What types of hospitals are you looking to join?
Enrollment is open to any hospital who would like to participate and has capacity to enter all COVID-19 patients sequentially. The registry is capturing clinical data, vitals and cardiac biomarkers in COVID-19 patients, therefore hospitals that routinely collect serial cardiac labs will be able to upload and track more patient outcomes in this quality improvement program. If your site has a strong interest, please reach out as soon as possible to email@example.com.
What is the patient population for the registry?
- All patients ≥18 years old who are hospitalized with an active COVID-19 infection (for example ICD-10-CM code U07.1 among discharge diagnosis).
- The infection should be confirmed with an RT-PCR test, either prior to or during the hospitalization, a positive IgM antibody test, or a clinical diagnosis using hospital specific criteria.
- Active infection encompasses:
- Patients who are diagnosed prior to hospitalization, but still symptomatic during the hospitalization.
- Patients with a positive test or diagnosis during hospital admission
- Patients who are symptomatic during the hospitalization and have a confirmed test available only after hospital discharge
- Patients < 18 years of age
- Patients who do not have a COVID-19 diagnosis
- Patients who previously tested positive or have been diagnosed with COVID-19, but without active infection as determined by COVID-19 related symptoms. This may include pts with positive IgG antibody tests without positive IgM antibody test.
Does my site need IRB approval to participate?
Similar to the American Heart Association’s existing suite of Get With The Guidelines® (GWTG) programs, the COVID-19 CVD registry is a Quality Improvement program and data are collected and submitted as part of standard healthcare operations.
Most hospitals participating in the other GWTG registries as well as this patient data registry do not require institutional review board (IRB) review. Your hospital IRB will determine whether participation is subject to or exempt from IRB oversight.
The U.S. Department of Health & Human Services offers a resource of Quality Improvement Activities FAQs that may help answer your specific IRB questions.
Can the registry be customized for states and regions?
The ability to customize data elements will vary depending on the extent of desired customizations and demands of the core product development. The AHA platform does offer sites the ability to configure elements for custom data capture. If AHA is not able to accommodate a customization to the core form, these custom elements may be an option for your region. For more information, contact firstname.lastname@example.org.
Joining the Registry
When will the registry launch?
The registry will launch in early May. However, paper case report forms (CRFs) along with detailed coding instructions will be provided to enrolled sites, so they can collect data in advance of the online registry go live date.
Is there a cost for the registry?
The registry is free.
How long will the registry be enrolling patients?
The registry plans to enroll patients through 12/31/2020. We will consider extending enrollment as needed.
How do hospitals enroll in the COVID-19 CVD Registry?
NON-Get With The Guidelines Hospitals:
Hospitals not already contracted for one of our GWTG programs will need to execute an online unified participation agreement (UPA).
Get With The Guidelines Hospitals:
Hospitals already contracted for GWTG through the AHA UPA may:
- sign a paper amendment to the UPA; or
- complete an online UPA solely for enrollment in the COVID-19 CVD Registry.
Hospitals who have Participating Hospital Agreements (PHA) with IQVIA may:
- execute an online UPA to bring all of their modules under the UPA; or
- complete an online UPA solely for enrollment in COVID-19 CVD Registry.
What data are you tracking in the registry?
The registry will capture patient clinical characteristics, medications, treatments, labs, vitals, biomarkers and outcomes in patients hospitalized with COVID-19. Below is a high-level summary of the types of data captured.
|Demographics||Sex, race/ethnicity, age|
|Medical history||Cardiac risk factors, COPD, autoimmune disorders|
|Medications (prior to admission)||ACE inhibitor, steroids|
|Admission vital signs||Height, weight, blood pressure, heart rate, oxygen saturation|
|Laboratory values (admission)||Creatinine, AST, ALT|
|Laboratory values (serial)||Troponin, BNP, lymphocyte count, creatinine|
|Medications (inpatient)||Hydroxychloroquine, lopinavir/ritonavir, Remdesivir|
|Inpatient interventions||Mechanical ventilation, mechanical circulatory support|
Mortality, stroke, heart attack, arrhythmia, DVT/PE
How much time does it take to abstract a patient record?
Abstraction times range from 30 – 45 minutes per record, which is in-line with abstraction time in our other GWTG programs. The CRF was designed to collect minimum necessary data to understand quality of care and outcomes in hospitalized COVID-19 patients. Abstraction time may vary depending on the individual case and how accessible the information is in your EMR.
I'm unsure if my institution collects all the information asked on the CRF. Can I still participate?
Yes - however, some elements are required and will need to be completed in order to mark the form complete. Response options such as "Unavailable" or "Not Documented" are offered throughout to aid in instances in which your hospital may not have access to certain information.
Can I sample patients or upload a subset of COVID-19 patients to the registry?
At this time, hospitals must commit to entering ALL patients hospitalized with a COVID-19 infection. This means completing consecutive case entry, not a sample. We recognize that this may be burdensome for some hospitals, however, allowing for sampling could introduce bias and confound later findings.
Is there a way to automate data from my hospital EMR, or other database to the registry?
Yes, there is an upload function, which allows you to upload a .csv or .xls file. Admission, hospitalization, and serial lab data can all be uploaded into the registry. It is importan to note that your upload file will need to be built to the specifications of the COVID-19 CVD Registry. Should you want to upload serial lab data into the registry, your file will need to be formatted as an .xls file, with admission and hospitalization data on the first tab, and serial lab data on the second tab. A detailed guide for how to build your upload file is available upon request.
Your hospital may use the uploader for retrospective data as far back as you have patients who meet inclusion criteria. Once uploaded, you have the ability to log-in and access uploaded records to update or complete data elements.
An overview of the IRP uploader workflow is listed below for your hospital.
- Uploader specifications provided to hospital
- IT Department or designated vendor creates an extract of data from your EHR or database (e.g., REDCap)
- IT Department converts that extract to a .csv or .xls file and formats to match the uploader specifications.
- IT Department or abstractor follows easy steps to upload the file on a pre-determined basis
Will cases I enter into other GWTG modules auto-populate overlapping fields in this registry?
At this time, cross module data population is not available in the COVID-19 CVD Registry. As we work to build out registry features and functions, we will explore cross module population and add it to the roadmap based on site feedback.
May I enter retrospective data?
Yes, we ask that hospitals begin entry with their first COVID patient and abstract forward from there.
Critically ill patients with COVID-19 may experience long hospitalizations which can affect interpretation of reported outcomes. What does this mean for a contributing site?
It is important that sites capture all COVID-19 discharges, even for patients with long lengths of stay. Sites should begin abstraction with basic information and admission date prior to patient discharge and complete the remainder of the information throughout the hospitalization and/or at the time of discharge. The registry platform generates reports for contributing hospitals based on discharge date, not admission date.
Accessing the Registry Data for Research
Will I be able to analyze data at the state or regional level?
Health System Super User Reporting is now available in the COVID-19 CVD registry. If interested, sites will need to complete an amendment to have this functionality added to their registry. Once configured for the Health System Super User, users will be able to run reports for their entire health system, both as an aggregate or as a comparison between hospitals in their system, in the same view.
When will data be available to research?
The COVID-19 CVD registry data are currently available for research. Sites are able to analyze and pull reports of their real-time data from the registry tool. See question below for how to gain access to the full national registry data
How can researchers access the data?
Researchers, at hospitals participating in the COVID-19 CVD Registry, may request access to the data by submitting a research proposal with specific aims to email@example.com. Proposals will be accepted on a rolling basis and will be reviewed by the COVID-19 Steering Committee. Proposals will be blindly scored on the basis of priority, novel contribution to the scientific literature, feasibility with the data collected, and merit. Approved projects will be paired with a COVID-19 Steering Committee mentor for the duration of the study.
Researchers have the option to complete the statistical analysis for their approved proposal on the Precision Medicine Platform (PMP) using the data set available at that time, or have the AHA Data Analysis Team of experts conduct the analysis in full collaboration with researchers.
No unauthorized data analyses may be conducted using the COVID-19 CVD registry without proposal review by the COVID-19 Steering Committee and written approval from the American Heart Association. All publications must follow AHA’s publication guidelines.
Common Data Abstraction Questions
Q1. A patient arrives at the hospital for another reason (UTI, pregnancy, other infection). During pre-admission or during admission, the patient tests positive for COVID. Should these patients be included in the registry?
A. Yes, regardless if the patient presents with another infection or for another reason, if they are found to have COVID-19, they should be entered into the registry.
Q2. A patient's initial hospitalization was COVID positive. The patient then has multiple hospitalizations and each new COVID test comes back positive. Do we enter multiple patient accounts for this patient in the registry?
A. Yes, as long as the patient is positive and considered to have an active infection each time, they would be entered into the registry. However, if they are positive for IgG antibodies, this is not considered an active infection in subsequent hospitalizations.
Q3. A patient's initial hospitalization was COVID positive. The patient then has multiple hospitalizations and each new COVID test comes back negative. Do we enter multiple patient accounts for this patient in the registry?
A. No, subsequent hospitalizations should not be entered into the registry with a negative COVID test.
Q4. If a patient is symptomatic for COVID during a hospitalization, but gets a COVID test done post-discharge, is there a time period that we should use to enter these patients into the registry if the test is positive?
A. Hospitals do not need to search for cases that test positive after discharge unless they were treated with the presumption of COVID during the hospitalization.
Q5. Can patients be added to the registry using ICD-10 B97.29 or is U07.1 the only accepted ICD-10 code?
A. The intent of the registry is to capture any positive COVID-19 patient that is hospitalized regardless of what brought them into the hospital. Patient eligibility is not limited to a single diagnosis code. The ICD-10-CM U07.1 code is an example of what can be pulled.
Q1. We abstract patients retrospectively. Some of our self-pay/uninsured patients have converted to HRSA (COVID) funded payor source after discharge. How should these patients be abstracted?
A. Self-pay should be selected as that was the patient's status during admission. Abstractors should only abstract what is known during admission, regardless if it changes in the future.
Diagnosis & Evaluation
Q1. If a patient is asymptomatic, what should we enter for documented symptoms? What data should we enter into the registry regarding COVID symptom onset?
A. Not documented should be selected for asymptomatic patients with COVID-19 regarding documented symptoms. The symptom onset date would be unknown, as the patient was not experiencing any active symptoms.
Q1. Our hospital does not use the options listed as values for troponin. How do I enter in these values?
A. You will need to do a conversion to select an available unit. If you collect in ng/mL now, you'll need to convert the data either to ng/mL (your value * 1000) or ug/L (this is a 1:1 comparison for ng/mL, e.g. 1 ng/mL = 1 ug/L) and then select the appropriate value. Please reach out to your hospital's laboratory for assistance.
Q2. A patient's lab value is outside of the upper limit in the COVID registry. How should this be abstracted?
A. Enter in the upper limit for that lab value.
Q3. My lab provides a threshold (i.e. <0.02) instead of a value. How do we abstract this?
A. If your lab provides a threshold rather than an actual lab value, report the value indicated in the threshold. In this case, it would be 0.02.
Q4. If a COVID positive patient is transferred to our facility and has the labs done at the referring site that same day, can the receiving site use those labs as the labs on admission?
A. Use the admission labs closest to admission at your facility. If the labs closest to the admission were done at the referring facility, use those. If an admission lab value is unavailable, select "Unavailable."
Q5. What are the serial lab entry requirements?
A. The goal of the registry is to capture serial labs for each day they are available. After day 14, hospitals may stop entering Serum Creatinine and Absolute Lymphocyte Count if they are the only labs being reported daily. However, be sure to document one last entry prior to discharge. The tool also has a limit of 50 days of serial labs. If you have a patient with an extended stay beyond 50 days, enter in 49 days of serial labs and on the 50th entry, report the last set of labs closest to discharge.
Q1. How do you abstract subsequent ICU stays or intubations for a patient during a single hospital stay? For example, a patient is admitted, sent to ICU, returned to a floor, but then was sent back to ICU within the same hospital stay.
A. The coding instructions request reporting the first documented instance for both ICU stay and intubation. If your hospital is interested in capturing subsequent ICU stays or intubations, these data can be captured in the custom fields.
Q2. How should we count the ICU days if a patient is in and out of the ICU? What about intubation days?
A. The registry collects the first event. The date of transfer from the ICU would be the first episode, not readmission to the ICU. Thus, the duration of the ICU event or intubation would only be from first stay (insertion) through first transfer out of ICU (extubation).
Q3. If a patient came in for surgery and was intubated, but COVID positive a week later and decompensates and was intubated again, which intubation event should we choose to abstract?
A. Enter the first COVID related intubation or event. In this case, do not abstract the surgical intubation.
Q1. What medications should be included in "Medications Prior to Admission"?
A. Only home medications should be included in “Medications Prior to Admission”. Do NOT include medications prescribed at an outside hospital prior to transfer.
Q2. If a patient is in a clinical trial for a medication (i.e. Remdesivir), and it is unclear whether or not the patient received the drug or a placebo, how should this be recorded?
A. Do not abstract this as a "Yes" for the trial medication. Document this in one of the "Other Medications" elements.