Tobacco Documents Online

N it~0531-)047 D R A F T- 3/26/63 These data show narked differences in recorded morta].ity rates from lung cancer between cities of different sizes, between cities of the saate aize, and between cities in different regims. 2here is no evidence of a sipificant correlation between per capita euwk:ng and these contrasting lang cancer mortality rates. 'Phere are, in fact, wide 8iscrepancxes. (Nmos, D.B. 1957) Qot onl3* are there wide and inoonaistent differences between morta7.ity rates I~rom tots.1 lung cancer recos:d.ed as both "primery" and "unapecilied," t4t.we are elso trid,e differences in the relation of recorded incidence of the two categories to one another in different cities. All these differences are uaexpisined. Mat do they mesn in terms of acciracy and reliability of pdeatb certificate" data4 Nhat do they unn in terms of nccuracy andL reliability of diagaosis sad 8if'Pesentiatf.oa between prisar;r and metastatic lung Lm,mrsY Mst do they mean in relation to raW possible association with any specific eavircnmentaall factor which mss}r be selected for atudy?