Implementation of a Computerized Decision Support System to Improve the Appropriateness of Antibiotic Therapy Using Local Microbiologic Data

Abstract

A prospective quasi-experimental study was undertaken in 218 patients with suspicion of nosocomial infection hospitalized in a polyvalent ICU where a new electronic device (GERB) has been designed for antibiotic prescriptions. Two GERB-based applications were developed to provide local resistance maps (LRMs) and preliminary microbiological reports with therapeutic recommendation (PMRTRs). Both applications used the data in the Laboratory Information System of the Microbiology Department to report on the optimal empiric therapeutic option, based on the most likely susceptibility profile of the microorganisms potentially responsible for infection in patients and taking into account the local epidemiology of the hospital department/unit. LRMs were used for antibiotic prescription in 20.2% of the patients and PMRTRs in 78.2%, and active antibiotics against the finally identified bacteria were prescribed in 80.0% of the former group and 82.4% of the latter. When neither LMRs nor PMRTRs were considered for empiric treatment prescription, only around 40% of the antibiotics prescribed were active. Hence, the percentage appropriateness of the empiric antibiotic treatments was significantly higher when LRM or PMRTR guidelines were followed rather than other criteria. LRMs and PMRTRs applications are dynamic, highly accessible, and readily interpreted instruments that contribute to the appropriateness of empiric antibiotic treatments.

Approaches to promoting the appropriate use of antibiotics through hospital electronic prescribing systems: a scoping review

“…Most evidence came from specialty settings (such as intensive care, emergency medicine and paediatrics); and from high‐risk patient groups (e.g. those at risk for or with infections); and from tertiary care …”

“…We found that most work has been conducted in relation to CPOE and CDS systems, focusing on interventions that targeted individual prescribers through prompts and reminders …”

“…The majority of systems were “home‐grown” standalone systems, implemented in single centres and extensively customised; but we also identified some work on packaged applications . Research activity surrounding more complex and distributed functionalities that integrated ePrescribing with medical notes, pharmacy systems and laboratory results was less common …”

“…In relation to practitioner performance, the studies reported improved turnaround time for first dose of antibiotic, improved selection of antibiotics (e.g. in line with local protocols), optimal dosing, improved adherence to guidelines/protocols, and a decrease in the total number of antibiotics prescribed . Some studies also reported a reduction in prescribing errors and increased de‐escalation, as well as improved antibiotic sensitivity and increase in sensitivity of surgical site infection detection .…”

“…There was limited evidence surrounding interventions that had a broader organisational focus and those that investigated approaches to re‐using data held within systems …”