scholarly journals Improving postpartum hemorrhage risk prediction using longitudinal electronic medical records

2021 ◽  
Author(s):  
Amanda B Zheutlin ◽  
Luciana Vieira ◽  
Shilong Li ◽  
Zichen Wang ◽  
Emilio Schadt ◽  
...  
Keyword(s):  
Electronic Medical Records ◽  
Postpartum Hemorrhage ◽  
Medical Records ◽  
Assessment Tool ◽  
Vital Signs ◽  
Area Under The Curve ◽  
Test Sample ◽  
Feature Model ◽  
Gradient Boosting ◽  
Clinical Feature

ABSTRACTObjectivePostpartum hemorrhage (PPH) remains a leading cause of preventable maternal mortality in the US. Our goal was to develop a novel risk assessment tool and compare its accuracy to those used in current practice.Materials and MethodsWe used a PPH digital phenotype we developed and validated previously to identify 6,639 cases from our delivery cohort (N=70,948). Using a vast array of known and potential risk factors extracted from electronic medical records available prior to delivery, we trained a gradient boosting model in a subset of our cohort. In a held-out test sample, we compared performance of our model to three clinical risk tools and one previously published model.ResultsOur 24-feature model achieved an area under the curve (AUC) of 0.71 (95% confidence interval [CI], 0.69-0.72), higher than all other tools (research-based AUC: 0.67 [95%CI, 0.66-0.69], clinical AUCs: 0.55 [95%CI, 0.54-0.56] to 0.61 [95%CI, 0.59-0.62]). Five features were novel including red blood cell indices and infection markers measured standardly upon admission. Additionally, we identified inflection points for several vital signs and labs where risk rose substantially. Most notably, patients with median intrapartum systolic blood pressure above 132mmHg had an 11% [interquartile range, 4%] median increase in relative risk for PPH.ConclusionsWe developed a novel approach for predicting PPH and identified clinical feature thresholds that can guide intrapartum monitoring for PPH risk. Our results suggest our model is an excellent candidate for prospective evaluation and could ultimately reduce PPH morbidity and mortality through early detection and prevention.

Do Electronic Medical Records Improve Advance Directive Documentation? A Systematic Review

2018 ◽  
Vol 36 (3) ◽  
pp. 255-263 ◽  
Author(s):  
Christopher Lemon ◽  
Michael De Ridder ◽  
Mohamed Khadra
Keyword(s):  
Systematic Review ◽  
Electronic Medical Records ◽  
Medical Records ◽  
Planning Process ◽  
Assessment Tool ◽  
Controlled Trial ◽  
Decision Aids ◽  
Automatic Identification ◽  
Cross Sectional ◽  
Identification Software

Background: Documentation rates of advance directives (ADs) remain low. Using electronic medical records (EMRs) could help, but a synthesis of evidence is currently lacking. Objectives: To evaluate the evidence for using EMRs in documenting ADs and its implications for overcoming challenges associated with their use. Design: Systematic review of articles in English, published from inception of databases to December 2017. Data Sources: PubMed, PsycINFO, EMBASE, and CINAHL. Methods/Measurements: Four databases were searched from inception to December 2017. Randomized and nonrandomized quantitative studies examining the effects of EMRs on creation, storage, or use of ADs were included. All featured an advance care planning process. Evidence was evaluated using the Cochrane Collaboration’s risk assessment tool. Results: Fifteen studies were included: 1 randomized controlled trial, 1 randomized pilot, 4 pre–post studies, 4 cross-sectional studies, 1 retrospective cohort study, 1 historical control study, 1 retrospective observational study, 1 retrospective review, and 1 evaluation of an EMR feature. Seven studies showed that EMR-based reminders, AD templates, and decision aids can improve AD documentation rates. Three demonstrated that EMR search functions, decision aids, and automatic identification software can help identify patients who have or need ADs according to certain criteria. Five showed EMRs can create documentation challenges, including locating ADs, and making some patients more likely than others to have an AD. Most studies had an unclear or high risk of bias. Conclusions: Limited evidence suggests EMRs could be used to help address AD documentation challenges but may also create additional problems. Stronger evidence is needed to more conclusively determine how EMR may assist in population approaches to improving AD documentation.

scholarly journals 883 A Closed-Loop Audit to Improve Admissions Documentation in A Busy ENT Department

2021 ◽  
Vol 108 (Supplement_6) ◽  
Author(s):  
A Connelly ◽  
K Law ◽  
A Williamson
Keyword(s):  
Electronic Medical Records ◽  
Medical Records ◽  
Closed Loop ◽  
Vital Signs ◽  
P Value ◽  
Medication History ◽  
Admission Source ◽  
Effective Care ◽  
User Friendly ◽  
Pro Forma

Abstract Aim Accurate and thorough admissions documentation is crucial for patient safety and effective care. We amended the admissions pro-forma used on a busy adult ENT ward to improve adherence to a modified version of Royal College of Surgeons of England guidelines. Method Baseline documentation of the 25 parameters of interest was assessed using electronic medical records for all emergency and pre-operative admissions over a 4-week period (n = 75). A new pro-forma was introduced, and the documentation over the following 4 weeks (n = 75) was assessed in the same way. Statistical analysis was done using Excel and RStudio (z-test for two proportions, p-value ≤ 0.05). Results The two groups were similar in age, gender, length of stay, and presenting complaint. The new pro-forma was completed for more admissions than the prior version (91% vs 77%) and resulted in documentation improvements in 19 out of 25 parameters. 9 of these were statistically significant, including initial vital signs and differential diagnosis. Parameters that improved, but not significantly, include admission source, medication history, and cognitive assessment. Across the 8 weeks, using a pro-forma (n = 126) significantly improved documentation of 11 parameters compared to freehand clerking (n = 24). Conclusions Adequate documentation at admission can help with immediate patient care, and act as a point of reference during extended stays. We were able to increase use of a pro-forma and produce meaningful documentation improvements quickly. Further work is required to assess why certain parameters are infrequently completed, and how future pro-forma iterations can become more user-friendly.

scholarly journals A method for systematic discovery of adverse drug events from clinical notes

2015 ◽  
Vol 22 (6) ◽  
pp. 1196-1204 ◽  
Author(s):  
Guan Wang ◽  
Kenneth Jung ◽  
Rainer Winnenburg ◽  
Nigam H Shah
Keyword(s):  
Electronic Medical Records ◽  
Medical Records ◽  
Adverse Drug Events ◽  
Area Under The Curve ◽  
Data Sources ◽  
Clinical Notes ◽  
Post Marketing Surveillance ◽  
Post Marketing ◽  
Harmful Effects ◽  
Positive Support

Abstract Objective Adverse drug events (ADEs) are undesired harmful effects resulting from use of a medication, and occur in 30% of hospitalized patients. The authors have developed a data-mining method for systematic, automated detection of ADEs from electronic medical records. Materials and Methods This method uses the text from 9.5 million clinical notes, along with prior knowledge of drug usages and known ADEs, as inputs. These inputs are further processed into statistics used by a discriminative classifier which outputs the probability that a given drug–disorder pair represents a valid ADE association. Putative ADEs identified by the classifier are further filtered for positive support in 2 independent, complementary data sources. The authors evaluate this method by assessing support for the predictions in other curated data sources, including a manually curated, time-indexed reference standard of label change events. Results This method uses a classifier that achieves an area under the curve of 0.94 on a held out test set. The classifier is used on 2 362 950 possible drug–disorder pairs comprised of 1602 unique drugs and 1475 unique disorders for which we had data, resulting in 240 high-confidence, well-supported drug-AE associations. Eighty-seven of them (36%) are supported in at least one of the resources that have information that was not available to the classifier. Conclusion This method demonstrates the feasibility of systematic post-marketing surveillance for ADEs using electronic medical records, a key component of the learning healthcare system.

scholarly journals Association Between Serum Procalcitonin Levels and Disseminated Intravascular Coagulation in Patients With Septic Shock

2021 ◽  
Author(s):  
Guang Fu ◽  
Xi-si He ◽  
Xiao-peng Cao ◽  
Hao-li Li ◽  
Hai-chao Zhan ◽  
...  
Keyword(s):  
Risk Factors ◽  
Intensive Care Unit ◽  
Septic Shock ◽  
Electronic Medical Records ◽  
Disseminated Intravascular Coagulation ◽  
Medical Records ◽  
Operating Characteristic ◽  
Early Stage ◽  
Area Under The Curve ◽  
Intravascular Coagulation

Abstract Objectives: To investigated the relationships between procalcitonin (PCT) and disseminated intravascular coagulation (DIC) during septic shock. Methods: A retrospective study was performed, which included septic shock patients admitted into intensive care unit (ICU) from January 1, 2015 to December 31, 2018. DIC was defined as international society of thrombosis and homeostasis criteria (ISTH≥5). PCT was based on the first value after admission into ICU and the routine biochemical coagulation data based on the worst value extracted from electronic medical records within 24 hours on admission into ICU.Results: Among 2156 patients screened, 164 patients with septic shock were included in the finally analysis and 35.4% (58/164) of whom developed DIC after admission. PCT level was significantly higher in septic shock patients who developed DIC than those who did not (54.6[13.6-200] vs12.6[2.4-53.3] ng/ml, P <0.001). Multivariable logistic regression revealed that PCT (OR=1.011, 95% CI 1.006- 1.016, P<0.001) was associated with DIC during septic shock. Curve fitting showed a positive correlation between PCT and DIC. The Receiver Operating characteristic (ROC) curve suggested that the cut-off point for PCT to predict DIC during septic shock was 42.1ng/ml, with sensitivity 60.34%, specificity72.74% and the area under the curve (AUC) 0.701(95% CI [0.619-0.784], P<0.001). Interestingly, PCT increased early detection of DIC during septic shock compared with other risk factors(P=0.012)Conclusions: Our data suggest that PCT level over 42.1ng/ml on admission is associated with DIC during septic shock, and PCT is a potential predictive factor of DIC induced by septic shock at early stage.

scholarly journals At the dawn of winter: comparing COVID-19 and influenza presentation and trajectory

2020 ◽  
Author(s):  
Anat Reiner Benaim ◽  
Jonathan Aryeh Sobel ◽  
Ronit Almog ◽  
Snir Lugassy ◽  
Tsviel Ben Shabbat ◽  
...  
Keyword(s):  
Electronic Medical Records ◽  
Medical Records ◽  
Acute Respiratory Infection ◽  
White Blood Cells ◽  
Vital Signs ◽  
White Blood Count ◽  
Respiratory Illnesses ◽  
Disease Trajectory ◽  
Severe Acute Respiratory Infection ◽  
Comorbidity Burden

BackgroundCOVID-19 is a newly recognized illness with a predominantly respiratory presentation. As winter approaches in the northern hemisphere, it is important to characterize the differences in disease presentation and trajectory between COVID-19 patients and other patients with common respiratory illnesses. These differences can enhance knowledge of pathogenesis and help in guiding treatment.MethodsData from electronic medical records were obtained from individuals admitted with respiratory illnesses to Rambam Health Care Campus, Haifa, Israel, between October 1st, 2014 and September 1st, 2020. Four groups of patients were defined: COVID-19 (693), influenza (1,612), severe acute respiratory infection (SARI) (2,292) and Others (4,054). The variable analyzed include demographics (7), vital signs (8), lab tests (38), and comorbidities (15) from a total of 8,651 hospitalized adult patients. Statistical analysis was performed on biomarkers measured at admission and for their disease trajectory in the first 48 hours of hospitalization, and on comorobidity prevalence.ResultsCOVID-19 patients were overall younger in age and had higher body mass index, compared to influenza and SARI. Comorbidity burden was lower in the COVID-19 group compared to influenza and SARI. Severely- and moderately-ill COVID-19 patients older than 65 years of age suffered higher rate of in-hospital mortality compared to hospitalized influenza patients. At admission, white blood cells and neutrophils were lower among COVID-19 patients compared to influenza and SARI patients, while pulse rate and lymphoctye percentage were higher. Trajectories of variables during the first two days of hospitalization revealed that white blood count, neutrophils percentage and glucose in blood increased among COVID-19 patients, while decreasing among other patients.ConclusionsThe intrinsic virulence of COVID-19 appeared higher than influenza. In addition, several critical functions, such as immune response, coagulation, heart and respiratory function and metabolism were uniquely affected by COVID-19.

scholarly journals ConCare: Personalized Clinical Feature Embedding via Capturing the Healthcare Context

2020 ◽  
Vol 34 (01) ◽  
pp. 833-840 ◽  
Author(s):  
Liantao Ma ◽  
Chaohe Zhang ◽  
Yasha Wang ◽  
Wenjie Ruan ◽  
Jiangtao Wang ◽  
...  
Keyword(s):  
Electronic Medical Records ◽  
Medical Records ◽  
Medical Literature ◽  
Personal Characteristics ◽  
Research Problem ◽  
Clinical Feature ◽  
Personal Health ◽  
Dynamic Features ◽  
Baseline Information ◽  
Fundamental Research

Predicting the patient's clinical outcome from the historical electronic medical records (EMR) is a fundamental research problem in medical informatics. Most deep learning-based solutions for EMR analysis concentrate on learning the clinical visit embedding and exploring the relations between visits. Although those works have shown superior performances in healthcare prediction, they fail to explore the personal characteristics during the clinical visits thoroughly. Moreover, existing works usually assume that the more recent record weights more in the prediction, but this assumption is not suitable for all conditions. In this paper, we propose ConCare to handle the irregular EMR data and extract feature interrelationship to perform individualized healthcare prediction. Our solution can embed the feature sequences separately by modeling the time-aware distribution. ConCare further improves the multi-head self-attention via the cross-head decorrelation, so that the inter-dependencies among dynamic features and static baseline information can be effectively captured to form the personal health context. Experimental results on two real-world EMR datasets demonstrate the effectiveness of ConCare. The medical findings extracted by ConCare are also empirically confirmed by human experts and medical literature.

scholarly journals Electronic Medical Records as Input to Predict Postoperative Immediate Remission of Cushing’s Disease: Application of Word Embedding

2021 ◽  
Vol 11 ◽  
Author(s):  
Wentai Zhang ◽  
Dongfang Li ◽  
Ming Feng ◽  
Baotian Hu ◽  
Yanghua Fan ◽  
...  
Keyword(s):  
Electronic Medical Records ◽  
Cushing’S Disease ◽  
Medical Records ◽  
Transsphenoidal Surgery ◽  
Remission Rate ◽  
Area Under The Curve ◽  
Word Embedding ◽  
Cushing's Disease ◽  
Free Text ◽  
College Hospital

BackgroundNo existing machine learning (ML)-based models use free text from electronic medical records (EMR) as input to predict immediate remission (IR) of Cushing’s disease (CD) after transsphenoidal surgery.PurposeThe aim of the present study is to develop an ML-based model that uses EMR that include both structured features and free text as input to preoperatively predict IR after transsphenoidal surgery.MethodsA total of 419 patients with CD from Peking Union Medical College Hospital were enrolled between January 2014 and August 2020. The EMR of the patients were embedded and transformed into low-dimensional dense vectors that can be included in four ML-based models together with structured features. The area under the curve (AUC) of receiver operating characteristic curves was used to evaluate the performance of the models.ResultsThe overall remission rate of the 419 patients was 75.7%. From the results of logistic multivariate analysis, operation (p &lt; 0.001), invasion of cavernous sinus from MRI (p = 0.046), and ACTH (p = 0.024) were strongly correlated with IR. The AUC values for the four ML-based models ranged from 0.686 to 0.793. The highest AUC value (0.793) was for logistic regression when 11 structured features and “individual conclusions of the case by doctor” were included.ConclusionAn ML-based model was developed using both structured and unstructured features (after being processed using a word embedding method) as input to preoperatively predict postoperative IR.

scholarly journals Estimated Date of Delivery with Electronic Medical Records by a Hybrid GBDT-GRU Model

2021 ◽  
Author(s):  
Yina Wu ◽  
Yichao Zhang ◽  
Xu Zou ◽  
Zhenming Yuan ◽  
Wensheng Hu ◽  
...  
Keyword(s):  
Pregnant Women ◽  
Mean Square Error ◽  
Electronic Medical Records ◽  
Hybrid Model ◽  
Medical Records ◽  
Prediction Models ◽  
Coefficient Of Determination ◽  
Gradient Boosting ◽  
Prediction Methods ◽  
Mean Square

Abstract Background: An accurate estimated date of delivery (EDD) helps pregnant women make adequate preparations before delivery and avoid the panic of parturition. EDD is normally derived from some formulates or estimated by doctors based on last menstruation period and ultrasound examinations. The main aim of this study was to develop a hybrid model to improve the accuracy of EDD and promote the health and safety of pregnant women and fetuses. Methods: This study attempted to combine antenatal examinations and electronic medical records to develop a hybrid model based on Gradient Boosting Decision Tree and Gated Recurrent Unit (GBDT-GRU). Besides exploring the features that affect the EDD, GBDT-GRU model obtained the results by dynamic prediction of different stages. The mean square error (MSE), mean absolute error (MAE) and coefficient of determination (R2) were used to compare the performance among the different prediction methods. In addition, we evaluated predictive performances of different prediction models by comparing the proportion of pregnant women under the error of different days. Results: The clinical data were collected with 33,222 pregnancy examination records from 5537 Chinese pregnant women who have given birth. Experimental results showed that the hybrid GBDT-GRU model outperformed other prediction methods with coefficient of determination (R2) of 0.84, mean square error (MSE) of 41.73. We also found that the GBDT-GRU model had a smaller deviation by comparing the bias between the actual delivery date and the EDD under different methods. Conclusions: In comparison with other prediction methods, the GBDT-GRU model provided better performance results. The results of this study are helpful for the development of guidelines for clinical delivery treatments, as it can assist clinicians in making correct decisions during obstetric examinations.

Sign in / Sign up

Export Citation Format

Share Document