Skip to main content
European Commission logo print header

Sustainable intensive pig and poultry production

Final Report Summary - PROHEALTH (Sustainable intensive pig and poultry production)

Executive Summary:
Overall the PROHEALTH project was a huge success and achieved all its goals and objectives. Over the duration of 5 years a significant amount of outcomes has been achieved by the 22 project partners. More specifically, the following work was performed in the different work packages (WP):

WP1: Risk factors for production related diseases have been identified. The time-related effects and associations between different health, welfare and performance parameters has been quantified.
Questionnaires for pigs, broiler and layer chickens were used to collect data on biosecurity, management and farm and batch/flock characteristics. Their analysis provided insights on the risk factors for a number of production diseases.
WP2: The role of sow gestation housing system on piglet viability and immune status has been addressed. In addition, the role of genetics and farrowing management on maternal competence and neonatal vitality has been studied. Work on the potential for adaptation of milk composition to optimise nurse sow fostering protocols during late lactation has been reported. Finally, an investigation on disinfection protocols on bacterial diversity of shells of hatching eggs was carried out and reported on, including recommendations.
WP3: A meta-analysis to identify the most relevant traits to describe animal responses during production diseases was reported. The large experimental programme under this WP included the investigations into the relationship between genotype and production diseases, and the development of strategies on how to reduce the susceptibility to leg disorders of pigs and poultry selected for high productive outputs. Such strategies included nutrition and management ones. Tissue, blood and faecal samples from the experimental programme were provided to WP5.
WP4: Daily reception and validation of data arising from a variety of environmental sensors received from a variety of farms has been accomplished. Associations between environmental components and production diseases have been reported. In an experimental programme it was possible to determine the role of the microbial environment on the temporal expression of production diseases.
WP5: Specific microbiota in pigs and chickens under different housing conditions and disease status have been identified. Micro-array analyses of samples from tissues of pigs and chickens under different housing conditions and disease status have been carried out and revealed. Biomarkers for detection of digestive disorders in broilers and respiratory diseases in pigs have been identified.
WP6: A set of large scale interventions on farms for reduction of production diseases of pigs and poultry were carried out. The interventions arose from the outcome of the previous WPs. Testing these interventions on far allowed for holistic consideration, including their economic and social acceptability consequences to be considered.
WP7: The characterisation of the socio-economic impacts of production diseases, including impacts related to animal welfare, ethical considerations, and impacts on costs and efficiency of production on the farm was addressed. Stakeholder surveys were carried out and helped to identify key health issues and future possibilities in animal health and the control of production diseases. The economic and whole chain consequences of the preferred interventions from WP6 were quantified.

Great effort has been made to disseminate the results from PROHEALTH as widely as possible: Eight editions of the electronic project newsletters were produced and distributed (most of them in multiple EU languages). A series of national stakeholder events has been organised in several European countries. A large number of presentations at national and international conferences have been made, including several special PROHEALTH sessions. Many peer-reviewed publications arising from the project have already appeared in the international literature and several more are in preparation or are currently under peer review.

Project Context and Objectives:
Rising demand for animal products, an increasing global population as well as global competition have increased level and intensification of animal production. This trend comes at the price of increasing the incidence of production diseases. Production diseases can be defined as 'Diseases which tend to persist in animal production systems and, typically, become more prevalent or severe, in proportion to the potential productivity of the system'. They compromise health and welfare, generating inefficiencies which negatively impact on profitability, environmental footprint and product quality. They may also increase the need to treat the affected animals with antibiotics.
Although not explicitly stated by consumers, production diseases are associated with most of the consumer concerns about the health and welfare of livestock systems. This is because production diseases are endemic; they are associated with the decreases in animal welfare and the continuous input of antibiotics in livestock systems. For example, neonatal mortality in pigs and lameness in both pigs and poultry are considered production diseases, because their prevalence increases in more intensive systems of production. During the course of the project the antibiotic use in both pig and poultry production systems, became increasingly the focus of consumer and policy maker concerns, and made the project significantly more topical.
During the last year there have been increased concerns about the environmental impacts of livestock systems, especially after the deliberations of the 2018 UN Conference held in Poland. It was not part of the PROHEALTH remit to address this issue. However, the outcomes of the project could be used as background data in exercises that aim to quantify the environmental impact of production diseases and the mitigation that can arise from the application of strategies to minimise the impact of production diseases in pigs and poultry.
The PROHEALTH project aimed to develop an understanding of the multi-factorial dimension of animal pathologies linked to the intensification of production and to use this new knowledge to develop, evaluate and disseminate effective management and control strategies.
The PROHEALTH project addressed each of the elements in the network of production disease causation. In particular, the specific objectives of the project have been to:
• identify the risk factors for production diseases and establish associations between diseases;
• explore the role of genetic and environmental factors on neonatal survival and in exerting longer-term developmental influences on health;
• evaluate the effects of genetic selection for productive traits on susceptibility and identify strategies to mitigate these;
• determine the role of variation in farm environment on the temporal expression of production diseases;
• characterise the microbio-immunological changes and identify pathological changes at the molecular level which take place during production diseases in order to develop diagnostic tools;
• synthesise strategies to reduce the impact of production diseases on a farm and assess the efficacy of improvement strategies in reducing disease prevalence or severity; and
• identify economically viable and socially acceptable ways to control pathologies, with emphasis on animal welfare implications.

Project Results:
The presentation of the main S&T results is organised along the seven S&T work packages (WP) of the project. Any interactions between WP are also detailed; WP6 and WP7 were the WPs that synthesised the work of the other WPs.


The objectives of WP1 were:
• To assess the current situation on health, welfare and performance in pigs and poultry across Europe
• To score biosecurity and management practices potentially related to poor health, welfare and performance in pig and poultry farms
• To quantify risk and protective factors in pigs regarding poor health, sow welfare and performance, neonatal and piglet mortality, lesions in slaughter pigs in a standardised way in diverse EU systems in 9 EU countries
• To quantify risk and protective factors in poultry in a standardised way in diverse EU systems in 7 EU countries regarding poor health, welfare and performance in broilers and laying hens

TASK 1.1 - Documentation of the current situation regarding health, welfare and performance in pig and poultry enterprises
Existing datasets on performance, health and welfare of poultry and pig farms from different EU countries with intensive production systems were used to investigate the occurrence of production diseases. The prevalence of performance, health and welfare parameters and the evolution of these parameters over time was investigated. The first part provided data on poultry: broilers (n=8 countries), broiler-breeders (n=7 countries) and laying hens (n=3 countries). The second part provided data on pigs: sows and piglets (n=7 countries) and fattening pigs (n=4 countries). The results of these data were presented in D1.1 (2nd periodic report)
TASK 1.2 - Assessment of biosecurity, management and farm and batch/flock characteristics in pig and poultry farms
Biosecurity and management practices of pig and poultry farms per country were collected. To obtain this information, web-based questionnaires were developed and then used to obtain data from different countries in a standardised way. Three different web-based questionnaires were developed: one each for pigs, broilers and laying hens. The results of these data were presented in D1.2 (2nd periodic report). The biosecurity was scored in sow herds from 6 different European countries: Belgium, Denmark, Finland, Germany, The Netherlands and Spain (coded, but not in this order, as A, B, C, D, E and F). A total of 236 sow farms completed the questionnaire. The overall external biosecurity score was highest in country B (87.5) and lowest in country A (67.3). Country C and D did not significantly differ from each other regarding overall external biosecurity score . With respect to internal biosecurity, country B had a significantly higher internal biosecurity (64.6) compared to countries A and C (55.0 and 51.2 respectively) . Country F had the lowest overall internal biosecurity score (46.4) compared to countries B, E and D (64.6 60.1 and 57.6 respectively) . The overall scores for external biosecurity were higher than the scores for internal biosecurity meaning it is easier for farmers to implement management measures for the prevention of pathogens entering a farm rather than to implement measures to prevent the spreading of pathogens within the farm. Apart from that, there was a large variation between countries, likely due to differences in farm and country characteristics and other factors. Since there was a large variation in scores between farms, there is still potential for improvement in many pig farms.
The biosecurity was also scored in 108 fattening pig farms in four European countries: Belgium, Poland, UK and Finland. External biosecurity was 67.1 (+/- 8.2) internal biosecurity was 64.8 (+/- 19.0) whilst there were notable differences among countries for both. A low positive significant correlation was seen between external biosecurity and the size of the farm. Low to moderate negative correlations were seen between the internal biosecurity scores and either the farmer experience in keeping pigs or the farm size.
The biosecurity was scored in 399 conventional broiler farms in 5 EU member states. The scoring system consisted of two main categories, i.e. external and internal biosecurity, which had eight and three subcategories, respectively. The participating broiler farms scored better for internal biosecurity (76.6) than for external biosecurity (68.4). The scores between countries ranged from 59.8 to 78.0 for external biosecurity and from 63.0 to 85.6 for internal biosecurity. Within the category of external biosecurity, the subcategory related to “infrastructure and vectors” had the highest mean score (82.4) while the subcategory with the lowest score related to biosecurity procedures for “visitors and staff” (51.5). Within the category of internal biosecurity, the subcategory “disease management” had the highest mean score (65.8). A significantly negative correlation was found between internal biosecurity and the number of employees and farm-size.
The results indicated a lot of variation for both external and internal biosecurity between broiler farms, suggesting that improvements are possible. Since the subcategory “visitors and staff” scored the lowest, better education of broiler-farmers and their staff may help to improve biosecurity on broiler farms in Europe.
Biosecurity was scored in 124 layer farms which included three housing types (enriched cages (n=67); deep litter system (n=32); perch system (n=25)). The internal biosecurity score (75.9%) was higher than the external biosecurity score (69.9%). There were no significant differences for biosecurity scores between the different countries. Clear correlations were found between overall biosecurity with both external and internal biosecurity. There was a high variation within countries for external, internal and overall biosecurity, suggesting that there is room for improvement. Biosecurity in laying hen farms with different production systems is known to be important in the epidemiology of several diseases. Nevertheless, the study did not demonstrate a significant difference for overall, external an internal biosecurity between different housing systems. One possible explanation for this could be that the variation in biosecurity scores between and within countries was larger than the variation between the three housing systems.

TASK 1.3 - Quantification of risk and protective factors for poor health, welfare and performance in pigs and poultry

In addition to the web-based, general questionnaires (see Task 1.2) more detailed questionnaires were developed. The questionnaires were filled in during a visit of the herds/flocks from which we collected data on health, welfare and production. The aim was to find possible associations between the results of the questionnaire-based studies (information from the general and the detailed questionnaires) and the health, welfare and production parameters of the herds was met. For pigs, univariable and multivariable linear models were employed to identify risk factors in pig herds. Litter index (litters/sow/year) was negatively associated with a higher weaning age of the piglets. Housing recently weaned sows to be inseminated in a separate unit from the gestation unit had a positive association with litter index. Repeat breeding was negatively associated with PRRS-free farms, farms that bred (raised) all gilts on the farm and farms that perform farrowing induction of sows. PRRS-free farms were also associated with a higher replacement rate. Farms that bred gilts on the farm and PRRS-free farms were negatively associated with pre-weaning mortality. Natural ventilation in the gestation unit was associated with fewer piglets born alive and with fewer weaned piglets. Closed type of farms was associated with less piglets born dead. The use of open box housing system for pregnant sows (provision of individual resting areas) was associated with more weaned piglets. In conclusion, several factors related to applying good farm and health management, and optimal housing conditions showed positive association with various sow and piglet performance parameters. Further studies will help to assess causal links for these factors.
For poultry, potential risk or protective factors were investigated in 358 conventional broiler farms in 7 EU member states. Different performance and health data was collected from at least six successive flocks. Eight multivariable linear mixed models were developed. Three models were used to investigate mortality (overall, first week, after first week), three for performance variables (growth, feed conversion, European production index) and two were related to slaughterhouse data (i.e. dead on arrival and condemnation rate). The following factors were associated with increased mortality: floor quality, neonatal E.coli infections, ventilation type and number of professional activities of the farmer. The factors associated with performance were chick sex, coccidiosis infections, necrotic enteritis, dysbacteriosis, light intensity adaptations, ventilation type, comparison of daily results by farmer, daily check of feed and water system and type of feed. For dead on arrival three risk factors were identified i.e. daily growth, type of light adaptation and type of drinking system. For condemnation rate seven risk factors were found, i.e. type of drinking system, daily growth, feed withdrawal time, type of ventilation, house size, E.coli after seven days and type of feed. The findings indicate that many different factors related to management, housing and health programs may improve broiler performance.
For laying hens housing and production management had clear effects on production parameters. The main risk factors for hens housed in enriched cage systems were: ventilation, feed and light management, type of hens and the level of professionalism of the farmer. Better production results and longevity occurred when cross x tunnel ventilation was used; when recirculation vents were present; when feed was provided before the lights went on; when white hens are kept and when the farmer has no other professional occupations. For deep litter systems the main risk factors for production and performance of hens were: structure of the floor, part of an integration, floor quality, the type of hens and the level of professionalism of the farmer. Better production results and longevity occurred when hens were housed on a half-slatted floor, on a smooth impervious floor, white hens were kept and when the farmer had no other professional occupations.

The objectives of WP2 were:
• To determine the role of sow gestation housing and diet on piglet viability and immune status (TASK 2.1)
• To determine the influence of genetics and farrowing management on maternal competence and neonatal vitality, and the long term consequences for pig health (TASK 2.2)
• To investigate the potential for adaptation of milk composition to optimise nurse sow fostering protocols (TASK 2.3)
• To investigate the vertical transmission pattern of E. coli and Enteroccocus faecalis, two major causes of first week mortality within the poultry industry, and to determine the optimal egg disinfection strategy to minimise this post hatch mortality an ability to cope with environmental challenges in chickens (TASK 2.4).

TASK 2.1 – Nutritional and environmental influences on pig fetal development, neonatal survival and immune status

This task was addressed in two experiments. In the first experiment (Task 2.1.1) two housing systems for gestating sows were compared: a conventional system on slatted flooring and an enriched system with deep straw litter and additional space. Three batches of 19 sows per housing system were used and traits of sow and piglet behaviour, physiology and immunity were measured. Compared with the enriched system, the conventional system increased indicators of stress in sows during gestation and was associated with increased neonatal mortality. At birth, mean piglet BW did not differ between groups; however, the percentage of light (< 1.2 kg) and heavy (≥ 2 kg) piglets was greater and lower, respectively, in C than in E litters. Plasma concentrations of cortisol, IGF-I, T4, T3, lactate, NEFA, fructose and albumin did not differ between C and E piglets, but the insulin to glucose ratio was greater in C than in E piglets. Compared with E piglets, C piglets had a lighter gut at birth and their glycogen content in longissimus muscle was lower. In this muscle, mRNA levels of PAX7, a marker of satellite cells, and of PPARGC1A, a transcriptional coactivator involved in mitochondriogenesis and mitochondrial energy metabolism, were greater, whereas the expression level of PRDX6, a gene playing role in antioxidant pathway, was lower in C than in E piglets. In conclusion, a stressful environment of sows during gestation had negative effects on piglet traits related to physiological maturity and energy homeostasis at birth. The experiment was reported in D2.1 (November 2015) . Faecal samples from sows and piglets in each system were supplied to WP5 for investigation of the effects on gut microbiota (P3).
Task 2.1.2 was set up on the basis of results from Task 2.1.1 (MS4, December 2015) to evaluate a strategy to reduce the negative impact of the conventional gestation housing system on neonatal survival by dietary modification and system enrichment. Three experimental treatments were compared: the conventional gestation system (C, on a slatted floor); the “enriched conventional” system (CE, on a slatted floor enriched with manipulable material and straw pellets), and the enriched system (E, on deep straw litter and with additional space per sow). In total, 83 crossbred Large White x Landrace sows were housed in groups until 105 days of gestation, and then all sows were transferred into farrowing pens and housed in identical individual crates on a slatted floor. At 101 days of gestation, the enrichment reduced sow stereotypies (ECE>C. Cortisol concentration in sow saliva, a marker of stress, differed among groups but differently according to days of gestation (C=CE>E, C>CE>E and C=CE=E at 14, 105 and 107 days, respectively). Granulocytes number in blood was greater in C than in CE and E sows , but CE sows had a greater oxidative stress compared to C and E sows. Piglet mortality during and within 12 h of birth was lower in the CE and E groups (6.6 and 6.3%) than in the C group (11.1%). In terms of body weight, CE piglets were close to C piglets, whereas in terms of metabolic status they tended to be intermediate between C and E piglets. No treatment effect was observed on piglet oxidative status. The results therefore indicate that enrichment strategies can partially ameliorate the negative effects of the conventional gestation housing system. The experiment was reported in D2.2 (November 2017).

TASK 2.2 – Genetic and management influences on maternal competence and neonatal vitality

This task was addressed in a genetic study and a separate management study. In the genetic study (Task 2.2.1) data on traits with potential relevance for piglet survival were collected on sows and piglets of known pedigree in a 750-sow multiplication herd over a 12-month period. A methodology was developed for automated characterisation of the posture changing movements of farrowing sows and a subset of sows also had movement data collected around farrowing via rump-mounted tri-axial accelerometers. Over 21,000 piglets had individual data collected on birth weight, sex, intrauterine growth retardation status (as defined by head shape) and reason, and date, of death. Analysis of data from sows fitted with accelerometers showed that sow conformation had an effect on the accelerometer-derived measures, e.g. the straightness of the hind legs when viewed from the back had an effect on the mean maximum acceleration during a posture changing transition. Heritabilities for accelerometer-derived measures ranged from low to high (0.00 to 0.6) depending on whether the analysis was conducted on the raw accelerometer measures (separate pre-and post- farrowing datasets) or on the mean accelerometer measures. Significant effects on number of piglets crushed by processing included an interaction between piglet weight and IUGR-status, and an interaction between farrowing floor type and IUGR-status. Additionally, sow conformation interacted with the accelerometer-derived measures to affect the probability of them being crushed, e.g. hind leg placement interacted with the mean rate of acceleration change, with normal leg placement under the body giving a lower probability of crushing piglets; hind leg shape interacted with the mean rate of side to side movement; and the shape of the hind pastern interacted with the mean rate of pitch change. The trait ‘quality of lying’ is suggested for consideration as a trait affecting piglet survival and which would be relatively easily collected on farm. The intrauterine growth retardation status of the piglet (as defined by head shape) interacted with birth weight to affect piglet survival and, specifically, the probability of being crushed. IUGR-status is a trait relatively easily collected on farm during the processing of piglets. Heritability estimates suggest that this information could also be used to select sows producing litters with low proportions of IUGR-piglets (i.e. more normal head shape piglets), allowing an easy, on-farm method of assessing uterine capacity. The experiment was reported in D2.3 (November 2016)

The management intervention (Task 2.2.2) investigated, in a factorial design, positive handling and classical music treatments in the farrowing house. The positive handling treatment used daily gentle scratching of the sows from placement in the farrowing unit until farrowing. However, since scratching takes time, an alternative method was also investigated - enrichment through sound, using classical music, from placement in the farrowing unit until five days post farrowing. The study was conducted in two herds (a 1,200 sow and a 700 sow herd) with loose housing throughout the mating, gestation and farrowing stages. Data on productivity traits and human avoidance score were recorded for 219 sows in herd 1 and 214 sows in herd 2. In addition, videos of the behaviour of 46 sows in herd 2 were recorded. Results demonstrated that positive handling (scratching) of sows resulted in a highly significant decrease in avoidance behaviour, whereas music had no significant effect. Personnel on farm, when asked about the effect of the two treatments, stated that they found sows in all treatment groups less reactive and easier to handle than sows in the non-treatment group. Also they did not consider the treatments as time-consuming or annoying. Although, it was not possible to document a reduction in neonatal piglet mortality when exposing the sows to positive handling, results demonstrated that positive handling has an impact on the human-animal relationship and the confidence of the farm personnel in working with loose housed sows. Therefore, it can be beneficial to include positive handling in daily management of loose housed sows. The experiment was reported in D2.4 (November 2016).
TASK 2.3 – Manipulation of milk composition to optimise nurse sow fostering protocols

Task 2.3 was addressed in a pilot study in conjunction with Agriculture and Agri-Food Canada (MS5), followed by the main experiment. This was carried out to test the hypothesis that temporary opening of the mammary tight junctions in late lactation sows, at a time when IgG levels in maternal plasma are high as a consequence of vaccination, would result in a significant increase in immunoglobulin concentration (including IgG) in the milk. Four lactating sows were immunised against ovalbumin and serum titres were achieved in all four at day 28 of lactation. At this time, the sows received a foster litter of 24h old piglets and were treated with a single supraphysiological dose of 100IU of oxytocin. Six hours later elevated levels of anti-ovalbumin IgG were detected in the milk of all four sows, as well as in the serum of the fostered piglets. The hypothesis was therefore supported and the data provide proof of principle for a method of more effectively using late lactation sows, which have reared their own litter, as foster mothers for young piglets. The experiment was reported in D2.5 (May 2016).

TASK 2.4 – Egg disinfection strategies to reduce post hatch mortality and ability to cope with environmental challenges

Task 2.4 was addressed by a microbial transmission field study and an egg disinfection intervention study. In the transmission study, four parent flocks were followed during the whole production period (20-60 weeks) by post mortem and bacteriological examination of randomly selected dead birds. Newly hatched chickens from each flock were swabbed in the cloaca (parent age 30, 40, 50, 60 weeks) and the bacterial flora analysed. Causes of first week mortality were determined pathologically and bacteriologically. E. coli isolates were selected for Pulsed-Field-Gel-Electrophoresis (PFGE) and Multi-Locus-Sequence-Typing (MLST). E. coli was the main cause of both salpingitis in parents and first week mortality in broilers, and E. coli dominated the bacterial flora of the cloaca of newly hatched chickens. PFGE of E. coli showed identical band patterns in isolates from the three different sources, indicating a vertical transmission of E. coli from salpingitis in parent birds to chicks and subsequent spread in the hatchery. A single PFGE type dominated in one parent farm and this type was also found in the newly hatched chicks and as the aetiology of first week mortality. Vertical transmission of E. faecalis from parent birds to chickens in three of the four farms was also indicated. The intervention study was performed to characterise the microbiome and aerobic bacterial load of hatching eggs before disinfection and during the subsequent disinfection steps. The study included both a group of clean and dirty eggs. For dirty eggs, an initial wash in chlorine was performed, and thereafter all eggs were submitted to two times fumigation and finally spray disinfection. The microbiome was characterised by sequencing of the total amount of 16S rRNA extracted from each sample, consisting of surface swabs of 5 eggs from the same group. In addition, the number of colony forming units (CFU) under aerobic conditions was established for each disinfection step. The disinfection procedure reduced bacterial load from more than 10^4 CFU (clean eggs) and 10^5 CFU (dirty eggs) to less than 10 CFU per sample for both groups of eggs after disinfection. The microbiome of both clean and dirty eggs had highest abundances of the phyla Firmicutes, Proteobacteria and Bacteroidetes. Within the phyla Firmicutes, the relative abundances of Clostridiales decreased while Lactobacillus increased from before to after final disinfection. The study introduced a new method for evaluation of the bacteria contaminating the surface of hatching eggs (the microbiome was characterised by sequencing of the total amount of 16S rRNA). The investigated disinfection procedure was effective in reducing CFU and, by adding a chlorine wash for dirty eggs, the microbiome of clean and dirty eggs had a highly similar microflora after the final disinfection step. The work and results from Task 2.4 were reported in D2.6 (November 2016).


For both pigs and chickens, the objectives of WP3 were:
• To perform a meta-analysis to identify the most relevant traits to describe animal responses, including animal-based welfare indicators, during production diseases (TASK 3.1)
• To define the relationship between genotype and production diseases, resulting from well-defined controlled challenges (TASK 3.2)
• To develop strategies on how to reduce the susceptibility to leg disorders of pigs and poultry selected for high production outputs through management, physical activity and nutrition (TASK 3.3)
• To describe and quantify the animal production response to defined challenges, to inform economic modelling (TASK 3.4)
• To provide blood, tissue and faeces samples from experiments with defined challenges to WP5 (TASK 3.5).

WP3 focused on major pig and poultry pathologies such as digestive, respiratory, leg and metabolic disorders and conducted experimental studies and systemic reviews of literature on the major categories of production diseases in both pigs and poultry. The experimental and literature approaches took into account the various aspects of the production system including genetic background, feeding and nutrition, health, animal based-welfare indicators and husbandry practices. WP3 was in close connection with other WP to supply data and data (WP5, WP7), to contribute to the selection of the improvement strategies tested in WP6 and to the production of scientific knowledge, including animal based indicators, on animal biology with WP2 and WP5. It benefited from the expertise of WP1 and WP4 for the characterisation of environment on animal responses.

TASK 3.1 – Meta-analysis to identify relevant animal health traits and animal-based welfare indicators relevant for production diseases

The objective was to identify the most relevant traits that can be used to characterise the physiological and health status of animals potentially affected by production diseases. The methodology was based on a systematic review of the scientific literature following a well-designed protocol in both species including a definition and a list of production diseases, the selection of inclusion and exclusion criteria, the construction of a database providing information on factors affecting the traits and a semi quantitative trait analysis. The methodology and main outputs are described in details in Deliverable D3.1 and a published scientific paper (Stravakakis et al. 2018). Briefly, this review evidenced a large variety of traits and highlighted the absence of a common core of traits across diseases and studies. Traits related to performance characteristics or immunological response are proposed as potential indicators for the prognosis and study of production diseases and for assessment of the efficacy of preventive or corrective practices. The review also revealed the gaps in the published research of production diseases, like for instance the lack of traits characterising animal welfare.

TASK 3.2 – Quantification of the relationship between genotype and production diseases (digestive, respiratory and leg disorders), resulting from well-defined controlled challenges

A panel of genetically different chicken and pig lines selected for productive traits, such as growth rate and feed efficiency were challenged experimentally in controlled environments.
The susceptibility to coccidiosis (D3.2) and to leg disorders (D3.3) was assessed in high performing broilers and broiler breeders. In pigs, the effect of genetic selection for productive traits (namely feed efficiency and muscle growth) on the susceptibility to post weaning diarrhoea (D3.6) respiratory and inflammatory diseases (D3.4) and locomotory (D3.3) disorders was evaluated.
Our work did not support the hypothesis that pigs and broilers selected for high productive traits were more affected by production diseases. Nevertheless, in controlled conditions, we showed that genetic lines can differ in their susceptibility to production diseases as what was observed in pigs selected for feed efficiency and housed in poor hygiene conditions (D3.4). However, our results revealed that management and housing conditions, nutrition and biosecurity are undoubtedly the major factors influencing the expression and severity of production diseases. In growing pigs, a poor hygiene of housing decreased growth performance by 20% and induced a systemic inflammatory response in pigs and respiratory diseases, such as pneumonia. These negative effects on health and performance persisted even after corrective solutions had been implemented later. Broilers affected by a production disease become more susceptible to other production diseases. For instance foot pad lesions may serve as port of entry for systemic or localised bacterial infections with increased mortality due to septicemic infections, with sepsis, endocarditis and arthritis as the major manifestations, over time. Coccidiosis lesions affect gut absorptive capacity as indicated by carotenoid concentrations and fat soluble vitamin status (Vit A and E), leading to reduced leg mineralisation and an increased risk of leg disorders with effects persisting after the birds had recovered from the pathogen whilst there was no interaction between infection and genotype used.

TASK 3.3 – Management and nutritional strategies to ameliorate bone quality in chickens and pigs of different production output potential

The objective was to conduct small scale experiments and review of the literature to develop strategies on how to reduce the susceptibility to leg and digestive disorders of pigs and poultry selected for high production outputs through management, physical activity and nutrition. Some of these strategies have been tested at larger scale in WP6.

Impact of management and physical activity on leg disorders

A systematic review of the literature was conducted to determine if there is evidence for a positive effect of some types of enrichment on leg health in broilers to be tested in WP6. The suggested mechanism behind the effect of environmental enrichment on leg health is that enrichment can increase activity levels, which has been shown to reduce leg problems. Stocking density, light program, intensity of light, introduction of perches or straw bales and increased distance between feed and water (separation of resources) were considered in the review. The review took into account different types of measures to assess leg disorders such as gait score, foot pad and hock dermatitis, tibial dyschondroplasia, leg abnormalities and deformation, bone density etc. The methodology and results are described in details in D3.5. Provision of some types of environmental enrichment can improve leg health. A lower stocking density and an intermittent lighting schedule are well-studied strategies that have proven to be effective. Provision of perches and a high intensity of light do not seem to be effective in improving leg health, except for possibly having an effect on the development of contact dermatitis. Provision of straw bales and separation of resources have only been studied to a limited extent, but both appear effective in improving leg health.
In growing pigs, increased physical activity was experimentally tested in group housed growing pigs on plain concrete floor. Physical activity was increased by forcing growing pigs to cross two times a sorter area before being allowed to enter the feeding area. In parallel, a control group of pigs with “normal” physical activity was included in the study. The methodology and results are described in details in D3.5. Lesions of osteochondrosis (OC) were assessed at slaughter to determine their correlation with lameness. The results confirmed that, in growing pigs, it is difficult to relate OC and lameness prevalence. There was no effect of increased physical activity on the incidence of leg disorders evaluated by either lameness or OC susceptibility.

Nutritional strategies to reduce digestive and leg disorders

One of the hypotheses tested in WP3 was that production diseases originate from competition for nutrient utilisation between “productive” and “non-productive” functions. In other words, animals selected for high productive traits may have higher requirement for certain nutrients to preserve their performance and health. These results provide the opportunity to reconsider current nutritional recommendations for genotypes selected for fast growth and high feed efficiency, thus addressing an important issue in the interaction between Genetics and Environment. Experimental trials were performed to test nutritional strategies to alleviate digestive disorders at weaning for piglets and leg disorders in both broilers and growing pigs with a focus on vitamin D on leg health.

Weaning of piglets is a critical period leading to an increase susceptibility of piglets to dysbiosis leading to digestive disorders, as well to systemic inflammation and oxidative stress. Feeding strategies contribute to reduce the negative consequences of weaning as well as the utilisation of antibiotics. The trial compared health and performance parameters of pigs after weaning when pigs were fed either a conventional dietary program or a complex feeding program based on diets formulated with highly digestible raw materials, protein concentrates in replacement of soya, prebiotics and antioxidant flavonoids. The complex feeding program reduced the incidence of digestive disorders and maintained growth performance of piglets after weaning. These effects observed in controlled and experimental conditions need to be confirmed in the field, where this feeding strategy is applied by farmers who do not use antibiotics or zinc oxide at pharmacological levels for several months (see WP6).

Vitamin D (vitD) is involved in skeletal integrity through the stimulation of the expression of genes in the small intestine which govern intestinal calcium and phosphorus absorption and in bone to osteoclast differentiation and calcium reabsorption promoting mineralisation of the bone matrix. Vitamin D deficiency causes poor mineralisation of skeleton. In commercial diets, vitD may be supplemented either in the form of cholecalciferol (D3), or in the form of 25-hydroxycholecalciferol (25-OH-D3), which is considered as the active metabolite.

The requirements for vitD for fattening pigs differ between countries (from 150 IU D3/ kg feed in the US to 460 IU D3/kg feed in Finland and Sweden). A question is arising on optimal vitD requirements for production and leg health. A study was conducted to determine the effects of dietary level (460 or 1380 IU/ kg) and source of vitD (25(OH)D3 or D3) on performance and leg health in growing boars. The methodology and results are described in details in D3.6. The results showed that increasing dietary content of 25-OH-D3 elevated vitD concentration in plasma of the growing boars more effectively than an equivalent increase in dietary content of D3. However, during the whole trial, vitD source had no effect on pig performance but the pigs tended to grow faster with the higher dietary inclusion of vitD. The dietary vitD source or level did not affect incidence or severity of OC in the humeri of the pigs. Bone bending moment which is often seen as the best indicator of phosphorus bioavailability in pigs, was improved by the higher inclusion of vitD. In conclusion, no practical advantages of 25-OH-D3 on leg health of the boars were found compared to D3 as the source of vitD. Increasing vitD supplementation had positive effects on growth performance and improved bone strength. The results show that the vitD values given in the nutrition recommendations for pigs describe the minimum need but they have not taken into account the optimisation of health and performance of pigs. More research about the effects of vitD on the prevalence of leg weakness and the potential to optimise pig health and productivity with supplemental D3 is needed.

In broilers, vitD dietary inclusion levels in commercial practice range between 2000 and 5000IU/kg of feed and nutrient recommendations by breeding companies for their fast growing genotypes suggest inclusion levels of 4000 IU/kg on average across ages. Currently, the inclusion of vitD and Ca and P additives in the drinking water, preventively as well as therapeutically, is a common practice in the broiler industry which is indicative of the uncertainties surrounding its optimum inclusion levels in commercial diets of modern genotypes. The objective of the study was to re-evaluate vitD requirements of 2 modern broiler genotypes with a special focus on leg health. The effects of dietary supplementation level (low level of D3 aimed at inducing a marginal vitamin D deficiency, a diet offering medium level of D3, close to what is used in commercial practice, a diet offering high levels of D3, which is above the EU limit of 5000 IU/kg) and source of vitD (partial substitution with 25-OH-D3 at commercial levels of supplementation) in two commonly used genotypes was examined. The methodology and results are described in detail in D3.6 and by Sakkas et al (2018). Performance was not affected by dietary treatments. However, walking capacity was improved by partial substitution of D3 with 25D3 and increased levels of bone mineralisation. These results together with those described in T3.2 (coccidiosis) suggest that a greater dietary inclusion level of vitD may be beneficial for leg health. These effects observed in controlled and experimental conditions need to be confirmed in the field (see WP6).

TASK 3.4 Quantification of the production consequences of defined challenges, to inform economic models

For Task 3.4 WP3 provided to WP7 data collected during the study described in D3.4 and showing that biosecurity, good hygiene and cleanliness of housing conditions were relevant intervention measures for prevention of production diseases. These data have been used by WP7 for evaluating the economic consequences of hygiene implementation (see Guidelines for Good Practice).

TASK 3.5 – Provision of blood and faeces samples from experiments with defined challenges to WP5

Task 3.5 aimed at providing blood, tissue and faeces that were collected from trials conducted on pigs and chicken to be used for the development of novel indicators for specific production diseases (see WP5).

The main objective of WP4 for the entire duration of the project was to determine the role of variability in the physical and microbial environment on the temporal expression of production diseases, specifically:
• Objective 1: To modify an existing protocol of collecting and summarising environmental data from pig and poultry farms (TASK 4.1).
• Objective 2: To determine the role of environmental factors on farm on the temporal expression of multifactorial production disease in pigs and poultry (broilers and layer chickens) (TASK 4.2).
• Objective 3: To determine the role of the microbial environment on the temporal expression of multifactorial production disease in pigs and poultry (broilers and layer chickens) (TASK 4.3).
• Objective 4: To develop IT systems for automation capture and processing of environmental and health data (TASK 4.4).

TASK 4.1 - The modification of existing protocols for collecting and summarising data from pig and poultry farms

To understand the temporal expression of disease over time within a farm environment, a system to monitor real time key environmental indicators (temperature, water intake, humidity and CO2) was developed. These factors were later related to the expression of multifactorial diseases (Task 4.2 and 4.3).
This was achieved through existing technology using sensors and wireless transmission via wireless connections (WiFi or cell phone SIM cards) adapted to farm requirements. This technology is already in place in numerous installations over the world, being affordable, robust and working independently of other control systems present on farm level.
A complementary system to collect data about clinical diseases and animal performance was also developed. It is based in a digital pen, paired with a Smartphone and digitised forms. Collected data were analysed using a classical statistical approach but were visualised as well using web-based dashboards by means of cloud processing (TASK 4.4). The system was already working in pig and poultry farms for obtaining metrics such as reproductive performance and has been adapted to record environmental and health collection data and monitoring within this project.
Both systems have been harmonised for the purposes of the project and specific work of development for both of them and their integration in a robust but user-friendly system has been performed.
The main value of the system and its differences from other products is its ability to integrate automatically recorded and real time environmental information with performance and health parameters. This was the basis for all data collection from animals and their environment and study their relationship with the temporal expression of farm diseases. The development of the new system for collecting and processing data automatically, related with environmental, health and performance at farm level was reported in deliverable D4.1.

TASK 4.2 - The role of environmental factors on farm on the temporal expression of production disease

From June 2015 to September 2016 different batches from different animal categories, either pig or poultry, were monitored to determine the role of environmental factors on farm on the temporal expression of production diseases. Environmental parameters such as temperature, humidity, CO2 levels and water consumption were measured using environmental sensors and were crossed with the productive information collected from the farms, according to the protocol defined in Task 4.1 and associated to the different animal categories:
• Pig: gestation, lactating, nursery, farrowing and gilts
• Poultry: broiler growers, broiler breeders, turkey growers, turkey breeders and layers
Fourteen farms located in Spain, Belgium, England and Cyprus took part in the monitoring period..
A logistic regression was performed for binary classification. Data collected by the environmental sensors involved hourly means of minute-resolution measures of batch room temperature, relative humidity, CO2 levels and water consumption. Sensor data were very noisy, with many null values, demanding a careful pre-processing in order to be incorporated in the model. However, it was possible to present a comparison of a commonly used classical statistical method optimised with the use of genetic algorithms and a deep learning-based model for the purpose of detecting anomalies related to increases in several diseases. The proposed methodology of a GRU-based auto encoder combined with a grid search-optimised classifier showed strong potential in its ability to detect anomalies.
The purpose of D4.2 submitted in January 2018, was to describe the material and methods used, to show a descriptive statistical analysis of the environmental parameters collected from the farms and to detail the complex statistical analysis carried out in the determination of the influence and detection of environmental factors pertaining to swine disease outbreaks in growing-finishing pigs. The main deliverables obtained from this task were the characterisation of the environmental and sanitary conditions of the experimental farms, including disease prevalence, and above all, a disease prediction algorithm.

TASK 4.3 - The role of the microbial environment on the temporal expression of multifactorial production disease in pigs and poultry

Task 4.3 investigated the relationship between the enteric microflora and the expression of production diseases in contemporary (within different airspace) and consecutive batches of pigs and flocks of poultry on the same site For pig categories, sample collection was performed in Spanish commercial production farms. Six-hundred and seventy-one samples were collected, cooled and sent in transport medium to the laboratory for phenotypic assessment as outlined previously (in Microbiological Testing in compliance with SOP MA17, screening for key indicator organisms such as Enterococcus spp, E. coli, Lactobacillus spp., Staphylococcus spp and Pseudomonas spp) as an index of flora variability and pathogen load. For poultry categories, 2500 poultry-origin samples were examined by the method described above over a 15-month period.
Statistical models were used to assess the difference in microflora composition between different sampling methods, consecutive samplings within the same batch/cycle, consecutive batches/cycles and between herds/farms, and the association between the changes in microflora and the expression of production diseases.
A potential relationship between the expression of production diseases and microflora-variation was investigated via Binary Logistic Regression analysis. The different categories that determined the microflora were first tested for normality and treated as independent categorical (ordinal) variables. The available level of performance or health data was very different between the different datasets. Overall, these key health and performance indicators were used as binary dependent variables in the analysis to predict a potential association between the microbial populations and the change in likelihood of the occurrence of production diseases. A possible association of the change in microflora with production disease was considered when the logistic regression model was significant and when at least one bacterial class was found to be a significant predictor in the model.
In poultry, residual bacterial contamination of farm environments showed clear positive and negative associations with particular flock traits relevant to production disease in broiler chickens. There were also significant associations between specific genera in boot swabs samples and relevant flock traits. The data-set supports the need for intervention strategies which focus broadly on the relationship between pathogen and beneficial organisms along the lines of the Weed/Seed/Feed-type strategies which are being used increasingly in the poultry industry.
In pigs, while interesting observations were made, bacteriological microflora evaluation alone was not a sufficiently consistent tool across farm types to systematically indicate health or production disease for the farms monitored. Further development and evaluation of these techniques may increase their utility and help reduce the need for antimicrobial treatments and deliver more sustainable and widely applicable strategies for health promotion in pigs.
The purpose of D4.3 submitted in May 2017, was to provide a very detailed description of the methodology used to determine the role of microflora profiles on the development of production diseases in pigs and poultry.

TASK 4.4 - Development of IT systems for automated capture of environmental and health data

Task 4.4 aimed to understand the temporal expression of disease over time within a farm environment, by developing a system which monitors key environmental indicators (temperature, water intake, humidity and CO2) in real time.
The dashboard developed in this task allows:
• The visualisation of the monitored data after their collection, giving a clear ‘picture’ about the environmental situation of each batch and its influence in productive and health parameters. The dashboard aims to give a service to the farmers to improve the environment, the health, and the productive data, that is to say, the dashboard has a didactic nature.
• Data comparison, integration and optimisation, through algorithms and parser for the aberrant data correction, that provides simplified tables where the information is gathered by animal age, gestation days and laying days. The unmanageable amount of collected data is simplified and it is possible to extract the tables obtained for later statistical analysis.
Design was based on requirements provided by PigCHAMP, with the major aim of staying consistent with the PROHEALTH project’s website. The style guide originating from the grant document was adjusted accordingly to prospective users, which include scientific and research community, farmers, veterinarians and general public.
The Deliverable D4.4 which describes in detail the different functions of the developed software, was finalised and submitted in February 2017. The software obtained in this deliverable is available at


All tasks were completed under one milestone required by the project (MS13).
The objectives of WP5 were:
• To use a microarray-based platform, deep sequencing and qRT-PCR and cellular analysis to monitor changes in host gene expression in pigs and poultry associated with production disease.
• Determine association between genotypes in pigs/chickens and disease/production phenotypes.
• Determine association between microbiota composition and disease/production phenotypes.
• Determine key changes associated with the onset of production diseases and use them for diagnostic purposes.
• To use information to suggest changes in diet and management which might ameliorate such key changes.

TASK 5.1 The aim of Task 5.1 was to develop an SOP for (a) the preparation of tissue and blood samples and (b) collection of faecal samples for microbiota characterisation (transport and processing).

An SOP was developed within the first few months of the project to ensure all tissues (intestinal and lung) were sampled from the same anatomical regions by all partners, prior to fixing in RNALater and shipment to Nottingham University (genomic analysis) and INRAT (Biomark qPCR). Equivalent faecal samples were also collected and fixed, as described in the SOP, prior to shipment to VRI for microbiota analysis (D5.1 month 3).

TASK 5.2 - Immunological, cellular and microbiological analysis

Studies performed at INRA Rennes (Partner 7a) were designed to test the hypothesis that hygiene of housing conditions could impact on growth performance and health. 80 (8 week old) Large white pigs were maintained for 6 weeks in unhygienic conditions (no cleaning of stalls) and 80 in clean conditions (cleaning of stalls). The genetic background of these pigs was one of high or low residual feed intake (H or LRFI) and the performance of the 80 pigs in total was analysed over the 3-month experimental period. There was a greater prevalence for pleurisy and pneumonia at slaughter in pigs housed in poor hygienic conditions and poor hygiene was also associated with lower growth. INRAR found that inflammatory markers (higher haptoglobin concentration and greater numbers of neutrophils in blood) were raised in poor hygiene.
Mesenteric lymph node (MLN), tracheobronchial lymph node (TLN) and Peyer’s patch (PP) tissues were removed at post-mortem and these were sent to INRA Tours to analyse 96 immune genes via a high through-put PCR (Biomark). This showed that genetic background (HRFI vs LRFI) had very little effect on expression of the 96 immune genes analysed but that rearing pigs in unhygienic conditions did. In unhygienic conditions most genes were down-regulated in the 3 tissues studied and a number of these genes (IFNA, TLR3, Mx2, MAVS, MDAP5 and OASL) were associated with viral infectionof lung tissues.
In aggregation of the PCA with observations and results from INRAR allowed INRAT to select MLN and TLN tissue from 13 animals for microarray analysis at UNOTT. Following microarray analysis, CD36 was the only to be differentially expressed by LRFI pigs reared in unhygienic conditions in either the MLN or TLN and was shown to be down-regulated in the TLN. Conversely, CD36 was up-regulated in the MLN of HRFI pigs reared in unhygienic conditions. CD36 is a scavenger receptor expressed on the surface of a number of different cell types including mononuclear phagocytes such as monocytes and macrophages. The remaining differential CD expression, therefore, only occurred in HRFI pigs reared in unhygienic conditions and 91 gene ontology pathways were affected in this group. Some of these pathways have an inflammatory phenotype (for example positive regulation of IL-12 production) while some have an anti-inflammatory phenotype (negative regulation of IL-8 production). In these pigs, a macrophage specific scavenger receptor (CD163) was up-regulated in the MLN but down-regulated in the TLN. CD163, expressing alveolar macrophages are one of the only cells to be permissive to PRRSV growth and transfection of non-permissive cells with porcine CD163 permits PRRSV growth in normally non-permissive cells. Therefore, down-regulation of CD163 in the TLN of HRFI pigs reared in unhygienic conditions may have reflected possible infections (pneumonia lesions) in these tissues. CD209 was also up-regulated in the MLN but down-regulated in the TLN. CD209 is better known as Dendritic Cell-Specific Intercellular adhesion molecule 3 (ICAM-3)-Grabbing Nonintegrin (DC-SIGN) and is a highly specific marker for DCs and in pigs is also associated with highly expressing CD163 DCs. Thus, gene analysis in HRFI pigs may suggest that unhygienic conditions promote an increase in macrophages and DCs or an increase in a CD163high/CD209high DC population in the MLN but a decrease in these populations in the TLN. CD74 in a number of mammalian species was regarded as being a component of HLA/class II and therefore expressed by all APCs. CD74 was only up-regulated in MLN of HRFI pigs reared in unhygienic conditions which may further indicate increased macrophage numbers in these tissues. Similarly, CD5L is mostly expressed by macrophages and is up-regulated in the MLN but down-regulated in the TLN of these pigs. Another co-stimulatory molecule (CD40) which is expressed by APCs, such as macrophages, DCs and B lymphocytes was up-regulated only in the intestines of HRFI pigs reared in unhygienic conditions.
Some lymphocyte specific markers were also differentially expressed in HRFI pigs but, once again, the majority of this expression occurred in the MLN of HRFI pigs reared in unhygienic conditions. In the TLN tissue of these pigs only CD3G was differentially expressed (down-regulated) while there was no differential expression in other genes which may have signalled changes in lymphocyte numbers. CD3G encodes the Gamma chain of the CD3-TCR complex and interestingly, although expression of CD3G was not altered in the MLN, CD3E (which encodes the Epsilon chain of the CD3-TCR complex) was increased. Furthermore, expression of genes associated with T cell subsets was also increased in HRFI pigs reared in unhygienic conditions. These included CD4, expressed by T-helper (Th) subsets and regulatory T cells (Tregs) and CD8B which encodes the Beta chain of the CD8 molecule expressed by cytotoxic T lymphocytes (CTLs). Since no other indicators of Treg activity were detected this would suggest that in HRFI pigs reared in unhygienic conditions there is an increase in Th and CTL populations in the MLN. A prominent B lymphocyte marker (CD79A) was also up-regulated in these tissues. CD79A genes encode the B-cell antigen receptor complex-associated protein alpha chain.
In a second study of pigs housed in old and new housing (tissue provided by agreement with TEAGASC IR), lung and intestinal samples were analysed histological and genomically 2 weeks post-wean. Pigs housed in the old environment failed to thrive and had clinical signs of respiratory infection. 17 pigs from each group (old and new housing) were analysed histologically. 7 pigs in the old environment had lesions consistent with supparative bronchopneumonia and a further 9 pigs showed thickening of the interalveolar septa, consistent with sub-acute interstitial pneumonia. The cells identified in the interalveolar septa were T lymphocytes and macrophages. Microarray analysis performed at Nottingham, determined that 49 gene ontology pathways were affected in the lungs of pigs housed in the old environment and indicated a T helper lymphocyte typ1/IFN-γ response with down-regulation of IL-4 (characteristic Th2 cytokine), consistent with histological findings. (D5.4 Month 36).
At INRAT, Immunological changes were studied in the intestine of fast growing Ross 308 and slower growing Ranger Classic broilers experimentally infected with 2.5x103 or 7X103 oocysts of Eimeria maxima obtained at day 6 and 13 post-infection (tissue provided by Newcastle University) and 28 day old Ross 308 broilers naturally infected with E. maxima and E. acervulina. In both studies, qPCR analysis indicated an inflammatory response in the intestine of broilers infected with Eimeria. Gene expression indicated an active Th1 response, with significantly increased expression of T cell (CD40 ligand), IL-17, IFN-γ and IFN-γR.
INRAT also measured immunological gene changes In Ross 308 broilers infected with 3 daily doses of Clostridium perfringens (5x108 CFU/ml) (tissue provided by Ghent University). Results from this study indicated that necrotic enteritis due to C. perfringens infection induced differential expression of 32 immune genes (all upregulated). These changes indicated an active Th1 response with increased expression of MHC1, CD28 and CD40L and innate immunity, with increased mannose binding lectin (opsonin) and pro-inflammatory chemokines (IL-6 and IL-8) which are chemoattractant for heterophils. (D5.4 Month 36).

TASK 5.3 Whole genomic gene expression

Microarray analysis was performed at the University of Nottingham using a 4x44k custom oligonucleotide microarray for pigs or chickens (Agilent, USA). 5 microarray analyses each were used to measure gene changes in the lungs of pigs maintained in hygienic or unhygienic conditions (INRAT), old or new housing (TEAGASC) or clinically diagnosed with pneumonia (Pig Champ, Spain). As well as pigs clinically diagnosed with pneumonia, pigs housed in unhygienic conditions or old housing showed signs of respiratory disease. Over 500 genes were differentially expressed (DE) in the lungs of pigs with pneumonia, maintained in unhygienic conditions or in old housing. Pathway analysis and molecular marker prediction from these samples will be discussed in Task 5,4 and 5.7 respectively.
A high residual feed intake (HRFI) or low residual feed intake (LRFI) background influenced gene expression when pigs were maintained in unhygienic conditions (data obtained from 5 microarrays each from MLN or TLN from HRFI or LRFI pigs maintained in either hygienic or unhygienic conditions). Gene expression in the MLN was also much greater than the TLN in pigs housed in unhygienic conditions from either genetic background.
When LRFI pigs were maintained in unhygienic conditions, 118 genes in the MLN were DE (67 up-regulated and 51 down-regulated) compared to MLN genes in LRFI pigs maintained in hygienic conditions. However, when HRFI pigs were maintained in unhygienic conditions, 3498 genes were DE in the MLN (856 up-regulated and 2642 down-regulated) compared to MLN tissues from pigs maintained in hygienic conditions. In comparison, When LRFI pigs were maintained in unhygienic conditions, 96 genes in the TLN were DE (25 up-regulated and 71 down-regulated) compared to TLN genes in LRFI pigs maintained in hygienic conditions. In HRFI pigs maintained in unhygienic conditions, 563 genes were DE (447 up-regulated and 116 down-regulated).
Gene expression from 5 microarrays per group were used to determine changes in the skeletal and cartilagenous tissues of pigs diagnosed with kyphosis, compared with age matched healthy pigs (tissue obtained from Newcastle University). In this study a very obvious difference was shown. In total, 1196 genes were differentially expressed in bone but only 348 genes were differentially expressed in cartilage, compared to healthy controls. The direction of gene expression was also very different in the two tissues. In bone 60% of genes which were differentially expressed >2 fold were down-regulated (40% upregulated) but in cartilage only 11% of genes which were differentially expressed >2 fold were down-regulated (89% up-regulated). Also in bone, 75% of genes which were differentially expressed >5 fold were down-regulated (25% up-regulated), whereas in cartilage only 2% of genes which were differentially expressed >5 fold were down-regulated (98% up-regulated). This may suggest a shift from bone to cartilage turn-over in kyphotic pigs. Evidence to indicate this was also seen by the fact that no common genes were up-regulated or down-regulated in both tissues; therefore gene expression was separate in both.
In broilers whole gene expression was measured by microarray in intestinal tissues from Ross 308 infected with 7X103 oocysts of Eimeria maxima obtained at day 6 post-infection (tissue provided by Newcastle University) and 28 day old Ross 308 broilers naturally infected with E. maxima and E. acervulina. 1033 genes were DE in broilers experimentally infected with E. maxima, 855 were up-regulated and 178 were down-regulated, compared to uninfected controls. In naturally infected Ross 308, 2875 genes were DE with 1498 up-regulated and 1377 down-regulated compared to non-clinical flock mates. However, it was hypothesised that the non-clinical flock mates would be sub-clinically infected. This was found to be the case and comparison of gene expression in tissue from clinical, sub-clinical and uninfected Ross308 has formed the basis of biomarker identification (see Task 5.7).
In Ross 308 broilers experimentally infected with C. perfringens, to induce necrotic enteritis (NE) (Ghent University), 2095 genes were DE in the intestine (1241 up-regulated and 854 down-regulated) compared to uninfected controls. This work has also led to development of a biomarker panel which may differentiate between coccidiosis and NE (see Task 5.7). Work completed within Task 5.3 was performed to meet D5.2 and D5.6.

TASK 5.4 Mapping gene expression to pathways

Changes in gene expression determined in Task 5.3 were analysed by GENESPRING software which determined fold change in expression and clustering via Principle Component Analysis. Gene ontology pathways affected by differential gene expression were analysed by Ingenuity pathway analysis (discussed in Task 5.3). Work was performed to meet and is detailed in D5.2 and D5.6.

TASK 5.5 Microbiota characterisation

Illuminseq high through-put sequence analyses of the variable (V3/V4) region of the 16s rRNA genes which were contained within the bacterial microbiota was performed at VRI. A comparison of the contigs produced with known reference sequences was then used to identify the bacterial species and any operational taxonomic units (OTUs) which were present in the faecal or caecal samples.
Fourty-eight newly hatched ISA Brown males of the egg laying chicken line were divided into 2 groups. Group 1 received a standard diet and Group 2 received a standard feed supplemented with liposoluble phenolic acid extract from Curcuma and hydrosoluble extract from Scutellaria. The supplemented diets were provided from d1 of age. Seven d later, half of the chickens in each group were infected with 1 × 107 cfu S. Enteritidis 147 spontaneously resistant to nalidixic acid.
Infection with S. Enteritidis affected microbiota composition at d4 post infection, i.e. in 11-day-old chickens, since in 4 chickens, Escherichia coli formed 30 to 70% of all microbiota. A significantly higher abundance of Escherichia coli and Clostridium XI occurred at the expense of Clostridium XlVa and Lachnospiracea incertae sedis which were lowered in S. Enteritidis infected chickens in comparison to uninfected controls. However, the combined effect of both S. Enteritidis infection and plant extract supplementation resulted in the most extensive changes in microbiota composition, both at 4 and 14 d post-infection. Microbiota of S. Enteritidis infected chickens fed a diet with plant extracts was characteristic with significantly increased levels of Lachnospiracea incertae sedis, Clostridium XI, Blautia, Flavonifractor, Coprococcus, Clostridium IV, Lactobacillus, and Faecalibacterium, and a decreased abundance of Clostridium XlVa) in comparison to microbiota of the uninfected chickens. Lactobacillus and Faecalibacterium are both considered to have a positive effect on gut health.
Microbiota sequencing of caecal contents in pigs with an HRFI or LRFI background also showed that Adaptation of microbiota to new diet after weaning was slower in LRFI than in HRFI pigs. Sanitary stress was of relatively minor influence on pig microbiota composition in both tested lines although abundance of Helicobacter increased in LRFI pigs subjected to stress. Selection for residual feed intake thus resulted in a selection of fecal microbiota of different composition. Furthermore, VRI also reported that Sows maintained during gestation on slatted floors had a different microbiota compared to sows maintained in an enriched system (deep straw bedding), thus showing that the production system influences microbiota composition. The microbiota of 1- and 4-day-old piglets differed from the microbiota of sows and sows therefore did not represent the most important source for their colonisation in early days of life

TASK 5.6 - Genome typing and identification of polymorphisms

Genome typing was performed by qPCR at University of Nottingham following multiple test correction of larger gene sets of interest determined in Tasks 5.3 and 5.4. Gene panels formed the basis of biomarker determination discussed in Task 5.7. Polymorphisms were not determined in the project since, in light of the large numbers of genes changed highlighted by infection or housing types etc., subtle changes in individual genes were considered probably less important.
MicroRNA profiling was performed at University of Nottingham using MioSeq 500 analysis. This has determined profiles associated with Mycoplasma infection in pigs and clinical versus subclinical Coccidiosis in broilers (see Task 5.7).

Task 5.7 - Integration of data and molecular marker prediction

Prediction of microbiota in healthy, susceptible and diseased chickens, as well as due to genetic background or housing types in pigs was performed at VRI, as discussed in Task 5.5.
At University of Nottingham gene expression associated with E. maxima infection in Ross 308 and Ranger Classic broilers were determined (Task 5.3 and 5.4) and a smaller panel of genes were chosen following MTC analysis, level of expression and known biological effect for qPCR validation. At day 6 post-infection with 2.5x103 oocysts (peak infection dose and time measured in the Newcastle study) four genes (ADD3, MLLT10, NAV2 and PLXNA2) were upregulated in Ross 308 but were not DE in Ranger Classic. Six genes (PTPRF, NCOR1, CSF3, SGK1, CROR and CD1B) were upregulated in both Ross 308 and Ranger Classic infected with 2500 oocysts at 6 dpi. This data has formed the basis of a potential biomarker panel of coccidiosis in the two lines and shows differential gene expression in these two lines following infection with equivalent dose (and time post-infection). Deep sequencing of miRNA has also detected novel miRNAs which are expressed in clinical and sub-clinical coccidiosis. Following false detection rate analysis, qPCR highlighted a panel of miRNA genes which could be used to differentiate clinical coccidiosis (gga-miR-122-5p and gga-miR-31-5p) from sub-clinical coccidiosis (gga-miR-122-5p, gga-miR-205b and gga-miR-144-3p).
Microarray analysis of intestinal tissues from Ross 308 broilers with experimental E. maxima infection and C. perfringens infection (to induce NE) (Discussed in Task 5.3). Following MTC a panel of genes were chosen based on expression and biological effect to further investigate by qPCR. Our data shows that 2 genes associated with low oxygen tension (ACE1 and LDHDA) were significantly upregulated in both diseases but 6 genes which were significantly down-regulated in NE (COL8A1, COL1A1, MYL4, MYLK, SERPINH1 and TUBB6) were significantly up-regulated in coccidiosis. This data may form the basis of a biomarker panel which differentiates these diseases early in infection.

Over 500 genes were differentially expressed (DE) in the lungs of pigs with pneumonia, maintained in unhygienic conditions or in old housing (Task 5.3). Pathway analysis indicated that only 4 genes were DE across all groups. These were ATPB1, GPX3, IRF9 (up-regulated) and GATAD2B (down-regulated) which may form the basis of a biomarker panel for respiratory disease in pigs.
Work described in Task 5.7 was performed to meet D5.6.

TASK 6.1 - Synthesis of project results and formulation of improvement strategies

The aim of this task was to evaluate results obtained in WPs 1-5 in a project workshop in conjunction with the scientific literature in order to select the most promising improvement strategies (outlined in tasks 6.2 6.3 and 6.4) for testing in large scale farm intervention studies. These have been outlined in D6.1 (M39).

TASK 6.2 – Quantification of the effects of the improvement strategies on the prevalence of different production diseases in pigs

In Task 6.2 three of the most promising strategies to reduce the prevalence of production diseases in pigs, were tested in field conditions. The deliverable constituted of three sub-tasks.

SUBTASK 6.2.1 – The impact of positive handling and music introduction and use of Pig Appeasing Pheromone in the farrowing unit.

In subtask 6.2.1 two trials were conducted investigating the effect of positive handling a) on sow behaviour and performance b) on post-weaning piglet performance, health and behaviour g c) whether it interacts with pheromone use on post-weaning piglet performance. The first experiment assessed the effect of sow positive handling (treated; music and back scratching) vs no treatment (control) in the farrowing unit. Overall, treated sows showed significantly reduced piglet mortality (≈2.08 % less) in comparison to non-treated sows. In addition, piglets from treated sows weighed more than piglets from control sows at weaning (6.1 kg vs 5.35 kg). The second trial investigated the long term effect of sow positive handling and pheromone use in the nursery on piglet health and performance during the post-weaning period. In total, 2592 weaned pigs from control and treated batches were allocated to 4 treatments: CC for piglets born from Control sows and not treated at weaning (no pheromone), CT for piglets born from Control sows and treated at weaning (pheromone), TT for piglets born from Treated sows and treated at weaning (pheromone), TC for piglets born from Treated sows and not treated at weaning (no pheromone). Statistics were not carried due to the lack of individual piglet weight and feed intake. Piglets originating from treated sows showed superior performance and reduced mortality. Likewise, piglets that received positive handling at the nursery performed better than control piglets. Overall TT piglets performed the best and CC the worse.

SUBTASK 6.2.2 – A retrospective evaluation of the efficacy of non-medicated pre-starter feed formulation on pig health and on antibiotic usage after weaning

Subtask 6.2.2 consisted of a retrospective survey which aimed at a) investigating the effects of moving from a medicated pre-starter feed after weaning to a non-medicated diet on pig health, productive performance and farm economic output b) to identify specific characteristics of farm management which use antibiotic-free pre-starter feed c) To evaluate if the decision to stop using antibiotics in pre-starter feed changed the use of antibiotics administration by other routes (drinking water and injectable form of antibiotics). Data from selected farms that fulfilled a list of criteria were collected over 2011-2016 and farms were divided in 3 groups: farms using an antibiotic free prestarter feed since 2011, farms having changed during the studied period to an antibiotic free feed and farms still using medicated prestarter feed (NM, T, M groups, respectively). The comparison of M and NM groups shows no difference in economical, technical and health outputs during the weaning to finishing period. However, they differed in terms of management practices, indicating a higher level of organisation for NM farms than M farms whilst T farms were intermediate. Decrease performance and increased loss rates were evident for T farms indicating that the transition to a non-medicated pre-starter feed may have negative and transient effects related to biosecurity and husbandry practices. On the other hand, farmers with a good biosecurity (NM group) have equivalent results to farmers using pre-starter feed with antibiotics and they produce at the same cost whilst achieving higher profit due to participation in quality programs that pays more for pigs.

SUBTASK 6.2.3 – The identification and specific nutrition of piglets with a low immunoglobulin (IgG) level in the blood.

Subtask 6.2.3 investigated the relationship between IgG concentration in colostrum and in piglet serum after colostrum intake, and productive performance and health status of pigs until slaughter in piglets from primiparous and multiparous sows and to evaluate the impact of dietary amino acid profile during the nursery and fattening periods on the productive performance of pigs with a reduced IgG concentration. Piglets born to primiparous sows had lower growth rate than to multiparous sows. IgG concentrations determined with ELISA were not correlated with the Brix refractometer which may allow rapid measurement of IgG. This result suggests that the Brix refractometer cannot be used as a valid quick method to evaluate the IgG concentration on farm.
Results obtained in these studies did not demonstrate that growth performance can be predicted from IgG concentration measured in piglet serum after colostrum intake. However, pigs with higher IgG serum concentration after colostrum intake showed higher growth rate than pigs with a lower concentration. Poor performance of more susceptible pigs (piglets born from primiparous sows, PP pigs) can be improved in the immediate post-weaning phase with a more adjusted feed. However, after the pre-starter phase, higher growth rate promoted by the enriched feed is mainly associated with higher voluntary feed intake, affecting the total feed efficiency. In pigs from multiparous sows, an enriched feed during the nursery phase did not provide any performance advantage compare with control high lysine feeds.

TASK 6.3 – Quantification of the effects of the improvement strategies on the prevalence of different production diseases in broilers, laying hens and turkeys

In Task 6.3 three of the most promising strategies, associated or not with additional treatments, were tested in field conditions. The deliverable constituted of three sub-tasks.

SUBTASK 6.3.1 – Field evaluation of nutritional treatments to reduce skeletal disorders in broilers

In subtask 6.3.1 three trials were performed assessing the efficacy of vitD metabolites. In the 1st trial, 2 broiler houses within the same farm were offered diets with D3 or diets where D3 is partially substituted by 25D3 (1:1) at commercially supplemented levels in broilers and reared to an older age (≈ d48) for the roaster market and being fed coccidiostat-free diets. The experiment was repeated over 4 production cycles. In the absence of effects on performance or carcass variables assessed at the slaughterhouse broilers fed diets with addition of 25D3 showed overall better walking capacity at the end of the trial due to a lower incidence of birds with impaired gait. In subtask 2 the same experimental design was applied but birds received diets with commercial levels of D3 or with additional 1,25-D3 according to manufacturer’s instructions, whilst birds were raised at a younger age (≈35d) and received diets with in feed coccidiostats. An improvement of weight gain was evident at d18 of age, but it was not statistically significant at the end of the trial. Addition of 1,25-D3 did not improve gait score, but increased the incidence of hock burns, although overall leg health was very good in this trial owing to the younger age at slaughter. Finally, the 3rd trial assessed the effects of partial substitution of D3 by 25D3 (1:1) at commercially supplemented levels in birds raised in fresh or re-used litter and receiving feeds with in feed coccidiostats. This experiment was carried in experimental facilities, albeit in large pens and at final stocking densities similar to these of commercial farms. Allocating birds in re-used litter provided a challenge to bird health and productivity and walking capacity with substantial differences in their underlying immune responses. Offering 25D3 improved performance, irrespective of litter substrate. Although, it did not result in an improvement of the gait score or an improvement of bone quality, 25D3 improved the foot pad dermatitis score. Collectively this sub-task illustrated the ability of vitD metabolites to improve aspects of performance, walking capacity and health of broilers in commercial settings. These results are not consistent in different production systems, whilst they seem to be more pronounced for birds raised to older ages.

SUBTASK 6.3.2 – Field trial comparison of the biological and economic performance of commercial broilers and pullets grown under an enhanced protocol treatment against normal farm practice

In subtask 6.3.2 the objective was to evaluate, in commercial farms, the effects of an enhanced protocol treatment (EPT) on health, productive performance of the broilers/pullets and the economic output for the farms. Two commercial Cypriot farm sites belonging to the same integrator and one commercial UK farm site were identified for testing the EPT in broilers. One commercial UK farm site was selected for testing EPT in laying hen pullets. The production costs broiler farms decreased, because of adaptations in the vaccination protocols and a decreased prevalence of clinical diseases as the biosecurity was improved on farm level. For laying hen/pullets implementation of the ETP protocol reduced mortality rate.

SUBTASK 6.3.3 – Environmental enrichment and leg health and bone properties in broiler chicken.

In subtask 6.3.3 the objective was to compare eight types of environmental enrichment and a non-enriched control in relation to effects on leg health in broiler chickens. The included outcomes were live bird observations including gait score, hock burns and leg deformities and post mortem observations including body condition score, various pathological conditions and bone and muscle measures of the leg. The hypothesis was that the provision of environmental enrichment would result in an improvement in leg health. The results from the study showed that the control, platforms, straw bales and roughage did not have a positive effect on leg health. On the other hand, lowering stocking density, introduction of vertical panels and increasing distance between resources, resulted in lowered gait scores, lower prevalence of arthritis and tenosynovitis and wider leg muscle and bones and may be employed to improve broiler leg health and welfare.

TASK 6.4 – Data on performance, disease prevalence and economic consequences of specific improvement strategies for parameterisation of WP7 models

The purpose of Task 6.4 was to summarise the main characteristics of data which were delivered to WP7 for further analysis from WP6. The datasets included parameters on performance, disease prevalence and economic consequences of specific improvement strategies implemented in WP6. Because of varying nature of experiments, each experiment provided somewhat different set of parameters. However, each dataset allowed economic analysis either by itself, or with complementary information gathered via secondary research. In the first experiment focusing on pigs, positive handling of the sows appeared to have a positive impact on piglet mortality, which can provide economic benefits. In the second experiment, the costs of feeding piglets from primiparous sows appeared to be costlier than feeding piglets from multiparous sows to the same live weight. Nutritional strategies to support problem-pigs with a high tryptophan-threonine feed appeared to influence the performance of the pigs, and to increase the costs of feeding. In the third experiment, the transition period from medicated to non-medicated pre-starter piglet feed appeared to have a short-term negative impact on loss rate and average daily gain. In broilers, the first experiment focused on vitD requirements in broilers and their impacts on leg health. VitD supplementation appeared to affect incidence of leg disorders, but no substantial effects on production performance was observed, and a negative effect was seen in the rate of mortality. In the second reported poultry experiment, several environmental enrichment and leg health interventions were investigated. The data indicated that there is a strong tendency for interventions to impact prevalence of foot pad dermatitis. The third poultry experiment focused on an enhanced protocol treatment (EPT). This experiment provided full data to investigate whether enhanced health care provided economic benefit to the farmer. Altogether these experiments will provide useful data for bioeconomic modelling and value chain analysis to be conducted in WP7 of PROHEALTH.


The objectives of WP7 were:
• To characterise socio-economic impacts of existing pathologies, including impacts related to animal welfare, ethical considerations, and impacts on costs and efficiency of production on the farm (TASK 7.1)
• To identify key issues and future possibilities in animal health (TASK 7.2)
• To estimate the financial and economic consequences of proposed interventions (TASK 7.3)
• To assess whether consumers in different countries accept the proposed interventions, and assess their priorities in relation to animal antibiotic use, human and animal health and information needs (TASK 7.4)
• To assess the value-added potential of interventions and new business opportunities and draw policy recommendations that can add value on the sustainability of pig and poultry systems in Europe (TASK 7.5).

TASK 7.1 – Characterisation of socio-economic issues, cost and efficiency impacts of pathologies

Meta-analyses and reviews to assess farm-level economic and productivity impacts of pathologies in pigs (N=130 studies) and poultry (N=127) and to understand what consumers and citizens think of production diseases (N=80) and to assess consumer willingness-to-pay (WTP, N=54) for interventions to reduce production diseases were conducted. Existing literature on the costs of production diseases was scarce, but it showed that the costs vary by disease incidence and severity. Endemic diseases in a pig herd facing disease problems may cost up to €30 to €40 per fattened pig. Porcine respiratory disease complex, mortality, reproductive disorders and lameness were among the costliest production diseases in pigs. For chickens, literature on nine health conditions was reviewed. Financial losses due to keel bone damage were estimated at €4 and infectious bronchitis at €3.2 per laying hen. In broilers, losses due to uncontrolled clostridiosis were about €1 and coccidiosis €0.21 per bird. Disease losses could be reduced by the intervention measures.
The review showed that the public are concerned about prophylactic use of antimicrobials and farm animal welfare (FAW) in modern production systems. Concerns on FAW varied by person’s age, gender, education and familiarity with farming. The review indicated a small WTP for FAW, varying by factors such as animal type and region. Knowledge gaps regarding the public’s attitudes towards specific production diseases and WTP for measures to reduce these diseases were identified. A mix of market and policy solutions appeared as the best way to improve FAW standards.

TASK 7.2 – Stakeholder consultation to identify key issues and future possibilities in animal health

The views of supply-side stakeholders along the food supply chain from five countries (UK, FI, ES, PL, DE) were elicited to identify potential solutions to key issues. Altogether N=100 full and N=70 abbreviated questionnaire (PL) responses were received. Gastrointestinal diseases (>61% of responses for broilers (DE/ES, PL, UK); laying hens (UK)), systemic bacterial infections (broilers (FI), laying hens (FI)) and reproductive disorders (laying hens (PL)) were ranked as the most important disease categories economically. The most often mentioned production disease in broilers was coccidiosis, whereas in laying hens it was salpingoperitonitis syndrome. The most prospective interventions for multiple disease control in chicken were improvements in ventilation, litter quality, biosecurity and hygiene, vaccination and adjustment in feed composition. Unnecessary medication was not preferred. For sows, reproductive disorders were overall the most important production diseases economically (>30% of responses). However, as per country, locomotory disorders ranked more important in FI and gastrointestinal disorders in PL. The most important disease related to sows was neonatal and perinatal mortality followed by lameness. For weaners, growers and fattening pigs, respiratory disorders were ranked as the most important diseases economically (>35% of responses). Porcine respiratory disease complex, post-weaning diarrhoea and porcine enteric disease complex were the most important individual diseases.

TASK 7.3 – Supporting farm-level solutions through bio-economic modelling

Three bioeconomic models to simulate the scale of disease, production performance and animal welfare issues in the herd/flock were developed. The pig production models utilised stochastic dynamic optimisation and the poultry model used linear programming. For fattening pigs, dirty housing (an uncleaned room, reduced ventilation and biosecurity) and clean housing (room cleaned before pigs arrive and daily thereafter, good biosecurity) were compared. The comparison was based on WP3 results. For piglet production, several management modifications were analysed: 1) buying sows with superior genetics (disease resistance or improved maternal abilities), and provision of assistance to piglets to reduce mortality; 2) the application of Mycoplasma hyopneumoniae vaccination for piglets; 3) the application of enrichment material for sows; 4) the use of mechanical ventilation in the pig house; and 5) changing the interior design of a farrowing unit, 6) positive handling of sows (backscratching the sows and playing them music for a period around farrowing, pig appeasing pheromone provided for pigs) and 7) nutritional strategies to support the growth and health of more sensitive subpopulations of piglets. Modifications 1-5 were based on WP1 and WP2 analyses and 6 and 7 on WP6 trials. For broilers production, WP3 and WP6 results were used to investigate 1) seven management modifications to prevent leg disorders, and including increasing the distance between feeders and drinkers up to 3.5m or up to 7m; having three pans of maize roughage, vertical panels, bales of straw, a platform raised by 30cm plus access ramps placed or a platform raised by 5cm but without ramps placed in the pen; and lowering stocking density (34Kg/m2), 2) vitD supplementation in broiler chicken to enhance their leg health, and 3) enhanced veterinary treatment protocols (ETP). The main components of ETP were structured microbiological monitoring, structured veterinary visits three times per flock lifetime, and monitoring ammonia in each house.
Several financially viable interventions were identified. The most viable interventions were improved hygiene and clean housing conditions for pigs (the financial net benefit estimated to range from €13 to €23 per pig), positive handling of sows (music and backscratching; €0.8 to €2.4 per piglet), interventions to improve piglet survival (€0.5 to €3.3 per piglet), providing enrichment to gestating sows (€0.2 to €1.1 per piglet), use of mechanical ventilation in the sow house (<€1 per piglet), maintaining a distance of 3.5 metres between drinkers and feeders in broiler houses (€0.99 per kg broiler liveweight), lowered stocking density in broiler houses (€0.26 per kg liveweight). For other interventions studied, the profitability was case-specific. The models could be used in further research to illustrate the role of interventions in different conditions and countries.

TASK 7.4 – Understanding consumer priorities and acceptance of proposed interventions

A citizen survey was conducted in 2017 in five countries (DE, ES, FI, PL, UK) covering layers (N=789 responses), broilers (N=790) and pigs (N=751). The results indicated that, the majority of respondents in all countries were unfamiliar with farming practices, with most respondents also either unsure, or disagreeing, that they purchased foods produced from intensive animal production systems. The results indicated the need for effective communication and assurance regarding regulations and measures to ensure the safety of the animal products. Several benefits of intensive production systems were acknowledged by the respondents, mainly in relation to resource and cost efficiency of these systems. Respondents viewed these systems also unfavourably, especially in relation to increased animal stress, increased risks of animal diseases and it being thought of as an unnatural production system. The most frequently mentioned concerns regarding intensive animal production systems were related to whether minimum welfare standards were being achieved, antibiotic use leading to antibiotic resistance, antibiotic residues and food safety. The results suggested that antibiotic usage and food safety are inextricably linked in the minds of consumers.

The respondents’ concerns were reflected in the disease mitigation strategies they prefer. The most preferred interventions included modifications to housing design, enhanced hygiene, providing enrichment materials and an environment where the animals can perform natural behaviours in pigs and poultry; and reducing stocking densities in poultry and; efficient monitoring of pigs and their housing and adjustments to the diets of pigs. These can be perceived as more natural and less invasive interventions than the other interventions proposed. Using antibiotics, medicines or vaccination were among the least preferred interventions. The survey data could be further analysed in subsequent studies because the dataset is extensive.

TASK 7.5 – Integrated quantitative analysis of value creation potential – a whole chain assessment

A value chain analysis (VCA) was carried out to assess the financial effects on various actors in the value chain, in particular farmers and consumers, and to identify how interventions could affect other actors in the pig and broiler value chains. For pigs, two disease mitigation scenarios were examined: 1) improved hygiene in pig fattening; and 2) better-targeted piglet feeding and sow genetics, management and handling measures reducing piglet mortality. For broilers, two management scenarios were examined: 1) enhanced veterinary treatment protocols; and 2) a measure to improve leg health in broilers. The interventions were analysed using qualitative and quantitative techniques, and in both generic and in country-specific context.

Among the four interventions, improved pig hygiene was seen to have the largest potential to add value by improving production efficiency and reducing costs. Financially, the benefit of improved hygiene to pig fattening farms was around +50% in gross margin, corresponding to up to 5% change in the consumer price for pork. Measures to reduce piglet mortality and positive handling of sows were estimated to increase gross margin by up to 3.3%. Of the two poultry interventions studied, increasing the distance between drinkers and feeders was estimated to increase net margin by 1-2%. Enhanced treatment protocols for broilers may reduce the financial performance in the short term because of the increased costs involved, but increase the robustness of the system, facilitate rapid response to emerging diseases and add value in the long term.

All interventions that were examined are preventative and, in this respect, they can add value to the products. There appeared to be market potential for interventions which are considered animal-friendly and carefully targeted, needs-based medication of the animals. Positive handling of sows and reduced piglet mortality, increased distance between broiler feeders and the drinkers, and improved hygiene appeared to be win-win interventions which are likely to add both value to the consumer and financial benefits to farmers. The positive handling intervention could be particularly important to consumers. These interventions could be incorporated into quality assurance schemes, thus providing new business opportunities to food sector operators.

Consumers were likely to be appreciative of all four interventions studied as they improved animal welfare without medication. Supply-side experts viewed the interventions quite positively, but were pondering the proportion of farmers who would be willing to adopt poultry interventions and sow positive handling intervention. Policy measures regarding studied interventions can be warranted because animal antimicrobial use, contributing to the risk of antimicrobial resistance, and food safety are societally important issues. Because both pig and poultry systems tend to be vertically integrated, there are opportunities to adopt interventions which look at animal health from the system perspective. The VCA results can be used as an example of how economic benefits of an animal health intervention can be analysed systematically.

Potential Impact:
Following the structure of the main S&T results, the presentation of the potential impact is also organised along the seven S&T work packages (WP) of the project. WP6 and WP7 were the WPs that synthesised the work of the other WPs. Clearly the impacts of individual WPs contribute towards the overarching impact of reducing the incidence of production diseases through the development, evaluation and dissemination of effective management and control strategies for these conditions. Data generated by PROHEALTH would lead to a more co-ordinated and holistic approach to improve animal health and welfare, reduce antimicrobial usage and resistance, and increase the profitability and sustainability of animal production. We have identified four classes of beneficiaries for the project impacts: academic beneficiaries, policy makers, stakeholder end users, such as farmers, livestock breeders, feed producers etc., and society at large.

WP1 focussed on 1) assessing health, welfare and performance over a three-year period (2011-2013), 2) scoring biosecurity in a standardised way and 3) identifying risk factors for poor health, welfare and production in pig and poultry farms with intensive production systems in different European countries. Impact has been achieved as follows.

Academic beneficiaries
Some interesting trends requiring further study were observed in the performance, health and welfare parameters from 2011 to 2013 (D1.1) (Van Limbergen et al. 2015; Klinkenberg et al. 2016a,b). Overall, performance parameters improved over time, both in pigs (mainly sows, less in fattening pigs) and poultry. However, trends in health and welfare traits over the years were less clear than those for performance, indicating room for improvement and the need for further study.
The biosecurity score we developed (D1.2) was successfully used in a large number of different pig and poultry farms across Europe. It can serve as an example for developing similar scoring systems in other animal species and also be used in farms outside Europe. The strengths and weaknesses in specific biosecurity areas were identified. As expected, there was a large variation in the level of biosecurity between farms. The scores indicate that in pig farms, there is more potential to improve internal biosecurity (i.e. measures for the prevention of the spreading of pathogens within a farm) than external biosecurity (i.e. measures for the prevention of pathogens entering a farm), whereas the opposite is true in poultry farms (D1.3) (Chantziaras et al. 2018a; Van Limbergen et al. 2018). Further research e.g. by following-up (specific) farms more closely over time and/or more in depth is needed to improve the biosecurity in pig and poultry farms. Such studies could also elucidate why some measures are not adopted by farmers.
The risk factor studies in the pig and poultry farms (D1.4) identified and quantified a large number of interesting factors (environment, housing, nutrition, management, health programs) that were associated with health, welfare and performance in pig and poultry farms (Pandolfi et al, 2018; Chantziaras et al. 2018b; Van Limbergen et al. 2018). The studies clearly elucidated the multifactorial nature of production diseases, and generated ample hypotheses for further studies to better understand the complex interplay of the associations and to assess possible causal links for these factors.

Policy makers
The biosecurity tool and the outcomes from its application during the project has the potential to identify areas where both internal and external biosecurity can be improved and the incidence of production diseases can be reduced. Although most of these improvement will arise from on farm changes, some may be embedded in the development of legislation across the EU. This will not only influence the incidence and impact of production diseases, but it is likely to yield benefits for other livestock (epidemic) conditions.

End Users
The results of D1.1 demonstrated that production diseases are still very common in intensive pig and poultry farms in Europe, and that for many parameters (mainly related to health and welfare) the prevalence remained more or less the same over a three-year period addressed. The results elucidated which areas of performance, health and welfare parameters need most attention in pig and poultry farms. This is important for the pig and poultry industry, as production diseases entail huge costs, as identified in WP7.
The development and successful application of the biosecurity scoring tool (D1.2) allows other pig and poultry farms to also use the tool and to objectively evaluate the biosecurity level in their farms. This will increase awareness, which is a first and very important first step for further control. In addition, it also allows benchmarking between farms, which is a further step to ultimately improve biosecurity in the pig and poultry industry in Europe. For example in Finland, PROHEALTH has contributed to the discussion which has led to the introduction of systematic biosecurity assessment in the Finnish livestock section.
The results of the biosecurity scoring (D1.3) suggest a large potential to improve biosecurity and revealed the most critical areas. There was a large variation between farms. In pig farms, internal biosecurity scores were lower than external biosecurity. In broiler and laying hen farms, external biosecurity has more possibilities for improvement than internal biosecurity.
The risk and protective factors associated with health, welfare and performance in pig and poultry farms (D1.4) can be used by farmers and stakeholders to implement proper measures. Most identified factors relate to environment, housing, nutrition, management, health programs, and can be optimized in the pig and poultry farms. In pig farms, measures to prevent the spread of pathogens within the pig farm, in particular, enhanced management practices during the farrowing and nursery phase, greater biosecurity between compartments and age groups, and more comprehensive cleaning and disinfection practices are recommended (Chantziaras et al. 2018b). In poultry farms, overall measures to prevent the pathogens from entering the farm, in particular improving biosecurity and hygiene with visitors and staff entering the farm, practices upon depopulating the broiler house and when purchasing day-old chicks are to be improved (Van Limbergen et al. 2018). The type of risk factor and the strength of the association with the outcome parameter are particularly important for the individual farmer. In addition to that, the occurrence (prevalence) of the risk factors in different European pig and poultry farms is mainly important for the entire pig and poultry sectors as that may help them to set priorities at country and at European level.

Socio-economic impact
Given the multifactorial nature of production diseases, a holistic approach has the best chances for success. Control of the risk factors may lead to major economic benefits. Healthy animals raised in animal welfare-friendly systems with high biosecurity standards is also beneficial for the consumer and the public perception, the environment, the export position of animal production in Europe and the commitment of Europe towards the international community. The biosecurity scoring tool and the results of the risk factor studies might be consulted and applied by animal health policy makers to optimise national and European legislation related to the prevention and control of infectious diseases, reduction of antibiotic resistance, food safety and zoonoses, animal welfare and environment (e.g. Animal Health Regulation (EU) 2016/429), new EU One Health Action Plan against Antimicrobial Resistance 2017; zoonoses Directive 2003/99/EC; animal welfare Council Directive 98/58/EC). As an example, the Scientific Committee of the Belgian Federal Agency for the Safety of the Food Chain has sent a newsletter to all Belgian veterinarians (November 2018) with an urgent advice regarding African Swine Fever. The article explicitly mentions the PROHEALTH results of WP1 regarding biosecurity on pig farms.


Academic Beneficiaries
WP2 of the PROHEALTH project has generated new scientific insights into gestational influences on piglet viability, the genetic basis of piglet intrauterine growth retardation, the implications of the human-animal relationship on maternal behaviour, the mechanism of mammary immunoglobulin transfer into colostrum and the vertical transmission of pathogenic microbes in poultry. This knowledge has been disseminated to the academic community through more than 20 presentations at international scientific conferences to different disciplinary audiences including meetings of the European Federation of Animal Science, The International Pig Veterinary Society, the International Society of Applied Ethology, the Société Française pour l’Etude du Comportement Animal, the European Veterinary Immunology Society, the International Society for NeuroImmunoModulation, the International Conference on Production Diseases in Farm Animals, the World Veterinary Poultry Association, and the European Poultry Conference. Seven papers have been published in top quartile, peer reviewed journals, and it is anticipated that a further seven which are in preparation will be submitted over the coming year.

Policy Makers
The work carried out in WP2 has addressed a number of subjects of current political debate. The demonstration that an enriched gestation environment for sows promotes improved piglet survival will lend strength to policy initiatives to improve animal welfare in the group housing systems, which are still relatively new in many member states. The demonstration of a genetic basis for the prevalence of intra-uterine growth retarded piglets offers a new approach to balance increased prolificacy with reduction in piglet mortality. The increased levels of piglet mortality observed as a consequence of genetic selection for prolificacy have become a growing cause of concern in many countries and this potential solution has been disseminated in a PROHEALTH Policy Briefing Note. The demonstration of the important role of vertical transmission of microbes in early mortality of chicks will lend weight to the policy objective of improving biosecurity measures in order to reduce the need for antimicrobial use. Overall, the elucidation of methods to reduce neonatal mortality in pigs and poultry will contribute towards the policy objectives of sustainable intensification of livestock production. Animal welfare can be improved, farm livelihoods enhanced, antibiotic use lowered to reduce the risk of proliferation of antibiotic resistant bacteria, and environmental impact reduced by improved efficiency of resource use when losses through mortality are reduced.

End Users
The demonstrated approaches for reduction in neonatal mortality will increase the profitability of livestock producers. PROHEALTH WP7 calculations suggest that genetic selection for robustness which reduces piglet pre- and post weaning mortality can increase profit by +€0.5 to +€3.3 per piglet. Similarly, work on positive handling of sows piloted in WP2 was subsequently shown in a WP6 study to reduce piglet mortality by 2.2% and improve post-weaning weight gain has been estimated to increase profit by +€0.4 to +€1.3 per piglet. Reduction in neonatal mortality will also increase job satisfaction of those working with livestock, both from keeping healthier animals and from the better public perception of their industry. It is noteworthy that the positive handling protocols for sows generated a favourable response amongst stockpersons on the farms where they were trialled. Results on the genetics of piglet survival have been carried out in collaboration with a major breeding company and disseminated to other breeders. Similarly, work on nutritional modification to enrich gestating sows have been carried out in collaboration with a major feed manufacturer and disseminated to other companies. PROHEALTH work has been highlighted in more than ~40 presentations to industry over the course of the project, including presentations at major industry conferences and exhibitions, and to local farmer groups. National and international scale events have included AnaPorc (Spain), Recherches Porcine en France, SPACE (France) and VIV Europe (Netherlands). Presentation on both pigs and poultry have been made at PROHEALTH industry workshops in Poland, Cyprus and Ghent. Furthermore, articles have been written for industry magazines including Pig Progress and Poultry Health Today.

Society at large
Members of society, whether consumers of animal products or not, have concerns about livestock welfare, environmental impact and food safety arising from antibiotic use. As described for Policy Stakeholders, the work in WP2 has addressed all of these concerns and WP7 work shows that the interventions which were initiated are viewed positively by the general public. Some members of society, in particular the NGOs, act as proxies for the interests of the animals themselves as important stakeholders. Clearly, the measures developed in WP2 to reduce mortality and improve health and welfare of both pigs and poultry are of great significance to them.


Academic Beneficiaries
WP3 contributes to a better understanding of the physiological basis of production diseases, with a special focus on the between genetic selection for high productive traits and the animal environment and management, such as the hygiene at housing, the diet and the degree of environmental enrichment.
We conducted a systematic review of the scientific literature and experiments in controlled conditions to quantify the impact of the major production diseases on several animal-based traits. An expected impact is the improvement of the content of the ontology AHOL, a shared tool of knowledge on the health of farm species ( to promote the use of a shared language understanding by computer and any users. We conducted a set of experiments in controlled conditions to assess, at the animal level, the interconnection between inflammation, stress response and metabolism and nutritional status in the expression and susceptibility to production diseases. This has improved understanding of the mechanisms involved in the pathogenesis of production diseases, including metabolism. This knowledge has been disseminated to the scientific community as 8 peer-reviewed articles (Animal, Veterinary Parasitology, Poultry Science, Journal of Animal Science, Veterinary Microbiology, British Poultry Science) and several presentations during internationally recognised conferences on animal production and health such as EAAP, ICPD, ESPN, WPC.

End users
Animal based traits identified by WP3 as relevant could serve as potential indicators for prognosis and identification of animals affected by production diseases. This will benefit farmers, livestock producers, allied industries and people involved in assessment of the efficacy of preventive or corrective practices. We confirmed that different pig and broiler lines differ in their susceptibility to be affected by production diseases suggesting that breeding programs should consider the genetic x environment interaction to select more robust animals in a wide variety of environments.
Our results revealed that management and housing conditions, feeding and nutrition, and biosecurity are undoubtedly the major factors influencing the expression and severity of production diseases. Generated knowledge is expected to help the design of multifactorial control strategies. For nutritionists and feeding companies, our results showed that dietary recommendations need to be revised to consider the positive impact of some nutrients on health, as suggested for vitamin D to support leg health in both pigs and broilers. Feeding programs need to be adapted at specific periods like the weaning in order to contribute to the reduction of antibiotic usage and antibiotic resistance. With WP7, we produced best practice guidelines to guide farmers to pay attention to the hygiene of the housing conditions to ensure good performance and health. Indeed, poor hygiene generates footpad lesions in broilers and the entry of pathogens is responsible for septicaemia. In pigs, poor hygiene is a risk factor for developing respiratory disease and depressed growth rate by up to 20%. The expected impact of having a good hygiene is a significant improvement of production, a reduction of morbidity, mortality and medication cost and thus increased incomes for producers up to €15-25 per pig.
WP3 work has been communicated to industry over the course of the project, including presentations at major industry conferences and exhibitions. National and international scale events have included AnaPorc (Spain), Journées de la Recherche Porcine en France, SPACE (France) and Scientific agricultural seminars (Finland). Presentation on both pigs and poultry have been made at PROHEALTH industry workshops in Poland, Cyprus and Ghent.

Society at large
The society has major concerns concerning antibiotic use by the livestock sector because of the risk of development of antibiotic resistance. In the pig sector, most of the antibiotics have been used during the postweaning period for controlling digestive disorders. WP3 and WP6 showed that an adapted feeding program would help to reduce antibiotic use during this critical period. In France, recent figures (Anses, IFIP) show that antibiotic sale and use during for weaned pigs has been reduced by 40 and 70%, respectively and PROHEALTH results have contributed to these encouraging figures. Overall, WP3 highlighted that good practices and good hygiene are efficient strategies to prevent production diseases and the need of medication like antibiotics. Such preventive strategies will be positively viewed by the society and contribute to restore the confidence of citizens and consumers.


WP4 focussed on 1) the development of a protocol of collecting and summarising environmental data from pig and poultry farms (2013-2014); 2) the determination of the role of environmental and microbial factors on farm on the temporal expression of multifactorial production disease in pigs and poultry (2014-2018); 3) the development of an IT system for automation, capture and processing of environmental and health data (2016-2017). Impact has been achieved as follows.

Academic beneficiaries:
WP4 quantified how the environment can result in disease occurrence in pig systems. It identified a lag time between the changes that occur in the environment and the occurrence (or the risk) of a production disease, meaning that there may be a window of opportunity during which animal keepers can act to mitigate the negative effect of diseases, such as porcine reproductive and respiratory syndrome (PRRS), a common and destructive disease endemic in pigs.
The main innovation of the WP was to develop the algorithms for predicting the incidence of disease according to the environmental conditions of the housing, by correlating the data collected through sensors with the animal health information. The WP demonstrated the use of machine learning algorithms for this purpose. At the time of conducing the project this was not a usual occurrence for livestock data, although the application of machine learning to develop algorithms for the prediction of risk for human conditions has been widespread. The results of these studies have been disseminated through the scientific articles (e.g. Cowton et al, 2018).
The importance of the massive use of data in the pig sector has been also summarises in a forthcoming paper: “Big (pig) data and the internet of the swine things. A new paradigm in the industry. Piñeiro C, Morales J, Rodríguez M, Aparicio M, García-Manzanilla E, Koketsu Y. 2019, Big (pig) data and the internet of the swine things. A new paradigm in the industry. Animal Frontiers” (Accepted), that build upon the PROHEALTH experience.

Policy Makers
In view of the results obtained in WP4, it is possible to conclude that the digitisation process that includes software, devices, systems, standard operating procedures, analytics and communications is already possible in the poultry and swine sectors, and enables the collection and use of large quantities of data. Further steps in this digitisation process will improve production efficiency and welfare on farms, or, as PROHEALTH has demonstrated, will improve animal health, under the quality standards that modern production requires. Over the next few years, this new digitisation process will generate a whole new set of knowledge in most of the relevant areas in poultry and swine production including nutrition, health management, reproduction, biosecurity, behaviour, welfare and even pollutant emissions. This will bring unprecedented changes and advantages to the industry and huge opportunities for the professionals in the sector worldwide. For this reason, policy makers should promote the development and on-farm use of digital tools and sensors to monitor the barn environment. These data should be integrated and used for launching pre-defined early interventions to mitigate diseases or even predict them.
A working group of the Spanish Government ( focused on the digitization of the livestock sector, after appreciating the promising results obtained through PROHEALTH. The working group concluded that for the effective development of digitisation in this field, actions aimed at favouring the development of pilot projects and demonstration actions should take place. This will allow the adaptation of existing technologies and services and initiatives for the benefit of the pig sector, avoiding the duplication of efforts in the development of technologies and methodologies already tested in other environments. The development of networks and the strengthening of collaboration between research centres, universities, technology centres and agricultural advisory bodies was recommended in order to effectively meet the demands of the sector and allow for the co-production of technologies and services.

End Users
The application of the digitisation process to the pig and poultry sectors offers enormous opportunities for improvement in these sectors. PROHEALTH was not a Precision Livestock Farming project. Instead it focused on a single aspect of application of this technology: the automated collection of environmental data from pig and poultry houses. The environmental and health data collection system developed through the PROHEALTH project, together with the promising results obtained, offer a great opportunity to get an information management system highly demanded by the pig and poultry sectors. The system should allow to send the collected information to a database, where the prediction algorithms developed through PROHEALTH (also others developed in the future, as the volume of data increases) will allow environmental alerts to be sent to farmers through a commercial app for smartphones.
The development of this commercial product would have a great impact to the sector, especially due to early intervention and even through targeted selective treatments to the houses of pens that need the intervention.

Society at large
WP4 findings will contribute towards the development of early warning systems that could raise the probability of a disease risk. Such early warning systems have the potential reduce antibiotic use in pig and poultry systems and improve animal welfare. Currently, it is unknown how citizens and consumers would react to the high degree of digitisation proposed by the livestock sector, especially in the dairy, pig and poultry sectors. However, there are H2020 collaborative projects that address exactly this question.


WP5 focussed on the development of potential molecular biomarkers of production disease and a deeper understanding of disease on different pig and broiler chicken genetic lines. WP5 Impact has been achieved as follows:

Academic beneficiaries
The use of molecular approaches for the diagnosis of disease, understanding pathogenesis and the complex interaction between bacterial microbiotas in human medicine is greatly advanced compared to information accrued in veterinary species. We have published and presented information which promotes the general use of, and highlights the future importance of, molecular biological platforms in production disease. (Giles et al., 2017. Res Vet Sci. 14: Pages 266-272; seminars at IPVS congress, Dublin 2016; WAFL, Wageningen, 2017). We have published data which shows differential gene responses in fast growing (Ross 308) and slow growing (Ranger Classic) broilers with coccidiosis (Giles et al., 2019. Vet Parasitol, In press; Giles et al., 2019. Parasite Immunol, submitted for publication). In some EU countries there is an increasing trend to use slower growing broilers, largely due to animal welfare concerns, but little has been published on how these respond to production disease compared to faster growing broiler lines. This information provides important new insight, which may influence future studies of pathogen biology, vaccinology and health management in broilers. In WP5 we have used Deep Sequencing to study microRNA expression in clinical and sub-clinical coccidiosis in broilers (Giles et al., 2019. BMC GENOMICS. Submitted for publication), and Mycoplasma infection in pigs. The expression and biology of miRNAs are of significant interest in molecular medicine and our studies may open new avenues of study in veterinary diagnostics and therapeutics.
Understanding the bacterial species which populate the intestinal microbiota, and how these species interact, is of interest to microbiologists, pharmacologists and immunologists and has potentially significant practical applications in diagnosis and treatment of disease. Studies in WP5 have reported data which shows the different species which make up the microbiota in layer hens and how this changes following Salmonella infection, thus ascertaining the beneficial and detrimental microbiota (Varmuzova et al., 2015. Poult Sci. 94:2049-58; seminar VIV Europe, Utrecht 2018). Transfer of microbiota from healthy adults to susceptible day old chickens was also shown to have a protective effect against Salmonella challenge (Varmuzova et al., 2016. Front Microbiol. 17;7:957). In pigs, work from WP5 has also shown different microbiotas in sows maintained on slatted floors compared with enriched housing with deep straw bedding and that the microbiota of piglets born from these sows is different to the sow within the first 1-4 days. This is thus suggesting that the environment may be used to promote different microbiotas in new born piglets (Kubasova et al., 2017. PLoS One. 12:e0170051).

End users
Coccidiosis remains the most important disease in the broiler industry and is a predisposing factor for necrotic enteritis (NE). These diseases usually require post-mortem diagnosis and may increase in future with reduced prophylactic and metaphylactic antibiotic treatment. Early detection of coccidiosis and NE in broiler flocks will facilitate targeted antibiotic treatment and reduce colonisation by other pathogens such as Salmonella. Overall, the effect of this will be to reduce antibiotic usage and AMR and reduce carriage of zoonotic pathogens into the human food chain. This has obvious industrial application and may inform better flock management. Although publication of this work has been withheld for potential IP, development of a biomarker which differentiates coccidiosis and NE in broilers has been presented at conferences, omitting the actual gene names. (PVSGEU, Cyprus 2017, VIV Europe, Utrecht 2018, EAAP, Dubrovnik, 2018).
Studies in WP5, which have shown differential gene response and immunity in fast and slow growing broiler lines with coccidiosis, (Giles et al., 2019. Vet Parasitol, In press; Giles et al., 2019. Parasite Immunol, submitted for publication) may influence broiler breeding in future as lines may be selected which have greater resistance to disease.
A group of genes were identified which were differentially expressed in both intestinal and lung tissues during intestinal disease. Some of these genes were used in a biomarker panel to diagnose coccidiosis and NE. This probably highlights a generalised mucosal response and suggests that some of these genes could be used to determine respiratory diseases in chickens, as well as coccidiosis and NE.
Research in WP5 has determined the dominant species in healthy and non-healthy gut microbiota of chickens and pigs (Varmuzova et al., 2015. Poult Sci. 94:2049-58; Varmuzova et al., 2016. Front Microbiol. 17;7:957; Kubasova et al., 2017. PLoS One. 12:e0170051). This has significant industrial applications since susceptible animals could be orally dosed with protective microbiota.

Socio-economic impact
Data generated in WP5 could lead to a more co-ordinated approach to husbandry, vaccination and antibiotic usage and will, therefore, impact on the EU code of practice (91/630/EEC) on protection from disease. Animal welfare in high intensity systems and antimicrobial resistance are major consumer concerns when purchasing meat. New initiatives, such as those reported in WP5, which promote animal health and reduce the need for antimicrobial interventions could have a very positive impact on consumer purchase. The effect of this would be increased sales which benefits the agricultural economy.
The EU uses high animal welfare standards as a point of differentiation from other exporting blocks and since the adoption of the ‘country of origin’ legislation in 2015 (EU directive 1169/2011) initiatives which reduce disease and decrease the need for antibiotic usage will increase the competitiveness of EU export.


The whole WP6 was dedicated to demonstration activities that aimed at applying the ‘best bet’ strategies to control production diseases, developed under WP1-5, on large scale facilities. As such, these demonstration activities aimed to bridge the usual gap between small scale, controlled trials under experimental conditions and large scale, industry based trials where significantly less control can be exercised. Therefore, the outcomes of this WP are directly relevant to End users and perhaps indirectly relevant to policy makers.
WP6 activities collected bio-economic data which were over and above data that could be collected under small, scale experimental activities. These data were passed on to the WP7 team in order to consider the economic consequences of the developed management strategies. In addition, the control strategies were considered in the consumer-based questionnaire about the preferred strategies to control production diseases. The impacts of both these exercises are detailed under WP7 Impacts below.
One of the major impacts that may arise from this WP is whether the developed controlled strategies will be adopted by the relevant Industries as methods of control of relevant production diseases. As the WP6 activities were only concluded relatively recently, the quantification of this impact should be performed over a 3 or 5–year horizon.


Academic Beneficiaries
WP7 provided novel information on how consumers and stakeholders view intensive production systems and the control of production diseases in these. This provides guidance for further research, as literature and citizen survey data can be exploited in further research. The bioeconomic models developed have a generic value and can be used in further analyses to illustrate the financial viability of animal health interventions in different conditions and countries. Materials generated during the value chain analysis provide academics with guidance on how economic benefits of an animal health intervention can be analysed systematically. The WP7 team will pursue further publication of results regarding the benefits of hygiene, piglet viability and consumer purchase intentions in peer-reviewed articles.

Policy makers
The socio-economic work of WP7 helped to develop better links between industry, policy and public perceptions. It provided foreground that can be exploited to ensure that appropriate minimum standards are in place and to design policies which increase consumer trust in animal production systems, leading to better alignment between intensive farming practices and the values, needs and expectations of society. The results will help stakeholders to address the mistrust and misconceptions the public have about production systems and interventions associated with production diseases, and urge them to increase the transparency or production practices. Besides the consumers, this will address some of the concerns of non-consumers (e.g. vegans). Improved hygiene and treatment protocols can reduce worries related to food safety and antimicrobial use.
WP7 identified both measures which were financially attractive to farmers as such and measures requiring additional incentives, because they increased farm costs, while providing public goods and increasing system robustness. Policy makers can target policies towards these effective health management measures, should they require additional incentives or extra effort such as certification schemes committed to cover additional costs to farmers. Promoting identified measures which systematically prevent production diseases can contribute to public trust towards intensive animal production and help to communicate fact-based information about production systems.
The partners will continue to disseminate WP7 findings to EU-level policy makers, such as DG Agri, and key stakeholder organisations such as Copa-Cogega and FVE, to help them to improve the sustainability of pig and poultry farming in Europe. Besides already completed publications, a follow-up article on stakeholder and public expectations towards intensive animal farming and value chain implications of animal health management will be submitted to EuroChoices, a journal reaching a wide range of policy makers and experts in Europe. WP7 results are scheduled to be disseminated to experts at events such as 12th BI forum on farm animal wellbeing in Czech Republic.

End Users
Besides already completed activities, the outputs of PROHEALTH will be further communicated to the industry end users at the EU and national level to ensure the project’s legacy. WP7 will inform activities in the Nuffield Council of bioethics working group on genomics and livestock in the UK and a variety of national farm animal welfare committees about the results and recommendations; for example a dialogue on how to achieve this has already started in Finland. WP7 will collaborate with farming organisations and industry-level animal health platforms to disseminate the results to the end users, even after project completion. For example, partnering with an information sharing project in Finland provides an access to a series of stakeholder workshops organised to deliver knowledge to pig and poultry farmers, industry experts and authorities and to motivate the adoption of biosecurity solutions with economics arguments arising from PROHEALTH.
WP7 results provide guidance on how to improve communication with the public and how the industry can respond to consumer expectations in an economically sustainable way, and encourage producers to tell the consumers why certain interventions and practices are applied. Food chain stakeholders will benefit from the improved dialogue and communication producers and the public and a better understanding of the perspectives of other stakeholders. Promoting better supply chain coordination and cooperation helps to ensure that all stakeholders are working together to support each other and convey consistent and accurate messages to the public. Supply-side stakeholders are encouraged to work with independent and trusted bodies such as animal welfare organisations and quality assurance schemes to raise standards. These recommendations will help to build greater confidence in the sustainability of EU food production systems.
WP7 results indicate that both the public and experts dislike medication-based interventions. Expert advisers can motivate farmers to improve biosecurity and to adopt sustainable, animal-friendly practices with economics arguments to adopt production practices. Principals, such as slaughterhouses, can utilise our results to develop procurement contracts and company best practice guidelines which incentivise systematic adoption of biosecurity, animal handling practices, and environment monitoring on farms. This is an opportunity to adopt standardised schemes and to increase societal acceptance of intensive animal production.
PROHEALTH results provide advice how livestock farms, slaughterhouses and food sector companies and other businesses can develop new business opportunities (see Table X). For instance, improved hygiene interventions, positive handling of and enrichment provided to sows, management and genetics to reduce piglet mortality, improved housing to reduce locomotory disorders can improve production efficiency and reduce production costs. Changing emphasis from volume to value, and increasing the market penetration of concept-based production where studied measures are applied, can build public trust and add value to intensive production systems. Measures which address animal psychological health, such as positive handling of pigs, can be particularly important to consumers. The reduction in disease will result in more uniform product of better quality, produced in systems that have less pharmaceutical input, which has a value as indicated by WP7 results.

Society at large
WP7 will continue to disseminate socio-economic results to the public in both traditional and social media to promote the dialogue between the industry, science and the public. PROHEALTH identified several management modifications which can reduce production diseases and the need to use antimicrobials on animals, and improve animal welfare in an economically sustainable way. Implementing measures such as those improving piglet viability can increase profits of farms, reduce environmental footprint of farming and address societal concerns related to pig farming, but as the results point out, the consumer ultimately benefits from effective control of production diseases through lower food prices and higher quality of food. Having access to more affordable, safer food has further benefits in terms of social equality. The measures can equally affect both genders.
All consumers stand to benefit from better quality products and improved regulations to maintain the high European standards in livestock production. Adopting measures contributes to public health and can have large benefits to the national economy over long time as antimicrobial resistance is a major societal challenge. Using the results to strengthen the quality assurance schemes and communication can reduce the stress people experience when thinking about animal production systems. Consumers and citizens will have greater assurance of livestock industry improving the sustainability of production systems, and also the animals will benefit through improved health and welfare (with positive impact on human health and welfare (cf. One Health, One Welfare).
List of Websites: