Electronic Health Record Data Sources Research Articles

Evaluation of Real-World Tumor Response Derived From Electronic Health Record Data Sources: A Feasibility Analysis in Patients With Metastatic Non-Small Cell Lung Cancer Treated With Chemotherapy.

Real-world data (RWD) holds promise for ascribing a real-world (rw) outcome to a drug intervention; however, ascertaining rw-response to treatment from RWD can be challenging. Friends of Cancer Research formed a collaboration to assess available data attributes related to rw-response across RWD sources to inform methods for capturing, defining, and evaluating rw-response. This retrospective noninterventional (observational) study included seven electronic health record data companies (data providers) providing summary-level deidentified data from 200 patients diagnosed with metastatic non-small cell lung cancer (mNSCLC) and treated with first-line platinum doublet chemotherapy following a common protocol. Data providers reviewed the availability and frequency of data components to assess rw-response (ie, images, radiology imaging reports, and clinician response assessments). A common protocol was used to assess and report rw-response end points, including rw-response rate (rwRR), rw-duration of response (rwDOR), and the association of rw-response with rw-overall survival (rwOS), rw-time to treatment discontinuation (rwTTD), and rw-time to next treatment (rwTTNT). The availability and timing of clinician assessments was relatively consistent across data sets in contrast to images and image reports. Real-world response was analyzed using clinician response assessments (median proportion of patients evaluable, 77.5%), which had the highest consistency in the timing of assessments. Relative consistency was observed across data sets for rwRR (median 46.5%), as well as the median and directionality of rwOS, rwTTD, and rwTTNT. There was variability in rwDOR across data sets. This collaborative effort demonstrated the feasibility of aligning disparate data sources to evaluate rw-response end points using clinician-documented responses in patients with mNSCLC. Heterogeneity exists in the availability of data components to assess response and related rw-end points, and further work is needed to inform drug effectiveness evaluation within RWD sources.

Methods for Identification of Blood Product Transfusions in a Cohort of Myelofibrosis Patients: An Assessment of Data Capture Using Insurance Claims and Electronic Health Record Abstraction

Introduction Transfusion data completeness is critical in assessing real-world response and progression in myeloid malignancies. In the real-world setting transfusion data can be collected via insurance claims or by Electronic Health Record (EHR) abstraction. Transfusion events captured from claims have previously been shown to be 71% sensitive and 92% specific compared to EHR abstraction in the community hospital setting, however this has not been evaluated in outpatient community oncology setting (Howard DH, et al. Transfusion Medicine. 2016;26:457-459). Our objective was to assess the completeness of transfusion data from claims in the The Komodo Healthcare Map™(KHM) to EHR abstraction in the Flatiron Health Research Database (FHRD) in a group of patients with myelofibrosis (MF). Methods This study used the nationwide Flatiron Health EHR-derived de-identified database and The Komodo Healthcare Map™, a database of de-identified claim-based healthcare encounters from insured patients in the US (2015-2022).This is a cross-sectional study of a cohort of patients with MF that underwent EHR abstraction (EHRC) (n = 507) from FHRD, capturing transfusion dates, and number of units from date of diagnosis up to data cutoff on 30 November 2022 (observation window). Those patients were linked to KHM to create a Linked Claims Cohort (LCC), and a sub-cohort of patients with any closed claims during the observation window to make a Closed Claims Cohort (CCC), where transfusions were assessed by reviewing transfusion-related HCPCS and ICD codes. Analyses were performed on the matched patients in the LCC and CCC, respectively. To assess the level of concordance between claims and EHR data sources on the number of transfusions per patient the intraclass correlation coefficient (ICC) was used. Also, percent overlap and sensitivity, specificity, and positive and negative predictive values were calculated. Results Out of a total of 507 in the EHRC, 498 were found in KHM and included in the LCC, and 245 had closed claims and comprised the CCC. Demographics are listed below (Table 1) and are similar between groups, however the CCC had a higher percentage of patients with more recent diagnosis. A total of 319 (64%) vs 168 (34%) in the LCC, and 143 (58%) vs 95 (39%) patients in the CCC had at least one transfusion identified by EHR and claims data sources, respectively (Table 2). In the LCC, there was 64.1% agreement of having a transfusion across data sources, with a kappa of 0.34 (p = 0), and in the CCC, there was 72.24% agreement, with a kappa of 0.47 (p = 0). The unweighted kappa for the number of transfusions recorded during the observation period was 0.11 (p = 0) in the LCC and 0.18 (p = 0) in the CCC. Treating the EHR source as the source of truth, the claims data source in LCC had a sensitivity of 48.28%, specificity of 92.18%, PPV of 91.67%, and NPV of 50% to identify transfusion events, while the claims in CCC had a sensitivity of 59.44%, specificity of 90.20%, PPV of 89.47%, and NPV of 61.33% Conclusions There was an observed low agreement between transfusions identified via EHR abstraction and Claims data sources, with the EHR having more transfusions identified, but a moderate agreement when selecting for patients with closed claims. As a source for transfusion data EHR provided more data points during the observation window for this group of patients. Future investigation of transfusions in patients with continuous closed claims during the observation window is warranted to determine if it improves data capture. This study demonstrates that EHR abstraction may have higher utility for identifying transfusions with lower attrition in a cohort of MF patients with long observation periods, and with variable insurance coverage in the outpatient oncology setting.

Forms or Free-Text? Measuring Advance Care Planning Activity Using Electronic Health Records.

Advance care planning (ACP) discussions seek to guide future serious illness care. These discussions may be recorded in the electronic health record by documentation in clinical notes, structured forms and directives, and physician orders. Yet, most studies of ACP prevalence have only examined structured electronic health record elements and ignored data existing in notes. We sought to investigate the relative comprehensiveness and accuracy of ACP documentation from structured and unstructured electronic health record data sources. We evaluated structured and unstructured ACP documentation present in the electronic health records of 435 patients with cancer drawn from three separate healthcare systems. We extracted structured ACP documentation by manually annotating written documents and forms scanned into the electronic health record. We coded unstructured ACP documentation using a rule-based natural language processing software that identified ACP keywords within clinical notes and was subsequently reviewed for accuracy. The unstructured approach identified more instances of ACP documentation (238, 54.7% of patients) than the structured ACP approach (187, 42.9% of patients). Additionally, 16.6% of all patients with structured ACP documentation only had documents that were judged as misclassified, incomplete, blank, unavailable, or a duplicate of a previously entered erroneous document. ACP documents scanned into electronic health records represent a limited view of ACP activity. Research and measures of clinical practice with ACP should incorporate information from unstructured data.

Ambient greenness, access to local green spaces, and subsequent mental health: a 10-year longitudinal dynamic panel study of 2·3 million adults in Wales

Living in greener areas, or close to green and blue spaces (GBS; eg, parks, lakes, or beaches), is associated with better mental health, but longitudinal evidence when GBS exposures precede outcomes is less available. We aimed to analyse the effect of living in or moving to areas with more green space or better access to GBS on subsequent adult mental health over time, while explicitly considering health inequalities. A cohort of the people in Wales, UK (≥16 years; n=2 341 591) was constructed from electronic health record data sources from Jan 1, 2008 to Oct 31, 2019, comprising 19 141 896 person-years of follow-up. Household ambient greenness (Enhanced Vegetation Index [EVI]), access to GBS (counts, distance to nearest), and common mental health disorders (CMD, based on a validated algorithm combining current diagnoses or symptoms of anxiety or depression [treated or untreated in the preceding 1-year period], or treatment of historical diagnoses from before the current cohort [up to 8 years previously, to 2000], where diagnosis preceded treatment) were record-linked. Cumulative exposure values were created for each adult, censoring for CMD, migration out of Wales, death, or end of cohort. Exposure and CMD associations were evaluated using multivariate logistic regression, stratified by area-level deprivation. After adjustment, exposure to greater ambient greenness over time (+0·1 increased EVI on a 0-1 scale) was associated with lower odds of subsequent CMD (adjusted odds ratio 0·80, 95% CI 0·80-0·81), where CMD was based on a combination of current diagnoses or symptoms (treated or untreated in the preceding 1-year period), or treatments. Ten percentile points more access to GBS was associated with lower odds of a later CMD (0·93, 0·93-0·93). Every additional 360 m to the nearest GBS was associated with higher odds of CMD (1·05, 1·04-1·05). We found that positive effects of GBS on mental health appeared to be greater in more deprived quintiles. Ambient exposure is associated with the greatest reduced risk of CMD, particularly for those who live in deprived communities. These findings support authorities responsible for GBS, who are attempting to engage planners and policy makers, to ensure GBS meets residents' needs. National Institute for Health and Care Research Public Health Research programme.

Generalizability of an EHR-network dataset to the United States for cardiovascular disease conditions: Comparison of Cerner real world data with the national inpatient sample

Electronic Health Record (EHR) data from health systems are increasingly being combined for clinical research purposes. Yet, it remains unclear whether these large EHR data sources provide a representative assessment of national disease prevalence and treatment. To evaluate this, we compared Cerner RealWorldData (CRWD), a large EHR data source, to those seen in the National Inpatient Sample (NIS) for three cardiovascular conditions (myocardial infarction (MI), congestive heart failure (CHF), and stroke. Adult patients (age ≥18 years) hospitalized with MI, CHF, and stroke were identified in both CRWD (86 health systems) and the NIS (4,782 hospitals). Patient demographics, comorbidities, procedures, outcomes (length of stay and in-hospital mortality) and hospital type (teaching or non-teaching) were compared between NIS and CRWD patients. Of 86 health systems participating in CRWD, 33 were excluded for potential data quality issues which accounted for about 11% of hospitalizations in the dataset, leaving 53 for inclusion in analysis which accounted for about 89% of hospitalizations in the dataset. Between January 1st 2017 and December 31st 2018, 116,956 MI, 188,107 CHF, and 93,968 stroke hospitalizations were identified in CRWD versus 2,245,300 MI, 4,310,745 CHF, and 1,333,480 stroke hospitalizations in the NIS. Patient demographics were similar among patients in CWRD and the NIS for all three cardiovascular groups except for ethnicity, with underrepresentation of Hispanic individuals in CRWD vs the NIS. Patients hospitalized in CRWD had a slightly higher proportion of coded co-morbidities compared with NIS hospitalizations due to a longer potential look-back period. For patients with MI, hospital mortality, length of stay, coronary artery bypass graft (CABG) rates, and percutaneous coronary intervention (PCI) rates were similar between CRWD and NIS. Additionally, there was similar in hospital mortality and length of stay for those with CHF and stroke hospitalizations between CRWD and NIS. On aggregate, characteristics of hospitalizations for MI, CHF, and stroke using EHR data from one nationwide EHR-derived database, CRWD, appears similar to characteristics of hospitalizations in the nationally representative NIS. Important limitations of CRWD include lack of geographic representativeness, under-representation of Hispanic adults, and the need to exclude health systems for missing data.

Incident atrial fibrillation and adverse clinical outcomes during extended follow-up of participants recruited to the remote heart rhythm sampling using the AliveCor heart monitor to screen for atrial fibrillation: the REHEARSE-AF study.

Atrial fibrillation (AF) is an important risk factor for stroke, which is commonly asymptomatic, particularly in older patients, and often undetected until cardiovascular events occur. Development of novel technology has helped to improve detection of AF. However, the longer-term benefit of systematic electrocardiogram (ECG) screening on cardiovascular outcomes is unclear. In the original REHEARSE-AF study, patients were randomized to twice-weekly portable electrocardiogram (iECG) assessment or routine care. After discontinuing the trial portable iECG assessment, electronic health record data sources provided longer-term follow-up analysis. Cox regression was used to provide unadjusted and adjusted hazard ratios (HR) [95% confidence intervals (CI)] for clinical diagnosis, events, and anticoagulant prescriptions during the follow-up period. Over the median 4.2-year follow-up, although a greater number of patients were diagnosed with AF in the original iECG group (43 vs. 31), this was not significant (HR 1.37, 95% CI 0.86-2.19). No differences were seen in the number of strokes/systemic embolisms or deaths between the two groups (HR 0.92, 95% CI 0.54-1.54; HR 1.07, 95% CI 0.66-1.73). Findings were similar when restricted to those with CHADS-VASc ≥ 4. A 1-year period of home-based, twice-weekly screening for AF increased diagnoses of AF for the screening period but did not lead to increased diagnoses of AF or a reduction in cardiovascular-related events or all-cause death over a median of 4.2 years, even in those at highest risk of AF. These results suggest that benefits of regular ECG screening over a 1-year period are not maintained after cessation of the screening protocol.

Suicide in Veterans with Post-Traumatic Headache: An Analysis of Health Systems Data (P14-12.001)

Suicide in Veterans with Post-Traumatic Headache: An Analysis of Health Systems Data (P14-12.001)

Large-scale evidence generation and evaluation across a network of databases for type 2 diabetes mellitus (LEGEND-T2DM): a protocol for a series of multinational, real-world comparative cardiovascular effectiveness and safety studies

IntroductionTherapeutic options for type 2 diabetes mellitus (T2DM) have expanded over the last decade with the emergence of cardioprotective novel agents, but without such data for older drugs, leaving a...

Identifying Patients With Hypoglycemia Using Natural Language Processing: Systematic Literature Review.

BackgroundAccurately identifying patients with hypoglycemia is key to preventing adverse events and mortality. Natural language processing (NLP), a form of artificial intelligence, uses computational algorithms to extract information from text data. NLP is a scalable, efficient, and quick method to extract hypoglycemia-related information when using electronic health record data sources from a large population.ObjectiveThe objective of this systematic review was to synthesize the literature on the application of NLP to extract hypoglycemia from electronic health record clinical notes.MethodsLiterature searches were conducted electronically in PubMed, Web of Science Core Collection, CINAHL (EBSCO), PsycINFO (Ovid), IEEE Xplore, Google Scholar, and ACL Anthology. Keywords included hypoglycemia, low blood glucose, NLP, and machine learning. Inclusion criteria included studies that applied NLP to identify hypoglycemia, reported the outcomes related to hypoglycemia, and were published in English as full papers.ResultsThis review (n=8 studies) revealed heterogeneity of the reported results related to hypoglycemia. Of the 8 included studies, 4 (50%) reported that the prevalence rate of any level of hypoglycemia was 3.4% to 46.2%. The use of NLP to analyze clinical notes improved the capture of undocumented or missed hypoglycemic events using International Classification of Diseases, Ninth Revision (ICD-9), and International Classification of Diseases, Tenth Revision (ICD-10), and laboratory testing. The combination of NLP and ICD-9 or ICD-10 codes significantly increased the identification of hypoglycemic events compared with individual methods; for example, the prevalence rates of hypoglycemia were 12.4% for International Classification of Diseases codes, 25.1% for an NLP algorithm, and 32.2% for combined algorithms. All the reviewed studies applied rule-based NLP algorithms to identify hypoglycemia.ConclusionsThe findings provided evidence that the application of NLP to analyze clinical notes improved the capture of hypoglycemic events, particularly when combined with the ICD-9 or ICD-10 codes and laboratory testing.

Practice habits in anesthesiology resident education: The impact of feedback using data analytics

BackgroundThe Accreditation Council for Graduate Medical Education (ACGME) has annually asked residents about several aspects of their education including whether they have received data about practice habits. However, the term “practice habits” is not clearly defined, allowing for varied interpretation. A working definition of practice habits was developed at the Renaissance School of Medicine Department of Anesthesiology at Stony Brook University, guided by educational program leadership utilizing information retrievable from the electronic health record (EHR). We used this information to provide practice habits data to anesthesiology residents. We hypothesized that the provision of practice habits feedback would lead to increased compliance on ACGME surveys. MethodsSeveral parameters were extracted from the EHR of every resident surgical case during their training years from 2017 to 2019. Each resident received feedback on those measures and de-identified aggregate data on trainees at the same training level. The ACGME survey data on practice habits were compared pre- to post-implementation of feedback provision. ResultsThe program scored compliance rates above the national mean for two consecutive years after implementation of the new feedback system: 79% vs 71% in 2017–2018, and 95% vs 71% in 2018–2019. ConclusionsExisting EHR data sources may serve as useful resources for residency programs to provide real-world feedback to trainees on their practice habits and inform program leadership about trainee progress and program gaps. Such feedback allows for peer-comparisons and identification of gaps in skills training and facilitates improved program compliance rate in the ACGME survey.

A Novel Patient Values Tab for the Electronic Health Record: A User-Centered Design Approach.

BackgroundThe COVID-19 pandemic has shined a harsh light on a critical deficiency in our health care system: our inability to access important information about patients’ values, goals, and preferences in the electronic health record (EHR). At Memorial Sloan Kettering Cancer Center (MSK), we have integrated and systematized health-related values discussions led by oncology nurses for newly diagnosed cancer patients as part of routine comprehensive cancer care. Such conversations include not only the patient’s wishes for care at the end of life but also more holistic personal values, including sources of strength, concerns, hopes, and their definition of an acceptable quality of life. In addition, health care providers use a structured template to document their discussions of patient goals of care.ObjectiveTo provide ready access to key information about the patient as a person with individual values, goals, and preferences, we undertook the creation of the Patient Values Tab in our center’s EHR to display this information in a single, central location. Here, we describe the interprofessional, interdisciplinary, iterative process and user-centered design methodology that we applied to build this novel functionality as well as our initial implementation experience and plans for evaluation.MethodsWe first convened a working group of experts from multiple departments, including medical oncology, health informatics, information systems, nursing informatics, nursing education, and supportive care, and a user experience designer. We conducted in-depth, semistructured, audiorecorded interviews of over 100 key stakeholders. The working group sought consensus on the tab’s main content, homing in on high-priority areas identified by the stakeholders. The core content was mapped to various EHR data sources. We established a set of high-level design principles to guide our process. Our user experience designer then created wireframes of the tab design. The designer conducted usability testing with physicians, nurses, and other health professionals. Data validation testing was conducted.ResultsWe have already deployed the Patient Values Tab to a pilot sample of users in the MSK Gastrointestinal Medical Oncology Service, including physicians, advanced practice providers, nurses, and administrative staff. We have early evidence of the positive impact of this EHR innovation. Audit logs show increasing use. Many of the initial user comments have been enthusiastically positive, while others have provided constructive suggestions for additional tab refinements with respect to format and content.ConclusionsIt is our challenge and obligation to enrich the EHR with information about the patient as a person. Realization of this capability is a pressing public health need requiring the collaboration of technological experts with a broad range of clinical leaders, users, patients, and families to achieve solutions that are both principled and practical. Our new Patient Values Tab represents a step forward in this important direction.

S0515 The Burden of Chronic Constipation and Its Treatment Among U.S. Nursing Home Residents

INTRODUCTION: Little is known about the current burden of chronic constipation (CC) among nursing home (NH) residents. This study describes the epidemiology of CC and utilization of drug therapy for constipation (DTC). METHODS: We conducted two retrospective cohort studies of long-stay NH residents ≥ 65 years utilizing (a) electronic health record (EHR) data from 2016 for 126 U.S. NHs [EHR cohort], and (b) national Medicare claims from 2014-2016; [Medicare cohort]. These data were linked with federally-mandated Minimum Data Set (MDS) resident assessments to obtain resident characteristics. The baseline period was defined as the 100 days prior to the index date (NH long-stay qualification date). Follow-up continued until death or the last available MDS assessment or drug order. CC was defined by 1) chronic DTC use, and/or 2) the MDS indicator for constipation. We calculated CC prevalence, baseline characteristics, and DTC use, as well as hospitalizations and estimated costs for the Medicare cohort. RESULTS: In the EHR cohort, 25,739 residents (71.8%) met the CC definition. Mean follow-up was 0.6 years, and the prevalence rate was 1.2 CC episodes per person-year (Table 1). Among individuals with prevalent CC, 86% received a DTC for an average of 18.7 days per person-month. In the Medicare cohort, 245,578 residents (37.5%) had CC. Mean follow-up was 1.5 years, with a prevalence rate of 0.23 per person-year. Relative to the EHR cohort, fewer individuals with prevalent CC (62%) received a DTC during an average of 9.7 days per person-month of follow-up; utilization also differed (Table 2). ∼8% had a constipation-related hospitalization during follow-up, with paralytic ileus being the most frequent diagnosis. The average cost of constipation-related admissions accounted for 16% of the average cost for all-cause hospitalizations. CONCLUSION: Relying on Medicare data alone likely undercounts CC cases and DTC use because Part D claims lack the OTC drug data that characterizes most constipation treatment. For this reason, EHR data sources are preferred for CC research in NH settings. CC and cognitive impairment frequently co-occur, adding to the burden of treatment that includes OTC laxative polypharmacy and time-consuming administration of products not well-accepted by NH residents with dementia (e.g. laxative powders requiring mixture with liquid, suppositories). An individualized approach to CC treatment could be more effective and reduce severe adverse events.Table 1Table 2

Approaches for combining primary care electronic health record data from multiple sources: a systematic review of observational studies

ObjectiveTo identify observational studies which used data from more than one primary care electronic health record (EHR) database, and summarise key characteristics including: objective and rationale for using multiple data...

Identifying children with Cystic Fibrosis in population-scale routinely collected data in Wales: A Retrospective Review.

IntroductionThe challenges in identifying a cohort of people with a rare condition can be addressed by routinely collected, population-scale electronic health record (EHR) data, which provide large volumes of data at a national level. This paper describes the challenges of accurately identifying a cohort of children with Cystic Fibrosis (CF) using EHR and their validation against the UK CF Registry.ObjectivesTo establish a proof of principle and provide insight into the merits of linked data in CF research; to identify the benefits of access to multiple data sources, in particular the UK CF Registry data, and to demonstrate the opportunity it represents as a resource for future CF research.MethodsThree EHR data sources were used to identify children with CF born in Wales between 1st January 1998 and 31st August 2015 within the Secure Anonymised Information Linkage (SAIL) Databank. The UK CF Registry was later acquired by SAIL and linked to the EHR cohort to validate the cases and explore the reasons for misclassifications.ResultsWe identified 352 children with CF in the three EHR data sources. This was greater than expected based on historical incidence rates in Wales. Subsequent validation using the UK CF Registry found that 257 (73%) of these were true cases. Approximately 98.7% (156/158) of individuals identified as CF cases in all three EHR data sources were confirmed as true cases; but this was only the case for 19.8% (20/101) of all those identified in just a single data source. ConclusionIdentifying health conditions in EHR data can be challenging, so data quality assurance and validation is important or the merit of the research is undermined. This retrospective review identifies some of the challenges in identifying CF cases and demonstrates the benefits of linking cases across multiple data sources to improve quality.

Using and improving distributed data networks to generate actionable evidence: the case of real-world outcomes in the Food and Drug Administration's Sentinel system.

The US Food and Drug Administration (FDA) Sentinel System uses a distributed data network, a common data model, curated real-world data, and distributed analytic tools to generate evidence for FDA decision-making. Sentinel system needs include analytic flexibility, transparency, and reproducibility while protecting patient privacy. Based on over a decade of experience, a critical system limitation is the inability to identify enough medical conditions of interest in observational data to a satisfactory level of accuracy. Improving the system's ability to use computable phenotypes will require an "all of the above" approach that improves use of electronic health data while incorporating the growing array of complementary electronic health record data sources. FDA recently funded a Sentinel System Innovation Center and a Community Building and Outreach Center that will provide a platform for collaboration across disciplines to promote better use of real-world data for decision-making.

Evaluating the Scope of Clinical Electronic Messaging to Coordinate Care in a Breast Cancer Cohort.

Care coordination has received attention as an opportunity to improve healthcare delivery. Current work to quantify provider coordination has primarily relied on identifying shared patients, but neglects to understand communication patterns. We applied social network analysis to electronic health record (EHR) secure messaging data to compare networks of providers who share patients and networks of providers who communicate about patients. We studied 2175 stage I-III breast cancer patients who received outpatient treatment from 1758 providers at a large academic medical center in the southeastern United States. Patients in our cohort were involved in 94324 appointments and were the subject of 307144 message threads. We found that 9.9% of provider-provider pairs that shared patients were mutually involved in electronic communication about their patients. EHR data sources can be used to evaluate provider communication across a clinical enterprise, which can help identify opportunities to improve collaboration and reduce provider burnout.

Development of a Controlled Vocabulary-Based Adverse Drug Reaction Signal Dictionary for Multicenter Electronic Health Record-Based Pharmacovigilance.

Integration of controlled vocabulary-based electronic health record (EHR) observational data is essential for real-time large-scale pharmacovigilance studies. To provide a semantically enriched adverse drug reaction (ADR) dictionary for post-market drug safety research and enable multicenter EHR-based extensive ADR signal detection and evaluation, we developed a comprehensive controlled vocabulary-based ADR signal dictionary (CVAD) for pharmacovigilance. A CVAD consists of (1) administrative disease classifications of the International Classification of Diseases (ICD) codes mapped to the Medical Dictionary for Regulatory Activities Preferred Terms (MedDRA® PTs); (2) two teaching hospitals' codes for laboratory test results mapped to the Logical Observation Identifiers Names and Codes (LOINC) terms and MedDRA® PTs; and (3) clinical narratives and ADRs encoded by standard nursing statements (encoded by the International Classification for Nursing Practice [ICNP]) mapped to the World Health Organization-Adverse Reaction Terminology (WHO-ART) terms and MedDRA® PTs. Of the standard 4514 MedDRA® PTs from Side Effect Resources (SIDER) 4.1, 1130 (25.03%), 942 (20.86%), and 83 (1.83%) terms were systematically mapped to clinical narratives, laboratory test results, and disease classifications, respectively. For the evaluation, we loaded multi-source EHR data. We first performed a clinical expert review of the CVAD clinical relevance and a three-drug ADR case analyses consisting of linezolid-induced thrombocytopenia, warfarin-induced bleeding tendency, and vancomycin-induced acute kidney injury. CVAD had a high coverage of ADRs and integrated standard controlled vocabularies to the EHR data sources, and researchers can take advantage of these features for EHR observational data-based extensive pharmacovigilance studies to improve sensitivity and specificity.

Impact of guidelines for the management of minor head injury on the utilization and diagnostic yield of CT over two decades, using natural language processing in a large dataset

ObjectivesWe investigated the impact of clinical guidelines for the management of minor head injury on utilization and diagnostic yield of head CT over two decades.MethodsRetrospective before-after study using multiple electronic health record data sources. Natural language processing algorithms were developed to rapidly extract indication, Glasgow Coma Scale, and CT outcome from clinical records, creating two datasets: one based on all head injury CTs from 1997 to 2009 (n = 9109), for which diagnostic yield of intracranial traumatic findings was calculated. The second dataset (2009–2014) used both CT reports and clinical notes from the emergency department, enabling selection of minor head injury patients (n = 4554) and calculation of both CT utilization and diagnostic yield. Additionally, we tested for significant changes in utilization and yield after guideline implementation in 2011, using chi-square statistics and logistic regression.ResultsThe yield was initially nearly 60%, but in a decreasing trend dropped below 20% when CT became routinely used for head trauma. Between 2009 and 2014, of 4554 minor head injury patients overall, 85.4% underwent head CT. After guideline implementation in 2011, CT utilization significantly increased from 81.6 to 87.6% (p = 7 × 10−7), while yield significantly decreased from 12.2 to 9.6% (p = 0.029).ConclusionsThe number of CTs performed for head trauma gradually increased over two decades, while the yield decreased. In 2011, despite implementation of a guideline aiming to improve selective use of CT in minor head injury, utilization significantly increased.Key Points• Over two decades, the number of head CTs performed for minor, moderate, and severe head injury gradually increased, while the diagnostic yield for intracranial findings showed a decreasing trend.• Despite the implementation of a guideline in 2011, aiming to improve selective use of CT in minor head injury, utilization significantly increased, while diagnostic yield significantly decreased.• Natural language processing is a valuable tool to monitor the utilization and diagnostic yield of imaging as a potential quality-of-care indicator.

The Pediatric Emergency Care Applied Research Network Registry: A Multicenter Electronic Health Record Registry of Pediatric Emergency Care.

Electronic health record (EHR)-based registries allow for robust data to be derived directly from the patient clinical record and can provide important information about processes of care delivery and patient health outcomes. A data dictionary, and subsequent data model, were developed describing EHR data sources to include all processes of care within the emergency department (ED). ED visit data were deidentified and XML files were created and submitted to a central data coordinating center for inclusion in the registry. Automated data quality control occurred prior to submission through an application created for this project. Data quality reports were created for manual data quality review. The Pediatric Emergency Care Applied Research Network (PECARN) Registry, representing four hospital systems and seven EDs, demonstrates that ED data from disparate health systems and EHR vendors can be harmonized for use in a single registry with a common data model. The current PECARN Registry represents data from 2,019,461 pediatric ED visits, 894,503 distinct patients, more than 12.5 million narrative reports, and 12,469,754 laboratory tests and continues to accrue data monthly. The Registry is a robust harmonized clinical registry that includes data from diverse patients, sites, and EHR vendors derived via data extraction, deidentification, and secure submission to a central data coordinating center. The data provided may be used for benchmarking, clinical quality improvement, and comparative effectiveness research.

Prospective validation of a near real-time EHR-integrated automated SOFA score calculator

ObjectivesWe created an algorithm for automated Sequential Organ Failure Assessment (SOFA) score calculation within the Electronic Health Record (EHR) to facilitate detection of sepsis based on the Third International Consensus Definitions for Sepsis and Septic Shock (SEPSIS-3) clinical definition. We evaluated the accuracy of near real-time and daily automated SOFA score calculation compared with manual score calculation. MethodsAutomated SOFA scoring computer programs were developed using available EHR data sources and integrated into a critical care focused patient care dashboard at Mayo Clinic in Rochester, Minnesota. We prospectively compared the accuracy of automated versus manual calculation for a sample of patients admitted to the medical intensive care unit at Mayo Clinic Hospitals in Rochester, Minnesota and Jacksonville, Florida. Agreement was calculated with Cohen’s kappa statistic. Reason for discrepancy was tabulated during manual review. ResultsRandom spot check comparisons were performed 134 times on 27 unique patients, and daily SOFA score comparisons were performed for 215 patients over a total of 1206 patient days. Agreement between automatically scored and manually scored SOFA components for both random spot checks (696 pairs, κ=0.89) and daily calculation (5972 pairs, κ=0.89) was high. The most common discrepancies were in the respiratory component (inaccurate fraction of inspired oxygen retrieval; 200/1206) and creatinine (normal creatinine in patients with no urine output on dialysis; 128/1094). 147 patients were at risk of developing sepsis after intensive care unit admission, 10 later developed sepsis confirmed by chart review. All were identified before onset of sepsis with the ΔSOFA≥2 point criterion and 46 patients were false-positives. ConclusionsNear real-time automated SOFA scoring was found to have strong agreement with manual score calculation and may be useful for the detection of sepsis utilizing the new SEPSIS-3 definition.