OBJECTIVES:

Intravenous (IV) to enteral transition of highly bioavailable antibacterial drugs is associated with improved safety and lower cost. We evaluated the impact of a bundle of stewardship-driven interventions (including in-person stewardship rounding, clinical pathways, and clinical pharmacist-driven enteral transition workflows) on IV versus enteral administration of highly bioavailable antibacterials at a freestanding children’s hospital.

METHODS:

We collected 2010–2018 inpatient usage data for clindamycin, levofloxacin, ciprofloxacin, metronidazole, rifampin, linezolid, and trimethoprim-sulfamethoxazole. We analyzed total use (in days of therapy [DOTs] per 1000 patient-days [PDs]) and the percentage of total use administered enterally, both hospital wide and stratified by unit subgrouping, specifically comparing use 1-year prestewardship implementation with year-5 postimplementation.

RESULTS:

Across the 8-year study window, clindamycin, fluoroquinolones, and metronidazole, together, accounted for 96% of IV DOTs for highly bioavailable antibacterials. Overall, clindamycin use decreased from 44.4 to 20.2 DOTs per 1000 PDs (P < .001), with the enteral percentage of total use increasing from 23% to 43% (P < .001) hospital wide. Overall, fluoroquinolone use decreased from 33.7 to 19.3 DOTs per 1000 PDs (P < .001), with the enteral percentage increasing from 40.7% to 55.9% (P < .001). Overall, metronidazole use increased, and the enteral percentage decreased (42.0% to 33.7%; P = .007). Low-IV-use antibacterials (rifampin, linezolid, and trimethoprim-sulfamethoxazole) showed no significant changes in total use or the enteral percentage of total use.

CONCLUSIONS:

Stewardship interventions were associated with decreased overall use and an increased enteral percentage of total use for both clindamycin and fluoroquinolones, although not metronidazole. These data provide an easy-to-collect benchmark for pediatric hospitals to compare IV with enteral use of highly bioavailable antibacterials within the context of overall antibacterial use.

Although many highly bioavailable antibacterials are given intravenously (IV) to hospitalized patients, enteral formulations can achieve similar serum and tissue concentrations. In these cases, enteral administration can achieve comparable clinical outcomes while being safer, more convenient, and less costly.1,2  For these reasons, the Centers for Disease Control and Prevention and national stewardship guidelines include IV-to-enteral transition as a goal of antimicrobial stewardship programs (ASPs).3,4  Recent quality improvement interventions have been aimed to address this.5,6  In this report, we describe the change in IV and enteral usage of 7 highly bioavailable antibacterials (clindamycin, levofloxacin, ciprofloxacin, metronidazole, rifampin, linezolid, and trimethoprim-sulfamethoxazole [TMP-SMX]) before and after implementation of multiple, bundled stewardship-driven interventions at a single, freestanding children’s hospital. We report usage data across all hospitalized patients (not just those eligible for IV-to-enteral transition) to establish an easy-to-measure metric and put use within the context of overall antibacterial use.

Our 444-bed freestanding, quaternary-care pediatric hospital in Colorado has ∼15 000 admissions and 97 000 patient-days (PDs) annually. We provide routine and subspecialty care as well as hematopoietic stem cell transplant, immunotherapy, and solid organ transplant (heart, liver, and kidney). Our ASP is a “prospective audit and feedback”–style program, termed “handshake stewardship,” notable for daily, in-person rounding by both an infectious disease physician and infectious disease pharmacist, without the use of prior authorization or other prescribing restrictions.7,8 

Both the rounding model and related stewardship-driven clinical pathways were developed over a 2-year planning period (2011–2013), with pathways for appendicitis, pneumonia, urinary tract infection, musculoskeletal infection, and fever in infants <60 days that address IV versus enteral antibacterial use. Handshake rounds, which began part-time in October 2013 and on all business days in July 2014, serves to answer provider questions, give patient-specific recommendations, and reinforce clinical pathways and policies. Beyond handshake rounds and guidance from clinical pathways, IV-to-enteral transition can also be initiated directly by the clinical pharmacists. Not only do pharmacists round with the inpatient medical teams and make suggestions during rounds, the hospital policy “Antimicrobials – Pharmacist Changing from Intravenous (IV) to Enteral (PO) Formulation” additionally allows clinical pharmacists to transition highly bioavailable antibacterials from IV to enteral using a standing order in the electronic health record (EHR) (Epic, Verona, WI). This change can occur without a physician order if patients meet certain inclusion criteria (age, type and/or severity of infection, and tolerance of enteral food and/or medications, as detailed in the Supplemental Information). This policy, first written in 2009, was minimally used until providers and pharmacists were educated and empowered during ASP planning and implementation.

This combination of daily ASP rounding, directive clinical pathways, empowering clinical pharmacists, and a standing EHR order together represent a bundled approach, in which improving IV-to-enteral transition is addressed in multiple, interconnected ways.

We extracted 2010–2018 inpatient antibacterial usage from the EHR for clindamycin, levofloxacin, ciprofloxacin, metronidazole, rifampin, linezolid, and TMP-SMX for all inpatient units except psychiatry and maternal fetal medicine. This time frame encompasses 1-year pre-ASP implementation (2010–2011), a 2-year planning period (2011–2013), and 5 years postimplementation (2013–2018). Use was measured in days of therapy (DOTs) per 1000 PDs on the basis of midnight census, stratified by IV versus any enteral route, aggregated monthly, then averaged across the time interval. For each antibacterial, 1 DOT is counted for each calendar day during which a unique patient received at least a single dose of that drug. On days in which a patient received both an IV and enteral dose of the same drug, 2 DOT are counted (1 IV and 1 enteral). All hospital PDs were included, even for patients likely ineligible for enteral transition because of nil per os (NPO) status. Year intervals were from October to September, on the basis of the ASP’s October 2013 start.

To understand variability across the hospital, we subdivided use among 3 unit subgroupings: pediatric and cardiac ICUs (excluding the NICU, given low antibacterial use despite many PDs), hematology-oncology (Heme/Onc), including hematopoietic stem cell transplant, and combined medical and surgical (Med/Surg) units.

For each antibacterial, the mean use (in DOTs per 1000 PDs) and 95% confidence intervals (CIs) were calculated for each time period, both hospital wide and by subgrouping. We compared the mean use (DOTs per 1000 PDs) and enteral percentage of total use between the 1-year period preimplementation (2010–2011) and year-5 postimplementation (2017–2018) using a 2-sided t test, with a significance level of 0.007 after Bonferroni correction, given comparisons across multiple time periods. This review was approved by the local organizational research risk and quality improvement review panel.

These 7 bioavailable drugs combine to account for 20.7% of the total hospital antibacterial days across the 8-year study interval. Among them, clindamycin, fluoroquinolones, and metronidazole together account for 96% of the IV antibiotic days because total IV usage for rifampin, linezolid, and TMP-SMX were each <2 DOTs per 1000 PDs. Of the total rifampin use, most was enteral, with a hospital-wide average across the study interval of 4.5 DOTs per 1000 PDs enteral versus 0.6 DOTs per 1000 PDs IV, without significant changes across time. In the setting of low overall use, the enteral percentage of total rifampin use changed from 76.9% (preimplementation) to 92.0% (year-5 postimplementation; P = .02). TMP-SMX, similarly, is used >95% enterally, averaged across the study interval, with no significant changes in use over time. The total use of linezolid was the lowest among all drugs, with combined IV and enteral use of 1.6 DOTs per 1000 PDs (preimplementation) and 1.7 DOT per 1000 PD (year-5 postimplementation), with no statistically significant changes in IV, enteral, or percentage enteral use.

From preimplementation to year-5 postimplementation, the enteral percentage of total clindamycin use hospital wide increased from 23.0% to 43.2% (P < .001), concurrently with a decrease in overall use from 44.3 to 20.2 DOTs per 1000 PDs (P < .001; Table 1; Fig 1). The largest increases in enteral percentage across the study interval were seen in Heme/Onc (14.5% to 35.2% enteral; P = .004) and the ICUs (8.9% to 25.1% enteral; P = .004). However, Med/Surg units accounted for 60% of the total clindamycin use across the study interval, so the Med/Surg increase from 29.7% to 55.8% enteral (P < .001) resulted in the most saved IV DOTs.

TABLE 1

Total Antibacterial Use and Enteral Percentage for Clindamycin, Fluoroquinolones, and Metronidazole by Unit Subgrouping from 2010 to 2018.

LocationPre (2010–2011)Plan (2011–2013)Post-1 (2013–2014)Post-2 (2014–2015)Post-3 (2015–2016)Post-4 (2016–2017)Post-5 (2017–2018)DOTs P Preand Post 5Enteral Percentage P Pre- and Post-5
DOTsa (95% CI)Enteral PercentageDOTsa (95% CI)Enteral PercentageDOTsa (95% CI)Enteral PercentageDOTsa (95% CI)Enteral PercentageDOTsa (95% CI)Enteral PercentageDOTsa (95% CI)Enteral PercentageDOTsa (95% CI)Enteral Percentage
Clindamycin                 
 Hospital wide 44.3 (40.5–48.2) 23.0% 38.7 (35.4–41.9) 31.9% 30.3 (27.1–33.4) 36.3% 27.6 (24.3–30.8) 40.0% 22.7 (20–25.4) 38.9% 24.5 (20.1–28.9) 41.9% 20.2 (17.2–23.2) 43.2% <.001 <.001 
 ICUs 50.3 (40.1–60.4) 8.9% 52.4 (41.7–63.2) 20.3% 44.3 (30.5–58.1) 28.3% 39.9 (32.6–47.1) 26.5% 41.4 (32.6–50.3) 20.4% 44.2 (28.8–59.6) 30.9% 36.5 (22.4–50.7) 25.1% .096 .004 
 Heme/Onc 114 (80.6–147) 14.5% 78.8 (55.8–102) 15.1% 38.3 (24.7–51.9) 25.3% 31.9 (22.8–41.1) 36.6% 24.7 (18.8–30.5) 34.7% 34.1 (27.3–41) 34.4% 32.9 (26.5–39.2) 35.2% <.001 .004 
Med/Surg units 47.4 (41.4–53.4) 29.7% 43.4 (41–45.8) 37.8% 37 (32.5–41.4) 40.0% 33.2 (29.1–37.2) 44.6% 26.5 (23.2–29.8) 50.1% 26 (21–31.1) 48.2% 20.1 (17.5–22.8) 55.8% <.001 <.001 
Fluoroquinolones                 
 Hospital wide 33.7 (30–37.4) 40.7% 28.5 (25.1–31.9) 51.3% 27.7 (23–32.5) 52.3% 32.1 (29.7–34.4) 45.1% 24.8 (21.3–28.4) 46.1% 20.1 (14.9–25.3) 42.7% 19.3 (16.7–21.9) 55.9% <.001 <.001 
 ICUs 61 (48.3–73.6) 18.9% 48.8 (40.3–57.4) 23.9% 48.5 (40.2–56.8) 33.7% 59 (52.9–65.2) 28.0% 41.5 (29–54) 27.0% 34.8 (24.2–45.5) 30.9% 33 (25.9–40.1) 31.5% <.001 .009 
 Heme/Onc 91.7 (69.7–114) 30.3% 88 (60.3–116) 43.0% 47.7 (29.1–66.2) 40.1% 59.3 (39.6–79) 29.1% 41.4 (30.1–52.6) 36.3% 36.2 (22.8–49.7) 35.2% 28 (16.8–39.2) 43.2% <.001 .19 
 Med/Surg units 29.3 (24.5–34.2) 57.0% 26.1 (22.6–29.5) 64.3% 29 (22.2–35.8) 62.9% 30.7 (27.9–33.5) 58.5% 25.9 (22.4–29.4) 61.0% 20.4 (15.1–25.7) 51.6% 20.9 (15.7–26.1) 70.4% .015 .012 
Metronidazole                 
 Hospital wide 30.1 (24.4–35.7) 42.0% 40.8 (37.3–44.3) 40.6% 42.7 (38.2–47.2) 35.7% 41.3 (36.9–45.6) 38.7% 49.4 (45.4–53.4) 39.5% 45.2 (40.7–49.8) 37.3% 45.9 (41.1–50.7) 33.7% <.001 .007 
 ICUs 38.9 (25.9–51.9) 14.8% 38.1 (30.9–45.3) 20.7% 45.8 (37–54.6) 18.8% 45.4 (34.3–56.6) 20.1% 54.8 (42.6–66.9) 19.9% 53.1 (39.7–66.6) 19.5% 70.4 (52.2–88.6) 13.9% .012 .88 
 Heme/Onc 94.3 (51.3–137) 38.3% 114 (85.6–143) 48.8% 79.2 (61.7–96.6) 50.8% 65.9 (53.2–78.6) 54.8% 71.4 (57.6–85.2) 67.1% 68.2 (54.7–81.6) 61.0% 66.1 (56.8–75.5) 60.6% .17 .002 
 Med/Surg units 26 (21.7–30.2) 54.9% 41.9 (36.5–47.4) 42.7% 46.4 (40.2–52.5) 33.3% 44.9 (39.2–50.6) 33.7% 56 (49.8–62.1) 34.6% 49.6 (46.2–53) 34.3% 48.8 (42.9–54.7) 32.7% <.001 <.001 
LocationPre (2010–2011)Plan (2011–2013)Post-1 (2013–2014)Post-2 (2014–2015)Post-3 (2015–2016)Post-4 (2016–2017)Post-5 (2017–2018)DOTs P Preand Post 5Enteral Percentage P Pre- and Post-5
DOTsa (95% CI)Enteral PercentageDOTsa (95% CI)Enteral PercentageDOTsa (95% CI)Enteral PercentageDOTsa (95% CI)Enteral PercentageDOTsa (95% CI)Enteral PercentageDOTsa (95% CI)Enteral PercentageDOTsa (95% CI)Enteral Percentage
Clindamycin                 
 Hospital wide 44.3 (40.5–48.2) 23.0% 38.7 (35.4–41.9) 31.9% 30.3 (27.1–33.4) 36.3% 27.6 (24.3–30.8) 40.0% 22.7 (20–25.4) 38.9% 24.5 (20.1–28.9) 41.9% 20.2 (17.2–23.2) 43.2% <.001 <.001 
 ICUs 50.3 (40.1–60.4) 8.9% 52.4 (41.7–63.2) 20.3% 44.3 (30.5–58.1) 28.3% 39.9 (32.6–47.1) 26.5% 41.4 (32.6–50.3) 20.4% 44.2 (28.8–59.6) 30.9% 36.5 (22.4–50.7) 25.1% .096 .004 
 Heme/Onc 114 (80.6–147) 14.5% 78.8 (55.8–102) 15.1% 38.3 (24.7–51.9) 25.3% 31.9 (22.8–41.1) 36.6% 24.7 (18.8–30.5) 34.7% 34.1 (27.3–41) 34.4% 32.9 (26.5–39.2) 35.2% <.001 .004 
Med/Surg units 47.4 (41.4–53.4) 29.7% 43.4 (41–45.8) 37.8% 37 (32.5–41.4) 40.0% 33.2 (29.1–37.2) 44.6% 26.5 (23.2–29.8) 50.1% 26 (21–31.1) 48.2% 20.1 (17.5–22.8) 55.8% <.001 <.001 
Fluoroquinolones                 
 Hospital wide 33.7 (30–37.4) 40.7% 28.5 (25.1–31.9) 51.3% 27.7 (23–32.5) 52.3% 32.1 (29.7–34.4) 45.1% 24.8 (21.3–28.4) 46.1% 20.1 (14.9–25.3) 42.7% 19.3 (16.7–21.9) 55.9% <.001 <.001 
 ICUs 61 (48.3–73.6) 18.9% 48.8 (40.3–57.4) 23.9% 48.5 (40.2–56.8) 33.7% 59 (52.9–65.2) 28.0% 41.5 (29–54) 27.0% 34.8 (24.2–45.5) 30.9% 33 (25.9–40.1) 31.5% <.001 .009 
 Heme/Onc 91.7 (69.7–114) 30.3% 88 (60.3–116) 43.0% 47.7 (29.1–66.2) 40.1% 59.3 (39.6–79) 29.1% 41.4 (30.1–52.6) 36.3% 36.2 (22.8–49.7) 35.2% 28 (16.8–39.2) 43.2% <.001 .19 
 Med/Surg units 29.3 (24.5–34.2) 57.0% 26.1 (22.6–29.5) 64.3% 29 (22.2–35.8) 62.9% 30.7 (27.9–33.5) 58.5% 25.9 (22.4–29.4) 61.0% 20.4 (15.1–25.7) 51.6% 20.9 (15.7–26.1) 70.4% .015 .012 
Metronidazole                 
 Hospital wide 30.1 (24.4–35.7) 42.0% 40.8 (37.3–44.3) 40.6% 42.7 (38.2–47.2) 35.7% 41.3 (36.9–45.6) 38.7% 49.4 (45.4–53.4) 39.5% 45.2 (40.7–49.8) 37.3% 45.9 (41.1–50.7) 33.7% <.001 .007 
 ICUs 38.9 (25.9–51.9) 14.8% 38.1 (30.9–45.3) 20.7% 45.8 (37–54.6) 18.8% 45.4 (34.3–56.6) 20.1% 54.8 (42.6–66.9) 19.9% 53.1 (39.7–66.6) 19.5% 70.4 (52.2–88.6) 13.9% .012 .88 
 Heme/Onc 94.3 (51.3–137) 38.3% 114 (85.6–143) 48.8% 79.2 (61.7–96.6) 50.8% 65.9 (53.2–78.6) 54.8% 71.4 (57.6–85.2) 67.1% 68.2 (54.7–81.6) 61.0% 66.1 (56.8–75.5) 60.6% .17 .002 
 Med/Surg units 26 (21.7–30.2) 54.9% 41.9 (36.5–47.4) 42.7% 46.4 (40.2–52.5) 33.3% 44.9 (39.2–50.6) 33.7% 56 (49.8–62.1) 34.6% 49.6 (46.2–53) 34.3% 48.8 (42.9–54.7) 32.7% <.001 <.001 
a

DOTs are reported per 1000 PDs.

FIGURE 1

Antibacterial use in DOTs per 1000 PDs is stratified vertically by antibacterial (note that the range on vertical axes varies by location, given differing usage rates across units) and horizontally by unit subgrouping. Bars are subdivided into IV (gray) and enteral (white) route of administration, with the enteral percentage of total use displayed in the base of each bar. The time intervals are pre (2010–2011), plan (2011–2013), and post 1 through post 5 (1-year intervals from 2013 to 2018). P values are displayed comparing DOTs pre and DOTs post 5. CICU, cardiac intensive care unit; HSCT, hematopoietic stem cell transplant.

FIGURE 1

Antibacterial use in DOTs per 1000 PDs is stratified vertically by antibacterial (note that the range on vertical axes varies by location, given differing usage rates across units) and horizontally by unit subgrouping. Bars are subdivided into IV (gray) and enteral (white) route of administration, with the enteral percentage of total use displayed in the base of each bar. The time intervals are pre (2010–2011), plan (2011–2013), and post 1 through post 5 (1-year intervals from 2013 to 2018). P values are displayed comparing DOTs pre and DOTs post 5. CICU, cardiac intensive care unit; HSCT, hematopoietic stem cell transplant.

Close modal

For these data, ciprofloxacin and levofloxacin use were combined, given similar usage trends. Total fluoroquinolone use decreased from 33.7 (preimplementation) to 19.3 (year-5 postimplementation) DOTs per 1000 PDs (P < .001). The hospital-wide enteral percentage of total fluoroquinolone use significantly increased from 40.7% preimplementation to 55.9% year-5 postimplementation (P < .001). Given statistical correction for multiple comparisons, changes in the enteral percentages across unit subgroupings were not statistically significant (Table 1).

Hospital-wide metronidazole use increased from 30.1 (preimplementation) to 45.9 (year-5 postimplementation) DOTs per 1000 PDs (P < .001), whereas the enteral percentage of total use decreased from 42.0% to 33.7% (P =.007) over the same interval. Metronidazole was the only drug with discordant usage trends across unit subgroupings. Heme/Onc percentage of enteral percentage increased from preimplementation to year-5 postimplementation (from 38.3% to 60.6% enteral; P = .002), whereas no significant change was seen in the ICUs (14.8% vs 13.9%; P = .88). On Med/Surg units, conversely, the enteral percentage of total use decreased from 54.9% enteral (preimplementation) to 33.3% (year-1 postimplementation; P < .001), with no additional change from year-1 (33.3%) to year-5 (32.7% enteral) postimplementation (P = .85).

Handshake stewardship has been shown to reduce hospital-wide antibacterial use in a cost-saving and sustainable way, from 891 DOTs per 1000 PDs (95% CI: 859–923) to 665 (95% CI: 637–694; P < .01) over an 8-year period.7,8  In the setting of this decreased overall use, here, we describe the impact of multiple, bundled interventions on IV versus enteral use of 7 highly bioavailable antibacterials. Given that system-wide methodologies to identify patients eligible for enteral transition are difficult to standardize, we report hospital-wide and unit-specific data (including all inpatient PDs regardless of NPO status) to provide a simple, standardized benchmark for comparing IV with enteral use across hospitals and in the context of overall use.

Across the study period, both clindamycin and fluoroquinolones were used less overall but, when used, were used enterally a higher proportion of the time. (Although the total enteral DOTs reveal minimal numeric change over time, this represents the decreased total use being counterbalanced by the increased enteral percentage of total use.) These improvements were seen across the hospital, including in units (ICUs and Heme/Onc) traditionally considered difficult for stewardship programs to impact and likely to have high rates of NPO patients.

Metronidazole, conversely, saw a hospital-wide decrease in the enteral percentage of total use. This change was largely driven by increasing IV use on Med/Surg units, with different trends in Heme/Onc (increased enteral percentage) and the ICUs (no significant change in enteral percentage). Interestingly, this shift largely occurred over a 3-month period (February to May 2012), coinciding with the implementation of a new appendicitis clinical pathway, changing empirical therapy from ertapenem to ceftriaxone plus metronidazole. On the surgical floor from February to May 2012, the average monthly use for metronidazole increased from 15.5 to 59.9 DOTs per 1000 PDs and, for ertapenem, decreased from 52.0 to 3.8 DOTs per 1000 PDs. Since that rapid mid-2012 change, there has not been a significant change in the enteral percentage of metronidazole use. Although IV metronidazole use still represents an opportunity for institutional improvement, IV metronidazole is preferable to ertapenem for infections in which a broader-spectrum agent is not required.

Provider acceptance of IV-to-enteral transitions initiated by clinical pharmacists was not formally tracked. However, provider acceptance is anecdotally good, and our program has seen 89% provider acceptance rates of formal stewardship recommendations.9 

Our enteral percentage of total use compares favorably with other freestanding children’s hospitals. Smith et al10  summarize highly bioavailable antibacterial use using the Pediatric Health Information System database. Including only patients who received another oral medication on the same calendar day, they report the percentage of enteral DOTs for clindamycin (21.7%), fluoroquinolones (53.4%), and metronidazole (38.4%). Although direct comparison is difficult, given our inclusion of NPO patients, we report a higher enteral percentage use for clindamycin (43.2%) and the fluoroquinolones (55.9%), although a lower enteral percentage for metronidazole (33.7%), although we used all patients in our denominator.

This retrospective report has multiple limitations. In single-center studies, applicability to other settings is limited, and adaption of our ASP model to other settings warrants further study. Usage rate trends across a subset of antibacterials must be interpreted cautiously, given that these 7 drugs are not used in isolation; their use must be understood in the context of other antibacterials that might be used in their place. Although metronidazole’s use relative to ertapenem is one example of this at our institution, there may be others that we have not identified. Additionally, because of limitations in the data set, we do not stratify use by diagnosis (certain subsets of diagnoses could drive most IV use) or provider type (medical and surgical patients share hospital floors, so we could not differentiate use between these groups), assess balance measures, or analyze length-of-stay data for these populations. However, because discharged patients are no longer included in our data set (and patients transitioned to enteral are presumably more likely to be discharged), these data likely underestimate stewardship’s impact.

Finally, these usage changes were also seen after multiple, overlapping interventions (handshake rounds, clinical pathway implementation, provider and pharmacist education, and the IV-to-enteral transition standing EHR order), and we cannot definitively uncouple the relative impacts of each part of this bundle. However, anecdotally, the clinical pharmacists initiate most IV-to-enteral transitions. This is supported by a previous analysis of our stewardship program, in which only 2% (72 of 3078) of formal stewardship recommendations involved IV-to-enteral transition,9  suggesting that interventions outside the formal stewardship team (eg, clinical pharmacists’ recommendations) likely accounted for most of this change. However, given that our ASP team educated and encouraged both providers and pharmacists about the IV-to-enteral transition and continues to do so, the stewardship team was instrumental in these practice changes.

Overall, these data reveal that IV versus enteral use as well as total use of highly bioavailable antibacterials varies widely across units. However, ASPs can successfully increase the enteral percentage of total use of these drugs through system-wide, bundled interventions. In this report, we also provide benchmark hospital-wide and unit-specific data for use across pediatric hospitals.

We thank Kelly Pearce for her contribution in data collection and Jason Child, PharmD, for his work developing and implementing the IV-to-enteral transition standing order.

Dr Haynes designed the study, collected and analyzed data, and drafted the initial manuscript; Drs Parker and MacBrayne assisted with study design and reviewed and revised the manuscript; and all authors approved the final manuscript as submitted.

FUNDING: No external funding.

1
Shah
SS
,
Srivastava
R
,
Wu
S
, et al
;
Pediatric Research in Inpatient Settings Network
.
Intravenous versus oral antibiotics for postdischarge treatment of complicated pneumonia
.
Pediatrics
.
2016
;
138
(
6
):
e20161692
2
Jumani
K
,
Advani
S
,
Reich
NG
,
Gosey
L
,
Milstone
AM
.
Risk factors for peripherally inserted central venous catheter complications in children
.
JAMA Pediatr
.
2013
;
167
(
5
):
429
435
3
Barlam
TF
,
Cosgrove
SE
,
Abbo
LM
, et al
.
Implementing an antibiotic stewardship program: guidelines by the Infectious Diseases Society of America and the Society for Healthcare Epidemiology of America
.
Clin Infect Dis
.
2016
;
62
(
10
):
e51
e77
4
Centers for Disease Control and Prevention
. Core elements of hospital antibiotic stewardship programs. Available at: www.cdc.gov/getsmart/healthcare/implementation/core-elements.html. Accessed May 28, 2020
5
Girdwood
SCT
,
Sellas
MN
,
Courter
JD
, et al
.
Improving the transition of intravenous to enteral antibiotics in pediatric patients with pneumonia or skin and soft tissue infections
.
J Hosp Med
.
2020
;
15
(
1
):
10
15
6
Downen
J
,
Jaeger
C
.
Quality improvement of intravenous to oral medication conversion using Lean Six Sigma methodologies
.
BMJ Open Qual
.
2020
;
9
(
1
):
e000804
7
MacBrayne
CE
,
Williams
MC
,
Levek
C
, et al
.
Sustainability of handshake stewardship: extending a hand is effective years later
.
Clin Infect Dis
.
2020
;
70
(
11
):
2325
2332
8
Hurst
AL
,
Child
J
,
Pearce
K
,
Palmer
C
,
Todd
JK
,
Parker
SK
.
Handshake stewardship: a highly effective rounding-based antimicrobial optimization service
.
Pediatr Infect Dis J
.
2016
;
35
(
10
):
1104
1110
9
Hurst
AL
,
Child
J
,
Parker
SK
.
Intervention and acceptance rates support handshake-stewardship strategy
.
J Pediatric Infect Dis Soc
.
2019
;
8
(
2
):
162
165
10
Smith
MJ
,
Thurn
C
,
Shah
SS
, et al
.
Route of administration for antibiotics with high oral bioavailability
.
Infec Control Hosp Epidemiol.
2019
;
40
:
248
249

Competing Interests

POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.

FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.

Supplementary data