Electronic but overly eclectic: Disciplined EHR data management is needed to automate MIPS reporting.

2019 ◽  
Vol 37 (15_suppl) ◽  
pp. e18074-e18074
Author(s):  
Anna E. Schorer ◽  
Jacob Koskimaki ◽  
Robert S. Miller ◽  
Wendy S. Rubinstein ◽  
Elmer Victor Bernstam ◽  
...  
Keyword(s):  
Data Exchange ◽  
Smoking Status ◽  
Vital Signs ◽  
Structured Data ◽  
Mandated Reporting ◽  
Incentive Payment ◽  
Care Plans ◽  
Report Quality ◽  
Oncology Practice ◽  
Data Elements

e18074 Background: Physician reimbursement for care delivered to Medicare beneficiaries fundamentally changed with the 2015 MACRA legislation, requiring eligible clinicians to report quality measures in the Merit-Based Incentive Payment System (MIPS). To determine whether structured data in electronic health records (EHRs) were adequate to report MIPS results, EHR data ingested by ASCO’s CancerLinQ (CLQ) was analyzed. Methods: Nineteen MIPS measures specified for medical oncology, including 8 shared by other specialties, were retrieved from qpp.cms.gov and systematically evaluated to determine data elements necessary to satisfy each measure. The existence of corresponding data fields and completion of these fields with clinical data was analyzed according to EHR implementation in de-identified and aggregated CLQ data. Results: Five clinician informaticists reviewed the 19 oncology MIPS measures, and identified a consensus list of 52 discrete EHR data elements (DEs) that would be needed. CLQ-processed data from 4 commercial EHR systems implemented at 47 CLQ practices found structured data fields for 84% (43 of 52) of the DE, but fewer than half (46%) of these fields were ever populated and only 32% (17 of 52) of DE were recorded for > 20% of cases. Only 3-5 of 19 MIPS measures could be reliably reported based on data element availability by most practices in this sample set. There were minimal differences between the EHRs ability to encode MIPS DE. Elements most likely to be encoded were those for registration (birthdate, gender), billing (diagnosis, meds), vital signs and smoking status, while those seldom or never encoded related to care plans (tobacco, alcohol, pain management). Other DE rarely encoded were patient events occurring outside the oncology practice (receipt/completion of consultations, dates of hospice enrollment and death), which would be dependent on data exchange between work units and practice entities or, more likely, re-entry by oncology practices. Conclusions: Only a minority of DE required to satisfy MIPS criteria are available as discrete data fields in current EHRs, limiting automated reporting efforts. Improved data quality and completeness is needed to satisfy mandated reporting.

scholarly journals Chasm Between Cancer Quality Measures and Electronic Health Record Data Quality

2022 ◽  
Author(s):  
Anna E. Schorer ◽  
Richard Moldwin ◽  
Jacob Koskimaki ◽  
Elmer V. Bernstam ◽  
Neeta K. Venepalli ◽  
...  
Keyword(s):  
Quality Measures ◽  
Health Technology ◽  
Structured Data ◽  
Malignant Neoplasms ◽  
Incentive Payment ◽  
Electronic Health Record Data ◽  
Health Records ◽  
Record Data ◽  
Electronic Health ◽  
Data Elements

PURPOSE The Medicare Access and CHIP Reauthorization Act of 2015 (MACRA) requires eligible clinicians to report clinical quality measures (CQMs) in the Merit-Based Incentive Payment System (MIPS) to maximize reimbursement. To determine whether structured data in electronic health records (EHRs) were adequate to report MIPS CQMs, EHR data aggregated by ASCO's CancerLinQ platform were analyzed. MATERIALS AND METHODS Using the CancerLinQ health technology platform, 19 Oncology MIPS (oMIPS) CQMs were evaluated to determine the presence of data elements (DEs) necessary to satisfy each CQM and the DE percent population with patient data (fill rates). At the time of this analysis, the CancerLinQ network comprised 63 active practices, representing eight different EHR vendors and containing records for more than 1.63 million unique patients with one or more malignant neoplasms (1.73 million cancer cases). RESULTS Fill rates for the 63 oMIPS-associated DEs varied widely among the practices. The average site had at least one filled DE for 52% of the DEs. Only 35% of the DEs were populated for at least one patient record in 95% of the practices. However, the average DE fill rate of all practices was 23%. No data were found at any practice for 22% of the DEs. Since any oMIPS CQM with an unpopulated DE component resulted in an inability to compute the measure, only two (10.5%) of the 19 oMIPS CQMs were computable for more than 1% of the patients. CONCLUSION Although EHR systems had relatively high DE fill rates for some DEs, underfilling and inconsistency of DEs in EHRs render automated oncology MIPS CQM calculations impractical.

Abstract P229: Is a Multidisciplinary Approach to Hypertension Better Than Traditional Care in Female Hypertensive Patients?

Hypertension ◽  
2015 ◽  
Vol 66 (suppl_1) ◽  
Author(s):  
Jessica A Weber ◽  
Shital C Shah ◽  
Sara Turley ◽  
Lynne T Braun ◽  
Erica R Kent ◽  
...  
Keyword(s):  
Nurse Practitioners ◽  
Multidisciplinary Approach ◽  
Smoking Status ◽  
Bivariate Analysis ◽  
Standard Practice ◽  
Personalized Care ◽  
Moderating Variables ◽  
Care Plans ◽  
Treatment Type

Background: Rush Heart Center for Women (RHCW) opened in October 2003 to provide a multidisciplinary approach (MDA) for female patients (pts). RHCW provided personalized care plans to address women’s heart health, with an emphasis on female-specific risk factors and symptoms. MDA including cardiologists, dietitians and nurse practitioners to treat female pts was compared to a similar cohort using standard practice in terms of HTN treatment. Methods: A retrospective study identified pts with HTN treated at RHCW using a MDA and compared to a cohort of pts treated by cardiologists and received a standard level of care consistent with guidelines and recommendations (standard practice). Pts seen between 2008 and 2014 with at least 2 visits between 9 and 15 months apart were studied. Demographic moderating variables were race, age, and insurance. Moderating clinical variables were baseline systolic BP, body mass index, diabetes, smoking status, history of coronary disease, stroke, and prior treatment of HTN. Bivariate and multivariate analyses were conducted to determine the effect of treatment type, with pts’ follow-up BP. Results: A total of 1486 pts were evaluated. Pts seen treated by MDA was younger by <3 years. Table shows descriptive statistics and bivariate analysis. Multivariate analysis revealed that pts treated with MDA had a significantly lower BP in their follow-up systolic BP, 3.8 mmHg less, compared to pts treated with standard practice. (p<0.002). Conclusion: The multidisciplinary approach had better BP control in female hypertensive pts. Prospective studies comparing MDA to standard practice may help to assess improved quality of life, compliance and outcomes.

scholarly journals Surveillance nursing diagnoses, ongoing assessment and outcomes on in-patients who suffered a cardiorespiratory arrest

2018 ◽  
Vol 51 (0) ◽  
Author(s):  
Maria-Eulàlia Juvé-Udina ◽  
Núria Fabrellas-Padrés ◽  
Jordi Adamuz-Tomás ◽  
Sònia Cadenas-González ◽  
Maribel Gonzalez-Samartino ◽  
...  
Keyword(s):  
Cardiac Arrest ◽  
Acute Care ◽  
Vital Signs ◽  
Cross Sectional ◽  
Care Plans ◽  
Care Facilities ◽  
Cross Sectional Design ◽  
Acute Care Facilities ◽  
Ongoing Assessment ◽  
Nursing Diagnoses

ABSTRACT Objective The purposes of this study were to examine the frequency of surveillance-oriented nursing diagnoses and interventions documented in the electronic care plans of patients who experienced a cardiac arrest during hospitalization, and to observe whether differences exist in terms of patients’ profiles, surveillance measurements and outcomes. Method A descriptive, observational, retrospective, cross-sectional design, randomly including data from electronic documentation of patients who experienced a cardiac arrest during hospitalization in any of the 107 adult wards of eight acute care facilities. Descriptive statistics were used for data analysis. Two-tailed p-values are reported. Results Almost 60% of the analyzed patients’ e-charts had surveillance nursing diagnoses charted in the electronic care plans. Significant differences were found for patients who had these diagnoses documented and those who had not in terms of frequency of vital signs measurements and final outcomes. Conclusion Surveillance nursing diagnoses may play a significant role in preventing acute deterioration of adult in-patients in the acute care setting.

scholarly journals The Pensoft Data Publishing Workflow: The FAIRway from articles to Linked Open Data

2019 ◽  
Vol 3 ◽  
Author(s):  
Lyubomir Penev ◽  
Teodor Georgiev ◽  
Viktor Senderov ◽  
Mariya Dimitrova ◽  
Pavel Stoev
Keyword(s):  
Open Data ◽  
Structured Data ◽  
Linked Open Data ◽  
Data Publishing ◽  
Knowledge Graph ◽  
Supplementary File ◽  
Biodiversity Data ◽  
Text Format ◽  
Biodiversity Knowledge ◽  
Data Elements

As one of the first advocates of open access and open data in the field of biodiversity publishiing, Pensoft has adopted a multiple data publishing model, resulting in the ARPHA-BioDiv toolbox (Penev et al. 2017). ARPHA-BioDiv consists of several data publishing workflows and tools described in the Strategies and Guidelines for Publishing of Biodiversity Data and elsewhere: Data underlying research results are deposited in an external repository and/or published as supplementary file(s) to the article and then linked/cited in the article text; supplementary files are published under their own DOIs and bear their own citation details. Data deposited in trusted repositories and/or supplementary files and described in data papers; data papers may be submitted in text format or converted into manuscripts from Ecological Metadata Language (EML) metadata. Integrated narrative and data publishing realised by the Biodiversity Data Journal, where structured data are imported into the article text from tables or via web services and downloaded/distributed from the published article. Data published in structured, semanticaly enriched, full-text XMLs, so that several data elements can thereafter easily be harvested by machines. Linked Open Data (LOD) extracted from literature, converted into interoperable RDF triples in accordance with the OpenBiodiv-O ontology (Senderov et al. 2018) and stored in the OpenBiodiv Biodiversity Knowledge Graph. Data underlying research results are deposited in an external repository and/or published as supplementary file(s) to the article and then linked/cited in the article text; supplementary files are published under their own DOIs and bear their own citation details. Data deposited in trusted repositories and/or supplementary files and described in data papers; data papers may be submitted in text format or converted into manuscripts from Ecological Metadata Language (EML) metadata. Integrated narrative and data publishing realised by the Biodiversity Data Journal, where structured data are imported into the article text from tables or via web services and downloaded/distributed from the published article. Data published in structured, semanticaly enriched, full-text XMLs, so that several data elements can thereafter easily be harvested by machines. Linked Open Data (LOD) extracted from literature, converted into interoperable RDF triples in accordance with the OpenBiodiv-O ontology (Senderov et al. 2018) and stored in the OpenBiodiv Biodiversity Knowledge Graph. The above mentioned approaches are supported by a whole ecosystem of additional workflows and tools, for example: (1) pre-publication data auditing, involving both human and machine data quality checks (workflow 2); (2) web-service integration with data repositories and data centres, such as Global Biodiversity Information Facility (GBIF), Barcode of Life Data Systems (BOLD), Integrated Digitized Biocollections (iDigBio), Data Observation Network for Earth (DataONE), Long Term Ecological Research (LTER), PlutoF, Dryad, and others (workflows 1,2); (3) semantic markup of the article texts in the TaxPub format facilitating further extraction, distribution and re-use of sub-article elements and data (workflows 3,4); (4) server-to-server import of specimen data from GBIF, BOLD, iDigBio and PlutoR into manuscript text (workflow 3); (5) automated conversion of EML metadata into data paper manuscripts (workflow 2); (6) export of Darwin Core Archive and automated deposition in GBIF (workflow 3); (7) submission of individual images and supplementary data under own DOIs to the Biodiversity Literature Repository, BLR (workflows 1-3); (8) conversion of key data elements from TaxPub articles and taxonomic treatments extracted by Plazi into RDF handled by OpenBiodiv (workflow 5). These approaches represent different aspects of the prospective scholarly publishing of biodiversity data, which in a combination with text and data mining (TDM) technologies for legacy literature (PDF) developed by Plazi, lay the ground of an entire data publishing ecosystem for biodiversity, supplying FAIR (Findable, Accessible, Interoperable and Reusable data to several interoperable overarching infrastructures, such as GBIF, BLR, Plazi TreatmentBank, OpenBiodiv and various end users.

scholarly journals From OpenEHR to FHIR and OMOP Data Model for Microbiology Findings

2021 ◽  
Author(s):  
Eugenia Rinaldi ◽  
Sylvia Thun
Keyword(s):  
Infection Control ◽  
Data Model ◽  
Data Exchange ◽  
Control Group ◽  
Common Data Model ◽  
University Hospitals ◽  
Data Set ◽  
Related Data ◽  
National Initiative ◽  
Data Elements

HiGHmed is a German Consortium where eight University Hospitals have agreed to the cross-institutional data exchange through novel medical informatics solutions. The HiGHmed Use Case Infection Control group has modelled a set of infection-related data in the openEHR format. In order to establish interoperability with the other German Consortia belonging to the same national initiative, we mapped the openEHR information to the Fast Healthcare Interoperability Resources (FHIR) format recommended within the initiative. FHIR enables fast exchange of data thanks to the discrete and independent data elements into which information is organized. Furthermore, to explore the possibility of maximizing analysis capabilities for our data set, we subsequently mapped the FHIR elements to the Observational Medical Outcomes Partnership Common Data Model (OMOP CDM). The OMOP data model is designed to support the conduct of research to identify and evaluate associations between interventions and outcomes caused by these interventions. Mapping across standard allows to exploit their peculiarities while establishing and/or maintaining interoperability. This article provides an overview of our experience in mapping infection control related data across three different standards openEHR, FHIR and OMOP CDM.

scholarly journals Canada's Progress to Make Smoking Cessation a First-Line Cancer Treatment

2018 ◽  
Vol 4 (Supplement 2) ◽  
pp. 90s-90s
Author(s):  
M. Halligan ◽  
D. Keen
Keyword(s):  
Smoking Cessation ◽  
Cancer Treatment ◽  
Cancer Patients ◽  
Smoking Status ◽  
Scale Up ◽  
Evidence Based ◽  
First Line ◽  
Quit Smoking ◽  
Substantial Progress ◽  
Oncology Practice

Background: Evidence indicates that smoking cessation improves the effectiveness of treatment and likelihood of survival among all cancer patients, not just those with tobacco-related disease, yet smoking is rarely addressed in oncology practice. Prior to 2016, only 3 provinces in Canada (out of a total of 10 provinces and three territories) reported implementation of smoking cessation for ambulatory cancer patients. Aim: Based on this evidence, the Canadian Partnership Against Cancer (CPAC) implemented a systems change initiative to promote adoption of evidence-based smoking cessation within provincial and territorial cancer systems across Canada. Methods: In 2016, CPAC funded seven provinces and two territories over a 15-month period to plan, implement or evaluate integration of evidence-based smoking cessation for ambulatory cancer patients within cancer systems. Funds were used to plan (2 provinces and 2 territories), implement (3 provinces) or evaluate (2 provinces) systematic, evidence-based approaches to smoking cessation within ambulatory cancer care settings (e.g., establishing routine systems for identification of smoking cancer patients and system to support patients to quit). Funds could not be used for direct service delivery (e.g., cessation counseling). Results: After 15-months of funding from CPAC, 6 provinces reported implementation of smoking cessation for ambulatory cancer patients. The remaining province and 2 territories funded by CPAC reported development of plans for adoption of smoking cessation for cancer patients in the future. Within provinces reporting implementation of smoking cessation for cancer patients, between 65%-97% of ambulatory cancer patients were screened for smoking status; 22%-80% of these patients were offered a referral to cessation services, and 21%-45% of cancer patients accepted a referral. Conclusion: Despite provincial and territorial variations in readiness to uptake evidence-based smoking cessation for cancer patients, CPAC's approach has led to substantial progress in adoption of this approach across Canada. While progress has been made, adoption of smoking cessation and relapse prevention by cancer systems is not yet widespread in Canada. Scale-up to remaining provinces and territory, and spread within existing provinces and territories is required to reach all cancer patients and families who require support to quit smoking. Framing smoking cessation as a therapeutic intervention, not prevention, and a routine part of cancer treatment will be critical for sustainability of this work.

scholarly journals The German Corona Consensus Dataset (GECCO): A standardized dataset for COVID-19 research

2020 ◽  
Author(s):  
Julian Sass ◽  
Alexander Bartschke ◽  
Moritz Lehne ◽  
Andrea Essenwanger ◽  
Eugenia Rinaldi ◽  
...  
Keyword(s):  
Data Exchange ◽  
Professional Associations ◽  
Case Report Form ◽  
Snomed Ct ◽  
University Hospitals ◽  
Research Activity ◽  
Response Options ◽  
Core Dataset ◽  
Icd 10 ◽  
Data Elements

Background: The current COVID-19 pandemic has led to a surge of research activity. While this research provides important insights, the multitude of studies results in an increasing segmentation of information. To ensure comparability across projects and institutions, standard datasets are needed. Here, we introduce the "German Corona Consensus Dataset" (GECCO), a uniform dataset that uses international terminologies and health IT standards to improve interoperability of COVID-19 data. Methods: Based on previous work (e.g., the ISARIC-WHO COVID-19 case report form) and in coordination with experts from university hospitals, professional associations and research initiatives, data elements relevant for COVID-19 research were collected, prioritized and consolidated into a compact core dataset. The dataset was mapped to international terminologies, and the Fast Healthcare Interoperability Resources (FHIR) standard was used to define interoperable, machine-readable data formats. Results: A core dataset consisting of 81 data elements with 281 response options was defined, including information about, for example, demography, anamnesis, symptoms, therapy, medications or laboratory values of COVID-19 patients. Data elements and response options were mapped to SNOMED CT, LOINC, UCUM, ICD-10-GM and ATC, and FHIR profiles for interoperable data exchange were defined. Conclusion: GECCO provides a compact, interoperable dataset that can help to make COVID-19 research data more comparable across studies and institutions. The dataset will be further refined in the future by adding domain-specific extension modules for more specialized use cases.

A content analysis of the Meaningful Use clinical summary: do clinical summaries promote patient engagement?

2015 ◽  
Vol 17 (03) ◽  
pp. 238-251 ◽  
Author(s):  
Karen Jiggins
Keyword(s):  
Content Analysis ◽  
Patient Engagement ◽  
Financial Incentives ◽  
Smoking Status ◽  
Vital Signs ◽  
Family Physicians ◽  
Meaningful Use ◽  
The United States ◽  
University Education ◽  
Care Practice

AimThis study analyzed Meaningful Use (MU) clinical summaries (CS) given to 100 older adults (⩾65) from 10 family physicians in an urban primary care practice.BackgroundIn the United States, MU was designed to promote and enhance patient engagement in hospitals and clinics across the country, providing financial incentives to physicians attesting to the Meaningful Use of a certified Electronic Health Record by meeting a series of measures and objectives. The CS is intended to support patient and family engagement by communicating elements discussed during the clinical encounter including an updated medication list, problem list, and plan of care (POC). Despite the $27.7 billion spent distributing MU payments to more than 418,000 Eligible Professionals in ambulatory care to date, there is little discussion in the scholarly literature supporting the use of the CS to facilitate patient engagement.MethodsTen CS were accessed from each of 10 family physicians during a regular practice week. Directed content analysis and descriptive statistics were used to evaluate the summaries. Key variables of analysis included diagnoses, medications, plan of care content, availability, completeness, health literacy, format, and readability.FindingsCS contained an average of 5.2 diagnoses and 10 medications. Summaries contained vital signs (98%), lab results (9%), smoking status (88%), professional care team members (4%), follow-up appointments (46%), and POC (67%); 37% of CS were judged to be incomplete. Readability scores indicated that a university education was required to understand the CS. CS support patient engagement by supplying information that supports behavior change and self-management, however barriers to patient engagement exist, including (a) access, (b) poor document readability, and (c) a lack of customization to the patient’s experience.

scholarly journals Using Electronic Health Records for Public Health Hypertension Surveillance

2018 ◽  
Vol 10 (1) ◽  
Author(s):  
Timothy D. McFarlane ◽  
Brian E. Dixon ◽  
P. Joseph Gibson
Keyword(s):  
Public Health ◽  
Population Health ◽  
Structured Data ◽  
Health Programs ◽  
Community Based ◽  
Marion County ◽  
Prevalence Estimates ◽  
Electronic Health ◽  
Data Elements ◽  
And Control

ObjectiveTo assess the equivalence of hypertension prevalence estimates between longitudinal electronic health record (EHR) data from a community-based health information exchange (HIE) and the Behavioral Risk Factor Surveillance System (BRFSS).IntroductionHypertension (HTN) is a highly prevalent chronic condition and strongly associated with morbidity and mortality. HTN is amenable to prevention and control through public and population health programs and policies. Therefore, public and population health programs require accurate, stable estimates of disease prevalence, and estimating HTN prevalence at the community-level is acutely important for timely detection, intervention, and effective evaluation. Current surveillance methods for HTN rely upon community-based surveys, such as the BRFSS. While BRFSS is the standard at the state- and national-level, they are expensive to collect, released once per year, and their confidence intervals are too wide for precise estimates at the local level. More timely, frequently updated, and locally precise prevalence estimates could greatly improve the timeliness and precision of public health interventions. The current study evaluated EHR data from a large, mature HIE as an alternative to community-based surveys for timely, accurate, and precise HTN prevalence estimation.MethodsTwo years (2014-2015) of EHR data were obtained from the Indiana Network for Patient Care for two major health systems in Marion County, Indiana, representing approximately 75% of the total county population (n=530,244). These data were linked and evaluated for prevalent HTN. Six HTN phenotypes were defined using structured data variables including clinical diagnoses (ICD9/10 codes), blood pressure (BP) measurements (HTN = ≥140mmhg systolic or ≥90mmHg diastolic), and dispensed HTN medications (Table 1). Phenotypes were validated using a random sample of 600 records, comparing EHR phenotype HTN to HTN as determined through manual chart review by a Registered Nurse. Each phenotype was further evaluated against BRFSS estimates for Marion County, and stratified by sex, race, and age to compare EHR-generated HTN prevalence measures to those known and in current use for chronic disease surveillance. Comparisons were made using the two one-sided statistical test (TOST) of equivalence, wherein the null hypothesis is the BRFSS and EHR prevalence estimates are different by +/-5% and the alternative is estimates differ by less than +/-5%. Rejection of the null resulted in the conclusion of equivalence of the estimates for use in population/public health.ResultsIn general, the performance of the EHR phenotypes was characterized by high specificity (>87%) and low to moderate sensitivity (range 25.4%-95.3%). The false positive rate was lowest among the phenotype defining HTN by both clinical diagnosis and BP measurements (0.3%), and sensitivity was greatest for the phenotype combining all three structured data elements (95.2%). The prevalence of HTN in Marion County, Indiana (2014-2015) for the EHR sample (n=530,244) ranged between 13.7% and 36.2%, compared to 28.4% in the BRFSS sample (Table 1). Only one EHR phenotype (≥1 HTN BP measurement) demonstrated equivalence with BRFSS prevalence at the county level (difference 0.9%, 90% CI for difference -2.3%-4.0%). HTN prevalence by sex, race, age, sex and age, and sex and race (n=120 comparisons) failed to demonstrate equivalence between EHR and BRFSS measures in all but two comparisons, both among females aged 18-39 years. Differences between EHR and BRFSS HTN prevalence at the subgroup level varied but were particularly pronounced among older adults. As suspected, HTN prevalence precision was improved in the EHR sample with the largest subgroup 95% CI width of 0.7% for male African Americans compared to the BRFSS sample 95% CI width of 29.6%.ConclusionsThe applicability of the tested HTN phenotypes will vary based upon which EHR structured data elements are available to public health (i.e., ICD10, vitals, medications). We found that HTN surveillance using a community-based HIE was not a valid replacement for the BRFSS, although the HIE-based estimates could be readily generated and had much narrower confidence intervals.ReferencesMozaffarian D, et al. Heart Disease and Stroke Statistics — 2016 Update. Circulation. 2016; 133: e38-e360.Yoon S, Fryar C, Carroll M. HTN Prevalence and Control Among Adults: United States, 2011–2014. NCHS Data Brief No. 220. 2015; Hyattsville, MD: National Center for Health Statistics, Centers for Disease Control and Prevention, US Dept of Health and Human Services. 

scholarly journals Unstructured clinical documentation reflecting cognitive and behavioral dysfunction: toward an EHR-based phenotype for cognitive impairment

2018 ◽  
Vol 25 (9) ◽  
pp. 1206-1212
Author(s):  
Andrea L Gilmore-Bykovskyi ◽  
Laura M Block ◽  
Lily Walljasper ◽  
Nikki Hill ◽  
Carey Gleason ◽  
...  
Keyword(s):  
Cognitive Impairment ◽  
Acute Care ◽  
Structured Data ◽  
Narrative Text ◽  
Unstructured Data ◽  
Cognitive Behavioral ◽  
Clinical Documentation ◽  
Behavioral Dysfunction ◽  
Increased Risk ◽  
Data Elements

Abstract Despite increased risk for negative outcomes, cognitive impairment (CI) is greatly under-detected during hospitalization. While automated EHR-based phenotypes have potential to improve recognition of CI, they are hindered by widespread under-diagnosis of underlying etiologies such as dementia—limiting the utility of more precise structured data elements. This study examined unstructured data on symptoms of CI in the acute-care EHRs of hip and stroke fracture patients with dementia from two hospitals. Clinician reviewers identified and classified unstructured EHR data using standardized criteria. Relevant narrative text was descriptively characterized and evaluated for key terminology. Most patient EHRs (90%) had narrative text reflecting cognitive and/or behavioral dysfunction common in CI that were reliably classified (κ 0.82). The majority of statements reflected vague descriptions of cognitive/behavioral dysfunction as opposed to diagnostic terminology. Findings from this preliminary derivation study suggest that clinicians use specific terminology in unstructured EHR fields to describe common symptoms of CI. This terminology can inform the design of EHR-based phenotypes for CI and merits further investigation in more diverse, robustly characterized samples.

Sign in / Sign up

Export Citation Format

Share Document