<< Back  

 Can't open the PDF? Click here for help.


Modifiable risk factors for depressed mood among farmers

Obiora E. Onwuameze, MB, BS, MS, PhD

Southern Illinois University School of Medicine, Department of Psychiatry, Springfield, IL, USA

Sergio Paradiso, MD, PhD

Una Mano per la Vita, Clinics and Association of Families and their Doctors, San Giovanni La Punta, Italy, Universidad Diego Portales, Santiago, Chile

Corinne Peek-Asa, PhD

University of Iowa College of Public Health, Department of Occupational and Environmental Health, Iowa City, IA, USA

Kelley J. Donham, DVM

University of Iowa College of Public Health, Department of Occupational and Environmental Health, Iowa City, IA, USA

Risto H. Rautiainen, PhD

University of Nebraska Medical Center, College of Public Health, Department of Environmental, Agricultural and Occupational Health, Omaha, NE, USA

BACKGROUND: Risk for depression among farmers is not fully understood. DSM-IV considers sadness or depressed mood a critical symptom of depression. The aim of this study was to examine risk factors for depressed mood among farmers using a longitudinal study design.

METHODS: Participants were principal farm operators in the Iowa Certified Safe Farm study. We identified risk factors for depressed mood by calculating relative risks (RR) using the generalized estimating equations method.

RESULTS: In the multivariate model, pesticide exposure (RR = 1.26; 95% CI: 1.04 to 1.53), having an additional job off the farm (RR = 1.32; 95% CI: 1.08 to 1.62), stress (RR = 3.09; 95% CI: 2.55 to 3.75), and previous injury (RR = 1.41; 95% CI: 1.05 to 1.89) prospectively increased the risk of depressed mood.

CONCLUSIONS: Consistent with earlier non-longitudinal studies, the results of this study suggest that reducing pesticide exposure, stress, and injury may reduce the risk of depression in the farm setting.

KEYWORDS: depression, pesticides, orthophosphates, farming, stress, serotonin



Farming is associated with increased risk of depression.1,2 The consequences of depression among farmers are far reaching. Although depression affects a farmer’s personal life and his or her family’s well-being,3 lost work days due to depression among farmers have a significant impact on the economy of urban and rural communities.4 In a study conducted between 1997 and 1999 that examined 29,400 individuals born between 1953 and 1957 in Hordaland County, Norway, agricultural workers showed the highest Hospital Anxiety and Depression Scale scores5 relative to all other occupational groups.6 In the United States, farming—together with unskilled labor and technical work—is among the occupations associated with greater depression.7 Estimated prevalence in the Midwest ranges from 7.4% to 35%, varying by study and state.8-10 Among male farmers, the prevalence of depression was 35% in Ohio,8 26.1% in Missouri,9 12.2% in Iowa,10 and 7.4% in Colorado.10 Rate variation may be due to approach—including assessment and definition of depression—recruitment criteria, and intrinsic characteristics of the farmers by state.

Elevated rates of depression among farmers predictably are associated with elevated suicide risk.11-18 Suicide among farmers and agricultural workers greatly contributes to higher suicide rates seen in rural vs urban populations.15 The suicide rate among Kentucky farmers is estimated at 42.2/100,000/year, compared with 19.2/100,000/year among the general US population.11 High suicide rates among farmers are not limited to the United States; in Scotland, suicide among farmers is estimated at 31.4/100,000/year compared with 27.5/100,000/year among the general Scottish population.17

Assessment of depression in farm studies has varied greatly. Widely used methods have ranged from participant-administered questionnaires including the Center for Epidemiologic Studies-Depression Scale,10,19-21 the Hospital Anxiety and Depression Scale,1 the general mental health subscale of the short form (36) health survey,7,22 to a single anamnestic question—eg, “Has a doctor ever diagnosed you with depression requiring medication?”23 or “Have you experienced feelings of sadness or depression within the last year?”24

There is considerable interest in the presence of sadness among farmers. Inability to resolve negative mood has been shown to increase vulnerability of developing clinical depression.24,25 Furthermore, sadness or low mood is a criterion, together with loss of pleasure, that is required for a depression diagnosis (and it will not change in DSM-5).26 Based on satisfactory sensitivity and specificity against depression severity measured with assessment scales (>65.0% and >87.0% relative to the Beck Depression Inventory27,28 and 86% and 78% relative to the Montgomery-Åsberg Depression Rating Scale29,30), assessment of sad mood is a widely used method to track changes in personal emotional status in longitudinal epidemiological studies.24,27,30,31

Previous cross-sectional studies have identified several psychosocial risk factors for depression among farmers including substantial income decline, being single, having poor general health, and loss of something of sentimental value.10 Farming exposes individuals to high stress and injuries,24,32 factors known to elevate depression risk.24 Biologic risk factors also have been identified, such as younger age, which also may be considered a psychosocial risk factor.33 Alcohol use has been linked to depression. Patten and Charney34 found that consuming 5 to 10 alcoholic drinks per week increased the risk of depression among farmers. Several cross-sectional surveys20,21,23,35,36 and 1 longitudinal study19 have reported the association between pesticide application and depression. Beseler et al19 reported that depression was associated with cumulative and chronic—but not acute—pesticide exposure.

Studies carried out to examine factors associated with depression among farmers have identified both nonmodifiable (eg, younger age), modifiable (eg, pesticide exposure), psychosocial (eg, income decline), and biologic factors (eg, alcohol use). Most studies have employed cross-sectional designs. The caveat with this design is that association among variables is synchronic, greatly limiting inference of causality. Understanding the temporal associations between risk factors and depression may suggest (albeit not prove) causation and help design appropriate interventions and preventive measures.

This study prospectively evaluated risk factors for depressed mood in a relatively large cohort of farmers. Based on prior literature, it was hypothesized that both psychosocial1,2,10,16,33 (including modifiable factors as occupational stress) and biologic factors19-21,23,34-36 (including modifiable factors as pesticide exposure) will be associated with depression.


This study examined data from the Iowa Certified Safe Farm (CSF) study, which is designed to evaluate the effectiveness of a multifaceted farm safety intervention. The methods for and a summary of this study have been described in detail.32,37,38


The Iowa CSF study targeted a 9-county area in northwestern Iowa. All principal farm operators who met the US Department of Agriculture farm criteria (>$1,000 in agricultural product sales per year) were eligible. Initially, participants were invited to participate through mailings and the media. Farmers who returned a card indicating their willingness to take part in the study were contacted by telephone until 300 farmers were recruited. They were pair matched based on farm size (crop acres), whether they raised livestock in general, raised pigs specifically, and previous injury experience. Pairs were randomly assigned to intervention or control groups. The intervention group received an on-farm safety review and educational intervention focusing on farm hazards including use of personal protective equipment (PPE). Other aspects of intervention included an annual health screening and monetary incentive. There was no significant association between the intervention group and depressed mood (RR [95% CI] = 0.94 [0.78 to 1.13], P > .5). Further recruiting was conducted to replace dropouts, and data from 257 farmers were available for analyses.

Data collection

Outcome and risk factor data were collected through quarterly phone calls, annual occupational history forms, annual on-farm safety review, and annual clinic screenings.37 Ten rounds of computer-aided, quarterly phone calls covered work exposures, injuries, and illnesses in a 3-year period. Calendars were provided for tracking health-related events including depression, injuries, skin conditions, hearing loss, and joint pain. All participants completed the annual occupational history forms and quarterly calls, and only farmers in the intervention group completed the clinic screenings and on-farm safety reviews because these procedures were major components of the intervention. The occupational history forms included demographic and farm production variables, as well as questions on safety behaviors and PPE use. The clinic screening form included questions about health outcomes.

Study variables

The outcome variable for this analysis was derived from the question asked during each quarterly interview: “How would you rate your level of depression in the last quarter?” Possible responses were: very low, low, average, high, or very high. Responses were combined based on clinical relevance: a dichotomous outcome variable “no depressed mood” and “depressed mood” was created. Responses from very low to average depression were believed to have unclear clinical relevance and represent normal mood fluctuations. Therefore very low, low, and average responses were parsimoniously classified as no depression. “High” and “very high” responses were classified as depression.

The demographic variables in this study included age, sex, and education. The age variable was assessed as a continuous variable, while education was dichotomized into ≤12 or >12 years of education. Occupational risk factors included raising livestock, exposure to pesticides, and exposure to agricultural chemicals, all of which were recorded using dichotomized (ie, “yes” or “no”) answers. Other dichotomized occupational variables included off-farm work and weekly farm hours (≤42 hours or >42 hours) as were included in previous studies.19 Lifestyle variables included drinking >9 alcoholic drinks per week and smoking (presence or absence). The General Health Status variable was recorded as 5 severity levels: very good, good, fair, poor, and very poor, and was dichotomized into poor—very poor and poor—vs good—very good, good, and fair. Presence or absence of a previous injury and stress also were predictor variables.


Data from quarterly calls, occupational history forms, farm reviews, and clinical screenings were merged to examine the associations between potential risk factors (exposures) and the presence of depression. The quarterly call dataset consisted of 10 repeated records of exposures and outcomes for each study patient. Variables from the other data collection instruments were merged into the quarterly dataset.

The generalized estimating equations (GEE) method was used to evaluate the associations between exposures and outcomes as well as to calculate relative risks (RR). This analytic procedure is able to handle repeated measure data where responses are correlated. GEE analyses were performed using generalized logit as the link factor, the number of months of observation as the offset factor, and the farm identification number (ie, the unique identification number for each participant) as the clustering factor.39 The temporal association between occurrence of depression and risk exposure was an essential feature of the present study. The following 2 methods were used to establish that the exposure occurred before the outcome, reported in the quarterly call data:

  1. Exposure from each quarter was matched with outcomes from the following quarter

  2. Most recent available annual occupational history, clinic screening, and farm review observations prior to the quarterly call outcome observation were added to the dataset.

This procedure allowed the independent variables (exposures) to be measured before the outcome measure (depressed mood) from the first to last quarter; depression in the first quarter was excluded. Model building included the following steps: 1) the association of depression with potential risk factors was first assessed in univariate analyses; 2) risk factors that had a P value of ≤.10 in the univariate analysis were selected for multivariate analyses; and 3) all risk factors that met the criteria were included in the multivariate model.


TABLE 1 describes the demographic characteristics of Iowa CSF farmers in comparison with Iowa farmers in general. On average, CSF cohort was slightly older than Iowa farmers in general and had more male farmers. Also, CSF farmers had larger crop acreages and larger hog operations compared with Iowa farmers.


Baseline characteristics of CSF farmers and the general Iowa farming population

Variables Iowa CSF farmers
(N = 257)
Iowa farmers in generala
(N = 61,935)
Mean age (years) 56 52
Sex (% male) 98% 92.7%
Ethnicity (% white) 100% 99.9%
Mean total acres farmed 628 350
Mean hog herdb 2,798 850
Mean corn acreage 318 187
Mean soybean acreage 302 176
aThe US Census of Agriculture 2002.
bPrincipal operators were exclusively hog-raising farmers.
CSF: Certified Safe Farm.

The average percentage of patients reporting “high” or “very high” depressed mood during the 10 examined quarters was 24.1% (TABLE 2). Nine of the 62 patients reported depressed mood in 2 different quarters and the remaining participants reported depression during 1 interview. This signifies that 96.5% of depressed farmers had depressed mood for an estimated duration of ≤4 months. There were no statistically significant differences between farmers with longer and shorter duration of depressed mood for demographic variables as well as pesticide exposure, previous injury, hearing loss, alcohol use, working longer hours (>42 hours per week), livestock farming, and poor general health.


Rates of depressed mood by severity

  High or very high depressed mood Very low, low, or average depressed mood
Frequency 62/257 195/257
Percentage (%) 24.1% 75.9%

TABLE 3 presents RR, 95% CI, and P values for the univariate regression of potential risk factors and depression. Farmers with pesticide exposure had 1.27 times risk of depression than nonexposed farmers (95% CI = 1.06 to 1.53). Three other variables significantly increased the risk for depressed mood: previous injury increased the risk by 1.53 (95% CI = 1.15 to 2.04); stress increased the risk by 3-fold (95% CI = 2.55 to 3.72); and having an additional job off the farm increased the risk by 1.22 (95% CI = 1.01 to 1.48). Hearing loss also met criteria to be included in the final model. Sex, age, education, smoking, hearing loss, alcohol use, working longer hours weekly (>42 hours per week), livestock farming, and poor general health did not predict depressed mood. To determine the extent to which mood changed with the season, effect of quarter was examined, but did not show association with depressed mood (TABLE 3). Marital status was expected to play a role in depression; however, the data did not allow this computation, because the percentage of single farm operators among the entire data set was <5%.


Univariate analysis for depression risk factors

Variables Present
(N = 257)
RR (95% CI) P
Education (≤12 years) 82 0.88 (0.72 to 1.07) .20
Sex (male) 251 0.87 (0.55 to 1.38) .55
Average weekly farm hours (>42) 108 1.09 (0.81 to 1.32) .29
Off-farm job 95 1.22 (1.01 to 1.48) .04
Pesticide exposure 94 1.27 (1.06 to 1.53) <.01
Agricultural chemical exposure 90 0.95 (0.78 to 1.17) .63
General health (poor) 5 1.29 (0.64 to 2.56) .48
Previous injury 28 1.53 (1.15 to 2.04) <.01
Stress 59 3.08 (2.55 to 3.72) <.01
Hearing loss 78 1.30 (0.96 to 1.77) .10
Alcohol (>9 drinks per week) 126 0.94 (0.79 to 1.13) .51
Smoking (current) 10 1.29 (0.84 to 1.98) .24
Livestock 158 0.97 (0.81 to 1.16) .76
Age (mean 56 years) 257 0.83 (0.68 to 1.02) .08
RR: relative risk.

TABLE 4 describes the RR, 95% CI, and P values for the multivariate regression analysis. The risk for depressed mood after pesticide exposure remained essentially unchanged (RR = 1.26; 95% CI: 1.04 to 1.53). Having an off-farm job, stress, and previous injury remained significant predictors for depressed mood.


Multivariate analysis of potential risk factors for depressed mood

Variables N RR (95% CI) P
Pesticide exposure 257 1.26 (1.04 to 1.53) .01
Off-farm job 257 1.32 (1.08 to 1.62) <.01
Stress 257 3.09 (2.55 to 3.75) <.01
Previous injury 257 1.41 (1.05 to 1.89) .02
RR: relative risk.


Negative or depressed mood among vulnerable individuals may initiate a cascade of psychological events leading to clinical depression.25 Our main focus was to identify modifiable and nonmodifiable, psychosocial, and biologic factors antecedent to depressed mood using data from the Iowa CSF study32,37,38 as a means to guide further secondary prevention studies. Hypotheses were based on several cross-sectional reports1,2,10,16,20,21,23,32-36 and a prospective study.19

We found several interesting results. Within the 3-year study period, approximately one-fourth of farmers admitted to depressed mood of potential clinical relevance. Risk factors for depressed mood were psychosocial—eg, stress, injury on the farm, holding an additional job off the farm—and biological—eg, exposure to pesticides. Most participants had an estimated duration of depressed mood of ≤4 months. Although some factors, including exposure to pesticides and injury, may be modifiable through improved injury prevention and farming practices, others, including need for a job off the farm to increment income, may be harder to modify in the current financial climate. Factors in cross-sectional studies previously reported to be associated with depression—eg, poor general health, younger age, and alcohol use—were not found to significantly predict depressed mood in the present study. Other factors that have shown association in previous studies—eg, income decline, being single, and loss of “something of sentimental value”—were not assessed.

Before discussing these findings, some specific issues need to be acknowledged. The present study outcome variable was based on a single question and on the conservative opinion that responses of “very low,” “low,” and “average” level of depression may represent normal mood variations while “high” and “very high” depressed mood may constitute relevant change. Although others have used a similar approach (eg, “Have you experienced feelings of sadness or depression within the last year?”),24 assessment type should be kept in mind when comparing the present study findings with studies of farmers using other methods to assess depression.1,19 Related to this issue is the transient nature of depression that might not meet clinical diagnostic threshold. Further studies should attempt to attain data on mental health service use during the longitudinal period to clarify the clinical significance of reported depressed mood. Remarkably, the 24.1% rate of depressed mood found in this study is comparable to the 24.0% reported in a previous study using a similar approach.24 As expected, this method yielded depression rates higher than a previous Iowa farmers study that assessed depression using the Center for Epidemiologic Studies Depression Scale (CES-D).10 However, 2 studies conducted in Ohio8 and Missouri9 using the CES-D yielded higher rates relative to Iowa, suggesting that rate variation is not exclusively based on depression assessment. Variation among depression rates may be related to the sample characteristics by study and state. Responses on mood changes were attained from phone interviews. In-person interviews may have resulted in different findings.

In addition, personal or family history of depression was not recorded to evaluate potential vulnerability to pesticide-induced depression. Farmers participating in the study operated larger farms than Iowa farmers in general, therefore results should be generalized with this caveat. Stress was found to be a predictor of depressed mood in the present study. Responses to the survey (eg, the need for working off the farm) may help to understand the types of stress farmers endure; future studies should more precisely characterize the types of stress affecting farmers (including variables such as farm income and total household income). Pesticide exposure in the present study was self-reported. This assessment is similar to previous studies20,21,23,35,36 but cannot objectively assess the severity and suddenness of exposure. There is a debate in the literature concerning the extent to which low-dose chronic or high-dose rapid (eg, poisoning) exposure can lead to depression. Future research should aim to improve the assessment method, because self-report has proven unable to distinguish the effects between cumulative high19 or low20,21,23,35,36 pesticide exposure on depression.

The prevalent duration of self-reported sad mood in the present study (≤4 months) was consistent with the lower end of the spectrum of duration of episodes of depressive illness (lasting from 2 months to several years, with an average of approximately 5 to 6 months)40 and with duration of depression among the general population (ie, 3 months).41 Interestingly, there were no differences reaching statistical significance between farmers with longer and shorter duration of depressed mood on demographic variables, as well as on pesticide exposure, previous injury, hearing loss, alcohol use, working longer hours (>42 hours per week), livestock farming, and poor general health.

Consistent with the findings in the present study, pesticide and chemical exposure have been reported as risk factors for depression in several cross-sectional surveys.20,21,23,35,36 Our study confirms the temporal antecedence of exposure to pesticide with respect to depression as reported in a study of Colorado farmers.19 Animal research suggests possible mechanisms explaining the relationship between depressed mood and pesticide exposure. One of the most used pesticides, chlorpyrifos organophosphate,42 has been shown to disrupt serotonin neurotransmission and lead to neurobehavioral changes.43-46 For example, exposure of newborn rats (postnatal days 1 to 4) to organophosphate pesticides caused rats grown into adulthood to spend more time in the open arms of the elevated plus maze and to show lower preference for chocolate milk relative to water, similar to behavioral alterations observed in animal models of depression.43 Other biologic mechanisms of depression among farmers may be stress mediated. A biologic link between stress and depression also has been posited. Stress-related decreased expression of brain-derived neurotrophic factor has been shown to contribute to limbic system atrophy including the hippocampus.47 In rodents, stress-induced reduction of hippocampal neurogenesis has been suggested as a potential mechanism of depression.48,49 Notably, the relationship between pesticide exposure and depression may be mediated by paraoxonase gene polymorphism.50 Human paraoxonase (especially paraoxonase-1) are serum enzymes protecting against exposure to organophosphorus pesticides by hydrolyzing their toxic oxon metabolites.51 The R allele has been shown the increase the risk of depression among patients with long-term exposure to organophosphate pesticides.52,53

It is known that stress is synchronically associated with depression,1,10,16,54-59 and this study is consistent with previous reports showing that farming is associated with increased stress.1,2 The present study adds the notion of temporal antecedence of stress relative to depressed mood to the existing literature of depression among farmers. Stress is known to induce interpersonal vulnerability, impaired self-worth, and reduced social connectedness.55 Farming may lead to depression because of farmers’ limited social network, which may be relevant to coping with stress.59 Several factors contributing to psychosocial stress—including loss of property with sentimental or vocational value, income decline, or financial strain (which may lead to the need to find an additional job besides farming)—have been considered as risk factors.10,59 In addition, farm work can be physically strenuous and most of it takes place outdoors, often during inclement weather, especially during planting and harvesting seasons.60 The emotional demands attached to meeting deadlines for planting and harvesting may contribute to high stress levels.10 Stress alters the perception of environmental and emotional demands potentially exceeding the farmers’ adaptive capacity58 and generating feelings that may be akin to hopelessness and helplessness.

Income decline or variability of farm income have been identified as primary motivations for working an additional job off the farm.61 The study’s finding is consistent with previous cross-sectional surveys that identified financial difficulties as a risk factor for depression among farmers.10,59 It should be noted that opportunity to access health benefits may be another reason for having an additional job off the farm, but consider that this also may be tied indirectly to poor finances. Our study showed that farm injury increased the risk of depressed mood, a finding consistent with the current literature.24 Further studies will need to examine the nature and severity of injury in the farm environment.

In the study, several factors previously shown to be independently associated with depressive illness among the general population including poor general health,20,33 alcohol use, and smoking34 did not show significant association with depressive mood. Not associating depression with general health may be an artifact of most participants’ good health (98%). Failure to show association with known risk factors for depression in the general population (eg, alcohol) suggests that depressive mood among farmers has an etiology that differs partially from that of depression among the general population.


This study found that pesticide exposure, stress, and injury were independent risk factors for depressed mood among farmers. Prior studies suggest that suicide is a significant concern among farmers11,17 and depression has far reaching economic consequences.4 Together with research showing the importance of negative mood in the mechanisms of clinical depression,25 the present study suggests potential modifiable factors including pesticide exposure, stress, and injury on the farm that can be targeted in secondary prevention studies. Conceivably, better farming practices aimed at reducing pesticide exposure, stress, and injury may have considerable effects on reducing the risk of depression among farmers.

DISCLOSURES: Drs. Onwuameze and Peek-Asa report no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products. Dr. Paradiso has received grant or research support from the Dana Foundation, the Mallinckrodt Foundation, NARSAD, and the National Institute on Aging (grant number 5K23AG027837). Dr. Donham receives grant or research support from the Centers for Disease Control and Prevention (CDC). Dr. Rautiainen receives grant or research support from the CDC/National Institute for Occupational Safety and Health and the state of Nebraska and is a consultant to MTT Agrifood Research Finland.

ACKNOWLEDGMENTS: This research was supported by the Iowa Injury Prevention Research Center, the Iowa Pork Producers Council, Iowa Wellmark Foundation, the National Institute for Occupational Safety and Health (NIOSH) (grant number UO6/CCU712193), the NIOSH-funded Heartland Education and Research Center (grant number T420H008491), the National Pork Board, and Pioneer Hi-Bred International Inc.

We gratefully acknowledge the AgriSafe Network Clinic (Spencer, IA, USA), and the University of Iowa investigators, namely Kelley Donham (Principal Investigator of the CSF study), Kendall Thu, Natalie Roy, Risto Rautiainen, Carol Hodne, LaMar Grafft, and Paul Whitten, who initiated the program, obtained the funding, provided coordination of research efforts, and trained health care providers and farm safety consultants in this study.

Finally, many thanks to Liz Smothers for her help with proofreading and typesetting of the manuscript.


  1. Sanne B, Mykletun A, Moen BE, et al. Farmers are at risk for anxiety and depression: the Hordaland Health Study. Occup Med (Lond). 2004;54:92–100.
  2. Simkin S, Hawton K, Fagg J, et al. Stress in farmers: a survey of farmers in England and Wales. Occup Environ Med. 1998;55:729–734.
  3. Blair A, Sandler DP, Tarone R, et al. Mortality among participants in the agricultural health study. Ann Epidemiol. 2005;15:279–285.
  4. Marcus M, Yasamy MT, van Ommeren M, et al; World Health Organization. Depression. A global public health concern. http://www.who.int/mental_health/management/depression/who_paper_depression_wfmh_2012.pdf. Accessed January 11 2013.
  5. Zigmond AS, Snaith RP. The hospital anxiety and depression scale. Acta Psychiatr Scand. 1983;67:361–370.
  6. Sanne B, Mykletun A, Dahl AA, et al. Occupational differences in levels of anxiety and depression: the Hordaland Health Study. J Occup Environ Med. 2003;45:628–638.
  7. Grosch JW, Murphy LR. Occupational differences in depression and global health: results from a national sample of US workers. J Occup Environ Med. 1998;40:153–164.
  8. Belyea MJ, Lobao LM. Psychosocial consequences of agricultural transformation: the farm crisis and depression. Rural Sociol. 1990;55:58–75.
  9. O’Brien DJ, Hassinger EW, Dershem L. Community attachment and depressive illness among residents of two rural Midwestern communities. Rural Sociol. 1994;59:255–265.
  10. Scarth RD, Stallones L, Zwerling C, et al. The prevalence of depressive symptoms and risk factors among Iowa and Colorado farmers. Am J Ind Med. 2000;37:382–389.
  11. Boxer PA, Burnett C, Swanson N. Suicide and occupation: a review of the literature. J Occup Environ Med. 1995;37:442–452.
  12. Malmberg A, Simkin S, Hawton K. Suicide in farmers. Br J Psychiatry. 1999;175:103–105.
  13. Page AN, Fragar LJ. Suicide in Australian farming 1988-1997. Aust N Z J Psychiatry. 2002;36:81–85.
  14. Roberts RE, Lee ES. Occupation and the prevalence of major depression alcohol, and drug abuse in the United States. Environ Res. 1993;61:266–278.
  15. Singh GK, Siahpush M. Increasing rural-urban gradients in US suicide mortality 1970-1997. Am J Public Health. 2002;92:1161–1167.
  16. Stallones L, Marx MB, Garrity TF. Prevalence and correlates of depressive symptoms among older U.S. adults. Am J Prev Med. 1990;6:295–303.
  17. Stark C, Gibbs D, Hopkins P, et al. Suicide in farmers in Scotland. Rural Remote Health. 2006;6:509.
  18. Zwerling C, Burmeister LF, Jensen CM. Injury mortality among Iowa farmers 1980-1988: comparison of PMR and SMR approaches. Am J Epidemiol. 1995;141:878–882.
  19. Beseler CL, Stallones L. A cohort study of pesticide poisoning and depression in Colorado farm residents. Ann Epidemiol. 2008;18:768–774.
  20. Stallones L, Beseler C. Pesticide poisoning and depressive symptoms among farm residents. Ann Epidemiol. 2002;12:389–394.
  21. Stallones L, Beseler C. Safety practices and depression among farm residents. Ann Epidemiol. 2004;14:571–578.
  22. Stewart AL, Hays RD, Ware JE Jr. The MOS short-form general health survey. Reliability and validity in a patient population. Med Care. 1988;26:724–735.
  23. Beseler C, Stallones L, Hoppin JA, et al. Depression and pesticide exposures in female spouses of licensed pesticide applicators in the agricultural health study cohort. J Occup Environ Med. 2006;48:1005–1013.
  24. Carruth AK, Logan CA. Depressive symptoms in farm women: effects of health status and farming lifestyle characteristics behaviors, and beliefs. J Community Health. 2002;27:213–228.
  25. Jeanne M, Gross JJ, Persons JB, et al. Mood matters: negative mood induction activates dysfunctional attitudes in women vulnerable to depression. Cognit Ther Res. 1998;22:363–376.
  26.  Diagnostic and statistical manual of mental disorders, 4th ed, text rev. Washington DC: American Psychiatric Association; 2000.
  27. Avasarala JR, Cross AH, Trinkaus K. Comparative assessment of Yale Single Question and Beck Depression Inventory Scale in screening for depression in multiple sclerosis. Mult Scler. 2003;9:307–310.
  28. Beck AT, Ward CH, Mendelson M, et al. An inventory for measuring depression. Arch Gen Psychiatry. 1961;4:561–571.
  29. Montgomery SA, Asberg M. A new depression scale designed to be sensitive to change. Br J Psychiatry. 1979;134:382–389.
  30. Watkins C, Daniels L, Jack C, et al. Accuracy of a single question in screening for depression in a cohort of patients after stroke: comparative study. BMJ. 2001;323:1159.
  31. Osborn DP, Fletcher AE, Smeeth L, et al. Performance of a single screening question for depression in a representative sample of 13 670 people aged 75 and over in the UK: results from the MRC trial of assessment and management of older people in the community. Fam Pract. 2003;20:682–684.
  32. Rautiainen RH, Lange JL, Hodne CJ, et al. Injuries in the Iowa Certified Safe Farm Study. J Agric Saf Health. 2004;10:51–63.
  33. Stallones L, Leff M, Garrett C, et al. Depressive symptoms among Colorado farmers. J Agric Saf Health. 1995;1:37–43.
  34. Patten SB, Charney DA. Alcohol consumption and major depression in the Canadian population. Can J Psychiatry. 1998;43:502–506.
  35. Reidy TJ, Bowler RM, Rauch SS, et al. Pesticide exposure and neuropsychological impairment in migrant farm workers. Arch Clin Neuropsychol. 1992;7:85–95.
  36. Savage EP, Keefe TJ, Mounce LM, et al. Chronic neurological sequelae of acute organophosphate pesticide poisoning. Arch Environ Health. 1988;43:38–45.
  37. Choi SW, Peek-Asa C, Sprince NL, et al. Hearing loss as a risk factor for agricultural injuries. Am J Ind Med. 2005;48:293–301.
  38. Donham KJ, Rautiainen RH, Lange JL, et al. Injury and illness costs in the Certified Safe Farm study. J Rural Health. 2007;23:348–355.
  39.  SAS. Cary, NC: SAS Institute; 2004.
  40. Lehmann HE. Clinical evaluation and natural course of depression. J Clin Psychiatry. 1983;44:5–10.
  41. Spijker J, de Graaf R, Bijl RV, et al. Duration of major depressive episodes in the general population: results from The Netherlands Mental Health Survey and Incidence Study (NEMESIS). Br J Psychiatry. 2002;181:208–213.
  42.  United States Environmental Protection Agency. Reregistration eligibility decision for chlorpyrifos. http://www.epa.gov/oppsrrd1/REDs/chlorpyrifos_red.pdf. Published July 31 2006. Accessed January 11, 2013.
  43. Aldridge JE, Levin ED, Seidler FJ, et al. Developmental exposure of rats to chlorpyrifos leads to behavioral alterations in adulthood, involving serotonergic mechanisms and resembling animal models of depression. Environ Health Perspect. 2005;113:527–531.
  44. Aldridge JE, Seidler FJ, Slotkin TA. Developmental exposure to chlorpyrifos elicits sex-selective alterations of serotonergic synaptic function in adulthood: critical periods and regional selectivity for effects on the serotonin transporter receptor subtypes, and cell signaling. Environ Health Perspect. 2004;112:148–155.
  45. Slotkin TA, Tate CA, Cousins MM, et al. Functional alterations in CNS catecholamine systems in adolescence and adulthood after neonatal chlorpyrifos exposure. Brain Res Dev Brain Res. 2002;133:163–173.
  46. Slotkin TA, Tate CA, Ryde IT, et al. Organophosphate insecticides target the serotonergic system in developing rat brain regions: disparate effects of diazinon and parathion at doses spanning the threshold for cholinesterase inhibition. Environ Health Perspect. 2006;114:1542–1546.
  47. Duman RS, Monteggia LM. A neurotrophic model for stress-related mood disorders. Biol Psychiatry. 2006;59:1116–1127.
  48. Dranovsky A, Hen R. Hippocampal neurogenesis: regulation by stress and antidepressants. Biol Psychiatry. 2006;59:1136–1143.
  49. Duman RS. Depression: a case of neuronal life and death? Biol Psychiatry. 2004;56:140–145.
  50. Lawlor DA, Timpson N, Harbord R, et al. Polymorphisms in the CRP gene and cardiovascular events. JAMA. 2007;297:1317; author reply 1317–1318.
  51. La Du BN. Human serum paraoxonase/arylesterase. In: Kalow W ed. Pharmacogenetics of drug metabolism. New York, NY: Pergamon Press; 1992:51-91.
  52. Browne RO, Moyal-Segal LB, Zumsteg D, et al. Coding region paraoxonase polymorphisms dictate accentuated neuronal reactions in chronic, sub- threshold pesticide exposure. FASEB J. 2006;20:1733–1735.
  53. Cherry N, Mackness M, Durrington P, et al. Paraoxonase (PON1) polymorphisms in farmers attributing ill health to sheep dip. Lancet. 2002;359:763–764.
  54. Cohen S, Janicki-Deverts D, Miller GE. Psychological stress and disease. JAMA. 2007;298:1685–1687.
  55. Hammen C. Stress and depression. Annu Rev Clin Psychol. 2005;1:293–319.
  56. Linn JG, Husaini BA. Determinants of psychological depression and coping behaviors of Tennessee farm residents. J Community Psychol. 1987;15:503–512.
  57. Mazure CM. Life stressors as risk factors in depression. Clin Psychol (New York). 1998;5:291–313.
  58. Mitchell PB, Parker GB, Gladstone GL, et al. Severity of stressful life events in first and subsequent episodes of depression: the relevance of depressive subtype. J Affect Disord. 2003;73:245–252.
  59. Thelin A. Psychosocial factors in farming. Ann Agric Environ Med. 1995;2:21–26.
  60.  United States Department of Labor. Bureau of Labor Statistics. Farming fishing, forestry, and transportation occupations. http://www.umsl.edu/services/govdocs/ooh20042005/www.bls.gov/OCO/reprints/ocor014.pdf. Accessed January 11, 2013.
  61. Mishra AK, Goodwin BK. Farm income variability and the supply of off-farm labor. Am J Agric Econ. 1997;79:880–887.

CORRESPONDENCE: Sergio Paradiso, MD, PhD Una Mano per la Vita Clinics and Association of Families and their Doctors via Cristoforo Colombo n. 13/ E 95030 San Giovanni La Punta CT Italy E-MAIL: paradiso.sp@gmail.com