Document 3Q9yK55OQn6XBejDN4RxYkXyx

Fe b r u ar y 22, 1962 Memo to From : Su b ject: D r. Robert A. Keh oe T. Ster lin g A n a ly sis of C a lifo r n ia T M L Study The ch an ges ob ser ved in the excretion of lead an d in the blood pictu r e of w o r k e r s (who sw itch ed fr o m h an dlin g T E L con tain in g p r o d u ct s to th ose con tain in g TM L) a r e of such con sist en cy an d m agn itude a s to preclu de an y ch an ce fa c t o r s. Th e pattern of r e su lt s would or d in ar ily for ce on ly one con clu sion on the in v est igat o r ; n am ely, th at un derlyin g ch an ges in ph ysiology o r body ch em istr y h ad taken p lace durin g the period su b jects h an dled TM L. It is tru e th at the data w ere collected in a som ew h at h ap h azar d an d cav alie r fash ion . H ow ever, th e m ost im portan t fin din gs a r e b ased on lab o r at o r y an aly se s of u rin e an d blood sam p les and m ay be above reproach . It is a gr e at m isfortu n e th at th e m ed ical follow -up data turn ed out to be u se le ss. Th e r e su lt s availab le now would m ake it ap p ear d e sir ab le to com plete the collection of a ll the data an d do it in the form origin ally plan n ed a s w ell a s to exten d the follow -up on the lab or ator y an aly se s fo r an addition al year. Lead Le v e ls in Urin e an d Blood 1. D ata on le a d in b lood an d u r in e w er e a lm o st com p let e. F iv e u r in e sam p le s w ere obtain ed durin g th e y e ar of stu dy wh ile two blood sam p les w ere taken , one in the begin n in g the oth er at the end of the study. Th ese data a r e p r esen t ed in a se r ie s of t ab le s a t the end of th is m em o. Con secu tive exam in ation s ar e iden tified a s "Rou n d s" one to fiv e . Th e n u m ber o f sa m p le s a r e given in th e secon d colum n of each tab le. St at ist ical evalu ation of d iffer en ces ob ser ved betw een con secu tive rou n ds a r e b ase d on A n aly se s of Varian ce. The statist ical an aly sis is m ade extrem ely com plex by the fact th at, while m ost m e a su m w ere m ade repeated ly on the sam e su b jects, not the sam e n um ber of su b jects w as tested each roun d. Th e an aly se s a r e n ot r ep r od u ced h e r e . When th e p r o b ab ilit y of fin din g, by ch an ce, a v ar iat ion am on g m ean s a s la r ge a s th e one V,n U v r UO 4 -J N9602