Dataset Information

A Bayesian Model Based on Local Phenotypic Resistance Data to Inform Empiric Antibiotic Escalation Decisions.

ABSTRACT:

Introduction

Clinicians commonly escalate empiric antibiotic therapy due to poor clinical progress without microbiology guidance. When escalating, they should take account of how resistance to an initial antibiotic affects the probability of resistance to subsequent options. The term "escalation antibiogram" (EA) has been coined to describe this concept. One difficulty when applying the EA concept to clinical practice is understanding the uncertainty in results and how this changes for specific patient subgroups.

Methods

A Bayesian model was developed to estimate antibiotic resistance rates in Gram-negative bloodstream infections based on phenotypic resistance data. The model generates a series of "credible" curves to fit the resistance data, each with the same probability of representing the true rate given the inherent uncertainty. To avoid overfitting, an integrated penalisation term adaptively smooths the curves given the level of evidence.

Results

Rates of resistance to empiric first-choice and potential escalation antibiotics were calculated for the whole hospitalised population based on 10,486 individual bloodstream infections, and for a range of specific patient groups, including ICU (intensive care unit), haematolo-oncology, and paediatric patients. The model generated an expected value (posterior mean) with 95% credible interval to illustrate uncertainty, based on the size of the patient subgroup. For example, the posterior means of piperacillin/tazobactam resistance rates in Gram-negative bloodstream infection are different between patients on ICU and the general hospital population: 27.3% (95% CI 18.1-37.2 vs. 13.4% 95% CI 11.0-16.1) respectively. The model can also estimate the probability of inferiority between two antibiotics for a specific patient population. Differences in optimal escalation antibiotic options between specific patient groups were noted.

Conclusions

EA analysis informed by our Bayesian model is a useful tool to support empiric antibiotic switches, providing an estimate of local resistance rates, and a comparison of antibiotic options with a measure of the uncertainty in the data. We demonstrate that EAs calculated for the whole hospital population cannot be assumed to apply to specific patient group.

SUBMITTER: Bamber R

PROVIDER: S-EPMC11343932 | biostudies-literature | 2024 Sep

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

A Bayesian Model Based on Local Phenotypic Resistance Data to Inform Empiric Antibiotic Escalation Decisions.

Bamber Ranjeet R Sullivan Brian B Gorman Léo L Lee Winnie W Y WWY Avison Matthew B MB Dowsey Andrew W AW Williams Philip B PB

Infectious diseases and therapy 20240718 9

<h4>Introduction</h4>Clinicians commonly escalate empiric antibiotic therapy due to poor clinical progress without microbiology guidance. When escalating, they should take account of how resistance to an initial antibiotic affects the probability of resistance to subsequent options. The term "escalation antibiogram" (EA) has been coined to describe this concept. One difficulty when applying the EA concept to clinical practice is understanding the uncertainty in results and how this changes for s ...[more]

PMID: 39023716

Similar Datasets

Project description:Background: Antibiotics are often prescribed empirically to treat infection syndromes before causative bacteria and their susceptibility to antibiotics are identified. Guidelines on empiric antibiotic prescribing are key to effective treatment of infection syndromes, and need to be informed by likely bacterial aetiology and antibiotic resistance patterns. We aimed to create a clinically-relevant composite index of antibiotic resistance for common infection syndromes to inform recommendations at the national level. Methods: To create our index, we used open-access antimicrobial resistance (AMR) surveillance datasets, including the ECDC Surveillance Atlas, CDDEP ResistanceMap, WHO GLASS and the newly-available Pfizer ATLAS dataset. We integrated these with data on aetiology of common infection syndromes, existing empiric prescribing guidelines, and pricing and availability of antibiotics. Results: The ATLAS dataset covered many more bacterial species (287) and antibiotics (52) than other datasets (ranges = 8-11 and 16-32 respectively), but had a similar number of samples per country per year. Using these data, we were able to make empiric prescribing recommendations for bloodstream infection, pneumonia and cellulitis/skin abscess in up to 44 countries. There was insufficient data to make national-level recommendations for the other six syndromes investigated. Results are presented in an interactive web app, where users can visualise underlying resistance proportions to first-line empiric antibiotics for infection syndromes and countries of interest. Conclusions: We found that whilst the creation of a composite resistance index for empiric antibiotic therapy was technically feasible, the ATLAS dataset in its current form can only inform on a limited number of infection syndromes. Other open-access AMR surveillance datasets are largely limited to bloodstream infection specimens and cannot directly inform treatment of other syndromes. With improving availability of international AMR data and better understanding of infection aetiology, this approach may prove useful for informing empiric prescribing decisions in settings with limited local AMR surveillance data.

Project description:In light of rising antibiotic resistance, better methods for selection of empiric antibiotic treatment based on clinical and microbiological data are needed. Most guidelines target specific clinical infections, and variably adjust empiric antibiotic selection by certain patient characteristics. Coverage estimates reflect the probability that an antibiotic regimen will be active against the causative pathogen once confirmed and can provide an objective basis for empiric regimen selection. Coverage can be estimated for specific infections using a weighted incidence syndromic combination antibiograms (WISCAs) framework. However, no comprehensive data combining clinical and microbiological data for specific clinical syndromes are available in Switzerland. We therefore describe estimating coverage from semi-deterministically linked routine microbiological and cohort data of hospitalised children with sepsis. Coverage estimates were generated for each hospital and separately pooling data across ten contributing hospitals for five pre-defined patient risk groups. Data from 1,082 patients collected during the Swiss Paediatric Sepsis Study (SPSS) 2011-2015 were included. Preterm neonates were the most commonly represented group, and half of infants and children had a comorbidity. 67% of neonatal sepsis cases were hospital-acquired late-onset whereas in children 76% of infections were community-acquired. Escherichia coli, Coagulase-negative staphylococci (CoNS) and Staphylococcus aureus were the most common pathogens. At all hospitals, ceftazidime plus amikacin regimen had the lowest coverage, and coverage of amoxicillin plus gentamicin and meropenem were generally comparable. Coverage was improved when vancomycin was included in the regimen, reflecting uncertainty about the empirically targeted pathogen spectrum. Children with community-acquired infections had high coverage overall. It is feasible to estimate coverage of common empiric antibiotic regimens from linked data. Pooling data by patient risk groups with similar expected pathogen and susceptibility profiles may improve coverage estimate precision, supporting better differentiation of coverage between regimens. Identification of data sources, selection of regimens and consideration of pathogens to target for improved empiric coverage is important.

Dataset Information

A Bayesian Model Based on Local Phenotypic Resistance Data to Inform Empiric Antibiotic Escalation Decisions.

Introduction

Methods

Results

Conclusions

Publications

A Bayesian Model Based on Local Phenotypic Resistance Data to Inform Empiric Antibiotic Escalation Decisions.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets