Aus dem Institut für Tierzucht und Tierhaltung

der Agrar- und Ernährungswissenschaftlichen Fakultät

der Christian-Albrechts-Universität zu Kiel

___________________________________________________________________________

Salmonella in pig production –

Risk factor analysis and modelling of transmission

Dissertation

zur Erlangung des Doktorgrades

der Agrar- und Ernährungswissenschaftlichen Fakultät

der Christian-Albrechts-Universität zu Kiel

vorgelegt von

Stefanie Hotes

aus Bremen

Dekanin: Prof. Dr. K. Schwarz

1. Berichterstatter: Prof. Dr. J. Krieter

2. Berichterstatter: Prof. Dr. Dr. C. Henning

Tag der mündlichen Prüfung: 10. Februar 2011

___________________________________________________________________________

Die Dissertation wurde mit dankenswerter finanzieller Unterstützung der

Innovationsstiftung Schleswig-Holstein angefertigt.

TABLE OF CONTENTS

GENERAL INTRODUCTION

.................................................................................................................................................. 1

CHAPTER ONE

Risk Factors for Salmonella Infection in Fattening Pigs – An Evaluation of Blood and

Meat Juice Samples

.................................................................................................................................................. 5

CHAPTER TWO

An individual-based model for vertical Salmonella transmission in pig production

................................................................................................................................................ 23

CHAPTER THREE

Salmonella control measures with special focus on vaccination and logistic slaughter

procedures

................................................................................................................................................ 45

CHAPTER FOUR

The additional costs of logistic slaughter procedures to decrease Salmonella prevalence

in pork

................................................................................................................................................ 67

GENERAL DISCUSSION

................................................................................................................................................ 79

GENERAL SUMMARY

................................................................................................................................................ 87

ZUSAMMENFASSUNG

................................................................................................................................................ 91

1

General Introduction

Bacteria of the genus Salmonella are ubiquitous and cause illnesses in humans and animals all

over the world. More than 2,500 serotypes are actually known and all serotypes can cause

illnesses in humans (WHO, 2005). But only about 80 are frequently involved in animals and

human diseases worldwide (de Freitas Neto et al., 2010). Human salmonellosis is generally

related to the consumption of contaminated food of animal origin like meat, eggs, milk or

chocolate (WHO, 2007). In the majority of cases, symptoms like diarrhoea, fever, abdominal

pain, nausea and vomiting will end after 1-2 days. But also thousands of human cases result in

death worldwide (WHO, 2005). Human salmonellosis is related with society costs like

medical costs, the value of time lost from work, the value of premature death etc. For the

United States the total cost associated with Salmonella is estimated at US$ 2.6 billion for the

year 2009 (Economic Research Service, 2010) and the European Union (EU) stated the

economic significance of Salmonella in humans as well as in animals. Although, the total

number of human salmonellosis in the EU decreased since 2004, there were more than

130,000 reported cases in 2008. About 42,909 of them were reported in Germany (European

Food Safety Authority, 2010). To raise the level of public and animal health, the European

Parliament and the Council determined the control of Salmonella and other specified food-

borne zoonotic agents within the regulation (EC) No 2160/2003 from 2003. It was agreed that

‘Zoonoses present at the level of primary production must be adequately controlled to ensure

that objectives of this Regulation are achieved’. The possible sources of Salmonella input to

pig producing and fattening farms are wide as well as the spread within a farm is affected by

specific farm characteristics. Funk and Gebreyes (2004) summarised risk factors associated

with Salmonella prevalence on swine farms: Humans as vectors, flooring types, housing

contamination, pig flow management, sow-to-pig transmission, other vertebrate species,

invertebrate species, feed, environment temperature and season, the stocking density and

marketing group effects as well as the general herd health status. To enable effective control

of Salmonella it is necessary to get detailed information about Salmonella transmission and

the relation between farm conditions and prevalence.

The aim of the present study was to develop strategies to decrease Salmonella prevalence

within the pork supply chain. Research contained the detection of the most important risk

factors associated with Salmonella in pigs as well as the evaluation of vertical and horizontal

transmission and respective Salmonella control measures. Two methodological approaches

2

were distinguished: (1) analysis of empirical data to analyse the link between farm

characteristics and prevalence. (2) a simulation approach, modelling the spread of bacteria

within and between farms. The former provided useful information for the simulation,

especially to consider appropriate control measures.

Chapter One contains analyses of blood and meet juice samples to reveal the most important

risk factors associated with Salmonella in pig. Data about sample results, husbandry and

management as well as about the hygiene conditions were collected from fattening farms

participating in a producers’ association in Northern Germany. Logistic-regression models

were used to assess risk factors associated with a positive sample results. Furthermore, the

blood sample analysis was compared to the meat juice model to reveal strengths and

weaknesses of both approaches.

Chapter Two represents the developed simulation model used for further research. The

simulation model described the spread of non-clinical Salmonella within the pork supply

chain. In addition to other simulation models (Ivanek et al., 2004; van der Gaag et al., 2004;

Wehebrink et al., 2007; Hill et al., 2008; Lurette et al., 2008) the presented model considered

several farrowing and finishing farms and simulated the production and contact structures in

detail. Hence, the current model allowed more detailed analyses of the transmission paths and

the importance of the respective production stages. The model was evaluated with a reflected

Plackett-Burman design allowing the consideration of three factor levels within a reasonable

number of runs.

Salmonella control measures to decrease slaughter pig prevalence are described in Chapter

Three. Considered control measures were the improvements of hygiene and husbandry

management systems as well as the effects of vaccination and logistic slaughter procedures.

Effectiveness of control measures was compared with regard to the production stage the

control measure was implemented in.

Chapter Four considers the cost analysis of logistic slaughter procedures. Calculations were

based on a producers’ association considering threshold prevalences for herd separation

according to prevalence of 40%, 20% and 10%, respectively. Sensitivity of cost increase was

assessed due to the consideration of an increased fuel price as well as by increasing the

proportion of high-prevalence farms.

3

References

de Freitas Neto, O.C., Penha Filho, R.A.C., Barrow, P., Berchieri Junior, A., 2010. Sources of

Human Non-Typhoid Salmonellosis: A Review. Brazilian Journal of Poultry Science

12, 1-11.

Economic Research Service, 2010. Foodborne Illness Cost Calculator. United Stated

Department of Agriculture, http://www.ers.usda.gov/data/foodborneillness/ (last

access: 6.11.2010).

European Food Safety Authority, 2010. The Community Summary Report on Trends and

Sources of Zoonoses, Zoonotic Agents and Food-borne Outbreaks in the European

Union in 2008. EFSA Journal 8(1): 1496.

Funk, J., Gebreyes, W.A., 2004. Risk factors associated with Salmonella prevalence on swine

farms. Journal of Swine Health and Production 12, 246-251.

Hill, A.A., Snary, E.L., Arnold, M.E., Alban, L., Cook, A.J.C., 2008. Dynamics of Salmonella

transmission on a British pig grower-finisher farm: a stochastic model. Epidemiology

and Infection 136, 320-333.

Ivanek, R., Snary, E.L., Cook, A.J.C., Grohn, Y.T., 2004. A mathematical model for the

transmission of Salmonella Typhimurium within a grower-finisher pig herd in Great

Britain. Journal of Food Protection 67, 2403-2409.

Lurette, A., Belloc, C., Touzeau, S., Hoch, T., Ezanno, P., Seegers, H., Fourichon, C., 2008.

Modelling Salmonella spread within a farrow-to-finish pig herd. Veterinary Research

39, 49.

van der Gaag, M.A., Vos, F., Saatkamp, H.W., van Boven, M., van Beek, P., Huirne, R.B.M.,

2004. A state-transition simulation model for the spread of Salmonella in the pork

supply chain. European Journal of Operational Research 156, 782-798.

Wehebrink, T., Kemper, N., Krieter, J., 2007. Simulation study on the epidemiology of

Salmonella spp. in the pork supply chain. Campylobacter spp., Yersinia spp. and

Salmonella spp. as Zoonotic Pathogens in Pig Production. Institute of Animal

Breeding and Husbandry, Christian-Albrechts-University, Kiel.

WHO, 2005. Drug-resistant Salmonella. Fact sheet N°139. World Health Organization,

http://www.who.int/mediacentre/factsheets/fs139/en/ (last access: 6.11.2010).

WHO, 2007. Food safety and foodborne illness. Fact sheet N°237. World Health

Organization, http://www.who.int/mediacentre/factsheets/fs237/en/ (last access:

9.8.2010).

4

5

CHAPTER ONE

Risk Factors for Salmonella Infection in Fattening Pigs –

An Evaluation of Blood and Meat Juice Samples

Stefanie Hotes1, Nicole Kemper1, Imke Traulsen1,

Gerhard Rave2 and Joachim Krieter1

1Institute of Animal Breeding and Husbandry

Christian-Albrechts-University

24098 Kiel, Germany

2Institute of Variation Statistics

Christian-Albrechts-University

24098 Kiel, Germany

Article published in Zoonoses and Public Health 57 (Suppl. 1) (2010) 30-38

6

Summary

The main objective of this study was to analyse potential herd-level factors associated with

the detection of Salmonella antibodies in fattening pigs. Two independent datasets, consisting

of blood and meat juice samples respectively, were used.

Additional information about husbandry, management and hygiene conditions was collected

by questionnaire for both datasets. The serological analysis showed that 13.8% of the blood

samples and 15.7% of the meat juice samples had to be classified as Salmonella-positive.

Logistic-regression models were used to assess statistically significant risk factors associated

with a positive sample result. The results of the statistical blood sample analysis showed that

the application of antibiotics increased the Odds Ratio (OR) by a factor of 5.21 (p < 0.001)

compared to untreated pigs. A fully slatted floor decreased the prevalence of Salmonella as

well as the use of protective clothing or the cleaning of the feed tube (ORs 0.35 – 0.54, p <

0.001). It was shown that a distance of less than 2km to other swine herds increased the

chance of a positive Salmonella result (OR = 3.76, p < 0.001).

The statistical analysis of the meat juice samples revealed the importance of feed aspects. The

chance of obtaining a positive meat juice sample increased by a factor of 3.52 (p < 0.001) by

using granulated feed instead of flour. It also became clear that liquid feeding should be

preferred to dry feeding (OR = 0.33, p < 0.001).

A comparison of the blood sample analysis to the meat juice model revealed that the latter

was less powerful because data structure was less detailed. The expansion of data acquisition

might solve these problems and improve the suitability of QS monitoring data for risk factor

analyses.

Keywords: Salmonella, fattening pigs, risk factors, blood samples, meat juice samples

7

1 Introduction

Infections with Salmonella are one of the most important sources of food-borne diseases.

With 40,000 to 50,000 reported human illnesses every year (Robert Koch Institute, 2007,

2008, 2009) Salmonella still represents a major public health problem in Germany. Pork is

one of the most frequent sources of infection. About 20% of human salmonellosis are

associated with contaminated pork products (Steinbach and Kroell, 1999). The European

Food Safety Authority (2008) reported that the prevalence of slaughtered pigs infected with

Salmonella is 10.9% for Germany, a little higher than the average for the European Union

(EU). The European regulation for the control of Salmonella and other specified food-borne

zoonotic agents (EC No 2160/2003) from 2003 states that “The protection of human health

against diseases and infections transmissible directly or indirectly between animals and

humans (zoonoses) is of paramount importance”. Furthermore, it regulates that national

control programmes have to be established, which provide the detection of zoonoses and

specify control measures. Such a control programme has existed in Germany on a voluntary

basis for some time. The QS Qualität und Sicherheit GmbH was founded in 2001 to create a

voluntary basis for a system of proven quality assurance (QS Qualität und Sicherheit GmbH,

2009). This so-called QS system or QS monitoring was developed for meat and meat products

and met all statutory requirements. The system distinguishes between three risk categories of

fattening farms:

I. Farms with a herd prevalence of less than 20%

II. Farms with a herd prevalence of between 20% and 40%

III. Farms with herd prevalence of more than 40%.

The classification is determined by a certain number of samples. The required number of

samples per year depends on the quantity of delivered market hogs. Farms producing more

than 200 hogs per year have to examine 60 pigs or carcasses. In the majority of cases, the

sampling takes place at the slaughterhouse in the form of meat juice samples. The monitoring

system is based on an enzyme-linked immunosorbent assay (ELISA) for the detection of

Salmonella antibodies. Results are given as optical density % (OD%). A sample is regarded as

Salmonella-positive if the cut-off of OD% 40 is exceeded. The data on sample taking and the

results have to be collected in a special Salmonella database. Fatteners belonging to Category

III are compelled to identify the sources of Salmonella and to establish targeted measures to

reduce the prevalence. Thus, the knowledge of risk factors associated with exposure to

Salmonella is essential.

8

The aim of the study was to identify risk factors associated with the detection of Salmonella

antibodies in fattening pigs. Two datasets, blood samples and meat juice samples, were

analysed by a logistic regression. A comparison of the analyses should emphasise the

strengths and weaknesses of the respective type of data. Recommendations were made for

further analysis and data collection.

2 Materials and methods

2.1 Collection of blood samples

The data for the blood sample dataset came from a previous project, within which more than

4,200 sows and fattening pigs were sampled (Meyer, 2004). The data collection took place

between March 2001 and April 2002 and was supported by the ZNVG

(Vermarktungsgemeinschaft für Zucht- und Nutzvieh). The ZNVG is one of the largest

production associations of Schleswig-Holstein with a market share of 40%. All sampled farms

are members.

In the current study, data from 32 conventional fattening farms including 59 fattening barns

was used. Statistical analysis was carried out at barn level considering the proportion of

positive samples per barn. The investigated farms were located throughout Schleswig-

Holstein. Farms were chosen using a stratified random sample based on spatial distribution.

Within the strata, the farms were selected by a lottery system. The median number of

fattening pigs was 965, with a minimum of 272 animals and a maximum of 2,810 animals.

The numbers of examined pigs differed between farms.

The calculation of Noordhuizen et al. (1997) was used to ensure that the sampling enabled the

estimation of Salmonella prevalence at herd level. The expected prevalence was set to 20%,

the absolute accuracy to 10%, and the probability of error to 5% (Meyer, 2004). The

calculated number of blood samples varied between 50 and 65 per farm. Pigs of all barns

were sampled. Several compartments were considered within each barn. The pig sampling

itself was randomised across pens and each animal was tested only once. The producers had

no prior information about the prevalence of Salmonella in fattening pigs. QS categorisation

had already become an issue but was not started until 2003. Therefore, the farmers had great

interest in the project and no-one refused participation. 1,836 fattening pigs were sampled in

total.

9

2.2 Analysis of blood samples

The blood samples were analysed for antibodies against Salmonella O-antigens with the

SALMOTYPE® meat juice ELISA. This ELISA is based on the detection of Salmonella O-

antigens 1, 4, 5, 6, 7 and 12. Samples with an OD% higher than 40 were considered positive

(Meyer, 2004).

2.3 Questionnaire for the blood sample dataset

A survey was carried out to determine possible relations between the herd-level prevalence

and husbandry practices at fattening. The veterinarian interviewed the farmer during the visit

and completed the questionnaire for each fattening barn. The questionnaire included 56

questions on farm size, pig purchase and housing system, housing conditions, manure storage,

drinking and feeding practice, pest control, cleaning and disinfection procedures,

environmental circumstances as well as hygiene and health aspects (questionnaire is available

from the corresponding author). Most of the questions were designed as multiple-choice to

obtain answers as concise as possible.

2.4 Collection of meat juice samples

The information on the meat juice samples came from the QS database of the ZNVG. The

selection of the farms was determined by the participation in the meat juice survey and a

sufficient reply to the questions. All QS monitoring samples, taken between July 2007 and

December 2008, were considered and the proportion of positive samples was calculated for

every farm. The final meat juice dataset contained 4,204 sample records from 37 fattening and

farrow-to-finishing farms from Schleswig-Holstein.

2.5 Questionnaire for the meat juice sample dataset

The data on the management, husbandry and hygiene aspects were obtained by the same

questions as described above. Unlike the blood sample survey, each farm participated in only

one questionnaire; regardless of whether they had more fattening barns. The analysis had to

be accomplished at farm level because a sample, taken at the slaughterhouse, could not be

traced back to the barn, just to the farm. The producers were to state the most common

practice. The questionnaire was sent by post dispatched by the ZNVG. Due to the fact that

10

only 14 of 88 farm managers answered, the remaining farms were called to complete the

questionnaire with the producer or an appropriate person. At the end of the survey, 67 farmers

had participated in the study, but only 37 could be considered for analysis. 14 farms were

excluded because the producers did not complete the questionnaire sufficiently. Additionally,

13 farms were not taken into account because they worked with a continuous-flow system.

These farms were neglected due to inconsistent answers throughout the survey. The producers

stated that they worked with a continuous-flow system but used the given answers for all-in

all-out production without satisfying specification. This problem did not appear for the blood

sample dataset when the questionnaire was carried out on-site. One farm was neglected

because the regular sample size of 60 samples was not achieved. Furthermore, two farms were

excluded because they used straw as bedding material. Due to the high number of excluded

farms, the farrow-to-finishing farms were left in the dataset. Finally, the meat juice dataset

considered 26 fattening farms and 11 farrow-to-finishing farms. The number of fattening pigs

varied between 400 and 2,500 per farm with a median of 950 animals.

2.6 Statistical analysis

The same statistical methods were applied to both datasets. Following Nyman et al. (2007),

continuous covariates were categorised using the quartiles or biologically important values as

cut-points. The distribution of the original categorical covariates was also checked. Categories

with too few observations were pooled when the new classification made biological or logical

sense. Otherwise, this covariate could not be considered in further analysis (Nyman et al.,

2007). Finally, 19 potential risk factors for the blood sample dataset and 17 for the meat juice

dataset remained (Table 1). The blood samples were analysed at barn level. Compartments

within the same barn but with heterogeneous equipment or different feeding systems were

considered as autonomous barns. Analyses of the blood and meat juice samples were carried

out with the proportion of positive samples as outcome variable.

A random farm or barn effect was assessed neither for the blood sample model nor for the

meat juice model. Both are observational studies; without an orthogonal study design risk

factors and random effect would be confounded. Hence, fitting the random effect would have

involved the risk of leaving significant covariates undetected. To account for the

ubiquitousness of Salmonella and its importance in public health, the risk of mistaking a

factor for significant seemed to be justified. Collinearity between all risk factors was analysed

pair-wise using an χ² test of independence. In cases where the expected cell frequencies were

11

so small that the χ² test may not have been valid, Fisher’s exact test was additionally

calculated to ensure the result. The effect size was assessed by calculation of the Phi

Coefficient or Cramer’s V, depending on whether both variables had only two categories or at

least one had more. The cut-off for a remarkable collinearity was set to a value higher than

0.80. No variable combination showed such an effect size – neither in the blood sample

variables, nor the meat juice data.

Table 1: Potential risk factors of the blood sample dataset and meat juice sample dataset

Definition of variablesa Levelb Definition of variablesa Levelb

Number of fattening pigs <= 350 Number of fattening pigs <= 750351 - 600 751 - 1200601 - 1000 1201 - 1500> 1000 > 1500

Floor Fully slatted floor Floor Fully slatted floorPartly slatted floor Partly slatted floor

Changing room Yes Changing room YesNo No

Pest occurrence Few rodents and no birds Pest occurrence Few rodents and no birdsIncreased rodents and birds Increased rodents and birds

Feeding system Trough feeding Feeding system Dry feedingLiquid feeding Liquid feedingMash feeding Mash feeding

"Mix" of several systemsFeed structure Granulated Feed structure Flour

Pellets GranulateFlour Pellets

"Mix" of several structuresAcidification of feed or water Yes Acidification of feed or water Yes

No NoApplication of antibiotics Yes Application of Antibiotics Yes

No NoHousing of diseased animals Special compartment/stable Housing of diseased animals Special compartment/stable

Special pen Special penNo special housing

Carcass disposal Disposal from yard Carcass Disposal Disposal from yardDisposal from road Disposal from road

Pigs suppliers Established pig suppliers Farm type Finishing farmVarying pig suppliers Farrow-to-finishing farm

Pen partition Latticed pen partition Number of fattening stables 1 fattening unitClosed pen partition 2 fattening units

Protective clothing Yes 3 fattening unitsNo 4 fattening units

Pets in stable Yes Number of persons with access 1 personNo to stable(s) (vet excluded) 2 persons

Cleaning feed tube Regularly > 2 personsSometimes Care of foreign livestock YesNever No

Cleaning walls Regularly Cleaning ventilation RegularlySometimes SometimesNever Never

Cleaning boots Regularly/sometimes Water origin Well waterNever Municipal water

Feed origin Solely bought Proximity to sewage Yes Solely/mainly own crop No

Proximity to other swine Closer than 2 kmherds Further away than 2 km

Blood sample dataset Meat juice sample dataset

Corresponding potential risk

Varying potential risk factors

a Variables in bold were included in the respective multivariable model b Only categories with observations are represented

12

Logistic regression models were fitted using automated stepwise selection procedure (SAS,

PROC LOGISTIC). The selection adds or removes a covariate to the model based on the χ²

score considering a significance level of p < 0.05. The selection process terminates if no

further covariate can be added to the model or the covariate admitted is the only one excluded

in the subsequent step (SAS Institute Inc., 2004). This procedure was additionally applied to

several, slightly changed sets of potential risk factors to assess the importance of the selected

covariates for the final model.

The goodness of fit of the final models was evaluated by Nagelkerke’s R² and by visual

examination of the standardised deviance residuals plotted against the linear predictor

(Collett, 2003).

3 Results

3.1 Salmonella prevalences

The serological analysis showed that 13.8% of the blood samples and 15.7% of the meat juice

samples had to be classified as Salmonella-positive. The QS classification, based on the

respective herd prevalence, is shown in Table 2. The relative occupancy of the risk categories

is very similar, comparing the frequencies for the blood and meat juice data.

Table 2: QS categorisation based on the serological results

QS risk group

Farms in % Farms in %

Category I ≤ 20% 23 71.88 25 67.57

Category II > 20% and ≤ 40% 7 21.88 9 24.32

Category III > 40% 2 6.25 3 8.11

Blood sample dataset Meat juice sample dataset Salmonella herd-prevalence

3.2 Model fitting

For the blood sample dataset, the model selection detected the covariates ‘Floor’, ‘Pest

occurrence’, ‘Application of antibiotics’, ‘Pen partition’, ‘Protective clothing’, ‘Cleaning feed

tube’, and ‘Proximity to other swine herds’ as significant independent from changes in the set

of potential risk factors. The residual plot of the blood sample estimation did not show any

trend. Nagelkerke’s R² reached a value of 19.7, indicating that the model was able to explain

about one fifth of the variance in Salmonella prevalence. For the meat juice model the

covariates ‘Feeding system’, ‘Feed structure’, and ‘Acidification of feed or water’ were

13

significant independent from changes in the set of potential risk factors. Furthermore, the

covariates ‘Application of antibiotics’; ‘Number of fattening barns’ and ‘Cleaning ventilation’

were significant except the risk factor ‘Housing of diseased animals’ was excluded from the

set of potential risk factors. On the one hand this might be due to the high amount of missing

values in the covariate ‘Housing of diseased animals’ or on the other hand to multicolliniarity

among the four covariates. Consequently, the estimates of the covariates ‘Application of

antibiotics’; ‘Number of fattening barns’ and ‘Cleaning ventilation’ were biased. The

residuals of the meat juice model were distributed randomly but the explanatory power of the

model was worse than for the blood sample estimation. Nagelkerke’s R² reached a value of

12.0%.

3.3 Risk factor analyses

The results of the logistic regression models are shown in Table 3. Due to missing answers for

the risk factors considered, the estimations were based on 55 barns (out of 59 barns) for the

blood sample analysis and 29 farms (out of 37 farms) for the meat juice analysis.

Results for the blood sample dataset showed that the Odds Ratio (OR) for the application of

antibiotics was 5.21 (see Table 3). Accordingly, pigs treated with antibiotics had a 5 times

higher chance of being tested positive than untreated pigs. Furthermore, the proximity to other

swine herds increased the chance of obtaining a positive blood sample. However, a fully

slatted floor, protective clothing and the – even irregular – cleaning of the feed tube reduced

the chance of a positively tested pig. The results suggest that a latticed partition between pens

and the increasing number of rodents and birds in the barn may decrease the chance of a

positive result, too.

Estimated ORs for the meat juice model revealed that liquid feeding should be preferred to a

dry feeding system (Table 3). Moreover, the feeding of flour decreased the chance of a

positive meat juice sample compared to a granulated feed structure. Estimations for the “mix”

groups cannot be interpreted. They comprised all farms with more than one feeding system or

different feed structures. The acidification of feed or water showed a significant impact on the

sample results. As described above, the remaining effects seemed to be biased.

14

Table 3: Most important risk factors associated with sero-positivity for Salmonella (different

letters within an effect show significant differences between categories with p <

0.05)

Effect OR 95% CI P valueFloor Fully slatted floor 0.51 0.35, 0.74 0.0005

Partly slatted floor 1 -Pest occurrence in stable(s) Few rodents and no birds 3.04 2.02, 4.56 <0.0001

Increased rodents and birds 1 -Application of antibiotics Yes 5.21 3.51, 7.73 <0.0001

No 1 -Pen partition Closed pen partition 0.56 0.40, 0.79 0.0010

Latticed pen partition 1 -Protective clothing Yes 0.54 0.38, 0.77 0.0007

No 1 -

Cleaning the feed tube Regularly 0.40a 0.27, 0.60 <0.0001

Sometimes 0.35a 0.24, 0.51Never 1 -

Proximity to other swine herds Closer than 2 km 3.76 2.47, 5.73 <0.0001Further away than 2 km 1 -

Feeding system Liquid feeding 0.33a 0.21, 0.53 <0.0001Mash feeding 1.21b 0.86, 1.69

"Mix" of several systems 2.28c 1.37, 3.79Dry feeding 1 -

Feed structure Granulate 3.52a 2.23, 5.55 <0.0001

Pellets 1.62b 0.93, 2.79

"Mix" of several structures 3.19a 1.85, 5.50Flour 1 -

Acidification of feed or water Yes 1.80 1.30, 2.49 0.0004No 1 -

Application of antibiotics Yes 0.72 0.56, 0.92 0.0080No 1 -

Number of fattening stables 1 fattening unit 2.20a 1.49, 3.24 <0.0001

2 fattening units 1.15b 0.73, 1.81

3 fattening units 2.21a 1.41, 3.474 fattening units 1 -

Cleaning ventilation Regularly 0.99a 0.74, 1.32 0.0001

Sometimes 0.56b 0.43, 0.74Never 1 -

Blood sample dataset

Meat juice sample dataset

4 Discussion

4.1 Salmonella prevalences

The blood and meat juice herd prevalence investigated was 13.8 % and 15.7 % respectively.

Tenhagen et al. (2009) reported a Salmonella prevalence of 13.5% of lymph nodes for

German pigs. Studies for Schleswig-Holstein are rare. The Federal Institute of Risk

15

Assessment (Bundesinstitut für Risikobewertung, 2008) quantified meat juice prevalence in

Schleswig-Holstein at 23.6%. This value is much higher than detected in the present study,

but could be explained by the use of a cut-off of OD% 20 instead of OD% 40.

The comparability of blood and meat juice results, as proved by Szabó et al. (2008) or Nielsen

et al. (1998), was an important assumption for the present study. Against this background,

Table 2 suggests that Salmonella herd prevalence did not obviously change between

2001/2002 and 2007/2008. Even more important seems to be the detection of risk factors

associated with Salmonella in fattening pigs.

4.2 Model fitting

The number of farms available for the blood sample analysis and the meat juice analysis were

32 and 37, respectively. The farm selection for the blood sample dataset was randomised with

regard to spatial distribution. Sampling was carried out in the year 2000/2001. Due to the fact

that the introduction of obligatory Salmonella monitoring was imminent, fatteners were

interested in their herd prevalence and all contacted farmers participated. Six years later, when

the meat juice survey started, only 14 of 88 producers answered at the first onset.

The present meat juice survey revealed the problem of obtaining satisfying and complete

answers about the current farm situation. Without any doubt, the questionnaire performed as a

telephone survey or sent by post was unfavourable in obtaining precise answers to all

questions. A personal interview in the form of a farm visit might have improved the

willingness of the farmers to participate but would have been expensive and time-consuming.

Regrettably, it was not possible for the advisors of the ZNVG to carry out the survey during

their regular farm visits. Due to the fact that the production of QS-distinguished pork requires

regular compliance audits, a simultaneous collection of farm data by the QS delegate might be

a possibility to receive a more detailed and convincing database. Such a continuous

acquisition enables long-term analysis as well. However, the question of how to deal with

mixed systems at a particular farm still remains difficult. Data collection at barn level,

including a certain traceability of sample results from slaughterhouse to barn, would be hard

to implement. But the results of the meat juice model emphasised the difficulties of mixed

groups. A meaningful interpretation of these groups is thus not possible.

Both blood and meat juice models were fitted without a random barn or farm effect because

data came from an observational study. Consequently, selection of farms was not based on an

orthogonal study design. Hence, random effect and risk factors might be confounded. Hosmer

16

and Lemeshow (2000) referred to a confounded relationship between risk factor and outcome

variable if a covariate is associated with both the outcome variable and an independent

variable or risk factor. Brill and Barbone (2004) pointed out that confounding is a major threat

to validity. For the present study, the fitting of the random effect had involved the risk of

leaving significant covariates undetected. On the other hand, some risk factors might be

mistaken as significant. Preliminary analysis considering a random effect in the model

confirmed the presented approach: Odds Ratios and Confidence Limits were almost identical

and risk factors remained significant.

4.3 Risk factor analyses

In accordance with other studies (Davies et al., 1997; Nollet et al., 2004; Vonnahme et al.,

2008) the results showed that a fully slatted floor is associated with a decreased risk of

seropositivity for Salmonella: contaminated faeces flow away much faster and have a minor

chance of infecting susceptible pigs in the pen. Regular cleaning of the feed tube prevents the

settlement and growth of Salmonella and decreases the risk of seropositivity as well. A

preventive effect was also revealed for protective clothing and a great distance to other swine

herds. These effects reduce the possibilities of Salmonella entrance and thereby decrease

Salmonella prevalence. Pest control is supposed to work in a similar manner (Funk and

Gebreyes, 2004) but the estimated OR did not show this. An increasing number of pests in the

barn was associated with Salmonella decrease. Farzan et al. (2006) was also unable to prove a

Salmonella-increasing effect caused by rodents. The difficulties might have been due to the

subjective assessment. The awareness and sensitivity towards rodents or birds in the barn is

certainly different between farm managers and consequently their answers did not underlie

the same magnitude of impartiality. In contrast to Lo Fo Wong et al. (2004), the analysis

could not point out that pigs which were able to have snout contact with neighboured pigs

because of a latticed or low pen separation had a higher chance of being tested sero-positive.

Conversely, we observed a protective effect of more open pen partitions. This result is

doubtful because more contacts increase the chance of an infectious contact. However it has

to be considered that nose-nose transmission is less frequently the reason for infection than

faecal-oral transmission (Schwartz, 1999).

In line with van der Wolf et al. (2001), the application of antibiotics was associated with

positive testing. The estimation showed a more than five times higher chance of a Salmonella-

positive blood sample. This effect can be explained by the disturbance of the endogenous

17

flora resulting in decreased colonisation resistance which in turn reduces the minimal

infectious or colonisation doses (Nurmi et al., 1992; van den Bogaard and Stobberingh, 1999).

Several studies reported a Salmonella-preventive effect of liquid feeding or a prevalence-

increasing impact of dry feeding systems respectively (van der Wolf et al., 2001; Bahnson et

al., 2006; Farzan et al., 2006; Benschop et al., 2008). These relations also became clear in the

meat juice analysis executed. Furthermore, it could be shown that flour decreased the amount

of positive samples compared to granulated feed but not by contrast with pellets. The

insignificant difference between pellets and flour is in line with Jørgensen et al. (2001) but in

contrast with Hansen et al. (2001), who was able to prove a significant difference between

pellets and meal feed. The positive effect of meal was explained by the more coherent

structure and a lesser tendency towards phase separation. These properties may have

stimulated lactobacilli and caused a high concentration of organic acids in the gastric content.

The resulting decrease in pH deteriorates the conditions for Salmonella. Accordingly,

Jørgensen et al. (2001) reported a relative risk of 2.7 for Salmonella in pens without

acidification of feed compared to pens in which acid was used. This result seems to be in

contrast with present examinations but it has to be considered that the use of acid is a

common recommendation for Salmonella prevention in Germany. In the years 2001/2002,

acid was used on eleven finishing farms of the blood sample dataset. All farms were low-risk

farms (prevalence smaller than 20%), but one was classed as Category II with a herd

prevalence of 20.7%. In comparison, six years later, 17 farms of the meat juice dataset used

acid in feed or water. Six of them were Category II farms and one had a high risk for

Salmonella infection with a herd prevalence of 61.2%. It seems that especially fatteners with

high herd prevalence use supplemental acid to achieve Salmonella reduction. Another

explanation for the association between acid additives and high Salmonella prevalence might

be the generation of acid tolerance. Foster (1995) described the adaptation of Salmonella

typhimurium to acidity below pH 4.0 if the organisms were first adapted to a moderate acid

pH. For the present study, time series data on the supplementation of acid would be necessary

to ascertain whether the use of additives was a reaction to an increasing prevalence and

whether Salmonella decrease was actually achieved.

Concluding the present study, important risk factors associated with Salmonella in fattening

pigs were detected. The analysis of the blood and meat juice samples demonstrated especially

the importance of hygienic principles and feeding aspects, respectively. Finally, capabilities to

improve data acquisition for risk factors analyses were shown. The obligatory Salmonella

monitoring system offers enormous information on Salmonella prevalences in German

18

fattening pigs. If this information could be connected with farm data more precisely, risk

factor analyses would be more comprehensive and convincing.

References

Bahnson, P.B., Fedorka-Cray, P.J., Ladely, S.R., Mateus-Pinilla, N.E., 2006. Herd-level risk

factors for Salmonella enterica subsp. enterica in U.S. market pigs. Preventive

Veterinary Medicine 76, 249-262.

Benschop, J., Stevenson, M.A., Dahl, J., French, N.P., 2008. Towards incorporating spatial

risk analysis for Salmonella sero-positivity into the Danish swine surveillance

programme. Preventive Veterinary Medicine 83, 347-359.

Brill, I., Barbone, F., 2004. A SAS® Program Computes the Crude, Stratified and Mantel-

Haenszel Odds Ratio for Case-Control Study Analysis of Nx2xK Tables. SUGI 29,

Statistics and Data Analysis, Paper 197-29. In, SAS Global Forum, Montréal, Canada,

pp. 1-9.

Bundesinstitut für Risikobewertung, 2008. Grundlagenstudie zur Erhebung der Prävalenz von

Salmonellen in Mastschweinen.

http://www.bfr.bund.de/cm/208/grundlagenstudie_zur_erhebung_der_praevalenz_von

_salmonellen_in_mastschweinen.pdf (last access: 16.4.2009).

Collett, D., 2003. Modelling Binary Data. Chapman&Hall/CRC Northwestern University,

Evanston, Illinois, USA.

Davies, P.R., Morrow, W.E.M., Jones, F.T., Deen, J., Fedorka-Cray, P.J., Harris, I.T., 1997.

Prevalence of salmonella in finishing swine raised in different production systems in

North Carolina, USA. Epidemiology and Infection 119, 237-244.

European Food Safety Authority, 2008. Report of the Task Force on Zoonoses Data collection

on the analysis of the baseline survey on the prevalence of Salmonella in slaughter

pigs, Part A, The EFSA Journal 135, 1-111.

Farzan, A., Friendship, R.M., Dewey, C.E., Warriner, K., Poppe, C., Klotins, K., 2006.

Prevalence of Salmonella spp. on Canadian pig farms using liquid or dry-feeding.

Preventive Veterinary Medicine 73, 241-254.

Foster, J.W., 1995. Low pH adaptation and the acid tolerance response of Salmonella

typhimurium. Critical Reviews in Microbiology 21, 215-237.

19

Funk, J., Gebreyes, W.A., 2004. Risk factors associated with Salmonella prevalence on swine

farms. Journal of Swine Health and Production 12, 246-251.

Hansen, C.F., Knudsen, K.E.B., Jensen, B.B., Kjærsgaard, H.D., 2001. Effect of meal feed,

potato protein concentrate, barley, beet pellets and zinc gluconate on Salmonella

prevalence, gastrointestinal health and productivity in finishers. In, 4th International

Symposium on the Epidemiology and Control of Salmonella and other Food Borne

Pathogens in Pork - Salinpork, Leipzig, pp. 103-105.

Hosmer, D.H., Lemeshow, S., 2000. Applied Logistic Regression. John Wiley & Sons, Inc.

New York.

Jørgensen, L., Kjærsgaard, H.D., Wachmann, H., Jensen, B.B., Knudsen, K.E.B., 2001. Effect

of pelleting and use of lactic acid in feed on Salmonella prevalence and productivity in

weaners. In, 4th International Symposium on the Epidemiology and Control of

Salmonella and other Food Borne Pathogens in Pork - Salinpork, Leipzig, pp. 109-

111.

Lo Fo Wong, D.M.A., Dahl, J., Stege, H., van der Wolf, P.J., Leontides, L., von Altrock, A.,

Thorberg, B.M., 2004. Herd-level risk factors for subclinical Salmonella infection in

European finishing-pig herds. Preventive Veterinary Medicine 62, 253-266.

Meyer, C., 2004. Qualitative und quantitative Risikofaktoren für die Einschleppung und

Verbreitung von Salmonellen in unterschiedlichen Produktionsverfahren beim

Schwein. Institute of Animal Breeding and Husbandry, Christian-Albrechts-

University. Kiel.

Nielsen, B., Ekeroth, L., Bager, F., Lind, P., 1998. Use of muscle fluid as a source of

antibodies for serologic detection of Salmonella infection in slaughter pig herds.

Journal of Veterinary Diagnostic Investigation 10, 158-163.

Nollet, N., Maes, D., De Zutter, L., Duchateau, L., Houf, K., Huysmans, K., Imberechts, H.,

Geers, R., de Kruif, A., Van Hoof, J., 2004. Risk factors for the herd-level

bacteriologic prevalence of Salmonella in Belgian slaughter pigs. Preventive

Veterinary Medicine 65, 63-75.

Noordhuizen, M., Frankena, K., Graat, E., 1997. Animal health care and public health issues.

In, World Congress on Food Hygiene, The Hague/Netherlands.

Nurmi, E., Nuotio, L., Schneitz, C., 1992. The competitive-exclusion concept: development

and future International Journal of Food Microbiology 15, 237-240.

Nyman, A.K., Ekman, T., Emanuelson, U., Gustafsson, A.H., Holtenius, K., Waller, K.P.,

Sandgren, C.H., 2007. Risk factors associated with the incidence of veterinary-treated

20

clinical mastitis in Swedish dairy herds with a high milk yield and a low prevalence of

subclinical mastitis. Preventive Veterinary Medicine 78, 142-160.

QS Qualität und Sicherheit GmbH, 2009. QS - Ihr Prüfsystem für Lebensmittel.

http://www.q-s.de/ (last access: 20.10.2009).

Robert Koch Institute, 2007. Epidemiologisches Bulletin - Aktuelle Daten und Informationen

zu Infektionskrankheiten und Public Health.

http://www.rki.de/cln_178/nn_264978/DE/Content/Infekt/EpidBull/Archiv/2007/03__

07,templateId=raw,property=publicationFile.pdf/03_07.pdf (last access: 15.10.2009).

Robert Koch Institute, 2008. Epidemiologisches Bulletin - Aktuelle Daten und Informationen

zu Infektionskrankheiten und Public Health.

http://www.rki.de/cln_178/nn_969736/DE/Content/Infekt/EpidBull/Archiv/2008/03__

08,templateId=raw,property=publicationFile.pdf/03_08.pdf (last access: 15.10.2009).

Robert Koch Institute, 2009. Epidemiologisches Bulletin - Aktuelle Daten und Informationen

zu Infektionskrankheiten und Public Health.

http://www.rki.de/cln_178/nn_1378492/DE/Content/Infekt/EpidBull/Archiv/2009/03_

_09,templateId=raw,property=publicationFile.pdf/03_09.pdf (last access: 15.10.2009).

SAS Institute Inc., 2004. Chapter 42 The LOGISTIC Procedure. User's Guide (release 9.1).

Cary, NC, USA.

Schwartz, K.J., 1999. Salmonellosis. In: Straw, B.E., D'Allaire, S., Mengeling, W.L., Taylor,

D.J. (Eds.), Diseases of swine. Iowa State University Press / Ames, Iowa U.S.A.,

Iowa, pp. 535-551.

Steinbach, G., Kroell, U., 1999. Salmonella infections in swine herds - Epidemiology and

importance for human illnesses. Deutsche tierärztliche Wochenschrift 106, 282-288.

Szabó, I., Scherer, K., Roesler, U., Appel, B., Nöckler, K., Hensel, A., 2008. Comparative

examination and validation of ELISA test systems for Salmonella typhimurium

diagnosis of slaughtering pigs. International Journal of Food Microbiology 124, 65-69.

Tenhagen, B.-A., Wegeler, C., Schroeter, A., Dorn, C., Helmuth, R., Käsbohrer, A., 2009.

Association of Salmonella spp. in slaughter pigs with farm management factors. In,

8th international Symposium - Epidemiology and Control of Foodborne pathogens in

pork - Safe Pork, Québec, Canada, pp. 232-235.

van den Bogaard, A.E., Stobberingh, E.E., 1999. Antibiotic usage in animals - Impact on

bacterial resistance and public health. Drugs 58, 589-607.

van der Wolf, P.J., Wolbers, W.B., Elbers, A.R.W., van der Heijden, H.M.J.F., Koppen,

J.M.C.C., Hunneman, W.A., van Schie, F.W., Tielen, M.J.M., 2001. Herd level

21

husbandry factors associated with the serological Salmonella prevalence in finishing

pig herds in The Netherlands. Veterinary Microbiology 78, 205-219.

Vonnahme, J., Kreienbrock, L., Beilage, E.g., 2008. Untersuchungen zur Identifikation von

Risikofaktoren für die Ausbreitung von Salmonellen in Aufzuchtbeständen für

Jungsauen. Berliner und Münchener Tierärztliche Wochenschrift 121, 33-40.

22

23

CHAPTER TWO

An individual-based model for vertical Salmonella

transmission in pig production

Stefanie Hotes, Imke Traulsen and Joachim Krieter

Institute of Animal Breeding and Husbandry

Christian-Albrechts-University

24098 Kiel, Germany

24

Abstract

The aim was to develop an individual-based model for the vertical transmission of Salmonella

within the pork supply chain in order to analyse the causative factors for Salmonella

prevalence of slaughter pigs. The supply chain considered refers to pig trading from farrowing

farm to fattening farm and to slaughterhouse. The present study concentrated on farrowing

sows as the initial source of Salmonella transmission. Varying rates of Salmonella shedding

sows led to 0.04% to 0.14% infected pigs after nursing. These smallest differences in

prevalence after nursing caused differences in slaughter pig prevalence from 0.50% to

11.95%. Results showed that the probability of effective contact, to restart shedding, the

shedding duration and the sow herd prevalences as well as their distribution across farrowing

farms determined the Salmonella prevalence at slaughter. Analysis of the slaughter process

itself revealed the chance to eliminate bacteria from carcasses by careful evisceration. The

percentage of pigs contaminated during the slaughter process fell below 2.32% even for the

worst case.

The model was evaluated with a Plackett-Burman design, which enabled the screening of the

most important factors for Salmonella transmission. Three levels (minimum, default,

maximum values) for all input factors were considered. The deviations from the default

prevalences caused by the extreme values did not balance each other for several significant

input factors. For these factors, the relation between input factor and regarded health states

was not linear.

Keywords: Stochastic modelling, Salmonella, pig production, pork, Plackett-Burman

25

1 Introduction

Salmonellosis is a major problem in most countries in the world (WHO, 2007). The World

Health Organization (WHO) has reported that diseases of zoonotic origin are of particular

concern in the European Union (EU). Recent estimates predict that temperature-sensitive

infectious disease, such as food-borne infections, will become more important in the coming

decades (WHO, 2010). Frequent cases of human salmonellosis in Germany have induced the

Federal Institute for Risk Assessment (BfR) to emphasise the avoidance of raw meat, in

particular raw pork in any form (BfR, 2005). Salmonella prevalence in slaughter pigs in the

EU is estimated at about 10%. Country prevalences range from no positive findings in

Finland up to a prevalence of 29% in Spanish slaughter pigs (European Food Safety

Authority, 2008). The 2003 European regulation for the control of Salmonella and other

specified food-borne zoonotic agents (EC No 2160/2003) regulates that national control

programmes have to be established to provide the detection of zoonosis and specific control

measures. The German implementation focuses on the finishing stage. Only if prevalence

exceeds 40%, the delivered pigs should be tested as well. However, pig deliveries are

certainly critical points for dynamics of Salmonella transmission. Bacteria can spread across

farms due to the enhanced contact structure. Previously published models have concentrated

on the spread within a single farm (Ivanek et al., 2004; Hill et al., 2008; Lurette et al., 2008)

or have not considered the production and contact structures in detail (van der Gaag et al.,

2004; Wehebrink et al., 2007).

The objective of the present study was to develop an individual-based model for the

transmission of non-clinical Salmonellosis within the pork supply chain. Modelling was to

reveal the most important factors for Salmonella spread with regard to the trade relationships

between the production stages.

2 Materials and methods

The developed simulation model describes the spread of non-clinical Salmonella infection

(henceforth referred to as “infection”) within the pork supply chain from farrowing stage to

slaughter. The model is stochastic with a discrete time step of one week. Comparable to all-in

all-out production, the one-week time step is defined as the basic unit for regrouping tasks and

moving pigs between branches of production. Only transports and processes at

26

slaughterhouses are detached from the one-week rhythm of the model. The present study

focused on the vertical transmission of Salmonella from sow to pig and subsequently from pig

to pig. The epidemiological process in the model describes the spread of Salmonella over a

52-week period with a prior burn-in also of 52 weeks. The burn-in preceded the simulation in

order to initiate the trade relationships and to fill the model with pigs. At starting point, the

model contains only farms and sows because output data has to be based on slaughter pigs

whose whole lives were to be monitored. Without burn-in, the first pigs would reach slaughter

age at week 28. The quantity and quality of the output data would be low. The burn-in ensures

a continuous flow of slaughter pigs and thereby improves the results. The program is written

in the object-oriented language C++. Routines from the NAG C library (NAG, 2001) are used

to generate random numbers.

2.1 Trade relationships

The model contains a network of farrowing and finishing farms as well as a slaughterhouse

which are linked by pig deliveries. The model considers two types of farrowing farms: 1)

conventional farrowing farms with supply relationships (PPS) to one (PPS_one), two

(PPS_two) or three finishing farms (PPS_three); 2) farrow-to-finishing farms without supply

relationships to a specialised finishing farm (PPF). At the finishing stage, fatteners buying pigs

from one (F_one), two (F_two) or three (F_three) conventional farrowing farms are

distinguished. Figure 1 illustrates several possible trade relationships. The only information

obligatory for the network concerns the farrowing farms and contains the sow herd size

(varying from 100 to 350 sows with a mean of 181 sows in the current study), the suckling

period (four weeks by default for the current study) and the production cycle (mix of one and

three weeks in the current study). Prior to every replication a regeneration of the trade

relationships is performed considering the stated distributions (Table 1, nos. 1-3). The

regeneration implicates the classification of the farrowing farms into farms with a low

(PPlowPrev), middle (PPmiddlePrev), or high sow herd prevalence level (PPhighPrev). Based on this

classification each sow on the farm has a certain probability to become a shedder (S_shedder)

or non-shedder (S_nonShedder) (Table 1; nos. 4 and 5). The health states of the sows are not

regarded in more detail and remain unchanged for the respective replication.

27

Farrow-to-finishing farm

Slaughterhouse

Finishing farm

Conventional farrowing farm

Conventional farrowing farm

Finishing farm

Conventional farrowing farm

Conventional farrowing farm

Finishing farm

Conventional farrowing farm

Finishing farm

Transport

Transport

Transport Transport Transport Transport Transport TransportTransport

Transport Transport

Farrowing stage

Finishing stage

Slaughtering stage

Farrow-to-finishing farm

Slaughterhouse

Finishing farm

Conventional farrowing farm

Conventional farrowing farm

Finishing farm

Conventional farrowing farm

Conventional farrowing farm

Finishing farm

Conventional farrowing farm

Finishing farm

Transport

Transport

Transport Transport Transport Transport Transport TransportTransport

Transport Transport

Farrowing stage

Finishing stage

Slaughtering stage

Figure 1: Illustration of possible trade relationships within the simulated pork supply chain

2.2 Health and contamination states

The program is individual-based i.e. each pig is followed from farrowing to chilling at

slaughter. During its lifetime the pig’s health status is updated every week. As described by

Lurette et al. (2008), four mutually exclusive health states are distinguished: susceptible pigs

free of Salmonella and thus non-shedding and seronegative (S-); seronegative, shedding pigs

(I-); seropositive, shedding pigs (I+), and seropositive non-shedding but Salmonella-bearing

pigs (C+) (Figure 2). The latent period between Salmonella ingestion and shedding falls

below the model basis of one week (Lurette et al., 2008) and is therefore neglected. The

seroconversion is assumed to last about two weeks (van der Gaag et al., 2004). Subsequent to

seroconversion, the pig stays seropositive. The possibility of recovery is not considered. Even

if a return to seronegativity exists, it is assumed to last longer than the lifetime of the pigs

(Lurette et al., 2008). The only change in health state based on the carrier stage is to restart

shedding and become I+ again. The risk that the carrier pigs restart shedding is always present

(Prs_basic) but increases if the pigs are exposed to stress (Prs_stress) (Table 1; no. 7). Whether

it is the first shedding period (I- until C+) or a subsequent one (starting from C+: I+ to C+),

the shedding duration is assumed to be Weibull-distributed (Hill et al., 2008) with a mean

shedding duration of about four weeks (Table 1, no. 12).

28

Table 1: Definition of the input proportions, probabilities and parameters used in the model

Notation Sourceproportions probabilities parameter

1 PPF 0.25 0.1875 0.3125 a

PPS 0.75 0.8125 0.6875 a

2 PPS_one 0.7 0.775 0.625 b

PPS_two 0.2 0.15 0.25 b

PPS_three 0.1 0.075 0.125 b

3 F_one 0.85 0.8875 0.8125 aF_two 0.14 0.105 0.175 aF_three 0.01 0.0075 0.0125 a

4 PPlowPrev 0.8 0.85 0.75 c

PPmiddlePrev 0.15 0.1125 0.1875 c

PPhighPrev 0.05 0.0375 0.0625 c

5Sl_nonShedder 0.99 0.9925 0.9875 b

Sl_shedder 0.01 0.0075 0.0125 b

Sm_nonShedder 0.95 0.9625 0.9375 b

Sm_shedder 0.05 0.0375 0.0625 b

Sh_nonShedder 0.9 0.925 0.875 b

Sh_shedder 0.1 0.075 0.125 b

6

the same pen pij = pw 0.0014 0.00105 0.00175 d

pens adjacent to each other pij = pb 0.0009 0.000675 0.001125 d

pens in same row but not adjacent pij = pb/3 0.0003 0.000225 0.000375 d

pens in opposite row pij = pb/100 0.000009 0.00000675 0.00001125 d

pm 0.0004 0.0003 0.0005 b

pl 0.0004 0.0003 0.0005 b

7basically Prs_basic 0.2 0.15 0.25 e

due to stress Prs_stress 0.4 0.3 0.5 e

8 Proportion of Salmonella units remaining at lairage rl 0.3 0.225 0.375 b

9 Probability to lose Salmonella at evisceration Pe 0.75 0.5625 0.9375 f

10 pc 0.1 0.075 0.125 f

11 rc 0.3 0.225 0.375 f

12 Weibull-distributed shedding duration withshape parameter α 2.36 2.36 2.36 dscale parameter β 27.8 15.05 40.6 d

Minimum values

Maximum values

Default values

Probability of effective contact at transport

Definition

sow herd prevalence is high

Proportion of farrowing farms with low, middle, and high sow herd prevalence

Proportion of farrow-to-finishing farms to conventional farrowing farms

Proportion of conventional farrowing farms selling their pigs to one, two, or three fatteners

Proportion of finishing farms buying the pigs from one, two, or three farrowing farms

No.

Probability of contaminating contact at slaughterhouse

Proportion of Salmonella units remaining at tools

Probability of effective contact within compartment between pen i and j if pigs are in…

Proportion of non-shedder to shedder sows if…

Probability to restart shedding…Probability of effective contact at lairage

sow herd prevalence is low

sow herd prevalence is middle

a)based on ZDS (2009), b) assumed, c) based on QS classification (QS GmbH, 2009), d) Hill et al. (2008), e) Lurette et al. (2008), f) based on van der Gaag et al. (2004)

28

29

All piglets are born susceptible to infection and infection is immediately possible within the

farrowing crate. The aspect of passive immunity of newborn piglets from a sow’s antibodies

is however neglected, due to the lack of precise information. The health state of the pig does

not change post-slaughter (Figure 2). But slaughter pigs lose bacteria due to careful

evisceration or become contaminated on their skin surfaces due to the slaughter sequence

(direct contact of carcasses) or contaminated tools (cross-contamination) (Table 1; nos. 8-11).

S- I- I+ C+Pre-slaughter states

contamination contamination

loss of bacteria

contamination

loss of bacteria

contamination

loss of bacteria

Post-slaughter states

seroconversion

seronegative

non-shedding

seronegative

shedding

seropositive

shedding

seropositive

non-shedding

Infected pigs

S- I- I+ C+Pre-slaughter states

contamination contamination

loss of bacteria

contamination

loss of bacteria

contamination

loss of bacteria

Post-slaughter states

seroconversion

seronegative

non-shedding

seronegative

shedding

seropositive

shedding

seropositive

non-shedding

Infected pigs

Figure 2: Health and contamination states of pigs pre- and post-slaughter

2.3 Probability of infection and contamination

The modelling of infection is based on the Reed-Frost epidemic model, which describes the

probability P(t) of a susceptible pig having effective contact with any of E(t) excreting pigs

during the period [t-1, t] (Rubel, 2005; Hill et al., 2008).

)()1(1)( tEptP −−= (1)

Effective contact is defined as contact between an infectious individual and a susceptible one,

which produces a newly infected individual (Bailey, 1975). The probability for an effective

contact is given by p.

The current program considers the sows of the farrowing farms as the initial sources of

Salmonella infection. Consequently, Salmonella is transmitted only vertically from sow to pig

and subsequently from pig to pig without Salmonella entry via rodents, birds, feed et cetera.

In the following, the possibilities for effective contact between two individuals are described

in detail.

30

2.3.1 Infection at farm

Assuming that all pigs housed in pens of the same compartment have contact with each other,

the probability of an effective contact increases with an increasing number of shedding pigs.

Contact can happen via faecal or airborne transmission. Formula 2 describes the probability

Pij of a susceptible pig in pen i becoming infected because of Ej(t) excreting pigs in pen j

during the period [t-1, t] (Hill et al., 2008):

)(

)1(1)(tE

ijij

jptP −−= (2)

The probability of effective contact between pens i and j (pij) depends on the distance between

these pens (Table 1; no. 6). The number of pens within a compartment varies between farms

but all compartments consist of two opposite rows. During nursing, transmission is assumed

to be reduced to pen level. This assumption is based on the limited contacting of new born

piglets to piglets of other pens.

2.3.2 Infection at transport

In an agglomerated pig producing area such as Northern Germany, transport from farrowing

farm to finishing farm does not exceed a few hours. Precise data about new infections during

transport are missing. Pigs are stressed due to transport and prior fasting whereby carrier pigs

could restart shedding. This is implemented in the program by an increased probability for

shedding reactivation (Prs_stress; Table 1; no. 7). Hence, in the model, the increased number

of shedding pigs increases the probability of a susceptible pig becoming infected at the

fattening compartment (Pij). This mainly compensates for the missing infection opportunity

during transport to the finishing stage.

In contrast, transport to slaughterhouse is not followed by an extensive housing period and

lasts much longer, which increases the probability of effective contact at the lorry (pm; Table

1; no. 6). The probability Pm of a susceptible pig becoming infected due to effective contact

with any of Em(t) excreting pigs at transport to slaughterhouse is represented by:

)()1(1)( tE

mmmptP −−= (3)

An increased probability of shedding restart is also considered at transport to slaughterhouse

(Table 1; no. 7).

31

2.3.3 Infection at lairage

After transport to the slaughterhouse, the pigs are housed at lairage until the slaughter process

starts. Comparable to infection at the farm and during transport the probability of infection at

lairage (pl; Table 1; no. 6) increases with an increasing number of Salmonella-shedding pigs

in the proximity. The lairage is neither cleaned intensively nor disinfected before a new group

of pigs is housed. Comprehensive cleaning is only possible at the end of a working day.

Insufficient cleaning during the day increases the probability of infection at lairage. In

addition to the number of shedding pigs in the current group (El(t)), the Salmonella output of

all previous groups has to be considered (Rl).

)()()1(1)( tRtE

llllptP

+

−−= (4)

with llll rtRtEtR ⋅−+−= ))1()1(()(

rl represents the proportion of Salmonella units remaining at lairage after a superficial

cleaning (Table 1; no. 8).

2.3.4 Loss of bacteria and contamination at slaughter line

After death, the pigs are no longer able to change their health state, however surface

contamination and loss of bacteria is possible. The probability of losing Salmonella (Pe)

depends on how carefully evisceration is performed (Table 1; no. 9).

Carcass contamination could happen via direct contact of carcasses or cross-contamination

due to soiled equipment and tools. Whether a pig becomes contaminated at the slaughter

process depends preliminarily on the probability of a contaminating contact (pc; Table 1; no.

10) and the Salmonella status of the previously slaughtered pigs (Ec(t)). Furthermore, pigs

already slaughtered could have contaminated the equipment and tools whereby bacteria could

skip carcasses or contaminate a whole array of subsequent carcasses.

)()()1(1)( tRtE

ccccptP

+

−−= (5)

with

=

1

0)(tEc

cccc rtRtEtR ⋅−+−= ))1()1(()(

Rc describes the impact of all previously slaughtered pigs with rc representing the proportion

of Salmonella units which move from carcass to carcass (Table 1; no. 11).

if Salmonella status of previous slaughtered pig = S-

if Salmonella status of previous slaughtered pig = I-, I+ or C+

32

2.4 Input data availability

For the present study most of the input data were obtained from literature or advisory services

(Table 1). No information was available about how many farrowing farms selling their pigs to

one, two or three fatteners, the respective sow herd prevalence, the probability of an effective

contact at transport and lairage and the proportion of Salmonella units remaining at lairage

(Table 1; no. 2, 5, 6 and 8). Hence, information was acquired by discussion with experts.

2.5 Number of replications

For the executed appraisal, the decision on how many replications were needed to obtain

sufficient model output , was based on the variance which would appear in the output data

due to a certain number of replications. First, the model was run with default, minimum, and

maximum values1 200 times each (see Table 1). Based on these 200 replications, 30 packages

of 5, 10, 20, 30, ... 100 replications were randomly sampled with replacement. This

bootstrapping was executed for minimum, maximum and default values. The variance of

every package was calculated. Afterwards, the standard deviation of every package size,

consisting of 30 variances each, was also calculated. Figure 3 shows the falling trend of

standard deviation with increasing package size.

0.E+00

1.E-04

2.E-04

3.E-04

4.E-04

5 10 20 30 40 50 60 70 80 90 100

Size of packages

Sta

ndar

d de

viat

ion

of v

aria

nces-

maximum values

default values

minimum values

Figure 3: Standard deviation of variances related to the size of packages – obtained via

bootstrapping method

1 Default, minimum, and maximum values represent the three factor levels of the Placket-Burman design explained later.

33

Finally, the package size with 30 replications was considered as sufficient. There was a

relatively sharp decline for the maximum graph up to 20 replications and for the default graph

up to 30 replications. The further trend slowed down in decline. This indicates that the

variance was no longer influenced by the number of replications.

2.6 Sensitivity analysis

A sensitivity analysis was performed to evaluate the model. A Plackett-Burman design (P-B

design) was used to identify the most important input factors and to estimate their impact on

the allocation of health states. Originally, P-B designs are two-level fractional factorial

designs of resolution III for studying up to k = N-1 components in N runs (Montgomery,

2005). To compare three factor levels in the analysis, the basic P-B design, considering

maximum (+) and default (0) values, had to be reflected. Therefore, the maximum values

were replaced with the minimum (-) values (Figure 4a and 4b) (Vanaja and Shobha Rani,

2007).

Assembly Assembly1 2 3 … 12 1 2 3 … 12

1 + 0 0 0 17 - 0 0 02 + + 0 + 18 - - 0 - 3 + + + 0 19 - - - 0

… …

16 0 0 0 0 32 0 0 0 0

Assembly Assembly1 2 3 … 12 1 2 3 … 12

33 0 + + + 49 0 - - -34 0 0 + 0 50 0 0 - 035 0 0 0 + 51 0 0 0 -

… …

48 + + + + 64 - - - -

Input factors Input factors

a) Basic design b) Reflected design

c) Reversion of the basic design d) Reversion of the reflected design

Input factors Input factors

Figure 4: Basic, reflected and reversed Plackett-Burman designs

In general, resolution III designs confound the main effects with two-factor interactions. To

ensure that the main effects are not confounded with two-factor interactions, these designs

have to be reversed (Barrentine, 1996; Montgomery, 2005). Hence, the basic and reflected

designs were carried out again but extreme and default values were switched (Figure 4c and

34

4d). This full fold-over technique broke the alias links between all the main effects and their

two-factor interactions and resulted in a design of resolution IV (Montgomery, 2005). All

together, 64 assemblies (four designs with 16 assemblies each) were performed considering

the default, minimum and maximum values of the twelve factors described in Table 1. Default

values of the probabilities were increased and decreased by 25% to obtain minimum and

maximum values, respectively. Concerning proportions, minor parts were increased by 25%

whereas the major part was decreased accordingly. For the Weibull distribution, the standard

deviation was subtracted and added to the default value. Output data of the P-B design was

analysed using the MIXED procedure of the SAS program package (SAS, 2003). Linear

models were fitted for the four health states, the transmission paths of infection and for the

carcass contamination as well as for the loss of Salmonella due to careful evisceration (Figure

5). For every health state six models were formulated to analyse the variances over the course

of time (Block A of Figure 5). To analyse Salmonella transmission paths, seven models were

formulated (Block B of Figure 5) and for the loss of bacteria and the carcass contamination

one model each was added (Block C of Figure 5).

Percentage of …

after

Percentage of pigs, which became infected during …

• nursing due to a pig of the same pen

• growing or fattening due to a pig of the same pendue to a pig of an adjacent pendue to a pig of a pen of the same row not adjacentdue to a pig of a pen of a the opposite row

• transport

• lairage

Percentage of carcasses …

• lost Salmonella at slaughter

• contaminated at slaughter

Respective response variable

(1) Proportion of pig producers with own fattening to pig producers selling pigs for fattening

(2) Proportion of pig producers selling their pigs to one, two, or three fatteners

(3) Proportion of fatteners buying the pigs from one, two, or three pig producers

(4) Proportion of farrowing farms with low, middle, and high sow herd prevalence

(5) Proportion of non-shedder to shedder sows

(6) Probability of effective contact

(7) Probability to restart shedding

(8) Proportion of Salmonella units remaining at lairage

(9) Probability to lose Salmonella at evisceration

(10) Probability of a contaminating contact at slaughter

(11) Proportion of Salmonella units remaining at tools

(12) Weibull-distributed shedding duration

Explanatory variables

Block A

Block B

Block C

24 models

7 models

2 models

• S- pigs nursing

growing

transport to fattener

fattening

transport to slaughter

lairage

• C+ pigs

• I+ pigs

• I- pigs

Percentage of …

after

Percentage of pigs, which became infected during …

• nursing due to a pig of the same pen

• growing or fattening due to a pig of the same pendue to a pig of an adjacent pendue to a pig of a pen of the same row not adjacentdue to a pig of a pen of a the opposite row

• transport

• lairage

Percentage of carcasses …

• lost Salmonella at slaughter

• contaminated at slaughter

Respective response variable

(1) Proportion of pig producers with own fattening to pig producers selling pigs for fattening

(2) Proportion of pig producers selling their pigs to one, two, or three fatteners

(3) Proportion of fatteners buying the pigs from one, two, or three pig producers

(4) Proportion of farrowing farms with low, middle, and high sow herd prevalence

(5) Proportion of non-shedder to shedder sows

(6) Probability of effective contact

(7) Probability to restart shedding

(8) Proportion of Salmonella units remaining at lairage

(9) Probability to lose Salmonella at evisceration

(10) Probability of a contaminating contact at slaughter

(11) Proportion of Salmonella units remaining at tools

(12) Weibull-distributed shedding duration

Explanatory variables

Block A

Block B

Block C

24 models24 models

7 models7 models

2 models2 models

• S- pigs nursing

growing

transport to fattener

fattening

transport to slaughter

lairage

• C+ pigs• C+ pigs

• I+ pigs• I+ pigs

• I- pigs• I- pigs

Figure 5: Outline of the formulated models for sensitivity analysis

All 33 linear models considered the same twelve explanatory variables which represented the

input factors of the simulation model. Analyses were done via F-Test, Least Square Means

(LS-Means) and confidence limits. Corresponding to the P-B designs, the analysis had to

35

consider three levels for every input factor. To meet these multiple comparisons, the p-values

and confidence limits were accomplished with the Bonferroni adjustment.

3 Results of the sensitivity analysis

First analyses considered only the assemblies 16, 48 and 64 of the P-B design, to receive an

impression of the prevalence generated by the model. These three assemblies described the

model output if all input factors were set to default, maximum and minimum values,

respectively. After nursing, the percentages of infected piglets were very similar, but

differences in prevalence increased in the course of time.

• Minimum values: prevalence after nursing: 0.04% after lairage: 0.50%

(assembly 64)

• Default values: prevalence after nursing: 0.08% after lairage: 4.84%

(assembly 16)

• Maximum values: prevalence after nursing: 0.14% after lairage: 11.95%

(assembly 48)

The proportion of infected pigs which lost Salmonella due to evisceration represented the

given probabilities of about 56% (minimum values), 75% (default values) and 94%

(maximum values). The carcass contamination never exceeded 2.32%. (Results not

presented.)

Further analyses were based on all 64 assemblies. The significance of the input factors for the

response variables are shown in Table 2. The presentation of the results for all 33 analyses

would be too expansive. Hence, the presentation of the health state models is limited to the

susceptible pigs (S-) (Table 2; Block A). This group represents exactly the opposite to the

infected pigs ((I-) + (I+) + (C+). Hence, factors with significant impact on the percentage of

susceptible pigs influenced the amount of infected pigs, respectively. Factors 4-6 in Table 2

show that from nursing to slaughter the “Proportion of farrowing farms with low, middle, and

high sow herd prevalence”, the “Proportion of non-shedder to shedder sows”, as well as the

“Probability of effective contact” have a significant impact on the percentage of susceptible

pigs (Table 2; Block A). While the first two factors determined the amount of Salmonella

units brought to model, the probability of effective contact effected the transmission dynamic

of the present bacteria.

36

Table 2: Results of the F-Test examining the significance of the input factors for the percentage of susceptible pigs (Block A), for the transmission

path (Block B), as well as for the percentage of Salmonella loss and carcass contamination (Block C)

No. Definition

nursing transport lairagenursing growing fattening lairage same pen same pen adjacent pen same row opposite row

1 Proportion of farrow-to-finishing farms to conventional farrowing farms

n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s.

2 Proportion of conventional farrowing farms selling their pigs to one, two, or three fatteners

n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s.

3 Proportion of finishing farms buying the pigs from one, two, or three farrowing farms

n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s.

4 Proportion of farrowing farms with low, middle, and high sow herd prevalence

<0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001

5 Proportion of non-shedder to shedder sows

<0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001

6 Probability of effective contact

<0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001

7 Probability to restart shedding

n.s. n.s. n.s. <0.0001 <0.0001 <0.0001 n.s. <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001

8 Proportion of Salmonella units remaining at lairage

n.s. n.s. n.s. n.s. n.s. 0.01 n.s. n.s. n.s. n.s. n.s. n.s. <0.0001 n.s. 0.006

9 Probability to lose Salmonella at evisceration

n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. <0.0001 n.s.

10 Probability of contaminating contact at slaughterhouse

n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. <0.0001

11 Proportion of Salmonella units remaining at tools

n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. <0.0001

12 Shedding duration n.s. <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 n.s. <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001

Block A (limited to S-) Block B Block C

transport to fattener*

transport to slaughter

Percentage of susceptible pigs after … Percentage of carcasses ...lost Salmonella

at slaughtercontaminated at slaughter

Percentage of pigs, which became infected during .. growing or fattening

significance level = 0.01 * New infections are not possible at transport to fattener. Hence, estimates for the percentage of susceptible pigs are identical to estimates after growing.

36

37

Due to an average shedding duration of four weeks, Factor No. 12 did not become significant

until growing. Even if pigs became infected at the week of birth, shedding duration probably

exceeded the nursing time. Subsequently, the probability to restart shedding (no. 7) cannot be

significant prior to that. Pigs have to finish shedding before they can restart. The factor

“Proportion of Salmonella units remaining at lairage” (no.8) became significant at lairage.

The similarities between the significant factors for the percentage of susceptible pigs (Table

2; Block A) and the transmission paths (Table 2; Block B) were reasoned. The transmission

paths were determined by effective contacts, which decreased the number of susceptible pigs.

Block C of Table 2 shows the significant input factors for the percentages of pigs which lost

Salmonella at evisceration or became contaminated during slaughter process, respectively.

Note that all pigs were considered to estimate the percentage of pigs which lost Salmonella;

even susceptible pigs. Hence, additionally to the “Probability to lose Salmonella at

evisceration” (no. 9) also factors influencing the percentage of infected pigs were significant.

The same applies to the percentage of contaminated carcasses. Next to the “Probability of

contaminating contact at slaughterhouse” (no. 10) and the “Proportion of Salmonella units

remaining at tools” (no.11), all factor groups influencing the amount of infected pigs were

significant.

To illustrate the progress of infection, Figure 6 shows the LS-Means and their confidence

limits for the percentage of susceptible pigs depending on factor groups 4-6 and 12. The most

effective contacts happened during fattening where pigs spent most of their lives (Figure 6a).

Figure 6b shows the importance of the sow herd prevalences at pig producing stage.

Comparing the deviations from the default values, it became clear that the decrease in

susceptible pigs caused by the maximum values exceeded the increasing effect of the

minimum values. The same applied to the proportion of non-shedder to shedder sows (Figure

6c). In contrast, the relation between shedding duration and percentage of susceptible pigs

was linear. Increasing and decreasing effects balanced each other out (Figure 6d). The

influence of transport-stress on the shedding reactivation is shown in Figure 7. The number of

seropositive, shedding pigs (I+) increased especially at transport to slaughter, which increased

the probability of a susceptible pig becoming infected.

38

(a) Probability of effective contact

90

92

94

96

98

100

after

nursing

after

growing

after

transport to

fattener

after

fattening

after

transport to

slaughter

after lairage

Sus

cept

ible

pig

s (S

-) in

%

minimum values

default values

maximum values

(b) Proportion of farrowing farms with low, middle, and high sow

herd prevalence

90

92

94

96

98

100

after

nursing

after

growing

after

transport to

fattener

after

fattening

after

transport to

slaughter

after lairage

Sus

cept

ible

pig

s (S

-) in

%

minimum values

default values

maximum values

(c) Proportion of non-shedder to shedder sows

90

92

94

96

98

100

after

nursing

after

growing

after

transport tofattener

after

fattening

after

transport toslaughter

after lairage

Sus

cept

ible

pig

s (S

-) in

%

minumum values

default values

maximum values

(d) Shedding duration

90

92

94

96

98

100

after

nursing

after

growing

after

transport to

fattener

after

fattening

after

transport to

slaughter

after lairage

Sus

cept

ible

pig

s (S

-) in

%

minimum values

default values

maximum values

Figure 6: LS-Means and confidence limits of the percentage of susceptible pigs depending

on (a) the probability of effective contact, (b) proportion of farrowing farms with

low, middle, and high sow herd prevalence, (c) the proportion of non-shedder to

shedder sows, and (d) the shedding duration.

0

1

2

3

4

after nursing after growing after transportto fattener

after fattening after transportto slaughter

after lairage

Ser

opos

itive

, she

ddin

g pi

gs (

I+),

-

carr

ier

pigs

(C

+),

and

se

rone

gativ

e, s

hedd

ing

pigs

(I-

) in

% -

seropositive, shedding pigs (I+)

carrier pigs (C+)

seronegative, shedding pigs (I-)

Figure 7: LS-Means and confidence limits of the percentage of infectious, seronegative pigs,

infectious, seropositive pigs, and carrier pigs depending on the probability to

restart shedding (default value)

39

4 Discussion

4.1 Transition model

The transition model considered both farm level and slaughter stage. The consideration of the

four pig’s health states was adequate for farm level. Until transport to slaughter, neglecting of

the latent period between Salmonella ingestion and bacteria shedding was considered to be

unproblematic since duration fell below the considered time step of one week. But Salmonella

ingestion at transport to slaughter resulted immediately in the shedding of bacteria since an

intermediate health state was missing. Hence, the chance of a susceptible pig having become

infected at lairage might have been overestimated. But infection at lairage represented about

5% in the maximum case. Overestimation was neither of little consequence for carcass

contamination. The passive immunity of newborn piglets was not considered either. Lurette et

al. (2008) described the maternal protective factor as one of the most influential parameters on

Salmonella prevalence in delivered pigs but precise information about this effect is missing.

In support, Nollet et al. (2005) could not prove the direct transmission of Salmonella from the

sows to their piglets at farrowing barn, which confirms passive immunity. But they

demonstrated similarities between the isolates found in the sows and those found during

growing, finishing and at slaughter. Hence, even if passive immunity of newborn piglets

exists, the sow may play a significant role in the indirect transmission of Salmonella to

growing and finishing pigs (Nollet et al., 2005). In the present study, not even 0.15% of the

piglets had effective contact during nursing. The differences between minimum, default, and

maximum values were very small after nursing but increased in the course of time. In the

maximum scenario, slaughter pig prevalences of 12% were reached. This prevalence seemed

to be very high, considering that the sows were the only initial infection source in the model

presented. But it has to be considered that within the model no infection remained undetected,

not even prior to seroconversion. In contrast, empirical prevalences based on antibody

detection may never represent all infected pigs at a particular time (Battenberg, 2007).

Nevertheless, the obtained slaughter pig prevalences emphasize the impact of the sow for

Salmonella entrance and the subsequent transmission. The transmission dynamic depends

extremely on how many piglets come into contact with Salmonella during nursing.

40

4.2 Number of replications

The determination of the number of replications is important and difficult at once. Vynnycky

and White (2010) emphasise that the number of replications is usually limited by the

realisation time. If several models are to be compared, the determinations rely on the number

of replications which are required to analyse statistical differences between the models

(Chung, 2004). But the present study did not compare models. Instead, the number of

replications was of concern, which ensured that enough random numbers were used to

represent the underlying distribution.

Usually, if this number of replications is obtained, the variance of the output does not

decrease due to additional replications. Variations are casual and balance each other. This

point was determined to be reached with 30 replications. Some might prefer 80 replications

due to further flattening within the trend of the maximum graph, but because of the already

small level of standard deviation the additional realisation time seemed to be unjustified.

4.3 Sensitivity analysis

Due to the relative high number of input factors a screening design was appropriate to identify

the most important factors and evaluate the model based on their estimated effects. P-B

designs are screening designs, which estimate unbiased main effects in the smallest design

possible (Vanaja and Shobha Rani, 2007). The limited number of runs minimised the

realisation time and offered the comparison of three levels instead of two. Thus, it could be

shown that the relationship between several factors and response variable is non-linear. An

increase in the probability of effective contact, a higher proportion of pig producers with

middle and high sow herd prevalence, or more shedder sows decreased the percentage of

susceptible pigs much more as could have been increased by the minimum values (Figure 6).

The problem that the main effects are confounded with two-factor interactions in a P-B design

was solved by reversing the designs. The resulting fold-over pairs were of resolution IV,

which did not confound the main effects and two-factor interactions (Box et al., 1978;

Montgomery, 2005). But in contrast to traditional designs of resolution IV, a P-B design does

not allow the estimation of interactions between factors (Vanaja and Shobha Rani, 2007).

Hence, it cannot be ruled out that there are significant relationships between factors which

remained undetected.

The estimated main effects of the model met expectations. To avoid that biologically

nonrelevant differences become significant, the significance level for the F-Test was reduced

41

to 1%. In simulation designs with large sample sizes smallest differences become significant.

According to Ivanek et al. (2004) the probability of effective contact (no. 6), the probability of

restarting shedding (no. 7), as well as the shedding duration (no. 12) were proven to be

significant. Hill et al. (2008) point out that the most effective control strategies are those that

reduce the probability of effective contact between pens. The relevance of the proportion of

farrowing farms with low, middle, and high sow herd prevalence (no. 4) as well as the

importance of the proportion of non-shedder to shedder sows (no. 5) cannot be compared

exactly with other studies, since previous models known by the authors did not consider the

sow as initial source of Salmonella infection in detail or concentrated on the spread within a

single farm. But van der Gaag et al. (2004) proved that a higher starting prevalence at

farrowing stage results in more infected animals. Furthermore, Ivanek et al. (2004) state that

Salmonella transmission is influenced by the prevalence among weaners.

For the contamination and loss of bacteria at slaughter, detailed information and simulation

studies are rare. Van der Gaag et al. (2004) describe the slaughterhouse as one of the most

important stages in the supply chain to reduce the prevalence of Salmonella-contaminated

carcasses. Accordingly, the present study showed that a lot of infected pigs lose bacteria due

to careful evisceration. But these pigs remain seropositive and increase the herd prevalence

even if there is no risk of humans becoming infected. The herd prevalence determines the

classification of farms within quality assurance systems and is therefore of paramount

importance, especially for the fatteners. The present study showed that efforts to decrease the

herd prevalence should be focused on Salmonella entry and the transmission via effective

contacts between pigs. A conceivable measure to reduce Salmonella entry via farrowing sows

might be the vaccination of sows and piglets as well the removal of faeces from the pen.

Furthermore, the susceptibility of pigs to present bacteria should be reduced. Common

recommendations are the acidification of feed or water, rodent control, intensive cleaning and

disinfection, et cetera. Subsequent studies will expand the presented model to horizontal entry

of Salmonella and will analyse the effectiveness of several prevalence-reducing strategies.

5 Conclusion

The described model contrasts with other programs due to the trade relationships and

transmission paths considered. The results of the sensitivity analysis with the Plackett-

Burman design approve the model and emphasise the most important factors for Salmonella

transmission as well as their often non-linear relations. Slaughter pig prevalences emphasise

42

the impact of the sow on Salmonella entry and the subsequent transmission from pig to pig.

Furthermore, the developed model offers the possibility to evaluate and compare strategies to

decrease Salmonella prevalence in pigs.

References

Bailey, N., 1975. The Mathematical Theory of Infectious Diseases. London and High

Wycombe: Charles Griffin Company Ltd.

Barrentine, L.B., 1996. Illustration of confounding in Plackett-Burman designs. Quality

Engineering 9, 11-20.

Battenberg, L., 2007. Versuch der Eintragsquellenanalyse von Salmonellen in ausgewählten

bayerischen Schweinehaltungsbetrieben. Veterinary Faculty. Ludwig-Maximilians-

University, München.

BfR, 2005. Salmonella in pork – still a risk. Federal Institute of Risk Assesment (BfR),

http://www.bfr.bund.de/cd/6075 (last access: 4.12.2010).

Box, G.E.P., Hunter, W.G., Hunter, J.S., 1978. Statistics for Experimenters - An Introduction

to Design, Data Analysis, and Model Building. John Wiley & Sons New York.

Chung, C.A., 2004. Simulation modeling handbook: a practical approach. CRC Press Florida.

European Food Safety Authority, 2008. Report of the Task Force on Zoonoses Data

Collection on the analysis of the baseline survey on the prevalence of Salmonella in

slaughter pigs, Part A. The EFSA Journal 135, 1-111.

Hill, A.A., Snary, E.L., Arnold, M.E., Alban, L., Cook, A.J.C., 2008. Dynamics of Salmonella

transmission on a British pig grower-finisher farm: a stochastic model. Epidemiology

and Infection 136, 320-333.

Ivanek, R., Snary, E.L., Cook, A.J.C., Grohn, Y.T., 2004. A mathematical model for the

transmission of Salmonella Typhimurium within a grower-finisher pig herd in Great

Britain. Journal of Food Protection 67, 2403-2409.

Lurette, A., Belloc, C., Touzeau, S., Hoch, T., Ezanno, P., Seegers, H., Fourichon, C., 2008.

Modelling Salmonella spread within a farrow-to-finish pig herd. Veterinary Research

39, 49.

Montgomery, D.C., 2005. Design and Analysis of Experiments. John Wiley & Sons, Inc.

Arizona State University.

43

NAG, 2001. The NAG C Library Manual - Mark 7. Numerical Algorithms Group Ltd,

Oxford, UK.

Nollet, N., Houf, K., Dewulf, J., Duchateau, L., De Zutter, L., De Kruif, A., Maes, D., 2005.

Distribution of Salmonella strains in farrow-to-finish pig herds: A longitudinal study.

Journal of Food Protection 68, 2012-2021.

QS GmbH, 2009. Jagd auf Salmonellen wird verschärft. http://www.q-s.de/mediacenter/qs-in-

dem-medien/medienecho/ (last access: 22.10.2010).

Rubel, F., 2005. Process Models in Veterinary Epidemiology. Department of Natural Sciences

Vienna.

SAS, 2003. SAS 9.1 SAS Institute Inc. Cary, NC, USA.

van der Gaag, M.A., Vos, F., Saatkamp, H.W., van Boven, M., van Beek, P., Huirne, R.B.M.,

2004. A state-transition simulation model for the spread of Salmonella in the pork

supply chain. European Journal of Operational Research 156, 782-798.

Vanaja, K., Shobha Rani, R.H., 2007. Design of Experiments: Concept and Applications of

Plackett Burman design. Clinical Research and Regulatory Affairs 24, 1-23.

Vynnycky, E., White, R.G., 2010. An Introduction to Infectious Disease Modelling. Oxford

University Press New York.

Wehebrink, T., Kemper, N., Krieter, J., 2007. Simulation study on the epidemiology of

Salmonella spp. in the pork supply chain. Campylobacter spp., Yersinia spp. and

Salmonella spp. as Zoonotic Pathogens in Pig Production. Institute of Animal

Breeding and Husbandry, Christian-Albrechts-University, Kiel.

WHO, 2007. Food safety and foodborne illness. Fact sheet N°237. World Health

Organization, http://www.who.int/mediacentre/factsheets/fs237/en/ (last access:

9.8.2010).

WHO, 2010. The European health report 2009 : health and health systems. World Health

Organization, http://www.euro.who.int/__data/assets/pdf_file/0009/82386/E93103.pdf

(last access: 9.8.2010).

ZDS, 2009. erzeugerring.info. http://www.erzeugerring.info

44

45

CHAPTER THREE

Salmonella control measures with special focus on

vaccination and logistic slaughter procedures

Stefanie Hotes, Imke Traulsen, Joachim Krieter

Institute of Animal Breeding and Husbandry

Christian-Albrechts-University

24098 Kiel, Germany

Article accepted to Transboundary and Emerging Diseases (Wiley-Blackwell)

46

Summary

The present study focussed on the effectiveness of Salmonella control measures to decrease

Salmonella prevalence at slaughter. Considered measures were the control of hygiene and

husbandry management as well as vaccination and logistic slaughter procedures. Results

emphasised the capabilities of the farrowing stage to influence slaughter pig prevalence.

Limited Salmonella entry by the implementation of hygiene control measures at farrowing

farms obtained a significant decrease in prevalence after lairage at slaughterhouse. In contrast,

hygiene control measures at finishing stage were less effective. Husbandry control measures,

preventing physical contacts between pigs, were proved to decrease slaughter pig prevalence

whether they were implemented at farrowing or finishing stage. Furthermore, the vaccination

of sows and piglets was an appropriate control measure to decrease slaughter pig prevalence,

if a large part of farms established this control measure. Simultaneous implementation of

control measures showed that vaccination and especially hygiene measures are mutually

supportive. Concerning logistic slaughter procedures it became obvious that with decreasing

prevalence, infections at transport and lairage become more and more important. The herd

status separation significantly decreased the percentage of infected pigs which became

infected at lairage.

Keywords: Salmonella, control measures, pig; vaccination, logistic slaughter

47

1 Introduction

Human salmonellosis is one of the most important food-borne diseases worldwide (WHO,

2005). The European Food Safety Authority (European Food Safety Authority, 2008)

reported that one in ten pigs is infected with Salmonella in Europe. The serotype detected

most frequently in pork was Salmonella typhimurium, which is the second most frequent

cause of human salmonellosis in Europe (Galanis et al., 2006). The European regulation for

the control of Salmonella and other specified food-borne zoonotic agents (EC No 2160/2003)

states that ‘The protection of human health against diseases and infections transmissible

directly or indirectly between animals and humans (zoonoses) is of paramount importance’.

Funk and Gebreyes (2004) summarised the results of several risk factor analyses and

specified eleven categories of Salmonella control aspects: humans as vectors, flooring types,

housing contamination, pig flow management, sow-to-pig transmission, other vertebrate

species, intervertebrate species, risk factors associated with feed, environmental aspects,

stocking density and the marketing group effect as well as the general state of health. These

clusters describe on the one hand risk factors associated with the entry of Salmonella to the

farm (hygiene aspects), and on the other hand risk factors associated with the spread of

bacteria within the farm (husbandry aspects). In addition to hygiene and husbandry control

measures, the vaccination of sows and piglets is proposed as a possibility to decrease

transmission. Springer et al. (2001) described that vaccinated pigs had a significantly lower

colonisation of the ileal and caecal mucosa than unvaccinated pigs. In addition to these

measures which aim to decrease infection at farm, logistic slaughter procedures should limit

Salmonella transmission between farms during transport or slaughter. Stärk et al. (2002)

highlighted that transport and lairage are related to important risk potential. Transport and

lairage represent the only infection source for pigs coming from Salmonella-free farms.

Hence, to analyse the relevance of control measures for public health, it is not sufficient to

analyse effectiveness on only one farm with special characteristics.

The current study estimated the effectiveness of control measures to decrease Salmonella

prevalence at slaughter within a trade network. This enabled the consideration of different

combinations of farm characteristics and trade relationships. Special regard was given to

logistic slaughter procedures as well as to the capabilities of vaccination and corresponding

interactions.

48

2 Materials and methods

2.1 The simulation model

The simulation model used described the spread of non-clinical Salmonella (henceforth

referred to as “infection”) within the pork supply chain. In addition to other simulation

models (Ivanek et al., 2004; Hill et al., 2008; Lurette et al., 2008), trade relations between

farrowing and finishing farms were considered. The model distinguished between

conventional farrowing farms with supply relationships to one, two or three finishing farms

and farrow-to-finishing farms without supply relationships to a specialised fattening farm. At

the finishing stage, fatteners buying pigs from one, two or three conventional farrowing farms

were considered. Furthermore, routing for the transport of slaughter pigs was included. From

birth at farrowing farm until lairage at slaughterhouse, pigs were defined by one of four

mutually exclusive health states. All piglets were born susceptible but infection was

immediately possible. After effective contact, an infected pig started Salmonella shedding.

Shedding duration was assumed to last four weeks on average. Two weeks post-infection

seroconversion finished and pigs remained seropositive until death. After the shedding period

pigs were assumed to bear Salmonella, which contained the risk of shedding restart (Lurette et

al., 2008). The risk that a Salmonella-bearing pig restarts shedding increased if the pig s were

exposed to stress at transport. The probability of infection differed between farms according

to hygiene and husbandry of the farm. Furthermore, the probability of an effective contact

increased with an increasing number of infectious pigs within the compartment. Figure 1

illustrates that infection is possible at every stage of production, except for transport from

farrowing to fattening farm. The simulation model assumed an agglomerated pig producing

area with relatively short distances between farms. Hence, new infections will not occur until

fattening. In contrast, transport to slaughterhouse was assumed to take considerably longer.

Hurd et al. (2002) demonstrated that rapid infection during transport is a major reason for

increased prevalence in market swine.

The health state of the pigs does not change post-slaughter. But slaughter pigs lose bacteria

due to careful evisceration or become contaminated on their skin surfaces due to the slaughter

sequence (direct contact of carcasses) or contaminated tools (cross-contamination).

The model output provided information about the percentage of pigs which became infected

or contaminated during several production sequences. Due to the fact that a recovery was

assumed to be unlikely, the prevalence after lairage represents the percentage of pigs which

became infected at a particular production stage (Figure 1). A detailed description of the

49

simulation model, the input parameters and the evaluation via sensitivity analysis is given by

Hotes et al. (2010b).

Pro

babi

lity

to b

eco

me

cont

amin

ated

Farrowing

Slaughter process

Transport

Lairage at slaughterhouse

Finishing

Percentage of pigs infected during farrowing

Percentage of carcasses contaminated during slaughter process

Percentage of pigs infected during transport

Percentage of pigs infected during time at lairage

Percentage of pigs infected during finishing

Prevalence after lairage

Pro

babi

lity

to b

eco

me

infe

cted

Pro

babi

lity

to b

eco

me

cont

amin

ated

Farrowing

Slaughter process

Transport

Lairage at slaughterhouse

Finishing

Percentage of pigs infected during farrowing

Percentage of carcasses contaminated during slaughter process

Percentage of pigs infected during transport

Percentage of pigs infected during time at lairage

Percentage of pigs infected during finishing

Prevalence after lairage

Pro

babi

lity

to b

eco

me

infe

cted

Figure 1: Overview of the production stages considered in the model as well as the related

output information

2.2 Salmonella control measures

In the present model four different control measures were considered: general hygiene

measures, husbandry control measures for piglets, growers and fattening pigs, vaccination of

sows and piglets as well as logistic slaughter procedures.

2.2.1 General hygiene measures

In general, hygiene measures aim especially at the decrease of Salmonella entry into barns.

Bacteria entry occurs via external sources such as rodents, birds, pets, equipment, staff,

previously housed pigs etc. Hence, hygiene measures summarise all measures restricting these

kinds of infiltration like pest control, no exchange of equipment between barns, minimised

access to barns or comprehensive cleaning. Depending on how many of these measures have

already been implemented, the probability of infection via external sources varies between

farms.

Figure 2 illustrates the three hygiene types of farms which were distinguished within the

current model. Hygiene types were modelled with the probability of infection via external

sources (pex). Probabilities were based on Lurette et al. (2008). Farms which are little

concerned with Salmonella entry, had the highest probability of transmission (pex = 0.0001).

50

Within the improved situation all these “high Salmonella entry” farms (farrowing or finishing

farms) were assumed to have implemented unexhausted hygiene control measures and

decrease thereby pex from 0.0001 to 0.00005 (Figure 2).

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Farrowing farmsdefault

Farrowing farmsimprovement

Fattening farmsdefault

Fattening farmsimprovement

Fra

ctio

n o

f fa

rms

acc

ord

ing t

o p

ex

0.0001

0.00005

0.00001

Figure 2: Fraction of farms according to the probability of infection due to external sources

(pex)

2.2.2 Husbandry control measures for piglets, growers and fattening pigs

Salmonella, which is already present at barn, has different transmission probabilities

depending on the husbandry conditions. For example: partially slatted floors increase the

resting time of faeces within reach of the pigs, which increases the risk for effective contacts.

Latticed pen partitions as well as regrouping pigs establish contacts and also increase the

probability of effective contacts. The current model distinguishes between contact-facilitating

and contact-limiting husbandry conditions. The corresponding probabilities are based on Hill

et al. (2008), who determined the probability of an effective contact within a pen (pw) with

0.0014 and between pens (pb) with 0.0009. Values were adjusted to facilitating conditions by

an enhancement of 25% and to limiting conditions by a reduction of 25%. By default, farms

were equally allocated to both groups. Within the improved situation all farrowing or

finishing farms were assumed to have transmission-limiting conditions (Figure 3).

pex:

51

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Farrowing farms

default

Farrowing farms

improvement

Fattening farms

default

Fattening farms

improvement

Fra

cti

on

of

farm

s a

cco

rd

ing

to

p

w a

nd

p b

0.00175 and0.001125

0.00105 and0.000675

Figure 3: Fraction of farms according to the probability of effective contact within a pen

(pw) and between pens (pb)

2.2.3 Vaccination of sows and piglets

The simulation of the vaccination was based on the assumption that farrowing farms with

high sow-herd prevalence use a vaccine to interrupt the chain of vertical transmission.

Furthermore it was assumed that antibodies induced by vaccination cannot be distinguished

from those induced after infection. Within model sows were vaccinated ante partum and the

piglets post partum. Reliable immunity for piglets was assumed to last until slaughter.

Therefore, the probability of effective contact was decreased by 20 % for vaccinated pigs.

With regard to vaccinated sows, it was assumed that a certain percentage of sows stop

Salmonella shedding due to vaccination. Two efficiencies of sow-immunisation were

regarded: 1) a decrease in Salmonella-shedding sows by 50% (low-efficiency) and 2) a

decrease in Salmonella-shedding sows by 90% (high-efficiency).

2.2.4 Logistic slaughter procedures

About 30% of the modelled slaughter pig transports considered pigs from more than one

farm. Salmonella bacteria had the chance to spread between pigs from different farms.

Furthermore, stress at transport was assumed to increase the probability to restart shedding for

carrier pigs. Hence, the probability for a susceptible pig to become infected at transport or

pw and pb:

52

lairage due to contact with a Salmonella excreting pig increased. Pigs from the same lorry

were slaughtered successively in random order. The risk for a carcass to become

contaminated was influenced by the Salmonella status of the previously slaughtered pigs. In

general, the idea of logistic slaughter procedures is to distinguish between groups of pigs with

low and high prevalence for transport and during slaughter to decrease Salmonella

transmission from pigs bearing Salmonella to susceptible pigs. Corresponding to the

recommendations of the German Salmonella monitoring system (QS), the model

distinguished between farms with more or less than 40% infected pigs (QS GmbH, 2010a).

Farms with more than 40% infected slaughter pigs were not transported and slaughtered until

all pigs of the low prevalence group were processed. Furthermore, the impact of a different

selectivity was estimated by distinguishing also between farms with more or less than 20%

infected pigs and more or less than 10% infected pigs prior to transport, respectively.

In addition to Salmonella prevalence after lairage, the percentage of infections during

transport to slaughter and during lairage were interesting with regard to logistic slaughter

procedures. Furthermore, the percentage of contaminated carcasses was analysed.

2.3 Scenarios

To evaluate the effect of the different control measures 21 scenarios were formulated (Table

1). All scenarios based on a simulation environment consisting of 50 farrowing farms with an

average herd size of 181 sows. About 25% of these farms were assumed to be farrow-to-

finishing farms. All other farrowing farms sold their pigs to a specialised finishing farm.

The basic scenarios (no 1-12) were used to assess the main effects of the four Salmonella

control measures. Each control measure was activated separately, meaning that all other

control measures were set to the default value. Only the hygiene management as well as the

husbandry situation were improved at farrowing and finishing stage simultaneously within

one scenario each. The allocation of the following values was transferred from the QS

classification of fattening farms (QS GmbH, 2009):

I. 80% farrowing farms with 1% of sows shedding Salmonella during nursing

II. 15% farrowing farms with 5% of sows shedding Salmonella during nursing

III. 5% farrowing farms with 10% of sows shedding Salmonella during nursing.

53

The subsequent scenarios (Table 1, no 13-21), focused on vaccination and their interactions

with hygiene and husbandry control measures. As the results show, vaccination was effective

at farms using this control measure. But due to the fact that these farms represent only 5% of

all farrowing farms, the effect did not became apparent regarding the prevalence within the

whole producers’ association. Hence, a simulation environment with more high-prevalence

farms was considered to analyse the interactions between vaccination and hygiene control

measures as well as between vaccination and husbandry control measures as follows:

I. 20% farrowing farms with 1% of sows shedding Salmonella during nursing

II. 30% farrowing farms with 5% of sows shedding Salmonella during nursing

III. 50% farrowing farms with 10% of sows shedding Salmonella during nursing.

The scenarios B_Def, B_Vac50% and B_Vac90% had to be repeated for the new simulation

environment to ensure that the results can be compared (-> I_Def, I_Vac50%, I_Vac90%).

Each scenario was repeated 30 times (Hotes et al., 2010b). Differences between scenarios

were compared via t-test. To meet multiple comparisons, the p-values and confidence limits

were accomplished with the Bonferroni adjustment. The significance level was set to 5%.

54

Table 1: Description of the simulated scenarios

Scenarios DescriptionFarrowing Finishing Farrowing Finishing 50% 90% 40% 20% 10%

Basic scenarios1

1 B_Def Default default default default default no no no no no2 B_HygFar Hygiene improvements at farrowing farms improved default default default no no no no no3 B_HygFin Hygiene improvements at finishing farms default improved default default no no no no no4 B_HygFar+Fin Hygiene improvements at farrowing and finishing farms improved improved default default no no no no no5 B_HusFar Transmisson improvements at farrowing farms default default improved default no no no no no6 B_HusFin Transmisson improvements at finishing farms default default default improved no no no no no7 B_HusFar+Fin Transmisson improvements at farrowing and finishing farms default default improved improved no no no no no8 B_Vac50% Vaccination (low-efficiency) default default default default yes no no no no9 B_Vac90% Vaccination (high-efficiency) default default default default no yes no no no

10 B_Log40% Logistic transport default default default default no no yes no no11 B_Log20% Logistic transport default default default default no no no yes no12 B_Log10% Logistic transport default default default default no no no no yes

Interaction scenarios2

13 I_Def Default default default default default no no no no no14 I_Vac50% Vaccination (low-efficiency) default default default default yes no no no no15 I_Vac50%+HygFar Hygiene improvements at vaccinating farrowing farms improved default default default yes no no no no16 I_Vac50%+HusFar Transmission improvements at vaccinating farrowing farms default default improved default yes no no no no17 I_Vac50%+

HygFar+HusFarHygiene and transmission improvements at vaccinating farrowing farms

improved default improved default yes no no no no

18 I_Vac90% Vaccination (high-efficiency) default default default default no yes no no no19 I_Vac90%+HygFar Hygiene improvements at vaccinating farrowing farms improved default default default no yes no no no20 I_Vac90%+HusFar Transmission improvements at vaccinating farrowing farms default default improved default no yes no no no21 I_Vac90%+

HygFar+HusFarHygiene and transmission improvements at vaccinating farrowing farms

improved default improved default no yes no no no

Hygiene measures Husbandry measures Vaccination Logistic slaughter

1Based on 50 farrowing-farms whereby 80% of farms have only 1% shedding sows, 15% of farms have 5% shedding sows, 5% of farms have 10% shedding sows 2Based on 50 farrowing-farms whereby 20% of farms have only 1% shedding sows, 30% of farms have 5% shedding sows, 50% of farms have 10% shedding sows

54

55

3 Results

The respective prevalence after lairage obtained by the 21 scenarios is shown in Figure 4 and

Figure 5. Results regarding the effect of logistic slaughter procedures on prevalence, on the

percentage of contaminated pigs as well as on the percentage of infected pigs during transport

and lairage are shown in Figures 6 to 8.

3.1 General hygiene measures

Results of the hygiene scenarios (B_HygFar, B_HygFin and B_HygFar+Fin) emphasised the

capabilities of the farrowing stage to influence slaughter pig prevalence (Figure 4). Scenario

B_HygFar assumed that all farrowing farms with high Salmonella entry via rodents, birds,

staff etc. limited this entry. Compared to the default scenario, prevalence decreased

significantly by about 2% points. A significant decrease could not be obtained via hygiene

improvements at finishing stage (B_HygFin). Also, the simultaneous improvement of hygiene

conditions at farrowing and fattening farms (B_HygFar+Fin) did not achieve an additional

decrease in prevalence compared to B_HygFar (Figure 4).

0%

2%

4%

6%

8%

10%

12%

14%

16%

18%

20%

B_Def B_Hyg B_Hyg B_Hyg B_Hus B_Hus B_Hus B_Vac B_Vac B_Log B_Log B_Log

Far Fin Far+Fin Far Fin Far+Fin 50% 90% 40% 20% 10%

Pre

va

len

ce

a abe b

c c

d

a ae a a a

Figure 4: Prevalence after lairage at slaughterhouse and confidence limits (different letters

indicate significant differences with p < 0.05)

56

3.2 Husbandry control measures for piglets, growers and fattening pigs

The importance of the farrowing stage to decrease Salmonella prevalence after lairage became

apparent once more (Figure 4) with regard to the husbandry control measures (B_HusFar,

B_HusFin and B_HusFar+Fin). The alteration from contact-facilitating to contact-limiting

husbandry conditions at farrowing farms decreased the prevalence after lairage by about 4%

(B_HusFar). But in contrast to the hygiene control measures, the same decrease was obtained

if husbandry control measures were implemented at finishing stage (B_HusFin). The

combination of improved husbandry conditions at farrowing and fattening farms

(B_HusFar+Fin) caused a further prevalence decrease; about one third to default.

3.3 Vaccination of sows and piglets

Regarding the vaccination scenarios of Figure 4 (B_Vac50% and B_Vac90%) it became

obvious that there was no significant decrease in prevalence compared to the default scenario

(B_Def). But if only those farms were considered which vaccinated their sows and piglets,

there was a prevalence decrease of 9% and 13% according to the scenario B_Vac50% and

B_Vac90% (results not presented). This effect did not become significant because of the

small number of farms which used vaccination in the basic simulation environment (Figure4).

The simulation environment with more high-prevalence farms was considered in order to

analyse the interactions of vaccination with other control measures. About 26% of all pigs of

this simulation environment were infected with Salmonella after lairage within the default

scenario (I_Def) (Figure 5). The scenario I_Vac50%, which assumes that 50% of the

Salmonella shedding sows stopped shedding, achieved a prevalence decrease of 5%. The

assumption that 90% of the Salmonella shedding sows stopped shedding yields a prevalence

decrease of a little more than 6% (I_Vac90%). The combination of vaccination and hygiene

measures at farrowing farms (I_Vac50%+HygFar and I_Vac90%+HygFar) resulted in the

same percentage of infected pigs as for the combination of vaccination and husbandry control

measures (I_Vac50%+HusFar and I_Vac90%+HusFar). The simultaneous implementation of

vaccination, hygiene and housing control measures showed a tendency to further decreases in

prevalence (I_Vac50%+HygFar+HusFar and I_Vac90%+HygFar+HusFar).

57

0%

5%

10%

15%

20%

25%

30%

I_Def I_Vac50% I_Vac50% I_Vac50% I_Vac50% I_Vac90% I_Vac90% I_Vac90% I_Vac90%

+HygFar +HusFar +HygFar +HygFar +HusFar +HygFar

+HusFar +HusFar

Pre

vale

nce

a

bc cd

defc

ef cdef

Figure 5: Prevalence after lairage at slaughterhouse and confidence limits within a

simulation environment with increased number of high sow-herd prevalence farms

(different letters indicate significant differences with p < 0.05)

3.4 Logistic slaughter procedures

Prevalences considering logistic slaughter procedures decreased with decreasing threshold

prevalence but were not significantly different from the default scenario (Figure 4). For

example, the difference between B_Def and B_Log10% represented only 0.35%. Figure 6

breaks down the average prevalence of all delivered pigs to the single results of the low- and

high-prevalence group.

Within the high-prevalence group of B_Log40% prevalence was about 67% but only 16% of

all farms belonged to this category. For comparison, the high-prevalence groups of

B_Log20% and B_Log10% obtained a prevalence of 53% and 44% and considered 27% and

36% of all farms, respectively. Regarding the percentage of contaminated carcasses (Figure

7), it became apparent that logistic slaughter procedures did not decrease the overall carcass

contamination significantly.

58

0%

10%

20%

30%

40%

50%

60%

70%

B_Def B_Log40% B_Log20% B_Log10%

Prev

ale

nce

All delivered pigs Low-prevalence group before transport High-prevalence group before transport

a a a a

b

c

d

e

f

g

Figure 6: Prevalence after lairage at slaughterhouse and confidence limits based on all

delivered pigs and separated into transport groups for logistic slaughter procedure

(different letters indicate significant differences with p < 0.05)

0%

2%

4%

6%

8%

10%

12%

B_Def B_Log40% B_Log20% B_Log10%

Percen

tag

e o

f co

nta

min

ate

d p

igs

All delivered pigs Low-prevalence group before transport High-prevalence group before transport

a a a a

b

c

d

e

f

g

Figure 7: Percentage of contaminated carcasses after slaughter procedures and confidence

limits based on all delivered pigs and separated into transport groups for logistic

slaughter procedure (different letters indicate significant differences with p < 0.05)

59

Figure 8 demonstrates the percentage of infected pigs which became infected during transport

and during lairage, respectively. It became apparent that the risk of becoming infected during

transport or lairage was relatively more important for the low-prevalence groups. For

example, within the scenario B_Log40%, 3% of all infected pigs of the low-prevalence group

had effective contact during transport while only 1% of the infected pigs of the high-

prevalence group became infected during transport. Furthermore, it can be shown that logistic

slaughter procedures decreased the proportion of infections during time spend at lairage

(Figure 8).

0%

1%

2%

3%

4%

5%

B_Def B_Log40% B_Log20% B_Log10%

All delivered pigs Low-prevalence group before transport High-prevalence group before transport

0%

1%

2%

3%

4%

5%

6%

7%

B_Def B_Log40% B_Log20% B_Log10%

All delivered pigs Low-prevalence group before transport High-prevalence group before transport

Per

cen

tage

of

infe

cted

pig

s w

hic

h b

ecam

e in

fect

ed a

t

lair

age

tran

sport

a af

b

c

af

d

eaf

d

f

abh

c

d

bh

e

f

g

b h

0%

1%

2%

3%

4%

5%

B_Def B_Log40% B_Log20% B_Log10%

All delivered pigs Low-prevalence group before transport High-prevalence group before transport

0%

1%

2%

3%

4%

5%

6%

7%

B_Def B_Log40% B_Log20% B_Log10%

All delivered pigs Low-prevalence group before transport High-prevalence group before transport

Per

cen

tage

of

infe

cted

pig

s w

hic

h b

ecam

e in

fect

ed a

t

lair

age

tran

sport

a af

b

c

af

d

eaf

d

f

abh

c

d

bh

e

f

g

b h

Figure 8: Percentage of infected pigs which became infected at transport or lairage and

confidence limits based on all delivered pigs and separated into transport groups

for logistic slaughter procedure (different letters indicate significant differences

with p < 0.05)

60

4 Discussion

The current study focussed on the impact of Salmonella control measures to decrease

prevalence after lairage at slaughter. Due to the fact that prevalence decrease at a single farm

does not meet public benefit, the impact of control measures was analysed considering several

farrowing and finishing farms with respective trade relationships. This allowed the

comparison between production stages with regard to effectiveness of control measures.

The implementation of the control measures to the simulation model was oriented towards

expert opinion but a complete insight regarding the impact of measures on the probability of

effective contacts was not obtained. Hence, the current study provides rather a relative

comparison than absolute statements about prevalence decrease.

4.1 General hygiene measures

The current analyses of hygiene control measures emphasised the capabilities of the farrowing

stage with regard to the decrease of slaughter pig prevalence. The limitation of Salmonella

entry at farrowing stage (B_HygFar) decreased the percentage of infected pigs after growing.

Hence, the dynamic of transmission during fattening was also decreased and fewer pigs

became infected. Transmission dynamics were also decreased if hygiene control measures

were implemented at fattening, but in this case, duration until slaughter was too short to

obtain a significant decrease in prevalence (B_HygFin). A previous study, which focussed on

vertical transmission, also emphasised the paramount importance of the infection rate at

farrowing stage (Hotes et al., 2010b). Ivanek et al. (2004) as well as van der Gaag et al.

(2004) confirmed that slaughter pig prevalence is influenced by the percentage of infected

pigs after farrowing stage. Furthermore, there is an international census about the importance

of general hygiene for Salmonella control (Stärk et al., 2002). In accordance with these

studies, the relevance of the farrowing stage as well as the importance of hygiene control

measures became obvious in the present study. Meerburg and Kijlstra (2007) recommended

that hygiene control measures should include pest control, control of wild birds and flies and

obligatory disinfection of boots/clothes and equipment.

4.2 Husbandry control measures for piglets, growers and fattening pigs

While hygiene conditions affect the entry of bacteria, husbandry aspects have an impact on

transmission between pigs. Analyses showed that in contrast to the hygiene measures,

61

husbandry improvements decreased the slaughter pig prevalence, regardless of whether they

were implemented at farrowing or fattening stage. Both resulted in a similar prevalence

decrease after lairage. Further reduction was obtained by implementing husbandry control

measures simultaneously at farrowing and finishing stage. Examples for prevalence-

decreasing housing conditions are slatted floors (Davies et al., 1997; Nollet et al., 2004;

Vonnahme et al., 2008; Hotes et al., 2010a) or pen divisions without open spaces (Lo Fo

Wong et al., 2004). These measures restrict contacts with contaminated faeces or with

infectious pigs of adjacent pens, respectively.

It is generally accepted that the acidification of feed or the use of wet feeding systems prevent

effective contacts (Jørgensen et al., 2001; van der Wolf et al., 2001; Belœil et al., 2004;

Bahnson et al., 2006; Farzan et al., 2006; Benschop et al., 2008). Acidification and wet

feeding systems do not prevent contacts between infectious and susceptible pigs per se but

decrease the probability of infection due to physical contact. These control measures are not

directly represented in the present simulation, but additional analyses showed that control

measures such as acid or wet feeding systems affect prevalence after lairage in a similar

manner to husbandry control measures. There was no significant difference whether control

measures were implemented at farrowing or finishing stage, whereas the combined

implementation at farrowing and finishing stage obtained further decrease in slaughter pig

prevalence.

4.3 Vaccination of sows and piglets

The results show that the vaccination of sows and piglets is an appropriate control measure to

decrease slaughter pig prevalence after lairage. But if a whole producers’ association is

considered, the benefit depends on how many farms use the vaccine. The widespread use of

vaccination is questionable as long as antibodies induced by vaccination cannot be

distinguished from those induced by field strains. Whilst a differentiation is not possible, only

farrowing farms with a high-than-average prevalence will probably be willing to vaccinate

sows and piglets to interrupt the chain of infection. Figure 5 shows that a successful

immunisation of 90% of the respective sows (I_Vac90%) decreased the slaughter pig

prevalence just a little more than the successful immunisation of only 50% of sows

(I_Vac50%). It can be presumed that the immune reaction of the piglets limited the vertical

transmission even for low effectiveness in the sows’ immunisation. This effect will not arise

if the serotype of the vaccine differs from the serotype shed by the sows. The slight

62

superiority of the 90% immunisation (I_Vac90%) compared to the 50% immunisation of the

sows (I_Vac50%) became totally lost if vaccination was combined with husbandry control

measures or with husbandry and hygiene control measures (I_Vac50%+HusFar and

I_Vac90%+HusFar or I_Vac50%+HygFar+HusFar and I_Vac90%+HygFar+HusFar). Due to

the combination with immunisation, hygiene measures became relatively more preferable

compared to husbandry control measures. The basic scenarios B_HygFar and B_HusFar

suggested that husbandry control measures are more effective than hygiene measures. This

difference was nullified in combination with vaccination. There was no significant difference

in prevalence decrease if hygiene control measures were combined with vaccination or if

husbandry control measures were combined with vaccination (I_Vac50%+HygFar and

I_Vac50%+HusFar as well as I_Vac90%+HygFar and I_Vac90%HusFar). One reason might

be that both immunity and husbandry control measures affect pig-to-pig transmission, which

decreased the marginal effectiveness of husbandry control measures.

The opportunity to vaccinate fattening pigs was not considered in the present study. The

assumption that antibodies induced by vaccination cannot be distinguished from those

induced after infection restricts the application on fattening pigs.

4.4 Logistic slaughter procedures

Logistic slaughter procedures are applied for poultry slaughterhouses participating in the

German Salmonella monitoring program QS (QS GmbH, 2010b) and are considered as

appropriate measures to decrease effective contacts between pigs during transport (QS GmbH,

2010a). Hurd et al. (2002) demonstrated that rapid infection during transport and holding is a

major reason for increased prevalence in pigs. But the present study did not show that

prevalence after lairage or the percentage of contaminated pigs could be significantly

decreased by logistic slaughter procedures (Figure 6 and 7). This was also the case, if the

number of slaughter pig transports considering more than one farm was increased from 30%

to 87% (result not presented). One reason might be that even a threshold prevalence of 10%

was too high to separate the better-than-average farms. More than 60% of farms had a pig

herd prevalence smaller than 10%. Another reason could be that the probability for effective

contact at transport was underestimated. Further research will consider these possibilities.

Nevertheless, the present study showed that the herd status separation significantly decreased

the risk of becoming infected during lairage at slaughterhouse (Figure 8). Pens at lairage were

less contaminated due to previously slaughtered pigs if slaughter groups were sorted by

63

prevalence. Similar results were obtained by Swanenburg et al. (2001), who demonstrated that

contamination of carcasses after slaughter was frequently caused by Salmonella-infected

herds that had been slaughtered beforehand. Furthermore, it could be shown that with

decreasing prevalence infections at transport and lairage became more and more important.

References

Bahnson, P.B., Fedorka-Cray, P.J., Ladely, S.R., Mateus-Pinilla, N.E., 2006. Herd-level risk

factors for Salmonella enterica subsp. enterica in U.S. market pigs. Preventive

Veterinary Medicine 76, 249-262.

Belœil, P.-A., Fravalo, P., Fablet, C., Jolly, J.P., Eveno, E., Hascoet, Y., Chauvin, C., Salvat,

G., Madec, F., 2004. Risk factors for Salmonella enterica subsp. enterica shedding by

market-age pigs in French farrow-to-finish herds. Preventive Veterinary Medicine 63,

103-120.

Benschop, J., Stevenson, M.A., Dahl, J., French, N.P., 2008. Towards incorporating spatial

risk analysis for Salmonella sero-positivity into the Danish swine surveillance

programme. Preventive Veterinary Medicine 83, 347-359.

Davies, P.R., Morrow, W.E.M., Jones, F.T., Deen, J., Fedorka-Cray, P.J., Harris, I.T., 1997.

Prevalence of salmonella in finishing swine raised in different production systems in

North Carolina, USA. Epidemiology and Infection 119, 237-244.

European Food Safety Authority, 2008. Report of the Task Force on Zoonoses Data

Collection on the analysis of the baseline survey on the prevalence of Salmonella in

slaughter pigs, Part A. The EFSA Journal 135, 1-111.

Farzan, A., Friendship, R.M., Dewey, C.E., Warriner, K., Poppe, C., Klotins, K., 2006.

Prevalence of Salmonella spp. on Canadian pig farms using liquid or dry-feeding.

Preventive Veterinary Medicine 73, 241-254.

Funk, J., Gebreyes, W.A., 2004. Risk factors associated with Salmonella prevalence on swine

farms. Journal of Swine Health and Production 12, 246-251.

Galanis, E., Lo Fo Wong, D.M.A., Patrick, M.E., Binsztein, N., Cieslik, A., Chalermchaikit,

T., Aidara-Kane, A., Ellis, A., Angulo, F.J., Wegener, H.C., 2006. Web-based

Surveillance and Global Salmonella Distribution, 2000-2002. Emerging Infectious

Diseases 12, 381-388.

64

Hill, A.A., Snary, E.L., Arnold, M.E., Alban, L., Cook, A.J.C., 2008. Dynamics of Salmonella

transmission on a British pig grower-finisher farm: a stochastic model. Epidemiology

and Infection 136, 320-333.

Hotes, S., Kemper, N., Traulsen, I., Rave, G., Krieter, J., 2010a. Risk Factors for Salmonella

Infection in Fattening Pigs - An Evaluation of Blood and Meat Juice Samples.

Zoonoses and Public Health 57, (Suppl. 1) 30-38.

Hotes, S., Traulsen, I., Krieter, J., 2010b. An individual-based model for Salmonella

transmission along the pig production chain. Submitted to Research in Veterinary

Science.

Hurd, H.S., McKean, J.D., Griffith, R.W., Wesley, I.V., Rostagno, M.H., 2002. Salmonella

enterica Infections in Market Swine with and without Transport and Holding. Applied

and Environmental Microbiology 68, 2376-2381.

Ivanek, R., Snary, E.L., Cook, A.J.C., Grohn, Y.T., 2004. A mathematical model for the

transmission of Salmonella Typhimurium within a grower-finisher pig herd in Great

Britain. Journal of Food Protection 67, 2403-2409.

Jørgensen, L., Kjærsgaard, H.D., Wachmann, H., Jensen, B.B., Knudsen, K.E.B., 2001. Effect

of pelleting and use of lactic acid in feed on Salmonella prevalence and productivity in

weaners. In, 4th International Symposium on the Epidemiology and Control of

Salmonella and other Food Borne Pathogens in Pork - Salinpork, Leipzig, pp. 109-

111.

Lo Fo Wong, D.M.A., Dahl, J., Stege, H., van der Wolf, P.J., Leontides, L., von Altrock, A.,

Thorberg, B.M., 2004. Herd-level risk factors for subclinical Salmonella infection in

European finishing-pig herds. Preventive Veterinary Medicine 62, 253-266.

Lurette, A., Belloc, C., Touzeau, S., Hoch, T., Ezanno, P., Seegers, H., Fourichon, C., 2008.

Modelling Salmonella spread within a farrow-to-finish pig herd. Veterinary Research

39, 49.

Meerburg, B.G., Kijlstra, A., 2007. Role of rodents in transmission of Salmonella and

Campylobacter. Journal of the Science of Food and Agriculture 87, 2774-2781.

Nollet, N., Maes, D., De Zutter, L., Duchateau, L., Houf, K., Huysmans, K., Imberechts, H.,

Geers, R., de Kruif, A., Van Hoof, J., 2004. Risk factors for the herd-level

bacteriologic prevalence of Salmonella in Belgian slaughter pigs. Preventive

Veterinary Medicine 65, 63-75.

QS GmbH, 2009. Jagd auf Salmonellen wird verschärft. http://www.q-s.de/mediacenter/qs-in-

dem-medien/medienecho/ (last access: 22.10.2010).

65

QS GmbH, 2010a. Programme zum Monitoring und zur Reduzierung von

lebensmittelassoziierten Zoonoseerregern im Rahmen des QS-Prüfzeichens: I.

Salmonellenmonitoring und -reduzierungsprogramm für die

Schweinefleischerzeugung.

http://www.q-s.de/fileadmin/QS_Fileadmin/downloads_v2/produktbereich_fleisch_

und_fleischwaren_inkl_futtermittel/schlachtung_zerlegung/salmonellenmonitoring_sc

hwein/leitfaden_salmonellenmonitoring_schwein/lf_salmo_sw_d_frei_100101.pdf

(last access: 22.10.2010).

QS GmbH, 2010b. „QS über Gesetz“ Anforderungen des QS-Systems, die über die

gesetzlichen Bestimmungen hinausgehen.

http://www.q-s.de/fileadmin/QS_Fileadmin/Mediacenter/kriterien_ueber_gesetz/

QS_Kriterien_ueber_Gesetz_fina0110.pdf (last access: 19.10.2010).

Springer, S., Lindner, T., Steinbach, G., Selbitz, H.J., 2001. Investigation of the efficacy of a

genetically-stabile live Salmonella Typhimurium vaccine for use in swine. Berliner

und Münchener Tierärztliche Wochenschrift 114, 342-345.

Stärk, K.D.C., Wingstrand, A., Dahl, J., Møgelmose, V., Lo Fo Wong, D.M.A., 2002.

Differences and similarities among experts' opinions on Salmonella enterica dynamics

in swine pre-harvest. Preventive Veterinary Medicine 53, 7-20.

Swanenburg, M., van der Wolf, P.J., Urlings, H.A.P., Snijders, J.M.A., van Knapen, F., 2001.

Salmonella in slaughter pigs: the effect of logistic slaughter procedures of pigs on the

prevalence of Salmonella in pork. International Journal of Food Microbiology 70,

231-242.

van der Gaag, M.A., Vos, F., Saatkamp, H.W., van Boven, M., van Beek, P., Huirne, R.B.M.,

2004. A state-transition simulation model for the spread of Salmonella in the pork

supply chain. European Journal of Operational Research 156, 782-798.

van der Wolf, P.J., Wolbers, W.B., Elbers, A.R.W., van der Heijden, H.M.J.F., Koppen,

J.M.C.C., Hunneman, W.A., van Schie, F.W., Tielen, M.J.M., 2001. Herd level

husbandry factors associated with the serological Salmonella prevalence in finishing

pig herds in The Netherlands. Veterinary Microbiology 78, 205-219.

Vonnahme, J., Kreienbrock, L., Beilage, E.g., 2008. Untersuchungen zur Identifikation von

Risikofaktoren für die Ausbreitung von Salmonellen in Aufzuchtbeständen für

Jungsauen. Berliner und Münchener Tierärztliche Wochenschrift 121, 33-40.

WHO, 2005. Drug-resistant Salmonella. Fact sheet N°139. World Health Organization,

http://www.who.int/mediacentre/factsheets/fs139/en/ (last access: 6.11.2010).

66

67

CHAPTER FOUR

The additional costs of logistic slaughter procedures to

decrease Salmonella prevalence in pork

Stefanie Hotes, Imke Traulsen, Joachim Krieter

Institute of Animal Breeding and Husbandry

Christian-Albrechts-University

24098 Kiel, Germany

68

Abstract

The current study focussed on the additional costs related to logistic slaughter procedures.

This control measure was assumed to be implemented within a producers’ association to

decrease Salmonella prevalence in pork. Calculations were based on the additional tours

caused by the separate transport of low- and high-prevalence farms and on the additional

transport distance caused by changed routing. The results showed that there is not necessarily

a considerable increase in the number of tours due to herd status separation for transport. The

number of routes changed due to logistic slaughter procedures varied between 43% and 69%

depending on the respective threshold prevalence. The additional costs per slaughtered pig

varied between 0.07€/pig and 0.58€/pig under the given assumptions. Costs were governed by

the percentage of changed routes and the additional distance of a changed route. Due to the

fact that the percentage of changed routes is related to the distribution of herd prevalence

within the producers’ association, there is no reasonable threshold in general. The current

study enables producers’ associations to evaluate the additional costs of logistic slaughter

procedures for their members.

Keywords: Salmonella, control measure, logistic slaughter, costs, pork

69

1 Introduction

Human salmonellosis is one of the most important food-borne diseases worldwide (WHO,

2005). Cases are frequently related to pork consumption. According to estimates by the

European Food Safety Association (EFSA), about 10% of slaughter pigs are infected with

Salmonella in Europe (European Food Safety Authority, 2008). Differences in prevalence

between countries are high. The EFSA found no pigs infected with Salmonella in lymph

nodes in Finland, whereas in Spain the prevalence was about 29%. Several studies reveal the

importance of transport to slaughter and lairage before slaughtering for prevalence in pigs and

pork. Berends et al. (1996) stated that the percentage of animals excreting Salmonella at the

end of the finishing period can double during transport and lairage. This is in accordance with

Hurd et al. (2002), who demonstrated that rapid infection during transport and holding is a

major reason for increased prevalence in swine. Especially the time spent in lairage before

slaughtering seems to play a crucial role in infection and contamination (Swanenburg et al.,

2001a; Belœil et al., 2004). Furthermore, if the proportion of Salmonella-free farms increases,

transport and lairage will be the only remaining infection sources for pigs originated from

these farms (Stärk et al., 2002). Hence, the most effective strategies reducing Salmonella

prevalence in pork have to include transportation and lairage (van der Gaag et al., 2004).

Logistic slaughter procedures affect transport and lairage due to herd separation according to

prevalence before transport and the preferred slaughtering of low-prevalence herds at the

beginning of slaughter day. Previous studies have shown that logistic slaughter procedures

have the capability to reduce Salmonella infection at lairage (Hotes et al., 2010) and are useful

to decrease the prevalence of Salmonella-contaminated pork after slaughter (Swanenburg et

al., 2001b). But whether logistic slaughter procedures are to implement as Salmonella control

measure depends not at least on the additional costs incurred by the separate transportation of

pigs from low- and high-prevalence farms. Hence, the objective of the present study was to

evaluate the additional costs related to logistic slaughter procedures.

70

2 Materials and methods

2.1 Applied simulation model

The present study was based on a stochastic simulation model which describes the spread of

non-clinical Salmonella within the pork supply chain within one year. The model considered a

trade network consisting of 50 farrowing farms with a mean of 180 sows per farm. The

simulation distinguished between farrow-to-finishing farms, conventional farrowing farms

and fattening farms. On average, 13 farrow-to-finishing farms and 45 specialised finishing

farms fattened the pigs to slaughter weight. All pigs were slaughtered at a central

slaughterhouse with 27 weeks of age. Within one year, approximately 227,800 pigs reached

slaughter age. Transport to slaughterhouse considered optimal routing, meaning that a lorry

loaded pigs from up to three farms if there was enough capacity (maximal loading capacity

was 180 pigs). Due to the fact that the simulation model did not consider different daily

growth, the average number of delivered pigs per farm was relatively high with 138 slaughter

pigs. Hence, only 30% of the tours considered more than one farm.

2.2 Logistic slaughter procedures

To estimate the impact of logistic slaughter procedures, herd separation according to

prevalence for transport and slaughter was implemented in the model. Pigs from low-

prevalence farms were slaughtered at the beginning of the slaughter day to prevent a

contamination of the slaughter line and the equipment. Three different threshold prevalences

were used to distinguish between low- and high-prevalence farms: (1) 40% at the scenario

Log40%, 20% at the scenario Log20% and 10% at the scenario Log10%. Slaughter pigs of

farms categorised as low-prevalence farms were not allowed to be transported with pigs of

high-prevalence farms. Table 1 represents the percentage of farms categorised as low- or

high-prevalence in accordance with the respective threshold prevalence.

Table 1: Percentage of farms categorised as low-or high-prevalence in accordance with the

respective threshold prevalence

Log40% Log20% Log10%Percentage of low-prevalence farms 84 73 64Percentage of high-prevalence farms 16 27 36

71

2.3 Calculation

To evaluate the impact of logistic slaughter procedures three questions were of interest:

1. Do logistic slaughter procedures increase the number of tours due to a decreased load

of the pig transporters?

2. How many routes will be changed due to separate transport?

3. What are the additional costs of logistic slaughter procedures?

To determine the number of additional tours, three simulation scenarios were analysed

considering threshold prevalences of 40% (scenario: Log40%), 20% (scenario: Log20%) and

10% (scenario: Log10%), respectively. The number of tours necessary to transport all pigs to

slaughterhouse was compared with the default scenario (without logistic transport) via t-test.

To meet multiple comparisons, the p-values were accomplished with the Bonferroni

adjustment. The significance level was set to 5%.

The percentage of routes which changed due to logistic slaughter procedures was assessed by

comparing the number of all routes possible with and without separate transport. Herd

separation according to prevalence caused the number of possible routes to decrease.

Established tours were also affected and alternative touring had to be established.

* 100Changed routes (%) = 1 -

Number of routes possible without logistic transport

Number of routes possible considering all low-prevalence farms

Number of routes possible considering all high-prevalence farms

+

* 100Changed routes (%) = 1 -

Number of routes possible without logistic transport

Number of routes possible considering all low-prevalence farms

Number of routes possible considering all high-prevalence farms

+

Changed routes (%) = 1 -

Number of routes possible without logistic transport

Number of routes possible considering all low-prevalence farms

Number of routes possible considering all high-prevalence farms

+

Usually, the alternative tours are of longer distance. So, to calculate the related costs,

assumptions about the additional distance had to be made. To account for different structures

of producers’ associations additional distances between 20km and 100km were assumed.

Normally, if the density of the association is high, distances between farms are short and

changed routes increase the km driven only moderately. In contrast, within a producers’

association with a low density of farms, the km driven increase much more. For the initial

calculation of the additional costs, an average fuel consumption of 35l/100 km was assumed.

The fuel price was set to 1.28€/l which represents the German price level of 2008

(Bundesverband Güterkraftverkehr Logistik und Entsorgung (BGL) e.V. , 2010). The hourly

wages of the truck driver were estimated at 14.29€/h considering incidental wage costs of

32% (Vereinigte Dienstleistungsgewerkschaft Nordrhein-Westfalen, 2009; Statistisches

72

Bundesamt Deutschland, 2010). Furthermore, depreciation, interest and repairs were

calculated approximately at 0.52€/km.

To gain an impression of the input-output relation, costs were also calculated considering a

fuel price of 1.50€/l. Furthermore, additional costs of logistic slaughter procedures were

calculated for a producers’ association consisting of twice as many high-prevalence farms.

3 Results

The average numbers of pigs per lorry as well as the average numbers of tours necessary to

transport all pigs to slaughterhouse are presented in Table 2. Additional costs of logistic

slaughter procedures considering the initial relation of low- and high-prevalence farms and a

fuel price of 1.28€/l (initial situation) are shown in Table 3. Figure 1 displays the additional

costs per pig considering an increased fuel price of 1.50€/l or an increased percentage of high-

prevalence farms in relation to the initial calculation.

The results show that logistic slaughter procedures lowered the average load of a transport

only moderately (Table 2). The load decreased by between one and three pigs per transporter,

which was associated with 17 to 28 additional tours. This represented barely more than 0.5

additional tours per week. Hence, the number of additional tours was regarded as negligible

for the calculation of costs. In contrast, many routes had to be changed due to the

implementation of logistic slaughter procedures. The separation of low- and high-prevalence

farms for transport to slaughter changed 43% to 69% of routes depending on the respective

threshold prevalence (Table 2).

Table 2: Impact of logistic slaughter procedures on transport-related parameters (different

letters indicate significant differences within parameter with p < 0.05)

Default (without logistic transport) 149a 1,529a

Logistic transport: Log40% 147b 1,546b 17 43 654

Log20% 146c 1,554c 25 60 915

Log10% 146c 1,557c 28 69 1,058

number of pigs per transporter

percentage of changed routes

changd routes per year

number of tours per year

additional tours per year

The results of the monetary evaluation of the changed routes are presented in Table 3. The

relation between additional distance and additional costs was linear for each of the three

logistic scenarios.

73

Table 3: Additional costs of separate transport depending on threshold prevalence and additional distance

km/tour km/year €/year €/year €/year €/year €/pig €/'low pig'1 €/'high pig'2

Log40% 20 13,071 5,854 3,736 6,767 16,357 0.07 0.09 0.4340 26,141 11,709 7,471 13,534 32,714 0.14 0.17 0.8760 39,212 17,563 11,207 20,301 49,071 0.22 0.26 1.3080 52,283 23,418 14,942 27,068 65,428 0.29 0.34 1.73

100 65,354 29,272 18,678 33,834 81,785 0.36 0.43 2.16Log20% 20 18,299 8,196 5,230 9,474 22,900 0.10 0.14 0.36

40 36,598 16,392 10,460 18,947 45,799 0.20 0.28 0.7360 54,897 24,589 15,690 28,421 68,699 0.30 0.42 1.0980 73,196 32,785 20,919 37,895 91,599 0.40 0.56 1.45

100 91,495 40,981 26,149 47,368 114,498 0.50 0.70 1.82Log10% 20 21,158 9,477 6,047 10,954 26,478 0.12 0.18 0.32

40 42,316 18,954 12,094 21,908 52,956 0.23 0.36 0.6560 63,475 28,431 18,141 32,862 79,433 0.35 0.54 0.9780 84,633 37,907 24,188 43,816 105,911 0.46 0.73 1.29

100 105,791 47,384 30,235 54,770 132,389 0.58 0.91 1.62

fuel costs driver costs additional total costsdepreciation,

interest, repairsadditional km

1 Additional costs divided by all pigs delivered from farms categorised as low-prevalence farm (farm prevalence ≤ threshold prevalence)

2 Additional costs divided by all pigs delivered from farms categorised as high-prevalence farm (farm prevalence > threshold prevalence)

73

74

Highest costs were obtained if the threshold prevalence was set to 10% (Log10%). The costs

per pig varied between 0.07€/pig and 0.58€/pig. Table 3 also includes the costs per ‘low pig’

and ‘high pig’, which represents that the additional costs have to be paid by the low-

prevalence farms or the high-prevalence farms, respectively. In these cases the additional

costs per ‘low pig’ varied between 0.09 €/pig and 0.91€/pig and additional costs per ‘high

pig’ varied between 0.32 €/pig and 2.16 €/pig.

Figure 1 illustrates the additional costs of logistic slaughter procedures assuming an additional

distance of 60km. Regarding the initial situation (fuel price of 1.28€/l and initial number of

high-prevalence farms) and the calculation considering an increased fuel price, it becomes

obvious that costs per pig increased by 6% independent of the threshold prevalence. Hence,

both calculations obtained the highest costs at the 10% threshold. In contrast, if a producers’

association with twice as many high-prevalence farms was considered, the lowest costs were

obtained at the 10% threshold prevalence (Figure 1). Furthermore, there was no continuous

trend if the number of high-prevalence farms was increased. Based on the threshold

prevalence of 40%, the 20% threshold increased the costs per pig while the 10% threshold

decreased the costs per pig.

0.200.220.240.260.280.30

0.320.340.360.380.40

40% 20% 10%

threshold prevalence

Ad

dit

ion

al

cost

s of

logis

tic

slau

gh

ter

pro

ced

ure

s p

er

slau

gh

ter

pig

(€/p

ig)

initial situation increased fuel price more high-prevalence farms

Figure 1: Additional costs of logistic slaughter procedures according to the initial situation,

an increased fuel price and twice as many high-prevalence farms. Calculation

considered an additional distance of 60km per changed tour.

75

4 Discussion

Previous studies showed that logistic slaughter procedures have the capability to reduce

Salmonella infection at lairage (Hotes et al., 2010) and to decrease the prevalence of

Salmonella-contaminated pork after slaughter (Swanenburg et al., 2001b).

The objective of the present study was to calculate the additional costs of logistic slaughter

procedures within a producers’ association. Optimal routing was assumed within the

simulation model regardless of whether herd separation according to prevalence was included

or not. It is doubtful whether optimal routing is always applied within reality. Certainly

routing considers further constrains in reality. But this assumption was necessary to estimate

the true costs of logistic slaughter procedures.

Results regarding the additional costs per pig varied between 0.07 and 0.58 €/pig. If costs

were allocated among pigs from low- or high-prevalence farms, the additional costs increased

due to the decreasing number of pigs. Van der Gaag et al. (2004) considered additional costs

of 0.62€/pig to evaluate the cost-effectiveness of the logistic supply. The present study

estimated lower costs if all pigs were considered. To improve data comparison, information

on threshold prevalence, assumed additional tours as well as information on extension of tours

would be necessary. The simulation model used in the present study allowed the estimation of

additional tours caused by the separate transport of low- and high-prevalence pigs. Results

showed that the number of additional tours is associated with the threshold prevalence. The

maximum of additional tours was obtained at the threshold prevalence of 10%. This scenario

resulted in 28 additional tours, which represented an increase of 1.8%. This is such a small

increase that it can be concluded that the respective costs are negligible. Whether this is also

true in practice depends on the quality of routing and distances between farms. The simulation

model could not consider the location of farms. Hence, an important constraint for routing

was not taken into account. If distances between members of the producers’ association are

homogeneous, simulation error will be small. But if there are single farms which are far away

from the majority of members, additional tours will probably increase much more. To take

this into account, a wide range of additional distance due to changed routing was considered.

Within the presented simulation environment about 30% of tours considered more than one

farm and the number of tours increased by 1.8% due to separate transport. Within a different

simulation environment considering that about 87% of transporter loaded pigs from more than

one farm (number of delivered pigs per fattening farm decreased from 138 pigs to 65 pigs on

76

average), the number of additional tours increased by 3.8% (result not presented). But this

still represents a very small increase.

The percentage of changed routes varied between 43% and 69% depending on the threshold

prevalence. The results showed: the smaller the threshold, the higher the percentage of

changed routes. But this is not a general relation. Decreasing the threshold implies that more

and more farms are categorised as high-prevalence farms. Once half of the farms are

categorised as high-prevalence farms, decreasing the threshold further decreases the number

of changed routes. This happened within the present study as twice as many high-prevalence

farms were considered (Figure 1). The highest percentage of changed routes was reached at

threshold prevalences between 40% and 20%. The 20% threshold categorised more than 50%

of the farms as high-prevalence farms. Hence, the further decrease in the threshold prevalence

from 20% to 10% decreased the percentage of changed routes and consequently the additional

costs. Generally speaking, the most cost-effective threshold prevalence is always associated

with the distribution of the herd prevalence. This implies that it is impossible to specify a

general, most cost-effective threshold prevalence. Furthermore, changes within a producers’

association affecting the relative frequencies of low- and high-prevalence farms can both

increase or decrease the costs of logistic slaughter procedures.

5 Conclusion

The additional costs of logistic slaughter procedures were calculated to range between

0.07€/pig and 0.58€/pig. Within the present study the number of additional tours increased

only slightly and was not taken into account for calculation. Based on constant fuel and labour

prices, the additional costs depended on two main aspects:

1. The percentage of changed routes, governed by the relation of low- and high-

prevalence farms as well as by the threshold prevalence.

2. The additional distance per restructured tour

Due to the relation between the percentage of changed routes and the relation of low- and

high-prevalence farms a most cost-effective threshold prevalence cannot in general be

determined. If logistic slaughter procedures with a determined threshold prevalence become

obligatory, different producers’ association will have different additional costs per transported

slaughter pig. The present study enables producers’ associations to evaluate the additional

costs for their members.

77

References

Belœil, P.-A., Chauvin, C., Proux, K., Madec, F., Fravalo, P., Alioum, A., 2004. Impact of the

Salmonella status of market-age pigs and the pre-slaughter process on Salmonella

caecal contamination at slaughter. Veterinary Research 35, 513-530.

Berends, B.R., Urlings, H.A.P., Snijders, J.M.A., Van Knapen, F., 1996. Identification and

quantification of risk factors in animal management and transport regarding

Salmonella spp. in pigs. International Journal of Food Microbiology 30, 37-53.

Bundesverband Güterkraftverkehr Logistik und Entsorgung (BGL) e.V. , 2010. Dieselpreis-

Information (Großverbraucher).

http://www.bgl-ev.de/images/downloads/initiativen/dieselpreisinformation.pdf (last

access: 12.11.2010).

European Food Safety Authority, 2008. Report of the Task Force on Zoonoses Data

Collection on the analysis of the baseline survey on the prevalence of Salmonella in

slaughter pigs, Part A. The EFSA Journal 135, 1-111.

Hotes, S., Traulsen, I., Krieter, J., 2010. Salmonella control measures with special focus on

vaccination and logistic slaughter procedures. Submitted to Transboundary and

Emerging Diseases.

Hurd, H.S., McKean, J.D., Griffith, R.W., Wesley, I.V., Rostagno, M.H., 2002. Salmonella

enterica Infections in Market Swine with and without Transport and Holding. Applied

and Environmental Microbiology 68, 2376-2381.

Stärk, K.D.C., Wingstrand, A., Dahl, J., Møgelmose, V., Lo Fo Wong, D.M.A., 2002.

Differences and similarities among experts' opinions on Salmonella enterica dynamics

in swine pre-harvest. Preventive Veterinary Medicine 53, 7-20.

Statistisches Bundesamt Deutschland, 2010. EU-Vergleich der Arbeitskosten und

Lohnnebenkosten für das Jahr 2009. Press release, no. 122

http://www.destatis.de/jetspeed/portal/cms/Sites/destatis/Internet/DE/Presse/pm/2010/

03/PD10__122__624,templateId=renderPrint.psml (last access: 11.11.2010).

Swanenburg, M., Urlings, H.A.P., Keuzenkamp, D.A., Snijders, J.M.A., 2001a. Salmonella in

the lairage of pig slaughterhouses. Journal of Food Protection 64, 12-16.

Swanenburg, M., van der Wolf, P.J., Urlings, H.A.P., Snijders, J.M.A., van Knapen, F.,

2001b. Salmonella in slaughter pigs: the effect of logistic slaughter procedures of pigs

on the prevalence of Salmonella in pork. International Journal of Food Microbiology

70, 231-242.

78

van der Gaag, M.A., Saatkamp, H.W., Backus, G.B.C., van Beek, P., Huirne, R.B.M., 2004.

Cost-effectiveness of controlling Salmonella in the pork chain. Food Control 15, 173-

180.

Vereinigte Dienstleistungsgewerkschaft Nordrhein-Westfalen, 2009. Lohntarifvertrag für die

gewerblichen Arbeitnehmer in der Speditions-, Logistik- und Transportwirtschaft

Nordrhein-Westfalen vom 11. Mai 2009. http://www.verdi-dunie.de/downloads/fb-

10_downloads/2009_fb10/VSL%20Lohntarifvertrag%2011-05-2009.pdf (last access:

29.10.2010).

WHO, 2005. Drug-resistant Salmonella. Fact sheet N°139. World Health Organization,

http://www.who.int/mediacentre/factsheets/fs139/en/ (last access: 6.11.2010).

79

GENERAL DISCUSSION

The aim of the present study was to develop strategies to control non-clinical Salmonella in

pigs. Two methodological approaches were considered: (1) analyses of observational data to

detect risk factors associated with antibody detection in fattening pigs and (2) the

development of a simulation model to analyse the spread and control of Salmonella bacteria

from farrowing farm until slaughter. The former indicated critical control points which were

subsequently considered within the simulation model. Special emphasis was given to logistic

slaughter procedures characterised by herd separation according to prevalence for transport to

slaughter.

Risk factor analysis

The aim of the risk factor analysis presented in Chapter One was to detect the most important

risk factors related to the detection of Salmonella antibodies in fattening pigs. Analyses were

based on two independent datasets, consisting of blood and meat juice samples respectively.

Blood samples were taken within a previous project (Meyer, 2004) supported by the ZNVG

(Vermarktungsgemeinschaft für Zucht- und Nutzvieh). Samples from pigs and information

regarding husbandry, management and hygiene conditions were collected on site. Data

collection as well as serological tests were time-consuming and expensive. The obligatory

sampling of fattening pigs within the German Salmonella monitoring system offered the

chance to obtain information about herd prevalence of many fattening farms without

additional sampling. The ZNVG provided information on the results of the meat juice

sampling. Information regarding husbandry, management and hygiene conditions was also

collected by questionnaire. Response rates emphasised that a telephone survey was preferable

in contrast to sending the questionnaire by post. In general, a telephone survey requires

trained personal to make the questioning itself comparable and to detect contradictions within

answers during the survey. The weakness of a telephone survey might be the subjective

description of farm conditions. This is of minor significance if the questions relate to the

feeding system but might cause a problem if farm managers are to assess the presence of

rodents. A standardised evaluation could only be accomplished if all farms were evaluated by

the same person. But the needlessness of farm visits was one of the main advantages of the

meat juice dataset in the present study.

80

Results of logistic regression models revealed important risk factors for the detection of

Salmonella antibodies. In accordance with other studies (Davies et al., 1997; van der Wolf et

al., 2001; Nollet et al., 2004; Bahnson et al., 2006; Farzan et al., 2006; Benschop et al., 2008;

Vonnahme et al., 2008), the floor type as well as feed-related aspects were observed as

important factors associated with the detection of Salmonella antibodies in fattening pigs. The

proximity to other swine herds as well as the lack of protective clothing seemed to increase

the probability of Salmonella entry and therefore Salmonella prevalence. Summarising the

results, the risk factor analyses revealed both the importance of factors facilitating the entry of

Salmonella bacteria as well as the importance of factors supporting the spread within a farm.

Results were useful for the modelling of Salmonella transmission and especially for the

implementation of Salmonella control measures.

The simulation model

The simulation model described the spread of non-clinical Salmonella within the pork supply

chain from farrowing until slaughter. In contrast to previously published models (Ivanek et

al., 2004; van der Gaag et al., 2004b; Hill et al., 2008; Lurette et al., 2008), the model

considered the production and contact structures in detail and considered trade relationships

between farrowing and fattening stage as well as routes for transport to slaughter. Modelling

was individual-based, keeping track to every pig born during the simulation. The information

obtained was most detailed, but this method is also the most computer-intensive. Vynnycky

and White (2010) list the discrete-time compartmental method and the continuous-time

compartmental method as alternatives to individual-based models. In contrast to the

individual-based approach, compartmental models keep track of groups of individuals.

Whereas the individual-based approach and the discrete-time stochastic compartmental model

assume that all transitions occur after the same fixed time step, the continuous-time

compartmental model considers the time-to-next-event which can vary (Vynnycky and White,

2010). Within the present study, the individual-based approach was applied to enable detailed

analyses at pig level, e.g. keeping track of the transmission path based on a certain infectious

pig. Due to the fact that analyses of the present study concentrated on the overall prevalence

within a producers’ association, individual-based information were summarised. Thus, the

processing of output-files was improved for analyses. Hence, for the present study a discrete-

time compartmental approach would have been sufficient. But the individual-based modelling

offers opportunity to analyse special aspects of Salmonella transmission in detail.

81

Validation of the simulation model

Validation determines whether a simulation model is a reasonable approximation of a real

system (Fishman and Kiviat, 1968). An important technique to validate components of the

model and to determine the factors with significant impact is sensitivity analysis (Law, 2007).

Within the present study, a Plackett-Burman design was used for sensitivity analysis and to

determine which factors, out of twelve, had the greatest effect on response. Originally,

Plackett-Burman designs are two-level fractional factorial designs of resolution III

(Montgomery, 2005). They represent the smallest designs possible to estimate the main

effects but do not allow the estimation of interactions between factors (Vanaja and Shobha

Rani, 2007). Nevertheless, main effects are aliased with two-factor interactions (Box et al.,

1978). In the present study, a full fold-over was executed to break this alias structure. Thereby

a two-level design of resolution IV was obtained, where the main effects were clear of two-

factor interactions (Montgomery, 2005). This design was executed with maximum and default

values. To compare three levels, the design was also executed considering minimum and

default values (Vander Heyden et al., 1993; Vanaja and Shobha Rani, 2007). Hence, a three-

level design of resolution IV was performed within 64 runs. A regular fractional factorial

design considering 12 factors would have been of 81 scenarios at least and would have been

able to estimate only a few clear two-factor interactions (Xu, 2005).

One of several alternatives to fractional factorial designs are group screening techniques.

These techniques combine individual factors into groups and treat them as if they were

individual factors (Bettonvil and Kleijnen, 1996). A relatively new approach of group

screening is sequential bifurcation, which requires comparatively few simulation runs

(Bettonvil and Kleijnen, 1996). Initially, all factors are in a single group. If testing determines

that this group has a significant effect on response, it is broken into two subgroups. Subgroups

are tested for significance again and are broken into subgroups if there is a significant effect.

This procedure continues until each unimportant group has been eliminated and only

important factors remain (Jacoby and Harrison, 1962; Bettonvil and Kleijnen, 1996; Kleijnen,

2005; Law, 2007). Sequential bifurcation is particularly useful if only a few significant factors

have to be screened out of numerous input factors. All group screening techniques assume

that signs for the main effect are known. Otherwise, it may happen that effects of individual

factors compensate each other (Bettonvil and Kleijnen, 1996; Kleijnen, 2005; Law, 2007).

This assumption is justifiable for the present study but may be not realistic in other scenarios

(Shen and Wan, 2009). A design which is also able to handle many factors but requires

minimal assumptions is the Latin hypercube approach (Kleijnen et al., 2005). Compared to

82

factorial designs, the primary advantage of Latin hypercube sampling is that sample design

does not grow exponentially with the number of features (Seaholm et al., 1988). In Latin

hypercube sampling, the range of each model parameter is divided into the same number of

segments. One segment of each parameter is randomly chosen without replacement.

Subsequently, a parameter value from each selected segment is randomly chosen. The

resulting combination of parameter values represents the first sample of the Latin hypercube

design. Random sampling continues until all segments are chosen. Within the Latin

hypercube design every segment is used only once (Reed et al., 1984; Seaholm et al., 1988).

This kind of sampling has many advantages, such as the fact that the entire range of each

factor is always represented or that the mean squared errors of the estimated variances of the

model output are smaller compared to random or stratified sampling (Seaholm et al., 1988). A

weakness of Latin hypercube designs is the potential confounding of effect interactions. The

use of more than one sample improves this situation (Seaholm et al., 1988), but also increases

the design and therefore the running time.

There are many more approaches for the validation of simulation models and also the

techniques discussed underlie further differentiations. The Plackett-Burman design used was

appropriate to validate the model and to determine the most important factors. Out of twelve

input factors the probability of effective contact, the probability to restart shedding, the

shedding duration and the sow herd prevalence as well as the distribution of sow herd

prevalences across farrowing farms were obtained to determine slaughter pig prevalence. In

contrast to Hill et al. (2008), Ivanek et al. (2004) or van der Gaag et al. (2004b), who did not

simulate the sow herd at farrowing stage, the present study was able to demonstrate the

importance of sows’ prevalence. In accordance with this result, Lurette et al. (2008)

emphasised the maternal protective factor of the piglets as one of the most influential

parameters on Salmonella prevalence in slaughter pigs. The aspect of passive immunity was

not considered in the present study due to missing information. For example, Nollet et al.

(2005) could not prove the direct transmission of Salmonella from sow to piglet on the one

hand but demonstrated similarities between the isolates found in sows and grown pigs on the

other hand. Further research regarding the vertical transmission between sow and piglet is

desirable. The simulation model presented enables detailed analyses, especially with regard to

the consideration of sows and piglets as individuals.

83

Control measures

The effects of Salmonella control measures are described in Chapter Three. Risk factor

analyses emphasised the importance of factors facilitating the entry of Salmonella bacteria

(hygiene-related aspects) as well as the importance of factors supporting the spread within a

farm (husbandry-related aspects). Hence, attention was paid to both the control of hygiene

and husbandry management. Furthermore, the vaccination of sows and piglets as well as the

possibility of logistic slaughter procedures were considered. The latter could only be analysed

because several farms were taken into account. But with respect to the remaining three control

measures, it had to be decided which kind of farms had implemented the respective control

measure. For example, cost-effectiveness analyses by van der Gaag et al. (2004a) were mainly

based on the assumption that all farms or firms within a stage implement the suggested

interventions. This assumption is unrealistic and inappropriate for the present study. Due to

the fact that the simulation model had already been designed to consider different qualities of

hygiene and husbandry management, interventions were limited to farms with insufficient

hygiene or contact-facilitating conditions. This has to be considered in the interpretation of

the results. With an increasing number of farms implementing a control measure, the overall

prevalence will decrease. Hence, the analyses concentrated on the differences in effectiveness

between control measures and regarded the differences in effectiveness between production

stages. For example, the results emphasised the capabilities of the farrowing stage to

influence slaughter pig prevalence. Hygiene control measures were more effective if

implemented at farrowing stage compared to finishing stage. The possibility to vaccinate the

sows and their piglets was only admitted for high prevalence farms. Due to the assumption

that antibodies induced by vaccination cannot be distinguished from those induced after

infection a general usage seemed to be unlikely. But if only a few farms use the vaccine, the

overall prevalence will not decrease. Results regarding the logistic slaughter procedures

showed that the herd level separation decreased the infection at lairage at slaughterhouse but a

significant prevalence decrease was not achieved. Nevertheless, the results confirmed that

with decreasing prevalence after fattening the risk of infection during transport and lairage

become more and more important. Whether logistic slaughter procedures are implemented as

Salmonella control measures does not at least depend on the additional costs incurred due to

the separate transportation of pigs from low- and high-prevalence farms.

In Chapter Four, the additional costs to be expected were calculated. Price for slaughter pig

transport increased between 0.07€/pig and 0.58€/pig depending on the additional distance per

tour and the respective prevalence threshold. For the producers’ association considered in the

84

study the additional tours caused due to herd separation according to prevalence were

negligible. The number of tours increased by only about 28 tours per year at maximum;

representing barely more than half a tour per week. But this could be different within another

producers’ association. Analyses indicated that the number of additional tours is associated

with the size pattern of the fattening farms. Hence, the calculated cost increase is only

representative for a producers’ association comparable to the simulated association. An

extensive sensitivity analysis would supply insightful information on the input-output relation

regarding logistic slaughter procedures. Therefore, the simulation model presented offers the

possibility to consider a wide range of different producers’ associations. Furthermore,

individual-based modelling could be beneficial to reveal differences in transmission related to

the producers’ association and logistic slaughter procedures.

References

Bahnson, P.B., Fedorka-Cray, P.J., Ladely, S.R., Mateus-Pinilla, N.E., 2006. Herd-level risk

factors for Salmonella enterica subsp. enterica in U.S. market pigs. Preventive

Veterinary Medicine 76, 249-262.

Benschop, J., Stevenson, M.A., Dahl, J., French, N.P., 2008. Towards incorporating spatial

risk analysis for Salmonella sero-positivity into the Danish swine surveillance

programme. Preventive Veterinary Medicine 83, 347-359.

Bettonvil, B., Kleijnen, J.P.C., 1996. Searching for important factors in simulation models

with many factors: Sequential bifurcation. European Journal of Operational Research

96, 180-194.

Box, G.E.P., Hunter, W.G., Hunter, J.S., 1978. Statistics for Experimenters - An Introduction

to Design, Data Analysis, and Model Building. John Wiley & Sons New York.

Davies, P.R., Morrow, W.E.M., Jones, F.T., Deen, J., Fedorka-Cray, P.J., Harris, I.T., 1997.

Prevalence of salmonella in finishing swine raised in different production systems in

North Carolina, USA. Epidemiology and Infection 119, 237-244.

Farzan, A., Friendship, R.M., Dewey, C.E., Warriner, K., Poppe, C., Klotins, K., 2006.

Prevalence of Salmonella spp. on Canadian pig farms using liquid or dry-feeding.

Preventive Veterinary Medicine 73, 241-254.

Fishman, G.S., Kiviat, P.J., 1968. The statistics of discrete-event simulation. Simulation 10,

185-195.

85

Hill, A.A., Snary, E.L., Arnold, M.E., Alban, L., Cook, A.J.C., 2008. Dynamics of Salmonella

transmission on a British pig grower-finisher farm: a stochastic model. Epidemiology

and Infection 136, 320-333.

Ivanek, R., Snary, E.L., Cook, A.J.C., Grohn, Y.T., 2004. A mathematical model for the

transmission of Salmonella Typhimurium within a grower-finisher pig herd in Great

Britain. Journal of Food Protection 67, 2403-2409.

Jacoby, J.E., Harrison, S., 1962. Multi-variable experimentation and simulation models.

Naval Research Logistics Quarterly 9, 121-136.

Kleijnen, J.P.C., 2005. An overview of the design and analysis of simulation experiments for

sensitivity analysis. European Journal of Operational Research 164, 287-300.

Kleijnen, J.P.C., Sanchez, S.M., Lucas, T.W., Cioppa, T.M., 2005. A user's guide to the brave

new world of designing simulation experiments. Informs Journal on Computing 17,

263-289.

Law, A.M., 2007. Simulation Modeling and Analysis. McGraw Hill Boston.

Lurette, A., Belloc, C., Touzeau, S., Hoch, T., Ezanno, P., Seegers, H., Fourichon, C., 2008.

Modelling Salmonella spread within a farrow-to-finish pig herd. Veterinary Research

39, 49.

Meyer, C., 2004. Qualitative und quantitative Risikofaktoren für die Einschleppung und

Verbreitung von Salmonellen in unterschiedlichen Produktionsverfahren beim

Schwein. Institute of Animal Breeding and Husbandry, Christian-Albrechts-

University. Kiel.

Montgomery, D.C., 2005. Design and Analysis of Experiments. John Wiley & Sons, Inc.

Arizona State University.

Nollet, N., Houf, K., Dewulf, J., Duchateau, L., De Zutter, L., De Kruif, A., Maes, D., 2005.

Distribution of Salmonella strains in farrow-to-finish pig herds: A longitudinal study.

Journal of Food Protection 68, 2012-2021.

Nollet, N., Maes, D., De Zutter, L., Duchateau, L., Houf, K., Huysmans, K., Imberechts, H.,

Geers, R., de Kruif, A., Van Hoof, J., 2004. Risk factors for the herd-level

bacteriologic prevalence of Salmonella in Belgian slaughter pigs. Preventive

Veterinary Medicine 65, 63-75.

Reed, K.L., Rose, K.A., Whitmore, R.C., 1984. Latin hypercube analysis of parameter

sensitivity in a large model of outdoor recreation demand. Ecological Modelling 24,

159-169.

86

Seaholm, S.K., Ackerman, E., Wu, S.C., 1988. Latin hypercube sampling and the sensitivity

analysis of a Monte-Carlo epidemic model. International Journal of Bio-Medical

Computing 23, 97-112.

Shen, H., Wan, H., 2009. Controlled sequential factorial design for simulation factor

screening. European Journal of Operational Research 198, 511-519.

van der Gaag, M.A., Saatkamp, H.W., Backus, G.B.C., van Beek, P., Huirne, R.B.M., 2004a.

Cost-effectiveness of controlling Salmonella in the pork chain. Food Control 15, 173-

180.

van der Gaag, M.A., Vos, F., Saatkamp, H.W., van Boven, M., van Beek, P., Huirne, R.B.M.,

2004b. A state-transition simulation model for the spread of Salmonella in the pork

supply chain. European Journal of Operational Research 156, 782-798.

van der Wolf, P.J., Wolbers, W.B., Elbers, A.R.W., van der Heijden, H.M.J.F., Koppen,

J.M.C.C., Hunneman, W.A., van Schie, F.W., Tielen, M.J.M., 2001. Herd level

husbandry factors associated with the serological Salmonella prevalence in finishing

pig herds in The Netherlands. Veterinary Microbiology 78, 205-219.

Vanaja, K., Shobha Rani, R.H., 2007. Design of Experiments: Concept and Applications of

Plackett Burman design. Clinical Research and Regulatory Affairs 24, 1-23.

Vander Heyden, Y., Khots, M.S., Massart, D.L., 1993. Three-level screening designs for the

optimisation or the ruggedness testing of analytical procedures. Analytica Chimica

Acta 276, 189-195.

Vonnahme, J., Kreienbrock, L., Beilage, E.g., 2008. Untersuchungen zur Identifikation von

Risikofaktoren für die Ausbreitung von Salmonellen in Aufzuchtbeständen für

Jungsauen. Berliner und Münchener Tierärztliche Wochenschrift 121, 33-40.

Vynnycky, E., White, R.G., 2010. An Introduction to Infectious Disease Modelling. Oxford

University Press New York.

Xu, H., 2005. A catalogue of three-level regular fractional factorial designs. Metrika 62, 259-

281.

87

GENERAL SUMMARY

The aim of the present study was to develop strategies to decrease Salmonella prevalence in

fattening pigs. The thesis considered two methodological approaches; the analysis of

empirical data to detect the most important risk factors and the development of a simulation

model to analyse the spread of Salmonella within the pork supply chain.

Chapter One represented the most important risk factors associated with antibody detection in

fattening pigs. Blood and meat juice sample results were analysed with regard to husbandry,

management and hygiene conditions of the respective farms. The advantages and

disadvantages of data acquisition became apparent. Blood samples were taken on site which

was time-consuming but allowed the simultaneously collection of husbandry, management

and hygiene data of the respective barns. In contrast, the results of the meat juice sampling

were obtained from the obligatory Salmonella monitoring system which obviate the need for

sampling but sample results could only traced back to the farm instead of the barn.

The results of the blood sample analysis revealed that a fully slatted floor, the use of

protective clothing or the cleaning of the feed tube decrease Salmonella prevalence in

fattening pigs. Furthermore, it was shown that a distance of less than 2km to other swine

herds increase the chance of positive sampling. The results of the meat juice analysis

emphasised the effect of the feed structure and the feeding system for seropositivity in

fattening pigs. General speaking, the analyses revealed both the importance of risk factors

facilitating the entry of Salmonella bacteria and risk factors supporting the spread within a

farm.

The simulation model was described in Chapter Two. Modelling considered the primary

production from farrowing to fattening as well as slaughter procedures until chilling. The

model was characterised by the consideration of several farrowing and fattening farms. Trade

relations between production stages were simulated as well as routing for transport to

slaughter. The first analyses focussed on vertical transmission and the validation of the model.

Sensitivity analysis using a Plackett-Burman design was performed to analyse the input-

output relation of the model and to screen input factors with the greatest effect on response.

Basically, Plackett-Burman designs are two-level fractional factorial designs of resolution III.

A full fold-over technique was applied to the Plackett-Burman design to obtain a resolution

IV design. This allowed the estimation of the main effects clear of two-factor interactions.

88

Furthermore, the two-level Plackett-Burman design was reflected to enable the consideration

of three factor levels. Thus, information was obtained on whether the relation between input

and output was linear within the regarded interval or non-linear. All together 64 scenarios

were performed. Sensitivity analysis approved the model. Results met expectations and were

in accordance with other studies. The probability for an effective contact, the probability that

the carrier restarts shedding, the shedding duration and the sow herd prevalence as well as

their distribution across farrowing farms had a significant effect on Salmonella prevalence

after lairage. A non-linear relation between the input value and response was proven for the

probability of the effective contact, the sow herd prevalence and the distribution of sow herd

prevalence across farrowing farms.

In Chapter Three regard was given to the effectiveness of Salmonella control measures. Risk

factor analyses emphasised the importance of both factors facilitating the entry of Salmonella

bacteria (hygiene-related factors) as well as factors supporting the spread within farm

(husbandry-related factors). Hence, hygiene control measures as well as husbandry control

measures were considered. Furthermore, the vaccination of sows and their piglets were

considered for farrowing farms with high sow herd prevalence to interrupt the chain of

infection. Due to the fact that the simulation model was based on a trade network, it was

possible to analyse the effect of logistic slaughter procedures. Logistic slaughter procedures

were characterised by the herd separation according to prevalence for transport to slaughter

and slaughter process. The results emphasised the capabilities of the farrowing stage to

decrease prevalence in slaughter pigs. For example, hygiene control measures decreased

prevalence significantly if implemented at farrowing stage but not if the control measure was

implemented at fattening stage. In contrast, husbandry control measures obtained a significant

decrease whether they were implemented at farrowing or finishing stage. Furthermore, the

vaccination of sows and piglets was an appropriate control measure to decrease slaughter pig

prevalence. Simultaneous implementation of control measures showed that vaccination and

especially hygiene measures are mutually supportive. Concerning logistic slaughter

procedures it became obvious that the herd separation according to prevalence decreased the

percentage of seropositive pigs infected at lairage.

The additional costs of logistic slaughter procedures were calculated in Chapter Four.

Calculations considered prevalence thresholds of 40%, 20% and 10%. Additional distances

between 20km and 100km were assumed to account for changed routing. Furthermore, it was

89

presumed that the implementation of logistic slaughter procedures increases the total number

of tours to slaughterhouse. But for the simulated producers’ association the additional tours

were negligible. Depending on the prevalence threshold and the respective additional

distance, transportation costs increased by between 0.07€/pig and 0.58€/pig due to herd

separation according to prevalence. Analysis showed that the most cost-effective threshold

prevalence depends on the allocation of prevalence across farms. Hence, if logistic slaughter

procedures become obligatory, additional costs will vary between producers’ associations.

90

91

ZUSAMMENFASSUNG

Ziel der vorliegenden Arbeit war die Entwicklung von Strategien zur Reduzierung der

Salmonellenprävalenz beim Schwein. Dabei wurden zwei methodisch unterschiedliche

Ansätze verfolgt. Zum einen wurden empirische Daten zur Ermittlung der bedeutendsten

Risikofaktoren für die Salmonellen-Ausbreitung ausgewertet. Zum anderen wurde ein

Simulationsmodell entwickelt, welches die Salmonellen-Ausbreitung innerhalb der

Schweineproduktion abbildete und umfassende Analysen vom Ferkelerzeuger bis zum

Schlachthof zuließ.

Kapitel Eins beschreibt die Ermittlung der bedeutendsten Risikofaktoren für die Ausbreitung

von Salmonellen in Mastbeständen. Für die Auswertung wurden neben serologischen

Testergebnissen von Blut- und Fleischsaftproben die betrieblichen Haltungs- und

Hygienebedingungen sowie relevante Aspekte des Managementsystems erfasst. Die

Entnahme der Blutproben war im Rahmen eines vorangegangenen Projektes auf den

jeweiligen Mastbetrieben erfolgt. Diese Form der Datenbeschaffung ermöglichte die

Erfassung sämtlicher Informationen auf Abteilebene. Für den Datensatz der Fleischsaftproben

konnten die serologischen Untersuchungsergebnisse des obligatorischen Salmonellen-

Monitorings genutzt werden. Dies hatte den Vorteil, dass zeitaufwendige Betriebsbesuche

nicht nötig waren. Allerdings konnten die Fleischsaftproben nur auf Betriebsebene

ausgewertet werden. Eine Rückverfolgbarkeit auf Stall- oder Abteilebene war nicht möglich.

Bei der Auswertung der Blutprobenergebnisse zeigte sich, dass ein höherer Spaltenanteil im

Boden, Schutzkleidung für bestandsfremde Personen sowie die Reinigung der

Futtermittelleitungen zur Prävalenzreduktion beitragen. Im Gegensatz dazu erhöhte die Nähe

zu weiteren Schweinebeständen die Chance eines positiven Antikörperbefundes. Im Bezug

auf die Fleischsaftproben wurde der Einfluss des Fütterungssystem und der

Futtermittelstruktur deutlich. Damit erwiesen sich sowohl haltungsrelevante Faktoren als auch

Aspekte der Stallhygiene als bedeutend für das Auftreten seropositiver Mastschweine.

Das in Kapitel Zwei dargestellte Simulationsmodell bildet die Salmonellenausbreitung von

der Ferkelerzeugung bis zum Schlachthof ab. Das Modell stellt die Lieferbeziehungen

zwischen Ferkelerzeugern und Mästern dar und berücksichtigt, dass beim Transport zum

Schlachthof mehrere Betriebe von einem LKW angefahren werden, sofern noch ausreichend

Ladekapazität besteht. Die Validierung des Modells erfolgte über eine Sensitivitätsanalyse

92

unter Anwendung eines Plackett-Burman Versuchsdesigns. Plackett-Burman Pläne gehören

zu der Klasse fraktioniert faktorieller Designs und ermöglichen die Selektion wichtiger

Einflussfaktoren. Um zu gewährleisten, dass die geschätzten Haupteffekte unabhängig von

allen Zweierinteraktionen sind, wurde eine sogenannte Faltung (fold-over) vorgenommen.

Dabei wird der ursprüngliche Versuchsplan invertiert und zusätzlich zum Originaldesign

ausgeführt. Die Auflösungsstufe des Designs erhöht sich dadurch von III auf IV. Um

nichtlineare Input-Output-Relationen aufzudecken, wurden bei der Sensitivitätsanalyse drei

Faktorstufen pro Inputparameter berücksichtigt (reflected design). Für die Umsetzung im

Versuchsdesign bedeutete dies die Verdopplung der Anzahl zu schätzender Szenarien auf 64.

Die Ergebnisse der Sensitivitätsanalyse bestätigten die Validität des Modells und zeigten

relevante Inputfaktoren auf. Als signifikant für die Prävalenz am Schlachthof erwiesen sich

die Ausscheidungsdauer sowie die Wahrscheinlichkeit, dass Carrier-Tiere erneut Salmonellen

ausscheiden. Auch die Wahrscheinlichkeit eines effektiven Kontaktes, die Prävalenz bei den

Sauen sowie die Verteilung dieser Prävalenz unter den ferkelerzeugenden Betrieben zeigten

einen signifikanten Einfluss auf die Salmonellenprävalenz am Schlachthof. Für die drei letzt

genannten Einflussfaktoren konnte ein nicht-linearer Zusammenhang zwischen

Inputparameter und erzielter Prävalenz festgestellt werden.

Kapitel Drei befasst sich mit der Effektivität verschiedener Bekämpfungsmaßnahmen zur

Salmonellenreduktion. Da sich bei der Analyse potentieller Risikofaktoren sowohl

haltungsrelevante Faktoren als auch Aspekte der Stallhygiene als bedeutend herausgestellt

hatten, wurden beide Bereiche in der Simulation berücksichtigt. Des Weiteren wurden die

Effektivität einer Impfung und der Einfluss der logistischen Schlachtung auf die

Salmonellenprävalenz betrachtet. Die logistische Schlachtung und der damit einhergehende

logistische Transport sind dadurch gekennzeichnet, dass Mastbetriebe mit erhöhter

Salmonellenprävalenz als Risikobetrieb klassifiziert werden. Um eine entsprechende

Kontamination zu vermeiden, werden diese Betriebe erst am Ende des Schlachttages

transportiert und geschlachtet.

Die Ergebnisse verdeutlichten die Möglichkeiten der Ferkelerzeuger zur Prävalenzreduktion

am Schlachthof beizutragen. So zeigte sich, dass Hygienemaßnahmen möglichst früh

innerhalb der Primärproduktion implementiert werden müssen, um eine signifikante

Prävalenzreduktion zu erzielen. In der durchgeführten Studie konnte keine signifikante

Senkung der Prävalenz festgestellt werden, wenn die Hygienemaßnahmen erst in der Mast

implementiert wurden. Im Gegensatz dazu bewirkten Veränderungen in den Haltungs-

93

bedingungen, unabhängig von der Produktionsstufe, eine signifikante Senkung der Prävalenz

am Schlachthof. Des Weiteren zeigte sich, dass die Impfung von Sauen und Ferkeln den

Infektionsdruck verringerte und somit die Prävalenz auf den Betrieben senkte. Als besonders

vorteilhaft erwies sich die Kombination aus Impfung und verbesserter Stallhygiene. Durch

den logistischen Transport und die logistische Schlachtung konnte der Anteil seropositiver

Tiere, der sich am Schlachthof infizierte, signifikant gesenkt werden.

Die Mehrkosten, die aufgrund des logistischen Transportes zu erwarten sind, wurden in

Kapitel Vier kalkuliert. Die Berechnungen stützten sich auf die in Kapitel Drei beschriebenen

Szenarien zum logistischen Transport. Dabei wurden drei unterschiedliche Grenzwerte zur

Definition der Risikogruppe berücksichtigt: 40%, 20% und 10%. Des Weiteren wurde

angenommen, dass sich die vom logistischen Transport betroffenen Touren um 20km bis

100km verlängern. Da die Simulationsergebnisse zeigten, dass für die zugrunde gelegte

Erzeugergemeinschaft kaum zusätzliche Touren durch den logistischen Transport entstehen,

wurde diese Größe bei der Kostenkalkulation nicht berücksichtig. Je nach Grenzwert und

Tourenverlängerung stiegen die Kosten zwischen 0.07€ und 0.58€ je Schwein. Mit dem

Grenzwert ändert sich auch der Anteil Risikobetriebe, was sich wiederum auf die

Tourenplanung auswirkt. Für die Praxis bedeutet dies, dass Erzeugergemeinschaften mit

unterschiedlicher Prävalenzstruktur auch mit unterschiedlichen Kosten durch den logistischen

Transport zu rechnen haben.

DANKSAGUNG

An dieser Stelle möchte ich mich bei allen Menschen bedanken, die zur Umsetzung und zum

Gelingen dieser Arbeit beigetragen haben.

Herrn Prof. Dr. J. Krieter danke ich für die Überlassung des Themas, die wissenschaftliche

Betreuung, die mir gewährten Freiräume bei der Erstellung der Arbeit sowie für die

Möglichkeit meine Forschungsergebnisse auf Tagungen im In- und Ausland zu

präsentieren.

Herrn Prof. Dr. Dr. C. Henning danke ich für die Übernahme des Koreferates.

Frau Dr. I. Traulsen danke ich für ihre fortwährende Hilfsbereitschaft und Unterstützung bei

der Erstellung dieser Arbeit. Die fachlichen Diskussionen und das entgegengebrachte

Vertrauen waren für mich von ganz besonderer Bedeutung.

Herrn Dr. G. Rave danke ich für die Anregungen und Unterstützung bei der statistischen

Auswertung.

Für die finanzielle Unterstützung danke ich der Innovationsstiftung Schleswig-Holstein.

Der Vermarktungsgemeinschaft für Zucht- und Nutzvieh (ZNVG, Neumünster) danke ich für

die Unterstützung bei der Auswahl der Betriebe und der Befragung der Betriebsleiter.

Des Weiteren möchte ich mich bei Frau Prof. Dr. N. Kemper, Frau Dr. A. Menrath, Frau R.

Preißler sowie Frau A. Albrecht für die Hilfe und Unterstützung bei

veterinärmedizinischen Fragestellungen bedanken.

Für die kollegiale und freundschaftliche Atmosphäre am Institut möchte ich mich bei allen

Kollegen bedanken. Ein besonderer Dank gilt Julia und Anna sowie Imke, Verena,

Bettina, Marrin und Lisa. – Ohne Euch wäre die Zeit am Institut nur halb so schön

gewesen.

Meiner Bürokollegin Regine, möchte ich für ihre Hilfsbereitschaft, die fachlichen und nicht

fachlichen Gespräche sowie die angenehme Atmosphäre im Büro danken.

Mein größter Dank gilt meiner Familie und Andreas, die mich stets unterstützt haben und nie

den Blick für das Wesentliche verlieren ließen.

97

LEBENSLAUF

Name Stefanie Hotes

Geburtsdatum 27.03.1983

Geburtsort Bremen

Staatsangehörigkeit deutsch

Familienstand ledig

Schulausbildung

1989 – 1993 Grundschule Hagen

1993 – 1995 Hermann-Allmers-Schule Hagen

1995 – 1999 Waldschule Hagen

1999 – 2002 Waldschule Schwanewede Abschluss: Allgemeine Hochschulreife

Studium

2002 – 2005 Studium der Agrarwissenschaften mit der Fachrichtung Agrarökonomie und Agribusiness an der Christian-Albrechts-Universität zu Kiel

Abschluss: Bachelor of Science

2005 – 2007 Studium der Agrarwissenschaften mit der Fachrichtung Agrarökonomie und Agribusiness an der Christian-Albrechts-Universität zu Kiel

Abschluss: Master of Science

Berufliche Tätigkeit

Juni 2007 – Mai 2008 Wissenschaftliche Mitarbeiterin am Institut für Ernährungswirtschaft und Verbrauchslehre der Christian-Albrechts-Universität zu Kiel bei Frau Prof. Dr. J. Roosen

seit Juni 2008 Wissenschaftliche Mitarbeiterin am Institut für Tierzucht und Tierhaltung der Christian-Albrechts-Universität zu Kiel bei Herrn Prof. Dr. J. Krieter