Tobacco Documents Online

Y _?-NOU-1y93 13. 21 FROM TO 010:212=t0~~.~53'72' P.001/00,?- _ 96,C--c- 417.7 FACSIo:'iiLE TO: Dr John Luik FROM: Dr David Rowland DATE: 22 November 1993 cc: Dr Kim Davis (LWD) Mr John Lepere (CECCM) Mr John McDonald (RBH) Dr Chris Proctor (BAT) Mr Matt Winokur (PM) SUBJECT: EPA : Tails and C nfide _. ce_ Leyets John, As mentioned to you earlier, I was in Kansas City last week and took the opportunity to discuss this issue with Bernie O'Neill of SHB. Unfortunately it didn't prove possible to contact you with the time available at the end of last week. Bernie O'Neill and I went through the EPA document. The key to the whole thing is a short paragraph on page 5-2. This reads: "Throughout this chapter, one-tailed tests of significance (p = 0.05) are used, which increases the statistical ability (power) to detect an effect. The 90% confidence intervais used for the anaiysvs performed are consistent with the use of the one-tailed test. The justification for this usage is based on the a priori hypothesis (from the plausibility of a lung cancer effect documented In Chapters 3 and 4) that a positive association exists between exposure to ETS and lung cancer." So the EPA used a one-tailed test of significance at p = 0.05 (ie the 95% confidence level). To effect this, they performed a two-tailed test with 90% confidence intervals, which is equivalent. Hence the source of confusion perpetuated by Stanton Glantz and other commentators. Since getting back to the office this morning, I see that Kim Davis of Lovell, White and Durrant has come to a similar conclusion after considering pages 5-34/35 and speaking to his contacts in the US. Thus the 90% argument is out: what has to be attacked Is the EPA's assumption that they were justified in using a one-tailed test. This you have covered. Further, points on this aspect were summarised in Chris Proctor's memo'. Kindest personal regards, ....................... AO'['1{SIADtS IRTERNATICUNAi 4ERt'14ES'LdStiTtD DGNHA1,tPLACE Follows : 1 pageViLLAGE RUAD Dt1xIIAM I:XIIRiDGC k4IpDLESEX LID9SIILTELEPSO:tiE DENHAM 40995) $34949 TELGX'925997 FAX (0893) 834781, RF.GISTF.RF.D OFFICE: 15 NILLliTRfLT LONUON WIX'7FR- REl:1CTERFD iN F.NGLANn (No-20137601