Cancer mortality amongelectric utility workers exposed topolychlorinated biphenyls

Objectives-To assesswhetherexcess mortalityfrom cancer,malignant melanomaoftheskin, andcancers ofthe brainandliver inparticular, isassociated withlongtermoccupational exposure to polychlorinated biphenyls (PCBs). Methods-Anepidemiological studyof mortality wasconducted among138905 men employedforatleastsixmonths between1950and1986atfiveelectrical powercompanies intheUnitedStates. Exposures wereassessed bypanels composedofworkers, hygienists, andmanagersateachcompany, whoconsidered tasks performed byworkers in28jobcategories andestimated weekly exposures inhours foreachjob. Poisson regression wasused toexaminemortality inrelation toexposuretoelectrical insulating fluids containingPCBs,controlling fordemographic andoccupational factors. Results-Neither allcausenortotal cancermortality wasrelated tocumulative exposureto PCB insulating fluids. Mortality frommalignant melanomaincreased withexposure; rateratios (RRs) relative tounexposed men formelanoma were1.23(95%confidence interval (95% CI)0.56to2.52), 1.71(0.68 to4.28) and 1.93(0.52 to7.14)formen with 2000-10 000,and>10000hoursofcumulative exposure toPCB insulating fluids, respectively, without consideration oflatency.Laggingexposureby 20 years yielded RRsof1.29(0.76 to2.18), 2.56(1.09 to5.97), and4.81(1.49 to15.50) forthe same exposurelevels. Mortality from braincancerwas modestlyincreased amongmen with 2000-10 000 hoursexposure (RR1.79, 95%CI0.81to 3.95), buttherewerenodeaths frombrain canceramongthemosthighly exposed men.A lagoffiveyearsyielded slightly increased RRs.Mortality fromliver cancerwasnotassociated withexposure to PCB insulating fluids. Conclusions-This studywaslarger and provided more detailed information on exposure thanpastinvestigations ofworkersexposed toPCBs.Theresults suggest thatPCBs causecancer, withmalignant melanomabeingofparticular concernin