Can evidence-based decision support tools transform antibiotic management? A systematic review and meta-analyses

Getting rapid diagnostic test data into the appropriate hands by leveraging pharmacy staff and a clinical surveillance platform: a case study from a US community hospital

OBJECTIVES

To outline the procedural implementation and optimization of rapid diagnostic test (RDT) results for bloodstream infections (BSIs) and to evaluate the combination of RDTs with real-time antimicrobial stewardship team (AST) support plus clinical surveillance platform (CSP) software on time to appropriate therapy in BSIs at a single health system.

METHODS

Blood culture reporting and communication were reported for four time periods: (i) a pre-BCID [BioFire® FilmArray® Blood Culture Identification (BCID) Panel] implementation period that consisted of literature review and blood culture notification procedure revision; (ii) a BCID implementation period that consisted of BCID implementation, real-time results notification via CSP, and creation of a treatment algorithm; (iii) a post-BCID implementation period; and (iv) a BCID2 implementation period. Time to appropriate therapy metrics was reported for the BCID2 time period.

RESULTS

The mean time from BCID2 result to administration of effective antibiotics was 1.2 h (range 0-7.9 h) and time to optimal therapy was 7.6 h (range 0-113.8 h) during the BCID2 Panel implementation period. When comparing time to optimal antibiotic administration among patients growing ceftriaxone-resistant Enterobacterales, the BCID2 Panel group (mean 2.8 h) was significantly faster than the post-BCID Panel group (17.7 h; P = 0.0041).

CONCLUSIONS

Challenges exist in communicating results to the appropriate personnel on the healthcare team who have the knowledge to act on these data and prescribe targeted therapy against the pathogen(s) identified. In this report, we outline the procedures for telephonic communication and CSP support that were implemented at our health system to distribute RDT data to individuals capable of assessing results, enabling timely optimization of antimicrobial therapy.

Complex Hospital-Based Electronic Prescribing-Based Intervention to Support Antimicrobial Stewardship: Qualitative Study

BACKGROUND

Antimicrobial resistance (AMR) represents a growing concern for public health.

OBJECTIVE

We sought to explore the challenges associated with development and implementation of a complex intervention designed to improve AMS in hospitals.

METHODS

We conducted a qualitative evaluation of a complex AMS intervention with educational, behavioral, and technological components in 5 wards of an English hospital. At 2 weeks and 7 weeks after initiating the intervention, we interviewed 25 users of the intervention, including senior and junior prescribers, a senior nurse, a pharmacist, and a microbiologist. Topics discussed included perceived impacts of different elements of the intervention and facilitators and barriers to effective use. Interviews were supplemented by 2 observations of ward rounds to gain insights into AMS practices. Data were audio-recorded, transcribed, and inductively and deductively analyzed thematically using NVivo12.

RESULTS

Tracing the adoption and impact of the various components of the intervention was difficult, as it had been introduced into a setting with competing pressures. These particularly affected behavioral and educational components (eg, training, awareness-building activities), which were often delivered ad hoc. We found that the participatory intervention design had addressed typical use cases but had not catered for edge cases that only became visible when the intervention was delivered in real-world settings (eg, variations in prescribing workflows across different specialties and conditions).

CONCLUSIONS

Effective user-focused design of complex interventions to promote AMS can support acceptance and use. However, not all requirements and potential barriers to use can be fully anticipated or tested in advance of full implementation in real-world settings.

Prediction of Antibiotic Resistance in Patients With a Urinary Tract Infection: Algorithm Development and Validation

BACKGROUND

The early prediction of antibiotic resistance in patients with a urinary tract infection (UTI) is important to guide appropriate antibiotic therapy selection.

OBJECTIVE

In this study, we aimed to predict antibiotic resistance in patients with a UTI. Additionally, we aimed to interpret the machine learning models we developed.

METHODS

The electronic medical records of patients who were admitted to Yongin Severance Hospital, South Korea were used. A total of 71 features extracted from patients' admission, diagnosis, prescription, and microbiology records were used for classification. UTI pathogens were classified as either sensitive or resistant to cephalosporin, piperacillin-tazobactam (TZP), carbapenem, trimethoprim-sulfamethoxazole (TMP-SMX), and fluoroquinolone. To analyze how each variable contributed to the machine learning model's predictions of antibiotic resistance, we used the Shapley Additive Explanations method. Finally, a prototype machine learning-based clinical decision support system was proposed to provide clinicians the resistance probabilities for each antibiotic.

RESULTS

The data set included 3535, 737, 708, 1582, and 1365 samples for cephalosporin, TZP, TMP-SMX, fluoroquinolone, and carbapenem resistance prediction models, respectively. The area under the receiver operating characteristic curve values of the random forest models were 0.777 (95% CI 0.775-0.779), 0.864 (95% CI 0.862-0.867), 0.877 (95% CI 0.874-0.880), 0.881 (95% CI 0.879-0.882), and 0.884 (95% CI 0.884-0.885) in the training set and 0.638 (95% CI 0.635-0.642), 0.630 (95% CI 0.626-0.634), 0.665 (95% CI 0.659-0.671), 0.670 (95% CI 0.666-0.673), and 0.721 (95% CI 0.718-0.724) in the test set for predicting resistance to cephalosporin, TZP, carbapenem, TMP-SMX, and fluoroquinolone, respectively. The number of previous visits, first culture after admission, chronic lower respiratory diseases, administration of drugs before infection, and exposure time to these drugs were found to be important variables for predicting antibiotic resistance.

CONCLUSIONS

The study results demonstrated the potential of machine learning to predict antibiotic resistance in patients with a UTI. Machine learning can assist clinicians in making decisions regarding the selection of appropriate antibiotic therapy in patients with a UTI.

Evaluating clinical decision support software (CDSS): challenges for robust evidence generation

OBJECTIVES

Computerized clinical decision support software (CDSS) are digital health technologies that have been traditionally categorized as medical devices. However, the evaluation frameworks for traditional medical devices are not well adapted to assess the value and safety of CDSS. In this study, we identified a range of challenges associated with CDSS evaluation as a medical device and investigated whether and how CDSS are evaluated in Australia.

METHODS

Using a qualitative approach, we interviewed 11 professionals involved in the implementation and evaluation of digital health technologies at national and regional levels. Data were thematically analyzed using both data-driven (inductive) and theory-based (deductive) approaches.

RESULTS

Our results suggest that current CDSS evaluations have an overly narrow perspective on the risks and benefits of CDSS due to an inability to capture the impact of the technology on the sociotechnical environment. By adopting a static view of the CDSS, these evaluation frameworks are unable to discern how rapidly evolving technologies and a dynamic clinical environment can impact CDSS performance. After software upgrades, CDSS can transition from providing information to specifying diagnoses and treatments. Therefore, it is not clear how CDSS can be monitored continuously when changes in the software can directly affect patient safety.

CONCLUSION

Our findings emphasize the importance of taking a living health technology assessment approach to the evaluation of digital health technologies that evolve rapidly. There is a role for observational (real-world) evidence to understand the impact of changes to the technology and the sociotechnical environment on CDSS performance.

Principles and Practice of Antimicrobial Stewardship Program Resource Allocation

Antimicrobial Stewardship Programs (ASP) improve individual patient outcomes and clinical care processes while reducing antimicrobial-associated adverse events, optimizing operational priorities, and providing institutional cost savings. ASP composition, resources required, and priority focuses are influenced by myriad factors. Despite robust evidence and broad national support, individual ASPs still face challenges in obtaining appropriate resources. Though understanding the current landscape of ASP resource allocation, factors influencing staffing needs, and strategies required to obtain desired resources is important, acceptance of recommended staffing levels and appropriate ASP resource allocation are much needed to facilitate ASP sustainability and growth across the complex and diverse health care continuum.

Physicians' Perspectives on AI in Clinical Decision Support Systems: Interview Study of the CURATE.AI Personalized Dose Optimization Platform

BACKGROUND

Physicians play a key role in integrating new clinical technology into care practices through user feedback and growth propositions to developers of the technology. As physicians are stakeholders involved through the technology iteration process, understanding their roles as users can provide nuanced insights into the workings of these technologies that are being explored. Therefore, understanding physicians' perceptions can be critical toward clinical validation, implementation, and downstream adoption. Given the increasing prevalence of clinical decision support systems (CDSSs), there remains a need to gain an in-depth understanding of physicians' perceptions and expectations toward their downstream implementation. This paper explores physicians' perceptions of integrating CURATE.AI, a novel artificial intelligence (AI)-based and clinical stage personalized dosing CDSSs, into clinical practice.

OBJECTIVE

This study aims to understand physicians' perspectives of integrating CURATE.AI for clinical work and to gather insights on considerations of the implementation of AI-based CDSS tools.

METHODS

A total of 12 participants completed semistructured interviews examining their knowledge, experience, attitudes, risks, and future course of the personalized combination therapy dosing platform, CURATE.AI. Interviews were audio recorded, transcribed verbatim, and coded manually. The data were thematically analyzed.

RESULTS

Overall, 3 broad themes and 9 subthemes were identified through thematic analysis. The themes covered considerations that physicians perceived as significant across various stages of new technology development, including trial, clinical implementation, and mass adoption.

CONCLUSIONS

The study laid out the various ways physicians interpreted an AI-based personalized dosing CDSS, CURATE.AI, for their clinical practice. The research pointed out that physicians' expectations during the different stages of technology exploration can be nuanced and layered with expectations of implementation that are relevant for technology developers and researchers.

Long-Term Impact of a Smartphone App on Prescriber Adherence to Antibiotic Guidelines for Adult Patients With Community-Acquired Pneumonia: Interrupted Time-Series Study

BACKGROUND

Mobile health platforms like smartphone apps that provide clinical guidelines are ubiquitous, yet their long-term impact on guideline adherence remains unclear. In 2016, an antibiotic guidelines app, called SCRIPT, was introduced in Auckland City Hospital, New Zealand, to provide local antibiotic guidelines to clinicians on their smartphones.

OBJECTIVE

We aimed to assess whether the provision of antibiotic guidelines in a smartphone app resulted in sustained changes in antibiotic guideline adherence by prescribers.

METHODS

We analyzed antibiotic guideline adherence rates during the first 24 hours of hospital admission in adults diagnosed with community-acquired pneumonia using an interrupted time-series study with 3 distinct periods post app implementation (ie, 3, 12, and 24 months).

RESULTS

Adherence increased from 23% (46/200) at baseline to 31% (73/237) at 3 months and 34% (69/200) at 12 months, reducing to 31% (62/200) at 24 months post app implementation (P=.07 vs baseline). However, increased adherence was sustained in patients with pulmonary consolidation on x-ray (9/63, 14% at baseline; 23/77, 30% after 3 months; 32/92, 35% after 12 month; and 32/102, 31% after 24 months; P=.04 vs baseline).

CONCLUSIONS

An antibiotic guidelines app increased overall adherence, but this was not sustained. In patients with pulmonary consolidation, the increased adherence was sustained.

Exploring Antimicrobial Stewardship Influential Interventions on Improving Antibiotic Utilization in Outpatient and Inpatient Settings: A Systematic Review and Meta-Analysis

Antimicrobial stewardship interventions are targeted efforts by healthcare organizations to optimize antimicrobial use in clinical practice. The study aimed to explore effective interventions in improving antimicrobial use in hospitals. Literature was systemically searched for interventional studies through PubMed, CINAHL, and Scopus databases that were published in the period between January 2010 to April 2022. A random-effects model was used to pool and evaluate data from eligible studies that reported antimicrobial stewardship (AMS) interventions in outpatient and inpatient settings. Pooled estimates presented as proportions and standardized mean differences. Forty-eight articles were included in this review: 32 in inpatient and 16 in outpatient settings. Seventeen interventions have been identified, and eight outcomes have been targeted. AMS interventions improved clinical, microbiological, and cost outcomes in most studies. When comparing non-intervention with intervention groups using meta-analysis, there was an insignificant reduction in length of stay (MD: -0.99; 95% CI: -2.38, 0.39) and a significant reduction in antibiotics' days of therapy (MD: -2.73; 95% CI: -3.92, -1.54). There were noticeable reductions in readmissions, mortality rates, and antibiotic prescriptions post antimicrobial stewardship multi-disciplinary team (AMS-MDT) interventions. Studies that involved a pharmacist as part of the AMS-MDT showed more significant improvement in measured outcomes than the studies that did not involve a pharmacist.

Digital interventions for antimicrobial prescribing and monitoring: a qualitative meta-synthesis of factors influencing user acceptance

OBJECTIVE

To understand and synthesize factors influencing user acceptance of digital interventions used for antimicrobial prescribing and monitoring in hospitals.

MATERIALS AND METHODS

A meta-synthesis was conducted to identify qualitative studies that explored user acceptance of digital interventions for antimicrobial prescribing and/or monitoring in hospitals. Databases were searched and qualitative data were extracted and systematically classified using the unified theory of acceptance and use of technology (UTAUT) model.

RESULTS

Fifteen qualitative studies met the inclusion criteria. Eleven papers used interviews and four used focus groups. Most digital interventions evaluated in studies were decision support for prescribing (n = 13). Majority of perceptions were classified in the UTAUT performance expectancy domain in perceived usefulness and relative advantage constructs. Key facilitators in this domain included systems being trusted and credible sources of information, improving performance of tasks and increasing efficiency. Reported barriers were that interventions were not considered useful for all settings or patient conditions. Facilitating conditions was the second largest domain, which highlights the importance of users having infrastructure to support system use. Digital interventions were viewed positively if they were compatible with values, needs, and experiences of users.

CONCLUSIONS

User perceptions that drive users to accept and utilize digital interventions for antimicrobial prescribing and monitoring were predominantly related to performance expectations and facilitating conditions. To ensure digital interventions for antimicrobial prescribing are accepted and used, we recommend organizations ensure systems are evaluated and benefits are conveyed to users, that utility meets expectations, and that appropriate infrastructure is in place to support use.

How to use new antibiotics in the therapy of ventilator-associated pneumonia

PURPOSE OF REVIEW

Ventilator-associated pneumonia (VAP) is a common nosocomial infection in critically ill patients requiring endotracheal intubation and mechanical ventilation. Recently, the emergence of multidrug-resistant Gram-negative bacteria, including carbapenem-resistant Enterobacterales, multidrug-resistant Pseudomonas aeruginosa and Acinetobacter species, has complicated the selection of appropriate antimicrobials and contributed to treatment failure. Although novel antimicrobials are crucial to treating VAP caused by these multidrug-resistant organisms, knowledge of how to optimize their efficacy while minimizing the development of resistance should be a requirement for their use.

RECENT FINDINGS

Several studies have assessed the efficacy of novel antimicrobials against multidrug-resistant organisms, but high-quality studies focusing on optimal dosing, infusion time and duration of therapy in patients with VAP are still lacking. Antimicrobial and diagnostic stewardship should be combined to optimize the use of these novel agents.

SUMMARY

Improvements in diagnostic tests, stewardship practices and a better understanding of dosing, infusion time, duration of treatment and the effects of combining various antimicrobials should help optimize the use of novel antimicrobials for VAP and maximize clinical outcomes while minimizing the development of resistance.

Clinical Decision Support Systems for Antibiotic Prescribing: An Inventory of Current French Language Tools

Clinical decision support systems (CDSSs) are increasingly being used by clinicians to support antibiotic decision making in infection management. However, coexisting CDSSs often target different types of physicians, infectious situations, and patient profiles. The objective of this study was to perform an up-to-date inventory of French language CDSSs currently used in community and hospital settings for antimicrobial prescribing and to describe their main characteristics. A literature search, a search among smartphone application stores, and an open discussion with antimicrobial stewardship (AMS) experts were conducted in order to identify available French language CDSSs. Any clinical decision support tool that provides a personalized recommendation based on a clinical situation and/or a patient was included. Eleven CDSSs were identified through the search strategy. Of the 11 CDSSs, only 2 had been the subject of published studies, while 9 CDSSs were identified through smartphone application stores and expert knowledge. The majority of CDSSs were available free of charge (n = 8/11, 73%). Most CDSSs were accessible via smartphone applications (n = 9/11, 82%) and online websites (n = 8/11, 73%). Recommendations for antibiotic prescribing in urinary tract infections, upper and lower respiratory tract infections, and digestive tract infections were provided by over 90% of the CDSSs. More than 90% of the CDSSs displayed recommendations for antibiotic selection, prioritization, dosage, duration, route of administration, and alternative antibiotics in case of allergy. Information about antibiotic side effects, prescription recommendations for specific patient profiles and adaptation to local epidemiology were often missing or incomplete. There is a significant but heterogeneous offer for antibiotic prescribing decision support in French language. Standardized evaluation of these systems is needed to assess their impact on antimicrobial prescribing and antimicrobial resistance.

OUP accepted manuscript

Background

Antimicrobial stewardship (AMS) programmes in hospitals support optimal antimicrobial use by utilizing strategies such as restriction policies and education. Several systematic reviews on digital interventions supporting AMS have been conducted but they have focused on specific interventions and outcomes.

Objectives

To provide a systematic overview and synthesis of evidence on the effectiveness of digital interventions to improve antimicrobial prescribing and monitoring in hospitals.

Methods

Multiple databases were searched from 2010 onwards. Review papers were eligible if they included studies that examined the effectiveness of AMS digital interventions in an inpatient hospital setting. Papers were excluded if they were not systematic reviews, were limited to a paediatric setting, or were not in English.

Results

Eight systematic reviews were included for data extraction. A large number of digital interventions were evaluated, with a strong focus on clinical decision support. Due to the heterogeneity of the interventions and outcome measures, a meta-analysis could not be performed. The majority of reviews reported that digital interventions reduced antimicrobial use and improved antimicrobial appropriateness. The impact of digital interventions on clinical outcomes was inconsistent.

Conclusions

Digital interventions reduce antimicrobial use and improve antimicrobial appropriateness in hospitals, but no firm conclusions can be drawn about the degree to which different types of digital interventions achieve these outcomes. Evaluation of sociotechnical aspects of digital intervention implementation is limited, despite the critical role that user acceptance, uptake and feasibility play in ensuring improvements in AMS are achieved with digital health.

An Electronic Health Record Integrated Decision Tool and Supportive Interventions to Improve Antibiotic Prescribing for Urinary Tract Infections in Nursing Homes: A Cluster Randomized Controlled Trial

OBJECTIVE

To investigate whether an electronic health record (EHR)-integrated decision tool, combined with supportive interventions, results in more appropriate antibiotic prescribing in nursing home (NH) residents with suspected urinary tract infection (UTI), without negative consequences for residents.

DESIGN

Cluster randomized controlled trial with NHs as the randomization unit; intervention group NHs received the EHR-integrated decision tool and supportive interventions, and control group NHs provided care as usual.

SETTING AND PARTICIPANTS

212 residents with suspected UTI, from 16 NHs in the Netherlands.

METHODS

Physicians collected data at index consultation (ie, UTI suspicion) and during a 21-day follow-up period (March 2019-March 2020). Overall antibiotic prescribing data at NH level, 12 months prior to and during the study, was derived from the electronic prescribing system. The primary study outcome was the percentage of antibiotic prescriptions for suspected UTI that was appropriate, at index consultation. Secondary study outcomes included changes in treatment decision, complications, UTI-related hospitalization, and mortality during follow-up; and pre-post study changes in antibiotic prescribing at the NH level.

RESULTS

295 suspected UTIs were included (intervention group: 189; control group: 106). The between-group difference in appropriate antibiotic prescribing was 13% [intervention group: 62%, control group: 49%; adjusted odds ratio (OR) 1.43, 95% CI 0.57-3.62]. In both groups, complications (2% vs 3%), UTI-related hospitalization (2% vs 1%), and possible UTI-related mortality (2% vs 2%) were rare. The pre-post study difference in antibiotic prescriptions per 1000 resident-care days was -0.95 in the intervention group NHs and -0.05 in the control group NHs (P = .02).

CONCLUSION AND IMPLICATIONS

Although appropriate antibiotic prescribing improved in the intervention group, this does not provide sufficient evidence for our multidisciplinary intervention. Despite this inconclusive result, our intervention could potentially still be effective, because we established a large reduction in the number of antibiotic prescriptions in the intervention group.

A Risk-Based Clinical Decision Support System for Patient-Specific Antimicrobial Therapy (iBiogram): Design and Retrospective Analysis

Background

There is a pressing need for digital tools that can leverage big data to help clinicians select effective antibiotic treatments in the absence of timely susceptibility data. Clinical presentation and local epidemiology can inform therapy selection to balance the risk of antimicrobial resistance and patient risk. However, data and clinical expertise must be appropriately integrated into clinical workflows.

Objective

The aim of this study is to leverage available data in electronic health records, to develop a data-driven, user-centered, clinical decision support system to navigate patient safety and population health.

Methods

We analyzed 5 years of susceptibility testing (1,078,510 isolates) and patient data (30,761 patients) across a large academic medical center. After curating the data according to the Clinical and Laboratory Standards Institute guidelines, we analyzed and visualized the impact of risk factors on clinical outcomes. On the basis of this data-driven understanding, we developed a probabilistic algorithm that maps these data to individual cases and implemented iBiogram, a prototype digital empiric antimicrobial clinical decision support system, which we evaluated against actual prescribing outcomes.

Results

We determined patient-specific factors across syndromes and contexts and identified relevant local patterns of antimicrobial resistance by clinical syndrome. Mortality and length of stay differed significantly depending on these factors and could be used to generate heuristic targets for an acceptable risk of underprescription. Combined with the developed remaining risk algorithm, these factors can be used to inform clinicians’ reasoning. A retrospective comparison of the iBiogram-suggested therapies versus the actual prescription by physicians showed similar performance for low-risk diseases such as urinary tract infections, whereas iBiogram recognized risk and recommended more appropriate coverage in high mortality conditions such as sepsis.

Conclusions

The application of such data-driven, patient-centered tools may guide empirical prescription for clinicians to balance morbidity and mortality with antimicrobial stewardship.

Hospital Pharmacists and Antimicrobial Stewardship: A Qualitative Analysis

Antimicrobial stewardship programmes (ASPs) in hospitals are predominantly led by specific ASP physicians and pharmacists. Limited studies have been conducted to appreciate non-ASP-trained hospital pharmacists' perspectives on their roles in antimicrobial stewardship. Focus group discussions (FGDs) were conducted with 74 pharmacists, purposively sampled from the 3 largest acute-care public hospitals in Singapore, to explore facilitators and barriers faced by them in antimicrobial stewardship. Applied thematic analysis was conducted and codes were categorised using the social-ecological model (SEM). At the intrapersonal level, pharmacists identified themselves as reviewers for drug safety before dispensing, confining to a restricted advisory role due to lack of clinical knowledge, experience, and empowerment to contribute actively to physicians' prescribing decisions. At the interpersonal level, pharmacists expressed difficulties conveying their opinions and recommendations on antibiotic therapy to physicians despite frequent communications, but they assumed critical roles as educators for patients and their caregivers on proper antibiotic use. At the organisational level, in-house antibiotic guidelines supported pharmacists' antibiotic interventions and recommendations. At the community level, pharmacists were motivated to improve low public awareness and knowledge on antibiotic use and antimicrobial resistance. These findings provide important insights into the gaps to be addressed in order to harness the untapped potential of hospital pharmacists and fully engage them in antimicrobial stewardship.

Computerized decision support system (CDSS) use for surveillance of antimicrobial resistance in urinary tract infections in primary care

BACKGROUND

Hospital-based surveillance of antimicrobial resistance may be irrelevant as a guide to antimicrobial use for urinary tract infections (UTIs) in primary care.

OBJECTIVES

To highlight the value of online computerized decision support systems (CDSS) in providing information on the surveillance of antimicrobial resistance in community-acquired UTIs.

METHODS

We collected the susceptibility profile for key antibiotics by type of UTI involving Escherichia coli from 2017 to 2020, using queries for UTI (Q-UTI) submitted to a French CDSS. We compared these results with those from the MedQual French surveillance system for community-acquired UTI and the European Antimicrobial Resistance Surveillance Network (EARS-NET) for invasive infections.

RESULTS

We collected 43 591 Q-UTI, of which 10 192 (23%) involved E. coli: 40% cystitis, 32% male-UTI, and 27% pyelonephritis. Resistance was 41.3% (95% CI, 40.3%-42.2%) for amoxicillin, 16.6% (95% CI, 15.9%-17.3%) for fluoroquinolones, 6.6% (95% CI, 6.1%-7.0%) for third-generation cephalosporins (3GC), and 5.7% (95% CI, 5.2%-6.1%) for aminoglycosides. Resistance to amoxicillin was lower than that reported in MedQual (42.7%, P value = 0.004), and in EARS-NET (55.2%, P value < 0.001). For fluoroquinolones, resistance was higher than in MedQual (12.0%, P value < 0.001) and EARS-NET (15.8%, P value = 0.041). In complicated pyelonephritis and male UTI, fluoroquinolone resistance peaked at ∼20%. For 3GC, all UTI had higher resistance than in MedQual (3.5%, P value < 0.001), but lower than in EARS-NET (9.5%, P value < 0.001). Aminoglycoside resistance was not reported by MedQual, and was lower than in EARS-NET (7.1%, P value < 0.001).

CONCLUSIONS

CDSS can inform prescribers in real-time about the ecology and surveillance of E. coli resistance in community-acquired UTI. In complicated upper UTIs, they can underline the risk of empirical use of fluoroquinolones and suggest preferential use of 3GC.

A computerized decision support system (CDSS) for antibiotic prescription in primary care-Antibioclic: implementation, adoption and sustainable use in the era of extended antimicrobial resistance

OBJECTIVES

To describe the implementation and use of a computerized decision support system (CDSS) for antibiotic prescription in primary care in France (Antibioclic). The CDSS targets 37 infectious diseases and has been freely available on a website since 2011.

METHODS

Description and implementation of the architecture of a CDSS for antibiotic prescription in general practice. Analysis of the queries made between 2012 and 2018 on the CDSS by GPs. Analysis of two cross-sectional studies of users in 2014 and 2019.

RESULTS

The number of queries increased from a median of 796/day [IQR, 578-989] in 2012 to 11 125/day [5592-12 505] in 2018. Unique users increased from 414/day [245-494] in 2012 to 5365/day [2891-5769] in 2018. Time taken to make a query was 2 min [1.9-2.1]. Among 3 542 347 queries in 2018, 78% were for adults. Six situations accounted for ≥50% of queries: cystitis; acute otitis media; acute sinusitis; community-acquired pneumonia; sore throat; and pyelonephritis. Queries concerned pathologies for which antibiotic prescription was necessary (64%), was conditional on additional clinical steps (34%) or was not recommended (2%). Most users (81%) were GPs, with median age of 38 years [31-52] and 58% were female. Among the 4016 GPs who responded to the surveys, the vast majority (96%) reported using the CDSS during the consultation, with 24% systematically using Antibioclic to initiate an antibiotic course and 93% having followed the CDSS recommendation for the latest prescription. Most GPs were comfortable using the CDSS in front of a patient.

CONCLUSIONS

Antibioclic has been adopted and is widely used in primary care in France. Its interoperability could allow its adaptation and implementation in other countries.

Factors That Impact the Adoption of Clinical Decision Support Systems (CDSS) for Antibiotic Management

The study evaluated individual and setting-specific factors that moderate clinicians’ perception regarding use of clinical decision support systems (CDSS) for antibiotic management. A cross-sectional online survey examined clinicians’ perceptions about CDSS implementation for antibiotic management in Australia. Multivariable logistic regression determined the association between drivers of CDSS adoption and different moderators. Clinical experience, CDSS use and care setting were important predictors of clinicians’ perception concerning CDSS adoption. Compared to nonusers, CDSS users were less likely to lack confidence in CDSS (OR = 0.63, 95%, CI = 0.32, 0.94) and consider it a threat to professional autonomy (OR = 0.47, 95%, CI = 0.08, 0.83). Conversely, there was higher likelihood in experienced clinicians (>20 years) to distrust CDSS (OR = 1.58, 95%, CI = 1.08, 2.23) due to fear of comprising their clinical judgement (OR = 1.68, 95%, CI = 1.27, 2.85). In primary care, clinicians were more likely to perceive time constraints (OR = 1.96, 95%, CI = 1.04, 3.70) and patient preference (OR = 1.84, 95%, CI = 1.19, 2.78) as barriers to CDSS adoption for antibiotic prescribing. Our findings provide differentiated understanding of the CDSS implementation landscape by identifying different individual, organisational and system-level factors that influence system adoption. The individual and setting characteristics can help understand the variability in CDSS adoption for antibiotic management in different clinicians.

Impact of a machine learning based decision support for Urinary Tract Infections: Prospective observational study in 36 primary care practices (Preprint)

Background

There is increasing attention on machine learning (ML)-based clinical decision support systems (CDSS), but their added value and pitfalls are very rarely evaluated in clinical practice. We implemented a CDSS to aid general practitioners (GPs) in treating patients with urinary tract infections (UTIs), which are a significant health burden worldwide.

Objective

This study aims to prospectively assess the impact of this CDSS on treatment success and change in antibiotic prescription behavior of the physician. In doing so, we hope to identify drivers and obstacles that positively impact the quality of health care practice with ML.

Methods

The CDSS was developed by Pacmed, Nivel, and Leiden University Medical Center (LUMC). The CDSS presents the expected outcomes of treatments, using interpretable decision trees as ML classifiers. Treatment success was defined as a subsequent period of 28 days during which no new antibiotic treatment for UTI was needed. In this prospective observational study, 36 primary care practices used the software for 4 months. Furthermore, 29 control practices were identified using propensity score-matching. All analyses were performed using electronic health records from the Nivel Primary Care Database. Patients for whom the software was used were identified in the Nivel database by sequential matching using CDSS use data. We compared the proportion of successful treatments before and during the study within the treatment arm. The same analysis was performed for the control practices and the patient subgroup the software was definitely used for. All analyses, including that of physicians’ prescription behavior, were statistically tested using 2-sided z tests with an α level of .05.

Results

In the treatment practices, 4998 observations were included before and 3422 observations (of 2423 unique patients) were included during the implementation period. In the control practices, 5044 observations were included before and 3360 observations were included during the implementation period. The proportion of successful treatments increased significantly from 75% to 80% in treatment practices (z=5.47, P<.001). No significant difference was detected in control practices (76% before and 76% during the pilot, z=0.02; P=.98). Of the 2423 patients, we identified 734 (30.29%) in the CDSS use database in the Nivel database. For these patients, the proportion of successful treatments during the study was 83%—a statistically significant difference, with 75% of successful treatments before the study in the treatment practices (z=4.95; P<.001).

Conclusions

The introduction of the CDSS as an intervention in the 36 treatment practices was associated with a statistically significant improvement in treatment success. We excluded temporal effects and validated the results with the subgroup analysis in patients for whom we were certain that the software was used. This study shows important strengths and points of attention for the development and implementation of an ML-based CDSS in clinical practice.

Trial Registration

ClinicalTrials.gov NCT04408976; https://clinicaltrials.gov/ct2/show/NCT04408976

Best practice: antibiotic decision-making in ICUs

PURPOSE OF REVIEW

A major challenge in the ICU is optimization of antibiotic use. This review assesses current understanding of core best practices supporting and promoting astute antibiotic decision-making.

RECENT FINDINGS

Limiting exposure to the shortest effective duration is the cornerstone of antibiotic decision-making. The decision to initiate antibiotics should include assessment of risk for resistance. This requires synthesis of patient-level data and environmental factors to determine whether delayed initiation could be considered in some patients with suspected sepsis until sensitivity data is available. Until improved stratification scores and clinically meaningful cut-off values to identify MDR are available and externally validated, decisions as to which empiric antibiotic is used should rely on syndromic antibiograms and institutional guidance. Optimization of initial and maintenance doses is another enabler of enhanced outcome. Stewardship practices must be streamlined by re-assessment to minimize negative effects, such as a potential increase in duration of therapy and increased risk of collateral damage from exposure to multiple, sequential antibiotics that may ensue from de-escalation.

SUMMARY

Multiple challenges and research priorities for antibiotic optimization remain; however, the best stewardship practices should be identified and entrenched in daily practice. Reducing unnecessary exposure remains a vital strategy to limit resistance development.

Use of stewardship smartphone applications by physicians and prescribing of antimicrobials in hospitals: A systematic review

BACKGROUND

Antimicrobial stewardship (AMS) programs promote appropriate use of antimicrobials and reduce antimicrobial resistance. Technological developments have resulted in smartphone applications (apps) facilitating AMS. Yet, their impact is unclear.

OBJECTIVES

Systematically review AMS apps and their impact on prescribing by physicians treating in-hospital patients.

DATA SOURCES

EMBASE, MEDLINE (Ovid), Cochrane Central, Web of Science and Google Scholar.

STUDY ELIGIBILITY CRITERIA

Studies focusing on smartphone or tablet apps and antimicrobial therapy published from January 2008 until February 28th 2019 were included.

PARTICIPANTS

Physicians treating in-hospital patients.

INTERVENTIONS

AMS apps.

METHODS

Systematic review.

RESULTS

Thirteen studies met the eligibility criteria. None was a randomized controlled trial. Methodological study quality was considered low to moderate in all but three qualitative studies. The primary outcomes were process indicators, adherence to guidelines and user experience. Guidelines were more frequently accessed by app (53.0% - 89.6%) than by desktop in three studies. Adherence to guidelines increased (6.5% - 74.0%) significantly for several indications after app implementation in four studies. Most users considered app use easy (77.4%->90.0%) and useful (71.0%->90%) in three studies and preferred it over guideline access by web viewer or booklet in two studies. However, some physicians regarded app use adjacent to colleagues or patients unprofessional in three qualitative studies. Susceptibility to several antimicrobials changed significantly post-intervention (from 5% decrease to 10% - 14% increase) in one study.

CONCLUSIONS

Use of AMS apps seems to promote access to and knowledge of antimicrobial prescribing policy, and increase adherence to guidelines in hospitals. However, this has been assessed in a limited number of studies and for specific indications. Good quality studies are necessary to properly assess the impact of AMS apps on antimicrobial prescribing. To improve adherence to antimicrobial guidelines, use of AMS apps could be considered.

Paving the Way for the Implementation of a Decision Support System for Antibiotic Prescribing in Primary Care in West Africa: Preimplementation and Co-Design Workshop With Physicians

Background

Suboptimal use of antibiotics is a driver of antimicrobial resistance (AMR). Clinical decision support systems (CDSS) can assist prescribers with rapid access to up-to-date information. In low- and middle-income countries (LMIC), the introduction of CDSS for antibiotic prescribing could have a measurable impact. However, interventions to implement them are challenging because of cultural and structural constraints, and their adoption and sustainability in routine clinical care are often limited. Preimplementation research is needed to ensure relevant adaptation and fit within the context of primary care in West Africa.

Objective

This study examined the requirements for a CDSS adapted to the context of primary care in West Africa, to analyze the barriers and facilitators of its implementation and adaptation, and to ensure co-designed solutions for its adaptation and sustainable use.

Methods

We organized a workshop in Burkina Faso in June 2019 with 47 health care professionals representing 9 West African countries and 6 medical specialties. The workshop began with a presentation of Antibioclic, a publicly funded CDSS for antibiotic prescribing in primary care that provides personalized antibiotic recommendations for 37 infectious diseases. Antibioclic is freely available on the web and as a smartphone app (iOS, Android). The presentation was followed by a roundtable discussion and completion of a questionnaire with open-ended questions by participants. Qualitative data were analyzed using thematic analysis.

Results

Most of the participants had access to a smartphone during their clinical consultations (35/47, 74%), but only 49% (23/47) had access to a computer and none used CDSS for antibiotic prescribing. The participants considered that CDSS could have a number of benefits including updating the knowledge of practitioners on antibiotic prescribing, improving clinical care and reducing AMR, encouraging the establishment of national guidelines, and developing surveillance capabilities in primary care. The most frequently mentioned contextual barrier to implementing a CDSS was the potential risk of increasing self-medication in West Africa, where antibiotics can be bought without a prescription. The need for the CDSS to be tailored to the local epidemiology of infectious diseases and AMR was highlighted along with the availability of diagnostic tests and antibiotics using national guidelines where available. Participants endorsed co-design involving all stakeholders, including nurses, midwives, and pharmacists, as central to any introduction of CDSS. A phased approach was suggested by initiating and evaluating CDSS at a pilot site, followed by dissemination using professional networks and social media. The lack of widespread internet access and computers could be circumvented by a mobile app with an offline mode.

Conclusions

Our study provides valuable information for the development and implementation of a CDSS for antibiotic prescribing among primary care prescribers in LMICs and may, in turn, contribute to improving antibiotic use, clinical outcomes and decreasing AMR.