Document 8VG0QEGw0VD7nVq25K95k0pwk

--f 1208 Update on 2,443 Exposure Bloemenet An Update of Mortality Among Chemical Workers Potentially Exposed to the Herbicide 2It,s4-DDeicrihvlaotriovepshenoxyacetic Acid and 1- I. q- -i: t" I" i Louis J. Bloemen, MPH Jack S. Mandel, PhD Gregory G. Bond, PhD Alex F. Pollock, MS Epidemiologc studies Of populations uith occupational exposure to phen. oxy herbicides have generated incon. sviasrtieonutsecvaidnecnercse, oinfcalusdsoincgiatsioofnt-s- wia sarcomas and the malignant lymph& JC .: I' j(eeL. ' Robert P.Vitek, MD Ralph R. Cook, MD, MPH Four .ican 01 cldditroniii i ~ i ~ ~ r r i i i ~ ! ~ ~th.r~uioigIhl 1l v~ ~S ~u~~ru~wipi i~i r r d t i v a previoiisly studied cohorr oj-8 78 chemical workers who were poienria1l.v exposed io 2.4-dichl0r0pheno.~yacetiaccid 12.4-0)and iis derivaiives beiween I945 and 1983. Observed morialiiy was compared wiih expecred levels based on dearh rarer rfrhe ITpqvclaririn und of 36 PO4 "une.yporc4" w r k e r ~/him iht' ';dItle l ~ l U f l l l ~ i c ~ f l t ~!Il)lcl'ay[li)Ii. . ~ ~ I J ~ i - f i d ~'ik/!l t~ l? l p h t l l t l U .vffL, )Id\ J parriLiiiar .tows q/iirc siuav oecmisr qt u mqgestea ussoi'iation w t i -'..I-D e.vposiire in Tome case-conrrol siiidies. For [he [oral obsenairon period. ihe siandardked rnorraliiy raiios for all causes and for malignani neoplasms mas.'-2b Recent reviews have con. IC cluded that there is currently not nt enough coherent evidence to confirm i iu that the phenoxy herbicides in gen- ' (rL eral. or 2.4dichlorophenoryaceti~ I , \ acid (2.4-D) in panicular. represent a ut carcinogenic hazard to humans.:'-B I ler .Additional follow-up of cohorts was suggested to help clarify the issue. In !Y88.two of the authors ( G G B and RRC) first reported a study" of Ithe cause-specilk mortality expen- ence of a cohort of 878 employees i' ;"~ T T who had been engaged in the manu- were 9: and 91. respecrively. .4nalyses using [he inlernal comparison group I,ielded virrua1l.v identical resitlis. The inirial sludy had found iwo dearhs $om .VHL. borh of which occurred under circumstances fie. short latency and modrsr l~.vpcisiirc:iI( hich made 11 less plaitsible rho! ihes w r e rrlu[t?dro 2.4-D exposure. .Vo nen' deaths from .VHL were ohseried in ihe extended facture. formulation, or p a c h u g of 2.4-D and its amines and esten. The current study provides 4 y e a n of additional mortality follow-up of this cohort through 1986. siqnrlicanr e.vcess lsrandardized rnorra1ir.v ratio. 196: 95 % contidence inrenal 24 10 '081 lbr [he iota1 r~hrenaiiiinperiod .4nal)ses bt, prodiictiiin u r t w u r d b?, no ditrrrent rncasures ot' e.vposure. iornhined i ~ lnhv dilfiwnt a p proaches 10 account t o r iatency. did not shorn,p a ~ e r n ssuggesiive c?Tucausal relaiionship beiween cvposure ro 2,4-D or iis derivaiives and any pariicitlar cause ot'deaih. From D o w Benelux \ \ Biomedical Research. 4530 14 Temeuzen. The \etherlands iDr Bloemcnl. The Universitv of Minnesota School o f Pubbc Hwlth. h w s i o n of Enbironrnenral and Occupational Health. Minneapolis. Minnesota (DrMandel). The D o w Chemical Cornpan\ Epidemiology Midland. Michigan (Dr Bond and Dr Cook]. Dow Chemical USA Michigan Dnision. Medical and industnal Hvgene Depmrncnts Midland. Michigan tDr Pollock DT \.itek) W96- 1736!93/35 I ? - I:08503 oO/O Copbnghi F bv 4mcncm Collcge of Occupational 3nd En\ironmenlal Medicme December 31. 1982 were identified from plant rosters Ernplo>ees' wok - I - ta11 Volume 35. Number 12, December 1993 n records were compiled and sure: duration and cumulative dose. b assignments were classified .Analysesby duration of exposure con- ee categories of potential time- sidered latency categorized in 5-year average exposure (ie. less intenals. Cumulative dose was ana- I. 0. I to I .O. or more than I .O lyzed with either no latency period or umulative exposures were laggrng exposures by a fixed interval each worker by multi- of I5 years.): Examination for trends e spent on each job by across standardized mortality ratios gned to its respective (SMRs)was done using a x' test with average value (ie. 0.05. 1 degree of freedom." 0.5. or 5 mg/m') and summing across Using a method of direct standard- ad J o b . ization." monalit! among [he 2.J-D -Vital status was successfully ascer- cohon was also compared with that . L x d fd:~I! CJhon membcrs t!ir"@ Jf fithe: \ \ a r k s \\ ha w r e also 3C- #J86 using records of the company. tively employed at this manuiactunng id Secunt! administration. Iocstion at an) time dunng I942 Sattonal Death Index. Death through 1982. Adjustment was made ta were coded according to for age. calendar year. and pay status he International Classifica- (re. hourly or salary). The internal sews' IICD) revision in ef- comparison group differed slightly e time of death. Cause of from the one used for the original death d e s mere then grouped ac- stud) as some temporar) workers w d i n g IO the catcgones o i the 3tn uere e\cluded. JS \+erefound? , A d - ieI ICD Roision for analyses using a ified life-table program.'" To cd- te expected numbers of deaths brn non- Hodgki n 's I y m phoma 1 (F;HL). US rates for 1940 to 19'9 I -.-o,.-a bu;nrd from the S 3 t i o n ~Cl enI *r for Health Statistics. Exact. test- u s irom 3 wtellitr operstion Results Dunng the additiond 1 \ears of idlo\\-up there occurred f o b c r d s ~ t h s than were expected from all causes -&e& i w o - d e d contidencs intenals dnd from total canker. uith :he !arter .?57 CI)were calculated according to difference achie\ing statistical signifi- Ibe method of Miettinen using the cance (Table I ) Oberall. for the totcll of Rothman and Boice.:' follow-up p e n 4 the all causes SMR were done by production u a s 92 (95% C1. 77 to 109).and the by two measures of expo- total cancer SMR uas 91 (95% CI. 61 1209 to 131). For no cause of death did employees experience a statistically significant increased risk compared to the US population, or to other. "unexposed" workers from the same location (Table 2). No new deaths from NHL were identified in the additional follow-up period. Overall, there were two deaths from NHL compared with approximately one expected (SMR. 196: 9 5 5 CI. 24 to 208). When allowance was made for a !?-year minimum latency penod. there were only small changes in the ShiRs. an exception was the SMR for NHL. which became zero because. as discussed in the original publication. both cases occurred within a relatively shon time after first exposure ( 3 and 10.5 years. respectively). .Analysesof the four individual produr:ion plmrs x m i n u e d ta s h w a statistic all^ signiticantl! clc~ation (SMR. 351: 955 C1. 116 to 833) in the categor) of "cancer of other and unspecified sites" among employees *ho worked in one of the four plants reierrrd to as the "2.J-D Phnt." hut rebealed no new findings of note. Ana1)sss b) duration of exposure or by estimated cumulative dose of exposure did not show any statistically significant trends. Tabla summarizing these anal!ses are available from the authors upon request. '- I- o TABLE1 >erved and Expected Deaths.' SMRs. and 95% Confidence intervals for Selected Causes of Death. Cohorl of 878 Workers "3250C :o 2 4-0 Cause of Death (ICDA-8) Updaled Follow-up Period. 1983-1986 Tola1 Follow-up Penod. 1945-1986 malqnant neoplasms (140-209) 3 Dgesbve wgans and pentoneurn (150-1 59) j iymwatic and hernalopoletlc system ,200-209) ! Yw-++dgwns lymphoma (200 202, I h e r o x l e r o l c hean dlsease (410-413) 21 32.2 65 40-100 2 8 3 24 3-87 0 2.0 -7 0 2.3 - t 0-184 0-160 1 0 9 111 3-619 I; 3 2 - T 6-1644 0 0 7 -t 0-527 0 0 3 -t 0-1230 9 1 0 3 87 40-166 2 ! 6 125 15-452 132 143.3 92 77-109 28 309 91 61-131 5 7.8 65 21-150 9 11 4 79 36-150 3 3.2 94 19-274 0 ; ; -7 C-226 5 3 2 157 51-367 . _2 10 < -19-6- In, 24-708 ?$-?,G 46 46 1 100 73-133 1210 JOM Volume 35, Number 12, December 1% ewnis which r e s i r i a the statiaiical power of the study to detect or rule. OUI mode31 increases in nsh. .\itnccg: Causa of Death (ICDA.8) ~~~ ~ All causes (001-9991 All malignant neop(aSmS (140-209) Dgestie organs and pentooevrn (150- 159) Aespratwy system (160-163) Re!a!ive 95% Obs Exp Rlrk Confidence Interval 132 1271 28 2 9 4 5 71 104 095 071 088-1 24 066-1 39 0 29-1 21 9 1 0 6 085 0 44-1 64 no new cases of NHL were found, he two cases ohsrned or+nall> c o m 2 : with about one expxted In this c,. hort. Taken at face value, this slight excess could be interpreted as in accord with other studiessugg&ng that any risk of NHL from exposm to phenoxy herbicides is small.'-:s Genitourinary system (165-189) Brain and other CNS tissws (191. 192) Lymphattc and hematopoletic system (200-2091 Non-Hodgktn s lymphoma (200 202) Other and u n s m " e c s'es (171 175 194- 199) Artenosclerotic heart Itsease ,410413) 3 2 7 1 1 1 0 35-3 46 0 1 3 -7 -t 5 3 2 1 5 7 0 65-3 78 2 0 7 303 0 78-11 85 6 3 2 0 1 0 91-4 46 46 44 9 103 0 77-1 37 Moreover, an expert group recently concluded:* that these two NHL deaths occurred -. . . under circum. stances [modest exposure. short latent). and in the presence oi con- tbunding exposures] shich seem unlikely to implicate 2.4-D exposure." Cerebrovasculardisease (430-438) Acudents (E800-E949) 9 6 3 1 4 2 0 73-2 75 17 1 2 3 1 3 9 0 86-2 25 A n excess of deaths in the catego? of 'cancer of other and unspecified * Number of expected deatts are basad on Mdlgan Divislcn male mortality rates adjusted for age. caIenuar year. and pay status t Values are inestimable sites" was observed for employees on one of the four production plants, This category of death has been consistently elevated among the general 4 :a Discussion Uonality was updated an additional 4 gears. through 1986. for 878 smplo!,ees who manufactured 2.4-D and who had been studied oripnall) by Bond et al." A vanety of causes of death were examined but special focus was directed at NHL because of a possible association with the use of phenol! herhicldes. such as 2.4-D. reponed from =\era1 case-control StUdieS.JI: iY:! but not supported by similar studies done elsewhere." " " I' A criticism of the cacc-control approach has been the potential for a differential misclassification of exposure arising from either "recall" or 'interviewer" bias." Methodologrcal quest:ons Ca\e k c n rmsed about whether farmers' next-of-kin can recall their past use of pesticides in sufficient detail to pennit valid estimates of exposure." Questions have 31%~ been raised about uncontrolled con- founding due lo other farm e x p surcs." and the implausibility of infrcquent. farm enbironment exposures causing substantial increases in cancer nsks. as suggested by some of the casecontrol studies." Welldesigned cohort studies of uhich presumably makes these studies less prone to misclassilication errors.'3-'u Add~tionall~th.e workers are !ikely to have received more frequent and intense exposures relative to farmers or other herbicide applicators. especially during the early years of production when industrial hygiene controls -ere not ab sophistiistcd. The manufactunng environment ~ l s o offers 3 different xt of potential confounding exposures than those found on the farm. thus providing a useful contrast. Overall mortality in this group of 2.4-D production workers was o b sewed to have been comparable uirh the US population (SMR. 921 and to other uorkcrs from this manulacturing site. and was slightly lower than th3t reponed initially" tSX1R. 1001. Similarl?. mortalit) from total cancer was comparable to expected le\ els 3nd showed no trends with duration of exposure. a measure of cumulative dose or latency. This suggests that these workers have. 31 least thus far. not experienced a substantial cancer hazard associated with their occupational exposure to 2.4-0. The interpretation of findings for work force at thls manufacturing site,3' and was found to be partially due to historical death certification practices."A review of the death certificates and job titles of the ~ffecteu workers in the current study revealed no funher information useful for interpreting whether they might bc related to workplace exposures. S o deaths were o k ~ e fdrom softtissue sarcam3. although again the :ahon is small and fewer than one case was expected. Nor were there any deaths o b x n e d from brain cancer, the cancer type that was statistically significantly elevated io male fats,but not female rats or mice of either sex, fed the highest dose levels of 2.4D.'l** Brain cancer has not been h n d in excess in an> o i the a h o n studies of phenon! herbicide-enposed uorhcrs conducted to date.Lnc'''" :I-:@ and docs not now seem likely to be a consequence oi human exposure to 2.4-D. In summary. this 4-year update of monality among uorkers who manufactured 2.4-D did not show patterns suggestive of a c a u u l association with any particular cause of death. including cancer. workers who manufactured the phenoxy herbicides offer the ad\antage of documented records of exposure. specific types of cancer. including NHL. is complicated by the small number of observed and expected EAcknowledgments We graiefull) acknoulcdge tcchnlcal sisunce provided b) Kenneth M. Bodner. 1 ,OM Volume 35, Number 12, December 1993 1211 8. Canmill. Dana L. Diamond. 14. Pcarcc NE. Phenoxyherbiades and non- A. KravaL and Else 4. McLaren. Hodglon's lymphoma in New Zealand Crtquenc) and duration of herbiadc use. References ' I. W ~ LI. Ih d n r o m A. CAsc-controi ! m d y : soft tisue wcomas and exposure Br J Ind .Ued. 1989:46: 143- 144. IS. G m n LM. Mortality analyssofOntano Hydro forestry tradesman cohon 19501982. Toronto. Ontano. Canada: Healrh 1! to phenoxyaccclcacids or chlorophcnols. BrJCancrr. l9'9:39:'11-71?. ' 2. Eriksson M. Hardell L. Berg NO. Moiler and Safety Dnision. Ontano Hydro. Y* vember 1986. 16. Woods JS. Poliswr L. Sevenon RK. T. 4xclson 0. Soti tissuc $ a r o m a and Heuser LS. Kulander BG. Soft tissue exposure to c h c s i i d subwnccs: 3 c3y- sarcoma and non-Hsd&n's I!mphoma ' referent stud!. Br J lnd .Med. 198 I.38: in relauon to phenox! hcrbinde 3nd 3-33. chlorophenol exposure in western Wash- -2. Hardell L. E n l w n >!. Lnncr P. Lund- grcn E. Malignant I)mphoma and expo- ington J.VCI l%-.'S.WJ-'Jl~l 17. Vcneis P. Terracini B. Cictone G. et d. SUR to c h e m i i a cspxiall) organic 501'. muchlorophenols3nd phenoxy acrds: $ a ase-control nud! Br J Cancer. I98 I: ' 43:169-176. Phenox! herbicides and soft tissuc y r comas in female nce weeders: a populauon-bued case-referent nud). Scand J Ubrk Enwon Health. 1987:13:9-17 4. On MC. Holder BB. O l v n RD. 4 mor- 18. Vineis P. Fagg~anoF. T e d c r h M. Cic- 1 tplrp analysis of ernplo?m engaged in cone (3. lnadence rates of lymphoma the manufacture of :.4.!-tnchlorophcn- 3nd soft tissue sarcomas 3nd cn\iron- oa\acctic acid J O C U ~\fed 1980:Z mental mcuurements of phcnorh her- J--?U. blcldes JLC.1 1uul.a; 3b2-363 I Riihimaki L'. 4sp S. H e r t r y S 54On3l- 19 P:mn B. Dahlander 4. Frednksson .\I. , it) of 2.4-dichlorophcnox)acetic acid Brage HN. Ohlson CC. 4xelron 0. Ma- , and 1.4.S-tnchlorophenoxyacetic acid lignant lymphoma and cccupatlonalex- . k t n c l d e appliaton in Finland. S c a d J Ubrk Environ Healih. 1982:8:37-42. posures Br J Ind .Wed. 1989:45: 516-520. 6 nolfe WH, Llishaiek JE. Miner JC. 20 Hardell L. Enksson 51. The 3sociation RAe 4. et A.ncaith m u s o i . u Force bctwcn wit I I ~ I J C sarcom.u rod expc- w!crans occupationall! crpovd to hcr- SUE to phcnox~aceocaads: 3 neu c x - 5 s i i h i n \ i c t ; . ~ n . I. P h > w i d >.edth. rcfcrcni stud). Cai;',r. !9SY.o1. JA.W.4 19W.?61~18?Ll831. 652-656. 7. Smith AH. P m YE.Fisher DO. Giles ? I Ott MC. Olson RA. Cook RR. Bond HJ.Tcague C.4. Houard JK. Soft tissue ' saxcoma and c x p o s u r ~io phcnox) hcr- CG Cohon monalit) stud) o f chemical workcn w t h porcntial exposure io the t k d e s 3nd chlorophcnols in Sew Zea- higher chlonnatcd dioains. JOccup.\lrd. ' bd.J.%CI 1984:73:1111-1117. , ?. Lyngc E. A follow-up nud) of cancer inadencc among workers in manufac- 1?8?:19.422-429 22. BondGG.McLaren EA. Lipps TE.Cook RR. Update of monaltty among chcmi- lure of phenox\ herbiiides in Denmark. cal uorhers uith potential esposure to ' .. :he higher ;hlunnatcd Jia\ins 1 ".:a; Csgon D. P~nnctB. \\inter PD. k h e - \led. l988.J I.12 1-;13 '? ED Bocw;. ! \I.-i!;i! i i ~ ~ ~ n c r -( * Bund GG '-\ c!tentrscni \ t i il,.-sh 6:. O w u d io :-mctn~,-Lihlorophrno\). JC?!I< acid. S.-:L j ij'or~. Em:run \lcL.~rcn 4 . Lippr TE. C ' m k RR C3uv spr:.!!ic m o n i i r ! me cmpicx- Htaiih. 198b.iZ U8-4'4 res engaged in the manuiacturc. formu- JO. Coggon D. Pannet B. winter P. M O ~ ~ I - lation. or packag~ngo?f .4dichlorophcn- , I. it?. and incldence of cancer 31 four fac- i tones making phenol! herbicides. Br J ;, Ind.\lrd. 1991:48:173-178. oayacetic acid and related salu.Br J Ind .\l<d.!958.4<.98-105. 24. Fingerhut .MA, Halpenn WE. Marlow , .**- G o t o r K.P Blair .4. 4gncultural Chem- DA. et al. Cancer momlitv in worhen i d s . Dnnk:ng U'arcr. 3nd Puhlis expovd :o ~.7.'.4-tetnchlilr~ihenrrr,l H:alth 4 n E;ldemio!opc Rc\icu Be- d i s h i n \ Lw/ J t/cd ! W l . J : 4 ihesdz. \ID \ationd Cmcer Institutc. 212-:1s !"Yb : ;i SK. ', i i ::!c, : ..%rn.c..-u.pltiu. r al I - hcrbiode _~ - ! 9 use :I"UC . ..... . --,- .,..;.<. .,&I _ _ .I,, ;-7, i i 3nd nsk ~r,1-m3 I-< ,I,w ..i * .'>. ~rm _.u r 't r i o nJ. SE. 4 S if ~ 4 l c IQQ6:?56.3!51\. p ~ l3d .h t h A I L 1 - Teague C.4. Uon-Hddgkin'r lymphoma M d fanning. 3n expanded sau-control ' SIUd! B r J C j n c . r r 19R' 39,155-161 , . .25 L h r r SH. \\'c:ccnhcreer DD. Babbitt :: _,I . . J K - i , ,. I ..-L,, , '8 .*.> * L A > \...;-A* .,.,.< I W-lV? ia.:Id ..i,,c 5;.X l l I C i k * . l ' x , ; a * l , , i;. R i d e l D. Ritter L. Momson HI, MJO f'. M o m i i b stud\ of Canadian male f3rm operators. non-Hodgkin's I!m- phoma mortality and agricultural practim in Saskatchewan. JYCI. 199032: 575-582. 27. Bond GG.Bodner ICM. Cook RR. Phen- oxy herbicides and cancer: insuffincnt epidemiologrc evidence for a causal Rlationship. Fund .4ppl Toxrcol 1989:12: 172-188. 28. Johnson ES. Association between soft tissue rarcomas. malignant lymphomas and phenox) hcrbiodes/chlorophcnols: evidence from occupariond cohon studies. Fund .4pp/ Toxrcol. 199O:IJ. 219-234. 19 Hamard School 01 Public Hcahh The weight of the evidence on the human carnnogenicit! a i :.tD Rcpon on workshop. Boson. MA: Program on Risk .Analysis and Ennronmentixl Health: 1990 30. Wonson RR. hnalvnsofrclati\esuMYal and proponional m o d i t ? . Compw Biomrd Res. 1974:':31!-?32. !I Rothman KJ. Boice I D Epidemiolopc anal\siruith 3 proprammablc c;~Icu!ator Boston. 41.4: Epidemiolop Rwurces Inc: 1982. 31. Enterline PE. Marsh CM. Canceramong workcn exposed to arsenic 3nd other substances in a copper smelter. .Am J Eprdrrniol 1981.116:895-91 I jj. Bresidu SE.Lubm JH. XtarcL. P. Lane holz B Multi~licnri\emodels 3nd the ~nd:)siurfsohofl 3 3 ~J.. l n : S:J:.L>L'.. 1983:78:I-12. 34. Hakulinen Th. 4 Mantel-Haenszcl statistic for taung the associauon between a polychotomous erposure and 3 rare outcome. .dm J Eprdrmrol. 1981:l 13: 192-196. 35 Blarr A. Zahm SH. Cantor KP. Stewan P4. Estimating exposure to pcsticldes in epidemiolo@c studies of cancer In. \\on8 RC\I. Frml!in Ti. Hcnc!cut! RC. Rcinrn IC. cds.Bii;lupial \!omtar. $ c g tor Prsti.d? F\ji<wre-\!u.u:. n e n t . Esiimation And Rish R c d a a o n \CS S!nipo,sium Sczcs H~sr.:ngion. Dc. .Amencan Chemical Socicr!. 1%Y: 3823846. 36. Johnson R4. Mandel JS. Gibson RW. Mandel JH. Bendor AP. Gundenon PD. Renter CM.Data on pnor pcsunde use mllstcd from self- and proxy respondents. Eprdemiolow 1993:4.157- 164 i' Smith 4 H . Batcs >I> Epidemiologic siudies of c ~ n c c ar nd wticide cipsure \lc.wmmcnt, E$:im.i!:on J A Risk Rc. n DC 4mencan Chemic31 Srrietv, 38. Pearce N. ReifJS. Epidemiolmc 5tudics of canar in agricultural workcn. .Am J lnd \/d 1990:IR I 33- I411 E 1212 Update on 2,443 Exposure Bloemen 39. Bond GG,McLaren E.4. Canmill JB. Wymer KT. Sobel W. Lipps TE,Cook RR. Cause-specific monality among male chemical w o r k m . Am J Ind Mea. 1987;1?.353-383. W . Bond C 6 ,Cdnrnlii Ja. Cm)h 24. in*c.i- tlgation of cxicss caner of other 2nd unspecified sites among chemical work- 1986. CR. Am J Ind Med. 1990:18:617-619. 42. Oncogenicity audy in mice wia k4 Lcner to the Editor. dichlorophenoxyacetic acid ( 2 , 4 ~ , 4 I . Combined loucll). and oricogsntcii). iinal rep.m. V'sliime I . Vienna vq' nub? i n rats, ? . ~ - ~ i c h ! c r ~ ; h c n o i ~ ~ c e t ~ c H. az-elton Laboratories .4mericq d;nd, ;Ci%C. \ 2;uT.L ; . \ .;'C?.J. V.4: Hazelton Labratones Amenca: I. 3 .- I? Erratum Several paragraphs of text were inadvertently omitted from the ACOEM Position Statement on National Health Care Reform. published in the June 1993 issue. page 623. The missing text. which should habe appeared after the fifth paragraph. reads as follows: The ACOEM Proposal The .American College of Occupational and En\ironmental Medicine proposes changes in six key areas related to health care reform: coverage, financing, suppl). demand, administration, and information management. Coverage A basic package of benefits that emphasize prevention, primary care,and continuity of care should be provided. The basic benefits package should be defined to meet medical practice guidelines. process efficiency standards. and outcome specifications rather than to cover types ofxrvice. Only proven, cost-cflkctive. and appropriate technology. YMCCS. and pharmaceuticals uould be cokered. Rewurces should be directed ana) from terminal care or low yield to more productive uses. h'orksite-based p n m a p care and health-promoting and preventibe senices should be reimbursed. These senices are often more cost effective and less disrupuve to workers' and SU~CMSO~S'schedules when provided at the worksite. To improve the quality of care and manage the disability of work-related illnesses and injuria. and to stop con shifting. we feel it is critical to create a uniform reimbursement and medical management system for occupationally related illness and injun. that pro\ides the same qualit!. etfictenc:. and el'rectiwness as general medical senices. One option is a merger of the two systems and diwbilitv coverage uith a focus on the appropnate care of the worker and appropnaie and effective management of disability, especially the provision of modified duty to accommodate impairments. Tius might require a Federal law preempting state laws. T h e 'work relatedness" test would bcrome irelevant. as would legal issues of compensability. hsability coverage for any absence would be combined. Fimq 411 Amcncans should tu required to purchase or ohtain hasic medical insurance for medically necessar): care from an employer-prowded benefits plan or from another (possibly government-organized) financing pool. Low-income Amencans would mei\e w credits or vouchers for this purchase. This is not intended io create a separate but unequal system ofcarc for those in small business or for the uncrnployd. The government pian is a nsk-pooling mechanism. We sincerely regret the omission.