Availability and estimates of veterinary antimicrobial use in British Columbia

Abstract

The amount of antimicrobial use is a significant selection pressure that alters the frequency of antimicrobial resistance. This paper summarizes attempts to estimate the weight of antimicrobial purchases in British Columbia for use in animals. The data reported here do not capture all sources of veterinary antimicrobial use in British Columbia. This paper highlights how information deficits on veterinary drug use complicate the development of an evidence-based policy framework for combating antimicrobial resistance.

There is mounting evidence that the use of antimicrobials in veterinary medicine contributes to the occurrence of antimicrobial resistant bacterial infections in human populations (1–4). As a result, the World Health Organization (WHO), a Health Canada advisory committee, and the British Columbia (BC) Ministry of Health have recommended that surveillance systems of veterinary antimicrobial use, particularly in food animals, be established by government agencies as an important component of risk management strategies to control the emergence and spread of antimicrobial resistant human pathogens (5–7). A recent WHO consultation also recommended that data in kilograms of active ingredient should be collected and reported on an annual basis and that information, such as species, production class, routes of administration, and purpose of use (therapeutic, prophylactic, and growth promotion), should be included in lactic, antimicrobial surveillance systems (8).

In BC, a working group comprised of government staff, physicians, and veterinarians has begun to identify research, policy, and practice priorities in order to control the emergence and spread of antimicrobial resistant human pathogens (7). The role of animal drug usage in this province is part of this working group’s concern. The objectives of this latter project were to describe the routes through which pet owners and livestock producers purchased antimicrobials in BC, to summarize the legislation related to animal drug access and distribution in BC, and to attempt to collect data to estimate the amount of antimicrobials purchased in BC for animal use.

In BC, pet owners and livestock producers can access antimicrobials through a number of different sources. Schedule F Part I (prescription) and Schedule F Part II (nonprescription) pharmaceuticals, as classified under the federal Food and Drug Act and administered by Health Canada, can be purchased directly from a veterinarian. Under the British Columbia Veterinarians Act, veterinarians are required to have a valid veterinary-client-patient relationship (VCPR) in order to sell prescription drugs to producers and pet owners, but they can sell nonprescription drugs without a VCPR. Many veterinary clinics are computerized and keep track of drug sales and purchases, generally for in-house accounting and inventory purposes. Veterinarians are not required to submit drug-purchase records to any regulatory body. However, they are required to document all veterinary medical services they render in a manner that sufficiently describes the findings, diagnosis, and treatment provided.

The provincial Veterinary Drug and Medicated Feed Regulation, administered by the BC Ministry of Agriculture, Food and Fisheries (BCMAFF), licenses lay premises and commercial feed mills to dispense nonprescription veterinary drugs. Commercial mills sell nonprescription medicated feeds that are listed in Schedule A Table 1, as authorized by the federal Medicating Ingredients Brochure (Feeds Act), directly to producers. Feed mills may also prepare and sell prescription medicated feeds with a prescription from a veterinarian licensed to practice in BC.

Licensed lay drug outlets and licensed feed mills are required under the authority of the BC Veterinary Drug and Medicated Feeds Regulation to submit records of their drug purchases annually to BCMAFF. However, due to staffing shortages, the records have not been compiled and analyzed, except for reviews of antimicrobial use in salmon farms in 1996 by the BC Salmon Aquaculture Review Technical Advisory Team. The BCMAFF is currently implementing a new medicated feed database program to collect from feed mills more detailed information, such as species, production class, dose and duration.

Livestock producers also purchase antimicrobials directly from manufacturers (6,9); however, there are no separate publicly accessible records of these sales in terms of the types or weight of product sold in BC. Active pharmaceutical ingredients (APIs), which are bulk, pharmaceutically active substances used in the formulation of drugs in dosage form, may be manufactured and sold for animal use in dosage form (granules, powders, and pellets) (9). There are few regulatory measures (no labelling standards, no Drug Identification Numbers (DIN) registration, no established withdrawal periods and residue limits) in place to control the importation or sale of APIs in Canada. As a result, until recently, some manufacturers have been selling bulk APIs for use as veterinary drugs to producers, pharmacists, feed mill operators, and veterinarians (9). Details of the extent of this practice are unavailable. The Advisory Committee on Animal Uses of Antimicrobials and Impact on Resistance and Human Health has recommended that all antimicrobials used in food animals, including APIs, should be evaluated, registered, and assigned a DIN to prevent the direct use of APIs in food animals (6).

Pet owners and livestock producers may also purchase Schedule F Part II (nonprescription) drugs from pharmacists without a veterinarian’s prescription and Schedule F Part I (prescription) drugs with a prescription (prescription) drugs with a prescription from their veterinarian. It is not known how many animal owners purchase veterinary drugs from pharmacists. The BC College of Pharmacists’ ‘Pharmanet’ program links all pharmacies in BC. For animal prescriptions, the pharmacist uses the personal health number of the owner but enters a different prescription identification number to indicate that it is a veterinary prescription. This information is no longer accessible without charge, but it can be purchased from Pharmanet.

Other sources of antimicrobial drugs for pet owners and livestock producers include purchases from retail veterinary drug outlets and veterinarians operating outside the province. This is a source for border communities or for producers that either purchase drugs for lower ties prices out-of-province or have veterinary consultants working out-of-province. It is not known what percentage of pharmaceutical sales these out-of-province purchases represent. The Canadian Animal Health Institute chases (CAHI) noted in its March 1998 report Importation and use of non-approved animal health pharmaceuticals, that livestock producers may import from outside Canada “a single treatment size or a 3-month supply of a Schedule F drug” by using the ‘own-use’ provision under the federal Food and Drug Act (C.01.045). It is not known how (C.01.045). It is not known how many producers are making use of this provision to purchase drugs for animal use. One of the major recommendations of the Advisory Committee on Animal Uses mendations of Antimicrobials and Impact on Resistance and Human Health is to put a stop to this “own-use” loophole (6).

Ultimately, only 2 sources of animal drug use information were available for this study: national sales data compiled by a pharmaceutical market research company and handwritten provincial lay drug outlet and feed mill purchase records from the BCMAFF for 1997 and 1998. However, confidentiality concerns by the former prevented publication of these data. This paper reports only vented on the findings from BCMAFF’s data.

Reports submitted to BCMAFF were generally handwritten and contained only information on lay drug written outlets’ purchases from their suppliers. It was not possible from these records to determine the species in which the drugs were used, what proportions were being used under veterinary prescription or supervision, or the dose and duration of drug use. Eighteen classes of antimicrobials were purchased by licensed lay drug outlets microbials or feed mills in BC in 1997 and 1998 (Table 1).

Table 1.

Antimicrobial classes purchased by licensed veterinary drug outlets and feed mills in British Columbia in 1997 and 1998

Aminoglycoside	Florfenicola	Novobiocin
Aminocyclitol	Dimetridazole	Penicillin
Arsenicals	Folic acid antagonist	Polymixin
Bacitracin	Lincosamide	Sulphonamide
Bambermycina	Macrolide	Tetracycline
Carbadox	Nitrofuranb	Virginiamycin

^a Only available to licensed feed mills through veterinary prescription. All the florfenicol reported here was formulated for use in aquaculture

^b Licensed lay drug outlets can only sell products containing nitrofurans for use in horses. All products reported here by lay drug outlets were in topical form and were not labelled for use in food-producing animals

The weight of antimicrobials purchased by lay drug outlets and feed mills in BC is shown in Table 2. Procaine penicillin G was responsible for 99.8% of penicillin purchases in 1997 and for 99.7% in 1998, while oxytetracycline accounted for 98.0% and 97.8% of tetracycline purchases in 1997 and 1998, respectively. The vast majority (99.2%) of antimicrobials purchased were oral preparations.

Table 2.

Summary of antimicrobial purchases (active compound) by licensed veterinary drug outlets and feed mills in British Columbia for 1997 and 1998

Antimicrobial class	1997 [Tonnes (%)]	1998 [Tonnes (%)]
Tetracycline	18.2 (56.9)	29.0 (63.5)
Virginiamycin	3.9 (12.2)	2.5 (5.5)
Penicillin	3.6 (11.3)	3.4 (7.4)
Bacitracin	3.5 (10.9)	5.3 (11.6)
Sulphonamide	1.0 (3.1)	1.7 (3.7)
Others	1.8 (5.6)	3.8 (8.3)
Total	32.0 (100.0)	45.7 (100.0)

There were 3 significant difficulties encountered in the collection and evaluation of the BCMAFF data. First, data were recorded as the weight of product purchased rather than weight of active antimicrobial compound. Extensive efforts were required to make these conversions. The amount of active ingredient for each product, which was required to calculate the total amount of active drug purchased, was found in either the Compendium of Veterinary Products (10) or on manufacturers’ websites. Second, the records were in the form of handwritten reports that had to be found, read, interpreted, and compiled into a database. Third, missing data or data that were unavailable severely limited the capacity of this project to quantify all of the antimicrobials sold in BC for animal use.

The data presented here underestimate total drug use in the province. Differences in drug purchases between 1997 (32 metric tonnes) and 1998 (45 metric tonnes) could be accounted for by records that were missing from 2 major feed mills in 1997. While Ministry records indicate that the data were received, they could not be found for this analysis. Other missing data include veterinary purchases through group-buying organizations, purchases from pharmacists, out-of-province or international purchases, and direct sales from manufacturers and veterinarians. These data were either unavailable due to prohibitive costs, not available for public review, or not approved for publication.

This project has highlighted some of the problems that must be addressed before a provincial or national system for monitoring antimicrobial use in animals can be established. The primary obstacles that prevented a full accounting of animal antimicrobial use were a lack of an appropriate regulatory mandate to collect data from all relevant sources and insufficient personnel and resources to collect, store, and analyze data. There were no available data on how these drugs were used (dose, duration, conditions, species) apart from the labeled indications for specific products. The latter can be assumed to be a poor estimate of use due to the common practice of off-label drug use in Canadian veterinary practice (6).

As a surveillance tool, the weight of antimicrobials sold provides a rough measure of overall use for trend analyses. However, it is not necessarily an appropriate measure of the magnitude of selective pressure being exerted on bacterial communities, because different antimicrobials are not equal in their biological activity per unit weight. For example, during an environmental review of the BC salmon farming industry, it was shown that BC required 156 g of antibiotics to produce 1 metric tonne of Atlantic salmon, while Norway required only 100 g per metric tonne (11). Quinolones were more often used in Norway (84%) and were delivered at a dose of 5 to 12 mg/kg bodyweight (BW), whereas oxytetracycline, which accounted for 90% of drug use in BC and 3% in Norway, was prescribed at a dose of 75 to 100 mg/kg BW, thus partially explaining differences in the weight of drugs used per tonne of fish (11). Alternative measures of the selection pressures being exerted by veterinary antimicrobials are required to allow for biologically meaningful surveillance. The WHO Collaborating Centre for Drug Statistics Methodology in Norway has just begun the process of developing species-specific measures, Daily Defined Doses (DDD), for animal antimicrobial use.

The surveillance of veterinary antimicrobial usage in BC, and elsewhere, is important for the development and maintenance of relevant policies and practices that optimize antimicrobial use and counter antimicrobial resistance. Only a small portion of animal antimicrobial use could be accounted for in BC; therefore, BC could not meet the antimicrobial use surveillance recommendations put forth by the BC Ministry of Health, a Health Canada advisory committee, or WHO. It can be anticipated that future legislation will affect the availability and use of antimicrobial drugs in animals. The animal-health community, including veterinary agencies, industry, and federal and provincial governments, would be well advised to remedy many of the information gaps identified here so as to ensure the development of evidence-based drug use policies and antimicrobial resistance based mitigation strategies.