• Text Resize A A A
  • Print Print
  • Share Share Share Share

Clinical Data Registries - OHRP Correspondence (2015)

June 25, 2015

Robert M. Portman, JD, MPP
Powers Pyles Sutter & Verville PC 
1501 M Street, N.W., Seventh Floor 
Washington, D.C.  20005-1700

Dear Mr. Portman:

This is in response to your communications with the Office for Human Research Protections (OHRP) on behalf of the Physician Clinical Registry Coalition regarding the application of the Department of Health and Human Services (HHS) protection of human subjects regulations at 45 CFR part 46, subpart A (the “Common Rule”) and the activities of clinical data registries.  In a letter sent to us on September 23, 2013, you requested a number of clarifications by OHRP related to the application of the Common Rule to clinical data registries.  Since that time we posted on the OHRP website the OHRP correspondence with Dr. Anthony Asher on behalf of the National Neurosurgery Quality and Outcomes Database, in which we responded to some of the same issues. And as noted below, we asked the Secretary’s Advisory Committee on Human Research Protections (SACHRP) to consider some of these issues, which it has now done.  While that correspondence and the SACHRP deliberations address some of your concerns, this letter replies directly to your specific requests.

The first request is that OHRP “confirm that providers and groups that submit clinical outcomes data to a data registry primarily for ‘health care operations purposes’ (as defined by the HIPAA privacy rules) are not engaging in human subjects research (as defined by the Common Rule).”  With regard to this issue, we note that any institution - including practice groups, hospitals or other clinical care providers - that solely releases private identifiable information that was obtained in the course of patient clinical care to a clinical data registry is not “engaged’ in human subjects research, and no regulatory requirements of the human subjects regulations apply.  This is true even if the clinical data registry will be conducting nonexempt human subjects research with that information.  You suggest that OHRP amend its Guidance on Engagement of Institutions on Human Subjects Research to specifically refer to this circumstance.  We note that this circumstance corresponds to scenario B(6) of non-engagement in the current guidance document although releasing information to clinical data registries is not specifically mentioned.  While we do not have current plans to revise that guidance document, we agree that it would be appropriate to add that specific example, and thus will keep this suggestion in mind with regard to possible future revisions.

The second request is “…that OHRP recognize that using a registry to aggregate and analyze data pertaining to care quality is not considered human subjects research under the Common Rule”.  OHRP’s view is that such a statement is overly broad, given the regulatory definition of “research.”  As indicated in the correspondence with Dr. Asher, some benchmarking activities may involve human subjects research, while other activities may not.  With this request in mind, OHRP asked SACHRP to consider this topic, which they did at their meetings on October 29-30, 2014 and March 24-25, 2015.  Their conclusions can be found in Attachment A to SACHRP's April 24, 2015 letter to the HHS Secretary.  We acknowledge that some benchmarking activities – for example, ones that are designed solely to measure performance levels by participating practitioners or institutions and not to produce new knowledge regarding the relationship between specific procedures and health outcomes -- will not satisfy the regulatory definition of research and thus will not activate the regulatory requirements.  We continue to look for better ways to clarify this distinction.

The third request is that OHRP “confirm that a clinical data registry conducting human subjects research would not be covered by [the] Common Rule as long as it were not receiving federal funding, conducting research pursuant to a federally-regulated activity, or conducting research under a Federal Wide Assurance where the entity has voluntarily chosen to have all research conducted under the Common Rule.”  We confirm that this is indeed the case. 

The fourth request is that OHRP comment on the use of a single IRB to review and approve a clinical data registry’s nonexempt human subjects research.  As noted above, institutions that are providing data to the clinical data registry, but are not engaged in the research activity, do not need any IRB review.  Thus, only review by a single IRB would be required if there is only one institution that is actually engaged in conducting the research.  But even if there are multiple institutions that are engaged in the conduct of the research, it is still possible to have a single IRB reviewing the study.  The participating institutions that are engaged in the research could put in place an Institutional Review Board (IRB) Authorization Agreement (IAA) in which they designate that single IRB as being responsible for satisfying all of the relevant regulatory requirements, including those that pertain to waiver of informed consent. In that circumstance, no other IRB review would be required.

The fifth request is that OHRP “confirm that the release of de-identified data by a clinical data registry for research analysis does not implicate the Common Rule because the use of de-identified data does not constitute human subjects research.”  It is true that the activity of releasing de-identified data for research does not meet the regulatory definition for research involving a “human subject,” and that consequently the requirements of the Common Rule do not apply to that release of data.  In addition, with regard to the entities that receive the data and analyze it, since that activity would not involve the use of a “human subject,” obtaining and analyzing the de-identified data for research also does not invoke any regulatory requirements of the Common Rule.

The sixth request is to discuss creating an exemption from the regulatory requirements for clinical data registries.  In response to this request, we asked SACHRP to consider this question.  Its conclusion can be found in the recommendation mentioned above (see Attachment A to SACHRP's April 24, 2015 letter to the HHS Secretary). OHRP would welcome continuing discussion of the advisability of creating such an exemption.

We hope this information is helpful.



Jerry Menikoff, M.D., J.D.
Office for Human Research Protections

Content created by Office for Human Research Protections (OHRP)
Content last reviewed on March 24, 2017