REVIEW 719  CONFIDENTIA
    ~ub _je_c_-_ - ne; _f8 d
    "Urinary cotinine as a measure of passive smoke exposure
    in asthmatic children"
    C J Ogbo:-., A K Duggan and C de Angelis
    Clinical Pediatrics (1994), 33, 220-226
    This study concerns 56 children with a prior diagnosis of asthma for
    whom urinary cotinine samples and parental reports of ETS exposure were
    collected firstly at the cime of an acute asz*.-uma attack and then, a few
    weeks later, when the child was symptom free. Ninecy-three children had
    originally been enrolled inco the study but follow-up information had not
    been collected for 37 of these. The children who did not complette the
    study were not s*gnificanzly different frcm those who did in zerms of
    various relevant variables studied.
    There were four main findings of the study:
    (1) There was very little difference in urinary cotinine or in urinary
    cotinine/creacinine ratio between the acute and the we-'- visics
    Is (Table 2). This finding was consisten: with ETS exposure not being
    *n%olved in the exacerbation of exis:ing asthma, a conclusion :he
    authors do not make in their summary.
    (ii) There was generally no significant difference between the acute and
    t'-e well visits in reported levels of ZTS exposure ("'able 3). The
    exception was self-reported amount (none, a little, some, a lot). Ul
    CD
    w~-ere at che acute visit amouncs were significantly higher than at CD
    co
    C\
    00
    .BATCo document for Legal Services : Health Canada 19 October 1999

    




    -2.
    the well visit. The authors believe t1his may be an artefact due to
    reduced re-porting of amount on repeated incerviews. Certainly it is
    difficult to see how it can be real, bearing in m1nd the lack of
    difference in cot-nine. The authors' explanation highlighis a
    weakness o! the sz-ady design in that the acuce visit always
    occurred before the well vis-it, thus lead~ng to an essential-
    confounding between the acute-well dif'ference and the first-second
    inter-4iew difference. A cross-over szudy design would have been
    preferable-
    (iii) There was quite a highly significant correlation between parental
    report and urzina-_.- cotinine. I have r--nged results in Table 4 which
    seem inconsistent with this. This appears to be because of a
    mislabelling of the results for urinary cozinine for "was child
    exposed?". That there is a sign . ficant correlation is totally to be
    expected - had there not been their quescionnaire or cotinine
    methods must have been up the pole. "tore relevant would have been
    to present results showing the frequency of major inconsistencies -
    high cotir.,-nes in "unexposed" children and vice versa.
    (iv) The authors, in their discussion, highlight and criticize the
    "high" levels of smoking in these families with asthmatic children.
    It should be pointed out that the "high" levels are uncontrolled
    observaticns - there is no valid population of non-asthmacics wic"n
    which to compare frequencies of household ETS exposure. The cases
    studied are virtually all black, inner-city children.
    (il
    CD
    CD
    Co
    CI\
    Co
    r_%.~
    Co
    CD
    BATCo document for Legal Services: Health Canada 19 October 1999

    




    ,he most interesting result of this study is that it presents
    find,--igs, albeit based on a rather small sample, showing no association
    of 7-.S exposure with exacerbation of asthma.
    P N '-ee
    2. S. --~- -
    0
    40
    Ln
    C>
    CD
    CL')
    c'II%
    Co
    cc-
    BATCo document for Legal Services : Health Canada 19 October 1999