Respiratory Medicine 115 (2016) 13e19

Contents lists available at ScienceDirect

Respiratory Medicine
journal homepage: www.elsevier.com/locate/rmed

High prevalence of obstructive lung disease in non-smoking farmers:

The Irish farmers lung health study
Breda Cushen a, *, Imran Sulaiman b, Noreen Donoghue c, Donna Langan a, Tara Cahill a,
Eimear Nic Dhonncha a, Olivia Healy a, Fiona Keegan a, Mark Browne d, Anthony O'Regan a
a
Department of Respiratory Medicine, Galway University Hospital, Galway, Ireland
b
Royal College of Surgeons in Ireland, Dublin, Ireland
c
Department of Respiratory Medicine, Bon Secours Hospital, Galway, Ireland
d
National University of Ireland, Galway, Ireland

a r t i c l e i n f o a b s t r a c t

Article history: Background: Mortality rates within the Irish farming community are increasing, whilst that of the
Received 25 August 2015 general population falls. The aim of this cross-sectional study was to determine the prevalence of res-
Received in revised form piratory disease amongst Irish farmers.
16 March 2016
Methods: All study participants were farming volunteers attending an agricultural exhibition. Data
Accepted 17 April 2016
collected by questionnaire included baseline demographics, respiratory history, presence of respiratory
Available online 20 April 2016
symptoms and occupational exposures. Spirometry was performed on all participants.
Results: Data from 372 farmers was analysed. The majority were male (76%) with median age of 55 years.
Keywords:
Occupational lung disease
61% were never smokers. 13% were previously diagnosed with airway disease (Chronic Obstructive
Obstructive lung disease Pulmonary Disease(COPD)/Asthma/Inhaler use) with 14% reporting hayfever/allergies. Almost two-thirds
COPD epidemiology reported one or more chronic respiratory symptom. Forty-four (12%) had obstructive spirometry using
fixed FEV1/FVC < 0.70 criterion and 29 (7.8%) using FEV1/FVC < 5% lower limit of normal. The majority,
two-thirds, were never smokers. Amongst never smokers with obstruction (13%), there was a signifi-
cantly higher proportion with a prior diagnosis of airway disease and hayfever/allergies. There was no
significant association between specific occupational exposures and obstruction.
Conclusion: The majority of Irish farmers are never smokers. They have a high prevalence of respiratory
symptoms. 13% of never smokers have airflow obstruction (FEV1/FVC < 0.70). The presence of airflow
obstruction is significantly associated with self-reported allergy history and prior airway disease. Further
studies are needed to identify the workplace factors accounting for these findings.
© 2016 Elsevier Ltd. All rights reserved.

1. Introduction
An estimated 25% of adults over 40 years of age have airflow
limitation [1] with a worldwide prevalence of 10% [2]. Whilst the
risk of airflow limitation secondary to smoke exposure is well
established, there is increasing recognition of other risk factors
such as air pollution and occupational exposures [1]. Respiratory
Abbreviations: COPD, Chronic Obstructive Pulmonary Disease; FEV1, Forced
Expiratory Volume in one second; FVC, Forced Vital Capacity; LLN, lower limit of
normal; SMR, Standardised Mortality Ratio; GOLD, Global Initiative of Chronic
Obstructive Lung Disease; ATS, American Thoracic Society; ERS, European Thoracic
Society.
* Corresponding author.
E-mail address: bredacushen@rcsi.ie (B. Cushen).
disease associated with agriculture was one of the first recognised
occupational hazards. As far back as 16th century, the potential
complications of inhaling dry grain dust were described [3]. Hy-
persensitivity pneumonitis occurring in agricultural environments
is well described but recent studies have shown its prevalence is
declining in the UK and Ireland [4], with a reported prevalence of
<3% [5]. This most probably reflects changes in farming practices
notably a decrease in the production of hay in favour of silage [4]. In
fact airflow limitation is more common amongst agricultural
workers [6e8]. This, in addition to high rates of respiratory
symptoms [9,10], likely impacts most on farmers' respiratory
health.
In Ireland, almost 200,000 people are employed by the primary
agriculture sector accounting for 5% of total population employ-
ment [11]. Overall, farmers are perceived to be healthier than the

http://dx.doi.org/10.1016/j.rmed.2016.04.006
0954-6111/© 2016 Elsevier Ltd. All rights reserved.
14 B. Cushen et al. / Respiratory Medicine 115 (2016) 13e19
average population due to their outdoor lifestyle and the physical
nature of their work with numerous studies finding overall reduced
mortality rates in this cohort when compared with the general
population [12,13]. However, a 2013 study analysing mortality
trends in Irish farmers over the last decade found that all-cause
mortality in this subpopulation is increasing with a standardised
mortality ratio (SMR) of 182.5 in 2006 compared with 132.4 in
2000. Farmers were found to be 5.1 times more likely to die from
any cause than salaried employees. This was mainly attributed to
increased circulatory disease (SMR 215.9 vs. 31.9) and cancers (SMR
156.6 vs. 44.3) as well as increased workplace accidents [14].
However, there appears to be little published data on the preva-
lence of respiratory disease in the Irish farming population. In view
of these findings and the high rates of airflow limitation and
chronic respiratory symptoms reported in farmers from other
countries, we undertook the Irish Farmer's Lung Health Study to
determine the prevalence of respiratory disease and airflow limi-
tation in the Irish farming population.
2. Methods and materials
2.1. Design
This cross-sectional study was conducted by the Respiratory
Department of the Galway University Hospitals Group, Ireland. It
was carried out over two consecutive days in September 2013 at the
Irish National Ploughing Championship. The National Ploughing
Championship is an annual agriculture and trade show which at-
tracts in the region of 190,000 visitors per year, the majority of
whom are farmers and agriculture workers. All study participants
were farming volunteers who were attending the event. Ethical
approval to conduct the study was granted by the Ethics Committee
of Galway University Hospital.
2.2. Questionnaire
The questionnaire, compiled by the investigators, incorporated
questions related to baseline demographics, self-reported past
respiratory history, self-reported respiratory symptoms, farmyard
practices (e.g. farming type, type of feed used, and chemical ex-
posures) as well as exposure to known risk factors for respiratory
disease, including tobacco smoke and burning of fossil fuels. The
questionnaire was delivered face-to-face by data collectors with
participants asked to provide a yes or no answer to the questions
posed.
2.3. Outcome definitions
A never smoker was defined as a person who denied having ever
smoked in their lifetime. Previous respiratory history included self-
reported physician diagnosis of Asthma, Chronic Obstructive Pul-
monary Disease (COPD) or Emphysema, Hayfever/Allergies,
Farmer's Lung, Tuberculosis or Sarcoidosis at any stage in their
lifetime. The prescription of inhaled medications was also deemed
indicative of a prior diagnosis of airways disease.
Participants were also asked to report whether or not they
experienced cough, dyspnoea, wheeze, nasal congestion, sinus
headache, anosmia, rhinorrhoea or post-nasal drip on a daily or
intermittent basis. For the purpose of analysis, investigators
grouped symptoms into three distinct categories: 1) Upper Respi-
ratory Tract, which included nasal congestion, sinus headache,
anosmia, rhinorrhoea and post-nasal drip 2) Dyspnoea and/or
Wheeze and 3) Cough.
To assess occupational exposures, farming practices were
grouped into three categories: 1) Type of Farming e Livestock/
Arable/Mixed; 2) Type of feed used e Hay/Haylage/Silage/Dry
Grain; and 3) Chemical exposures e chemical spraying (e.g. pesti-
cides) on the farm.
Participants also reported whether they burned fossil fuels in
the home in particular wood and turf and if they wore personal
protective equipment during their work on the farm.
2.4. Spirometry
Onsite height and weight was recorded followed by spirometry
measurement carried out by trained respiratory scientists using
calibrated MicrolabTM spirometers. All recordings were subse-
quently reviewed by a consultant respiratory physician. We used a
fixed FEV1/FVC ratio of <0.70 to define airflow limitation as rec-
ommended by the Global Initiative of Chronic Obstructive Lung
Disease (GOLD) [15]. In addition we evaluated the FEV1/FVC <5%
lower limit of normal (LLN) using Hankinson et al. reference
equations [16]. Only those which were deemed acceptable and
reproducible in line with ERS/ATS spirometry guidelines [17] were
included in analysis.
2.5. Presentation of results and statistical analysis
Descriptive statistics are presented as counts and percentage for
categorical data and mean ± standard deviation(SD) for continuous
variables, unless otherwise stated. The primary outcome measure
was airflow obstruction on spirometry (Fixed FEV1/FVC <0.70).
Secondary outcomes included FEV1, % predicted FEV1, FVC, % pre-
dicted FVC and respiratory symptoms. Respiratory symptoms were
grouped into upper respiratory tract, lower respiratory tract or
cough as outlined previously. Further, those with a history of
Asthma or COPD or inhaled therapy were grouped as “prior airways
disease”. Initial analysis was carried out in the whole cohort with
subsequent analysis in the never smoker cohort only.
Chi-squared test or Fishers Exact test, where appropriate, were
used to compare categorical variables between groups (airflow
obstruction vs. normal spirometry; presence or absence of respi-
ratory symptoms). Between group comparisons of continuous data
were carried out using one way analysis of variance (ANOVA) and
Kruskal Wallis test, for parametric and non-parametric data
respectively.
Univariate logistic regression analysis was performed to identify
any association between individual variables and categorical out-
comes. Multivariate analysis was performed using logistic regres-
sion. Age and gender were included in the model as were all
variables whose p value was <0.25 on univariate analysis. These
included BMI, Prior airway disease, Hayfever/allergies, Burning of
turf, Arable farming, Use of dry grain, silage and haylage feed, Lower
respiratory tract symptoms and Upper respiratory tract symptoms.
Multiple linear regression analysis was used to identify associ-
ations between variables and continuous secondary outcome
measures.
Findings were deemed statistically significant at p value < 0.05.
All statistical analyses were performed using Stata version 13
(StataCorp. 2011. Stata Statistical Software: Release 13. College
Station, TX: StataCorp LP).
3. Results
3.1. Baseline demographics and respiratory health
In all, 400 volunteers completed questionnaires and spirometry.
Data from 11 non-farmers was excluded. A further 17 were dis-
counted due to: incomplete data (n ¼ 8) and inadequate spirometry
(n ¼ 9). Data from the remaining 372 was included in analysis.
 B. Cushen et al. / Respiratory Medicine 115 (2016) 13e19 15

Baseline demographics and self-reported respiratory history of The prevalence of airflow obstruction amongst never smokers
the cohort are shown in Table 1. The median age was 55 years (IQR was high and did not differ significantly from smokers. 13% of never
47e62years) and the majority were male (76%) and livestock smokers studied had obstruction as compared to 10% of smokers
farmers (80%) in keeping with the known demographic of Irish when fixed ratio was applied, p ¼ 0.48. When LLN was used as a
farmers. threshold for FEV1/FVC, 7.5% of never smokers and 8% of smokers
The majority, 89%, of those studied were overweight or obese, demonstrated airflow obstruction, p ¼ 0.78. The majority of those
median BMI 29 kg/m2, IQR 26-32 kg/m2. studied had mild obstruction (GOLD Stage 1) with FEV1>80% pre-
Interestingly, 61% of those questioned had never smoked dicted, Table 2.
(n ¼ 227) with only 9% current smokers. This is much lower than Using the fixed ratio (FEV1/FVC < 0.70), never smokers
the general Irish population smoking prevalence of 19.5% [18] but comprised 66% of those with obstruction as compared to 59% using
consistent with previously published data on smoking trends in <5%LLN criteria.
agriculture workers [19]. Fifty volunteers (13%) reported a prior
diagnosis of airways disease, or had been prescribed inhalers by 3.3. Respiratory symptoms
their GP, Table 1. The majority reported Asthma (9%) rather than
COPD (1%). Less than 1% (n ¼ 3) had been diagnosed with Farmer's Self-reported respiratory symptoms for the entire cohort are
Lung in the past. shown in Table 3. The prevalence of chronic respiratory symptoms
was high at 62%. Forty percent of subjects reported upper respira-
3.2. Spirometric analysis tory tract (URT) symptoms; 33% reported cough, and 32% reported
dyspnoea and/or wheeze. There was no significant difference in the
Of the 372 acceptable spirometric recordings, airflow obstruc- numbers reporting respiratory symptoms amongst the smoking
tion was identified in 44 subjects (12%) using fixed FEV1/FVC < 0.70. and never smoking cohorts.
When FEV1/FVC < 5%LLN was applied, prevalence of airflow Of the never smokers, a significantly higher proportion of those
obstruction was lower at 7.8% (n ¼ 29). Overall mean FEV1 values with obstruction (FEV1/FVC <0.70) reported lower respiratory tract
were 3.26 ± 0.75 L (L) (FEV1% predicted 99 ± 16%, mean ± SD) with symptoms (52%), as compared to those with normal spirometry
FVC values of 4.21 ± 0.93L (FVC % predicted 104 ± 15%). Obstructed (27%), p < 0.01 (Table 3). Obstructed never smokers reported
smokers had lower mean FEV1 recordings compared with the never significantly increased wheeze (34% vs. 16%, p ¼ 0.01). There was no
smoking group, 2.53 ± 0.89L (79 ± 21% predicted) and 2.91 ± 0.8L significant difference in the prevalence of cough or upper respira-
(86 ± 14% predicted), respectively, however this was not statisti- tory tract symptoms between the two groups.
cally significant, p ¼ 0.17. There was no significant difference in age Of those never smokers whose FEV1/FVC ratio was above LLN
between those with airflow obstruction as defined by FEV1/ but less than the fixed 0.70 cut-off, 50% reported symptoms of daily
FVC < 0.70 only (n ¼ 15; median 55 years, IQR 50-58) or both FEV1/ or intermittent dyspnoea, wheeze or cough compared to 27% with
FVC < 0.70 and <5%LLN criteria (n ¼ 29; median 54 years, IQR 46- normal spirometry.
64), p ¼ 0.71. Those reporting upper and lower respiratory tract symptoms
had significantly lower FEV1 (%), Fig. 1, and FVC (%) values. On
Table 1 multivariate analysis, there was trend towards lower FEV1 in those
Baseline demographics of the farming cohort. reporting lower respiratory tract symptoms (p ¼ 0.05) but no
Number of farming participants ¼ 372
further association between symptoms and other outcome
Age, years measures.
Median (IQR) 55 (47e62) There was no significant association between respiratory
Sex symptoms and the farming exposures assessed on univariate or
Male 284 (76)
multivariate logistic regression analysis.
Female 88 (24)
Farming Type, %
Livestock 80 3.4. Risk factors for airflow obstruction in non-smoking farmers
Arable 4
Mixed 16
Smoking Status
Univariate analysis revealed several significant non-smoking
Never smoker 227 (61)
Current/ex-smoker 145 (39) Table 2
Body Mass Index (n ¼ 371) Distribution of FEV1 by GOLD grade of the entire cohort and never smoking and
Underweight 0 smoking subgroups.
Normal 38 (10)
Gold grade All Never smoker Smoker
Overweight 169 (45)
n ¼ 372 n ¼ 227 n ¼ 145
Obese 157 (42)
Morbidly Obese 7 (2) FEV1/FVC ratio FEV1/FVC ratio FEV1/FVC ratio
Level of Education (n ¼ 368)
<0.70 <5%LLN <0.70 <5%LLN <0.70 <5%LLN
Primary 43 (12)
n ¼ 44 n ¼ 29 n ¼ 29 n ¼ 17 n ¼ 15 n ¼ 12
Post-Primary 188 (51)
Third level 137 (37) 1 29 (66) 15 (52) 20 (69) 9 (53) 9 (60) 6 (50)
Prior Respiratory History FEV1 > 80%
Prior Airways Diseasea 50 (13) 2 13 (30) 12 (41) 9 (31) 8 (47) 4 (27) 4 (33)
Asthma 34 (9) 50% < FEV1 < 80%
COPD 4 (1) 3 2 (4) 2 (7) 0 0 2 (13) 2 (17)
Farmer's Lung 3 (<1) 30% < FEV1 < 50%
Allergic Rhinitis 51(14) 4 0 0 0 0 0 0
Other lung diseaseb 4(<1) FEV1 < 30%
Inhaled medication 12 (3)
Count and proportion, n(%), of participants with airflow obstruction by FEV1/
a
COPD/Asthma/Inhaler use. FVC < 0.70 and FEV1/FVC < 5% lower limit of normal criteria stratified by Gold Grade
b
Tuberculosis/Sarcoidosis. Data are presented as count (%) unless and grouped by smoking status. GOLD-Global Initiative of Chronic Obstructive Lung
otherwise stated. Disease.
16 B. Cushen et al. / Respiratory Medicine 115 (2016) 13e19

Table 3 association between FEV1 (%), FVC (%) and age, p < 0.01, with fe-
Prevalence of Respiratory Symptoms of all participants, and never smokers grouped males demonstrating significantly lower values, p < 0.05. However,
by FEV1/FVC ratio.
neither age nor gender was found to be significantly associated
Whole Never Never p value with presence of obstruction.
group smokers smokers None of the potential exposures assessed accounted for the
N ¼ 372 FEV1/ FEV1/
FVC FVC0.70
presence of airflow obstruction in the cohort nor was there any
<0.70 n ¼ 198 significant relationship between these variables and FEV1 (%) or
n ¼ 29 FVC (%). These included farm type (livestock vs. arable), feed used
n % n % n % (grain vs. hay vs. silage), chemical exposures, use of Personal Pro-
tective Equipment and burning of fossil fuels, Fig. 2.
Any Respiratory Symptom 231 62 20 69 116 59 0.28
Cough 124 33 10 34 68 34 0.98 In multivariate analysis, following adjustment for age and
Lower Respiratory Tract 119 32 15 52 53 27 <0.01 gender, there was a significant association between prior history of
Wheeze 70 19 10 34 31 16 0.01 airways disease and obstruction, p < 0.01(OR 5.03, 95%CI 1.7e14.8),
Dyspnoea 95 26 10 34 42 21 0.11 FEV1, p < 0.05, and % predicted FEV1, p ¼ 0.01, Fig. 1.
Upper Respiratory Tract 150 40 16 55 77 39 0.09
Nasal Congestion 96 26 8 28 55 28 0.93
Sinus Headache 52 14 5 17 31 16 0.72 4. Discussion
Anosmia 65 17 4 14 37 19 0.81
Rhinnorrhoea 55 15 6 21 32 16 0.55
Post-Nasal Drip 32 9 3 10 21 11 1.00 In this study, the first comprehensive study of the respiratory
health of a large cohort of Irish farmers, we have shown a high rate
Count and proportion of participants reporting respiratory symptoms. Significant p
values highlighted in bold for chi-squared test or Fisher's exact test (for n < 5),
of non-smoking related obstructive physiology with associated
comparing never smokers with obstruction (FEV1/FVC < 0.70) to never smokers respiratory symptoms. Our results, while in keeping with studies of
with normal spirometry (FEV1/FVC  0.70). farmers in other countries, demonstrate the highest reported
prevalence of respiratory symptoms in a farming population.
We found that over 60% of Irish farmers report one or more
risk factors for airflow obstruction in this cohort. These are pre- chronic respiratory symptom. There was a broad range of symp-
sented in Table 4. Obstruction was significantly more common in toms with a slight predominance for those pertaining to the upper
never smokers with a prior diagnosis of airways disease (Odds Ratio respiratory tract. An increased prevalence of respiratory symptoms,
(OR) 6.5, 95%CI 2.76e15, p < 0.01) and in those reporting hayfever chronic bronchitis and airflow obstruction in farmers, in particular
or allergies (OR 2.5, 95%CI 1.04e6.05, p ¼ 0.03. There was a negative amongst livestock farmers, have been reported in other European

Fig. 1. FEV1 (%) and Respiratory Symptoms/Disease in a Non-Smoking Farming Population. Univariate linear regression analysis identified significant reductions in % predicted FEV1
values in those reporting lower respiratory tract symptoms, p < 0.01 (a), upper respiratory tract symptoms, p < 0.05 (b), prior history of airways disease, p < 0.01 (c) and hayfever/
allergy, p < 0.05 (d). The association with prior history of airways disease remained significant on multivariate analysis, p < 0.05.
 B. Cushen et al. / Respiratory Medicine 115 (2016) 13e19 17

Table 4
e Risk factors for airflow obstruction in the never smoking cohort.

FEV1/FVC < 0.70 (n ¼ 29) FEV1/FVC  0.70 (n ¼ 198) p value

Age, years
Median (IQR) 52 (47e58) 55 (47e62) 0.26
Sex
Male 22 (76) 144 (73) 0.72
Female 7 (24) 54 (27)
Level of Education
Primary 1 (3) 21 (11) 0.83
Second level 18 (62) 99 (50)
Tertiary 10 (35) 77 (39)
BMI (kg/m2)
Mean ± SD 27.8 ± 4.02 28.9 ± 3.71 0.15
Prior airway disease 13 (45) 22 (11) <0.01
Hayfever/Allergies 9 (31) 30 (15) 0.03
Farm Type
Livestock 26 (90) 159 (80) 0.77
Arable 1 (3) 5 (3) 0.25
Mixed 2 (7) 33 (17) 0.18
Feed used
Silage 22 (76) 168 (85) 0.22
Haylage 4 (14) 14 (7) 0.22
Hay 12 (41) 77 (39) 0.79
Dry Grain 6 (21) 74 (37) 0.08
Chemical Spraying 18 (62) 129 (65) 0.71
Use of personal protective equipment 14 (48) 92 (46) 0.77
Fossil fuel exposure
Any 22 (76) 174 (89) 0.08
Turf only 2 (7) 25 (13) 0.21
Wood only 12 (41) 77 (39) 0.59
Turf þ wood 8 (28) 70 (35) 0.41

Univariate logistic regression analysis, comparing never smokers with obstruction, FEV1/FVC < 0.70, to never smokers with normal spirometry, FEV1/FVC  0.70.
Data presented as count (%) unless otherwise stated; Significant p values (p < 0.05) highlighted in bold.

Fig. 2. FEV1 (%) and Feed Used in a Non-Smoking Farming Population. There was no significant relationship between the type of feed used and % predicted FEV1 measurements on
univariate or multivariate analysis, following adjustment for age, sex and smoking status.
18 B. Cushen et al. / Respiratory Medicine 115 (2016) 13e19
studies [9,10,20e23]. The high prevalence of respiratory symptoms
associated with farming has been attributed to long-standing ex-
posures to airway irritants and inflammatory response triggers
such as endotoxins, fungal spores and inorganic dusts as well as to
biological fumes [8]. Eduard, in his study of farmyard exposures in
Norwegian farmers, found much higher levels of these agents on
livestock farms, particularly swine and poultry, as compared with
crop farms. In our study the majority were livestock farmers with
only 4% involved in arable farming alone. This may account for the
very high percentage of reported respiratory symptoms.
Farmers have been identified as a group with significant airway
disease and airflow obstruction and our findings mirror those
found in other studies in this area [6e8]. Twelve per cent of all
study participants had obstructive physiology and, of those,
remarkably 66% were never smokers. The overall prevalence of
obstructive airway disease in our total never smoking cohort was
13% (FEV1/FVC < 0.70) and 7.5% using LLN criteria, which was
similar to the prevalence observed amongst our smokers, 10% and
8% respectively. Although high levels of airflow obstruction have
been reported in the general population relatively few of the sub-
jects with obstructive physiology are non-smokers in these studies.
For example, in the large cohort studies, BOLD [24] and NHANES III
[25], using unselected populations, up to 12% percent of partici-
pants were diagnosed with obstructive disease but non-smokers
comprised only 28% and 23% respectively of those patients with
obstructive physiology. A recent Canadian study of more than 5000
participants found a COPD prevalence in the never-smoking cohort
(n ¼ 2295) using LLN criteria, of 6.4% [26]. Other farming studies in
this area report levels of chronic bronchitis and COPD ranging from
3 to 30% [6e8,22,27]. It is likely therefore, that farmers have non-
smoking related risk factors for obstructive lung disease.
The reported prevalence of airflow obstruction in farmers is
highly variable. This may reflect differences in the definition used
to define obstruction. The prevalence of airflow obstruction in
Eduard's study [8], for example, excluding those with a prior
diagnosis of Asthma, was 13.5% using pre-bronchodilator FEV1/FVC
<5% LLN criteria. However, prevalence was almost half (6.9%) when
the GOLD fixed ratio (FEV1/FVC < 0.70) was applied. Conversely,
similar to other studies [24,29], we found that the prevalence
decreased when LLN criteria was applied. We chose to use the
GOLD recommended fixed FEV1/FVC < 0.70 to define obstruction.
This definition has also been used in large population-based studies
of never-smokers [27,28]. The use of fixed ratio over the LLN criteria
to define obstruction carries the risk of overdiagnosis of obstruction
amongst older age groups [28]. Interestingly, we found that the
median age of those who had obstruction by fixed ratio but not by
LLN criteria was not significantly different from those with
obstruction defined by both criteria, suggesting that older age is not
an explanation for the high proportion of obstruction in this study.
The use of LLN exclusively can lead to underdiagnosis [29] and a
missed opportunity to treat those with early stage COPD. Previous
prospective studies have found an increased risk of death and
COPD-hospitalisations in those who met the criteria for obstruction
based on a fixed ratio cut-off but not by LLN as compared to those
normal spirometry [30,31]. This is significant when one considers
that 50% of those who were deemed to have airflow obstruction by
GOLD criteria but not by LLN in our study reported symptoms
suggestive of airways disease.
Although farmers' lung was previously a common diagnosis, in
our study, as is typical of other recent studies [4,5], this was a rare
pre-existing diagnosis. In fact it is felt that airways disease is a more
likely explanation for respiratory symptoms in the farming popu-
lation. In our study, the non-smoking cohort with obstruction was
significantly more likely to describe dyspnoea and wheeze, and had
a higher, though not significant, rate of upper respiratory tract
symptoms suggestive of rhinitis. In addition there was a significant
association with previous diagnosis of airways disease, and self-
reported atopic disease. Based on these associations it seems
likely that airways disease, such as asthma and bronchitis, is the
most likely explanation for the obstruction in this group of subjects.
Obesity is a significant risk factor for asthma and obstructive
lung disease. We found a strikingly high incidence of being over-
weight and obese in the entire study population. This observation
is concerning given the recently described higher than average
mortality and prevalence of cardiovascular disease in Irish farmers
[14].
We could not detect significant environmental exposures to
explain the high rate of airflow obstruction in our study. This is not
surprising as we studied a mixed group of farmers using self-
reported farming practices, and the focus of our study was pri-
marily to identify the prevalence of respiratory disease rather than
identify the causative agent. As noted above, livestock farming is
associated with higher rates of respiratory symptoms and lower
FEV1 as compared to crop farms. Given that 96% of farmers in our
study kept livestock we cannot exclude an association between
obstruction and farm type. We found no significant difference in
exposures between those never smokers with obstruction and
those with normal spirometry. Previous studies note associations
between low FEV1 and ammonia, hydrogen sulphide and inorganic
dust exposure [8] but these could not be assessed in the limits of
this study. Working predominantly indoors has also been shown to
be associated with increased levels of airway disease in farmers,
with one study reporting a prevalence of 17% [32]. It was difficult to
accurately and reliably capture the percentage of time farmers
spent indoors in this cross-sectional study. A prospective study
which allows for measurement of these exposures is needed to
accurately evaluate their role in the pathogenesis of lung disease in
this cohort.
There is increased recognition of the influence occupational
exposures have on the clinical course of COPD. Those with COPD
who experience continued occupational exposure have been
shown to have poorer quality of life, poorer functional status and
have an increased exacerbation risk [33]. We did not assess these
issues but those with airflow obstruction were more likely to
describe chronic respiratory symptoms inferring associated
morbidity.
There are several limitations in our study design. The study
cohort was a self referred cohort attending a large farming event.
While we may have selected those with respiratory complaints, it is
likely that the cohort is representative of the general farming
community given our position on the main thoroughfare of the
meeting and the volume of subjects studied. However, it is possible
that those with severe disease and therefore unfit to attend the
exhibition may have been excluded. Regardless, when compared to
other population studies conducted in a similar manner, the key
findings of higher chronic respiratory symptoms and in particular
non-smoking related airflow obstruction remains valid. We did not
measure post-bronchodilator spirometry and as such we report our
findings as airflow obstruction rather than COPD. A post-
bronchodilator study would be helpful in future investigations.
5. Conclusion
In this study we have identified that Irish farmers are a group
with high levels of non-smoking related obstructive lung disease
and with a significant degree of respiratory symptoms. Given the
lack of data on the respiratory health of Irish farmers, one can only
assume that this high burden of lung disease is largely unrecog-
nised. These results need to be considered in the context of high
age-adjusted mortality rates in Irish farmers. Further studies are
 B. Cushen et al. / Respiratory Medicine 115 (2016) 13e19
needed to more accurately determine the specific workplace ex-
posures to account for our findings and to facilitate the develop-
ment of preventative strategies.
Competing interests
The authors have no competing interests to declare.
Funding
No financial contributions received.
Acknowledgments
The authors would like to thank Teagasc Ireland and the Health
and Safety Authority for their support in conducting this study by
providing us with a location at the National Ploughing Champion-
ship at which we could carry out the study.
References
[1] E. Diaz-Guzman, D.M. Mannino, Epidemiology and prevalence of chronic
obstructive pulmonary disease, Clinics Chest Med. 35 (1) (2014) 7e16, http://
dx.doi.org/10.1016/j.ccm.2013.10.002.
[2] R.J. Halbert, J.L. Natoli, A. Gano, et al., Global burden of COPD: systematic re-
view and meta-analysis, Eur. Respir. J. 28 (3) (2006) 523e532.
[3] Respiratory health hazards in agriculture, Am. J. Respir. Crit. Care Med. 158 (5
Pt 2) (1998 Nov) S1eS76.
[4] A. Arya, K. Roychoudhury, C.P. Bredin, Farmer's lung is now in decline, Ir. Med.
J. 99 (7) (2006 Jul-Aug) 203e205.
[5] J.V. Pether, F.B. Greatorex, Farmer's lung disease in Somerset, Br. J. Ind. Med.
33 (4) (1976 Nov) 265e268.
[6] J.C. Dalphin, D. Debieuvre, D. Pernet, et al., Prevalence and risk factors for
chronic bronchitis and farmer's lung in French dairy farmers, Br. J. Ind. Med.
50 (10) (1993) 941e944.
[7] A. Vergnenegre, X. D'arco, B. Melloni, et al., Work related distal airway
obstruction in an agricultural population, Occup. Environ. Med. 52 (9) (1995)
581e586.
[8] W. Eduard, N. Pearce, J. Douwes, Chronic bronchitis, COPD, and lung function
in farmers: the role of biological agents, Chest 136 (3) (2009) 716e725, http://
dx.doi.org/10.1378/chest.08-2192.
[9] E. Monso
, R. Magarolas, K. Radon, et al., Respiratory symptoms of obstructive
lung disease in European crop farmers, Am. J. Respir. Crit. Care Med. 162 (4 Pt
1) (2000) 1246e1250.
[10] J.A. Hoppin, D.M. Umbach, S. Long, et al., Respiratory disease in United States
farmers, Occup. Environ. Med. 71 (7) (2014) 484e491, http://dx.doi.org/
10.1136/oemed-2013-101983.
[11] CENSUS 2011. www.cso.ie.
[12] N. Thelin, S. Holmberg, P. Nettelbladt, et al., Mortality and morbidity among
farmers, nonfarming rural men, and urban referents: a prospective
population-based study, Int. J. Occup. Environ. Health 15 (1) (2009 JaneMar)
21e28.
[13] J.K. Waggoner, G.J. Kullman, P.K. Henneberger, et al., Mortality in the agri-
cultural health study, 1993-2007, Am. J. Epidemiol. 173 (1) (2011 Jan 1)
19
71e83, http://dx.doi.org/10.1093/aje/kwq323.
[14] B. Smyth, D.S. Evans, A. Kelly, et al., The farming population in Ireland:
mortality trends during the 'Celtic Tiger' years, Eur. J. Public Health 23 (1)
(2013) 50e55, http://dx.doi.org/10.1093/eurpub/cks017.
[15] www.goldcopd.org.
[16] J.L. Hankinson, J.R. Odencrantz, K.B. Fedan, Spirometric reference values from
a sample of the general US population, Am. J. Respir. Crit. Care Med. 159 (1)
(1999 Jan 1) 179e187.
[17] M.R. Miller, J. Hankinson, V. Brusasco, et al., ATS/ERS Task Force. Stand-
ardisation of spirometry, Eur. Respir. J. 26 (2) (2005 Aug) 319e338.
[18] www.hse.ie: Smoking in Ireland 2014- Synopsis of Key Patterns. National
Tobacco Control Office, Health Service Executive.
[19] M.C. Alavanja, D.P. Sandler, S.B. McMaster, et al., The Agricultural Health
Study, Environ. Health Perspect. 104 (4) (1996 Apr) 362e369.
[20] J.P. Hayes, J. Rooney, The prevalence of respiratory symptoms among mush-
room workers in Ireland, Occup. Med. Lond. 64 (7) (2014 Oct) 533e538,
http://dx.doi.org/10.1093/occmed/kqu110.
[21] A. Guillien, M. Puyraveau, T. Soumagne, S. Guillot, F. Rannou, D. Marquette,
P. Berger, S. Jouneau, E. Monnet, F. Mauny, J.J. Laplante, Prevalence and risk
factors for COPD in farmers: a cross-sectional controlled study, Eur. Respir. J.
47 (1) (2016 Jan 1) 95e103, http://dx.doi.org/10.1183/13993003.00153-2015.
[22] O. Omland, Exposure and respiratory health in farming in temperate zones-a
review of the literature, Ann. Agric. Environ. Med. 9 (2) (2002 Jan 1) 119e136.
[23] K. Radon, E. Monso, C. Weber, et al., Prevalence and risk factors for airway
diseases in farmersesummary of results of the European Farmers' project,
Ann. Agric. Environ. Med. 9 (2) (2002) 207e213.
[24] B. Lamprecht, M.A. McBurnie, W.M. Vollmer, et al., BOLD Collaborative
Research Group, COPD in never smokers: results from the population-based
burden of obstructive lung disease study, Chest 139 (4) (2011 Apr)
752e763, http://dx.doi.org/10.1378/chest.10-1253.
[25] B.R. Celli, R.J. Halbert, R.J. Nordyke, B. Schau, Airway obstruction in never
smokers: results from the third national health and nutrition examination
survey, Am. J. Med. 118 (12) (2005) 1364e1372.
[26] W.C. Tan, D.D. Sin, J. Bourbeau, et al., Characteristics of COPD in never-
smokers and ever-smokers in the general population: results from the Can
COLD study, Thorax 70 (9) (2015 Sep 1) 822e829.
[27] B. Lamprecht, L. Schirnhofer, B. Kaiser, M. Studnicka, A.S. Buist, Farming and
the prevalence of non-reversible airways obstructiondresults from a
population-based study, Am. J. Ind. Med. 50 (6) (2007 Jun 1) 421e426.
[28] C.A. Vaz Fragoso, J. Concato, G. McAvay, P.H. Van Ness, C.L. Rochester,
H.K. Yaggi, T.M. Gill, The ratio of FEV1 to FVC as a basis for establishing chronic
obstructive pulmonary disease, Am. J. Respir. Crit. Care Med. 181 (5) (2010
Mar 1) 446e451, http://dx.doi.org/10.1164/rccm.200909-1366OC.
[29] F.A. Hoesein, P. Zanen, J.W. Lammers, Lower limit of normal or FEV 1/FVC<
0.70 in diagnosing COPD: an evidence-based review, Respir. Med. 105 (6)
(2011 Jun 30) 907e915, http://dx.doi.org/10.1016/j.rmed.2011.01.008.
[30] D.M. Mannino, A.S. Buist, W.M. Vollmer, Chronic obstructive pulmonary dis-
ease in the older adult: what defines abnormal lung function? Thorax 62 (3)
(2007 Mar 1) 237e241.
[31] S. Zaigham, P. Wollmer, G. Engstro €m, Lung function, forced expiratory volume
in 1 s decline and COPD hospitalisations over 44 years of follow-up, Eur.
Respir. J. (2015 Dec 3), http://dx.doi.org/10.1183/13993003.01575-2015. ERJ-
01575.
[32] E. Monso
, E. Riu, K. Radon, et al., Chronic obstructive pulmonary disease in
never-smoking animal farmers working inside confinement buildings, Am. J.
Ind. Med. 46 (4) (2004 Oct) 357e362.
[33] L.M. Paulin, G.B. Diette, P.D. Blanc, et al., SPIROMICS Research Group, Occu-
pational exposures are associated with worse morbidity in patients with
COPD, Am. J. Respir. Crit. Care Med. 191 (5) (2015 Mar 1) 557e565, http://
dx.doi.org/10.1164/rccm.201408-1407OC.