Document k6kQJ1pqrQzmwO7Gbrzexjk1B

Cancer Causer and Control,4, 153 - 156 Pesticide exDosures and multide II myeloma in Iowa men Linda Morris Brown, Leon F. Burmeister, George D. Everett, and Aaron Blair (Received 22 October 1992; accepted in ret.iredform 21 December 1992) A population-based case-control study of 173 White men with multiple myeloma (MM) and 6% controls was conducted in Iowa (United States), an area with a large farming population, to evaluate the association between MM, agricultural risk factors. and exposure to individual pesticides. A slight nonsignificantly elevated risk for MM was seen among farmers (odds ratio [OR] = 1.2,95 percent confidence interval [CI] = C.8-1.7). Although slight excesses were observed, there were no significant associations between MM and handling either classes of pesticides o r specific pesticides. Thus, this study found little evidence t o suggest a n association between risk of MM and farming o r pesticides. Key words:Case-control study, epidemiology, etiology, farming, multiple myeloma, pesticides, United States. Introduction Excess risk for multiple myeloma (MM) has been associated with farming in a number of studies.'.' Although pesticides as a generic class of chemicals have been associated with MM, information on risk from use of individual pesticides is We analyzed detailed pesticide data from a case-control study among White men in Iowa (United States) to investi- gate whether risk of MM was related to the use of speci- fic agricultural pesticides. A previous analysis of these data included only classes of pesticides.' Methods Concurrent, population-based, case-control, interview studies of leukemia, non-Hodgkin's lymphoma (NHL), and MM in Iowa and leukemia and NHL in Minnesota were conducted during 1981-84 using the same questionnaire and controls. Presented in this paper are results for the MM cases and Iowa controls. Details of the pesticide analyses for leukemia and N H L have been published elsewhere.j.h Included in this study were all cases of MM among White men aged 30 or older diagnosed during 1981-84. Cases were identified from the Iowa Health Registry, a member of the Surveillance, Epidemiology, and End Results (SEER) program. To confirm the diagnosis of MM, pathologic material and laboratory reports were reviewed by an expert pathologist. Twenty-five cases of MM for whom adequate information for diagnostic confirmation were unavailable are not included in the analyses. Three sources of controls were used to select a popu- lation-based stratified sample of White men without lymphatic or hematopoietic cancer: random digit dialing [RDD] (living controls under age 65);- Medicare records provided by the Health Care Financing M s Brown and Dr Blairare with the Epidemiologj and Biostatistics Program, iVational Cancer Institute, Bethesda. MD, USA. Dr Burmeister is i;ith the Department of Preventive Medicine, bni-,.ersiri of IoGa, l o z a G t ) , IA, USA Dr Everett is with thc Department of Internal Medicine, Orlando Regional Medical Center, Orlando, FL. L'S.4 .4ddress correspondence to M s Brown, Epidemio1og)i and Biostatistics Program, ,Vattonal Cancer Institute, Executive Plaza North, Room 415, Berhesda, MD, USA. This project 'was supported in part bj a grant from the ?rational Institute of Environmental Health Sciences (ES03099) Cancer Cause\ and Control bo1 4 1993 153 L. hi. Brown et a1 Administration (living controls aged 65 or over); and state death certificate files (deceased controls). Controls were frequency-matched to the cases by five-year age group and vital status at time of interview. Since the study was designed to evaluate risks for farming and farm-related exposures, we did not match controls to cases o n geographic area. A standardized questionnaire was used to obtain detailed information on general farm activities and the use on the farm of24 animal insecticides, 34 crop insecticides, 38 herbicides, and 16 fungicides. Questions o n pesticide use included whether the subject personally mixed, handled, or applied the pesticide; whether the subject usually used protective equipment when handling the pesticide; and the first and last year the pesticide was used. Information on the frequency of pesticide use was not obtained. In-person interviews lasting approximately 50 min were conducted with subjects or with close relatives, if subjects were deceased. Interviews were completed for 84 percent of the MM cases and 78 percent of the controls. The response rate for the controls includes the 87.5 percent response rate for the RDD controls at the household screening phase. Included in the analysis were 173 (101 alive, 72 deceased) MM cases and 650 (452 alive, 198 deceased) controls. Logistic modelss were used co calculate odds ratios (OR) for MM from agricultural exposures and from individual pesticides that were handled personally by at least five cases. This restriction based on the much smaller case-group was made to assure adequate numbers of subjects in each cell for analysis. All ORs were calculated using nonf'armers as the referent group because they were not exposed to any farm-related Table 1. Risk of multiple myeloma according to number of years farmed and ever use of types of pesticides No. of No. of ORa cases controls CCI): Never farmed Ever farmed 62 272 111 378 1.2 (0.8-1 7 ) Farmed (yrs) 1-9 '0-29 30 - 44 45 Unknown 40 92 2.0 11.2-3 2) 38 l e 1 1.5 (1.0-241 21 93 1.o (0.6-1.7) 8 85 0.3 10.2-0.7) 47 Pesticides Fungicides Herbicides Insecticides 93 318 7 35 54 194 91 308 1.2 (0.8-1.8) 1.o (0.4-2.4) 1.2 (0.8- 1.9) 1 2 (0.9- 1.8) a All odds ratio (OR) values relativeto subjects who never farmed. All O R s adjusted for vital status and age in a logistic analysls. CI = 95% confidence interval Table 2. Risk of multiple myeloma for mixing, handling, or applying specific pesticides Pesticides No. of No. of ORa cases controls (CIP Animal insecticides Chlordane Coumaphos DDT Dichlorvos Lindane Malathion Nicotine 9 5 20 7 16 6 6 29 1.6 (0.7-3.6) 16 2.0 (0.7 5.8) ~ 84 1.1 (0.6- 1.9) 21 2.0 (0.8-5.0) 75 1.1 (0.6-2.0) 44 0.8 (0.3-1.9) 22 1.4 (0.5-3.6) Crop insecticides Aldrin Carbofuran Carbaryl DDT Fonofos Lindane Malathion Phorate Terbufos 22 12 6 20 5 5 8 11 10 89 1.1 (0.7 - 2.0) 49 1.2 (0.6-2.5) 20 1.5 (0.6-3.9) 52 1.7 (0.9-3.1) 22 1.4 (0.5-3.8) . 19 1.2 (0.4-3.4) 24 1.9 (0.8-4.6) 41 1.4 (0.7-3.1) 33 1.8 (0.8-4.0) Herbicides Alachlor Atrazine Bentazon Butylate Chloramben Cyanazine 2.4-0 Dicamba Glyphosate Metribuzen 2.4,s-T Trifluralin 13 73 0.9 (0.5- 1.7) 12 74 0.8 (0.4- 1.6) 9 38 1.3 (0.6-2.8) 8 36 1.3 (0.6-3.1) 6 47 0.6 (0.3- 1.6) 11 51 1.2 (0.6- 2.4) 35 150 1.o (0.6-1.6) 10 43 1.3 (0.6-2.8) 11 40 1.7 (0.8- 3.6) 7 36 1.2 (0.5-3.0) 7 39 0.9 (0.4-2.1) 17 69 1.2 (0.7-2.3) All odds ratios (OR) values relativeto nonfarmers(62 cases, 272 controls). All ORs adjusted for vltal status and age in a logistic analysis. CI = 95% confidence interval. activities. We felt that subjects who farmed bur reported no use ot pesticides might be musual and tarmers who did not use a specific pesticide might have been exposed to other potentially carcinogenic pesticides. Vital status (alive,deceased) and age (< 45,45-64. 65 +), were included in models to adjust for potential confounding. Other tactors such as smoking and education were svaluated and tound not to bc confounders of agricultural risk factors. Results Some farming activity was reported by 64 percent of the cases and 58 percent of the controls ( O R = 1.2; Table 1). The farming-associated risk was greater for subjects less than 65 years (OR = 1.4,95 percent confidence interval [CI] = 0.7-2.8) than for subjects 65 years 154 CmL.erC u s e s i n d Control \ o i 4 1993 Multiple myeloma and pesticides of a g e o r o l d e r ( 0 R = 1.1. CI = 0.7-1.7). Years engaged no significant associations between MM and the hand- in farming appeared to be inversely associated with risk ling of pesticides in general, classes of pesticides, or of MM (P for trend < 0.05),even though the median specific pesticides. Risks were not elevated for year first (1932) and last farmed (1968) were the same exposure to phenoxy herbicides as has been reported for cases and controls. Similar patterns in risk were for MM in a recent Swedish study' and for NHL in a seen for both younger (< 65 years) and older (65 + number of studies.r8-2T' he slightly elevated risks seen years) subjects (not shown). The O R s were not elev- for a number of animal and crop insecticides were often ated significantly among farmers reporting the use of greater among deceased than living subjects, suggest- any pesticide (OR = 1.2), any fungicide (OR = l.O), ing that other factors such as recall bias or chance, due any herbicide (OR = 1.2), or any insecticide to relatively small numbers of cases and controls or (OR = 1.2; Table 1). multiple statistical comparisons, mav play a role. Risks for MM were not increased significantly for This study, where over half of the subjects were farmers who personally mixed, handled, or applied any farmers, found little evidence of an association between specific insecticide or herbicide (Table 2 ) . Nonsignifi- risk af MM and farming. These data also do not pro- cantly elevated ORs of I .5 or greater were seen for the vide strong support for an association with specific animal insecticides chlordane, coumaphos, and agricultural chemicals. dichlorvos; the crop insecticides carbaryl, dichloro- diphenyl-trichloroethane (DDT), malathion, and ter- bufos; and the herbicide glyphosate. Failure to use pro- References tective equipment was associated with a higher risk of MM for chlordane ( O R = 1.9), coumaphos ( O R = 2.4), D D T ( O R = 1.8), terbufos ( O R = 2.1), and glyphosate ( O R = 1.9), but not for dichlorvos ( O R = 2.0), carbaTl ( O R = 1.4), and malathion ( O R = 1..I)R.isks were generally higher for deceased than alive subjects and for older than for younger subjects. k s k s were not elevated for subjects who handled the phenoxy herbicides 2,4-dichlorophenoxvacetic acid (2,4-D) and 2,4,5-trichlorophenoxyaceticacid (2,4,5-T). As previously reported for these data,' ORs for farmers exposed to any chemical class of pesticides (e.g., chlorinated hydrocarbon or organophosphate insecticides) were not significantly elevated. Discussion This study of White men in Iowa was designed to evaluate the association betweeen MM and agricultural risk factors, as well as exposure to individual pesticides. 1 . Blair ,4, Zahrn SH. Cancer among tarmcrs. OK .Wed: Stare o f r h e tZrr R e v i e w 1991; 6: 335-54. 2. Riedel DA, Potrern LM. The epidemiology of multiple myeloma. Hem/Oncol Clinic Xorrh America 1992; 6: 225-47. 3. Eriksson M, Karlsson M. Occupational and other environmental factors and multiple myeloma: a popu- lation based case-control study. B r ] Ind .Wed 1992; 49: 95-133. 4. Burrneister LF. Cancer in Iowa farmers: Recent results. .4m] Ind Med 1990; 18: 295-301. 5 . Brown LM, Blair A, Gibson R, et af. Pesticide exposures and other agricultural risk factors for leukemia among men in Iowa and Minnesota. Cancer Res 1990; 53: 6585-91. 6. Cantor KP. Blair A. Everett G. et af.Pesticide exDosures I~ ~I and other agricultural risk factors for non-Hodgkin's lymphoma among men in Iowa and Minnesota. C m c e r Res 1992; 52: 2447-55. 7. WaksbergJ. Sampling methods for random digit dialing. J Am Stat Assoc 1978; 73: 40-6. 8. Dixon WJ, ed. BMDP Statistical Software Manual, Vol- u m e 2. Berkeley, CA: University of California Press, ' There was a slight, nonsignificantly elevated risk for 1990: 1013-77. MM among farmers ( O R = 1.2), with the greatest risk 9. Pearce NE, Smith AH, Fisher DO. Malignant lym- seen for farmers of shortest duration (< 10 years, phoma and multiple myeloma linked with agricultural occupations in a New Zealand cancer registry-based O R = 2.0). In other studies where positive associations study. Am J Epzdemiol 1985; 121: 225-37. with farming have been reported, risks generally have 10. Brownson RC, Reif JS. A cancer registry-based study of been in the range of 1.2 to 2.2.13.' However, two recent occupational risk for lymphoma, multiple myeloma and studies conducted among White men in MissourilCand Nebraska" found no association between MM and leukemia. Inr J Epidemiol 1988; 17: 27-32. 1 1 . Zahm SH, Blair A, Weisenburger DD. Sex differences in the risk of multiple myeloma associated with agriculture. farming. Various farm-related exposures have been Br J Ind Med 1992; 49: 815-6. suggested as potential risk factors for Ivmphatic and 12. Pearce NE. Reif JS. Epidemiologic studies of cancer in hematopoietic cancers including pesticides, viruses, or agricultural workers. Am J Ind Med 1999; 18: 133-48. chronic antigenic stimulation.l2 Positive associations between MM and use of pesti- cides in general have been reported previously in some 13. Boffetta P, Stellman SD, Garfinkel L. A case-control study of multiple myeloma nested in the American Cancer Society prospective study. Inr ] Cancer 1989; 43: 554-9. StUdieS,>.I>-bIhut not in others." In our study, we found 14. Cantor KP, Blair A. Farming and mortality from mul- Cancer Causes and Control. V o l 4 1993 155 L. M.Brown et a1 tiple myeloma: a case-control study with the use of death certificates. JNCI 1894; 72: 25 1-5. 15. Flodin U, Fredriksson M,Persson B. Multiple myeloma and engine exhausts, fresh wood, and creosote: a casereferent study. A m I I n d Med 1987; 12: 519-29. 16. Morris PD, Koepsell TD, Daling JR, et al. Toxic substance exposure and multiple mveloma: a case-control study.JNCI 1986; 76: 987-94. 17. Pearce NE, Smith AH, Howard JK, Sheppard RA, Giles HJ, Teague CA. Case-control study of multiple myeloma and farming. B r ] Cancer 1986; 54: 493-500. 18. Hardell L, Eriksson M,Lenner P. Malignant lymphoma and exposure to chemicals, especially organic solvents, chlorphenols and phenoxy acids: a case-control study. BY] Cancer 1981; 43: 169-76. 19. Hoar SK, Blair A, Holmes FF, et af.Agricultural herbi- cide use and risk of lymphoma and soft-tissue sarcoma. j Am Med Assoc 1986; 2 5 6 1 141-7. 20. Wigle DT, Semenciw RM, Wilkins K, et af. Mortalitv study of Canadian male farm operators: non-Hodgkin's lymphoma mortality and agricultural practices in Saskatchewan. JNCI 1990; 82: 575-82. 21. Zahm SH, Weisenburger DD, Babbitt PA, et al. A casecontrol study of non-Hodgkin's lymphoma and the herbicide 2,4-dichlorophenoxyacetic acid (2,4-D) in Eastern Nebraska. Epidemiology 1990; 1: 349-56. 2 156 Cancer Causes and Control. Voi 4. 1993