Antibiotic use in eastern europe: a cross-national database study in coordination with the who regional office for europe

Antibiotic use in eastern Europe: a cross-national
database study in coordination with the WHO Regional
Offi
ce for Europe
Ann Versporten, Ganna Bolokhovets, Lilit Ghazaryan, Vafa Abilova, Galina Pyshnik, Tijana Spasojevic, Irma Korinteli, Lul Raka, Baktygul Kambaralieva, Lidija Cizmovic, Angela Carp, Vesela Radonjic, Nargis Maqsudova, Hatice Demet Celik, Marina Payerl-Pal, Hanne Bak Pedersen, Nina Sautenkova, Herman Goossens, on behalf of the WHO/Europe-ESAC Project Group Summary
Background There are no reliable data on antibiotic use in non-European Union (EU) southern and eastern European Lancet Infect Dis 2014
countries and newly independent states. We aimed to collect valid, representative, comparable data on systemic Published Online
antimicrobial use in these non-EU countries of the WHO European region.
Methods Validated 2011 total national wholesale antibiotic-use data of six southern and eastern European countries
and regions and seven newly independent states were analysed in accordance with the WHO anatomical therapeutic See Online/Comment
chemical (ATC)/defi ned daily doses (DDD) method and expressed in DDD/1000 inhabitants per day (DID).
Laboratory of Medical
Findings Total (outpatients and hospital care) antibiotic use ranged from 15·3 DID for Armenia to 42·3 DID for Turkey. Microbiology, Vaccine and
Co-amoxiclav was mainly used in Georgia (42·9% of total antibiotic use) and Turkey (30·7%). Newly independent Infectious Disease Institute
states used substantial quantities of ampicillin and amoxicillin (up to 55·9% of total antibiotic use in Azerbaijan). (VAXINFECTIO), University of
Antwerp, Antwerp, Belgium
Montenegro and Serbia were the highest consumers of macrolides (15·8% and 19·5% of total antibiotic use, (A Versporten MPH,
respectively), mainly azithromycin. Parenteral antibiotic treatment is common practice: 46·4% of total antibiotic use in Prof H Goossens PhD); Health
Azerbaijan (mainly ampicillin; 5·3 DID) and 31·1% of total antibiotic use in Tajikistan (mainly ceftriaxone; 4·7 DID).
Technologies and
Pharmaceuticals, Division of
Health Systems and Public
Interpretation This study provides publicly available total antibiotic-use data for 13 non-EU countries and areas of the Health, WHO Regional Offi ce
WHO European region. These data will raise awareness of inappropriate antibiotic use and stimulate policy makers for Europe, Copenhagen,
to develop action plans. The established surveillance system provides a method to develop quality indicators of Denmark (G Bolokhovets PhD,
H Bak Pedersen MSc, antibiotic use and to assess the eff ect of policy and regulatory actions.
N Sautenkova MPharm);
Scientifi c Centre of Drug and
Funding Netherlands Ministry of Health, Welfare, and Sport, and EU.
Medical Technology Expertise
of the Ministry of Health,
Yerevan, Armenia
Copyright 2014. World Health Organization. Published by Elsevier Ltd/Inc/BV. All rights reserved.
(L Ghazaryan MPharm); Ministry
of Health of Azerbaijan
2011, to ESAC-Net of the European Centre for Disease Republic, Analytical Expertise
Increasing levels of antimicrobial resistance have been Prevention and Control (ECDC).8 For European Union Centre for Medicines, Baku,
Azerbaijan (V Abilova MPharm);
spawned by rampant antibiotic use as shown both at the (EU) member countries and two European Economic Department on Organisation
population1 and individual level.2 The European Antibiotic Area/European Free Trade Association countries of Medicines Provision,
Resistance Surveillance network reported a Europe-wide (Norway and Iceland), ESAC-Net currently collects and Ministry of Health, Minsk,
increase of antimicrobial resistance to Gram-negative analyses antimicrobial consumption data for both the Belarus (G Pyshnik); Agency for
Medicines and Medical Devices
pathogens, with alarming evidence of increasing resistance community and the hospital sector.9 For the remaining of Bosnia and Herzegovina,
to third-generation cephalosporins, fl uoroquinolones, and 11 southern and eastern European countries and Banja Luka, Bosnia and
carbapenems in Escherichia coli and Klebsiella pneumoniae
12 former countries of the Soviet Union (excluding the Herzegovina
in 2011.3 This threat is reinforced by globalisation and three Baltic states) of the WHO European region, valid (T Spasojevic MPharm); JSC "My
family Clinic", Tbilisi, Georgia
migration because it favours spread of the resistance antibiotic-use data are not available. To address this gap, (I Korinteli MD); National
problem. The World Economic Forum's Global Risks 2013
the WHO Regional Offi ce for Europe (WHO Europe) and Institute of Public Health of
report4 concluded that "while viruses may capture more the Laboratory of Medical Microbiology of the University Kosovo and Faculty of
headlines, arguably the greatest risk of hubris to human of Antwerp, Belgium, established a surveillance network Medicine, University of
Pristina, Pristina, Kosovo
health comes in the form of antibiotic-resistant bacteria". on antimicrobial consumption in non-EU countries of (L Raka PhD); CitiHope
Sally Davies, England's chief medical offi
cer, stressed the the WHO European region. The overall aim of the project International, Bishkek,
importance of international collaboration and the need for was to set up a sustainable network of national anti- Kyrgyzstan
a global approach to contain antibiotic resistance.5 microbial surveillance systems to collect valid, (B Kambaralieva MSc); Agency
for Medicines and Medical
Standardised and feasible methods to survey anti- representative, and comparable data on antimicrobial Devices of Montenegro,
microbial use have been developed by the former use in non-EU countries of the WHO European region.
Podgorica, Montenegro
European Surveillance of Antimicrobial Consumption Our aims were to report on the method of data collection (L Cizmovic MPharm); Agency of
(ESAC) project.
Medicines, Chisinau, Moldova
6,7 This project was transferred in June, employed and the encountered pitfalls; describe the www.thelancet.com/infection Published online March 20, 2013 http://dx.doi.org/10.1016/S1473-3099(14)70071-4
(A Carp MPharm); Medicines
characteristics of data sources, providers, and type of data We calculated the defi ned daily dose per package and Medical Devices Agency of
available for the participating countries; assess data (DPP=[unit strength×pack size]/DDD). The DPP at Serbia, Belgrade, Serbia
validity and representativeness at national level; do a product level was then multiplied with the (V Radonjic PhD); Avicenna Tajik
State Medical University,
cross-national comparison of 2011 antibiotic-use rates of corresponding number of nationally reported packages Dushanbe, Tajikistan
12 non-EU European countries and Kosovo; and provide of antimicrobial drugs brought and sold on the market (N Maqsudova MPharm);
region-specifi c quality targets to improve antibiotic use. in 1 year (number of DDD at product level). Ministry of Health of Turkey,
All references, including in the reference list, to "Kosovo" Denominator data used were the total number of Turkish Medicines and Medical
Devices Agency, Ankara, Turkey
mean "Kosovo (in accordance with UN Security Council inhabitants per year of a country (mid-year population) (H Demet Celik MScPharm); and
resolution 1244 [1999])".
as provided by the respective national statistical offi Croatian Committee for
or the United Nations Development Program.11 For Antibiotic Resistance
Serbia, we consulted the CIA World Factbook because it Surveillance, Croatian Academy
for Medical Sciences, Zagreb,
Participating countries
provided denominator data for Serbia only (omitting Croatia (M Payerl-Pal MD)
The ministries of health of the participating countries Kosovo; appendix). We subsequently calculated the Correspondence to: nominated national representatives to participate in this outcome measurement unit, DDD/1000 inhabitants per Prof Herman Goossens, WHO/Europe-ESAC project. Medicine agencies of health day (DID), at product level.
Laboratory of Medical ministries from nine newly independent states (Armenia, Microbiology, Vaccine and Infectious Disease Institute Azerbaijan, Belarus, Georgia, Kyrgyzstan, Moldova, Data aggregation
(VAXINFECTIO), Faculty of Tajikistan, Ukraine, and Uzbekistan) and six south and Data aggregation was done in accordance with the ATC Medicine and Health Science, eastern European countries (Albania, Bosnia and classifi cation.10 For macrolides, we attributed a classifi cation University of Antwerp—CDE, Herzegovina, Macedonia, Montenegro, Serbia, Turkey), according to the mean plasma elimination half-life Universiteitsplein 1, Room S6.26, B-2610 Antwerp, plus Kosovo, agreed to participate. We added validated subdividing them into short-acting (half-life <4 h), 2011 data for Croatia, a southeastern European country intermediate-acting (half-life 4–24 h), and long-acting (half- not reporting 2011 data to ESAC-Net.
life >24 h) macrolides.12 The quinolone substances were classifi ed according to three generations based on their For the CIA World Factbook see
Data collection
chemical structure and antimicrobial activity.13 Overall, The participating country representatives constructed an 144 unique antibiotic substances were used in 2011, exhaustive validated national antimicrobial drug register ranging from 41 substances in Kosovo to 72 in Turkey. See Online for appendix
and use database, including detailed information (unit Those substances were aggregated into ten pharmacological
strength, pack size, galenic form, and route of subgroups (ATC third level) and 35 chemical subgroups
administration) for all antimicrobial products available on (ATC fourth level) for descriptive analyses.
the market. The database contained antibacterials for
systemic use (anatomical therapeutic chemical [ATC] Data validation
subgroup J01), antimycotics (J02), antifungals (D01BA) Data validation included thorough checking of every
and antivirals for systemic use (J05), amantadine used as reported drug in the drug register database to ensure the
an anti-infl uenza drug (N04BB01), antibiotics for treatment
WHO ATC/DDD classifi cation method was correctly of tuberculosis (J04AB and J04AM), oral and rectal applied. We sought online supplementary information in nitroimidazole derivatives (P01AB), and antibiotics used as case of poorly defi ned product labels (unit strength and intestinal anti-infectives (A07AA). Antimicrobials for pack size), for example, on Russian products. National topical and vaginal use were excluded. Each medicinal representatives were supplied with a standard validation product was classifi ed in accordance with the WHO report providing longitudinal total and proportional standardised and internationally recognised ATC coding antimicrobial volumes of use, expressed in DID. Results system, classifying drugs according to their main were then revised, corrected, or justifi ed (gaps, drop or therapeutic use. We further assigned to each product the increase of use over time). Reference data from ESAC-WHO defi ned daily dose (DDD), a unit of measurement Net were used to asses and interpret the data.14that is an international compromise of the assumed We report on validated national antimicrobial wholesale average maintenance dose per day for a drug used for its data of ATC group J01 for 2011 collected from 12 non-EU main indication in adults.10 The implementation of the European countries and Kosovo. The appendix summarises WHO ATC/DDD method enabled us to construct a all data available by type of data, data suppliers and database for measuring and comparing antimicrobial use. coverage, reported aggregation levels, and population data; In close collaboration with the WHO Collaborating Centre information that was collected by means of a questionnaire. for Drug Statistics Methodology of the Norwegian Institute Eight countries delivered total care data, Kyrgyzstan and of Public Health, we assigned provisional ATC codes and Montenegro supplied data separately for the community DDDs to products for which this value was not yet and hospital sector, and Turkey and Georgia provided attributed.10 These products mainly included combinations ambulatory care data only. Turkey used the Information For more on the Information
of drugs reported by the newly independent states—eg, Management System database, which included complete,
Management System database
ciprofl oxacin and tinidazole, ornidazole, or metronidazole; not extrapolated, data covering the whole ambulatory care see http://www imshealth com/ and tetracycline and oleandomycin.
sector. Armenia, Azerbaijan, Belarus, Turkey, and Kosovo www.thelancet.com/infection Published online March 20, 2013 http://dx.doi.org/10.1016/S1473-3099(14)70071-4
supplied quarterly data, allowing the investigation of seasonal variation. Data coverage was 100% for ten countries, greater than 98% for two countries, and, for political reasons, 70% for Georgia.
We analysed country-specifi c total and proportional antibiotic use expressed in DID for the year 2011. The data were further compared with 2011 total care (ambulatory plus hospital care) ESAC-Net data of the DDD/1000 inhabitants per day ECDC. ESAC-Net data are publicly available at the ATC third and fourth level (ESAC-Net interactive database).14 Supplementary data on the macrolides12 and quinolones subgroups13 were provided by the ECDC.
Antibiotic use in countries that could not yet deliver Bosnia and Herzegovina data (Albania, Macedonia, and Uzbekistan) or obtained poor data coverage (Ukraine) are not reported. Other antibacterials (J01X) Tetracyclines (J01A) Antibacterial combinations (J01R) Quinolones (J01M) Aminoglycosides (J01G) Macrolides, lincosamides, and streptogramins (J01F) Role of the funding source
Amphenicols (J01B) Other β-lactam antibacterials, cephalosporins (J01D) The funder had no role in study design, data collection, Sulfonamides and trimethoprim (J01E) β-lactam antibacterials, penicillins (J01C) data analysis, data interpretation, or writing the report. Figure 1: Total antibiotic use in 12 European countries and Kosovo, 2011
The corresponding author had full access to all the data The category (WHO anatomical therapeutic chemical subgroup) "Other β-lactam antibacterials, cephalosporins" in the study and had fi nal responsibility for the decision includes carbapenems and monobactams. "Other antibacterials" includes glycopeptide antibacterials, polymyxins, to submit for publication, following agreement from all fusidic acid, imidazole derivates, nitrofuran derivates, and other antibacterials. DDD=defi ned daily doses. authors.
*Reported only outpatient antibiotic use.
Penicillins with extended spectrum (broad-spectrum penicillins; J01CA) Figure 1 depicts total antibiotic use (ATC group J01) β-lactamase-sensitive penicillins (narrow-spectrum penicillins; J01CE)Combinations of penicillins, including β-lactamase inhibitors (J01CR) expressed in DID in 12 non-EU European countries and β-lactamase-resistant penicillins (J01CF) Kosovo (year 2011). Antibiotic use among the participating countries diff ered signifi cantly, ranging from 15·3 DID in Armenia to 42·3 DID in Turkey. We also compared anti- biotic use to 29 ESAC-Net countries in 2011 (appendix).
Penicillins (ATC group J01C) were the most commonly DDD/1000 inhabitants per day used antibiotics in all countries. Highest proportional use of penicillins of total antibiotic use was noted for Georgia (67·6%; 14·2 DID), and then Azerbaijan (65·9%; Kosovo Azerbaijan 11·4 DID). Highest total penicillin use was reported for Tajikistan (18·2 DID; fi gure 2) and lowest for Armenia Bosnia and Herzegovina (6·1 DID). Narrow spectrum penicillin use, mainly phen-oxymethylpenicillin, was low and varied from 0·02 DID in Figure 2: Total penicillin (J01C) use subdivided into four main subgroups in 12 European countries and
Belarus to 1·3 DID in Montenegro. Amoxicillin and Kosovo, 2011
DDD=defi ned daily doses. *Reported only outpatient antibiotic use.
ampicillin were very commonly used in Tajikistan (15·9 DID; 45·6% of total antibiotic use) and Azerbaijan (9·7 DID; 55·9%) and amoxicillin in Montenegro Fourth-generation cephalosporins (J01DE) (9·1 DID; 23·6%). Highest use of combinations of Third-generation cephalosporins (J01DD) penicillins (mainly co-amoxiclav) was noted for Turkey Second-generation cephalosporins (J01DC) First-generation cephalosporins (J01DB) (13·0 DID; 30·7% for co-amoxiclav of total antibiotic use) and Georgia (9·0 DID; 42·9%), and lowest use was noted for Azerbaijan (0·5 DID) and Kyrgyzstan (0·6 DID). Penicillin (ATC group J01C) use was also compared with 29 ESAC-Net countries (appendix).
DDD/1000 inhabitants per day Highest total cephalosporin (ATC group J01D) use was noted for Turkey (14·1 DID; 33·4% of total antibiotic use) and lowest for Azerbaijan (0·8 DID; 4·3%; fi gure 3). Highest fi rst-generation cephalosporin use was reported by Bosnia and Herzegovina Montenegro (2·9 DID), Serbia (2·3), Kosovo (2·1), and Figure 3: Total cephalosporin use subdivided into four main subgroups in 12 European countries and
Kyrgyzstan (2·0), and lowest was reported by Georgia (0·02). Kosovo, 2011
Highest use of second-generation cephalosporins was DDD=defi ned daily doses. *Reported only outpatient antibiotic use.
www.thelancet.com/infection Published online March 20, 2013 http://dx.doi.org/10.1016/S1473-3099(14)70071-4
use of long-acting macrolides (azithromycin) was noted for Montenegro (3·4 DID), Serbia (2·7), and Short-acting macrolides Georgia (1·2). Macrolide (ATC group J01FA) and Long-acting macrolides lincosamide (J01FF) use was also compared with 29 ESAC-Net countries (appendix). Streptogramin (J01FG) use was not reported.
Highest quinolone (ATC group J01M) use was noted for DDD/1000 inhabitants per day Montenegro (4·4 DID; 11·4% of total antibiotic use) and lowest for Azerbaijan (0·7 DID; 4·3%; fi gure 5). Armenia Azerbaijan Tajikistan Montenegro and Serbia reported highest use of fi rst- generation quinolones (1·5 and 1·2 DID, mainly pipemidic Figure 4: Total macrolide (J01FA) and lincosamide (J01FF) use subdivided into four main subgroups in
acid) and then Georgia (0·8 DID, mainly norfl oxacin). 12 European countries and Kosovo, 2011
Highest use of second-generation quinolones was reported DDD=defi ned daily doses. *Reported only outpatient antibiotic use.
by Tajikistan, Turkey, Kyrgyzstan, Montenegro, and Kosovo (3·4–2·6 DID, mainly cipro fl oxacin). Highest use of third- generation quinolones was noted for Turkey (0·5 DID, mainly gemifl oxacin) and then Croatia (0.1 DID, mainly moxifl oxacin); minor use was reported for all other countries—no use was reported for Serbia. Quinolone (ATC group J01M) use was also compared with 29 ESAC- Net countries (appendix). Seasonal variation of quinolone DDD/1000 inhabitants per day use showed an increased use of levofl oxacin during the winter season in Turkey, Armenia, and Azerbaijan. Higher volumes of moxi fl oxacin use were reported during the winter season in Turkey; no increase during winter season was not for the other quinolones (appendix).
Figure 5: Total quinolone (J01M) use subdivided into three main subgroups in 12 European countries and
Highest tetracycline (ATC group J01A) use was reported Kosovo, 2011
DDD=defi ned daily doses. *Reported only outpatient antibiotic use.
for Belarus (3·0 DID; 16·9% of total antibiotic use), and then Serbia (2·3; 9·0%) and Armenia (2·0; 13·0%), and reported by Turkey (mainly cefuroxime; 9·0 DID) and then lowest for Georgia (0·5 DID; 2·3%; appendix, also Kosovo (1·5) and lowest by Tajikistan (0·02). Highest third- includes 29 ESAC-Net countries).
generation cepha losporin use was reported by Tajikistan Highest sulphonamide and trimethoprim (ATC group (4·9 DID), Turkey (4·2), and Montenegro (3·2) and lowest J01E) use, mainly sulfamethoxazole and trimethoprim, by Bosnia and Herzegovina (0·2). Most newly independent was noted for Tajikistan (2·0 DID; 5·6% of total antibiotic states consumed in total and proportionally more third- use) and Kyrgyzstan (1·9; 7·5%), and lowest for Belarus generation cephalosporins than the other countries (mainly (0·1; 0·4%).
ceftriaxone; up to 91% of total cephalosporin use in Georgia Substantial use of amphenicols (ATC group J01B) was and 80% in Tajikistan, Azerbaijan, and Belarus). reported for Azerbaijan, Armenia, Kyrgyzstan, Belarus, and Montenegro mainly consumed cefi xime for oral use Tajikistan (0·4–0·6 DID). Combination products that were (2·1 DID). Southern and eastern European countries con- not yet listed in the ATC/DDD classifi cation system added sumed mainly fi rst-generation cepha losporins (cephalexin). overall 0·013 DID in Georgia to 0·44 DID in Moldova.
Turkey was the only country that reported third-generation Highest total parenteral antibiotic use was noted for cephalosporins cefdinir and cefditoren for oral use Tajikistan (11·5 DID; 31·1% of total antibiotic use), and (1·6 DID, 11·3% of total cephalosporin use). Cephalosporin ceftriaxone alone accounted for 4·7 DID (12·7% of total (ATC group J01D) use was also compared with 29 ESAC- antibiotic use; appendix). However, highest proportional Net countries (appendix).
parenteral use was noted in Azerbaijan (8·0 DID; 46·4% Highest macrolide (ATC group J01FA) and of total antibiotic use), and ampicillin alone accounted lincosamide (J01FF) use was noted for Montenegro (6·1 for 5·3 DID or 31% of total antibiotic use. Lowest total DID; 15·8% of total antibiotic use) and Serbia (5·0 DID; parenteral use was noted for Turkey (0·9 DID; 2·1%).
19·5%), and lowest for Tajikistan (0·7 DID; 2·1%; fi gure 4). Montenegro reported the highest use of short- acting macrolides (2·2 DID, mainly erythromycin), with To our knowledge, this study for the fi rst time presents Kosovo and Serbia the next highest (0·9 and 0·8 DID, reliable total antibiotic-use data for 13 southern and eastern respectively). Highest use of intermediate-acting European countries, Kosovo, and newly independent macrolides was reported by Turkey and Serbia (2·8 and states, expressed in DID (panel). We were able to compare 1·2 DID, respectively, mainly clarithromycin), whereas their antibiotic-use patterns with those of 29 ESAC-Net use in the other countries was less than 1 DID. Highest countries because we used the same methods as developed www.thelancet.com/infection Published online March 20, 2013 http://dx.doi.org/10.1016/S1473-3099(14)70071-4
in the former ESAC project1 and the validation process was comprehensive medicines reimbursement system leaving based on standard reports developed by ESAC. Our main medicines as out-of-pocket payments. The reliance on fi ndings are that total antibiotic use ranged from 15·3 DID direct out-of-pocket payments is serious a problem in for Armenia to 42·3 DID for Turkey; co-amoxiclav was many of the countries included in this study, undermining commonly used in Georgia and Turkey; newly independent the principle of equity with respect to both fi nancing and states used substantial quantities of ampicillin and access to health care. High rates of self-medication with amoxicillin; Montenegro and Serbia were the highest antibiotics might go along with the underuse of health consumers of macrolides, mainly azithromycin; and services18 or might lead to diagnostic and health-care parenteral treatment with antibiotics is common practice system seeking delays.19 Therefore, restriction of over-the-in the newly independent states.
counter use of antibiotics is urgently needed, and could be Compared with ESAC-Net countries, Turkey had the partly achieved by implementing national regulatory highest antibiotic use in Europe, and on the basis of this instruments,20 public awareness campaigns,21 or enhancing fi nding, the Turkish government already published a eff orts in educating health-care providers towards Rational Drug Use National Action plan 2013–2017, with appropriate prescribing.19 Aggressive promotion by For the Turkish rational drug
quantitative targets to reduce antibiotic use (by 2 DID industry and lack of trust towards doctors' skills is another use plan see http://www.
annually between 2014 and 2017). Armenia had very low reason for self-medication in some of these countries, and akilciilac.gov.tr/
antibiotic use, similar to northern EU countries. However, regaining this trust should also lead to improve patient
this low antibiotic use might relate to underuse because of
limited access to drugs for a substantial proportion of the
Panel: Research in context
population, particularly in rural regions and among poor people, as shown in a recent survey on health inequalities in Armenia.15 Belarus is also a European country in which We searched Pubmed with keywords related to antibiotics, subgroups of antibiotics, surveys, antibiotic use was low, but here health services have and countries involved in our study. We did not identify any specifi c published works remained aff ordable for virtually everyone after the collapse providing an overview of common patterns of antibiotic use. Increasing levels of of the Soviet model of health care, which sought to achieve antimicrobial resistance correlate with inappropriate antibiotic use as shown at the universal, free access to basic health services.16 This low population1 and individual level.2 Monitoring of antimicrobial use is a crucial component antibiotic use might illustrate the eff ect of better coverage to identify targets for improving antimicrobial use and to further correlate with or more rational use of medicines and state budget.
antimicrobial resistance surveillance programmes.1 The internationally recognised WHO Because of poor information systems in most surveyed anatomical therapeutic chemical (ATC)/defi ned daily doses (DDD) method allows the countries and lack of universal coverage in these countries, measurement and comparison of drug use in populations.10 This standardised method has data collectors had to go through several sources of been employed to survey antimicrobial use within Europe by the European Surveillance of information, such as wholesalers' data (available from the Antimicrobial Consumption (ESAC) project,6,7 and currently by ESAC-Net of the European ministry of health in most countries; appendix). However, Centre for Disease Prevention and Control.14 For other southern and eastern European the advantage of collecting sales data over reimbursement countries and the former Soviet Union countries of the WHO European region, valid data is the inclusion of antibiotics procured over the antimicrobial use data were not available.
counter without a prescription.17 For countries with centralised procurement of hospital medicines (such as Kyrgyzstan), ambulatory care data have been complemented For the fi rst time, validated data on antibiotic use in seven newly independent states with hospital reimbursement data. Data on humanitarian (Armenia, Azerbaijan, Belarus, Georgia, Kyrgyzstan, Moldova, Tajikistan), fi ve southern and assistance and local manufacturers were also considered eastern European countries (Bosnia and Herzegovina, Croatia, Montenegro, Serbia, Turkey), for countries with big volumes of donations (such as and Kosovo, have been collected and analysed in accordance with the WHO ATC/DDD Armenia and Tajikistan) and big manufacturers (such as method. Findings were benchmarked to ESAC-Net data.
Serbia, Turkey, and Ukraine). For most countries, reliable We identifi ed substantial diff erences in the quantity and quality of antibiotic use, with Turkey denominator data were available; however, estimates were using most antibiotics (42·3 DDD/1000 inhabitants per day) and Armenia the least used for Bosnia and Herzegovina, and Kosovo.
(15·3 DDD/1000 inhabitants per day). In general, broad-spectrum penicillins (amoxicillin and One of the main problems in the countries included in ampicillin) were most commonly used, but co-amoxiclav was most used in Georgia and this study is the widespread practice of selling antibiotics Turkey. Montenegro and Serbia particularly used the long-acting macrolide azithromycin. over the counter. Outdated (2001) data are available for Remarkably high parenteral antibiotic use was noted for all newly independent states, with eight newly independent states, showing that on average Tajikistan showing the highest total parenteral use (mainly ceftriaxone) and Azerbaijan 21·8% of the adult population purchased medicines showing the highest proportional parenteral use (mainly ampicillin). This high use of without a prescription.16 However, according to a survey injectable antibiotics relates to the inappropriate use of antibiotics in outpatients.
done in 2012 among the WHO/Europe-ESAC project These data for the fi rst time allow the auditing of antimicrobial use, help identify targets for group members, more than 50% of antibiotics are sold quality improvement, and aid the development of national action plans to enhance judicious over the counter in most of their countries. There are antibiotic use. This study provides the foundation for a sustainable long-term surveillance several reasons for the unauthorised over-the-counter sales network on antimicrobial use in this part of the WHO European region. Consequently, long- of antibiotics, including lack of sustainable health-care term antimicrobial use data will aid the assessment and improvement of future action plans.
systems, poor enforcement of regulation, and lack of a www.thelancet.com/infection Published online March 20, 2013 http://dx.doi.org/10.1016/S1473-3099(14)70071-4
satisfaction and prescription-based antibiotic use.15 that only 2% of outpatient antibiotics in 20 European However, it is clear that only strengthening (both widening countries participating in ESAC were used for parenteral and deepening) of health coverage can become the basis treatment.28 Although our study includes data from both for rational use of any prescription medicines, including hospital and community care, injectable antibiotics given antibiotics. In the future, reimbursement data should also in hospitals solely cannot account for the recorded high be collected in these countries because the diff erence parenteral use. Indeed, most antibiotics are used in with sales data might provide a rough estimate of outpatients and for countries participating in the former non-reimbursed (principally over-the-counter) sales.17 ESAC project that reported combined hospital and Amoxicillin was widely used in all newly independent outpatient use, the proportions of hospital use were below states (except Georgia), southern and eastern European 10% for most countries.29 Therefore, we think that the countries, and Kosovo. Although the recommended anti- remarkably high parenteral use is explained by the high biotic treatment for community-acquired lower-respiratory- outpatient use of injectable antibiotics (mainly ampicillin tract infections in Europe is amoxicillin or a tetracycline,22 and ceftriaxone), certainly in the newly independent strikingly high use of this antibiotic might still relate to states. In Italy, outpatient parenteral treatment, mainly unnecessary use. Indeed, antibiotics virtually never benefi t ceftriaxone, was popular because physicians and patients patients with acute cough, which is one of the most thought parenteral administration of antibiotics to be common reasons for consulting in primary care and more eff ective than oral administration of these drugs, prescribing antibiotics.23,24 The combination of amoxicillin even for treatment of benign infections in primary care.30 and a β-lactamase inhibitor, which overcomes some types of More in-depth studies are needed to explore indications of resistance, is not a fi rst-line agent for empirical treatment parenteral use and to explain the success of these for most commonly encountered infections in primary antibiotics among patients and physicians.
care. Yet, in Turkey and Georgia this drug is used extensively, The purpose of collecting indicators of antibiotic use is to which raises concern regarding its appropriate use.
identify inappropriate prescribing and to provide a means The southern and eastern European countries, Kosovo, to measure the eff ect of interventions. Benchmarking, by Kyrgyzstan, and Moldova consumed high volumes of comparison of antibiotic use between countries, is an fi rst-generation cephalosporins, similar to the use in important stimulus to quality improve ment.31,32 Our study northern EU countries.25 The newly independent states identifi es opportunities for quality improvement (eg, and Montenegro had much higher third-generation reduce total use of antibiotics, reduce use of co-amoxiclav cephalosporin (mainly ceftriaxone) use compared with and azithromycin in southern and eastern European the rest of Europe. Because of its long half-life and once- countries and Kosovo, and reduce parenteral use in newly daily dosing requirement, ceftriaxone is an attractive independent states).31,32 Policy makers and medical option for outpatient parenteral therapy.26 However, third- professionals should use these data to trigger actions and to generation cephalosporins might select for bacteria that develop, implement, and assess national guidelines.
produce extended-spectrum β-lactamases, so antibiotic In conclusion, our study shows striking diff erences of resistance should be monitored in these countries.27 antibiotic prescribing in non-EU countries and Kosovo The long-acting macrolides (mainly azithromycin) are in the WHO European region and should be considered responsible for high total macrolide use (mainly in as a fi rst step to improve antibiotic use in these countries
Montenegro and Serbia), but also proportionally within and areas. Our eff ort should be continued and expanded
this group of antibiotics (in most countries included in to other countries of the region; we should also survey
this study). Our fi ndings suggest that this subgroup of antibiotic resistance, because surveillance provides an
antibiotics is still prescribed inappropriately in this part essential component for policy development and
of Europe. Use of so-called respiratory quinolones containment of antibiotic resistance.
(levofl oxacin and moxifl oxacin) increased during the Contributors
winter season in those countries where quarterly data NS and HG initially proposed the idea for this study. AV was responsible
were provided. Because quinolones are not recommended
for data collation, validation, and analysis, and contributed in writing the as fi rst-line therapy for the treatment of many infectious article. HG was responsible for the analysis and interpretation of data and writing of the article. GB was responsible for running the network and was diseases, their high use noted in some of the countries involved in data validation and interpretation. All authors contributed to included in this study raises concern, especially for the interpretation of the data and the write up. All authors and members of prevention and control of multiply and extensively drug- the WHO/Europe-ESAC Project Group critically reviewed the report.
resistant Mycobacterium tuberculosis.
WHO/Europe-ESAC project group
An interesting fi nding of this study is untypically high use Albania Pellumb Pipero (Ministry of Health, Tirana). Armenia Lilit Ghazaryan, Hasmik Martirosyan (Scientifi c Centre of Drug and of amphenicols (chloramphenicol in particular) in some of Medical technology Expertise, Yerevan). Azerbaijan Vafa Abilova, the newly independent states. This fi nding could partly be Farid Aliyev (Ministry of Health of Azerbaijan Republic Analytical explained by the fact that chloramphenicol had been widely Expertise Centre for Medicines, Baku). Belarus Galina Pyhsnik used in these countries for treatment of diarrhoea.
(Department on Organisation of Medicines Provision, Ministry of Health, Minsk). Bosnia and Herzegovina Tijana Spasojevic (Agency for Medicines Finally, a remarkable fi nding of our study is the high and Medical Devices of Bosnia and Herzegovina, Banjaluka). Georgia Irma parenteral use. In the former ESAC project, it was reported Korinteli (JSC My family Clinic, Tbilisi); Karaman Pagava (Tbilisi State www.thelancet.com/infection Published online March 20, 2013 http://dx.doi.org/10.1016/S1473-3099(14)70071-4
Medical University, Tbilisi). Kyrgyzstan Baktygul Kambaralieva (CitiHope 11 UNDP. Human development trends data by indicator: population, International, Bishkek); Ainura Mamasheva (Mandatory Health Insurance total both sexes (thousands). https://data.undp.org/dataset/ Fund, Bishkek). Macedonia Gobulinska Bosevska (Institute of Public Population -total-both-sexes-thousands-/a6kh-7xde (accessed Jan 28, Health and Kristina Hristova, Health Insurance Fund, Skopje). Montenegro Lidija Cizmovic (Agency for Medicines and Medical Devices of 12 Adriaenssens N, Coenen S, Versporten A, et al. European Montenegro, Podgorica). Moldova Angela Carp (Agency of Medicines, Surveillance of Antimicrobial Consumption (ESAC): outpatient Chisinau); Elena Chitan (Department of Social Pharmacy "Vasile macrolide, lincosamide and streptogramin (MLS) use in Europe
(1997–2009). J Antimicrob Chemother 2011; 66 (suppl 6): vi37–45.
Procopisin", State University of Medicine and Pharmacy "Nicolae Testemitanu", Chisinau). Kosovo Lul Raka, Arianit Jakupi, Valdet Uka, 13 Adriaenssens N, Coenen S, Versporten A, et al. European Surveillance of Antimicrobial Consumption (ESAC): outpatient Shefqet Hajdari, Besian Sejdiu (Kosovo Medicine Agency, Pristina). Serbia quinolone use in Europe (1997–2009). J Antimicrob Chemother 2011; Vesela Radonjic (Medicines and Medical Devices Agency of Serbia, 66 (suppl 6): vi47–56.
Beograd). Switzerland Catherine Plüss-Suard (Service of Hospital 14 ECDC. ESAC-Net interactive database. http://www.ecdc.europa.eu/ Preventive Medicine, Lausanne University Hospital, Lausanne). Tajikistan Nargis Maqsudova, Salomudin Isupov (Avicenna Tajik State Medical database.aspx (accessed Feb 25, 2014).
University, Dushanbe). Ukraine Larisa Iyakovleva (National 15 Tonoyan T, Muradyan L. Health inequalities in Armenia—analysis Pharmaceutical University, Kharkov); Kateryna Posokhova, Elena Matvieva of survey results. Int J Equity Health 2012; 11: 32.
(State Centre of the Ministry of Health of Ukraine, Kiev). Turkey 16 Balabanova D, McKee M, Pomerleau J, Rose R, Haerpfer C. Health Hatice Demet Celik, Hakki Gursoz, Ali Alkan, Bahar Melik, Fatma Savur, service utilization in the former Soviet Union: evidence from eight Mesil Aksoy, Cem Seckin (Ministry of Health of Turkey, Turkish Medicines countries. Health Serv Res 2004; 39: 1927–50.
and Medical Devices Agency, Ankara). Croatia Marina Payerl-Pal (Croatian 17 Campos J, Ferech M, Lazaro E, et al. Surveillance of outpatient Committee for Antibiotic Resistance Surveillance, Croatian Academy for antibiotic consumption in Spain according to sales data and Medical Sciences, Zagreb); Arjana Tambic Andrasevic (Department of reimbursement data. J Antimicrob Chemother 2007; 60: 698–701.
Clinical Microbiology, University Hospital for Infectious Diseases, Zagreb).
18 Clark DV, Ismayilov A, Bakhishova S, et al. Under-utilization of health care services for infectious diseases syndromes in rural Declaration of interests
Azerbaijan: a cross-sectional study. BMC Health Serv Res 2011; 11: 32.
We declare that we have no competing interests.
19 Rabin AS, Kuchukhidze G, Sanikidze E, Kempker RR, Blumberg HM. Prescribed and self-medication use increase delays Funding of this work was provided as a voluntary donation to WHO from in diagnosis of tuberculosis in the country of Georgia. the Netherlands' Ministry of Health, Welfare and Sport. We thank Klaus Int J Tuberc Lung Dis 2013; 17: 214–20.
Weist, ECDC Stockholm, Sweden for providing ESAC-Net data and for 20 WHO Europe. Joint annual review meeting on national health critical reading of the reoport, Hege Salvesen Blix and Irene Litleskare, strategy of Tajikistan. http://www.euro.who.int/en/countries/ University of Oslo, Norway for providing advice on missing ATC codes or DDD values. The work done by the Kosovo team was supported from the national-health-strategy-of-tajikistan (accessed Jan 28, 2014).
EU grant "Research Capacity Development in Kosovo".
21 Huttner B, Goossens H, Verheij T, Harbarth S. Characteristics and outcomes of public campaigns aimed at improving the use of antibiotics in outpatients in high-income countries. Lancet Infect Dis Goossens H, Ferech M, Vander SR, Elseviers M. Outpatient antibiotic 2010; 10: 17–31.
use in Europe and association with resistance: a cross-national 22 Woodhead M, Blasi F, Ewig S, et al. Guidelines for the management database study. Lancet 2005; 365: 579–87.
of adult lower respiratory tract infections. Eur Respir J 2005; Malhotra-Kumar S, Lammens C, Coenen S, Van Herck K, Goossens H. Eff ect of azithromycin and clarithromycin therapy on 23 Butler CC, Hood K, Verheij T, et al. Variation in antibiotic prescribing pharyngeal carriage of macrolide-resistant streptococci in healthy and its impact on recovery in patients with acute cough in primary volunteers: a randomised, double-blind, placebo-controlled study. care: prospective study in 13 countries. BMJ 2009; 338: b2242.
Lancet 2007; 369: 482–90.
24 Little P, Stuart B, Moore M, et al. Amoxicillin for acute ECDC. Antimicrobial resistance surveillance in Europe 2011. lower-respiratory-tract infection in primary care when pneumonia Annual Report of the European Antimicrobial Resistance is not suspected: a 12-country, randomised, placebo-controlled Surveillance Network (EARS-Net). Stockholm: European Centre trial. Lancet Infect Dis 2013; 13: 123–29.
for Disease Prevention and Control, 2012.
25 Versporten A, Coenen S, Adriaenssens N, et al. European Surveillance WEF. Insight report: global risks 2013, 8th edn. Geneva: World of Antimicrobial Consumption (ESAC): outpatient cephalosporin use in Economic Forum, 2013. http://www3.weforum.org/docs/WEF_ Europe (1997–2009). J Antimicrob Chemother 2011; 66 (suppl 6): vi25–35.
GlobalRisks_Report_2013.pdf (accessed May 22, 2013).
26 Guglielmo BJ. Chapter e1: anti-infective chemotherapeutic and Gulland A. Science ministers discuss the growing threat of antibiotic agents. In: Papadakis MA, McPhee SJ, Rabow MW, eds. antimicrobial resistance. BMJ 2013; 346: f3921.
Current Medical Diagnosis and Treatment 2013. New York, NY: Vander Stichele RH, Elseviers MM, Ferech M, Blot S, Goossens H. McGraw-Hill, 2013.
European surveillance of antimicrobial consumption (ESAC): 27 Livermore DM, Hawkey PM. CTX-M: changing the face of ESBLs in data collection performance and methodological approach. the UK. J Antimicrob Chemother 2005; 56: 451–54.
Br J Clin Pharmacol 2004; 58: 419–28.
28 Coenen S, Muller A, Adriaenssens N, Vankerckhoven V, Adriaenssens N, Coenen S, Versporten A, et al. European Surveillance Hendrickx E, Goossens H. European Surveillance of Antimicrobial of Antimicrobial Consumption (ESAC): outpatient antibiotic use in Consumption (ESAC): outpatient parenteral antibiotic treatment in Europe (1997–2009). J Antimicrob Chemother 2011; 66 (suppl 6): vi3–12.
Europe. J Antimicrob Chemother 2009; 64: 200–05.
ECDC. European surveillance networks/ECDC disease specifi c 29 Vander Stichele RH, Elseviers MM, Ferech M, Blot S, Goossens H. Hospital consumption of antibiotics in 15 European countries: results of the ESAC Retrospective Data Collection (1997–2002). networks.aspx (accessed Jan 28, 2014).
J Antimicrob Chemother 2006; 58: 159–67.
ECDC. Surveillance of antimicrobial consumption in Europe, 2010. 30 Esposito S, Noviello S, Leone S, et al. Outpatient parenteral Stockholm: European Centre for Disease Prevention and Control, antibiotic therapy (OPAT) in diff erent countries: a comparison. Int J Antimicrob Agents 2004; 24: 473–78.
31 Goossens H, Guillemot D, Ferech M, et al. National campaigns to (accessed Sept 16, 2013).
improve antibiotic use. Eur J Clin Pharmacol 2006; 62: 373–79.
10 WHO. Anatomical Therapeutic Chemical (ATC) classifi cation 32 Coenen S, Ferech M, Haaijer-Ruskamp FM, et al. European system: guidelines for ATC classifi cation and DDD assignment. Surveillance of Antimicrobial Consumption (ESAC): quality Oslo: WHO Collaborating Centre for Drug Statistics Methodology, indicators for outpatient antibiotic use in Europe. Qual Saf Health Care 2007; 16: 440–45.
the_atc_ddd_system/ (accessed May 14, 2013).
www.thelancet.com/infection Published online March 20, 2013 http://dx.doi.org/10.1016/S1473-3099(14)70071-4

Source: http://graduateinstitute.ch/files/live/sites/iheid/files/sites/globalhealth/ghp-new/Documents/Events/Policy%20Dialogues/Anti-Microbial%20Database%20Study.pdf