.
3
Q. They say: Finally, CPS 2 participants" whatever those "are
4 on average more educated and affluent than the US
population as a whole. Greater access to medical care and
5 screening may contribute to their lower mortality rates
from colon cancer. While these differences may influence
6 the ability to generalise our findings to groups of low
socioeconomic status, they are unlikely to compromise
7 internal validity."
A. Yes. I think what I would take from this study is
8 that it is a study which originally looked at 764,343
people which is an extremely large study. If you speak to
9 statisticians about any assessment of cancer treatment or
cancer risk, the larger the number of people that you have
10 the less likely are you to have errors which may creep in
because of possible chance. We were discussing this
11 morning about statistics. So the greater the size of the
study the less are the chances that the result will appear
12 purely chance.
13 Q. In the lights of that I would, if you will forgive me,
just like, I think you have answered the question I was
14 going to ask but nonetheless I would like to read the
abstract, if I may. "Background - diet, physical obesity,
15 asprin use and family history may all modify the risk of
colon cancer". Dr. Arnott, the word "modify" in this
16 context does not mean reduce, does it? It does not mean
moderate?
17 A. No, because they are talking about obesity which may
in fact increase the risk. Family history of course may
18 be relevant in terms of increasing the risks.
19 Q. It means qualify it one way or the other?
A. Absolutely, yes.
20
Q. "Modify the risk of colon cancer, but few epidemiologic
21 studies are large enough to examine these factors
simultaneously. Purpose -- we have respectively assessed
22 the relationship of diet and other factors to the risk of
fatal colon cancer. Methods -- using data from cancer
23 prevention study II an ongoing prospective mortality
study, we studied" as you said, Dr. Arnott "764,343
24 adults who in 1982 completed a questionnaire on diet and
other risk factors and did not report cancer or other
25 major illness. We assessed mortality through August 1988
and identified 1150 deaths from colon cancer.
26 Multivariated analyses were used to compare these case
persons with 5,746 matched control subjects drawn from the
27 cohort. Results -- risk of fatal colon cancer decreased
with more frequent consumption of vegetables and high
28 fibre grains. P for trend equals .031 and .0012 in women.
The relative risk RR for the highest versus quintile of
29 vegetable intake was .76 in men 95 per cent confidence
interval equals .57 to 1.02, and .62 in women 95 per cent
30 CR equals .45 to .86.
