Respiratory disorders are not more common in farmers. Results from a study on Icelandic animal farmers

Abstract

Rationale

Respiratory disorders are more common among farmers than in other populations, despite less atopy and a lower prevalence of smoking; this is likely due to environmental exposure to organic dust and irritants. The current prevalence of respiratory disorders and atopy in Icelandic farmers is unknown, but a significantly greater rate of respiratory symptoms than the general population has been reported in the past. Modern farming practices, which effectively decrease the generation of respirable dust, have been implemented in Iceland in the past decade.

Objective

The aim of this study was to estimate the prevalence of respiratory disorders in a nationwide study of Icelandic farmers, and to identify any environmental or occupational associations with these disorders.

Methods

We conducted a questionnaire-based study of all Icelandic farmers with an age-matched comparison group randomly selected from the national citizen registry of Iceland. The questionnaire included items regarding respiratory symptoms and working environment.

Results

Out of 2042 farmers invited to participate, 1108 responded (54%), as did 689 of 1500 controls (46%). Farmers were slightly older and more likely to be male (77% vs. 47%). Smoking rates were significantly lower among farmers than among controls. The prevalence of asthma was not significantly different between the two groups, with a lifetime prevalence of 9.4% (n=104) among farmers compared to 10.2% (n=70) among controls. Medication use for asthma did not significantly differ between the farmer (6.5%) and control (4.8%) groups. The prevalence of self-reported, physician-diagnosed chronic bronchitis and emphysema likewise did not significantly differ between the groups, but self-reported “hay fever” was significantly more prevalent among farmers.

Conclusion

Concomitant with changes in farming practices the prevalence of respiratory disorders among Icelandic farmers has decreased. This suggests that modernization of the agricultural environment has a positive effect on workers health.

Introduction

Agricultural workers are reported to have a higher prevalence of respiratory disease than the general population ^1–4, despite generally lower rates of smoking ⁵. This has been ascribed to the result of chronic inhalation exposure to organic dust and other pollutants in their working environment ⁶. Agricultural dusts consist of a complex bioaerosol with major health effects caused by exposure to organic compounds ^1,2. Chronic inhalation of agricultural dusts is associated with an increased prevalence of respiratory symptoms and decline in lung function, some of which can be irreversible^1,2,7–10.

Agriculture in Iceland is centered on livestock (primarily cattle and sheep) for dairy, meat and wool; crop farming is uncommon. The atmosphere is temperate and somewhat humid, winters are mild and summers cool. Iceland is 103,000 km² (39,756 square miles), consisting of 23.8% agricultural land, 2.8% lakes, 11.9% glaciers and 64.5% non-tillable wasteland ²⁴. Weather conditions vary considerably with regards to average temperature, yearly rainfall and humidity. While the southwest is the most temperate region, and is where the greatest amount of farming activity is concentrated, the northwest and northeast parts of the island are also home to significant agriculture. Both of those regions have colder winters and lower yearly average rainfall than the southwest.

Previous studies of Icelandic farmers, conducted almost 20 years ago, found a high prevalence of respiratory discomfort while working with hay and other forage, not associated with a positive skin test for common allergens in hay ^22,23. Numerous changes in Icelandic agricultural practices and patterns have occurred since these studies were conducted, with a decline in the number of Icelandic farmers, consolidation of farms into larger units, increased automation and the introduction of large-scale animal confinement operations for the manufacturing of pork and poultry. Importantly, the processing of hay (a mainstay of animal feeds in Icelandic livestock operations) has markedly altered in the past decade; previously, dried hay was stored in enclosed barns, whereas now the use of vacuum-sealed plastic wraps has been adopted throughout the country, in part with the aim of decreasing the amount of dust in the agricultural environment. Because of the nearly universal use of this process, it is not currently possible to compare the effects of hay processing methodology.

The purpose of this study was to assess the respiratory health of Icelandic farmers (compared with a control group and with historical controls), and to consider the impact of changes in agricultural practice.

Methods

Study Design

We performed a survey of all registered farmers in Iceland operating a farm of more than 100 sheep units (40 animal units; (1 animal unit = 1,000 lbs milk cow)), with a comparison group, aged 25 – 70 years old, randomly selected from the national citizen registry of Iceland. The study was reviewed and approved by the National Bioethics Committee in Iceland the Icelandic Data Protection Agency and the Institutional Review Board of the University of Iowa.

Study Groups

A total of 2,042 Icelandic farmers are registered with the Icelandic Farmers Association (IFA) as operating a farm larger than 100 sheep units. This cutoff point is utilized by the IFA to designate farmers whose main income stems from farming activities and thus are likely to spend most of their working time on the farm or in farm related activities. No workers on Concentrated Animal Feeding Operations (CAFO) operations were included in the study. The individuals included in the study were sent a questionnaire, accompanied by a letter explaining the purpose and aim of the study. The comparison group consisted of 1,500 individuals from the general population who were sent a similar questionnaire and letter. Participants were asked to indicate on the questionnaire whether they would agree to be contacted in the future regarding later phases of the study. Responding to the questionnaire was considered informed consent for the purpose of this study. A postcard was mailed to non-responders asking them to complete the questionnaire. Advertisements were also taken out in local media in an attempt to increase response rates.

Questionnaire

The questionnaire was based on the Icelandic version of the European Respiratory Health Survey instrument ²⁵ and on similar questionnaires from earlier studies ^22,26. The main emphasis of the survey was on pulmonary and allergy symptoms and history as well as past medical history, work environment, mental health and occupational accidents. Basic demographics were obtained as well as smoking status, length of employment in current job, size of farm, type of livestock and type of crops.

Units of Measurement

We measured the prevalence of respiratory disorders as answered by the farmers on the questionnaire. Although we did ask for illnesses diagnosed by a physician, any history of a disorder was self reported. Respiratory symptoms were also self reported. Chronic bronchitis was diagnosed as productive cough at least three months out of the year. An index of respiratory symptoms was calculated, combining dyspnea, wheezing, chest tightness and cough into a numeric value between 0 and 4.

Statistics

The chi square test was used for dichotomous variables. Logistic regression models were created using the Statistical Package for the Social Sciences (SPSS) version 12.01. We controlled for smoking, age and gender in all models.

Results

Demographics

A total of 1107 of 2050 farmers responded (54%), while 689 out of 1500 (46%) individuals in the comparison group replied. The study group (farmers) was predominately male (77%) while the control group included more women (55%). The study group was older than the control group (p<0.0001) (Figure 1). Of the comparison group 47% (n=300) reported living in an urban area. This group, while better educated (Table 1), was comparable to rural controls in all other aspects, including smoking. When stratified by gender, smoking rates were significantly lower among farmers than among controls. This was true for both current smoking (p=0.026 for men and <0.0001 for women) and for any history of smoking (p=0.005 for men and <0.0001 for women) (Figure 2).

Table 1.

Demographics of study participants

	Farmers, %	Non-farmers, %	p value
Age >50 years	51	39	<0.001
Gender males	77	48	<0.001
Current smoking	13	20	<0.001
Current smoking (men)	14	20	<0.026
Current smoking (women)	8	22	<0.0001
Smoked >1 year	36	50	<0.001
Married or cohabitating	83	82	ns
Number of children <18 years	2.18 (sd 1.3)	2.05 (sd 1.13)	<0.042
Spouse works out of home	44	82	<0.001
Education elementary	43	30	<0.001

ns, not significant.

Overall, 85.4% (n=946) of the responding farmers reported raising livestock (sheep, cattle, poultry, or swine) as their main farming activity. Raising sheep was the most common main farming activity (n=516, 46.6%) and dairy farming (n=193, 17.4%) the second most frequent. The greatest number of farms ranged in size between 21 and 40 hectares, 30.5% (n=179). The mean number of animal units per farm was 133 (s.dev. 6.2). The participating farming population was evenly distributed throughout the country.

Respiratory disorders

The prevalence of asthma was not significantly different between the two groups. The lifetime prevalence of asthma was 9.4% (n=104) among farmers compared to 10.2% (n=70) among controls. Among farmers, 2.8% (n=31) reported symptoms of asthma in the past 12 months compared to 3.0% (n=21) among controls. Medication use for asthma was similar between the study (6.5%, n=63) and control (4.8%, n=29) groups (Figure 3). There was a sharp drop in the prevalence of asthma symptoms among farmers in the age group between 50 and 59 years of age (Figure 4).

After adjusting for smoking history, the prevalence of self reported physician-diagnosed chronic bronchitis and emphysema was not significantly different between the study and control groups, 2.9% vs. 2.3% (n=32 vs. n=16) for emphysema and 3.0% vs. 3.6% (n=33 vs. n=25) for chronic bronchitis (Figure 3). No difference was found between groups in a logistic regression model (Table 2.) The prevalence of respiratory disorders was the same regardless of residence in the north, south, west or east of Iceland. Furthermore the prevalence of respiratory disorders did not differ by type of farming.

Table 2.

Respiratory disorders and symptoms among study participants

Disorder	Farmers, %	Non-farmers, %	p value
Lifetime asthma	9.4	10.2	ns
Current asthma	3.2	3.5	ns
Chronic bronchitis	3.9	4.3	ns
Emphysema	3.8	2.8	ns
Medication use	6.5	4.8	ns
Symptom
Wheezing	9.4	10.6	ns
Chest tightness	7.8	12.8	<0.001
Morning cough	5.8	5.1	ns
Dyspnea	4.3	5.4	ns

ns, not significant.

Prevalence of respiratory symptoms

The prevalence of wheezing was similar in both groups: 9.4% in the study group and 10.6% in the control group. Nocturnal chest tightness was significantly increased in the control group. The occurrence of dyspnea was similar in both groups. The prevalence of productive morning cough at least 3 months out of the year was also similar between both groups, and substantially higher than the prevalence of self-reported physician-diagnosed chronic bronchitis (Table 2).

Respiratory symptoms and asthma prevalence were not significantly affected by stratification of the control group into urban and rural populations. The lifetime prevalence of asthma, prevalence of current wheezing and dyspnea was not significantly different between any of the groups (Figure 5).

A logistic regression model taking into account age, gender and smoking history found no difference between groups with regards of respiratory symptoms (Table 3). A total of 342 responders indicated any respiratory symptoms on the questionnaire. Using the index of respiratory symptoms as the outcome in linear model, using age, smoking history, or gender as exploratory variables did not identify any difference in the prevalence of respiratory symptoms between the two groups (Table x)

Allergic disorders and symptoms

There was no difference in the frequency of physician diagnosed allergic rhinitis between the two groups, 19% for farmers vs. 20.5% (n=185 vs. n=126). However symptoms of rhinitis were significantly less common among farmers than controls, 27.1% vs. 33.3% (n=266 vs. n=206) (p=0.008). The self reported occurrence of fever and systemic symptoms after exposure to hay was more common among farmers than controls, 8.3% vs. 3.0% (n=77 vs. n=16) (p<0.0001) (Figure 6).

Discussion

In this nationwide study of all Icelandic farmers we found no significant difference in the prevalence of respiratory disorders and symptoms between the study group and controls. We found that smoking was less common in farmers than in the control group, which is in agreement with previous reports ^5,27. The finding of self reported fever after exposure to organic dust was more common in farmers than in the control group. This is not surprising as these symptoms suggest either Organic Dust Toxic Syndrome or Hypersensitivity Pneumonitis, both of which would be expected to be more common in the farming population²⁸.

Finding no significant difference in the prevalence of respiratory disorders between the study group and controls is contrary to previous studies which suggest that respiratory disorders are markedly more common among farmers and agricultural workers in general^1–4,6,25. The European Community Respiratory Health Survey (EHSRC) found farming activities to be a significant risk factor for asthma although a relatively small number of farmers were included in the study²⁵. Previous studies on Icelandic farmers from two decades ago found high prevalence rates of obstructive lung disease and hypersensitivity pneumonitis among farmers in all age groups^22,23. Farmers’ poor respiratory health in the past has been associated with inhalation of organic dust, a complex mixture of environmental pollutants that contains a number of inflammatory mediators. Lipopolysaccharides (endotoxin) and bacterial DNA are both associated with acute and chronic inflammation (evidence coming from both murine and human studies^{11, 12, 20}); interestingly, both of these agents have also been linked to the protective effects that have led to the “hygiene hypothesis,” and may represent a double-edged sword of agricultural life^{4, 7, 19, 21}. Thus exposures to agricultural pollutants may both irritate and inflame the lung but also may lead to protection against inflammatory disorders, including asthma and atopy, inflammatory bowel disease, and multiple sclerosis.

There was no variation in respiratory symptoms among different regions in Iceland nor was there any difference based on type of livestock used. This would indicate that any improvement in respiratory health of farmers is likely to be due to factors impacting on the Icelandic farming population in general rather than regional factors, including climate. One such factor is the uniform change in Icelandic farming practices over the past two decades towards utilizing vacuum sealed plastic wraps for storing hay. Other significant changes include increased automation, the consolidation of smaller farms into larger units and increased levels of education. While improved methods of storing hay and decreased exposure to dust with increased automation are attractive explanation to the findings of our study, nothing is currently known about dust levels in Icelandic animal barns and the potential length of exposure among farm workers.

This study has a number of weaknesses. A relatively low response rate, albeit in a study population including the whole population of farmers in Iceland, raises the possibility of under representation of individuals with lung disease in the study group or over representation in the control group. Low smoking rates among farmers may result in lower prevalence of respiratory disease and a healthy worker effect may also be a factor in this study. However a healthy worker effect would likely have been present in other studies and low smoking rates are common in farming populations.

It is of interest that the symptoms of current wheezing and chest tightness are more common than self reported physician diagnosed asthma. Furthermore the occurrence of productive morning cough is significantly higher than self reported diagnosis of chronic bronchitis and emphysema. This suggests that patients who may meet diagnostic criteria for obstructive lung disease are not being recognized as such, and raises the concern that these individuals may thus not be receiving appropriate treatment.

The high prevalence of respiratory disorders commonly present among the generally healthy low smoking population of agricultural workers found in numerous other studies was not present in our study of all Icelandic farmers. Although the potential for bias certainly exists, this improvement is most likely due to widespread changes in farming practices that have taken place over the past two decades. The public health implication of our findings is considerable as the considerable volume of respiratory ailments historically linked with agricultural work can be avoided with modernization of farming practices.