Unlocking the Potential of Secondary Data for Public Health Research: Retrospective Study With a Novel Clinical Platform.

BACKGROUND: Clinical routine data derived from university hospitals hold immense value for health-related research on large cohorts. However, using secondary data for hypothesis testing necessitates adherence to scientific, legal (such as the General Data Protection Regulation, federal and state protection legislations), technical, and administrative requirements. This process is intricate, time-consuming, and susceptible to errors. OBJECTIVE: This study aims to develop a platform that enables clinicians to use current real-world data for testing research and evaluate advantages and limitations at a large university medical center (542,944 patients in 2022). METHODS: We identified requirements from clinical practitioners, conceptualized and implemented a platform based on the existing components, and assessed its applicability in clinical reality quantitatively and qualitatively. RESULTS: The proposed platform was established at the University Medical Center Hamburg-Eppendorf and made 639 forms encompassing 10,629 data elements accessible to all resident scientists and clinicians. Every day, the number of patients rises, and parts of their electronic health records are made accessible through the platform. Qualitatively, we were able to conduct a retrospective analysis of Parkinson disease over 777 patients, where we provide additional evidence for a significantly higher proportion of action tremors in patients with rest tremors (340/777, 43.8%) compared with those without rest tremors (255/777, 32.8%), as determined by a chi-square test (P<.001). Quantitatively, our findings demonstrate increased user engagement within the last 90 days, underscoring clinicians' increasing adoption of the platform in their regular research activities. Notably, the platform facilitated the retrieval of clinical data from 600,000 patients, emphasizing its substantial added value. CONCLUSIONS: This study demonstrates the feasibility of simplifying the use of clinical data to enhance exploration and sustainability in scientific research. The proposed platform emerges as a potential technological and legal framework for other medical centers, providing them with the means to unlock untapped potential within their routine data.

Addressing Information Biases Within Electronic Health Record Data to Improve the Examination of Epidemiologic Associations With Diabetes Prevalence Among Young Adults: Cross-Sectional Study.

Background: Electronic health records (EHRs) are increasingly used for epidemiologic research to advance public health practice. However, key variables are susceptible to missing data or misclassification within EHRs, including demographic information or disease status, which could affect the estimation of disease prevalence or risk factor associations. Objective: In this paper, we applied methods from the literature on missing data and causal inference to assess whether we could mitigate information biases when estimating measures of association between potential risk factors and diabetes among a patient population of New York City young adults. Methods: We estimated the odds ratio (OR) for diabetes by race or ethnicity and asthma status using EHR data from NYU Langone Health. Methods from the missing data and causal inference literature were then applied to assess the ability to control for misclassification of health outcomes in the EHR data. We compared EHR-based associations with associations observed from 2 national health surveys, the Behavioral Risk Factor Surveillance System (BRFSS) and the National Health and Nutrition Examination Survey, representing traditional public health surveillance systems. Results: Observed EHR-based associations between race or ethnicity and diabetes were comparable to health survey-based estimates, but the association between asthma and diabetes was significantly overestimated (OREHR 3.01, 95% CI 2.86-3.18 vs ORBRFSS 1.23, 95% CI 1.09-1.40). Missing data and causal inference methods reduced information biases in these estimates, yielding relative differences from traditional estimates below 50% (ORMissingData 1.79, 95% CI 1.67-1.92 and ORCausal 1.42, 95% CI 1.34-1.51). Conclusions: Findings suggest that without bias adjustment, EHR analyses may yield biased measures of association, driven in part by subgroup differences in health care use. However, applying missing data or causal inference frameworks can help control for and, importantly, characterize residual information biases in these estimates.

Languages other than English in a pediatric ED: Documentation, extended stays, and revisits.

OBJECTIVE: Given the increasing proportion of patients and caregivers who use languages other than English (LOE) at our institution and across the U.S, we evaluated key workflow and outcome measures in our emergency department (ED) for patients and caregivers who use LOE. METHODS: This was a retrospective, cross-sectional study of patients and caregivers who presented to a free-standing urban pediatric facility. We used electronic health record data (EHR) and interpreter usage log data for our analysis of language documentation, length of stay, and ED revisits. We assessed ED revisits within 72-h using a multivariable logistic regression model adjusting for whether a primary care provider (PCP) was listed in the EHR, whether discharge was close to or on the weekend, and insurance status. We restricted our analysis to low-acuity patient encounters (Emergency Severity Index (ESI) scores of 4 and 5) to limit confounding factors related to higher ESI scores. RESULTS: We found that one in five patients and caregivers who use LOE had incorrect documentation of their language needs in the EHR. Using interpreter usage data to most accurately capture encounters using LOE, we found that patient encounters using LOE had a 38-min longer length of stay (LOS) and twice the odds of a 72-h ED revisit compared to encounters using English. CONCLUSION: These results highlight the need for better language documentation and understanding of factors contributing to extended stays and increased revisits for pediatric patients and caregivers who use LOE.

Blockchain-enabled EHR access auditing: Enhancing healthcare data security.

In the realm of modern healthcare, Electronic Health Records EHR serve as invaluable assets, yet they also pose significant security challenges. The absence of EHR access auditing mechanisms, which includes the EHR audit trails, results in accountability gaps and magnifies security vulnerabilities. This situation effectively paves the way for unauthorized data alterations to occur without detection or consequences. Inadequate EHR compliance auditing procedures, particularly in verifying and validating access control policies, expose healthcare organizations to risks such as data breaches, and unauthorized data usage. These vulnerabilities result from unchecked unauthorized access activities. Additionally, the absence of EHR audit logs complicates investigations, weakens proactive security measures, and raises concerns to put healthcare institutions at risk. This study addresses the pressing need for robust EHR auditing systems designed to scrutinize access to EHR data, encompassing who accesses it, when, and for what purpose. Our research delves into the complex field of EHR auditing, which includes establishing an immutable audit trail to enhance data security through blockchain technology. We also integrate Purpose-Based Access Control (PBAC) alongside smart contracts to strengthen compliance auditing by validating access legitimacy and reducing unauthorized entries. Our contributions encompass the creation of audit trail of EHR access, compliance auditing via PBAC policy verification, the generation of audit logs, and the derivation of data-driven insights, fortifying EHR access security.

Automated System to Capture Patient Symptoms From Multitype Japanese Clinical Texts: Retrospective Study.

BACKGROUND: Natural language processing (NLP) techniques can be used to analyze large amounts of electronic health record texts, which encompasses various types of patient information such as quality of life, effectiveness of treatments, and adverse drug event (ADE) signals. As different aspects of a patient's status are stored in different types of documents, we propose an NLP system capable of processing 6 types of documents: physician progress notes, discharge summaries, radiology reports, radioisotope reports, nursing records, and pharmacist progress notes. OBJECTIVE: This study aimed to investigate the system's performance in detecting ADEs by evaluating the results from multitype texts. The main objective is to detect adverse events accurately using an NLP system. METHODS: We used data written in Japanese from 2289 patients with breast cancer, including medication data, physician progress notes, discharge summaries, radiology reports, radioisotope reports, nursing records, and pharmacist progress notes. Our system performs 3 processes: named entity recognition, normalization of symptoms, and aggregation of multiple types of documents from multiple patients. Among all patients with breast cancer, 103 and 112 with peripheral neuropathy (PN) received paclitaxel or docetaxel, respectively. We evaluate the utility of using multiple types of documents by correlation coefficient and regression analysis to compare their performance with each single type of document. All evaluations of detection rates with our system are performed 30 days after drug administration. RESULTS: Our system underestimates by 13.3 percentage points (74.0%-60.7%), as the incidence of paclitaxel-induced PN was 60.7%, compared with 74.0% in the previous research based on manual extraction. The Pearson correlation coefficient between the manual extraction and system results was 0.87 Although the pharmacist progress notes had the highest detection rate among each type of document, the rate did not match the performance using all documents. The estimated median duration of PN with paclitaxel was 92 days, whereas the previously reported median duration of PN with paclitaxel was 727 days. The number of events detected in each document was highest in the physician's progress notes, followed by the pharmacist's and nursing records. CONCLUSIONS: Considering the inherent cost that requires constant monitoring of the patient's condition, such as the treatment of PN, our system has a significant advantage in that it can immediately estimate the treatment duration without fine-tuning a new NLP model. Leveraging multitype documents is better than using single-type documents to improve detection performance. Although the onset time estimation was relatively accurate, the duration might have been influenced by the length of the data follow-up period. The results suggest that our method using various types of data can detect more ADEs from clinical documents.

Extracting Critical Information from Unstructured Clinicians' Notes Data to Identify Dementia Severity Using a Rule-Based Approach: Feasibility Study.

BACKGROUND: The severity of Alzheimer disease and related dementias (ADRD) is rarely documented in structured data fields in electronic health records (EHRs). Although this information is important for clinical monitoring and decision-making, it is often undocumented or "hidden" in unstructured text fields and not readily available for clinicians to act upon. OBJECTIVE: We aimed to assess the feasibility and potential bias in using keywords and rule-based matching for obtaining information about the severity of ADRD from EHR data. METHODS: We used EHR data from a large academic health care system that included patients with a primary discharge diagnosis of ADRD based on ICD-9 (International Classification of Diseases, Ninth Revision) and ICD-10 (International Statistical Classification of Diseases, Tenth Revision) codes between 2014 and 2019. We first assessed the presence of ADRD severity information and then the severity of ADRD in the EHR. Clinicians' notes were used to determine the severity of ADRD based on two criteria: (1) scores from the Mini Mental State Examination and Montreal Cognitive Assessment and (2) explicit terms for ADRD severity (eg, "mild dementia" and "advanced Alzheimer disease"). We compiled a list of common ADRD symptoms, cognitive test names, and disease severity terms, refining it iteratively based on previous literature and clinical expertise. Subsequently, we used rule-based matching in Python using standard open-source data analysis libraries to identify the context in which specific words or phrases were mentioned. We estimated the prevalence of documented ADRD severity and assessed the performance of our rule-based algorithm. RESULTS: We included 9115 eligible patients with over 65,000 notes from the providers. Overall, 22.93% (2090/9115) of patients were documented with mild ADRD, 20.87% (1902/9115) were documented with moderate or severe ADRD, and 56.20% (5123/9115) did not have any documentation of the severity of their ADRD. For the task of determining the presence of any ADRD severity information, our algorithm achieved an accuracy of >95%, specificity of >95%, sensitivity of >90%, and an F1-score of >83%. For the specific task of identifying the actual severity of ADRD, the algorithm performed well with an accuracy of >91%, specificity of >80%, sensitivity of >88%, and F1-score of >92%. Comparing patients with mild ADRD to those with more advanced ADRD, the latter group tended to contain older, more likely female, and Black patients, and having received their diagnoses in primary care or in-hospital settings. Relative to patients with undocumented ADRD severity, those with documented ADRD severity had a similar distribution in terms of sex, race, and rural or urban residence. CONCLUSIONS: Our study demonstrates the feasibility of using a rule-based matching algorithm to identify ADRD severity from unstructured EHR report data. However, it is essential to acknowledge potential biases arising from differences in documentation practices across various health care systems.

The Development of an Electronic Medical Record System to Improve Quality of Care for Individuals With Type 1 Diabetes in Rwanda: Qualitative Study.

BACKGROUND: Electronic medical record (EMR) systems have the potential to improve the quality of care and clinical outcomes for individuals with chronic and complex diseases. However, studies on the development and use of EMR systems for type 1 (T1) diabetes management in sub-Saharan Africa are few. OBJECTIVE: The aim of this study is to analyze the need for improvements in the care processes that can be facilitated by an EMR system and to develop an EMR system for increasing quality of care and clinical outcomes for individuals with T1 diabetes in Rwanda. METHODS: A qualitative, cocreative, and multidisciplinary approach involving local stakeholders, guided by the framework for complex public health interventions, was applied. Participant observation and the patient's personal experiences were used as case studies to understand the clinical care context. A focus group discussion and workshops were conducted to define the features and content of an EMR. The data were analyzed using thematic analysis. RESULTS: The identified themes related to feature requirements were (1) ease of use, (2) automatic report preparation, (3) clinical decision support tool, (4) data validity, (5) patient follow-up, (6) data protection, and (7) training. The identified themes related to content requirements were (1) treatment regimen, (2) mental health, and (3) socioeconomic and demographic conditions. A theory of change was developed based on the defined feature and content requirements to demonstrate how these requirements could strengthen the quality of care and improve clinical outcomes for people with T1 diabetes. CONCLUSIONS: The EMR system, including its functionalities and content, can be developed through an inclusive and cocreative process, which improves the design phase of the EMR. The development process of the EMR system is replicable, but the solution needs to be customized to the local context.

Natural Language Processing Versus Diagnosis Code-Based Methods for Postherpetic Neuralgia Identification: Algorithm Development and Validation.

Background: Diagnosis codes and prescription data are used in algorithms to identify postherpetic neuralgia (PHN), a debilitating complication of herpes zoster (HZ). Because of the questionable accuracy of codes and prescription data, manual chart review is sometimes used to identify PHN in electronic health records (EHRs), which can be costly and time-consuming. Objective: This study aims to develop and validate a natural language processing (NLP) algorithm for automatically identifying PHN from unstructured EHR data and to compare its performance with that of code-based methods. Methods: This retrospective study used EHR data from Kaiser Permanente Southern California, a large integrated health care system that serves over 4.8 million members. The source population included members aged ≥50 years who received an incident HZ diagnosis and accompanying antiviral prescription between 2018 and 2020 and had ≥1 encounter within 90-180 days of the incident HZ diagnosis. The study team manually reviewed the EHR and identified PHN cases. For NLP development and validation, 500 and 800 random samples from the source population were selected, respectively. The sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), F-score, and Matthews correlation coefficient (MCC) of NLP and the code-based methods were evaluated using chart-reviewed results as the reference standard. Results: The NLP algorithm identified PHN cases with a 90.9% sensitivity, 98.5% specificity, 82% PPV, and 99.3% NPV. The composite scores of the NLP algorithm were 0.89 (F-score) and 0.85 (MCC). The prevalences of PHN in the validation data were 6.9% (reference standard), 7.6% (NLP), and 5.4%-13.1% (code-based). The code-based methods achieved a 52.7%-61.8% sensitivity, 89.8%-98.4% specificity, 27.6%-72.1% PPV, and 96.3%-97.1% NPV. The F-scores and MCCs ranged between 0.45 and 0.59 and between 0.32 and 0.61, respectively. Conclusions: The automated NLP-based approach identified PHN cases from the EHR with good accuracy. This method could be useful in population-based PHN research.

Processing of Short-Form Content in Clinical Narratives: Systematic Scoping Review.

BACKGROUND: Clinical narratives are essential components of electronic health records. The adoption of electronic health records has increased documentation time for hospital staff, leading to the use of abbreviations and acronyms more frequently. This brevity can potentially hinder comprehension for both professionals and patients. OBJECTIVE: This review aims to provide an overview of the types of short forms found in clinical narratives, as well as the natural language processing (NLP) techniques used for their identification, expansion, and disambiguation. METHODS: In the databases Web of Science, Embase, MEDLINE, EBMR (Evidence-Based Medicine Reviews), and ACL Anthology, publications that met the inclusion criteria were searched according to PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines for a systematic scoping review. Original, peer-reviewed publications focusing on short-form processing in human clinical narratives were included, covering the period from January 2018 to February 2023. Short-form types were extracted, and multidimensional research methodologies were assigned to each target objective (identification, expansion, and disambiguation). NLP study recommendations and study characteristics were systematically assigned occurrence rates for evaluation. RESULTS: Out of a total of 6639 records, only 19 articles were included in the final analysis. Rule-based approaches were predominantly used for identifying short forms, while string similarity and vector representations were applied for expansion. Embeddings and deep learning approaches were used for disambiguation. CONCLUSIONS: The scope and types of what constitutes a clinical short form were often not explicitly defined by the authors. This lack of definition poses challenges for reproducibility and for determining whether specific methodologies are suitable for different types of short forms. Analysis of a subset of NLP recommendations for assessing quality and reproducibility revealed only partial adherence to these recommendations. Single-character abbreviations were underrepresented in studies on clinical narrative processing, as were investigations in languages other than English. Future research should focus on these 2 areas, and each paper should include descriptions of the types of content analyzed.

Standardizing Corneal Transplantation Records Using openEHR: Case Study.

BACKGROUND: Corneal transplantation, also known as keratoplasty, is a widely performed surgical procedure that aims to restore vision in patients with corneal damage. The success of corneal transplantation relies on the accurate and timely management of patient information, which can be enhanced using electronic health records (EHRs). However, conventional EHRs are often fragmented and lack standardization, leading to difficulties in information access and sharing, increased medical errors, and decreased patient safety. In the wake of these problems, there is a growing demand for standardized EHRs that can ensure the accuracy and consistency of patient data across health care organizations. OBJECTIVE: This paper proposes the use of openEHR structures for standardizing corneal transplantation records. The main objective of this research was to improve the quality and interoperability of EHRs in corneal transplantation, making it easier for health care providers to capture, share, and analyze clinical information. METHODS: A series of sequential steps were carried out in this study to implement standardized clinical records using openEHR specifications. These specifications furnish a methodical approach that ascertains the development of high-quality clinical records. In broad terms, the methodology followed encompasses the conduction of meetings with health care professionals and the modeling of archetypes, templates, forms, decision rules, and work plans. RESULTS: This research resulted in a tailored solution that streamlines health care delivery and meets the needs of medical professionals involved in the corneal transplantation process while seamlessly aligning with contemporary clinical practices. The proposed solution culminated in the successful integration within a Portuguese hospital of 3 key components of openEHR specifications: forms, Decision Logic Modules, and Work Plans. A statistical analysis of data collected from May 1, 2022, to March 31, 2023, allowed for the perception of the use of the new technologies within the corneal transplantation workflow. Despite the completion rate being only 63.9% (530/830), which can be explained by external factors such as patient health and availability of donor organs, there was an overall improvement in terms of task control and follow-up of the patients' clinical process. CONCLUSIONS: This study shows that the adoption of openEHR structures represents a significant step forward in the standardization and optimization of corneal transplantation records. It offers a detailed demonstration of how to implement openEHR specifications and highlights the different advantages of standardizing EHRs in the field of corneal transplantation. Furthermore, it serves as a valuable reference for researchers and practitioners who are interested in advancing and improving the exploitation of EHRs in health care.

Association of a Novel Electronic Form for Preoperative Cardiac Risk Assessment With Reduction in Cardiac Consultations and Testing: Retrospective Cohort Study.

BACKGROUND: Preoperative cardiac risk assessment is an integral part of preoperative evaluation; however, there is significant variation among providers, leading to inappropriate referrals for cardiology consultation or excessive low-value cardiac testing. We implemented a novel electronic medical record (EMR) form in our preoperative clinics to decrease variation. OBJECTIVE: This study aimed to investigate the impact of the EMR form on the preoperative utilization of cardiology consultation and cardiac diagnostic testing (echocardiograms, stress tests, and cardiac catheterization) and evaluate postoperative outcomes. METHODS: A retrospective cohort study was conducted. Patients who underwent outpatient preoperative evaluation prior to an elective surgery over 2 years were divided into 2 cohorts: from July 1, 2021, to June 30, 2022 (pre-EMR form implementation), and from July 1, 2022, to June 30, 2023 (post-EMR form implementation). Demographics, comorbidities, resource utilization, and surgical characteristics were analyzed. Propensity score matching was used to adjust for differences between the 2 cohorts. The primary outcomes were the utilization of preoperative cardiology consultation, cardiac testing, and 30-day postoperative major adverse cardiac events (MACE). RESULTS: A total of 25,484 patients met the inclusion criteria. Propensity score matching yielded 11,645 well-matched pairs. The post-EMR form, matched cohort had lower cardiology consultation (pre-EMR form: n=2698, 23.2% vs post-EMR form: n=2088, 17.9%; P<.001) and echocardiogram (pre-EMR form: n=808, 6.9% vs post-EMR form: n=591, 5.1%; P<.001) utilization. There were no significant differences in the 30-day postoperative outcomes, including MACE (all P>.05). While patients with "possible indications" for cardiology consultation had higher MACE rates, the consultations did not reduce MACE risk. Most algorithm end points, except for active cardiac conditions, had MACE rates <1%. CONCLUSIONS: In this cohort study, preoperative cardiac risk assessment using a novel EMR form was associated with a significant decrease in cardiology consultation and testing utilization, with no adverse impact on postoperative outcomes. Adopting this approach may assist perioperative medicine clinicians and anesthesiologists in efficiently decreasing unnecessary preoperative resource utilization without compromising patient safety or quality of care.

Early Diagnosis of Hereditary Angioedema in Japan Based on a US Medical Dataset: Algorithm Development and Validation.

BACKGROUND: Hereditary angioedema (HAE), a rare genetic disease, induces acute attacks of swelling in various regions of the body. Its prevalence is estimated to be 1 in 50,000 people, with no reported bias among different ethnic groups. However, considering the estimated prevalence, the number of patients in Japan diagnosed with HAE remains approximately 1 in 250,000, which means that only 20% of potential HAE cases are identified. OBJECTIVE: This study aimed to develop an artificial intelligence (AI) model that can detect patients with suspected HAE using medical history data (medical claims, prescriptions, and electronic medical records [EMRs]) in the United States. We also aimed to validate the detection performance of the model for HAE cases using the Japanese dataset. METHODS: The HAE patient and control groups were identified using the US claims and EMR datasets. We analyzed the characteristics of the diagnostic history of patients with HAE and developed an AI model to predict the probability of HAE based on a generalized linear model and bootstrap method. The model was then applied to the EMR data of the Kyoto University Hospital to verify its applicability to the Japanese dataset. RESULTS: Precision and sensitivity were measured to validate the model performance. Using the comprehensive US dataset, the precision score was 2% in the initial model development step. Our model can screen out suspected patients, where 1 in 50 of these patients have HAE. In addition, in the validation step with Japanese EMR data, the precision score was 23.6%, which exceeded our expectations. We achieved a sensitivity score of 61.5% for the US dataset and 37.6% for the validation exercise using data from a single Japanese hospital. Overall, our model could predict patients with typical HAE symptoms. CONCLUSIONS: This study indicates that our AI model can detect HAE in patients with typical symptoms and is effective in Japanese data. However, further prospective clinical studies are required to investigate whether this model can be used to diagnose HAE.

Automated Identification of Postoperative Infections to Allow Prediction and Surveillance Based on Electronic Health Record Data: Scoping Review.

BACKGROUND: Postoperative infections remain a crucial challenge in health care, resulting in high morbidity, mortality, and costs. Accurate identification and labeling of patients with postoperative bacterial infections is crucial for developing prediction models, validating biomarkers, and implementing surveillance systems in clinical practice. OBJECTIVE: This scoping review aimed to explore methods for identifying patients with postoperative infections using electronic health record (EHR) data to go beyond the reference standard of manual chart review. METHODS: We performed a systematic search strategy across PubMed, Embase, Web of Science (Core Collection), the Cochrane Library, and Emcare (Ovid), targeting studies addressing the prediction and fully automated surveillance (ie, without manual check) of diverse bacterial infections in the postoperative setting. For prediction modeling studies, we assessed the labeling methods used, categorizing them as either manual or automated. We evaluated the different types of EHR data needed for the surveillance and labeling of postoperative infections, as well as the performance of fully automated surveillance systems compared with manual chart review. RESULTS: We identified 75 different methods and definitions used to identify patients with postoperative infections in studies published between 2003 and 2023. Manual labeling was the predominant method in prediction modeling research, 65% (49/75) of the identified methods use structured data, and 45% (34/75) use free text and clinical notes as one of their data sources. Fully automated surveillance systems should be used with caution because the reported positive predictive values are between 0.31 and 0.76. CONCLUSIONS: There is currently no evidence to support fully automated labeling and identification of patients with infections based solely on structured EHR data. Future research should focus on defining uniform definitions, as well as prioritizing the development of more scalable, automated methods for infection detection using structured EHR data.

Scalable and successful patient portal lifestyle coaching training for primary care clinical staff.

Maintaining a healthy weight postintentional weight loss is crucial for preventing chronic health conditions, yet many regain weight postintervention. Electronic health record (EHR) portals offer a promising avenue for weight management interventions, leveraging patient-primary care relationships. Our previous research demonstrated that coaching alongside self-monitoring improves weight maintenance compared to monitoring alone. Integrating weight management into routine clinical practice by training existing staff could enhance scalability and sustainability. However, challenges such as inconsistent staff qualifications and high coach turnover rates could affect intervention effectiveness. Standardizing services, training, and coaching continuity seem crucial for success. To report on developing, testing, and evaluating an EHR-based coaching training program for clinical staff, guided by an implementation tool for the MAINTAIN PRIME study. Conducted across 14 University of Utah primary care sites, we developed, tested, and evaluated a coaching training for clinical staff. Guided by a planning model and the Predisposing, Enabling, and Reinforcing (PER) tool, stakeholders actively participated in planning, ensuring alignment with clinic priorities. All clinical staff were invited to participate voluntarily. Evaluation measures included staff interest, training effectiveness, confidence, and readiness. Data collection utilized REDCap, with survey results analyzed using descriptive statistics. Despite increased clinical workload and reassignments posed by coronavirus disease 2019, we were able to train 39 clinical staff, with 34 successfully coaching patients. Feedback indicated high readiness and positive perceptions of coaching feasibility. Coaches reported satisfaction with training, support, and enjoyed establishing connections with patients. The PER strategies allowed us to implement a well-received training program found effective by primary care coaches.

This report describes a training program for medical staff like nurses and medical assistants. The goal is to teach them how to coach patients through an online portal to help them keep their weight off after making healthy lifestyle changes. We worked with different clinic groups and used a planning tool called PER worksheet (predisposing, enabling, and reinforcing) to set up the training program. From September 2021 to March 2023, we offered the training in 14 clinics, and most interested staff completed it. The results showed that the training worked well. People who took part felt they learned enough to coach patients and felt ready to coach. They liked the training and found it helpful. This study suggests that we can teach coaching skills in just four hours of training and that ongoing support and mentorship are important to the trained coaches. Furthermore, this training set-up allows new staff to be trained as they join, which is especially important in places where staff changes frequently. Overall, using the PER tool enabled us to create a training program that staff can use in outpatient clinics to help patients improve their weight management.

Disease characteristics influence the privacy calculus to adopt electronic health records: A survey study in Germany.

Background: The electronic health record (EHR) is integral to improving healthcare efficiency and quality. Its successful implementation hinges on patient willingness to use it, particularly in Germany where concerns about data security and privacy significantly influence usage intention. Little is known about how specific characteristics of medical data influence patients' intention to use the EHR. Objective: This study aims to validate the privacy calculus model (PCM) regarding EHRs and to assess how personal and disease characteristics, namely disease-related stigma and disease time course, affect PCM predictions. Methods: An online survey was conducted to empirically validate the PCM for EHR, incorporating a case vignette varying in disease-related stigma (high/low) and time course (acute/chronic), with N = 241 participants, aged 18 years and older residing in Germany with no previous experience with the diseases mentioned in the respective medical reports. Participants were randomized (single-blinded) into four groups in parallel: high stigma and acute time course (n = 74), high stigma and chronic time course (n = 56), low stigma and acute time course (n = 62) and low stigma and chronic time course (n = 49). The data were analyzed using structural equation modeling with partial least squares. Results: The model explains R² = 71.8% of the variance in intention to use. The intention to use is influenced by perceived benefits, data privacy concerns, trust in the provider, and social norms. However, only the disease's time course, not stigma, affects this intention. For acute diseases, perceived benefits and social norms are influential, whereas for chronic diseases, perceived benefits, privacy concerns, and trust in the provider influence intention. Conclusions: The PCM validation for EHRs reveals that personal and disease characteristics shape usage intention in Germany. The need for tailored EHR adoption strategies that address specific needs and concerns of patients with different disease types. Such strategies could lead to a more successful and widespread implementation of EHRs, especially in privacy-conscious contexts.

Automated Medical Records Review for Mild Cognitive Impairment and Dementia.

Objectives: Unstructured and structured data in electronic health records (EHR) are a rich source of information for research and quality improvement studies. However, extracting accurate information from EHR is labor-intensive. Here we introduce an automated EHR phenotyping model to identify patients with Alzheimer's Disease, related dementias (ADRD), or mild cognitive impairment (MCI). Methods: We assembled medical notes and associated International Classification of Diseases (ICD) codes and medication prescriptions from 3,626 outpatient adults from two hospitals seen between February 2015 and June 2022. Ground truth annotations regarding the presence vs. absence of a diagnosis of MCI or ADRD were determined through manual chart review. Indicators extracted from notes included the presence of keywords and phrases in unstructured clinical notes, prescriptions of medications associated with MCI/ADRD, and ICD codes associated with MCI/ADRD. We trained a regularized logistic regression model to predict the ground truth annotations. Model performance was evaluated using area under the receiver operating curve (AUROC), area under the precision-recall curve (AUPRC), accuracy, specificity, precision/positive predictive value, recall/sensitivity, and F1 score (harmonic mean of precision and recall). Results: Thirty percent of patients in the cohort carried diagnoses of MCI/ADRD based on manual review. When evaluated on a held-out test set, the best model using clinical notes, ICDs, and medications, achieved an AUROC of 0.98, an AUPRC of 0.98, an accuracy of 0.93, a sensitivity (recall) of 0.91, a specificity of 0.96, a precision of 0.96, and an F1 score of 0.93 The estimated overall accuracy for patients randomly selected from EHRs was 99.88%. Conclusion: Automated EHR phenotyping accurately identifies patients with MCI/ADRD based on clinical notes, ICD codes, and medication records. This approach holds potential for large-scale MCI/ADRD research utilizing EHR databases.

Development and Validation of Electronic Health Record Measures of Safety Planning Practices as Part of Zero Suicide Implementation.

OBJECTIVE: Safety planning for suicide prevention is an important quality metric for Zero Suicide implementation. We describe the development, validation, and application of electronic health record (EHR) programs to measure uptake of safety planning practices across six integrated healthcare systems as part of a Zero Suicide evaluation study. METHODS: Safety planning was documented in narrative notes and structured EHR templates using the Stanley Brown Safety Planning Intervention (SBSPI) in response to a high-risk cutoff score on the Columbia Suicide Severity Rating Scale (CSSRS). Natural Language Processing (NLP) metrics were developed and validated using chart review to characterize practices documented in narrative notes. We applied NLP to measure frequency of documentation in the narrative text and standard programming methods to examine structured SBSPI templates from 2010-2022. RESULTS: Chart reviews found three safety planning practices documented in narrative notes that were delivered to at least half of patients at risk: professional contacts, lethal means counseling for firearms, and lethal means counseling for medication access/storage. NLP methods were developed to identify these practices in clinical text with high levels of accuracy (Sensitivity, Specificity, & PPV ≥ 82%). Among visits with a high-risk CSSRS, 40% (Range 2-73% by health system) had an SBSPI template within 1 year of implementation. CONCLUSIONS: This is one of the first reports describing development of measures that leverage electronic health records to track use of suicide prevention safety plans. There are opportunities to use the methods developed here in future evaluations of safety planning.

Measuring safety planning delivery in real-world systems to understand quality of suicide prevention care is challenging.Natural Language Processing (NLP) methods effectively identified some safety planning practices in electronic health records (EHR) from all notes ensuring a comprehensive measurement, but NLP will require updates/testing for local documentation practices.Structured safety planning templates in the EHR using the Stanley Brown Safety Planning Intervention improve ease and accuracy of measurement but may be less comprehensive than NLP for capturing all instances of safety planning documentation.

Developing a validated methodology for identifying clozapine treatment periods in electronic health records.

BACKGROUND: Clozapine is the only recommended antipsychotic medication for individuals diagnosed with treatment-resistant schizophrenia. Unfortunately, its wider use is hindered by several possible adverse effects, some of which are rare but potentially life threatening. As such, there is a growing interest in studying clozapine use and safety in routinely collected healthcare data. However, previous attempts to characterise clozapine treatment have had low accuracy. AIM: To develop a methodology for identifying clozapine treatment dates by combining several data sources and implement this on a large clinical database. METHODS: Non-identifiable electronic health records from a large mental health provider in London and a linked database from a national clozapine blood monitoring service were used to obtain information regarding patients' clozapine treatment status, blood tests and pharmacy dispensing records. A rule-based algorithm was developed to determine the dates of starting and stopping treatment based on these data, and more than 10% of the outcomes were validated by manual review of de-identified case note text. RESULTS: A total of 3,212 possible clozapine treatment periods were identified, of which 425 (13.2%) were excluded due to insufficient data to verify clozapine administration. Of the 2,787 treatments remaining, 1,902 (68.2%) had an identified start-date. On evaluation, the algorithm identified treatments with 96.4% accuracy; start dates were 96.2% accurate within 15 days, and end dates were 85.1% accurate within 30 days. CONCLUSIONS: The algorithm produced a reliable database of clozapine treatment periods. Beyond underpinning future observational clozapine studies, we envisage it will facilitate similar implementations on additional large clinical databases worldwide.

Five-Feature Models to Predict Preeclampsia Onset Time From Electronic Health Record Data: Development and Validation Study.

BACKGROUND:  Preeclampsia is a potentially fatal complication during pregnancy, characterized by high blood pressure and the presence of excessive proteins in the urine. Due to its complexity, the prediction of preeclampsia onset is often difficult and inaccurate. OBJECTIVE:  This study aimed to create quantitative models to predict the onset gestational age of preeclampsia using electronic health records. METHODS:  We retrospectively collected 1178 preeclamptic pregnancy records from the University of Michigan Health System as the discovery cohort, and 881 records from the University of Florida Health System as the validation cohort. We constructed 2 Cox-proportional hazards models: 1 baseline model using maternal and pregnancy characteristics, and the other full model with additional laboratory findings, vitals, and medications. We built the models using 80% of the discovery data, tested the remaining 20% of the discovery data, and validated with the University of Florida data. We further stratified the patients into high- and low-risk groups for preeclampsia onset risk assessment. RESULTS:  The baseline model reached Concordance indices of 0.64 and 0.61 in the 20% testing data and the validation data, respectively, while the full model increased these Concordance indices to 0.69 and 0.61, respectively. For preeclampsia diagnosed at 34 weeks, the baseline and full models had area under the curve (AUC) values of 0.65 and 0.70, and AUC values of 0.69 and 0.70 for preeclampsia diagnosed at 37 weeks, respectively. Both models contain 5 selective features, among which the number of fetuses in the pregnancy, hypertension, and parity are shared between the 2 models with similar hazard ratios and significant P values. In the full model, maximum diastolic blood pressure in early pregnancy was the predominant feature. CONCLUSIONS:  Electronic health records data provide useful information to predict the gestational age of preeclampsia onset. Stratification of the cohorts using 5-predictor Cox-proportional hazards models provides clinicians with convenient tools to assess the onset time of preeclampsia in patients.

A GPT-based EHR modeling system for unsupervised novel disease detection.

OBJECTIVE: To develop an Artificial Intelligence (AI)-based anomaly detection model as a complement of an "astute physician" in detecting novel disease cases in a hospital and preventing emerging outbreaks. METHODS: Data included hospitalized patients (n = 120,714) at a safety-net hospital in Massachusetts. A novel Generative Pre-trained Transformer (GPT)-based clinical anomaly detection system was designed and further trained using Empirical Risk Minimization (ERM), which can model a hospitalized patient's Electronic Health Records (EHR) and detect atypical patients. Methods and performance metrics, similar to the ones behind the recent Large Language Models (LLMs), were leveraged to capture the dynamic evolution of the patient's clinical variables and compute an Out-Of-Distribution (OOD) anomaly score. RESULTS: In a completely unsupervised setting, hospitalizations for Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) infection could have been predicted by our GPT model at the beginning of the COVID-19 pandemic, with an Area Under the Receiver Operating Characteristic Curve (AUC) of 92.2 %, using 31 extracted clinical variables and a 3-day detection window. Our GPT achieves individual patient-level anomaly detection and mortality prediction AUC of 78.3 % and 94.7 %, outperforming traditional linear models by 6.6 % and 9 %, respectively. Different types of clinical trajectories of a SARS-CoV-2 infection are captured by our model to make interpretable detections, while a trend of over-pessimistic outcome prediction yields a more effective detection pathway. Furthermore, our comprehensive GPT model can potentially assist clinicians with forecasting patient clinical variables and developing personalized treatment plans. CONCLUSION: This study demonstrates that an emerging outbreak can be accurately detected within a hospital, by using a GPT to model patient EHR time sequences and labeling them as anomalous when actual outcomes are not supported by the model. Such a GPT is also a comprehensive model with the functionality of generating future patient clinical variables, which can potentially assist clinicians in developing personalized treatment plans.