1088 — Association of Characteristics of VA Antimicrobial Stewardship Programs with Antibiotic Use

Jones MM, VA Salt Lake City HCS; Graber CJ, VA Greater Los Angeles HCS; Chou AF, Oklahoma City VA; Zhang Y, VA Salt Lake City HCS; Goetz MB, VA Greater Los Angeles HCS; Madaras-Kelly K, Boise VA Medical Center; Samore MH, VA Salt Lake City HCS; Glassman PA, VA Greater Los Angeles HCS;

Objectives:
Suboptimal use of antimicrobial therapy occurs 30-50% of the time in inpatient settings. Antimicrobial stewardship (AS) programs have been implemented to promote appropriate antibiotic use thus, minimizing adverse events, optimizing patient outcomes, and improving patient safety. However, little information is available about AS program structures and processes that may predict effective antibiotic use. We used data from the 2012 VA Healthcare Analysis and Informatics Group (HAIG) survey on antimicrobial stewardship to map AS characteristics of individual facilities to their antibiotic use.

Methods:
The HAIG survey was administered at 130 VA facilities that provide acute care. Factor analyses identified 32 organizational factors, mapped to the Promoting Action on Research Implementation in Health Services framework: (1) Evidence; (2) Contexts; and (3) Facilitation. Bivariate analyses evaluated associations between organizational characteristics and antibiotic usage per 1000 patient days, using data from the Veterans Informatics and Computing Infrastructure (VINCI).

Results:
One factor in the evidence domain, 6 factors in context, and 2 factors in facilitation were associated with decreased inpatient antibiotic use: presence of an on-site infectious diseases consult service (Evidence)(p = 0.02), presence of postgraduate training programs (Context)(p = 0.04), disease-specific stewardship policies (Context)(p = 0.03), infectious diseases fellow/attending involvement in antibiotic approvals (Context)(p = 0.002), facility complexity (Context)(p = 0.002), degree and duration of physician/pharmacist involvement in ASP (Context)(p = 0.04), systematic review for antibiotic de-escalation (Context)(p = 0.01), presence of infectious diseases attendings or clinical pharmacists on acute care teams (Facilitation)(p = 0.002), and infectious diseases training of the ASP pharmacist (Facilitation)(p = 0.006). Three factors in the context domain were associated with increased antibiotic use: a perceived need for organizational support for stewardship (p = 0.02), non-infectious diseases physician involvement in ASP (p = 0.005), and having antibiotic stop orders in place (p = 0.04). A confirmatory analysis restricted to antibiotic use among patients with admissions for a primary non-infectious etiology demonstrated similar findings.

Implications:
We identified several AS characteristics that are associated with lower inpatient antibiotic use.

Impacts:
In further development of AS Programs within the VA, clear associations between program components and antibiotic use should inform program design and policy. Along with findings from pending multivariable analyses that will include additional antibiotic use outcomes, this study will help inform future AS Program implementation.