Federal Exemption Categories

“Established or commonly accepted educational settings” primarily means schools, but can apply to any setting where a specified curriculum is implemented. This includes educational courses delivered through online platforms, but not general informational websites.

“Normal education practices” are defined as activities that typically occur in a classroom or other educational setting.  May involve the use of technologies such as computer software or handheld devices. When assessing a "normal" educational practice, the IRB will consider whether the proposed research involves an educational "best practice" that should benefit students and support the curriculum already required in that setting. In most cases, randomizing students to different conditions is not an normal educational practice and the research project would likely not qualify for exemption.

If part of the research, test/interviews/survey questions must be uploaded. Surveys/interviews are most important because questions may go beyond the assessment of the educational practice, which would likely fall outside of Exemption 1.

See also

• U-MIC presentation
• "Exemption #1 Tips and Exemples" at HRPP Guidance webpage
• Mockup of Section 12 in eResearch for Exemption 1

Exemption 1 can apply to educational research. Activities in the educational setting that do not meet the regulatory definition of “research” [45 CFR 46.102(l)] require no IRB oversight. These may include teaching evaluations.

This exemption is most commonly used for survey research with adult subjects. Researchers should confirm a survey respondent is an adult before proceeding with the main survey questions (i.e. as an eligibility criterion).

For minor (child) subjects, Exemption 2(i) and 2(ii) are limited [45 CFR 46.104(b)(3)] to

• research involving educational tests
• research involving observation of public behavior when the investigator(s) do not participate in the activities being observed.

Research involving surveys of children, or observation of public behavior where the investigator participates in the activities being observed, is not eligible for Exemption 2. Research involving children may not invoke Exemption 2(iii) with limited IRB review.

Data collection methods for this exemption are explicitly limited. Studies that involve any interventions, or that link to other individual-level data, do not qualify for Exemption 2.

Federal Exemption 2 may apply to research involving

1. data recorded anonymously (identifiers are never stored with the research data for analysis, not even temporarily), or
2. data with identifiers that will not place subjects at risk if disclosed (including risk of social or professional opprobrium as well as physical or legal risk), or
3. research involving storage of identifiable data that could place subjects at risk if disclosed, so long as there is also an IRB-approved data security plan in place (IRB determines via limited IRB review that the privacy of subjects and confidentiality of data are adequately protected).

Examples of information that might put a subject at risk include disclosure of illegal activities, sexual behaviors, sensitive genetic or medical information, or embarrassing behaviors in the workplace. U-M also requires for Exemption 2(iii) that any informed consent document(s) and/or recruitment materials are submitted with the application. Please use the Exempt consent template available through the IRBMED Specialty Informed Consent Templates webpage; the Standard consent template is not appropriate.

See also

• U-MIC presentations
  • Anonymous, Coded, and De-identified Data
  • Exemption 2
• Exemption #2 Tips and Examples at HSBS guidance webpage
• Mockup of Section 12 in eResearch for Exemption 2

Activities involving surveys, observation, or other Exemption 2 data collection methods but that do not meet the regulatory definition of “research” [45 CFR 46.102(l)] require no IRB oversight. These may include customer service surveys.

"Behavioral interventions" means that clinical or medical procedures, such as blood draws, even if low risk, are not included in exemption 3.

A "benign" intervention may include ordinary, mild, transient forms of discomfort, such as the stress associated with completing a timed cognitive task, anxiety about performance, and boredom. In assessing whether the research might reasonably be offensive or embarrassing to subjects, the researcher should consider the characteristics of the subject population (e.g. are there particular cultural norms that may be active?) and the research context (e.g. how private is the setting?).

"Brief in duration" is intended to refer to the intervention as opposed to the intervention and the data collection activities together. Thus, the data collection activities could proceed over a longer period of time.

Data may only be collected using verbal or written responses or audiovisual recording, which excludes data collection by Fitbit, heart rate monitors, or any other devices placed on the subjects' bodies.

A subject's "prospective agreement" must be meaningful, but it is not the same as a requirement for explicit informed consent.  For example, individuals who are simply made aware that research data will be collected as they voluntarily complete a computer task may be understood as having agreed to participate.

(These notes are based on the SACHRP Committee 2017 "Recommendations on Benign Behavioral Interventions.")

Like Exemption 2(iii) above, limited IRB review for Exemption 3(i)(C) must determine that the privacy of subjects and confidentiality of data are adequately protected. U-M also requires for Exemption 3(i)(C) that any informed consent document(s) and/or recruitment materials are submitted with the application. Please use the Exempt consent template available through the IRBMED Specialty Informed Consent Templates webpage; the Standard consent template is not appropriate.

See also

• U-MIC presentations
  • Anonymous, Coded, and De-identified Data
  • Exemption #3 Tip Sheet at HSBS guidance webpage
• Mockup of Section 12 in eResearch for Exemption 3

This is the most common IRBMED Exemption.

4(i) “Publicly available” is defined as information shared without conditions on use. Datasets requiring a Data Use Agreement (DUA or DSA) are not “publicly available.” Per federal guidance, publicly available "can apply to secondary research use of archives in a public library, for example, or to government or other institutional records where public access is provided on request, or from a commercial entity if the information is provided to members of the public on request or if the only requirement for obtaining the information is paying a user fee, registering or signing in as a visitor to an archive. It would also apply if a commercial entity made identifiable biospecimens publicly available to anyone on request or for a fee."

4(ii) Data recorded from non-public sources must be de-identified during data collection by the study team, such that no one could “readily ascertain” subjects' identities based on the research dataset: not the researchers, nor the data provider, nor anyone else. Research utilizing multiple datasets is not generally eligible for Exemption 4, since the researchers maintain identifiers temporarily to collate the data.

HIPAA Exemption 4(iii) "Use" has a HIPAA-specific technical definition, meaning the sharing/access/analysis/etc. of the Protected Health Information (PHI) is confined 'inside' the covered entity (CE) that maintains the information. Interdisciplinary studies at U-M may include study team members both 'inside' and 'outside' the Michigan Medicine CE: such studies disclose as well as use PHI, and such studies are not eligible for the HIPAA Exemption. Per federal guidance, since "HIPAA also provides protections in the research context for the information that would be subject to this exemption (e.g., clinical records)..." it is not beneficial to apply the full set of IRB regulations as an additional set of protections.

4(iv) This Exemption is expected to be used rarely, if at all, at U-M.

See also

• U-MIC presentation Anonymous, Coded, and De-identified Data

Analysis of publicly available datasets without identifiers does not meet the regulatory definition of “involving human subjects,” and as such requires no IRB oversight. “Publicly available” is defined as information shared without conditions on use. Datasets requiring a Data Use Agreement (DUA or DSA) are not “publicly available.” A “publicly available” dataset may still require payment of a fee to gain access.

Analysis of non-public data/biospecimens where the investigators never obtain identifiable private information about the data/biospecimen human donors often does not meet the regulatory definition of “involving human subjects.” In these cases, no IRB oversight is required, although a Data Use Agreement (DUA or DSA) or Materials Transfer Agreement (MTA) is very likely required. DUAs and MTAs are processed through the Unfunded Agreement (UFA) system in eResearch Proposal Management (eRPM). If data received includes HIPAA identifiers (in the form of a “limited data set”), an eResearch application is required.

See also

• OHRP Guidance on Research Involving Coded Private Information or Biological Specimens
  • Heading “Research not Involving Human Subjects Versus Exempt Human Subjects Research”
  • Heading “Comparison to the HIPAA Privacy Rule”
• NIH Frequently Asked Questions – Human Subjects
  • Exemptions
  • Human Specimens and Cell Lines

Unlike other categories, this exemption must be granted by the HRPP Director rather than IRB staff. IRB staff provide preliminary processing of an application for Exemption 5, then forward the information to U-M Office of Research (UMOR) for additional review. The application should include explicit documentation from the federal sponsor that the activity is “public benefit or service programs” research meeting the Exemption 5 criteria, such as reference in the Notice of Award to Exemption 5 or a reference to Exemption 5 in a federal rule, such as Medicaid Section 1115 Demonstrations.

This exemption is rarely used at U-M.

U-M Exemption 5 is a flexibility initiative extending the applicability of Exemption 5 permitting the application of the category to State of Michigan programs.

This is the only exemption category also recognized by FDA, per 21 CFR 56.104(d).

This exemption is rarely used at U-M.