Respiratory tract infections (RTIs) are common in febrile children visiting the general practitioner (GP) or the emergency department (ED).1–3 4–6 7
Several national and international guidelines are in place to guide physicians in the management of children with RTIs.8–14 1 , 15–18 19 , 20
To rationalize and reduce the prescription of antibiotics, we need accurate data on the management of RTIs. Comparative studies on this subject at different levels of healthcare have not been performed. This study offers a unique possibility to investigate this topic.
The aim of this study was to compare clinical characteristics and antibiotic management of children with RTIs in healthcare facilities representative of the 3 levels of healthcare in 1 geographic area in The Netherlands.
METHODS
Study Design
This study is part of the MOFICHE (Management and Outcome of Fever In Children) study. MOFICHE is a prospective observational study aiming to assess the management and outcome of children with fever who seek medical treatment and present to EDs across Europe. The MOFICHE study is part of the PERFORM study (Personalised Risk Assessment in Febrile illness to Optimise Real-life Management across the European union), which aims to improve the diagnosis and management of febrile children across Europe.
Data of face-to-face patient contacts from 3 healthcare facilities representative of the 3 levels of healthcare within the adherence area of the university hospital Radboudumc, Nijmegen, The Netherlands, were collected between January 1 and December 31, 2017. Sensitive search strategies were applied on the patient databases of different healthcare settings to make sure all relevant cases were selected.
In The Netherlands, children with RTIs are primarily seen by a GP at their practice or at an out of office hours GPs center. In general, patients will only be seen at the ED in a general hospital after referral by a GP. Some patients will bypass the GP and present to the ED directly. General hospital pediatricians will refer complex or critically ill patients to a tertiary care ED in a university hospital. In general, more children with complex comorbidities will be seen at the tertiary care ED.
For the primary care setting, patients from a large general practice: Thermion, Nijmegen-Lent were included. This practice has 8 GPs, responsible for 2882 children <18 years registered in the study period. The practice participates in the Family Medicine Network (FaMe-Net) which totals 6 practices with 26 GPs. Antibiotic prescription in the study practice is representative of the other 5 practices: the prescription rate for respiratory infections for all ages in the study practice was 50.2/1000 person years (py) comparing to an average prescription rate of 48.2/1000 py in FaMe-Net.
For the secondary care setting, patients from a general hospital, the Canisius Wilhelmina Hospital, Nijmegen were included. This hospital has 14 pediatricians, 25 pediatric beds and a catchment area including 88,124 children <18 years in the study period.
For the tertiary care setting, patients from a university hospital: Radboudumc, Nijmegen were included. This hospital has 68 pediatricians and 88 pediatric beds covering all different pediatric subspecialties and a catchment area including 408,240 children <18 years in the study period.
The national guideline for otitis media in children advocates restrictive prescription of antibiotics: only children under 6 months old, with risk factors for a complicated course or with ill appearance or children with persistent symptoms for more than 48 hours are treated. The antibiotic of first choice is amoxicillin.11
For pharyngitis/tonsillitis, the national guideline restricts antibiotic prescription to cases with suspicion of a peritonsillar abscess, lymphadenitis and risk factors for a complicated course or ill appearance. The antibiotic of first choice is pheneticillin for primary care and pheneticillin or amoxicillin for the hospital setting.12 , 13
The national primary care guideline on cough/pneumonia advises prescription of amoxicillin as the first choice in children with suspected pneumonia. In hospital guidelines, amoxicillin is the first choice for community-acquired pneumonia with mild symptoms and cefuroxime or amoxicillin/clavulanic acid for cases with moderate to severe symptoms.8 , 10 , 13
The national guidelines on bronchiolitis, asthma, rhinitis and laryngitis do not support the use of antibiotics.8 , 14
For our study, children <18 years of age with: (1) fever; temperature ≥ 38 °C, measured at the GP or ED or (2) history of (measured) fever less than 3 consecutive days before presentation were eligible for inclusion. Children could participate in more than one GP or ED contact. Only children with a RTI were included. Exclusion criteria were as follows: (1) complex comorbidities defined as the presence of a chronic condition in 2 or more body systems or the presence of a malignancy (described earlier by Simon et al21
Clinical data of patient contacts were entered in an electronic database. For every contact, the presumed focus of infection was noted including otitis media; pharyngitis/tonsillitis; other upper RTIs (URTIs: rhinitis, sinusitis and laryngitis) and lower RTIs (LRTIs). For the primary and tertiary care setting, all patient contacts during the study period were included. For secondary care, all patient contacts during the first 2 weeks of every month were included.
The following variables were extracted from the patient file: sex, age, body temperature, duration of febrile illness, ill appearance, comorbidity, respiratory symptoms (coughing, other signs of RTI, increased work of breathing, tachypnea and cutaneous saturation). Management was categorized as: any further investigation (laboratory, microbial and radiological), performing of C-reactive protein, white blood cell count or chest radiograph, referral to a higher level of healthcare and admission to hospital following the ED visit.
Antibiotic prescription was divided in total prescriptions and prescriptions of the antibiotic of first choice according to the national guideline.
Data Analysis
For the descriptive analyses, data are expressed as medians and interquartile range or as frequencies and percentages for categorical variables. A comparison of baseline characteristic and antibiotic prescriptions between the healthcare settings was made using X2 and Mann-Whitney U tests by the closed testing procedure; a P < 0.05 was considered significant. We estimated absolute numbers of consultations and antibiotic prescriptions for RTIs per 100,000 children <18 years old based on the numbers of children in the catchment areas of the different healthcare facilities as described earlier. For secondary care, we corrected for the inclusion during 2 weeks/month.
RESULTS
A total of 1815 patient contacts met criteria for evaluation. Cases with complex comorbidities and incomplete data on antibiotic prescription were excluded, leaving 892 cases available for analysis (Fig. 1 ).
FIGURE 1.: Selection of eligible patient contacts.
Patient Characteristics and Management
Table 1 presents the baseline characteristics of the 892 cases in the 3 healthcare settings. Median age was 2 years, most of the cases; 725 (81%), were children under 5 years of age. In the majority of secondary (43%) and tertiary care (55%) cases, febrile illness existed less than 1 day, whereas in primary care, this was significantly longer: mostly more than 2 days (in 48%, P < 0,001). Ill appearance and increased work of breathing were more prevalent in secondary and tertiary care. In primary care, cutaneous oxygen saturation was recorded in 6.6% (33/497) of the cases, in contrast to around 90% in the 2 other healthcare settings. When measured in primary care, cutaneous oxygen saturation was reduced in 24% (8/33) compared with 17% (31/182) and 7% (12/177) in secondary and tertiary care (P = 0.002).
TABLE 1. -
Baseline Characteristics
Primary Care (n = 497)
Secondary Care (n = 199)
Tertiary Care (n = 196)
Significance*
General characteristics
Age in years, median (IQR)
1.9 (1.0–4.2)
1.6 (0.8–3.5)
2.2 (0.9–4.4)
P = 0.077
Male sex, n (%)
267 (54)
113 (57)
104 (53)
P = 0.711
Duration fever in days, n (%)
P < 0.001
<1
129 (26)
85 (43)
107 (55)
P < 0.001 † ‡ §
1–2
71 (14)
20 (10)
20 (10)
P = 0.174
>2
240 (48)
66 (33)
51 (26)
P < 0.001 † ‡
Unknown
57 (11)
28 (14)
18 (9)
P = 0.314
Temperature, median (IQR)
37.9 (37.3–38.5)
38.4 (37.6–39.2)
38.3 (37.7–39.0)
P < 0.001 † ‡
Medical care same complaint in last 5 d, n (%)
43 (9)
75 (38)
77 (39)
P < 0.001 † ‡
Referral by GP, n (%)
NA
177 (89)
101 (52)
P < 0.001 §
Symptoms, n/nrep (%)
Ill appearance
137/403 (34)
93/185 (50)
69/165 (42)
P = 0.001 †
Coughing
311/360 (86)
144/167 (86)
122/155 (79)
P = 0.069
Other signs
341/357 (96)
142/156 (91)
147/170 (86)
P = 0.001 † ‡
Work of breathing
31/208 (15)
89/149 (60)
60/174 (34)
P < 0.001 † ‡ §
Tachypnea for age
31/74 (42)
92/156 (59)
59/137 (43)
P = 0.008 † §
Saturation < 95%
8/33 (24)
31/182 (17)
12/177 (7)
P = 0.002 ‡ §
Diagnostics, n (%)
Any diagnostics performed
23 (5)
83 (42)
88 (45)
P < 0.001 † ‡
Blood investigations
16 (3)
59 (30)
77 (39)
P < 0.001 † ‡ §
Imaging
1 (0.2)
40 (20)
29 (15)
P < 0.001 † ‡
Other tests¶
6 (1)
35 (18)
45 (23)
P < 0.001 † ‡
Further management, n (%)
Supplemental oxygen therapy
0
32 (16)
16 (8)
P < 0.001 † ‡ §
Nebulized medication
Bronchodilatators
28 (6)
57 (29)
30 (15)
P < 0.001 † ‡ §
Adrenaline
0
2 (1)
1 (0.5)
P = 0.105
Budesonide
0
1 (0.5)
3 (2)
P = 0.025 ‡
Referral to higher level of care
13 (3)
NA
NA
—
Admission to hospital
NA
94 (47)
56 (29)
P < 0.001 §
IQR, interquartile range; NA, not applicable.
*P values are derived from closed testing procedures. Significant differences are shown in bold (defined as P < 0.05) and found between:
†Primary and secondary care .
‡Primary and tertiary care .
§Secondary and tertiary care .
¶Included cultures, urine, cerebrospinal fluid, stools or naropharyngeal tests.
Further investigations were performed in 42% (83/199) of cases in secondary and 45% (88/196) of cases in tertiary care. Bronchodilators were significantly more often prescribed in the secondary care setting, in 29% (57/199) of cases. Three percent (13/497) of cases seen in primary care were referred to a higher level of healthcare. The admission rate was highest in secondary care (47%, 94/199).
Data on the focus of the RTIs are shown in Figure 2 . Otitis media was most frequently diagnosed in primary care. Other URTIs were mostly diagnosed in primary and tertiary care setting, while cases with LRTIs were more common in secondary care.
FIGURE 2.: Focus of RTIs in different levels of healthcare.
Antibiotic Prescription
The overall antibiotic prescription rate for all RTIs in our study was 29%. This rate did not differ between settings: 29%, 27% and 31% in, respectively, primary, secondary and tertiary care. By extrapolating these rates, we estimated the absolute numbers of antibiotic prescriptions for RTIs in children <18 years per setting to be 5031, 131 and 15 antibiotic prescriptions per 100,000 children in the population per year for primary, secondary and tertiary care, respectively (Table 2 ).
TABLE 2. -
Consultations and Antibiotic Prescriptions for RTIs per 100,000 Population <18 Years of Age per Year
Kolom1
Primary Care
Secondary Care
Tertiary Care
Consultations for RTIs
17,280
493
48
Antibiotic prescription for RTIs (%)
5031 (29)
131 (27)
15 (31)
Figure 3 presents antibiotic prescription rates per focus of infection per healthcare setting. For otitis media, antibiotic prescription was 55%, 60% and 62% in primary, secondary and tertiary care. In 92%, 58% and 63%, the antibiotic of first choice (amoxicillin) was prescribed. Other antibiotics prescribed were amoxicillin/clavulanic acid, azithromycin and antibiotic ear drops. After exclusion of children under 6 months and/or with ill appearance, the prescription rate was 43% in primary care (numbers too low in other settings).
FIGURE 3.: Antibiotic prescription per focus in different levels of healthcare.
For pharyngitis/tonsillitis, the prescription rate was 39%, 15% and 67% in primary, secondary and tertiary care: a significantly lower rate in secondary care as compared with primary (P = 0.038) and tertiary care (P = 0.002). In 47%, 50% and 50%, the antibiotic of first choice was prescribed in primary, secondary and tertiary care: pheneticillin and for hospital setting also amoxicillin. Amoxicillin/clavulanic acid and azithromycin were the other antibiotics prescribed.
Prescription rates for other URTIs were 4%, 10% and 17% in primary, secondary and tertiary care, respectively.
LRTI cases were divided in cases with and without need for nebulization. In the group without nebulization, prescription rates were 83%, 47% and 65% for primary, secondary and tertiary care, respectively, with a significantly lower rate in secondary care compared with primary care (P < 0.001). In primary care, the antibiotic of first choice (amoxicillin) was prescribed in 87%. This rate was lower in secondary care (73%) and tertiary care (33%) where a broad range of alternative antibiotics were prescribed: amoxicillin/clavulanic acid, azithromycin, ceftriaxone and clindamycin. In LRTI cases with need for nebulizations, lower prescription rates were found: 55%, 19% and 39% for primary, secondary and tertiary care, respectively.
DISCUSSION
We analyzed the management of children with RTIs in one region in The Netherlands in primary, secondary and tertiary care. We found that antibiotic prescription rates were not significantly different among the 3 levels of healthcare. Extrapolating this to absolute numbers per setting we can conclude that primary care is responsible for the vast majority of antibiotic prescriptions per 100,000 children per year (Table 2 ). We believe that this underpins the need for antibiotic stewardship not only in the hospital setting but most certainly also in primary care.22 , 23
We found an overall antibiotic prescription rate for RTIs of 29%, which is comparable with the 27% found in an earlier Dutch study in primary care.24 25 , 26
We focused on antibiotic prescription for different RTIs in the 3 levels of healthcare.
For otitis media, in all settings, antibiotic prescription was high: 55%–62%. This rate is higher than rates reported in other studies conducted in The Netherlands (45%–50%) and comparable or lower compared with other countries: 82% in the United Kingdom, 62% in Belgium, 75% in Sweden and 58% in Norway.17 , 24 , 25 , 27–30
For pharyngitis/tonsillitis, primary and tertiary care had high prescription rates of 39% and 67%, respectively, which was significantly higher than 15% in secondary care. Other Dutch studies show a rate of 50%–60%, whereas rates in other countries are higher: 93% in the United Kingdom, 78% in Belgium, 86% in Sweden and 69% in Norway.17 , 24 , 25 , 27–29
We observed a significant number of URTIs in tertiary care (66% of total cases) with a relatively high prescription rate of 17%. Patients with comorbidities in this setting are often instructed to be seen at the tertiary care ED with a lower threshold (only 52% were referred by GP, compared with 89% in secondary care). Although these cases were excluded in our study, this triaging habit and associated defensive prescribing by pediatricians looking after this population may have contributed to the high number of URTIs and frequent prescription of antibiotics as compared with other settings.
In this study, LRTIs were defined as cases with pneumonia, bronchiolitis and asthma exacerbations: the former with an indication for antibiotic treatment and the latter with no indication. Therefore, cases were selected based on need for nebulization.
In the group without nebulizations—thus with a likely diagnosis of pneumonia—antibiotic prescription rates were high in primary care (81%) and lower in secondary (47%) and tertiary care (63%). Other Dutch studies also show high prescription rates in primary care of 70% and a lower rate of 29% for admitted patients in a secondary care setting.24 , 27 , 28 25 , 30–32
Adherence to the national guideline was good in primary care: in 87% of cases amoxicillin was prescribed. It could be improved in secondary (73%) and tertiary care (33%) where other antibiotics were prescribed: amoxicillin/clavulanic acid, azithromycin, ceftriaxone and clindamycin. Although hospital guidelines allow prescription of broader spectrum antibiotics in severe cases of LRTIs, the proportion of amoxicillin prescription in the hospital settings seems too low and can be improved.
In the group of LRTI with nebulizations, antibiotic prescription decreased to 55%, 19% and 39% for primary, secondary and tertiary care, respectively, meaning that still a significant number of these children received antibiotics, especially in the primary care setting. Previous Dutch studies show prescription rates of 46%–54% for bronchiolitis in primary care, comparable to what we found.24 , 27 , 28 27 , 29
We chose to exclude cases with complex comorbidities to make an objective and unbiased comparison between the 3 levels of healthcare. Because guidelines on the management of RTIs often exclude patients with (complex) comorbidities, it could be that antibiotics are prescribed with lower thresholds in this group of patients.
Despite this correction, as expected, the case mix in primary versus secondary or tertiary care setting was different. Patients with a RTI presenting at the ED were more ill appearing and more often had breathing difficulties than children presenting to the GP. Also, the duration of fever before presentation was shorter in these settings, indicating more acute illness. Patients in secondary care had increased work of breathing and required oxygen and bronchodilators more often compared with patients in tertiary care. Also, the number of lower RTIs was higher in secondary care as compared with tertiary care; we believe that this reflects the higher number of patients with bronchiolitis and asthma presenting to this setting. More otitis media was found in primary care, which confirms that, in the Netherlands, otitis media is typically managed by the GP.
STRENGTHS AND LIMITATIONS
To our knowledge, this is the first study to investigate and compare the management of children with RTIs in all levels of healthcare in 1 region in 1 year. A thorough overview of the cases was achieved by collecting detailed data on various important clinical aspects.
We only collected data from 2 weeks per month in secondary care, potentially making these data less robust. However, as the total number of cases is comparable with the other settings and all seasons were included, we think comparisons between settings can still be made reliably.
Although the database was quite extensive, some details were lacking: for example presence of wheeze in LRTI to differentiate pneumonia from bronchiolitis/asthma. Data on temperature and clinical symptoms may have been incompletely registered by physicians. Also factors known to influence antibiotic prescription including parental concern or experience level of the physician were not available.
CONCLUSIONS
The antibiotic prescription rates for RTIs in our study are in line with other Dutch studies and in general lower than the rates described in other countries. However, there is still room for improvement: antibiotic prescription rates in children with RTIs are similar in primary, secondary and tertiary care. As the absolute number of children with RTIs managed in primary care is much higher than in secondary and tertiary care, the vast majority of antibiotics are prescribed in primary care. Therefore, it is crucial to implement antibiotic stewardship in the primary care setting and not only in hospital settings to have a significant impact on antibiotic consumption for RTIs in children.
GPs better adhere to prescription of narrow-spectrum antibiotics advocated by the guidelines than pediatricians, suggesting antibiotic stewardship efforts in hospital settings should focus on the adherence to guidelines concerning the choice of antibiotics.
REFERENCES
1. Stam J, van Stuijvenberg M, Grüber C, et al.; Multicenter Infection Prevention Study 1 (MIPS 1) Study Group. Antibiotic use in infants in the first year of life in five European countries. Acta Paediatr. 2012;101:929–934.
2. Yorita KL, Holman RC, Sejvar JJ, et al. Infectious disease hospitalizations among infants in the United States. Pediatrics. 2008;121:244–252.
3. de Bont EG, Lepot JM, Hendrix DA, et al. Workload and management of childhood fever at general practice out-of-hours care: an observational cohort study. BMJ Open. 2015;5:e007365.
4. Tsolia MN, Psarras S, Bossios A, et al. Etiology of community-acquired pneumonia in hospitalized school-age children: evidence for high prevalence of viral infections. Clin Infect Dis. 2004;39:681–686.
5. Kumar P, Medigeshi GR, Mishra VS, et al. Etiology of acute respiratory infections in infants: a prospective birth cohort study. Pediatr Infect Dis J. 2017;36:25–30.
6. Kool M, Monteny M, van Doornum GJ, et al. Respiratory virus infections in febrile children presenting to a general practice out-of-hours service. Eur J Gen Pract. 2015;21:5–11.
7. Veldhoen ES, Wolfs TF, van Vught AJ. Changes in infectious disease mortality among children in the Netherlands. Eur J Pediatr. 2009;168:465–468.
8. Verheij TJM, Hopstaken RM, Prins JM, et al. NHG-Standaard Acuut hoesten (Eerste herziening). Huisarts Wet. 2011;54:68–92.
9. National Institute for Health and Care Excellence (NICE). NICE guideline: Cough (acute): antibiotic prescribing.2019. Available at:
https://www.nice.org.uk/guidance/ng120 . Accessed December 20, 2020.
10. Terheggen-Lagro SWJ, de Vreede I, Tuut MK, et al. Richtlijn diagnostiek en behandeling van ongecompliceerde pneumonie bij kinderen in de tweede en derde lijn.2015. Available at:
https://www.nvk.nl/themas/kwaliteit/richtlijnen/richtlijn?componentid=7864320&tagtitles=Infectieziekten%252ben%252bImmunologie%2CIntensive%252bCare%2CLongziekten . Accessed December 20, 2020.
11. Damoiseaux RAMJ, Venekamp RP, H. Eekhof JA, et al. NHG-standaard otitis media acuta bij kinderen (Derde herziening). Huisarts Wet. 2104;57:648.
12. Dagnelie CF, de Jongh E, Lemmen WH, et al. NHG-standaard acute keelpijn (Derde herziening). Huisarts Wet. 2015;58:422–429.
13. SWAB.Dutch Working Party on Antibioic Policy.2019. Available at:
https://children.swabid.nl/ . Accessed December 20, 2020.
14. de Kruiff CC, Boluyt N, Rings EHHM, et al. Evidence-Based Richtlijnen voor 5 acute problemen in de kindergeneeskunde: Bronchiolitis, Acuut astma, Laryngitis subglottica, Epileptische aanvallen/status epilepticus > 1 maand, Dehydratie. 2012. Available at:
https://www.nvk.nl/themas/kwaliteit/richtlijnen/richtlijn?componentid=6356993&tagtitles=Infectieziekten+en+Immunologie,Intensive+Care,Longziekten . Accessed December 20, 2020.
15. Bandell RAM, Dekkers T, Semmekrot BA, et al. Macrolide prescription in Dutch children: compliance with guidelines. Eur J Clin Microbiol Infect Dis. 2019;38:675–681.
16. Damoiseaux RA, de Melker RA, Ausems MJ, et al. Reasons for non-guideline-based antibiotic prescriptions for acute otitis media in The Netherlands. Fam Pract. 1999;16:50–53.
17. Uijen JH, Bindels PJ, Schellevis FG, et al. ENT problems in Dutch children: trends in incidence rates, antibiotic prescribing and referrals 2002-2008. Scand J Prim Health Care. 2011;29:75–79.
18. Otters HB, van der Wouden JC, Schellevis FG, et al. Trends in prescribing antibiotics for children in Dutch general practice. J Antimicrob Chemother. 2004;53:361–366.
19. Gerber JS, Ross RK, Bryan M, et al. Association of broad- vs narrow-spectrum antibiotics with treatment failure, adverse events, and quality of life in children with acute respiratory tract infections. JAMA. 2017;318:2325–2336.
20. World Health Organization. WHO factsheet antibiotic resistance.2020. Available at:
https://www.who.int/news-room/fact-sheets/detail/antibiotic-resistance . Accessed December 20, 2020.
21. Simon TD, Cawthon ML, Stanford S, et al.; Center of Excellence on Quality of Care Measures for Children with Complex Needs (COE4CCN) Medical Complexity Working Group. Pediatric medical complexity algorithm: a new method to stratify children by medical complexity. Pediatrics. 2014;133:e1647–e1654.
22. Fletcher-Lartey S, Yee M, Gaarslev C, et al. Why do general practitioners prescribe antibiotics for upper respiratory tract infections to meet patient expectations: a mixed methods study. BMJ Open. 2016;6:e012244.
23. Colliers A, Coenen S, Bombeke K, et al. Understanding general practitioners’ antibiotic prescribing decisions in out-of-hours primary care: a video-elicitation interview study. Antibiotics (Basel). 2020;9:115.
24. Dekker ARJ, Verheij TJM, van der Velden AW. Antibiotic management of children with infectious diseases in Dutch primary care. Fam Pract. 2017;34:169–174.
25. O’Brien K, Bellis TW, Kelson M, et al. Clinical predictors of antibiotic prescribing for acutely ill children in primary care: an observational study. Br J Gen Pract. 2015;65:e585–e592.
26. Lee GC, Reveles KR, Attridge RT, et al. Outpatient antibiotic prescribing in the United States: 2000 to 2010. BMC Med. 2014;12:96.
27. Tyrstrup M, van der Velden A, Engstrom S, et al. Antibiotic prescribing in relation to diagnoses and consultation rates in Belgium, the Netherlands and Sweden: use of European quality indicators. Scand J Prim Health Care. 2017;35:10–18.
28. Ivanovska V, Hek K, Mantel Teeuwisse AK, et al. Antibiotic prescribing for children in primary care and adherence to treatment guidelines. J Antimicrob Chemother. 2016;71:1707–1714.
29. Rebnord IK, Sandvik H, Mjelle AB, et al. Factors predicting antibiotic prescription and referral to hospital for children with respiratory symptoms: secondary analysis of a randomised controlled study at out-of-hours services in primary care. BMJ Open. 2017;7:e012992.
30. Van Deursen AM, Verheij TJ, Rovers MM, et al. Trends in primary-care consultations, comorbidities, and antibiotic prescriptions for respiratory infections in The Netherlands before implementation of pneumococcal vaccines for infants. Epidemiol Infect. 2012;140:823–834.
31. Quaak CH, Cové E, Driessen GJ, et al. Trends in paediatric inpatient antibiotic therapy in a secondary care setting. Eur J Pediatr. 2018;177:1271–1278.
32. Gotta V, Baumann P, Ritz N, et al.; ProPAED Study Group. Drivers of antibiotic prescribing in children and adolescents with febrile lower respiratory tract infections. PLoS One. 2017;12:e0185197.
