Welcome to the ESPID 2021 Meeting Calendar

Background

The number of paediatric patients visiting the Emergency Department (ED) with non-urgent problems is increasing, which contributes to poor patient flow and ED crowding. Implementation of a fast track can address this problem. We aimed to identify which low urgent triaged children are suitable for a fast track.

Methods

This study is part of the MOFICHE study (Management and Outcome of Febrile children in Europe), an observational study including routine data of febrile children (0-18 years) attending European EDs. Children triaged as low urgent by the Manchester Triage System in nine EDs were included. Children are suitable for a fast track, defined as a lower level of care, if they have minimal resource use and are discharged home. Multivariable logistic regression analyses regarding presenting symptom and management (blood tests, imaging and admission) were performed. Covariates included patient characteristics, referral, previous medical care or antibiotic use, visit hours and ED. Presenting symptoms were categorized into: neurological, (n=234) respiratory (n=8376), gastrointestinal (n=1943) and others (n=3430, reference category).

Results

We included 13,983 children (median age 2.7, IQR 1.3-5.2). The majority was self-referred (68%), had respiratory symptoms (60%) and 31% received antibiotics. The neurological group underwent imaging more often (aOR 1.8, 95%CI 1.1-2.9) and were admitted more frequently (aOR 1.9, 95%CI 1.4-2.6). The respiratory group had less blood tests performed (aOR 0.6, 95%CI 0.6-0.7), more imaging (aOR 1.8, 95%CI 1.6-2.0) and were less frequently admitted (aOR 0.6, 95%CI 0.5-0.7). The gastro-intestinal group had more blood tests performed (aOR 1.2, 95%CI 1.1-1.4) and were admitted more frequently (aOR 1.4, 95%CI 1.2-1.6).

Conclusions

Our study shows that children with respiratory symptoms are most suitable for a fast track, in contrast to the neurological group who underwent more extensive management.

Clinical Trial Registration

This study is not a Clinical Trial.

Background

Over prescription of antibiotics in paediatric Emergency Departments (EDs) leads to increased antimicrobial resistance. Optimisation of antibiotic prescription is a critical goal for antimicrobial stewardship initiatives.

Methods

Using the European PERFORM (www.perform2020.org) BIVA-database of febrile children attending the ED who had blood tests performed, cases were phenotyped using the PERFORM bacterial/viral probability algorithm. We determined empiric antibiotic use in children in view of the individual’s final phenotype of definite bacterial (DB) or definite viral (DV) infection. Antibiotics prescribed were classified according to WHO AWaRe (Access, Watch, Reserve).

Results

Of 1080 febrile children with a definite final diagnosis, 582 were assigned a DB and 498 a DV final phenotype. Of note, initial working diagnoses were largely similar between DB and DV phenotypes, except urinary tract infection and respiratory tract infection.

A total of 542 (93.1%) DB and 281 (57.0%) DV were prescribed empiric antibiotics during admission. In the DB group, 55 (10.2%) children received oral and 487 (89.9%) intravenous/intramuscular (IV/IM) antibiotics. In comparison, 67 (23.8%) children with a DV phenotype received oral and 214 (76.2%) IV/IM antibiotics (p<0.00001). The top 3 antibiotics were third-generation cephalosporins, penicillins and penicillin/beta-lactamase inhibitor combinations in both DB and DV. A total of 408 (75.3%) DB and 212 (75.4%) DV had ≥ 1 WHO Watch antibiotics prescribed.

Conclusions

Differentiating bacterial/viral aetiology of febrile illness is difficult on initial presentation to the ED. A significant proportion of children with a final DV phenotype received antibiotics during admission, predominantly classified as WHO Watch. Rapid and accurate point-of-care tests in the ED differentiating between DB and DV could significantly reduce antibiotic prescribing, thereby improving antimicrobial stewardship.

Acknowledgements

This project received funding from the European Union’s Horizon2020 programme under grant agreement 668303.

Clinical Trial Registration

Not applicable.

Background

Lack of diagnosis information in the claims databases poses a limit in the assessment of prescribing appropriateness in Italian primary care. We aimed to establish a baseline of the current antibiotic prescription patterns over the years by age and diagnosis. Secondly, we want to describe the switching/prolongation patterns between different types of antibiotics.

Methods

This retrospective cohort study assesses antibiotic prescriptions retrieved from Pedianet, a paediatric primary-care database, from 1st January 2012 to 31st December 2018. Descriptive diagnoses were manually classified, according to ICD-9CM code or free text, and then linked to the specific PI. Descriptive analysis was performed and then stratified by diagnosis class, calendar year, and by age class. Prevalence of prescription index (PI), antibiotic index (AI) and treatment switch and prolongation were the outcomes considered. Mann-Kendall Test and Poisson regression were used to assess trend quantification.

Results

In total 611,352 AI were included. From 2012 to 2018 the AI rate decreased significantly (MK test; p=0.004) from 1.67 AI/person-years (95% CI: 1.66-1.68) in 2012 to 1.22 AI/ person-years (95% CI: 1.21- 1.23) in 2018 by 6% yearly (RR: 0.94; 95% CI 0.93-0.94). PIs were associated with an upper respiratory tract infection diagnosis in 23% of cases followed by pharyngitis (21%), bronchitis and bronchiolitis (12%), and acute otitis media (12%). In total, 8% of treatment episodes were prolonged or switched in class (mainly co-amoxiclav, macrolides, and III-gen cephalosporins, Figure1).

Conclusions

To our knowledge, this is the first study assessing antibiotic prescribing practices in pediatric primary care in Italy by years, childre's age, and diagnosis. Our study confirmed that broad-spectrum antibiotics prescriptions remain high in Italian primary-care setting. Estimating the drivers for antibiotic prescriptions allows defining the area of intervention for antibiotic stewardship policies in primary care.