Duration of Antibiotic Therapy for General Medicine and General Surgery Patients Throughout Transitions of Care: An Antibiotic Stewardship Opportunity for Noninfectious Disease Pharmacists

Introduction

Reductions in antibiotic duration are at the cornerstone of decreasing antimicrobial resistance.^1,2 Limiting exposure to antibiotics decreases risk of developing resistance. Shorter durations of antibiotic therapy have shown noninferior outcomes compared with longer traditional durations in hospitalized patients with pyelonephritis (5-7 vs 10-14 days), community-acquired pneumonia (CAP, 5 vs 7 days), and hospital-acquired pneumonia (HAP, 7-8 vs 10-15 days). ^{3 -6} Similar outcomes have been demonstrated with shorter durations of antibiotics in a multitude of common infections in the inpatient setting, leading the Centers for Disease Control and Prevention (CDC) to require Antimicrobial Stewardship Programs (ASPs) to develop and implement initiatives to target durations of therapy.^6,7

Effective January 2017, requirements for ASP were defined for U.S. acute care hospitals; however, not all hospitals have infectious disease (ID)-trained pharmacists to lead all possible stewardship endeavors. In large academic medical centers, incorporation of non-ID-trained pharmacists to assist in stewardship is necessary due to the large patient populations receiving antibiotics. ⁸ To date, there are limited data demonstrating antimicrobial stewardship opportunities at transitions of care for non-ID-trained pharmacists. Inappropriate use of antibiotics at hospital discharge has been documented; however, data assessing total inpatient and outpatient durations of antibiotic therapy are lacking.^9,10 Electronic discharge prescriptions have been shown to be a reliable source to capture intended postdischarge antibiotic durations and thus may be a target for prospective interventions to optimize antibiotic therapy at this transition of care.

The purpose of this study is to assess the total duration of antibiotics at the transition of care from inpatient to an outpatient setting in the management of urinary tract infections (UTIs), complicated urinary tract infection (cUTI), CAP, and HAP. We hypothesize longer durations of therapy are prescribed at this transition of care compared with current clinical guideline and primary literature recommendations. Our secondary outcome is to identify opportunities for stewardship interventions by non-ID pharmacists at the inpatient to discharge transition of care.

Patients and Methods

This was a single-center, retrospective cohort analysis evaluating durations of therapy of inpatient and outpatient antibiotic therapy in adult patients admitted to general surgery and general medicine services at a large academic medical center between January 1, 2017 and September 20, 2017. The institution is an academic medical center with more than 1500 beds including both cancer and transplant programs. The medical center has a well-established ASP. Non-ID pharmacists provide clinical services to the general medicine and general surgical patients and all pharmacists use the ASP and national guidelines to make recommendations. This study was approved by the University institutional review board (Study ID 2019E0184).

Patients were identified from the Information Warehouse using the following criteria: age ≥18 years and ≤89 years, who were discharged from a general medicine or general surgery service with an electronic prescription for an oral or intravenous antimicrobial. Only patients discharged into the community and nursing or rehabilitation facilities on an antibiotic and with International Classification of Diseases, Tenth Revision (ICD-10) codes for one the following diagnoses were included: uncomplicated UTI, cUTI, CAP, and HAP. Those discharged with antiviral, antiretroviral, antifungal, or antimalarial agents were excluded, even if they were combined with antibiotics for one of the specified indications. Prisoners and pregnant women were also excluded. Manual electronic medical record (EMR) review then excluded patients with 2 concurrent infections or with positive cultures for organisms with guideline evidence or existence of controversial data supporting longer durations of antibiotics (ie, methicillin-resistant Staphylococcus aureus pneumonia). Patients on prophylactic antibiotic regimens for immunocompromised states were also excluded. Patients were also excluded upon manual review for the following reasons: culture, radiographic, or documentation from the physician of an infection that was different than what was included in the study (ie, potential evidence of ICD-10 miscoding) or those patients who were not discharged on antibiotics per the discharge summary (ie, evidence of a prescription being ordered in the EMR and then canceled).

Baseline demographic data collected included age, sex, infection diagnosis, culture data, antibiotic(s), and inpatient and outpatient prescribed durations of therapy. Total duration of antibiotic therapy was calculated by adding the inpatient and outpatient durations. Appropriate duration of therapy was defined by institutional ASP guidelines which are based on national guidelines and current literature. Durations were 3 days for uncomplicated UTI, 5 days for CAP, and 7 days for cUTI or HAP.

Descriptive statistics were used with data presented as percent for nominal data and median (25%-75% interquartile range [IQR]) for continuous data. All statistical analysis was performed using Statistical Package for Social Science 24.0 (SPSS INC, Chicago, IL.)

Results

A total of 101 patients were included for analysis (Figure 1). The patients had a median age of 65 (55–79) years and 49% were female. Overall, 42.5% had CAP, 31.7% had HAP, 14.9% had UTI, and 10.9% had cUTI. Cultures were obtained in 40.5% of patients; 21.7% were positive.

OPEN IN VIEWER

Figure 1.

Inclusion and exclusion criteria.

Overall, most of the patients (81%) had antibiotics for longer than recommended with only 3% receiving less than the recommended duration. The overall median of duration beyond the current guideline recommendation was 4 (2–7) days, but varied by infection (Table 1). The median duration was more than 3 times as long as recommended for UTI; the median duration was 5 days longer than recommended for cUTI. For CAP and HAP, the median durations were 2 and 3 days longer than recommended, respectively. The median duration for UTI and cUTI was longer than for CAP or HAP (6.5 [4–11] vs 3 [2–5.5] days, P < .001). Overall, the recommended duration of antibiotics was completed while inpatient in 34.7%, but varied by infection. More patients with a UTI completed therapy while inpatient vs for either CAP or HAP (53.8% vs 28%, P = .03).

OPEN IN VIEWER

Table 1.

Treatment duration.

Infection	Guideline compliant treatment, d	Positive culture, %	Median inpatient duration, d	Completed duration inpatient, %	Median outpatient duration, d	Median total duration, d	Median duration of treatment in excess of recommended duration (patient days)
UTI n = 15	3	26.7	5 [3–7]	80	5 [2.5–7]	10 [7–14]	7.5 [4–11.5]
cUTI n = 11	7	54.5	4 [3–5.5]	18.2	7 [4.5–8]	12 [7.5–12.5]	5 [4–6.5]
CAP n = 43	5	16.3	3 [2–4.5]	25.6	5 [3–5]	7 [7–9]	3 [2–4]
HAP n = 32	7	15.6	5.5 [3.5–7.5]	31.2	5 [2.5–7]	10 [8–12]	3 [2–7]

UTI = uncomplicated urinary tract infections; cUTI = complicated urinary tract infections; CAP = community-acquired pneumonia; HAP = hospital-acquired pneumonia.

The median antibiotic duration was shorter in patients for whom there were no cultures obtained or when there were negative cultures compared with those patients with positive cultures for all indications: UTI 10.3 vs 10.8 days, cUTI 9 vs 12 days, CAP 8 vs 9.1 days, and HAP 10.5 vs 19.8 days.

Discussion

We identified that most (81%) of the general medicine and general surgery patients received total durations of antibiotics for longer than what is currently recommended by national guidelines. The number of days of excessive treatment of any UTI was significantly longer than for CAP and HAP. Likewise, those with a UTI or a cUTI were most likely to complete therapy as an inpatient and not require outpatient therapy; however, it was prescribed in many instances.

Transitions of care represent opportunities to re-review all discharge medications and, specifically, the desired total duration of antibiotics. Postdischarge antibiotic prescriptions can be limited to only what is necessary to align with national evidence-based recommendations (ie, prescribe the exact number of tablets to complete the total course). In our study, more than three-quarters of these study patients were given electronic prescriptions and received antibiotic courses that exceeded 7 days. Similarly, a recent evaluation of total duration of inpatient and outpatient antimicrobial therapy indicated that when accounting for discharge durations, 55% of patients received antibiotics beyond guideline recommendations (median [IQR], pneumonia: 10 days [8–15], UTI: 11 days [8–15], skin and soft tissue: 13 days [10–17], intra-abdominal infection: 13 days [10–18]). ⁹ These authors showed that, on average, patients received 3.8 days of unnecessary antibiotics, thereby posing risks for adverse effects and added costs. Our study also identified excess numbers of antibiotic days. Each unnecessary dose of antibiotic contributes to development of resistance, potential for adverse events, and changes in the microbiome.

It is well proven that ASPs are successful in reducing the incidence of antibiotic-resistant infections and the associated complications from antimicrobial regimens. ¹¹ While ASPs operate and guide use within an institution, it is imperative that non-ID pharmacists become extensions of the ASP to support appropriate use and total durations of therapy for antibiotics. Several “low-hanging fruit” ASP initiatives have been carried out by non-ID pharmacists. ¹² Non-ID pharmacists in South Africa were able to implement antibiotic stewardship for CAP across 47 hospitals and decreased overall antibiotic use by 18%. ¹³ In addition, Wenzler et al ⁸ showed that non-ID pharmacists were able to improve compliance with bundled recommendations for the successful management of Staphylococcus aureus bacteremia and decrease mortality 6-fold by leveraging the EMR to identify infected patients in real-real time. Total antibiotic duration at transitions of care presents yet another opportunity for all pharmacists to intervene. Based on our results, using the medical record to identify patients with specific infections may pose an opportunity for non-ID pharmacists to review antibiotic stop dates and automatic electronic stop dates may also encourage better adherence to guideline recommendations for durations of therapy. This is especially important given most general medicine and general surgery services in our institution do not have a clinical pharmacist on patient care rounds. Re-education of prescribers is also a necessary component for gaining buy-in for shorter durations and prevention of unnecessary adverse events. Non-ID pharmacists are in an opportune position to provide this valuable education and feedback. By leveraging previous success of completion of evidence-based hospital-specific local guidelines and available EMR technology combined with non-ID pharmacists, it is our hope that improvements in antimicrobial stewardship at these transitions of care can be achieved.

Limitations of this study include it being a single-center, retrospective review for these selected diagnoses among only general medicine and general surgical patients during this time frame. Diagnosis in most patients’ included in the study was based on clinical diagnosis vs microbiological evidence. Wide variances in treatment duration of UTI and cUTI may have been attributed to a lack of national guidelines and conflicting primary literature supporting short vs longer durations. These data may not be generalizable; however, other studies have also identified total durations of antibiotic therapy as being excessive.⁶

Conclusions

Excessive total duration of inpatient and outpatient antibiotic therapy for multiple infections was identified. Transition from inpatient hospital care to discharge poses an opportunity for non-ID pharmacists to participate and achieve improved antibiotic stewardship, by ensuring proper duration of antibiotic therapy.