In the last decade, concerns about inappropriate antibiotic use (AU) and emergence of antibiotic resistance have led to the establishment of formal antibiotic stewardship (AS) programs based on recommendations made by the Infectious Diseases Society of America.1
A Canadian group published a consensus paper on 4 standard metrics suitable for pediatric AS, including days of therapy per 1000 patient-days, total antibiotic days, 30-day readmission rates and adherence to AS recommendations.2 However, variations in information systems across hospital sites make capturing comparable data to achieve these metrics challenging. Nevertheless, benchmarks for AU in pediatric hospitals remain an important goal.3 Point prevalence surveys (PPS) of antimicrobial use represent a potential convenient method to measure AU across pediatric institutions with variable clinical information systems.4,5
The objectives of the current study were to describe and compare AU between all Canadian pediatric hospitals in both the pediatric and neonatal units and assess the appropriateness of AU for community-acquired pneumonia (CAP) for which guidelines exist.6–8
MATERIALS AND METHODS
Participating hospitals are all part of the Pediatric Investigators Collaborative Network on Infections in Canada. All hospitals are publicly funded, have emergency departments and have full-time pediatric infectious disease specialists. The 15 sites provide secondary and tertiary care capacity with specialists, cancer care and neonatal units. Five sites, designated as “quaternary” also provide specialized services such as bone marrow, liver and/or heart transplants through a referral basis. Two sites had ≤50 beds, 10 sites had 90–150 beds and 3 sites had >200 beds. Twelve neonatal intensive care units (NICU) were mixed units with both inborn and outborn patients and are a mix of level 2, 3 and 4 nurseries. Three NICUs, who accept transfers only from other institutions, were considered outborn only and are level 3 and 4 units.
Two single-day PPS of AU were conducted in each of the 15 participating hospitals; between November 6–13, 2018 and February 12–18, 2019. All inpatients younger than 18 years of age who were admitted (excluding patients in birthing units and psychiatric wards) were included. Infectious disease physicians and site pharmacists collected the data on the day of the survey. Data sources included paper-based and electronic medical charts. Using the case report form developed by investigators for a previous pilot study, data were entered into a centralized RedCap database (see Supplemental Digital Content 1, http://links.lww.com/INF/E385).
Data collected included admitting diagnosis, age, admitting service, documented pathogen (if available) and antibiotic(s) prescribed. For each prescribed antibiotic, the route, start date, indication for use and targeted site of infection was recorded. For the analysis, we excluded drugs only used for treatment of tuberculosis (eg, isoniazid), antivirals and topical or inhalational routes of administration. Oral and intravenous antibiotics and antifungals were grouped together as systemic antimicrobials.
Antibiotics were then grouped according to the 2019 World Health Organization (WHO) AWaRe classification (https://adoptaware.org).9,10 The core Access group is considered narrower spectrum with less potential for toxicity and includes among others penicillins, aminopenicillins, cloxacillin, first-generation cephalosporins, aminoglycosides trimethoprim-sulfamethoxazole and metronidazole. The Watch group (eg, second and third-generation cephalosporins, piperacillin-tazobactam, carbapenems, quinolones, macrolides, and glycopeptides) is considered broader spectrum with more risk of toxicity and development of resistance. The Reserve groups (eg, daptomycin, polymyxins, linezolid, fourth- or fifth-generation cephalosporins) are considered to have higher resistance potential or toxicity concerns.10
Empiric therapy was defined as therapy given for a clinical syndrome (eg, CAP or febrile neutropenia) where an etiologic agent had not yet been identified whereas targeted therapy was defined as therapy given for an identified pathogen. These were combined for the analysis. Prophylaxis was defined as being given for prevention of infection. Assessment of appropriateness of therapy was not done.
One day of therapy (DOT) was defined as receipt of at least one dose of an antibiotic on 1 day. To determine median length of therapy (as of the day of the survey), the DOT per patient (date of survey minus start date plus one for each antibiotic) was used to determine the median DOT per site. Days admitted were defined as the difference between the date of survey completion minus the date of admission plus 1 day. Two sites were not able to provide the full date of admission and thus total days admitted could not be calculated for those sites. Data on infection or colonization with antibiotic-resistant organisms were collected only during the second survey.
Approval from the local Research Ethics Boards at all sites was obtained before conducting the PPS.
All statistical analyses were performed using R statistical software version 184.108.40.206 Means with SDs for normally distributed continuous variables and median with interquartile range (IQR) for non-normally distributed variables were used. Categorical variables were described using frequencies and percentages. A Wilcoxon test was performed to test for differences in the days admitted by center type (quaternary vs. tertiary and outborn vs. inborn). A χ2 test was performed to test for differences in proportion of empiric and pathogen-directed AU as well as Access and Watch group antibiotics use between center types. Funnel plots were used to examine variability in proportion of patients receiving antimicrobials between sites.12 A horizontal line depicts the percentage of antimicrobial use amongst the entire cohort. The 95% and 99.8% approximate control limits around this line are used to account for the sample size at each site.
Medical, Surgical and Intensive Care Services (Excluding Neonatal Intensive Care Units)
Over the 2 PPS days, 3924 patients-days were surveyed. In total, 2729 (69.5%) children were admitted to medical/surgical units, or pediatric intensive care units (PICUs) of which 1850 (68.1%) had an underlying illness. The median age was 4.0 years (IQR 0.7–11.0). The median number of days of admission before the survey day was 6 (IQR 3.0–17.0) with little evidence of a difference between tertiary and quaternary sites (P = 0.14). Patients admitted to general pediatric services constituted the biggest group (n = 1125, 41.2%), with 327 (12%) to pediatric intensive care, 739 (27.1%) to medical subspecialties and 538 (19.7%) to surgical services. In total, 179 (6.5%) patients reported an allergy to penicillin.
Of the 2729 patients, 1210 (44.3%) received antimicrobials; 762 (63.0%) patients received 1 antimicrobial, 301 (24.9%) received 2, 125 (10.3%) received 3, 19 (1.6%) received 4 and 3 (0.2%) received 5 antimicrobials. Among the 1210 patients who received antimicrobials, 386 (31.9%) received them for prophylaxis and 945 (78.1%) received them for empiric or pathogen directed therapy. Of the 945 who received for empiric or pathogen directed treatment, 165 (17.5%) patients received oral antimicrobials. Of the 181 oral antimicrobials given to these patients, the most commonly administered were: aminopenicillin/penicillins (n = 55), amoxicillin-clavulanic acid (n = 24), cephalexin (n = 20), azithromycin/clarithromycin (n = 17), antifungals (n = 14), fluoroquinolones (n = 12) and metronidazole (n = 10).
Overall, a total of 1830 antimicrobials were prescribed. Of these, 1352 (73.9%) were given for empiric or pathogen-directed therapy. Clinical conditions for which antimicrobials were prescribed are listed in Table 1. For 9 antimicrobials, the reason (“empiric or pathogen-directed or prophylaxis”) was missing. Overall, 469 (25.6%) antimicrobials were given for prophylaxis and of these, the most commonly prescribed were: trimethoprim-sulfamethoxazole n = 183 (39.0%), antifungals n = 104 (22.2%) and first-generation cephalosporins, n = 76 (16.2%).
TABLE 1. -
Clinical Indications for Which Empiric or Pathogen Directed Antimicrobials Are Being Prescribed in Non-NICU Patients
||Total Number of Antimicrobials Given for This Clinical Indication
||Proportion of the 945 Patients Receiving an Antimicrobial for This Clinical Indication, n (%)
|Lower respiratory tract infection
|Skin, soft tissue infection
|Urinary tract infection
|Central nervous system infection
|Upper respiratory tract infection (eg, sinusitis)
|Device related infection
A total of 1371 antibiotics were given for these clinical indications.
Among the 5 quaternary sites, the overall proportion of children who received antimicrobials for empiric or pathogen-directed treatment was 489/1514 (32.3%) and ranged from 28.7% to 41.8%. Among the 10 tertiary sites, the overall proportion of children who received antimicrobials was higher, 456/1215 (37.5%) and varied from 33.3% to 44.4% (P = 0.005).
The proportion of Access group antibiotics at quaternary sites was 213/489 (43.6%) and ranged from 37.1% to 46.5%. The proportion of Access group antibiotics at tertiary care sites was 220/456 (48.2%), in comparison with quaternary centers (P = 0.17) and varied between 32.7% and 60% (Fig. 1). The proportion of Watch group antibiotics at quaternary sites was 313/489 (64.0%) (varied between 61.3 % and 67.2 %). The proportion of Watch group antibiotics at tertiary care sites was 287/456 (62.9%), (varied between 39.7% and 80%) and was no different from that observed in quaternary centers (P = 0.78) (Fig. 1).
Figure 2 illustrates the median days of Access and Watch group antibiotics use by site. The overall median (IQR) days of Access and Watch therapy were 4.0 (2.0–7.0) and 4.0 (2.0, 8.0), respectively, and were similar for quaternary and the tertiary centers.
In addition to antibiotics that had been classified as Access or Watch group, a small number of patients received a Reserve group antibiotics: linezolid or daptomycin (n = 10) and colistin (n = 3).
Neonatal Intensive Care Units
Of the 1195 patients admitted to the NICU, both mixed and outborn units, 236 (19.7%) received at least 1 antimicrobial; 92 (39.0%) received 1, 117 (49.6%) received 2, 24 (10.2%) received 3 and 3 (1.3%) received 4 antimicrobials. Among the 236 who received antimicrobials, 57 (24.2%) patients received antimicrobials for prophylaxis and 196 (83.1%) received antimicrobials for empiric or pathogen-directed therapy.
Of the 410 antimicrobials prescribed, 336 (81.9%) were given for empiric or pathogen-directed therapy and 70 (17.1%) were for prophylaxis. For 3 antimicrobials, the reason for prescribing was missing. Of the 70 antimicrobials prescribed for prophylaxis, the most common types were: antifungals 28 (40.0%), penicillins or aminopenicillins 14 (20.0%), and aminoglycosides 7 (10.0%).
Of the 196 (16.4%) NICU patients receiving antimicrobials for empiric or pathogen directed reasons, the most common clinical indications for therapy were fever without a source, intra-abdominal infections and bloodstream infections 71 (36.2%), 35 (17.9%) and 33 (16.8%) of patients, respectively.
Among the 3 outborn NICU sites, the proportion of infants who received antimicrobials for empiric or pathogen-directed treatment was 35/131 (26.7%) and ranged from 8% to 35%. Among the 12 mixed NICU units, the proportion was lower [161/1064 (15.1%)] with a range of 5.6%–31.4% (P = 0.001).
The overall proportion of Access group antibiotics was 125/196 (63.8%) (Fig. 3). The proportion of Access group antibiotic at outborn units was 22/35 (62.9%) and 103/161 (64.0%) at mixed NICU units (P = 1.00). The overall proportion of Watch group antibiotics was 89/196 (45.4%) (Fig. 3). The proportion of Watch group antibiotics at outborn units was 16/35 (45.7%) and 73/161 (45.3%) at mixed NICU units (P = 1.00). The overall median (IQR) days of Access group antibiotics was 5.0 (3.0–8.0) and 7.0 (4.0–11.0) days for Watch group antibiotics. Median (IQR) days Access group antibiotics in outborn units was 4.5 (3.0–6.0) and 5.0 (3.0–8.0) in mixed units. Median (IQR) days of Watch group antibiotics in outborn units was 5.0 (4.0–9.0) and 7.0 (3.0–12.0) in mixed units.
Of the 2729 patients surveyed, 153 (5.6%) had a diagnosis of CAP. Antibiotics prescribed for this indication varied among hospitals (Fig. 4). Overall, 50 (32.7%) patients received only penicillin or aminopenicillins, 8 (5.2%) received only azithromycin or clarithromycin, 48 (31.4%) received a third-generation cephalosporin and 24 (15.7%) received another antibiotic.
Other antibiotics used for CAP included amoxicillin-clavulanic acid, second-generation cephalosporins, clindamycin, piperacillin-tazobactam, carbapenems and vancomycin. Most patients treated for CAP were admitted to general pediatrics [109 (71.2%)] with 31 (20.3%) requiring PICU admission. Of those admitted to PICU, a total of 39 antimicrobials were given; 16 (51.6%) patients received only a third-generation cephalosporin, 3 (9.7%) received only amoxicillin/ampicillin/penicillin, 1 (3.2%) received only azithromycin, 6 (19.4%) received a third-generation cephalosporin in combination with 1 or 2 other antibiotics and 5 (16.1%) received other antibiotics.
Among those who were part of the second PPS, 36/606 (5.9%) were colonized or infected with methicillin-resistant Staphylococcus aureus, Enterobacter cloacae, an extended-spectrum beta-lactamase-producing Gram-negative organism or vancomycin-resistant enterococcus. A total of 513 (84.7%) were not known to be colonized or infected while the status was unknown for 57 (9.4%) patients.
Comparing AU between hospitals for the purposes of “benchmarking” is poses challenges due to different information systems hospitals use for collecting days of therapy.3,13 Significant variability among children’s hospitals has been documented in the United States but benchmarks have not yet been defined.14 PPS in adult sites performed concurrently with determination of defined daily doses have been shown in 1 study to correlate.15 We chose to focus mainly on measurement of empiric or pathogen-directed antimicrobial therapy as this is a current target of most pediatric ASP programs. Prophylaxis is often driven by specialty guidelines and varies by types of patient populations.
This multisite PPS of Canadian pediatric hospitals revealed that among tertiary and quaternary care hospitals surveyed, the overall proportion of children who received antimicrobials for therapy was 34.6% and varied from 28.7% to 44.4%. However, the proportion was slightly higher at tertiary care sites, 37.5% and 32.3%, respectively, possibly due to different patient populations. A prior PPS from the Canadian Nosocomial Infection Surveillance Program performed between 2002 and 2009 noted that approximately 40% of patients in pediatric hospitals received antimicrobials.16 This overall proportion is comparable with the 33%–60% reported in several studies from the United States, 40.9% in a United Kingdom study and 46% in an Australian study.14,17–20
The use of broad-spectrum antibiotics, including the use of third-generation cephalosporins has been increasing in inpatients (including children) in the United States, and are considered part of the WHO “Watch” group of antibiotics but not the core Access group.10,19,21 Our study identified that Watch group antibiotics accounted for 67.5% of all antibiotics used in hospitalized children. This proportion is concerning as it is higher than what is reported for the majority of the countries in the WHO report.22 We postulated that since tertiary care centers in Canada potentially would have less complicated patients compared with quaternary care centers, their use of Watch group antibiotics would be lower, however, the proportions were no different between tertiary and quaternary centers. The reasons for overall similar prescribing patterns between quaternary and secondary/tertiary pediatric hospitals are not known but could reflect similar teaching and prescribing cultures, fear of undertreating a possible resistant pathogen, imprecision in initially defining clinical syndromes, and less priority given to potential toxicity and local antimicrobial susceptibility patterns. A pediatric PPS in Australia noted that half of antimicrobials were broad-spectrum, and postulated that this may be due to lack of guidelines for specific clinical syndromes.20 Other studies have also shown that smaller or secondary community hospitals and larger or tertiary care hospitals have similar use of broad-spectrum antibiotics despite a less complex patient population.23,24
Although only noted in the second survey, the proportion of patients known to be colonized with a resistant pathogen was only 5.9%, thus probably not justifying such a large percentage of broad-spectrum (or Watch group) AU. Development of guidelines that are flexible enough to encourage use of narrower spectrum agents could potentially help guide less broad-spectrum empiric therapy in patients who do not have central nervous system infections or who are not critically ill.
Duration of therapy also has an impact on resistance and may be influenced by comprehensive ASP. To estimate a duration of therapy, we compared calculated median days of therapy for both core Access and Watch group antibiotics. Most guidelines recommend narrowing antibiotic spectrum if a resistant pathogen is not identified but the median days of therapy were similar for both narrow and broad-spectrum antibiotics (4 days). The factors which affect duration of Watch group antibiotics could be related to a delay in narrowing the spectrum of antibiotics when prescribing for clinical syndromes (in the setting lack of definitive microbiology) or variability in discharging patients with outpatient intravenous therapy. This metric has not previously been reported in PPS but could be used as a benchmark of broad-spectrum (Watch group) AU.
Given the small number of NICU patients and the large variability in percentage of patients receiving antibiotics, it is difficult to make robust comparisons for this group. Among large numbers of NICUs in California, the benchmark for antibiotic usage was estimated at approximately 14% which is lower than the overall 19% in this study.25 The variability in the percentage of infants receiving Watch group antibiotics was large given the small sample sizes; 4 mixed sites were above the 45.4% overall proportion of Watch group antibiotics. Compared with the Australian data where only 4% of prescriptions in the NICU were inappropriate, this data suggests that appropriateness should be assessed in these settings.26 It is unclear whether different benchmarks should apply to these different patient populations as not enough data is available. Recent studies have noted that longer duration of antibiotics in newborns may contribute to unfavorable outcomes.27,28
In Canada, the Canadian Pediatric Society has published a practice point for management of CAP.6,7 Our PPS revealed that only 32.7 % of patients diagnosed with CAP were given only aminopenicillins or penicillin alone and practice was extremely variable among centers. This is surprising given the Canadian Pediatric Society's recommendation for ampicillin in the majority of cases (reserving empiric third-generation cephalosporin and macrolides for specific situations). In contrast, the Netherlands, Switzerland and Ireland report that over 80% of the antibiotics given for lower respiratory tract infections are from the Access group which are generally narrow spectrum.22 The likely reasons for variation in CAP antibiotic treatment between centers (excluding 2 smaller sites) is possibly lack of precise local guidelines based on current recommendations and the presence or absence of comprehensive audit and feedback programs on hospital wards.
There are several strengths of the current study. First, inclusion of all Canadian children’s hospitals does represent the majority of pediatric inpatient beds in Canada and provides a large nationally representative sample. Second, the survey was completed at 2 different times of the year (fall and winter), which are most often associated with the highest number of admissions due to community respiratory infections such as RSV and influenza. Furthermore, this study allows for general comparison between sites within Canada based on similar size and complexity of care. The major limitation is the lack of assessment of appropriateness.20 This metric however could prove challenging with a paucity of national guidelines for infectious conditions in inpatients. Given the nature of a PPS, an inherent bias is that patients with longer lengths of stay (eg, oncology or complex care patients) would more likely be captured in a PPS. Importantly, the PPS does not measure outpatient antibiotics which were given at discharge which also contributes to burden of use.
In summary, measurement of overall antimicrobial use as well as broad-spectrum AU among pediatric hospitals in Canada is feasible using a PPS. Important metrics such as the proportion of WATCH group antibiotics and data on median durations should be used by local AS programs to influence practice changes locally.29 Ultimately this data should be actionable and used to establish benchmarking goals such as decreasing proportion of broad spectrum of antimicrobial use, guideline adherence and probably decreasing antibiotic prophylaxis in Canadian pediatric quaternary and tertiary pediatric hospitals.
Canadian Pediatric Antimicrobial Stewardship Group: Members include Julie Blackburn, MD, Department of Laboratory Medicine, Centre Hospitalier Universitaire Sainte-Justine, Université de Montréal, Montreal, QC; Michelle Science, Department of Pediatrics, The Hospital for Sick Children, University of Toronto, Toronto, ON; Kathryn Timberlake, PharmD, Department of Pharmacy, The Hospital for Sick Children, Toronto, ON; Helene Roy, PharmD, Department of Pharmacy, Centre Hospitalier Universitaire Sainte-Justine, Montreal, QC; Alena Tse-Chang, MD, Stollery Children’s Hospital, Department of Pediatrics, University of Alberta, Edmonton, AB; Joseph V. Vayalumkal, MD, Alberta Children’s Hospital, Department of Pediatrics, University of Calgary, Calgary, AB; Cora M. Constantinescu, MD, Alberta Children’s Hospital, Department of Pediatrics, University of Calgary, Calgary, AB; Deonne Dersch-Mills, PharmD, Pharmacy Services, Alberta Health Services, Calgary, AB; Ashley Roberts, MD, Women and Children’s Hospital of British Columbia, Vancouver, BC, Department of Pediatrics, University of British Columbia, Vancouver, BC; Vanessa Paquette, PharmD, Women and Children’s Hospital of British Columbia, Vancouver, BC; Natasha Kwan, PharmD, Children’s & Women’s Health Centre of British Columbia, Vancouver, BC; Roseline Thibeault, MD, Service d’infectiologie pédiatrique, Centre Mère-Enfant-Soleil du CHU de Québec, Université Laval, Québec, QC; Isabelle Viel-Thériault, MD, Service d’infectiologie pédiatrique, Centre Mère-Enfant-Soleil du CHU de Québec, Université Laval, Québec, QC; Dominik Mertz, MD, Department of Medicine, McMaster University, Hamilton, ON; Sarah Khan, MD, Department of Medicine, McMaster University, Hamilton, ON; Sergio Fanella, MD, Max Rady College of Medicine, University of Manitoba, Winnipeg, MB; Ashley N. Walus, BScPharm, Winnipeg Regional Health Authority, Winnipeg, MB; Marie-Astrid Lefebvre, MD, Montreal Children’s Hospital, McGill University Health Centre, Montreal, QC, Department of Pediatrics, McGill University Health Centre, Montreal, QC; Michelle Barton, MD, Department Paediatrics, London Health Sciences Centre, Western University, London, ON; Venita Harris, PharmD, Pharmacy Services, London Health Sciences Centre, London, ON, Department of Paediatrics, Western University, London, ON; Athena McConnell, MD, Pediatric Infectious Diseases, University of Saskatchewan, Saskatoon, SK; Blair W. Seifert, PharmD, Department of Pharmaceutical Services, Royal University Hospital—Saskatoon Saskatchewan Health Authority, Saskatoon, SK; Jeannette L. Comeau, MD, IWK Health Center, Department of Pediatrics, Dalhousie University, Halifax, NS; Kathryn Lynn Slayter, PharmD, IWK Health Center, Department of Pediatrics, Dalhousie University, Halifax, NS; Cheryl Foo, MD, Janeway Children’s Health and Rehabilitation Centre, Department of Pediatrics, Memorial University, St. John’s, NL; Kirk Leifso, MD, Kingston Health Sciences Centre, Department of Pediatrics Queen’s University, Kingston, ON; Jennifer Bowes, MSc, Children’s Hospital of Eastern Ontario Research Institute, Ottawa, ON and Nicole Le Saux, MD, Department of Pediatrics, University of Ottawa, Ottawa, ON, Children’s Hospital of Eastern Ontario, University of Ottawa, Ottawa, ON.
1. Barlam TF, Cosgrove SE, Abbo LM, et al. Implementing an antibiotic stewardship program: guidelines by the Infectious Diseases Society of America and the Society for Healthcare Epidemiology of America. Clin Infect Dis. 2016;62:e51–e77.
2. Science M, Timberlake K, Morris A, et al.; Groupe Antibiothérapie en Pédiatrie Canada Alliance for Stewardship of Antimicrobials in Pediatrics
(GAP Can ASAP). Quality metrics for antimicrobial stewardship programs. Pediatrics
3. Ibrahim OM, Polk RE. Benchmarking antimicrobial drug use in hospitals. Expert Rev Anti Infect Ther. 2012;10:445–457.
4. Fridkin SK, Srinivasan A. Implementing a strategy for monitoring inpatient antimicrobial use among hospitals in the United States. Clin Infect Dis. 2014;58:401–406.
5. Versporten A, Bielicki J, Drapier N, et al.; ARPEC project group. The Worldwide Antibiotic Resistance and Prescribing in European Children (ARPEC) point prevalence survey: developing hospital-quality indicators of antibiotic prescribing for children. J Antimicrob Chemother. 2016;71:1106–1117.
6. Le Saux N, Robinson JL; Canadian Paediatric Society, Infectious Diseases and Immunization Committee. Uncomplicated pneumonia in healthy Canadian children and youth: practice points for management. Paediatr Child Health. 2015;20:441–450.
7. Chibuk T, Cohen E, Robinson J, Mahant S, Hartfield D. Paediatric complicated pneumonia: Diagnosis and management of empyema. Canadian Pediatric Society. Published 2018. https://www.cps.ca/en/documents/position/complicated-pneumonia-empyema
. Accessed January 20, 2021.
8. Bradley JS, Byington CL, Shah SS, et al. The management of community-acquired pneumonia in infants and children older than 3 months of age: clinical practice guidelines by the pediatric infectious diseases society and the infectious diseases society of America. Clin Infect Dis. 2011;53:e1–e52.
9. World Health Organization (WHO). The 2019 WHO AWaRe classification of antibiotics for evaluation and monitoring of use. Geneva: World Health Organization. Published 2019. https://www.who.int/medicines/news/2019/WHO_releases2019AWaRe_classification_antibiotics/en/
. Accessed December 15, 2020.
10. Sharland M, Pulcini C, Harbarth S, et al.; 21st WHO Expert Committee on Selection and Use of Essential Medicines. Classifying antibiotics in the WHO essential medicines list for optimal use-be AWaRe. Lancet Infect Dis. 2018;18:18–20.
11. R Core Team. R: A Language and Environment for Statistical Computing. Https://Www.r-Project.Org/
. R Foundation for Statistical Computing; 2019. https://www.r-project.org/
. Accessed June 6, 2020.
12. Spiegelhalter DJ. Funnel plots for comparing institutional performance. Stat Med. 2005;24:1185–1202.
13. Porta A, Hsia Y, Doerholt K, et al. Comparing neonatal and paediatric antibiotic prescribing between hospitals: a new algorithm to help international benchmarking. J Antimicrob Chemother. 2012;67:1278–1286.
14. Gerber JS, Newland JG, Coffin SE, et al. Variability in antibiotic use at children’s hospitals. Pediatrics
15. Barney GR, Felsen CB, Dumyati GK. One-day point prevalence as a method for estimating antibiotic use in nursing homes. Infect Control Hosp Epidemiol. 2019;40:221–223.
16. Taylor G, Gravel D, Saxinger L, et al.; Canadian Nosocomial Infection Surveillance Program. Prevalence of antimicrobial use in a network of Canadian hospitals in 2002 and 2009. Can J Infect Dis Med Microbiol. 2015;26:85–89.
17. Pakyz AL, Gurgle HE, Ibrahim OM, et al. Trends in antibacterial use in hospitalized pediatric patients in United States academic health centers. Infect Control Hosp Epidemiol. 2009;30:600–603.
18. Gharbi M, Doerholt K, Vergnano S, et al. Using a simple point-prevalence survey to define appropriate antibiotic prescribing in hospitalised children across the UK. BMJ Open. 2016;6:1–8.
19. Levy ER, Swami S, Dubois SG, et al. Rates and appropriateness of antimicrobial prescribing at an academic children’s hospital, 2007-2010. Infect Control Hosp Epidemiol. 2012;33:346–353.
20. Osowicki J, Gwee A, Noronha J, et al. Australia-wide point prevalence survey of the use and appropriateness of antimicrobial prescribing for children in hospital. Med J Aust. 2014;201:657–662.
21. Baggs J, Fridkin SK, Pollack LA, et al. Estimating national trends in inpatient antibiotic use among US hospitals from 2006 to 2012. JAMA Intern Med. 2016;176:1639–1648.
22. Hsia Y, Lee BR, Versporten A, et al.; GARPEC and Global-PPS networks. Use of the WHO Access, Watch, and Reserve classification to define patterns of hospital antibiotic use (AWaRe): an analysis of paediatric survey data from 56 countries. Lancet Glob Health. 2019;7:e861–e871.
23. Stenehjem E, Hersh AL, Sheng X, et al. Antibiotic use in small community hospitals. Clin Infect Dis. 2016;63:1273–1280.
24. Yan K, Xue M, Ye D, et al. Antibiotic prescribing practices in secondary and tertiary hospitals in Shaanxi province, western China, 2013-2015. PLoS One. 2018;13:e0207229.
25. Schulman J, Profit J, Lee HC, et al. Variations in neonatal antibiotic use. Pediatrics
26. Osowicki J, Gwee A, Noronha J, et al.; ANZPID-ASAP group (Australian and New Zealand Paediatric Infectious Diseases-Australasian Stewardship of Antimicrobials in Paediatrics). Australia-wide point prevalence survey of antimicrobial prescribing in neonatal units: how much and how good? Pediatr Infect Dis J. 2015;34:e185–e190.
27. Ting JY, Roberts A, Sherlock R, et al.; Canadian Neonatal Network Investigators. Duration of initial empirical antibiotic therapy and outcomes in very low birth weight infants. Pediatrics
28. Cotten CM, Taylor S, Stoll B, et al.; NICHD Neonatal Research Network. Prolonged duration of initial empirical antibiotic treatment is associated with increased rates of necrotizing enterocolitis and death for extremely low birth weight infants. Pediatrics
29. Malcolm W, Nathwani D, Davey P, et al. From intermittent antibiotic point prevalence surveys to quality improvement: experience in Scottish hospitals. Antimicrob Resist Infect Control. 2013;2:3.