SCOPE 49 - Methods to Assess Adverse Effects of Pesticides on Non-target Organisms

 7.1 INTRODUCTION

The use of pesticides in industrialized and developing nations has led to concern over the effects of pesticide exposures on humans and other NTOs. There is little doubt that such exposures can result in illness or death. The World Health Organization (WHO) has estimated that approximately 500 000 human poisonings occur each year worldwide (WHO, 1981) and the number of fatalities has been estimated at greater than 10 000 per year (Loevinsohn, 1987). Although epidemiological data indicate that the primary hazards of exposures to pesticides are acute toxic reactions, these exposures have also been implicated as a possible risk factor for cancer in agricultural workers (American Medical Association, 1988).

The goal of this chapter will be to examine existing methods and approaches for assessment of exposure of humans and other NTOs to pesticides. In general, there are three principal sources of human exposure to pesticidesoccupational (including agricultural), non-occupational (for example non-commercial treatments of dwellings and workplaces, home gardening, etc.), and dietary, that is indirect exposure of the population through the use of pesticides on agricultural commodities. Each of these sources of exposure will be discussed separately, and the methods for assessing each type of exposure will be examined.

7.2 OCCUPATIONAL EXPOSURES TO PESTICIDES

The major types of occupational exposures, in terms of the magnitude and frequency of exposure, are agricultural (mixers, loaders, applicators, and harvesters), professional pesticide applicators who treat dwellings and workplaces, and workers in pesticide manufacturing plants. In general, these types of exposures are regulated by national agencies, who set permissible limits on exposure and determine whether a particular use has an adequate margin of safety for the worker.

Two principal approaches are currently employed for evaluating occupational exposures to pesticidespassive dosimetry and biological monitoring. A third area of monitoring is `physiological monitoring', and involves the observation of exposed populations for changes in biological/physiological parameters. Each approach has significant strengths and weaknesses, and no single approach will provide a complete estimate of exposure.

7.2.1 PASSIVE DOSIMETRY

The most common passive type of method used to estimate occupational exposure is a combination of patches on the skin and clothing and hand washes to estimate dermal exposure, and air sampling to estimate inhalation exposure (USEPA, 1987a; Reinert et al., 1986; Durham and Wolfe, 1962). In general, absorbent patches are placed on the skin in sufficient number and location as to estimate total dermal exposure. Patches are also frequently placed on the outside of garments to assess the effectiveness of protective clothing. After exposure to the pesticide under conditions of field use, the patches are removed and analysed for pesticide content. Exposure to the hands can be estimated by swabbing, hand washes, or absorbent gloves. Since the major portion of occupational pesticide exposure occurs to the hands, the method of assessment of this type of exposure is critical to the overall estimate of exposure. No single technique is fully satisfactory for determining exposures to the hands, as washing will not recover residues that are absorbed into skin (Wester and Maibach, 1985), and absorbent gloves may significantly overestimate exposure to the hands, because the gloves absorb or trap more pesticide than would be found in contact with bare skin (Reinert et al., 1986; Davis et al., 1983).

A variety of techniques exist for measuring inhalation exposure, ranging from modified respirators, which contain an absorbent material, to powered air sampling devices placed near the breathing zone of the monitored individual (Lewis, 1976). The choice of a method by which to estimate inhalation exposure will depend in large part on the physicalchemical nature of the specific pesticide and the specific activity of the monitored individual.

7.2.2 BIOLOGICAL MONITORING

This approach involves the direct measurement of pesticides and/or metabolites in blood, urine, and occasionally tissues (Coye et al., 1986a; Reinert et al., 1986). It is, therefore, theoretically possible to estimate the actual absorbed dose, providing that sufficient information on the pharmacokinetics of the specific chemical is available.

Another common approach in this area is to monitor some physiological or biochemical parameter as an index of exposure, for example measurement of decreases in serum or erythrocyte acetylcholinesterase activity after exposure to organophosphate insecticides. Although frequently referred to as `biological monitoring', this approach might be more correctly categorized as `physiological monitoring' or `health surveillance', since a response rather than actual biological dose is measured. For example, measurement of serum (pseudo-) cholinesterase activity is a common method for estimating exposure to organophosphate insecticides. Even though one may infer exposure by this technique, it is difficult to estimate the dose that an individual has received, and one could not identify the specific chemical involved by this technique alone. In addition, because of the wide variability in blood cholinesterase activity among humans, the utility of this technique depends on accurate pre-exposure baseline measurements in the monitored individual.

Other physiological techniques that have been used to assess pesticide exposure include monitoring of peripheral blood for chromosomal abnormalities and assessment of pupillary reflexes. Immunological techniques such as enzyme-linked immunosorbent assays (ELISA) hold promise as reliable, rapid, and inexpensive methods for `real-time' personal monitoring of pesticide exposures under field conditions (Hammock et al., 1987; Mumma and Brady, 1987). Although widely used in laboratories for many years, recent technical advances in this area have resulted in products with potential application for monitoring field exposure. Such techniques may be particularly useful in developing countries or rural areas where the sophisticated laboratory equipment necessary for chemical measurements of pesticides is not readily accessible. At least one such product has been developed for paraquat and parathion (`QUIK-CARD', Granite Division, Environmental Diagnostics, Inc., Burlington, North Carolina, USA). The card contains an indicator which will change colour in the presence of the particular chemical, and is a rapid, semi-quantitative method for screening exposures to these two highly toxic pesticides. In addition to monitoring personal exposure, immunological techniques may also provide economical methods for monitoring field residues to prevent worker re-entry hazard, and to monitor agricultural runoff so as to prevent surface water contamination.

7.2.3 PASSIVE DOSIMETRY VERSUS BIOLOGICAL MONITORING 

The major advantage of passive techniques is that the contribution of inhalation and dermal routes, and of the different areas of the body to total dermal exposure, can be independently estimated (Reinert et al., 1986). In addition, activities which produce a high potential for exposure can be identified. Over time, a database can be developed that will allow prediction of exposure under a given set of conditions. The major disadvantage of this technique is that the actual dose is unknown, and assumptions must be made to estimate dermal penetration and extrapolate from patches to total body surface area. In addition, patch measurements seem to have inherent limitations in terms of reproducibility (Fenske et al., 1987).

Biological monitoring has the advantage of potentially allowing for the measurement of the actual dose of a pesticide. The disadvantages of this type of approach are that some knowledge of the metabolism and pharmacokinetics of the pesticide is required, and it is generally not possible to distinguish between discrete work activities as sources of exposure. Even when pharmacokinetic data are known, extrapolations to estimate exposure can be confounded by extraneous variables such as concomitant drug or chemical exposure (Borm and de Barbanson, 1988).

Physiological monitoring and health surveillance suffer from similar limitations, in that it is not possible to quantify the dose or identify discrete work activities that produce exposures. The most relevant types of measurements generally require that a pre-exposure baseline be established and the exposure monitored through sequential measurements (Richter et al., 1986).

None of these techniques alone can provide a complete estimate of exposure, and the use of any of these methods to estimate exposure will require certain assumptions. Combining passive dosimetry and biological monitoring, although costly in terms of resources and time, may provide the best overall estimate of occupational exposure and thereby potential risk (Chester et al., 1987; Grover et al., 1986; Carmen et al., 1982; Leavitt et al., 1982).

7.2.4 METHODS FOR REDUCING OCCUPATIONAL PESTICIDE EXPOSURE

In general, absorption through the skin is the most significant route of occupational exposure to pesticides, at least during outdoor operations (USEPA, 1987b; Grover et al., 1986; Leavitt et al., 1982; Atallah et al., 1982). Of all areas of the skin, the hands generally receive the greatest exposure for most occupational activities involving pesticides. Depending on the specific pesticide and occupational activity, estimates of exposure through the hands range from 25 to 98 per cent of total dermal exposure (USEPA, 1987a; Reinert et al., 1986; Grover et al., 1986; Putnam et al., 1983; Leavitt et al., 1982). Therefore, by simply instituting the use of protective gloves that are impermeable to the pesticide, exposure can be greatly reduced. Other types of protective clothing such as long-sleeve shirts, pants, head gear and face shields may also significantly reduce exposure (Putnam et al., 1983; Atallah et al., 1982; Davies et al., 1982). In indoor operations, such as greenhouse pesticide applications, and for outdoor applications of certain chemicals, the inhalation route of exposure may be most significant, since for many chemicals the rate of absorption through the lung is likely to be higher than the rate of absorption through the skin, even though the total dermal dose is higher (Fenske et al., 1987). The use of a respirator in addition to other protective gear would be indicated in those situations.

Although protective clothing can greatly decrease dermal exposure to a pesticide, care must be taken to launder clothing properly to prevent saturation of the clothing material with a pesticide and subsequent chronic dermal exposure. In addition, the type of clothing material seems to have an effect on pesticide retention. For most situations, cotton material appears to be desirable in terms of protective efficacy and ease of cleaning balanced against the need for prevention of heat stress in the wearer (Chiao-Cheng et al., 1988; Davies et al., 1982). Washing the skin after pesticide exposure may not fully remove the chemical, reiterating the need to prevent exposure through appropriate protective measures (Wester and Maibach, 1985).

The level of occupational exposure is also directly dependent on the type of activity involved. For example, individuals who direct aerial spray operations from the ground (so-called `flaggers' or `flagmen'), may receive a dermal dose of pesticide that is several thousand times higher than other workers, such as mixer-loaders or applicators (Atallah et al., 1982). The use of a mechanical device would eliminate the need for this type of exposure.

The method of mixing or applying the pesticide also has a significant effect on the level of exposure to the mixer/loader/applicator and other NTOs. Pouring rather than pumping a pesticide formulation from a mixing tank into the application apparatus (i.e. open versus closed systems) can greatly increase exposure. Depending on the specific chemical and type of activity, the use of a wettable powder as compared to an emulsifiable concentrate may also increase exposure to the worker due to the generation of dusts which may be inhaled by the worker or drift onto clothing and other areas (Putnam et al., 1983). Similarly, the use of closed-cab ground application systems can greatly limit exposure to the applicator (Cowell et al., 1987; Carmen et al., 1982). Airblast or aeroplane spraying operations have the greatest potential for drift of the pesticide away from the application area, resulting in exposures in areas where pesticide application was not intended. Thus, minimization of this type of application in favour of ground-based methods can also reduce the extent of inadvertent exposure to NTOs. In terms of long-term impact on worker health, reduction of exposures through engineering controls may prove superior to personal protective measures.

7.3 NON-OCCUPATIONAL PESTICIDE EXPOSURES

Although a significant effort has been devoted to the quantification of occupational exposures, particularly in the agricultural sector, relatively little information exists on the overall level of pesticide exposure in the general population. In the US, attempts are under way to evaluate incidental homeowner exposures to pesticides through the EPA's Non-Occupational Pesticide Exposure Study (NOPES) (USEPA, 1987b). The monitoring methods for this study involve the use of personal air monitors, analysis of drinking water, and gloves (to assess dermal hand exposures) in cases where household applications of pesticides were involved (Lewis et al., 1988; Lewis, 1988; USEPA, 1987b). Although the data from the full study are still being analysed, the results of the pilot study suggest that incidental pesticide exposures in and around the home (at least in the US) are below the threshold of toxicological concern. Limited experimental data in this area also suggest that normal homeowner use of pesticides produces exposures well below those encountered in agricultural uses (Weisskopf et al., 1988; Everhart and Holt, 1982). The use of highly persistent organochlorine insecticides in dwellings and on food has created a great deal of interest in many countries regarding the body burden of these chemicals in the general population. In the US, the National Health and Nutrition Examination Survey (NHANES II) evaluated adipose and blood samples for residues of pesticides and other substances. More than 90 per cent of the adipose samples had detectable residues of DDT, chlordane, heptachlor epoxide, dieldrin, BHC, and HCB, which exist in equilibrium with fluid compartments. The median concentrations of these compounds in serum ranged from 1.4 p.p.b. for oxychlordane and trans-nonachlor to 11.8 p.p.b. for DDE (Murphy and Harvey, 1985; Murphy et al., 1983).

Evaluation of human adipose tissue for mirex, an organochlorine insecticide used in the southern US to control fire ants, indicated that about 10 per cent of the population in that region would be predicted to carry residues of mirex of about 0.3 p.p.m. in body fat (Kutz et al., 1985). Compilations of data for tissue and/or serum residues of organochlorine compounds suggest that levels are quite variable from region to region, and definite trends in the human body burden over time are difficult to establish. The variation in detected residue levels may be the result of differences in analytical techniques among laboratories, but undoubtedly also reflects variations in usage among the different regions. In the US, data compiled from sampling of adipose tissues obtained over the years 1970-1977 indicated that total adipose DDT + metabolite levels have been declining, from around 8 p.p.m. in 1970 to around 4 p.p.m. in 1977. However, the frequency of detection of DDT and/or its metabolites in adipose tissue remained constant over these years at 100 per cent. Chlordane, heptachlor, transnonachlor, and dieldrin concentrations remained fairly constant, although at much lower concentrations than for DDT (around 0.1 p.p.m.). As for DDT, the frequency of detection of these other organochlorine compounds approached 100 per cent in most cases. Experimental studies in man, and computer modelling of residues suggest that adipose concentrations of these compounds should slowly decline over time in the absence of further usage. Obviously, in areas where these compounds are still used, no reductions would be expected (Strassman and Kutz, 1981; Matsumara, 1985).

One area of concern related to the persistence of these compounds in the human body has been the transfer of organochlorine insecticides (and other persistent chemicals such as PCBs) to the infant via the placenta and breast milk. Lipophilic pesticides such as the organochlorines will transfer from adipose tissue to milk fat during lactation, and the levels in either medium are comparable (Travis et al., 1988; Skaare et al., 1988). These insecticides can potentially translocate to the fetus also, as evidenced by chemical residues detected in cord blood (Skaare et al., 1988). Although it is generally difficult to identify the source of exposure, at least one report has correlated the use of chlordane in termite treatments of dwellings with the appearance of chlordane in human milk (Taguchi and Yakushiji, 1988).

Monitoring of adipose tissue or other biological samples from humans provides valuable information regarding the total exposure of the population to certain pesticides; however, it is not possible to determine the route or circumstance by which that exposure occurred. These data illustrate the need for managing and assessing exposures as they occur, since residues may persist in the environment, food chain, and human body long after the use of the chemical has been discontinued.

7.4 DIETARY EXPOSURES TO PESTICIDES

In terms of the number of people potentially exposed to pesticides, the dietary route is the most significant route of exposure, even though the magnitude of such exposure is generally lower than that resulting from occupational or household use of pesticides. Virtually everyone receives at least occasional low doses of pesticide via the diet. The entry of pesticides into the diet results primarily from the use of these chemicals in agricultural production, but may also result from treatments of warehouses, food-processing and food-preparation facilities, and households. A variety of methods exist for estimating dietary exposures to pesticides; however, all existing methods are hampered by the relative lack of accurate data on residues of pesticides in food at the time of consumption.

In the US, dietary exposure estimates are derived from the EPA Tolerance Assessment System (TAS). This system is a computerized database that contains consumption estimates derived from the 19771978 USDA Nationwide Food Consumption Survey. The TAS can estimate exposures to the average population, or to any of 22 subgroups of that population based on age, sex, ethnicity, geographic region, or season of the year. Any valid residue estimate can be used to estimate exposure (Saunders, 1987; Saunders and Petersen, 1986).

Historically, estimation of dietary exposure in the US was based on an estimate of residue in food, called the `tolerance'. The tolerance is actually an enforcement value, and represents the upper limit of residue that is permitted on a crop as it enters commerce. (The tolerance is similar in concept to the maximum residue limit, MRL, used by the WHO/FAO for recommendation to the CODEX.) By definition, the residues on food at the time of consumption will be less than the tolerance or MRL value, and in fact data submitted to the EPA generally demonstrate that actual residue values in food as eaten are far below the tolerance value. When actual residue levels are used to estimate exposure, the calculated dietary risk for a particular chemical may be several orders of magnitude less than that calculated from tolerances (Petersen and Chaisson, 1988; Saunders, 1987).

For this reason, the EPA has developed the concept of `anticipated residue', which may be operationally defined as any residue value other than a tolerance which is used for purposes of calculating exposure and estimating risk. In practice, this value may be calculated by applying correction factors to the tolerance to account for the per cent of crop treated with a chemical (generally less than 100 per cent of a crop is treated with any single chemical), by using field-trial data to calculate an average field residue, and by applying processing factors to the mean field residue (preferably) or to the tolerance if adequate field-trial data are not available to estimate a valid mean (Saunders, 1987). Processing factors can take into account activities such as washing, peeling, canning, and cooking, which will generally decrease residues of pesticide in food at the time that it is consumed. If the application of appropriate factors still results in an unacceptable level of exposure, monitoring studies are conducted to determine the actual levels of chemical present in representative foods at the consumer level.

For most chemicals, the effects that may be of potential human health concern are identified in chronic feeding studies conducted in experimental animals. In assessing dietary exposures to these chemicals, the appropriate measure is an estimate of the average daily intake of the chemical. However, certain pesticides, notably some carbamate insecticides, are acutely toxic, and do not have chronic toxicity per se. In these cases, the relevant measure of dietary exposure is not average daily intake, but rather the maximum amount that an individual can consume in a single day. The TAS can calculate this value by determining the distribution of single-day exposures. In this manner, the regulator can determine whether a portion of the population will be at risk due to acute dietary exposure to the pesticide. As was discussed previously, tolerances, anticipated residues, and/or monitoring data can be used to calculate acute dietary exposures.

WHO is currently in the process of developing a computerized model for predicting dietary exposures to pesticides. This model will be based on Food and Agricultural Organization food consumption estimates (called `food balance sheets') and will be used by the WHO/FAO Joint Meeting on Pesticides Residues (JMPR) to assess the impact of MRLs on established ADIs. This proposed system will also have the capability of applying processing and/or production factors to the MRL when calculating potential dietary exposures (WHO, 1988). Upon completion, this system will provide most nations with a rapid and simple method for screening dietary exposures and identifying sources of potential concern relative to the diet. However, since the FAO food balance sheets are based on gross estimates of food consumption for the population as a whole, this technique does not permit estimation of food consumption or dietary pesticide to subgroups of potential concern such as infants and children.

A research product of this investigation will be a compilation of food consumption data on a worldwide basis, which may also be useful for estimates of dietary pesticide exposure in many countries.

At the present, a comprehensive compilation of pesticide residues in food does not exist, and it has only been in recent times that various efforts have commenced to address this issue. One source of such data in the US has been the FDA Total Diet Study (Gartrell et al., 1986). Although limited by the number of samples analysed, this study does provide a useful indication of pesticide residues in the diet. The study has consistently shown over the past several years that concentrations for most pesticides are quite low (the majority of composite averages are between 0.1 and 10 p.p.b.). When compared to the acceptable daily intakes (ADIs) for pesticides established by USEPA or WHO, the calculated exposures do not appear to represent an appreciable risk.

Monitoring for pesticide residues in food is the most desirable method of generating data for estimating dietary pesticide exposures, as this approach is likely to provide the most accurate estimate of residues in food at the time of consumption. Unfortunately, this type of data is expensive and time-consuming to generate. One approach that has been used in the US to increase the efficiency of this type of data collection has been the Surveillance Index, developed by the FDA to target chemicals for monitoring that are likely to pose a significant health threat based on their toxicities, use patterns, or potential for high dietary exposure (Reed, 1985). A similar ranking method has been proposed for monitoring the intake of toxicants via seafood (Brown et al., 1988).

An area of potential exposure to pesticides that is of concern to many nations is the contamination of drinking water by pesticides and/or other chemicals (WHO, 1987). Indeed, contamination of drinking water supplies with a xenobiotic toxicant has been implicated or suspected in several episodes of disease (Rajput et al., 1987; Armstrong et al., 1984). Many countries have initiated programs to monitor the quality of drinking water and to identify regions which have been contaminated (Cohen et al., 1987). An accurate assessment of dietary pesticide exposure should include an estimate of the contribution of drinking water to the total dietary burden of pesticides.

7.5 DISCUSSION

In terms of the magnitude of potential exposures to the individual, it is clear that occupational exposures should be of greatest concern, and that exposures in this area must be carefully monitored to prevent serious disease. The risks from these types of exposures have been well documented by epidemiological reports of illness in agricultural workers (Edmiston and Maddy, 1987; Xue, 1987; Saunders et al., 1987; Coye et al., 1986b).

Many of the existing methods can estimate exposure; however, calculations of dose are difficult. A better understanding of the metabolism and pharmacokinetics of pesticide absorption (dermal, oral, and inhalation) and elimination would allow better estimation of actual doses, and, therefore, better risk management. Although some efforts have been made in this direction (Shehata-Karam et al., 1988; Wester and Maibach, 1985; Wester et al., 1983), the database is not large in terms of the number of different chemicals evaluated. For many chemicals, it may be necessary to extrapolate from experimental animal data in order to estimate human dose.

Construction of models of exposure may provide a valuable tool for predicting and managing occupational (or other types) of exposure, and alleviating the great expense of conducting studies for each chemical under each set of potential uses and exposure conditions. Research efforts in this area suggest that this approach holds promise (e.g. Chester and Hart, 1986; Nigg et al., 1984), but more research is needed to develop models that may be applied under more than a single set of conditions.

With respect to non-occupational exposures, existing data are scanty. Studies such as the EPA's NOPES are very useful as a means for estimating the total burden of pesticides that is likely to occur in the general population as a result of incidental household pesticide exposure. Additional studies of this type are desirable on an international basis, particularly in areas where the level of exposure may be high due to local pest pressures or agricultural practices.

Dietary exposures to pesticides remain an area of significant concern to many people, although evidence for adverse health effects resulting from exposure to low levels of pesticides in the diet is generally lacking. In fact, it has been suggested that natural constituents of food may actually present more of a hazard than pesticides or other chemical contaminants in food (Ames et al., 1987). The expansion of monitoring efforts such as the FDA Total Diet Study to provide better statistical representation and reliability would yield much useful information, and allow better estimation of the potential health effects of pesticides in the diet.

Although most calculations of dietary exposures have focused on estimates that are population-based average exposures, concern has been expressed over the potential effects of pesticide exposure in infants and children. These two subgroups are of particular interest, because they may be more sensitive to some toxic effects (e.g., lead) and because children and infants are well known to consume more food relative to body weight than adults. Therefore, if residue levels are constant, children and infants will receive a higher dose of pesticides in the dietin fact, 2-5 times higher, depending on the crops involved (Saunders, 1987). In addition, ingestion of household dust may also represent a significant route of pesticide exposure for small children (Lewis, 1988). These phenomena raise an interesting and important question: how should the data from chronic feeding studies conducted in experimental animals be extrapolated to derive estimates of risk in infants and children? This is an area that requires more research, in terms of both the extrapolation of long-term animal toxicity studies to pesticide exposures in children, and the identification of effects to which infants and children may be uniquely susceptible.

In summary, occupational pesticide exposures seem to present the greatest risk in terms of the magnitude of individual exposure, and attention should be focused on that area to reduce individual risks. In addition, adequate methods must exist to allow regulatory agencies to protect the quality of the food supply, as is rightly demanded by the public. However, it is important that the data derived from the various exposure methodologies be viewed objectively and that national priorities and resources be focused on those areas that represent the most significant hazard.

7.6 RECOMMENDATIONS

1. Occupational exposures can be reduced through changes in application methodology. Specifically, ground-based applications rather than airblast or airborne applications should be used; and high-exposure occupations such as flagging should be eliminated by the substitution of mechanical devices. In addition, further advances in protective clothing and the use of closed application systems should be advocated.

2. The development of immunoassay methods, which hold the promise of providing rapid and inexpensive methods for monitoring exposures, should be encouraged. These techniques will be most useful in rural areas where access to sophisticated laboratory equipment is limited. In addition, further development and refinement of methodologies for assessing occupational exposure are desirable.

3. A better understanding of the manner in which pesticides are metabolized and excreted in humans is desirable. Such knowledge will permit better extrapolations of toxicity data from experimental animals to man, and allow for a better understanding of the significance of exposures to pesticides in man. 

4. Improved methods are necessary on an international scale to provide developing nations with the capability of screening dietary exposures and identifying sources of potential hazard in the diet. The effort currently in progress at the WHO is a promising first step in that direction. In addition, better monitoring of residues as they occur in food and water at the time of consumption is necessary so that adequate data are available to correctly assess dietary risk.

5. Finally, more attention should be given to the effects of pesticide exposures on subgroups, such as infants and children, who are known to be exposed to relatively higher concentrations of pesticides through dietary and household exposures. Data on the specific effects of such exposures in these two important groups is relatively lacking.

7.7 REFERENCES

American Medical Association, Council of Scientific Affairs (1988) Cancer risk of pesticides in agricultural workers. J. Am. Med. Assoc. 260, 959-966.

Ames, B. N., Magaw, R. and Gold, L. S. (1987) Ranking possible carcinogenic hazards. Science 236, 271-280.

Armstrong, C. W., Stroube, R. B., Rubio, T., Siudyla, E. A. and Miller, G. B. (1984) Outbreak of fatal arsenic poisoning caused by contaminated drinking water. Arch. Environ. Health 39, 271-279.

Atallah, Y. H., Cahill, W. P. and Whitacre, D. M. (1982) Exposure of pesticide applicators and support personnel to O-ethyl-O-(4-nitrophenyl) phenylphosphonothioate (EPN). Arch. Environ. Contam. Toxicol. 11, 210-225.

Borm, P. J. A. and de Barbanson, B. (1988) Bias in biologic monitoring caused by concomitant medication. J. Occup. Med. 30, 214-223.

Brown, H. S., Goble, R. and Tatelbaum, L. (1988) Methodology for assessing hazards of contaminants in seafood. Reg. Toxicol. Appl. Pharmacol. 8, 76-101.

Carmen, G. E., Iwata, Y., Pappas, J. L., O'Neal, J. R. and Gunther, F. A. (1982) Pesticide applicator exposure to insecticides during treatment of citrus trees with oscillating boom and air-blast units. Arch. Environ. Contam. Toxicol. 11, 651-659. 

Chester, G. Hatfield, L. D., Hart, T. B., et al. (1987) Worker exposure to, and absorption of, cypermethrin during aerial application of `ultra low volume' formulation to cotton. Arch. Environ. Contam. Toxicol. 16, 69-78.

Chester, G. and Hart, T. B. (1986) Biological monitoring of a herbicide applied through backpack and vehicle sprayers. Toxicol. Lett. 33, 137-149.

Chiao-Cheng, J. G., Regan, B. M., Bresee, R. R., Meloan, C. E. and Kadoum, A. M. (1988) Carbamate insecticide removal in laundering from cotton and polyester fabrics. Arch. Environ. Contam. Toxicol. 17, 87-94.

Cohen, S. Z., Eiden, C. and Lorber, M. N. (1987) Monitoring ground water for pesticides in the U.S.A. Schriftenr. Ver. Wasser Boden Lufthyg. (Ger.) 68, 265-295. 

Cowell, J. E., Danhaus, R. G., Kuntzman, J. L., et al. (1987) Operator exposure from closed system loading and application of alachlor herbicide. Arch. Environ. Contam. Toxicol. 16, 69-78.

Coye, M. J., Lowe, J. A. and Maddy, K. J. (1986a) Biological monitoring of agricultural workers exposed to pesticides. I. Cholinesterase activity determinations. J. Occup. Med. 28, 628-636.

Coye, M. J., Barnett, P. G., Midtling, J. E., et al. (1986b) Clinical confirmation of organophosphate poisoning of agricultural workers. Am. J. Ind. Med. 10, 399-409. 

Davies, J. E., Freed, V. H., Enos, H. F., et al. (1982) Reduction of pesticide exposure with protective clothing for applicators and mixers. J. Occup. Med. 24, 464-468. 

Davis, J. E., Stevens, E. R. and Staiff, D. C. (1983) Potential exposure of apple thinners to azinphosmethyl and comparison of two methods for assessment of hand exposure. Bull. Environ. Contam. Toxicol. 31, 631-638.

Durham, W. F. and Wolfe, H. R. (1962) Measurement of the exposure of workers to pesticides. Bull WHO 26, 75-91.

Edmiston, S. and Maddy, K. T. (1987) Summary of illnesses and injuries reported in California by physicians in 1986 as potentially related to pesticides. Vet. Human Toxicol. 29, 391-397.

Everhart, L. P. and Holt, R. F. (1982) Potential benlate exposure during mixer/loader operations, crop harvest, and home use. J. Agric. Food Chem. 30, 222-227. 

Fenske, R. A., Hamburger, S. J. and Guyton, C. L. (1987) Occupational exposure to fosetyl-A1 fungicide during spraying of ornamentals in greenhouses. Arch. Environ. Contam. Toxicol. 16, 615-621.

Gartrell, M. J., Craun, J. C., Podrebarac, D. S. and Gunderson, E. L. (1986) Pesticides, selected elements, and other chemicals in adult total diet samples, October 1980-1982. J. Assoc. Off. Anal. Chem. 69, 146-161.

Grover, R., Cessna, A. J., Muri, N. I. et al. (1986) Factors affecting the exposure of ground-rig applicators to 2,4-D dimethylamine salt. Arch. Environ. Contam. Toxicol. 15, 677-686.

Hammock, B. D., Gee, S. J. Cheung, P. Y. K., Miyamoto, T., Goodrow, M. H., et al. (1987) Utility of immunoassay in pesticide trace analysis. In: Greenhalgh, R. and Roberts, T. R. (Eds) Pesticide Science and Biotechnology, Blackwell Science, Oxford, pp.309-316.

Kutz, F. W., Straussman, S. C., Stroup, C. R., Carra, J. S., Leininger, C. C., Watts, D. L. and Sparacino, C. M. (1985) The human body burden of mirex in the Southeastern United States. J. Toxicol. Environ. Health 15, 385-394.

Leavitt, J. R., Gold, R. E., Holcslaw, T. and Tupy, D. (1982) Exposure of professional pesticide applicators to carbaryl. Arch. Environ. Contam. Toxicol. 11, 57-62.

Lewis, R. G. (1976) Sampling and analysis of airborne pesticides. In: Lee, R. E., Jr. (Ed.) Air Pollution from Pesticides and Agricultural Processes, CRC Press, Cleveland, Ohio, pp. 51-94.

Lewis, R. G. (1988) Human Exposure to Pesticides Used in Air and Around the Household. Report prepared for the Task Force on Environmental Cancer and Heart and Lung Disease, Working Group on Exposure, US Environmental Protection Agency, Office of Research and Development, Environmental Monitoring Systems Laboratory, Research Triangle Park, North Carolina.

Lewis, R. G., Bond, A. E., Johnson, D. E. and Hsu, J. P. (1988) Measurement of atmospheric concentrations of common household pesticides: a pilot study. Environ. Monit. Assess. 10, 59-73.

Loevinsohn, M. E. (1987) Insecticide use and increased mortality in rural Phillipines. Lancet 8546, 1359-1362.

Matsumara, F. (1985) Insecticide residues in man. In: Matsumura, F. (Ed.) Toxicology of Insecticides (2nd edn), Plenum Press, New York, London, pp. 547-568. 

Mumma, R. O. and Brady, J. F. (1987) Immunological assays for agrochemicals. In: Greenhalgh, R. and Roberts, T. R. (Eds) Pesticide Science and Biotechnology, Blackwell Science, Oxford, pp. 309-316.

Murphy, R. S., Kutz, F. W. and Strassman, S. C. (1983) Selected pesticide residues or metabolites in blood and urine specimens from a general population survey. Environ. Health. Perspect. 48, 81-86.

Murphy, R. and Harvey, C. (1985) Residues and metabolites of selected persistent halogenated hydrocarbons in blood specimens from a general population survey. Environ. Health Perspect. 60, 115-120.

Nigg, H. N., Stamper, J. H. and Queen, R. M. (1984) The development and use of a universal model to predict tree crop harvester pesticide exposure. Am. Ind. Hyg. Assoc. J. 45(3), 182-186.

Petersen, B. and Chaisson, C. (1988) Pesticides and residues in food. Food Technol. 42(7), 59-64.

Putnam, A. R., Willis, M. D., Binning, L. K. and Boldt, P. F. (1983) Exposure of pesticide applicators to nitrofen: influence of formulation, handling systems, and protective garments. J. Agric. Food Chem. 31, 645-650.

Rajput, A. H., Uitti, R. J., Stern, W., et al. (1987) Geography, drinking water chemistry, pesticides and herbicides and the etiology of Parkinson's disease. Can. J. Neurol. Sci. 14, 414-418.

Reed, D. V. (1985) Chemical contaminants monitoringthe FDA surveillance index for pesticides: establishing food monitoring priorities based on potential health risk. J. Assoc. Off. Anal. Chem. 68(1), 122-124.

Reinert, J. C., Nielsen, A. P., Lunchik, C., Hernandez, O. and Mazzetta, D. M. (1986) The United States Environmental Protection Agency's guidelines for applicator exposure monitoring. Toxicol. Lett. 33, 183-191.

Richter, E. D., Rosenvald, Z., Kaspi, L. and Gruener, N. (1986) Sequential cholinesterase tests and symptoms for monitoring organophosphate absorption in field workers and in persons exposed to pesticide spray drift. Toxicol. Lett. 33, 25-32.

Saunders, D. S. and Petersen, B. P. (1986) Introduction to the Tolerance Assessment System, US Environmental Protection Agency, Office of Pesticide Programs, Residue Chemistry Branch, Washington DC.

Saunders, D. S. (1987) Briefing paper on the Tolerance Assessment System for Presentation to the FIFRA Scientific Advisory Panel, US Environmental Protection Agency, Office of Pesticide Programs, Residue Chemistry Branch, Washington, DC.

Saunders, L. D., Ames, R. G., Knaak, J. B. and Jackson, R. J. (1987) Outbreak of Omite-CR-induced dermatitis among orange pickers in Tulare County, California. J. Occup. Med. 29, 409-413.

Shehata-Karam, H., Monteiro-Riviere, N. A. and Guthrie, F. E. (1988) In vitro penetration of pesticides through human newborn foreskin. Toxicol. Lett. 40, 233-239. 

Skaare, J. U., Tuveng, J. M., and Sande, H. A. (1988) Organochlorine pesticides and polychlorinated biphenyls in maternal adipose tissue, blood, milk, and cord blood from mothers and infants living in Norway. Arch. Environ. Contam. Toxicol. 17, 473-478.

Strassman, S. C. and Kutz, F. W. (1981) Trends of organochlorine pesticide residues in human tissue. In: Khan, M. A. Q. and Stanton, R. H. (Eds), Toxicology of Halogenated Hydrocarbons: Health and Ecological Effects, Pergamon Press, New York, pp. 38-49.

Taguchi, S. and Yakushiji, T. (1988) Influence of termite treatment in the home on the chlordane concentration in human milk. Arch. Environ. Contam. Toxicol. 17, 65-71. 

Travis, C. C., Hattermer-Frey, H. A. and Arms, A. D. (1988) Relationship between dietary intake of organic chemicals and their concentrations in human adipose tissue and breast milk. Arch. Environ. Contam. Toxicol. 17, 473-478.

USEPA (1987a) Pesticide Assessment Guidelines. Subdivision U. Applicator Exposure Monitoring, US Environmental Protection Agency, Office of Pesticide Programs, Exposure Assessment Branch, Washington, DC.

USEPA (1987b) Non-Occupational Pesticide Exposure Study (NOPES)Phase II: Jacksonville, Florida, Summer, 1986, Interim report, US Environmental Protection Agency, Office of Research and Development, Environmental Monitoring System Laboratory, Research Triangle Park, North Carolina.

Weisskopf, C. P., Sieber, J. N., Maizlish, N. and Schenker, M. (1988) Personal exposure to diazinon in a supervised pest eradication program. Arch. Environ. Contam. Toxicol. 17, 201-212.

Wester, R. C. and Maibach, H. I. (1985) In vivo percutaneous absorption and decontamination of pesticides in humans. J. Toxicol. Environ. Health 16, 25-37.

Wester, R. C., Maibach, H. I, Bucks, D. A. W. and Guy, R. H. (1983) Malathion percutaneous absorption and repeated administration to man. Toxicol. Appl. Pharmacol. 68, 116-119.

WHO (1981) Pesticide deaths. What's the toll? Ecoforum 6, 10.

WHO (1987) Drinking-Water Quality. Guidelines for Selected Herbicides, Environmental Health Series # 27, World Health Organization, Regional Office for Europe, Copenhagen, Denmark.

WHO (1988) Guidelines for Predicting Dietary Intake of Pesticide Residues (prepared by the joint UNEP/FAO/WHO Food Contamination Monitoring Programme in collaboration with the Codex Committee on Pesticide Residues), World Health Organization, Geneva.

Xue, S. Z. (1987) Health effects of pesticides: a review of epidemiologic research from the perspective of developing nations. Am. J. Ind. Med. 32, 269-279.