AMERICAN JOURNAL OF INDUSTRIAL MEDICINE 36:444±449 (1999)

Work-Related Respiratory Disorders and

Farming Characteristics Among Cattle
Farmers in Northern Germany

Katja Radon, MCE,1,4 Ulrike Opravil, MD,2 Jo

Èrg Hartung, PhD,3
Dieter Szadkowski, PhD, and Dennis Nowak, PhD,1,4
4

Background The aim of this cross-sectional study was to investigate the prevalence of
respiratory symptoms among Northern German farmers in relation to cattle farming
characteristics.
Methods 1735 farmers were visited on their farms and were interviewed using a
standardized questionnaire on work-related respiratory symptoms and farming details.
Results 84.6% of the farmers were cattle farmers. The prevalence of work-related
respiratory symptoms was 40.3%. In a multiple logistic regression model adjusting for
response rate, age, gender, and smoking habits, work-related respiratory symptoms were
shown to be signi®cantly associated with the ventilation of the cattle house (OR
(ventilation via the wall): 0.57), feeding management (OR (feeding once daily): 0.53),
and plant crop (OR: 0.75). Farmers, living inland, showed a signi®cant higher prevalence
of work-related respiratory symptoms (OR: 1.34).
Conclusions The use of ventilation via the wall might be recommended for new cattle
houses in regions with warm winters. Am. J. Ind. Med. 36:444±449, 1999. ß 1999
Wiley-Liss, Inc.

KEY WORDS: cattle farming; respiratory symptoms; farming characteristics;

prevention; cross-sectional study

INTRODUCTION 1995]. Additionally, epidemiological studies have shown a

Biological material like grain, animal fodder, and
bedding material are potential substrates for the growth of
bacteria, molds, and mites in farming environments.
Therefore, immunologically potent substances such as
allergens, endotoxins and glucans have been found in
agricultural dust [Attwood et al., 1987; Donham et al.,
1
Institute of Occupational and Environmental Medicine, Ludwig-Maximilians-University,
Munich, Germany
2
Centre of Work-Related Diseases, Brunsbuettel, Germany
3
Institute of Animal Hygiene and Animal Protection, School of Veterinary Medicine
Hanover, Germany
4
Institute of Occupational Medicine, University of Hamburg, Hamburg, Germany
This study is part of a thesis of U.O. for a medical doctorate.
*Correspondence to: K. Radon, D. Nowak Institute of Occupational and Environmental
Medicine, Klinikum Innenstadt, Ludwig-Maximilians-University Munich, Ziemssenstr. 1, D -
80336 Munich, Federal Republic of Germany. E-mail: K.Radon@lrz.uni-muenchen.de
Accepted 23 March1999
high frequency of symptoms of chronic bronchitis and
asthma in farmers [Iversen et al., 1988; Terho, 1990;
Melbostad et al., 1997; Melbostad et al., 1998]. Further-
more, Vogelzang et al. [1996] have studied pig farm
characteristics associated with the development of respira-
tory morbidity. Many of these studies focused on grain
workers [Gimenez et al., 1995; Post et al., 1998] or swine
farmers [Attwood et al. 1987; Crook et al., 1991; Donham et
al., 1995; Vogelzang et al., 1996; Nowak, 1998]. Cattle
farmers, keeping livestock, veal, or dairy, have been less
extensively studied although it is known that cattle farmers
have a higher risk of respiratory morbidity than controls
[Choudat et al., 1994] or farmers without livestock
[Melbostad et al., 1997, 1998].
Houses for cattle in Northern Germany are only used in
winter. In general, there are three different kinds of
ventilation in cattle houses without automatic ventilation:
ventilation through windows and doors, eave-ridge ventila-

ß 1999 Wiley-Liss, Inc.

 Prevention of Respiratory Symptoms in Cattle Farmers
tion and wall ventilation. Wall ventilation means either no
walls but draught-excluder/porch nets (Windfangnetze)
around the stable, three walls and one open side, or gaps
between the planks. Ventilation by open or partly open walls
is typical only for regions in Germany with warm winter.
The aim of the present study was to determine cattle
farming characteristics which correlated with the develop-
ment of work-related respiratory symptoms and ultimately
to minimize the risk of respiratory morbidity in cattle
farmers in Northern Germany. Therefore, 1735 farmers
including 1468 cattle farmers in Schleswig-Holstein (North-
ern Germany) participated in this questionnaire survey
focusing on work-related respiratory symptoms, personal
and farming characteristics.
SUBJECTS AND METHODS
Study Population and Questionnaire
A letter requesting participation was mailed to 1247
farms with a minimum size of 74.1 acres in 221
municipalities in 4 districts of Schleswig-Holstein (North-
ern Germany) by the local professional farm organization.
This farm size was chosen to favor selection of full-time
farmers. Most of the farmers were farm owners and their
family members (spouse, parents, children). Additionally,
farming inspectors who regularly visit the farms for other
reasons asked the farmers to participate. Therefore, a total
1735 farmers on 834 farms (66.9%) who expressed interest
in the study were visited by 7 trained interviewers. The
medical part of the questionnaire contained questions on
chronic respiratory symptoms, work-related respiratory
symptoms, and the time of onset of the complaints.
Questions on personal characteristics included age, smoking
habits, work history, and full-time or part-time work. A
questionnaire on characteristics and methods of farm
operation had to be completed for every animal house.
According to the differences in type of farming three
different questionnaires were used for cattle barns as well as
sheep and horse houses, swine con®nement buildings, and
poultry houses.
Analysis
Participants were classi®ed as cases (n ˆ 698) when
they showed one or more of ®ve work-related respiratory
symptoms (shortness of breath, cough without phlegm,
cough with phlegm, wheezing, nasal irritation). The
questionnaire contained items on various characteristics of
the animal houses. Due to the high numbers of animal
houses (up to 8 cattle barns, 9 swine con®nement buildings,
and 2 poultry houses) per farm, a mean value for each item
was calculated for every farmer, e.g., if half of the cattle
houses at one farm had slatted ¯oors and the other half was
445
without slatted ¯oors the value for slatted ¯oor was
calculated to 0.5.
Computations were completed with the aid of a
statistical package for personal computers (Statistica 5.1,
StatSoft, Tulsa, USA). Standard methods of statistical
analysis were used, such as cross-tabulation, t-test for
dependent variables, and multiple logistic regression
analysis. All statistical tests were done at  ˆ 0:05
signi®cance level.
RESULTS
Study Population
Response rates varied across municipalities from a
minimum of 4.4% up to a maximum of 100%. 83.6% of the
respondents were full-time farmers. 636 of the 1735 farmers
were female (36.7%). The mean age  SD of the farmers
was 45.0  13.3 years. 20.2% were currently smoking while
14.1% had stopped smoking at least 1 month ago. 5.9% of
the farmers used respiratory protection at work.
Most of the farmers had livestock and plant crop
(76.2%) while 118 were only cultivating plants (6.8%).
1468 farmers were cattle farmers (84.6%), 54.9% kept cattle
only (n ˆ 952). While 1062 of the farmers lived near the
North Sea or the Baltic Sea (61.2%), 673 farmers lived on
farms that were located inland.
Respiratory Symptoms
As shown in Table I, 40.3% of the farmers reported at
least one work-related respiratory symptom (breathlessness,
cough without phlegm, cough with phlegm, wheezing, nasal
irritation). The prevalence of work-related respiratory
symptoms increased with age as shown by the Odds Ratio
(OR) [95% CI]: 1.01 [1.00±1.02] per year of age. The
prevalence of work-related respiratory symptoms did not
differ signi®cantly between current smokers, ex-smokers
and never smokers while women complained less frequently
of work-related respiratory symptoms (OR [95% CI]: 0.81
TABLE I. Prevalence of Work-related Respiratory Symptoms Among 1735 Farmers in
Northern Germany
Do you have during your work one or Prevalence
more of the following complaints [%]
Breathlessness 13.0
Cough without phlegm 17.4
Cough with phlegm 11.3
Wheezing 6.8
Nasal irritation 23.0
One or more of these symptoms 40.3
Flu-like illness 7.6
446 Radon et al.
FIGURE 1. Time onset of the complaints in the ten years prior to the study (*P < 0:05; ***P < 0:0001 compared to10 years prior to
the study).
[0.66±0.99]). Full-time farmers did not report more work-
related respiratory symptoms than part-time farmers (OR
[95% CI]: 1.19 [0.90±1.60]).
There was no signi®cant difference in the prevalence of
work-related respiratory symptoms between farmers produ-
cing cattle only (42.0%), farmers producing cattle and other
animals (39.5%), and farmers without cattle production
(35.8%; P ˆ 0:18). The frequency of work-related respira-
tory symptoms increased signi®cantly within the ten years
prior to the study (Fig. 1).
Farm Characteristics
Due to the fact that cattle farming is most common in
this part of Germany, the evaluation of farm characteristics
was focused on cattle houses. Table II summarizes farm
characteristics explored for any association with work-
related respiratory symptoms with respect to cattle produc-
tion.
80.6% of the farmers were producing cattle and crops,
primarily animal feed (grass (41.2%), corn (31.2%) and/or
wheat (27.4%). Most of the farmers had mixed production
of cattle, veal and dairy (64.8%). The median number of
cattle houses per farm was 2 (range 0±8); 57 of the cattle
farmers (3.9%) reported open air farming only. Cattle
farmers of the study group worked in animal houses with a
size range from 30 to 2000 m2. The number of cattle per
farm was only available for 501 farmers. The median
number of animals in these farms was 100 (range: 12±514).
Most of the barns had no provision for heating (99.2%).
Most cattle houses were naturally ventilated (54.6% with
only natural ventilation). Usually, animals were fed by hand
(58.9% hand feeding in all barns) and more than once daily
(91.9%). Food combinations of hay, milk, and silage were
mostly used depending on the season. Typically, cleaning
and disinfection intervals were more than one year (95.4%).
Respiratory Symptoms and Farming
Characteristics
In Table II, associations between cattle barn character-
istics and the prevalence of one or more work-related
respiratory symptoms are given after adjusting for response
rate in the municipality, age, gender, and smoking status.
For the logistic regression model the area and volume of the
animal houses had to be log-transformed in order to achieve
normal distribution due to the wide variety. The reporting of
any work-related respiratory symptom was signi®cantly
associated with growing crops, the log-transformed area of
the barns, heating, air outlet via the wall, feeding once daily,
and feeding root crop or corn silage. A weak but not signi®-
cant association was also seen for air inlet in the eaves.
As various of these characteristics may be strongly
interrelated, associations were analyzed stratifying for
relevant variables (Table III). The strong relationship
between air outlet via the wall and work-related respiratory
symptoms was stronger after strati®cation while the
association with crop farming or feeding once daily and
work-related respiratory symptoms was unchanged. A weak
relationship was seen for heating and the log-transformed
size of the cattle house. Feeding of root crop, corn silage and
air inlet in the eaves did not reach signi®cance in the
multivariate model.
For the 501 farms with given animal numbers the
median volume (38.8 m3; range: 5.2±331.7 m3) and area
(9.4 m2; range: 1.4±69.1 m2) per animal was calculated. For
 Prevention of Respiratory Symptoms in Cattle Farmers 447

TABLE II. Distribution of Farm Characteristics and Univariate Associations Between Farm Characteristics and the Prevalence of One or
More Work-related Respiratory Symptoms. Corrected for Response Rate, Age, Gender, and Smoking Status

No. of farmers with

characteristic of one
Characteristic Median cattle house minimuma Odds Ratio 95% CI

Plant crop 285 0.71b 0.54^0.92

Number of hours worked in the barn per day 2.5 ^ ^ ^
per hour 0.99 0.93^1.06
Median size of the barns
per log (m2) of area 400.0 ^ 0.59 0.38^0.92
per log (m3) of volume 1646.4 ^ 0.79 0.57^1.10
Mean number of cattlec 100 ^
per log (animal) 0.60 0.33^1.10
Number of barns 2.0
per barn ^ 1.01 0.92^1.11
Temperature
Heating 0.0 11 9.66 1.21^77.85
Ventilation
Natural 1.0 1259 0.86 0.63^1.18
Location of air inlet
Eaves 0.0 397 0.69 0.47^1.01
Location of air outlet
Wall 0.3 785 0.66 0.51^0.86
Roof 0.8 507 1.04 0.79^1.36
Floor
Solid, concrete floor 0.7 1219 0.91 0.67^1.23
Feeding system
Automatic 0.0 579 0.94 0.69^1.30
Feeding once daily 0.0 114 0.47 0.25^0.87
Food (combinations possible)
Root crop 0.0 128 1.74 1.10^2.74
Grain 0.0 334 1.12 0.84^1.50
Hay 0.5 1076 1.11 0.84^1.50
Milk 0.3 972 0.81 0.58^1.11
Silage
Grass 1.0 1294 0.92 0.64^1.32
Corn 0.5 783 0.74 0.58^0.93
Use of bedding
Straw 1.0 1371 0.80 0.53^1.22
Cleaning of the cattle houses
Yearly 1.0 1339 0.84 0.55^1.29
Living Inland ^ 587 1.40 1.13^1.74

a
Based on 1406 cattle farmers
b
Not adjusted for smoking
c
Based on 501cattle farmers

symptomatic cattle farmers (n ˆ 186) the median volume
was slightly but not signi®cantly higher than for farmers
without symptoms (41.8 vs. 36.0 m3; PMann±Whitney U-Test ˆ
0.17). Regarding the area per cattle no difference could be
observed (9.7 vs. 9.2 m2; PMann±Whitney U-Test ˆ 0.74). The
same results were seen in a multiple regression model after
adjusting for response rate, age, gender, and smoking habits
(OR per log m3/cattle: 1.48 [0.85±2.61]; OR per log m2/
cattle: 1.26 [0.68±2.33]).
Regional Differences
Regarding work-related respiratory symptoms, farmers
working on farms near the sea had a signi®cantly lower
448 Radon et al.
TABLE III. Multivariate Associations Between Farm Characteristics and the Prevalence
of One or More Work-related Respiratory Symptoms. Corrected for Response Rate, Age, Gen-
der, and Smoking Status.
OR [95% CI]a
Plant crop 0.75 [0.56^1.00]
Size of the cattle barn (log) 0.66 [0.41^1.05]
Heating 7.08 [0.85^58.85]
Feeding once daily 0.53 [0.28^1.00]
Feeding of root crop 1.57 [0.97^2.54]
Feeding of corn silage 0.84 [0.64^1.08]
Air inlet: eaves 0.74 [0.49^1.11]
Air outlet: wall 0.57 [0.43^0.75]
a
Based on 1402 cattle farmers
prevalence of work-related respiratory symptoms (37.2%
vs. 44.9%, respectively; P ˆ 0:002). After adjusting for
response rate, age, smoking, gender, plant crop, feeding
once daily, and outlet in the wall this difference remained a
signi®cant predictor of work-related respiratory symptoms
(OR [95 % CI]: 1.34 [1.07±1.67]).
DISCUSSION
This study has shown that the main factors associated
with the development of work-related symptoms in North-
ern German cattle farmers are (1) ventilation not via the
wall, (2) feeding more frequently than once daily, (3) no
plant crop, and (4) living inland. Additionally, farmers with
a lower volume of air per animal had a tendency towards
more work-related respiratory symptoms and heating might
also be a risk factor for the development of this symptoms.
To our knowledge, no studies have reported on the link
between speci®c characteristics of animal farming and
respiratory symptoms except two studies of Bongers et al.
[1987] and Vogelzang et al. [1996, 1997]. These studies
focused on swine con®nement units and, therefore, are not
directly comparable to our investigation. Farming and farm
characteristics remain stable for a long duration and change
only slowly due to modernization. Therefore, using a
questionnaire on these characteristics re¯ects exposure
during recent years when the symptoms asked for have
developed. Likewise, direct exposure assessments of dust,
endotoxin, glucan, allergen assessment vary from day-to-
day and are not necessarily representative of long-term
exposure.
Farming characteristics were based on the mean value
for all cattle barns on each farm despite the kind of the barn
(dairy, veal, cattle) or the time the farmer spent in this
animal house. In addition, the type of operation in each
animal house was not taken into account. Therefore, some
misclassi®cation might be expected in this study and could
lead to an underestimation of effects. However, this kind of
classi®cation seemed to be more meaningful than focusing
on the cattle house, where the farmer spent the most of his or
her time. Although all farmers of each farm participated in
this study, it is likely that women work more frequently
inside the animal houses (milking, feeding) while men are
expected to work more often with plant crop. This was taken
into account in the logistic regression model by adjusting for
gender.
The strong association between age and prevalence of
work-related respiratory symptoms found in several studies
[e.g., Iversen et al., 1988; Vogelzang, 1996; Melbostad et al.,
1998] indicates a dose-response relationship. Due to the fact
that farm owners stay in their jobs for many years [Thelin
and HoÈglund; 1994], age and number of years in farming
were strongly correlated.
The most common work-related respiratory symptom
was nasal irritation (23.0%). Up to now it is not well
understood whether nasal irritation may precede severe
respiratory morbidity. The overall prevalence of work-
related respiratory symptoms in the farming population
under study is high even in farmers with mainly cattle
production. Therefore, it is of special interest to determine
risk factors for the development of respiratory morbidity
among these cattle farmers in order to get preventive farm
characteristics.
The most pronounced and remarkable effect was the
lower prevalence of work-related respiratory symptoms
among farmers with a ventilation outlet in the wall. Most of
the farmers with a ventilation outlet in the wall also had a
ventilation inlet in the wall (98.4%) as described above. The
high air exchange in this type of animal houses probably
leads to lower concentration of dusts. Additionally, the
temperatures in these cattle barns are lower especially in
winter.
A trend for an effect of temperature in cattle houses on
respiratory morbidity was also seen in respect to heating of
the cattle barns. Barns without heating are expected to have
a lower temperature in winter than animal houses with
heating even if the animal house is closed on all sides. In our
analysis, heating failed the level of signi®cance in the
multivariate analysis but this might be mainly due to the fact
that heating was used on only 11 farms. Therefore, it may be
speculated that a lower concentration of mites, endotoxins
and glucans in cattle houses with lower average tempera-
tures in winter lead to a lower prevalence of respiratory
morbidity among farmers.
After adjusting for response rate, age, smoking, gender,
plant crop, feeding once daily, and air outlet in the wall,
fewer symptoms were also found among farmers living
closer to the coast than farmers living inland. It does not
seem likely that farming characteristics are the reasons for
this ®nding. Thus, it might be an indication of a dose-
response effect because the closer the farms are located to
the sea the higher the air velocity may be and therefore, a
 Prevention of Respiratory Symptoms in Cattle Farmers
higher air exchange rate within open/partly open animal
houses reduces the dust concentration.
Farmers who also produce plants complained signi®-
cantly less often of respiratory symptoms. Evidence of the
prevalence of work-related respiratory symptoms among
farmers with plant crop is con¯icting. Vohlonen et al. [1987]
found that farmers with livestock production had more
symptoms of chronic bronchitis than farmers in grain
production while Iversen et al. [1988] found that farmers
with no animal production resembled dairy farmers with
respect to the prevalence of most symptoms. To our
knowledge, the combined effect of animal production and
plant crop on respiratory health has not yet been studied.
Since farmers with and without plant crops spend similar
amounts of time in the cattle barn (2.7 vs. 2.6 hr/day;
P ˆ 0:30), the lower rate of respiratory symptoms suggests
a ``protective'' effect for those cultivating plants.
Feeding of the animals once daily was signi®cantly
associated with a lower prevalence of respiratory symptoms.
Contrary, automated feeding was not associated with a
lower prevalence of work-related respiratory symptoms.
Vogelzang et al. [1997] even found a higher risk of mild
bronchial hyperresponsiveness and automated dry feeding.
Bongers et al. [1987] showed a slightly lower lung function
for farmers with automated feeding systems. These
observations are concordant with the fact that automated
dry feeding is a source of high levels of dust-exposure.
Nevertheless, this is also supported by our observation of a
relevant increase in the prevalence of work-related respira-
tory symptoms in the ten years prior to the study. Besides the
age-related rise in symptoms, this may be partly contribu-
table to an increased use of automated dry feeding.
The described effects of age of the farmer, ventilation,
volume, and heating of the cattle barn as well as the kind of
farming, on respiratory symptoms seem to overcome the
hypothesized dose-response effect of hours per day spent in
a cattle barn. Additionally, one has to bear in mind that
hours per day in the cattle barn re¯ects average values over
the year. Many of the farmers work inside the cattle barn
only in winter.
In summary, this study has shown a preventive effect of
modern, big cattle barns with natural ventilation via the wall
which is probably due to a higher air exchange rate.
Furthermore, lower temperature has to be recommended for
cattle houses. Further research is necessary to support these
®ndings by data of measured exposure.
ACKNOWLEDGMENT
This study was supported by the Bundesministerium
fuÈr Arbeit und Sozialordnung, the Schleswig-Holsteinische
Landwirtschaftliche Berufsgenossenschaft, and the Eur-
opean Union. The authors thank the farmers for their
449
participation and Brit Walter, Gisela Reinke, Wiebke Tesch,
Olaf Opravil, Sven Mathiessen, Lorenz Groû, Dr. Christine
Kallenberg, and Detlef Glomm for their assistance.
REFERENCES
Attwood P, Brouwer R, Ruigewaard P, Versloot P, De Wit R, Heederik
D, Boleij JSM. 1987. A study of the relationship between airborne
contaminants and environmental factors in Dutch swine con®nement
buildings. Am Ind Hyg Assoc J 48:745±751.
Bongers P, Houthuijs D, Remijn B, Brouwer R, Biersteker K. 1987.
Lung function and respiratory symptoms in pig farmers. Br J Ind Med
44:819±823.
Choudat D, Goehen M, Korobaeff M, Boulet A, Dewitte JD, Martin
MH. 1994. Respiratory symptoms and bronchial reactivity among pig
and dairy farmers. Scand J Work Environ Health 20:48±54.
Crook B, Robertson JF, Travers Glass SA, Botheroyd EM, Lacey J,
Topping MD. 1991. Airborne dust, ammonia, microorganisms, and
antigens in pig con®nement houses and the respiratory health of
exposed farm workers. Am Ind Hyg Assoc J 52:271±279.
Donham K, Reynolds S, Whitten P, Merchant J, Burmeister L,
Popendorf J. 1995. Respiratory dysfunction in swine production
facility workers: dose-response relationships of environmental
exposures and pulmonary functions. Am J Ind Med 27:405±418.
Gimenez C, Fouad K, Choudat D, Laureillard J, Bouscaillou P, Leib E.
1995. Chronic and acute respiratory effects among grain mill workers.
Int Arch Occup Environ Health 67:311±315.
Iversen M, Dahl R, Korsgaard J, Hallas T, Jensen EJ. 1988. Respiratory
symptoms in Danish farmers: an epidemiological study of risk factors.
Thorax 43:872±877.
Melbostad E, Eduard W, Magnus P. 1997. Chronic bronchitis in
farmers. Scand J Work Environ Health 23:271±280.
Melbostad E, Eduard W, Magnus P. 1998. Determinants of asthma in a
farming population. Scand J Work Environ Health 24:262±269.
Nowak D. 1998. Health effects of airborne pollutants particularly in
swine con®nement houses ± a review focusing on aspects of
occupational medicine. [orig. german] Dtsch tieraÈrztl Wschr
105:225±234.
Post W, Heederik D, Houba R. 1998. Decline in lung function related
to exposure and selection processes among workers in the grain
processing and animal feed industry. Occup Environ Med 55:349±355.
Terho EO. 1990. Work-related respiratory disorders among ®nnish
farmers. Am J Ind Med 18:269±272.
Thelin A, HoÈglund S. 1994. Change of occupation and retirement
among Swedish farmers and farm workers in relation to those in other
occupations. Soc Sci Med 38:147±151.
Vogelzang PFJ, van der Gulden JWJ, Preller L, Heederik D, Tielen
MJM, van Schayck CP. 1996. Respiratory morbidity in relationship to
farm characteristics in swine con®nement work: possible preventive
measures. Am J Ind Med 30:212±218.
Vogelzang PFJ, van der Gulden JWJ, Preller L, Tielen MJM, van
Schayck CP, Folgering H. 1997. Bronchial hyperresponsiveness
and exposure in pig farmers. Int Arch Occup Environ Health 70:
327±333.
Vohlonen I, Tupi K, Terho EO, Husman K. 1987. Prevalence and
incidence of chronic bronchitis and farmer's lung with respect to the
geographical location of the farm and to the work of farmers. Eur J
Respir Dis Suppl 152:37±46.