A global action plan on antimicrobial resistance was adopted by the World Health Organization (WHO) in 2015 (24), with support from the Food and Agriculture Organization of the United Nations (FAO) (25) and the World Organization for Animal Health (OIE) (26), with responsible and prudent use of these medicines in human and animal health as a key goal. The OIE has outlined key strategies for prudent use of antimicrobials in support of these global efforts, including improved AMR awareness and understanding, strengthened knowledge through surveillance and research, the support of good governance and capacity building, and encouragement for implementation of international standards (27). The WHO has identified criteria for classifying antimicrobials of importance to human medicine (19) and similar guidelines have been produced for food producing animals (28).

There is international acceptance of the need for a One Health approach given the substantial use of antimicrobial agents in both human medicine and food animal production. This approach also ensures coordination across all relevant sectors, both to reduce antimicrobial usage (10, 29) and to limit the emergence and spread of AMR (30–32). Based on findings from a recent systematic review, reducing the level of antimicrobial use in livestock populations is likely to be a beneficial strategy for both animals and humans (33). Further, there are examples of linkages between AMR in food animals and humans, through the acquisition of resistant bacteria or, more importantly, through the spread of resistance genes (34). Although the mechanisms for cross-species transmission of resistant bacteria and their genetic elements are not fully understood, it seems clear that the health of humans, animals, and the ecosystem are intricately linked, and that an interdisciplinary and multi-sectoral approach will be required to address the problem of AMR (10).

There is a particular focus on antimicrobial stewardship, which refers to the efforts made to ensure that antimicrobials are used only when necessary and appropriate. It was first established as a set of “responsible use” policy measures to combat AMR in human hospitals (35) and is now used widely in human medicine (36, 37). In food animal production, antimicrobial stewardship similarly refers to a commitment to judicious use of antimicrobials (38–40), including efforts to limit inappropriate usage, to optimise the choice, dose rate, route, and duration of therapy to maximise clinical cures, and to minimise the emergence and spread of AMR.

In Europe, the European Commission developed guidelines on the prudent use of antimicrobials in veterinary medicine in 2015 (41). In 2017, the Commission adopted the EU One Health Action Plan against AMR (42), with key objectives including making the EU a best-practice region, boosting research, development and innovation, and shaping the global agenda. Subsequently, the Farm to Fork Strategy (as one of the policy areas within the European Green Deal) was adopted in 2020 as a tool to help share the EU's path toward sustainable food systems (43). The Farm to Fork Strategy seeks to accelerate the transition within the EU to a sustainable food system, given the linkages between healthy people, healthy societies and a healthy planet. AMR is listed as an area of particular concern within this strategy, with a key target being a 50% reduction in overall EU sales of antimicrobials for farmed animals and for agriculture by 2030. Regulations 2019/4 (on medicated feed) (44) and 2019/6 (12) provide a wide range of measures to fight AMR and promote a more prudent and responsible use of antimicrobials in animals (20).

In Ireland, the National Action Plan on Antimicrobial Resistance (termed iNAP) was established in 2017. The plan recognises the urgent and growing problem of antimicrobial resistance for human health worldwide, and aims to implement policies and actions to prevent, monitor and combat AMR across the health, agricultural and environmental sectors (45). iNAP objectives relevant to mastitis control and intramammary antimicrobial stewardship in the dairy industry include:

In the food animal industry, a code of good practice regarding the responsible prescribing and use of antimicrobials has been developed (46). In the dairy industry, Animal Health Ireland coordinates CellCheck, Ireland's national mastitis control programme (47), and have produced guidelines for the use of selective DCT (48) and prudent intramammary antimicrobial prescribing (49).

The development of the CellCheck programme was substantially informed by Countdown Downunder, the Australian national mastitis control programme (51), and a memorandum of understanding is in place between these two programmes. Established in 1998, the Countdown Downunder programme placed particular emphasis on resource development, including the Countdown Downunder Farm Guidelines for Mastitis Control, in which mastitis control information was arranged according to stage of lactation (calving, lactation, late lactation, drying-off, dry period), a dry cow consult to support decision making at drying off, and a Mastitis Focus Report providing an overview of udder health in an individual herd (52). Training is also critical, leading to skills development among veterinarians, field officers and milk quality staff, herd improvement personnel, milking machine technicians and other dairy advisors. Key features of the programme have included clear, consistent industry-agreed messages, a regional advisory capacity for mastitis control, delivering extension messages through local advisers and using a team approach when dealing with mastitis issues. A 10-year review confirmed initial progress toward industry cell count goals, however, this trend was subsequently reversed as a result of severe environmental conditions (51).

CellCheck follows a similar resource and training model, seeking to increase farmer and advisory awareness of appropriate mastitis control strategies, and to facilitate access to resources to assist with on-farm mastitis control. The programme is informed by the CellCheck TWG, which has guided the development of the Farm Guidelines for Mastitis Control, the CellCheck Dry Cow Strategy, and the Farm Summary Report (which provides an overview of udder health in an individual herd). There is a CellCheck Implementation Group, to facilitate industry engagement and ownership (53). Improvement in national milk quality as measured by BMSCC was observed initially but has subsequently plateaued (Figure 1). In 2020, 65% of herds in Ireland had an annual unadjusted geometric mean SCC <200,000 cells/mL compared with 39% of herds in 2013 (54).

A range of measures have been introduced in Denmark and the Netherlands to address antimicrobial usage and improve antimicrobial stewardship in food animal production. As highlighted in Table 1 (with further detail in the Supplementary Material), the first of the listed measures was introduced in 1995 in Denmark and in 2012 in the Netherlands. Denmark introduced legislation in 1995 to “decouple” antimicrobial prescribing and sales, limiting veterinary profits from direct sales to 5–10% (72). In the same year, veterinary advisory service contracts were introduced on a voluntary basis, and subsequently mandated in 2010 for larger herds. This requires frequent veterinary visits and a 1-to-1 relationship between the farmer and the veterinarian. Further, farmer access to antimicrobials is linked to the level of farm oversight that is provided by the veterinarian (61). Treatment and control measures are underpinned by an understanding of the aetiologic agent (i.e., bacterial culture), patterns of udder infections and antimicrobial resistance in each herd, and the use of narrow-spectrum antimicrobials and selective DCT has become the norm. This approach is central to basic veterinary training in Denmark (and other Nordic countries), and the same message is conveyed to dairy farmers (67). Farm-level benchmarking has facilitated the introduction of the yellow card system, first in the pig industry, identifying those farmers, for remedial actions, with the highest consumption of antimicrobials per animal produced (73). In the Netherlands, independent evaluation of the concept of decoupling was conducted in 2010. It was concluded that economic motives to prescribe would best be addressed by introducing a strict 1-to-1 farmer-veterinarian relationship (in place from 2010) and prescriber benchmarking (from 2013) (62). Since 2012, there has been a mandatory requirement to register all antimicrobials supplied by veterinarians (74), and benchmarking, both of farmers and prescribers, has been identified as a critical factor for success in prudent prescribing. Benchmarking is generally viewed positively by Dutch veterinarians (75). In 2012, prophylactic use of antimicrobials was banned in the Netherlands, and the use of HP-CIAs (3rd and 4th generation cephalosporins, fluoroquinolones) in farm animals was prohibited, apart from exceptional circumstances (62). Since 2012, SDa (Autoriteit Diergeneesmiddelen, Netherlands Veterinary Medicines Institute) have produced an annual report of antimicrobial usage in agricultural livestock (56). Guidelines for the implementation of selective DCT were introduced in 2014 (76).

In Denmark, measures to reduce antimicrobial use were mainly established and carried out by the government, following consumer concerns about antimicrobial use (62). A strong level of trust among farmers, consumers, educators, researchers and governmental agencies has enabled strict legislation and recommendations on antimicrobial use in food animal production to be introduced and implemented. This approach has received strong societal support, with progress reliant on the willingness of all stakeholders to cooperate (67). In the Netherlands, multiple events contributed to national change, including consumer concern following the discovery of a reservoir of multidrug resistant S. aureus (MRSA) in the livestock sector (77). The Dutch government acted as a facilitator for change, coupled with self-regulation by private parties. In the Netherlands, the government set national reduction targets, co-funded SDa, incorporated private regulations into legislation, intensified inspection and enforcement of legislation and supported the strengthening of the independent position of veterinarians through the introduction of technical measures of antimicrobial prescribing and usage (62). For comparison, in New Zealand, where selective DCT is now widely practiced, change was driven solely by industry, without any regulatory oversight (78).

In Ireland during 2018, there was a cessation in the sale of intramammary products containing HP-CIAs from many points of retail and the development of a national policy on the use of HP-CIAs in food animals (46). Further, prescribing guidelines for private veterinary practitioners (PVPs) were introduced in 2022 (49) (Table 1).

BMSCCs provide a general indication of the level of intramammary infection (IMI) in dairy herds. It is a crude measure of cow-composite SCCs, and therefore is a crude measure of the prevalence of IMI within a herd (80). A farm with BMSCC consistently above 200,000 cells/mL (the industry's most common threshold) or either a sharp or gradual increase can signal the existence of a mastitis problem (81), especially in herds with a contagious mastitis problem (82).

Multiple factors influence the association between BMSCC estimates and within-herd IMI prevalence, including the udder pathogen(s) involved (for example, Streptococcus agalactiae will more strongly stimulate a cellular reaction than Staphylococcus aureus), the strictness with which milk from cows with clinical mastitis is kept out of the bulk tank, the age of the cows, the stage of lactation and the average herd production level (83). Although care is needed with interpretation, in the context of udder pathogens such as S. aureus, higher annual BMSCC estimates are generally indicative of herds with suboptimal mastitis control.

BMSCC thresholds vary between countries. In a meta-analysis investigating the efficacy of selective and blanket DCT, the most common herd-level selection criteria was to have a BMSCC below a predetermined threshold (ranging from 250,000 to 400,000 cells/ml) (84). Ruegg (81) suggested that consistent production of milk with BMSCC values <200,000 cells/mL is an achievable goal for most dairy farms.

Many tests have been evaluated for the diagnosis of subclinical mastitis, which is defined as the presence of inflammation with a normal appearance of the mammary gland and visibly normal milk (85). Reflecting the inflammatory status of the mammary gland, milk SCC is used extensively to monitor udder health and milk quality (85, 86). It is associated with the risk of IMI, both at the level of the quarter and the cow (87), and a cow-composite SCC of >200,000 cells/mL is a strong indicator of mastitis (81). As reported by McDougal (87), cow-composite SCC has a diagnostic sensitivity (Se) of 0.3–0.85 and specificity (Sp) of 0.67–0.9 of predicting IMI, given a cut point of 200,000 cells/mL. The California mastitis test (CMT) is commonly used to diagnose IMI (17). However, if the CMT were used to aid selection of cows suitable for selective DCT, a large percentage of uninfected quarters would receive DCT (23–46% of cows without infection with major udder pathogens were detected by CMT) (88, 89). Culture (at the laboratory and on-farm) and real-time multiplex polymerase chain reaction (PCR) can be useful in the diagnosis of an IMI. On-farm culture systems generally seek to categorise udder pathogens into broad categories, as an aid to treatment rather than a means to make species-level pathogen diagnoses (85). Compared with culture, PCR is both faster and more sensitive, but more costly and with the potential to detect DNA from dead bacteria (85). Machine learning algorithms, to aid analysis of the large amounts of farm data that are generated, offer a promising new tool to support farm-level decision-making (86), including predictive algorithms for intramammary infection status in late-lactation cows (90).

Mastitis is often associated with low levels of hygiene during milking (e.g., wearing of gloves for milking, milking clean teats and appropriate post milking teat disinfection) or at dry-off (3, 91, 92) as well as with general farm management practices (93). In comparison to higher BMSCC herds, lower BMSCC herds are more likely to implement management practices that are conducive to improved mastitis control (including attention to hygiene; cleaner cubicles, drinking buckets and cows; cleaner calving pens and drier bedding for lactating cows in maternity pens; post-milking disinfection; DCT) (94). In the case of contagious mastitis, ongoing infection pressure will facilitate within-herd transmission, potentially placing younger, non-infected cows at risk, even with careful attention to detail during milking. This discussion is particularly important in the context of the Regulation, which states that antimicrobials “shall not be applied routinely nor used to compensate for poor hygiene, inadequate animal husbandry or lack of care or to compensate for poor farm management” (12). The majority of antimicrobial usage in dairy herds is associated with mastitis treatment and control (95), and optimal mastitis control is central to efforts to reduce on-farm antimicrobial usage.

Based on the most-recent estimates from 2020, 35% of Irish herds had an annual unadjusted geometric mean BMSCC of 200,000 cells/mL or greater, suggestive of the potential for suboptimal mastitis control in these herds. National progress toward improved mastitis control has slowed in recent years (Figure 1).

S. aureus, Streptococcus uberis and Escherichia coli are commonly isolated udder pathogens on Irish dairy farms (96, 97). There is a high prevalence of S. aureus infection on Irish farms.

The level of milk recording in Ireland (in 2017: 33% of herds and 48% of cows) is low relative to international counterparts (98). Milk recording provides animal-level information to inform both mastitis control and prudent prescribing.

Dairy production is highly seasonal on most Irish farms, which adds complexity to herd management at specific periods of the year (99). An estimated 1 million dairy cows are dried off in Ireland over an 8-week period.

There are ongoing difficulties in sourcing experienced labour (on dairy farms, as milking technicians, for bulk tank servicing), exacerbated by a period of recent expansion in the Irish dairy industry. These challenges can substantially complicate farm management, including those activities contributing to mastitis control and intramammary antimicrobial stewardship.

In 2017, 32.9% of surveyed Irish dairy farmers provided <1 cubicle per cow throughout the winter housing period (100). Only 2.5% of surveyed farmers had more than 1 cubicle per cow (101). These housing constraints have implications for cow welfare and mastitis control.

Detailed mastitis investigations are not routinely conducted in response to on-farm mastitis problems, either by veterinary practitioners or other farm advisors. Consequently, on-farm recommendations can be generic in nature, rather than informed by a detailed understanding of the epidemiology of infection or of broader drivers for mastitis on the farm in question. This is changing, in part with the introduction of the Graduate Certificate in Dairy Herd Health for veterinary practitioners at University College Dublin (102).

A series of data adjustments and interpretation are applied in Ireland to determine herd eligibility to supply raw milk for processing of dairy products. Legislation is currently being interpreted in a manner that facilitates ongoing supply (103, 104), which may reduce the imperative to sustainably address mastitis issues in herds with suboptimal mastitis control.

Most Irish dairy farmers are members of the Bord Bia Sustainable Dairy Assurance Scheme (SDAS), which is a national quality assurance programme that includes a range of private animal health and welfare standards (105, 106). Although SDAS has the potential to motivate farmers toward improved milk quality, this potential has not yet been realised because the SDAS standards do not exceed the legislative baseline with respect to milk quality (106).

National BMSCC data is not yet available for public good research. Consequently, it is not possible to answer key questions to help inform national decision-making on progress in national mastitis control, the CellCheck programme or antimicrobial stewardship in Ireland.

In Ireland between 2007 and 2022, remote prescribing of intramammary antimicrobial products was allowed under national legislation (107), leading to the potential for very limited veterinary oversight of these products. Although this legislation is now revoked (108), there may be a legacy impact on the attitudes and behaviours of farmers and others within the dairy industry, with implications for future efforts toward improved intramammary antimicrobial stewardship.

In Ireland, prescribing must be undertaken by a veterinary practitioner; this is generally the attending veterinary practitioner (or group of veterinary practitioners) who have been given responsibility for the primary care of a herd by the designated keeper in the context of a client-patient-practice relationship (CPPR) (“an agreement between an animal owner (or designated keeper) and a veterinary practitioner(s) within a veterinary practice to provide veterinary services that demonstrate real and ongoing clinical veterinary practitioner/animal contact”) (109). However, a herdowner may have more than one CPPR in place, thereby allowing for the potential for antimicrobials to be sourced from more than one prescriber. In addition, between 2007 and 2022, remote prescribing of intramammary antimicrobial products was allowed.

Ireland is currently reliant on national sales data to estimate on-farm intramammary antimicrobial usage. Given the nature of these data, farm-level usage cannot be calculated, and prescriber and farm-level benchmarking are not currently possible.

Based on ongoing analyses of national sales data, there has been minimal shift to this point from blanket to selective DCT on Irish dairy farms (66).

There is evidence of ongoing usage of EMA category B (“Restrict”) antimicrobials in in-lactation therapy and of increasing usage of these antimicrobials in DCT (66).

Under the Regulation, prophylactic use must not be routinely carried out. Further, “antimicrobials should not be applied routinely nor used to compensate for poor hygiene, inadequate animal husbandry or lack of care or to compensate for poor farm management” (12). In compliance with this Regulation, DCT should be limited to therapeutic use in cows known to be infected. Further, the EMA guidelines should be followed, providing an overview of the categorization of antimicrobials for use in animals for prudent and responsible use (18).

The use of selective DCT may not increase the risk of intramammary infection at calving if internal teat sealants are used for all cows (110). As highlighted previously, high standards in hygiene will be critical during the administration of teat sealant (23), and throughout the dry period and around calving. Although infection risks relating to the milking parlor are removed during the dry period, higher BMSCC herds and those with high incidence of infections may also have suboptimal management conditions which may be associated with an increase in infectious challenge during the dry period and around calving.

There is a need for individual cow information to determine infection status at drying off. Currently, milk recording is recommended every 4–6 weeks in Ireland, with a minimum of 6 recordings throughout the lactation including one shortly prior to drying off and one shortly following calving. At this frequency, data are sufficient to support both mastitis control and prudent prescribing. The TWG recognise that other options are also possible (such as bacterial culture, PCR testing, California mastitis testing) (80, 111, 112), but with disadvantages. Further, studies from the Netherlands (113) and New Zealand (87) suggest that a single milk recording, taken within 4–6 weeks of drying off, can provide useful information about infection status at drying off. However, this limited information would be insufficient to guide mastitis control on Irish herds throughout the year.

There is also a need for herd-level information, including knowledge and oversight by the prescriber of all antimicrobials prescribed and used, and of the rationale and strategy for antimicrobial prescribing and use. This is best achieved, consistent with international best-practice in antimicrobial stewardship, with a single prescriber (or single prescribing practice) for each farm. There is also a need for a detailed understanding by the prescriber of the farm, including the herd, the people, the facilities, and farm management (in general, during lactation and at drying off). This information is central to mastitis management, including the investigation of suboptimal mastitis control, to gain an improved understanding of the epidemiology of infection and of factors contributing to suboptimal mastitis control based on a detailed on- and off-farm investigation, to develop and agree a plan with the farmer to robustly and sustainably address each of these factors, including agreed actions and timelines and objective measures to monitor progress, and for ongoing and regular assessment and review. It is important that the prescriber has an understanding of milk quality trends, the mastitis pathogen challenge and antimicrobial sensitivity/resistance patterns, to inform prescribing.

Guidelines for prescribers to support prescribing and mastitis control decisions at drying off are presented in Table 2. In addition, see (49).

Animal- and herd-level information will again be required. The TWG recommends ongoing collection and testing of milk samples from animals with clinical or subclinical mastitis, ensuring a mixture of young and old cows, with evidence of both recent and chronic infections, based on sampling conducted at different points throughout lactation, to identify the causative pathogen(s). This is to guide individual clinical decisions and, equally importantly, as part of the broader assessment and monitoring of mastitis pathogen challenge(s) and antimicrobial resistance patterns on the farm.

The CellCheck TWG recommendations to support prescribing and mastitis control decisions during lactation are presented in Table 3. In addition, see (49).