Day 022 - 12 Sep 94 - Page 17


     
     1        and the National Research Council paper in 1982?
     2        A.  Which are old papers.
     3
     4   Q.   I was going to ask:  They qualify as old, do they?
     5        A.  I think so.  One of the problems in a lot of the
     6        studies which were originally carried out was that some of
     7        them were analyses of papers which were already in
     8        existence.  If one reads a lot of the papers on trying to
     9        estimate what diet is, going back to the 1970s, the
    10        authors of these papers actually state clearly that the
    11        methodology was actually in its infancy at that time.  And
    12        there is no doubt about it that in recent years
    13        methodology has improved and also people have begun to
    14        realise the problems of bias that might creep in with the
    15        so-called case control studies; in other words, the
    16        retrospective analyses where you look at a group of people
    17        who have cancer, and you try to match them with a group of
    18        people who do not have cancer.  These are studies which
    19        are particularly open to bias.
    20
    21        Therefore, people have tried to run the so-called
    22        prospective epidemiological studies where they take two
    23        groups of people who are healthy at the time the study
    24        starts, and then follow them through the years to see what
    25        sorts of illnesses these people develop.  They, therefore,
    26        look at their dietary intakes at the beginning of the
    27        studies.  This means that the diet these people describe
    28        is much less likely to be influenced by factors such as
    29        illness or enthusiasm, which is a danger of the case
    30        control studies, that the control group often are
    31        enthusiastic people.  The other difficulty with controlled
    32        studies is that it is often done on a mail shot basis.
    33        They send a questionnaire around to several hundreds or
    34        thousands of people sometimes.  There is a poor
    35        recruitment rate.  In other words, in some studies, only
    36        55 per cent of the people to whom a questionnaire is sent
    37        actually respond.
    38
    39        This immediately raises doubts about the validity of the
    40        study of that nature.
    41
    42   MR. MORRIS:  I think we will come later to specific methodology
    43        and have a look at them and compare.  I am just trying to
    44        look at what, you know, responsible bodies are saying.  If
    45        we go on to the ones I have given to you by hand?  You can
    46        put the Grey Book away now.  Thank you very much.  If we
    47        look at the one that is headed:  "NATO Advanced Research
    48        Workshop on Advanced Technologies".  NATO, presumably, is
    49        North Atlantic Treaty Organisation.  Something intrigued
    50        me -- do they have these kinds of conferences that look 
    51        into nutritional matters?  I was surprised when I saw 
    52        that. 
    53        A.  I have to say I was not aware of that, but -----
    54
    55   Q.   All right. So some countries ----?
    56        A.  I mean, nutrition is obviously something which is of
    57        great interest to a lot of people and a number of
    58        organisations are quite likely to have the facilities
    59        therefore to organise a congress such as NATO may have
    60        done.

Prev Next Index