Guidance on IRB Review of Research Involving Existing Data Sets

Updated May 2007 (formerly called "Policy on Public Use Data Sets")

The University of Chicago Social and Behavioral Sciences (SBS) Institutional Review Board (IRB) recently reviewed its policies and procedures for reviewing research involving existing data sets and data archives. The IRB recognizes that some research involving existing data sets and archives may not meet the definition of “human subjects” research requiring IRB review; some may meet definitions of research that is exempt from the HHS regulations at 45 CFR 46; and some may require IRB review.  This document is intended to provide guidance on IRB review policies and review procedures and to reduce burdens associated with IRB review for investigators whose research involves only the analysis of existing data sets and archives.

The IRB considers most research involving existing data sets and archives to fall within the following categories:

• Analysis of de-identified, publicly available data
• Analysis of non-publicly available data with restricted access to participant identifiers (coded private information)
• Analysis of publicly available data with private identifiable information or of non-publicly available data with private identifiable information where researchers will not record individual identifiers
• Analysis of non-publicly available data containing private identifiable information

Because the information accessed in these forms of analysis varies, the IRB has instituted review procedures that reflect differences among data sources.  Investigators conducting research involving existing data sources are encouraged to consider these review procedures (below) when preparing their studies for possible review by the IRB.

Analysis of de-identified, publicly available data

The IRB recognizes that the analysis of de-identified, publicly available data does not constitute human subjects research as defined at 45 CFR 46.102 and that it does not require IRB review.  Many studies utilize data made available through large data consolidation bureaus and consortiums.  To reduce burdens on investigators, the IRB maintains a list of data holders whose archives include only publicly available, de-identified data.  The IRB has reviewed the data use procedures of these data holders and recognizes that the data that they provide for analysis does not constitute “human subjects” information as defined at 45 CFR 46.102 or in the SBS IRB’s policies and procedures. 

1. Inter-University Consortium for Political and Social Research (ICPSR)
2. U.S. Bureau of the Census
3. National Center for Health Statistics
4. National Center for Education Statistics
5. National Election Studies

Accordingly, research projects that are limited to the analysis of data held by these organizations do not require IRB review.  The IRB no longer requires the registration or review of studies involving the analysis of data held by these organizations unless a project merges multiple data sets and in so doing enables the identification of individuals whose data is analyzed.

Additional data sets and archives may quality for inclusion on this list. Investigators who wish to have a specific data set or data archive considered for inclusion on this list should submit the following information to the SBS IRB:

1. The name of data set or data archive; and
2. The URL for the data set/archive or other specific information on how to obtain the data set; and
3. An abstract that describes the content and potential uses of the data set/archive.

Analysis of non-publicly available data with restricted access to participant identifiers (coded private information)

The IRB recognizes that increasing numbers of studies involve the analysis of non-publicly available datasets that include coded private information or that are provided to researchers after the removal of identifying information.  Further, a number of data holding organizations (e.g., Bureau of Labor Statistics, Centers for Disease Control, etc.) provide statistical analysis services for investigators that generate aggregated data from data sets with private identifying information.  Many of these studies do not involve human subjects.  Data use permissions vary widely across data sets and holders; as such, the IRB requests that investigators conducting studies using coded private information or contracting data holders for statistical analyses of data sets involving private information seek a determination from the IRB as to whether the study constitutes human subjects research.  This request is made in accordance with guidance from OHRP concerning research involving coded private information.

For studies involving the analysis of coded private information or the analysis of private information by a third party on behalf of a research team, the IRB requests that investigators submit the following information to the IRB by email:

1. Researcher's name, title, and department; and
2. Title and brief description of the research study; and
3. Name of data set/archive that will be used; and
4. Description of the data access or security plan to be implemented or required by the data holder

If the IRB determines that the project does not constitute human subjects research a message outlining this determination will be sent to the investigator.  If the IRB determines that the project does involve human subjects research the investigator will be asked to submit a protocol for consideration by the IRB.

Analysis of publicly available data with private identifiable information or of non-publicly available data with private identifiable information where researchers will not record individual identifiers

Research involving the analysis of publicly available data containing private identifiable information or the analysis of non-publicly available data that will not be recorded by the investigator in a manner that allows the direct or indirect identification of individuals is exempt from IRB review (45CFR46.101(b)4).  University and SBS IRB policies identify the IRB as the unit responsible for determining exemption from IRB review.  As such, research involving these analyses requires the submission of a protocol for consideration by the IRB. 

Investigators submitting protocols involving these research procedures are asked to provide the following information in their protocol submissions to aid the IRB in making a determination of exemption.

1. Description of data set and availability; and
2. Description of data to be accessed for analysis; and
3. Copies of data use agreements required by data holder.

Analysis of non-publicly available data containing private identifiable information

Research involving the analysis of non-publicly available data that contains private identifiable information about living individuals is considered by the IRB to constitute human subjects research that is not exempt from 45 CFR 46 and University review requirements.  Studies involving analysis of this form of data require review by the IRB.  This review is conducted under expedited or full board review procedures in accordance with the IRB’s review policies. 

Investigators submitting protocols involving these research procedures are asked to include the following information in their protocol submissions to aid the IRB in its review.

1. Description of data set(s) to be analyzed; and
2. Copies of data use or security agreements required by data holder; and
3. Description of data security and access procedures.

The IRB Staff is available to answer any questions regarding this guidance by phone (773-834-7835) or email.