Gender Disparities in Electronic Health Record Usage and Inbasket Burden for Internal Medicine Residents.

BACKGROUND: Studies have demonstrated patients hold different expectations for female physicians compared to male physicians, including higher expectations for patient-centered communication and addressing socioeconomic or emotional needs. Recent evidence indicates this gender disparity extends to the electronic health record (EHR). Similar studies have not been conducted with resident physicians. OBJECTIVE: This study seeks to characterize differences in EHR workload for female resident physicians compared to male resident physicians. DESIGN: This study evaluated 12 months of 156 Mayo Clinic internal medicine residents' inbasket data from July 2020 to June 2021 using Epic's Signal and Physician Efficiency Profile (PEP) data. Excel, BlueSky Statistics, and SAS analytical software were used for analysis. Paired t-tests and analysis of variance were used to compare PEP data by gender and postgraduate year (PGY). "Male" and "female" were used in substitute for "gender" as is precedent in the literature. SUBJECTS: Mayo Clinic internal medicine residents. MAIN MEASURES: Total time spent in EHR per day; time in inbasket and notes per day; time in notes per appointment; number of patient advice requests made through the portal; message turnaround time. KEY RESULTS: Female residents received more patient advice requests per year (p = 0.004) with an average of 86.7 compared to 68, resulting in 34% more patient advice requests per day worked (p < 0.001). Female residents spent more time in inbasket per day (p = 0.002), in notes per day (p < 0.001), and in notes per appointment (p = 0.001). Resident panel comparisons revealed equivocal sizes with significantly more female patients on female (n = 55) vs male (n = 34) resident panels (p < 0.001). There was no difference in message turnaround time, total messages, or number of results received. CONCLUSIONS: Female resident physicians experience significantly more patient-initiated messages and EHR workload despite equivalent number of results and panel size. Gender differences in inbasket burden may disproportionally impact the resident educational experience.

Capture of real-time data from electronic health records: scenarios and solutions.

Background: The integration of real-time data (RTD) in the electronic health records (EHRs) is transforming the healthcare of tomorrow. In this work, the common scenarios of capturing RTD in the healthcare from EHRs are studied and the approaches and tools to implement real-time solutions are investigated. Methods: Delivering RTD by representational state transfer (REST) application programming interfaces (APIs) is usually accomplished through a Publish-Subscribe approach. Common technologies and protocols used for implementing subscriptions are REST hooks and WebSockets. Polling is a straightforward mechanism for obtaining updates; nevertheless, it may not be the most efficient or scalable solution. In such cases, other approaches are often preferred. Database triggers and reverse proxies can be useful in RTD scenarios; however, they should be designed carefully to avoid performance bottlenecks and potential issues. Results: The implementation of subscriptions through REST hooks and WebSocket notifications using a Fast Healthcare Interoperability Resources (FHIR) REST API, as well as the design of a reverse proxy and database triggers is described. Reference implementations of the solutions are provided in a GitHub repository. For the reverse proxy implementation, the Go language (Golang) was used, which is specialized for the development of server-side networking applications. For FHIR servers a python script is provided to create a sample Subscription resource to send RTD when a new Observation resource for specific patient id is created. The sample WebSocket client is written using the "websocket-client" python library. The sample RTD endpoint is created using the "Flask" framework. For database triggers a sample structured query language (SQL) query for Postgres to create a trigger when an INSERT or UPDATE operation is executed on the FHIR resource table is available. Furthermore, a use case clinical example, where the main actors are the healthcare providers (hospitals, physician private practices, general practitioners and medical laboratories), health information networks and the patient are drawn. The RTD flow and exchange is shown in detail and how it could improve healthcare. Conclusions: Capturing RTD is undoubtedly vital for health professionals and successful digital healthcare. The topic remains unexplored especially in the context of EHRs. In our work for the first time the common scenarios and problems are investigated. Furthermore, solutions and reference implementations are provided which could support and contribute to the development of real-time applications.

The Potential for Information Sharing Between an Electronic Medical Record System and the Moroccan Transplantation and Dialysis Registry: Examining Semantic Interoperability.

Background and aim In 2005, the Moroccan Ministry of Health established Magredial, a registry to track and monitor patients with end-stage renal disease (ESRD), with the aim of improving healthcare outcomes. After achieving initial success, Magredial's activity decreased, leading to its inactivity by 2015. Currently, efforts are underway to revive Magredial's use. The main goal of this study is to investigate the feasibility of data transfer between the electronic medical records (EMRs) of Hassan II Hospital of Fes, Morocco, and the registry by achieving semantic interoperability between the two systems Materials and methods The initial phase of this study involved a detailed review of existing literature, highlighting the importance of registries, especially in nephrology. This part of the study also aims to emphasize the role of semantic interoperability in facilitating the sharing of data between EMRs and registries. Following that, the study's second phase, which centered on the case study, conducted a detailed analysis of the data architectures in both Magredial and the EMR of the nephrology department to pinpoint areas of alignment and discrepancy. This step required cooperative efforts between the nephrology and IT departments of Hassan II Hospital. Results Our findings indicate a significant interoperability gap between the two systems, stemming from differences in their data architectures and semantic frameworks. Such discrepancies severely impede the effective exchange of information between the systems. To address this challenge, a comprehensive restructuring of the EMR is proposed. This strategy is designed to align disparate systems and ensure compliance with the interoperability standards the Health Level 7 Clinical Document Architecture (HL7-CDA) set forth. Implementing the proposed medical record approach is complex and time-consuming, necessitating healthcare professional commitment, and adherence to ethical standards for patient consent and data privacy. Conclusions Implementing this strategy is expected to facilitate the seamless automation of data transfer between the EMR and Magredial. It introduces a framework that could be a foundational model for establishing a robust interoperability framework within nephrology information systems in line with international standards. Ultimately, this initiative could lead to creating a nephrologist-shared health record across the country, enhancing patient care and data management within the specialty.

Charting Tomorrow's Healthcare: A Traditional Literature Review for an Artificial Intelligence-Driven Future.

Electronic health record (EHR) systems have developed over time in parallel with general advancements in mainstream technology. As artificially intelligent (AI) systems rapidly impact multiple societal sectors, it has become apparent that medicine is not immune from the influences of this powerful technology. Particularly appealing is how AI may aid in improving healthcare efficiency with note-writing automation. This literature review explores the current state of EHR technologies in healthcare, specifically focusing on possibilities for addressing EHR challenges through the automation of dictation and note-writing processes with AI integration. This review offers a broad understanding of existing capabilities and potential advancements, emphasizing innovations such as voice-to-text dictation, wearable devices, and AI-assisted procedure note dictation. The primary objective is to provide researchers with valuable insights, enabling them to generate new technologies and advancements within the healthcare landscape. By exploring the benefits, challenges, and future of AI integration, this review encourages the development of innovative solutions, with the goal of enhancing patient care and healthcare delivery efficiency.

Implementation of an electronic health record system during global surgical outreach: A prospective evaluation of success and sustainability.

BACKGROUND: The burden of musculoskeletal conditions continues to grow in low- and middle-income countries. Among thousands of surgical outreach trips each year, few organizations electronically track patient data to inform real-time care decisions and assess trip impact. We report the implementation of an electronic health record (EHR) system utilized at point of care during an orthopedic surgical outreach trip. METHODS: In March 2023, we implemented an EHR on an orthopedic outreach trip to guide real-time care decisions. We utilized an effectiveness-implementation hybrid type 3 design to evaluate implementation success. Success was measured using outcomes adopted by the World Health Organization, including acceptability, appropriateness, feasibility, adoption, fidelity, and sustainability. Clinical outcome measures included adherence to essential quality measures and follow-up numerical rating system (NRS) pain scores. RESULTS: During the 5-day outreach trip, 76 patients were evaluated, 25 of which underwent surgery beforehand. The EHR implementation was successful as defined by: mean questionnaire ratings of acceptability (4.26), appropriateness (4.12), feasibility (4.19), and adoption (4.33) at least 4.00, WHO behaviorally anchored rating scale ratings of fidelity (6.8) at least 5.00, and sustainability (80%) at least 60% follow-up at 6 months. All clinical quality measures were reported in greater than 80% of cases with all measures reported in 92% of cases. NRS pain scores improved by an average of 2.4 points. CONCLUSIONS: We demonstrate successful implementation of an EHR for real-time clinical use on a surgical outreach trip. Benefits of EHR utilization on surgical outreach trips may include improved documentation, minimization of medical errors, and ultimately improved quality of care.

Development and Implementation of an Integrated Standard e-Prescription Model in Alignment with Iranian National EHR.

The implementation of an Electronic Prescribing (EP) system offers numerous advantages in enhancing the efficiency of prescribing practices. To ensure successful implementation, a comprehensive understanding of the workflow in paper-based prescribing is crucial. In Iran, the Ministry of Health, and Medical Education (MOHME) has been actively involved in developing an EP system since 2011. The pilot results within MOHME have garnered significant support from all basic insurance organizations, primarily due to the importance of addressing financial considerations. As a result, these insurance organizations have taken the lead in the national development of the EP system, as responsibilities have shifted. The development of an Integrated Care Electronic Health Record (ICEHR or EHR) and the approach adopted by MOHME have paved the way for the creation of a standardized set of Application Programming Interfaces (APIs) based on openEHR and ISO13606 standards. These APIs facilitate the secure transfer of consolidated data from the EP systems, stored in the data warehouses of basic insurance organizations, to the Iranian EHR. This model follows an ICEHR architecture that emphasizes the transmission of this information to the Iranian EHR. This paper provides a detailed discussion of the various aspects and accomplishments related to these developments.

Implementation of the electronic health record in the German healthcare system: an assessment of the current status and future development perspectives considering the potentials of health data utilisation by representatives of different stakeholder groups.

Introduction: The digitalisation of the German healthcare system enables a wide range of opportunities to utilize healthcare data. The implementation of the EHR in January 2021 was a significant step, but compared to other European countries, the implementation of the EHR in the German healthcare system is still at an early stage. The aim of this paper is to characterise the structural factors relating to the adoption of the EHR in more detail from the perspective of representatives of stakeholders working in the German healthcare system and to identify existing barriers to implementation and the need for change. Methods: Qualitative expert interviews were conducted with one representative from each of the stakeholder groups health insurance, pharmacies, healthcare research, EHR development and panel doctors. Results: The interviews with the various stakeholders revealed that the implementation process of the EHR is being delayed by a lack of a viable basis for decision-making, existing conflicts of interest and insufficient consideration of the needs of patients and service providers, among other things. Discussion: The current status of EHR implementation is due to deficiency in legal regulations as well as structural problems and the timing of the introduction. For instance, the access rights of various stakeholders to the EHR data and the procedure in the event of a technical failure of the telematics infrastructure are remain unclear. In addition, insufficient information and communication measures have not led to the desired acceptance of EHR use among patients and service providers.

Quantitative evaluation of the impact of relaxing eligibility criteria on the risk-benefit profile of drugs for lung cancer based on real-world data.

INTRODUCTION: Restrictive eligibility criteria in cancer drug trials result in low enrollment rates and limited population diversity. Relaxed eligibility criteria (REC) based on solid evidence is becoming necessary for stakeholders worldwide. However, the absence of high-quality, favorable evidence remains a major challenge. This study presents a protocol to quantitatively evaluate the impact of relaxing eligibility criteria in common non-small cell lung cancer (NSCLC) protocols in China, on the risk-benefit profile. This involves a detailed explanation of the rationale, framework, and design of REC. METHODS: To evaluate our REC in NSCLC drug trials, we will first construct a structured, cross-dimensional real-world NSCLC database using deep learning methods. We will then establish randomized virtual cohorts and perform benefit-risk assessment using Monte Carlo simulation and propensity matching. Shapley value will be utilized to quantitatively measure the effect of the change of each eligibility criterion on patient volume, clinical efficacy and safety. DISCUSSION: This study is one of the few that focuses on the problem of overly stringent eligibility criteria cancer drug clinical trials, providing quantitative evaluation of the effect of relaxing each NSCLC eligibility criterion. This study will not only provide scientific evidence for the rational design of population inclusion in lung cancer clinical trials, but also establish a data governance system, as well as a REC evaluation framework that can be generalized to other cancer studies.

Decentralizing Health Care: History and Opportunities of Web3.

This paper explores the relationship between the development of the internet and health care, highlighting their parallel growth and mutual influence. It delves into the transition from the early, static days of Web 1.0, akin to siloed physician expertise in health care, to the more interactive and patient-centric era of Web 2.0, which was accompanied by advancements in medical technologies and patient engagement. This paper then focuses on the emerging era of Web3-the decentralized web-which promises a transformative shift in health care, particularly in how patient data are managed, accessed, and used. This shift toward Web3 involves using blockchain technology for decentralized data storage to enhance patient data access, control, privacy, and value. This paper also examines current applications and pilot projects demonstrating Web3's practical use in health care and discusses key questions and considerations for its successful implementation.

Strengths, Weaknesses, Opportunities, and Threats (SWOT) Analysis of Hemodialysis Electronic Health Record Implementation.

BACKGROUND AND AIM: The Nephrology Department of Hassan II Hospital in Fez, Morocco, has implemented an Electronic Medical Record (EMR) system for managing patients undergoing acute hemodialysis. This initiative aims to digitize patient monitoring and enhance the management of acute dialysis within the department. Conducting strengths, weaknesses, opportunities, and threats (SWOT) analysis - assessing strengths, weaknesses, opportunities, and threats - was crucial to identifying and understanding the internal strengths and weaknesses, as well as the external opportunities and threats. This article outlines the SWOT analysis findings that may impact the project's success and shape decision-making. It also discusses strategies that could be implemented to allocate resources, mitigate risks, and capitalize on potential advantages. MATERIALS AND METHODS: This study involved a multidisciplinary team, including professors, nephrologists, nephrology residents, and a healthcare information system engineer. Brainstorming sessions were held during the specification drafting phase to pinpoint both internal and external factors affecting the project. User feedback during testing further refined these factors, ensuring the project's alignment with real-world needs and challenges. RESULTS: The study identifies the project's strengths as providing safe and immediate access to information, along with strong communication between the department (application users) and the project manager. The significant EMR weakness is the lack of logistical resources and the absence of a long-term maintenance plan for the application. The opportunity presented by this EMR implementation is its functionality's potential to evolve, enabling the solution to be deployed in other dialysis centers across the region. The project's threat is the potential abandonment of EMR use by future practitioners. CONCLUSION: These SWOT analysis findings enable the development and implementation of strategies to reduce the current deployment's vulnerabilities and ensure the success of future HIS implementations in the nephrology network of the Fez-Meknes region, Morocco.

An interpretable predictive deep learning platform for pediatric metabolic diseases.

OBJECTIVES: Metabolic disease in children is increasing worldwide and predisposes a wide array of chronic comorbid conditions with severe impacts on quality of life. Tools for early detection are needed to promptly intervene to prevent or slow the development of these long-term complications. MATERIALS AND METHODS: No clinically available tools are currently in widespread use that can predict the onset of metabolic diseases in pediatric patients. Here, we use interpretable deep learning, leveraging longitudinal clinical measurements, demographical data, and diagnosis codes from electronic health record data from a large integrated health system to predict the onset of prediabetes, type 2 diabetes (T2D), and metabolic syndrome in pediatric cohorts. RESULTS: The cohort included 49 517 children with overweight or obesity aged 2-18 (54.9% male, 73% Caucasian), with a median follow-up time of 7.5 years and mean body mass index (BMI) percentile of 88.6%. Our model demonstrated area under receiver operating characteristic curve (AUC) accuracies up to 0.87, 0.79, and 0.79 for predicting T2D, metabolic syndrome, and prediabetes, respectively. Whereas most risk calculators use only recently available data, incorporating longitudinal data improved AUCs by 13.04%, 11.48%, and 11.67% for T2D, syndrome, and prediabetes, respectively, versus models using the most recent BMI (P < 2.2 × 10-16). DISCUSSION: Despite most risk calculators using only the most recent data, incorporating longitudinal data improved the model accuracies because utilizing trajectories provides a more comprehensive characterization of the patient's health history. Our interpretable model indicated that BMI trajectories were consistently identified as one of the most influential features for prediction, highlighting the advantages of incorporating longitudinal data when available.

Quality of care in patients with hypertension: a retrospective cohort study of primary care routine data in Germany.

BACKGROUND: Hypertension is a leading cause of morbidity and mortality if not properly managed. Primary care has a major impact on these outcomes if its strengths, such as continuity of care, are deployed wisely. The analysis aimed to evaluate the quality of care for newly diagnosed hypertension in routine primary care data. METHODS: In the retrospective cohort study, routine data (from 2016 to 2022) from eight primary care practices in Germany were exported in anonymized form directly from the electronic health record (EHR) systems and processed for this analysis. The analysis focused on five established quality indicators for the care of patients who have been recently diagnosed with hypertension. RESULTS: A total of 30,691 patients were treated in the participating practices, 2,507 of whom have recently been diagnosed with hypertension. Prior to the pandemic outbreak, 19% of hypertensive patients had blood pressure above 140/90 mmHg and 68% received drug therapy (n = 1,372). After the pandemic outbreak, the proportion of patients with measured blood pressure increased from 63 to 87%, while the other four indicators remained relatively stable. Up to 80% of the total variation of the quality indicators could be explained by individual practices. CONCLUSION: For the majority of patients, diagnostic procedures are not used to the extent recommended by guidelines. The analysis showed that quality indicators for outpatient care could be mapped onto the basis of routine data. The results could easily be reported to the practices in order to optimize the quality of care.

Security and Privacy of Online Record Access: A Survey of Adolescents' Views and Experiences in Sweden.

PURPOSE: Ensuring security of online health records and patients' perceptions of security are concerns in adolescent healthcare. Little is known about adolescents' perceptions about healthcare's ability to protect online health records. This article explores adolescents' perspectives on security and privacy of their online health records, potential differences based on gender and health, attitudes to sharing information, and perceptions of what constitutes sensitive information. METHODS: This study included a subset of items from a national online patient survey conducted in Sweden (January-February 2022), focusing on respondents aged 15-19 years. Gender and health status differences were calculated using the Kruskal-Wallis test. RESULTS: Of 218 adolescent respondents (77.1% female), a minority had security and privacy concerns. A notable proportion (41.3%) wished to control who could see their records, and those who reported better perceived health were more likely to want to manage access to their electronic health record (H = 13.569, p = .009). Most had not experienced unauthorized access to their records (75.2%) and had never shared health information on other online applications (85.8%). More than half (56.0%) perceived some information as sensitive, where mental health was the most common (76.0%). Most felt that reading their notes improved their trust for their healthcare professional (65.6%) and supported better communication with healthcare professionals (66.5%). DISCUSSION: In this national survey, adolescents generally reported few concerns about patient portals. Findings emphasize the need for security and privacy protection and to empower adolescents with greater control over access to their health information housed in electronic health record systems.

Risk factors for allergy documentation in electronic health record: A retrospective study in a tertiary health center in Switzerland.

BACKGROUND: Most hospitals use electronic health records (EHR) to warn health care professionals of drug hypersensitivity (DH) and other allergies. Indiscriminate recording of patient self-reported allergies may bloat the alert system, leading to unjustified avoidances and increases in health costs. The aim of our study was to analyze hypersensitivities documented in EHR of patients at Lausanne University Hospital (CHUV). METHODS: We conducted a retrospective study on patients admitted at least 24 h to CHUV between 2011 and 2021. After ethical clearance, we obtained anonymized data. Because culprit allergen could be either manually recorded or selected through a list, data was harmonized using a reference allergy database before undergoing statistical analysis. RESULTS: Of 192,444 patients, 16% had at least one allergy referenced. DH constituted 60% of all allergy alerts, mainly beta-lactam antibiotics (BLA) (30%), NSAID (11%) and iodinated contrast media (ICM) (7%). Median age at first hospitalization and hospitalization length were higher in the allergy group. Female to male ratio was 2:1 in the allergic group. Reactions were limited to the skin in half of patients, and consistent with anaphylaxis in 6%. In those deemed allergic to BLA, culprit drug was specified in 19%, 'allergy to penicillin' otherwise. It was impossible to distinguish DH based on history alone or resulting from specialized work-up. CONCLUSIONS: Older age, longer hospital stays, and female sex increase the odds of in-patient allergy documentation. Regarding DH, BLA were referenced in 4% of inpatient records. Specific delabeling programs should be implemented to increase data reliability and patient safety.

A pragmatic methodology to extract anesthetic and physiological data from the electronic health record.

BACKGROUND/AIMS: Traditional manual methods of extracting anesthetic and physiological data from the electronic health record rely upon visual transcription by a human analyst that can be labor-intensive and prone to error. Technical complexity, relative inexperience in computer coding, and decreased access to data warehouses can deter investigators from obtaining valuable electronic health record data for research studies, especially in under-resourced settings. We therefore aimed to develop, pilot, and demonstrate the effectiveness and utility of a pragmatic data extraction methodology. METHODS: Expired sevoflurane concentration data from the electronic health record transcribed by eye was compared to an intermediate preprocessing method in which the entire anesthetic flowsheet narrative report was selected, copy-pasted, and processed using only Microsoft Word and Excel software to generate a comma-delimited (.csv) file. A step-by-step presentation of this method is presented. Concordance rates, Pearson correlation coefficients, and scatterplots with lines of best fit were used to compare the two methods of data extraction. RESULTS: A total of 1132 datapoints across eight subjects were analyzed, accounting for 18.9 h of anesthesia time. There was a high concordance rate of data extracted using the two methods (median concordance rate 100% range [96%, 100%]). The median time required to complete manual data extraction was significantly longer compared to the time required using the intermediate method (240 IQR [199, 482.5] seconds vs 92.5 IQR [69, 99] seconds, p = .01) and was linearly associated with the number of datapoints (rmanual = .97, p < .0001), whereas time required to complete data extraction using the intermediate approach was independent of the number of datapoints (rintermediate = -.02, p = .99). CONCLUSIONS: We describe a pragmatic data extraction methodology that does not require additional software or coding skills intended to enhance the ease, speed, and accuracy of data collection that could assist in clinician investigator-initiated research and quality/process improvement projects.

The Iowa Health Data Resource (IHDR): an innovative framework for transforming the clinical health data ecosystem.

IMPORTANCE: This manuscript will be of interest to most Clinical and Translational Science Awards (CTSA) as they retool for the increasing emphasis on translational science from translational research. This effort is an extension of the EDW4R work that most CTSAs have done to deploy infrastructure and tools for researchers to access clinical data. OBJECTIVES: The Iowa Health Data Resource (IHDR) is a strategic investment made by the University of Iowa to improve access to real-world health data. The goals of IHDR are to improve the speed of translational health research, to boost interdisciplinary collaboration, and to improve literacy about health data. The first objective toward this larger goal was to address gaps in data access, data literacy, lack of computational environments for processing Personal Health Information (PHI) and the lack of processes and expertise for creating transformative datasets. METHODS: A three-pronged approach was taken to address the objective. The approach involves integration of an intercollegiate team of non-informatics faculty and staff, a data enclave for secure patient data analyses, and novel comprehensive datasets. RESULTS: To date, all five of the health science colleges (dentistry, medicine, nursing, pharmacy, and public health) have had at least one staff and one faculty member complete the two-month experiential learning curriculum. Over the first two years of this project, nine cohorts totaling 36 data liaisons have been trained, including 18 faculty and 18 staff. IHDR data enclave eliminated the need to duplicate computational infrastructure inside the hospital firewall which reduced infrastructure, hardware and human resource costs while leveraging the existing expertise embedded in the university research computing team. The creation of a process to develop and implement transformative datasets has resulted in the creation of seven domain specific datasets to date. CONCLUSION: The combination of people, process, and technology facilitates collaboration and interdisciplinary research in a secure environment using curated data sets. While other organizations have implemented individual components to address EDW4R operational demands, the IHDR combines multiple resources into a novel, comprehensive ecosystem IHDR enables scientists to use analysis tools with electronic patient data to accelerate time to science.

Introduction to School Medicaid for School Nurses.

School nurses are important participants in School Medicaid (SM) programs nationally. Yet, the complexity of SM programs and required documentation are major barriers to implementing this program for nurses. School nurses are often required to participate in the SM program without having a clear understanding of the purpose and components of the overall program. With the expansion of SM programs due to the Centers for Medicare and Medicaid Services "free care" guidance change, nursing services and documentation are receiving more attention and scrutiny as more nursing services become eligible for reimbursement in states across the country. This article presents a clear overview of SM, its history, different components of state SM programs, school nursing documentation and challenges, and resources for school nurses who are interested in more information.

GRU-D-Weibull: A novel real-time individualized endpoint prediction.

BACKGROUND: In the era of healthcare digital transformation, using electronic health record (EHR) data to generate various endpoint estimates for active monitoring is highly desirable in chronic disease management. However, traditional predictive modeling strategies leveraging well-curated data sets can have limited real-world implementation potential due to various data quality issues in EHR data. METHODS: We propose a novel predictive modeling approach, GRU-D-Weibull, which models Weibull distribution leveraging gated recurrent units with decay (GRU-D), for real-time individualized endpoint prediction and population level risk management using EHR data. EXPERIMENTS: We systematically evaluated the performance and showcased the real-world implementability of the proposed approach through individual level endpoint prediction using a cohort of patients with chronic kidney disease stage 4 (CKD4). A total of 536 features including ICD/CPT codes, medications, lab tests, vital measurements, and demographics were retrieved for 6879 CKD4 patients. The performance metrics including C-index, L1-loss, Parkes' error, and predicted survival probability at time of event were compared between GRU-D-Weibull and other alternative approaches including accelerated failure time model (AFT), XGBoost based AFT (XGB(AFT)), random survival forest (RSF), and Nnet-survival. Both in-process and post-process calibrations were experimented on GRU-D-Weibull generated survival probabilities. RESULTS: GRU-D-Weibull demonstrated C-index of ~0.7 at index date, which increased to ~0.77 at 4.3 years of follow-up, comparable to that of RSF. GRU-D-Weibull achieved absolute L1-loss of ~1.1 years (sd≈0.95) at CKD4 index date, and a minimum of ~0.45 year (sd≈0.3) at 4 years of follow-up, comparing to second-ranked RSF of ~1.4 years (sd≈1.1) at index date and ~0.64 years (sd≈0.26) at 4 years. Both significantly outperform competing approaches. GRU-D-Weibull constrained predicted survival probability at time of event to smaller and more fixed range than competing models throughout follow-up. Significant correlations were observed between prediction error and missing proportions of all major categories of input features at index date (Corr ~0.1 to ~0.3), which faded away within 1 year after index date as more data became available. Through post training recalibration, we achieved a close alignment between the predicted and observed survival probabilities across multiple prediction horizons at different time points during follow-up. CONCLUSION: GRU-D-Weibull shows advantages over competing methods in handling missingness commonly encountered in EHR data and providing both probability and point estimates for diverse prediction horizons during follow-up. The experiment highlights the potential of GRU-D-Weibull as a suitable candidate for individualized endpoint risk management, utilizing real-time clinical data to generate various endpoint estimates for monitoring. Additional research is warranted to evaluate the influence of different data quality aspects on prediction performance. Furthermore, collaboration with clinicians is essential to explore the integration of this approach into clinical workflows and evaluate its effects on decision-making processes and patient outcomes.

Derivation of paediatric blood pressure percentiles from electronic health records.

BACKGROUND: Identification of abnormal blood pressure (BP) in children requires normative data. We sought to examine the feasibility of using "real-world" office BP data obtained from electronic health records (EHR) to generate age-, sex- and height-specific BP percentiles for children. METHODS: Using data collected 01/01/2009-8/31/2021 from eight large children's healthcare organisations in PEDSnet, we applied a mixed-effects polynomial regression model with random slopes to generate Z-scores and BP percentiles and compared them with currently used normative BP distributions published in the 2017 American Academy of Paediatrics (AAP) Clinical Practise Guidelines (CPG). FINDINGS: We identified a study sample of 292,412 children (1,085,083 BP measurements), ages 3-17 years (53% female), with no chronic medical conditions, who were not overweight/obese and who were primarily seen for general paediatric care in outpatient settings. Approximately 45,000-75,000 children contributed data to each age category. The PEDSnet systolic BP percentile values were 1-4 mmHg higher than AAP CPG BP values across age-sex-height groups, with larger differences observed in younger children. Diastolic BP values were also higher in younger children; starting with age 7 years, diastolic BP percentile values were 1-3 mmHg lower than AAP CPG values. Cohen's Kappa was 0.90 for systolic BP, 0.66 for diastolic BP, and 0.80 overall indicating excellent agreement between PEDSnet and 2017 AAP CPG data for systolic BP and substantial agreement for diastolic BP. INTERPRETATION: Our analysis indicates that real-word EHR data can be used to generate BP percentiles consistent with current clinical practise on BP management in children. FUNDING: Funding for this work was provided by the Preserving Kidney Function in Children with Chronic Kidney Disease (PRESERVE) study; Patient-Centred Outcomes Research Institute (PCORI) RD-2020C2020338 (Principal Investigator: Dr. Forrest; Co-Principal Investigator: Dr. Denburg).