Development of an Evidence-Based Antimicrobial Stewardship Smartphone App in a Tertiary Academic Pediatric and Women’s Health Centre in Canada

Abstract

Session: 75. Stewardship: Program Implementation Thursday, October 5, 2017: 12:30 PM Background. Smart phone use by medical professionals is ubiquitous. In a recent survey, > 90% of health care providers were interested in locally developed antimicro-bial stewardship (AMS) and infectious diseases applications ("apps"). We describe the process by which our antimicrobial stewardship program (ASP) developed an app to provide guidance regarding empiric antimicrobial choice, and education about antimi-crobials and pathogens, integrating local laboratory data. We also describe early app uptake. Methods. The IWK Health Centre is a 271-bed tertiary care Pediatric and Women's health centre serving the Maritime Provinces in eastern Canada. Using the Spectrum Mobile Health platform, our ASP developed an app in consultation with pediatric and women's health clinical divisions. Through collaboration with the microbiology laboratory, the app was integrated with our laboratory information system (LIS) allowing real-time access to local antibiogram results. The iPhone-and Android-compatible app was introduced to health care providers through presentations , hospital intranet, email, and word of mouth. Following the official launch, uptake was monitored both in number of app downloads and number of hits. Adherence to empiric treatment guidelines included in the app will be assessed utilizing our existing ASP prospective audit and feedback service. Results. From December 2015 to March 2017, the ASP created content for the IWK AMS App. Three sections were developed. (1) Syndromes: evidence-based empiric treatment guidelines for common syndromes. (2) Antimicrobials: spectrum of activity, dosing regimens, drug monitoring, common usage, adverse effects, drug interactions and pharmacology. (3) Pathogens: information on precautions, local suscepti-bilities through linkage with our recently developed virtual antibiogram, associated syndromes, and epidemiology. In May 2017, the app was launched. Within the first 24 hours, it was downloaded 157 times and accessed 1,193 times. Session: 75. Stewardship: Program Implementation Thursday, October 5, 2017: 12:30 PM Background. Accuracy of vancomycin trough monitoring has come into question. We evaluated an area under the curve (AUC) monitoring protocol and 3 different dosing calculators at a single center. Methods. Adult inpatients with vancomycin AUC monitoring from 5/2016-1/2017 were included. We excluded those with peaks drawn less than 1 hour after infusion. AUC was calculated with Sawchuck-Zaske (SZ) methodology. This was compared with two publicly available online calculators: ClinCalc and UCSF's infectious disease monitoring program (IDMP). Paired t-tests were used to compare AUCs from ClinCalc and IDMP to SZ. We collected renal function, infection, microbiology, and dosing data. Clinical outcomes included survival to discharge, discharge disposition , rate of acute kidney injury (AKI) per Risk-Injury-Failure-Loss-End Stage Renal Disease criteria, and bacterial clearance. Results. 29 subjects were included. Median age was 48 years, 59% were male, median weight was 80.4 kg. Median daily dose was 3000 mg (32.4 mg/kg). No patient had renal impairment at baseline. Skin and soft-tissue infections were most common , 11 (38%). Six subjects had bacteremia, 2 had confirmed endocarditis. MRSA was isolated in 14 cases (48%). Median duration of vancomycin was 11 days. Mean 24 hour AUC (standard deviation) was 654 (203) mg/L for SZ, 536 (278) mg/l for ClinCalc (P = 0.02) and 556 (187) mg/L (P = 0.004) for IDMP. AUC differences of at least 30% compared with SZ were identified in 14 (48%) and 6 (21%) subjects evaluated with ClinCalc and IDMP respectively. AKI occurred in three subjects: two risk and one injury. All survived to discharge; 52% discharged home, 41% to a skilled nursing facility, 7% left against medical advice. Twenty (69%) had bacterial clearance, 2 (7%) had persistently positive cultures, 7 (24%) were treated empirically. Conclusion. Vancomycin AUC varies with calculation methodology. The SZ method was impacted by dose and duration of infusion. ClinCalc showed greater variability in higher weight patients. ClinCalc and IDMP calculated lower AUCs than SZ, and recommended higher doses to target an AUC:MIC ratio of at least 400. As