Acta Pædiatrica ISSN 0803-5253
REGULAR ARTICLE
A worrying trend in weight-adjusted paediatric antibiotic use in a Norwegian tertiary care hospital Ragnhild Raastad ([email protected])1,2, Ingunn Fride Tvete3, Tore G. Abrahamsen2,4, Dag Berild1,2, Truls M. Leegaard2,5, Mette Walberg6, Fredrik M€uller2,7 1.Department of Infectious Diseases, Oslo University Hospital, Oslo, Norway 2.Institute of Clinical Medicine, University of Oslo, Oslo, Norway 3.The Norwegian Computing Center, Oslo, Norway 4.Department of Pediatrics, Oslo University Hospital, Oslo, Norway 5.Department of Microbiology and Infection Control, Akershus University Hospital, Lørenskog, Norway 6.Microbiology Section, Laboratory Centre, Vestre Viken Hospital Trust, Rud, Norway 7.Department of Microbiology, Oslo University Hospital, Oslo, Norway
Keywords Antibacterial agents, Defined daily dose, Drug utilisation, Hospitalised children, Recommended daily dose Correspondence R Raastad, Department of Infectious Diseases, Oslo University Hospital, Box 4956 Nydalen, N-0424 Oslo, Norway. Tel: +4799704347 | Fax: +4722991581 | Email: [email protected] Received 6 October 2014; revised 29 December 2014; accepted 5 March 2015. DOI:10.1111/apa.12994
ABSTRACT Aim: The World Health Organization recommends the defined daily dose (DDD) as the standard unit of measurement for antibiotic use, but this is not applicable in children. We aimed to assess paediatric antibiotic use in a Norwegian tertiary care hospital using a novel weight-adjusted method. Methods: We obtained antibiotic purchase data from the hospital pharmacy and administrative data for all admissions from 2002 to 2009 to the paediatric wards at Oslo University Hospital, Rikshospitalet. Recommended daily doses per 100 kg days (RDDs/kg days) were calculated based on national guidelines for paediatric antibiotic use, length of stay and estimated weight for sex and age using national growth references. Results: Total antibiotic use increased significantly from 51.8 to 65.5 RDDs/100 kg days. We found statistically significant annual increases in the consumption of carbapenems (18.0%), third-generation cephalosporins (6.0%) and imidazole derivatives (6.6%) and a considerable difference between total antibiotic use measured in RDDs/100 kg days and DDDs/100 bed days for neonates. Conclusion: Weight-adjusted antibiotic use provided a more meaningful description of the quantities of antibiotics consumed than DDDs/100 bed days, particularly for neonates. Total antibiotic use, use of meropenem, third-generation cephalosporins and imidazole derivatives increased significantly despite low prevalence of antibiotic-resistant pathogens.
INTRODUCTION Although antibiotics are frequently prescribed for hospitalised children, data on antibiotic consumption are scarce. The World Health Organization (WHO) recommends using the defined daily dose (DDD) as the standard unit of measurement in drug utilisation research (1). The DDD is defined as the assumed average maintenance dose per day for a drug used for its main indication in adults where an adult is considered to weigh 70 kg. In children, dose recommendations vary according to age and body weight and the DDD method is therefore not applicable. Several alternative methods have been suggested to overcome this shortcoming (2–7), but all of these require access to data at the individual patient level. In most European countries, Abbreviations ATC classification, Anatomical Therapeutic Chemical classification; DDD, Defined daily dose; ICU, Intensive care unit; RDD, Recommended daily dose.
including Norway, electronic prescribing databases are not yet implemented and obtaining individual patient-level data on antibiotic consumption therefore requires considerable time and effort.
Key notes
©2015 Foundation Acta Pædiatrica. Published by John Wiley & Sons Ltd 2015 104, pp. 687–692
The method recommended by the World Health Organization (WHO) for quantifying antibiotic use is not applicable in children, and we present an alternative method that adjusts for body weight. Total antibiotic use, use of third-generation cephalosporins, meropenem and imidazole derivatives in hospitalised children between 2002 and 2009 increased significantly. There was a considerable difference between total antibiotic use measured using the two methods for neonates.
687
Weight-adjusted paediatric antibiotic use
Raastad et al.
The consumption of antibiotics and the occurrence of antibiotic resistance are low in Norway compared to most other European countries (8,9). However, national data indicate that there has been a gradual increase in hospital use of antibiotics known to promote antibiotic resistance (10). Whether this is also the case in the paediatric population is unknown. To enable analysis of temporal trends in the paediatric wards of a Norwegian tertiary care university hospital, we developed a method for calculating weight-corrected antibiotic use in hospitalised children. Local microbiology data were assessed to detect changes that may have contributed to altered prescription patterns.
METHODS Population The study was conducted at Oslo University Hospital, Rikshospitalet, a tertiary care facility that, at the time of the study, also served as a secondary care hospital for a population of approximately 39 000 children up to 16 years of age. The study population included all children admitted to the two paediatric wards and the neonatal ward from 2002 to 2009. The paediatric intensive care unit (ICU) was excluded because antibiotics delivered to this unit were inseparable from deliveries to the adult ICU. One of the wards was an oncology ward, which also provided treatment for liver disease and transplants. The other ward
Dates of admission and discharge, age and sex for all patients included in the study were extracted from the hospital’s administrative database. Estimated median weight for the individual patient’s sex and age in weeks was obtained from the growth references for Norwegian children (15). Incidence of bacteria isolated from blood cultures and prevalence of selected resistant isolates were obtained from the laboratory’s routine database. All individual data were aggregated into monthly data. Information concerning indications for antibiotic therapy was not electronically available. Because the administrative data were anonymised, we were unable to exclude readmittance of the same patient. The weight-adjusted method was first suggested by Bennet et al. (5). To obtain more accurate estimates, we assigned weight by the sex and age in weeks instead of only age in months for children under the age of 1 year and age in years for children who were 1 year or older. Furthermore, we analysed monthly instead of annual data. Calculation The calculation of monthly, weight-corrected antibiotic consumption as RDDs/100 kg days was performed based on the recommended daily dose, the individual length of stay and the estimated median weight for the sex and age of each patient as follows:
[No. of vials][No. of miligrams per vial] RDD [RDD] ¼ PNo: of patients 100 kg days ð½Length of stay i ½Estimated median weighti Þ 100 Patient i
served as a general paediatric ward as well as a national centre for kidney and heart transplants. Of the 22 beds in the neonatal ward, 10 were dedicated to intensive care. There were no organisational changes, and the number of hospital beds reserved for patients with malignancies was constant throughout the study period. The guidelines for antibiotic use and the recommended treatment of neutropenic fever remained unchanged. Age classification was performed according to international recommendations for clinical trials of medicinal products in paediatric populations (11). Data collection Monthly in-hospital deliveries and returns of all antibacterial drugs for systemic use – subgroup J01 of the Anatomical Therapeutic Chemical (ATC) classification – were extracted from the hospital pharmacy’s database. Returned drugs were unopened vials due to ordering or delivery errors. Data were converted into DDDs and recommended daily doses (RDDs). The 2009 version of the WHO guidelines for ATC classification and DDD assignment was used (12). The RDD for a drug was defined as the recommended daily maintenance dose per kg for its main indication and was based on national guidelines (13,14).
688
where ½RDD ¼ ½Recommended dosage in mg/kg ½No. of recommended dosage per day and ½Length of stayi ¼ ½Date of dischargei ½Date of admissioni
Statistical analysis A linear regression model was used to determine the magnitude and significance of the change in antibiotic use during the study period. We corrected for seasonal variation by including 11 indicators representing the months from February through December, with January as the baseline. The statistical analyses were performed using R version 2.14.2 (16). A p-value of
REGULAR ARTICLE
A worrying trend in weight-adjusted paediatric antibiotic use in a Norwegian tertiary care hospital Ragnhild Raastad ([email protected])1,2, Ingunn Fride Tvete3, Tore G. Abrahamsen2,4, Dag Berild1,2, Truls M. Leegaard2,5, Mette Walberg6, Fredrik M€uller2,7 1.Department of Infectious Diseases, Oslo University Hospital, Oslo, Norway 2.Institute of Clinical Medicine, University of Oslo, Oslo, Norway 3.The Norwegian Computing Center, Oslo, Norway 4.Department of Pediatrics, Oslo University Hospital, Oslo, Norway 5.Department of Microbiology and Infection Control, Akershus University Hospital, Lørenskog, Norway 6.Microbiology Section, Laboratory Centre, Vestre Viken Hospital Trust, Rud, Norway 7.Department of Microbiology, Oslo University Hospital, Oslo, Norway
Keywords Antibacterial agents, Defined daily dose, Drug utilisation, Hospitalised children, Recommended daily dose Correspondence R Raastad, Department of Infectious Diseases, Oslo University Hospital, Box 4956 Nydalen, N-0424 Oslo, Norway. Tel: +4799704347 | Fax: +4722991581 | Email: [email protected] Received 6 October 2014; revised 29 December 2014; accepted 5 March 2015. DOI:10.1111/apa.12994
ABSTRACT Aim: The World Health Organization recommends the defined daily dose (DDD) as the standard unit of measurement for antibiotic use, but this is not applicable in children. We aimed to assess paediatric antibiotic use in a Norwegian tertiary care hospital using a novel weight-adjusted method. Methods: We obtained antibiotic purchase data from the hospital pharmacy and administrative data for all admissions from 2002 to 2009 to the paediatric wards at Oslo University Hospital, Rikshospitalet. Recommended daily doses per 100 kg days (RDDs/kg days) were calculated based on national guidelines for paediatric antibiotic use, length of stay and estimated weight for sex and age using national growth references. Results: Total antibiotic use increased significantly from 51.8 to 65.5 RDDs/100 kg days. We found statistically significant annual increases in the consumption of carbapenems (18.0%), third-generation cephalosporins (6.0%) and imidazole derivatives (6.6%) and a considerable difference between total antibiotic use measured in RDDs/100 kg days and DDDs/100 bed days for neonates. Conclusion: Weight-adjusted antibiotic use provided a more meaningful description of the quantities of antibiotics consumed than DDDs/100 bed days, particularly for neonates. Total antibiotic use, use of meropenem, third-generation cephalosporins and imidazole derivatives increased significantly despite low prevalence of antibiotic-resistant pathogens.
INTRODUCTION Although antibiotics are frequently prescribed for hospitalised children, data on antibiotic consumption are scarce. The World Health Organization (WHO) recommends using the defined daily dose (DDD) as the standard unit of measurement in drug utilisation research (1). The DDD is defined as the assumed average maintenance dose per day for a drug used for its main indication in adults where an adult is considered to weigh 70 kg. In children, dose recommendations vary according to age and body weight and the DDD method is therefore not applicable. Several alternative methods have been suggested to overcome this shortcoming (2–7), but all of these require access to data at the individual patient level. In most European countries, Abbreviations ATC classification, Anatomical Therapeutic Chemical classification; DDD, Defined daily dose; ICU, Intensive care unit; RDD, Recommended daily dose.
including Norway, electronic prescribing databases are not yet implemented and obtaining individual patient-level data on antibiotic consumption therefore requires considerable time and effort.
Key notes
©2015 Foundation Acta Pædiatrica. Published by John Wiley & Sons Ltd 2015 104, pp. 687–692
The method recommended by the World Health Organization (WHO) for quantifying antibiotic use is not applicable in children, and we present an alternative method that adjusts for body weight. Total antibiotic use, use of third-generation cephalosporins, meropenem and imidazole derivatives in hospitalised children between 2002 and 2009 increased significantly. There was a considerable difference between total antibiotic use measured using the two methods for neonates.
687
Weight-adjusted paediatric antibiotic use
Raastad et al.
The consumption of antibiotics and the occurrence of antibiotic resistance are low in Norway compared to most other European countries (8,9). However, national data indicate that there has been a gradual increase in hospital use of antibiotics known to promote antibiotic resistance (10). Whether this is also the case in the paediatric population is unknown. To enable analysis of temporal trends in the paediatric wards of a Norwegian tertiary care university hospital, we developed a method for calculating weight-corrected antibiotic use in hospitalised children. Local microbiology data were assessed to detect changes that may have contributed to altered prescription patterns.
METHODS Population The study was conducted at Oslo University Hospital, Rikshospitalet, a tertiary care facility that, at the time of the study, also served as a secondary care hospital for a population of approximately 39 000 children up to 16 years of age. The study population included all children admitted to the two paediatric wards and the neonatal ward from 2002 to 2009. The paediatric intensive care unit (ICU) was excluded because antibiotics delivered to this unit were inseparable from deliveries to the adult ICU. One of the wards was an oncology ward, which also provided treatment for liver disease and transplants. The other ward
Dates of admission and discharge, age and sex for all patients included in the study were extracted from the hospital’s administrative database. Estimated median weight for the individual patient’s sex and age in weeks was obtained from the growth references for Norwegian children (15). Incidence of bacteria isolated from blood cultures and prevalence of selected resistant isolates were obtained from the laboratory’s routine database. All individual data were aggregated into monthly data. Information concerning indications for antibiotic therapy was not electronically available. Because the administrative data were anonymised, we were unable to exclude readmittance of the same patient. The weight-adjusted method was first suggested by Bennet et al. (5). To obtain more accurate estimates, we assigned weight by the sex and age in weeks instead of only age in months for children under the age of 1 year and age in years for children who were 1 year or older. Furthermore, we analysed monthly instead of annual data. Calculation The calculation of monthly, weight-corrected antibiotic consumption as RDDs/100 kg days was performed based on the recommended daily dose, the individual length of stay and the estimated median weight for the sex and age of each patient as follows:
[No. of vials][No. of miligrams per vial] RDD [RDD] ¼ PNo: of patients 100 kg days ð½Length of stay i ½Estimated median weighti Þ 100 Patient i
served as a general paediatric ward as well as a national centre for kidney and heart transplants. Of the 22 beds in the neonatal ward, 10 were dedicated to intensive care. There were no organisational changes, and the number of hospital beds reserved for patients with malignancies was constant throughout the study period. The guidelines for antibiotic use and the recommended treatment of neutropenic fever remained unchanged. Age classification was performed according to international recommendations for clinical trials of medicinal products in paediatric populations (11). Data collection Monthly in-hospital deliveries and returns of all antibacterial drugs for systemic use – subgroup J01 of the Anatomical Therapeutic Chemical (ATC) classification – were extracted from the hospital pharmacy’s database. Returned drugs were unopened vials due to ordering or delivery errors. Data were converted into DDDs and recommended daily doses (RDDs). The 2009 version of the WHO guidelines for ATC classification and DDD assignment was used (12). The RDD for a drug was defined as the recommended daily maintenance dose per kg for its main indication and was based on national guidelines (13,14).
688
where ½RDD ¼ ½Recommended dosage in mg/kg ½No. of recommended dosage per day and ½Length of stayi ¼ ½Date of dischargei ½Date of admissioni
Statistical analysis A linear regression model was used to determine the magnitude and significance of the change in antibiotic use during the study period. We corrected for seasonal variation by including 11 indicators representing the months from February through December, with January as the baseline. The statistical analyses were performed using R version 2.14.2 (16). A p-value of
![[Paediatric dermatology emergencies in a tertiary hospital].](/assets/img/hold.png)
