Document 1QyBQ4qg4Vbk091qBoJBo9zKj

:iVc ExItji '*ny Cw jigifittnng rrxrcn. ; ibc hfiJlb = DHHS 0. Inrjtvtc f li>W2 n'h Ocyp A-7K. for a : A m e rica n Juurtud litdii.sirial Medicine llilif - O & tIV92( A Cohort Study on Cancer Incidence Among Danish-Gardeners Eva Siaiirup Hartsfcft, mo, PhD, Henrik Haste, mo, and Flemming Lander, MD This study u*3b performed to examine the voncer risk of Djnlih yanier.irs hitvine bisn highly exposed lo pesticides. Wc have followed a cohort of 4.015 employ id sardine r* (S59 females and 3.156 males) from Mj \ 1975 until me end or 19S4 with regard to cancer incidence. The observed incidence was Compared with expected numbers calcu lated from national incidence rates, fo r all cancer sites combined, the standardized morbidity ratio (SNibR) was 104, Among male gardeners a significantly increased incidence was seen for soft tissue sarcoma iSMbR = 5 2 6 .95ft cuftfidsncc internal (Cl j: IO y -L i38),-and chronic lymphatic leukemia (SMbR *= 275. 95ft Cl; ll)!-599i, The incidence of.nun-Hodgkin's lymphoma was twice that which was expected (SMbR 200. 9 5 5 CL 56-393), Wc suggest that some of the pesticides to which the gardencri have been exposed are capable of iniuatmg or promoting the development of malignant neoplasms in tissues o i mesenchymal origin. C 19$; Wik; *ti, lrx\ Key words: lymphatic leukemia. non-Hodkin* lymphoma, occupational cancer, pciticides, soft tissue sarcoma, agricultural exposures, gardeners INTRODUCTION Gardeners arc exposed to many different substance* including insecticides, herbicides, fungicides, disinfectants, growth regulating substances, and fertilizers. Thu international Agency for R esearch on Cancer (1ARC) has reviewed me evidence for the carcinog*enicity* o f several o f those acents and has concluded that the folle vinV pesticides may be carcinogenic to hum ans tIARC-group B): amhrole. chioidecone; chlorophenoxy acetic com pounds, 1.1 .l-irich]oro-2,2-bis(p.ch)orophenyi)ethane tD D T), 1.2-dibrom o-3-chloropropane iD B C PL hexachlorobenzene, hexachlorocYo lhcxnc {including lindane), rr.irex. toxaphene. and 2 .3 .7 .S-wichloro-dibcnkodioxin (TCDD) tInternational A gency for Research On Cancer. 197*4. 1976. I$77, 1979. 1953. 19S6. 19S7-3- In addition, the Cancer risk associated with exposure to arsenicalcontaining pesticides is well docum ented [International Agency for Research on Cancer, 1980). Institute rvf Social Medicine. UnKer>it\ of Copenpajcn. Coper.hujeji. Denmuri (E.S.H.i. Inmwte of Community Health. L'nr.eiVii;, of . Odens. D ::::l-. tH H.), Department of Occupaioral Medicine. Odense l'r.h en.iy Hospital. Odense. Denmark tF.L.). Address rcp:lnl requests to Dr. va Stotirup Hansen, institute of Sixbi Medicine, P.<r.,;:r. Insuis. Bleedaimsvcj 3. DK-220Q Copenhagen N. DtrnmurL Accepted for publication September 5. 1991. 19*>! Wjlcy-l.its, lne. I 1 i \ -L .'H3 652 Hnn$en et at. Only a few epidemiological studies have focused on gardeners* health. A study from the United Kingdom reported an increased risk of soft tissue sarcoma among market gardeners [Balarajan and Acheson, 19&4], A Swedish registry study on per sons working in horticulture found no increase in either soft tissue sarcoma or ma lignant lymphoma [WUdund and Holm, 1986; Wiklnnd ct al.( 1988a], whereas a case-control study on soft tissue sarcoma in tire southeast of Sweden revealed a fourfold increase among gardeners [Wingrsn ci al,, 1990]. Other occupational groups which may be exposed ro pesticides include forestry workers, flour industry workers, farmers, pesticide applicators, and workers cmployed in pesticide manufacturing plants. Several studies on these groups or Oft persons living in pesticide contaminated communities have indicated an increased risk of lung cancer [Ott ct al., 1974; Barthel, 1976, Banhel, 1981; Lynge, 1985], maUgnam neoplasms of lymphatic and hematopoietic tissue [Hardell et ah, I9S1; Burmeisrer et aL, 1982, 1*983: Cantor, 1982; Pearce etal., 1985, 1986a,b, 1987; Schu macher, 1985; Hoar et al,, 1986; Woods et al., 1987; Dubrow et a t , 1988; Rodin et al., 1988; Mcl-sughlin ct al., 1988; Corrao et al., 1989; Godon ct al., 1989; La Vccchiaetah. 19S9; Persson ct al., 1989; Alav&njaet a]., 1990; Wcisenburger, 1990; W igleetal., 1990; Zabro et al., 1990; Coggon ct al., 1991; Vineis et al,, 1991). and soft tissue sarcomas [Hardell and Sandstrom, 1979; Eriksson et al., 1981, 1990; Lynge, 1985; Vineis et al., 1986; Zahm et al., 1988; Fingerbut ct a!., 1991], This study was set up to investigate further the possible relationship between occupational exposure to pesticides and the risk of malignant neoplasms, primarily those of mesenchymal origin. The study considered cancer incidence among em ployed gardeners throughout a 10-year period. PERSONS AND METHODS A historical cohort of employed gardeners was followed from May l. 1975 through January I, 1985, with regard ro cancer incidence. The observed incidence was compared wjth that which would have been expected had the sex-, age-, and period-specific cancer incidence rates of the total Danish population applied to the study cohort. The study Cohort The cohort was defined to include every person who, on May I. 1975. was a member of one of the ten local trade unions of gardeners, associated with the Danish Union of General Workers. The members of the study cohort were identified by using an old version of the membership file, which by coincidence bad been kepi on a magnetic tape, containing the members' civil registration number and the code of the local trade union to which the person in question was affiliated. The cohort comprised 4,015 persons, of whom 859 were women and 3,156 men. Follow-Up Procedures The study cohort was followed until Januaiy 1, 1985, with regard to vital status and emigration. The follow-up was performed by our searching for the individual civil registration number of each of the cohort members in the Danish Central Pop ulation Register; this procedure enabled us to trace everybody. Each person accumu lated person-years at risk from May 1, 1975, until the date of death, the date of | ] | ; ! j I i 1 j  I *0 l! study iion i per: ma^as a 3ed a l cm* ir on frisk maBur* *chu- -\n el La 990; , and 990; -ecn ariJy cm- . i* 1975 .ence and ? the as a jiish -sitig ona f the ised :snis dual ?opmue of Cancer Among Danish Gardeners Table I. Number of Person*Years ui Risk in the OinUb Gardener Cohort, 19?S Through 1584 Ape (years) Mai Feinal Ttiu) 15-29 30-44 45-54 55-64 65-74 75-79 S0-&4 15-84 4,743 y.oe? 4.5SD 6,467 4.2S8 210 S 29.36R 1,315 2,484 I.9B0 1,806 431 10 1 8,025 6.058 11,554 6.560 8.275 4.7i$ 219 9 37,393 653 emigration, or January 1, 1985, whichever occurred first. The study covered A iota! of 37,393 per$no-yesxs at risk {Table I). Information on the cancer incidence jo the cohort was obtained from the Danish Cancer Registry, which U a nationwide register that has been in operation since 1943 [Clemmesen, 1974], For each member of the Study cohort, the individual civil reg istration number was searched for in the files of the Danish Cancer Registry covering the period from the Study start on May 1, 1975 until the most recent Updating of the registry, January 1, 1985. Following the identification of cancer cases, we attempted to get additional exposure data for those study persons who had developed soft tissue sarcoma or malignant neoplasms of the blood- and lymph-forming tissues. For most of these persons, the local unions w-ere able to provide data on work area and/or duration of union membership. Datn-Arialysis The site-specific cancer incidence observed in the study cohort was compared with the expected incidence based on the national incidence rates, specific for 5 year age groups, sex, and periods. To allow for an induction/laiency lime of 10-15 years, we curtailed the analyses to cover only persons aged 30 years or mors, as aimosi all gardeners start in the trade when they arc young (personal communication, head of the gardeners' section of the Danish Union of General Workers). Albdt a crude substitute, analysis by age was the only alternative to ignoring rtducliOn/laLenev lime, We also disregarded the data for the oldest age group [i.e., 80-84 ye/ir$ of age) because of the extremely small number (nine) of person-years at risk in this group. The statistical evaluation was based ori the assumption that the observed number follows a Poisson-distriburion. Confidence intervals (Cl) for the estimated standard ized morbidity ratio (SMbR) were based on exact Poisson-1imiis [Bailar and Edercr, 1964; Ciba-Geigy, 1975], Exposure Employed Danish gardeners constitute three fairly separate groups according to work area, namely workers in 1) greenhouses, 2) nursery' gardens, and 3) public parks, gardens, and cemc;crie$, In general, the gardeners display a low job-turnover and a pronounced stability of employment. According to information obtained from the head of the gardeners* section of the Danish Union of General Workers, almost all of the female members and j0-2072 "r r .1IF"L'W . V '.M iI*. iI , l- r i iittt! t I iafc*. litv IKif- 64 Hansen et &L of the male. members were greenhouse workers at the time the cohort was established Reportedly, the major part of the rest of the gardeners--almost exclusively males-- were working in public parks, gardens, and cemeteries. Only a few percent were working in nursery gardens. Both the greenhouse workers and the outdoor gardeners have been regularly exposed ro a mixture of pesticides during their active work life. The union people estimate that about one-third of the gardeners, almost exclusively males, have been directly exposed by spraying pesticides, whereas the remainder of the group have been indirectly exposed, primarily by skin contact when bundling newly sprayed plants or growing medium mixed with pesticides, and by inhalation when working in newly sprayed areas (personal communication, head of the garden ers' section of the Danish Union of General Workers). Workers in greenhouses have primarily been exposed to fungicides and insec ticides, including the various chlorinated compounds (e.g-, DDT, chlordane. and lindane), organophosphorous compounds, and nicotine. Gardeners holding outdoor jobs have primarily been exposed to herbicides, including the phenoxy acetic acids (e.g., 2,4-D. 2,4,5-T, and MCPA) and amitrolc. This exposure has taken place regularly throughout the growing season, i.e,, from April Through October. In addi tion, the outdoor jobs have involved some exposure to insecticides and fungicides. It is worth noting that DDT had been extensively used in Denmark until the late 1970s. On the other hand, th c ageniral insecticides have never been used on a large scale in Denmark. In this cohort study, data on individual job histories have not been obtained except for a subset of the cancer cases (see above). However, in the main, gender may be taken as a rough substitute of type of exposure, with a predominance of indirect exposure ro insecticides among female gardeners, and of direct or indirect exposure to herbicides among male gardeners. RESULTS A total of 219 cancer esses was observed in the study cohort. Of these, 217 case occurred among persons aged ?0 to 79 years. The SMbRs for the ni2in groups of diagnoses are shown in Table II. N'one of these ratios differed significantly from unity. Separate analyses for malignancies of mesenchymal origin showed a signifi cantly increased incidence of soft tissue sarcoma and chronic lymphatic leukemia among males, and a non-significant increase of non-Hodgkin`s lymphomas in both males and females (Tables III. TV). For those study subjects who developed soft tissue sarcoma, non-Hodgkin's lymphoma, or chronic lymphatic leukemia, the available exposure data are shown in Table V. DISCUSSION The results of this 10-year follow-up of employed gardeners seem to indicate that, on the whole, the cancer incidence among gardeners does not differ much from thai of the total population. However, significant increases were recorded for soft tissue sarcoma and lymphatic leukemia along with a non-significant increase of non-Hodgkin's lymphoma. As regards the potential bias in this study, the cohort of gardenerx represents a healthy subset of the total Danish population, as the study persons wrre all occupa- C anctr Among Danish G ardcD frt 655 TABLE II. Cancer Inndcnee Amanj Danish Garttmers Aged 30 to 79 Years* Male* Fratti Total Carve/ site* Total (140-205) Eureal cavity ana ph&rvr.k (140-148; Dimeni *e and ptrifoo^utn (150-159) Rtspiri'-ory syycm (160-165) B.-cLit (J70) Frittale gcfiiul organs (171--176) Male OfgatK (177-179) Obi1 cap 1*4/172 05 6/5 M 42/40.78 41/42.58 GO 29 _ -- 2QU7.8 SMbR (95SfcU) 107 (y:-i24j Ot/V exp 33/35 66 114 (42-247) GO 40 103 (74-139) 7/5.72 % (69-131) 0(0-1272) --. __ 2/2 82 109 33 7/7.75 115 (70-178) -- -- SMbR <95*CT) 93 ((4-130) 0 (0-9221 122 (49-252) 71 (9-256) 107 (51-211) 90(36-186) Otrt.' SMbR P (9550) 217/207.71 1CU (91-120) 6S.6S 106(39-230) 49/46.50 105 (78-139) 43/4S 40 94 (69-128) 10*9.62 IO- (50--205) 7/7.75 90 (36-166) 20/17.38 M5 (70-178) n so -isn St?n (190 191) Other specified sites (192-197) Unspecified siiti (198-199) L>mpfmJc and heouiopoictic tissue (20C-20S) 18/20.67 \P (52-158) 31/24.T7 127 (86-181) 6/6.43 93 (54-203) 5/4.11 122 (40-2W) 2/1.54 130 (16--469) 1/4.45 22 (1-125) 1/1.48 68 (2-376) I/U.71 14) (4-785) 2073.21 93(55-139) 32/28.82 111 (76-IS?) 7/7.91 83 (26-182) 64.82 124 (46-271) 15/10.45 )44 (SO-237) Id .46 137 (17-495) 17/11.91 143 (83-229) "Obscr-cd ne. (OW . txpcctcd nrv (rvp), si^drdized mortndity ratio (SMbR). and 95** cot>riccncc interval (957* Cl) for the SVfhR. 'Tne code of the International Ciass'rcation of Discos (1CD), 7th revision, in parcnihe^i. TABLE III. Incidence of Soil Tissue Sarcoma Among Danish Gardenm*,fc obs m SMbR (95?c Cl) Males hemak! Total 0 3 526 (109-1.538) 0 (0-4,099) 455 (94-1.324) `Ofc--rved no. (OfcJ). Standardized morbidity rado (SMbR), and 95^ confidence mtcAi! (95^ Cl) for the $V(bR. Mode No. 197 in the ICD. 7tn revision. tionally active a: the time of enrollment into the study. The cohort constitutes--also with respect to urbanization--a non-representative sample of the total Danish popu lation as only a few city-dwellers are included. Generally, the cancer incidence rates arc higher in cities than the national average. Tnis difference is most pronounced for tobacco-related neoplasms [Clcmmessn. 1974; Danish Cancer Registry. 1987.]. Ac cording to the unpublished revolts of Lander, gardeners generally smoke less than the average population. Tnis feature would create a negative confounding as far as tobacco-rclated cancers are concerned. As regards the diagnoses that were seen in excess among the gardeners, i.e., soft tissue sarcoma, ehronic lymphatic leukemia, I o :o O' 6S6 HaniHi et al. TABLE IV. Inodance or MaTigrinrU Neoplasms of Lymphatic and Hematopoietic Tissut Among Danish GanLmers* Males Females Total Diagnom Ohs SMbR (9S Cl) Obi SKfbR (9591 a> Obs SMbR (95 Cl) TociJ3 13 144 (80-23?) 2 137 (17-495) 17 145 (63-229) Ncn-Hodelan** lymphoma* 6 173 (63-376) 2 364 (44-1,314) 6 200 (86-393) Chronic lymphatic leukemia* 6 273 (101-599) 0 0 (0-1,677) 6 251 (92-546) Other leukemias* 3 144 (30-419) _0 (0-1.116) 3 123 (25-361) Observed no, <Ot>5). standardized morbidity ratio (SMbR), m i 95"4 confidence interval (Of% Q) for the SMbR. *ICD, 7th revision, no*- 200-205. ^ICD, 7ih revision, nos. 200, 202, and 205. elCD, 7ih revision, no- 204.0. National incidence ra: for cijicmic lymphatic leukemia have not been available, but have been estimated from the Incidrnte rates for all leukemias and the proportion of chronic lymphatic leukemia among the incident cases of leukemia, 1S73 through 1977, stratified by age and sa*. '`ICD, 7th revision, nos. 204.1-204.4, TABLE V. Duration of Evposure Prior to Diagnosis for Those Gardeners Who Developed Soft Tissue Sarcoma (STS), Noo-HodgWiTs Lymphoma (NHL), or Chronic Lymphatic Let&emla (CLL) Diagnosis (jqw y STS (197) STS (197) STS (197) NHL (200) NHL (200) NHL (200) NHL (200) NHL (200) NHL (200) NHL (200) NHL (205) C U (204.0) O L (204.0) CLL (204.0) CLL (204.0) CLL (204.0) CLL (204,0) Age (ye*n) at diagnosis 31 60 71 38 75 71 54 54 65 M 70 56 69 66 56 68 a s? Male Male Male Female Female Male Male Male Mole Mate Male Male Male Male Male Male Male Duration of exposure (years) 9 13 42 S 10 10 17 28 35 45 Unknown 32 33 44 Unknown Unknown Unknown *71)6 code of the 1CD, 7th revision, in parentheses, and non-H^dgkin's lymphoma, the national incidence rales have not displayed any region?! fiends within the period studied [Danish Cancer Registry, 19S2], and there is no reason to suspect that the observed increase among the gardeners has been due to confounding by urbanization. As regards the quality of diagnosis, this has not differed between the twu groups cumpared, because access to medical C2ru was free and equal ihruughqui the period studied. All in all, the estimated SMbR for soft tissue Cancer Among Danish Gardeners- 657 sarcoma, chronic lymphatic leukemia, and non-Hodgkin's lymphoma seem to be reasonably free from bias. On ihe other hand, the SMbR-estimates for tobacco-related cancere art certainly biased in a negative direction. The diagnostic category soft tissue.^artoma (International Classification of Diseas, /in revision (1CD-7), code number 197) jncjudrAalLcjitraskeletaJ sarcomas exchidiflgjympho- and rrticulosarcomas. However, some of the cases of soft tissue sarcoma may have been coded according to their topographic localization rather than according to their histology. The extent of this type of misc)assiftcation has been the topic of a validation study performed by the Danish Cancer Registry, This examina* rion revealed that, during the years 1978 through 1982, the registered incidence of extrasfceletaJ sarcomas excluding lyrhpho- and rericulosarcomas had been 3.4 limes the registered incidence of ICD-7 code No. 197 R-ynge et al., 1987], Nevertheless, the national incidence rates for soft tissue sarcoma are based on the cases given code No. 197, and the comparison of observed and expected numbers (Table Ul) have to be restricted to the cases given this code. For completeness, however, we checked the list of gardener cancer cases for extraskeletal sarcomas. For this purpose, we made use of a special sarcoma code, which the Danish Cancer Registry hS employed since it started registration procedures in 1942 [Clemmesen. 1965], By this procedure, we found no additional cases of soft tissue sarcoma among the gardeners. We found a fivefold increased incidence of soft tissue sarcoma among Danish gardeners. This observation is consistent with the findings in several studies of occupational groups exposed to herbicides [Hardell and Sandsttom, 1979; Eriksson et al.* 1981, 1990;LvnEe, 1985; Vineis etal., 19B6;Zahmetal.. 1988; Wingreen et al., 1990; Fingeihut et al-, 1991]. Other studies have failed to demonstrate an association with herbicide exposure [H oard al., 1986; Wiklundand Holm, 1986; Wiklund et al,, 1988b; Green, 1991: Vineis al,, 1991], This inconsistency may reflect method ological differences affecting study sensitivity or differences between the populations studied as to the type and the level of occupational exposure. Our observation of an increased incidence of chronic lymphatic leukemia and of non-Hodgkin's lymphoma is in agreement with the results of several other studies, which have pointed tp chlorinated pesticides such as the phinoxy acetic compounds and DDT as being possible risk factors for lymphopoietic neoplasms [Hardcli ct al,. 1981; Hoar et al.. 19B6; Pearce et al., 19S6a,b; Woods ct al,, 1987; Flodin et al., 1988; Penson et al., 1&S9; Alavanja et al.t 1990; W'eisenburger, 1990; Wigle et al., 1990; Zahtn et al., 1990; Coggon et al., 1991; Vineis et al,. 1991]. In this context* it seems worth mentioning the evidence for a cytological relationship between chronic lymphatic leukemia and the various types of nor-ltrwjftkiTiVlyTMphorffln flf joriry of cases in both diagnostic categories are of R-cell origin [DeVita et al.. 19891. TR^a^-tharinO'st cases oTchronic lymphatic leukemia and of pnp-Hndakin^ tvmpfioma represent mamtestationS oLft rnnl'TttPt trensfonaarioiuin a R-Ivnqpfrnfytii progenitorcefl could imply a common eiio1ogyJhrJhA_dise.ases in question. CONCLUSIONS Our finding of an increased incidence of soft tissue sarcoma and lymphatic neoplasms in the pesticide exposed cohort agreed with our expectations based on the scientific literature relating to this topic. Our results indicate that some pesticides may be capable of initiating or promoting the development of malignant neoplasms orig- _g j T - j * ----- - / ]ffl T-niA- .ir- -~v' -jm*m f > 65S Hansen et al. inaiing in mesenchymal tissues. It is tempting to suggest that some of the carcino genic substances such as amhrole or the halogenated compounds have played t part. ACKNOWLEDGMENTS The authors thank Mr. Knud Ove Pedersen and Mr. Jcsper Lund Lirsen for collecting data orv individual job histories as well as providing information about the changes over the years in the gardening trade, and of typical exposure situations. We also want to thank Ms. Marianne Fur Foulscn, Ms. Lotte Jorgensen, and Ms. Inge Olsen for their assistance in the preparation and linguistic correction of the manu script. REFERENCES AUvwja MCR. Blair A, Masters MN (19-50): Cancer mortality in the U.S. flour industry'. J Mail Can: lust 82:840-848. Bxilar JC, Edaer F (1904); Significance factor* for the ratio or a Poisson variable lo itx expectation. Biometrics 20:639-643. B?hrajan R, Arheion FD (1984): Soft tissue sarcomas ih asrifulturt and forestiy worker*. J Epidemiol Commun Health 38:113-116. Birthd E (1976); Gehuftes Vorkommen vgn Bronchialkrebs bet beruflicher Penlaidcxpositon in der Landwirtschaft. Z Erk/unk Ann-Org 146:266-274. Barthel E (19BI): Increased risk of lung cancer in pesticide-exposed male agricultural workers, J Toxirol Environ Health 8:1027-1040. Burweister LF. Van Lier $F, Isjcsnn P (19S2); Leukemia and farm practices in Iowa. Am J Epidemiol 115:720-728. Butmeistfij LF. Evcrcn CD. Vim Lier $F. Isacsan P (1983): Selected cancer mortality ar.d farm practice* In lows. Am J Epidemiol 118:72-77. Cantor KP (1982): Farming and mortality fram mm-Hod^kin's lymphoma: A case-comro! study. Im J Cancer 29:239-247. Ciba-Geigy (1975): "Pocumsma Geigy, Mathematics and Statistics." Copenhagen: Cita-Ofilgy, p. 107. ` Clemmcsen 3 (1965): " StaiiMica! Studies in the Aetiology of Malignant Neoplasms: I: Review and Results." Copenhagen: Munksgaard, Clcmmtsen J (1974): Statistical studies in the aetiology of maligrumi neoplasms: IV. Denmark 1943-67. Acta pithol Microbiol Scand [Suppl] 247:1-266. Cuggtm D, Pannen B. Wittier P (1991): Mortality anti incidence of canter at four factories making phenoxy herbicides. Br J Ind Med 4S;I7?-17S. Corrao G, Callcri M. Carl? F. Russo R. Eosia S. Piccioni P (1989); Cancer risk ;n a cohop of licensed pesticide users. Scand J Work Environ Health 15:203-209. Danish Caocer Regisuy (l 982): " Incidence of Cancer in Denmark 1973-1977." Copenhagen: Danish Cancer Registry. DiVita Vl`, Jafft $, Mauch P. Longo DL {19S9): Lymphocytic lymphomas. In DeVita VT. Heilman S, Rossiiberg SA (eds); "Cancer, Principles and Practice of Oncology." Philadelphia: Lippirvcoa. pp 1741-1798. Dubrow R. Paulson JO. Tndinti RW (19SS). Fwminj and malignant lymphoma in Hancock Coumy, Ohio. BrJlnd Med 45:25-28. Eriksson M. Harden L, Birg NO. Mller T. Axclson 0 (1931): Sori-ssue sarcomas and exposure to domical substances: a case-rcfcrcnt study. Dr J lnd Med 38:27-33. Eriksson M. Harden L. Adami HO (1990): Exposure id dioxins i a risk Dcior for soft tissue wrroms: A population-based case-comrol study, J Natl Cancer Inst 82:486-490. Finpcrhut MA, Halpcrin W`. Marlow Da . Piacittlli LA. Honchar FA. Su.-ccr.cj. m h , Greife AL, Dill PA, Siecn'nnd K. Suruda AJ (159Il; Cancer mortality in workers exposed to 2.3.7.8-tetraehIOrvcliherv.n'P'itiftrin. N Engl J Med 324:212-2)8. Cancer Aitumg Dunlch Gardeners 659 Rodin U, Prcdrikswr, M. Perston. B, Axctwn O (1983): Chronic lymphatic leukaemia and cn-ins exhausts, fresh wood, and DDT: a ease-referent study. Br J Ind Mrd 45:33-38. Oodon D. Tbow-2. J-P. L-ajove P, Nadeau D (f9S9): Incidence of cancers of the brain, the lymphatic tissue?, add of leukemia and the use of pesticides *nong Qucbcc'i rural farm population ]9$21983. GcOraphi3 Mecfiea 19:213-231 Cnsen LAI (1991): A cohort monaJity study Of forestry workers exposed to phinoty herbicides. Br i Ind Med 43:234-238. HwtleU L, SandstrSm A (1929): CaM-contral study: Sofr-dssuc sarcoma aid c^pWlKi to phihMyaietie ride or chiwoptwtiols- Br J Cartier 39:711-717. Hanlell L, Eriksson M, LennerP, Lundgrcn E (19$1): Malignant lymphoma and exposurt to chemicals, ptciaHy organic solvents, chlorophcnols and phenaxy acids-. A case-control study. Br J Cancer 43:169-176. Hoar SK. BJair A, Holmes FF, Boysen CD, Robcl RJ, lloovcr ft., Fraumeni JF (1986): Agricultural herbicide use and risk of lymphoma and soft-tissue sarcoma JAMA 256:1141-1147, International Agency for Research on Cancer (1974): " Some OrgahoehloriM Pcsticldii " Lyon: brer* national Agcnty for Research on Cancer (1ARC monographs on the evaluation of the tareinogcnic risk of chemicals to humans: Vql 5). International Agency for Research on Cancer 0976): " Some Cuibamates, TmocarblfOites and Cliba- 2id ." Lyon: International Agency for Research aa Cancer (lARC monographs on the evaluation of the carcinogenic risk of chemicals to humans; Vot 12). Intimations' Agency for Research on Cancer (1977): "Some Fumigants, the Herbicides 2,4-D and 2 ^ 4 -T Chlorinated Dibcnzodioxins and Misceliancop? Industrial Chemicals.*' Lyon: Interna tional Agency for Research on Cancer (LARC monographs on the evaluation of die carcinogenic risk of chemical* tu humans; Vol 15). lnlcmaiioaal Agency for Research on Cancer fJ979): ' `Some Halogefiated Hydn.*sri>on$*\ Lyon: Inter- national Agency for Research on Cancer (IARC monographs on the evaluation of the carcinogenic risk of chemicals to humans; Vol 20). International Agency for Research on Cancer (1980): "Some Metals and Metallic Compounds." Lyon: IfiKnuttional Agency for Research on Cancer (lARC monographs Oh the evaluation or the carcincamuC risk of chemicals to humans; vqf 23). International Agency for Research on Caocer (1583): "Miscellaneous Pesticides." Lyon: International Agency lor Research on Cancer (1ARC monographs on the evaluation of the carcinogenic risk oi chemical* 10 humans; Vol 3(1). IctCThitional Agency for Research on C*ncer (1956): " Some Halogcnaied Hydrocarbons and pesticide Exposures." Lyon: International Agency for Research on Cancer (IAKC monographs on the evaluation of the earvirtOgeriic risk of chemicals to humans; Vol 41), International Ageney for Research on Cancer (1987): "Overall Eva)nations of Carcinogenicity: An updating of LARC Mono^rapht Volumes 1 to 42." Lyon: International Agency for Research on Cancer (IaRC monographs on the evaluation of the carcinogenic risks to humans; suppl. No. 7). La Vecchis C, NTeri E, D*Avamo B, Franceschi S (1$$9): Occupation and lymphoid neoplasms. Bf J CttAcer #h3B5-388. Lyngc (1985): A follow-up study of cancer incidence among workers in manufacture of phenoxy herbicides ir. Denmark. Br J C^- 52^259-270. Lyngc E, Storm HH. 3enscn OM (1987): The evaluation of trends in soft 65*114 iiiconu according to diagAostic criteria and consumption of phenoxy herbicides. Cancer 60:1896--1901. McJ^ughlio JK. Linei MS, Stone Bf. Blot WJ, Fraumeni JF, M&lter HSR, Weiner JA, Ericsson 1LE (l9$S): Multiple myeloma and occupation in Sweden. Arch Environ Health 43:7-10. Ott MC. Holder BB, Gordon HL (1974): Respiratory cancer and occupations' exposure to araifticMs. Arch EnWron Health 29:250-255. Pearce N. Smith AH, Fisher DO (1985): Malignant lymphoma and multiple myeloma linked with agricultural occupations in &New Zealand cancer registry-based Study. Am I Epidemic! 121: 225-237. Pearce NE, Smith AH. Howard JK, Sheppard P*A, Giles H). Tcagu; CA (1936a): Non-Hodskin's lymphoma anJ exposure to phenoxyhcrbicitirt. chlwopbenols. fencing work, mid mc.V works employment: a cawt-control study. Br J ir.d Med 43:75-83. Pearce NE. Smith aH, Howard JK- Sheppard Ra , Giles HJ. CA (1986b): Case-control study of multiple myeloma ant! fanning. Br J Cancer 54:493-300.  i: } j: t Ak. ii -- -- 1 - b ------ r . * *. ' ' L; ? l, , l t ' , l . l L f ' ~ ' . " . I - -- * _r --------. l-> - - -* -~ - * -1- - -< -- J .. I s _ -T icl D ll *? - " 660 Hansen et aL Peirce NE, Shepfwd RA. Smith AH. Teague CA (19S7): Noo-Hodgkin's lymphoma nd farming: An expanded cose-control study, lm ) C>n.'er 39:I55-)61. Person B. Dahiandei A-M. Fredriki-son M. Bn;g; HN, OoLson CG. Atclsoo 0 1959): Malignant hrr.pbeitris and oocyps.tior.al exposures. 2r J Ind Med 46:5)6-520. Schumacher MC (19S5): Farming occupations a^d nvrf-j*Jfty from non Hodgkin4* lycr-fhonfla in Utah. J Occup Med 273S0-5M. Vincis P. Tcrrtvcinl B, Cicro*xe G. Cap^tt; A, Colombo E, Oootu A. MfTi L. fhs-a R Ricci P, Tahiti! E. Coroba P (1986); Pher.oxy herbicides and vaft-iiisue sarcomas in female rue feeders. Scand J Wort; Environ Health 13:9-1?. Vincis P. fas^isoo F, Tcdcsctii M, Ciccone G (1991): Incidence mte* of lymphomas and soft-tissue sarcomas ardc^vintfi.-jvjttnl rruasursnvcr.is of phcnnxv herbicides. J Nail CancerInu 83:362-363. Weisanburger DD (1990). EnvimrmienLa! epidemiology of nomHtvJgViit'i lymphoma !r. tssiern Ne braska. Am J Ind Med 18:303-305. Wjgle DT. Senxociw RM, Wilkins K. Riedel D. Ritter L, Morrison HI. Mao Y (1990): Monaliry study of Canadian male farm operators: Nofi Hod^kin's lymphoma morality md agricultural pmct>:e$ in Siskaichc>iaft. J Nad Cmee: Ir.st S2-_575-5&2. V/ikiund K, Holm L-E (19S6): Soft tissue catcema risk in Swedish afrieuliu'a! and forestry v.orkett. J Karl Cancc icy 76.229-234, WikJund K, I-jriiefors B-M. Holm L-E (19! Js): Risk of malignant lymphoma ir Swedish 3|ricelti)ral and forestry wotmti. Br J lnd Med 45:15-24. V/iUund K, t>isrh J. Holm L-E (I9$!b): Soft tissue sarcoma risk in Su-dish licensed peiricidc applicators J Occup Med 30:501-804. Wmgmn C. Fmdnk.son M, Brige HN. Ncrdaaskjuld B. Axelsor. 0 (1990): Soft tissue sarcoma aod occupations! exposures. Cancer 66:806-811. Woods JS. Pobssar L. Severson RK, Heuscr LS, Kulindrr BG (I9S7)- Soft tissue sarcoma and non Hodgkin's lymphoma in relation to pher.oxyhrrbicide and chlorinated phenol exposure in -cstem Washington. J Nad Carver Lnsi 7$ S99-910. Zahrr. SH, BUh A. Holmr.j FF, Boysen CD. Robe! R) (19SS): A ctse-referen* stud) of soft-tissue sarcoma and Hodgkin's disease. Scar'd ) Wort. Environ Health 14:224-230. Zahxn SH. Wci.seburger DD. Bobbin PA, SaaJ RC. Vrucht JB. Cantor KR. Blair A (1990)- \ cafe- control study of ron-Kodrkin's lymphoma and the herbicide 2.4-dichlorophenoxyftceiiea:id (2.4- D) in Nebraska. Epidmiiology 1:349-356.