Antibiotic Consumption in Non-Teaching Lebanese Hospitals 619

Home , Monobactam

Journal of Infection and Public Health (2016) 9, 618—625

View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector

Antibiotic consumption in non-teaching

Lebanese hospitals: A cross-sectional study

a,∗ b c

Katia Iskandar , Pierre A. Hanna , Pascale Salameh ,

Etwal B. Raad

Lebanese International University, Department of Pharmacy, Beirut, Lebanon

Lebanese University, Department of Medicine, Mount Lebanon, Lebanon

Lebanese University, Department of Pharmacy, Mount Lebanon, Lebanon

Received 5 August 2015; received in revised form 17 December 2015; accepted 26 December 2015

KEYWORDS Summary The rising threat of antibiotic resistance is linked to patterns of antibi-

Antibiotic; otic use in hospital settings where global efforts are undertaken to encourage

Consumption; reporting and benchmarking antibiotic consumption in an attempt to improve pre-

Resistance; scription regimens. In Lebanon, where data concerning the level of antibiotic

Hospitals; consumption in hospitals is scarce, the aim of our paper is to track the inten-

sity of antibiotic consumption in order to identify potential evidence of antibiotic

Deﬁned Daily Dose

misuse or abuse. The study is conducted in 2012 for a period of 12-month using

data from pharmacy records in 27 non-teaching Lebanese hospitals according to

the Anatomical, Therapeutic and chemical classiﬁcation system and Deﬁned Daily

Dose (ATC/DDD) recommended by the World Health Organization and compiling data

on ABC Calc software version 3.1. Results show that the average antibiotic con-

sumption excluding pediatric cases is 72.56 Deﬁned Daily Dose per 100 Bed-Days

(DDD/100BD). Total broad spectrum antibiotic consumption is 12.14 DDD/100BD with

no signiﬁcant difference found between public and private hospitals (p > 0.05 for

all). The most commonly used antibiotics were Amoxycillin/Clavulanic acid, Ceftria-

xone, Amoxycillin and Cefuroxime for parenteral use. Consumption of beta-lactams,

Cephalosporins, Carbapenems, Monobactams and quinolones did not vary signiﬁ-

cantly by region, occupancy rate, number of beds including the number of intensive

Abbreviations: ATC/DDD, Anatomical, Therapeutic and chemical classification system/Defined Daily Dose; DDD/100BD, Defined

Daily Dose per 100 Bed-Days; WHO, World Health Organization.

∗

Corresponding author. Tel.: +961 3 476262; fax: +961 1 511488.

E-mail addresses: [email protected] (K. Iskandar), [email protected] (P.A. Hanna), [email protected]

(P. Salameh), [email protected] (E.B. Raad).

http://dx.doi.org/10.1016/j.jiph.2015.12.013

Antibiotic consumption in non-teaching Lebanese hospitals 619

care unit beds. Our data ﬁndings provides baseline information on patterns of antibi-

otic consumption in Lebanon and the issue calls for concerted efforts to encourage

data reporting on national basis and to correlate future ﬁndings with results of antibi-

otic susceptibility testing which can provide insights and tools needed to assess the

public health consequences of antimicrobial misuse and to evaluate the impact of

antibiotic resistance containment interventions.

Introduction area of hyper-endemicity for multiresistant hospi-

tal pathogens [11]. In Lebanon, there is a lack of

information concerning the level of antibiotic con-

Antibiotic resistance is a growing global health

sumption in the hospitals. The objective of the

threat of broad concern where increased antibiotic

study is to address this issue and focus on the

consumption is driving resistance [1]. Consequently,

intensity of antibiotic use in participating hospitals

antibiotics are becoming less effective or even inef-

and benchmark with published data expressed in

fective, resulting in an accelerating health security

daily divided dose per 100-bed-days (DDD/100BD) in

emergency that is rapidly outpacing available treat-

neighboring Mediterranean in particular and other

ment options [2]. Surveillance of antimicrobial

available data worldwide in general. Our aim is to

resistance tracks changes in microorganisms and

give an insight of antibiotic prescribing patterns and

allows the early detection of resistant strains of

provide a baseline data for future benchmark and

public health importance, while surveillance of

correlation with changes in antibiotic susceptibility

antibiotic consumption allows the quantiﬁcation of

testing and trends of antimicrobial resistance in the

the selection pressure on microbial populations and

hospital settings.

serves as an outcome measure of antibiotic stew-

ardship programs. According to the World Health

Organization (WHO), linking the surveillance ﬁnd-

ings to patterns of antibiotic consumption has Material and methods

proven to be a crucial factor driving political

commitment to successful resistance containment Study design

campaigns. In this context, hospitals represent ‘hot

spots’ for selective pressure on micro-organisms This is an observational cross-sectional study con-

[3] where the lack of control of antimicrobial use ducted for a period of 12-month in 2012 using

will inevitably lead to overuse, poor outcomes data from pharmacy records aggregated at hospital

and higher healthcare costs [4]. Numerous initia- level.

tives in recent years have encouraged hospitals

to conduct surveillance of antimicrobial consump- Data collection

tion in order to identify possible overuse and

misuse [5]. In fact, in high income countries, Following the approval of the Institutional Review

networks such as European Surveillance on Antibi- Board, ﬁfty two hospitals were asked to ﬁll out

otics Consumption Network database (maintained an anonymous questionnaire. Administrative data

by European Centre for Disease Control for Euro- consisted of the hospital number of beds and the

pean Union (EU) countries) [6] and resistance map occupancy rate for a period of 12 month during the

in the United States of America have enabled year 2012 allowing the determination of the number

greater understanding of antibiotic use; however, of Bed-days, a standardized ﬁgure that provides a

there are still gaps in data worldwide, especially degree of comparison among different institutions.

in resource limited settings [7] Few published Other requested data included, the hospital sta-

descriptions [5,8] or comparisons of antibiotic tus, number of intensive care unit (ICU) beds and

consumption are available [9,10,34,35] particu- the availability of a transplantation and/or oncol-

larly in the Mediterranean region, identiﬁed as an ogy unit, considered primary areas of focus due to

620 K. Iskandar et al.

high rates of antibiotic usage [12—16]. Data con- Hospital antibiotic consumption was divided

cerning the consumption of antibiotics for systemic into eight main antibiotic groups: penicillins

use were collected from the hospital pharmacy (J01C); cephalosporins, carbapenems and

computer records representing the total number monobactams (J01D); tetracyclines (J01A);

of antibiotics per unit dose per category, actually macrolides + lincosamides + streptogramins (J01F);

prescribed and consumed during the study period quinolones (J01M); sulphonamides (J01E); amino-

[31]. The validity of the data was not tested since glycosides (J01G); and ‘others’ [20], including

all collected data were retrieved from computer- (J01XA) glycopeptides, (J01XD) imidazoles, (J01XX)

ized reports however we highlight the presence of linezolid and (J01XB) polymyxin. Broad spectrum

bias due to peculiarities of prescribing systems and antibiotics included antipseudomonal penicillins

the possibility of including an element of ambula- (J01CA) fourth generation cephalosporins (J01DE),

tory care mix in the consumption record. Using ABC carbapenems (J01DH), glycopeptides (J01XD) and

Calc. software version 3.1, antibiotic consumption quinolones (J01M).

data were aggregated at the ﬁfth level of the ATC

classiﬁcation and expressed in Deﬁned Daily Doses

(DDD). Statistical analysis

Data was entered and analyzed using SPSS version

Hospital demographic data

17.0. In the descriptive analysis, frequency and

percentages were presented for nominal variables,

Twenty seven general private and public hospitals

while means and standard deviations were used for

participated in the survey from different regions in

continuous variables. For bivariate analysis, ANOVA

Lebanon. Patient’s age ranged from 18 to 85 years

was used to compare means between three groups

old. All participating hospitals have both medical

or more, while Student test was used to com-

and surgical units, but the number of beds including

pare between two groups, provided distribution was

ICU beds, the occupancy rate as well as the avail-

normal and variances were homogeneous; if not,

ability of an oncology and/or a transplantation unit

non-parametric tests were used: Kruskall—Wallis

varied.

and Mann—Whitney tests, respectively. In all cases,

a p-value <0.050 was considered statistically signif-

Antibiotic consumption icant.

Antibiotic consumption data retrieved from phar-

macy records, were collated on a standardized ﬁle

derived from the ABC Calc software version 3.1 Results

[developed by the Danish Statens Serum Institute

(http://www.escmid.org/)] on a Microsoft Excel Hospital characteristics

application (Microsoft Corporation, Redmond, WA,

USA) together with administrative data. The Excel According to the order of Hospitals, there are

application included all antibacterial drugs mar- one hundred thirty eight hospitals in Lebanon

keted in Lebanon for systemic use J01 ATC group where nine are classiﬁed as teaching hospitals.

[17,18]. As recommended by the WHO, ABC Calc Among one hundred nineteen non-teaching hos-

reports hospital antibiotic consumption as a number pitals, nineteen are considered long-term care

of DDD/100BD. settings and the remaining one hundred ten are

For a speciﬁc drug, the DDD corresponds to the short — stay acute care hospitals. Tw o types of

assumed average daily dose for its main indica- hospitals were excluded from the study: teaching

tion in adults. ABC Calc uses the 2006 version of hospitals because of extensively long and compli-

the ‘‘ATC (Anatomical Therapeutic Chemical) Index cated Institutional Board Review procedures and

with DDDs’’. Antibiotics were recorded by their long-term stay hospitals. Among 52 randomly con-

non-proprietary names. Each drug was then given tacted hospitals from different regions in Lebanon

its chemical name and a code according to the ATC including Beirut, Mount Lebanon, Bekaa, North and

classiﬁcation that comprises 5 levels. The ﬁrst level South Lebanon, only 27 general hospitals partici-

is the anatomical group (e.g. anti-infective for sys- pated in the survey among which 89% were private

temic use), the second is the therapeutic group and 11% public hospitals. All hospitals required to

(antibiotics for systemic use), the third is a ther- remain anonymous and refused to give any data

apeutic subgroup, the fourth gives the chemical concerning their annual report of susceptibility pro-

form and the ﬁfth is a chemical subgroup. For the ﬁles of bacterial strains. Hospitals participating in

purpose of our study, all ﬁve levels were used [19]. the survey were from different urban regions in

Antibiotic consumption in non-teaching Lebanese hospitals 621

Table 1 Hospitals description and characteristics. Table 2 Antibiotic utilization.

Characteristic Frequency (%) M (SD)

Region N = 27 Antibiotic use

Beirut 4 (14.8%) DDD/100 bed-days 72.56 (18.08)

Bekaa 5 (18.5%) Beta lactam 19.04 (6.75)

Mount Lebanon 13 (48.1%) Cephalosporins, monobactam & 27.41 (8.55)

North Lebanon 2 (7.4%) carbapenems

South Lebanon 3 (11.1%) Aminoglycosides 3.31 (1.89)

Carbapenems 5.12 (2.52)

Oncology unit

Penicillins with antipseudomonal 2.15 (1.27)

Absent 8(29.6%) activity/BLinh

Present 19(70.4%)

Fourth generation cephalosporins 2.27 (1.64)

Transplantation unit Glycopeptides 2.61 (1.56)

Absent 26(96.3%) Total Broad spectrum 12.14 (5.25)

Present 1(3.7%) Most commonly used drug

Amoxicillin/clavulanate IV 16 (59.3%)

Number of beds

Ceftriaxone IV 9 (33.3%)

80 beds or less 10(37.0%)

Amoxicillin IV 1 (3.7%)

81 to 120 beds 9(33.3%)

Cefuroxime IV 1 (3.7%)

More than 120 beds 8(29.6%)

Statistical analysis: descriptive analysis, frequency and

Occupancy rate

percentages are presented for nominal variables and means

53% or less 9(33.3%)

and standard deviations are used for continuous variables;

54 to 70% 10(37.0%)

p-value <0.05 is considered statistically signiﬁcant.

More than 70% 8(29.6%)

ICU beds

consumption between public and private hospitals

6 ICU beds or less 10(37.0%)

(p > 0.05 for all).

7 to 8 ICU beds 9(33.3%)

The consumption of beta-lactams,

9 beds or more 8(29.6%)

cephalosporins, carbapenems,monobactams

and quinolones did not vary signiﬁcantly by region,

occupancy rate, number of beds including the num-

Lebanon, 48% from Mount Lebanon, 19% from the

ber of ICU bed or the availability of an oncology or

Bekaa region, 15% from Beirut, 11% from the South

a transplantation unit.

and 7% from the North. Among them, 37% of hospi-

The consumption of antipseudomonalpeni-

tals have less than 80 beds and 63% have more than

cillinsvaried by region with the highest value noted

7 ICU beds while 70% have an oncology unit and

in Beirut (p = 0.044). In addition, the use of amino-

96% do not have a transplantation unit. The occu-

glycosides (p = 0.034), carbapenems (p = 0.042)

pancy rate was less than 53% in 33% of the hospitals

and glycopeptides (p = 0.019)increased signiﬁcantly

and ranged between 54% and 70% in 37% of hospi-

when the occupancy rate decreased in the hospitals

tals while it was 70% in more than 30% of hospitals (Table 3).

(Table 1).

Hospital characteristics associated with Discussion

antibiotic consumption

In Lebanese non-teaching hospitals, our data

The average antibiotic consumption in the hos- showed that antibiotic consumption did not vary

pitals is 72.56 DDD/100 BD with a total broad signiﬁcantly by region, occupancy rate, and num-

spectrum antibiotic consumption of 12.14 DDD/100 ber of beds including number of ICU bed or the

BD. The most commonly used antibiotics were availability of an oncology or a transplantation unit.

Amoxycillin/Clavulanic acid, Ceftriaxone, Amoxy- Results from our survey demonstrated that

cillin and Cefuroxime for parenteral use with average antibiotic consumption in Lebanese non-

Amoxycillin/Clavulanic acid consisting of 59% of teaching hospitals excluding pediatric cases was

total use (Table 2). 72.55 DDD/100 BD. This value was lower com-

The average antibiotic consumption in public pared to the ﬁndings of the ARMed project [21]

hospitals is 66 DDD/100 BD whereas it is 73 DDD/100 that benchmarked antibiotic use prospectively

BD in private hospitals with a p value of 0.51. No in hospitals from southern and eastern Mediter-

signiﬁcant difference was found for any antibiotic ranean countries and showed that the median total

622 K. Iskandar et al. (2.73) (1.46) (0.85) (1.63) (0.41) (1.74) (1.83) (1.55) (1.79) (1.34) (1.67) (1.82) (1.32) (0.72) (1.92) (1.00) (1.49)

0.019

variables; =

p J01XA 2.92 2.30 2.03 1.68 1.93 4.50 2.54 2.73 2.39 2.70 3.71 2.30 1.75 2.80 1.93 3.13

continuous (1.73) 3.33 (1.71) (1.27) (1.10) (1.62) (1.59) 2.90 (1.83) (1.55) (1.92) (1.41) (1.67) (1.65) (1.65) (1.82) (1.90) (0.58) (1.74)

0.050 0.073

for = =

p J01DE 2.35 0.90 1.07 2.64 p 5.40 2.15 2.55 2.03 2.18 2.08 2.45 2.25 2.66 1.27 2.90

used

are

(1.28) 3.13 (1.20) (1.70) (0.40) (0.69) (1.38) 2.41 (0.90) 1.94 (1.27) (1.55) (0.91) (1.35) (1.68) (1.08) (1.00) (1.53) (1.15) (1.02)

0.044

p J01CA 1.20 1.57 1.26 3.50 2.10 1.91 2.30 2.28 2.40 2.24 1.75 2.06 1.81 2.64 deviations

(2.56) 3.50 (2.14) 2.35 (2.33) (1.74) (3.73) (2.00) 2.33 (3.62) 1.71 (2.36) (2.38) (2.93) (2.50) (2.36) (2.44) (2.03) (2.18) (1.78) (3.47)

0.050 0.041 0.042

standard

= = =

p p J01DH 6.35 2.90 4.48 p 10.10 4.92 4.91 5.45 6.23 5.53 3.34 4.79 4.54 6.16 and

means

(5.21) 5.49 (5.93) (4.97) 5.26 (6.30) 4.76 (5.31) (5.59) 5.03 (5.35) (4.85) (5.93) (4.87) (5.55) (5.54)

(4.72) 5.73 (0.35) (0.96) (4.22)

(4.59)

and

J01M 7.35 6.40 12.70 16.10 11.12 10.36 13.46 9.5 10.75 8.76 13.19 12.36

variables

(2.47) 8.70 (2.06) 13.31 (1.56) (1.25) (0.94) (1.77) 11.58 (2.18) 10.64 (1.91) (2.03) 10.43 (1.84) (1.74) (1.76) 13.80 (2.17) (0.61) (2.00) (2.05) (1.00)

0.034

nominal p J01G 2.00 2.48 2.10 3.35 2.94 2.88 3.49 1.99 4.19 2.25

for

(12.92) 4.28 (5.88)(7.78) 3.66 (11.59) 3.10 (9.68) (7.31)(11.58) 3.09 3.83 (8.71) (11.10) 3.98 (6.62) (7.60) (10.80) 4.28 (5.36) (7.23) (10.81) 3.36 (6.81) (7.94)

0.092 presented

p J01D 26.03 24.02 25.80 27.47 28.86 22.08 28.96 are

(2.81)(8.17) 23.35 (3.96) 29.03 (1.19) 35.50 (6.90) (5.27)(9.31) 27.46 27.28 (6.88) (5.20)(4.12) 26.05 (9.48) 27.62 (8.62)(5.14) 30.72 (5.19) 28.69 (5.18)(9.82) 26.11 (3.79) 27.47

percentages a

16.83 20.40 21.00 19.96 20.19 J01C and

(7.15)(4.49) 14.45 (0.21) 20.73 17.10 (5.78) (4.90)(5.93) 17.93 21.66 (4.82) (5.33)(4.87) 16.44 18.64 (5.41)(5.12) 22.52 17.50 (5.34) 16.91 (6.01) broad

(4.22) (6.17) 22.73 (4.17) 17.04 (3.32) 20.38

frequency

0.024

characteristics. =

Total 10.06 23.50 11.70 14.83 spectrum p signiﬁcant

analysis,

BD hospital

(28.72) 15.68 (16.62) 13.12 (8.84)(9.49) 10.75 (16.11) 7.57 (14.36) 12.90 (25.77) 10.34 (18.41) (19.11) 12.22 (16.02) 11.63 (17.95) 12.6 (16.52) 14.42 (21.20) 12.52 (8.15) 9.09 (15.36) 12.31 (16.50) 9.56 (22.49)

and

statistically

0.057

= Descriptive

use

71.84 p DDD/100 83.33 73.48 76.08 77.40 72.37 considered

less 72.78 analysis:

rate beds or less 79.38

120 70% 62.85

Antibiotics

beds 65.61 more 79.01 unit

beds 66.57 of or

<0.05 or less 83.20

Lebanon 69.45

LebanonLebanon 55.10 73.26 3

beds 120 70% 70.74

than than

ICU or beds

Statistical

beds to to

a beds unit ICU Beds

p-value 80 Table Characteristic Region Present 71.07 Bekaa Mount North South Oncology Absent Transplantation Number 81 More Occupancy 54 More ICU 7—8 9 Present 6 Beirut Absent 53%

Antibiotic consumption in non-teaching Lebanese hospitals 623

antibiotic use was 112 DDD/100BD, with an inter- countries and showed that the median total antibi-

quartile range of 84—428 DDD/100BD. This may otic use was 112 DDD/100BD, with an inter-quartile

be due to the hospitals type included in the study range of 84—428 DDD/100BD. This may be due to

or to a more rigid implementation of antibiotic the hospitals type included in the study or to a

restriction programs as stressed by the hospital more rigid implementation of antibiotic restriction

accreditation process undertaken by the Ministry programs as stressed by the hospital accredita-

of Health in Lebanon in 2011. Nevertheless, our tion process undertaken by the Ministry of Health

ﬁnding is comparable to the USA 2002—2003 study in Lebanon in 2011. Nevertheless, our ﬁnding

(79 DDD/100BD) [22], while in Europe antibiotic is comparable to the USA 2002—2003 study (79

use was lower as reported in France by Dumartin DDD/100BD) [22], while in Europe antibiotic use

et al., 37—39 DDD/100BD [23] and 43.5 RDD/100 was lower as reported in France by Dumartin et al.,

patient days (median) with an interquartile range 37—39 DDD/100BD [23] and 43.5 RDD/100 patient

of 36—48 RDD/100 — corresponding to a median days (median) with an interquartile range of 36—48

of 64.4 DDD/100 (interquartile range, 53—73 RDD/100 — corresponding to a median of 64.4

DDD/100) in a published study in Germany in 2015 DDD/100 (interquartile range, 53—73 DDD/100) in a

[29]. In Sweden a range of 56—59 DDD/100BD in published study in Germany in 2015 [29]. In Sweden

[24] and 58—91 DDD/100BD in Denmark [25] and a range of 56—59 DDD/100BD in [24] and 58—91

ﬁnally 50—70 DDD/100BD in the Netherlands [26]. DDD/100BD in Denmark [25] and ﬁnally 50—70

In addition, our data ﬁndings showed that DDD/100BD in the Netherlands [26].

regional differences only affected the consump- In addition, our data ﬁndings showed that

tion of Penicillins with antipseudomonal activity regional differences only affected the consump-

and that the decrease in the occupancy rate tion of Penicillins with antipseudomonal activity

was inversely proportional to the consumption of and that the decrease in the occupancy rate

Carbapenems, Glycopeptides and aminoglycosides. was inversely proportional to the consumption of

Broad spectrum antibiotic use, speciﬁcally car- Carbapenems, Glycopeptides and aminoglycosides.

bapenems and fourth generation cephalosporins Broad spectrum antibiotic use, speciﬁcally car-

increased signiﬁcantly with the availability of a bapenems and fourth generation cephalosporins

transplantation unit, a service only available in <4% increased signiﬁcantly with the availability of a

of participating hospitals, while only Carbapenems transplantation unit, a service only available in <4%

use was signiﬁcantly higher in case of availability of of participating hospitals, while only Carbapenems

an oncology unit. use was signiﬁcantly higher in case of availability of

Number of beds, including the number of ICU an oncology unit.

beds did not signiﬁcantly affect the consumption Number of beds, including the number of ICU

of antibiotics including broad spectrum antibi- beds did not signiﬁcantly affect the consumption

otics which means that antibiotic consumption data of antibiotics including broad spectrum antibi-

should preferably be collected at ward level in otics which means that antibiotic consumption data

order to have more accurate results to benchmark. should preferably be collected at ward level in

order to have more accurate results to benchmark.

Discussion Current and future developments

Currently, the Lebanese Ministry of Health in collab-

Limitations

oration with the WHO is mandating the hospitals to

submit their yearly susceptibility testing results and

In Lebanese non-teaching hospitals, our data

to harmonize antimicrobial breakpoints according

showed that antibiotic consumption did not vary

to international guidelines. It will be important to

signiﬁcantly by region, occupancy rate, and num-

correlate reported data with patterns of antibiotic

ber of beds including number of ICU bed or the

consumption in order to monitor progress toward a

availability of an oncology or a transplantation unit.

more prudent antibiotic use [29,32].

Results from our survey demonstrated that

average antibiotic consumption in Lebanese non-

teaching hospitals excluding pediatric cases was

72.55 DDD/100 BD. This value was lower com- Conclusion

pared to the ﬁndings of the ARMed project [21]

that benchmarked antibiotic use prospectively in Our study provides baseline information on antibi-

hospitals from southern and eastern Mediterranean otic consumption and emphasis on the need to

624 K. Iskandar et al.

generalize the ﬁndings to encompass all hospitals [9] Vander Stichele RH, Elseviers MM, Ferech M, Blot S,

teaching and non-teaching through the implemen- Goossens H. Hospital consumption of antibiotics in 15

European countries: results of the ESAC Retrospective

tation of a national database for the periodic

Data Collection (1997—2002). J Antimicrob Chemother

surveillance and benchmark of antibiotic consump-

2006;58(July (1)):159—67.

tion and its correlation with changes in patterns of

[10] Vincent JL, Rello J, Marshall J, et al. International study of

antibiotic susceptibility testing. Continuous moni- the prevalence and outcomes of infection in intensive care

toring of antibiotic use will eventually results in units. J Am Med Assoc 2009;302, 2323e2329.

[11] Gur D, Unal S. Resistance to antimicrobial agents in Mediter-

optimizing antibiotic prescription and the imple-

ranean countries. Int J Antimicrob Agents 2001;17:21—6.

mentation of national strategies to ﬁght antibiotic

[12] Vincent JL, Bihari DJ, Suter PM, et al. The prevalence

resistance and improve patient safety.

of nosocomial infection in intensive care units in Europe:

results of the European Prevalence of Infection in Intensive

Care (EPIC) Study EPIC International Advisory Committee. J

Funding Am Med Assoc 1995;274:639—44.

[13] Hanberger H, Garcia-Rodriguez JA, Gobernado M, et al.

Antibiotic susceptibility among aerobic gram-negative

None.

bacilli in intensive care units in 5 European countries.

French and Portuguese ICU Study Groups. J Am Med Assoc

1999;281:67—71.

[14] Archibald L, Phillips L, Monnet D, McGowan Jr JE, Ten-

Competing interests

over F, Gaynes R. Antimicrobial resistance in isolates from

inpatients and outpatients in the United States: increas-

None declared.

ing importance of the intensive care unit. Clin Infect Dis

1997;24:211—5.

[15] Nourse C, Murphy H, Byrne C, et al. Control of a nosocomial

outbreak of vancomycin resistant Enterococcus faecium in

Ethical approval

a pediatric oncology unit: risk factors for colonization. Eur

J Pediatr 1998;157:20—7.

Not required.

[16] Kollef MH. Inadequate antimicrobial treatment: an impor-

tant determinant of outcome for hospitalized patients. Clin

Infect Dis 2000;31(Suppl 4):S131—8.

Acknowledgments [17] Hutchinson JM, Patrick DM, Marra F, et al. Measurement of

antibiotic consumption: a practical guide to the use of the

Anatomical Therapeutic Chemical classiﬁcation and Deﬁned

We thank all the healthcare professionals in hospi- Daily Dose system methodology in Canada. Can J Infect Dis

tals involved in the Survey. 2004;15(January—February (1)):29—35.

[18] WHO Collaborating Centre for Drug Statistics Methodol-

ogy. ATC Index with DDDs. Retreived from: http://www.

References whocc.no/ [accessed on 17.04.10].

[19] Dumartin C, L’Heriteau F, Pefau M, et al. Antibiotic use in

530 French hospitals: results from a surveillance network at

[1] Laxminarayan R, Heyman DL. Challenges of drug resistance

hospital and ward levels in 2007. J Antimicrob Chemother

in the developing world. Br Med J 2012;344:e1567. 2010;65(9):2028—36.

[2] WHO. Antimicrobial resistance: global report on surveil-

[20] Peˇsić G, Jović Z, Vasić K. Application of the ATC/DDD

lance; 2014. www.who.int/drugresistance/documents/

methodology to compare antibiotic utilization in two

surveillancereport/en/.

university hospital surgical departments. Med Biol

[3] Lopez-Lozano JM, Monnet DL, Yague A, et al. Modelling and 2005;12(3):174—8.

forecasting antimicrobial resistance and its dynamic rela-

[21] Borg MA, Scicluna E. Antibiotic resistance in the Mediter-

tionship to antimicrobial use: a time series analysis. Int J

ranean region: the AR Med project. J Antimicrob Chemother

Antimicrob Agents 2000;14, 21e31.

2008;62(October (4)):830—6.

[4] WHO. How to investigate antimicrobial use in hospi-

[22] Polk RE, Fox C, Mahoney A, Letcavage J, MacDougall C. Mea-

tals: selected indicators; 2012, February, apps.who.int/

surement of adult antibacterial drug use in 130 US hospitals:

medicinedocs/./s21031en.pdf.

comparison of deﬁned daily dose and days of therapy. Clin

[5] Carrie AG, Metge CJ, Zhanel GG. Antibiotic use in

Infect Dis 2007;44:664—70.

a Canadian province, 1995—1998. Ann Pharmacother

[23] Cizman M, Slovenian Consumption Study Group. Nationwide

2000;34:459—64.

hospital antibiotic consumption in Slovenia. J Antimicrob

[6] Adriaenssens N, et al. European Surveillance of Antimi-

Chemother 2011;66:2189—91.

crobial Consumption (ESAC): outpatient antibiotic

[24] SWEDRES. A report on Swedish antibiotic utilisation and

use in Europe (1997—2009). J Antimicrob Chemother

resistance in human medicine; 2009. http://www.strama.

2011;66(December (Suppl 6)), vi3—12. se/uploads/docs/Swedres%202009%20ﬁnal%20version.pdf.

[7] Laxminarayan R, Van Boeckel TP. The value of track-

[25] DANMAP. Use of Antimicrobial Agents and Occurrence of

ing antibiotic consumption. Lancet Infect Dis 2014;14(May

Antimicrobial Resistance in Bacteria From Food Animals

(5)):360—1.

Foods and Humans in Denmark; 2009.

[8] Janknegt R, Oude Lashof A, Gould IM, van der Meer JW.

[26] NETHMAP. Consumption of Antimicrobial Agents and Antimi-

Antibiotic use in Dutch hospitals 1991—1996. J Antimicrob

crobial Resistance Among Medically Important Bacteria in

Chemother 2000;45:251—6.

The Netherlands; 2010.

Antibiotic consumption in non-teaching Lebanese hospitals 625

[29] Kern WV, Fellhauer M, Hug M, Hoppe-Tichy T, Först G, [34] Schweickert B, Eckmanns T, Bärwolff S, Wischnewski

Steib-Bauert M, de With K. Recent antibiotic use in N, Meyer E. Surveillance of antibiotic consumption in

German acute care hospitals — from benchmarking to hospitals: tasks of the Public Health Service. Bundes-

improved prescribing and quality care. Dtsch Med Wochen- gesundheitsblatt Gesundheitsforschung Gesundheitsschutz

schr 2015;140(November (23)). 2014;57(April (4)):399—405.

[31] Tan C, Ritchie M, Alldred J, Daneman N. Validating hospital [35] Buccellato E, Biagi C, Melis M, Lategana R, Motola D, Vac-

antibiotic purchasing data as a metric of inpatient antibiotic cheri A. Use of antibacterial agents in Italian hospitals:

use. J Antimicrob Chemother 2015, November. a 2004 to 2011 drug utilization survey in the Emilia-

[32] Reddy SC, Jacob JT, Varkey JB, Gaynes RP. Antibiotic use in Romagna region. Expert Rev Anti Infect Ther 2014;12(March

US hospitals: quantiﬁcation, quality measures and steward- (3)):383—92.

ship. Expert Rev Anti Infect Ther 2015;13(July (7)):843—54.

Available

online at www.sciencedirect.com ScienceDirect