Setting
The intervention took place over 12 months from January 2018 to December 2018 at an 850-bed quaternary-care teaching hospital. For the first 3 months, the intervention was developed and tested on two medicine units, and was expanded to include two additional medicine units for the remaining 9 months from April 2018 to December 2018.
At our hospital, each medicine unit has an IDR team comprised of a hospitalist, nursing staff, clinical pharmacist, case managers, and either advance practice providers (APP) or medical residents. Most medicine units at our facility are mainly employed by APP, which include nurse practitioners and physician assistants, with the exception of one unit that is driven by medical residents on hospitalist-led teams. Hospitalists provide oversight of IDR, though the attending-of-record for a given patient may be different.
Prior to the implementation of the intervention, the existing ASP activities at our institution included review of sterile site cultures for appropriate antibiotic prescribing, prospective audit and feedback focusing on costly agents or those reserved for resistant pathogens, and IV to PO conversions conducted by clinical pharmacists for select antibiotics with good oral bioavailability. This central approach allows for oversight of prescribing at our institution and track trends.
The Intervention: Handshake Stewardship
We implemented a version of handshake stewardship that was different than that of Hurst et al [5]. In their study, the pharmacist-physician ID team would physically locate the teams during unit rounds (15 teams total), and communicated recommendations to providers. In our study, the hospitalist and clinical pharmacist who are non-ID, unit-based clinicians would serve as the stewards on each unit, and thus we had four groups implementing ASP interventions simultaneously.
The goals of our intervention were similar to those cited in the literature: discontinuation of antimicrobials when a source of infection is not present, reduction in excess duration of antimicrobials, and antimicrobial de-escalation based on culture data.
Potential targets for intervention included patients with uncomplicated cystitis, community-acquired pneumonia, or patients on empiric therapy who were stabilized without a clear infection after approximately 3 days. These were identified prior to the intervention by the central ASP team as having the highest potential for improving antibiotic utilization, but did not exclude other opportunities for antibiotic de-escalation.
Following a process map developed by the central ASP team (figure 1), pharmacists reviewed patients who were prescribed antibiotics using a criteria-based list on the electronic medical record. These lists were reviewed manually to monitor antibiotics used, culture results, and therapy duration. This review is conducted prior to IDR, so that the pharmacist can present this information during IDR.
The hospitalist leads IDR as an opportunity to discuss the overall plan for antibiotics. After IDR, the APP, resident, or pharmacist would communicate stewardship recommendations to the attending-of-record.
Analysis
A 10-question survey was distributed to clinicians that comprise IDR teams on each of the units to understand how subgroups rated the impact of the initiative.
We also measured the change in target inpatient antibiotic utilization (antibiotic days per 1000 patient days) on each of the 4 medicine units before and after intervention (2017 vs 2018). This was a composite measure of the 16 antibiotic agents selected for observation in all groups (figure 3).