We all probably agree that a successful antimicrobial stewardship program (AMS) should aim to improve patient treatment outcomes, improve patient safety, and reduce the impact of antimicrobial resistance (AMR) in patients and the wider society. Yet how to implement a successful, resilient, AMS program is less well understood.
Programs to restrict access to antimicrobials have been shown to limit inappropriate antimicrobial usage and improve patient outcomes in the short-term, however the impact this intervention has on the development of a partnership with AMS teams and front-line clinicians long term remains unclear. A divisive relationship can sometimes occur and result in long term loss of engagement.
Recent sepsis campaigns, new NICE Sepsis guidance in the UK, and litigious consequences of missed infected diagnosis, have understandably introduced a cautious approach to empirical antimicrobial prescribing. Diagnostic insufficiency, heterogeneous presentation of many infections and increased awareness of multi-drug resistant organisms make empiric prescribing a leap of faith for many clinicians.
When the aim is to do no harm, balancing the needs of the patient in front of you against the societal impact of prescribing ultra-broad spectrum antimicrobials is challenging. Thus AMS programs need robust surveillance systems to better identify patients where empirical antimicrobials are inappropriate and promptly change to targeted therapy based on current or recent microbiology results.
Challenges of antimicrobial stewardship programs
With one in three patients expected to be receiving systemic antimicrobials at any one time within a hospital, the scale of a hospital-based AMS program becomes problematic. AMS teams in the main do not have the resource to review all patients to ensure appropriateness of prescribing.
Audit and feedback, typically on 3 – 12 month cycles, may improve discrete process measures but this feedback loop is still too long for true behavioral change. Instead, real-time feedback may drive improvements in prescribing habits whilst correcting, and thus minimizing the impact of, inappropriate treatment on patients promptly.
As healthcare evolves and adapts to technological advances, AMS programs must also change. The availability of Electronic Prescribing and Medicines Administration (EPMA) systems should provide this stimulus for change; now with access to real-time prescribing information and patient specific measures (including inflammatory markers, recent microbiology & susceptibility results, and physiological observations) to make informed decisions on antimicrobial prescribing decisions. Using computerized decision support systems (CDSS), an AMS team can quickly identify patients for targeted review.
Testing computerized decision support systems
We recently introduced such a CDSS (ICNET® Pharmacy) into practice to test this hypothesis. Using automated alerts, targeted patients are ‘pushed’ to the AMS team for timely review. Using live feeds for example, a patient with an ESBL E.coli infection started by the clinical team on an empiric cephalosporin in line with local guidelines can be promptly identified and prescription amended within minutes by the AMS team.
Any off-guideline prescribing can be quickly identified and the AMS team can then discuss concerns with the parent team; rarely is off-guideline prescribing done in disregard or contempt for guidelines but usually in response genuine concern for patients.
The CDSS we trialed is web-based, requires minimal technological input and ‘pushes’ patients to the team for review.
Here, the AMS team can work with the parent team, advise the most appropriate treatment, and reassure the prescriber. This collaborative working enables the parent team to build their AMS knowledge and confidence, empowers them to take more responsibility for their patients’ AMR risks, and changes AMS behaviors within the organization.
The CDSS we trialed is web-based, requires minimal technological input and ‘pushes’ patients to the team for review. Time taken to identify patients in need of review is minimal, enabling the AMS team to evolve from office-based activity to a ward-based service. End-of-bed consults with the patient and clinical team are now the norm. Expansion of the service to the whole hospital including the less-traditionally AMS serviced areas of maternity and pediatric wards has been possible with no additional changes in staffing resource through efficiencies made using the CDSS.
In addition to the objective data presented in our evaluation published in ARIC, we feel the CDSS empowered our AMS team to progress from ‘policing’ to ‘enabling’ our clinical teams in their antimicrobial prescribing. Interventions recorded for correcting inappropriate prescribing dropped significantly indicative of behavioral change.
Instead the AMS team are making interventions to optimize dosing, escalate therapy in early treatment failure, and facilitating early discharges. These patient-orientated interventions are aligned closely with trust priorities thus further supporting the AMS team activities.
The landscape of eHealth and mHealth (mobile health) is rapidly evolving, with both in-house and commercial CDSS being developed for AMS and for other healthcare areas. Whilst we have high hopes for these systems improving patient outcomes and impacting AMR trends, we are staunch advocates of there being multi-modal evaluations of such systems before widespread adoption. Our paper in Antimicrobial Resistance & Infection Control helps detail part of this landscape, and we look forward to further evaluating the evolution and integration of CDSS tools into clinical practice.
Comments