The NC TraCS Institute Informatics and Data Science (IDSci) team has been working with many UNC investigators on data requests for COVID-19 related projects.

We are able to support UNC researchers through local data requests, i2b2, and national collaborations, including the National COVID Cohort Collaborative (N3C).

CDW Data Requests

The Carolina Data Warehouse supports all research requests for clinical data from UNCHCS. Researchers may submit a request for consult or collaboration at tracs.unc.edu/request. To support efficiency, consistency, and quality in COVID-19 requests, we encourage requesters to align their projects with our COVID-19 datamart and COVID phenotypes whenever possible.

UNC COVID-19 Datamart for Research

As part of our PCORnet and N3C work, we maintain a COVID-19 datamart in the PCORnet Common Data Model format. This datamart is available for many COVID related requests. It is refreshed twice weekly.

The COVID-19 datamart includes Epic data covering the following domains: demographics (including vital status and zip code), encounters, diagnosis, procedures, vitals, labs, prescribing, medication administration, smoking/vaping, inpatient vitals, ICU status (Y/N), ventilator status (Y/N). See the UNC COVID-19 CDM documentation (UNC login required) for a full list of available data.

COVID Phenotype

We maintain a regularly updated COVID-19 phenotype. The phenotype was developed in collaboration with the research community and remains open for public comment. We invite investigators to incorporate this phenotype in their requests — either in its entirety or by selecting one of the phenotype categories, depending on the needs of your project. The phenotype includes four categories:

1. lab-confirmed positive cases
2. lab-confirmed negative cases
3. suspected positive cases
4. possible positive cases

In many cases, if a COVID request only requires data available in the PCORnet CDM and uses the one or more of the COVID phenotype categories, we can complete the request more quickly and with lower effort. As the datamart is a subset of the Carolina Data Warehouse, regular request submission, triage, and approval processes apply.

GET STARTED

Request a consult or submit a data request: tracs.unc.edu/request

National Partnerships

While we recognize the interest in analyzing the determinants of improvement or complications in UNC Health patients, there is great value in large, national scale research. We are engaged in two national efforts that will support larger sample sizes, more diverse populations, and more generalizable results.

National COVID Cohort Collaborative (N3C)

The National COVID Cohort Collaborative (N3C), led by NIH through NCATS, is a centralized, national electronic health data resource that the research community can use to study COVID-19. N3C is now available to the UNC research community. We have compiled important information and tips about N3C and the process for requesting access to the data: go.unc.edu/N3CFAQ.

PCORnet, the Patient Centered Research Network

PCORnet is a large, distributed network of integrated delivery systems, which have harmonized their EHR-derived data to facilitate sharing of data across many institutions. As a distributed network, the data remains at each institution but is available for aggregate data querying and project-specific data sharing. UNC has been a member of this network for some years. PCORnet is leveraging its established infrastructure to support COVID-19 research activities. Get started by requesting a consult with TraCS.