Hospital Pharmacy - June 2019 - 177

177

Mogle et al
Table 1. Demographics of Survey Respondents.
Demographics
Level of training
PGY-1
PGY-2
PGY-3
PGY-4, PGY-5, or fellow
Attending physician
Practice area(s)
Internal medicine
General surgery
Emergency medicine
Orthopedic
Infectious diseases
Neurology
Pulmonary critical care
Physical medicine and rehabilitation
Time caring for patients in ICU
0%
1%-25%
26%-50%
51%-75%
76%-100%
Average weekly use of carbapenems
0 patients
1-5 patients
6-10 patients

n (%)
17 (17.9)
26 (27.4)
19 (20.0)
14 (14.7)
19 (20.0)
68 (71.6)
14 (14.7)
9 (9.5)
2 (2.1)
2 (2.1)
1 (1.1)
1 (1.1)
1 (1.1)
14 (14.7)
57 (60.0)
18 (18.9)
3 (3.2)
3 (3.2)
44 (46.3)
46 (48.4)
5 (5.3)

Note. Respondents were permitted to select multiple practice areas. PGY
= postgraduate year; ICU = intensive care unit.

Answers to multiple-choice questions assessing baseline
therapeutic knowledge are shown in Table 2. In assessing
provider knowledge of penicillin allergy and beta-lactam
cross reactivity, 51.5% of providers correctly selected that
carbapenems and advanced-generation cephalosporins have
similar and low (ie, ≤1%) rates of cross reactivity.5,6 Based
on these responses, approximately one-third of providers
favored the use of carbapenems over other beta-lactams in
the setting of a penicillin allergy.
According to the institutional antibiogram, susceptibility
rates of Pseudomonas aeruginosa to cefepime, piperacillintazobactam, and meropenem are 94%, 87%, and 85%,
respectively. Aztreonam susceptibilities are not routinely
performed. The institutional antibiogram is widely accessible through the electronic medical record. Overall, 29.5% of
respondents correctly identified cefepime as the drug with
the highest percentage of susceptible P aeruginosa isolates at
our institution. Nearly one-fifth (18.9%) of providers selected
meropenem as having the greatest susceptibility rate for P
aeruginosa, which is incorrect based on the institutional
antibiogram.
Finally, familiarity with comparative efficacy data
between carbapenems and beta-lactam/beta-lactamase

inhibitors (BLBLIs) for the treatment of bacteremia due to
ESBL-producing Enterobacteriaceae from low inoculum
sources of infection was assessed. Although carbapenems
remain the mainstay of treatment for severe ESBL infections,
recent literature suggests similar outcomes when BLBLIs are
utilized for the treatment of bacteremia due to ESBLproducing Enterobacteriaceae from low inoculum sources of
infection.7 However, more than half (52.6%) of providers
considered carbapenems to be superior to piperacillin-tazobactam in this setting. Knowledge deficits in each of these 3
areas were apparent across all levels of training. The discrimination index for multiple-choice questions 1, 2, and 3 were
0.57, 0.37, and 0.44, respectively, indicating a reasonably
good to very good discrimination index for each question.8

Discussion
This is the first study we are aware of investigating factors
that influence empiric carbapenem prescribing in acute care
hospitals. This study not only illustrates the factors considered most important to providers when prescribing carbapenems empirically but also highlights misconceptions stemming
from knowledge deficits that may lead to unnecessary prescribing. A previous systematic review evaluating antibiotic
prescribing in long-term care facilities (LTCFs) identified
that misconceptions and knowledge deficits among providers played a significant role in antibiotic prescribing.9
However, this study was specific to the LTCF context and
not specific for factors influencing carbapenem use, and
thus, may not be generalizable as such.
At our institution, providers indicated that the factor most
strongly influencing their decision to prescribe a carbapenem
empirically was the suspected pathogens at the site of infection. Because it was not directly assessed, we cannot draw
conclusions as to which pathogens providers are targeting
when choosing to prescribe carbapenems empirically.
However, it is logical to assume that a carbapenem would be
employed in scenarios where MDR bacteria including P
aeruginosa and ESBL-producing Enterobacteriaceae are
suspected, and knowledge pertaining to empiric coverage
and treatment of these pathogens was objectively assessed
by the survey.
A significant proportion of providers did not correctly
identify the antimicrobial agent providing the greatest
likelihood of susceptibility for P aeruginosa at our institution. Nearly 1 out of every 5 providers chose meropenem
as the agent with greatest susceptibility rate for P aeruginosa despite having poorer susceptibility rates compared
with cefepime and piperacillin-tazobactam. This suggests
an unjustified preference for carbapenems among these
providers, presumably based on the perception that carbapenems possess greater broad-spectrum activity or
because providers may be unfamiliar with how to access or
interpret the institutional antibiogram. In addition, more
than half of respondents appeared to be unfamiliar with
Hospital Pharmacy - June 2019

Table of Contents for the Digital Edition of Hospital Pharmacy - June 2019
