Workflow Differences Affect Data Accuracy in Oncologic EHRs: A First Step Toward Detangling the Diagnosis Data Babel

PURPOSE Diagnosis (DX) information is key to clinical data reuse, yet accessible structured DX data often lack accuracy. Previous research hints at workflow differences in cancer DX entry, but their link to clinical data quality is unclear. We hypothesized that there is a statistically significant relationship between workflow-describing variables and DX data quality. METHODS We extracted DX data from encounter and order tables within our electronic health records (EHRs) for a cohort of patients with confirmed brain neoplasms. We built and optimized logistic regressions to predict the odds of fully accurate (ie, correct neoplasm type and anatomic site), inaccurate, and suboptimal (ie, vague) DX entry across clinical workflows. We selected our variables based on correlation strength of each outcome variable. RESULTS Both workflow and personnel variables were predictive of DX data quality. For example, a DX entered in departments other than oncology had up to 2.89 times higher odds of being accurate ( P < .0001) compared with an oncology department; an outpatient care location had up to 98% fewer odds of being inaccurate ( P < .0001), but had 458 times higher odds of being suboptimal ( P < .0001) compared with main campus, including the cancer center; and a DX recoded by a physician assistant had 85% fewer odds of being suboptimal ( P = .005) compared with those entered by physicians. CONCLUSION These results suggest that differences across clinical workflows and the clinical personnel producing EHR data affect clinical data quality. They also suggest that the need for specific structured DX data recording varies across clinical workflows and may be dependent on clinical information needs. Clinicians and researchers reusing oncologic data should consider such heterogeneity when conducting secondary analyses of EHR data.

Download Full-text

Necessity of Clinical Information in Surgical Pathology

Archives of Pathology & Laboratory Medicine ◽

10.5858/1999-123-0615-nociis ◽

1999 ◽

Vol 123 (7) ◽

pp. 615-619 ◽

Cited By ~ 3

Author(s):

Raouf E. Nakhleh ◽

Gordon Gephardt ◽

Richard J. Zarbo

Keyword(s):

Clinical Data ◽

Clinical Information ◽

The United States ◽

Surgical Pathology ◽

Lower Percentage ◽

Malignant Neoplasms ◽

Anatomic Site ◽

Improvement Program ◽

Additional Information ◽

Inadequate Information

Abstract Objectives.—To examine the frequency and nature of problems caused by inadequate clinical data provided on surgical pathology requisition forms. Design.—Participants in the 1996 Q-Probes voluntary quality improvement program of the College of American Pathologists were asked to document prospectively all surgical pathology cases with inadequate information. Inadequate clinical information was defined as the pathologist's need for additional clinical information before a diagnosis could be rendered, regardless of the amount of information already present on the requisition slip. Cases that had no clinical information on a requisition slip were not counted if the lack of history did not hinder diagnosis. The study concluded when 3 months had elapsed or 40 surgical pathology cases were documented. The following data were recorded for each case: anatomic site, type of procedure, nature of disease, method of obtaining additional information, importance of obtained information, and the length of delay in the final diagnosis. Participants.—Three hundred forty-one laboratories, 322 of which were from the United States. Results.—A total of 5594 cases (0.73%) required additional clinical information for diagnosis (10th through 90th percentile range, 3.01% to 0.08%). Institutions with greater average occupied bedsize, a greater number of cases accessioned per year, and a greater number of pathologists had a lower percentage of cases with inadequate clinical data (P < .05). Sixty-eight percent of these cases had no delay in completion of a case, 16.2% had a delay of 1 day or less, and 15.1% of cases were delayed more than 1 day. In 59.4% of cases, the additional clinical information obtained confirmed the initial diagnostic impression. In 25.1%, the information was not relevant to the pathologic diagnosis. In 6.1% there was a substantial change in the diagnosis or a revised report was issued, and in 2.2% no additional information could be obtained. Specific anatomic sites that correlated with a higher rate of changed diagnoses or revised reports in cases with inadequate information included the small bowel, the bronchus/lung, and the ovary. Resection specimens were also significantly associated with a higher rate of changed diagnoses or revised reports when additional information was obtained, as were malignant neoplasms and therapy-induced changes. Conclusions.—This study establishes an aggregate rate of cases with inadequate clinical information for diagnosis (0.73%) and documents the extent of problems caused by inadequate clinical information. The criticality of appropriate clinical information provided to the pathologist is identified for specific anatomic sites and disease processes and is reflected in changed diagnoses or revised reports.

Download Full-text

Opportunities and Challenges for Analyzing Cancer Data at the Inter- and Intra-Institutional Levels

JCO Precision Oncology ◽

10.1200/po.19.00394 ◽

2020 ◽

pp. 743-756

Author(s):

Julie Wu ◽

Jordan Bryan ◽

Samuel M. Rubinstein ◽

Lucy Wang ◽

Michele Lenoue-Newton ◽

...

Keyword(s):

Clinical Data ◽

Information Exchange ◽

Cancer Center ◽

Bipartite Network ◽

Genotypic Differences ◽

Cancer Type ◽

Sample Type ◽

Metastatic Tumors ◽

Institutional Level ◽

Cancer Data

PURPOSE Our goal was to identify the opportunities and challenges in analyzing data from the American Association of Cancer Research Project Genomics Evidence Neoplasia Information Exchange (GENIE), a multi-institutional database derived from clinically driven genomic testing, at both the inter- and the intra-institutional level. Inter-institutionally, we identified genotypic differences between primary and metastatic tumors across the 3 most represented cancers in GENIE. Intra-institutionally, we analyzed the clinical characteristics of the Vanderbilt-Ingram Cancer Center (VICC) subset of GENIE to inform the interpretation of GENIE as a whole. METHODS We performed overall cohort matching on the basis of age, ethnicity, and sex of 13,208 patients stratified by cancer type (breast, colon, or lung) and sample site (primary or metastatic). We then determined whether detected variants, at the gene level, were associated with primary or metastatic tumors. We extracted clinical data for the VICC subset from VICC’s clinical data warehouse. Treatment exposures were mapped to a 13-class schema derived from the HemOnc ontology. RESULTS Across 756 genes, there were significant differences in all cancer types. In breast cancer, ESR1 variants were over-represented in metastatic samples (odds ratio, 5.91; q < 10−6). TP53 mutations were over-represented in metastatic samples across all cancers. VICC had a significantly different cancer type distribution than that of GENIE but patients were well matched with respect to age, sex, and sample type. Treatment data from VICC was used for a bipartite network analysis, demonstrating clusters with a mix of histologies and others being more histology specific. CONCLUSION This article demonstrates the feasibility of deriving meaningful insights from GENIE at the inter- and intra-institutional level and illuminates the opportunities and challenges of the data GENIE contains. The results should help guide future development of GENIE, with the goal of fully realizing its potential for accelerating precision medicine.

Download Full-text

Update on Data Reuse in Health Care

Yearbook of Medical Informatics ◽

10.15265/iy-2017-013 ◽

2017 ◽

Vol 26 (01) ◽

pp. 24-27 ◽

Cited By ~ 2

Author(s):

C. Safran

Keyword(s):

Clinical Data ◽

Legal Framework ◽

Data Reuse ◽

Policy Initiatives ◽

Clinical Databases ◽

Real World Evidence ◽

Policy Frameworks ◽

Potential Risks ◽

Barriers To Use

Summary Objective: Reuse of clinical data has broad use in clinical, research, governmental, and business settings. This summary provides an update on the benefits, barriers to use with large clinical databases, policy frameworks that have been formulated, and challenges. Methods: This report highlights some recent publications on the diverse uses of clinical data and some policy initiatives to promote reuse. It also contains the opinions of the author. Results: Although many examples of the benefits of data reuse have been documented, this summary also reviews why the quality of clinical data needs to be the focus of future informatics work. Conclusion: The promise of reusing data outweighs potential risks, but concerns about privacy and the need to modernize our legal framework will be necessary to realize the full benefits of real-world evidence.

Download Full-text

Combining a convolutional neural network with autoencoders to predict the survival chance of COVID-19 patients

Scientific Reports ◽

10.1038/s41598-021-93543-8 ◽

2021 ◽

Vol 11 (1) ◽

Author(s):

Fahime Khozeimeh ◽

Danial Sharifrazi ◽

Navid Hoseini Izadi ◽

Javad Hassannataj Joloudari ◽

Afshin Shoeibi ◽

...

Keyword(s):

Clinical Data ◽

Data Augmentation ◽

Clinical Information ◽

Ct Images ◽

Classification Performance ◽

Survival Chance ◽

Average Accuracy ◽

Novel Method ◽

Aided Diagnosis ◽

Augmentation Procedure

AbstractCOVID-19 has caused many deaths worldwide. The automation of the diagnosis of this virus is highly desired. Convolutional neural networks (CNNs) have shown outstanding classification performance on image datasets. To date, it appears that COVID computer-aided diagnosis systems based on CNNs and clinical information have not yet been analysed or explored. We propose a novel method, named the CNN-AE, to predict the survival chance of COVID-19 patients using a CNN trained with clinical information. Notably, the required resources to prepare CT images are expensive and limited compared to those required to collect clinical data, such as blood pressure, liver disease, etc. We evaluated our method using a publicly available clinical dataset that we collected. The dataset properties were carefully analysed to extract important features and compute the correlations of features. A data augmentation procedure based on autoencoders (AEs) was proposed to balance the dataset. The experimental results revealed that the average accuracy of the CNN-AE (96.05%) was higher than that of the CNN (92.49%). To demonstrate the generality of our augmentation method, we trained some existing mortality risk prediction methods on our dataset (with and without data augmentation) and compared their performances. We also evaluated our method using another dataset for further generality verification. To show that clinical data can be used for COVID-19 survival chance prediction, the CNN-AE was compared with multiple pre-trained deep models that were tuned based on CT images.

Download Full-text

Impact of medical mistrust (MM) on perceptions of tumor genomic profiling (TGP) among African American (AA) cancer patients: Application of perceptual mapping (PM).

Journal of Clinical Oncology ◽

10.1200/jco.2021.39.15_suppl.e22527 ◽

2021 ◽

Vol 39 (15_suppl) ◽

pp. e22527-e22527

Author(s):

Michael J. Hall ◽

Paul D'Avanzo ◽

Yana Chertock ◽

Jesse A Brajuha ◽

Sarah Bauerle Bass

Keyword(s):

Cancer Patients ◽

Test Performance ◽

Information Needs ◽

University Hospital ◽

Urban Site ◽

Cancer Center ◽

Medical Mistrust ◽

Chi Square ◽

Perceptual Mapping ◽

Chi Square Analysis

e22527 Background: TGP is widely used to identify targetable mutations for precision cancer treatment and clinical trials. Many patients have poor understanding of TGP and are unaware of possible secondary hereditary risks. Lack of clarity regarding the relevance of informed consent and genetic counseling further magnify risks for patients. AA patients have lower genetic knowledge and health literacy and higher MM than Caucasian patients, making them especially vulnerable in the clinical setting. Perceptions of TGP in AA cancer patients have not been well-characterized. Methods: 120 AA pts from 1 suburban and 1 urban site (Fox Chase Cancer Center[FCCC] and Temple University Hospital[TUH]) were surveyed. A k-means cluster analysis using a modified MM scale was conducted; chi-square analysis assessed demographic differences. Perceptual mapping (PM)/multidimensional scaling and vector modeling was used to create 3-dimensional maps to study how TGP barriers/facilitators differed by MM group and how message strategies for communicating about TGP may also differ. Results: Data from 112 analyzable patients from FCCC (55%) and TUH (45%) were parsed into less MM (MM-L, n = 42, 37.5%) and more MM (MM-H, n = 70, 72.5%) clusters. MM-L and MM-H clusters were demographically indistinct with no significant associations by sex (p = 0.49), education (p = 0.3), income (p = 0.65), or location (p = 0.43); only age was significant (older = higher MM, p = 0.006). Patients in the MM-H cluster reported higher concerns about TGP, including cost (p < 0.001), insurance discrimination (p < 0.001), privacy breaches (p = 0.001), test performance/accuracy (p = 0.001), secondary gain by providers (p < 0.001) and provider ability to explain results (p = 0.04). Perceptual mapping identified both shared and contrasting barriers between MM clusters (Table). Conclusions: More than 2/3 of AA patients comprised a MM-H cluster. Communication strategies should focus on concerns about family and how to discuss TGP with an oncologist. PM can identify distinct and shared information needs of vulnerable populations undergoing TGP. [Table: see text]

Download Full-text

The MOBEL Project

Clinical Technologies ◽

10.4018/978-1-60960-561-2.ch412 ◽

2011 ◽

pp. 1171-1190

Author(s):

Inger Dybdahl Sorby ◽

Line Melby ◽

Yngve Dahl ◽

Gry Seland

Keyword(s):

Communication Systems ◽

Information Needs ◽

System Development ◽

Point Of Care ◽

User Involvement ◽

Clinical Information ◽

Problem Area ◽

Patient Centered ◽

Mobile Information ◽

Information And Communication

This chapter presents results and experiences from the MOBEL (MOBile ELectronic patient record) project at the Norwegian University of Science and Technology (NTNU) in Trondheim, Norway. MOBEL was a multidisciplinary research project established in 2000. The problem area of the project was communication and information needs in hospital wards, and the aim of the project was to develop and explore methods and prototypes for point of care clinical information systems (PoCCS) that support clinicians in their patient-centered activities. The chapter summarizes four sub studies performed during the project. Each study presents different approaches to user-centered design of PoCCS. Findings from these studies confirm the need for mobile information and communication technology (ICT) in hospitals. Furthermore, the studies demonstrate how more user involvement and complementary approaches to traditional requirements engineering (RE) and system development methods can be useful when developing mobile information and communication systems for clinicians.

Download Full-text

The MOBEL Project

Handbook of Research on Advances in Health Informatics and Electronic Healthcare Applications ◽

10.4018/978-1-60566-030-1.ch004 ◽

2010 ◽

pp. 52-73

Author(s):

Inger Dybdahl Sorby ◽

Line Melby ◽

Yngve Dahl ◽

Gry Seland

Keyword(s):

Communication Systems ◽

Information Needs ◽

System Development ◽

Point Of Care ◽

User Involvement ◽

Clinical Information ◽

Problem Area ◽

Patient Centered ◽

Mobile Information ◽

Information And Communication

Download Full-text

Clinical data quality: a data life cycle perspective

Biostatistics & Epidemiology ◽

10.1080/24709360.2019.1572344 ◽

2019 ◽

Vol 4 (1) ◽

pp. 6-14 ◽

Cited By ~ 1

Author(s):

Chunhua Weng

Keyword(s):

Life Cycle ◽

Data Quality ◽

Clinical Data ◽

Life Cycle Perspective ◽

Data Life Cycle

Download Full-text

An Ontology-based Mediator of Clinical Information for Decision Support Systems

Methods of Information in Medicine ◽

10.3414/me9126 ◽

2008 ◽

Vol 47 (06) ◽

pp. 549-559 ◽

Cited By ~ 8

Author(s):

K. Ohe ◽

Y. Kawazoe

Keyword(s):

Decision Support ◽

Decision Support System ◽

Clinical Data ◽

Support System ◽

Adverse Drug Events ◽

Clinical Information ◽

Temporal Information ◽

Temporal Abstraction ◽

Alert System ◽

Data Repositories

Summary Objective: We have been developing a decision support system that uses electronic clinical data and provides alerts to clinicians. However, the inference rules for such a system are difficult to write in terms of representing domain concepts and temporal reasoning. To address this problem, we have developed an ontologybased mediator of clinical information for the decision support system. Methods: Our approach consists of three steps: 1) development of an ontology-based mediator that represents domain concepts and temporal information; 2) mapping of clinical data to corresponding concepts in the mediator; 3) temporal abstraction that creates high-level, interval-based concepts from time-stamped clinical data. As a result, we can write a concept-based rule expression that is available for use in domain concepts and interval-based temporal information. The proposed approach was applied to a prototype of clinical alert system, and the rules for adverse drug events were executed on data gathered over a 3-month period. Results: The system generated 615 alerts. 346 cases (56%) were considered appropriate and 269 cases (44%) were inappropriate. Of the false alerts, 192 cases were due to data inaccuracy and 77 cases were due to insufficiency of the temporal abstraction. Conclusion: Our approach enabled to represent a concept-based rule expression that was available for the prototype of a clinical alert system. We believe our approach will contribute to narrow the gaps of information model between domain concepts and clinical data repositories.

Download Full-text

On the Way to Close the Loop in Information Logistics: Data from the Patient — Value for the Patient

Yearbook of Medical Informatics ◽

10.1055/s-0038-1667076 ◽

2018 ◽

Vol 27 (01) ◽

pp. 091-097 ◽

Cited By ~ 1

Author(s):

Werner Hackl ◽

Alexander Hoerbst ◽

Keyword(s):

Information Systems ◽

Medical Informatics ◽

Clinical Information ◽

Data Reuse ◽

Clinical Information Systems ◽

Free Text ◽

Search Terms ◽

Information Logistics ◽

Free Text Search ◽

Selection Of

Objective: To summarize recent research and to propose a selection of best papers published in 2017 in the field of Clinical Information Systems (CIS). Method: Each year a systematic process is carried out to retrieve articles and to select a set of best papers for the CIS section of the International Medical Informatics Association (IMIA) Yearbook of Medical Informatics. The query aiming at identifying relevant publications in the field of CIS was refined by the section editors during the last years. For three years now, the query is stable. It comprises search terms from the Medical Subject Headings (MeSH) thesaurus as well as additional free text search terms from PubMed and Web of Science®. The retrieved articles were categorized in a multi-pass review carried out by the two section editors. The final selection of candidate papers was then peer-reviewed by Yearbook editors and external reviewers. Based on the review results, the best papers were then selected by the IMIA Yearbook editorial board. Text mining, and term co-occurrence mapping techniques were used to get an overview on the content of the retrieved articles. Results: The query was carried out in mid-January 2018, yielding a consolidated result set of 2,255 articles which had been published in 939 different journals. Out of them, 15 papers were nominated as candidate best papers and four of them were finally selected as best papers in the CIS section. Again, the content analysis of the articles revealed the broad spectrum of topics which is covered by CIS research. Conclusions: Modern clinical information systems serve as backbone for a very complex, trans-institutional information logistics process. Data that is produced by, documented in, shared via, organized in, presented by, and stored within clinical information systems is more and more reused for multiple purposes. We found a lot of examples showing the benefits of such data reuse with various novel approaches implemented to tackle the challenges of this process. We also found that the patient moves in the focus of interest of CIS research. So the loop of information logistics begins to close: data from the patients is used to produce value for the patients.

Download Full-text