PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
Am J Ind Med. Author manuscript; available in PMC 2014 May 20.
Published in final edited form as:
PMCID: PMC4028316
NIHMSID: NIHMS578793

Pesticides Present in Migrant Farmworker Housing in North Carolina

Thomas A. Arcury, PhD,1,* Chensheng Lu, PhD,2 Haiying Chen, MD, PhD,3 and Sara A. Quandt, PhD4

Abstract

Background

Migrant farmworkers are exposed to pesticides at work. Housing provided to migrant farmworkers may also expose them to pesticides, increasing their health risks. This analysis (1) describes the presence of organophosphorous (OP) and pyrethroid pesticides in North Carolina migrant farmworker houses, and (2) delineates associations of farmworker camp characteristics with pesticide detection and concentration.

Methods

In 2010, 186 migrant farmworkers camps in NC were recruited (participation rate of 82.3%); pesticide wipe samples for 176 houses were analyzed. Tobacco is the predominant hand-harvested crop in this region. Two farmworkers per camp completed interviews; a third assisted with a housing inspection. Gas chromatography–mass spectrometry was used to detect OP and pyrethroid pesticides. Covariates of pesticide detection and concentration were determined with ANOVA and Tobit regression.

Results

OPs were found in 166 of 176 houses (average of 2.4/house); pyrethroids were found in 171 houses (average of 4.3/house). The number of different OPs detected in each camp and concentrations of these OPs were not associated with camp and housing characteristics. The number of different pyrethroids detected in each camp and concentrations of these pyrethroids were associated with camps having residents with H2-A visas, a posted North Carolina Department of Labor Certificate of Inspection, no barracks, fewer residents, no bedroom weather protection or floor violations, and no roaches.

Conclusions

Farmworkers are exposed to pesticides where they live. Policy on removing pesticides from farmworker houses is needed. Reducing pesticides in farmworker houses will reduce one health risk confronted by this vulnerable population.

Keywords: health disparities, vulnerable population, occupational exposures, environmental exposures, agricultural workers, immigrant workers

INTRODUCTION

Farmworkers are exposed to a broad range of pesticides. High levels of pesticide biomarkers have been measured for farmworkers [Arcury et al., 2010] and for members of their families [Coronado et al., 2004; Bradman et al., 2011; Huen et al., 2012]. Exposure to these pesticides can affect the immediate and long-term health of migrant farmworkers [Reigart and Roberts, 1999]. Long-term health effects include increased risks for cancer [Alavanja et al., 2004], respiratory disease [Valcin et al., 2007], neurological disease [Starks et al., 2012], and reproductive problems [Swan, 2006; Perry et al., 2011]. Farmworkers experience pesticide exposure in their houses as well as at work, adding to the potential health effects. Pesticides have been measured in the houses of farmworker families in California [Harnly et al., 2009; Quirós-Alcalá et al., 2011], Oregon [McCauley et al., 2006], Washington [Lu et al., 2000; Curl et al., 2002; Coronado et al., 2010], and western North Carolina (NC) [Quandt et al., 2004]. The specific pesticides in farmworker family houses have changed as the number of organophosphorous (OP) pesticides registered for use has declined and the use of pyrethroid pesticides has increased.

Pesticide exposure in migrant farmworker housing in the eastern US has not been examined. Migrant farmworkers differ from seasonal farmworkers in that migrant farm-workers move to temporary houses at least 75 miles from their permanent houses to work seasonally in agriculture [Carroll et al., 2005]. Migrant farmworkers have little control of their housing; therefore, it deserves special consideration. Housing provided for migrant farmworkers by their employers is often in extremely poor condition [Ziebarth, 2006; Vallejos et al., 2011]. NC migrant farmworker housing generally violates current regulations establishing minimum quality standards [Arcury et al., 2012a], and often does not provide farmworkers conditions conducive to basic safety, hygiene, or privacy [Arcury et al., 2012b]. NC migrant farmworker camps have an average of 11.4 housing violations, with a range of 4–22 violations. Kitchens in these camps often have pest infestation; refrigerators in two-thirds do not maintain an appropriate temperature [Quandt et al., 2013]. Drinking water in one-third of the camps is contaminated [Bischoff et al., 2012].

Documenting the extent to which environmental pesticides are present in migrant farmworker housing is important. Clinically, residential as well as occupational exposure should be considered when evaluating migrant farmworker health. Pesticide exposure among the members of vulnerable populations, such as farmworkers, is a matter of environmental justice. Farmworkers have no control of the pesticides applied at their work [Arcury et al., 2002]. Migrant farmworkers are not provided with information about pesticides applied in housing supplied by their employers. Although they face considerable risk from exposure, farmworkers receive relatively low rewards for the pesticide hazards they experience [Robinson et al., 2011].

This analysis has two aims. First, it describes the presence of OP and pyrethroid pesticides in NC migrant farmworker houses. Second, it delineates the associations of migrant farmworker camp and housing characteristics with the presence of these pesticides.

METHODS AND MATERIALS

Participants

This study was conducted in 16 counties in east-central NC where a large number of farmworkers are employed. Tobacco is the predominant hand-harvested crop in this region, with farmworkers also employed for planting and harvesting cucumbers, sweet potatoes, and other vegetable and fruit crops. The counties are served by the organizations that participated in the research: NC Farmworkers Project, Carolina Family Health Center, Kinston Community Health Center, and Piedmont Health Services. The study focused on employer-provided housing occupied by migrant farm-workers referred to as camps. Lists of camps were obtained from partner organizations. These lists were expanded as new camps were encountered during data collection. All indentified camps were asked to participate in the study. A total of 186 camps was enrolled in the study, with a participation rate of 82.3% (186/226). Residents in 36 camps declined to participate, and the grower or contractor refused to permit participation in four other camps. Camps that participated in the study were given a volley ball as a token of appreciation. Residential pesticide data could not be collected in 10 of the 186 camps, so the final sample included 176 camps.

The project was described to the residents of each camp, and the data collectors asked that three camp residents volunteer to participate. Two farmworkers completed interviews that provided information on potential health characteristics associated with housing quality, as well as their perceptions of the housing in which they lived. One farmworker was needed to guide the data collectors in conducting an inspection of the camp housing; this included an inspection of the general camp grounds, kitchen and dining facilities, bathing facilities, laundry facilities, and bedrooms [Arcury et al., 2012a,b]. The number of buildings inspected varied with the size of the camp, with most camps (61.1%) having only one building, 20.0% having two buildings, and 18.9% having three or more buildings. All the participants spoke Spanish; 68 (18.3%) also spoke an indigenous language. The final sample included 371 men who completed interviews and 182 men who assisted in the housing assessments; 231 men refused to participate when asked. The participation rate of 70.5% (553/784) could have been lower as individuals who did not want to participate could avoid the recruiters. Farmworkers who completed the interviews and who helped with the assessment were each given a $30 cash incentive. This research was approved by the Wake Forest School of Medicine IRB. All study participants provided written informed consent.

Data Collection

Data were collected from June through October 2010. Data are from three study components: (1) interviewer administered questionnaires; (2) housing assessments; and (3) wipe samples to assess residential pesticide exposure. Data collection forms were developed in English and translated into Spanish by a native Spanish speaker. The forms were reviewed by staff members of the community partners who were native Spanish speakers. Revised forms were field tested.

Interviews and housing assessments were completed by trained staff members who were fluent Spanish speakers [Arcury et al., 2012a]. Interviews assessed demographic information, housing features, and perceptions of housing quality; they took approximately 90 min to complete. Housing assessments were completed with the assistance of a farmworker. The housing assessment form assessed compliance with standards summarized in the NC Department of Labor (NCDOL) Introduction to Migrant Housing Inspections [NCDOL, 2008].

Up to three wipe samples were collected in each camp: one each in uncarpeted floor areas in the entry way of the bedrooms of the two interviewed farmworkers and one in a camp common area. These areas were selected as they were areas commonly used by the farmworkers. Wipe samples are a standard method for documenting the presence of dislogeable environmental toxicants on surfaces [Quandt et al., 2004; Deziel et al., 2011; Lu et al., 2013]. A 50 cm × 50 cm template was placed in each area, and each area was wiped with two sterile 4″ × 4″ dressing gauze laced with 2 ml of pesticide grade isopropanol. All gauze pads were placed into the same sterile amber glass bottle to which 4 ml of pesticide grade toluene was added. Bottles were placed in a cooler with blue-ice until they were placed in a freezer at −20°C. All the containers were packaged in insulated shipping containers with dry-ice and shipped to the laboratory for analysis. In additon, 30 field blanks and 30 matrix spikes were made. Field blanks and matrix spikes were prepared by placing clean gauze pads in sample jars on the selected days of sampling. These were brought to the field, and, for matrix spikes, known amounts of pesticides were added. These were then brought back to the lab along with field samples.

Laboratory Procedures

Sample preparation

Established procedures were used for the analysis [Lu et al., 2013]. Known amounts of mixed internal standards including D6-dichlorvos, D10-diazinon, and D6-trans-permethrin were spiked into sample bottles. 80 ml of ethyl acetate was added into each bottle and it was shaken for 30 min on a shaker table. The extraction solvent was then poured into to a centrifuge tube. Another 80 ml of ethyl acetate was added to each sample bottle for a second extraction for 30 min on a shaker table. The extraction solvents were combined (approximately 160 ml) and evaporated to near dryness at 40°C under gentle nitrogen in a TurboVap evaporator. The extract was reconstituted with 0.8 ml ethyl acetate and toluene (3:1 by volume). The dissolved sample extract was then transferred to a 2-ml dispersive-SPE centrifuge tube (Agilent Technologies, Santa Clara, CA), which contained 150 mg of MgSO4, 50 mg of PSA, 50 mg of C18, and 7.5 mg of graphite carbon for cleanup. The d-SPE tube was vortexed for 1 min, and centrifuged at 7,500 rpm for 5 min. Four hundred microliter of supernatant was transferred into an auto sampler vial for GC–MS analysis.

GC–MS analysis

Pesticides were separated and identified using an Agilent 6890 gas chromatograph coupled with an Agilent 5973 mass selective detector (MSD). One microliter of sample was injected in splitless. The injection inlet temperature was kept at 270°C. Pesticides were separated through a 30 m × 0.25 mm DB-5 ms column. The oven temperature program was started and held at 60°C for 2 min, ramped to 180°C at 25°C/min and held at 180°C for 4 min, then ramped to 250°C at 10° C/min and held for 4 min, then to 270°C at 3°C/min and held for 4 min, and finally to 300°C at 20°C/min, plus 3 min post-run at 300°C. The carrier gas was helium, and the flow rate was constantly at 1.2 ml/min. The MSD was kept at 230°C for source, 150°C for quad, and 280°C for Aux-2 temperature. One quantitation ion and two to three confirmation ions were monitored for each chemical using the SIM model. The SIM masses and retention times of measured pesticides are listed in Table AI.

Sample analysis

Field blank wipes were prepared by spiking internal standards into brand new grade gauze pads and then following the extraction steps described above. Blanks contained trace level (<1 ng per analysis) of bifenthrin, allethrin, tefluthrin, and chlorpyrifos, and 19 ng of resmethrin per analysis. Target compounds were identified based on the quantitation ion, confirmation ions and retention time, while the concentrations were quantified based on the quantitation ion only. Target compounds were calculated using linear regression generated from a 10-point external calibration curve prepared in solvent. Reported results were corrected by blank levels and the recoveries generated from matrix spiked calibration standards (n = 7).

Measures

Eleven OP and 14 pyrethroid pesticides were detected in laboratory analysis at greater than the limit of quantification (LOQ). LOQ is the lowest concentration for each analyte that can be accurately measured with great confidence, as defined by the noise (the background) to signal ratio greater than 3. LOQ was calculated using three times the standard deviation of the Limit of Detection. In general, the impact of using LOQ versus LOD only matters to a very small fraction of data that falls into this range, but not the overall distribution of the data set. Three sets of pesticide measures were created: (1) the concentration of each pesticide was reported in μg/m2 for standardization as all samples from a camp were analyzed together; (2) a dichotomous indicator for the presence of each individual pesticide in each camp; and (3) the total number of OPs found in a camp, the total number of pyrethroids found in a camp (where total-permethrin was included in the total but cis-permethrin and trans-permethrin were not counted), and the total number of any pesticides found in a camp.

Camp characteristics included the presence of any workers with H-2A visas in the camp; housing type (no barracks present, barracks present); number of residents in camp (1–10; 11–20; 21 or more); the presence of female residents in the camp; whether the camp violated the regulation on crowding; and data collection period (early season of June through mid-July; mid season of mid-July through August; late season of September or October). Additional characteristics were included in the analysis that could affect the presence of pesticides in the houses. These included whether a NCDOL Certificate of Inspection was posted; a weather protection violation was present in either bedroom; a floor violation was present in either bedroom; roaches were present in either bedroom; and roaches were present in the camp included [Arcury et al., 2012a]. The method described by Bradman et al. [2005] was used to determine the presence of roaches; this method relies on the observation of living or dead roaches, of roach egg cases, or of roach feces in an area.

Statistical Analysis

Frequencies and percentages were used to summarize camp and housing characteristics as well as the prevalence of a pesticide (defined as greater than LOQ) in the wipe samples. For each pesticide that had at least 20% values above the LOQ, geometric means (GMs), geometric standard deviations (GSD), and quartiles were estimated using maximum likelihood methods based on the proportion of values below the LOQ as well as the actual values above the LOQ. The association of camp and housing characteristics with the number of OP, pyrethroid and total pesticides were assessed using analysis of variance (ANOVA) models. We selected pesticides that had at least 20% values above the LOQ and evaluated the effect of the camp and housing characteristics on the concentration of these pesticides using Tobit regression models to account for the left-censoring of the data. Models were fit on the natural log scale to assess bivariate associations and the regression coefficients were back-transformed to the original scale for ease of interpretation of results. The reported regression coefficients (beta) represent the fold change in pesticide concentrations for one unit increase in a predictor. The standard errors (SE) of the back-transformed regression coefficients were obtained using the Delta method [Oehlert, 1992; Rice, 1994]. All analyses were performed by SAS 9.3 (Cary, NC) and P-values less than 0.05 were considered significant.

RESULTS

Camp and Housing Quality Characteristics

Over two-thirds of the camps had workers with H-2A visas present (Table I). The housing in 70.5% of the camps included barracks. Almost half (48.3%) of the camps had 10 or fewer residents, 26.7% had 11–20 residents, and a quarter had 21 or more residents. About a quarter had a crowding violation, and a quarter had female residents. Data were collected in 27.8% of the camps in early season, 45.5% in mid-season, and 26.7% in late season. One-third of the camps had a NCDOL Certificate of Inspection posted. Over half (55.6%) were found to have a bedroom weather protection violation, and 48.3% had a bedroom floor violation. Roaches were present in 54.3% of the bedrooms and 72.6% of the camps.

TABLE I
Migrant Farmworker Camp Characteristics, Eastern North Carolina, 2010 (n = 176)

Pesticides in Farmworker Houses

OP pesticides and pyrethroid pesticides were found in a large number of houses (Table II). Among the OPs, chlorpyrifos was present in almost three-quarters of the migrant farmworker houses (GM 0.11 μg/m2), and malathion was present in 64.2% of the houses (GM 0.12 μ/m2). Phosmet was present in 18.8% of the houses. Registration for use in the US has been discontinued for fenthion (43.8%), diazinon (16.5%), and mevinphos (14.2%); however, each was present in a substantial number of houses, although their concentrations were relatively low. Dichlorvos was present in 7.4% of the houses. Phosalone was present in 5.7% of the houses; this pesticide is not registered for use in NC. Telfuthrin, coumaphos, and fonophos had limited occurrences.

TABLE II
Organophosphorous and Pyrethroid Pesticides Prevalence Greater Than the Limit of Quantification (LOQ) and Concentrations (μg/m2), Migrant Farmworker Housing, Eastern North Carolina, 2010 (n = 176)

Many pyrethroid pesticides were found. Cis-permethrin, trans-permethrin, and total-permethrin were each present in about 90% of houses (GMs 1.05 μ/m2, 1.50 μ/m2, 2.41 μ/m2, respectively). Tetramethrin was found in 80.0% of the houses (GM 0.19 μ/m2), with bifenthrin present in 72.0% (GM 0.11 μ/m2), cypermethrin in 59.1% (GM 0.70 μ/m2), allethrin in 36.4% (GM 0.02 μ/m2), resmethrin in 32.4% (GM <0.01 μ/m2), and cyhalothrin in 19.3%. Two pyrethroids not registered for use in NC were found in the samples: cypermethrin was common, being present in 59.1% of the houses; fenvalinate was found in 8.5% of the houses.

No OPs were detected in 10 houses, and no pyrethroids were detected in five houses, but only one house was devoid of any pesticide (Table III). As many as six different OPs were found in the houses (mean = 2.4). As many as nine different pyrethroids were found in the houses (mean = 4.3). Up to 15 total pesticides were found in these houses (mean = 6.6). None of the camp or housing characteristics was significantly associated with the number of OPs found. Four characteristics were associated with the number of pyrethroids. Camps with H2-A workers had fewer pyrethroids (mean = 4.0, SE = 0.16) than those without H-2A workers (mean = 4.5, SE = 0.24; P = 0.04). Camps with barracks had more pyrethroids (4.8, SE = 0.24) than those without barracks (3.9, SE = 0.16; P < 0.01). Camps with fewer residents had a smaller number of pyrethroids present; those with 1–10 residents had an average of 3.7 pyrethroids (SE = 0.10), those with 11–20 residents had an average of 4.3 pyrethroids (SE = 0.25), and those with 21 or more residents had an average of 4.8 pyrethroids (SE = 0.26; P < 0.01). Finally, camps with bedroom floor violations had more pyrethroids (4.6, SE = 0.19) than those without such violations (3.8, SE = 0.19; P < 0.01). Multivariate analysis that included H2-A visa status, barracks presence or absence, number of residents, and floor violations resulted in only floor violations having a significant association with number of pyrethroids: houses with no bedroom floor violations had a mean of 4.0 different pyrethroids, and those with violations had a mean of 4.9 different pyrethroids (P < 0.01).

TABLE III
Number of Organophosphorous, Pyrethroid, and Total Pesticides Above the Limit of Quantification Found in Each Camp (n = 176)

Camp and Housing Quality Characteristics Associated With Pesticides

Four OPs (chlorpyrifos, malathion, fenthion, phosmet) with a sufficient number of detections greater than the LOQ were included in the bivariate analysis comparing camp characteristics differences in concentrations of pesticides detected. Three of these 44 comparisons attained statistical significance (P < 0.05). Concentrations of chlorpyrifos for camps with no crowding were 60% less than camps with crowding (β = 0.40, SE = 0.17). Concentrations of fenthion for camps with no barracks were 49% higher than in camps with barracks (β = 1.49, SE = 0.31). Concentrations of phosmet were 79% lower in camps measured at mid season than in camps at which samples were collected in late season (β = 0.21, SE = 0.13).

Nine pyrethroids (cis-permethrin, trans-permethrin, total-permethrin, tetramethrin, bifenthrin, cypermethrin, allethrin, resmethrin, cyhalothrin) had values greater than the LOQ for at least 20% of the wipe samples. Bivariate comparisons of the associations of the concentrations of these nine pyrethroids with camp characteristics were calculated (Table IV). Three of the covariates, crowding violation, female residents and no camp roaches present, had no significant associations with any of the pyrethroid concentrations.

TABLE IV
Significant Associations of Migrant Farmworker Camp Characteristics With Concentrations (μg/m2) of Pyrethroid Pesticides Found in Each Camp (n = 176)

Camps with no H-2A workers had concentrations of bifenthrin that were 140% higher (β = 2.40, SE = 0.98) and of resmethrin that were 521% higher (β = 6.21, SE = 4.81) than those with H-2Aworkers. Those camps with no barracks present had concentrations of bifenthrin that were 72% lower (β = 0.28, SE = 0.11) and of resmethrin that were 80% lower (β = 0.20, SE = 0.16) than those with barracks. Camps with 1–10 residents had concentrations of bifenthrin that were 65% lower (β = 0.35, 0.16) than those with 21 or more residents. Camps at which data were collected in mid-season had concentrations of bifenthrin that were 154% greater (β = 2.54, SE = 1.19) than those with data collection in the late season. Camps with the NCDOL Certificate of Inspection posted had concentrations of cis-permethrin that were 58% lower (β = 0.42, SE = 0.13), of trans-permethrin that were 59% lower (β = 0.41, SE = 0.13), and of total-permethrin that were 62% lower (β = 0.38, SE = 0.11), but concentrations of allethrin that were 354% greater (β = 4.54, SE = 3.50) than camps with an NCDOL Certificate of Inspection posted.

Camps with no weather protection violations had concentrations of cis-permethrin that were 50% lower (β = 0.50, SE = 0.14), of trans-permethrin that were 53% lower (β = 0.47, SE = 0.14), of total-permethrin that were 55% lower (β = 0.45, SE = 0.14), and of tetramethrin that were 55% lower (β = 0.45, SE = 0.15). Camps with no floor violations had concentrations of tetramethrin that were 53% lower (β = 0.47, SE = 0.16), of bifenthrin that were 52% lower (β = 0.48, SE = 0.18), and of resmethrin that were 85% lower (β = 0.15, SE = 0.11). Houses without roaches present in the bedroom had concentrations of cis-permethrin that were 43% lower (β = 0.57, SE = 0.16), of trans-permethrin that were 48% lower (β = 0.52, SE = 0.15), of total-permethrin that were 47% lower (β = 0.53, SE = 0.17), of cypermethrin that were 70% lower (β = 0.30, SE = 0.15), and of cyhalothrin that were 77% lower (β = 0.23, SE = 0.15) than those with roaches present.

DISCUSSION

Migrant farmworkers in NC face environmental exposure to a substantial number of OP and pyrethroid insecticides in their housing. OPs are limited for use in agricultural production. The number of OPs found in these migrant farmworker houses and the concentrations of specific OPs are not associated with camp characteristics. Pyrethroids are labeled for residential and agricultural use. The number and concentrations of pyrethroids present in the migrant farmworker houses are associated with specific camp characteristics. Camps having workers with H-2A visas, a posted NCDOL Certificate of Inspection, no barracks, fewer residents, and no bedroom weather protection or floor violations, and no roaches had lower levels of pyrethroids.

Comparison of these results with other research needs to be undertaken with caution. Earlier studies of agricultural worker housing were limited largely to analyses of a few specific OPs, such as azinphos-methyl, diazinon and chlorpyrifos; and several completed sample collection over a decade ago [Lu et al., 2000; Harnly et al., 2009]. Azinphos-methyl has not been widely used in the eastern US. Diazinon was withdrawn from use at the end of 2004. Chlorpyrifos was withdrawn from non-agricultural use in 2001. Therefore, these pesticides should be present in farmworker houses less often and at lower concentrations in 2010 (when our data were collected) than in studies in which data were collected in the 1990s and early 2000s. Many earlier analyses used different methods to collect environmental samples, with many collecting dust samples rather than the wipe samples used for this research. Finally, most of the published research has focused on houses inhabited by farmworker families with children. This analysis focused on migrant men in camps; only a few camps had women or children present.

However, comparison of these results to other research documents the relatively high levels of residential pesticides to which migrant farmworkers are exposed. Based on dust samples collected in 2006 in Salinas, California, Quirós-Alcalá et al. [2011] report that 55% of the 29 samples collected in 15 farmworker homes and 36% of the 25 samples collected in 13 Oakland urban homes contained chlorpyrifos, compared to 72.7% of the samples collected for this study. They also report that 7% of their farmworker samples and 12% of their urban samples contained malathion, compared to 64.2% of the samples collected in this study. They did find that 79% of the samples from farmworker houses and 52% of the samples from urban houses contained diazinon, compared to 16.5% of the migrant farmworker houses in this study; however, their samples were collected closer to the time when diazinon was still registered. Similar to this study, Quirós-Alcalá et al. [2011] report the presence of pyrethroid pesticides in all of the samples they collected in farmworker and urban houses.

Quandt et al. [2004] collected wipe samples from 41 farmworker family houses located in western NC and Virginia in 2001. Seventy-eight percent of their wipe samples contained chlorpyrifos (compared to 72.7% in this study), and 34% contained diazinon (compared to 16.5%). Detections of cis-permethrin (66%) and trans-permethrin (93%) were about the same as for the migrant farmworker houses in this study (90.3% and 89.8%, respectively). Finally, Harnly et al. [2009], based on dust samples collected in 1999–2000, reported that 91% of 197 agricultural worker family houses in California had chlorpyrifos in their homes (compared to 72.7% for this study), 86% had diazinon (compared to 16.5%), 7.2% had phosmet (compared to 18.8%), 98% had cis-permethrin (compared to 90.3%), and 98% had trans-permethrin (compared to 89.8%).

Most studies examining the presence of pesticides in farmworker houses measured covariates of exposure that are not comparable to those relevant in this analysis. For example, measures of proximity to agricultural fields and variation in socioeconomic status have been relevant in other studies [Lu et al., 2000; Quandt et al., 2004; Coronado et al., 2011], but are not relevant to this analysis, as all migrant farmworker camps in NC are adjacent to fields and all migrant farmworkers have similar incomes. Residents with H2-A visas and NCDOL certificates of inspection were relevant to this study, but have not been relevant to other studies. However, several analyses of farmworker family houses found indicators of cleanliness related to pesticide presence and concentrations. Houses that were difficult to clean or that were less clean had greater levels of pesticides [Quandt et al., 2004; Harnly et al., 2009]. Similarly, recent use of pesticides in the home [Quirós-Alcalá et al., 2011], as well as storing pesticides and work shoes in the home [Harnly et al., 2009], were related to greater levels of pesticides. Several housing measures in this study reflect this association; houses with more weather and floor violations had greater levels of pesticides, and those with no roaches had lower levels of pesticides.

OP and pyrethroid insecticides are toxic chemicals with known health effects. In addition to the number and concentrations of OP and pyrethroid insecticides found in the migrant farmworker houses, specific characteristics of several of these pesticides raise larger concerns for ongoing pesticide exposure of farmworkers. For example, the OP fenthion was found in samples from 48.1% of the migrant farmworker houses. This OP is no longer registered for use in the US. Two other OPs, diazinon and mevinphos, which also are no longer legal to use in the US, are each found in about 15% of the migrant farmworker houses. Therefore, while migrant farmworkers may no longer be exposed to these pesticides while working, they are exposed to them where they sleep. The same is true for the pyrethroids cypermethrin (found in 59.1% of the samples) and fenvalinate (found in 8.5% of the samples); neither is registered for use in NC, but NC migrant farmworkers are exposed to both in their houses.

The substantial prevalence of three OP and pyrethroid pesticides not registered for use is an important concern. It is doubtful that these pesticides are being purchased and applied illegally. The cause of the continued presence of these pesticides in the residential environment is likely more insidious. It results from their persistence in indoor environments even years after they are no longer used in agriculture. Totally removing them from the migrant farmworker houses (and the houses of others in rural communities) may require a concerted effort over years. We can expect that the remnants of other pesticides, such as chlorpyrifos and malathion, will be in rural houses for years after these pesticides are removed from commercial use. The public health community needs to consider methods that can be implemented to significantly reduce residential pesticide exposure in farmworker housing.

Residential exposure to these pesticides compounds workplace exposure. Rather than having a safe refuge from occupational pesticide exposure, these farmworkers continue to be exposed in their houses. The specific health implications of this residential pesticide exposure have not been delineated. The Agricultural Health Study indicates that chronic exposure to even small amounts of pesticides increases risk for significant health outcomes, including cancer and neurological deficient [Starks et al., 2012; Koutros et al., 2013]. Current policy does not address the types and levels of pesticides to which farmworkers are exposed in their houses. The US Environmental Protection Agency [1992] Work Protection Standard addresses work place pesticide use and storage; required training includes discussion of take home routes of pesticide exposure. However, no requirements for testing farmworker houses for pesticide exposure are included in this policy. Understanding the individual and synergistic components of occupational and residential pesticide exposure for farmworkers and other worker populations will require an integrated research program with access to individual participants at work and at home.

The results of this study should be evaluated in terms of its limitations. The data come from one region of one state in 1 year. Only grower-owned migrant farmworker camps are included in the study. Generalizations of results to other regions should be made with caution. This study uses a cross-sectional design; therefore, the causal relationships can only be inferred. Although all identified migrant farmworker camps were asked to participant in the study, not all migrant farmworker camps may have been identified; the conditions in unidentified camps may differ from those in the camps identified for the study. Data collectors were not allowed into a number of camps; the conditions in these camps may differ from the camps that participated in the study. The limit of detection for pesticides measured by this study is limited to current limits on technology. However, the study has a number of strengths, including its large sample size and high participation rate, extensive collection of information, and use of state-of-the-art technology in measuring pesticides.

Migrant farmworkers are exposed to substantial numbers of pesticides when they work. This analysis shows that they continue to be exposed to pesticides in the houses in which they live. Some of the pesticides to which workers are exposed in their houses are pesticides that are no longer legal to use in the workplace. Indicators of housing conditions are related to the number and concentrations of pyrethroid pesticides found in migrant farmworker houses. Exposure to these pesticides increases risks for immediate and long-term health effects for these migrant farmworkers. Efforts to reduce pesticides in farmworker houses will reduce one health risk that the members of this vulnerable population confront. Policy on removing pesticides from migrant farmworker houses is needed. Further research on the number and concentrations of pesticides to which migrant farm-workers are exposed in their houses and the potential health effects of this exposure is also needed.

Acknowledgments

Contract grant sponsor: National Institute of Environmental Health Sciences; Contract grant number: R01-ES012358.

Appendix

TABLE AI

Pesticide Name, GC–MS Retention Times, Quantitation Ion (Q ion) and Confirmation Ions (C Ion)

PesticideRetention time (min)Q ionC1 ion (%Q1/T)C2 ion (%Q2/T)C3 ion (%Q3/T)
Organophosphorous pesticides
 Bromophos15.42331329333125
 Chlorpyrifos14.88314258286208
 Coumaphos24.62362226210334
 Diazinon12.31304227276248
 Dichlorvos6.63109185145220
 Fenthion15.00278279169125
 Fonophos12.28246109137110
 Mevinphos7.81127192109164
 Phosmet20.42160161133317
 Phosalone21.69182184367154
 Pirimphos-ethyl15.41318333304290
 Pirimphos-methyl14.40290276305233
 Resmethrin19.72123143171128
 Tefluthrin12.70177178197199
 Tetramethrin20.56164123165107
Pyrethroid pesticides
 Allethrin16.04136123107168
 Bifenthrin20.45181166165182
 Cyfluthrin25.79–26.35226127163199
 Cyhalothrin22.1–22.48181197208199
 Cypermethrin26.64–27.2163165181209
 Fenpropathrin20.78181265125
 Fenvalinate29.09–29.75419181225167
 Flucythrinate27.14–27.7199225209181
 Fluvalinate29.55–29.81250252181251
 cis-permethrin24.36183127163165
 trans-permethrin24.71183127163165
 Phenothrin21.34–21.55123183184124
 Resmethrin19.72123143171128
 Tefluthrin12.70177178197199
 Tetramethrin20.56164123165107

Footnotes

Disclosure Statement: The authors report no conflicts of interests.

References

  • Alavanja MC, Hoppin JA, Kamel F. Health effects of chronic pesticide exposure: Cancer and neurotoxicity. Annu Rev Public Health. 2004;25:155–197. [PubMed]
  • Arcury TA, Quandt SA, Russell GB. Pesticide safety among farmworkers: Perceived risk and perceived control as factors reflecting environmental justice. Environ Health Perspect. 2002;110(Suppl 2):233–240. [PMC free article] [PubMed]
  • Arcury TA, Grzywacz JG, Talton JW, Chen H, Vallejos QM, Galván L, Barr DB, Quandt SA. Repeated pesticide exposure among North Carolina migrant and seasonal farmworkers. Amer J Ind Med. 2010;53:802–813. [PMC free article] [PubMed]
  • Arcury TA, Weir M, Chen H, Summers P, Pelletier LE, Galván L, Bischoff WE, Mirabelli MC, Quandt SA. Migrant farmworker housing regulation violations in North Carolina. Am J Ind Med. 2012a;55:191–204. [PMC free article] [PubMed]
  • Arcury TA, Weir MM, Summers P, Chen H, Bailey M, Wiggins MF, Bischoff WE, Quandt SA. Safety, security, hygiene and privacy in migrant farmworker housing. New Solut. 2012b;22:153–173. [PMC free article] [PubMed]
  • Bischoff W, Weir M, Summers P, Chen H, Quandt SA, Liebman AK, Arcury TA. The quality of drinking water in North Carolina farmworker camps. Am J Public Health. 2012;102:e49–e54. [PMC free article] [PubMed]
  • Bradman A, Castorina R, Barr DB, Chevrier J, Harnly ME, Eisen EA, McKone TE, Harley K, Holland N, Eskenazi B. Determinants of organophosphorus pesticide urinary metabolite levels in young children living in an agricultural community. Int J Environ Res Public Health. 2011;8:1061–1083. [PMC free article] [PubMed]
  • Bradman A, Chevrier J, Tager I, Lipsett M, Sedgwick J, Macher J, Vargas AB, Cabrera EB, Camacho JM, Weldon R, Kogut K, Jewell NP, Eskenazi B. Association of housing disrepair indicators with cockroach and rodent infestations in a cohort of pregnant Latina women and their children. Environ Health Perspect. 2005;113:1795–801. [PMC free article] [PubMed]
  • Carroll DJ, Samardick R, Gabbard SB, Hernandez T. Research Report No 9. U.S. Department of Labor Office of the Assistant Secretary for Policy Office of Programmatic Policy; 2005. Findings from the National Agricultural Workers Survey (NAWS) 2001–2002: A demographic and employment profile of United States farm workers.
  • Coronado GD, Thompson B, Strong L, Griffith WC, Islas I. Agricultural task and exposure to organophosphate pesticides among farmworkers. Environ Health Perspect. 2004;112:142–147. [PMC free article] [PubMed]
  • Coronado GD, Griffith WC, Vigoren EM, Faustman EM, Thompson B. Where’s the dust? Characterizing locations of azinphos-methyl residues in house and vehicle dust among farmworkers with young children. J Occup Environ Hyg. 2010;7:663–671. [PMC free article] [PubMed]
  • Coronado GD, Holte S, Vigoren E, Griffith WC, Barr DB, Faustman E, Thompson B. Organophosphate pesticide exposure and residential proximity to nearby fields: Evidence for the drift pathway. J Occup Environ Med. 2011;53:884–891. [PMC free article] [PubMed]
  • Curl CL, Fenske RA, Kissel JC, Shirai JH, Moate TF, Griffith W, Coronado G, Thompson B. Evaluation of take-home organophosphorus pesticide exposure among agricultural workers and their children. Environ Health Perspect. 2002;110:A787–A792. [PMC free article] [PubMed]
  • Deziel NC, Viet SM, Rogers JW, Camann DE, Marker DA, Heikkinen MS, Yau AY, Stout DM, 2nd, Dellarco M. Comparison of wipe materials and wetting agents for pesticide residue collection from hard surfaces. Sci Total Environ. 2011;15:4442–4448. [PMC free article] [PubMed]
  • Harnly ME, Bradman A, Nishioka M, McKone TE, Smith D, McLaughlin R, Kavanagh-Baird G, Castorina R, Eskenazi B. Pesticides in dust from homes in an agricultural area. Environ Sci Technol. 2009;43:8767–8774. [PubMed]
  • Huen K, Bradman A, Harley K, Yousefi P, Barr DB, Eskenazi B, Holland N. Organophosphate pesticide levels in blood and urine of women and newborns living in an agricultural community. Environ Res. 2012;117:8–16. [PubMed]
  • Koutros S, Beane Freeman LE, Lubin JH, Heltshe SL, Andreotti G, Barry KH, DellaValle CT, Hoppin JA, Sandler DP, Lynch CF, Blair A, Alavanja MC. Risk of total and aggressive prostate cancer and pesticide use in the Agricultural Health Study. Am J Epidemiol. 2013;177:59–74. [PMC free article] [PubMed]
  • Lu C, Fenske RA, Simcox NJ, Kalman D. Pesticide exposure of children in an agricultural community: Evidence of household proximity to farmland and take home exposure pathways. Environ Res. 2000;84:290–302. [PubMed]
  • Lu C, Adamkiewicz G, Attfield KM, Kapp M, Spengler JD, Tao L, Xie SH. Household pesticide contamination from indoor pest control applications in urban low-income public housing dwellings: A community-based participatory research. Environ Sci Technol. 2013;47:2018–2025. [PMC free article] [PubMed]
  • McCauley LA, Travers R, Lasarev M, Muniz J, Nailon R. Effectiveness of cleaning practices in removing pesticides from home environments. J Agromed. 2006;11:81–88. [PubMed]
  • North Carolina Department of Labor. [accessed March 24, 2011];Introduction to migrant housing inspections in North Carolina (with revisions through January 2008) 2008 Available: http://www.nclabor.com/ash/ash_blue_book.pdf.
  • Oehlert GW. A note on the delta method. Am Stat. 1992;46:27–29.
  • Perry MJ, Venners SA, Chen X, Liu X, Tang G, Xing H, Barr DB, Xu X. Organophosphorous pesticide exposures and sperm quality. Reprod Toxicol. 2011;31:75–79. [PMC free article] [PubMed]
  • Quandt SA, Arcury TA, Rao P, Mellen BG, Camann DE, Doran AM, Yau AY, Hoppin JA, Jackson DS. Agricultural and residential pesticides in wipe samples from farmworker family residences in North Carolina and Virginia. Environ Health Perspect. 2004;112:382–387. [PMC free article] [PubMed]
  • Quandt SA, Summers P, Bischoff WE, Chen H, Wiggins MF, Spears CR, Arcury TA. Cooking and eating facilities in migrant farmworker housing in North Carolina. Am J Public Health. 2013;103:e78–e84. [PMC free article] [PubMed]
  • Quirós-Alcalá L, Bradman A, Nishioka M, Harnly ME, Hubbard A, McKone TE, Ferber J, Eskenazi B. Pesticides in house dust from urban and farmworker households in California: An observational measurement study. Environ Health. 2011;10:19. [PMC free article] [PubMed]
  • Reigart JR, Roberts JR. Recognition and management of pesticide poisoning. 5. Washington, DC: United States Environmental Protection Agency; 1999. EPA 735-R-98-003.
  • Rice J. Mathematical statistics and data analysis. 2. Belmont, CA: Duxbury; 1994.
  • Robinson EN, Nguyen HT, Isom S, Quandt SA, Grzywacz JG, Arcury TA. Wages, wage violations, and pesticide safety experienced by migrant farmworkers in North Carolina. New Solut. 2011;21:251–268. [PMC free article] [PubMed]
  • Starks SE, Gerr F, Kamel F, Lynch CF, Alavanja MC, Sandler DP, Hoppin JA. High pesticide exposure events and central nervous system function among pesticide applicators in the Agricultural Health Study. Int Arch Occup Environ Health. 2012;85:505–515. [PMC free article] [PubMed]
  • Swan SH. Semen quality in fertile US men in relation to geographical area and pesticide exposure. Int J Androl. 2006;29:62–68. [PubMed]
  • US Environmental Protection Agency. [accessed April 2, 2010];Worker protection standard: Final rule. 40 C.F.R. § 170. 1992 Available: http://www.epa.gov/pesticides/health/worker.htm.
  • Valcin M, Henneberger PK, Kullman GJ, Umbach DM, London SJ, Alavanja MC, Sandler DP, Hopin JA. Chronic bronchitis among non-smoking farm women in the agricultural health study. J Occup Environ Med. 2007;49:574–583. [PMC free article] [PubMed]
  • Vallejos QM, Quandt SA, Grzywacz JG, Isom S, Chen H, Galván L, Whalley L, Chatterjee AB, Arcury TA. Migrant farmworkers’ housing conditions across an agricultural season in North Carolina. Am J Ind Med. 2011;54:533–544. [PMC free article] [PubMed]
  • Ziebarth A. Housing seasonal workers for the Minnesota processed vegetable industry. Rural Sociol. 2006;71:335–357.