1799. Impact of Real-Time Electronic Notifications to Pharmacists of Rapid Diagnostic Blood Culture Results. (26th November 2018)

• Record Type: Journal Article
• Authors: Caulder, Liz; Beardsley, James; Palavecino, Elizabeth; Van Dyke, Erica; Johnson, James; Ohl, Christopher; Luther, Vera; Williamson, John
• Is Part Of: Open forum infectious diseases. Volume 5(2018)Supplement 1
• Page Start: S510
• Page End: S510
• Publication Date: 2018-11-26
• Subjects: Communicable diseases -- Periodicals; Medical microbiology -- Periodicals; Infection -- Periodicals; 616.9
• DOI: 10.1093/ofid/ofy210.1455
• Languages: English
• ISSNs: 2328-8957

Abstract

Abstract: Background: Rapid diagnostic tests that utilize multiplex PCR technology provide faster time to pathogen identification, but maximizing the impact on outcomes is dependent upon who is available to respond to test results. In June 2017, pharmacists began receiving in-basket notifications of positive results from the institution's FilmArray BCID assay. The objective of this study was to determine the impact on antibiotic utilization associated with this method of communicating results. Methods: This was a retrospective, observational, before-and-after study at an academic medical center with an established stewardship program. Inclusion criteria: Adult patients age ≥18 admitted to an ICU or oncology unit with ≥1 positive blood culture containing a gram-positive organism identified by FilmArray BCID. Patients with polymicrobial infection, concomitant infection caused by a different organism, antibiotics started before admission, or death prior to organism identification were excluded. Data were collected during a 4-month period before (PRE) and a 4-month period after (POST) implementation of in-basket notifications. Stewardship metrics and other outcome measures were compared between the two groups. Pharmacists received no targeted stewardship training on how to respond to results. Results: Ninety-two patients met study criteria (49 PRE and 43 POST). Patients were age 62 ± 16, male (55%), and 77 (84%) were located in an ICU. Median Charlson Comorbidity Index was 4 and Pitt Bacteremia Score was 1. Sixty-seven patients were considered to have noncontaminant bloodstream infection. Median results for these patients are listed in the table. Patients with contaminants ( n = 25) had 3.5 and 7 antibiotic-free days in the PRE and POST groups, respectively ( P = 0.34). Conclusion: In-basket notifications did not significantly improve antibiotic utilization or clinical outcomes. Active interventions and antimicrobial stewardship initiatives are needed in combination with rapid diagnostic tests. Disclosures: All authors: No reported disclosures.