Deployment of Real-time Natural Language Processing and Deep Learning Clinical Decision Support in the Electronic Health Record: Pipeline Implementation for an Opioid Misuse Screener in Hospitalized Adults (Preprint) DOI
Majid Afshar, Sabrina Adelaine, Felice Resnik

et al.

Published: Dec. 11, 2022

BACKGROUND The clinical narrative in electronic health records (EHRs) carries valuable information for predictive analytics; however, its free-text form is difficult to mine and analyze decision support (CDS). Large-scale natural language processing (NLP) pipelines have focused on data warehouse applications retrospective research efforts. There remains a paucity of evidence implementing NLP at the bedside care delivery. OBJECTIVE We aimed detail hospital-wide, operational pipeline implement real-time NLP-driven CDS tool describe protocol an implementation framework with user-centered design tool. METHODS integrated previously trained open-source convolutional neural network model screening opioid misuse that leveraged EHR notes mapped standardized medical vocabularies Unified Medical Language System. A sample 100 adult encounters were reviewed by physician informaticist silent testing deep learning algorithm before deployment. An end user interview survey was developed examine acceptability best practice alert (BPA) provide results recommendations. planned also included human-centered feedback BPA, cost-effectiveness, noninferiority patient outcome analysis plan. RESULTS reproducible workflow shared pseudocode cloud service ingest, process, store as Health Level 7 messages from major vendor elastic computing environment. Feature engineering used engine, features fed into algorithm, returned BPA EHR. On-site demonstrated sensitivity 93% (95% CI 66%-99%) specificity 92% 84%-96%), similar published validation studies. Before deployment, approvals received across hospital committees inpatient operations. Five interviews conducted; they informed development educational flyer further modified exclude certain patients allow refusal longest delay because cybersecurity approvals, especially exchange protected between Microsoft (Microsoft Corp) Epic (Epic Systems vendors. In testing, resultant provided within minutes provider entering note CONCLUSIONS components detailed tools other systems benchmark. deployment artificial intelligence routine presents important yet unfulfilled opportunity, our close gap intelligence–driven CDS. CLINICALTRIAL ClinicalTrials.gov NCT05745480; https://www.clinicaltrials.gov/ct2/show/NCT05745480

Language: Английский

Deployment of Real-time Natural Language Processing and Deep Learning Clinical Decision Support in the Electronic Health Record: Pipeline Implementation for an Opioid Misuse Screener in Hospitalized Adults DOI Creative Commons
Majid Afshar, Sabrina Adelaine, Felice Resnik

et al.

JMIR Medical Informatics, Journal Year: 2023, Volume and Issue: 11, P. e44977 - e44977

Published: April 20, 2023

Background The clinical narrative in electronic health records (EHRs) carries valuable information for predictive analytics; however, its free-text form is difficult to mine and analyze decision support (CDS). Large-scale natural language processing (NLP) pipelines have focused on data warehouse applications retrospective research efforts. There remains a paucity of evidence implementing NLP at the bedside care delivery. Objective We aimed detail hospital-wide, operational pipeline implement real-time NLP-driven CDS tool describe protocol an implementation framework with user-centered design tool. Methods integrated previously trained open-source convolutional neural network model screening opioid misuse that leveraged EHR notes mapped standardized medical vocabularies Unified Medical Language System. A sample 100 adult encounters were reviewed by physician informaticist silent testing deep learning algorithm before deployment. An end user interview survey was developed examine acceptability best practice alert (BPA) provide results recommendations. planned also included human-centered feedback BPA, cost-effectiveness, noninferiority patient outcome analysis plan. Results reproducible workflow shared pseudocode cloud service ingest, process, store as Health Level 7 messages from major vendor elastic computing environment. Feature engineering used engine, features fed into algorithm, returned BPA EHR. On-site demonstrated sensitivity 93% (95% CI 66%-99%) specificity 92% 84%-96%), similar published validation studies. Before deployment, approvals received across hospital committees inpatient operations. Five interviews conducted; they informed development educational flyer further modified exclude certain patients allow refusal longest delay because cybersecurity approvals, especially exchange protected between Microsoft (Microsoft Corp) Epic (Epic Systems vendors. In testing, resultant provided within minutes provider entering note Conclusions components detailed tools other systems benchmark. deployment artificial intelligence routine presents important yet unfulfilled opportunity, our close gap intelligence–driven CDS. Trial Registration ClinicalTrials.gov NCT05745480; https://www.clinicaltrials.gov/ct2/show/NCT05745480

Language: Английский

Citations

18
Implementing TeamMAPPS: Formative Qualitative Findings from the Dissemination and Implementation Study of a New Evidence-Based Team Science Intervention DOI Creative Commons
Stephen Molldrem, Heidi Luft,

Jeffrey S. Farroni

et al.

Journal of Clinical and Translational Science, Journal Year: 2025, Volume and Issue: 9(1)

Published: Jan. 1, 2025

Abstract Introduction: Team Methods to Advance Processes and Performance in Science (TeamMAPPS) is an evidence-based competency model intervention. TeamMAPPS was developed by experts the of with translational teams mind. focuses on three core teamwork competencies: (1) psychological safety, (2) awareness exchange, (3) self-correction adaptation. In 2023, framework operationalized into five online training modules that can be used train whole or individuals, without facilitation, any order. This article reports formative findings from pre-implementation stage Dissemination Implementation (D&I) study. Methods: We conducted 27 interviews participant-observation fieldwork 23 individuals involved conceptualization, design, implementation (four were interviewed twice). All implementers affiliated a Clinical Translational Award (CTSA) hub. Data collected during pre-implementation, when being tested early-stage trained. D&I theories frameworks structure study, analyze interview data, recommend strategies. Findings: “Adoption,” “reach,” “effectiveness” emerged as key outcomes. perceived evidence-based, highly adaptable, intervention offering unique benefits. draw participants’ responses expert recommendations suggest Conclusions: CTSAs other organizations use varied strategies implement TeamMAPPS. The flexibility its rootedness evidence-base synthesized leaders make appealing for seeking enhance their team offerings.

Language: Английский

Citations

0
Applying the lessons of implementation science to maximize feasibility and usability in team science intervention development DOI Creative Commons
Betsy Rolland, Felice Resnik, Sarah D. Hohl

et al.

Journal of Clinical and Translational Science, Journal Year: 2021, Volume and Issue: 5(1)

Published: Jan. 1, 2021

The Science of Team (SciTS) has generated a substantial body work detailing characteristics effective teams. However, that knowledge not been widely translated into accessible, active, actionable, evidence-based interventions to help translational teams enhance their team functioning and outcomes. Over the past decade, field Implementation rapidly developed methods approaches increase translation biomedical research findings clinical care, providing roadmap for mitigating challenges developing while maximizing feasibility utility. Here, we propose an approach intervention development using constructs from two frameworks, Consolidated Framework Research, Reach, Effectiveness, Adoption, Implementation, Maintenance, extend Wisconsin Interventions framework described in Rolland et al. 2021. These can SciTS researchers design, build, test, disseminate meet needs both adopters, institutional leadership decides whether adopt intervention, implementers, those actually intervention. Systematically considering impact design decisions on usability may lead quickly move prototype pilot test pragmatic trials assess impact.

Language: Английский

Citations

19
Enhancing translational team effectiveness: The Wisconsin Interventions in Team Science framework for translating empirically informed strategies into evidence-based interventions DOI Creative Commons
Betsy Rolland, Sarah D. Hohl, LaKaija J. Johnson

et al.

Journal of Clinical and Translational Science, Journal Year: 2021, Volume and Issue: 5(1)

Published: Jan. 1, 2021

Abstract Achieving the clinical, public health, economic, and policy benefits of translational science requires integration application findings across biomedical, behavioral health policy, thus, collaboration experts in these areas. To do so, teams need skills, knowledge, attitudes to mitigate challenges build on strengths cross-disciplinary collaboration. Though competencies are not innate teams, they can be built through implementation effective strategies interventions. The Science Team (SciTS) has contributed robust theories evidence empirically-informed best practices enhance Yet field lacks methodological approaches rigorously translate those into evidence-based interventions improve collaborative research. Here, we apply lessons from Implementation Human-Centered Design & Engineering describe Wisconsin Interventions (WITS) framework, a process for translating established team bolster effectiveness. illustrate our use WITS, how University Wisconsin’s Institute Clinical Translational Research translated existing Collaboration Planning framework robust, scalable, replicable intervention. We conclude with recommendations future SciTS research refine test framework.

Language: Английский

Citations

14
Dissemination and implementation science resources, training, and scientific activities provided through CTSA programs nationally: Opportunities to advance D&I research and training capacity DOI Creative Commons
Rachel C. Shelton, Rowena J Dolor, Jonathan N. Tobin

et al.

Journal of Clinical and Translational Science, Journal Year: 2022, Volume and Issue: 6(1)

Published: Jan. 1, 2022

Abstract Introduction: Clinical and Translational Science Award (CTSA) Program hubs are well-positioned to advance dissemination implementation (D&I) research training capacity nationally, though little is known about what D&I support services CTSAs provide. To address this gap, the CTSA Dissemination, Implementation, Knowledge Transfer Working Group conducted an environmental scan of (2017–2018). Methods: Of 67 institutions, we contacted 43 that previously reported delivering services. experts from these institutions were emailed a survey assessing resources, services, training, scientific projects. Responses categorized double-coded by study authors using content analysis approach. Results: Thirty-five (81.4%) responded. Challenges in developing supporting science activities related inadequate workforce (45.7%) lack understanding (25.7%). Services provided included consultation/mentoring programs (68%), pilot funding/grants (50%), workshops/seminars/conferences (46%). Training development frequently identified as future priorities. Recommendations increase meet demand (68.6%), accessible tools/resources (34.3%), greater visibility/awareness methods consultation (22.9%), expand (22.9%). Conclusions: have tremendous potential advancement impact across translational continuum. Despite growing presence CTSAs, continued commitment prioritization needed institutional leadership raise awareness its value, demands, develop necessary infrastructure for conducting science.

Language: Английский

Citations

8
Using a Community-Informed Translational Model to Prioritize Translational Benefits in Youth Concussion Return-to-Learn Programs DOI
Julian Takagi‐Stewart, Aspen Avery, Shyam J. Deshpande

et al.

Health Promotion Practice, Journal Year: 2023, Volume and Issue: 25(3), P. 383 - 390

Published: Jan. 26, 2023

The Translational Science Benefit Model (TSBM) was developed to broadly capture systematic measures of health and societal benefits from scientific research, beyond traditional outcome measures. We aimed develop a process for the application TSBM then provide an example novel ongoing Return-to-Learn (RTL) after youth concussion project involving partnerships with community stakeholders.

Language: Английский

Citations

3
Focus Group Study of Medical Stakeholders to Inform the Development of Resilient Together for Dementia: Protocol for a Postdiagnosis Live Video Dyadic Resiliency Intervention DOI Creative Commons
Sarah Bannon, Julie Brewer, Talea Cornelius

et al.

JMIR Research Protocols, Journal Year: 2023, Volume and Issue: 12, P. e45533 - e45533

Published: Feb. 21, 2023

Alzheimer disease and related dementias (ADRD) are increasingly common conditions that disrupt the lives of persons living with dementia their spousal care partners. At time ADRD diagnoses, many couples experience challenges produce emotional distress relationship strain. present, there no interventions to address these early after diagnoses promote positive adjustment.

Language: Английский

Citations

2
Feasibility Meets Implementation Science: Narrowing the Research-To-Practice Gap for Exercise Activity in Multiple Sclerosis DOI Creative Commons
James Smith, Onno van der Groen, Yvonne C. Learmonth

et al.

International Journal of Qualitative Methods, Journal Year: 2023, Volume and Issue: 22

Published: June 9, 2023

Background There is a need to identify why multiple sclerosis exercise research not translating into real-world participation. To lay the foundations of strong clinical research, considering translational element implementation science at feasibility phase trial vital. Methods Document analysis was used examine document sources on activity interventions designed for people living with sclerosis. focused that incorporated prescription elements and behaviour change were studies incorporating aspects science. Results Implementation should come much earlier than efficacy or effectiveness pipeline. An alternate view outlined where meet based case examples have yet shown effectiveness. Findings from our key themes indicate cyclical iterative approach process. Multiple how it can be assessed using an lens support more successful are provided. The determination in involve as drawn theory development, optimising intervention design quality strategies, identifying those delivering before conducting research. Conclusions methodology underused qualitative appropriate use very resource, time-efficient unobtrusive process could track development explore integration phase, findings indicating introduced better.

Language: Английский

Citations

2
Synoptic Operative Reports: Can Form Follow Function in Surgery? DOI Open Access
Sharon S. Lum,

Halley Vora

Annals of Surgical Oncology, Journal Year: 2022, Volume and Issue: 29(11), P. 6515 - 6517

Published: April 5, 2022

Language: Английский

Citations

4
Use of implementation mapping in the planning of a hybrid type 1 pragmatic clinical trial: the BeatPain Utah study DOI Creative Commons
Julie M. Fritz, Bryan Gibson, David W. Wetter

et al.

Implementation Science Communications, Journal Year: 2024, Volume and Issue: 5(1)

Published: Jan. 5, 2024

Abstract Background Considerable disparities in chronic pain management have been identified. Persons rural, lower income, and minoritized communities are less likely to receive evidence-based, nonpharmacologic care. Telehealth delivery of nonpharmacologic, evidence-based interventions for persons with is a promising strategy lessen disparities, but implementation comes many challenges. The BeatPain Utah study hybrid type 1 effectiveness-implementation pragmatic clinical trial investigating telehealth strategies provide care from physical therapists back receiving ommunity health centers (CHCs). CHCs primary all regardless ability pay. This paper outlines the use mapping develop multifaceted plan study. Methods During planning year trial, we developed comprehensive logic model including five-step process informed by additional frameworks theories. five iterative steps were addressed year: (1) conduct needs assessments involved groups; (2) identify outcomes, performance objectives, determinants; (3) select strategies; (4) produce protocols materials; (5) evaluate outcomes. Results CHC leadership/providers, patients, identified as groups. Barriers assets across groups which identification objectives necessary implement two key processes: electronic referral patients clinics team connecting providing telehealth. Determinants each group our choice focused on training, education, clinician support, tailoring therapy cultural competency. We selected outcomes success strategies. Conclusions Implementation provided systematic approach an during phase ongoing trial. will be able used inform future efforts other settings. Trial registration ClinicalTrials.gov Identifier: NCT04923334 . Registered June 11, 2021.

Language: Английский

Citations

0