Methodological Considerations for Large Language Model-Based Symptom Extraction in Neuro-Oncology Electronic Health Records.

Implementation of an Automated Pediatric Malnutrition Screen Using Anthropometric Measurements in the Electronic Health Record

IntroductionAcross the U.S., the prevalence of opioid use disorder (OUD) and the rates of opioid overdoses have risen precipitously in recent years. Several effective medications for OUD (MOUD) exist and have been shown to be life-saving. A large volume of research has identified a confluence of factors that predict attrition and continued substance use during substance use disorder treatment. However, much of this literature has examined a small set of potential moderators or mediators of outcomes in MOUD treatment and may lead to over-simplified accounts of treatment non-adherence. Digital health methodologies offer great promise for capturing intensive, longitudinal ecologically-valid data from individuals in MOUD treatment to extend our understanding of factors that impact treatment engagement and outcomes.MethodsThis paper describes the protocol (including the study design and methodological considerations) from a novel study supported by the National Drug Abuse Treatment Clinical Trials Network at the National Institute on Drug Abuse (NIDA). This study (D-TECT) primarily seeks to evaluate the feasibility of collecting ecological momentary assessment (EMA), smartphone and smartwatch sensor data, and social media data among patients in outpatient MOUD treatment. It secondarily seeks to examine the utility of EMA, digital sensing, and social media data (separately and compared to one another) in predicting MOUD treatment retention, opioid use events, and medication adherence [as captured in electronic health records (EHR) and EMA data]. To our knowledge, this is the first project to include all three sources of digitally derived data (EMA, digital sensing, and social media) in understanding the clinical trajectories of patients in MOUD treatment. These multiple data streams will allow us to understand the relative and combined utility of collecting digital data from these diverse data sources. The inclusion of EHR data allows us to focus on the utility of digital health data in predicting objectively measured clinical outcomes.DiscussionResults may be useful in elucidating novel relations between digital data sources and OUD treatment outcomes. It may also inform approaches to enhancing outcomes measurement in clinical trials by allowing for the assessment of dynamic interactions between individuals' daily lives and their MOUD treatment response.Clinical Trial RegistrationIdentifier: NCT04535583.

BackgroundA commitment to Electronic Health Record (EHR) systems now constitutes a core part of many governments’ healthcare reform strategies. The resulting politically-initiated large-scale or national EHR endeavors are challenging because of their ambitious agendas of change, the scale of resources needed to make them work, the (relatively) short timescales set, and the large number of stakeholders involved, all of whom pursue somewhat different interests. These initiatives need to be evaluated to establish if they improve care and represent value for money.MethodsCritical reflections on these complexities in the light of experience of undertaking the first national, longitudinal, and sociotechnical evaluation of the implementation and adoption of England’s National Health Service’s Care Records Service (NHS CRS).Results/discussionWe advance two key arguments. First, national programs for EHR implementations are likely to take place in the shifting sands of evolving sociopolitical and sociotechnical and contexts, which are likely to shape them in significant ways. This poses challenges to conventional evaluation approaches which draw on a model of baseline operations → intervention → changed operations (outcome). Second, evaluation of such programs must account for this changing context by adapting to it. This requires careful and creative choice of ontological, epistemological and methodological assumptions.SummaryNew and significant challenges are faced in evaluating national EHR implementation endeavors. Based on experiences from this national evaluation of the implementation and adoption of the NHS CRS in England, we argue for an approach to these evaluations which moves away from seeing EHR systems as Information and Communication Technologies (ICT) projects requiring an essentially outcome-centred assessment towards a more interpretive approach that reflects the situated and evolving nature of EHR seen within multiple specific settings and reflecting a constantly changing milieu of policies, strategies and software, with constant interactions across such boundaries.

BackgroundHealth care databases are natural sources for estimating prevalence and incidence of chronic conditions, but substantial variation in estimates limits their interpretability and utility. We evaluated the effects of design choices when estimating prevalence and incidence in claims and electronic health record databases.MethodsPrevalence and incidence for five chronic diseases at increasing levels of expected frequencies, from cystic fibrosis to COPD, were estimated in the Clinical Practice Research Datalink (CPRD) and MarketScan databases from 2011 to 2014. Estimates were compared using different definitions of lookback time and contributed person-time.ResultsVariation in lookback time substantially affected estimates. In 2014, for CPRD, use of an all-time vs a 1-year lookback window resulted in 4.3–8.3 times higher prevalence (depending on disease), reducing incidence by 1.9–3.3 times. All-time lookback resulted in strong temporal trends. COPD prevalence between 2011 and 2014 in MarketScan increased by 25% with an all-time lookback but stayed relatively constant with a 1-year lookback. Varying observability did not substantially affect estimates.ConclusionThis framework draws attention to the underrecognized potential for widely varying incidence and prevalence estimates, with implications for care planning and drug development. Though prevalence and incidence are seemingly straightforward concepts, careful consideration of methodology is required to obtain meaningful estimates from health care databases.

An important goal of medical education is to teach students to use an electronic health record (EHR) safely and effectively. The purpose of this study is to examine medical student accounts of EHR use during their core inpatient clinical clerkships using a national sample. Paper health records (PHRs) are similarly examined. An online survey about health record use within the inpatient component of six core clerkships was administered to medical students after they completed Step 2 Clinical Knowledge of the United States Medical Licensing Examination. The sample included 17,202 U.S. medical students graduating between 2012 and 2016. Mean percentages of clerkships in which students engaged in various health record activities were computed, and analysis of variance was used to examine differences. The mean percentages of clerkships in which a student accessed or entered information into an EHR increased from 78% to 93% and 59% to 72%, respectively. For students who used an EHR, the mean percentage of clerkships in which they entered information remained constant at 76%. Students entered notes during the majority of their clerkships, with increases over time. However, students entered orders in less than a quarter of their clerkships, with decreases over time. The percentage of clerkships in which students used PHRs was lower and declining. Although students used an EHR in the majority of their inpatient core clerkships, they received limited educational experiences related to order and note writing, which could translate into a lack of preparedness for future training and practice.

Electronic health record data for assessing risk of hospitalization for COVID-19: Methodological considerations applied to multiple sclerosis

Few population-based systems exist to monitor child obesity prevalence in Canada. Data from electronic medical records (EMR) have been used in a small number of juridstictions worldwide to estimate obesity prevalence in children. To determine the frequency of height and weight documentation in EMRs in children, to describe the prevalence of child overweight and obesity, by age, and sex using data from the Electronic Medical Record Administrative data Linked Database (EMRALD) database in Ontario, and to determine if there are differences in prevalence based on visit type (well-child visit vs. other). We abstracted height and weight in children zero to 19 years of age in EMRALD who had at least one well-child visit from January 2010 to December 2011. Using the most recent visit with both a documented height and weight, we reported the proportion and 95% CIs of subjects defined as overweight, and obese, by age group and sex, using the WHO growth reference standards. We compared the proportion or overweight and obese children by visit type for all age groups, using χ2 tests. There were 28,083 well-child visits in 7705 children over this study period. 84.7% of children who attended well-child visits had both a height and weight documented. The prevalence of overweight and obesity, varied by age group from 12% to 32%, and 2% to 12%, respectively. Obesity rates were significantly higher in one- to four-year-olds compared to children <1 year of age (6.1% vs. 2.3%), and in 10- to 14-year-olds compared to five- to nine-year-olds (12% vs. 9%). Both one- to four-year-old (7.2% vs. 4.9%) and 10- to 14-year-old boys (14.5% vs. 9.6%) had higher obesity rates. The proportion of overweight and obese children was higher using heights and weights reported from other child visit types, compared to well-child visits, for all age groups (P<0.05), except for children less than one year of age (P=0.45). We documented a high rate of overweight and obesity in children. EMR may be a useful tool to conduct population-based surveillance of child overweight and obesity in Canada. The selection of visit type may be an important methodological consideration.

Purpose: Retrospective chart review (RCR) is the process of manual patient data review to answer research questions. Large study populations and heterogeneous data make this a tedious, biased and error-prone process1. The authors therefore designed and developed ChartSweep, a HIPAA-compliant Windows (Microsoft, WA, USA) application leveraging the Python coding language to streamline and expedite the RCR process while remaining faithful to its methodological rigor as outlined by Matt and Matthew. ChartSweep is open-source and can be customized for use with any electronic medical record system as part of any study requiring retrospective chart review. Methods: ChartSweep is a tool developed at the Massachusetts General Hospital. It uses the Selenium Python library to pull information from electronic medical records and securely store it in.csv,.txt,.pdf or.jpeg format. As a proof-of-concept, a retrospective review of 172 patient records stored in Epic (electronic medical record storage) was performed to identify subjects who had undergone radiofrequency ablation (RFA) of the greater or lesser occipital nerves (for treatment of migraine headache). The first search was conducted manually according to standard RCR methodology, the second automatically using ChartSweep. Automated ChartSweep output was then reviewed and patient charts describing RFA in other contexts (lumbar ablation, endometrial ablation) were manually excluded. Total time required for each review and discrepancies between data output were evaluated. Results: Overall manual review time was 1,371 minutes (23 hours) with a mean evaluation time per medical record of 8 minutes. Automated ChartSweep review was significantly faster requiring 56 minutes overall, and 0.3 minutes per patient record (P< 0.0001). Time saved was 7.6 minutes per chart and 1,315 minutes (21.9 hours) total. Both reviews identified 16 patients who had undergone RFA out of 172 total patients. Conclusion: Open-source Python libraries as leveraged by ChartSweep significantly accelerate the retrospective chart review process in plastic surgery research. Quality of data review is not compromised. Further analyses with larger study populations are required to validate ChartSweep as a reliable research tool. 1Matt, V., Matthew, H. J. J. o. e. e. f. h. p. The retrospective chart review: important methodological considerations. 2013;10.

BackgroundA major aim of the i2b2 (informatics for integrating biology and the bedside) clinical data informatics framework aims to create an efficient structure within which patients can be identified for clinical and translational research projects.Our objective was to describe the respective roles of the i2b2 research query tool and the electronic medical record (EMR) in conducting a case-controlled clinical study at our institution.MethodsWe analyzed the process of using i2b2 and the EMR together to generate a complete research database for a case–control study that sought to examine risk factors for kidney stones among gastrostomy tube (G-tube) fed children.ResultsOur final case cohort consisted of 41/177 (23%) of potential cases initially identified by i2b2, who were matched with 80/486 (17%) of potential controls. Cases were 10 times more likely to be excluded for inaccurate coding regarding stones vs. inaccurate coding regarding G-tubes. A majority (67%) of cases were excluded due to not meeting clinical inclusion criteria, whereas a majority of control exclusions (72%) occurred due to inadequate clinical data necessary for study completion. Full dataset assembly required complementary information from i2b2 and the EMR.Conclusionsi2b2 was critical as a query analysis tool for patient identification in our case–control study. Patient identification via procedural coding appeared more accurate compared with diagnosis coding. Completion of our investigation required iterative interplay of i2b2 and the EMR to assemble the study cohort.

BackgroundAddressing Patient Reported Outcomes (PROs) is essential for patient-centred care, shared decision making and improved health outcomes. Value-based health care systems in New South Wales (NSW) have a growing focus on collecting and using PROs that matter most to patients to improve their healthcare outcomes. Developing oral health patient reported outcomes measures (OH-PROM) is a first step towards value-based oral health care. This paper describes the development process of an adult and child OH-PROM tool that can be piloted for NSW public dental patients.MethodsAn expert panel was assembled to undertake a systematic process of developing OH-PROMs for NSW Health. Key methodological considerations included: (1) forming an expert panel to specify the target population and context of implementation, (2) rapid literature review and environmental scan to identify existing validated OH-PROM tools for adults and children. (3) consensus gathering with the expert panel (4) consumer feedback, and (5) finalisation of the tool for electronic oral health record (eOHR) integration to establish a set of questions, that were relevant, context-appropriate, and important to oral healthcare outcomes for patients using public dental services.ResultsThe panel considered a total of 59 questions from two child (15), and four adult (44) Oral Health Related Quality of Life (OHRQoL) questionnaires used to collect OH-PROMs. These questions were mapped to the four key dimensions of OHRQoL for OH-PROMs: Oral Function, Orofacial Pain, Orofacial Appearance, and Psychosocial Impact. The consensus resulted in seven questions that aligned with these four dimensions to form two new NSW OH-PROM tools: one for adults and one for children. The tools were tested with consumers for understandability and usefulness before being incorporated into the electronic oral health record system, in readiness for future pilot testing.ConclusionThe process for developing new OH-PROMs for NSW public dental services took a pragmatic approach that combined literature appraisal, expert consensus, and consumer consultation. Future work will assess the implementation of the OH-PROM tool and test its validity for broader use as an outcome measure for value-based oral healthcare.

Facilitating biomedical researchers’ interrogation of electronic health record data: Ideas from outside of biomedical informatics

Real-world evidence (RWE) is increasingly used to complement findings from randomized controlled trials (RCTs), contextualizing the effectiveness and safety of medical interventions as delivered in routine clinical practice. Advances in the curation and accessibility of electronic health record (EHR) data present the opportunity to utilize real-world data (RWD) to investigate therapeutic areas including oncology, where administrative healthcare claims databases alone are often not fit-for-purpose. The RCT DUPLICATE initiative has previously evaluated when RWE can most appropriately draw causal conclusions by emulating trials for nononcology indications. Here, we present the design and trial selection for the emulation of comparative oncology trials with real-world evidence (ENCORE) project, which extends this work to oncology. ENCORE is designed to emulate 12 RCTs in four oncology-specialized EHR databases across four different cancer indications, specifically non-small-cell lung cancer, breast cancer, colorectal cancer, and multiple myeloma. It will place special emphasis on systematic evaluation of fitness of data in relation to the study design and statistical analysis for a particular research question and preregistration of study protocols prior to initiation and analysis. Prespecified criteria will assess agreement of treatment effect estimates between RCTs and their respective emulations. Through extensive sensitivity analyses benchmarked against RCT results, the ENCORE project aims to inform understanding of how measurement, design, and analytic decisions influence the interpretation of results from emulated oncology trials using RWD.

This paper presents the history of the use of the computer for maintaining patient medical care information. An electronic record generated with a computer, which is non-specific for any healthcare specialty, is referred to as the electronic health record. The electronic health record was previously called the computer-based patient record. "Electronic" replaced the earlier term "computer-based" because "electronic" better describes the medium in which the patient record is managed. The electronic health record and its application to dentistry are discussed. The electronic health record is a "database" of patient information that has been entered by any healthcare provider; the electronic oral health record is an "electronic record" of oral health information that has been entered by an oral healthcare provider. The significant differences between the electronic health record and the electronic oral health record are outlined and highlighted. Included is a template describing a procedure to be used by dental personnel during the decision making process of purchasing an electronic oral health record. A brief description of a practice template is also provided. These completed templates can be shared with dental software vendors to clarify their understanding of and to clearly describe the needs of today's dental practice. The challenge of introducing information technology into educational institutions' curricula is identified. Finally, the potential benefit of using electronic technology for managing oral healthcare information is outlined.

Real-world studies (RWSs) have emerged as a transformative force in oncology research, complementing traditional randomized controlled trials (RCTs) by providing comprehensive insights into cancer care within routine clinical settings. This review examines the evolving landscape of RWSs in oncology, focusing on their implementation, methodological considerations, and impact on precision medicine. We systematically analyze how RWSs leverage diverse data sources, including electronic health records (EHRs), insurance claims, and patient registries, to generate evidence that bridges the gap between controlled clinical trials and real-world clinical practice. The review underscores the key contributions of RWSs, including capturing therapeutic outcomes in traditionally underrepresented populations, expanding drug indications, and evaluating long-term safety and effectiveness in routine clinical settings. While acknowledging significant challenges, including data quality variability and privacy concerns, we discuss how emerging technologies like artificial intelligence are helping to address these limitations. The integration of RWSs with traditional clinical research is revolutionizing the paradigm of precision oncology and enabling more personalized treatment approaches based on real-world evidence.

Introduction