A comparison of antimicrobial usage in human and veterinary medicine in France from 1999 to 2005

Journal of Antimicrobial Chemotherapy (2008) 62, 617 625 doi:10.1093/jac/dkn213 Advance Access publication 19 May 2008 A comparison of antimicrobial usage in human and veterinary medicine in France from 1999 to 2005 Gérard Moulin 1, Philippe Cavalié 2, Isabelle Pellanne 2, Anne Chevance 1, Arlette Laval 3, Yves Millemann 4, Pierre Colin 5 and Claire Chauvin 6 * on behalf of the Antimicrobial Resistance ad hoc Group of the French Food Safety Agency 1 AFSSA ANMV French Agency for Veterinary Medicinal Products, Fougères, France; 2 AFSSAPS French Health Products Safety Agency, St Denis, France; 3 National Veterinary School of Nantes, Nantes, France; 4 National Veterinary School of Alfort, Maisons-Alfort, France; 5 ESMISAB, University of Brest, Plouzané, France; 6 AFSSA French Food Safety Agency, Ploufragan, France Received 10 January 2008; returned 22 February 2008; revised 21 April 2008; accepted 27 April 2008 Objectives: The antimicrobials allowed and amounts sold in veterinary and human medicine in France were compared to see if the same antimicrobial drugs are used in veterinary and human medicine, and to the same extent. Methods: Registers of all approved antimicrobial commercial products, kept by the French Agency for Veterinary Medicinal Products (AFSSA ANMV) for animals and the French Health Products Safety Agency (AFSSAPS) for humans, were compared to determine whether the same antimicrobials were approved in 2007 for use in both human and animal populations. Sales data were collected from pharmaceutical companies between 1999 and 2005 by the AFSSA ANMV and AFSSAPS. Usage of the different antimicrobial anatomical therapeutic chemical (ATC) classes in human and veterinary medicines was recorded. Data were expressed in tonnes of active ingredients and were then related to the animal and human biomasses to compare usages expressed in mg/kg. Results: All antimicrobial ATC classes were used in both human and veterinary medicine. Tetracyclines accounted for the most sales in veterinary medicine. b-lactams predominated in human medicine. A decrease in the amounts consumed by both human and animal populations was observed during the study. In 2005, 760 tonnes were used in human medicine and 1320 tonnes in veterinary medicine, corresponding to 199 and 84 mg/kg of live weight in human and animal populations, respectively. Conclusions: The same antimicrobial drugs were used in human and veterinary medicines but the quantitative patterns of use were different. Expression of antimicrobial usage is a key point to address when comparing usage trends. Keywords: anti-infectives, drug use surveillance, data collection, measurement unit Introduction Surveillance of antimicrobial usage has been widely recommended internationally in recent decades and data collection promoted in all sectors of use (human medicine, veterinary medicine and agriculture). 1,2 In respect of these recommendations, two independently managed systems to monitor antimicrobials sales in France were implemented in human 3 and veterinary medicines 4 (agricultural usage being banned in France). A first joint quantitative analysis of rough data pertaining to animal and human antimicrobials usage was initiated in the... *Correspondence address. AFSSA French Food Safety Agency, LERAP, BP 53, Zoopole, 22440 Ploufragan, France. Tel: þ33-2-96-01-62-66; Fax: þ33-2-96-01-62-95; E-mail: c.chauvin@afssa.fr Members are listed in the Acknowledgements section.... 617 # The Author 2008. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved. For Permissions, please e-mail: journals.permissions@oxfordjournals.org

Moulin et al. European Union in the late 1990s. 5 The aim was to compare the tonnage of active ingredients used in European human and animal populations and thereby indirectly estimate the respective contributions to antimicrobial resistance. 6 In the following years, comparisons in some countries such as Norway 7 and the United Kingdom 8 were based on the volume of antimicrobials delivered. More recently, as part of the benefit/risk analysis of antimicrobial use in veterinary medicine, international organizations initiated a joint qualitative analysis of antimicrobial usage in human and veterinary medicines, based on the critical antimicrobial concept. 9 In 2003, the French Food Safety Agency was asked to prepare a report on the veterinary use of antimicrobials, antimicrobial resistance and its public health impact, 10 in which antimicrobial usage in humans and animals was measured and compared for the first time. The updated and detailed results of this comparison of human and veterinary use of antimicrobials in France are presented here. The antimicrobials authorized in veterinary and human medicine were checked and quantities used from 1999 to 2005 were compared to see if the arsenal and patterns of use differed across populations or during the period. Materials and methods Inventory of antimicrobials approved in human and veterinary medicine Antimicrobials for use in humans and animals in France are only available as prescription drugs. To be prescribed and delivered by pharmacies and veterinarians, commercial antimicrobial products must have previously obtained authorization. A register of all approved commercial formulations is kept and regularly updated by national agencies, namely the French Agency for Veterinary Medicinal Products (AFSSA ANMV) for animals and the French Health Products Safety Agency (AFSSAPS) for humans. The antimicrobial classes approved in human and veterinary medicines, respectively, were checked against the Anatomical Therapeutic Chemical (ATC) classification of human products, 11 and the ATC veterinary classification (ATCvet) for veterinary products. 12 All ATC and ATCvet groups have been considered (i.e. groups for systemic use, alimentary tract, genitourinary system, dermatological use and sensory organs). Subgroups were considered within some ATC classes, as detailed in Table 1. Whenever an antimicrobial was approved for human or animal usage in 2007, its whole class or subgroup has been ticked as approved for human or animal usage. Sales of antimicrobials We used data collected from systems already existing in France to monitor the annual sales of veterinary and human antimicrobials. 3,4 Antimicrobials sales and antimicrobials usage or consumption are therefore intended to be considered as synonyms in this study. Sales of antimicrobials for veterinary use are monitored according to the OIE guidelines on Monitoring of Antimicrobials, 13 which recommend that Member Countries monitor the amounts of antimicrobials used in animal husbandry to evaluate usage patterns by antimicrobial class. This monitoring programme is coordinated on a voluntary basis by the AFSSA ANMV, in collaboration with the French Veterinary Medicine Industry association. It has been carried out, with financial support from the French Ministry of Agriculture, since 1999. The methodology is based on a questionnaire sent to every applicant who has registered a veterinary medicinal product containing antimicrobials. Every December, a letter is sent from the AFSSA ANMV to the marketing authorization holder requesting details of the number of units of each commercial formulation of each veterinary medicinal product sold. These figures must be supplied for the period between 1 January and 31 December of the year in question. The sales of each formulation are cross-referenced with data in the AFSSA ANMV (qualitative and quantitative composition, pharmaceutical form and target species). French data on antimicrobial use in human outpatients and inpatients were collected to meet national and European requirements. The follow-up of antimicrobial usage is one of the indicators adopted by the French Public Health Law to measure the development of collective exposure to antimicrobials. France also contributes to the ESAC project (European Surveillance of Antimicrobial Consumption), created in 2001 with the aim of collecting comparable and reliable data concerning antimicrobial use in Europe. 14 In both cases, the same methodology is used to measure these consumption levels and their trends. The French processed data are independent and derived from annual statements that pharmaceutical companies have to send to AFSSAPS for a tax based upon turnovers. These comprehensive data provide accurate sales information for each formulation and distinguish between in- and outpatient sectors. They can thus be used to monitor in- and outpatient antimicrobial usage and include all the information (i.e. dose, packaging size and annual sales of each packaging) required to calculate the amount of defined daily doses (DDDs) or weight of active ingredient in tonnes. All routes of administration, including topical ones, were considered in this study. The antimicrobial sales data collected from 1999 to 2005 were considered. To manage data from both monitoring systems, the ATC and ATCvet drug classifications were used, in the same way as for the qualitative comparison of approved antimicrobials. The unit of measurement chosen to express the amounts of active ingredient sold for both human and veterinary use was tonnage. Antimicrobials expressed in International Units were converted to mass of active ingredient, according to the WHO standard values. With veterinary antimicrobials it is difficult to express the results for each species, as the same veterinary medicinal product can be indicated for several species and the sales per species can only be estimated. 4 For the purpose of this study, all animal species were considered together but a distinction was made between pets and farm animals whenever possible. Human and animal antimicrobial usages were compared by expressing the data as the quantity of active component sold in relation to the biomass liable to be treated. This can be considered as an indicator of selection pressure. The biomass denominator was calculated annually from a population census combined with live weight estimates. In veterinary medicine, all potentially treated animals were considered, including pets and minor species. Annual detailed national inventories from the French Ministry of Agriculture were used for major species such as swine, cattle, sheep, goats, poultry species and horses, 15 professional declarations were used for farmed fish and game, and survey results were considered for pet populations. 16 For animals slaughtered during the year, the amount of live weight slaughtered was considered. For animals in place at the end of the year, such as sows, boars and lactating cows, calculation of the biomass liable 618

Animal and human antimicrobial usages in France Table 1. Comparison of antimicrobial classes and subgroups approved for human and veterinary medicine in France in 2007 Active substances approved in France in 2007 Antimicrobial classes human medicine veterinary medicine Aminoglycosides yes yes Amphenicols yes yes b-lactams yes yes b-lactamase sensitive penicillins yes yes penicillin G yes yes penicillin V yes yes b-lactamase resistant penicillins yes yes penicillins M yes yes penicillins with extended spectrum yes yes penicillins A yes yes carboxypenicillins yes no ureidopenicillins yes no Other b-lactams yes yes first-generation cephalosporins yes yes second-generation cephalosporins yes yes third-generation cephalosporins yes yes fourth-generation cephalosporins yes yes monobactams yes no carbapenems yes no Fluoroquinolones yes yes Other quinolones yes yes Macrolides, lincosamides, streptogramins yes yes macrolides a yes yes a lincosamides yes yes streptogramins yes no pleuromutilins no yes Nitrofuran derivatives yes yes Polymixins yes yes Sulfonamides yes yes Tetracyclines yes yes Trimethoprims yes yes Other antimicrobials glycopeptides yes no imidazole derivatives yes yes anti-tuberculous b yes yes c fusidic acid yes yes bacitracin yes yes clofazimine yes no dapsone yes no daptomycin yes no fosfomycin yes no novobiocin no yes mupirocin yes no linezolid yes no thiostrepton no yes a The new macrolides drugs named ketolides are not approved in veterinary medicine. b Including rifampicin and rifamycin, also used for indications other than tuberculosis. c One molecule (rifamycin) belonging to the antituberculous antimicrobial class is used in veterinary medicine as an intramammary suspension. to be treated was based on the numbers combined with live weight. In human medicine, data from the annual census of the French population by age and sex (from the National Institute for Statistics and Economic Studies) 17 were combined with the corresponding average body weights per age and sex obtained from health insurance statistics. 18 619

Moulin et al. Results Comparison of antimicrobials approved in human and veterinary medicine Results from the comparison of approved antimicrobials are given in Table 1. No antimicrobial class is specific to veterinary or human use, only subclasses or particular antimicrobials. The antimicrobials used only in human medicine are carboxypenicillins, ureidopenicillins, monobactams, carbapenems, ketolides and glycopeptides, together with other antimicrobials such as clofazimine, dapsone, daptomycin and fosfomycin. The antimicrobials sold for use only in veterinary medicine are pleuromutilins, thiostrepton and novobiocin. Usage in veterinary medicine Figure 1 presents the sales of antimicrobial classes recorded in veterinary medicine between 1999 and 2005 [from data provided in Table S1; available as Supplementary data at JAC Online (http://jac.oxfordjournals.org/)]. The annual weight of antimicrobials sold in veterinary medicine was 1300 tonnes. Four antimicrobial classes (tetracyclines, sulphonamides/trimethoprim, b-lactams and aminoglycosides) accounted for more than 80% of the antimicrobials sold. Tetracyclines alone represented around half of all sales. The antimicrobials belonging to newer classes represented relatively low amounts (cephalosporins: 0.64% in 2005, fluoroquinolones: 0.33%), but increased considerably between 1999 and 2005. Cephalosporins sales rose by 38.4% during monitoring, and fluoroquinolones sales by 31.6%. The oral route was the main administration route, accounting for more than 88% of sales, the estimated sales for parenteral use being 10.5%. According to the applicants declarations, at least 92% of antimicrobial tonnage was intended for foodproducing animals. Almost 64% of cephalosporins were intended for pets. Usage in human medicine Figure 2 presents the sales of antimicrobial classes in human medicine between 1999 and 2005 [from data provided in Table S2; available as Supplementary data at JAC Online (http:// jac.oxfordjournals.org/)]. The breakdown of consumption levels in tonnes reveals the prominence of b-lactams, which constituted more than 50% of the total consumption and even rose slightly during the study period. Cephalosporin consumption dropped, however (which accounts for the total consumption of other b-lactams in outpatients). Formulations for local application accounted for a minor part of the total consumption (6.9% in 1999 and 7.4% in 2005). Consumption by outpatients accounted for 88% on average of the total human consumption. Comparison of sales data The contribution of human and veterinary medicines to total use of the different antimicrobial classes recorded in France in 2005 are represented in Figure 3. Some classes were used almost exclusively either in animals (aminoglycosides, amphenicols, polymyxins, tetracyclines) or in humans (nitrofuran derivatives and other antimicrobials). Tetracyclines and sulphonamides/trimethoprim accounted for the highest sales in veterinary medicine (50.4% and 18.8%, respectively, of total veterinary antimicrobial sales in 2005). These two classes represented just 1.7% and 2.9%, respectively, of the total antimicrobial sales in human medicine. Aminoglycosides and polymyxins also accounted for significant sales in veterinary medicines (5.9% and 4.9%, respectively, of total veterinary antimicrobial sales in 2005), whereas they represented just 0.2% of total sales in human medicine. b-lactams accounted for the highest sales in human medicine (51.6% of total sales) but only 8.3% of total sales in veterinary medicine. Macrolide sales in human medicine were quite high at 14.8% (9% of total antimicrobial sales in veterinary medicine). Cephalosporins sold for human use accounted for 88.2% of this class sold in France for human and veterinary medicine. Fluoroquinolones for human use accounted for 89.6% of the antimicrobials of this class sold in France for human and veterinary medicine. In contrast, the other quinolones sold in veterinary medicine accounted for 82.1% of the antimicrobials of this class sold in France for human and veterinary medicine. The veterinary sector accounted for 60% of the total amount of antimicrobials sold in France in 2005 and the human sector to 40%. However, the contributions of the two sectors differed with regard to the total animal and human biomasses calculated between 1999 and 2005 ( presented in Table 2). In 2005, the calculated ratio of mg of active ingredient sold per kg of live weight of potential users was 2.4 times higher in human medicine than in veterinary medicine. Despite similar sales trends in both sectors (Table 2), with a decrease between 2000 and 2004 and followed by a slight increase in 2005, changes in biomass resulted in a difference in ratio trends. Human biomass increased during the study period, whereas animal biomass has decreased since 2001, leading to a relatively stable ratio expressed in mg/kg of live weight. Discussion This study was based on existing monitoring systems with no modifications or complementary data acquisition. Comparison of antimicrobial drugs allowed was performed using data gathered from national agencies registers for the year 2007. As no changes have occurred, the qualitative comparison can be considered to accurately reflect the therapeutic antimicrobial veterinary and human arsenal currently in use. The quantitative comparison was based on antimicrobial sales surveys in human and veterinary medicine. Sales were interpreted as representing antimicrobial consumption. Nevertheless, annual sales do not necessarily correspond to annual usage, especially when reforms interfere with drug distribution, as may have been the case in human medicine during 2005. Indeed, a reform of the distribution margins system in France (which came into effect in January 2006) prompted the mass purchase of generics, including antimicrobials, by chemists at the end of 2005. This resulted in a potential discrepancy between sales of antimicrobials and actual human consumption in 2005. A small proportion of antimicrobials registered for human usage may also have been prescribed by veterinarians to treat pets and might therefore be misclassified and erroneously attributed to human usage. Although the sales data for human and veterinary antimicrobials were acquired independently, the methodologies were similar. The survey for human antimicrobial usage was based on official 620

Animal and human antimicrobial usages in France Antimicrobial consumption (in tonnes) 700 650 600 550 500 450 400 350 300 250 200 150 100 50 0 Aminoglycosides Amphenicols b-lactams Other b-lactams Fluoroquinolones Other quinolones Macrolides, lincosamides and streptogramins Figure 1. Sales trends, in tonnes, from 1999 to 2005 of different antimicrobial classes (according to the nomenclature detailed in Table 1) used in veterinary medicines (AFSSA ANMV). Years from 1999 to 2005 are represented from the left to the right, with 1999 being the darkest bars to 2005 being the lightest bars. Antimicrobial consumption (in tonnes) 500 450 400 350 300 250 200 150 100 50 0 Aminoglycosides Amphenicols b-lactams Other b-lactams Fluoroquinolones Other quinolones Macrolides, lincosamides and streptogramins Nitrofuran derivatives Nitrofuran derivatives Polymyxins Polymyxins Sulfonamides and trimethoprim Sulfonamides and trimethoprim Tetracyclines Tetracyclines Other antimicrobials Other antimicrobials Figure 2. Sales trends, in tonnes, from 1999 to 2005 of different antimicrobial classes (according to the nomenclature detailed in Table 1) used in human medicine (AFSSAPS). Years from 1999 to 2005 are represented from the left to the right, with 1999 being the darkest bars to 2005 being the lightest bars. sales declarations to AFSSAPS, whereas the survey of antimicrobials in animals was based on voluntary declarations to AFSSA from marketing authorization holders. Nevertheless, the voluntary declaration of antimicrobial sales in veterinary medicine is exhaustive and covers all medicinal products sold in the French market. The methods were harmonized to obtain comparable results, allowing for the antimicrobials to be grouped into the same classes, as different standards were usually used in the two surveys. All pharmaceutical forms and administration routes were included in both sectors, although topical forms are not usually included in human surveys. These harmonization efforts explain differences between these results and those in previously published reports. 4,19 Moreover antimicrobial human sales, usually expressed in DDD, were converted into tonnage to 621

Moulin et al. 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Total Aminoglycosides Amphenicols b-lactams Other b-lactams Fluoroquinolones Figure 3. Contributions to total sales recorded in France in 2005 of antimicrobials approved for human (in white) and animal (in grey) usage (using the class nomenclature detailed in Table 1). Table 2. Antimicrobial consumption and biomasses estimated in human and animal populations in France, from 1999 to 2005 Amount of antimicrobial drugs sold (tonnes) Other quinolones Macrolides, lincosamides and streptogramins Population body mass (tonnes) Nitrofuran derivatives Polymyxins Sulfonamides and trimethoprim Tetracyclines Other antimicrobials Antimicrobials sold related to biomass (mg/kg of live weight) human veterinary humans animals human veterinary 1999 896.20 1316.31 3 597 843 17 122 220 249.1 76.9 2000 861.08 1386.22 3 634 603 17 410 934 236.9 79.6 2001 845.18 1380.31 3 671 731 17 864 462 230.2 77.3 2002 809.44 1331.53 3 709 154 17 268 049 218.2 77.1 2003 755.90 1304.76 3 747 650 16 722 083 201.7 78.0 2004 716.24 1273.14 3 781 175 15 966 824 189.4 79.7 2005 759.67 1320.10 3 810 215 15 795 105 199.4 83.6 permit comparison with antimicrobials used in veterinary medicine. The DDD indicator, 11 now widely used in international studies of human drug consumption, was not adopted for the quantitative comparison of usages, because the veterinary consumption of antimicrobials cannot be converted into a number of DDDs. To date, no international standard corresponding to the human DDD has been set for veterinary medicine. 20 Moreover, this would have to be defined by species and consumption cannot be related to the different species with precision. Another standard, i.e. tonnage, was therefore chosen to express both human and veterinary antimicrobial consumption. The weight of active ingredient limits data interpretation as it does not reflect the number of treatments received by either animals or humans, due to large differences between antimicrobials in the dose prescribed. 21 An indirect estimate of usage intensity in animal and human populations was obtained by expressing the amount of antimicrobials sold annually in milligrams in relation to the calculated biomass of humans and animals for the corresponding year. The biomass calculation was based on census. Age and sex were taken into account in the human population to assign the correct average live weight to each population category. In the animal population, all species likely to be treated were considered, namely all farm and pet species, including minor species such as farm game. The heterogeneity of the probability of exposure to antimicrobials within the population was not included in the calculation. This has been well characterized in the human population, 22 but would have been difficult to quantify for each animal species. As a result, all kilograms of live weight were considered to present the same probability of being 622

Animal and human antimicrobial usages in France treated, thereby overestimating the biomass actually exposed to antimicrobials. The qualitative comparison of antimicrobials used in human and veterinary medicine shows that the great majority of antimicrobial classes are used in both sectors. But some antimicrobials used in both populations may be limited to a specific use in animals, e.g. chloramphenicol and nitrofuran derivatives are banned for veterinary use in food-producing animals. For antimicrobials in the antibiotics for treatment of tuberculosis group in the ACT and ATCvet classifications (J04AB and QJ04AB), only one product (rifamycin) is authorized and for intramammary use only (quoted QJ54AB). Nevertheless, an evident overlap exists between the human and veterinarian antimicrobial arsenal. This overlap is consistent with the recent analysis of human and veterinary arsenal based on the critical antibiotic concept. 9 The international organizations involved in the analysis stated that this overlap highlights the need to use all antimicrobials with prudence and also to obtain data on the extent of antimicrobial usage in animals so that priorities, in the risk assessment of non-human use of antimicrobials, can be ranked. 9 This approach should consider the quantitative use of antimicrobials in animals as well as criteria such as the perceived importance of antimicrobial use in humans. The veterinary usage trends and patterns were evaluated from the surveillance of antimicrobial usage implemented since 1999. Veterinary antimicrobial drug sales in total tonnage were globally stable with similar values in both 1999 and 2005. Slight fluctuations could be linked to factors such as changes in animal demography, epidemiological situation, appearance or disappearance of products on the market, cost of the products and actions for the prudent use of antimicrobials. The main variations during the observation period were for cephalosporins, fluoroquinolones and macrolides (sales increase of 38.4%, 31.6% and 28.1%, respectively) and for sulphonamides/trimethoprim and other quinolones (sales decrease of 18.6% and 17%, respectively). The increased sales of macrolides between 1999 and 2002 were probably brought about by the withdrawal of most antimicrobials used as growth promoters, as observed in other countries, 23 although this effect has tended to decrease since 2002. Sales trends in human medicine show a regular decrease. This highlights the impact of campaigns and initiatives in France for a few years now to prevent antimicrobial resistance and promote prudent use of these drugs. 24 Although not yet corroborated by prescription data, everything suggests that the two phenomena occurred simultaneously i.e. reduced antimicrobial consumption (antibacterial for systemic use, J01 ATC code), together with a partial shift from cephalosporins consumption (i.e. other b-lactam class) to broad-spectrum antimicrobials (b-lactams alone or in combination such as amoxicillin combined with a specific enzyme inhibitor). The apparent increase observed in 2005 does not reflect a reversal of tendency but mainly results from epidemiological factors (influenza-like diseases) and economic factors such as the reform of distribution margins which affected generic purchases and led to increased acquisition by chemists of antimicrobials such as amoxicillin, amoxicillin in combination with clavulanic acid, and so on. The proportion of topical forms rose slightly during the study period. This can be attributed to a slower decrease in consumption of topical forms than of other forms (i.e. antibacterial for systemic use, J01 ATC code). Withdrawal of marketing authorizations, after reassessment of the benefit/risk ratio, concerning formulations for local application (especially nose and throat) by the French Health Products Safety Agency in 2003 and mainly 2005, should have affected the global consumption of topical forms in subsequent years. The quantitative data show that veterinary sales exceeded human sales in latter years. In 2005, the total antimicrobial sales in veterinary medicine (1320 tonnes) were 1.7 times higher than the total antimicrobial sales in human medicine (760 tonnes), representing 63.5% of the two combined. However, the patterns of antimicrobials usage differed markedly between humans and animals. The main classes were different (i.e. tetracyclines and sulphonamides in veterinary medicine and b-lactams in human medicine), and sales of the newer classes of antimicrobials cephalosporins (quoted as other b-lactams) and fluoroquinolones were markedly higher in human medicine than in veterinary medicine. Furthermore, in addition to differences in patterns, the ratio between the total tonnage of antimicrobial sales and the potential weight to be treated in human and veterinary medicine was also different. The ratio in milligrams of antimicrobial per kilogram of weight per year was 220 mg/kg/ year for humans and 80 mg/kg/year for animals. As a result, humans potentially receive almost three times more antimicrobials than animals. Similar estimates [56 mg/kg of body weight (bw) used in animals and 241 mg/kg bw in humans] were obtained in a previous study comparing human and veterinary antimicrobial consumption in the European Union in 1997. 6 A detailed comparison of international studies remains difficult, given the different approaches and methodologies used. Antimicrobial sales surveys are conducted in various other European countries (Denmark, 25 Finland, 26 the Netherlands, 27 the UK, 8 Norway 28 and Sweden 29 ) using different methodologies for veterinary sales monitoring. The distinctions between antimicrobial classes are not necessarily the same (e.g. distinction between fluoroquinolones and other quinolones or between combinations), and topical forms or formulations strictly intended for pets are not always included. Moreover, human usage is rarely reported in weight of active ingredient, as the DDD unit is usually employed. Nevertheless, the main French observations have also been reported in other countries, such as the predominance of animal antimicrobial sales in total usage in tonnes which has been reported in Denmark 25 and the UK. 8 Similarly, differences between the patterns of human and animal usage have been constantly reported, with aminoglycosides, tetracyclines and sulphonamides being mainly used in animals and b-lactams in humans. 7,8 Usage patterns in human medicine have already been compared between European union countries as part of the ESAC programme. 30 Comparison of antimicrobials usage in veterinary medicine is more difficult as the term animal includes different species. The global animal usage of antimicrobials is influenced by the relative importance of the different species and their own patterns of consumption. Moreover, expressing antimicrobial use in terms of the quantity of active substance does not directly represent the number of treatments received by animals or humans. Data concerning usage in animals should therefore be considered as a compartment estimate in an antimicrobial resistance 31 or environmental risk analysis 32 context rather than as estimates of practices and selective pressure intensity. The interest of these surveys is also to reveal trends and evaluate the results of management policies set up to reduce antimicrobial use. The decrease in human usage 623

Moulin et al. can be considered to result from the promotion of a prudent and rational usage of antimicrobials. Efforts should now be pursued for the prudent and rational use of antimicrobials in both human and veterinary medicine. Acknowledgements The French Food Safety Agency expert group for Antimicrobial Resistance was chaired by Didier Guillemot (Institut Pasteur, Paris), coordinated by Corinne Danan (AFSSA, Maisons-Alfort) and its members were as follows: Hélène Aubry-Damon (AFSSA, Maisons-Alfort), Anne Bouchardon (AFSSA, Ploufragan), Anne Brisabois (AFSSA, Maisons-Alfort), Hubert Brugère (Ecole Nationale Vétérinaire, Toulouse), Elisabeth Chaslus-Dancla (INRA, Nouzilly), Claire Chauvin (AFSSA, Ploufragan), Pierre Colin (AFSSA, Brest), Henri Dabernat (Centre Hospitalier Universitaire Purpan, Toulouse), Muriel Eliaszewicz (AFSSA, Maisons-Alfort), Jean François Guillot (Université François-Rabelais, Tours), Isabelle Kempf (AFSSA, Ploufragan), Arlette Laval (Ecole Nationale Vétérinaire, Nantes), Roland Leclercq (Centre Hospitalier Universitaire, Caen), Francis Megraud (Hôpital Pellegrin, Bordeaux), Danièle Meunier (AFSSA, Lyon), Yves Millemann (Ecole Nationale Vétérinaire, Maisons-Alfort), Gérard Moulin (AFSSA ANMV, Fougères), Isabelle Pellanne (AFSSAPS, Paris), Pascal Sanders (AFSSA, Fougères), Benoît Schlemmer (Hôpital Saint-Louis, Paris) and Pierre-Louis Toutain (Ecole Nationale Vétérinaire, Toulouse). Funding No external or specific funds were used for this study. Transparency declarations None to declare. Supplementary data Tables S1 and S2 are available as Supplementary data at JAC Online (http://jac.oxfordjournals.org/). References 1. First Joint FAO/OIE/WHO Expert Workshop on Non-Human Antimicrobial Usage and Antimicrobial Resistance: Scientific Assessment. Geneva, Switzerland, 1 5 December 2003. http://www. who.int/foodsafety/publications/micro/nov2003/en/ (21 April 2008, date last accessed). 2. Second Joint FAO/OIE/WHO Expert Workshop on Non Human Antimicrobial Usage and Antimicrobial Resistance: Management Options. Oslo, Norway, 15 18 March 2004. http://www.who.int/foodborne_disease/resistance/fdb_antimicrobial_mar04.pdf (21 April 2008, date last accessed). 3. Analyse des ventes de médicaments aux officines et aux hôpitaux en France 1995 2005. http://afssaps.sante.fr/pdf/5/rapport_vente_ medicament_chiffre_2006.pdf (21 April 2008, date last accessed). 4. Suivi des ventes de médicaments vétérinaires contenant des antibiotiques en France en 2005. http://www.afssa.fr/documents/ ANMV-Ra-antibiotiques2005.pdf (21 April 2008, date last accessed). 5. European Federation of Animal Health (FEDESA). Antibiotic Use in Farm Animals does not threaten Human Health. FEDESA/FEFANA Press release. Brussels, Belgium, 13 July 2001. 6. Ungemach FR. Figures on quantities of antibacterials used for different purposes in the EU countries and interpretation. Acta Vet Scand 2000; 93: 89 98. 7. Grave K, Lingaas E, Bangen M et al. Surveillance of the overall consumption of antibacterial drugs in humans, domestic animals and farmed fish in Norway in 1992 and 1996. J Antimicrob Chemother 1999; 43: 243 52. 8. Overview of Antimicrobial Usage and Bacterial Resistance in Selected Human and Animal Pathogens in the UK, 2004. http://www. dardni.gov.uk/vet-meds.pdf (21 April 2008, date last accessed). 9. FAO/WHO/OIE Joint FAO/WHO/OIE Expert Meeting on Critically Important Antimicrobials. Report of a meeting held in FAO, Rome, Italy, 26 30 November 2007. http://www.fao.org/ag/ agn/agns/files/prepub_report_cia.pdf (21 April 2008, date last accessed). 10. French Food Safety Agency. Overview of report Veterinary Uses of Antibiotics, Antimicrobial Resistance and Consequences on Human Health. January 2006. http://www.afssa.fr/documents/ SANT-Sy-ABREN.pdf (21 April 2008, date last accessed). 11. World Health Organization. WHO Collaborating Centre for Drug Statistics methodology. http://www.whocc.no/atcddd/ (21 April 2008, date last accessed). 12. World Health Organization. WHO Collaborating Centre for Drug Statistics Methodology. Guidelines for ATC Classification and DDD Assignment. Oslo, Norway, 2007. http://www.whocc.no/atcvet (21 April 2008, date last accessed). 13. Office International des Epizooties. Guidelines for the monitoring of the quantities of antimicrobials used in animal husbandry. In: OIE ed. OIE International Standards on Antimicrobial Resistance. Paris, 13 16 2003. http://www.oie.int/eng/normes/mcode/en_chapitre_3.9.2. htm (21 April 2008, date last accessed). 14. Vander Stichele RH, Elseviers MM, Ferech M et al. European Surveillance of Antimicrobial Consumption (ESAC): data collection performance and methodological approach. Br J Clin Pharmacol 2004; 58: 419 28. 15. Ministère de l Agriculture et de la Pêche. Agreste la statistique agricole. http://www.agreste.agriculture.gouv.fr/ (21 April 2008, date last accessed). 16. La population animale. Enquête FACCO/TNS Sofres. http:// www.facco.fr/resultats.htm (21 April 2008, date last accessed). 17. INED Institut National d Etudes Demographiques. http://www. ined.fr/population-en-chiffres/france (21 April 2008, date last accessed). 18. IRSA CPAM Institut inter-régional pour la santé, prévention et santé publique, examens IRSA CPAM, 2001 âge 19 74 ans, poids en fonction de l âge, 32 352 hommes et 36 138 femmes, Distributions de poids, taille et corpulence de 6 à 9 ans et de 10 à 18 ans. 19. FARM 2003 2004. French Antimicrobial Resistance Monitoring in Bacteria of Animal Origin. http://www.afssa.fr/documents/ SANT-Ra-FARM.pdf (21 April 2008, date last accessed). 20. Anonymous. Defined Daily Dose animal (DDD animal ) Status January 2006. ATCvet Working Group, Oslo, Norway, January, 2006. http://www.whocc.no/filearchive/word/dddanimal_statement_jan_2006. doc (21 April 2008, date last accessed). 21. Chauvin C, Madec F, Guilemot D et al. The crucial question of standardisation when measuring drug consumption. Vet Res 2001; 32: 533 43. 624

Animal and human antimicrobial usages in France 22. Sommet A, Sermet C, Boelle PY et al. No significant decrease in antibiotic use from 1992 to 2000, in the French community. J Antimicrobial Chemother 2004; 54: 524 8. 23. Vigre H, Larsen PB, Andreasen M et al. The effect of discontinued use of antimicrobial growth promoters on the risk of therapeutic antibiotic treatment in Danish farrow-to-finish pig farms. Epidemiol Infect 2008; 136: 92 107. 24. Goossens H, Guillemot D, Ferech M et al. National campaigns to improve antibiotic use. Eur J Clin Pharmacol 2006; 62: 373 9. 25. DANMAP 2006. Use of antimicrobial agents and occurrence of antimicrobial resistance in bacteria from food animals, foods and humans in Denmark. http://www.danmap.org/pdffiles/danmap_2006. pdf (21 April 2008, date last accessed). 26. FINRES-Vet 2005 2006. Finnish Veterinary Antimicrobial Resistance Monitoring and Consumption of Antimicrobial Agents. http:// www.palvelu.fi/evi/files/55_519_523.pdf (21 April 2008, date last accessed). 27. MARAN 2005. Monitoring of Antimicrobial Resistance and Antibiotic Usage in Animals in the Netherlands In 2005. http://library. wur.nl/file/wurpubs/wurpublikatie_i00361303_001.pdf (21 April 2008, date last accessed). 28. NORM/NORM-VET 2006. Usage of Antimicrobial Agents and Occurrence of Antimicrobial Resistance in Norway. www.vetinst.no/nor/ content/download/531/4382/file/norm%20norm-vet%202006.pdf (21 April 2008, date last accessed). 29. SVARM 2006. Swedish Veterinary Antimicrobial Resistance Monitoring. http://www.sva.se/upload/pdf/tj%c3%a4nster%20och%20 produkter/trycksaker/svarm2006.pdf (21 April 2008, date last accessed). 30. Muller A, Coenen S, Monnet DL et al. European Surveillance of Antimicrobial Consumption (ESAC): outpatient antibiotic use in Europe, 1998 2005. Euro Surveill 2007; 12: E071011.1. 31. McEwen SA, Singer RS. Stakeholder position paper: the need for antimicrobial use data for risk assessment. Prev Vet Med 2006; 73: 169 76. 32. Kool SAE, Moltmann JF, Knacker T. Estimating the use of veterinary medicines in the European union. Regul Toxicol Pharmacol 2008; 50: 59 65. 625