FAQs – RADx℠ Radical

All
CDEs
Consent/Protocol Language
Data sharing
IRB
Viral standards

In the consent forms, is it made clear that eventually the de-identified data will go to the NIH in terms of transparency with the participant? This program is a federal database and I don’t know if ethically we have to enter that in the consent form or not – is that customary?

It is customary to let the prospective participants know what data are being collected, how and whether data will be either anonymized or handled in terms of confidentiality – all of this is very much a part of any normal consent. How we convey this information needs to be accurate, accessible, and transparent so that…

A lot of the data that you’re asking for is information that I can collect off of a patient’s chart. Are we required by the NIH to post this as a survey to the patients? I understand asking them if we can share the zip code but in terms of much of the information – if it’s de-identified and we collect it anyway, is there a requirement for the NIH about how we collect this information? Not all of the information may be available from all of the patients. How do you want us to go about that in terms of data collection?

If you don’t have that information, you can explain why you were unable to collect it. Whether you get data from a chart or from the patient themselves depends on your IRB-approved protocol.

How can data collectors gain the trust of underserved communities and overcome selection biases?

Have a good relationship with the patients or the participants in the community, make sure that they’re treated with respect, and listen to them. There is so much that goes into developing a trusted relationship.

We are going to use the database that our hospital is using (EPIC), I don’t know if reformatting is possible from EPIC to the exact wording of the CDEs as the data in EPIC is collected with every patient in the hospital and I don’t know that the hospital will or can change to match the CDE wording.

Submit as much data as you can and, if possible, conforms to the CDEs. If needed, we can have a one-on-one meeting with the particular project member, and just answer why it’s not possible to collect a certain element – that’s why we’re here.

How can we make sure we’re sending the right kind of data to the DCC?

We have created a checklist to help you make sure that the data you plan to share in the repository does not violate the Safe Harbor requirements. You would apply the checklist tool to the data before you ever share it with the DCC. You can find it here: https://www.radxrad.org/resource/hipaa-safe-harbor-checklist-from-for-radx-rad/.

Will personal information be shared to outside entities?

The DCC will only accept de-identified data. Sites are responsible for de-identification of data prior to sharing with the DCC. We will determine a global study ID to link across the various repositories.

Will data be analyzed for research purposes, public health monitoring, or something else?

The projects funded by RADx-rad are research studies. These studies are designed for the purpose of developing and evaluating novel approaches to testing, diagnoses and surveillance. The research conducted by the RADx-rad awardees may contribute to better public health practices and surveillance models in the future. The de-identified data deposited in the NIH repository may be used for purposes…

If we have an existing/ongoing study that is currently not collecting all required CDEs will we need to submit an IRB amendment to adjust our data collection?

Each awardee can discuss with their respective Program Officers whether a waiver for a particular CDE item is appropriate or if an amendment will be needed.

Where should questions about the CDEs be sent?

The central contact for DCC-related CDE questions is Kirk Roberts: kirk.roberts@uth.tmc.edu. Please also “cc” the RADx-rad central email address at: radx-rad-admin@ucsd.edu.

Will the CDEs that are being mandated impact selection bias in underserved communities?

The latest set of minimum CDEs has been updated to reflect many of the concerns related to underserved communities. There are still questions related to race, ethnicity, disability, insurance status, etc. These are included precisely because NIH wants to gauge the involvement of underserved communities in the RADx-rad projects.

Can we add prior smell or taste disorders to the required medical history questions? This is an important variable – nearly 16 million people in the US already have a smell or taste loss and trying to understand the relationship between those two variables could be really important. Many of those in the chemosensory group will be doing this, but it would be very helpful if everybody were collecting this information.

Within the chemosensory group it makes sense to add that prior smell or taste disorders to medical history for all participants. The purpose of the data standards meetings is for each project to agree on things that would be scientifically interesting to have across the project.

Our project is composed mostly of retrospective data and some prospective data. We are not planning on consenting participants since we will be analyzing publicly available data, aggregate data and doing chart reviews. Do we need to integrate the minimum CDE survey questions to our project?

What we are recommending is to collect data according to the CDEs and not in a different format. Prospective data are those data that come into existence after your study begins and may require informed consent or a waiver of consent by the IRB. Consult with your IRB as needed.

Is there template language for the consent form and/or protocol to be added to the IRB application that will convey how we will share data with entities and that data will be accessible for secondary use?

Please see sample language below: “The data collected from you may be used for research that is not yet defined. If that occurs, your information will not be connected to your identity. We call that ‘de-identified. De-identified, individual-level, personal health data will be shared with the UC San Diego Data Coordinating Center. Since this information about…

If we are conducting an observational study and we are consenting patients, are these CDEs still required even if we have an observational study?

CDEs are required for observational or interventional studies.

Where should questions about IRB submissions be sent?

Questions can be sent to the RADx-rad central email address: radx-rad-admin@ucsd.edu

What are we supposed to do if we have already been collecting data before the CDEs were identified and approved?

The CDEs are required for prospective data. Please contact your Program Officer to discuss the possible impact of including the minimum CDE changes on your project.

Are we going to have a list of CDEs for pediatric projects?

The DCC conducted a substantial process to identify which CDEs should change for pediatric projects. Ultimately, there was no real consensus beyond the wording of some CDE questions. At least one pediatric project has successfully petitioned for a waiver for CDE change. Please email Kirk Roberts (kirk.roberts@uth.tmc.edu) for an example that can be used for a similar waiver request.

Is there going to be a central survey point or a website by DCC to direct patients to enter their data?

No. Each project should have its own data collection mechanism. If there is a website that asks for viable/PHI information, it would be protected at the institution that is conducting the study.

My study doesn’t involve human subjects, as we are detecting viruses from wastewater. What is the plan of CDEs for non-human subject studies?

As described in the above question, we will work with awardees and NIH to develop recommended CDEs for each research area including those do not involve human subjects (e.g., wastewater). Please note those that additional CDEs are not required, but highly recommended, in order to improve data interoperability across studies in the same research area.

My study involves human subjects; but it uses datasets that already exist (e.g., derived from existing electronic health records). Shall we follow the minimum CDEs requirement?

For study variables in existing datasets that can be mapped to CDEs in the minimum set, DCC will work with awardees to transform those study variables to CDEs. However, it is not required that awardees re-contact subjects to collect minimum CDEs that are missed in the existing dataset.

What language is recommended for use in our Informed Consent documents to convey information about the CDEs?

We will provide examples of statements that can be adapted for use in consent documents to convey the rationale for requesting CDEs and how those data will be stored and shared.

When can we get variants?

We are selecting and producing variants now. Please reach a member of the Diagnostic Core or send your request to radx-rad-admin@ucsd.edu

When can we get viral standards?

Please reach a member of the Diagnostic Core or send your request to radx-rad-admin@ucsd.edu

What are the recommended CDEs for additional variables collected in our study? How do I learn more or contribute to the CDEs recommendation?

We will work together with awardees and the NIH RADx Data Hub team to develop recommendations to CDEs for each FOA. We are setting up meetings with awardees for each FOA group, to discuss CDEs suitable for each research area. Please join us at those meetings so that we can hear your thoughts. We can…

What ID shall we use for the “Identity” field in the minimum CDEs document?

You can use your own internal Identifier (ID). Or you may use the tool in https://www.radxrad.org/resource/rtb/ to generate Study IDs.

Do we need to amend our IRB to include the minimum CDEs?

Yes, you may have to amend your IRB protocols to include the minimum CDEs, if they are not already included in your existing protocols. Feel free to join our weekly office hours on IRB issues.