Journal of Occupational & Environmental Medicine:
Job Control, Psychological Demand, and Farmworker Health: Evidence From the National Agricultural Workers Survey
Grzywacz, Joseph G. PhD; Alterman, Toni PhD; Gabbard, Susan PhD; Shen, Rui PhD; Nakamoto, Jorge PhD; Carroll, Daniel J. BSc; Muntaner, Carles PhD
From the Department of Human Development and Family Science (Dr Grzywacz) and Center for Family Resilience (Dr Grzywacz), Oklahoma State University, Tulsa; Division of Surveillance, Hazard Evaluation, and Field Studies (Dr Alterman), National Institute for Occupational Safety and Health, Cincinnati, Ohio; Aguirre Division (Drs Gabbard and Nakamoto), JBS International, Burlingame, Calif; Emergint Technologies (Dr Shen), Cincinnati, Ohio; Employment and Training Administration (Mr Carroll), US Department of Labor, Washington, DC; and Dalla Lana School of Public Health and Bloomberg Faculty of Nursing (Dr Muntaner), University of Toronto, Toronto, Ontario, Canada.
Address correspondence to: Joseph G. Grzywacz, PhD, Department of Human Development and Family Science, Oklahoma State University, 700 N Greenwood Ave, Main Hall #2120, Tulsa, OK 74106 (firstname.lastname@example.org).
The authors declare no conflicts of interest.
Objective: Improve understanding of the potential occupational health impact of how agricultural jobs are organized. Exposure to low job control, high psychological demands, and high job strain were hypothesized to have greater risk for poor self-rated physical health and elevated depressive symptoms.
Methods: Cross-sectional data (N = 3691) obtained using the Work Organization and Psychosocial Factors module of the US National Agricultural Workers Survey fielded in 2009–2010.
Results: More than one fifth (22.4%) of farmworkers reported fair/poor health, and 8.7% reported elevated depressive symptoms. High psychological demand was associated with increased risk of fair/poor health (odds ratio, 1.73; 95% confidence interval, 1.4 to 2.2) and elevated depressive symptoms (odds ratio, 2.6; 95% confidence interval, 1.9 to 3.8).
Conclusions: The organization of work in field agriculture may pose risks for poor occupational health outcomes among a vulnerable worker population.
Farmworkers in the United States, the vast majority of whom in the United States are immigrants from Mexico, are a health-disparate population in a hazardous occupation.1–3 Every day, farmworkers across the country confront occupational exposures that pose both immediate and long-term health threats. Farmworkers regularly work in extreme temperature conditions that pose risk for heat illness and fatality.4,5 They experience repeated low-dose exposure to known neurotoxicants,6,7 and they may work with machines or livestock that pose risk for traumatic injury.8,9 These previously identified and visible occupational health threats are important and likely contribute to occupational health inequalities.
More insidious are exposures whose origins lie in the way farmwork is organized, or how farmwork jobs are designed and managed.10 Initiatives in both the Unites States and Europe are drawing attention to psychosocial aspects of jobs that create and exaggerate health inequalities.11,12 The Job Demands–Control Model13 is among the most widely used models used to study the health implications of psychosocial aspects of jobs. The Job Demands–Control Model posits that the absence of control or the inability to make decisions about the way work is performed affects worker health. Indeed, several studies have linked control with various health outcomes, ranging from intima-media thickness of the carotid arteries14,15 to sickness absence,16 self-reported poor health,17 and disablement.18,19 Likewise, exposure to high levels of psychological demand, typically understood as the patterned and unexpected psychological stressors arising from the timing and pace of work, has been linked with several health outcomes.20–22 Unlike the temperature and chemical exposures that can have immediate health impact, the health effects of job control and psychological demands are likely latent, perhaps not manifesting in poor health for months or years.
Very little research has examined the potential health implications of job control and psychological demands among farmworkers. Grzywacz and colleagues10 noted in a small regional sample that farmworkers were exposed to modest to moderate levels of psychological demand, and they reported having a low level of control. In this and subsequent research, these investigators found that demands and control were associated with mental health-related quality of life10 and with pesticide exposure indicated by the presence of urinary diaklyphosphate metabolites.23 Unfortunately, this research was completed in small regional samples, which although important, does not offer insight into the role job control and psychological demands may play in the broader health of farmworkers.
The goal of this study was to improve understanding of the potential population health impact of how jobs in agriculture are organized. To achieve this goal, we draw on core propositions of the Job Demands–Control Model13 to test three basic hypotheses: (1) workers exposed to low job control have greater risk for poor self-rated physical health and elevated depressive symptoms, (2) workers exposed to high psychological demand are at greater risk for poor self-rated health and elevated depressive symptoms, and (3) workers in high-strain jobs (ie, those with low job control and high psychological demand) are at elevated risk for poor self-rated health and elevated depressive symptoms.
The data for this analysis are from a supplement added to the 2009–2010 National Agricultural Workers Survey (NAWS). The NAWS is the primary source of data on US farmworkers. Each year since federal fiscal year 1989, the NAWS has conducted interviews with a national probability sample of field workers employed in crop agriculture, not including workers with a temporary work permit (H2A visa). The US Department of Labor sponsors the NAWS, and it is fielded by the Aguirre Division of JBS International.
A detailed description of the NAWS sampling and weighting can be found at the Department of Labor's Web site (see http://www.doleta.gov/agworker/naws.cfm) and has been reported in other articles,24–26 but is summarized here. The NAWS uses a multistage sampling design to account for seasonal and regional fluctuations in the level of farm employment. The year is divided into three interviewing cycles, each lasting 4 months, to capture seasonal fluctuations in the agricultural workforce. The number of interviews allocated to each cycle is proportional to the amount of crop activity at that time of the year.
During each interview cycle, four levels of sample selection are used: region, county cluster, employer, and field worker. At the highest level, the NAWS sampling scheme divides the continental United States into 12 regions. Each region in turn consists of clusters of counties that have similar farm labor usage patterns. County selection is made from a standing roster of randomly selected county clusters. For every cycle, in each region, the staff draws a random sample of county clusters from the roster and uses a random sorting procedure to order counties within clusters. The penultimate sampling stage is the selection of agricultural employers using simple random sampling. The NAWS staff compiled a list of agricultural employers from public agency records. Field staff review, supplement, and update the lists annually using local information.
All NAWS data are collected through interviewer-administered survey questionnaires in a face-to-face interview by trained interviewers. Before approaching workers, interviewers are trained to contact the selected growers or contractors, explain the purpose of the survey, and obtain access to the worksite to schedule interviews. Interviewers then go to the farm, ranch, or nursery and select a random sample of workers by using field-sampling techniques designed by a statistician. As such, the sample includes only workers actively employed in agriculture at the time of the interview. Before collecting data, interviewers explain the purpose of the survey to the workers, ask them to participate, and obtain informed consent. Interviewers administer the questionnaire in the location and language (ie, English or Spanish) of the worker's choice. Workers receive a $20 honorarium for their participation.
A supplemental module titled Work Organization and Psychosocial Factors was added to the NAWS survey in 2009–2010. The content of the Work Organization and Psychosocial Factors module was based on recommendations from a multidisciplinary panel of experts convened in 2004. The supplemental module was available in Spanish and English.
Self-rated health was assessed using the standard global health question, “In general, how would you describe your health?” Individuals who self-rated their health as poor or fair were coded as 1; those who self-reported good, very good, or excellent health were coded 0. Elevated depressive symptoms were assessed with a 10-item version of the Center for Epidemiologic Studies Depression Scale (CES-D) used in previous studies of Latino farmworkers. 27–29 The internal consistency of the short CES-D in the current sample was 0.76. Depressive symptoms were summed and classified such that individuals with a score of 10 or higher were characterized as having elevated depressive symptoms.28
Three key concepts of the Job Demands–Control Model are considered in this analysis. All independent variables were assessed using selected items from the Job Content Questionnaire.13 Results from previous research suggest that short measures of psychological demand and job control perform comparably to longer measures.30 The Job Content Questionnaire items used in this study were modified in two ways. First, they were modified to narrow participants' consideration to only their agriculture-related job (eg, “In your current job in farmwork...”). Second, the response options were changed from affective response options (ie, strongly disagree to strongly agree) to frequency-based response options (ie, “never” to “always”). This modification is consistent with previous research in this population10 and with the results obtained from cognitive testing, indicating that Latino farmworkers have greater difficulty using affective response categories than frequency-based responses.31
Psychological demands, which are generally understood as the psychological stressors that arise while carrying out tasks and responsibilities associated with a job, were defined with two items asking, “In your current job in farmwork, how often ... (1) Does your job require you to work very hard? and (2) Are you asked to an excessive amount of work?” Participants were assigned a value of 1 or as having elevated psychological demand if the sum of the two items was greater than or equal to 2, suggesting that they responded “sometimes” to both questions, or they indicated “very often” to one of the questions. Individuals whose summed score was less than 2 were coded 0, as not having elevated psychological demand.
Control or the relative degree of freedom workers have over which tasks are performed and how those tasks are accomplished was also defined using four items modified from the Job Content Questionnaire (eg, “In your current job in farmwork, how often do you have the freedom to decide how you do your work?”). The four items were summed. Individuals were given a value of 1, indicating low control, if the summed score was less than or equal to 2 and no single response to the four items had a value of 2 or higher (indicating “very often” or “always”). All others were coded 0, indicating the absence of low control. Job strain is coded such that individuals with both elevated psychological demands and low control are coded 1, 0 otherwise.
Predictors of elevated psychological demands, low control, and job strain were organized into four basic groups. The first group was personal characteristics, including age (continuous), sex (females = 1), whether the participant is traveling alone or with family members (unaccompanied = 1, 0 otherwise), and whether the individual is authorized to work in the United States (no = 1). Additional covariates presumed to capture different aspects of acculturation are ethnicity (Latino, non-Latino White, and non-Latino other), an indigenous (Native American) language was spoken in the participants' home as a child (yes = 1), and self-reported ability to read English well (yes = 1). Other covariates include educational attainment (categorized into no formal education, 1 to 9, 10 to 12, and more than 12 years of formal education), years in agriculture (categorized into less than 5, 5 to 15, and more than 15 years), and region of the country (California, Northwest, Southwest, Midwest, East, and Southeast). A single dichotomous variable reflecting whether the participant was interviewed during the high season was constructed to account for possible variation in self-reported control and psychological demand imposed by more intense production expectations.
The analytic sample included 3691 farmworkers. Descriptive statistics were calculated for all the characteristic variables and the health outcomes. Two multivariate logistic regression models were fit for each outcome (ie, fair/poor self-rated health and elevated depressive symptoms). The first model explored the simple unadjusted associations of low job control, high psychological demand, and job strain with each outcome. Counter to our hypothesis, job strain was not associated with either outcome and was subsequently dropped from all analyses. The second model tested the hypothesized associations of low job control and high psychological demand with each outcome, controlling for personal characteristics as well as acculturative and occupational factors. All analyses were unweighted, but accounted for the complex survey design, including the clustering of farmworkers within region and within data collection cycle. All the analyses were done using SAS 9.3 (SAS Institute, Cary, NC).
On average, participants in the sample were about 38 years of age (SD = 12.9 years), most were males (17.3% females), and 44.6% were in the United States without a spouse/partner or another family member (Table 1). More than two -thirds of the sample reported having less than 9 years of education, and more than one half of them (51.3%) did not have authorization to work in the United States legally. Four of five participants in the sample (83.4%) were Latino, although 12.6% of participants were non-Latino White. Approximately 7% of the sample reported that an indigenous language was spoken in the home during childhood, and nearly one quarter of the sample reported being able to read English well.
Participants had worked in agriculture for about 15 years (SD = 11.3 years) and had performed 228 days (SD = 80.0 days) of fieldwork in the past year. One half of the sample was interviewed during “high season.” Approximately one in five workers (22.4%) reported fair/poor health, and 8.7% of the sample reported elevated depressive symptoms.
Results from simple, unadjusted analyses provided mixed results for the hypothesis that low control is associated with poorer health outcomes (Table 2). In contrast to our hypothesis, low control did not have a statistically significant association with fair/poor self-rated health. Nevertheless, the odds of reporting elevated depressive symptoms were 87% higher for individuals with low control than for those without low control. As hypothesized, farmworkers with elevated psychological demand were 38% more likely than those without elevated psychological demand to report fair/poor health, and 2.3 times more likely to report elevated depressive symptoms. Preliminary analyses yielded no evidence that job strain was associated with fair/poor health or elevated depressive symptoms.
Multivariate analyses were generally consistent with the pattern of associations observed in simple models (Table 3). Low control did not have a statistically significant association with fair/poor health after controlling for other covariates, but the odds of fair/poor health were 73% greater (95% confidence interval, 1.4 to 2.2) for farmworkers exposed to elevated psychological demand relative to those unexposed. Low control did not have a statistically significant association with fair/poor health after controlling for other covariates, but the odds of elevated depressive symptoms were more than 2.6 times (95% confidence interval, 1.9 to 3.8) greater for those exposed to elevated psychological demand relative to those unexposed.
Other predictors of fair/poor health also warrant attention (Table 3). Controlling for exposure to low control and elevated psychological demand, every 1-year increase in farmworker age was associated with a 2% increase in the odds of reporting fair/poor health. The odds of fair/poor health were 28% greater for females relative to males. The odds of fair/poor health decreased with increased educational attainment.
The odds of reporting elevated depressive symptoms were nearly two times greater for female than for male farmworkers (odds ratio, 1.86), controlling for exposure to low control and high psychological demand (Table 3). An inverse graded relationship was observed between educational attainment and elevated depressive symptoms. Being unaccompanied and lacking legal authorization to work in the United States were both associated with increased odds of reporting elevated depressive symptoms. Risk for elevated depressive symptoms was 2.5 times greater for non-Latino other (predominantly Black) than for Latinos. Every additional day spent in the fields was associated with a 1% decrease in the odds of elevated depressive symptoms.
Farmworkers are frequently described as a medically underserved and health-disparate population.1–3 Although appropriate attention has been given to clear hazards that pose immediate morbidity and mortality threats to farmworker health, substantially less research has considered more insidious threats to the health and well-being of this health-disparate workforce. The goal of this study was to improve understanding of the potential population health impact of how jobs in agriculture are organized. To achieve this goal, we analyzed data from the Work Organization and Psychosocial Factors module added to the 2009–2010 NAWS to test three basic hypotheses from the Job Demands–Control Model.13 Results of hypothesis testing make several contributions to the literature.
Counter to expectations, there was no evidence that low control was associated with poorer physical or mental health in this sample of farmworkers, the vast majority of whom are Latino. Several reviews suggest that low control is associated with poorer health outcomes; 11,14,16–19 however, other applications of the demands-control model in farmworker samples also reported few main effects for low control.10,23 The absence of statistically significant associations between low control and health among farmworkers relative to other samples is compelling in that it raises questions about the universality of the presumed beneficial effects of worker control. Restricted range is another possible explanation for the absence of association between control and health outcomes. Further research is warranted.
As hypothesized, we found robust associations indicating that exposure to high psychological demand was associated with poorer physical and mental health among farmworkers. These associations are consistent with the broader literature,20–22 but yet are inconsistent with results obtained from previous studies with farmworkers.10,23 The diverging results observed in this study relative to previous research with farmworker samples may be attributed to two possible explanations. The first is statistical power: the previous studies were performed with substantially smaller samples that may have had insufficient power to detect the deleterious health effects of psychological demand. A second explanation is that the absence of statistically significant associations in previous studies was an artifact of restricted range in exposure. That is, previous research relied on samples of farmworkers primarily involved in tobacco production, whereas this study included workers across a wider variety of crops and agricultural activities. Similarly, whereas the previous results were obtained from samples of predominantly H2A workers, H2A workers are not included in the NAWS. A final explanation is more substantive; previous studies simultaneously modeled the effects of both psychological and physical job demands and found that physical job demands may have greater consequences for farmworker health. The NAWS supplement did not measure physical job demands; consequently, we are unable to account for this possible spuriousness. More research with diverse samples of farmworkers across various crops and more complete assessment of demands confronted by farmworkers is needed.
Results beyond those focused on the focal hypotheses also contribute to the literature. Although other studies have documented injury and illness or have characterized the health of farmworkers in discrete regions,32–35 this is the first study to provide estimates of self-rated health and depressive symptoms of farmworkers from a national perspective. More than one fifth (22.4%) of farmworkers in this predominantly Latino sample reported fair/poor health, which is higher than the 15.8% of foreign-born Hispanics from pooled National Health Interview Survey data reporting fair/poor health.36 The elevated prevalence of fair/poor health among farmworkers relative to the general population combined with our evidence that the risk of fair/poor health increases with each additional year in agriculture supports the general notion that farmworkers are health-disparate and vulnerable worker population.1–3 The observed rate of elevated depressive symptoms (ie, 8.7%) is substantially lower than that reported in previous research: Alderete and colleagues37 reported that 20% of farmworkers in Fresno County, California, had elevated depressive symptoms, as measured by the full 20-item CES-D, and Grzywacz and colleagues29 found that nearly one quarter of farmworkers in eastern North Carolina had elevated depressive symptoms, as measured by the 10-item CES-D used in the current study. The apparent discrepancy in rates of elevated depressive symptoms among NAWS participants and those observed in previous farmworker research is unclear. The North Carolina findings29 offer some potential insight in that these suggest that elevated depressive symptoms were highest early in the agriculture season (18.5%) and lowest in the middle (5.0%) and later (7.8%) portions of the season. In light of the fact that the average worker was in the field 228 days in the past 12 months (see Table 1) and about half of the sample was interviewed during “high season,” it is possible that the low rate of elevated depressive symptoms in this sample reflects the possibility that many participants were interviewed during middle to later stages of the agricultural season. Temporal variation in the physical and mental health of farmworkers across the agricultural season is an important area for future research.
The contributions of this research must be interpreted in light of its limitations. Most obvious is the cross-sectional study design, which impairs the ability to make any causal inferences. A second set of limitations surround measurement issues. Appraisals of psychological demands, at least those connected to the intensity of work, have been documented to vary across the agricultural season,29 but our observations were obtained across six interview cycles spanning 2 years, which likely contributes to diluted estimates of exposure. Furthermore, despite promising results from cognitive interviews of our survey items with Latino agricultural workers,31 measurement equivalence is an ongoing challenge in studying how work organization contributes to health,11 and this challenge is compounded in the ethnically diverse NAWS sample. Other measurement weaknesses, including both comparatively shallow outcomes assessment and narrow predictor assessments of how farmwork is organized, must be recognized. Although deep assessment is counter to the NAWS's surveillance mission, it nevertheless raises potential problems like misclassification and confounding that require results to be interpreted with caution. Finally, although the NAWS provides the largest possible surveillance of agricultural workers, it does have limitations in that it does not include workers with temporary work visas, and its sampling strategy based on receiving permission to interview workers in the field likely results in biases favoring operations that are more humane to workers.38
Limitations notwithstanding, the results of this study contribute to the agricultural health literature. Most notably, these are the first estimates from a national sample of farmworkers suggesting that elevated exposure to psychological demand, or the patterned and unexpected psychological stressors that arise while carrying out the tasks and responsibilities of a job, is associated with poorer health outcomes. Although replication research is needed, these results suggest that the organization of work in field agriculture poses risks for poor occupational health outcomes among a vulnerable worker population.
1. Arcury TA, Quandt SA. Delivery of health services to migrant and seasonal farmworkers. Annu Rev Public Health. 2007;28:345–363.
2. Villarejo D, Baron SL. The occupational health status of hired farm workers. Occup Med. 1999;14:613–635.
3. Villarejo D. The health of U.S. hired farm workers. Annu Rev Public Health. 2003;24:175–193.
4. Luginbuhl RC, Jackson LL, Castillo DN, Loringer KA. Heat-related deaths among crop workers—United States, 1992–2006. MMWR Morb Mortal Wkly Rep. 2008;57:649–653.
5. Culp K, Tonelli S, Ramey SL, Donham K, Fuortes L. Preventing heat-related illness among Hispanic farmworkers. AAOHN J. 2011;59:23–32.
6. Arcury TA, Grzywacz JG, Talton JW, et al. Repeated pesticide exposure among North Carolina migrant and seasonal farmworkers. Am J Ind Med. 2010;53:802–813.
7. Arcury TA, Grzywacz JG, Isom S, et al. Seasonal variation in the measurement of urinary pesticide metabolites among Latino farmworkers in eastern North Carolina. Int J Occup Environ Health. 2009;15:339–350.
8. Shah DJ, Shipp EM, Cooper SP, et al. Hand problems in migrant farmworkers. J Agric Saf Health. 2009;15:157–169.
9. Douphrate DI, Rosecrance JC, Wahl G. Workers' compensation experience of Colorado agriculture workers, 2000–2004. Am J Ind Med. 2006;49:900–910.
10. Grzywacz JG, Quandt SA, Arcury TA. Immigrant farmworkers' health-related quality of life: an application of the Job Demands–Control Model. J Agric Saf Health. 2008;14:79–92.
11. Landsbergis PA, Grzywacz JG, LaMontagne AD. Work organization, job insecurity, and occupational health disparities. Am J Ind Med. [published ahead of print October 16, 2012] doi:10.1002/ajim.22126.
12. Siegrist J, Benach J, Bislimovska J. Review of Social Determinants of Health and the Health Divide in the WHO-European Region Report of Task Group 2: Employment and Working Conditions Including Occupation, Unemployment and Migrant Workers. Copenhagen, Denmark: World Health Organization Regional Office for Europe; 2011.
13. Karasek R, Theorell T. Healthy Work: Stress, Productivity, and the Reconstruction of Working Life. New York, NY: Basic Books; 1990.
14. Fujishiro K, Diez Roux AV, Landsbergis P, et al. Associations of occupation, job control and job demands with intima-media thickness: the Multi-Ethnic Study of Atherosclerosis (MESA). Occup Environ Med. 2011;68:319–326.
15. Muntaner C, Nieto FJ, Cooper L, Meyer J, Szklo M, Tyroler HA. Work organization and atherosclerosis: findings from the ARIC study. Atherosclerosis risk in communities. Am J Prev Med. 1998;14:9–18.
16. Duijts SF, Kant I, Swaen GM, van den Brandt PA, Zeegers MP. A meta-analysis of observational studies identifies predictors of sickness absence. J Clin Epidemiol. 2007;60:1105–1115.
17. Laaksonen M, Rahkonen O, Martikainen P, Lahelma E. Associations of psychosocial working conditions with self-rated general health and mental health among municipal employees. Int Arch Occup Environ Health. 2006;79:205–212.
18. de Croon EM, Sluiter JK, Nijssen TF, et al. Work ability of Dutch employees with rheumatoid arthritis. Scand J Rheumatol. 2005;34:277–283.
19. IJzelenberg W, Burdorf A. Risk factors for musculoskeletal symptoms and ensuing health care use and sick leave. Spine (Phila Pa 1976). 2005;30:1550–1556.
20. Kamarck TW, Muldoon MF, Shiffman S, Sutton-Tyrrell K, Gwaltney C, Janicki DL. Experiences of demand and control in daily life as correlates of subclinical carotid atherosclerosis in a healthy older sample. Health Psychol. 2004;23:24–32.
21. LaMontagne AD, Keegel T, Vallance D, Ostry A, Wolfe R. Job strain—attributable depression in a sample of working Australians: assessing the contribution to health inequalities. BMC Public Health. 2008;8:181.
22. Nahrgang JD, Morgeson FP, Hofmann DA. Safety at work: a meta-analytic investigation of the link between job demands, job resources, burnout, engagement, and safety outcomes. J Appl Psychol. 2011;96:71–94.
23. Grzywacz JG, Quandt SA, Vallejos QM, et al. Job demands and pesticide exposure among immigrant Latino farmworkers. J Occup Health Psychol. 2010;15:252–266.
24. Georges A, Alterman T, Gabbard S, et al. Depression, social factors, and farmworker health care utilization. J Rural Health. 2013;29(suppl 1):s7–s16.
25. Hoerster KD, Mayer JA, Gabbard S, et al. Impact of individual-, environmental-, and policy-level factors on health care utilization among US farmworkers. Am J Public Health. 2011;101:685–692.
26. Wang S, Myers JR, Layne LA. Injuries to hired crop workers in the United States: a descriptive analysis of a national probability survey. Am J Ind Med. 2011;54:734–747.
27. Grzywacz JG, Alterman T, Muntaner C, et al. Mental health research with Latino farmworkers: a systematic evaluation of the short CES-D. J Immigr Minor Health. 2010;12:652–658.
28. Grzywacz JG, Hovey JD, Seligman LD, Arcury TA, Quandt SA. Evaluating short-form versions of the CES-D for measuring depressive symptoms among immigrants from Mexico. Hispanic J Behav Sci. 2006;28:404–424.
29. Grzywacz JG, Quandt SA, Chen H, et al. Depressive symptoms among Latino farmworkers across the agricultural season: structural and situational influences. Cultur Divers Ethnic Minor Psychol. 2010;16:335–343.
30. Fransson EI, Nyberg ST, Heikkila K, et al. Comparison of alternative versions of the job demand–control scales in 17 European cohort studies: the IPD-Work consortium. BMC Public Health. 2012;12:62.
31. Grzywacz JG, Alterman T, Muntaner C, Gabbard S, Nakamoto J, Carroll DJ. Measuring job characteristics and mental health among Latino farmworkers: results from cognitive testing. J Immigrant Minor Health. 2009;11:131–138.
32. Arcury TA, O'Hara H, Grzywacz JG, Isom S, Chen H, Quandt SA. Work safety climate, musculoskeletal discomfort, working while injured, and depression among migrant farmworkers in North Carolina. Am J Public Health. 2012;102(suppl 2):S272–S278.
33. Quandt SA, Feldman SR, Vallejos QM, et al. Vision problems, eye care history, and ocular protection among migrant farmworkers. Arch Environ Occup Health. 2008;63:13–16.
34. Quandt SA, Hiott AE, Grzywacz JG, Davis SW, Arcury TA. Oral health and quality of life of migrant and seasonal farmworkers in North Carolina. J Agric Saf Health. 2007;13:45–55.
35. Villarejo D, McCurdy SA, Bade B, Samuels S, Lighthall D, Williams D III. The health of California's immigrant hired farmworkers. Am J Ind Med. 2010;53:387–397.
36. Dey AN, Lucas JW. Physical and Mental Health Characteristics of US- and Foreign-Born Adults: United States, 1998–2003. Hyattsville, MD: National Center for Health Statistics; 2006.
37. Alderete E, Vega WA, Kolody B, Aguilar-Gaxiola S. Depressive symptomatology: prevalence and psychosocial risk factors among Mexican migrant farmworkers in California. J Community Psychol. 1999;27:457–471.
38. Schenker M. NAWS—what you count counts. J Agric Saf Health. 2005;11:291–292.
Copyright © 2014 by the American College of Occupational and Environmental Medicine