Article Text
PDF
Abstract
Objective Describe the occupational characteristics of farmer and non-farmer workers and investigate critical occupational risk factors for mental disorders in sugarcane farmers in Peru.
Method We conducted a cross-sectional study with occupational health and safety focus among farmers and non-farmers. Mental disorder symptoms were evaluated through the local validated version of the 12-Item General Health Questionnaire (GHQ-12). We explored the association between mental disorder symptoms, work conditions and known occupational risk factors (weekly working hours, pesticide exposures, heat stress and heavy workload). Negative binomial regression models were fitted, and 95% CIs were calculated.
Results We assessed 281 workers between December 2019 and February 2020. One hundred and six (37.7%) respondents identified themselves as farmworkers. The mean GHQ-12 scores for farmers and non-farmers were 3.1 and 1.3, respectively. In the fully adjusted multivariable model, mental disorder symptom counts among farmers were more than twice as high as those of non-farmers (β: 2.11; 95% CI: 1.48 to 3.01). The heavy workload increased the mean number of mental disorder symptoms by 68% (95% CI: 21% to 133%), and each additional working hour per day increased the mean number of mental disorder symptoms by 13% (95% CI: 1% to 25%).
Conclusion Farmers have higher mental disorder symptoms than non-farmers. A heavy workload and more working hours per day are independently associated with more mental disorder symptoms. Our findings highlight the importance of including mental health within occupational programmes and early interventions tailored to sugarcane industrial mill workers in the Latin American context.
- epidemiology
- public health
- mental health
- occupational & industrial medicine
Data availability statement
No data are available.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
Strengths and limitations of this study
We discussed a critical but unresolved issue with one of the main task forces in Peru and other low-income and middle-income countries.
We used a locally validated version of the 12-Item General Health Questionnaire as a screening instrument for mental disorders.
Our sample size was relatively small for detecting more occupational risk factors, but the statistical power was enough to support the main conclusions.
Introduction
Every year, more than 450 million people develop a mental disorder globally. Mental disorders represent a critical proportion of the global disease burden and disability-adjusted life years.1 About 75% of people affected by mental disorders live in low-income and middle-income countries (LMICs), and most have no access to appropriate treatment.2 Per a recent global review that included evidence from 27 countries, farmers have higher rates of suicide, depression and anxiety than the general population.3 In many LMICs, agriculture and farming remain the principal source of income4; however, farmers’ mental health usually receive poor attention from employers and limited care from health systems.3
Understanding the effects of occupational risk factors on farmers’ mental health at an epidemiological level is essential to determine prevention strategies that may help to avoid long-term mental health issues. For example, farmers are disproportionately exposed to work-related health risk factors4 such as lower salaries,5 pesticides,6 heat stress7 and heavier workloads.5 These factors can contribute to a higher risk of developing physical and mental diseases. Farmers can also be more likely to develop common mental disorders than non-farmers working in the same industry.8 However, to our knowledge, the problem of mental disorders due to agricultural work conditions have been barely studied in LMICs and especially in a Latin American context.3
The available evidence on this topic, especially for this population group is lacking. Our study compared the prevalence of mental health disorders among sugar cane farmers and non-farm workers and explored its relationship with sociodemographic and work characteristics. There is an urgent need to have an evidence-based understanding of mental health risk factors for high occupational exposure groups in farming communities to improve the prevention efforts. We aim to describe the occupational characteristics of farmers and non-farmers, determine differences in mental health status screening between these groups, and identify occupational risk factors associated with mental disorders. We hypothesised that farmers are more at risk of developing mental disorders than non-farmers in this population.
Methods
Study design
We analysed the baseline data of a prospective cohort of Peruvian farmers and non-farmers from the cane industry. That study, ‘Evaluating the effects of exposure to sugarcane industry work on kidney function in farmers’,9 compared the time trends of kidney damage biomarkers with three assessments over 12 months in both occupational groups.
Setting
This study was developed in Centro Poblado San Jacinto, a small village in the north of Peru, economically dependent on the local sugarcane industry. San Jacinto has a population of 12 000 inhabitants, of which approximately 70% have worked or are currently working in agriculture-related activities. The sugar industry has more than 9000 cultivated acres between 21 m and 429 m above sea level. Although the sugar industry provides primary occupational healthcare by law,10 most of the workers' healthcare in San Jacinto is provided through EsSalud and MINSA health centres. However, in both centres, mental healthcare is minimal or practically non-existent in rural places such as San Jacinto.
Participants
We detail the sample size calculations and sampling procedures for the main study in online supplemental file 1. We included 281 out of 291, 175 farmers and 106 non-farmers, and this allowed us to achieve 100% power to detect a difference of 1.2 between farmers and non-farmers with a significance level of 0.05 online supplemental file 2.
Supplemental material
According to the main study’s selection criteria, male participants between the ages of 18 and 60 and habitual residents in the study area (last 12 months) were eligible participants. Participants with a diagnosis of high blood pressure, diabetes mellitus and chronic kidney disease were excluded from this analysis, as they are considered to have known causes of chronic kidney disease. Also, we excluded participants working on more than one job as the effect of specific occupational exposures could not be estimated.
Farmer workers are subcontracted by the sugar company; their wages depend on the amount of sugarcane they cut or plant and will usually work long hours. Non-farmer workers are contracted directly by the sugar company and do not have the same heavy workload as field workers. They perform management activities, logistic processes, product quality assessment, and supervise production team operations.
Variables
Main outcome
Mental disorder symptoms were measured using a locally validated version of the 12-Item General Health Questionnaire (GHQ-12).11 This tool assesses the worker’s mental health status by asking 12 questions about how they have felt during the past week on various symptoms. The symptoms include problems with sleep and appetite, subjective experiences of stress, tension or sadness, mastery of daily problems, taking decisions and self-esteem. For each symptom, the person can respond less than usual, no more than usual, more than usual and much more than usual. We assigned a score equal to 0 for the first two options and a score equal to 1 for the latter two. Thus, GHQ-12 ranged from 0 to 12 symptoms; a score ≥5 would mean that the worker is at risk of having depression.12
Occupational groups
The work activity (ie, farmer and non-farmer) was the studied exposure. The farmer roles included cane cutters, seeders and seed cutters (exposed group). The non-farmer roles were defined as performing a factory or administrative activity (non-exposed group).
Covariates and occupational risk factors
Sociodemographic variables collected included age (years), level of education (<7 years of education, >7 years of education), monthly salary (low <US$480, high ≥US$480) and civil status (without union, with union). Occupational risk factors: the occupational heat stress index (formula: wet-bulb balloon temperature (WBGT)=0.7 wet bulb temperature + 0.2 globe temperature + 0.1 dry bulb temperature),13 hours of work per day,14 type of contract (fixed-term contract, indefinite contract), time of work in the industry (years), rest time during the working day (minutes), working hours per week, heavy workload (no, yes),5 use of shade during work break (no, yes) and exposure to pesticides (no, yes).6 Lifestyle covariates: tobacco consumption (at least one cigarette per day), alcohol consumption (self-reported consumption of ≥6 beers or its equivalent in alcohol with other beverages on the same occasion at least once a month), body mass index (normal: BMI >18.5 kg/m2 and <25 kg/m2, overweight/obesity: BMI ≥25 kg/m2) and self-rated health (poor, good).
Data collection
Questionnaires
After a prescreening and informed consent process, the participants were invited to participate in the study voluntarily. Once a written consent of participation was signed, the research staff surveyed them through an online questionnaire on tablets. The research team was trained on questionnaire application by the principal investigator, and research bioethics and responsible conduct in research by QUIPU—Centro Andino de Investigación y Entrenamiento en Informática para la Salud Global.15 The questionnaire sections included: demographics, employment, work history,16 and mental disorders.
Ambient measurements
Between 3 February and 21 February 2021, we recorded the air temperature and relative humidity every 15 min between 08:00 and 14:00 across the sugarcane fields at 1.25 m above the ground, using a WBGT and two 800036 WBGT laptops (Sper Scientific, China) independently to ensure data quality. We reported the mean results of the two devices. We calculated the heat index following the US Occupational Safety and Health Administration assessments and indications.13
Statistical analyses
The baseline characteristics of the study population were tabulated overall and according to work activity. To describe the data, we used percentages for categorical variables and median and interquartile ranges for continuous variables.
Mental disorder symptoms were treated as a count variable (0–12 symptoms) and summarised by showing the mean and SD for farmers and non-farmers. We fitted a negative binomial regression to the model count of symptoms as an outcome, setting work activity as the unique predictor. This allowed to formally compare the expected number of symptoms (mean) in non-farmers over the expected number of symptoms in farmers. In other words, we estimated a ratio of means (RM) between both groups.17 As with other ratio measures, RM >1 implies more risk of suffering depressive symptoms, RM <1 less risk and RM=1 equal risk. We preferred negative binomial regression instead of Poisson regression because the first can be used for overdispersed count data (online supplemental file 3), as in this case.18 We also fitted two adjusted models. Model 1 included the most critical work-related factors identified in the literature: monthly salary, exposure to pesticides and working hours per week. In Model 2, we adjusted for the same factors plus the type of contract, time of work in the industry, occupational heat stress index and heavy workload. Both models were also adjusted for age and work activity, the latter because it could still include other inherent risk factors we did not measure (occupational and non-occupational).
We adopted an exploratory approach for the last objective, analysing the full sample (independently of the work activity). Similar negative binomial regression models were fitted with sociodemographics, lifestyle and occupational risk factors as predictors and mental disorders symptoms as the outcome (ie, one unadjusted model per factor). Then, we joined those factors with a significant unadjusted association with mental disorders symptoms in one multivariable model. The factor selection and last estimated association allowed us to detect the main factors.
We calculated 95% CIs and considered p values <0.05 as significant. The statistical analysis was performed with Stata V.16.1 for Windows (Stata Corporation, College Station, Texas).
Patient and public involvement
No patients were involved.
Results
Characteristics of farmer and non-farmer participants
We surveyed 281 male workers between December 2019 and February 2020. A total of 106 (37.7%) respondents were identified as farmers, while 172 (62.3%) were non-farmers. The farmers group was slightly older (mean: 42 years) compared with non-farmers (mean: 40 years). Farmers had a lower monthly salary and had achieved fewer education levels than non-farmers.
Regarding occupational risk factors, the group of non-farmers had, on average, 11 years working in the sugarcane industry. One out of every four farmers had a fixed-term contract/service lease, compared with non-farmers who had permanent contracts/direct employment with the company. The farmers were exposed to a higher index of occupational heat stress (28.3°C, IQR±0.6), they worked 8.5 hours per day (IQR±1.5), they rested 12.9 fewer minutes in a workday, they worked +55 hours (IQR±8.0) during the week and had a heavier workload, compared with non-farmers.
Regarding lifestyle, the farmer’s group had a lower prevalence of tobacco consumption, alcohol consumption and overweight/obesity than non-farmers. The mean GHQ-12 score for farmers was 3.1 and 1.3 for non-farmers (table 1).
Characteristics of the study participants (N=281)
Differences in mental disorder symptoms between farmers and non-farmers
Farmers got 2.3 (95% CI: 1.71 to 3.09) times the mean number of mental disorder symptoms than non-farmers. After adjusting for the variables described in the first model (RM: 2.27; 95% CI: 1.69 to 3.06) and the second model (RM: 2.11; 95% CI: 1.48 to 3.01), the mean number of mental disorders symptom for farmers compared with non-farmers were still more than double (table 2).
Mental disorder symptoms (GHQ-12) among farmers and non-farmers (N=281)
Occupational risk factors and mental disorder symptoms
We detected three factors associated with symptoms of mental disorders. Having a heavy workload increased 68% of the mean number of mental disorders symptoms (95% CI: 21% to 133%). On average, each extra working hour per day increased the same outcome by 13% (95% CI: 1% to 25%). We detected a marginally-protective effect of having a shaded work break against symptoms of mental disorders (27%, 95% CI: −47% to 0%) (table 3).
Sociodemographic, occupational and lifestyle risk factors associated with mental disorders symptoms (GHQ-12) (N=281)
Discussion
We assessed mental disorder symptoms and potential risk factors on farmers and non-farmers from the industrial sugarcane mill in a rural Peruvian context. We found that the farmers had more mental disorders symptoms compared with non-farmers and that for any worker in this study, having a heavy workload and working more hours per day was associated with a higher risk of having mental disorder symptoms. There was a lack of association between pesticides exposure and a higher scoring in the heat stress index with mental disorders symptoms, opposed to reported evidence of these factors in other studies.6 7
Our study focused on active sugarcane industry workers and compared the occupational characteristics among the farmers' and non-farmers' groups. Our farmers' sample was younger than the mean age of participants reported in other studies.19 According to Wang et al, younger farmers experienced higher stress-related symptoms, while elderly farmers experienced more mental disabilities.20 We also found that many farmers worked under a fixed-term contract/service lease with fewer benefits. Insecurity related to future employment can negatively affect workers' health.4 A previous Norwegian study found that male agricultural workers had the highest HADS-D (Hospital Anxiety and Depression Scale) level of all occupational groups, and job insecurity may be a possible explanation.21 Due to their labour instability, farmers tend to overwork many more hours than is legally allowed (Law 27 671 rules the working day, hours and over time, established by the Peruvian government).22 Despite this, farmers have a lower average monthly salary than non-farmers, as it is considered unskilled labour where the only requirement is previous experience. Financial challenges negatively impact farmers’ mental health, for example, psychological distress, depression and less satisfaction with life, particularly in those settings where agriculture represents the main source of income.23 Also, farmers had heavier workloads compared with non-farmer workers. Kallioniemi et al found that stressors related to workload were associated with stress and burnout symptoms in Finland’s farmers.5 These results support our study findings.
In our setting, farmers were responsible for the planting, harvesting of the crops and sugarcane cutting. These activities involve a high physical and mental toll and are always carried out under the sun, often without choice or protection. Surprisingly, we did not observe an increased effect of heat stress on mental disorders. However, in the last 20 years, the average environmental temperature in Peru has increased due to global warming.24 This increase has been linked to an increase in depression, bipolar disorder and post-traumatic stress disorder cases, which indicates the severity of farmers' mental health. These trends are likely due to seasonal variations in serotonin levels in the brain, which are affected by temperature and light. As constant sun exposure decreases, serotonin levels in the brain slowly return to baseline.25 This phenomenon is called acclimatisation, and it can explain the protective effect of having a shaded work break against mental disorder symptoms.
Farmers presented more symptoms of mental disorders given the nature of their extremely demanding physical activities and their working conditions. In support of our claim, Hounsome et al in the UK found a difference of 1.21 in the GHQ-12 score between farmers and non-farmers.26 The farmer’s working conditions are a plausible explanation for our results. For instance, the intense, heavy-duty working shifts beyond the allowed legal limits are striking signs of precarious agricultural employment, especially in Peru. Although the agricultural sector in the country contributes to 9% of the gross domestic product and represents 24.7% of its economically active population,27 the farmers’ contract modality is notably diverse, and many times they are paid on a daily performance basis. Due to this and other factors, the agricultural sector has the highest poverty prevalence in Peru and, therefore, has poorer mental health consequences, as has been established elsewhere.28 Farmers from our study also had limited access to work-related social security benefits. This happens because many farmers have temporary contracts or do not have formal contracts.29 Similar results have been found across seasonal farmers in Ethiopia, where a higher prevalence of common mental disorders was reported .30
In our study, symptoms of mental disorders increased with additional hours of excessive work. This finding is consistent with the North American study reported by Kearney et al, where 60% of farmers who worked >40 hours per week reported being very stressed.31 Excessive working hours in stressful environments and poor working conditions have been found associated with increased mental health disorder symptoms. In Brazil, it is highlighted that the heavy workload is a definite farmer’s stressor.32 The Occupational Health and Safety guidelines recommend that farmers should work 75% of the time and rest 25% of it when carrying out heavy load activities in high ambient temperatures to avoid adverse health effects. In Peru, agricultural work is ruled by the Special Labor Regime Law (Law 27360-Promotion of the Agrarian Sector), which holds up to a maximum of 48 hours the farmer’s working week.27 However, this limit is usually not followed by their employers, which will not be often audited for labour law compliance or receive any sanctions from the Government.
Strengths and limitations
We explored a critical yet postponed issue among one of the main task forces in Peru and other LMICs. We used a locally validated version of the GHQ-12 as a screening instrument for mental disorders in our study population due to its satisfactory reliability sensitivity and specificity.33 Also, our study has some limitations that must be considered. The sample size was relatively small for our third objective. However, we tried to be conservative when fitting models related to this objective, for example, adjusting only for key potential confounders. Given the external evidence discussed above, we can be conclusive on the heavy workload and working hours per day. However, we cannot reach conclusions regarding pesticides exposure, occupational heat stress and shaded work breaks. Finally, we acknowledge that we did not use a random sampling and are aware of the possibility of sampling bias. However, the characteristics of age, level of education and low economic income described in our study are similar to those described in Peru’s National Agricultural Census,34 implying that our findings are representative of Peruvian farmers.
Occupational health implications
Good practices that protect and promote mental health in the workplace should bring together the implementation of social safety nets with health facilities to protect workers' mental health. The Peruvian government created community mental health centres in mental health reform (through Law 29889, in 2015) to ensure the provision of outpatient and specialised care for people with mental health disorders.35 In theory, farmers can and should be referred for specialised care. However, in practice, access to the nearest health centre is complicated, there are no strategies for early detection of mental health symptoms by the industry’s occupational health staff and farmers are afraid to report them due to fear of future repercussions. These will hold a serious barrier to access to timely treatment of mental disorders among agricultural workers.
Our results highlight that good practices for protecting and promoting mental health in the workplace should consider the following: the implementation and enforcement of health and safety policies and practices, including the identification of distress, drinking enough fluids, wearing appropriate clothing and scheduling work activities and breaks in the shade; informing staff that support is available; and organisational practices that support a healthy work–life balance.
Conclusion
Sugarcane farmers have higher mental disorder symptoms than their non-farmer peers. A heavy workload and more working hours per day are independently associated with more mental disorder symptoms. Our findings highlight the importance of including mental health within occupational programmes and early interventions tailored to sugarcane industrial mill workers in the Latin American context.
Data availability statement
No data are available.
Ethics statements
Patient consent for publication
Ethics approval
This study was approved by the Institutional Review Board of the Universidad Peruana Cayetano Heredia (ID code 19018). All data analysed for this study were deidentified and stored in one encrypted device. Participants gave informed consent to participate in the study before taking part.
Acknowledgments
The authors want to thank Essalud San Jacinto and the participants in the study. This report is independent research supported by the National Institute for Health Research ARC North Thames. The views expressed in this publication are solely those of the author(s) and not necessarily those of the institute.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Contributors JCB-A and JB-P conceived and designed the overall study. JB-P and JB supervised the overall study. JB-P and JCB-A developed the idea for this manuscript, led the statistical analysis and drafted the first version of the paper. EF provided important intellectual content and, with JB-P, JCB-A, and JB, drafted the manuscript. All authors participated in giving final approval of the submitted version. JCB-A is the guarantor.
Funding This study was funded by the Peruvian Consejo Nacional de Ciencia, Tecnología e Innovación Tecnológica through the Fondo Nacional de Desarrollo Científico, Tecnológico y de Innovación Tecnológica (contract number 171-2018-FONDECYT-BM-IADT-SE).
Competing interests None declared.
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting or dissemination plans of this research.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.