Table VI
            Parameter Estimates for Selected Causes of Death for
          X-10/Y-12 Subcohort (N = 28,347) Using Main Effects ERR
                 Model with External Rates and Unadjusted
                    Cumulative Dose with Ten Year Lag

                         All         All     Circulatory    External
                       Causes      Cancer      System       Causes

  Term               Est   SE    Est   SE     Est   SE     Est   SE

 
 
A1(Age-52.5)/100    .35   .17   -.27   .38   -0.1   .26    .83   .51

  <1900           -28.8   6.2   -9.2  13.4  -19.5   8.7  -56.4  25.2
   1900 - 09      -17.0   4.8   -4.3  10.4   -9.3   6.9  -20.2  16.1
B2-1910 - 19      -20.9    .5  -10.1   9.4  -20.2   6.5  -27.8  14.1  
   1920 - 29      -30.2   5.0  -25.0  10.3  -31.8   7.5  -18.9  14.5  
   1930+          -63.1   7.8  -66.2  17.9  -71.3  14.4  -47.3  17.3 
 
S3-Nonmonthly        0     .      0     .      0     .     0    .   
  Mth vs NonMth   -42.6   4.1  -41.4   8.3  -37.6   5.8  -55.5  12.82

L3- 1 Year +         0     .      0     .      0     .     0    .   
    < 1 Year        9.9   4.3   11.3   9.1    1.3   6.3   24.9  12.3 

IG3,4-EN             0      .     0     .     0  .  0  .   
      EM           -2.4   3.7    4.3   7.5   -5.3   5.3   -1.5  12.5  
      NE            7.3   4.1    2.9   8.7    7.9   5.8   34.5  13.0 

F3-X-10              0     .     0      .      0     .      0   .   
   Y-12 vs X-10    12.4   4.2   15.3   8.7   10.8   6.2   12.2  12.3   
   Mult vs X-10     8.4   3.8    5.9   7.8    3.1   5.4    8.5  12.4  
 
D5-Dose             .31         1.45	       0             06 
   95% Cl7     (-0.16, 1.01) (0.15, 3.48)   (<0, 0.86)    (<0, 1.99)

    

  1. This term describes (and adjusts for) any systematic age-related difference 
in the external rates (U.S. White Males) and the study cohort in percent
per year units --- see Eq. 3 ----and the Appendix for further discussion.
    2. 


  2.  The estimates for each birth cohort are SMRs (in L% units) for the
reference level of each of the other terms in the model, e.g. for the 1910-19
cohort the estimated all cause SMR = exp (-.208) = .81.  The reference
group is nonmonthly workers employed at least 1 year at X-10 and who were
eligible but not monitored for internal radiation exposure.
    3. 


  3. For S, L, IG, and F the estimates are of relative risks in L% units, 
i.e. for all cancer the relative risk for monthly vs. nonmonthly workers 
is exp (-.414) = 0.66. 
    4. 


  4. IG-Internal Radiation: EN - eligible and not monitored, EM = eligible 
and monitored, and NE = not eligible for monitoring.
    5. 


  5. Dose-response for the ERR per Sv (see Eq. 4) which is equivalent to 
percent per 10 mSv (rem).
    6. 


  6. Zero estimates indicates that likelihood was maximized at value
that  would lead to negative relative risk estimate.
    7. 


  7. Likelihood ratio based confidence interval for ERR.
    8.