Environmental Pesticide Exposure as a Risk Factor for Alzheimer\'s Disease: A Case-Control Study

Environmental Research Section A 86, 37}45 (2001) doi:10.1006/enrs.2001.4254, available online at http://www.idealibrary.com on

Environmental Pesticide Exposure as a Risk Factor for Alzheimer’s Disease: A Case-Control Study Eric Gauthier,* Isabel Fortier,- Franiois Courchesne,* Paul Pepin,- Jim Mortimer,? and Denis Gauvreau*DeH partement de GeH ographie, UniversiteH de MontreH al, MontreH al, QueH bec, H3L 3E3, Canada; -Alge` ne Biotechnologies Corp. Quebec, Canada; and ?Institute of Aging, University of Tampa, Tampa, Florida Received August 17, 2000

Some studies identi7ed a major decrease of choline acetyltransferase and acetylcholinesterase (AchE) in AD cases (4, 5). Others reported oxidative stress by free radicals (FR) as a risk factor for AD (6). Ikeda et al. (1994) (7) observed that the FR are highly toxic and can initiate lipid peroxidation whose level is increased in NFT. The etiology of AD has not yet been resolved. It has been suggested that AD could result from a multifactorial process involving both a genetic predisposition and an exposure to environmental factors modulated by the biological aging process (8). To date, AD has been linked to at least 7ve chromosomes. In the case of autosomal dominant presenile AD, three genes have been identi7ed: the 7rst, localized on chromosome 21 (9), encodes the amyloid precursor protein (10); the second is localized on chromosome 14 (11) and encodes presenilin 1 (12); and the third encodes presenilin 2 (13) and is localized on chromosome 1 (14). Along with these dominant genes, other genes are risk factors for AD. It was shown that the e4 allele of the apolipoprotein E (ApoE) gene, localized on chromosome 19 (15), was an important risk factor for senile (16, 17) and presenile (18) AD. Recently, a team of investigators established that chromosome 12 was linked to AD (19). In addition, many studies have revealed a strong association between family history and AD development (20). However, known genetic components account for only a portion of AD cases (21), suggesting that other genes or risk factors, such as environmental exposure, could be involved in the etiology of the disease. Epidemiological studies revealed a higher prevalence of AD in rural environments than in urban settings (22, 23). For many decades, pesticides have been used more intensively in rural than in urban areas and their use has been increasing with the growth of productivity and the specialization of

The aim of this study was to evaluate the inBuence of pesticide exposure on the development of Alzheimer’s disease (AD), taking into account the potentially confounding factors (genetic, occupational exposure, and sociodemographic). The 1924 study participants (>70 years old) were randomly selected in the Saguenay}Lac Saint-Jean region (Quebec, Canada). The AD diagnosis was established in three steps according to recognized criteria. Sixty-eight cases were paired with a nondemented control for age (ⴞ2 years) and sex. Structured questionnaires addressed to subjects and proxy respondents allowed a description of the sociodemographic characteristics, lifestyle characteristics, and residential, occupational, familial, and medical histories. Assessment of environmental exposure to pesticides was based on residential histories and the agriculture census histories of Statistics Canada (1971+1991) for herbicide and insecticide spraying in the area. Statistical analyses were performed with a logistic regression, adjusting for potential confounding factors. The results failed to show a signiAcant risk of AD with an exposure to herbicides, insecticides, and pesticides. However, future investigations are needed to establish more precisely the identiAcation, measurement, mobility, and bioavailability of neurotoxic pesticide residues in relation to AD.  2001 Academic Press

Key Words: Alzheimer’s disease; pesticides exposure; confounding factors; residential history. INTRODUCTION

Alzheimer’s disease (AD) is the fourth leading cause of death in the elderly (1). AD is a neurodegenerative disease de7ned by speci7c neuropathological and neurochemical features. The main lesions observed in the brain of AD cases are senile plaques (2) and neuro7brillary tangles (NFT) (3). 37

0013-9351/01 $35.00 Copyright  2001 by Academic Press All rights of reproduction in any form reserved.

38

GAUTHIER ET AL.

cultures (24). Many pesticides have known neurotoxic properties (25). Toxicological studies revealed that some pesticide families (carbamates, organophosphates, organochlorines, bipyridyles) can cause serious damage to the nervous system (26). The organophosphorous compounds such as parathion and fenitrothion and carbamates such as aldicarb are known inhibitors of AchE (27, 28). Bipyridyles such as paraquat can generate FR that are able to cross the blood}brain barrier and in8ict irreversible damage to neurons (29, 30). Organochlorines such as DDT and dieldrin can disturb the functioning of the mitochondrial system and produce FR (31, 32). Thus, exposure to pesticides may contribute to the cholinergic system de7ciency and the production of FR observed in AD. Epidemiological studies in the workplace have identi7ed some damage to the central nervous system and neuropsychological disorders due to pesticide exposure (33}36). Others have shown signi7cant relationships between exposure to pesticides and neurodegenerative diseases such as Parkinson’s disease (32, 37}39) and AD (40, 41). The Canadian Study of Health and Aging has estimated an odds ratio (OR) of 2.17 (95% CI, 1.18}3.99) associated to AD with respect to occupational exposure to pesticides (41). However, the interpretation of this result is limited because the study did not consider long-term environmental exposure to pesticides. In addition, occupational exposure to pesticides was established from answers provided by contacts to a simple binary question on whether the subjects were exposed or not exposed. This does not allow the quanti7cation of exposure. Finally, the Canadian study of Health and Aging (1994) (41) did not consider potential confounding factors such as age, sex, education level, occupation, and genetic factors that might in8uence the relationship between pesticides and AD. The objectives of this study were (1) to document the history of pesticides (insecticides and herbicides) used in the region of Saguenay}Lac Saint-Jean (SLSJ), Quebec (Canada), (2) to evaluate the longterm exposure to pesticides according to residential and occupational itineraries for a group of cases and controls, and (3) to evaluate the in8uence of pesticide exposure on AD development, taking into account the potentially confounding factors (genetic, occupational, and sociodemographic factors). MATERIALS AND METHODS

Population and Study Design This study is part of an epidemiological survey program to study relationships between environmental

and genetic risk factors in AD. It was conducted in the Saguenay}Lac Saint-Jean region in the province of Quebec (Canada). The interest of the SLSJ region for this kind of study stems from the fact that it is geographically isolated. As a consequence, the residents have been living in the region for most of their lives, facilitating the reconstitution of their residential history, the evaluation of long-term environmental and occupational exposure, and the reconstitution of genealogy for family cases. The 1924 study participants were randomly selected from the 7les of the provincial health plan of Quebec, and the sample was equally strati7ed according to age (70}74, 75}79, 80 years and over) and sex. AD diagnosis was established in three steps according to recognized criteria. The 7rst step involved screening for cognitive impairment and an overview of the health status of the participants. Neurofunctional evaluation included sensory tests and a brief cognitive assessment (3MS). Results of the 3MS (42) test were used to select subjects with a possible cognitive de7cit. Subjects who obtained a score of 78 or less on the scale of 100 of the 3MS test were considered as possibly cognitively impaired and went through the second step for an assessment of their cognitive abilities. Subjects with a score higher than 78 were considered free from cognitive impairment. The second step aimed to determine whether the subject had dementia. This evaluation included a clinical interview with a proxy respondent and a series of neuropsychological tests (CERAD battery, Benton’s test of verbal 8uency) (43) with the subject. The diagnosis of dementia was based on the overall results of the interviews and was made in accordance with established criteria (DSM IV) (44). Subjects considered demented according to these criteria were referred to the third evaluation step, the medical exam. This last step was used to identify the disease responsible for the dementia with NINCDSADRDA criteria for probable and possible AD (45). The interview was conducted by a neurologist and included an exam following the ICD10 criteria (46). Each case was paired for age ($2 years) and sex with a nondemented control and evaluated through the three phases of diagnosis. Structured questionnaires addressed to the subjects and to proxy respondents of cases and controls allowed a description of the sociodemographic characteristics, lifestyle characteristics, and residential, occupational exposure ( job title and duration; chemical type, intensity and frequency of exposure), familial, and medical histories. To assess the quality of data provided by proxy respondents, an analysis of agreement between the controls and their respondents was

PESTICIDE EXPOSURE AND ALZHEIMER’S DISEASE

conducted and a good concordance was observed for variables under study (47, 48). Blood samples were obtained from cases and controls for genetic genotyping following strict ethics procedures and the genotype of ApoE was determined for each subject with standardized laboratory procedures (16). Exposure to Pesticides The agroforestry industry is a major economic activity in SLSJ. Many neurotoxic pesticides (lead arseniate, calcium arseniate, DDT, phosphamidon, fenitrothion, Novathion, dimethoate, Matacil) have been used in forested areas since the beginning of the 1930s to 7ght spruce budworm and Swaine jack pine saw8y (49}52). Furthermore, the SLSJ region spans an area of about 23,000 km2 of which 1450 km2 are suitable for agricultural activities. Intensive monoculture has caused an increase in the use of pesticides to control weeds and insects (53). Historical studies on agriculture in SLSJ established that the use of pesticides started seemingly after the 1970’s with the development of agroindustries (large-scale production, monoculture, 7nancial support by government) (54). The data of the Agriculture Census of Statistics Canada for 1971, 1976, 1981, 1986, and 1991 and the data of the Quebec Department of Forestry were used for each municipality to establish the area sprayed with herbicides and insecticides (55, 52). The area sprayed was used as a potential indicator for environmental exposure to pesticides since it was closely linked to quantities applied and to the density of products used (56, 57). Potential exposure to insecticides and herbicides was estimated with Eq. (1) by combination of the subject’s residential history with the data on the area sprayed with pesticides in the municipalities of residence,

focused on exposure to pesticides for different time windows before the onset. This type of analysis is done by going back in time by multiyear intervals starting from the onset so that we can consider multiple environmental exposures of the controls and matched cases (58). Statistical analyses were performed with the SAS package (59). Logistic regression was used to estimate the relationship between exposure to pesticides and AD, taking into account different potential confounding factors such as education level, the presence of family cases, ApoE e4 allele, and occupational exposure. The education level was measured as a continuous variable but was entered in the regression models as a binary variable with a threshold at the seventh-grade level (grade school). The number of family cases among the 7rst-degree relatives was also considered. The presence or absence of at least one case of dementia, cognitive impairment, or AD in the family was considered as a binary variable in the regression analysis. The ApoE was included in the model as a binary variable: the presence of at least one ApoE e4 or no e4 allele. The occupational exposure to neurotoxic substances (solvent, lead, mercury, pesticides, aluminum, asbestos, silicates, etc.) was assessed. Occupational histories were obtained from proxy respondents for both cases and controls. Exposure was assessed by an occupational hygienist, blinded to the case or control status of each subject, using the occupational histories and the algorithm as a job index (60}63) and the job-exposure matrix of the National Institute for Occupational Safety and Health (64). The structure of the algorithm takes into account all available details about each job (the type of job and length, intensity, and frequency of exposure) reported by the informant and gives an estimate for each participant of a cumulative exposure

Pest (1971)x1#Pest (1976)x2#Pest (1981)x3#Pest (1986)x4#Pest (1991)x5 Exposure" , x G where, x is 1, if year of census4onset and the 

subject had lived in this municipality of residence; x is 0, otherwise; and Pest is exposure to insec 



ticides and herbicides for 1971, 1976, 1981, 1986, and 1991. For the cases, only the exposure during the period preceding the appearance of AD symptoms (onset) was considered. In the control group, we considered only the exposure before the onset of the case to which the control was matched. The analysis

39

(1)

assessment (CEA) to neurotoxic substances for their entire working life (65). The CEA provide four different global classes of probability of exposure (high, medium, low, and no exposure). Furthermore, the occupational exposure to pesticides was also evaluated by use of a standard job classi7cation index for workers who were exposed for at least 10 years (65). In the logistic regression analysis, the exposure to pesticides was de7ned as a binary variable, where

40

GAUTHIER ET AL.

the upper quartile of subjects was considered to have been exposed. We have selected the upper quartile of the subjects to decrease the probability of falsely considering a subject exposed (66). The use of a binary variable for exposure is more ef7cient than the use of a continuous variable since it increases the statistical power of the logistic regression (67). RESULTS

Description of the Study Population A total of 3826 names were selected in the SLSJ region. Of this number, 473 were dead at the time of recruitment, 1037 refused to participate, 274 could not be reached, and 118 lived outside the region. The total study population therefore consisted of 1924 individuals. In this sample, 122 cases of AD were diagnosed and each was matched to a control of the same age ($2 years) and sex for a total of 244 subjects for our study. Excluding pairs for which information was missing for one of the variables of interest (complete information on education level, family cases, ApoE genotyping, residential itinerary, time of onset), the 7nal sample consisted of 67 pairs of cases and controls. The sample was made up mostly of women (69%) and the majority of subjects were 580 years old (66%). Signi7cant relationships were observed between some variables and AD. We noted that an education level higher or equal to seventh grade had a tendency to lower the risk of AD (OR 0.55; 95% CI 0.27}1.09). We also observed a strong association between a family history of AD or other dementia in 7rst-degree relatives and the development of AD (OR 5.87; 95% CI 2.57}13.40). Furthermore, we observed that the presence of ApoE e4 was a risk factor for AD (OR 4.83; 95% CI 2.28}10.34). Occupational exposure to neurotoxic substances was not signi7cantly related (o"1.0) to AD. The same relation was observed for occupational exposure to pesticides and AD (o"0.81). Description of the Exposure to Pesticides The analysis of historical data on aerial spraying of insecticides on the SLSJ forest from 1930 to 1994 indicates that the pesticides did not seem to reach residential areas. In fact, the spraying of pesticides was performed on average at more than 50 km from residential areas and according to strict norms of application (maximum altitude of 150 m and maximum wind speed of 10 km/h) (52, 68). Studies on chemical insecticide derivatives outside treated areas indicate a maximum dispersion perimeter of

10 km from the point of application, in the direction of the dominant winds (69}71). Furthermore, studies on the persistence of residual pesticides and their major metabolites in the environment (streams, aquatic and terrestrial organisms) showed that the residual concentrations were around the detection limit 14 days after the application and that the maximum dispersion was 33 km from the treated area (72}74). Finally, the follow-up of pesticide residues in drinking water revealed that the concentrations were lower than the detection limit (75, 76). In this context, it is reasonable to assume that the subjects included in our study were not exposed to pesticides used in forestry. Data from Statistics Canada indicate that the use of pesticides in farming increased since 1971 (Fig. 1). Herbicides are mainly responsible for the increase in pesticide use and they represent a large portion of the total areas sprayed. In 1971, the total area sprayed with pesticides totaled 26.69 ha of which 56.3% were spayed with herbicides. From 1971 to 1991, the areas sprayed showed a 10-fold increase and herbicides accounted for more than 87% of the pulverization. Areas sprayed with insecticides increased from 1971 to 1981 to reach an average of 37.17 ha and stabilized from 1981 to 1991. Since 1971, the number of municipalities using pesticides almost doubled to reach more than 78% of all municipalities in the SLSJ area. Furthermore, the use of pesticides has largely progressed toward urban centers, hence potentially exposing a greater number of people. Data on pesticide sales in Quebec also show an increase in pesticide use since 1978 (77}79). More than 38 neurotoxic pesticides were used in Quebec since 1970 (80}82, 77}79). Risk Analysis The results for the long-term exposure to herbicides and insecticides (1971 to 1991) did not show any signi7cant effect on AD development (Table 1). The average exposures for people considered exposed and nonexposed were 75.61 ha ($18.68 ha) and 19.51 ha ($13.42 ha), respectively. From 1971 to the time of onset, the subjects lived on average in 1.3 ($0.6) different residences and stayed 15.17 years ($5.13 years) in each. Furthermore, the level of agreement between proxy respondents and controls for residential histories between 1971 and onset was 88.6% and increased progressively until onset, to attain 98.6%. We propose that the level of agreement between cases and their proxy respondents could be considered similar. According to Landis and Koch (47) and the Canadian Study of

PESTICIDE EXPOSURE AND ALZHEIMER’S DISEASE

41

FIG. 1.

Health and Aging (41), a level of agreement over 80% is considered good for statistical purposes. DISCUSSION

Our study did not establish a signi7cant relation between long-term exposure to pesticides, insecticides, or herbicides and development of AD. However, it is possible that the risk was underestimated. The following reasons could be involved. First, our study may include certain nondifferential biases because questionnaires were answered by proxy respondents. Individual exposure is based on residential

itinerary as established by the questionnaire. The dif7culties in use of retrospective data are well known, and the use of information supplied by proxy respondents may increase the degree of misclassi7cation bias. Furthermore, the method for determining the individual exposure is similar for cases and controls. Thus, the existence of nondifferential bias could result in an underestimation of the OR values (83). Second, the use of areas sprayed with pesticides as an index of exposure remains a global environmental indicator. Sprayed area values incorporated the areas sprayed with both neurotoxic and

42

GAUTHIER ET AL.

TABLE 1 Risk of Alzheimer’s Disease According to Exposure to Pesticides Exposure between 1970 and onset (n"68 pairs) Variables [threshold]

Unadjusted OR (CI 95%)

Adjusted OR* (CI 95%)

FIG. 2.

Herbicides [186 ha] Insecticides [54 ha] Pesticides [230 ha]

1.08 (0.49}2.37) 1.73 (0.79}3.78) 1.00 (0.45}2.21)

1.07 (0.39}2.54) 1.62 (0.64}4.11) 0.97 (0.38}2.41)

that the presence of ApoE e4 might induce a de7ciency in cell membrane regeneration and thus an inability to repair damages resulting from the toxic effects of FR originating from organochlorine pesticides such as DDT. In addition, ApoE e4, organophosphorous compounds, and carbamates might act together at the level of the cholinergic system de7cit. In conclusion, future investigations are needed to establish more precisely the identi7cation, measurement, mobility, and bioavailability of neurotoxic pesticide residues in relation to AD. Evaluation of exposure to domestic pesticides is also warranted. From a public health perspective, even if the results of this study are not signi7cant, we should encourage caution toward better control in pesticide use and conduct further investigations to study the dose}effect relationship of long-term pesticide exposure on AD.

Note. OR, odds ratio; (CI 95%), con7dence intervals with a signi7cance level of 95%; [threshold], the subjects de7ned as exposed were those found in the upper quartile of the subjects; pesticides, sum of insecticides and herbicides. * Adjusted for education level ((7, 57 years of education), presence of family cases (no, one, or more), and presence of at least one ApoE e4 allele.

nonneurotoxic products, which can reduce the OR. About 265 different active substances (54% herbicides, 20% insecticides, 26% others) were used in the SLSJ during the period between 1970 and 1991 (80}82, 77}79). Only 38 of them were considered neurotoxic. Among those 38 neurotoxic pesticides, 36 were insecticides (7 carbamates, 24 organophosphates, 5 organochlorines) and 2 were herbicides from the bipyridyle family. Third, when we compare unadjusted OR and adjusted OR, we observe that they are similar. The education level, the presence of ApoE e4, and the presence of family cases do not seem to be acting as confounding factors in the relationship between pesticide exposure and AD. In general terms, for a variable to confound a relationship, it must be associated with both the exposure and the risk of disease (Fig. 2). If there is no association between the exposure and the potential confounder, or conversely, if the potential confounder has no relationship with the risk of disease, then the variable cannot be interpreted as a confounding factor (84). However, considering that the exposure estimate is a global indicator including all the pesticides used, it is not impossible that these factors could potentially be confounding. For instance, it is plausible that certain pesticides interact with the same action site as ApoE e4. In fact, it is known that APoE e4 is normally produced in numerous tissues and is involved in the membrane lipid metabolism responsible for the repair and modulation of cell growth (85). Furthermore, researchers have suggested that ApoE e4 might be involved in the AchE de7cit in the development of AD (86). A plausible hypothesis is

ACKNOWLEDGMENTS We acknowledge the 7nancial support of the Alzheimer Society of Canada, the Fond FCAR, l’UniversiteH de MontreH al, the National institute of Health (NIH), and Algene Biotechnology Corp. We thank Alain Dufresne of les laboratoires LSD Inc. for technical help.

REFERENCES 1. Hill, G. (1994). De la recherche a` la pratique. Actes de l’atelier sur le deH 7 de la deH mence au Canada. Mal. Chron. Can. 15, S8}S9. 2. Duyckaerts, C., Hauw, J. J., Bastenaire, F., and Piette, F. (1986). Laminar distribution of neocartical senile plaques in senile dementia of Alzheimer type. Acta Neuropathol. 70, 249}256. 3. Ball, M. J. (1977). Neuronal loss, neuro7brillary tangles and granulovacuolar degeneration in hippocampus with ageing and dementia. Acta Neuropathol. 37, 111}118. 4. Davis, P., and Maloney, J. R. (1976). Selective loss of central cholinergic neurons in Alzheimer’s disease. Lancet 2, 1403. 5. Perry, E. K., Perry, R. H., Blessed, G., and Tomlinson, B. E. (1977). Necropsy evidence of central cholinergic de7cits in senile dementia. Lancet 1, 189. 6. Frolich, L., and Riederer, P. (1995). Free radical mechanisms in dementia of Alzheimer type and the potential for antioxidative treatment. Arzhneimittel-Forschung 45, 443}446.

PESTICIDE EXPOSURE AND ALZHEIMER’S DISEASE 7. Ikeda, Y., Toda, S., Wang, M., and Nakazawa, S. (1994). Early changes of blood}brain barrier and superoxide scavenging activity in rat cryogenic brain injury. Acta Neurochirurg. 60, 136}138. 8. Gauvreau, D. (1987). Le paradigme de la maladie d’Alzheimer. Interface 8, 16}21. 9. St-George-Hyslop, P. H. (1987). The genetic defect causing familial Alzheimer’s disease maps on chromosome 21. Science 238, 664}666. 10. Mullan, M., Houlden, H., and Windelspecht, M. (1992). A locus for familial early onset Alzheimer’s disease on the long arm of chromosome 14 proximal to the alpha1 antichymotrypsin gene. Nat. Genet. 2, 340. 11. Schellenberg, G. D., Bird, T. D., and Wijsman, E. M. (1992). Genetic linkage evidence for a familial Alzheimer’s disease locus chromosome 14. Science 258, 668}671. 12. Sherrington, R., Rogaev, E. I., and Liang, Y. (1995). Cloning of a gene bearing missense mutations in early- onset familial Alzheimer’s disease. Nature 375, 754}760. 13. Rogaev, E. I., Sherrington, R., Rogaev, E. A., Levesque, G., Ikeda, M., Liang, Y, Chi, H., Holman, K., Tsuda, T., Mar, L., Sorbi, S., Nacmias, B., Placentini, S., Amaducci, L., Chumakov, I., Cohen, D., Lannfelt, L., Fraser, P. E., Rommens, J. M., and St-George-Hyslop, P. H. (1995). Familial Alzheimer’s disease in kindreds with missense mutations in a gene on chromosome 1 related to the Alzheimer’s disease type 3 gene. Nature 376, 775}778. 14. Levy-Lahad, E., Wissman, E. M., Nemens, E., Anderson, L., Goddard, K. A., Weber, J. L., Bird, T. D., and Schellenberg, G. D. (1995). A familial Alzheimer’s disease locus on chromosome 1. Science 269, 970}973. 15. Pericak-Vance, M. A., Bebout, J. L., and Gaskell, P. C. (1991). Linkage studies in familial Alzheimer disease: Evidence for chromosome 19 linkage. Am. J. Hum. Genet. 48, 1034}1050. 16. BeH tard, C., Robitaille, Y., Gee, M., Tiberghien, LarriveH e, D., Roy, P., Mortimer, J. A., and Gauvreau, D. (1994). ApoE allele frequencies in Alzheimer’s disease, Lewy body dementia, Alzheimer’s disease with cerebrovascular disease and vascular dementia. NeuroReport 5, 1893}1896. 17. Strittmater, W. J., Saunders, A. M., Schmechel, D., PericakVnce, M., Enghild, J., Salvesen, G. S., and Roses, A. D. (1993). Apolipoprotein E: High-avidity binding to beta-amyloid and increased frequency of type 4 allele in late-onset familial Alzheimer disease. Proc. Natl. Acad. Sci. USA 90, 1977}1981. 18. Perez-Tur, J., Campion, D., Martinez, M., Brice, A., Tardieur, S., Hannequin, D., Agid, Y., Delacourte, A., Clerget-Darpoux, F., and Chartier Harlin, M. C. (1995). Evidence for ApoE e4 association in early-onset alzheimer’s disease patients with late onset relatives. Am. J. Med. Genet. 60, 550}553. 19. Stephenson, J. (1997). Researchers 7nd evidence of a new gene for late-onset Alzheimer disease. J. Am. Med. Assoc. 277, 775}776. 20. Van Djuin, C. M., Clayton, D., and Chandra, V. (1991). Familial aggregation of Alzheimer’s disease and related disorder: A collaborative reanalysis of case-control studies. Int. J. Epidemiol. 20, S13. 21. Van Broeckhoven, C. (1995). Molecular genetics of Alzheimer disease: Identi7cation of genes and gene mutations. Eur. Neurol. 35, 8}19.

43

22. Shibayama, H., Kasahara, Y., and Kobsayashi, H. (1986). Prevalence of dementia in japanese elderly population. Acta Psychol. Scand. 74, 144}151. 23. Jean, H., Emard, J. F., Thouez, J. P., Houde, L., Robitaille, Y., Mathieu, J., Boily, C., Daoud, N., Beaudry, M., Cholette, A., Bouchard, R., Veilleux, F., and Gauvreau, D. (1996). Alzheimer’s disease: Preliminary study of spatial distribution at birth place. Social Sci. Med. 42, 871}878. 24. Brundtland, G. H. (1989). ‘‘Notre Avenir a` Tous.’’ Commission mondiale sur l’environnement et le deH veloppement. Edition du Fleuve, Les publication du QueH bec. 25. Corriveau, L., and Lajoie, P. (1987). ‘‘Les Pesticides les plus UtiliseH s en Agriculture au QueH bec et la SanteH .’’ Recueil des donneH es toxicologiques et eH pideH miologiques, DeH partement de santeH communautaire du Centre Hospitalier de l’universiteH Laval. 26. Blain, P. G. (1990). Aspects of pesticide toxicology. Adverse Drug React. Acute Poison Rev. 9, 37}68. 27. Fukuto, T. R. (1990). Mechanism of action of organophosphorus and carbamate insecticides. Environ. Health Perspect. 87, 245}254. 28. Bardin, P. G., Van Eden, S. F., Moolman, A., Foden, A. P., and Joubert, J. R. (1994). Organophosphate and carbamate poisoning. Arch. Intern. Med. 154, 1433}1441. 29. Elroy-Stein, O., Bernstein, Y., and Groner, Y. (1986). Overproduction of human Cu/Zn-superoxide dismutase in transfected cells: Extenuation of paraquat-mediated cytoxicity and enhancement of lipid peroxidation. EMBO J. 5, 615}622. 30. Corasaniti, Mt., De7lippo, R., and Rodino, P. (1991). Evidence that paraquat is able to cross the blood brain-barriere to a different extent in rats of various age. Funct. Neurol. 6, 385}391. 31. Kimbrough, R. D., Gaines, T. B., and Linder, R. E. (1971). The ultrastructure of livers of rats fed DDT and Dieldrin. Arch. Environ. Health 22, 460}467. 32. Fleming, L., Mann, J. B., Bean, J., Thomas, B., and SanchezRamos, J. R. (1994). Parkinson’s disease and brain levels of organochlorine pesticides. Ann. Neurol. 36, 100}103. 33. Boyd, C. A., Weiler, M. H., and Porter, W. P. (1990). Behavioral and neurochemical changes associated with chronic exposure to low-level concentration of pesticide mixture. J. Toxicol. Environ. Health 30, 209}221. 34. Reidy, T. J., Bowler, R. M., Rauch, S. S., and Pedroza, G. I. (1992). Pesticide exposure and neuropsychological impairment in migrant farm workers. Arch. Clin. Neuropsychol. 7, 85}95. 35. Steenland, K., Jenkins, B., Ames, R. G., O’Malley, M., Chrislip, D., and Russo, J. (1994). Chronic neurological sequelae to organophosphate pesticide poisoning. Am. J. Public Health 84, 731}736. 36. Stephens, R., Spurgeon, A., Calvert, I. A., Beach, J., Levy, L. S., Berry, H., and Harrington, J. M. (1995). Neuropsychological effects of long-term exposure to organophosphate in sheep dip. Lancet 345, 1135}1139. 37. Barbeau, A., Roy, M., Cloutier, T, Plasse, L., and Paris, S. (1986). Environmental and genetic factors in the etiology of Parkinson’s disease. Adv. Neurol. 45, 299}306. 38. Semchuk, K. M., Love, E. J., and Lee, R. (1991). Parkinson’s disease and exposure to rural environmental factors: A population based case-control study. Can. J. Neurol. Sci. 18, 279}286.

44

GAUTHIER ET AL.

39. Butter7eld, P. G., Valanis, B. G., Spencer, P. S., Lindeman, C. A., and Nutt, J. G. (1993). Environmental antecedents of young-onset Parkinson’s disease. Neurology 43, 1150}1158.

55. Statistique Canada. (1995). ‘‘Enque9 te sur le Pro7l Agricole du SLSJ de 1931 a` 1992.’’ Base de donneH es con7dentielles, Statistique Canada, division recensements agricole.

40. Cannas, A., Costa, B., Tacconi, P., Pinna, L., and Fiaschi, A. (1991). ‘‘Dementia of Alzheimer Type (DAT) in a Man Chronically Exposed to Pesticides.’’ Institute of Neurology, Univ. of Calgary.

56. Godon, D., Lajoie, P., and Nadeau, D. (1987). ‘‘Atlas de l’Utilisation des Pesticides en Agriculture au QueH bec.’’ DeH partement de santeH communautaire du centre hospitalier de l’UniversiteH Laval.

41. Canadian Study of Health and Aging. (1994). Canadian Study of Health and Aging: Risk factors for Alzheimer disease in Canada. Can. Med. Assoc. J. 44, 2073}2080.

57. Westman, W. E. (1985). ‘‘Ecology Impact Assessment and Environmental Planning.’’ Wiley, New York.

42. Teng, E. L., and Chui, H. C. (1987). The modi7ed Mini-Mental State (3MS) examination. J. Clin. Psychiatry 48, 314}318. 43. Mirra, S. S., Heyman, A., and McKell, D. (1991). The consortium to establish a registry for Alzheimer’s disease (CERAD). Part II: Standardisation of the neuropathological assessment of Alzheimer’s disease. Neurology 41, 479}485. 44. APA. (1994). ‘‘Diagnostic and Statistical Manual of Mental Disorders: DSM-IV,’’ 4th ed. Am. Psychiatr. Assoc. Washington, DC. 45. McKhann, G., Drachman, D., Folstein, M., Katzman, R., Price, D., and Stadlan, E. M. (1984). Clinical diagnosis of Alzheimer’s disease: Report of the NINCDS-ADRDA work group under the auspices of department of health and human services task force on Alzheimer’s disease. Neurology 34, 939}944.

58. Moulton, L. H., and LE: , M. G. (1991). Latency and timedependent exposure in a case-control study. J. Clin. Epidemiol. 44, 915}923. 59. SAS Institute. (1996). SAS companion for the microsoft Windows environment: Version 6. SAS Institute, 2nd ed., Cary, NC. 60. CSST. (1988). ‘‘Renseignement sur les Contaminants Chimiques et Biologiques Re` glementeH s.’’ Commission de la santeH et de la seH curiteH du travail, Service du reH pertoire toxicologique, Direction du soutien scienti7que et technique, 3 Vol. 61. Danse, I. R. (1991). ‘‘Common Sense Toxics in Workplace: A Manual for Doctors, Nurses, Emergency Responders, Employers, Industrial Hygienists, Risk Managers, Claims Adjusters and Lawyers.’’ Van Nostrand Reinhold, New York. 62. ACGIH. (1996). ‘‘Guide to Occupational Exposures Values;1996.’’ American Conference of Governmental Industrial Hygienists, Ohio.

46. WHO. (1993). ‘‘The ICD-10 Classi7cation of Mental and Behavioural disorders, Diagnostic Criteria EOR Research.’’ WHO, Geneva.

63. O’Donoghue, J. L. (1985). ‘‘Neurotoxicity of Industrial and Commercial Chemicals,’’ 2 Vol. CRC Press, Boca Raton, FL.

47. Landis, J. R., and Koch, G. C. (1977). The measurement of observer agreement for categorical data. Biometrics 33, 159}174.

64. Johnsen, H. (1992). ‘‘Occupational Neurotoxicity: Criteria Document for Evaluation of Existing Data.’’ Nordic Council of ministers of Denmark, National Institute of Occupational Health.

48. Chong, J. P., Haines, T., Farnworth, H., and Julian, J. (1989). Concordance of occupational and environmental exposure information elicited from patients with AD and surrogate respondents. Am. J. Indust. Med. 15, 73}89.

65. WHO. (1986). ‘‘Principles and Methods for Assessment of Neurotoxicity Associated with Exposure to Chemicals.’’ World Health Organization, Hygiene criteria 60.

49. MTFQ (1953). ‘‘Rapport sur la Protection des Fore9 ts dans la Province de QueH bec Durant l’anneH e 1951.’’ Ministe` re des terres et fore9 ts du QueH bec (MTFQ), Service de la protection.

66. Jenicek, M., and CleH roux, R. (1987). ‘‘ED pideH miologie: Principes, Techniques et Applications.’’ Edisem, Inc.

50. MTFQ (1939). ‘‘Les Insectes Forestiers du Que& bec en 1938.’’ Ministe` re des terres et fore9 ts du QueH bec (MTFQ), Service d’entomologie forestie` re, Bull. No. 3.

68. Sundaram, K. M. S. (1995). Devenir des insecticides dans l’environnement forestier. In ‘‘Insectes Forestiers Ravageurs au Canada’’ (J. A. Armstrong and W. G. H. Ives, Eds.), pp. 551}572. Ressources naturelles Canada, Service canadien des fore9 ts, Direction des sciences et du deH veloppement durable.

51. McLeod, J. M. (1969). ReH sultats de traitements au phosphamidon, fenitrothion et Novathion contre la tenthre` de de Swaine, Neodiprion swainei Middleton, au QueH bec en 1967. Ann. Soc. Entomol. QueH bec 14, 55}68. 52. Dorais, L., Auger, M., Pelletier, M., Chabot, M., Bordeleau, C., and Cabana, J. (1995). La reH pression des insectes au QueH bec, 1974}1987. In ‘‘Insectes Forestiers Ravageurs au Canada’’ (J. A. Armstrong and W. G. H. Ives, Eds.), pp. 668}678. Ressources naturelles Canada, Service canadien des fore9 ts, Direction des sciences et du deH veloppement durable.

67. Hosmer, D. W., and Lemeshow, S. (1989). ‘‘Applied Logistic Regression.’’ Wiley, New York.

69. Marotte, P. M. (1984). ‘‘Evaluation Qualitative et Quantitative des DeH pots d’Insecticides.’’ Ministe` re de l’eH nergie et des ressources, Direction de la conservation, service des eH tudes environnementales, rapport 26.

53. Savard, M. (1989). ‘‘Pour que Demain Soit: L’ED tat de l’Environnement au Saguenay-Lac-St-Jean, pour un DeH veloppement Durable.’’ Le Griffon d’argile.

70. Marotte, P. M., Major, L., and Maisonneuve, C. (1995). Surveillance des reH sidus d’insecticide dans le cadre des programmes opeH rationnels de protection contre les ravageurs forestiers. In ‘‘Insectes Forestiers Ravageurs au Canada’’ (J. A. Armstrong and W. G. H. Ives, Eds.), pp. 574}587. Ressources naturelles Canada, Service canadien des fore9 ts, Direction des sciences et du deH veloppement durable.

54. Gauthier, M. (1975). L’Agriculture au Saguenay-Lac-St-Jean (P.Q.);ED tude GeH ographie. Travaux GeH ographiques du Saguenay, No 1., Centre de recherche du moyen-nord, DeH partement de geH ographie, UniversiteH du QueH bec a` Chicoutimi.

71. Marotte, P. M., Delisle, S., Dostie, R., and Parent, G. (1985). ‘‘Etude de la DeH rive des Insecticides Chimiques ‘l’ExteH rieur des aires TraiteH es.’’ Ministe` re de l’eH nergie et des ressources, Direction de la conservation, service des eH tudes environnementales, rapport 46.

PESTICIDE EXPOSURE AND ALZHEIMER’S DISEASE 72. Morin, R., Gaboury, G., and Dostie, R. (1982). ‘‘ReH sidus de FeH nitrothion et d’Aminocarbe dans l’Environnement.’’ Ministe` re de l’eH nergie et des ressources, Direction de la conservation, service des eH tudes environnementales, rapport 30. 73. Mamarbachi, G., Dostie, R., and Marotte, P. M. (1985). ‘‘Persistance des ReH sidus d’Aminocarbe et de Fenitrothion dans les Eaux Naturelles en Milieu Forestier.’’ Ministe` re de l’eH nergie et des ressources, Direction de la conservation, service des eH tudes environnementales, rapport 45. 74. Mamarbachi, G., Marotte, P. M., and Dostie, R. (1987). ‘‘ReH sidus de FeH nitrothion, d’Aminocarbe et de Leurs MeH tabolites dans des Organismes ExposeH s aux PulveH risations AeH riennes Contre la Tordeuse des Bourgeons de l’ED pinette au QueH bec en 1983 et 1984.’’ Ministe` re de l’eH nergie et des ressources, Direction de la conservation, service des eH tudes environnementales. 75. Delisle, S., and Dostie, R. (1986). ‘‘Concentrations d’Insecticides Chimiques deH tecteH es dans l’Eau Potable des Bases d’OpeH rations en 1985’’. Ministe` re de l’eH nergie et des ressources, Direction de la conservation, service des eH tudes environnementales, rapport 59. 76. MEF. (1994). ‘‘Base de DonneH es sur la QualiteH de l’Eau Potable.’’ Ministe` re de l’environnement et Faune du QueH bec, Direction des politiques du secteur municipale, 7chier informatique. 77. St-Jean, R., and PareH , J. (1980). ‘‘Pesticides EmployeH s en Agriculture au QueH bec en 1978.’’ Ministe` re de l’environnement du QueH bec, Bureau d’eH tude sur les substances toxiques.

45

78. Reiss, R., Perron, F., PareH , J., and St-Jean, R. (1984). ‘‘Les Pesticides en Agriculture au QueH bec en 1982.’’ Environnement Canada et Ministe` re de l’environnement du QueH bec. 79. Gorse, I., Bouchard, L., PareH , J., and Veillette, C (1995). ‘‘Pesticides: Bilan des Ventes de Pesticides au QueH bec en 1992.’’ Ministe` re Environnement et de la Faune du QueH bec, Direction des politiques des secteurs agricole et naturel. 80. CCHST. (1996). ‘‘Base de DonneH es CD-ROM sur la Toxicologie des Pesticides HomologueH s au Canada.’’ Centre Canadien d’hygie` ne et de seH curiteH au travail (CCHST), disk CCINFO. 81. MAPAQ. (1994). ‘‘ReH pertoire des Pesticides RecommandeH s par le Ministe` re de l’Agriculture, des Pe9 cheries et de l’Alimentation du QueH bec.’’ Ministe` re de l’agriculture, des pe9 cheries et de l’alimentation du QueH bec(MAPAQ), Direction de l’environnement et du deH veloppement durable. 82. MEF. (1992). ‘‘Bilan des Ventes de Pesticides au QueH bec en 1992.’’ Environnement et Faune QueH bec, Direction des politiques des secteurs agricole et naturel. 83. Bernard, P. M., and Lapointe, C. (1987). ‘‘Mesures Statistiques en ED pideH miologie.’’ Presses de l’UniversiteH du QueH bec. 84. Hennekens, C. H., and Buring, J. E. (1987). ‘‘Epidemiology in Medicine.’’ Little and Brown, Boston. 85. Berr, C. (1995). Apolipoproteine E et maladie d’Alzheimer. Rev. Neurol. (Paris) 151, 3}5. 86. Poirier, J., Delisle, M. C., Quirion, R., Aubert, I., Farloe, M., Lahiri, D., Hui, S., Bertrand, P., Nalbantoglu, J., and Gil7x, B. M. (1995). Apolipoprotein E4 allele as a predictor of cholinergic de7cits and treatment outcome in Alzheimer disease. Proc. Nat. Acad. Sci. USA 92, 12260}12264.