Protocols for COVID-19 Data Analysis
Last updated May 28, 2020
For details, expand each category.
JH-CROWN: The COVID-19 PMAP Registry | PI: Brian Garibaldi
The goal for the use of this PMAP registry/repository is to maintain a curated database of clinical information for patients evaluated at Johns Hopkins Medical Institutions for suspected or confirmed COVID-19. It will be used to support IRB-approved research projects, recruitment efforts into clinical trials, quality improvement projects, and pilot projects. The applications for this data include studying the natural history of COVID-19, defining clinical biomarkers that predict trajectory and outcome, as well as determining the clinical effectiveness of various interventions and healthcare services provided to such patients.
Data Analyses for COVID-19 Resource | PI: Christopher Chute
This study will create a PMAP registry of SARS-CoV-2 tested persons in the State of Maryland and Washington DC, regardless of result. The goal is to use EHR data from the Chesapeake Regional Information System for our Patients (CRISP) to first identify predictors of various stages of COVID-19 severity and then utilize that information for clinical mitigation, risk profiling, and disease and resource modeling.
National COVID-19 Cohort Collaborative (N3C): A National Resource for Shared Analytics | PI: Christopher Chute
This study establishes a central registry of patients who have been tested for or have a clinical diagnosis of COVID-19. It will contain information specific to individual patients, as a limited (protected) dataset of EHR data at a national level. The goals of this registry are to support novel machine learning analytics and discovery of important predictors associated with emergency visits, hospitalizations, ICU transfer, ventilator dependency, and death, amongst a myriad of related outcomes.
Pilot Test Data for the National COVID-19 Cohort Collaborative | PI: Christopher Chute
Generate HIPAA Safe Harbor de-identified test data from the Hopkins PCORNet, TriNetX, and ACT registries for COVID-19 tested or diagnosed patients for 300 persons.
Coronavirus and SD Reporting RegistrySecure-SD Registry, Surveillance Epidemiology of Coronavirus (COVID-19) Under Research Exclusion | PI: Sophie Lanzkron
This registry is designed to capture pediatric and adult COVID-19 cases that are occurring across the world in patients living with sickle cell disease. The goal of the registry is surveillance to allow reporting on outcomes of cases of COVID-19 in this population of patients. The registry contains only de-identified data.
PCORnet COVID-19 Surveillance and Tracking Plan | PI: Harold Lehmann
This study will create a resource consisting of information from more than 500 patient records that has already been collected under an existing PaTH PCORI-funded Clinical Data Research Network (CDRN) Health Systems Database. The data will be used to characterize the cohort of COVID-19 patients and provide detailed information on demographics and pre-existing conditions as part of the PCORnet program. Potential participants will be identified by querying Epic for testing and diagnostic coding terminologies which are indicative of SARS-CoV-2 infection.
Clinical Characterization Protocol for Severe Infectious Diseases (CCPSEI) | PI: Lauren Sauer
This is a standardized protocol for the rapid, coordinated, clinical investigation of severe or potentially severe acute infections from SARS-CoV-2. Participants with acute illness known or suspected to be caused SARS-CoV-2 will be enrolled. This protocol has been designed to enable data and biological samples to be prospectively collected and shared rapidly in a globally-harmonized data collection and sampling schedule. Multiple independent studies can be easily aggregated, tabulated and analyzed across many different settings globally.
The primary objective for this study is to describe the clinical features of patients presenting with COVID-19 and to determine the temporal correlation of viral load with disease state and severity.
The main secondary objectives for this study is to develop a data and specimen repository that acts as a resource for the investigation and analysis of downstream research questions specific to the clinical course of COVID-19, the development of medical countermeasures and diagnostic tests, and, basic biology questions associated with SARS-CoV-2.
Use, Safety and Effectiveness of Treatments for COVID-19 Infection | PI: Caleb Alexander
We propose to conduct a diverse set of investigations examining the role of therapeutics in COVID-19 infection. This data will provide urgently needed information regarding the use, safety and effectiveness of products used among patients with COVID-19. Initially, we propose to focus on three areas: (1) hydroxychloroquine; (2) ACE Inhibitors/ARBs; and (3) immunosuppressive drugs such as TNF-alpha and Interleukin inhibitors. For the sake of illustration, we focus on hydroxychloroquine for the remainder of this proposal; analyses examining other products may be similarly structured but will depend our ability to adequately observe and quantify products of interest.
For hydroxychloroquine, our research objectives are:
- To characterize the association between hydroxychloroquine exposure and intubation, length of stay and hospital mortality; and
- To quantify the association between hydroxychloroquine exposure and potential toxicities such as QT prolongation
Clinical Presentations and Outcomes of COVID-19 Infection in Solid Organ Transplant (SOT) Recipients and Non-SOT Patients: A Retrospective Observational Cohort Study of the JH-CROWN Registry | PI: Robin Avery
This protocol proposes to: a) understand comparative outcomes and risks for death in solid organ transplant (SOT) patients as compared to the general population (non-SOT) hospitalized with COVID, and b) compare inflammatory markers and other laboratory parameters in SOT recipients and non-SOT patients to explore whether inflammatory responses are blunted in SOT recipients, and whether acute kidney injury and transaminitis are more frequent in SOT recipients
Investigate Drug Repurposing Efforts in COVID-19 via Review of Patient Medication Lists | PI: Chetan Bettegowda
This study seeks to investigate drug repurposing efforts in COVID-19 by determining whether the outcomes of patients are impacted by patient intake of common medications prescribed for other indications (e.g. underlying patient comorbidities). Example medications include prazosin (alpha-blocker), which are prescribed for hypertension and benign prostate hyperplasia.
The Impact of COVID-19 on Maternal and Neonatal Outcomes | PI: Irina Burd
The novel coronavirus, COVID-19, has caused a worldwide pandemic. There is still much that is unknown regarding the virus, especially its effects on pregnancy, the fetus, and the neonate. This study seeks to evaluate adverse pregnancy and neonatal outcomes related to COVID-19 infection.
Clinical Characteristics, Sedation Requirements, and Clinical Outcomes within the First 24 Medical ICU (MICU) Patients Infected with COVID-19 at Johns Hopkins Hospital (JHH) | PI: Mahendra Damarla
Characterize the clinical characteristics, sedation requirements, and clinical outcomes that we have observed within the first 24 MICU patients infected with COVID-19 at JHH for whom we have provided clinical care by performing a retrospective chart review.
Prone Positioning Before Intubation in COVID-19 Patients | PI: Mahendra Damarla
A recent study describing the respiratory physiology of mechanically ventilated COVID-19 associated ARDS patients showed low respiratory system compliance in the supine positon, however prone positioning increased lung recruitment and improved oxygenation. Given physiological benefits of prone positioning in COVID-19 patients, we hypothesized patients with respiratory distress who were not yet intubated but at high risk for intubation might benefit from prone positioning.
Reactive Plasmacytosis in COVID-19 | PI: Brian Garibaldi
Describe cases of reactive plasmacytosis in COVID-19 by performing a manual EPIC review to publish a case series on 6-10 patients. Evaluate timeline of inflammatory markers and clinical course relative to plasmacytosis.
Statistical Modeling to Predict Outcome in COVID-19 | PI: Brian Garibaldi
This request is to conduct a series of 4 statistical analyses designed to develop and illustrate simple to more complex statistical methods to describe the longitudinal trajectories of the Johns Hopkins COVID-19 Registry cohort and to test specific hypotheses about baseline and time-varying factors that affect progression. The organizing causal model is that baseline patient and viral characteristics cause an individual’s progression of disease as measured by the 8-level WHO COVID-19 severity index, and that this progression is mediated through pathobiological pathways that are reflected in a moderate number of biomarkers and vital signs. The four analyses will be conducted to produce novel insights about disease progression and to provide statistical packages, illustrative analyses, and on-line teaching vignettes to make it easier for other groups to address their own specific hypotheses involving longitudinal predictors of disease progression.
Demographics of initial 900 patients in PMAP COVID-19 registry | PIs: Brian Garibaldi and Scott Zeger
The core personnel, including the PI and co-investigators on the JH-CROWN registry, will utilize this resource to perform preliminary analysis to identify specific areas of inquiry that require in-depth exploration. The core personnel may also conduct preliminary analysis to define the research agenda related to COVID-19.
RAPID: Prediction of Cardaic Dysfunction in COVID-19 Patients Using Machine Learning | PI: Allison Hays
We will identify laboratory, imaging, and electrocardiographic features of early cardiac injury in COVID-19 patients, with the goal of developing a classification algorithm to identify COVID-19 patients at risk for cardiac dysfunction and sudden cardiac death. Data will be collected from patients admitted to Johns Hopkins Hospital with a confirmed diagnosis of COVID-19 based on nucleic acid or polymerase chain reaction testing. Follow-up data will also be used, for 21 days or until hospital discharge, whichever comes first. Electrocardiography (ECG), time series physiological data (heart rate, systolic and diastolic blood pressure), computed tomography (CT) images, echocardiographic images, and laboratory values specific for cardiac injury obtained in the course of routine care will be analyzed.
Features that will serve as input into the machine learning classifier will be extracted from both time series (ECG, cardiac-specific laboratory values, continuously-obtained vital signs) and imaging data (CT, echocardiography). Feature selection will be used to extract only the most predictive variables to be incorporated into the risk-prediction classifier. Missing data for predictor variables present in >50% of the patient cohort will be imputed using the mean of available data.
Obesity as a Risk Factor for Adverse Outcome from COVID-19 Infection | PI David Kass
Approved by CADRE to receive a de-identified set of standard clinical data from our EMR regarding COVID-19 inpatients including age, BMI sex, race/ethnicity, and pre-existing conditions.
Automatic Assessment and Monitoring of Disease Severity and Risk for Respiratory Support among COVID-19 Patients Admitted to the Johns Hopkins Health Care System | PI: Joao Lima
With this proposal, we set out to identify and stratify individuals who are at risk of becoming critically ill and require mechanical ventilation in a large health care system. The premises of our proposal are the following: the availability of a system that can continuously integrate clinical data acquired in the course of standard patient care will learn to distinguish which patients will require critical care versus those who will not. In addition, such system will also determine which patient characteristics, risk factors and interventions associate with favorable and unfavorable outcomes thus helping to guide not only resource allocation but also clinical conduct once a critical mass of data has been processed by running classification and predictive algorithms. Finally, such a system could also be used for the early detection of changing trends in clinical patterns and resource utilization during the course of the epidemics due to viral drug resistance development, virulence mutations, changes in individual or health system behaviors, as well as allocation of heath care resources.
Evaluation of Chest Imaging Characteristics for Patients with COVID-19 Pneumonia: Imaging Diagnosis and Association with Outcomes | PI: Cheng Lin
Clinical, diagnostic, and imaging (chest radiographs and computed tomography scans) characteristics of patients suspected or confirmed to have COVID-19 will be assessed to test 2 hypotheses: 1) characteristic chest CT features of COVID-19 pneumonia can be diagnostic for the disease and predictive of severe disease; and 2) radiomic features can be utilized to improve the sensitivity and specificity of COVID-19 diagnosis, as well as allow for an automated method of quantifying and characterizing lung abnormalities.
Diagnostic Performance of Lung Point of Care Ultrasound (POCUS) for COVID-19 | PI: Gigi Liu
The purpose of this study is to assess the ability of Lung Point of Care Ultrasound (POCUS) to diagnose COVID-19. COVID-19 RT-PCR will be used as the standard against which the sensitivity, specificity, positive predictive and negative predictive values associated with lung POCUS will be assessed. This study will be retrospective in an observational case series of adults who present to the Johns Hopkins Hospital with respiratory symptoms suggestive of COVID-19.
International Registry on Thoracic Cancer Patients with COVID-19 | PI: Jarushka Naidoo
This study will evaluate the following endpoints: major demographic features of thoracic cancer patients with COVID-19 (e.g. age, sex, place of residence); prevalence of major comorbidities in thoracic cancer patients with COVID-19; proportion of thoracic cancer patients experiencing a severe events overall and by severe events including deaths; proportion of thoracic cancer patients by COVID-19 clinical course; proportion of thoracic cancer patients with COVID-19 who received chemotherapy, surgery, radiotherapy, immune check point inhibitors in the last 2 months before COVID-19 infection; predictive factors of severe events in thoracic cancer patients with COVID-19 including cancer-related treatment; prognostic factors of thoracic cancer patients with COVID-19 including cancer-related treatment
Pulmonary Physiology and Outcomes in COVID-19 Respiratory Failure Requiring Mechanical Ventilation | PI: Eric Nolley
The novel coronavirus SARS-CoV2 causes hypoxemic respiratory failure with diffuse bilateral infiltrates consistent with acute respiratory distress syndrome (ARDS). Preliminary observations suggest that although patients’ meet criteria for ARDS, they have higher lung compliance and greater dead space than expected for patients with ARDS. No studies yet have rigorously characterized the physiology, management, and outcomes of COVID-19-related ARDS. We aim to characterize COVID-19 ARDS in patients undergoing mechanical ventilation at JHH and compare to a historical cohort of patients with similar severity ARDS. A better understanding COVID-19 ARDS physiology and outcomes may help guide therapeutic strategies for these patients requiring mechanical ventilation.
Clinical Features of COVID-19 Disease in Critically Ill Patients in the US: A Multicenter Study | PI: Chirag Parikh
The coronavirus disease-19 (COVID-19) pandemic has had a devastating impact around the world. Combating this pandemic will require a multidisciplinary approach from the medical research community, including translational studies to understand the pathogenesis of disease, randomized controlled trials of novel and repurposed pharmacotherapies, and rigorously conducted epidemiologic studies that include granular patient-level data. We will conduct a multi-center observational study of the clinical features and outcomes of critically ill patients with COVID-19 admitted to intensive care units (ICUs) across the U.S. We will determine the independent risk factors for hospital mortality and acute organ injury.
North American COVID-19 ST-Segment Elevation Myocardial Infarction (NACMI) Registry | Jon Resar
Many of the COVID-29 patients have underlying cardiovascular co-morbidities. Early reports indicate that among COVID-19 positive patients with typical ST-elevation myocardial infarction (STEMI), emergent angiography has revealed a surprising variety of results including classic coronary occlusion, non-obstructive coronary artery disease, angiographically normal epicardial coronary arteries, and/or left ventricular dysfunction. Compiling a registry with information about COVID-19 patients with cardiovascular symptoms will be beneficial in identifying patterns of myocardial injury, developing a risk model for cardiac complications, understanding short and long-term major adverse cardiac events, and designing clinical trials testing different treatment modalities.
The plan consists of reviewing medical records of patients who are COVID-19 positive or COVID-19 suspected who present with ST segment elevation or new left bundle branch block on ECG at baseline and one year. The primary objective is to create a multi-center database of case reports of patients who present with ST segment elevation or new left bundle branch block (LBBB) on ECG and are COVID-19 positive or have suspected COVID-19. Secondary objectives are: to compare the demographics, clinical findings, clinical outcomes, and clinical management of patients identified in primary objective to a historical control of over 15,000 consecutive STEMI activation patients from the Midwest STEMI Consortium; to characterize this population demographically and clinically with the goal of developing data-driven treatment plans, guidelines, and diagnostic acumen regarding these unique patients; to characterize and compare delivery of care and outcomes in COVID-19 positive/suspect patients presenting with STEMI in Canada and the US.
North American registry of the digestive manifestations of COVID-19 | PI: Vikesh Singh
We aim to characterize the digestive manifestations of COVID-19 by developing a registry of affected patients at medical centers across North America. More than 5,000 cases of COVID-19 have been diagnosed in North America with many more unconfirmed due to inadequate testing. Emerging evidence suggests that the novel coronavirus (SARS-CoV-2) infects the gastrointestinal tract in addition to the respiratory system and may be spread by fecal-oral transmission, particularly at the time of endoscopy. Little is known about the digestive manifestations of COVID-19 or the contribution of these manifestations to the spread and virulence of disease, especially in North America. A comprehensive understanding of the GI manifestations of this disease as it emerges may have important implications in the care of affected patients and in informing public health initiatives to extinguish the pandemic.
Multi-center Retrospective Observational Trial of Children Hospitalized with COVID-19 Infection | PI: Rebecca Riggs
This is a retrospective multi-center observational study capturing demographic data, medical support, and outcomes of children hospitalized with COVID-19. The lead-site (shown below) will only receive a deidentified data set. There are 41 Pediatric Institutions who will contribute the above data for all pediatric patients admitted to their institution during the past 6 weeks of this COVID-19 Pandemic.
COVID-19: Surgical Implications Study| PI: Ken Stevens
The COVID-19 pandemic is a rapidly expanding global challenge, but there is currently little evidence to inform the management of surgical patients during the pandemic. There is an urgent need to understand the outcomes of COVID-19 infected patients who undergo surgery. Capturing real-world data and sharing international experience will inform the management of this complex group of patients who undergo surgery throughout the COVID-19 pandemic, improving their clinical care.
Our purpose is to describe the management and outcomes of confirmed COVID-19 patients who have a surgical issue (operative and non-operative management).
We will be collaborating on a GlobalSurg project [https://globalsurg.org/] on COVID19, which will be open to ALL centers worldwide who have COVID+ patients requiring an operation. This project is extremely important to get the data out there on how each hospital is managing these novel cases, surgeons around the world stand to benefit from this.
Case Series of the COVID-19 Population at Johns Hopkins | PI Zishan Siddiqui
Approved by CADRE to receive past medical history, vitals including use of oxygen/high flow oxygen, lab results, and readmission data from the COVID-19 Precision Center of Excellence registry.
Healthcare engagement during the COVID-19 pandemic | PI: Casey Overby Taylor
The objective of this research is to examine the dynamics of healthcare engagement (i.e., missed healthcare events and continuity of care measures) following the COVID-19 pandemic. Healthcare events include medication refills, scheduled medical procedures, and scheduled appointments. Continuity of care measures include proportion of visits with a given practitioner, number of practitioners consulted, distribution of visits to practitioners, and same practitioner from one visit to the next.
ExtraCorporeal Membrane Oxygenation for 2019 Novel Coronavirus Acute Respiratory Disease (The ECMOCARD Study) | PI: Glenn Whitman
In response to the COVID-19 outbreak and to assist in pandemic planning both locally and globally, a research collaborative has been assembled. The study aims to describe clinical features; severity of pulmonary dysfunction; incidence of ICU admission and use of mechanical ventilation and ECMO; ECMO technical characteristics; duration of ECMO; complications; and survival of patients with COVID-19.
Prevalence of Elevated Liver Enzymes in Patients with Novel Coronavirus 2019 (COVID-19) Infection and their Clinical Characteristics and Outcomes | PI: Tinsay Woreta
The novel coronavirus disease 2019 (COVID-19) is a rapidly evolving health crisis throughout the world which mainly affects lower the respiratory tract. The incidence of hepatic abnormalities significantly increases after infection with COVID-19 and during the course of the disease, which may indicate the effect of SARS-CoV-2 on the liver or side effects of medications used to treat COVID-19 infection. Early series from China suggested that up to 14-53% of patients with COVID-10 infection have elevated liver enzymes. Studies from China also show that 2-11% of patients infected with COVID-19 have pre-existing liver conditions. There are currently no published studies that describe the prevalence of elevated liver enzymes in patients with COVID-19 infection in the US. Furthermore, the prevalence of chronic liver disease among patients with COVID-19 infection in the US is not known. The main aim of the present study is to evaluate the prevalence of elevated liver enzymes in patients with COVID-19 infection in a large U.S. cohort and compare the clinical characteristics and outcomes of patients with and without liver injury.
JH-CROWN: The COVID PMAP Registry Statistical Analyses | PI: Scott Zeger
Conduct a series of 4 statistical analyses designed to develop and illustrate simple to more complex statistical methods to describe the longitudinal trajectories of the Johns Hopkins COVID-19 PMAP Registry cohort and to test specific hypotheses about baseline and time-varying factors that affect progression.