Antimicrobial Use in Canadian Cow–Calf Herds

Simple Summary As of 1 December 2018, all medically important antimicrobials (MIA) for veterinary use in Canada required a prescription from a veterinarian for purchase. Cow–calf herds, while numerous and an important component of the food supply chain, have not been included in Canadian national on-farm antimicrobial-use surveillance. Studies of antimicrobial-use practices on extensively managed cow–calf operations are limited, and prior to this study, the effects of the regulatory changes on antimicrobials in cow–calf herds had not been reported. The objectives of this study were to describe antimicrobial use practices on cow–calf operations and any changes associated with the imposed regulations. Overall, antimicrobial use patterns in 2020 remain relatively unchanged in cow–calf herds after the regulatory changes and compared to previous comparable data from 2014. While almost all herds used antimicrobials, the frequency of antimicrobial use within cow–calf operations was low. Antimicrobials were primarily used for the treatment of respiratory disease in calves, neonatal diarrhea, and lameness in adults; however, most herds treated less than 5% of animals for these reasons. The most frequently reported antimicrobials, tetracyclines and florfenicol, were not classified as either high or very high importance to human health. Abstract Despite growing concern surrounding antimicrobial use (AMU) and the importance of cow–calf herds to the Canadian livestock industry, surveillance of AMU in cow–calf herds to inform antimicrobial stewardship programs has been sporadic. Producers from the Canadian Cow–Calf Surveillance Network (87%, 146/168) provided data and almost all reported AMU in at least one animal (99%, 145/146 herds) in 2019–2020. The most common reasons for AMU were treatment of respiratory disease in nursing calves in 78% of herds and neonatal diarrhea in 67% of herds, as well as for lameness in cows in 83% of herds. However, most herds treated <5% of animals for these reasons. Less than 2.5% of herds treated more than 30% of calves for either bovine respiratory disease or neonatal diarrhea and no herds treated more than 30% of cows for lameness. The most frequently reported antimicrobial was oxytetracycline in 81% of herds, followed by florfenicol in 73% of herds. Antimicrobials with very high importance to human health, such as ceftiofur, were used at least once by 20% of herds but were only used in >30% of nursing calves from one herd. Similarly, while 56% of herds used macrolides at least once, within-herd use was the highest in nursing calves where <4% of herds reported use in >30% of animals. Herds using artificial insemination and calving in the winter were more likely (p = 0.05) to treat >5% of nursing calves for respiratory disease, suggesting the importance of vaccination programs for herds at risk. Overall, AMU was similar to previous Canadian studies; however, the percentage of herds using macrolides had increased from a comparable study in 2014.


Introduction
Antimicrobial use (AMU) and antimicrobial resistance (AMR) are sources of increasing concern for both human health and the livestock industry [1]. In Canada, two governmentsupported programs estimate livestock exposure to antimicrobials. The first initiative focuses on antimicrobial sales and includes livestock commodities important to the Canadian Risk factors explored included location of the herd (Western vs. Eastern Canada), herd size, number of animals purchased into the herd in 2020, whether a producer backgrounded calves, producer age, type of record-keeping system, season calving began, location of calving (pasture vs. any type of confinement system), use of community pasture, the use of breeding technologies including artificial insemination (AI), and herd type.
The number of animals purchased by each herd during the first half of the year was categorized as above or below the median for all herds and was considered a relative indicator of biosecurity risk for AMU for calf diarrhea and respiratory disease. For evaluating risk factors for AMU for cow lameness, herds were classified depending on whether they purchased more cattle than the median value for all herds throughout the year. Potential risk factors were selected based on previous publications [6,13] and expert opinion.
The secondary objective of the study was to compare the data collected in this publication to the previously collected baseline data to determine if there have been any changes within the industry since the imposition of the new regulations [4]. The relative frequency of specific outcomes of interest was compared between the 2020 and 2014 data [6] using Fischer's exact tests.

Description of Participating Herds
Returned surveys numbered 146 of 168 AMU, for a response rate of 87%: 30% of respondents were from Alberta, 20% from Saskatchewan, 17% from Ontario, 14% from Quebec, 12% from Manitoba, 5% from British Columbia, and 1% from both New Brunswick and Nova Scotia. Two-thirds of herds (67%, 98/146) were in Western Canada (British Columbia, Alberta, Saskatchewan, and Manitoba) and 33% (48/146) were in Eastern Canada (Ontario, Quebec, New Brunswick, and Nova Scotia). Out of the 56 who had previously participated in the WCCCSN [6], 52 returned a completed AMU survey.
The median number of cows calving per herd was 124 (range: 14 to 1044) and the median number of heifers calving was 18 (range: 0 to 210). The median number of cows calving per herd in Western Canada was 131 (range: 23 to 1044) and heifers calving was 19 (range: 0 to 210). Eastern herds were typically smaller with a median number of cows calving of 67 per herd (range: 14 to 578) and heifers of 8 (range: 0 to 109). The median number of bulls was 7 per herd (5th percentile, 2, 95th percentile 31).

Overall Summary of AMU
Antimicrobial use data were collected for nursing calves, weaned calves, cows and bulls (Tables 1-4). A greater proportion of producers reported treating adult cows and nursing calves at least once (Tables 1, 3, A1 and A3) compared to weaned calves or bulls (Tables 2, 4, A2 and A4). Almost all producers reported AMU at least once in a cow (95%, 139/146) or nursing calf (95%, 139/146). At least one cow was treated in 96% (94/98) of herds in the west and 92% (44/48) of herds in the east (p = 0.44). Similarly, 98% (96/98) of western herds reported AMU in nursing calves compared to 88% (42/48) in the east (p = 0.02). Antimicrobial use was reported at least once in weaned calves in 66% (96/146) of operations and for bulls in 65% (95/146) of herds; 71% (70/98) in the west and 54% (26/48) in the east for weaned calves (p = 0.04), and 74% (73/98) in the west and 46% (22/48) in the east for bulls (p < 0.001). a Health Canada 2009, underlined product defines the category. b Rows and columns might not add up to the number of unique herds as herds can use one antimicrobial for different reasons at more than one frequency and more than one antimicrobial at the same frequency.   Medically important antimicrobials were used at least once in 99% (145/146) of herds within the study, with most herds using more than one category of antimicrobials (Table 5). Category I antimicrobials were used at least once by 33% (48/146) of herds, while category II products were used at least once by 80% (116/146) of herds, and category III by 94% (137/146) of herds. Use of category I antimicrobials at least once was similar in the west (36%, 35/98) and in the east (27%, 13/48) (p = 0.35). Category II antimicrobial use at least once was also not significantly different in the east (83%, 40/48) than in the west (78%, 76/98) (p = 0.62). No MIAs or category I-III antimicrobials [12] were administered orally to adult cows or bulls (Tables A7 and A8). All category IV antimicrobials were delivered orally (PO) across all animal classes. Category IV products (Table 6) were used by 14% of herds (21/146). More category IV use was reported in herds from Western Canada (17%, 17/98) than from Eastern Canada (8%, 4/48) (p = 0.21). Adult cows and weaned calves were most likely to receive in-feed category IV antimicrobials at least once. The proportion of animals within a herd treated with a category IV antimicrobial was not collected since this class of antimicrobial is provided in-feed for disease prevention and, therefore, typically given to all animals in a particular animal class. Forty-five percent of herds reported the use of at least one MIA in less than 5% of animals in either nursing calves or cows (Table A15). Most herds (86%, 126/146) reported the use of MIA in less than 30% of nursing calves or cows. Twelve (8%) herds reported treatment of more than 70% of their nursing calves or cows with at least one MIA.
Ceftiofur was the most used category I injectable antimicrobial and was reported as being used at least once in 20% of herds as compared to danofloxacin (3%) and enrofloxacin (2%) ( Table 5). Ceftiofur was reported as being used at least once in 23% (23/98) of herds in the west and 13% (6/48) in the east (p = 0.13) (Table A10).
Category I drug use [12] in more than 30% of nursing calves was reported in one (<1%) herd where ceftiofur was used to treat respiratory disease (Tables 1-4). Category I drugs were not used in more than 30% of animals in weaned calves, cows, or bulls (Tables 1-4).
Of the 146 herds reporting, 4% (6/146) of herds reported the use of tetracyclines in more than 30% of nursing calves (Tables 1 and A11). Tetracyclines were used by 3% (5/146) of herds in more than 30% of weaned calves (Tables 2 and A12), and less than 1% (1/146) of herds in more than 30% of bulls (Tables 4 and A14). No herds used tetracyclines in more than 30% of cows (Tables 3 and A13). Florfenicol was used in greater than 30% of weaned calves in 1% (2/146) of herds, with less than 1% (1/146) of these herds using florfenicol in more than 70% of the weaned calves (Table 2). Florfenicol was not used in greater than 30% of nursing calves, cows, or bulls in any herd (Tables 1, 3 and 4). Only 2% (3/146) of herds used sulfonamide boluses in greater than 30% of nursing calves (Tables 1 and A11). No sulfonamide boluses were used in cows or bulls.
Lameness was the most common reason for the use of antimicrobials across all animal classes with 89% (130/146) of herds reporting that they treated for lameness at least once (Table 5). Respiratory infection was the second-most-common reason for treatment across all animal classes with 87% (127/146) of herds reporting treatment at least once ( Table 5). The least-common reason for treatment across all animal classes was used for diseases that did not fall into the categories provided to producers and were identified as "other". Only 5% (7/146) of herds reported "other" diseases, which included infection, injury, and dystocia (Table 5).
Products considered as MIA were used for the prevention of diseases such as diarrhea and pneumonia in 18% (26/146) of herds and all animal classes except for bulls (Tables A1-A4). AMU for disease prevention was highest for calves before weaning, with 12% (18/146) of herds reporting the use of MIA for disease prevention (Table A1); 15% (7/48) of herds from the east and 11% (11/98) of herds from the west used MIA for disease prevention (p = 0.60). For weaned calves, 5% (8/146) of herds (Table A2) and for cows, 1% (2/146) of herds (Table A3) reported using MIA for disease prevention.
Sixty-two percent (90/146) of producers reported using remote delivery systems such as dart guns, crossbows, and pole syringes at least once to treat an animal. Remote delivery was more common in the west with 75% (73/98) of herds reporting use at least once compared to only 35% (17/48) of eastern herds reporting the use of a remote delivery system at least once. Dart guns were the most popular with 29% (42/146) of herds reporting use. Almost all dart gun use was in the west with 41% (41/98) of herds reporting dart gun use while only 2% (1/48) of herds used a dart gun in the east (p < 0.001). Most herds using dart guns (52%, 22/42) delivered less than 10% of total treatments using the dart gun. Nine producers reported using their dart gun to deliver greater than 50% of total treatments for sick animals.

AMU in Nursing Calves
Ceftiofur, a category I antimicrobial, was used in nursing calves for the treatment of diarrhea in 8% (12/146) of herds and for respiratory infections in 6% (8/146) of herds (Table A1). Danofloxacin use was reported in 4% (6/146) of herds for treating respiratory infections in nursing calves (Table A1). Enrofloxacin use was reported in 2% (3/146) of herds; in one herd for treating nursing calves for respiratory disease, in one herd for diarrhea, and in one herd for navel infections (Table A1).
The category II drugs most likely to be used at least once in nursing calves were trimethoprim/sulfadoxine combinations, followed by penicillins and macrolides (Tables 1 and A10). Trimethoprim/sulfadoxine combinations were most used to treat diarrhea, whereas penicillins were used for navel infections and arthritis (Table A1). Macrolides, most commonly tulathromycin and tilmicosin, were used for treating respiratory disease (Table A1).
The category III antimicrobials that were most likely to be used at least once in nursing calves were florfenicol followed by oxytetracycline (Table 1). Florfenicol was used primarily for respiratory disease, followed by navel infections (Table A1). Oxytetracycline was used for arthritis, followed by respiratory disease and navel infections (Table A1).
The number of herds reporting antimicrobial treatment for respiratory disease in calves before weaning varied by region (p < 0.001), with 86% (84/98) and 63% (30/48) of western and eastern herds, reporting this use, respectively. The number of herds reporting AMU for diarrhea was similar in Eastern and Western Canada (p = 0.99), with both 67% of eastern (32/48) and western (66/98) herds reporting AMU for diarrhea.
Respiratory infection was the most common reason for reporting AMU at least once in nursing calves (78%, 114/146) followed by diarrhea (67%, 98/146) (Tables 5 and A1). Respiratory infection in nursing calves was most likely to be treated at least once with florfenicol or oxytetracycline. Trimethoprim/sulfadoxine followed by sulfonamide boluses were the most common antimicrobials used to treat diarrhea (Table A1).
Oxytetracyclines were the most common antimicrobials used for disease prevention in nursing calves (8%, 11/146) ( Table A1). Benzylpenicillin procaine and benzathine penicillin were the second-most-reported MIA used for disease prevention in this age group, with 3% (4/146) of herds reporting use (Table A1). Two herds also reported using tulathromycin (1%) and one (<1%) herd reported the use of tildipirosin for disease prevention (Table A1).
The most common route of administration in nursing calves was subcutaneous (SQ) injection, followed by antimicrobial products being delivered orally (PO) ( Table A5). Treatment of diarrhea via a bolus was the primary reason for oral AMU in nursing calves (43%, 64/146) ( Table A1). The frequency of bolus use was similar across regions; 43% (43/98) in the west and 46% (21/48) in the east (p = 0.99).
Ceftiofur, a category I antimicrobial, was used for treating weaned calves with arthritis and eye (topical) infections in three herds (Table A2). Enrofloxacin use was reported in one herd for treating weaned calves for respiratory disease (Table A2). Penicillins were most commonly reported for the treatment of arthritis and eye infections (Table A2). Respiratory disease and arthritis were the main reason for the use of the category III antimicrobials tetracycline and florfenicol in weaned calves.
Similar to what was reported in nursing calves many of the farms using antimicrobials treated less than 5% of weaned calves (Table 2). Three percent (5/146) of herds reported treating more than 71% of weaned calves for various conditions including eye and respiratory disease with at least one category II or III product including tetracyclines, macrolides (gamithromycin, tulathromycin), and florfenicol.
Overall, respiratory disease was the most common reason for treatment in weaned calves (56%, 81/146) followed by treatment of arthritis (34%, 49/146) (Table A2). Florfenicol was the most reported drug for the treatment of respiratory disease, while oxytetracycline was the most frequent treatment for arthritis. Herd owners also reported treating weaned calves for eye disease. For disease prevention, the most reported conditions were pneumonia and diarrhea.
The most common method of administering antimicrobials to weaned calves was by SQ injections followed by intramuscular (IM) injections (Table A6), reflecting the drugs selected for use.

AMU in Breeding Females and Bulls
Parenteral ceftiofur was used to treat respiratory disease, reproductive disorders, and lameness in cows (Table A3). Similarly, bull lameness was the main reason for ceftiofur use (Table A4). Danofloxacin, another category I antimicrobial, was used for treating mastitis in <1% (1/146) of herds (Table A3); fluoroquinolones were not used for any other reason in cows nor were fluoroquinolones used in bulls for any reason. The only other category I drug reported was Polymyxin B which is a component of a commercial intramammary (IMM) preparation for mastitis. Fifteen percent of respondents (22/146) used this product at least once for mastitis. Producers also reported the topical use of intramammary preparations containing ceftiofur for treating eye infections in cows and bulls (Tables A3 and A4).
Penicillins were the most common category II antimicrobials given to cows (Tables 3 and A13). Penicillins were primarily used to treat lameness and reproductive-tract infections. However, the use of macrolides was also reported for treating lameness (<10% of herds) and reproductive tract infections (<3% of herds) in cows and bulls (Tables A3 and A4). Macrolides were more likely to be used in cows from the west to treat lameness as compared to cows from the east (p < 0.001) (Table A10).
Oxytetracycline, a category III antimicrobial, was by far the most used antimicrobial in cows and bulls and was administered for a variety of reasons (Tables A3 and A4). Florfenicol was also used to a lesser extent and was primarily administered for lameness. In addition to oxytetracycline, no additional MIAs were used for disease prevention in cows and no MIAs were used for disease prevention in bulls (Tables A3 and A4).
Lameness was the most common reason for AMU in adult cows and bulls (Tables A3 and A4). Treatment of lameness was reported at least once by 83% (121/146) of herd owners for cows and 58% (84/146) for bulls. Treatment of lameness was reported at least once for cows by 87% (85/98) of herd owners in the west compared to 75% (36/48) in the east (p = 0.10). Treatment of lameness in bulls was reported in 68% (66/98) of western herds compared to 35% (16/48) of eastern herds (p < 0.001).
Lameness in adult animals was most frequently treated with oxytetracycline (Tables A3 and A4). Oxytetracycline was also used for treating cows for eye and reproductive-tract infections (Table A3). Eye infection, often reported as pink eye, was the second-most-common reason for AMU in bulls followed by reproductive-tract and respiratory infections (Table A4).
The most popular route of administration in cows and bulls, as with weaned calves, was SQ injection followed by IM injection (Tables A7 and A8).

Factors Associated with AMU
Antimicrobial use by antimicrobial class and percentage of the herd treated was summarized for the three most-common treatment scenarios: preweaning respiratory disease and diarrhea, as well as cow lameness ( Table 7). The antimicrobial used by the greatest number of herds for the treatment of respiratory disease in nursing calves was florfenicol, an injectable phenicol. The antimicrobial reported as used by the greatest number of herds for the treatment of diarrhea in nursing calves was injectable trimethoprim sulfonamides and for cow lameness injectable tetracyclines (Table 7). Macrolides were the only antimicrobials used in >70% of animals within a herd for any of the three mostcommon diseases (Table 7). Macrolide use in >70% of calves was reported in one herd for the treatment and prevention of respiratory disease in nursing calves.
a Health Canada 2009, underlined product defines the category. b Rows and columns might not add up to the number of unique herds as herds can use different antimicrobials within the same class at more than one frequency for the same reason or more than one class of antimicrobials for a specific reason at the same frequency.

Comparison to 2014 Results
Twenty-one percent (30/146) of herds reported treating >5% of nursing calves with at least one antimicrobial for respiratory disease. Based on unconditional analysis (Table A16), herds calving earlier in the year and those herds using artificial insemination (AI) were more likely to treat >5% of nursing calves for respiratory disease. The use of breeding synchronization programs and having a record system also had p < 0.20 in the unconditional analysis. In the final multivariable model, which controlled for the calving season and the use of herd records, herds that used AI were more likely (OR 2.4, 95% CI: 1.0 to 5.9, p = 0.05) to treat >5% of their calves for respiratory disease compared to herds that did not use AI. Herds that calved in the winter (Dec to Feb) were also more likely (OR 2.5, 95% CI: 0.99 to 6.1, p = 0.05) to treat >5% of nursing calves for respiratory disease compared to herds that calved in the spring (March to May).
Diarrhea in nursing calves was the second-most-common reason for at least one calf to be treated prior to weaning (Tables 7 and A1). However, calf diarrhea was the most common reason for treating at least 5% of the herd; 22% (32/146) of herds treated >5% of calves with at least one antimicrobial. Based on the unconditional analysis, no factors examined in this study were significantly associated with the treatment of >5% of nursing calves for diarrhea (Table A16). Only the calving season had a p < 0.20 in the unconditional analysis.
Lameness in adult cows was the most common reason for treatment at least once across all animal classes for all diseases in the study (Tables 7 and A3); however, only 12% (18/146) of herds reported treatment of >5% of cows with an MIA (Table 7). Based on the unconditional analysis (Table A16) herds with a decision maker < 30 years old, those primarily seedstock producers, and those using community pastures were more likely to treat >5% of the herd (p < 0.20). In the final multivariable model, no factors were shown to lead to an increased likelihood of treating >5% of cows for lameness.
A summary table was developed to allow for comparison between this study from 2020, and a similar baseline study in 2014 (Table 8). The table contains the results from the 2020 study, which represents the entire country of Canada as well as the western-only data to allow for a more direct comparison with the 2014 study which only included Western Canada. Overall, AMU was similar between the studies. However, there were some differences including an increase in macrolide use when comparing the 2020 western-only data to the 2014 data (61% vs. 44%) (p = 0.02) ( Table 8). Macrolide use was also more often reported for treating respiratory disease in nursing calves in 2020 compared to 2014 (p = 0.03) ( Table 8). The number of herds treating cows for lameness with a macrolide was nearly double for the 2020 study and higher for the western-only herds in the 2020 study versus the 2014 study (p = 0.002) ( Table 8).

Discussion
This is the first report of AMU data in Canadian cow-calf herds since 2013/2014 [5,6], the first since the new prescription-only regulations in 2018 [4], and the first to provide AMU data from cow-calf herds in Eastern Canada since 2008 [7]. The current study summarizes AMU from July 2019 to June 2020 in a sample of herds from across Canada. The current study provides information from Eastern Canada as well as from Western Canada. The more current data from Western Canada allows for discussion of any differences from the previous report [6] which also focused on a sample of Western Canadian cow-calf herds for the same period in 2013 and 2014.
Oxytetracycline and florfenicol were the most frequently reported antimicrobials used across all animal classes. These antimicrobials are considered of medium importance to human health (category III) by Health Canada [12]. Use of category III antimicrobials more frequently than categories of higher importance to human medicine in cow-calf herds supports antimicrobial stewardship by reducing selection pressure on antimicrobials of high and very high importance needed for treating more difficult infections.
Use of ceftiofur, a category I antimicrobial [12], was reported at least once in 20% of herds in at least one class of cattle, whereas category I fluoroquinolones were reported less frequently with 5% of herds reporting their use. Similar to the 2014 study [5,6], no herds reported the use of enrofloxacin or danofloxacin (injectable fluoroquinolones) in more than 5% of weaned calves or cows. Based on the most recent 2017 USDA beef study, the proportion of herds using injectable cephalosporins and fluoroquinolones is less than what was reported for Canada with only 1.6% and 0.9% of cow-calf herds in the USA reporting their use respectively [9]. However, in the 2017 USDA beef survey, producers were only asked to report their primary antibiotic, whereas, in the current study, they were asked to report up to three treatment options for each condition, making secondary drugs used in sicker animals or in the case of treatment failures more likely to be included. While in Canada, the proportion of animals for which category I antimicrobials are used within individual farms is relatively low, it would be prudent to carefully consider whether they are necessary for all herds where they are currently being used.
Trends in the primary diseases identified and the approaches to treatment in nursing beef calves have remained stable between the 2014 and 2020 studies. Respiratory disease and diarrhea were the two most-common reasons for using an antimicrobial at least once in nursing calves in both 2020 and 2014. While trends across the years were similar, there was a difference across the country in the 2020 data with the respiratory disease being treated more frequently in herds from Western Canada than from Eastern Canada. In the current study, 78% and 67% of herds treated at least one calf with an antimicrobial for respiratory disease or diarrhea, respectively, with a higher proportion of herds treating respiratory disease in the west than in the east. The relative importance of these diseases was similar to what has been recently observed in the American cow-calf industry, although the percentage of herds reporting treatment in the current study was higher. In the 2017 USDA beef study, 36% of medium-sized herds (50-199 cows) and 50% of larger herds (>200 cows) reported treating at least one calf for respiratory disease and 31% of medium-sized herds and 55% of larger herds for diarrhea [9].
Top treatment choices for respiratory disease and diarrhea did not change between 2014 [6] and 2020 with florfenicol most likely to be used for respiratory disease and oral sulfonamides followed by injectable sulfonamide-trimethoprim combinations most likely for diarrhea.
Macrolides are a category II drug considered of high importance to human health; however, macrolide use for treating respiratory disease in nursing calves was reported by a significantly higher proportion of herds in 2020 than in 2014. This change might be associated with the increasing availability of generic forms of one macrolide and the associated decrease in cost to producers, as well as the emergence of new macrolides. The use of macrolides for treating bovine respiratory disease (BRD) in nursing calves was also more common in 2020 in herds from the west than from the east.
Oral administration of MIAs was almost exclusively limited to the treatment of diarrhea in nursing calves via bolus administration. The proportion of herds using an oral MIA for the treatment of diarrhea in calves prior to weaning was 52% in 2014 [6] and 41% in 2020. There was, however, a significant decrease in the use of category II neomycin boluses from 11% in 2014 to 3% in 2020. In the 2017 USDA beef study, 11% of operations reported giving oral antimicrobials to nursing calves as the first-line treatment for diarrhea [9], which is much lower than what was seen either in the 2014 or the current study. The 2017 USDA beef study data were not stratified by herd size, and the study was dominated by small herds (<50 cows); however, the proportion of herds treating calves for scours was substantially higher in larger herds based on information from subsequent tables.
Lameness was the most reported reason for AMU in cows and bulls in 2020. Lameness was reported by relatively more western than eastern herds for both cows and bulls, with nearly double the herds treating bulls for lameness in the west as compared to the east. This difference is likely due to the relatively larger herd size and increased number of animals at risk in western herds. The 2017 USDA beef study [9] and the 2007-2008 USDA beef study [14] also found that lameness was the most-reported reason for treatment in adult cows. Lameness in adult cows was reported in 35% of medium-sized operations and 52% of large operations with approximately 80% of these herds reporting treatment with an antimicrobial [9]. The current study found that 83% of herds reported having and treating lame cows. This is considerably higher than the 35% and 52% reported by the mediumand large-sized herds in the USDA studies.
There are a number of reported differences in the proportion of herds treated for specific diseases between the American and Canadian studies. The reasons for the difference in the percentage of herds treated for respiratory disease, diarrhea, and lameness require further investigation. However, some possibilities could include climate differences in the primary cow/calf rearing areas, farm size, the season of calving, management intensity, recording keeping, and the approach to data collection and analysis.
The use of MIAs in 18% of herds for disease prevention (19% from the west) was similar in the present study to the 2014 study (21%) [6]. Disease prevention was most common in calves before weaning and was unchanged in 2020 with 12% (11% from the west) of herds compared to 11% in 2014 [5]. The most-common antimicrobial choices for preventative reasons also remained the same with oxytetracyclines and penicillins as the top choices. Unlike in 2014, no herds used category I products for disease prevention in 2020. While this finding could be due to additional veterinary oversight due to MIAs being prescription only, to definitively answer this question a targeted study would be needed.
In the previous 2014 study, 23% of cow-calf producers had reported the use of dart guns (18%), crossbows (10%), and pole syringes (1%) to treat on pasture where handling facilities were not available [6]. In the present study, the use of remote delivery systems was higher for the use of dart guns (31%) and pole syringes (9%). The use of crossbows (12%) was unchanged. Regardless of the remote delivery system used, less than 10% of treatments were delivered in this manner. Interestingly, there was a regional preference between Western and Eastern Canada with dart guns being significantly more commonly used in herds from the west. The 2017 USDA beef study [8] reported the use of pneumatic darts in 16% of medium-sized operations and 33% of large operations which was consistent with what was observed in Canada.
The main challenge with the use of remote delivery systems is that the total volume delivered by pneumatic darts is often less than 10 mL. This substantially limits the treatment options and typically results in the use of newer MIAs because often newer MIAs have a lower volume per unit of body weight dose [8]. Tulathromycin and tilmicosin were the two most-popular choices for pneumatic drug delivery in the 2017 USDA beef study [8]. The lower volume requirement might, in part, be why there has been an increase in the use of macrolides for treating lameness in cows. Lameness in cows is most common on summer pasture where access to handling facilities is limited and remote delivery tools can be a practical option for treatment.
While several herd attribute and management variables were examined, only a very small number were associated with the likelihood that more than 5% of animals treated for any of the three most-common reasons. The only factor significantly associated with the risk of treating more than 5% of nursing calves for respiratory disease in the final multivariable model was the use of AI. The reason for this association could be that, typically, calves are separated from their mother and comingled with other calves while producers process the cows, temporarily increasing animal stress and density, thereby potentially increasing the risk of pathogen transmission. This finding is consistent with previous reports from other studies [13,15,16] that identified the increasing number of times cow-calf pairs were gathered before turning out to summer pasture and estrus synchronization as risk factors for respiratory disease in calves before weaning.
Throughout the study, differences between Western and Eastern Canada were noted, including a higher frequency of western herds reporting AMU in nursing calves, weaned calves, and bulls. The variation in approaches to AMU could be due to differences in herd attributes and management practices. There were clear differences in herd size between eastern and western herds. Eastern herds reported a median of 75 bred cows and heifers whereas western herds had a median of 150 bred cows and heifers. Larger herds often require different management and could encounter different disease pressures than smaller herds. Herds from Western Canada were also more likely to maintain both individual-level records than herds located in Eastern Canada possibly due to less of an economic incentive for record maintenance in smaller operations [17]. Less record-keeping could mean more reliance on recall, potentially contributing to under-reporting.
Other studies have also reported differences between Western and Eastern Canadian cow-calf herds. In a recent summary of management practices, interventions near the time of birth were higher in eastern herds than in western herds [18]. Administration of intranasal respiratory vaccines and vitamins, and adequate colostrum intake were more common in Eastern Canada; however, overall herd vaccination levels were slightly lower in eastern herds [18]. While this recent BCRC report shows a greater uptake of these practices in Eastern Canada, a 2013 publication contains contradictory information and indicates that higher adoption of these practices was seen in Western Canadian herds, with the exclusion of overall vaccination levels for which similar data were reported [19].
A limitation of this study and all survey studies is recall bias. In this study, most producers reported maintaining herd (64%) and/or individual animal-treatment records (79%). The utilization of receipts and other sources of information was also encouraged to assist in the completion of the survey. However, historically, the quality/extent of records maintained by cow-calf operations has been questioned [7,17]. To address the issue of recall bias, measures were implemented in this study that were targeted to situations where records were not available or incomplete. For example, surveys were administered to producers in July 2020 immediately following calving season when most AMU is typically reported, the target time frame for the study, ensuring that the information if not recorded would still be relatively fresh in the producers' minds. Additionally, producers were provided with an "antimicrobial handbook," which included all licensed antimicrobials organized by route of administration, common names, generic names, and colour photographs of the product to aid recall regarding specific product use. Finally, when answers were unclear, participants were contacted by study personnel for clarification. Errors in reporting the choices of drugs used for particular diseases were expected to be relatively infrequent given the combined aids to recall, as were reports of the frequency of drug use. Rather than asking producers for specific treatment rates, producers were simply asked to report whether a specific product was used for a few animals (<5%), some animals (5-30%), many animals (30-70%), or most animals (>70%).
Social pressures including increased public scrutiny and stigmas in the beef industry around the use of antimicrobials in food production could have led to inaccurate data being provided by producers due to a perceived negative image or consequences related to what could be deemed "unnecessary use". However, the impact of any hesitation to report use was expected to be minimal, given the need for a veterinary prescription for MIA, the commitment of the producers to completing surveys for this network over many years, the reporting of a range of antimicrobials, including very high importance category I products, and consistency of results across surveys, including surveys before AMU in the livestock industry was a target of public scrutiny.
Since the herds enrolled in the study were participants in a longitudinal surveillance network, the sample population is potentially representative of relatively more progressive, intensively managed herds than might be observed in a random sample. While the study gives insight into AMU practices on Canadian cow-calf operations, the data reflects a segment of Canadian herds of similar size under comparable management.
The percentage of herds and herd sizes enrolled from each province was, however, representative of all herds within Western and Eastern Canada. The herd sizes included in this study represent the cows from greater than 75% of all beef herds in Canada reported to the 2021 Agriculture Census [20]. Within Western Canada, the proportion of herds from each province was reflective of the 2021 Agriculture Census data reflecting the numbers of beef cows: Alberta 44%, Saskatchewan 30%, Manitoba 11%, and British Columbia 5% [20]. The proportion of herds from Eastern Canada in the present study was higher than that suggested by the distribution of beef cow-calf herds in Canada, Ontario 6%, Quebec 3%, and Maritimes 1% [20], to ensure there were sufficient numbers to provide information to regional producer groups.
Opportunities exist for improvement in the use of antimicrobials in cow-calf operations. The proportion of herds that used any antimicrobial at least once in more than 30% of either cows or nursing calves was relatively small at only 12%. By better understanding the reasons for use in these herds, it might be possible to target intervention strategies to address some of the primary reasons for use. The most apparent target for use reduction is the 12% of herds that reported the use of MIA for disease prevention in nursing calves.

Conclusions
The use of antimicrobials for the treatment of lameness was very common but is not necessary or beneficial in all cases. Noninfectious causes are generally responsible for most lameness in beef cattle [21]. For example, sand cracks are a known noninfectious cause of lameness in cattle. Studies have found that the prevalence of sand cracks in some Western Canadian herds resulted in up to 60% of the herd being affected [22]. Treatment of sand cracks with antimicrobials is typically not warranted.
Furthermore, the use of antimicrobials for the treatment of diarrhea in calves is another area where it might be possible to implement additional stewardship practices. Viruses can cause diarrhea in calves and viral causes of diarrhea will not respond to antimicrobials; however, it is challenging to differentiate viral and bacterial causes of diarrhea [23,24]. Veterinarians, producers, researchers, and the industry need to continue to work together to develop appropriate treatment guidelines and protocols and enhance biosecurity measures and vaccine uptake to help further reduce AMU in cow-calf herds. Some treatment guidelines and decision trees have been developed and are available to Canadian veterinarians through the Canadian Veterinary Medical Association [25] but at the time of this report, the decision tree was not yet available to producers. Future studies will be critical to inform stewardship protocols to meet the needs of this evolving industry.
In summary, there were few changes between 2014 and the 2020 AMU data presented in the current study. While there was little use of category I antimicrobials of very-high importance to human health in cow-calf herds, macrolide use did increase slightly between 2014 and 2020 in Western Canadian herds for respiratory disease in nursing calves and lameness in cows and bulls. Other regional differences were identified in the frequency of AMU for respiratory disease in nursing calves, the types of products used for treatment, and the use of remote delivery systems. The relative importance of different reasons for treatment is similar to the most recent surveillance report from the United States. However, the proportion of herds reporting use in the current study was higher than that for similar medium-sized herds in the US report.
Author Contributions: C.L.W. was responsible for the conceptualization, development of methodology, supervision, and project administration. C.L.W. and J.D.F. were responsible for the investigation, formal analysis, and data curation. J.D.F. prepared the original draft of the manuscript. C.L.W., S.P.G., N.E. and J.R.C. reviewed and edited the manuscript. All authors have read and agreed to the published version of the manuscript. Informed Consent Statement: Informed consent was obtained from all study participants.
Data Availability Statement: Data can not be shared except as in the aggregate form provided in the supplementary tables due to agreements with study participants.

Conflicts of Interest:
The authors declare no conflict of interest.