Day 254 - 22 May 96 - Page 33


     
     1        adult.
     2
     3   Q.   Basically, all they can say from this study, looking at the
     4        previous studies, is that, as in the conclusion, "In the
     5        context of the Western lifestyle lowering the total intake
     6        of fat in mid-life is unlikely to reduce the risk of breast
     7        cancer substantially".  That is about as much as they can
     8        say from this?
     9        A.  That is true.
    10
    11   Q.   Incidentally, in that table, where they adjusted the
    12        relative risks in the table 2 on page 358 for these
    13        variables, which included fibre intake, presumably that
    14        means that they consider that the level of fibre is
    15        something which affects the likelihood of getting breast
    16        cancer?
    17        A.  No, I do not think that is the case.  I think that we
    18        know from evidence which we have talked about in the past
    19        that fibre has been a factor which has been stated to
    20        influence the development of breast and bowel cancers, and
    21        so in a statistical analysis like that they try to look at
    22        other factors which may possibly be influential in the
    23        development of breast cancer.  Because, remember, they are
    24        looking only at the influence of fat or they are trying to
    25        look only at the influence of fat, and therefore they try
    26        to compensate for other factors which may have an
    27        influence.  That is the reason behind this statistical
    28        analysis.  They are not looking at all these other factors,
    29        what they are trying to do is to exclude them from the
    30        analysis so that they can look simply at fat as a risk
    31        factor in the development of breast cancer.
    32
    33   MR. JUSTICE BELL:  Would it be fair to say that they are trying
    34        to exclude the effect of anything which, looking at all the
    35        learning on the subject, might conceivably affect the
    36        results?
    37        A.  That is correct, yes.
    38
    39   Q.   And eliminate those apart from fat?
    40        A.  Yes.
    41
    42   Q.   They are not forming a judgment on whether it probably
    43        does?
    44        A.  Not at all.
    45
    46   Q.   Or there is a strong possibility.  It is anything which
    47        conceivably might confound the results?
    48        A.  Exactly, my Lord, because somebody else is looking at
    49        it.
    50 
    51   Q.   I see they have "High School graduation" there, for 
    52        instance? 
    53        A.  Yes.  Other people reading this paper, for example, if
    54        they had not corrected these, other variables might
    55        say, "Well, you have not only looked at fat but what about
    56        the influence of age at menarche for example?", so I think
    57        they have actually had to look at all of these other
    58        variables and try to get groups of patients, women, who
    59        were at similar risk in terms of each of these other
    60        variable factors and look at them.  It is a very complex,

Prev Next Index