CovRNN - A recurrent neural network model for predicting outcomes of COVID-19 patients: model development and validation using EHR data

BackgroundPredicting outcomes of COVID-19 patients at an early stage is critical for optimized clinical care and resource management, especially during a pandemic. Although multiple machine learning models have been proposed to address this issue, based on the need for extensive data pre-processing and feature engineering, these models have not been validated or implemented outside of the original study site. MethodsIn this study, we propose CovRNN, recurrent neural network (RNN)-based models to predict COVID-19 patients outcomes, using their available electronic health record (EHR) data on admission, without the need for specific feature selection or missing data imputation. CovRNN is designed to predict three outcomes: in-hospital mortality, need for mechanical ventilation, and long length of stay (LOS >7 days). Predictions are made for time-to-event risk scores (survival prediction) and all-time risk scores (binary prediction). Our models were trained and validated using heterogeneous and de-identified data of 247,960 COVID-19 patients from 87 healthcare systems, derived from the Cerner(R) Real-World Dataset (CRWD). External validation was performed using three test sets (approximately 53,000 patients). Further, the transferability of CovRNN was validated using 36,140 de-identified patients data derived from the Optum(R) de-identified COVID-19 Electronic Health Record v. 1015 dataset (2007-2020). FindingsCovRNN shows higher performance than do traditional models. It achieved an area under the receiving operating characteristic (AUROC) of 93% for mortality and mechanical ventilation predictions on the CRWD test set (vs. 91{middle dot}5% and 90% for light gradient boost machine (LGBM) and logistic regression (LR), respectively) and 86.5% for prediction of LOS > 7 days (vs. 81{middle dot}7% and 80% for LGBM and LR, respectively). For survival prediction, CovRNN achieved a C-index of 86% for mortality and 92{middle dot}6% for mechanical ventilation. External validation confirmed AUROCs in similar ranges. InterpretationTrained on a large heterogeneous real-world dataset, our CovRNN model showed high prediction accuracy, good calibration, and transferability through consistently good performance on multiple external datasets. Our results demonstrate the feasibility of a COVID-19 predictive model that delivers high accuracy without the need for complex feature engineering.

Comprehensive characterization of COVID-19 patients with repeatedly positive SARS-CoV-2 tests using a large US electronic health record database

BackgroundIn the absence of genome sequencing, two positive molecular SARS-CoV-2 tests separated by negative tests, prolonged time, and symptom resolution remain the best surrogate measure of possible re-infection. MethodsUsing a large electronic health record database, we characterized clinical and testing data for 23 patients with repeatedly positive SARS-CoV-2 PCR test results [≥]60 days apart, separated by [≥]2 consecutive negative test results. Prevalence of chronic medical conditions, symptoms and severe outcomes related to COVID-19 illness were ascertained. ResultsMedian age was 64.5 years, 40% were Black, and 39% were female. 83% smoked within the prior year, 61% were overweight/obese, 83% had immune compromising conditions, and 96% had [≥]2 comorbidities. Median interval between the two positive tests was 77 days. Among the 19 patients with 60-89 days between positive tests, 17 (89%) exhibited symptoms or clinical manifestations indicative of COVID-19 at the time of the second positive test and 14 (74%) were hospitalized at the second positive test. Of the four patients with [≥]90 days between two positive tests, two had mild or no symptoms at the second positive test and one, an immune compromised patient, had a brief hospitalization at the first diagnosis, followed by ICU admission at the second diagnosis three months later. ConclusionsOur study demonstrated a high prevalence of immune compromise, comorbidities, obesity and smoking among patients with repeatedly positive SARS-CoV-2 tests. Despite limitations, including lack of semi-quantitative estimates of viral load, these data may help prioritize suspected cases of reinfection for investigation and continued surveillance. ImportanceComprehensive characterization of clinical and SARS-CoV-2 testing data for patients with repeatedly positive SARS-CoV-2 tests can help prioritize suspected cases of reinfection for investigation in the absence of sequencing data and for continued surveillance for potential long-term health consequences of SARS-CoV-2 infection.

Analysis of Prescription Dosage for Discharged Patients in Our Hospital / 中国药师

Objective: To analyze the prescription dosage for discharged patients in our hospital and investigate the potential risks.Methods: A retrospective research method was adopted to statistically analyze 55 872 discharge instructions from July to September in 2015,and more attention was paid to the prescriptions with dosage over 4 weeks,expecially the ones with 26-week dosage, and the related influencing factors were studied as well.

Analysis on Prescription Writing Norms of Narcotic Drugs for Injection in Operation Room and Inpatient Area of Our Hospital / 中国药师

Objective:To statistically analyze the prescription writing norms of narcotic drugs for injection in the operation room and inpatient area of our hospital to improve the prescription writing quality and the efficiency of management. Methods:A retrospec-tive research method was adopted, and 17 103 prescriptions for narcotic drugs for injection in the operation room and inpatient area were collected and analyzed from June to August in 2014. Results:The irregular proportion of prescriptions of narcotic drugs for injec-tion was 9. 05%, and the abnormal item was lack of the text. The proportion of prescriptions using residual amount and needing to be marked was 4. 01%. Conclusion: Improving the management and unifying the prescriptions of narcotics, or applying the electronic prescription if allowed will improve the quality of prescriptions of narcotics and achieve the purpose of fine management.

Preliminary Investigation of Residual Narcotics for Injection in Wards of Our Hospital / 中国药师

Objective:To investigate the residual narcotics for injection after used in wards in order to provide evidence for the use standardization of narcotics for injection. Methods:A retrospective research method was adopted. Totally 5 288 prescriptions including narcotic drugs for injection in wards were collected and analyzed from June to August in 2014. Results:Only four kinds of narcotics for injection were used in wards, the residual proportion of morphine hydrochloride injection and pethidine hydrochloride injection was 2%, and that of fentanyl citrate injection was 0. 2%. Conclusion:The residue of narcotics for injection in wards is very little, and the specifi-cations of the medication in our hospital meet the demands of clinics. The research is beneficial to improving the management efficiency.

Preliminary Investigation on Residue of Narcotics for Injection in Operation Rooms in Our Hospital / 中国药师

Objective:To count and analyze the residue of narcotics for injection after the use in operation rooms in order to find the possibility of risk avoiding and management standardization. Methods:A retrospective research method was adopted to count and analyze 10268 prescriptions including narcotics for injection in operation rooms from June to August in 2014. Results:Ketamine hydro-chloride injection had the largest residue and its remaining amount also occupied the largest proportion of medication. The second one was pethidine hydrochloride injection and the third one was remifentanil hydrochloride for injection. The highest ratio of remaining a-mount in each prescription was ketamine hydrochloride injection, and the second one was morphine hydrochloride injection and pethi-dine hydrochloride injection ranked the third. Conclusion:It is suggested to reduce the specification of medication with the highest ra-tio of remaining amount in each prescription such as ketamine hydrochloride injection, and improve the management efficiency for the use of narcotics in operation rooms.

Artificial dura materials in neurosurgical procedures:An analysis of 100 cases / 中国组织工程研究

BACKGROUND:The meningeal defect patients can be treated with artificial dural materials combined neurosurgical techniques which can protect the integrity of brain tissue. OBJECTIVE:To analyze the biocompatibility and application feasibility of various artificial dura mater materials. METHODS:The artificial dura materials were classified according to the different sources, and the biocompatible and application feasibility of various artificial dura mater materials were analyzed, as wel as the advantages and disadvantages of various materials. The anticoagulant and hemolytic properties of the materials were analyzed through recalcification test and hemolysis test. 100 patients receiving col agen membrane repair selected from Department of Neurosurgery, Dongfang Hospital of Beijing University of Chinese Medicine were observed. RESULTS AND CONCLUSION:The dural repair materials mainly include autologous material, al ograft materials, dissimilar materials, synthetic materials and natural materials. Each kind of material has its advantages and disadvantages. Meningeal defect complications include cerebrospinal fluid leakage, pseudo meninges, infection and arachnoiditis. There are many problems of artificial dural materials to be solved, and the novel biological dural materials and absorbable materials are the direction of dural material research.