Implementation Of Clinical Decision Support Research Articles

Qualitative study of challenges with recruitment of hospitals into a cluster controlled trial of clinical decision support in Australia

ObjectiveTo identify barriers to hospital participation in controlled cluster trials of clinical decision support (CDS) and potential strategies for addressing barriers.DesignQualitative descriptive design comprising semistructured interviews.SettingFive hospitals in New South...

Clinical decision support for gastrointestinal panel testing.

This study aimed to assess the impact of clinical decision support (CDS) to improve ordering of multiplex gastrointestinal polymerase chain reaction (PCR) testing panel ("GI panel"). Single-center, retrospective, before-after study. Tertiary care Veteran's Affairs (VA) Medical Center provides inpatient, outpatient, and residential care. All patients tested with a GI panel between June 22, 2022 and April 20, 2023. We designed a CDS questionnaire in the electronic medical record (EMR) to guide appropriate ordering of the GI panel. A "soft stop" reminder at the point of ordering prompted providers to confirm five appropriateness criteria: 1) documented diarrhea, 2) no recent receipt of laxatives, 3) C. difficile is not the leading suspected cause of diarrhea, 4) time period since a prior test is >14 days or prior positive test is >4 weeks and 5) duration of hospitalization <72 hours. The CDS was implemented in November 2022. Compared to the pre-implementation period (n = 136), fewer tests were performed post-implementation (n = 92) with an IRR of 0.61 (p = 0.003). Inappropriate ordering based on laxative use or undocumented diarrhea decreased (IRR 0.37, p = 0.012 and IRR 0.25, p = 0.08, respectively). However, overall inappropriate ordering and outcome measures did not significantly differ before and after the intervention. Implementation of CDS in the EMR decreased testing and inappropriate ordering based on use of laxatives or undocumented diarrhea. However, inappropriate ordering of tests overall remained high post-intervention, signaling the need for continued diagnostic stewardship efforts.

ICU Utilization After Implementation of Minor Severe Pneumonia Criteria in Real-Time Electronic Clinical Decision Support.

To determine if the implementation of automated clinical decision support (CDS) with embedded minor severe community-acquired pneumonia (sCAP) criteria was associated with improved ICU utilization among emergency department (ED) patients with pneumonia who did not require vasopressors or positive pressure ventilation at admission. Planned secondary analysis of a stepped-wedge, cluster-controlled CDS implementation trial. Sixteen hospitals in six geographic clusters from Intermountain Health; a large, integrated, nonprofit health system in Utah and Idaho. Adults admitted to the hospital from the ED with pneumonia identified by: 1) discharge International Classification of Diseases , 10th Revision codes for pneumonia or sepsis/respiratory failure and 2) ED chest imaging consistent with pneumonia, who did not require vasopressors or positive pressure ventilation at admission. After implementation, patients were exposed to automated, open-loop, comprehensive CDS that aided disposition decision (ward vs. ICU), based on objective severity scores (sCAP). The analysis included 2747 patients, 1814 before and 933 after implementation. The median age was 71, median Elixhauser index was 17, 48% were female, and 95% were Caucasian. A mixed-effects regression model with cluster as the random effect estimated that implementation of CDS utilizing sCAP increased 30-day ICU-free days by 1.04 days (95% CI, 0.48-1.59; p < 0.001). Among secondary outcomes, the odds of being admitted to the ward, transferring to the ICU within 72 hours, and receiving a critical therapy decreased by 57% (odds ratio [OR], 0.43; 95% CI, 0.26-0.68; p < 0.001) post-implementation; mortality within 72 hours of admission was unchanged (OR, 1.08; 95% CI, 0.56-2.01; p = 0.82) while 30-day all-cause mortality was lower post-implementation (OR, 0.71; 95% CI, 0.52-0.96; p = 0.03). Implementation of electronic CDS using minor sCAP criteria to guide disposition of patients with pneumonia from the ED was associated with safe reduction in ICU utilization.

Clinical Decision Support Implementation in Health Information Systems Across Radiology Departments

Introduction: Understanding the economic landscape is pivotal for gauging the feasibility and sustainability of Clinical Decision Support (CDS) integration across radiology, nursing, and laboratory departments. This review will delve into the economic considerations associated with CDS implementation, shedding light on clinical impact and cost-effectiveness. Methods: The systematic review employed a robust methodology, combining controlled vocabulary and free-text keywords in a comprehensive search across multiple databases. The inclusion criteria encompassed original research articles, systematic reviews, and meta-analyses in English, focusing on Clinical Decision Support (CDS) implementation in radiology, nursing, and laboratory departments within Health Information Systems. The two-step screening process, detailed data extraction, and methodological quality assessment were conducted with rigor by two reviewers, resolving discrepancies through discussion or consultation with a third reviewer. Results: The systematic review incorporated seven intervention studies spanning radiology, nursing, and laboratory departments within Health Information Systems (HIS). Findings revealed a broad range of sample sizes, from 152 to 805 participants, showcasing the diversity of healthcare professionals involved. Across these studies, CDS interventions demonstrated substantial positive impacts, particularly in radiology with a risk ratio of 1.75 (95% CI: 1.42-2.10) for improved diagnostic accuracy, in nursing with a 58% risk reduction in medication errors (95% CI: 0.30-0.58), and in laboratory services with a 65% lower risk of unnecessary tests (95% CI: 0.24-0.51). These consistent themes highlight the effectiveness of CDS interventions but underscore the need for ongoing customization to meet department-specific needs. Conclusions: The systematic review underscores the significant positive impact of Clinical Decision Support (CDS) implementation across radiology, nursing, and laboratory departments within Health Information Systems, as evidenced by improved diagnostic precision, medication management, and laboratory efficiency, while emphasizing the importance of continuous customization to address department-specific nuances.

Physicians’ Perceptions of Clinical Decision Support to Treat Patients With Heart Failure in the ED

Clinical decision support (CDS) could help emergency department (ED) physicians treat patients with heart failure (HF) by estimating risk, collating relevant history, and assisting with medication prescribing if physicians' perspectives inform its design and implementation. To evaluate CDS usability and workflow integration in the hands of ED physician end users who use it in clinical practice. This mixed-methods qualitative study administered semistructured interviews to ED physicians from 2 community EDs of Kaiser Permanente Northern California in 2023. The interview guide, based on the Usability Heuristics for User Interface Design and the Sociotechnical Environment models, yielded themes used to construct an electronic survey instrument sent to all ED physicians. Main outcomes were physicians' perceptions of using CDS to complement clinical decision-making, usability, and integration into ED clinical workflow. Seven key informant physicians (5 [71.4%] female, median [IQR] 15.0 [9.5-15.0] years in practice) were interviewed and survey responses from 51 physicians (23 [45.1%] female, median [IQR] 14.0 [9.5-17.0] years in practice) were received from EDs piloting the CDS intervention. Response rate was 67.1% (51 of 76). Physicians suggested changes to CDS accessibility, functionality, and workflow integration. Most agreed that CDS would improve patient care and fewer than half of physicians expressed hesitation about their capacity to consistently comply with its recommendations, citing workload concerns. Physicians preferred a passive prompt that encouraged, but did not mandate, interaction with the CDS. In this qualitative study of physicians who were using a novel CDS intervention to assist with ED management of patients with acute HF, several opportunities were identified to improve usability as well as several key barriers and facilitators to CDS implementation.

Impact of Predictive Text Clinical Decision Support on Imaging Order Entry in the Emergency Department

PurposeImaging clinical decision support (CDS) is designed to assist providers in selecting appropriate imaging studies and is now federally required. The aim of this study was to understand the effect of CDS on decisions and workflows in the emergency department (ED). MethodsThe authors’ institution’s order entry platform serves up structured indications for imaging orders. Imaging orders are scored by CDS on the basis of appropriate use criteria (AUC). CDS triggers alerts for imaging orders with low AUC scores. Because free text alone cannot be scored by CDS, an artificial intelligence predictive text (AIPT) module was implemented to guide the selection of structured indications when free-text indications are entered. A total of 17,355 imaging orders in the ED over 6 months were retrospectively analyzed. ResultsCDS alerts for low AUC scores were triggered for 3% of all imaging study orders (522 of 17,355). Providers spent an average of 24 seconds interacting with alerts. In 18 of 522 imaging orders with alerts, alternative studies were ordered. After AIPT implementation, the percentage of unscored studies significantly decreased from 81% to 45% (P < .001). ConclusionsIn a quaternary academic ED, CDS alerts triggered by low AUC scores caused minimal increase in time spent on imaging order entry but had a relatively marginal impact on imaging study selection. AIPT implementation increased the number of scored studies and could potentially enhance CDS effects. CDS implementation enables the collection of novel data regarding which imaging studies receive low AUC scores. Future work could include exploring alternative models of CDS implementation to maximize its impact.

Advances in Clinical Decision Support Systems: Contributions from the 2022 Literature.

To summarize significant research contributions published in 2022 in the field of clinical decision support (CDS) systems and select the best papers for the Decision Support section of the International Medical Informatics Association (IMIA) Yearbook 2023. A renewed search query for identifying CDS scholarship was developed using Medical Subject Headings (MeSH) terms and related keywords. The query was executed in PubMed in January 2023. The search results were reviewed in three stages by two reviewers: title-based triaging, followed by abstract screening, and then full text review. The resulting articles were sent for external review to identity best paper candidates. A total of 1,939 articles related to CDS were retrieved. Of these, 11 articles were selected as candidates for best papers. The general themes of the final three best papers are (1) reducing documentation burden through in-line guidance for clinical notes, (2) clinician engagement for continuous improvement of CDS, and (3) mitigating healthcare-related carbon emissions using scalable and accessible CDS, respectively. The field of clinical decision support remains highly active and dynamic, with innovative contributions to a range of clinical domains from primary to acute care. Interoperability issues, documentation burden, clinician acceptance, and the need for effective integration into existing healthcare workflows are among the prominent challenges and areas of interest faced by CDS implementation efforts.

Examining primary care provider experiences with using a clinical decision support tool for pain management.

To evaluate primary care provider (PCP) experiences using a clinical decision support (CDS) tool over 16 months following a user-centered design process and implementation. We conducted a qualitative evaluation of the Chronic Pain OneSheet (OneSheet), a chronic pain CDS tool. OneSheet provides pain- and opioid-related risks, benefits, and treatment information for patients with chronic pain to PCPs. Using the 5 Rights of CDS framework, we conducted and analyzed semi-structured interviews with 19 PCPs across 2 academic health systems. PCPs stated that OneSheet mostly contained the right information required to treat patients with chronic pain and was correctly located in the electronic health record. PCPs used OneSheet for distinct subgroups of patients with chronic pain, including patients prescribed opioids, with poorly controlled pain, or new to a provider or clinic. PCPs reported variable workflow integration and selective use of certain OneSheet features driven by their preferences and patient population. PCPs recommended broadening OneSheet access to clinical staff and patients for data entry to address clinician time constraints. Differences in patient subpopulations and workflow preferences had an outsized effect on CDS tool use even when the CDS contained the right information identified in a user-centered design process. To increase adoption and use, CDS design and implementation processes may benefit from increased tailoring that accommodates variation and dynamics among patients, visits, and providers.

Supporting Evidence-Based Care Through Informatics Innovation to Address the Overdose Epidemic.

Supporting Evidence-Based Care Through Informatics Innovation to Address the Overdose Epidemic.

Clinician Perspectives on Clinical Decision Support for Familial Hypercholesterolemia.

Familial Hypercholesterolemia (FH) is underdiagnosed in the United States. Clinical decision support (CDS) could increase FH detection once implemented in clinical workflows. We deployed CDS for FH at an academic medical center and sought clinician insights using an implementation survey. In November 2020, the FH CDS was deployed in the electronic health record at all Mayo Clinic sites in two formats: a best practice advisory (BPA) and an in-basket alert. Over three months, 104 clinicians participated in the survey (response rate 11.1%). Most clinicians (81%) agreed that CDS implementation was a good option for identifying FH patients; 78% recognized the importance of implementing the tool in practice, and 72% agreed it would improve early diagnosis of FH. In comparing the two alert formats, clinicians found the in-basket alert more acceptable (p = 0.036) and more feasible (p = 0.042) than the BPA. Overall, clinicians favored implementing the FH CDS in clinical practice and provided feedback that led to iterative refinement of the tool. Such a tool can potentially increase FH detection and optimize patient management.

Veterans Health Administration: Implementation of pharmacogenomic clinical decision support with statin medications and the SLCO1B1 gene as an exemplar.

In an effort to expedite the publication of articles, AJHP is posting manuscripts online as soon as possible after acceptance. Accepted manuscripts have been peer-reviewed and copyedited, but are posted online before technical formatting and author proofing. These manuscripts are not the final version of record and will be replaced with the final article (formatted per AJHP style and proofed by the authors) at a later time. To describe the implementation of clinical decision support tools for alerting prescribers of actionable drug-gene interactions in the Veterans Health Administration (VHA). Drug-gene interactions have been the focus of clinicians for years. Interactions between SCLO1B1 genotype and statin medications are of particular interest as these can inform risk for statin-associated muscle symptoms (SAMS). VHA identified approximately 500,000 new users of statin medications prescribed in VHA in fiscal year 2021, some of whom could benefit from pharmacogenomic testing for the SCLO1B1 gene. In 2019, VHA implemented the Pharmacogenomic Testing for Veterans (PHASER) program to offer panel-based, preemptive pharmacogenomic testing and interpretation. The PHASER panel includes SLCO1B1, and VHA utilized Clinical Pharmacogenomics Implementation Consortium statin guidelines to build its clinical decision support tools. The program's overarching goal is to reduce the risk of adverse drug reactions such as SAMS and improve medication efficacy by alerting practitioners of actionable drug-gene interactions. We describe the development and implementation of decision support for the SLCO1B1 gene as an example of the approach being used for the nearly 40 drug-gene interactions screened for by the panel. The VHA PHASER program identifies and addresses drug-gene interactions as an application of precision medicine to reduce veterans' risks for adverse events. The PHASER program's implementation of statin pharmacogenomics utilizes a patient's SCLO1B1 phenotype to alert providers of the risk for SAMS with the statin being prescribed and how to lower that risk through a lower dose or alternative statin selection. The PHASER program may help reduce the number of veterans who experience SAMS and may improve their adherence to statin medications.

Evidence-Based Clinical Decision Support to Improve Care for Patients Hospitalized With Acute Myocardial Infarction.

Clinical decision support in the EHR is an innovation that can support guideline adherence in acute myocardial infarction. Cardiac rehabilitation referral and left ventricular systolic function assessment are part of evidence-based clinical practice guidelines associated with reduced morbidity and mortality following acute myocardial infarction. Effective clinical decision support is sustained by evidence-based principles for design and implementation. This quality improvement project evaluated the impact of practice advisories designed using principles of effective clinical decision support design to improve performance of left ventricular systolic function assessment and ambulatory referral to cardiac rehabilitation for patients hospitalized with acute myocardial infarction. Performance in cardiac rehabilitation referral and left ventricular systolic function assessment was measured for a 3-month interval pre- and post-intervention. Pre-implementation, cardiac rehabilitation referral or valid documented reason for non-referral was 80.3%. Rehabilitation referral or documented valid reason for non-referral increased to 98.4% post-implementation (P < .001). Left ventricular systolic function assessment increased from 94.2% to 100% following clinical decision support implementation (P = .120). This quality improvement project supports the positive impact of effective clinical decision support design and implementation to improve outcomes for patients hospitalized with acute myocardial infarction.

Implementation strategies to address the determinants of adoption, implementation, and maintenance of a clinical decision support tool for emergency department buprenorphine initiation: a qualitative study

BackgroundUntreated opioid use disorder (OUD) is a significant public health problem. Buprenorphine is an evidence-based treatment for OUD that can be initiated in and prescribed from emergency departments (EDs) and office settings. Adoption of buprenorphine initiation among ED clinicians is low. The EMBED pragmatic clinical trial investigated the effectiveness of a clinical decision support (CDS) tool to promote ED clinicians’ behavior related to buprenorphine initiation in the ED. While the CDS intervention was not associated with increased rates of buprenorphine treatment for patients with OUD at intervention ED sites, attending physicians at intervention EDs were more likely to initiate buprenorphine at least once over the duration of the study compared to those in the usual care arms (44.4% vs 34.0%, P = 0.01). This suggests the CDS intervention may be associated with increased adoption of buprenorphine initiation. As a secondary aim, we sought to identify the determinants of CDS adoption, implementation, and maintenance in a variety of ED settings and geographic locations.MethodsWe purposively sampled and conducted semi-structured, in-depth interviews with clinicians across EMBED trial sites randomized to the intervention arm from five healthcare systems. Interviews elicited clinician experiences regarding buprenorphine initiation and CDS use. Interviews were analyzed using directed content analysis informed by the Practical, Robust Implementation and Sustainability Model (PRISM). We used a hybrid approach (a priori codes informed by PRISM and emergent codes) for codebook development. ATLAS.ti (version 9.0) was used for data management. Coded data were analyzed within individual interview transcripts and across all interviews to identify major themes. This process involved (1) combining, comparing, and making connections between codes; (2) writing analytic memos about observed patterns; and (3) frequent team meetings to discuss emerging patterns.ResultsTwenty-eight interviews were conducted. Major themes that influenced the successful adoption, implementation, and maintenance of the EMBED intervention and ED-initiated BUP were organizational culture and commitment, clinician training and support, the ability to connect patients to ongoing treatment, and the ability to tailor implementation to each ED. These findings informed the identification of implementation strategies (framed using PRISM domains) to enhance the ED initiation of buprenorphine.ConclusionThe findings from this qualitative analysis can provide guidance to build better systems to promote the adoption of ED-initiated buprenorphine.

Practice facilitation for scale up of clinical decision support for hypertension management: study protocol for a cluster randomized control trial

BackgroundOnly half of patients with hypertension have adequately controlled blood pressure. Clinical decision support (CDS) has the potential to overcome barriers to delivering guideline-recommended care and improve hypertension management. However, optimal strategies for scaling CDS have not been well established, particularly in small, independent primary care practices which often lack the resources to effectively change practice routines. Practice facilitation is an implementation strategy that has been shown to support process changes. Our objective is to evaluate whether practice facilitation provided with hypertension-focused CDS can lead to improvements in blood pressure control for patients seen in small primary care practices. Methods/designWe will conduct a cluster randomized control trial to compare the effect of hypertension-focused CDS plus practice facilitation on BP control, as compared to CDS alone. The practice facilitation intervention will include an initial training in the CDS and a review of current guidelines along with follow-up for coaching and integration support. We will randomize 46 small primary care practices in New York City who use the same electronic health record vendor to intervention or control. All patients with hypertension seen at these practices will be included in the evaluation. We will also assess implementation of CDS in all practices and practice facilitation in the intervention group. DiscussionThe results of this study will inform optimal implementation of CDS into small primary care practices, where much of care delivery occurs in the U.S. Additionally, our assessment of barriers and facilitators to implementation will support future scaling of the intervention.ClinicalTrials.gov Identifier: NCT05588466

Using clinical decision support to improve urine testing and antibiotic utilization.

Urine cultures collected from catheterized patients have a high likelihood of false-positive results due to colonization. We examined the impact of a clinical decision support (CDS) tool that includes catheter information on test utilization and patient-level outcomes. This before-and-after intervention study was conducted at 3 hospitals in North Carolina. In March 2021, a CDS tool was incorporated into urine-culture order entry in the electronic health record, providing education about indications for culture and suggesting catheter removal or exchange prior to specimen collection for catheters present >7 days. We used an interrupted time-series analysis with Poisson regression to evaluate the impact of CDS implementation on utilization of urinalyses and urine cultures, antibiotic use, and other outcomes during the pre- and postintervention periods. The CDS tool was prompted in 38,361 instances of urine cultures ordered in all patients, including 2,133 catheterized patients during the postintervention study period. There was significant decrease in urine culture orders (1.4% decrease per month; P < .001) and antibiotic use for UTI indications (2.3% decrease per month; P = .006), but there was no significant decline in CAUTI rates in the postintervention period. Clinicians opted for urinary catheter removal in 183 (8.5%) instances. Evaluation of the safety reporting system revealed no apparent increase in safety events related to catheter removal or reinsertion. CDS tools can aid in optimizing urine culture collection practices and can serve as a reminder for removal or exchange of long-term indwelling urinary catheters at the time of urine-culture collection.

Improving Pediatric Intensive Care Unit Discharge Timeliness of Infants with Bronchiolitis Using Clinical Decision Support.

Identifying children ready for transfer out of the pediatric intensive care unit (PICU) is an area that may benefit from clinical decision support (CDS). We previously implemented a quality improvement (QI) initiative to accelerate the transfer evaluation of non-medically complex PICU patients with viral bronchiolitis receiving floor-appropriate respiratory support. Design a CDS tool adaptation of this QI initiative to further accelerate transfer evaluation of appropriate patients. The original initiative focused on identifying for transfer evaluation otherwise healthy children admitted to the PICU with bronchiolitis who had been receiving floor-appropriate levels of respiratory support for at least 6 hours. However, this initiative required that clinicians manually track the respiratory support of qualifying patients. We designed an electronic health record (EHR)-based CDS tool to automate identification of transfer-ready candidates. The tool parses EHR data to identify children meeting prior QI initiative criteria and alerts clinicians to assess transfer readiness once a child has been receiving floor-appropriate respiratory support for 6 hours. We compared time from reaching floor-appropriate support to placement of the transfer order ("time-to-transfer"), PICU length of stay (LOS), and hospital LOS between patients admitted prior to our QI initiative (December 1, 2018-October 19, 2019, "pre-QI phase"), during the initiative but before CDS tool implementation (October 20, 2019-February 7, 2022, "QI phase"), and after CDS implementation (February 8-November 11, 2022, "CDS phase"). CDS-phase patients (n = 131) had a shorter median time-to-transfer of 5.23 (interquartile range [IQR], 3.38-10.0) hours compared with QI-phase patients (n = 304) at 5.93 (IQR, 4.23-12.2) hours (p = 0.04). PICU and hospital LOS values decreased from the pre-QI (n = 150) to QI phase. Though LOS reductions were sustained during the CDS phase, further reductions from QI to CDS phase were not statistically significant. An EHR-based CDS adaptation of a prior QI initiative facilitated timely identification of PICU patients with bronchiolitis ready for transfer evaluation. Such tools might allow PICU clinicians to focus on other high-acuity tasks while accelerating transfer evaluation of appropriate patients.

The Application of Dental Fluoride Varnish in Children: A Low Cost, High-Value Implementation Aided by Passive Clinical Decision Support.

Fluoride is vital in the prevention of dental caries in children. In 2014, the U.S. Preventive Services Task Force deemed fluoride varnish a recommended preventive service (grade B). Electronic health record-based clinical decision support (CDS) tools have shown variable ability to alter physicians' ordering behaviors. This study aimed to increase the application of fluoride varnish in children while analyzing the effect of two passive CDS tools-an order set and a note template. Data on outpatient pediatric visits over an 18-month period before and after CDS implementation (October 15, 2020-April 15, 2022) were queried, while trends in application rate of fluoride were examined. We constructed a multiple logistic regression model with a primary outcome of whether a patient received fluoride at his/her visit. The primary predictor was a "phase" variable representing the CDS implemented. Physician interaction with CDS as well as the financial effects of the resulting service use were also examined. There were 3,049 well-child visits of children aged 12 months to 5 years. The addition of a fluoride order to a "Well Child Check" order set led to a 10.6% increase in ordering over physician education alone (25.4 vs. 14.8%, p = 0.001), while the insertion of fluoride-specific text to drop-down lists in clinical notes led to a 6.2% increase (31.5 vs. 25.4%, p = 0.005). Whether a patient received topical fluoride was positively associated with order set implementation (odds ratio [OR] = 5.87, 95% confidence interval [CI]: 4.20-8.21) and fluoride-specific drop-down lists (OR = 7.81, 95% CI: 5.41-11.28). Female providers were more likely to use order sets when ordering fluoride (56.2 vs. 40.9% for males, p ≤ 0.0001). Added revenue totaled $15,084. The targeted use of order sets and note templates was positively associated with the ordering of topical fluoride by physicians.

Analytics to monitor local impact of the Protecting Access to Medicare Act's imaging clinical decision support requirements.

This study aimed is to: (1) extend the Integrating the Biology and the Bedside (i2b2) data and application models to include medical imaging appropriate use criteria, enabling it to serve as a platform to monitor local impact of the Protecting Access to Medicare Act's (PAMA) imaging clinical decision support (CDS) requirements, and (2) validate the i2b2 extension using data from the Medicare Imaging Demonstration (MID) CDS implementation. This study provided a reference implementation and assessed its validity and reliability using data from the MID, the federal government's predecessor to PAMA's imaging CDS program. The Star Schema was extended to describe the interactions of imaging ordering providers with the CDS. New ontologies were added to enable mapping medical imaging appropriateness data to i2b2 schema. z-Ratio for testing the significance of the difference between 2 independent proportions was utilized. The reference implementation used 26327 orders for imaging examinations which were persisted to the modified i2b2 schema. As an illustration of the analytical capabilities of the Web Client, we report that 331/1192 or 28.1% of imaging orders were deemed appropriate by the CDS system at the end of the intervention period (September 2013), an increase from 162/1223 or 13.2% for the first month of the baseline period, December 2011 (P = .0212), consistent with previous studies. The i2b2 platform can be extended to monitor local impact of PAMA's appropriateness of imaging ordering CDS requirements.

Pharmacogenomic Clinical Decision Support: A Scoping Review.

Clinical decision support (CDS) is often cited as an essential part of pharmacogenomics (PGx) implementations. A multitude of strategies are available; however, it is unclear which strategies are effective and which metrics are used to quantify clinical utility. The objective of this scoping review was to aggregate previous studies into a cohesive depiction of the current state of PGx CDS implementations and identify areas for future research on PGx CDS. Articles were included if they (i) described electronic CDS tools for PGx and (ii) reported metrics related to PGx CDS. Twenty of 3,449 articles were included and provided data on PGx CDS metrics from 15 institutions, with 93% of programs located at academic medical centers. The most common tools in CDS implementations were interruptive post-test alerts. Metrics for clinical response and alert response ranged from 12-73% and 21-98%, respectively. Few data were found on changes in metrics over time and measures that drove the evolution of CDS systems. Relatively few data were available regarding support of optimal approaches for PGx CDS. Post-test alerts were the most widely studied approach, and their effectiveness varied greatly. Further research on the usability, effectiveness, and optimization of CDS tools is needed.

Multisite implementation of a workflow-integrated machine learning system to optimize COVID-19 hospital admission decisions

Demand has outstripped healthcare supply during the coronavirus disease 2019 (COVID-19) pandemic. Emergency departments (EDs) are tasked with distinguishing patients who require hospital resources from those who may be safely discharged to the community. The novelty and high variability of COVID-19 have made these determinations challenging. In this study, we developed, implemented and evaluated an electronic health record (EHR) embedded clinical decision support (CDS) system that leverages machine learning (ML) to estimate short-term risk for clinical deterioration in patients with or under investigation for COVID-19. The system translates model-generated risk for critical care needs within 24 h and inpatient care needs within 72 h into rapidly interpretable COVID-19 Deterioration Risk Levels made viewable within ED clinician workflow. ML models were derived in a retrospective cohort of 21,452 ED patients who visited one of five ED study sites and were prospectively validated in 15,670 ED visits that occurred before (n = 4322) or after (n = 11,348) CDS implementation; model performance and numerous patient-oriented outcomes including in-hospital mortality were measured across study periods. Incidence of critical care needs within 24 h and inpatient care needs within 72 h were 10.7% and 22.5%, respectively and were similar across study periods. ML model performance was excellent under all conditions, with AUC ranging from 0.85 to 0.91 for prediction of critical care needs and 0.80–0.90 for inpatient care needs. Total mortality was unchanged across study periods but was reduced among high-risk patients after CDS implementation.