Reducing Antibiotic Overuse: An Intervention with Positive Outcomes
Abstract & Commentary
By Leslie A. Hoffman, RN, PhD
Professor Emeritus, Nursing and Clinical & Translational Science, University of Pittsburgh
SYNOPSIS: Active, daily communication between infectious disease and critical care practitioners significantly reduced antibiotic overuse without increasing mortality.
SOURCE: Rimawi RH, et al. Impact of regular collaboration between infectious diseases and critical care practitioners on antimicrobial utilization and patient outcome. Crit Care Med 2013;41:2099-2107.
This study was undertaken to test the potential that routine daily assessment of antibiotic use by an infectious disease (ID) specialist (fellow) could further improve best practices for ICU patients. At the time the study was instituted, the institution had implemented an antimicrobial stewardship program and rounds included pharmacist consultation. However, chart reviews indicated "considerable opportunity" for improvement. Baseline data were collected during a 3-month preintervention period followed by the 3-month intervention that was scheduled 1 year after baseline data collection to avoid seasonal discrepancy. The ID fellow reviewed charts of all medical ICU (MICU) patients receiving antibiotic therapy, discussed complex cases with the ID attending, and rounded with the critical care team to provide recommendations. Antibiotic use was recorded as days of therapy. In this calculation, each antibiotic given on a single day was recorded as 1 day of therapy (DOT), i.e., 1 antibiotic/day = 1 DOT and 3 antibiotics/day = 3 DOT. DOTs were divided by length of stay (LOS); the ratio was multiplied by 1000 and expressed as DOT/1000 patient days to allow comparison between cohorts with differing LOS. Antibiotics given prior to MICU admission or after MICU discharge were not included.
A total of 246 charts were reviewed: 123 in the preintervention phase and 123 in the post-intervention phase. There were no significant between-group differences in age, gender, race, APACHE II scores, or types of infections. Fewer patients received antibiotics that did not correspond to guidelines during the intervention period (P < 0.0001). There was a significant reduction in antibiotic use during the intervention period (1590 vs 1420 DOT/1000 patient days; P = 0.03274). There was also a 17-fold increase in the use of narrow-spectrum penicillins (e.g., penicillin G and nafcillin) in the intervention phase (67 vs 4 DOT/1000 patient days; P = 0.0322). All-cause hospital mortality was significantly lower in the intervention phase (P = 0.0367) with no difference in all-cause MICU mortality (P = 0.4970). There was a significant reduction in days of mechanical ventilation (6.07 vs 10.1; P = 0.0053) and MICU LOS (7.78 vs 10.29; P = 0.0188) in the intervention period, compared to the preintervention period. Cost of antibiotics in the preintervention period exceeded that in the intervention period by a difference of $22,486.
Efforts to better match antibiotic prescriptions with patient needs are clearly warranted. In this institution, an antibiotic stewardship program had already been instituted. However, despite existence of this program and the inclusion of a pharmacist on rounds, there continued to be a need for improvement, based on chart reviews. The ID specialist made 180 recommendations during the 3-month course of the study. Specific recommendations related to shortening (n = 77), stopping (n = 53), narrowing (n = 34), or broadening therapy (n = 4), as well as converting to oral administration (n = 8) and changing therapy due to an adverse effect (n = 4). Thus, most (72%) recommendations related to more timely cessation of therapy.
The authors posed several reasons for this outcome. Rather than lack of knowledge, they suggested that reluctance to change therapy was potentially related to hesitation to abruptly change antibiotics after a handoff of care, even when there was awareness that a change was indicated, until familiar with the case. Because the ID specialist was actively involved in discussions during MICU rounds, there could be "consensus discussions" leading to change. When proposing the intervention, there was concern that disputes might arise between specialties. This did not occur and the volume of ID consultations was not affected.
When calculating cost saving, the analysis only considered hospital pharmaceutical acquisition costs. No costs were attributed to the ID specialist, who was a fellow, since the experience was considered part of training. The time requirement for providing consultation (estimated at 2 hours/day) was not excessive and the experience was judged an excellent academic fellowship opportunity. The study, therefore, met its goal of achieving better antibiotic stewardship with no untoward consequences. The benefits of an ID specialist participating in rounds on patients with sepsis on clinical units have been well documented. The current study extends these benefits to the ICU setting.