This page is meant to be used, not just read once. It's a working reference for the chair (or any reviewer) at the point of actually classifying a protocol. Each step links to the corresponding OHRP decision chart, which walks through the same logic visually and in more detail than we will cover here. Keep the full chart set bookmarked: OHRP Human Subject Regulations Decision Charts, 2018 Requirements (also available as a single PDF: download the full chart set).

Step 1: Is it research?

"Research" under 45 CFR 46.102(l) means a systematic investigation, including research development, testing, and evaluation, designed to develop or contribute to generalizable knowledge. A few activities are explicitly excluded from this definition even if they look similar on the surface: certain scholarly and journalistic activities, some public health surveillance, and some criminal justice or intelligence activities listed at 46.102(l)(1)-(4). Case reports, most quality-improvement projects internal to a single course or program (or college service, such as registrar or food services), and normal classroom pedagogy that isn't intended to produce generalizable knowledge typically fall outside this definition. If a faculty member wants to try a new teaching method and see how their own students do, with no intent to publish or generalize beyond that classroom, that's likely not research at all, not because it's exempt, but because it never meets the "research" threshold definition.

Note: Journalistic or history research methods do not qualify as research in this way. Both fields have their own methods for documenting sources and ethically collecting information for review and reporting. They should be held to those standards and not social/behavioral science research conventions. If there is the possibility of real harm (e.g., a history or communications project involving victims/perpetrators of crime that could be damaging if the identities of the subjects are revealed), the IRB might be asked to suggest ways to preserve confidentiality. The IRB can offer suggestions to help the researchers based on our expertise, but this board has no oversight responsibilities in those situations.

Step 2: Does it involve human subjects?

If it's research, ask whether it involves human subjects as defined at 45 CFR 46.102(e). Under the revised rule, this includes both living individuals about whom an investigator obtains identifiable private information or biospecimens through intervention or interaction, and secondary use of already-existing identifiable data or specimens for research purposes. Research using only publicly available or properly de-identified data generally does not involve human subjects at all.

Chart to consult: Chart 01, Is an Activity Human Subjects Research Covered by 45 CFR Part 46?

Is an Activity Human Subjects Research Covered by 45 CFR Part 46?

OHRP Decision Chart 1

If the answer to Step 1 or Step 2 is no, the activity does not require IRB review under 45 CFR 46, though it may still need to be documented as a formal "not human subjects research" determination depending on institutional requirements. For example, a faculty or staff member may work with an outside collaborator who needs documentation from the IRB that a survey does not need to be reviewed. We have a template for that documentation when necessary.

Step 3: Does it qualify for an exemption?

If it's human subjects research, check it against the eight exemption categories at 45 CFR 46.104(d). A study can be exempt only if it fits one of these categories cleanly; when in doubt, it doesn't. Two things apply across all eight categories: the determination has to be made by someone other than the investigator, and research involving prisoners as subjects cannot be exempt (with narrow exceptions under Subpart C for research aimed at a broader population that only incidentally includes prisoners).

Category	What it covers	Common on our campus	OHRP chart
46.104(d)(1)	Research on normal educational practices in established educational settings	Comparing teaching methods in a course	Chart 03
46.104(d)(2)	Educational tests, surveys, interviews, or observation of public behavior	Anonymous student surveys; observational studies in public campus spaces	Chart 04
46.104(d)(3)	Benign behavioral interventions with adult subjects, paired with data collection by survey, interview, or observation	Brief psychology lab tasks (e.g., a short cognitive or attitude-priming task) followed by a survey	Chart 05
46.104(d)(4)	Secondary research using identifiable data or specimens that does not require consent	A sociology thesis using existing, properly governed institutional or public datasets	Chart 06
46.104(d)(5)	Certain federally supported public benefit or service programs	Rare for us; mostly applies to federal or state program evaluation	Chart 07
46.104(d)(6)	Taste and food quality evaluation or consumer acceptance studies	Taste-testing studies; Lab studies involving food or flavor stimuli	Chart 08
46.104(d)(7)	Storage or maintenance of identifiable data/specimens for future secondary research, with broad consent	Less common at a small college without a biospecimen repository	Chart 09
46.104(d)(8)	Secondary research using identifiable data/specimens collected under broad consent	Same, applies mainly where broad consent was obtained upstream	Chart 10

Chart to consult first: Chart 02, Is the Research Involving Human Subjects Eligible for Exemption Under 45 CFR 46.104(d)?, then the specific chart for the category that seems to fit.

Limited IRB Review: This is actually a distinct fourth type of review the revised rule created in 2018, sitting between a plain exemption and expedited review. Four exemption categories require it: 45 CFR 46.104(d)(2)(iii), (d)(3)(i)(C), (d)(7), and (d)(8). All four cover research that would otherwise qualify for exemption, except that the data involved is identifiable rather than anonymous or de-identified. Because of that, the exemption isn't automatic; someone still has to make one added finding under 45 CFR 46.111(a)(7), that there are adequate provisions to protect participant privacy and maintain data confidentiality. For (d)(7) and (d)(8) specifically, the reviewer also has to confirm that broad consent was properly obtained and that it actually covers the proposed use of the data.

In practice, this determination is made by the chair or a designated reviewer using the same expedited-style procedure (one reviewer, no convened meeting), and like expedited review, it can be used to approve research but never to disapprove it. Campus example: a psychology survey that would otherwise fit the benign-behavioral-intervention exemption, except participants provide their names or student emails so they can be entered in a raffle for participating. That single detail makes the data identifiable, so the study needs the added privacy and confidentiality finding under limited IRB review rather than a plain exemption determination.

Step 4: If not exempt, does it qualify for expedited review?

Expedited review requires both conditions to be true: the research involves no more than minimal risk, and it fits one of the nine categories on the federal expedited review list (45 CFR 46.110, drawing on the list published at 63 FR 60364, November 9, 1998). Categories 1 through 7 apply to both initial and continuing review; categories 8 and 9 apply only to continuing review of research already underway.

Category	What it covers	Common on our campus
1	Clinical studies of certain drugs or medical devices already exempt from IND/IDE requirements	Rare here; would potentially come up if a kinesiology or nutrition study tested a marketed supplement or over-the-counter product within its approved use
2	Blood draws by finger stick, heel stick, or venipuncture, within defined volume and frequency limits	Kinesiology exercise physiology labs measuring blood lactate or glucose via finger stick during exercise testing
3	Collection of biological specimens by noninvasive means (hair, nail clippings, saliva, sweat, etc.)	Psychology studies collecting saliva samples to measure cortisol as a stress marker
4	Noninvasive data collection using procedures already routine in clinical or educational practice (excluding x-ray or microwave), such as sensors, weighing, sensory testing, EEG, or moderate exercise testing	Kinesiology body-composition, flexibility, and submaximal exercise testing; psychology or neuroscience research using EEG or basic physiological sensors
5	Use of data, documents, records, or specimens originally collected for a nonresearch purpose	A project using existing course exam results or previously collected institutional records, where the records weren't gathered for this study originally
6	Collection of voice, video, digital, or image recordings made for research purposes	Recorded focus groups or interviews in sociology or education research; video-recorded classroom observation; collection of recorded utterances for analysis
7	Research on individual or group behavior, perception, cognition, or communication, or research using survey, interview, oral history, or focus group methods	The majority of non-exempt psychology, sociology, and education survey or interview research that isn't anonymous enough to qualify as exempt
8	Continuing review of previously full-board-approved research that has closed to new enrollment, completed all interventions, or is limited to data analysis	Longitudinal studies in their follow-up-only phase
9	Continuing review of research not covered by categories 2 through 8, where the convened board has already determined and documented that the risk is no more than minimal	General continuing review of straightforward minimal-risk studies

List to consult for full regulatory detail: Categories of Research That May Be Reviewed by IRB through an Expedited Review Procedure (1998 list, as referenced by 45 CFR 46.110)

Note that the 2018 decision chart set does not include a dedicated expedited-review flowchart the way the pre-2018 charts did. The determination rests on reading the study against the minimal-risk definition (45 CFR 46.102(j)) and the category table above directly. If either condition is unclear, that's a signal to send the protocol to full board rather than force it into expedited.

Step 5: If neither exempt nor expedited, it goes to full board

This is the default when the research is more than minimal risk, involves a vulnerable population in a way that calls for full committee discussion, or simply doesn't fit any defined category cleanly. There's no chart for this step because it isn't really a determination; it's what happens when Steps 3 and 4 don't resolve the question.

Other determinations

The five steps above cover the most common determination the board makes: what tier does a new protocol get. But several other determinations come up regularly, after initial approval or as part of specific protocol designs. These come up often enough in psychology, sociology, and education research that the board should be comfortable with the actual criteria.

Is continuing review required?

Under the revised rule, not every full-board study needs an annual renewal. Continuing review is not required, unless the IRB specifically decides otherwise, in three circumstances (45 CFR 46.109(f)(1)):

The research is eligible for expedited review under 45 CFR 46.110, even if it was actually reviewed by the full board for other reasons. Unless there are changes to the protocol or unless specified by the IRB, the approval is ongoing.
The research falls under a limited IRB review category (exempt with extra review for privacy and confidentiality - see above).
The research has progressed to the point where all that remains is data analysis (including analysis of identifiable data or biospecimens) or accessing follow-up clinical or academic data from procedures subjects would undergo anyway as part of standard practice.

Campus example: a two-year longitudinal education study that finished collecting data last spring and is now only running statistical analysis on the dataset it already has does not need continuing review, even though it's still technically an "active" study on paper. The chair still has to make and document that determination; it doesn't happen automatically just because the circumstances fit.

Chart to consult: Chart 11, Is Continuing Review Required Under 45 CFR 46.109(f)?

OHRP Continuing Review

Waiver or alteration of consent, general standard (45 CFR 46.116(f))

This is the one most relevant to our work, because psychology research in particular sometimes depends on participants not knowing the full study design in advance. To waive or alter any of the required elements of consent, the IRB has to find and document all of the following:

The research involves no more than minimal risk to subjects.
The research could not practicably be carried out without the waiver or alteration.
If the research involves identifiable private information or identifiable biospecimens, it could not practicably be carried out without using that information in identifiable form.
The waiver or alteration will not adversely affect the rights and welfare of subjects.
Whenever appropriate, subjects will be provided with additional pertinent information after participation.

That fifth criterion is potentially the most important: it's the regulatory basis for requiring debriefing in deception research. If a psychology study can't tell participants the true purpose upfront without ruining the data (a classic example: telling participants a task measures something other than what it actually measures), the waiver can only be granted if participants are told the truth afterward, and given a chance to withdraw their data if they want to.

Chart to consult: Chart 13, When Can Informed Consent Be Waived or Altered Under 45 CFR 46.116(f)?

Waiver or alteration of consent, public benefit and service programs (45 CFR 46.116(e))

This is a narrower version of the waiver above, limited to research conducted by or subject to approval of state or local government officials, studying public benefit or service programs themselves. It requires only two findings: minimal risk, and that the research couldn't practicably be done without the waiver. This is unlikely to come up in most of our protocols, but it's worth knowing it exists and is a distinct standard from 46.116(f), since the two are easy to conflate.

Chart to consult: Chart 12

Waiver of documentation of consent (45 CFR 46.117(c))

This is a different question from waiving consent itself: it asks whether the researcher can skip getting a signature, while still going through the actual consent conversation or written information. The IRB can waive the signed-form requirement if it finds any one of the following:

The only thing linking a participant to the research would be the signed consent form, and the main risk is a breach of confidentiality if that link were exposed. In this case, each participant still gets asked whether they want their signature kept as a record, and their answer governs.
The research is no more than minimal risk and doesn't involve any procedures that would normally require written consent outside of a research context. This is typical for anonymous surveys that provide all the consent information and tell participants to indicate consent by continuing with the survey.
Participants are members of a distinct cultural group where signing forms isn't the norm, the research is no more than minimal risk, and there's an appropriate alternative way to document that consent happened.

Campus example: a sociology survey on a sensitive topic, distributed anonymously online, where a signed consent form would be the only document tying a specific respondent to sensitive answers. Waiving the signature requirement here (while still presenting the full consent information as a cover page or introduction) actually protects participants better than requiring a signature would.

Chart to consult: Chart 14, Can Documentation of Informed Consent Be Waived Under 45 CFR 46.117(c)?

A caution about the Charts (because OHRP says it explicitly): these charts are generalizations and guides. They're a strong starting point for a determination, not a substitute for reading the actual regulatory text when a protocol sits close to a line. The chair of the IRB should help guide the group to the appropriate regulatory language when necessary.

Next up, you'll find a quiz over the material in this module. After the quiz, there's one more page that serves as training for incoming IRB chairs. The full committee is encouraged to view it for informational purposes.