Electronic Health Record Access Research Articles

BCAS: Blockchain-based secure access and sharing scheme for EHR data

Sharing Electronic Health Record (EHR) data is critical for improving medical decision-making and emergency care. However, EHRs contain sensitive patient information, making data breaches and misuse a significant concern during sharing. To tackle this challenge, we propose the Blockchain-Based Secure Access and Sharing (BCAS) scheme, a patient-centered, secure, and efficient solution for large-scale EHR access and sharing that intelligently combines blockchain with proxy re-encryption. Four smart contracts are designed for the BCAS to achieve effective user identity registration, precise virtual identity verification, strict access control, and secure data management. The adoption of hybrid on-chain and off-chain storage structures, coupled with the four smart contracts, effectively optimizes the efficiency of the blockchain in BCAS. Proxy re-encryption further enhances data confidentiality and streamlines EHR sharing among authorized users. The security analysis indicates that BCAS can resist DDoS, spoofing, and man-in-the-middle attacks, ensuring the privacy and security of patient EHR data. The performance evaluation demonstrates that the CPU usage of smart contracts does not exceed 15%. The on-chain and off-chain storage structures require less than 30s for uploading and downloading an 800MB file. Additionally, the proxy re-encryption scheme can process a 10MB file in less than 43ms. These results indicate that BCAS offers efficient performance without compromising data security and privacy. Compared with existing blockchain-based EHR sharing systems, BCAS offers superior performance, a more comprehensive solution, and increased resistance to attacks.

PatCen: A blockchain-based patient-centric mechanism for the granular access control of infectious disease-related test records.

The recent global outbreaks of infectious diseases such as COVID-19, yellow fever, and Ebola have highlighted the critical need for robust health data management systems that can rapidly adapt to and mitigate public health emergencies. In contrast to traditional systems, this study introduces an innovative blockchain-based Electronic Health Record (EHR) access control mechanism that effectively safeguards patient data integrity and privacy. The proposed approach uniquely integrates granular data access control mechanism within a blockchain framework, ensuring that patient data is only accessible to explicitly authorized users and thereby enhancing patient consent and privacy. This system addresses key challenges in healthcare data management, including preventing unauthorized access and overcoming the inefficiencies inherent in traditional access mechanisms. Since the latency is a sensitive factor in healthcare data management, the simulations of the proposed model reveal substantial improvements over existing benchmarks in terms of reduced computing overhead, increased throughput, minimized latency, and strengthened overall security. By demonstrating these advantages, the study contributes significantly to the evolution of health data management, offering a scalable, secure solution that prioritizes patient autonomy and privacy in an increasingly digital healthcare landscape.

Reliable generation of privacy-preserving synthetic electronic health record time series via diffusion models.

Electronic health records (EHRs) are rich sources of patient-level data, offering valuable resources for medical data analysis. However, privacy concerns often restrict access to EHRs, hindering downstream analysis. Current EHR deidentification methods are flawed and can lead to potential privacy leakage. Additionally, existing publicly available EHR databases are limited, preventing the advancement of medical research using EHR. This study aims to overcome these challenges by generating realistic and privacy-preserving synthetic EHRs time series efficiently. We introduce a new method for generating diverse and realistic synthetic EHR time series data using denoizing diffusion probabilistic models. We conducted experiments on 6 databases: Medical Information Mart for Intensive Care III and IV, the eICU Collaborative Research Database (eICU), and non-EHR datasets on Stocks and Energy. We compared our proposed method with 8 existing methods. Our results demonstrate that our approach significantly outperforms all existing methods in terms of data fidelity while requiring less training effort. Additionally, data generated by our method yield a lower discriminative accuracy compared to other baseline methods, indicating the proposed method can generate data with less privacy risk. The proposed model utilizes a mixed diffusion process to generate realistic synthetic EHR samples that protect patient privacy. This method could be useful in tackling data availability issues in the field of healthcare by reducing barrier to EHR access and supporting research in machine learning for health. The proposed diffusion model-based method can reliably and efficiently generate synthetic EHR time series, which facilitates the downstream medical data analysis. Our numerical results show the superiority of the proposed method over all other existing methods.

Survey paper on Secured Data Retrieval from a distributed database using custom blockchain and Role Base Access Control.

These days, there has been a paradigm shift in the upkeep of Electronic Medical Records (EMRs) in mobile cloud environments. This involves integrating cloud computing with mobile devices to facilitate the transmission of healthcare information between patients and healthcare professionals. With the help of this innovative approach, healthcare may be provided with a high degree of flexibility, minimal overhead costs, and electronic health record access. This new strategy also raises concerns about the security of the networks that enable e-health systems and the privacy of patient data. Finding a solution to efficiently distribute EMRs across mobile users in mobile cloud settings while also upholding the highest security standards is a challenging task. In this paper, explored two different distributed blockchain and cloud frameworks for EMR sharing that combine blockchain technology with a mobile cloud's Decentralised Interplanetary File System (DIFS). Based on the empirical findings, it seems that the idea provides a practical way to protect critical patient data from possible threats and guarantee the dependability of data transfers on mobile clouds. There are notable performance improvements in areas like data privacy protections, minimal network latency, maximum security levels, and lightweight access control when comparing the results of this security analysis and performance assessment to those of earlier data sharing models.

Closing the Gap: Foster Parents' Lived Experiences and Recommendations for Interdisciplinary Health-Care Providers.

Purpose: This study explored barriers and facilitators to utilizing health-care services for foster youth, from the perspective of foster parents. Background: There are nearly 437,283 youths in the United States foster care system. Youths living in locations that are different from their familial home situations have a significantly higher occurrence of physical and mental health illnesses than children who remain in their familial homes, increasing risk for homelessness, substance use and abuse, chronic long-term illnesses, and incarceration, thus placing additional burden on the health-care system. Design and Methods: This study utilized a qualitative descriptive design with purposive sampling, with video focus groups to explore 15 foster parents' experiences in accessing health-care services for foster youth. Results: Barriers and facilitators to accessing health care and care coordination were identified, including logistics, transportation, electronic health record access, location of health care, and the impact of the social worker. Practice Implications: Nurses are well-positioned to aid in closing gaps in care for foster youth. Conclusions: These experiences provide important insights and implications for health-care providers and for future research and policy development.

A review of deep learning models and online healthcare databases for electronic health records and their use for health prediction

A fundamental obstacle to healthcare transformation continues to be the acquisition of knowledge and insightful data from complex, high dimensional, and heterogeneous biological data. As technology has improved, a wide variety of data sources, including omics data, imaging data, and health records, have been available for use in healthcare research contexts. Electronic health records (EHRs), which are digitalized versions of medical records, have given researchers a significant chance to create computational methods for analyzing healthcare data. EHR systems typically keep track of all the data relating to a patient’s medical history, including clinical notes, demographic background, and diagnosis details. EHR data can offer valuable insights and support doctors in making better decisions related to disease and diagnostic forecasts. As a result, several academics use deep learning to forecast diseases and track health trajectories in EHR. Recent advances in deep learning technology have produced innovative and practical paradigms for building end-to-end learning models. However, scholars have limited access to online HER databases, and there is an inherent need to address this issue. This research examines deep learning models, their architectures, and readily accessible EHR online databases. The goal of this paper is to examine how various architectures, models, and databases differ in terms of features and usability. It is anticipated that the outcomes of this review will lead to the development of more robust deep learning models that facilitate medical decision-making processes based on EHR data and inform efforts to support the selection of architectures, models, and databases for specific research purposes.

Comparing South Dakota Pharmacist Perspectives of Pharmacy Services in Rural versus Urban settings

BackgroundAccess to healthcare services is a major barrier to residents of the rural state of South Dakota. As a highly accessible member of the healthcare team, outpatient pharmacists can play a key role in a patient's healthcare journey. There is a need to identify the unique barriers and facilitators pharmacists in both rural and urban areas face to maximize the impact of their role. ObjectiveThe objective of this work was to compare perceptions of rural and urban pharmacists regarding the facilitators and barriers to providing patient care in South Dakota. MethodsThis qualitative project highlights results from interviews and focus group sessions with a convenience sample of South Dakota pharmacists. Participants were recruited using a referral word-of-mouth system, contracts with healthcare market research agencies, newspaper advertisements, and posters displayed in public locations in South Dakota. Practice location was characterized as rural or urban based on United States Department of Agriculture definitions. Findings from interviews and focus group sessions were coded and analyzed using content analysis by two student researchers. ResultsParticipants included 12 rural-practicing and 21 urban-practicing pharmacists in South Dakota. In both rural and urban areas, key barriers included communication with providers (50% rural; 50% urban), lack of electronic health record access (25% rural; 14% urban), not enough staff (22% rural; 20% urban), and patient misunderstanding the scope of pharmacy (22% rural; 40% urban). Barriers specific to rural areas included time to provide services (22%), having smaller facilities (27%) and provider hesitation regarding collaborative practice agreements (29%). There were no urban-specific barriers. Facilitators specific to urban areas included frequent communication with patients (6.1%) and good quality support staff (9.1%). There were no rural-specific facilitators. ConclusionsNext steps include increasing awareness of pharmacy-based patient care services, researching further to identify the extent to which facilitators and barriers influence the ability to initiate and sustain pharmacy services in rural and urban areas, and providing support to pharmacies to overcome barriers and leverage facilitators.

The Role of Health Information Technology During Hospital to Home Transitions.

The use of healthcare information technology (HIT) is vital for storing and exchanging health information during patient transitions, playing a significant role in care coordination for sepsis survivors. The critical role of HIT was evident during the pre-implementation phase of a study to implement an evidence-based protocol supporting the timely transition of sepsis survivors to home health and outpatient care. Through 61 semi-structured interviews involving 91 stakeholders, over half of the 33 identified themes were related to HIT. Notably, electronic health record (EHR) alert systems led to over-capture and alarm fatigue. Efficient information transfer during HHC referral highlighted the need for improved EHR access. The study underscores HIT's importance and potential while emphasizing the need for collaborative policy and interface development to promote effective transitions in care.

The Premier Medical Education Model: Improving Preceptor Recruitment in Underserved Areas.

Building on research highlighting the success of tribal, rural, and underserved clerkships to increase students' intention to practice family medicine in these areas, we explored the perspectives of prospective precepting physicians and administrators to develop an optimal structure to facilitate recruitment of external preceptors. We conducted semistructured interviews with family physicians (N=14) and health system administrators (N=14) working in tribal, rural, and underserved areas. Discussions were recorded, transcribed verbatim, and coded independently by two researchers. Applying rapid assessment qualitative research methods, we used a framework method to identify emergent themes that were applied to improve the recruitment of external preceptors. Physicians identified key facilitating factors and barriers to serving as a preceptor, which paralleled those common within the existing literature. However, administrators were motivated to serve as a precepting site to increase the potential of recruiting future physicians. We developed the Premier Medical Education Hub model to align these different but compatible interests with the goal to increase preceptor participation. In this model, each host site dedicates staff and adopts standardized procedures to manage rotations, hosts at least five students annually, provides housing, has procedures to facilitate electronic health record access, and offers student immersion experiences. As practice ownership shifts from physician-owned to health system ownership, administrators become the gatekeepers for prospective preceptors. Our findings demonstrate that integrating the compatible interests between physicians and administrators allows for the creation of a synergistic model that facilitates preceptor recruitment.

Machine learning for outcome prediction of atrial fibrillation recurrence after catheter ablation

Abstract Background Outcome prediction after catheter ablation for atrial fibrillation (AF) based on electronic health records (EHR) using machine learning (ML) is not yet established. Purpose The aim of the study was to assess the value of ML methods to predict the risk of AF recurrence after catheter ablation based on easily accessible EHR. Methods We analyzed 1362 patients from our prospective registry. Only patients undergoing a first AF ablation were studied. Follow-up was performed at 3, 6 and 12 months after the ablation including 24-hour and 7-day Holter electrocardiogram (ECG). Four different analyses with increasing complexity out of a set of overall 22 features using seven simple and three ensemble method ML algorithms were performed: model 1 (including age, sex, height, weight, and established anamnestic risk factors: AF-type, coronary artery disease, myocardial infarction, stroke, heart failure, hypertension, current smoker, diabetes, renal insufficiency), model 2: model 1 plus additional AF-history features (duration of AF in months, previous typical atrial flutter, number of previous electrical cardioversions, number of failed antiarrhythmic drugs), model 3: model 2 plus simple echocardiographic parameters (left ventricular ejection fraction, LA diameter), model 4: model 3 plus biomarkers (CRP, creatinine, NTproBNP). To evaluate the performance of the prediction, we report C-statistics, accuracy, and recall (sensitivity). The C-statistics of five risk scores for AF recurrence (APPLE, DR-FLASH, BASE-AF2, ATLAS, CHA2DS2-Vasc) and a logistic regression with a forward selection were calculated for comparison. Results Of the 1362 patients, 996 (73%) were male, mean age was 62±10 years and 60% presented with paroxysmal AF. Recurrence of AF during 1-year follow-up was observed in 458 patients (34%). The results for the four models using the eleven different ML algorithms are summarized in the Figure. In detail, the C-statistics were modest with a maximal value of 0.602 for the random forest (RF) classifier for the most complex model 4. The performance for the risk scores was maximal for the APPLE score (C-statistics 0.572) and comparable for the forward-selection logistic regression including five features with a C-statistics of 0.607 (95% CI 0.573-0.641). Conclusion In conclusion, outcome prediction of AF recurrence after catheter ablation with easily accessible EHR data remains challenging and cannot be improved simply by applying machine learning methods. Whether a more comprehensive feature selection, such as more complex ECG parameters instead of the selection of easily available features from the EHR, a combination with deep neural network-based feature extraction, or unsupervised ML might improve the result in a clinically relevant way needs further investigation.Results of the eleven ML algorithms

Generating Synthetic Electronic Health Record Data Using Generative Adversarial Networks: Tutorial.

Synthetic electronic health record (EHR) data generation has been increasingly recognized as an important solution to expand the accessibility and maximize the value of private health data on a large scale. Recent advances in machine learning have facilitated more accurate modeling for complex and high-dimensional data, thereby greatly enhancing the data quality of synthetic EHR data. Among various approaches, generative adversarial networks (GANs) have become the main technical path in the literature due to their ability to capture the statistical characteristics of real data. However, there is a scarcity of detailed guidance within the domain regarding the development procedures of synthetic EHR data. The objective of this tutorial is to present a transparent and reproducible process for generating structured synthetic EHR data using a publicly accessible EHR data set as an example. We cover the topics of GAN architecture, EHR data types and representation, data preprocessing, GAN training, synthetic data generation and postprocessing, and data quality evaluation. We conclude this tutorial by discussing multiple important issues and future opportunities in this domain. The source code of the entire process has been made publicly available.

Enhancing availability of services to control hypertension through a team-based care approach that includes pharmacists

BackgroundPrimary care physician (PCP) shortages are expected to increase. The Michigan Medicine Hypertension Pharmacists’ Program uses a team-based care (TBC) approach to redistribute some patient care responsibilities from PCPs to pharmacists for patients with diagnosed hypertension. ObjectiveThis evaluation analyzed whether the Michigan Medicine Hypertension Pharmacists’ Program increased the availability of hypertension management services and described facilitators that addressed barriers to program sustainability and replicability. MethodsWe conducted a retrospective observational study that used a mixed methods approach. We examined the availability of hypertension management services using the number of pharmacists’ referrals of patients to other services and the number of PCP appointments. We analyzed qualitative interviews with program staff and site-level quantitative data to examine the program’s impact on the availability of services, the impact of TBC that engaged pharmacists, and program barriers and facilitators. ResultsPatients who visited a pharmacist had fewer PCP visits over 3- and 6-month periods compared to a matched comparison group that did not see a pharmacist and were 1.35 times more likely to receive a referral to a specialist within a 3-month period. Support from leaders and physicians, shared electronic health record access, and financial backing emerged as leading factors for program sustainability and replicability. ConclusionAdding pharmacists to the care team reduced the number of PCP appointments per patient while increasing the availability of hypertension management services; this may in turn improve PCPs’ availability. Similar models may be sustainable and replicable by relying on organizational buy-in, accessible infrastructure, and financing.

WIC staff and healthcare professional perceptions of an EHR intervention to facilitate referrals to and improve communication and coordination with WIC: A qualitative study.

Participation in the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) has numerous benefits, yet many eligible children remain unenrolled. This qualitative study sought to explore perceptions of a novel electronic health record (EHR) intervention to facilitate referrals to WIC and improve communication/coordination between WIC staff and healthcare professionals. WIC staff in three counties were provided EHR access and recruited to participate. An automated, EHR-embedded WIC participation screening and referral tool was implemented within 8 healthcare clinics; healthcare professionals within these clinics were eligible to participate. The interview guide was developed using the Consolidated Framework for Implementation Research to elicit perceptions of this novel EHR-based intervention. Semi-structured interviews were conducted via telephone. Interviews were recorded, transcribed, coded, and analyzed using thematic analysis. Twenty semi-structured interviews were conducted with eight WIC staff, seven pediatricians, four medical assistants, and one registered nurse. Most participants self-identified as female (95%) and White (55%). We identified four primary themes: (1) healthcare professionals had a positive view of WIC but communication and coordination between WIC and healthcare professionals was limited prior to WIC having EHR access; (2) healthcare professionals favored WIC screening using the EHR but workflow challenges existed; (3) EHR connections between WIC and the healthcare system can streamline referrals to and enrollment in WIC; and (4) WIC staff and healthcare professionals recommended that WIC have EHR access. A novel EHR-based intervention has potential to facilitate healthcare referrals to WIC and improve communication/coordination between WIC and healthcare systems.

Design of a Bilingual and Bicultural Dementia Screening Toolkit Integrated in the Electronic Health Record

AbstractBackgroundBased predominantly on longitudinal cohort studies, the prevalence of Alzheimer’ disease (AD) is estimated to be 3‐10% in individuals over 65 years old and has increased risk (1.5‐2.0 times) in Black and Hispanic/Latinx individuals. In our prior analyses of the University of California Los Angeles (UCLA) electronic health record (EHR), we found that only 1% of patients over 65 years old are diagnosed with AD, and AD is less likely to be diagnosed in Black and Hispanic/Latinx patients compared to white patients (relative risk 0.86‐0.9). One solution to the underdiagnosis of AD in real‐world clinical care, is to screen for dementia.MethodsRecent work by several University California AD Centers and the California Department of Public Health developed a short (<5 min) dementia screening tool (DST). The DST includes a three‐question questionnaire for the patient and an informant if present, followed by the Mini‐Cog if applicable (Figure 1). The questions ask about changes in language, memory, and personality (Figure 2). The DST is considered positive if any question is answered “Yes” by the patient or informant, or if the Mini‐Cog score is <3 out of 5 points. To improve care for non‐English speaking patients, we developed the questionnaire in Spanish, which was adapted to Hispanic/Latinx culture.ResultsThe DST is integrated in the EHR. The questionnaires are sent to patients prior to the visit via the messaging system. Answers to the questionnaire and the Mini‐Cog are entered into a flowsheet in an easily accessible EHR tab. We also created a documentation link that automatically loads the DST results, final score, and recommendation into provider notes. The DST is live at a diverse UCLA family medicine clinic, which sees HL and Black patients. In the first three weeks, 178 patients were screened with the toolkit and eight patients were referred for specialty care.ConclusionsWe have developed a cultural and linguistic responsive DST that is integrated in the EHR. We will continue evaluating its ability to increase appropriate diagnosis of dementia in diverse racial and ethnic groups, and share our DST including EHR integration so it can be implemented elsewhere.

Machine Learning to Predict Risk of Venous Thromboembolism Among Patients with Myeloproliferative Neoplasms in the Intensive Care Unit

INTRODUCTION Myeloproliferative neoplasms (MPN) increase the risk of venous thromboembolism (VTE) and constitute important thrombotic risk factors. VTE risk is especially high if these patients require the intensive care unit (ICU). Precise and user-friendly diagnostic tools that incorporate complex interactions between clinical data and laboratory results to predict VTE among patients with MPN in the ICU are lacking. We hypothesize that machine-learning algorithms could be used to predict VTE and guide thromboprophylaxis in this population. METHODS This was an observational, retrospective study using routine clinical practice data from MIMIC-IV, a freely accessible electronic health record database. We analyzed patients with an MPN diagnosis who were admitted to the ICU. ICD-9 and ICD-10 codes were utilized to identify incident VTE events during ICU stay (I82X, 453X). Descriptive statistics were reported for all variables of interest. Various machine-learning models were utilized to predict the incident VTE as the outcome variable. Patients who had expired as discharge dispositions were excluded. Models accounted for demographic characteristics, laboratory results, comorbidity burden, and other domains. Model performance was evaluated by comparing the accuracy and Area Under the Receiver Operating Characteristic Curve (AUROC) values. RESULTS We analyzed 378 cases that met the inclusion criteria. Among these, 160 (42.3%) were female. The majority were White (65.9%), and 151 (39.9%) had Medicare insurance. The median [Q1, Q3] age of the sample was 64.0 [52.2, 76.8] years. VTE occurred in 27 patients (7.1% incidence). ICU mortality was reported in 26 (6.9%) cases. The proportion of ER admissions was 48.1% among those cases who had VTE and 47.6% among those who didn't have VTE in the ICU (p > 0.05). Patients who developed VTE were more likely males (66.7% vs. 33.3%), but there was no difference in age (median age 61 vs. 64 years; p > 0.05). The machine learning model results showed that the Naive Bayes classifier performed best in predicting VTE in the ICU among patients with MPNs, with an AUROC value of 84.36%, followed by the deep learning model, with an AUROC value of 68.3%. The deep learning model showed that the top ten variables associated with VTE by feature importance were oxygen saturation, length of stay, blood urea nitrogen, other insurance type, anion gap, respiratory rate, bicarbonate, pressure of oxygen, pH value, and race other than White. CONCLUSION Predicting VTE in the ICU among patients with MPNs may be achieved by machine learning methods. The deep learning algorithm outperformed other models in terms of discrimination and calibration. If validated and applied in clinical practice, these methods may optimize thromboprophylaxis strategies. Future research is needed to test this model in more extensive clinical settings.

1369. Effect of COVID-19 on Antibiotic Consumption: An Analysis of the COVID-19 Community Research Partnership

Abstract Background The COVID-19 Community Research Partnership (CCRP) is a multisite healthcare system-based study of the COVID pandemic. Community residents in this study provided consent for electronic health record (EHR) access allowing for analysis of antibiotic prescribing. Methods We conducted a retrospective, self-controlled cohort study to determine antibiotic consumption in the CCRP pre-COVID (Jan 2018 - Dec 2019) and post-COVID (Jan 2020 - Jan 2022). Participants were ≥18 years with an EHR record in the pre-COVID period. The primary outcome was the number of antibiotic prescriptions (AP) per 1000 patient care encounters in each time period. Secondary outcomes included the AP location of care, analysis of antibiotic classes, and COVID attributable antibiotic consumption, defined as an AP written ≤14 days before or after an EHR documented or self-reported positive SARS-CoV-2 test or COVID diagnosis. Results 34,234 participants were included representing 2,775,711 encounters. Median age was 53 years with 70% female and 88% white. 41% lived in a rural or suburban area. There were 26,121 APs among 25,433 participants. Mean rate of AP per 1000 encounters was 12.1 and 7.6 pre vs post-COVID (p< 0.001). A decline in AP occurred at onset of the pandemic persisting through the post-COVID period (Fig 1). The prescribing rate for all antibiotics, except linezolid, decreased during the pandemic (p< 0.001, Fig 2). The largest declines were seen with antibiotics typically prescribed for respiratory illnesses. The decline in antibiotic consumption was driven by outpatient encounters (Fig 3). 6,341 participants were diagnosed with COVID and 333 APs were COVID attributable. These participants were more likely to receive a macrolide (45% vs. 22%, p< 0.001), live in a rural area (31% vs. 20%, p< 0.001), and receive care in an emergency department (12% vs. 6.1%, p< 0.001). Figure 1. Antibiotic Prescriptions During The Study Time Period Figure 2. Analysis Of Antibiotic Classes Figure 3. Location of Antibiotic Prescribing Conclusion The COVID pandemic was associated with a broad decline in antibiotic consumption. These results may be due to declines in other respiratory viruses, anchoring bias for COVID during the pandemic, improved antibiotic stewardship practices, or social isolation. Antibiotics for respiratory illnesses declined the most, suggesting this may be a target for outpatient stewardship efforts. Disclosures All Authors: No reported disclosures

Machine Learning to Detect Disparities in ICU Mortality Among Patients Receiving Direct Oral Anticoagulants

INTRODUCTION The current prescribing practices of direct oral anticoagulants (DOACs) in intensive care unit (ICU) patients and the associated clinical outcomes, including the incidence of major bleeding episodes and mortality, are significant clinical challenges. There is a need for early detection and intervention among patients taking DOACs to prevent adverse clinical outcomes. We hypothesized that machine-learning algorithms could be used to create a more precise and user-friendly diagnostic tool that incorporates a variety of clinical and laboratory data and takes complex interactions into account. METHODS This was an observational, retrospective study using routine clinical practice data from MIMIC-IV, a freely accessible electronic health record database. The study included patients who were admitted to the ICU with a DOAC (apixaban, rivaroxaban, dabigatran, or edoxaban) listed as one of the active medications at the time of hospital admission. ICD-9 and ICD-10 codes were utilized to identify cases (Z7901, V5861). Descriptive statistics were reported for all variables of interest. Various machine-learning models were utilized to predict the outcome variable of ICU mortality after accounting for variables from the demographic, laboratory, comorbidity, etc. domains. Model performance was evaluated by comparing the accuracy and Area Under the Receiver Operating Characteristic Curve (AUROC) values. RESULTS Overall, there were 3597 cases that met the inclusion criteria. Among these, 1517 (42.2%) were female. The majority of the sample was White (70.5%), and 2093 (58.2%) cases had Medicare insurance. The median [Q1, Q3] age of the sample was 74 [65, 83] years. ICU mortality was reported in 297 cases (8.2%). The mean ( SD) ICU length of stay was 3.5 (4.6) days among those cases who were discharged alive and 4.2 (4.7) days among those who expired (p < 0.012). The median blood urea nitrogen among those discharged alive and expired was 20.0 and 29.5 (p<0.001), respectively. Intracranial hemorrhage was reported in 186 (5.6%) and 59 (19.9%) of those discharged alive and expired, respectively (p<0.001). The machine learning model results showed that the Random forest classifier performed best in predicting mortality in the ICU, with an AUROC value of 85.3%. The top ten variables associated with ICU mortality by feature importance were pH, WBC count, platelet count, blood urea nitrogen, heart rate, anion gap, partial pressure of carbon dioxide, oxygen saturation, age, and respiratory rate. CONCLUSION The Random forest classifier algorithm improved discrimination and calibration compared to other ML models for predicting mortality in the ICU among patients taking DOACs. It has the potential to reduce delayed diagnosis and overtreatment in clinical practice. Future studies will validate this model in broader clinical settings.

Differences in Health Professionals’ Engagement With Electronic Health Records Based on Inpatient Race and Ethnicity

US health professionals devote a large amount of effort to engaging with patients' electronic health records (EHRs) to deliver care. It is unknown whether patients with different racial and ethnic backgrounds receive equal EHR engagement. To investigate whether there are differences in the level of health professionals' EHR engagement for hospitalized patients according to race or ethnicity during inpatient care. This cross-sectional study analyzed EHR access log data from 2 major medical institutions, Vanderbilt University Medical Center (VUMC) and Northwestern Medicine (NW Medicine), over a 3-year period from January 1, 2018, to December 31, 2020. The study included all adult patients (aged ≥18 years) who were discharged alive after hospitalization for at least 24 hours. The data were analyzed between August 15, 2022, and March 15, 2023. The actions of health professionals in each patient's EHR were based on EHR access log data. Covariates included patients' demographic information, socioeconomic characteristics, and comorbidities. The primary outcome was the quantity of EHR engagement, as defined by the average number of EHR actions performed by health professionals within a patient's EHR per hour during the patient's hospital stay. Proportional odds logistic regression was applied based on outcome quartiles. A total of 243 416 adult patients were included from VUMC (mean [SD] age, 51.7 [19.2] years; 54.9% female and 45.1% male; 14.8% Black, 4.9% Hispanic, 77.7% White, and 2.6% other races and ethnicities) and NW Medicine (mean [SD] age, 52.8 [20.6] years; 65.2% female and 34.8% male; 11.7% Black, 12.1% Hispanic, 69.2% White, and 7.0% other races and ethnicities). When combining Black, Hispanic, or other race and ethnicity patients into 1 group, these patients were significantly less likely to receive a higher amount of EHR engagement compared with White patients (adjusted odds ratios, 0.86 [95% CI, 0.83-0.88; P < .001] for VUMC and 0.90 [95% CI, 0.88-0.92; P < .001] for NW Medicine). However, a reduction in this difference was observed from 2018 to 2020. In this cross-sectional study of inpatient EHR engagement, the findings highlight differences in how health professionals distribute their efforts to patients' EHRs, as well as a method to measure these differences. Further investigations are needed to determine whether and how EHR engagement differences are correlated with health care outcomes.

Involving Older Users in Design of Patient Accessible Electronic Health Records: Exploring the Potential of Open‐Ended Questions

ABSTRACTDespite older adults being one of the most important groups of users of eHealth services, including patient accessible electronic health records, these are claimed to not be designed to suit older users, and to not necessarily fulfill their information needs. User‐centered design methods often utilized when developing eHealth services have also been claimed to not work well for older adults. This paper suggests an alternative way of collecting ideas and feedback on eHealth from older adults, by including open‐ended questions in the data collection of larger studies on eHealth information behavior. When data are collected by methods like focus group interviews and surveys, older adults can be requested to envision an eHealth service that suits their needs. Results from studies in Finland are presented, and show that current eHealth services do not fulfill all information needs of older adults, as answers to the question in both focus group interviews and surveys contain practical suggestions for additional and more comprehensible information, easier ways of searching for needed information, and alternative channels for communication. Open‐ended questions can be a cost‐effective and timesaving way to help in the design and development of patient accessible electronic health records and other types of eHealth services.

331 Implementation of Integrated Electronic Health Record Access for Out-of-Hospital Clinicians

331 Implementation of Integrated Electronic Health Record Access for Out-of-Hospital Clinicians