Acta Pædiatrica ISSN 0803-5253
REGULAR ARTICLE
Individual maternal and child exposure to antibiotics in hospital – a national population-based validation study 1 € C Almqvist ([email protected])1,2, AK Ortqvist , T Gong1, A Wallas3, KM Ahlén1, W Ye1, C Lundholm1
1.Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden 2.Astrid Lindgren Children’s Hospital, Lung and Allergy Unit, Karolinska University Hospital, Stockholm, Sweden 3.Institute of Environmental Medicine, Karolinska Institutet, Stockholm, Sweden
Keywords Antibiotics, Bacterial infections, Pregnancy, Prematurity, Viruses Correspondence Prof. Catarina Almqvist, MD, Department of Medical Epidemiology and Biostatistics, KarolinskaInstitutet, Box 281, SE-171 77 Stockholm, Sweden. Tel: +46 8 524 86193 | Fax: +46 8 31 49 75 | Email: [email protected] Received 4 August 2014; revised 7 November 2014; accepted 19 December 2014. DOI:10.1111/apa.12916
ABSTRACT Aim: Exposure to antibiotics in early life may affect future health. Most antibiotics are prescribed in outpatient care, but inpatient exposure is also important. We estimated how specific diagnoses in hospitals corresponded to individual antibiotic exposure. Methods: All pregnant women and children from birth to 5 years of age with infectious diseases and common inpatient diagnoses between July 2005 and November 2011 were identified from the Swedish National Patient Register. Random samples of individuals from predefined groups were drawn, and medical records received from the clinics were manually reviewed for antibiotics. Results: Medical records for 4319 hospital visits were requested and 3797 (88%) were received. A quarter (25%) of children diagnosed as premature had received antibiotics, and in children from one to 5 years of age, diagnoses associated with bacterial infections were more commonly treated with antibiotics (62.4–90.6%) than those associated with viruses (6.3–22.2%). Pregnant women who had undergone a Caesarean section were more likely to be treated with antibiotics than those who had had a vaginal delivery (40.1% versus 11.1%). Conclusion: This study defines the proportion of new mothers and young children who received individual antibiotic treatment for specific inpatient diagnoses in Sweden and provides a useful basis for future studies focusing on antibiotic use.
INTRODUCTION The development of antibiotics has enabled treatment of previously life-threatening bacterial infections, and antibiotics are now commonly used to treat children and adults worldwide. Children may be exposed to antibiotics antenatally, during delivery, the neonatal period or first years of life. As antibiotics given during early life may affect the future health of a child (1–4), it is important to study how children are exposed to them, both in hospitals and in outpatient care. The majority of antibiotics distributed in Sweden are prescribed in outpatient care, with hospitals accounting for 11% of total antibiotic use (5). During 2010, 22% of the general population and 27% of children from birth to 6 years of age received at least one prescription of antibiotics in outpatient care (5), and approximately one of three inpatients receive antibiotics during hospitalisation (6). Information regarding dispensed antibiotics at an individual level in outpatient care is available through the Swedish Prescribed Drug Register. There are currently no register data covering the individual administration of antibiotics to inpatients during hospitalisation, and the full individual exposure to antibiotics is therefore unavailable at present. All main and secondary diagnoses during inpatient care are reported to the National Patient Register according
392
to the International Classification of Diseases (ICD) system. Exposure to antibiotics during hospitalisation in relation to infectious diseases and other common inpatient diagnoses can be assessed through medical records. The aim of this national population-based study was to estimate how diagnoses of infectious diseases and other common diagnoses in the National Patient Register corresponded to the use of antibiotics in hospitalised pregnant women and young children.
Key notes
This study assessed the proportion of pregnant women and children with specific inpatient diagnoses who were treated with antibiotics based on data from 3797 medical records. A quarter of premature children received antibiotics and their use was more common in bacterial infections (62.4–90.6%) than viruses (6.3–22.2%) in children from one to 5 years of age. Pregnant women were more likely to receive antibiotics following Caesarean sections (40.1%) than vaginal deliveries (11.1%).
©2014 Foundation Acta Pædiatrica. Published by John Wiley & Sons Ltd 2015 104, pp. 392–395
Almqvist et al.
Individual antibiotic exposure in hospital
MATERIAL AND METHODS A cohort of all pregnant women and children from birth to 5 years of age with a discharge date between July 2, 2005 and November 30, 2011 was identified from the Swedish National Patient Register. The start date of each pregnancy was estimated from the Swedish Medical Birth Register, as the date of birth minus the gestational age in days. The universal use of the Swedish personal identity number enables researchers to establish links between the Medical Birth Register, Swedish Prescribed Drug Register and the National Patient Register (7). Links between inpatient diagnoses of infectious disease and other common diagnoses in the National Patient Register and dispensed antibiotics, using code J01 from the anatomical therapeutic chemical (ATC) classification system in the Swedish Prescribed Drug Register, were performed at the National Board of Health and Welfare. To limit the study population to those who had only been given antibiotics in hospitals, all individuals who received outpatient-dispensed prescriptions for antibiotics on the day of discharge or the day after, according to the Swedish Prescribed Drug Register, were excluded. Four groups of children were identified for medical record retrieval, based on the children’s age at discharge (birth to 12 months and 1–5 years); prematurity (ICDcodes P072 and P073); malformations, factors influencing health status and cerebral palsy (ICD-codes Q, Z and G80)
and inpatient diagnoses. The fifth group comprised the pregnant women. The sample size needed for each group and diagnosis was calculated based on clinical assumptions. We aimed at 95% confidence intervals of 5% from the estimated proportions (percentages), and for the calculations, we assumed the proportions in the population were between 10 and 90%, depending on the group and diagnosis. The groups were mutually exclusive. Group 1 comprised children from birth to 12 months with a diagnosis of prematurity (n = 320). Group 2 comprised children from birth to 12 months without a main or secondary diagnosis of prematurity and without a main diagnosis of ICD-code Q or Z (n = 320). Children aged from one to 5 years with a main inpatient diagnosis of infectious disease and other common diagnoses, as specified in Table 1, made up Group 3 (n = 2641), and children aged from one to 5 years without the main diagnoses specified in Table 1 made up Group 4 (n = 320). Group 5 comprised pregnant women with the main diagnoses specified in Table 1 (n = 718). Simple random samples (without replacement) of each of the five groups and diagnosis were drawn by the National Board of Health and Welfare. We were provided with the personal identity number, date of discharge and name of the hospital so that we could request each individual’s medical records. The National Board of Health and Welfare also provided the total number of hospital visits for each
Table 1 Diagnoses and ICD-10 codes used to define groups one to five and descriptive statistics on the numbers exposed to antibiotics, along with corresponding numbers with main and secondary diagnoses Q and/or G80 excluded Exposed to antibiotics
Group
Specific diagnosis
Children 0–12 months 1 Prematurity 2 Diagnoses other than prematurity Children 1–5 years 3 Appendicitis Chronic tonsillitis Gastroenteritis Bacterial sepsis Meningitis Unspecific viral infection Pneumonia Asthma Acute bronchitis and URTI Skin infection Seizures Unspecific abdominal pain Cystitis and pyelonephritis 4 Other main diagnoses than those in group 3 Pregnant women 5 Maternal diagnoses Vaginal delivery Caesarian Section (CS) CS (twins) and multiple delivery
ICD 10 codes
P072, P073 All ICD codes† K35 J35 A08, A09 A40, A41 A87, G00, G02,G03 B34 J18 J45, J46 J20, J21, J06 L08 R56 R10 N30, N10
O268B, K80, K81, K82, K83, O23, N30, N39, N10, O85 O80 O820, O821 O842, O848
Yes n (%)
69 (25.1) 49 (18.1)
Missing n (%)
7 (2.6) 11 (4.1)
29 1 26 43 88 64 242 6 64 57 14 13 269 57
(67.4) (3.2) (8.8) (89.6) (62.4) (22.1) (84.9) (4.5) (22.2) (82.6) (6.1) (6.3) (90.6) (21.4)
3 2 11 2 3 7 5 4 1
46 5 71 55
(16.4) (11.1) (40.1) (43.7)
Yes* n (%)
60 (24.1) 46 (17.6)
6 12 5 15
(7.0) (6.5) (3.7) (4.2) (2.1) (2.4) (1.8) (3.0) (0.4) – (2.6) (5.8) (1.7) (5.6)
28 1 24 38 88 59 222 6 53 55 14 13 245 52
(66.7) (3.2) (8.4) (88.4) (62.4) (21.2) (84.5) (4.6) (20.0) (83.3) (6.1) (6.4) (91.1) (21.0)
20 5 14 9
(7.1) (11.1) (7.9) (7.1)
46 5 70 55
(16.4) (11.1) (39.8) (43.7)
*Main and secondary diagnosis of malformations and cerebral palsy (ICD-10 Q and G80) excluded. † Except P072, P073 as main or secondary diagnoses, and Q or Z as main diagnoses.
©2014 Foundation Acta Pædiatrica. Published by John Wiley & Sons Ltd 2015 104, pp. 392–395
393
Individual antibiotic exposure in hospital
Almqvist et al.
specified group and diagnosis during the study period (Table S1). Starting in January 2012, the study participants’ medical records, corresponding to the date of discharge from the hospitals for the specific diagnosis of interest, were requested by a letter to the head of the ward where the patient had been treated. Two reminder letters were sent out before the end of data collection in June 2013. Each medical record was manually reviewed, and all available information regarding diagnosis and exposure to antibiotics was extracted from text and drug administration records. Antibiotics were further categorised into groups based on ATC-codes when the substance name was available. We requested medical records for 4319 hospital visits based on a study population of approximately 1.2 million children and pregnant women. Permission for the study was obtained from the Regional Ethical Review board in Stockholm, Sweden.
RESULTS We received 3797 medical records of the 4319 requested, which corresponded to a response rate of 88%. The 522 missing medical records could not be found on the wards or the wards did not respond to our request. Table 1 describes the diagnoses and ICD-10 codes used to identify each group and to identify medical records at hospitals in combination with the discharge date. Table 1 also displays the number and proportion of hospital visits with information on treatment with antibiotics, broken down into yes and missing. Table S1 displays diagnoses and ICD-10 codes along with descriptive statistics on hospital visits in the total population, requested and received medical records. In total, 17 750 children between birth and 12 months had a diagnosis of prematurity. A quarter (25%) of the reviewed medical records reported treatment with antibiotics, which corresponded to 4437 children in the total population. About three quarters (74.6%) of the cases that recorded antibiotic use also provided information on type of antibiotics that could be related to an ATC-code. In the group of children aged from one to 5 years, diagnoses that were commonly associated with bacterial infections were more commonly treated with antibiotics; appendicitis (67.4%), bacterial sepsis (89.6%), meningitis (62.4%), pneumonia (84.9%) and pyelonephritis (90.6%). Diagnoses more often associated with viruses were treated less often with antibiotics; gastroenteritis (8.8%), unspecific viral infections (22.1%), acute bronchitis and upper respiratory tract infections (22.2%) and unspecific abdominal pain (6.3%). The proportion of children treated with antibiotics was not different in the group of children with diagnoses of malformations or cerebral palsy. Of the 973 that recorded antibiotic use, 87.2% also provided information on the ATC-code for the type of antibiotic. We reviewed a sample of 45 medical records for the 334 520 pregnant women who had a vaginal delivery, and this showed that 11.1% had been treated with antibiotics. In
394
addition, 16.1% of the women with the most common diagnoses during pregnancy and 40.1% of those who received a Caesarean section had been treated with antibiotics. The majority (83.6) of the records that mentioned antibiotics use also provided information on the type of antibiotics.
DISCUSSION Our results, based on the National Patient Register and medical records, show that a quarter of children below 1 year of age, who were hospitalised with a diagnosis of prematurity, and a high proportion of children with diagnoses of infectious diseases, such as bacterial sepsis, pneumonia and pyelonephritis were treated with antibiotics in hospital. Children diagnosed with virus-associated infections were less likely to be treated with antibiotics, which reflect treatment guidelines and recommendations (8). We also show that antibiotic treatment was more common in pregnant women delivered by Caesarean section than vaginal delivery and that approximately one-sixth of the women with the more common diagnoses during pregnancy were treated with antibiotics. Previous population-based studies have reported point prevalence of antibiotics treatment in inpatient care, but these were not specific to the individual (9,10). The Swedish Strategic Program against Antimicrobial Resistance studied all hospital admissions for children from birth to 16 years of age in selected hospitals during five two-week periods, and this showed that on average one third of children was exposed to antibiotics (9). Although this method is suitable for assessing the risk for antibiotic resistance and the adherence to clinical guidelines on a population level, it is not able to evaluate individual antibiotic exposure for longterm follow-up or to study a selected group of patients. There are also ongoing efforts to develop a better understanding of how to interpret viral findings and to reduce the unnecessary use of antibiotics, as infections such as acute respiratory illness account for a large proportion of all visits to paediatric health facilities (11). Our study is a population-based study to assess individual antibiotic treatment in hospitals, based on a thorough collection and review of a sample of medical inpatient records from children and pregnant women in national Swedish registers with complete coverage. Although there was some missing information on the type of antibiotics and the exact ATC-codes, we were able to assess the proportion of antibiotic treatment per diagnosis. This will allow future studies on the association between antibiotics and subsequent disorders to take separate diagnoses into account (1–4).
ACKNOWLEDGEMENTS We would like to acknowledge Gerd Agerberg and Michael Broms for their excellent data collection and database € rn management and Associate Professors Helle Kieler, Bjo € ve Lindemalm and Professor Jonas F Wettermark, Synno
©2014 Foundation Acta Pædiatrica. Published by John Wiley & Sons Ltd 2015 104, pp. 392–395
Almqvist et al.
Ludvigsson for valuable discussions. We are also indebted to the hospital wards that provided medical records.
CONFLICT OF INTEREST None.
FUNDING Financial support was provided by the Swedish Research Council (grant no 2011-3060) and the Swedish Initiative for Research on Microdata in the Social And Medical Sciences (SIMSAM) framework grants no. 80748301 and 340-20135867, with other grants provided by the Stockholm County Council (ALF project), the Swedish Heart Lung Foundation and the Strategic Research Program in Epidemiology at Karolinska Institutet.
Individual antibiotic exposure in hospital
6.
7.
8.
9.
10.
References 1. Alm B, Erdes L, Mollborg P, Pettersson R, Norvenius SG, Aberg N, et al. Neonatal antibiotic treatment is a risk factor for early wheezing. Pediatrics 2008; 121: 697–702. 2. Marra F, Marra CA, Richardson K, Lynd LD, Kozyrskyj A, Patrick DM, et al. Antibiotic use in children is associated with increased risk of asthma. Pediatrics 2009; 123: 1003–10. 3. Almqvist C, Wettermark B, Hedlin G, Ye W, Lundholm C. Antibiotics and asthma medication in a large register-based cohort study – confounding, cause and effect. Clin Exp Allergy 2012; 42: 104–11. 4. Ortqvist AK, Lundholm C, Kieler H, Ludvigsson JF, Fall T, Ye W, et al. Antibiotics in fetal and early life and subsequent childhood asthma: nationwide population based study with sibling analysis. BMJ 2014; 349: g6979. 5. Hellman J, Olsson-Liljeqvist B, Bengtsson B, Greko C, eds (2012) SWEDRES-SVARM 2012. Use of antimicrobials and
11.
occurrence of antimicrobial resistance in Sweden. Solna/ Uppsala: Strama. €nds antibiotika p Hur anva a sjukhus? STRAMAS punktprevalensstudier (PPS) 2003, 2004, 2006, 2008 och 2010. [Internet]. Strama-Slutenv ard and Folkha¨lsomyndigheten. Available from http://www.strama.se/dyn/,84,.htm. Ludvigsson JF, Otterblad-Olausson P, Pettersson BU, Ekbom A. The Swedish personal identity number: possibilities and pitfalls in healthcare and medical research. Eur J Epidemiol 2009; 24: 659–67. Craig JC, Williams GJ, Jones M, Codarini M, Macaskill P, Hayen A, et al. The accuracy of clinical symptoms and signs for the diagnosis of serious bacterial infection in young febrile children: prospective cohort study of 15 781 febrile illnesses. BMJ 2010; 340: c1594. Stramas punktprevalensstudier 2003-2010 (2011). [Internet]. Strama, SMI. Available from http://www.folkhalsomyndighe ten.se/documents/smittskydd-sjukdomar/antibiotika-vardhyg ien/statistik-fran-stramas-PPS-2003(2010-fohm2014).pdf Kimland E, Nydert P, Odlind V, Bottiger Y, Lindemalm S. Paediatric drug use with focus on off-label prescriptions at Swedish hospitals – a nationwide study. Acta Paediatr 2012; 101: 772–8. Rhedin S, Lindstrand A, Rotzen-Ostlund M, Tolfvenstam T, Ohrmalm L, Rinder MR, et al. Clinical utility of PCR for common viruses in acute respiratory illness. Pediatrics 2014; 133: e538–45.
SUPPORTING INFORMATION Additional Supporting Information may be found in the online version of this article: Table S1. Hospital visits in the total population for diagnoses and ICD-10 codes during the study period, along with requested and received medical records.
©2014 Foundation Acta Pædiatrica. Published by John Wiley & Sons Ltd 2015 104, pp. 392–395
395
Copyright of Acta Paediatrica is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
REGULAR ARTICLE
Individual maternal and child exposure to antibiotics in hospital – a national population-based validation study 1 € C Almqvist ([email protected])1,2, AK Ortqvist , T Gong1, A Wallas3, KM Ahlén1, W Ye1, C Lundholm1
1.Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden 2.Astrid Lindgren Children’s Hospital, Lung and Allergy Unit, Karolinska University Hospital, Stockholm, Sweden 3.Institute of Environmental Medicine, Karolinska Institutet, Stockholm, Sweden
Keywords Antibiotics, Bacterial infections, Pregnancy, Prematurity, Viruses Correspondence Prof. Catarina Almqvist, MD, Department of Medical Epidemiology and Biostatistics, KarolinskaInstitutet, Box 281, SE-171 77 Stockholm, Sweden. Tel: +46 8 524 86193 | Fax: +46 8 31 49 75 | Email: [email protected] Received 4 August 2014; revised 7 November 2014; accepted 19 December 2014. DOI:10.1111/apa.12916
ABSTRACT Aim: Exposure to antibiotics in early life may affect future health. Most antibiotics are prescribed in outpatient care, but inpatient exposure is also important. We estimated how specific diagnoses in hospitals corresponded to individual antibiotic exposure. Methods: All pregnant women and children from birth to 5 years of age with infectious diseases and common inpatient diagnoses between July 2005 and November 2011 were identified from the Swedish National Patient Register. Random samples of individuals from predefined groups were drawn, and medical records received from the clinics were manually reviewed for antibiotics. Results: Medical records for 4319 hospital visits were requested and 3797 (88%) were received. A quarter (25%) of children diagnosed as premature had received antibiotics, and in children from one to 5 years of age, diagnoses associated with bacterial infections were more commonly treated with antibiotics (62.4–90.6%) than those associated with viruses (6.3–22.2%). Pregnant women who had undergone a Caesarean section were more likely to be treated with antibiotics than those who had had a vaginal delivery (40.1% versus 11.1%). Conclusion: This study defines the proportion of new mothers and young children who received individual antibiotic treatment for specific inpatient diagnoses in Sweden and provides a useful basis for future studies focusing on antibiotic use.
INTRODUCTION The development of antibiotics has enabled treatment of previously life-threatening bacterial infections, and antibiotics are now commonly used to treat children and adults worldwide. Children may be exposed to antibiotics antenatally, during delivery, the neonatal period or first years of life. As antibiotics given during early life may affect the future health of a child (1–4), it is important to study how children are exposed to them, both in hospitals and in outpatient care. The majority of antibiotics distributed in Sweden are prescribed in outpatient care, with hospitals accounting for 11% of total antibiotic use (5). During 2010, 22% of the general population and 27% of children from birth to 6 years of age received at least one prescription of antibiotics in outpatient care (5), and approximately one of three inpatients receive antibiotics during hospitalisation (6). Information regarding dispensed antibiotics at an individual level in outpatient care is available through the Swedish Prescribed Drug Register. There are currently no register data covering the individual administration of antibiotics to inpatients during hospitalisation, and the full individual exposure to antibiotics is therefore unavailable at present. All main and secondary diagnoses during inpatient care are reported to the National Patient Register according
392
to the International Classification of Diseases (ICD) system. Exposure to antibiotics during hospitalisation in relation to infectious diseases and other common inpatient diagnoses can be assessed through medical records. The aim of this national population-based study was to estimate how diagnoses of infectious diseases and other common diagnoses in the National Patient Register corresponded to the use of antibiotics in hospitalised pregnant women and young children.
Key notes
This study assessed the proportion of pregnant women and children with specific inpatient diagnoses who were treated with antibiotics based on data from 3797 medical records. A quarter of premature children received antibiotics and their use was more common in bacterial infections (62.4–90.6%) than viruses (6.3–22.2%) in children from one to 5 years of age. Pregnant women were more likely to receive antibiotics following Caesarean sections (40.1%) than vaginal deliveries (11.1%).
©2014 Foundation Acta Pædiatrica. Published by John Wiley & Sons Ltd 2015 104, pp. 392–395
Almqvist et al.
Individual antibiotic exposure in hospital
MATERIAL AND METHODS A cohort of all pregnant women and children from birth to 5 years of age with a discharge date between July 2, 2005 and November 30, 2011 was identified from the Swedish National Patient Register. The start date of each pregnancy was estimated from the Swedish Medical Birth Register, as the date of birth minus the gestational age in days. The universal use of the Swedish personal identity number enables researchers to establish links between the Medical Birth Register, Swedish Prescribed Drug Register and the National Patient Register (7). Links between inpatient diagnoses of infectious disease and other common diagnoses in the National Patient Register and dispensed antibiotics, using code J01 from the anatomical therapeutic chemical (ATC) classification system in the Swedish Prescribed Drug Register, were performed at the National Board of Health and Welfare. To limit the study population to those who had only been given antibiotics in hospitals, all individuals who received outpatient-dispensed prescriptions for antibiotics on the day of discharge or the day after, according to the Swedish Prescribed Drug Register, were excluded. Four groups of children were identified for medical record retrieval, based on the children’s age at discharge (birth to 12 months and 1–5 years); prematurity (ICDcodes P072 and P073); malformations, factors influencing health status and cerebral palsy (ICD-codes Q, Z and G80)
and inpatient diagnoses. The fifth group comprised the pregnant women. The sample size needed for each group and diagnosis was calculated based on clinical assumptions. We aimed at 95% confidence intervals of 5% from the estimated proportions (percentages), and for the calculations, we assumed the proportions in the population were between 10 and 90%, depending on the group and diagnosis. The groups were mutually exclusive. Group 1 comprised children from birth to 12 months with a diagnosis of prematurity (n = 320). Group 2 comprised children from birth to 12 months without a main or secondary diagnosis of prematurity and without a main diagnosis of ICD-code Q or Z (n = 320). Children aged from one to 5 years with a main inpatient diagnosis of infectious disease and other common diagnoses, as specified in Table 1, made up Group 3 (n = 2641), and children aged from one to 5 years without the main diagnoses specified in Table 1 made up Group 4 (n = 320). Group 5 comprised pregnant women with the main diagnoses specified in Table 1 (n = 718). Simple random samples (without replacement) of each of the five groups and diagnosis were drawn by the National Board of Health and Welfare. We were provided with the personal identity number, date of discharge and name of the hospital so that we could request each individual’s medical records. The National Board of Health and Welfare also provided the total number of hospital visits for each
Table 1 Diagnoses and ICD-10 codes used to define groups one to five and descriptive statistics on the numbers exposed to antibiotics, along with corresponding numbers with main and secondary diagnoses Q and/or G80 excluded Exposed to antibiotics
Group
Specific diagnosis
Children 0–12 months 1 Prematurity 2 Diagnoses other than prematurity Children 1–5 years 3 Appendicitis Chronic tonsillitis Gastroenteritis Bacterial sepsis Meningitis Unspecific viral infection Pneumonia Asthma Acute bronchitis and URTI Skin infection Seizures Unspecific abdominal pain Cystitis and pyelonephritis 4 Other main diagnoses than those in group 3 Pregnant women 5 Maternal diagnoses Vaginal delivery Caesarian Section (CS) CS (twins) and multiple delivery
ICD 10 codes
P072, P073 All ICD codes† K35 J35 A08, A09 A40, A41 A87, G00, G02,G03 B34 J18 J45, J46 J20, J21, J06 L08 R56 R10 N30, N10
O268B, K80, K81, K82, K83, O23, N30, N39, N10, O85 O80 O820, O821 O842, O848
Yes n (%)
69 (25.1) 49 (18.1)
Missing n (%)
7 (2.6) 11 (4.1)
29 1 26 43 88 64 242 6 64 57 14 13 269 57
(67.4) (3.2) (8.8) (89.6) (62.4) (22.1) (84.9) (4.5) (22.2) (82.6) (6.1) (6.3) (90.6) (21.4)
3 2 11 2 3 7 5 4 1
46 5 71 55
(16.4) (11.1) (40.1) (43.7)
Yes* n (%)
60 (24.1) 46 (17.6)
6 12 5 15
(7.0) (6.5) (3.7) (4.2) (2.1) (2.4) (1.8) (3.0) (0.4) – (2.6) (5.8) (1.7) (5.6)
28 1 24 38 88 59 222 6 53 55 14 13 245 52
(66.7) (3.2) (8.4) (88.4) (62.4) (21.2) (84.5) (4.6) (20.0) (83.3) (6.1) (6.4) (91.1) (21.0)
20 5 14 9
(7.1) (11.1) (7.9) (7.1)
46 5 70 55
(16.4) (11.1) (39.8) (43.7)
*Main and secondary diagnosis of malformations and cerebral palsy (ICD-10 Q and G80) excluded. † Except P072, P073 as main or secondary diagnoses, and Q or Z as main diagnoses.
©2014 Foundation Acta Pædiatrica. Published by John Wiley & Sons Ltd 2015 104, pp. 392–395
393
Individual antibiotic exposure in hospital
Almqvist et al.
specified group and diagnosis during the study period (Table S1). Starting in January 2012, the study participants’ medical records, corresponding to the date of discharge from the hospitals for the specific diagnosis of interest, were requested by a letter to the head of the ward where the patient had been treated. Two reminder letters were sent out before the end of data collection in June 2013. Each medical record was manually reviewed, and all available information regarding diagnosis and exposure to antibiotics was extracted from text and drug administration records. Antibiotics were further categorised into groups based on ATC-codes when the substance name was available. We requested medical records for 4319 hospital visits based on a study population of approximately 1.2 million children and pregnant women. Permission for the study was obtained from the Regional Ethical Review board in Stockholm, Sweden.
RESULTS We received 3797 medical records of the 4319 requested, which corresponded to a response rate of 88%. The 522 missing medical records could not be found on the wards or the wards did not respond to our request. Table 1 describes the diagnoses and ICD-10 codes used to identify each group and to identify medical records at hospitals in combination with the discharge date. Table 1 also displays the number and proportion of hospital visits with information on treatment with antibiotics, broken down into yes and missing. Table S1 displays diagnoses and ICD-10 codes along with descriptive statistics on hospital visits in the total population, requested and received medical records. In total, 17 750 children between birth and 12 months had a diagnosis of prematurity. A quarter (25%) of the reviewed medical records reported treatment with antibiotics, which corresponded to 4437 children in the total population. About three quarters (74.6%) of the cases that recorded antibiotic use also provided information on type of antibiotics that could be related to an ATC-code. In the group of children aged from one to 5 years, diagnoses that were commonly associated with bacterial infections were more commonly treated with antibiotics; appendicitis (67.4%), bacterial sepsis (89.6%), meningitis (62.4%), pneumonia (84.9%) and pyelonephritis (90.6%). Diagnoses more often associated with viruses were treated less often with antibiotics; gastroenteritis (8.8%), unspecific viral infections (22.1%), acute bronchitis and upper respiratory tract infections (22.2%) and unspecific abdominal pain (6.3%). The proportion of children treated with antibiotics was not different in the group of children with diagnoses of malformations or cerebral palsy. Of the 973 that recorded antibiotic use, 87.2% also provided information on the ATC-code for the type of antibiotic. We reviewed a sample of 45 medical records for the 334 520 pregnant women who had a vaginal delivery, and this showed that 11.1% had been treated with antibiotics. In
394
addition, 16.1% of the women with the most common diagnoses during pregnancy and 40.1% of those who received a Caesarean section had been treated with antibiotics. The majority (83.6) of the records that mentioned antibiotics use also provided information on the type of antibiotics.
DISCUSSION Our results, based on the National Patient Register and medical records, show that a quarter of children below 1 year of age, who were hospitalised with a diagnosis of prematurity, and a high proportion of children with diagnoses of infectious diseases, such as bacterial sepsis, pneumonia and pyelonephritis were treated with antibiotics in hospital. Children diagnosed with virus-associated infections were less likely to be treated with antibiotics, which reflect treatment guidelines and recommendations (8). We also show that antibiotic treatment was more common in pregnant women delivered by Caesarean section than vaginal delivery and that approximately one-sixth of the women with the more common diagnoses during pregnancy were treated with antibiotics. Previous population-based studies have reported point prevalence of antibiotics treatment in inpatient care, but these were not specific to the individual (9,10). The Swedish Strategic Program against Antimicrobial Resistance studied all hospital admissions for children from birth to 16 years of age in selected hospitals during five two-week periods, and this showed that on average one third of children was exposed to antibiotics (9). Although this method is suitable for assessing the risk for antibiotic resistance and the adherence to clinical guidelines on a population level, it is not able to evaluate individual antibiotic exposure for longterm follow-up or to study a selected group of patients. There are also ongoing efforts to develop a better understanding of how to interpret viral findings and to reduce the unnecessary use of antibiotics, as infections such as acute respiratory illness account for a large proportion of all visits to paediatric health facilities (11). Our study is a population-based study to assess individual antibiotic treatment in hospitals, based on a thorough collection and review of a sample of medical inpatient records from children and pregnant women in national Swedish registers with complete coverage. Although there was some missing information on the type of antibiotics and the exact ATC-codes, we were able to assess the proportion of antibiotic treatment per diagnosis. This will allow future studies on the association between antibiotics and subsequent disorders to take separate diagnoses into account (1–4).
ACKNOWLEDGEMENTS We would like to acknowledge Gerd Agerberg and Michael Broms for their excellent data collection and database € rn management and Associate Professors Helle Kieler, Bjo € ve Lindemalm and Professor Jonas F Wettermark, Synno
©2014 Foundation Acta Pædiatrica. Published by John Wiley & Sons Ltd 2015 104, pp. 392–395
Almqvist et al.
Ludvigsson for valuable discussions. We are also indebted to the hospital wards that provided medical records.
CONFLICT OF INTEREST None.
FUNDING Financial support was provided by the Swedish Research Council (grant no 2011-3060) and the Swedish Initiative for Research on Microdata in the Social And Medical Sciences (SIMSAM) framework grants no. 80748301 and 340-20135867, with other grants provided by the Stockholm County Council (ALF project), the Swedish Heart Lung Foundation and the Strategic Research Program in Epidemiology at Karolinska Institutet.
Individual antibiotic exposure in hospital
6.
7.
8.
9.
10.
References 1. Alm B, Erdes L, Mollborg P, Pettersson R, Norvenius SG, Aberg N, et al. Neonatal antibiotic treatment is a risk factor for early wheezing. Pediatrics 2008; 121: 697–702. 2. Marra F, Marra CA, Richardson K, Lynd LD, Kozyrskyj A, Patrick DM, et al. Antibiotic use in children is associated with increased risk of asthma. Pediatrics 2009; 123: 1003–10. 3. Almqvist C, Wettermark B, Hedlin G, Ye W, Lundholm C. Antibiotics and asthma medication in a large register-based cohort study – confounding, cause and effect. Clin Exp Allergy 2012; 42: 104–11. 4. Ortqvist AK, Lundholm C, Kieler H, Ludvigsson JF, Fall T, Ye W, et al. Antibiotics in fetal and early life and subsequent childhood asthma: nationwide population based study with sibling analysis. BMJ 2014; 349: g6979. 5. Hellman J, Olsson-Liljeqvist B, Bengtsson B, Greko C, eds (2012) SWEDRES-SVARM 2012. Use of antimicrobials and
11.
occurrence of antimicrobial resistance in Sweden. Solna/ Uppsala: Strama. €nds antibiotika p Hur anva a sjukhus? STRAMAS punktprevalensstudier (PPS) 2003, 2004, 2006, 2008 och 2010. [Internet]. Strama-Slutenv ard and Folkha¨lsomyndigheten. Available from http://www.strama.se/dyn/,84,.htm. Ludvigsson JF, Otterblad-Olausson P, Pettersson BU, Ekbom A. The Swedish personal identity number: possibilities and pitfalls in healthcare and medical research. Eur J Epidemiol 2009; 24: 659–67. Craig JC, Williams GJ, Jones M, Codarini M, Macaskill P, Hayen A, et al. The accuracy of clinical symptoms and signs for the diagnosis of serious bacterial infection in young febrile children: prospective cohort study of 15 781 febrile illnesses. BMJ 2010; 340: c1594. Stramas punktprevalensstudier 2003-2010 (2011). [Internet]. Strama, SMI. Available from http://www.folkhalsomyndighe ten.se/documents/smittskydd-sjukdomar/antibiotika-vardhyg ien/statistik-fran-stramas-PPS-2003(2010-fohm2014).pdf Kimland E, Nydert P, Odlind V, Bottiger Y, Lindemalm S. Paediatric drug use with focus on off-label prescriptions at Swedish hospitals – a nationwide study. Acta Paediatr 2012; 101: 772–8. Rhedin S, Lindstrand A, Rotzen-Ostlund M, Tolfvenstam T, Ohrmalm L, Rinder MR, et al. Clinical utility of PCR for common viruses in acute respiratory illness. Pediatrics 2014; 133: e538–45.
SUPPORTING INFORMATION Additional Supporting Information may be found in the online version of this article: Table S1. Hospital visits in the total population for diagnoses and ICD-10 codes during the study period, along with requested and received medical records.
©2014 Foundation Acta Pædiatrica. Published by John Wiley & Sons Ltd 2015 104, pp. 392–395
395
Copyright of Acta Paediatrica is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
