Applied Clinical Informatics最新文献

Background: The use of Electronic Health Records (EHRs) in research demands robust, interoperable systems. By linking biorepositories to EHR algorithms, researchers can efficiently identify cases and controls for large observational studies (e.g., Genome-Wide Association Studies (GWAS)). This is critical for ensuring efficient and cost-effective research. However, the lack of standardized metadata and algorithms across different EHRs complicates their sharing and application. Our study presents an example of a successful implementation and validation process.

Objective: To implement and validate a rule-based algorithm from a tertiary medical center in Tennessee to classify cases and controls from a research study on rotator cuff tear nested within a tertiary medical center in North Texas and to assess the algorithm's performance.

Methods: We applied a phenotypic algorithm (designed and validated in a tertiary medical center in Tennessee) using EHR data from 492 patients enrolled in case-control study recruited from a tertiary medical center in North Texas. The algorithm leveraged ICD (International Classification of Diseases) and CPT (Current Procedural Terminology) codes to identify case and control status for degenerative rotator cuff tears. A manual review was conducted to compare the algorithm's classification with a previously recorded gold standard documented by clinical researchers.

Results: Initially the algorithm identified 398 (80.9%) patients correctly as cases or controls. After fine-tunning and corrections of errors in our gold standard dataset, we calculated a sensitivity of 0.94 and specificity of 0.76.

Discussion: The implementation of the algorithm presented challenges due to the variability in coding practices between medical centers. To enhance performance, we refined the algorithm's data dictionary by incorporating additional codes. The process highlighted the need for meticulous code verification and standardization in multi-center studies.

Conclusion: Sharing case-control algorithms boosts EHR research. Our rule-based algorithm improved multi-site patient identification and revealed 12 data entry errors, helping validate our results.

Background: Global efforts aimed at ending human immunodeficiency virus (HIV) incidence have adapted and evolved since the turn of the century. The utilization of machine learning incorporated into an electronic health record (EHR) can be refined into prediction models that identify when an individual is at greater HIV infection risk. This can create a novel and innovative approach to identifying patients eligible for preventative therapy.

Objectives: This study's aim was to evaluate the effectiveness of an HIV prediction model in clinical workflows. Outcomes included pre-exposure prophylaxis (PrEP) prescriptions generated and the model's ability to identify eligible patients.

Methods: A prediction model was developed and implemented at the safety-net hospital in Dallas County. Patients seen in primary care clinics were evaluated between July 2020 to June 2022. The prediction model was incorporated into an existing best practice advisory (BPAs) used to identify potentially eligible PrEP patients. The prior, basic BPA (bBPA) displayed if a prior sexually transmitted infection was documented and the enhanced BPA (eBPA) incorporated the HIV prediction model.

Results: A total of 3,218 unique patients received the BPA during the study time period, with 2,346 ultimately included for evaluation. There were 678 patients in the bBPA group and 1,666 in the eBPA group. PrEP prescriptions generated increased in the post-implementation group within the 90-day follow-up period (bBPA:1.48 v. eBPA:3.67 prescriptions per month, p<0.001). Patient demographics also differed between groups, resulting in a higher median age (bBPA:36[IQR 24] v. eBPA:52[QR 19] years, p<0.001) and an even distribution between birth sex in the post-implementation group (female sex at birth bBPA:62.2% v. eBPA:50.2%, p=<0.001).

Conclusions: The implementation of a HIV prediction model yielded a higher number of PrEP prescriptions generated and was associated with the identification of twice the number of potentially eligible patients.

Objective: To review pediatric artificial intelligence (AI) implementation studies from 2010-2021 and analyze reported performance measures.

Methods: We searched PubMed/Medline, Embase CINHAL, Cochrane Library CENTRAL, IEEE and Web of Science with controlled vocabulary.

Inclusion criteria: AI intervention in a pediatric clinical setting that learns from data (i.e., data-driven, as opposed to rule-based) and takes actions to make patient-specific recommendations; published between 01/2010 to 10/2021; must have agency (AI must provide guidance that affects clinical care, not merely running in background). We extracted study characteristics, target users, implementation setting, time span, and performance measures.

Results: Of 126 articles reviewed as full text, 17 met inclusion criteria. Eight studies (47%) reported both clinical outcomes and process measures, six (35%) reported only process measures, and two (12%) reported only clinical outcomes. Five studies (30%) reported no difference in clinical outcomes with AI, four (24%) reported improvement in clinical outcomes compared to controls, two (12%) reported positive effects on clinical outcomes with use of AI but had no formal comparison or controls, and one (6%) reported poor clinical outcomes with AI. Twelve studies (71%) reported improvement in process measures, while two (12%) reported no improvement. Five (30%) studies reported on at least 1 human performance measure.

Conclusions: While there are many published pediatric AI models, the number of AI implementations is minimal with no standardized reporting of outcomes, care processes, or human performance measures. More comprehensive evaluations will help elucidate mechanisms of impact.

Background: Cerebral aneurysms (CAs) affect 3-5% of the general population, with saccular aneurysms being the most common type. Despite advances in treatment, patient understanding of CAs and associated procedures remains limited, impacting informed consent and treatment outcomes.

Objectives: This pilot study aims to evaluate the effectiveness of mixed reality (MR) technology in enhancing patient education and understanding of cerebral angiograms and aneurysm treatment, thereby improving the patient-surgeon communication process.

Methods: A non-randomized single-center prospective study was conducted with 16 patients diagnosed with intracranial aneurysms. Participants used a Microsoft HoloLens to view an interactive 3D presentation about cerebral angiograms and aneurysm treatments. Pre- and post-intervention surveys assessed their knowledge and anxiety levels using a 5-point Likert scale. The Wilcoxon signed-rank test was used for statistical analysis.

Results: Post-intervention, the total survey scores improved significantly (average increase of 6.7 points, p<0.05). Seven out of eight survey questions showed significant knowledge improvement. The mean perceived ability to explain aneurysm treatment improved by 1.38 points and understanding of access points for procedures increased by 1.31 points (both p<0.05). The question regarding understanding of treatment risks did not show significant change (p>0.05). Anxiety levels decreased, with 75% of participants reporting reduced anxiety post-intervention.

Conclusions: MR technology significantly enhances patient understanding and reduces anxiety regarding cerebral angiogram procedures and aneurysm treatments. These findings support the integration of MR in patient education to improve clinical outcomes and patient satisfaction. This approach offers a promising direction for future healthcare communication strategies, especially in complex procedures requiring detailed patient comprehension.

Objective To develop a measure of Health Information Exchange (HIE) for characterizing the density of inter-hospital HIE connections and identify regional characteristics associated with HIE network density Materials and Methods HIE network density was measured as the proportion of hospital pairs within a region that are connected through HIE. The 2022 American Hospital Association's Information Technology Supplement survey was used to calculate HIE network density for US hospital referral regions (HRRs). Bivariate tests and multivariable regression were used to characterize hospital, electronic health record (EHR) vendor, and resident characteristics associated with HIE network density. Results Data on 2,509 hospitals across 274 HRRs were included in the study, with 92% of hospitals participating in at least 1 HIE. On average, hospitals participated in two HIEs and there were 7 HIEs present in each region. HIE network density ranged from 0.0 to 1.0, with a median of 0.78 and an interquartile range of 0.51 to 1.00. Hospital and vendor characteristics associated with greater HIE network density include: more HIEs per hospital, a higher proportion of non-profit hospitals, greater Epic marketshare, and more concentrated hospital and EHR vendor markets. Resident characteristics associated with greater HIE network density include: higher home values, more educated residents, and higher median household incomes. Conclusion We found that, on average, 7 out of 10 hospital-pairs within a given hospital referral regions are connected via at least one HIE, with lower HIE network density in regions with lower socioeconomic status. This measure can be used to track the impact of the Trusted Exchange Framework and Common Agreement on area-level interoperability.

Background: The "Cancer Risk Calculator" mobile application aims to inform patients about their personal risks of cancer and their risk factors influencingsaid risks. The present analysis examines the responses to a questionnaire submitted by oncology patients treated with radiotherapy or their family members.

Objective: The primary objective was to determine the effectof the app on the user's awareness and potential habit changes related to cancer risk. Further, the study aimed to discern any relationships between respondent characteristics and their questionnaire responses.

Methods: A total of 162 patients were included in the analysis. Each patient's dataset comprised gender, date of birth, entry date, respondent type, type of cancer, and responses to 12 application-related questions. Statistical methods such as multiple regression models were employed to identify any effects of the respondent's characteristics on their responses. Statistical significance was set at p<0.05.

Results: Responding to the survey questions, 67.1% of respondents found the application useful, and 63.4% reported learning something new. More than half (52.5%) indicated a willingness to change their habits based on the information provided. Respondents also indicated that they were surprised by the number of risk factors shaping their risks and the large influence of some of these risk factors. Variables such as breast cancer diagnosis (p=0.044) and age (p=0.049) influenced specific question responses.

Conclusions: The "Cancer Risk Calculator" app appears to have a significant utility in educating its users about cancer risk and potentially influencing habit change.

Background: Engagement of clinicians who understand clinical workflows and technology constraints can accelerate the development and implementation of better electronic health record (EHR) designs that improve quality and reduce burnout. Provider builder programs can accelerate clinical informatics education for a broader coalition of clinical specialties.

Objective: In this State of the Art / Best Practice paper, we aim to (1) propose a provider builder maturity model informed by the experience of three institutions using a single EHR vendor (Epic Systems©) and (2) describe the program elements and relationships necessary to advance along this model to yield organizational benefits.

Methods: We used a modified version of the Glaser State-of-the-Art approach, gathering consensus among a small group of experts at institutions with successful provider builder programs. The model was updated through meetings with a larger group of experts and then feedback from presentation at national conferences and the American Medical Informatics Association's Maturity Model Working Group.

Results: The final maturity model describes the characteristics and suggested next steps beginning from Planting the Seed (Stage 0) and progressing through Lone Wolves (Stage 1), a Community of Builders (Stage 2), Organizational Structure (Stage 3), a Council of Builders (Stage 4), and Informatics in the Room Where it Happens (Stage 5). We also describe the journeys of 3 organizations through these stages.

Conclusions: A provider builder maturity model can help guide organizations on their journey engaging clinicians in collaborative EHR design to promote quality and safety and reduce burnout.

Background Reliable, precise, timely, and clear documentation of diagnoses is difficult. Poor specificity or the absence of diagnostic documentation can lead to decreased revenue and increased payor denials, audits, and queries to providers. Nuance's Dragon Medical Advisor (DMA) is a computer-assisted physician documentation (CAPD) product. Natural language processing is used to present real-time advice on diagnostic specificity during documentation. Objectives This study assessed the feasibility, acceptability, and preliminary efficacy of real-time CAPD in improving diagnostic specificity and in turn reducing clinical documentation improvement burden. Methods This prospective, crossover trial recruited 18 hospitalists employed by Lifespan Health System and assigned them randomly to two groups. Each group first completed documentation using either traditional clinical documentation improvement (CDI) methods or CDI + DMA real-time advice for eight weeks and then crossed over. Metrics from Epic's EMR and Nuance administrative tools as well as anonymous surveys and one-on-one interviews were collected and analyzed. Results Hospitalists had 29% fewer standard CDI queries using DMA with CDI (IRR: 0.71; 95% CI: 0.37,1.39). Self-reported ability to predict clarification requests improved by 1 point on average (1.00; 95% CI: 0.32,1.67) on the Likert scale. This benefit was kept even after DMA was stopped and the group reverted back to CDI only. Qualitative survey reports indicated overall ease of use and educational benefits. Additional work needs to be done to determine if there is significant increase in note-writing time or reimbursement. Conclusions Hospitalists using DMA spent less time responding to In-basket queries. There was a strong educational opportunity, and the tool was easy to use. DMA offers promise for improving diagnostic specification while minimally impacting provider workflow.

Objectives:  This study aimed to (1) empirically investigate current practices and analyze ethical dimensions of clinical data sharing by health care organizations for uses other than treatment, payment, and operations; and (2) make recommendations to inform research and policy for health care organizations to protect patients' privacy and autonomy when sharing data with unrelated third parties.

Methods:  Semistructured interviews and surveys involving 24 informatics leaders from 22 U.S. health care organizations, accompanied by thematic and ethical analyses.

Results:  We found considerable heterogeneity across organizations in policies and practices. Respondents understood "data sharing" and "research" in very different ways. Their interpretations of these terms ranged from making data available for academic and public health uses, and to health information exchanges; to selling data for corporate research; and to contracting with aggregators for future resale or use. The nine interview themes were that health care organizations: (1) share clinical data with many types of organizations, (2) have a variety of motivations for sharing data, (3) do not make data-sharing policies readily available, (4) have widely varying data-sharing approval processes, (5) most commonly rely on Health Insurance and Portability and Accountability Act (HIPAA) de-identification to protect privacy, (6) were concerned about clinical data use by electronic health record vendors, (7) lacked data-sharing transparency to the general public, (8) allowed individual patients little control over sharing of their data, and (9) had not yet changed data-sharing practices within the year following the U.S. Supreme Court 2022 decision denying rights to abortion.

Conclusion:  Our analysis identified gaps between ethical principles and health care organizations' data-sharing policies and practices. To better align clinical data-sharing practices with patient expectations and biomedical ethical principles, we recommend updating HIPAA, including re-identification and upstream sharing restrictions in data-sharing contracts, better coordination across data-sharing approval processes, fuller transparency and opt-out options for patients, and accountability for data-sharing and consequent harms.

Background:  There is a need to integrate informatics education into medical training programs given the rise in demand for health informaticians and the call on the Accreditation Council for Graduate Medical Education and the body of undergraduate medical education for implementation of informatics curricula.

Objectives:  This report outlines a 2-year longitudinal informatics curriculum now currently in its seventh year of implementation. This report is intended to inform U.S. Graduate Medical Education (GME) program leaders of the necessary requirements for implementation of a similar program at their institution.

Methods:  The curriculum aligns with the core content for the subspecialty of clinical informatics (CI) and is led by a multidisciplinary team with both informatics and clinical expertise. This educational pathway has a low direct cost and is a practical example of the academic learning health system (aLHS) in action. The pathway is housed within an internal medicine department at a large tertiary academic medical center.

Results:  The curriculum has yielded 13 graduates from both internal medicine (11, 85%) and pediatrics (2, 15%) whose projects have spanned acute and ambulatory care and multiple specialties. Projects have included clinical decision support tools, of which some will be leveraged as substrate in applications seeking extramural funding. Graduates have gone on to CI board certification and fellowship, as well as several other specialties, creating a distributed network of clinicians with specialized experience in applied CI.

Conclusion:  An informatics curriculum at the GME level may increase matriculation to CI fellowship and more broadly increase development of the CI workforce through building a cadre of physicians with health information technology expertise across specialties without formal CI board certification. We offer an example of a longitudinal pathway, which is rooted in aLHS principles. The pathway requires a dedicated multidisciplinary team and departmental and information technology leadership support.