|Home | About | Journals | Submit | Contact Us | Français|
To assess water quality in migrant farmworker camps in North Carolina (NC), and determine associations of water quality with migrant farmworker housing characteristics.
Data were collected in 181 farmworker camps in eastern NC during the 2010 agricultural season. Water samples were tested using the Total Coliform Rule (TCR) and housing characteristics were assessed using NC Department of Labor (NCDOL) standards.
A total of 61 (34%) of 181 camps failed the TCR. Total coliform bacteria were found in all 61 camps, with E. coli also being detected in 2. Water quality was not associated with farmworker housing characteristics or with access to registered public water supplies. Multiple official violations of water quality standards had been reported for the registered public water supplies.
Water supplied to farmworker camps often does not comply with current standards and poses a great risk to the physical health of farmworkers and surrounding communities. Expansion of water monitoring to more camps and changes to the regulations such as testing during occupancy and stronger enforcement are needed to secure water safety.
Water is the essence of human life, part of every cell and vital for every function of our body. The World Health Organization (WHO) has declared access to a regular supply of safe water a basic human right which has to be respected, protected, and fulfilled1. The consequences of failing to do so are alarming: diarrheal diseases causing two million annual deaths worldwide; outbreaks of cholera, legionella and other waterborne pathogens; and cancer and tooth and skeletal damage due to unsafe levels of arsenic and fluoride2. Developing countries carry the main burden of these diseases. However, provision of safe water is a global responsibility shared by individuals as well as local, state and federal governments. For this purpose international and local guidelines and regulations have been created to ensure drinking water quality3–5.
Migrant farmworkers are a particularly vulnerable population within the US for diseases resulting from unsafe drinking water, as well as other environmental and occupational hazards6. The large number of migrant farmworkers that labor in the US has little control of their living environments, including the water available for drinking, bathing, and laundry7. Although federal and state regulations have been developed to ensure adequate and safe housing for migrant farmworkers, the standards are often not met8. The few studies published on water quality show that contamination of drinking water is a continuing concern9. While pesticides and nitrates contained in drinking water represent a recognized problem, this study focuses on fecal contamination of water10. Total coliform were found in drinking water of half of 30 farmworker camps tested in North Carolina in 198911. An U.S. Environmental Protection Agency (US-EPA) investigation into the drinking water quality of migrant camps in Wisconsin revealed that water in 67% of camps contained total coliform bacteria in 199712. Two of five farmworker camps located in Colorado were also tested positive for total coliform in 200213. Breaches in water safety are persistent and widespread9.
This study describes the quality of drinking water and tests its association to housing characteristics and living conditions in 181 migrant farmworker camps located in eastern North Carolina during the 2010 agricultural season. The results are further compared to previously published data of water quality11.
This project was based on a conceptually-based model of community-based participatory research for involving community members in the assessment of farmworker housing and water quality, and using research results to address public health policy. This model of community-based participatory research operationalizes participation, including differentiating modes of community participation in research, and delineating the domains of research in which community members are involved, such as consultation, strategic planning, implementation, and dissemination. It defines community-based participatory research as translational science in which results are provided to community members in a usable format, and are used to address public health policy14, 15.
A cross-sectional study design was used in this study to document housing quality and exposures to contaminated water, pesticides, and allergens in farmworker camps and homes. The presented data focus on the quality of water provided to migrant farmworkers and the characteristics of their respective residences.
A migrant farmworker is defined by having established a temporary, mostly seasonal residence for work in agriculture. While the exact number of migrant farmworkers is not known, approximately 1.01 million hired farmworkers work in the US based on data from the 2006 Current Population Survey16. The majority are Hispanic or Latino, with 84% of migrant and seasonal farmworkers in the US self-identified as Hispanics, and 75% of all farmworkers having been born in Mexico, 23% in the US, 2% in Central America, and 1% in other countries17. The challenges facing migrant farmworkers are multifaceted; lack of social support, stress of adaptation, discrimination, economic and material difficulties, rootlessness, language and literacy barriers, and limited or no access to healthcare all contribute to the vulnerability of this group6–8, 11. These factors are further amplified by the unauthorized work status of about half of farmworkers 18.
A total of 186 migrant farmworker camps were surveyed during the 2010 growing season (June to October 2010) in 16 eastern North Carolina counties: Caswell, Craven, Cumberland, Duplin, Edgecombe, Greene, Halifax, Harnett, Johnston, Lenoir, Nash, Person, Sampson, Wake, Wayne, and Wilson. Camps were defined as residence areas occupied largely by unaccompanied male workers. These included barracks and other communal residences or clusters of residences where workers shared housing facilities. Lists of camps were obtained from the North Carolina Farmworkers Project and partnering clinics (Center Carolina Family Health Center, Kinston Community Health Center, Piedmont Health Services, Inc.). Over the course of data collection, field supervisors expanded the list as they encountered new camps. All identified camps were contacted to participate in the study by approaching the farmworkers residing in the camp. The farmworkers were informed about the study including the goals, involvement expected, and risks of participation. Individuals agreeing to participate gave written consent. Reasons for refusal of individual farmworkers were documented, if provided.
Interviews, housing assessments, and water sampling were completed by trained staff members who were fluent Spanish speakers. Interviews were performed with two residents in each participating camp, and a camp assessment was conducted with the assistance of a third resident farmworker. Interviews assessed demographic information, housing features, and perceptions of housing quality. Data collection forms were developed in English and translated into Spanish by a native Spanish speaker familiar with Mexican Spanish. The forms were reviewed by staff members of the community partners who were native Spanish speakers. Revised forms were field tested, with the interview questionnaires being pretested with four male migrant farmworkers. All materials were revised based on the field test.
Interviews assessed participant personal characteristics, housing features, and perceptions of housing quality. Farmworkers who completed the interviews helped with an assessment of their sleeping rooms. Interviews took approximately 90 minutes to complete.
For the housing assessment, data collectors observed, asked questions, and used instruments such as a flashlight and an extending mirror to inspect inside cabinets and behind appliances for signs of pest infestation and exposed wires. The housing assessment included a 129-item housing assessment form to assess compliance with 32 North Carolina Department of Labor (NCDOL) Migrant Housing Inspection standards18. Thirteen additional NCDOL standards were not assessed due to seasonality (e.g. heating) or practicability (e.g. hot water supply testing requiring prior usage). The 32 standards tested were divided into the categories: (1) housing site, (2) structures, (3) kitchen, (4) toilets, (5) laundry and bathing, (6) heating, (7) water and sewer, (8) garbage, (9) pests (cockroach, rodents), (10) health, and (11) general duty. The posting of the migrant housing certificate in a place accessible to all the migrant residents as mandated by Section 0200 of the North Carolina Administrative Code was documented18. The data collectors also assessed 49 non-mandated housing items such as public registered water source, proximity of residences to livestock, standing water on the camp grounds, and overall camp cleanliness.
Water samples were collected according to the guidelines laid out by the Public Water Supply Section of the North Carolina Department of Environment and Natural Resources19. Samples were collected from the kitchen faucet or another faucet if the kitchen faucet was not accessible. Data collectors first cleaned the faucet with an alcohol wipe and then let the water run for five minutes before sampling. Water was collected in sterile laboratory-provided containers. The containers were opened at the time of collection, and data collectors were trained to avoid contamination by not touching the inside of the lid, rim or inside of the container with either their fingers or the faucet itself. Housing assessment and water collection took approximately 60 minutes.
The outcome measure for this study is water quality. The water samples were delivered to state-certified laboratories within 24 hours, where the samples were tested for total coliforms and E. coli following standard method 9223 (total coliform rule [TCR])20. The labs used a selective and differential medium for the determination of the presence or absence of total coliforms and E. coli in drinking water based on enzyme activity (Colilert [IDEXX Laboratories, Inc, Westbrook, Maine], Colitag [CPI International, Santa Rosa, CA])21, 22. Both tests are United States Environmental Protection Agency (EPA) certified and standardized.
Quality control procedures consisted of a total of 23 duplicates (13% of the entire sample) collected across the counties. Duplicate samples were obtained at the same time as the actual samples, and then a special ID was created to blind the laboratory. The purpose of the repeat measures was to test the laboratories’ accuracy and the potential of sample contamination.
Camp water sources were identified as non-transient, non-community public (NTNC) water systems and other systems using the web-accessible water registry data and geographic information system (GIS) coordinates for the individual camps19. A NTNC public water system is not a community system, but regularly serves at least 25 of the same people for more than six months per year. They are regulated by the Safe Drinking Water Act (SDWA) of the EPA and the North Carolina Rules Governing Public Water Systems4, 23. The identification of NTNC and other systems using GIS data and public records allowed comparison of the water quality data by system set-up without revealing the identity of the camp locations to second parties. The web-accessible data also provided data on previously reported violations in water quality for each NTNC24.
A rainfall data set was requested from the North Carolina Retrieval and Observations Network of the Southeast Database (CRONOS) for all the counties in the study (http://www.ncclimate.ncsu.edu/cronos). The data set contained information from various weather sensors spread across the study counties. A near analysis in ESRI ArcInfo (Version 9. 3 ESRI Redlands, CA) was performed to determine which sensor was the closest to each respective camp location. The near analysis was conducted in the GIS lab at the Center for Community Safety at Winston-Salem State University. A rainfall summation (total in inches) for the seven days prior to data collection per closest sensor for each camp was calculated and compared to the water quality data. Furthermore, potential clustering of camps failing the water testing was conducted by mapping the geographic information for each camp.
Several additional camp characteristic measures were included in the analysis as potential correlates of water quality. Data collection time had the values of early season (June through mid-July), mid season (mid-July through August), and late season (September and October) for when in the season the camp was assessed. Housing type was based on the presence or absence of barracks in the camp. Camps with barracks could also have non-barrack housing, such as houses and trailers. Non-barracks camps had only houses or trailers. Number of camp residents was divided into three categories: 1 to 10; 11 to 20; and 21 or more. Number of housing units in the camp had the values, 1, 2, and 3 or more. Presence of female residents was a dichotomous measure. H-2A status was a dichotomous measure indicating whether any farmworkers with H-2A visas were living in the camp. The H-2A program is currently the only agricultural guest-worker program in the US allowing individuals’ employment in specific agricultural areas for a pre-determined time period. While the number of H-2A visa holders varies by state, North Carolina has a relatively large group of farmworkers working under this visa program. North Carolina Department of Labor (NCDOL) inspection certificate posted was a dichotomous measure. All camps inspected by the NCDOL should post the inspection certificate; as all of the camps included in this study housed migrant farmworkers, all should have been inspected. Camp violations of NCNOL standards had the values, 4 to 10, 11 or 12, and 13 or more25. Overall cleanliness of dwelling was a dichotomous measure based on the data collector’s evaluation. Pest infestation was a dichotomous measure indicating the presence of cockroaches or rodents in the dwelling. Standing water in camp was a dichotomous measure indicating whether the data collector observed standing water. Proximity to livestock shelters was a dichotomous measure indicating whether livestock shelters were visible from the camp.
Descriptive statistics (frequencies and percentages) were used to describe the camp characteristics. Bivariate associations between camp characteristics and water quality were examined using chi-square tests. A p-value of less than 0.05 was considered statistically significant. All analyses were performed using SAS 9.2 (Cary, NC).
A total of 186 camps were enrolled in the study. Residents in an additional 36 camps declined to participate, and the grower refused to permit participation in another four camps. The resulting camp participation rate was 82.3% (186/226). In five of the 186 participating camps data collection was not completed due to the intervention by the grower. Therefore, the final sample for this analysis included 181 (80%) camps. Reasons given by individual farmworkers for non-participation ranged from being too tired or not interested, cooking/eating/drinking to no reason.
Table 1 summarizes the characteristics of the enrolled camps. Camps having a NTNC public water source had more housing units (NTNC: 76% ≥ 3 housing units vs. Non-NTNC 11% ≥ 3 housing units), and more occupants (NTNC: 80% > 20 occupants vs. Non-NTNC 18% > 20 occupants) living in barracks (NTNC: 90% barracks vs. Non-NTNC 23% barracks) than non-NTNC camps (p<0. 05). No difference was detected in posting frequency of the NCDOL certificate (NTNC: 33% posted vs. Non-NTNC 35% posted; p>0.05).
A total of 61 (34%) out of 181 camps failed the North Carolina water quality requirements. Total coliform bacteria were found in the water of all 61 camps with E. coli also being detected in 2 of them. The 23 quality control samples matched the results (14 repeated samples negative, 9 repeated samples positive) found in the original probes.
Bivariate analyses of the water testing results (outcome) and camp characteristics are summarized in table 1. None showed statistically significant associations including NCDOL certification or having a NTNC public water source. Eighteen of the 21 (86%) camps that received water from NTNC system had at least one monitoring violation, and 15 (71%) had at least one positive TCR test results for the years 2009 and 2010. Only one NTNC camp passed the water testing requirements. No geographic clustering of camps failing the water testing was found.
Testing the quality of drinking water is essential to protect the public from communicable diseases4. The purpose of this project was to assess the housing conditions of migrant farmworkers in eastern North Carolina. As part of the assessment the drinking water available to workers in their residences was tested. The Migrant Housing Act of North Carolina § 95–225 (c) Adoption of standards and interpretations outlines that “the Commission for Public Health shall adopt and the Department of Environment and Natural Resources shall enforce rules that establish water quality and water sanitation standards for migrant housing under this Article”. Providers of the housing are responsible for supplying adequate and sanitary water (North Carolina Migrant Housing Act Article 19 §95–223)18. Drinking water was tested using the Total Coliform Rule (TCR) monitoring for total coliform and E. coli contamination18, 26. Using the standardized water testing procedure during the growing season, our study found that one-third of residences failed the TCR. Coliform bacteria and E. coli were detected indicating water contamination with human pathogens. This affects not only farmworkers but also puts the surrounding community consuming water from the same sources at risk.
The literature describing water quality in migrant farmworker camps is sparse9, 11, 12, 13. Vela-Acosta et al. reported total coliform in 2 out of 5 farmworker camps located in Colorado in 200213. In 1988 and 1989 Ciesielski et al. found total coliform in 44% and 50% (second year) of 27 and 30 farmworker camps in North Carolina11. Our study confirms the findings and indicates the persistence of this problem over two decades.
The consequences of consuming contaminated water are dire. In a review of drinking water-associated outbreaks from 1971 to 2006 in the US, over 577,000 persons were affected in 780 outbreaks27. Most suffered from gastrointestinal illnesses (685 [88%]) such as diarrhea or vomiting, followed by hepatitis A (29 [4%]), and acute respiratory illnesses (24 [3%]) (e.g., legionellosis). There were also rare illness types (15%) ranging from change in hair color to miscarriages. Water samples contained parasites (18%), non-legionella bacteria (14%), chemicals (12%), viruses (8%), legionella (3%), and mixed causes (<1%). However, no cause could be detected in most outbreaks (45%). Unfortunately, information regarding how farmworkers are affected by unsanitary water is very limited. One documented outbreak of typhoid fever occurred in a migrant farmworker camp in Dade County, Florida in 197328. Two-hundred and forty-six individuals required hospitalization during the outbreak but, fortunately, none died. The water supply system was later identified as the source. The lack of reports does not rule out the existence of outbreaks. Reasons may be small camp sites resulting in low case counts, the often self-limiting character of the diseases, the legal status of farmworkers combined with barriers to access health care, or the non-existence of a public reporting and documentation system for these settings.
Using the data collected during housing assessments, associations between camp site characteristics and water system failures were considered. None showed a significant connection with the water testing outcomes. This is surprising for camps with NCDOL standard certification, and also for registered NTNC public water supply sources. NCDOL standards require water testing before occupancy. However, display of the NCDOL pre-occupancy certification did not ensure sanitary water quality during occupancy. Over two decades ago Ciesielski et al. reported similar findings11. Only one camp inspected by NCDOL before occupancy was cited for water violations followed by 44% and 50% camp failure rates during the two seasons studied. Receiving water from a NTNC registered source was also expected to provide higher quality water due to more frequent testing and control. However, there was no difference between failure rates of NTNC (7 out of 21) and other water supplies (55 out of 160). Citations for water violations in the NTNC sources further revealed that over 70% of the camps had at least one positive TCR result and one or more monitoring violations (TCR tests not performed) in 2009 and 2010. Only one of the 21 NTNC camps passed the requirements.
Why the two regulatory tools, NCDOL and NTNC standards, failed to protect farmworkers from contaminated water is not fully understood. Two factors may have contributed: timing of testing and enforcement of the rules. Inspecting camps only before occupancy did not ensure water safety. Therefore, testing during the growing season may improve compliance with the standards. For NTNC water supplies, more frequent testing is already part of the regulations. But enforcement appears to be weak allowing providers to ignore water safety breaches or just not test at all. The scale of the penalties should urge providers to choose protection of consumers over indifference. Finding contaminated water in registered public water supply systems may also point toward substantial risks for the general community using similar sources.
There are limitations. The primary goal of this project was to assess the housing characteristics of migrant farmworkers in North Carolina. Interviews of the occupants and extensive sampling of the living spaces and exposure risks from infestation to pesticides were the principal means. However, the water supply set-up was not documented in detail. This leaves open why water systems failed and will require further investigations. The risk of water sample contamination during collection was minimized by following the standardized water collection protocol provided by the state-approved testing laboratories. The repeat samples confirmed previous results rendering contamination unlikely. The cross-sectional design provides a snapshot of the conditions found in the camps in one region over one season. Generalizations of the outcomes to other regions and seasons should be made with caution. However, the high number of recent TCR violations noted in registered NTNC camps over a two year period, and similar results found in an investigation in North Carolina executed two decades ago point to a persistence of these problems11.
This study revealed a major breach in housing safety for migrant farmworkers in rural North Carolina. One third of the camp sites investigated failed the TCR indicating a high likelihood of exposure to contaminated drinking water. The risk of physical harm such as diarrhea, vomiting, or acute respiratory illnesses further threatens the already vulnerable population of migrant farmworkers but also residents in adjacent communities using similar water supply systems. Based on a previous assessment in the 1980s and our current findings in a much larger sample, it appears that existing regulations continue to fall short in securing mandated water safety requirements9, 11. Covering all migrant housing by NCDOL inspections, amending the regulations to water quality testing not only before but also during occupancy and strengthening enforcement of these regulations may be the first steps to improve the water quality of migrant farmworker housing and their surrounding communities.
Funding for the study was provided by the National Institute of Environmental Health Sciences (Grant number: R01-ES012358).
Contributors: W. E. Bischoff developed the study aim for water testing within the governing housing project, guided the data analysis and interpretation, and served as the lead author and editor of the article. M. Weir was the project manager, coordinated data collection, coding, and summaries, and critically reviewed and revised drafts of the manuscript. P. Summers collected data, performed the GIS location analysis, and contributed to the revision of the manuscript. H. Chen served as the biostatistician, performed the data analysis and drafted the statistical analysis section. S.A. Quandt contributed to the study design and interpretation of the findings, and critically reviewed and revised the manuscript. A. K. Liebman provided essential insights in migrant farmworker policies and their interpretation, and critically reviewed and revised the manuscript. T.A. Arcury was the principal investigator of the governing housing project, developed the study idea, design, and methodology, provided guidance of the data collection and analysis, and critically reviewed and provided feedback on the article.
Human Participant Protection: Institutional review board approval was obtained from the Wake Forest School of Medicine.