Day 022 - 12 Sep 94 - Page 16

     
     1        some of these studies are particularly well conducted.
     2        They are prospective studies where they have tried to
     3        reduce or avoid bias creeping into the results.  If you
     4        have been involved with any kind of clinical trial or
     5        experiment, it is very easy for bias to creep into these
     6        things, so prospective studies are the ones where one gets
     7        the best information about possible linkages with diet and
     8        cancer.
     9
    10   Q.   Yes.
    11        A.  The previous document ----.
    12
    13   Q.   What I am saying is, as far as this pamphlet is concerned,
    14        they seem to be putting their cards on the table as
    15        regards fat intake as being particularly associated?
    16        A.  I think you have to read the whole thing in context
    17        because it says, "An association has been described
    18        between dietary factors".  In other words -- and then they
    19        go on to say, "Fat intake has been particularly associated
    20        with cancer of the breast" and so on.  So it is relating
         to the previous sentence, which is that an association has
    22        been described.  That is what they are commenting on.
    23        They then go on to say that the evidence has been derived
    24        from animal studies and epidemiological data.
    25
    26   Q.   So the evidence of links between fat intake and those
    27        cancers comes from animal studies and human
    28        epidemiological data?
    29        A.  That is right.
    30
    31   Q.   But there is another area of evidence which is what you
    32        called "prospective"?
    33        A.  Yes.
    34
    35   Q.   Prospective?
    36        A.  Prospective, yes.
    37
    38   Q.   Which is not conclusive; it is controversial.
    39
    40   MR. JUSTICE BELL:  Is that not epidemiological as well?
         A.  It is epidemiological as well.  All the data relating
    42        diet to cancer are epidemiological, which are
    43        observational data.
    44
    45   MR. MORRIS:  They seem pretty -- so their position is slightly
    46        more certain than yours then?
    47        A.  I do not think you can say that from this.  They are
    48        saying it has been reported that there is an association,
    49        and they say that the evidence to support this comes from
    50        animal studies and epidemiological data. 
    51 
    52   Q.   Right. 
    53        A.  They are just saying that there has been a suggestion
    54        there might be an association; they are not saying any
    55        more than that.
    56
    57   Q.   That may be one interpretation of it; I will not go into
    58        the other details.
    59
    60   MR. JUSTICE BELL:  The references they give are Doll in 1975