AMU in livestock farming, particularly in poultry and cattle, poses a significant challenge due to its contribution to the development of AMR. Understanding the patterns and behaviors associated with AMU is crucial in devising effective interventions to rationalize its use and mitigate the public health threat of AMR.

Key study findings indicate that all respondents reported using antimicrobials on their farms, which aligns with previous findings [7, 25]. Half of them reported the use of colistin, of which 55% reported they used it for disease prevention and growth promotion. Alarmingly, quite often, participants do not leave a WP between using antimicrobials and slaughtering or using animal products. Poultry farms had significantly higher rates of non-therapeutic AMU, total colistin use, and non-therapeutic colistin use than cattle farms. Our results also showed that the farm personnel’s attitudes towards non-therapeutic AMU and their perceived rates of non-therapeutic AMU in similar farms were associated with their actual non-therapeutic AMU.

During this study, several Egyptian websites were found to provide fixed schedules for growing poultry, which included routine antimicrobial administration. Those schedules repeatedly included ciprofloxacin and colistin administration to poultry from day 1. That practice might have had repercussions reflected in the high resistance rates of E. coli against these two antimicrobials found in our previous work in Egypt (81% and 44% for ciprofloxacin and colistin, respectively) [31]. It is concerning that many participants did not consistently observe WP before slaughtering or using animal products, a practice that may lead to antimicrobial residues entering the food chain. Similarly, Xu et al. in China found that 26% of poultry farmers did not follow a WP [25]. The lack of a WP results in antimicrobial residues reaching consumers and exposing them to hazardous adverse effects [35].

AMU in poultry farms was more injudicious than in cattle farms. Van Boeckel et al. statistically predicted a global average rate of AMU in poultry three times higher than in cattle [24]. Moreover, Xu et al. and Glasgow et al. found that 78% and 88% of poultry farmers in China and Grenada, respectively, used antimicrobials for non-therapeutic purposes, and the latter reported routine antimicrobial administration to chickens since early life [25, 37]. Poultry is thought to be more prone to Salmonella and E. coli infections; hence, antimicrobials targeting gram-negative bacteria are heavily used in poultry farms [38]. Additionally, in the studied poultry farms, large-scale farms had a significantly higher rate of non-therapeutic AMU (87%) than small-scale ones (56%). That comes in line with the poor AMU practices reported with intensive animal farming, where it might not be feasible to individualize AMU and where routine mass antimicrobial administration is a common practice [38]. The situation is more challenging in large-scale poultry farms, probably because they have an even larger number of animals than cattle farms [34].

The WHO guidelines for appropriate AMU in food animals recommend a total cessation of using medically important antimicrobials as growth promoters and restricting their therapeutic use to medically diagnosed infections. Moreover, critically important antimicrobials should not be used for prophylaxis, and HPCIAs should be avoided entirely [39]. However, in our study, participants reported that the most commonly used antimicrobials on their farms were penicillins, macrolides, and tetracyclines. Penicillins are classified as highly important for human medicine, tetracyclines are critically important, and macrolides are classified as HPCIAs [18]. In addition, half of the studied farms used colistin, mostly for non-therapeutic purposes. It is concerning that colistin is also one of the HPCIAs and is used as a last resort for treating serious antimicrobial-resistant infections [39]. Furthermore, poultry farms had significantly higher rates of both total and non-therapeutic colistin use than cattle farms. That might be explained by the observed online dissemination of information on social media (Facebook groups, blogs, websites, etc.), recommending colistin several times during the rearing cycle in poultry. Also, colistin is legally sold in Egypt for veterinary use and is available via several pharmaceutical companies, making it accessible and relatively easy to procure [40].

Another important factor is the route of antimicrobial administration. Oral administration was the main route for all antimicrobial classes, except for cephalosporins. Also, adding antimicrobials to the animal’s drinking water was the single most common route for colistin, macrolides, tetracyclines, polypeptides, penicillins, lincosamides, and quinolones. In animal farming, especially in large-scale settings, this sometimes can be the most feasible route. However, it poses a risk of improper dosing due to seasonal variations; for example, during summer, higher temperatures lead to water evaporation, increased antimicrobial concentration, and higher consumption by animals. That is added to the risk of selective pressure on gut microbiota induced by oral antimicrobial administration, thus contributing to the development of AMR [19]. The European guidelines recommend against mass AMU and state that individual drug administration should be applied whenever possible [41].

Although all the participants reported vaccinating their animals, there was vast variability in vaccine types. The Egyptian Central Administration for Veterinary Quarantine often carries out vaccination campaigns, against foot-and-mouth disease, Rift Valley fever, and avian influenza, in cattle and poultry farms and livestock markets. However, these campaigns usually take place in response to epidemics [42]. There is currently no fixed vaccination schedule for farm animals in Egypt.

Among methods used to measure livestock farm productivity, FCR is one of the most common indicators. A lower FCR indicates greater farm productivity [16]. In the participating farms, the FCR values were consistent with the normal ranges reported in Canada: below 2 for poultry and 4.5–7 for cattle [43]. It is noteworthy that no significant difference was found in median FCR between farms that used antimicrobials therapeutically and those that used them nontherapeutically, for both poultry and cattle. That is consistent with a report published in the USA, where there was no significant difference regarding the FCR, the annual production, and mortality rate between poultry farms that used antimicrobials for non-therapeutic purposes versus those that did not. The report suggested a decline in the efficacy of antimicrobials as growth promoters since 2000 [16].

It is essential to understand the drivers behind AMU in livestock [27]. We found a significant association between the participants’ attitudes towards non-therapeutic AMU and the actual use rate in their farms. Also, there was a highly significant association between the participants’ perception of the rate of non-therapeutic AMU in other farms (subjective norm) and the non-therapeutic AMU in their own farms. Non-therapeutic AMU was significantly higher among those who thought that most other livestock farms used antimicrobials for non-therapeutic purposes. A similar association was reported by Visschers et al. and Om and McLaws, who found that antimicrobial was influenced by how the farm personnel perceived the AMU rates in other farms [29, 30].

It is promising that most participants agreed to get training, particularly on proper AMU in animal farming, protecting animals against infections, and AMR. This agreement may indicate their awareness of the problem and a willingness to implement future interventions to rationalize AMU on their farms. It is worth mentioning that Xu et al. found that prior training of livestock farm personnel was associated with better AMU [25]. Future training can be tailored to accommodate different farm personnel through a variety of approaches, such as online courses, social media campaigns, and awareness campaigns at the farms. They can include reaching out to persons managing Facebook pages and blogs with a large audience of farm personnel and incorporating awareness campaigns into vaccination campaigns conducted by the Egyptian Central Administration for Veterinary Quarantine at the farms. Topics to be covered can include essential biosafety measures, the AMR implications, and the farm personnel’s roles in preventing and combating AMR. To promote more effective and sustainable farming practices, interventions should be developed to educate farmers on the benefits of judicious AMU and alternative growth-promoting methods, as well as the repeatedly reported finding that non-therapeutic AMU does not improve feed efficiency as intended. Furthermore, providing incentives such as subsidies or tax breaks to farms that adhere to proper AMU and biosafety measures and conducting follow-ups on their production rates can help ensure their financial sustainability. These strategies can be augmented by sharing success stories from similar farms to encourage the adoption of best farming practices.

Finally, the current study was the first to shed light on the patterns of AMU by applying a theory-driven approach. The study sample exhibited a distribution of farm locations that mirrored the actual farm distribution in Egypt, as reported by Egypt’s Central Agency for Public Mobilization and Statistics (CAPMAS) in 2019. The CAPMAS report highlights a distinct concentration of poultry production within the Nile Delta governorates, while cattle production tends to be centralized in other regions, similar to the distribution observed in this study (58% of poultry farms vs 36% of cattle farms were in the Nile Delta) [44].

This study had limitations that are worth discussing. One was the absence of an official farm list in Egypt, necessitating a convenience sampling approach that could introduce selection bias. To mitigate this, we selected farms from various governorates across the country and ensured that the chosen farm types (cattle/poultry) matched the distribution reported in the CAPMAS report [44]. Additionally, there was a possibility of underreporting inappropriate AMU due to social desirability bias, where some participants may have tended to choose more socially favorable answers instead of the true ones. To minimize this, the instrument was self-administered by participants, and confidentiality was assured.

Another limitation is the potential underrepresentation of farms without registered veterinarians. According to the FAO, almost 50% of the poultry farms in Egypt are unregistered, and consequently not subject to governmental supervision [34]. In our study, most of the participants (91%) were veterinarians. This might not be the actual case in most farms in Egypt, where farmers often avoid hiring veterinarians to reduce costs and attempt to treat their animals themselves [7]. If having a veterinarian on staff contributes to reducing injudicious AMU, the actual rates of AMU could be even higher than what was reported here, raising further concerns.

The regulatory framework for AMU in livestock in Egypt is evolving slowly. A recent ministerial decree mandates veterinarian supervision for newly registered farms [45], a step in the right direction, which should be complemented by other policies to reduce the injudicious AMU in animal farms. This requires collaborative efforts of the involved institutions, such as the Ministry of Agriculture and Land Reclamation (MALR), the Ministry of Health and Population (MOHP), and international collaborations with organizations such as the WHO and the FAO. In 2018, Egypt developed a National Action Plan (2018–2022) in collaboration with the WHO to combat AMR using a One Health approach. While some of the plan’s goals related to human health have been achieved, there is currently no evidence that any of the objectives for livestock have been met [46].