RIT-specific IRB Recommendations
Are you planning on writing an IRB protocol specifically to obtain data from Research IT? Read on for some helpful tips and recommendations.
Submitting an IRB protocol is quite straightforward, even if it is your first time. However, if you have any questions, we may be able to offer limited assistance. Please remember that an important function of the IRB is to protect patients from inappropriate contact from researchers. For example, if you are conducting a data driven retrospective study with no plans to contact patients, you will be prompted to re-state that information in various instances throughout the form. A new IRB protocol can be created at https://eprotocol.stanford.edu/irb/. Click on each of the tabs down the left-hand side and answer all the questions; when you get to the "Protocol Information" tab note there are 14 pages arranged as tabs across the top.
In the Resources tab, for most studies we support, the expected answer to question c is something along the lines of "Traditional office space, no wet lab facilities" - if this is correct for your study.
The rest of these suggestions pertain to the Protocol Information tab:
- Participant Population: You can of course use the approximate counts gleaned from the cohort tool, but you should also use language to the effect that these numbers are approximate. Note that you only need to break down your numbers by age gender and so on if you are specifically looking to recruit a particular subpopulation distinguished by the supplied demographic. Many data-driven retrospective research studies use language along the lines of "all eligible adults regardless of race or ethnicity". Regarding patient contact, you can state something like "N/A. We have no direct patient contact. We are only studying data drawn from the clinical record."
- Confidentiality and Privacy Protections: We highly recommend you use https://med.stanford.edu/webtools/secure/forms/irbworksheet to generate your language for this section. You should specify all health information you will be viewing or using for your study, referring to the list of identifiers embedded in the wording of the question as well as listing any source documents, e.g. transcription notes, radiology reports, etc. The answer should also explain the extent of the planned disclosure, e.g. "we plan to obtain: patient full name and telephone numbers but none of this identifying information will be disclosed to anyone other than designated research coordinators affiliated with the study". If you are using either the data review tool or the secure workbench you can use this phrase "The data will be housed within the Stanford Center for Clinical Informatics STRIDE project infrastructure, complete with offsite secure backup, physical and electronic security, and role based user access so that the minimum necessary information is exposed to each researcher."
- Waiver of HIPAA Authorization: Here again we highly recommend you use https://med.stanford.edu/webtools/secure/forms/irbworksheet to generate your language for this section. The language in these sections (if relevant to your protocol) is supposed to be consistent with the language in the confidentiality and privacy sections. In some cases there will be fewer data elements listed here but often the same language used in both sections.
We are unable to approve the release of any data that is not included in your IRB-approved protocol. Our attestation forms generate the same language that is used by https://med.stanford.edu/webtools/secure/forms/irbworksheet in an effort to make the approval process work better. The language in the confidentiality, privacy and waiver of HIPAA authorization sections of your protocol must match the language generated on the Research IT data request and attestation forms. If there is a disconnect between these sections you will have to modify your data request or otherwise submit a revised protocol to the IRB requesting the additional data elements be included in your protocol. So save yourself the headache—and unnecessary delay to your research—and review all sections of your protocol before completing and submitting your Research IT data request.
If you still have questions, contact us! We're here to help. Please let us know through our consultation request form how we can assist you.