[ RadSafe ] Urine tests for uranium exposure typing?

[Steven Dapra]

         (My error -- earlier I posted this in the wrong thread.  This is 
the correct thread.)


On March 28 I wrote:

         Concerning the incidence of birth defects it [Kang et al.] says:

" . . . male Gulf veterans reported infants with likely birth defect(s) at 
nearly twice the rate of controls (adjusted OR = 1.94; 95% CI = 1.37-2.74). 
. . . Female Gulf veterans were almost three times more likely to have a 
child with a likely birth defect than non-Gulf females, after adjusting for 
other factors (adjusted OR = 2.97; 95% CI = 1.47-5.99)." (p. 507) (The 
controls were non-Gulf veterans, as noted on p. 505.)

         On March 31, James Salsman wrote:

         "P.S. To Steve: The Khan [sic] et al. (2001) ratios were 1.8 for
offspring of males, and 2.8 for offspring of females, not
1.9 and 2.9 -- do you know something I don't?"

MY REPLY:

         Yes, I know how to read primary source material.  The source 
is:  Annals of Epidemiology; 2001; 11:504-511.

         The figures I quoted are found in the text on page 507.  The text 
refers readers to Table 4 on page 508.  The Odds Ratios quoted above are 
the Adjusted ORs for "Likely defects" (under the heading of  "Outcomes").

         The OR's Salsman lists are the Adjusted OR's for "Moderate to 
severe" Outcomes.  Reading from Table 4 (Adjusted OR's at 95% CI):

         Males --- 1.78 (1.19-2.66)
         Females --- 2.80 (1.26-6.25)

         Note also that Salsman has (apparently) rounded 1.78 to 1.8, and 
that he has changed 2.80 to 2.8.

         Tell us something, James ---  and I would like a clearly stated 
answer.  Have you read this paper in the Annals of Epidemiology, or are you 
relying on a secondary source?  No mumbo jumbo.  A clearly stated answer 
will suffice, please and thank you.

Steven Dapra
sjd at swcp.com