WPC 2 ZBn Xx6X@7X@Bibliogrphy66lBibliographyN=(N{*Nx{$E Doc Init{66lInitialize Document Style{*Nx{$E    I. 1. A. a.(1)(a) i) a)DocumentҲa1DocumentE+N xDocument StyleoN=(N{!*NF *  ׃  2qea2DocumentE+N xDocument StyleoN=(N{!*N*    a3DocumentE+N xDocument StyleoN=(N{!*N0     a4DocumentE+N xDocument StyleoN=(N{!*N   . a5DocumentE+N xDocument StyleoN=(N{!*N  2Ke p p ca6DocumentE+N xDocument StyleoN=(N{!*N  a7DocumentE+N xDocument StyleoN=(N{!*N ` ` ` a8DocumentE+N xDocument StyleoN=(N{!*N ` ` ` Tech Init"66lInitialize Technical Style{"*Nx{$E  1 .1 .1 .1 .1 .1 .1 .1 Technical2 } -^ a1Technical+N xTechnical Document StyleN=(N{#*N 4!     a2Technical+N xTechnical Document StyleN=(N{#*N *    a3Technical+N xTechnical Document StyleN=(N{#*N'   a4Technical+N xTechnical Document StyleN=(N{#*N&   28 $  .  a5Technical+N xTechnical Document StyleN=(N{#*N&   . a6Technical+N xTechnical Document StyleN=(N{#*N&!"  . a7Technical+N xTechnical Document StyleN=(N{#*N&#$  . a8Technical+N xTechnical Document StyleN=(N{#*N&%&  . 2o4j 0Pleading{$66lHeader for numbered pleading paper*Nx{$E'(   ,#x6X@7X@# X  y*dddyy*dddy HH1 HH2 HH3 HH4 HH5 HH6 HH7 HH8 HH9 H10 H11 H12 H13 H14 H15 H16 H17 H18 H19 H20 H21 H22 H23 H24 H25 H26 H27 H28   Ӽa1Right Par+N xRight-Aligned Paragraph NumbersN=(No*N8)*@   a2Right Par+N xRight-Aligned Paragraph NumbersN=(No*NA+,@` `  ` ` ` a3Right Par+N xRight-Aligned Paragraph NumbersN=(No*NJ-.` ` @  ` `  2Na4Right Par+N xRight-Aligned Paragraph NumbersN=(No*NS/0` `  @  a5Right Par+N xRight-Aligned Paragraph NumbersN=(No*N\12` `  @hh# hhh a6Right Par+N xRight-Aligned Paragraph NumbersN=(No*Ne34` `  hh#@( hh# a7Right Par+N xRight-Aligned Paragraph NumbersN=(No*Nn56` `  hh#(@- ( 2 0a8Right Par+N xRight-Aligned Paragraph NumbersN=(No*Nw78` `  hh#(-@pp2 -ppp Ыx6X@7X@<6X9`("Courier 10cpiXx6X@7X@<6X9`("Courier 10cpiX`\  PQP)\  ` TmsRmn 10pt (F)`\  PQP)\  ` TmsRmn 10pt (F)`\  PQP)\  ` TmsRmn 10pt (F)`\  PQP)\  ` TmsRmn 10pt (F)23|6#`\  PQP#Њ &APPENDIX B METHOD FOR CORRECTING RELATIVE RISK FFOR SMOKER MISCLASSIFICATION ă   S!#`\  PQP#p)BSX` hp x (#%'0*,.8135@8:@C 1Additive model, Lee's Table 3. His additive model was chosen because it is similar to our additive model for passive smoking effects on smokers. 2EPA's misclassification factors developed in Section B.4., namely, 1.09% of current regular smokers, 24.2% of current occasional smokers, and 11.7% of exsmokers, when weighted for their respective prevalence and relative risk, are equivalent to about 1.5% of average selfreported eversmokers. EPA used these rates for all studies except FONT, which is a special case. Lee used 2.0% of selfreported eversmokers for all studies. 3Lee used 49% eversmokers vs. our 43% based on the case age distribution. Our misclassification rates for current smokers, m2 (4.3%) and m3 (0%), were developed as noted in Section B4, except that 2 out of 3.5 expected misclassified occasional smokers had been eliminated by cotinine tests, leaving 1.5/35 = 0.043 for m2 in this study. For m1, we assumed that it was the same percentage (41%) of 0.117 as 10% was of 24.2% for m2. `&.Q(continued on the following page)    `  hp x(08"$&@)+-H024P79@C#`\  PQP# B6   Table B2. (continued) 4A female smoker risk of 3.58 (U.S. DHHS, 1986) and smoker prevalence of 22% (Hammond, 1966) for age distribution of cases. Lee used 8.0 and 49%. 5EPA estimates a smoker risk of 6 and a smoking prevalence of 34% for the time period 197181 vs. Lee's values of 8.0 and 49%. 6The main difference is in the assumed smoker misclassification rate, but Lee's assumption of 49% smoking prevalence vs. 42% assumed by EPA increases the bias estimate from 1.09 to 1.15. 7Lee assumed 58% smoking prevalence vs. 47%, which EPA got from the paper itself. Lee assumed a lower smoker risk (9.5) vs. EPA's 12.4; the effect of this was offset by Lee's higher misclassification rates.    3'  8 Table B3. Misclassification of female current smokers h ddx . #" ddx . #"h "p@@@@Pp@@@@P" Selfreported smoking status number"SEEEEUSEEEEU" StudyCotinine "level1ă  Number  Never Former Current""Coultas et al. (1988)2!1030 ,"30+ F"AllX` hF~ap x (#%7 5 3873 8 79 184""Cummings (1990)3!1030 ,"30+ F"All0 2 2251 0 143 116""Pierce et al. (1987)4!1030 ,"30+ F"All9 3 2324 3 79 167""Subtotal!1030 ,"30+ F"All16 10 8448 11 301(67% never) (48% never) 467""Lee (1986)5!1030 ,"30+ F"All3 3 3332 2 125 256""Haddow et al. (1986)5!1030 ,"30+ F"All1 0 1741 1 58 64""Haddow et al. (1988)5!1030 ,"30+ F"All15 1 1,1287 1 380 503""Riboli (1991)5,6ĩU.S.7!1030 ,"30+ F"All1 0 2240 0 81 143""Riboli (1991)5,6ĩEast Asia8!1030 ,"30+ F"All1 1 3251 0 25 77""Riboli (1991)5,6ĩGreece9!1030 ,"30+ F"All0 0 960 0 5 15""Total!1030 ,"30+ F"All37 15 3,12419 15 975 1,525""Proportion misclassified10 1030% 30+24.2% 1.09%12.4% 1.09% X` hp x (#%'0*,.8135@8: