Comparison of physical examination and laboratory data between a clinical study and electronic health records

Abstract Accurate display and interpretation of clinical laboratory test results is essential for safe and effective diagnosis and treatment. In an attempt to ascertain how well current electronic health records (EHRs) facilitated these processes, we evaluated the graphical displays of laboratory test results in eight EHRs using objective criteria for optimal graphs based on literature and expert opinion. None of the EHRs met all 11 criteria; the magnitude of deficiency ranged from one EHR meeting 10 of 11 criteria to three EHRs meeting only 5 of 11 criteria. One criterion (i.e., the EHR has a graph with y-axis labels that display both the name of the measured variable and the units of measure) was absent from all EHRs. One EHR system graphed results in reverse chronological order. One EHR system plotted data collected at unequally-spaced points in time using equally-spaced data points, which had the effect of erroneously depicting the visual slope perception between data points. This deficiency could have a significant, negative impact on patient safety. Only two EHR systems allowed users to see, hover-over, or click on a data point to see the precise values of the x–y coordinates. Our study suggests that many current EHR-generated graphs do not meet evidence-based criteria aimed at improving laboratory data comprehension.

Download Full-text

Abstract MP23: Completeness in the Abstraction of Cardiac Biomarkers and Cardiac Pain Data From Electronic Health Records (EHR). Findings From the Atherosclerosis Risk in Communities (ARIC) Study

Circulation ◽

10.1161/circ.137.suppl_1.mp23 ◽

2018 ◽

Vol 137 (suppl_1) ◽

Author(s):

Anna Kucharska-Newton ◽

Manuela Bullo ◽

Matthew Loop ◽

Carlton Moore ◽

Stephanie W Haas ◽

...

Keyword(s):

Electronic Health Records ◽

Gold Standard ◽

Troponin I ◽

Laboratory Data ◽

Cardiac Biomarkers ◽

Kappa Statistics ◽

Health Records ◽

Electronic Health ◽

Cardiac Pain

Background: Calibration of case-finding algorithms from electronic health records (EHR) against established disease surveillance protocols is key to avoiding misclassification bias when using EHR data in epidemiological research. We examined the agreement in the classification of troponin I levels and identification of cardiac pain in hospital EHR data against manually abstracted charts for hospitalizations observed by the ARIC community surveillance of cardiovascular events. Methods: A structured data request for laboratory data and provider notes was submitted to hospitals in the ARIC community surveillance program. Computer programs were developed to extract dates of service, type of laboratory assays performed, and individual assay values for days 1-4 of each hospitalization. Presence of cardiac pain was extracted from provider notes using natural language processing protocols. We calculated percent agreement for troponin I values, kappa statistics for their classification as abnormal (values ≥ twice upper limit normal (ULN)), equivocal (values ≥ULN, but < twice ULN) normal (<ULN), and incomplete, and validity statistics for cardiac pain. Abstraction of information from the medical records by trained abstractors was considered the “gold standard” for comparisons. The analysis sample consisted of all events eligible for full abstraction discharged from one hospital in 2014. Analytical code was created using a “training” dataset randomly-selected from the analysis sample, with the final results computed using a validation sample. Results: Of the 126 EHRs, 104 were eligible for abstraction of cardiac biomarkers and pain information. Agreement in the troponin I values was 75.5% (95%CI: 65.8%, 83.6%) for day 1 of the hospitalization, decreasing thereafter to 62.5% (95%CI: 24.5%, 91.5%) for Day 4. The kappa coefficient for the classification of troponin I values was 0.96 (95% CI: 0.90, 1.00), We observed a high sensitivity in the abstraction of information on cardiac pain (0.99 (95%CI: 0.94, 1.0)). The specificity of cardiac pain information was 0.24 (95% CI: 0.16, 0.35) when extracted from all note types, increasing to 0.90 (95%CI: 0.75, 0.97) if extracted from discharge notes. Conclusion: Troponin I values and manifestation of ischemia such as cardiac pain are critical to the classification of acute coronary events. Therefore, the observed excellent agreement with the gold standard ARIC abstraction shows promise for the use of EHRs in the surveillance of acute cardiovascular disease.

Download Full-text

Real-time prediction of mortality, readmission, and length of stay using electronic health record data

Journal of the American Medical Informatics Association ◽

10.1093/jamia/ocv110 ◽

2015 ◽

Vol 23 (3) ◽

pp. 553-561 ◽

Cited By ~ 44

Author(s):

Xiongcai Cai ◽

Oscar Perez-Concha ◽

Enrico Coiera ◽

Fernando Martin-Sanchez ◽

Richard Day ◽

...

Keyword(s):

Length Of Stay ◽

Electronic Health Records ◽

Real Time ◽

Characteristic Curve ◽

Laboratory Data ◽

Patient Specific ◽

Electronic Health Record Data ◽

Health Records ◽

Prediction Of Mortality ◽

Electronic Health

Objective To develop a predictive model for real-time predictions of length of stay, mortality, and readmission for hospitalized patients using electronic health records (EHRs). Materials and Methods A Bayesian Network model was built to estimate the probability of a hospitalized patient being “at home,” in the hospital, or dead for each of the next 7 days. The network utilizes patient-specific administrative and laboratory data and is updated each time a new pathology test result becomes available. Electronic health records from 32 634 patients admitted to a Sydney metropolitan hospital via the emergency department from July 2008 through December 2011 were used. The model was tested on 2011 data and trained on the data of earlier years. Results The model achieved an average daily accuracy of 80% and area under the receiving operating characteristic curve (AUROC) of 0.82. The model’s predictive ability was highest within 24 hours from prediction (AUROC = 0.83) and decreased slightly with time. Death was the most predictable outcome with a daily average accuracy of 93% and AUROC of 0.84. Discussion We developed the first non–disease-specific model that simultaneously predicts remaining days of hospitalization, death, and readmission as part of the same outcome. By providing a future daily probability for each outcome class, we enable the visualization of future patient trajectories. Among these, it is possible to identify trajectories indicating expected discharge, expected continuing hospitalization, expected death, and possible readmission. Conclusions Bayesian Networks can model EHRs to provide real-time forecasts for patient outcomes, which provide richer information than traditional independent point predictions of length of stay, death, or readmission, and can thus better support decision making.

Download Full-text

Machine-learning based prediction of hypo- and hyperglycemia from electronic health records (Preprint)

10.2196/preprints.36176 ◽

2022 ◽

Author(s):

Harald Witte ◽

Christos Theodoros Nakas ◽

Lia Bally ◽

Alexander Benedikt Leichtle

Keyword(s):

Electronic Health Records ◽

Prediction Model ◽

Hospital Admissions ◽

Laboratory Data ◽

Significant Risk ◽

Routine Data ◽

Therapeutic Interventions ◽

Classification Model ◽

Health Records ◽

Electronic Health

BACKGROUND The increasing need for blood glucose (BG) management in hospitalized patients poses high demands on clinical staff and health care systems alike. Acute decompensations of BG levels (hypo- and hyperglycemia) adversely affect patient outcomes and safety. OBJECTIVE Acute BG decompensations pose a frequent and significant risk for inpatients. Ideally, proactive measures are taken before BG levels derail. We have generated a broadly applicable multiclass classification model for predicting decompensation events from patients’ electronic health records to indicate where adjustments of patient monitoring and/or therapeutic interventions are required. METHODS A retrospective cohort study was conducted of patients hospitalized at a tertiary hospital in Bern, Switzerland. Using patient details and routine data from electronic health records (EHRs), a multiclass prediction model for BG decompensation events (< 3.9 mmol/L (hypoglycemia), or > 10, > 13.9, or > 16.7 mmol/L (representing different degrees of hyperglycemia)) was generated, based on a second-level ensemble of gradient-boosted binary trees. RESULTS 63’579 hospital admissions of 33’212 patients were included in this study. The multiclass prediction model reached a specificity of 93.0%, 98.5%, and 93.6% and a sensitivity of 69.6%, 63.0%, and 65.5%, for the main categories of interest. i.e., non-decompensated cases, hypo- or hyperglycemia, respectively. The median prediction horizon was seven and four hours for hypo- and hyperglycemia, respectively. CONCLUSIONS EHRs hold the potential to reliably predict all kinds of BG decompensations. Readily available patient details and routine laboratory data can support the decisions for proactive interventions and thus help to reduce the detrimental health effects of hypo- and hyperglycemia.

Download Full-text