Background: Approximately 40% of all antibiotics related to an acute-care hospital stay are prescribed at the time of hospital discharge. However, there is no metric to compare hospitals on their antibiotic-prescribing at this transition of care. In this study, we sought to build a risk-adjusted metric for comparing hospitals on their overall post-discharge antibiotic use. Methods: This was a retrospective study across all acute-care admissions within the Veterans Health Administration during 2018-2021. For patients discharged to home or self-care, data on antibiotics administered while inpatient and those prescribed at discharge were collected. To predict post-discharge antibiotic use (days of therapy, DOT), we built a zero-inflated negative binomial model with a random intercept for each VA medical center. Data were split into training and testing sets to measure model performance and absolute error. Covariates included patient demographics, medical specialty at discharge, comorbidities, discharge diagnoses of infection, and the length of inpatient antibiotic therapy. Outliers, defined as DOT ≥ 30, were excluded, and the predicted random intercept was used to determine hospital performance. To compare hospitals with a positive versus negative random intercept in our model (i.e. higher vs. lower than expected overall post-discharge use, respectively), we calculated mean total antibiotic duration (inpatient + post-discharge) for two uncomplicated infection types: community-acquired pneumonia (CAP) and skin and soft tissue infections (SSTI). Results: 1,804,400 patients were discharged to home or self-care across 130 hospitals. The mean age was 67.8 (SD 12.9), and 93.7% were male. Antibiotics were prescribed to 41.5% while hospitalized and 19.5% at discharge. The median number of post-discharge DOT among those prescribed post-discharge antibiotics was 7 (IQR 4-12). The predictive model detected post-discharge antibiotic use with fidelity, including accurate identification of any post-discharge antibiotic exposure (area under the precision-recall curve=0.97 ) and reliable prediction of the number of post-discharge DOT in those who were exposed (mean absolute error = 1.65; Figure 1). At negative versus positive random intercept hospitals (Figure 2), antibiotic duration for CAP and SSTIs was 7.3 versus 8.1 days (p < 0 .001) and 9.4 vs. 10.2 days (p < 0 .001), respectively. Conclusion: A model using electronically available data was able to accurately predict antibiotic use prescribed at hospital discharge. Hospitals with lower than expected overall post-discharge antibiotic use also prescribed shorter courses of antibiotic therapy for uncomplicated cases of CAP and SSTI, which may reflect more robust processes at these sites to reduce antibiotic overuse at discharge.

Disclosure: Michi Goto: Contracted Research – Merck

Background: Many urologists continue antibiotics after common urologic procedure beyond the timeframes recommended by professional guidelines. In this study, we sought to evaluate the association between postprocedural antibiotic use and patient outcomes. Methods: We identified all patients who underwent 1 of 3 urologic procedures (transurethral resection of bladder tumor [TURBT], transurethral resection of prostate [TURP], and ureteroscopy) within the Veterans’ Health Administration (VHA) between January 1, 2017, and June 30, 2021. A postprocedural antibiotic was any antibiotic potentially used for a urinary tract–related indication that was prescribed for administration after the day of the procedure. Outcomes were captured within 30 days of the procedure and included (1) return visits, defined as any emergency department or urgent care encounter or hospital readmission, and (2) Clostridium difficile infection (CDI), defined as a positive test for C. difficile and the prescription of an anti-CDI antibiotic. We used log-binomial models with risk adjustment to determine the association between postprocedural antibiotic use and outcomes. We constructed hospital-level observed-to-expected ratios for postprocedural antibiotic use, and we used these models to calculate the probability of each patient receiving postprocedural antibiotics. Results: Overall, we identified 74,629 patients; 98% were male; the mean age was 70 years (SD, 10). Among them, 50% underwent TURBT, 28% underwent TURP, and 23% underwent ureteroscopy. A postprocedural antibiotic was prescribed to 25,738 (35%) cases for a median duration of 3 days (IQR, 3–6). Return visits occurred in 13,489 patients (18%), and CDI occurred in 104 patients (0.1%). Patients exposed to postprocedural antibiotics had 16% more return visits (RR, 1.16; 95% CI, 1.13–1.20) and more than twice as much CDI (RR, 2.22; 95% CI, 1.51–3.26) than patients not exposed to postprocedural antibiotics. In log-binomial risk-adjusted analysis, the risk of return visits did not differ between the 2 groups (RR, 1.00; 95% CI, 0.97–1.04) but the risk of CDI was higher in patients who received post-procedural antibiotics (RR, 1.87; 95% CI, 1.00–3.51). Hospitals (n = 105) varied widely in their observed-to-expected ratios for prescribing postprocedural antibiotics, and the frequency of return visits was similar regardless of the frequency at which postprocedural antibiotics were prescribed (Table 1). Conclusions: Postprocedural antibiotics were prescribed beyond recommended intervals after more than one-third of common urologic procedures, with a large degree of variability across hospitals. The use of postprocedural antibiotics was not associated with fewer return visits but was associated with a nonsignificant increase in CDI risk. Efforts to reduce postprocedural antibiotics are needed.

Funding: Yes

Disclosures: This work was funded, in part, by the Merck Investigator Studies Program. This work was also supported by a Career Development Award (DJL) from the VA Health Services Research and Development Service (CDA 16-204) and by the Iowa City VA Health Care System, Department of Pharmacy Services.

We evaluated antibiotic-prescribing across 111 mental health units in the Veterans’ Health Administration. We found that accurate diagnosis of urinary tract infections is a major area for improvement. Because non–mental-health clinicians were involved in most antibiotic-prescribing decisions, stewardship interventions for mental health patients should have a broad target audience to be effective.

Background: Early postoperative and acute prosthetic joint infection (PJI) may be managed with debridement, antibiotics, and implant retention (DAIR). Among patients with nonstaphylococcal PJI, an initial 4–6-week course of intravenous or highly bioavailable oral antibiotics is recommended in the Infectious Diseases Society of America (IDSA) guidelines, with disagreement among committee members on the need for subsequent chronic oral antimicrobial suppression (CAS). We aimed to characterize patients with nonstaphylococcal PJI who received CAS and to compare them to those who did not receive CAS. Methods: This retrospective cohort study included patients admitted to Veterans’ Affairs (VA) hospitals from 2003 to 2017 who had a PJI caused by nonstaphylococcal bacteria, underwent DAIR, and received 4–6 weeks of antimicrobial treatment. PJI was defined by Musculoskeletal Infection Society (MSIS) 2011 criteria. CAS was defined as at least 6 months of oral antibiotics following initial treatment of the PJI. Patients were followed for 5 years after debridement. We used χ2 tests and t tests were used to compare patients who received CAS with those who did not receive CAS. Results: Overall, 561 patients had a nonstaphylococcal PJI treated with DAIR, and 80.6% of patients received CAS. The most common organisms causing PJI were streptococci. We detected no significant differences between patients who received CAS and those who did not receive CAS, except that modified Acute Physiology and Chronic Health Evaluation (mAPACHE) scores were higher among patients who did not receive CAS (Table 1). Conclusion: Patients not on CAS were more severely ill (by mAPACHE) than those on CAS. Otherwise, the 2 groups were not different. This finding was contrary to our hypothesis that patients with multiple comorbidities or higher mAPACHE scores would be more likely to get CAS. A future analysis will be conducted to assess treatment failure in both groups. We hope to find a specific cohort who may benefit from CAS and hope to deimplement CAS in others who may not benefit from it.

Funding: No

Disclosures: None

Table 1.

Group Name: VHA Center for Antimicrobial Stewardship and Prevention of Antimicrobial Resistance (CASPAR) Background: Antimicrobial stewardship programs (ASPs) are advised to measure antimicrobial consumption as a metric for audit and feedback. However, most ASPs lack the tools necessary for appropriate risk adjustment and standardized data collection, which are critical for peer-program benchmarking. We created a system that automatically extracts antimicrobial use data and patient-level factors for risk-adjustment and a dashboard to present risk-adjusted benchmarking metrics for ASP within the Veterans’ Health Administration (VHA). Methods: We built a system to extract patient-level data for antimicrobial use, procedures, demographics, and comorbidities for acute inpatient and long-term care units at all VHA hospitals utilizing the VHA’s Corporate Data Warehouse (CDW). We built baseline negative binomial regression models to perform risk-adjustments based on patient- and unit-level factors using records dated between October 2016 and September 2018. These models were then leveraged both retrospectively and prospectively to calculate observed-to-expected ratios of antimicrobial use for each hospital and for specific units within each hospital. Data transformation and applications of risk-adjustment models were automatically performed within the CDW database server, followed by monthly scheduled data transfer from the CDW to the Microsoft Power BI server for interactive data visualization. Frontline antimicrobial stewards at 10 VHA hospitals participated in the project as pilot users. Results: Separate baseline risk-adjustment models to predict days of therapy (DOT) for all antibacterial agents were created for acute-care and long-term care units based on 15,941,972 patient days and 3,011,788 DOT between October 2016 and September 2018 at 134 VHA hospitals. Risk adjustment models include month, unit types (eg, intensive care unit [ICU] vs non-ICU for acute care), specialty, age, gender, comorbidities (50 and 30 factors for acute care and long-term care, respectively), and preceding procedures (45 and 24 procedures for acute care and long-term care, respectively). We created additional models for each antimicrobial category based on National Healthcare Safety Network definitions. For each hospital, risk-adjusted benchmarking metrics and a monthly ranking within the VHA system were visualized and presented to end users through the dashboard (an example screenshot in Figure 1). Conclusions: Developing an automated surveillance system for antimicrobial consumption and risk-adjustment benchmarking using an electronic medical record data warehouse is feasible and can potentially provide valuable tools for ASPs, especially at hospitals with no or limited local informatics expertise. Future efforts will evaluate the effectiveness of dashboards in these settings.

Funding: No

Disclosures: None

Figure 1.

Figure 2.

Figure 3.

On-site surveys of weed populations provide information on the relative occurrence and density of weeds that can be useful to growers in that region. Data generated by weed surveys can aid in the management of weed issues by monitoring the movement of problem weeds and forecasting areas susceptible to infestations. Currently, on-site surveys are often performed on a small scale, within single fields or counties. Questionnaire surveys are helpful for assessing relative abundance but do not always provide detailed information on weed distribution in time or space. A survey was conducted annually in Ohio from 2013 through 2017 in 49 counties with soybean [Glycine max (L.) Merr.] production to assess the late-season occurrence of horseweed [Conyza canadensis (L.) Cronquist]. The objectives of this research were to: (1) determine the frequency, level of infestation, and distribution of C. canadensis in soybean fields in the primary soybean-producing Ohio counties over 5 yr; and (2) identify significant spatial clusters or movement trends over time. Conyza canadensis was encountered in each county from 2013 through 2017. Spatial cores of interest, or counties identified as having significant levels of C. canadensis infestations or a lack thereof relative to surrounding counties, were identified in all years except 2017. The lowest frequency of C. canadensis encountered at all rating levels occurred in 2017, which coincided with second-highest frequency of infestations (highest density level) among years. There was no distinct distribution or pattern of C. canadensis movement within the state from year to year, but there was an increase in counties with infestations over time compared with the early years of the survey when many counties had few to no infestations. These results suggest that C. canadensis persists as a common and troublesome threat to Ohio soybean producers and that growers should continue making C. canadensis management a priority when developing weed control programs.

We evaluated the relationship between local MRSA prevalence rates and antibiotic use across 122 VHA hospitals in 2016. Higher hospital-level MRSA prevalence was associated with significantly higher rates of antibiotic use, even after adjusting for case mix and stewardship strategies. Benchmarking anti-MRSA antibiotic use may need to adjust for MRSA prevalence.

In this cohort of Escherichia coli and Klebsiella spp hospital-onset bacteremia, isolated fluoroquinolone resistance had a larger relative impact on mortality than other phenotypic resistance patterns. This finding may support stewardship efforts targeting unnecessary fluoroquinolone use and increased attention from infection prevention and control departments.

In this single-center study, the standardized antimicrobial administration ratio (SAAR) for total antimicrobial use decreased in response to a stewardship intervention. Antimicrobial prescribing and clinical outcomes were stable or improved during the period of lower SAARs. Our findings suggest that SAAR values of ~0.8 can be safely achieved.

The nucleation and growth of Al on 7 × 7 and $\sqrt 3 \times \sqrt 3$R30 Al reconstructed Si(111) that result in strain-free Al overgrown films grown with an atomically abrupt metamorphic interface are compared. The reconstructed surfaces and abrupt strain relaxations are verified using reflection high-energy electron diffraction. The topography of evolution is examined with atomic force microscopy. The growth of Al on both the surfaces exhibits 3D island growth, but the island evolution of growth is dramatically different. On the 7 × 7 surface, mounds formed are uniformly distributed across the substrate, and growth appears to proceed uniformly. Alternatively, on the $\sqrt 3 \times \sqrt 3$R30 surface, Al atoms exhibit a clear preference to form mounds near the step edges. During Al growth, mounds increase in size and number, expanding out from step edges until they cover the whole substrate. Consistent expression of a mounded nucleation and growth mode imparts a physical limitation to the achievable surface roughness that may impact the ultimate performance of layered devices such as Josephson junctions that are critical components of superconducting quantum circuits.

Our objective is to map dynamic provinces and investigate dynamic changes in Jakobshavn Isbræ, Greenland. We use an approach that combines structural glaciology and remote-sensing data analysis, facilitated by mathematical characterization of generalized spatial surface roughness that provides parameters related to ice dynamics, deformation and interaction of the ice with bed topography. The approach is applied to derive time series of elevation and roughness changes and to attribute changes during rapid retreat. Different dynamic types of fast- and slow-moving ice can be mapped from ICESat Geoscience Laser Altimeter System data (2003–09) and Airborne Topographic Mapper data, using spatial roughness characterization, validated with ASTER and bed-topographic data. Results of comparative analysis of elevation changes and roughness changes of Jakobshavn south ice stream indicate (1) surface lowering of 10–15 m a-1 between 2004 and 2009 and (2) no change in surface roughness and dynamic types. These findings are consistent with a front retreat as part of a fjord-glacier cycle or following warming of fjord water and with climatic warming, but not with an internal dynamic acceleration as a cause of the observed changes during rapid retreat. Relationships to changes in basal water pressure are discussed. All glaciodynamic changes appear to have initiated near the front and propagated up-glacier.

We investigated the frequency and determinants of guideline-discordant antibiotic prescribing in outpatients with respiratory infections or cystitis. Antibiotic prescribing was guideline discordant in 60% of patients. The most common reason for discordance was prescribing an antibiotic when not indicated. In a multivariate analysis, physicians in training had the highest likelihood of guideline-concordant antibiotic prescribing.

Infect Control Hosp Epidemiol 2017;38:724–728