Modelling the burden of Mycoplasma hyopneumoniae infection for enhanced decision support on fattening farms
Abstract
This study presents a stochastic, individual-based model developed with EMULSION software to simulate Mycoplasma hyopneumoniae infection dynamics in growing pigs on a Dutch fattening farm. The model evaluates three key burdens: clinical signs and lung lesions, antimicrobial usage, and financial performance. It incorporates multiple transmission levels (within-pen, between-pen, indirect) and calculates economic impacts such as feed and treatment costs, penalties for slaughter weight deviations, and carcass revenues. Parameters were sourced from literature, surveys, and expert elicitation, with sensitivity analysis applied. Model outputs illustrate infection progression, antibiotic treatment patterns, and economic losses under different scenarios. This tool supports decision-making for preventive and control strategies and can be extended to assess co-infections with other respiratory pathogens. This work was presented as a poster at the European Symposium of Porcine Health Management (ESPHM) 2025, held in Bern, Switzerland (Reference: BBD-PP-46).
Full text
Modelling the burden of Mycoplasma hyopneumoniae infection for enhanced decision support on fattening farms Boeters, M1, Garcia-Morante, B2,3, Picault, S4, van Schaik, G1, Sibila M2,3, Segalés, J3,5, Steeneveld, W1 1 Department of Population Health Sciences, section Farm Animal Health, Faculty of Veterinary Medicine, Utrecht University, Utrecht, the Netherlands. 2IRTA, Animal Health, Centre de Recerca en Sanitat Animal (CReSA), Campus de la Universitat Autònoma de Barcelona (UAB), 08193 Bellaterra, Catalonia, Spain. 3 Unitat mixta d’investigació IRTA-UAB en Sanitat Animal, Centre de Recerca en Sanitat Animal (CReSA), Campus de la Universitat Autònoma de Barcelona (UAB), 08193 Bellaterra, Catalonia, Spain. 4 Oniris, INRAE, BIOEPAR, 44300, Nantes, France. 5 Departament de Sanitat i Anatomia Animals, Facultat de Veterinària, Universitat Autònoma de Barcelona (UAB), Campus de la UAB, Bellaterra, Spain. INTRODUCTION RESULTS Epidemiological modelling can aid in explaining, predicting, and guiding the prevention and control of infectious diseases in animals, while also addressing associated risks to animal health, welfare, productivity, and economic sustainability. Despite its importance, modelling the transmission of Mycoplasma hyopneumoniae in industrial pig production remains limited. OBJECTIVE This study modelled M. hyopneumoniae infection in growing pigs on a Dutch fattening farm, focusing on multiple burdens by quantifying the impact of infection on pig health indicators, antibiotic use, production performance, and economic results. MATERIALS & METHODS A stochastic, individual-based model was developed using EMULSION software (https://sourcesup.renater.fr/www/emulsion-public/) to simulate infection dynamics, production performance, coughing and pneumonia development (Fig. 1.). Antibiotic treatment initiation based on the detection of coughing was also included in the model. Figure 2. Visualization of the organizational levels included in the model: pen, pen group (two neighboring pens sharing a feeding trough) and compartment (one room in which all pigs share the same air-space) As an example of model outputs, we showcase a M. hyopneumoniae outbreak simulation based on 50 repetitions and an initial prevalence at the start of fattening of 11% (Fig. 3.). •Across the entire fattening period, all pigs (N = 144) became infected. •24% of pigs remained subclinical (range: 19 –29%). •76% of pigs developed pneumonia (range: 71 –81%). •1.4% of pigs had unresolved lesions at slaughter (range: 0 –5%). •Antibiotic treatment was administered in all simulations, with a median initiation on day 16 post-entry (range: day 15 –20) and a median total dosage of 445 g (range: 336 –480 g). •Compared to a non-infected cycle, the outbreak led to a total loss of 8.70 € per fattening pig (range: 8 –10€): •Total slaughter weight: -230 Kg (range: 180 –290 Kg). •Revenues: -440€ (range: 350 – 560€). •Feed costs: +725€ (range: 655 – 796€). •Treatment costs: +88€ (range: 86 – 91€). For any comments or inquiries, please contact [email protected] DISCUSSION & CONCLUSIONS •This innovative multi-level modelling approach (EMULSION) simultaneously incorporates multiple burdens within a single simulation. •The current model offers valuable insights into the spread of infection within a compartment. •This model serves as a foundation for simulating individualand pen-level interventions and evaluating their effectiveness across multiple burdens, providing critical advancements for developing more effective management strategies for M. hyopneumoniae infections. This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No. 101000494. Figure 1. Overview of the modeled state machines: infection dynamics, development of lung lesions, development of clinical sigs, and production performance. Yellow arrows indicate interdependencies, meaning an individual can only make the transition within the state machine if it is also currently in the state from which the yellow arrow originates Figure 3. Infection progression from 7 to 28 days post-entry into the fattening compartment The model simulates an all-in/all-out Dutch fattening farm, where 144 pigs (12 pigs per pen) are raised from 10 weeks until slaughter. It integrates multiple organizational levels, enabling the differentiation of transmission dynamics within-pen, between-pen, and indirect transmission pathways (Fig. 2.). To estimate the financial impact of disease, an outbreak cycle was compared to a non-infected cycle, and average costs and prices in the Netherlands for 2024 were used. Actuació del Pla estratègic de la PAC 2023-2027 cofinnçada per: