|Home | About | Journals | Submit | Contact Us | Français|
We used community-based ethnography and public health risk assessment to assess beliefs about pesticide exposure risks among farmworkers in the Lower Yakima Valley of Washington State.
We used unstructured and semistructured interviews, work-site observation, and detailed field notes to gather data on pesticide exposure risks from 99 farmworkers.
Farmworkers’ pesticide-relevant beliefs and attitudes could be grouped into 5 major themes: (1) dry pesticides are often perceived as a virtually harmless powder, (2) farmworkers who identify themselves as allergic to pesticides are more acutely affected by exposure, (3) the effect of pesticide exposure is more severe for those perceived as physically weak, (4) protective equipment is used selectively in response to financial pressure to work rapidly, and (5) some farmworkers delay decontamination until they find water deemed an appropriate temperature for handwashing.
We elucidated farmworkers’ pesticide-relevant beliefs regarding perceived danger and susceptibility to pesticides, the need to put safety second to financial considerations, and reasons for delaying decontamination. Researchers and policymakers should incorporate these data in study designs and legislation concerned with farmworker exposure to pesticides.
Farmworkers are routinely exposed to a variety of health hazards in the course of their work. Some of those hazards may be explained by farmworkers’ own beliefs about hazards and risks. Qualitative research on exposure to pesticides indicates that hazard is associated with perception of harm and shows that some farmworkers do not perceive exposure to pesticides as a cause for concern for themselves, other farmworkers, or their living or unborn children.1–4 Additional qualitative research indicates that farmworkers associate pesticide exposure with sensory detection. For example, if workers do not feel, taste, see, or smell a chemical, they do not feel that they have been exposed.5 Some reports even reveal that farmworkers in rural Mexico taste pesticides to measure dilution, and some farmworkers believe that those who do not fall ill as a result of conducting these taste tests are strong and healthy.6
Researchers have used qualitative methods to learn more about pesticide risks, but information remains limited. Here, we will provide more information on pesticide-related beliefs, behaviors, and risk perceptions among farmworkers. Our research questions were guided by a review of scientific literature, as well as standard principles of the Health Belief Model (perceived susceptibility to hazard, perceived severity of risk, perceived barriers to reducing the risk, and modifying factors such as sociocultural beliefs and practices). We combined Health Belief Model principles and ethnographic research methodology to develop and implement a comprehensive strategy for assessing the perceptions of risk associated with pesticide exposure, enabling us to learn about the contexts of vulnerability, differential susceptibility, and perceived risk among farmworkers.
Our exploratory research questions were the following: What belief systems influence farmworkers’ perceived risk of pesticide exposure? What are farmworkers’ perceptions regarding differential severity of or vulnerability to pesticide exposure? What observed work-related and sociocultural factors modify pesticide exposure risks?
The principal investigator (S. A.S) and a trained bilingual field assistant (B.N.) collected all data for this investigation. To ensure proper use of ethnographic methods, the principal investigator and field assistant lived in the study region (the Lower Yakima Valley of Washington State) during all phases of data collection, which took place from March 2005 to February 2006.
Eligibility criteria for participants included a minimum age of 18 years, year-round residency in the Lower Yakima Valley, and employment as an agricultural worker. Inperson recruitment took place at local community organizations, churches, grocery stores, and work sites. Interested individuals were asked to provide their names and telephone numbers for later contact. Potential participants were contacted by telephone for in-home visits to discuss details of the study.
Open-ended, semistructured, and unstructured interviews were performed in participants’ homes. Unstructured interviews were used to gain rapport with participants. There was no interview guide or list of questions for unstructured interviews. Instead, unstructured interviews were informal discussions. Participants, the principal investigator, and the field assistant often used this time to make plans to attend local community events, churches, or special family occasions.
In semistructured interviewing, an interview guide of 19 primary questions (not including probing questions) was used. Questions focused on knowledge of pesticides, beliefs about health risks associated with pesticide exposure, safety practices in the field, safety practices at home, and beliefs about work-related conditions that contribute to exposure (e.g., provision of water). Semistructured interviews were audio recorded, and a unique identifier (ID) was recorded in place of the participant’s name. All interviews were conducted in Spanish and lasted between 1.5 and 3 hours (average=2 hours).
Observational data were collected using direct and participant observation.7–9 In participant observation, the principal investigator worked alongside farmworkers in apple and cherry orchards in the Lower Yakima Valley. The investigator took field notes on what she saw, heard, and experienced during the workday. For example, the principal investigator took notes on the configuration of work groups (e.g., men vs women), what work was performed by whom, and the types of exposure protection farmworkers used, if any. The investigator also recorded observational data on the conditions inside farmworkers’ homes, decontamination activities (e.g., showers, hand washing, changing work clothes), family transfer of pesticide residue, and participants’ demeanor.
The field assistant transcribed all taped interviews and translated them into English. Having a single individual transcribe and translate all interviews ensured consistency and reliability of the Spanish-to-English translations. Moreover, because the field assistant was present during the interviews, she was able to salvage inaudible parts of the transcripts (for example, when 2 or more persons were speaking at the same time) using detailed field notes and debriefing summaries.
The principal investigator then reviewed the interview transcripts, debriefing summaries, and detailed field notes in their entirety. This approach strengthened the text-analysis process, allowing the data to be reviewed with greater familiarity and within the context in which it was collected. English interview transcriptions and field notes were then imported into the qualitative analysis software package N’Vivo version 2.0.10 N’Vivo was used to code, search for, and assign attributes (e.g., gender, age, and years working in agriculture)to each interview. N’Vivo also enabled investigators to organize and sort information, make queries, and run reports.
Each interview was coded using exploratory “open coding,” in which emergence of an exhaustive list of thematic incidences were characterized line by line in the free-flowing text.11–13 Next, a unified list of codes was condensed for a second analysis of the data line by line, using the list of unified codes. The codes from all of the interviews were then grouped into categories by similarity, and the transcripts were recoded using the unified, grouped set of codes based on the code categories. We then reviewed sentences, phrases, and paragraphs surrounding coded text, in conjunction with memos and field notes, to identify common themes emerging from the data. A comprehensive table was created showing the codes, the code categories, and interview passages that appeared to illuminate emergent themes. Each theme that emerged was then rechecked against the data to ensure that the theme reflected the data. Finally, we selected narratives that portrayed each theme for inclusion here in order to give voice to the participants’ stories.
Ninety-nine farmworkers participated in this investigation. Nearly all of the farmworkers (98%) were born in Mexico; however, all participants identified themselves as being of Mexican origin. Other demographic characteristics of farmworkers in our sample are listed in Table 1. Analysis revealed that farmworkers’ pesticide-relevant beliefs and attitudes could be grouped into 5 major themes: (1) dry pesticides are often perceived as a virtually harmless powder, (2) farmworkers who identify themselves as allergic to pesticides are more acutely affected by exposure, (3) the effect of pesticide exposure is more severe for those perceived as physically weak, (4) protective equipment is used selectively in response to financial pressure to work rapidly, and (5) some farmworkers delay decontamination until they find water deemed appropriate for handwashing or as a result of humoral-medicine beliefs about water temperature. The 5 themes and representative quotes illustrating each theme are listed in Table 2.
The most compelling theme that emerged was that farmworkers considered dry substances, which farmworkers called powder, to be relatively harmless compared with wet substances (i.e., pesticide sprays and liquids), which they considered to be harmful. When farmworkers were asked to define “powder,” many indicated that it was the white residue that appeared on fruits. Farmworkers also defined “powder” as dirt that was blown over the fruit by heavy winds. Many farmworkers used the Spanish words for “powder” (polvo or polvito) and “dirt” (tierra) interchangeably, but participant observation confirmed use of the term “powder” to describe dirt blown into the trees. One person described “powder” this way: “Like when the dirt gets loose, the wind blows and the dirt raises and the trees get full [of dirt]. And when we are cutting the fruit, we are working, and we move the powder and it all falls on us.”
Farmworkers described powder as being different from pesticide sprays, and some of them said they did even not consider powder a pesticide. Of the small number of people who said that powder might be linked to pesticides, there was uncertainty about the actual composition of powder.
[W]ell, I do not know in reality what the powder is. But sometimes in the lanes of the field [orchard] there are cars [tractors] running on them.… They make powder and it goes on the trees and then when we are cutting the apple or the thinning or the harvest the tree is moving and … the powder flies. Or it could also be the same spray that they put on it when they spray—it could be that also. We do not ask … all we want is to fill bins [of fruit] and get out of there.
A vast majority of farmworkers believed some people are allergic to pesticides: those who report allergy-like symptoms, such as reddish, watery eyes; itchy eyes; rashes; sneezing; scratchy throat; and a runny nose. Farmworkers described their allergies as being linked to the pesticide residue or powder.
Well, there are some that say that their vision gets irritated in their eyes. Others have allergies like a cold. Only during the time of the thinning … or their eyes get red sometimes. That is what I have noticed in my coworkers.
To treat the allergy symptoms, individuals reported using over-the-counter allergy medicines (e.g., antihistamine pills, lotions, and creams). Observation confirmed farmworkers taking antihistamine pills while at work and using anti-itch lotions and creams after returning home. Workers were also seen using prescription-strength antihistamines prescribed by physicians in addition to over-the-counter medications.
Wearing gloves, goggles, and masks was associated with decreased daily yield and reduced pay. Thus, farmworkers chose not to wear protective gear when they were paid by yield. In most cases, no protective equipment was provided by the employer; but when protective equipment was provided, farmworkers often chose not to use it. Instead, they focused on completing their jobs (thinning or harvesting crops) as quickly as possible to maximize their income. During participant observation, field study staff saw farmworkers decide not to use available protective equipment.
Some farmworkers also reported feeling pressure from managers to keep up with other workers when gloves, goggles, and masks prevented them from doing so: “[S]ometimes, yes, almost all the time I have gloves. But sometimes it is just that they get in my way and so I take them off.”
When farmworkers were paid an hourly wage, however, more people decided to use protection, as demonstrated in the following exchange between interviewer and participant:
Farmworker: “When I am picking cherries, I do not wear glasses or gloves.’
Interviewer: “Why not?”
Farmworker: “I do not use them because they get in my way. What happens is that, when I am [working] by contract, it is best for me that nothing get in my way.… Without the gloves I can do things faster. But when I am [working] by the hour and not by contract, then I do wear the gloves and glasses.”
Farmworkers believed that children have a more severe reaction to pesticides than do adults:
[C]hildren are affected [more] because they are small. They have fewer defenses than us [adults] because we are grown. Because we are grown, we can resist things more. We are stronger than them because they are smaller. Well, that is my thinking. They might have less defenses than us because we are grown.
Farmworkers also believed that men were less likely to be negatively affected by pesticide exposure than women:
Interviewer: “Do you think that the men are affected [by pesticides] differently than the women?”
Farmworkers: “Well, I think that they do.”
Farmworker: “Well, from what I see there at work, they say that we [women] are weaker than the men… that we have less resistance than them for that [pesticide exposure].”
Farmworkers also generally believed that the elderly were at higher risk of experiencing a negative effect from pesticide exposure than younger people:
The difference is greater if they are older [for this interview, a mother and sister-in-law are heard agreeing in the background]. They get red from the spray from the trees. If older workers work for two days, then two days after they are there [at work] with a fever and with very red eyes.
Finally, some farmworkers felt that those who had weak body types were more susceptible to the negative effects of pesticide exposure. The Spanish term used for “body type” in the interviews was organismo, which literally translates as “organism” in English. To keep the translation in context, we used the term “body type” as the English translation of organismo. Farmworkers who believed that having a weak body type was a determinant of pesticide susceptibility also described weak individuals as those who did not take care of their health:
[T]here are some people that, how do you call it, they do not have their defenses right and anything will affect them. They are not taking care of their health very well and that is what causes the effects.
An important theme emerged regarding proper and improper uses of water. When farm owners provided water for farmworker use, farmworkers categorized the water into 2 mutually exclusive categories: water for drinking only and water for washing hands only. Drinking water was considered to be water that was chilled with ice, with drinking cups provided. Water for washing hands was considered to be unchilled and located near or inside a restroom, with paper towels and soap provided.
During participant observation, study staff noticed that water was available in only 1 location for farmworkers to use as they chose, either for drinking or for hand washing. However, farmworkers chose not to use this water for either purpose because it was provided in a large, opaque container. Farmworkers said they were afraid to use the water because they did not know what was inside the opaque containers or whether the containers had been properly cleaned.
Farmworkers also expressed a belief that the body must be given time to cool before entering a shower, even a warm shower, because the impact of cold water on a hot body could cause pain in their bones and joints:
If I get home and get wet … my hands would hurt a lot because I was [working] with the hoe all day. For example, right now since I am thinning, my hands get very hot. If I get home and get wet, after a while it hurts my hands. That is why I do not get home and get wet; because it harms me.
There are several limitations and strengths of this study that should be considered. Nearly 80% of our interview sample was female, which may serve as a significant limitation in light of national statistics indicating that farmworkers are typically male.14 Our recruitment efforts targeted men and women equally. Despite rigorous efforts to increase the number of men in the study, males repeatedly indicated that their wives performed the same work and asked if their wives could participate in their place. It is unclear whether the gender proportions of our participant sample biased our results about risk, but we found no striking differences in interview responses between the men and women who participated in our study. This aspect of our study may also point to an important new area for future research, as the US farmworker population may become proportionately more female over the next several years.
Farmworkers in this study indicated having allergic reactions to powdery pesticide residues, which were often perceived as mere dirt. This finding has several important implications. Because some farmworkers perceived powder to be virtually harmless, they may have chosen to wear less protective gear when working around these substances. Further, because farmworkers described experiencing allergy-like symptoms that were a result of their exposure to powder, allergy symptoms may be associated with low perceived risk and subsequent reduced use of protection. It is also possible that farmworkers who suffer from allergies will use protection to reduce their symptoms. More research is needed to understand the impact of beliefs about powder and allergies on protection use. Our findings about powder and allergies also suggest the critical need for education about the forms and dangers of pesticides.
To our knowledge, one previous investigation has elucidated the concept of allergies among farmworkers. That study, however, was limited to skin illness. Individuals in that study experienced skin problems as a result of exposure to certain chemicals, plants, or the sun.15 The authors also reported that most of their participants did not admit susceptibility to allergic skin reactions.15
However, the descriptions of allergies given by participants in our study were quite different. First, farmworkers in our study described allergic reactions beyond skin-related illness. In addition to skin rashes, farmworkers described having watery eyes, red eyes, itchy eyes, sneezing, scratchy throat, and a runny nose. Moreover, participants in our sample used prescription antihistamine medications to treat their allergy symptoms. Finally, farmworkers in our sample were not reluctant to admit their own allergy susceptibility. Men and women were equally likely to suffer from allergies. Also, despite the belief that some farmworkers have a weak body type, having allergic reactions to pesticides was not considered a weakness. Although weakness and vulnerability to pesticides based on gender, age, and body type are discussed in previous reports,1,6 our results indicate that farmworkers believe that both weak- and strong-bodied individuals can be allergic to pesticides. This finding is interesting and complex, because there is a perceived distinction among farmworkers between allergies and weakness.
Other themes that emerged in our work showed that farmworkers sometimes chose not to wear gloves, goggles, and masks because these items decreased their yield and thus their pay. The significance of this finding is 2-fold. First, we discovered that the primary influence that impeded farmworkers’ use of safety equipment was financial. Additionally, our study revealed that farmworkers felt that their employers pressured them not to use forms of protection. Other investigations have shown similar results, indicating that farmworkers did not wear eye protection because it interfered with work16 and that working by contract was related to markers for green tobacco sickness among tobacco farmworkers.17
This finding sheds light on larger issues regarding the vulnerability of farmworkers. Snipes et al. previously18 reported that farmworkers in the Lower Yakima Valley feel that employers are unjust. Other studies have also indicated the existence of policy and environmental justice issues related to farmworker health throughout the United States.19–25 This literature suggests there is a need for better enforcement of farmworker safety regulations and may indicate a need for new policies regarding farmworker payment regulations, given that farmworkers’ decisions about protection were based upon how it would affect their pay.
Finally, we found that farmworkers classified water in the field by its intended use: for drinking only or for hand washing only. Because availability of water for hand washing is critical to farmworkers’ decontamination after pesticide exposure, farmworkers’ beliefs regarding availability and appropriateness of water are key to pesticide safety practices. A recent study showed that farmworkers who washed their hands with soap and wore gloves, long pants, long-sleeved shirts, closed-toe shoes, and a hat decreased their pesticide metabolite levels by nearly 50%.26 Other research has indicated that farmworkers have reported having more water for drinking than for hand washing while in the fields.19 Additional research has reported that farmworkers complain that there is seldom separate water for washing and drinking.20 The findings from this study as well as from others indicate that farmworkers perceive some water as unfit for hand washing, which has important implications for how farmworkers engage in pesticide safety practices.
The other theme that emerged from our data regarding water concerned farmworkers’ delay of showering after returning home from the fields. An early report of this practice was made by Arcury et al., who observed it in North Carolina.20 We observed, however, that farmworkers who held this belief were prone to immediately remove clothing after returning home from work, and they carried their own water into the fields for drinking and washing their hands. Thus, farmworkers’ delay in showering after work served as a contradictory safety behavior. On one hand, delay of showering was associated with postponed decontamination after pesticide exposure. Conversely, farmworkers who delayed showering after work engaged in increased safety behavior, such as washing hands while at work and removing work clothes immediately. Further research is needed to explore the degree to which delay in showering is linked to removing clothes after work or hand washing while in the field.
We urge investigators to incorporate the following key recommendations in future research, education, and interventions for farmworker pesticide safety: (1) implement education and other interventions to address farmworkers’ beliefs about differential susceptibility to the harmful effects of pesticides (e.g., those who are “allergic” vs those who are not); (2) design research questions, questionnaires, and surveys to address farmworkers’ perceptions of wet versus dry pesticide substances (“powder” vs pesticides); (3) include rate of pay as a potential barrier to reduction in risk of pesticide exposure (e.g., whether farmworkers are paid hourly or by yield); and (4) incorporate beliefs about delayed hand washing or showering in interventions for reduction in the risk of pesticide exposure (e.g., whether farmworkers deem water appropriate for hand washing and whether farmworkers delay showering after exposure).
The results of our study also have several policy implications. We recommend that all guidelines of the Environmental Protection Agency’s Worker Protection Standard, such as requirements for training and protective equipment, be extended to harvesters, thinners, and pruners, because they are frequently exposed to daily low levels of pesticide residues. Currently, Worker Protection Standard training and protection guidelines are only mandated for agricultural workers who spray pesticides or for workers who reenter fields after spraying occurs.27 Other agricultural workers are not protected under this standard. Thus, extending Worker Protection Standard protection to all farmworkers could play a significant role in reducing pesticide exposures.
We also recommend that federal and state governments impose regulated payment schedules for farmworkers, because payment by yield sometimes prevents farmworkers from taking recommended breaks, wearing recommended protective gear, or washing their hands. Finally, we recommend greater oversight of mandated provision of materials and facilities for decontamination (specifically soap, water without ice, and paper towels). Following all of the recommendations listed above may have important implications for farmworker safety and reduced risk of pesticide exposure.
This research was made possible by grants from the National Science Foundation (award ID BCS 0424447), the National Institutes of Health (ES09601), and the Environmental Protection Agency (R826886). Writing support for this research was contributed by the W. K. Kellogg Foundation’s Health Scholars Program, Multi-disciplinary Track (grant P0117943), and by an EXPORT Center of Excellence grant provided by the National Center on Minority Health and Health Disparities, National Institutes of Health (5 P60 MD000503).
We would like to thank Gloria Coronado, PhD, Ilda Islas, Genoveva Ibarra, and other project staff from “For Healthy Kids!” who were instrumental in the success of this research project. We would also like to thank the farmworkers who participated in this study; without them, this research would not have been possible.
ContributorsS.A. Snipes, B. Thomson, K. O’Connor, and B. Shell-Duncan collaborated in developing the ethnographic community-based research protocol. S. A. Snipes also collected all data through unstructured interviews, semistructured interviews, and participant observation; analyzed the data; and was the lead author of the article. B. Thomson, K. O’Connor, and B. Shell-Duncan assisted in writing initial drafts of the article. D. King and A. P. Herrera contributed to successive drafts of the article. B. Navarro assisted with ethnographic data collection and performed all Spanish-to-English translations.
Human Participant Protection
All study procedures were approved by the human subjects board at the University of Washington and the institutional review board at the Fred Hutchinson Research Center in Seattle, WA. Informed consent was obtained from all participants before enrollment.
Reprints can be ordered at http://www.ajph.org by clicking on the “Reprints/Eprints” link.