Applied Clinical Informatics最新文献

This study aimed to evaluate the effect of optimizing the ambulatory medication preference list on provider efficiency in medication ordering.Using electronic health record (EHR) vendor data, a multidisciplinary informatics team optimized the general ambulatory medication preference list to better align with providers' ordering patterns. We conducted a pre-postintervention analysis assessing time-in-orders per encounter and number of manual changes per order.Postintervention, average manual changes per order decreased from 4.12 to 3.00 (p < 0.01), and median time spent in the orders activity per encounter decreased from 3.1 to 2.3 minutes (p < 0.01).Optimizing the ambulatory medication preference list reduced time spent and clicks needed by providers when ordering medications. This is relevant to ongoing efforts to address EHR-related burden.

The adoption of electronic health records (EHRs) into clinical practice has changed clinical workflows and, in some cases, increased documentation burden and clinician burnout. Identifying factors associated with perceived EHR usability after the implementation of a new EHR may guide efforts to reduce burden and burnout.This study measured: (1) group-level perceptions of EHR usability pre- and postimplementation of a new EHR; (2) adaptation to the new EHR; and (3) the effects of clinical role, setting, and specialty on these measures.Pre- and postimplementation surveys were sent to clinical staff at two academic medical centers (AMC A and AMC B), each part of the same Northeast health system where one instance of a new EHR was implemented starting in 2020. The surveys measured constructs from the Health Information Technology Usability Evaluation Scale (Health-ITUES) and Health Information Technology Adaptation survey. Unpaired t-tests assessed changes in group-level scores from pre- to postimplementation, and multiway analyses of variance with post hoc pairwise t-tests with Bonferroni's correction were used to assess differences in scores by clinical role, setting, and specialty.Average Perceived Usefulness (PU) and adaptation scores were higher at AMC B than at AMC A, but similar pre- to postimplementation trends were observed at both sites. Perceptions of Quality of Work Life (QWL), PU, and User Control (UC) improved across both sites postimplementation, whereas Perceived Ease of Use and Cognitive Support and Situational Awareness declined. Ordering Providers, Registered Nurses, clinicians practicing in the Emergency Department setting, and Emergency Medicine, and Critical/Intensive Care specialists had statistically different scores across various constructs.After implementation of a new EHR system at two AMCs, clinical staff perceptions of quality of work life (QWL), perceived usefulness (PU), and user control (UC) generally improved, although perceptions of perceived ease of use and cognitive support declined.

Federated network studies allow data to remain locally while the research is conducted through the sharing of analytical code and aggregated results across different health care settings and countries. A large number of databases have been mapped to the Observational Medical Outcomes Partnership (OMOP) Common Data Model (CDM), boosting the use of analytical pipelines for standardized observational research within this open science framework. Transparency, reproducibility, and robustness of results have positioned federated analyses using the OMOP CDM within the European Health Data and Evidence Network (EHDEN) as an essential tool for generating large-scale evidence.We conducted large-scale federated analyses involving 52 databases from 19 countries using the OMOP CDM. In this State-of-the-Art/Best Practice article, we aimed to share key lessons and strategies for conducting such complex, large multidatabase analyses.Meticulous planning, establishing a strong community of collaborators, efficient communication channels, standardized analytics, and strategic division of responsibilities are essential. We highlight the benefits of network engagement, cross-fertilization of ideas, and shared learning. Further key elements contributing to the study's success included an inclusive, incremental implementation of the analytical code, timely engagement of data partners, and community webinars to discuss and interpret study findings.We received predominantly positive feedback from data custodians about their participation, and included input for further improvements for future large-scale federated network studies from this shared learning experience.

This study aimed to assess the effect of a passive, opt-in electronic health record (EHR)-based clinical decision support (CDS), the Chronic Pain OneSheet, on guideline-recommended chronic pain management in primary care.A pragmatic randomized controlled trial with a parallel group design was conducted between October 2020 and May 2022. Participants were 137 primary care clinicians (PCCs) treating qualifying patients with chronic pain at 25 primary care clinics within two academic health systems in the United States. PCCs were randomized in the EHR to have access to OneSheet or usual care. OneSheet aggregates guideline-relevant information in a single view and provides shortcuts to guideline-recommended actions (e.g., ordering urine drug screening [UDS] for patients prescribed opioids). We constructed five visit-level binary outcomes: (1) documenting pain-related goals; (2) documenting pain and function via Pain, Enjoyment of Life and General Activity (PEG) scale; (3) reviewing prescription drug monitoring programs (PDMPs); (4) ordering UDS; and (5) ordering naloxone. Analysis used generalized linear mixed models for each outcome.OneSheet access minimally increased rates of pain-related goal documentation (0.2 percentage point increase, p = 0.013), PEG scale documentation (0.7 percentage point increase, p < 0.001), and UDS orders (2.2 percentage point increase, p = 0.006). OneSheet access decreased the rate of ordering naloxone (0.5 percentage point decrease, p < 0.001). OneSheet access did not affect PDMP review rates (0.5 percentage point decrease, p = 0.382).OneSheet access did not result in clinically significant improvements in guideline-recommended management of chronic pain in primary care despite a robust user-centered design incorporating clinician input and EHR integration. Several factors likely limited OneSheet effectiveness, including limited ability to target certain patient visits, workflow limits on data collection and ordering, and evolving COVID-19 and opioid epidemic-related policies and procedures. These findings highlight specific limitations of OneSheet and the broader challenges of implementing effective EHR-based CDS in complex health care environments.

Overuse and misuse of antibiotics is an urgent health care problem and one of the key factors in antibiotic resistance. Validated clinical prediction rules have shown effectiveness in guiding providers to an appropriate diagnosis and identifying when antibiotics are the recommended choice for treatment.We aimed to study the relative ability of registered nurses using clinical prediction rules to guide the management of acute respiratory infections in a simulated environment compared with practicing primary care physicians.We evaluated a case-based simulation of the diagnosis and treatment for acute respiratory infections using clinical prediction rules. As a secondary outcome, we examined nursing self-efficacy by administering a survey before and after case evaluations. Participants included 40 registered nurses from three academic medical centers and five primary care physicians as comparators. Participants evaluated six simulated case studies, three for patients presenting with cough symptoms, and three for sore throat.Compared with physicians, nurses determined risk and treatment for simulated sore throat cases using clinical prediction rules with 100% accuracy in low-risk sore throat cases versus 80% for physicians. We found great variability in the accuracy of the risk level and appropriate treatment for cough cases. Nurses reported slight increases in self-efficacy from baseline to postcase evaluation suggesting further information is needed to understand correlation.Clinical prediction rules used by nurses in sore throat management workflows can guide accurate diagnosis and treatment in simulated cases, while cough management requires further exploration. Our results support the future implementation of automated prediction rules in a clinical decision support tool and a thorough examination of their effect on clinical practice and patient outcomes.

Clinical decision support systems (CDSS) have been suggested to be helpful in detecting and preventing drug-related problems such as adverse drug events (ADEs). However, patient participation systems monitoring self-reported data, such as symptoms, are still sparsely described in the literature.This study aimed to investigate if the use of a patient participating CDSS (PCDSS) can facilitate early detection of ADEs, thereby contributing to safer drug treatment in older adults.A 1-year prospective observational study of elderly patients using a free web-based PCDSS to register symptoms over time at home. Initially, the PCDSS analyzed the extent and quality of the patient's drug use, based on a Swedish national set of criteria, and assessed drug-related symptoms using a standardized scale (PHASE-20). Thereafter, the patients recorded symptoms at home for 1 year-the first 6 months in free text, the second 6 months selecting from 19 predefined symptoms. The PCDSS signaled when symptoms were registered on three occasions in a 3-week period. The patient was then asked to contact his/her nurse at the healthcare center (HCC) for assessment of the symptoms and decisions on further contacts with the nurse or doctor. We analyzed the extent of signals generated, accompanying contacts, and associated medication reviews and adjustments.The 48 study participants registered 1,275 symptoms during the monitoring period, 61% by women. The PCDSS generated a total of 171 signals, of which 58% from women. Seventy-one percent (121) occurred under the first registration (free text) period. Of all signals, 44% (75) led to activities at the HCC, of which 48% (36) were physician contacts. In total, they contributed to medication reviews in 42% (15) and medication adjustments in 64% (23), with a total of 33 adjustments.Patient participation by self-reporting symptoms via a PCDSS can contribute to safer drug use.

The German Federal Ministry of Health introduced the Pflegepersonalregelung 2.0 (PPR 2.0) to address the nursing staffing crisis. It establishes a framework to determine personnel requirements, ensuring adequate staffing. However, the required daily classification of patient care levels imposes a significant administrative burden on nursing staff. Digitizing this process may reduce documentation time and enhance efficiency, but effectiveness depends on usability and acceptance.This study evaluates the acceptance and usability of a direct digitization of the analog PPR 2.0 classification catalog into a digital user interface-the PPR 2.0 Calculator.A mixed-methods approach was used, combining quantitative assessment using the Technology Acceptance Model 3 (TAM 3) and the System Usability Scale (SUS), with qualitative insights from a semistructured interview. Fifteen nursing staff members from a pediatric rheumatology clinic in Germany participated.The PPR 2.0 Calculator was rated highly usable, with strong scores for Perceived Ease of Use (4.00) and Computer Self-Efficacy (4.09). Participants required minimal technical support, indicating an intuitive interface. However, Perceived Usefulness (2.82) and Job Relevance (2.53) scores were lower, suggesting limited value in daily workflows. The SUS score (65.50) was slightly below the benchmark of 68, indicating good usability with moderate room for improvement.Digitizing the analog PPR 2.0 catalog resulted in good usability, but significant challenges regarding practical relevance and workflow integration remained. Directly adopting the catalog content negatively affected perceived usefulness and job relevance, revealing limitations in the classification framework itself. Refinement of the PPR 2.0 framework is needed to reflect real-world clinical nursing tasks. Seamless integration with existing infrastructures and structured documentation is also critical. Future improvements should go beyond simple digitization and explore automated classification features.

Effective communication in the intensive care unit (ICU) is essential, particularly for patients with non-English language preference, yet timely access to professional interpreters remains limited. While artificial intelligence (AI)-based translation tools have been explored in outpatient and nonacute care settings, studies evaluating their use in acute care, environments such as the ICU remain limited. To address this gap, we developed AI-TransLATE (AI-enhanced Transition to Language-Agnostic Transcultural Engagement), a speech-based translation tool designed for multilingual communication in critical care settings.This study aimed to assess the interpretation quality of AI-TransLATE across four languages-Spanish, Chinese, Arabic, and Turkish-using scripted ICU scenarios.We created ICU communication scripts and recorded bilingual research team members simulating clinical interactions. Two independent bilingual evaluators assessed interpretation quality using a 5-point Likert scale across fluency, adequacy, meaning preservation, and severity of errors. Clarity and cultural appropriateness were also rated. Percentage agreement was used to assess interrater agreement.AI-TransLATE achieved acceptable composite scores (≥16/20) across all languages. Spanish and Turkish performed consistently well; Chinese and Arabic showed variability due to omissions and terminology errors.AI-TransLATE shows promise as a clinical communication tool, but further evaluation in real-world, unscripted ICU settings is needed.

This study aimed to leverage a large language model (LLM) to improve the efficiency and thoroughness of medical record documentation. This study focused on aiding clinical staff in creating structured summaries with the help of an LLM and assessing the quality of these artificial intelligence (AI)-proposed records in comparison to those produced by doctors.This strategy involved assembling a team of specialists, including data engineers, physicians, and medical information experts, to develop guidelines for medical summaries produced by an LLM (Llama 3.1), all under the direction of policymakers at the study hospital. The LLM proposes admission, weekly summaries, and discharge notes for physicians to review and edit. A validated Physician Documentation Quality Instrument (PDQI-9) was used to compare the quality of physician-authored and LLM-generated medical records.The results showed no significant difference was observed in the total PDQI-9 scores between the physician-drafted and AI-created weekly summaries and discharge notes (p = 0.129 and 0.873, respectively). However, there was a significant difference in the total PDQI-9 scores between the physician and AI admission notes (p = 0.004). Furthermore, there were significant differences in item levels between physicians' and AI notes. After deploying the note-assisted function in our hospital, it gradually gained popularity.LLM shows considerable promise for enhancing the efficiency and quality of medical record summaries. For the successful integration of LLM-assisted documentation, regular quality assessments, continuous support, and training are essential. Implementing LLM can allow clinical staff to concentrate on more valuable tasks, potentially enhancing overall health care delivery.

Sepsis is a heterogeneous syndrome with high morbidity and mortality. Despite extensive clinical trials, therapeutic progress remains limited, in part due to the absence of actionable sepsis subtypes.This study aimed to evaluate the feasibility of using HL7 Fast Healthcare Interoperability Resources (FHIR) for prerandomization sepsis subtyping to support clinical trial enrichment across multiple health systems.Data from 765 encounters at two academic medical centers were analyzed. FHIR-based resources were extracted from both research data warehouses (RDWs) and electronic health records (EHRs). A Python implementation of the Sepsis Endotyping in Emergency Care (SENECA) sepsis subtyping algorithm was developed to query and assemble FHIR resources for subtype classification.Open-source Python code for the SENECA algorithm is provided on GitHub. Experiments demonstrated: (1) successful sepsis subtyping across both health systems; (2) concordance between the original R implementation and the new Python implementation; and (3) discrepancies when comparing RDW-derived versus EHR-integrated FHIR APIs, primarily due to query and filtering limitations. Missing data were common and influenced by both clinical practice and FHIR API constraints. We provide five recommendations to address these challenges.FHIR can support multi-institutional sepsis subtyping and trial enrichment, though technical and governance challenges remain.