The European Union’s (EU) food safety policy aims to ensure high standards of consumers’ health protection (1, 2). Despite increased efforts to reduce pathogens in the food production chain, foodborne outbreaks and illnesses continue to occur, and food safety has been increasingly recognized as one of the main aspects of public health (3–5). Pigs represent the largest livestock category reared across the EU and pork accounts for nearly half of the total meat produced in the Common Market (6). Pigs are an important reservoir of certain Salmonella enterica spp. serovars (SAL) which are known causes of foodborne diseases in humans (7). While Commission Regulation (EC) 2073/2015 defines the relevant food safety procedures, including process-hygiene criteria, and general instructions for surveillance of SAL are given through Council Regulation (EC) 2,160/2003, specific regulations demanding mandatory programs for SAL in pigs do not exist across the EU (8). Nevertheless, illnesses in humans caused by SAL in Europe are associated with substantial economic losses resulting in up to €90 million annually (9). Additionally, SAL has been one of the most commonly diagnosed foodborne pathogens for decades (7). While the total number of human salmonellosis cases exceeded 60,000 in 2020, only infections with specific serovars, such as S. typhimurium or S. Derby, have been attributed to the consumption of pork products (7). However, the pig reservoir has been reported as the second most important source of human salmonellosis after laying hens in the EU (10, 11). Salmonellosis in humans, which is usually characterized by gastroenteritis is mostly foodborne (12), but people having direct contact with pigs are also at increased risk of contracting the disease (13).

SAL contamination may occur at different points of the production cycle, either at the primary production level or in further processing, including slaughtering (10). Biosecurity standards and hygiene management in pig farms play a very important role in preventing the introduction of many pathogens as well as reducing their spread within the farm once the agent has been introduced (14, 15). Broadly, biosecurity relates to the implementation of measures that can reduce the transmission, introduction, establishment or survival of a pathogen (16). External and internal BSMs target either aspects of farm management, such as replacements, breeding strategies, and wildlife near farms, or focus on herd practices, carcass disposal, cleaning protocols and personnel hygiene (15, 17). When deciding on employment of specific BSMs, farm characteristics, such as location, facilities, production type and herd size need to be taken into account in addition to country-specific production and management strategies (14, 17, 18). Nevertheless, biosecurity plans should be designed for individual farms considering that the BSM’s effectiveness is usually recommended in general terms (14, 19), and evaluated in field studies based on identifying correlations with decreased (sero-) prevalence. While agreement exists for the positive impact of some BSMs, such as a low number of potential sources for SAL, e.g., for the purchase of new livestock (14), or quarantine protocols for breeding sows (17), the overall effectiveness of many BSMs has not yet been described in the scientific literature. In addition, there is a lack of standardized on-farm BSM implementation protocols (17). Considering the shift in pig production from smaller to large holdings, reliable disease prevention has become increasingly more relevant (15). Nevertheless, the employment of sustainable biosecurity strategies and their continuous improvement remain challenging tasks for many pig farms (20).

The effectiveness of BSMs to mitigate specific pathogens is poorly understood (21). Epidemiological models are recognized as valuable tools that can assist decision-makers in identifying and evaluating strategies for disease control (13, 22, 23). Only a few simulation models have been used to investigate the effect of specific BSMs on the occurrence of infectious diseases in the pork production chain (PPC) (24–27). To assist EU Member States, a quantitative microbiological risk assessment (QMRA) model for SAL was developed and established (13). Dependent on country-specific production systems and infection prevalence in herds, it allows assessment of the effectiveness of on-farm and slaughterhouse interventions in reducing SAL in pigs and humans. In addition, it explores the epidemiological processes in the PPC and offers valuable insight into a pathogen’s transmission, including exposure to humans (13, 25, 28).

The implementation of BSMs at various specific points of the PPC is associated with considerable costs due to, e.g., required manpower, equipment, material, installation, and maintenance. Identifying and quantifying these costs are important to justify and prioritize investments and to inform stakeholders (29). The benefits associated with biosecurity efforts are reflected in their potential to reduce the pathogen’s prevalence and thereby prevent losses caused by animal and human diseases. The various aspects of the associated human diseases are often captured by the Cost of Illness (COI) methodology. It encompasses monetary factors, including direct, e.g., diagnostic or treatment, and indirect expenditures that result from, e.g., premature death or disability to work (30, 31). Different economic methods are available to determine the cost-effectiveness of BSM implementation. The cost-benefit analysis (CBA), which has been widely used in economic assessments of animal disease control programs and policies, represents a validated approach providing information on economic efficiency (32–34). It weighs the total benefit expressed in monetary terms against the total cost spent on BSM implementation and evaluates, through the benefit-cost ratio (BCR), the cost-effectiveness of a respective BSM (24, 35–37).

This study was conducted within the One Health EJP BIOPIGEE Group, which focuses on the improvement of biosecurity practices in pig farming across Europe. The BSMs evaluated in this study were identified through a literature review and meta-analysis carried out in the recently finalized corresponding sub-project (38). The objective of this study was to conduct a comprehensive economic assessment of the costs and benefits associated with those BSMs proven as useful in reducing SAL prevalence and to evaluate their cost-effectiveness along the PPC. This has been done for two European countries with different levels of SAL prevalence and livestock characteristics, i.e., Austria (AT) and the United Kingdom (UK).

The study presented here aimed to assess the cost-effectiveness of specific BSMs that are shown to be effective in reducing SAL prevalence along the PPC. The results of the analysis indicated the cost-effectiveness of four of these BSMs in AT and five in the UK.

The economic analysis conducted in this study is built on the epidemiological outputs of the QMRA model and reflects the monetary impact resulting from the implementation of the BSMs. Their effectiveness inputted to the model is based on the results of the above-mentioned meta-analysis focusing on biosecurity in pig farms (38). The ORs delivered from this analysis were converted to RRs to meet the technical requirements of the QMRA model (28), although this conversion resulted in a reduction in the total number of BSMs available for the economic evaluation. While ORs and RRs are common risk identifiers in epidemiological or clinical studies (53, 54), some authors used other methods to define reduction values in the simulation models (36, 55, 56). Using the RRs of the BSMs introduced additional uncertainty into the study, which was reflected in the UA. The UA results showed that an inaccurate estimate of the BSM’s effectiveness could reverse its cost-effectiveness. The RRs only consider the direct effect of the interventions but exclude possible implementation shortcomings, which could not be included in the epidemiological evaluation. To reduce uncertainty, more comprehensive information on the effectiveness of the BSMs should be obtained. In addition, the effectiveness of some BSMs has been evaluated at the herd or farm level, i.e., not at the individual pig level. Since the individual farms in which the effectiveness of BSMs was evaluated differ in their structure, management and hygiene, despite national and international regulations the uncertainty associated with these differences cannot be eliminated at this point. However, the resulting BCRs are still informative, since they represent a best-case scenario, based on the understanding that the RR takes its maximum value when evaluated at the individual pig level (28). Hence, BSMs identified as not cost-effective under the current considerations would not be expected to have a BCR > 1, even if RRs at the individual pig level were available (28).

The design and input data requirements of the QMRA model have a major impact on the structure of the presented economic analysis and the certainty of its output. The architecture of the model allows the evaluation of only one specific pathogen at a time (13, 28). However, it is expected that BSMs would benefit a wide range of pathogens (including endemic diseases of pigs), making positive reductions in both pig and human cases of disease. Models encompassing the transmission risk of multiple pathogens would enable the evaluation of an expected higher cost-effectiveness of BSMs that are effective for more than one pathogen. This could promote a holistic approach to such analyses, as for example public health data already indicates a correlation between SAL and other immunomodulatory viruses, with relevance to general human and animal health (57). Further, in our study, only a 100% implementation rate of BSMs was considered in the QMRA model, which has previously been indicated as a constraint in the cost assessment of BSMs (24). Continuous application of BSMs to 100% of farms or animals has a direct impact on the total cost, which could be reduced by their implementation in targeted herds only, e.g., those contributing most to the overall prevalence. However, QMRA does not have the capability to simulate this, because it treats all herds as having the same prevalence. Reducing the countrywide implementation rate in QMRA would lead to a linear scaling of the cost and, similarly, benefits. An option to consider only large farms would allow comparison of complex implementation strategies and support cost-sensitive decisions more effectively. Furthermore, the implementation rate derived from the above-mentioned questionnaire was not country-specific but estimated for both evaluated countries together. Such an approach ignores the regional characteristics of farms and the impact of national legal frameworks for establishing biosecurity strategies. For future epidemiological-economic studies with focus on higher-risk farms, the necessary baseline data should be collected at the country level.

While it is debatable whether the use of organic acid to reduce specific gastrointestinal pathogens can be considered as a BSM (16), the application appears to be cost-effective at certain production stages (24, 58). According to our analysis both BSMs using organic acids, either in feed for weaners or added to water, achieved a BCR > 1 even if applied to 100% of the slaughter pigs. The evaluated costs per pig for adding organic acid into the weaner’s feed (€0.14/pig AT; €0.22/pig UK) are the lowest costs incurred per pig among the analyzed BSMs. While the country-specific length of the weaning period, as well as the required amount of water or feed for this production stage, were considered along with the price of the acid, additional costs due to possible corrosive damage of the feeding or watering pipes could not be included in the analysis. An underestimation of the total cost can therefore not be excluded. Moreover, possible negative effects such as the development of acid tolerance (59) were not considered in the CBA, since the risk for its development is not conclusively supported in the literature (60). Due to data gaps in the assessment of benefits based on performance indicators, potentially improved weight gain due to organic acid intake (61) was not assessed in our study.

According to our results, disinfection of farrowing pens was shown to be a cost-effective BSM when implemented in AT or the UK. However, carrying out disinfection at the entire farm level has previously been identified as not cost-effective (24, 25). In our study, we specifically calculated the costs of disinfecting farrowing pens after every use for each breeding sow (AT €9.8/sow per year; UK €8.27/sow per year). In this cost estimate, the number of all breeding sows in a country and their reproductive performance were taken into account, but the national replacement rates were not considered due to a lack of specific data. Therefore, an overestimation of the total cost is possible.

Anal plugging, the purpose of which is to effectively reduce fecal contamination during the slaughter process (62), was identified as another cost-effective BSM to reduce SAL along the PPC and is the only BSM analyzed to be implemented in the slaughterhouse. In general, BSMs seem to be more cost-effective when applied during the slaughter process (25, 26, 36, 55). Various associated costs, including labor as well as general and specific equipment, were taken into account to determine the total cost per slaughter pig (€0.38/pig AT; €0.35/pig UK). Since anal plugging was not studied under field conditions, data were lacking to consider a potential reduction in the processed carcasses per day and additional costs. Furthermore, the QMRA based estimation of the effect of anal plugging assumed that the plugs seal the anus completely in 100% of the pigs. Thus, the current estimate might overestimate the benefit of this BSM. Future evaluations should include these considerations to reduce the uncertainty of the estimates. If vaccination could be applied in a more targeted manner, such as at nucleus and multiplier farms at the top of a PPC, then this may improve the cost-effectiveness.

Although the effectiveness of the other BSMs evaluated to reduce SAL was demonstrated in the meta-analysis (38), their respective BCRs were <1 for both countries. Even though the effectiveness of vaccination of pigs to reduce the exposure to humans was proven, i.e., a case reduction of 63% (41), the calculated BCR suggests that implementation in 100% of animals is not cost-effective. This BSM resulted in the highest total cost for a one-year implementation when applied to all sows and all piglets within a 1 year cycle in the UK (€39.1 million).

For the important factors related to the benefits, total benefit is most influenced by the costs associated with human salmonellosis, i.e., AT (~97%) and UK (~57%). Although the results for some parameter variations within the COI analysis are not presented in the UA results section, the importance of these factors should still be discussed due to the direct relative change in the economic output. The URF (7.3) used for human SAL cases is not country-specific (48). European studies have used different multiplication factors ranging from 4.7 to 57.5 to estimate the total number of salmonellosis cases in the community (36, 52, 63–66). Several aspects may have an influence, such as the national health system, the accessibility and availability of resources, and the individual’s willingness to seek medical attention (51, 67). Disease pyramids including critical points where positive cases could be missed (51) should be evaluated for each country individually based on national information from various sources. The SAF used in this study for salmonellosis (47) has been calculated for specific geographic regions only, i.e., for Western Europe (34.1%) and Northern Europe (10.6%), which introduces uncertainty in our country-specific analysis. Other CBA studies used higher values for the UK (24) and lower values for AT (48). Different consumer habits and national product availability, such as for cured pork sausages, influence the national SAF as well. While individual studies on national SAF values are available for some countries (68–70), they were not identifiable for the countries evaluated within this study. Country-specific assessments rely on disease-outbreak data and comparison of SAL serovars in the potential source and human cases when a microbial subtyping is applied (47). Future studies should attempt to establish country-specific SAFs, taking local consumption patterns into account. Additional factors, notably antibiotic resistance, which is a rising concern in SAL serovars common in pigs (71), economic changes on the European pork market (72), and the consideration of multiple affected livestock species as sources (51), should be included in future animal health economics research. Moreover, increased consumer risk awareness and improved hygiene in households during meat processing could contribute to the reduction in the number of human infections.

Any major disease outbreak in meat-producing livestock affects the meat market and may have consequences for trade. The EU is a single market without borders and therefore special considerations should be included when assessing the impact of foodborne diseases on trade in Europe (72). Surveillance frameworks are essential for such a market system, however, there are no standardized and regulated surveillance systems for SAL along the PPC in place in both evaluated countries, and there is a lack of information on the associated expenditure. Therefore, the effectiveness of surveillance and the associated costs and benefits could not be considered in our study. Similarly, no data is available on the number and magnitude of pork-attributed outbreaks across Europe, nor domestic product recalls, which would be necessary to investigate possible corresponding market changes, e.g., in supply, demand and prices. Despite the existence of the Common market, there is a considerable variation in pork market characteristics between Member States. Moreover, pork prices are characterized by seasonality, randomly occurring fluctuations and various short- and long-term trends (73). Therefore, respective considerations of market effects resulting from SAL occurrence on pig farms are subject to high uncertainty.

One of the general challenges when implementing biosecurity systems is the undeniable discrepancy between the stakeholder carrying the costs versus those benefiting from the outcome (36). Farmers make relevant decisions based on potential value added and the ease and costs of the implementation (74). The possibility of considering multiple pathogens, mentioned above, could help move these discussions forward. While penalties for high SAL prevalence have been introduced in some European countries (75, 76), visible financial benefits within their production system might still be the most convincing factor for farmers. Other diseases in pigs, the incidence of which can be reduced by the implementation of BSMs, such as post-weaning multi-systemic wasting syndrome and porcine circovirus 2, can result in high losses for farmers (77), and therefore the potential additive reduction-effect of BSMs needs to be evaluated.

Nevertheless, the motivation to prevent Salmonellosis in humans through biosecurity should not be determined solely by the cost-effectiveness of the BSMs based on the COI approach, which is limited to measuring the losses in monetary terms. It is arguable, whether personal disease burdens such as suffering, pain, and loss in productivity can be captured in monetary terms, and therefore methods estimating the burdens in non-monetary ways should be considered too. Such methods are, e.g., Disability-Adjusted Life Year (DALY) or Quality-Adjusted Life Year (QALY), which however, do not include aspects such as the costs associated with the use of health care services (9). When deciding on the implementation of BSMs, stakeholders should therefore consider different approaches to achieve the desired outcomes.

Overall, the current spread of the African swine fever virus in Europe (78) highlights the need for increased biosecurity along the PPC. The concerns on the associated serious economic losses could be an incentive for national authorities and farmers to invest collaboratively in improving biosecurity. In return, targets to reduce SAL in the pig population could be achieved at the same time. Hereby, our results contribute to the discussion on increasing biosecurity along the PPC in order to reduce foodborne disease outbreaks and prevent the spread of infectious animal diseases in general.

The original contributions presented in the study are included in the article/Supplementary material, further inquiries can be directed to the corresponding author.

CB: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Validation, Writing – original draft, Writing – review & editing. AK: Conceptualization, Funding acquisition, Methodology, Project administration, Resources, Supervision, Validation, Writing – review & editing. NW: Data curation, Formal analysis, Investigation, Methodology, Software, Validation, Visualization, Writing – review & editing. GCC: Data curation, Formal analysis, Investigation, Methodology, Software, Validation, Writing – review & editing. TM: Conceptualization, Formal analysis, Methodology, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing.