Antimicrobial resistance (AMR) is a steadily increasing threat to public and livestock health around the world (1). Since the introduction in the 20th century of antimicrobial drugs (such as antibiotics) to treat infections, the rate of emergence and spread of AMR has increased substantially (2, 3). This has led to a situation where many common bacterial infections are, once again, becoming difficult to control. AMR is a multi-factorial problem that will require multi-sectorial action to address. Reduction of over-reliance on and inappropriate use of antimicrobials has been outlined by leading health researchers and agencies as the main strategy available to tackle the challenge of increasing AMR globally. This applies to all sectors; for instance, the WHO Global Action Plan stated a need ‘to optimise the use of antimicrobial medicines in human and animal health’ (4) and a ‘One Health’ approach, combining human, animal and environment aspects has been advocated (5) as ‘the main strategy required to reduce the development of antimicrobial resistance’.

The potential risk to human health from the use of antimicrobials in food animals has been highlighted by many researchers (6–16). More recent evidence is suggesting that while there is still a risk of spread between species, including humans, most AMR is and remains species specific (17, 18). Nevertheless, it is incumbent on all sectors to use antibiotics responsibly to preserve their efficacy for as long as possible. In addition to therapeutic use and metaphylactic (i.e., administration to clinically healthy animals that belong to a herd or flock with clinical signs) antimicrobial substances have been, or are being, used in livestock in some parts of the world not only for routine prophylaxis (treatment of animals at risk of infection in the absence of clinical signs) but also for enhancement of production (e.g., growth promoters); in other areas these latter two uses have been banned, or restricted. Responsible use is about as little as possible but as much as necessary (to protect animal welfare). While AMR is a global challenge, the degree of control and regulation of antimicrobial use varies widely despite some practices being major contributory factors to the development of AMR in livestock (19–21). The economics of production within a sector may contribute to the ability to adopt and to implement policies to reduce AMU.

Pork meat is one of the major sources of human food with over 40% of total meat consumed worldwide coming from pigs (22). As in other livestock sectors (1, 23, 24), the use of antibiotics (the main antimicrobial agent) has been a common practice on pig farms. The pig and poultry sectors are often considered to use more antimicrobials compared with other food animal production systems. An effort to estimate global consumption of antimicrobial in food animals suggests that the pig sector uses almost four times more antimicrobial per kilogramme meat produced (25). In the United Kingdom (UK), a quarter of all the antimicrobials sold in 2017 were used in farm animals (26). Around 36% of the total antimicrobials sold in 2022 for animal use were bought by pig farmers (27). The reported frequency of evidence for AMR from pigs and pork products varies. For instance, a study in China showed 91% of E. coli isolates from 1871 samples from pigs and their breeding environment contained E. coli resistant to major antimicrobials (22). In the UK, an abundance of AMR genes has been demonstrated in a single herd (28). At a population-level, it is reported (27) that there has been an encouraging increase in fully susceptible E. coli. There has been a concomitant increase in the percentage of pigs carrying ESBL/AmpC¹ - producing E. coli bacteria (27). This increase is not accompanied by an increase in clinical cases of resistance (Nevel pers. comm).

Antibiotic stewardship programmes, co-ordinating efforts to reduce antimicrobial use and encourage responsible and prudent use in the UK livestock sectors have been developed and implemented (29). This should help to preserve the efficacy of antibiotics for as long as possible. These activities have been especially effective in the pig sector. They are proving worthwhile with the UK pig sector having achieved, in 2022, its aim to reduce antimicrobial use (AMU) by 30% compared to the year 2020 levels (29, 30). This aim was initially set to be achieved by the year 2024. Indeed, since antimicrobial use in the pig sector started being recorded in 2015, the sector has reduced its use by 75% as of 2022 (30).

Pig producers will need to continue their efforts to use antibiotics responsibly thereby protect the efficacy of currently available products and protecting the reputation of the sector. Responsible antibiotic use is an important issue for trade. Actions will be needed at both industry-wide and individual farm level. There are several studies of alternative management options that can be used to reduce antimicrobial use on livestock farms (2, 31). To reduce antimicrobial usage without adversely affecting farm production and farm profitability is a challenge and it is necessary to make (holistic) changes, such as improvements in facilities, management, and health status. To replace antimicrobials, farmers need to minimize production losses by adopting strategies aimed at preventing disease occurrence on farms. These strategies could include adopting vaccination programmes, adding supplements to feed to boost immune systems and changing farm management to improve hygiene and biosecurity on farms (31–39) (40). A list of alternative strategies to reduce antimicrobial use on farms, as extracted from the studies cited, are:

These management options need to be economically and practically feasible to be successfully implemented on farms. There are many economic studies that analyze the decision making on farms due to certain changes in policies or markets (41–44). Generally, the economic consequences of implementing alternative management strategies on farms are one of the major barriers for farmers to change their farm management strategies (45–47). There are few studies published that investigate the economics of AMU and AMR in livestock species (48, 49, 50, 51) and specifically on pig farms (39). In this manuscript, we describe the use of a farm level economic pig production model (FEPM) to assess the economic impacts on profitability in the immediate term, of the adoption of some common options included in three scenarios. A sensitivity analysis illustrates, for the first time, the relationship between the impacts of these scenarios and other factors that influence the profitability of UK pig farms.

In this study, a static, farm level, economic pig production model (FEPM) was used to explore the economic effects of the adoption of alternative antibiotic use management options on two farms with different profitability levels over a calendar year. Farm gross marign was used as a measure to compare the economic impact of alternative management options with a baseline scenario. A sensitivity analysis was included in the model to examine the effect of price changes on the modeling outputs.

Farmers make decisions on management strategies to achieve farm targets, such as to expand production, to increase profitability, to reduce greenhouse gas (GHG) emissions, or as assumed in this paper, to reduce antimicrobial use on farms. One of the main drivers influencing farm level decisions on different production systems is the economic aspect of those decisions (70–72). Economics become much more important in decision making on commercial farms like poultry, pig and dairy farms (39, 59, 73–75). Understanding the economic feasibility of farm level decision making provides essential insight into the pros and cons of management strategies on farms. This paper looked at the economic feasibility of different AMU reduction scenarios using a static economic pig production model (FEPM). The modeled period was 1 year. The FEPM is a modified version of ScotFarm, a farm level economic model developed at SRUC to conduct impact assessment of external shocks to an agricultural production system. The ScotFarm model has previously been used to assess economic impacts of policies (57, 60, 76), to explore management scenarios reducing GHG emissions (56, 77), and to determine the financial burdens of livestock diseases on farms (58, 59). The FEPM model was run on two types of pig production systems in the UK. These farm types were selected on the assumption that economic impacts of management changes would be different for different farm types (72, 78). Matheson et al. (79) also found that the reduction in amount of antimicrobial use between 2017 and 2018 was different between pig farms in the UK based on their characterisation.

This type of modeling study does not and should not be used to provide direct advice for individual farms; it explores generically the comparative effects of adopting different scenarios in a simplified, representative, system. The three alternative AMU reduction scenarios used in this study were identified by the pig experts as the most plausible scenarios for implementation on pig farms in the UK. These three scenarios of implementation are relatively common across the industry, though it is important to note that not all scenarios are equally applicable – individual farms need to decide whether any of these scenarios are feasible and suitable within their context. For example, depopulation option as considered under the ‘AMU Depop’ scenario may not be adequate to prevent a higher risk of reinfection if there are neighboring farms which are not adopting similar measures to prevent disease on farms.

The economic and production parameters under these scenarios were based on published literature and were checked with experts to be appropriate, (i.e., valid), values. The parameters used in the model are very similar to the parameters used for Irish pig production (61). It was assumed that production would increase by 10% due to improved hygiene and biosecurity in all three scenarios. It can be argued that some farms may have more than 10% increase in production level due efficient and improved management conditions on farms, and due to the overall higher health status, particularly in the AMU Depop scenario. Indeed, Sasaki et al. (68), documented the effect of depopulating and restocking on three farrow-to-finish Japanese pig farms on reproductive and production performance. These authors used 2 years’ worth of data before the depop and 2 years’ worth of data after the repop. They found that postweaning mortality and age at slaughter (days) reduced by 58 and 11% respectively, while average daily gain (g) increased by 12%. However, we kept the 10% improvement assumption in all scenarios for this paper because, firstly, this assumption was based on studies for pig farms undertaken on European pig production system that are similar to the UK pig farms used in this paper and secondly, we did not find any other peer reviewed paper apart from the Sasaki paper to suggest a higher increase in productivity.

This study showed the monetary benefits of making management improvements while reducing AMU by 35% on farm. The other two AMU reduction scenarios appear to have led to decreased farm gross margin. However, there are a couple of important considerations that must be made to understand these results. First, we used a static model for this work, meaning that all costs were incurred in one time point and the returns must be seen in the context that the time frame for the profitability estimate is one calendar year. The two less-profitable scenarios required major investments, such as improvements in management and biosecurity, or the depopulation and repopulation of the farm. The economic return of such interventions is likely to be seen over time and to increase with each additional production cycle run after the changes were implemented. For instance, a Dutch study showed a reduction of around £50 in animal health costs per sow over the years by implementing de-populationmanagement system (80). Second, we assumed that the farm’s health status is stabilized with the improved management and specific pathogen free claims and the likelihood of disease outbreaks is reduced substantially over the years. In the field, for individual farms this may not be the case, for instance, the benefits of depop-repop can be lost immediately if disease recurs through a breach in biosecurity.

The results also showed that reducing AMU under the described scenarios would have a slightly higher adverse impact on a ‘Top third’ farm compared to a ‘Mid-range’ average farm. A ‘Top third’ farm has, in general, higher input costs compared to an average ‘Mid-range’ farm. Farms with higher input costs are known to be more vulnerable to increased variable costs compared to farms with lower input costs (81).

This study highlighted the impacts of changes in pig revenues and feed costs. The high volatility of these costs and prices make decision making on farms very difficult. There has been an increase in average pig market price in 2023 by almost 38% compared to 2018 pig prices in the UK (62). However, feed costs also increased by a substantial 78% due to higher grain prices (62). Although, these changes in market prices are substantially higher than the changes used in the sensitivity analysis (±5 to ±15%) in this study, the results show that an un-favorable pig price and feed costs would make famers’ adoption of these AMU management scenarios financially challenging. These aspects need to be considered seriously when developing AMU reduction strategies and policies. Due to recent high uncertainty in market prices, there have been recommendations to reduce feed costs on pig farms by increasing feed conversion efficiency, reducing slaughter weight and the use of alternative diets (82). These recommendations are however beyond the scope of this study and hence were not included in the model.

In conclusion, it seems to be possible to reduce AMU while still maintaining a profitable farm, as long as the AMU reduction is done in conjunction with farm and management improvements. Under the assumptions of this model, a short term decision making management changes coupled with a small AMU reduction on farm is attainable and translates into higher margins over 1 year. Although our model predicted that more drastic AMU reductions incurred a loss of farm gross margin compared to the baseline model, it is likely that over time the economic return will be favourable and translates into higher margins. Market fluctuations and feed prices are major factors affecting pig production and the profitability of pig enterprises currently; they must be considered when making decisions about antimicrobial use reduction strategies.

Beyond market fluctuations, our model did not account for the training and behavioral changes needed so that farm staff learn new procedures and improve management overall. Farmers’ attitude toward AMU including their knowledge of AMR, capabilities, social pressure and motivation play a role in adopting AMU reducing options (83). Regan et al. (84) argued that a bottom-up behavior change from an individual farmer to organizational and societal level interventions alongside implementation of regulatory policies are required for a wider adaptation of AMU reducing practices. We acknowledge that analysis of the uptake of AMU options based solely on farm gross margin is limiting. However, economic feasibility is always one of the main concerns of farmers and plays a major role in making decisions to uptake new farm practices (85). It is clear from the literature consulted and expert input that just reducing AMU (without any other measures) is not going to increase margins and could have a negative impact on animal welfare. Each farm must consider what is the most feasible and suitable set of improvements to accompany AMU reduction and should see this endeavor as a holistic attempt to improve health and welfare on farm. The approach chosen (i.e., improvements and management changes to be implemented) should be a joint decision between farmers, their staff, and the farm veterinarian.

The data analyzed in this study is subject to the following licenses/restrictions: The data used in the model is taken from the e-MB owned by AHDB, UK and is restricted to protect confidentiality of the UK pig farmers. Requests to access these datasets should be directed to https://emb-pigs.ahdb.org.uk/.

The animal study was approved by the Veterinary Medicine Directorate. The study was conducted in accordance with the local legislation and institutional requirements.

SS: Writing – review & editing, Writing – original draft, Validation, Methodology, Investigation, Funding acquisition, Formal analysis, Conceptualization. MC: Writing – review & editing, Validation, Methodology, Formal analysis. CC-G: Writing – review & editing, Methodology, Funding acquisition, Conceptualization. AN: Writing – review & editing, Validation, Data curation. AM: Writing – review & editing, Validation, Data curation. ST: Writing – review & editing, Validation, Supervision, Resources.