INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Farming is a well-known risk factor for bronchial asthma, chronic bronchitis, and organic dust toxic syndrome (ODTS) (1). For farmers exposed to livestock, the role of animal exposure in the development of respiratory symptoms has been well characterized, but the impact of crop farming on respiratory health remains ill defined. The development of symptoms in crop farmers may be related to exposure to a variety of well-known etiological agents present in agricultural settings (4): pollens, grain dust, moulds, mites, glucans, endotoxin, ozone, and pesticides.

We present the results of a cross-sectional study of 4,793 crop farmers from four European countries, with the aim of determining the prevalence of respiratory symptoms of obstructive lung disease in farmers producing different plants and to assess the role of different crops as risk factors for the appearance of such symptoms. Additionally, we also investigated whether work inside greenhouses might be related to an increase in the prevalence of respiratory symptoms.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Design

As part of the European Union Concerted Action "Prevalence and risk factors for airway obstruction in farmers" (9), a cross-sectional study on the prevalence of respiratory symptoms in Western European crop farmers was performed with the assessment of different farming characteristics and of the association between plant crop exposures and occurrence of symptoms.

Population

Farmers from four European countries (Denmark, Germany, Switzerland, and Spain) participated in the European Union Concerted Action between 1995 and 1997. In each country, a region located within 300 miles of the participating occupational health center that fulfills the requirement of a population > 50,000 farmers was defined as the area to be studied. In every region, a target population sample (> 1,500 subjects) was selected at random from the most recent available census record of farmers in every region (state census in Switzerland and farmer organization censuses in Denmark, Germany, and Spain). In Denmark, Switzerland, and Spain, the farmers were contacted by mail and asked to fill out a self-administered questionnaire on respiratory symptoms and occupational characteristics. Up to two postal reminders followed the first letter. In Spain, after the second postal reminder, nonresponders were interviewed by phone or visited at home if they could not be reached, at which time the questionnaire was filled out. In Germany, questionnaires were filled out through home visits in all cases.

Questionnaire

A self-administered questionnaire with 16 questions was used. Ten questions were about age, sex, smoking habits, and respiratory symptoms within the last year. Wheezing, shortness of breath at night, asthma attacks, nasal allergy, chronic phlegm (phlegm > 3 mo a year), symptoms at work (cough, wheezing, and shortness of breath), and symptoms of ODTS were considered. Six questions on occupational exposure asked the subject if his or her work involved livestock and/or plant cultivation, what plants he or she produced regularly (grain, vegetables/tomatoes, root crops, oil plants, tobacco, hops, mushrooms, fruits/nuts, or flowers), if his or her work was accomplished inside greenhouses, and the number of hours per day that were worked inside.

Questions about symptoms were obtained from the European Community Respiratory Health Survey (ECRHS) questionnaire (10, 11), and occupational questions were adapted from the questionnaire evaluating organic dust exposure (12). Previously validated local language (Danish, German, Spanish, or Catalan) versions of the questions were used in each country (13), or, when a validated version did not exist, the questions were translated from English to the local language and then retranslated.

Statistical Analysis

All data were introduced in a database and analyzed using the SPSS statistical software package (Chicago, IL). Descriptive statistics were performed for the whole group and for each country separately, expressing the results as proportions or as means with standard deviations when appropriate. Associations between variables were assessed through inferential statistics for the whole sample. Respiratory symptoms were considered dependent variables, and the different plant crops were independent variables. The symptoms considered were the following: (1) wheezing, (2) asthma attacks, (3) nasal allergy, (4) chronic phlegm, (5) ODTS, and (6) symptoms at work. Cross-tabulations were performed first, using the chi-square distribution to compare prevalence rates of categoric variables. To assess the relative importance of the different exposures as risk factors, multivariate analyses were performed using logistic regression modeling, with the different respiratory symptoms as dependent variables and the following independent variables: age, sex, smoking, livestock exposure, country, and plant crops. The results were given as adjusted odds ratios (OR), with 95% confidence intervals (CI). Finally, to investigate a possible relation between work inside greenhouses and symptoms, we assessed the dose-response relationship between hours per day worked in the greenhouse and prevalence of symptoms, for the subsample of crop farmers with more than 5% of the workforce working inside greenhouses (farmers cultivating vegetables/tomatoes, fruits/ nuts, or flowers). Cross-tabulations were performed after categorizing the number of hours worked inside; and the associations were also analyzed using logistic regression, adjusting for age, sex, smoking, livestock exposure, and country. All statistical tests were two sided, and a p value equal to or less than 0.05 was reported as significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Population

In Denmark, Germany, Switzerland, and Spain, a sample of 7,674 farmers was the target population. After adjusting for census error (wrong address, change of residence, or death) and misclassification (no livestock and/or plant crop exposure at work), a population sample of 6,751 farm workers was defined. Of these, 5,761 farmers answered the questionnaire (overall response rate: 85.3%), with local response rates varying from a minimum of 80.6% to a maximum of 97.8%. In Germany, a second target population was defined that included the workforce of 1,247 farms registered in the local farm census of Schleswig-Holstein. A total of 834 such farms agreed to participate (response rate: 66.9%), thus giving a workforce of 1,735 farmers to study. From the 7,496 farmers who participated in the study, 4,793 subjects were full-time crop farmers at the time of the survey (63.9%), and this was the sample presented in this study.

Exposures

Slight age, sex, and smoking differences were observed from country to country, along with clear-cut differences in the prevalences of the different crop exposures (Table 1). Tobacco, hops, and mushrooms were seldom cultivated (n < 35 subjects in every case), and for this reason these crops were not analyzed.

Association between Respiratory Symptoms and Crop Exposure

No significant differences in the prevalence of respiratory symptoms were found between farmers cultivating vegetables/tomatoes and fruits/nuts versus others. The cultivation of flowers or oil plants, however, was associated with two or more respiratory symptoms. Flower farmers had significantly higher prevalences of wheezing (20.1%), asthma (5.4%), chronic phlegm (16.1%), and ODTS (19.4%), when compared with nonexposed farmers (chi-square, p < 0.05). Work with oil plants was associated with ODTS (18.3%) and symptoms at work (26.5%) (chi-square, p < 0.001). A significantly higher prevalence of ODTS was also observed in workers cultivating grain (16.1%) and root crops (17.8%) (chi-square, p < 0.01) (Table 2).

Logistic regression models were created to confirm the relevance of each crop as a risk factor for respiratory symptoms, and work with flowers and oil plants appeared as the main risk factors after adjusting for all covariates. Flower cultivation was a clear-cut risk factor for asthma (OR 2.1. 95% CI 1.1- 3.9), and the production of oil plant crops was significantly associated with ODTS (OR 1.5, 95% CI 1.3-1.9), symptoms at work (OR 1.4, 95% CI 1.2-1.7), and chronic phlegm (OR 1.3, 95% CI 1.1-1.6) (Table 3).

Work inside Greenhouses

The possible role of greenhouse crop farming as risk factor for respiratory symptoms was assessed in the subsample of crop farmers with more than 5% of the workforce working inside greenhouses (n = 1,363). Of this subsample, 210 subjects worked inside (15.4%) (vegetables/tomatoes n = 150, fruits/ nuts n = 51, flowers n = 85). In the analysis, no dose-response relationship between the number of hours spent inside the greenhouse and respiratory symptoms was observed (chi-square, n.s.). In the logistic regression models, however, such greenhouse work appeared to be a marginal risk factor for asthma (OR 2.1, 95% CI 0.9-4.8) (Table 4).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In the present cross-sectional study, a sample of crop farmers from four European countries has been examined to assess the prevalence of respiratory symptoms of obstructive lung disease in such farmers, and the association of symptoms with the exposure to different plant crops. After adjusting for covariates, flower growing emerged as the most important risk factor for asthma, and the cultivation of oil plants for acute respiratory symptoms and chronic phlegm. Work inside greenhouses was a marginal risk factor for asthma.

Assessments of lung health in rural populations where farming is a main occupational activity have shown that respiratory symptom prevalence ranges widely (14). Differences may be related to the inclusion of populations with different age ranges, or to heterogeneity of the occupational exposures (17). Iversen and coworkers, in a cross-sectional study of Danish farmers, found a prevalence of 27% for chronic bronchitis and a prevalence of 8% for bronchial asthma (18). A lower prevalence of asthma (5.3%) was found by Dalphin and coworkers in French farmers (19). When farmers have been compared to nonfarming control populations from the same regions, an increased prevalence of respiratory symptoms has often been reported in currently exposed farm workers (19). The high prevalence of respiratory symptoms in farmers may not always be evident, however, because symptomatic farm workers may leave farming more often than asymptomatic ones. The rate of change of occupation has been reported to be low among farm owners, however (22).

Farming has often been associated with respiratory symptoms in workers exposed to livestock (23), but crop production has been much less studied. Much of the work done on workers exposed to plants has focused on the manufacture and use of plant products: specifically, respiratory symptoms induced by grain dust have been well described in grainhandlers (26, 27), and byssinosis has been documented in cotton mill workers (28). Likewise, endotoxin is a proven cause of respiratory symptoms in both livestock handling and in food plant processing (29). A study of crop farmers cultivating cereals revealed a 10% prevalence of chronic phlegm, twice the level found in a control population from the same French region (30). Vergnenegre and coworkers found a similar prevalence of chronic phlegm (7.7%), as well as a higher risk of obstructive lung disease in agricultural workers employed on small farms (31). Gramsky and coworkers, in a study of 759 Californian agricultural farm workers, found that the risk for respiratory symptoms increased when the workers were exposed to the field crops more than 8 mo per year (32). In the same region, McCurdy and coworkers found a 7.1% prevalence of bronchial asthma in male agricultural workers growing rice (33).

In our study, we examined a sample from a population of over 200,000 European farmers growing crops that are common in the four countries studied, finding that 15.0% reported wheezing, 3.2% asthma, 14.5% nasal allergies, and 12.2% chronic phlegm. These figures are similar to the prevalences found for the general European population (34) and do not suggest a higher prevalence of symptoms of obstructive lung disease among European agricultural workers. The analysis of specific crops, however, showed an unexpectedly higher prevalence of wheezing (20.5%) and asthma (5.1%) in workers who grew flowers (p < 0.05). Working with flowers continued to be associated with a high risk of bronchial asthma (OR 2.1, 95% CI 1.1-3.9) after adjustment for covariates. Occasional cases of occupational asthma in flower workers have been described (35), but to our knowledge, no epidemiological association between flower production and bronchial asthma has been previously reported.

A high prevalence of acute symptoms was found among oil plant farmers (ODTS 18.3%, symptoms at work 26.5%), suggesting an acute impact of work with this plant crop on lung health. This effect was confirmed by multivariate analysis, which demonstrated that oil plant cultivation was not only a risk factor for ODTS (OR 1.5, 95% CI 1.3-1.9) and symptoms at work (OR 1.4, 95% CI 1.2-1.7), but also for chronic phlegm (OR 1.3, 95% CI 1.1-1.6). ODTS is considered a self-limiting disease that rarely extends beyond 36 h. Influenza-like symptoms, including malaise, myalgias, fatigue, and fever, appear, sometimes with a decrease in lung volumes. Most episodes are related to the handling of mouldy vegetables, but the etiology of the syndrome is unclear.

We have found an association between ODTS and the cultivation of grain (OR 1.5, 95% CI 1.1-2.1). This effect of grain crops on lung health was not unexpected, considering the well-known relation between grain processing and respiratory symptoms. The reported higher prevalence of symptoms in plant manufacture workers is probably related to exposure to grain stored after harvest, when bacteria and moulds overgrow and endotoxin and glucans appear as byproducts of these microorganisms. Major inflammatory substances that induce respiratory symptoms will thus be present at much higher levels in storage than during harvest.

Work inside greenhouses emerged as only a marginal risk factor for respiratory disease in the present study. A tendency for a higher prevalence of symptoms with the increase in the number of hours spent in the greenhouse was not seen, but some impact of greenhouse work on symptoms was suggested by multivariate analysis, which showed a trend for increased risk of asthma (OR 2.1, 95% CI 0.9-4.5).

In summary, flower production seems to be a risk factor for bronchial asthma in European crop farmers. Cultivating oil plants increases the risk of acute respiratory symptoms, such as ODTS, and of chronic phlegm. These observations suggest that workers handling these specific plant crops need to be carefully monitored and that further work on the mechanisms of particular agents is warranted.