IRB Requirements for Special Populations

Some groups of people cannot give fully autonomous consent, or face pressures that could compromise their voluntary participation, or bear disproportionate risks from research. Federal regulations require the IRB to apply additional protections when these populations are enrolled. This guide summarizes Subparts B, C, and D of 45 CFR 46, plus commonly encountered populations that do not have a dedicated subpart but nonetheless require extra care.

Research with any special population almost always means full board review. Budget the timeline accordingly.

Pregnant individuals, fetuses, and neonates — Subpart B

45 CFR 46 Subpart B applies to research involving pregnant women, human fetuses, and neonates. Approval requires that all of the following be met:

• Preclinical and clinical studies, including those on non-pregnant women, have been conducted to the extent possible and provide data for assessing potential risks
• Risk to the fetus is caused solely by interventions or procedures that hold out the prospect of direct benefit to the pregnant woman or fetus; or, if there is no such prospect of benefit, the risk to the fetus is not greater than minimal and the purpose is the development of important biomedical knowledge not otherwise obtainable
• Any risk is the least possible for achieving the study's objectives
• If the research holds out the prospect of direct benefit to the pregnant woman only, the pregnant woman's consent is obtained
• If the research holds out the prospect of direct benefit to the fetus or to both, the consent of both the pregnant woman and the father is required (with exceptions for unavailability, incompetence, or pregnancy resulting from assault)
• No inducements are offered to terminate a pregnancy

Pregnancy alone does not disqualify participants from most research. Blanket exclusion of pregnant people from benign behavioral or survey research is an equity violation, not a protection.

Prisoners — Subpart C

"Prisoner" under 45 CFR 46.303(c) means any individual involuntarily confined or detained in a penal institution, including pretrial detention and court-ordered commitment. If a participant becomes a prisoner during the study, the IRB must be notified promptly — the study cannot continue enrolling them or analyzing their data without re-review.

Permissible research with prisoners is limited under 45 CFR 46.306 to four categories:

1. Study of the possible causes, effects, and processes of incarceration and criminal behavior, if the research presents no more than minimal risk and no more than inconvenience
2. Study of prisons as institutional structures or of prisoners as incarcerated persons, if the research presents no more than minimal risk and no more than inconvenience
3. Research on conditions particularly affecting prisoners (e.g., hepatitis, drug addiction, sexual assault), subject to Secretary notification requirements
4. Research on practices that may improve the health or well-being of prisoners, subject to Secretary notification

Approval also requires the IRB to include a prisoner or prisoner representative and to affirmatively find that advantages offered to prisoners are not so large that they impair the ability to weigh risks freely (45 CFR 46.305).

Children — Subpart D

45 CFR 46 Subpart D defines "children" as persons who have not attained the legal age for consent to treatments or procedures involved in the research under the applicable law of the jurisdiction — generally under 18 in most US states. Four categories of research with children are permissible:

• 46.404 — Not greater than minimal risk. Permissible with parental permission and child assent as appropriate.
• 46.405 — Greater than minimal risk with prospect of direct benefit. The benefit must be at least as favorable as available alternatives; parental permission and assent required.
• 46.406 — Greater than minimal risk, no prospect of direct benefit, but likely to yield generalizable knowledge about the child's disorder or condition. Requires a minor increase over minimal risk, commensurate with the child's situation, and permission from both parents where reasonably feasible.
• 46.407 — Research not otherwise approvable. Requires Secretary-level review and public comment. Rare.

Assent and parental permission

Assent means the child's affirmative agreement. Mere failure to object is not assent. The IRB determines whether assent is required based on the ages, maturity, and psychological state of the children involved (45 CFR 46.408). For most research, parental permission is also required; for greater-than-minimal-risk research without direct benefit, both parents must generally sign unless one is unavailable, incompetent, or not reasonably available. See the consent form guide for assent language examples.

Cognitively impaired adults

The Common Rule does not have a dedicated subpart for cognitively impaired adults, but 45 CFR 46.111(b) requires additional safeguards when research involves subjects who are "likely to be vulnerable to coercion or undue influence." Recommended protections:

• Capacity assessment by a qualified clinician before enrollment
• Legally authorized representative (LAR) consent under state law when the individual cannot consent
• Ongoing assent from the participant, with withdrawal honored even if the LAR wishes to continue
• Re-consent if capacity is regained
• Independent study monitor for greater-than-minimal-risk research

Students as participants

Enrolling students from your own course or program is a classic coercion risk. Students may feel that refusing would affect their grades or relationship with the instructor. IRBs typically require:

• A third party (not the instructor) to recruit and consent students
• Recruitment explicitly stating that participation is unrelated to grades
• Data collection after grades are submitted, if feasible
• Anonymous procedures so the instructor cannot identify participants
• Alternative extra-credit options of equivalent value if credit is offered for participation

Employees as participants

Workplace research raises similar concerns. Employees may fear career consequences for refusing or for disclosing unfavorable views. Safeguards include:

• Recruitment through a neutral party rather than a supervisor
• Assurance that participation (or non-participation) will not affect employment
• Aggregate-only reporting so individual responses cannot be identified
• Data storage outside employer-controlled systems

Economically or socially vulnerable groups

Research with populations in economically precarious situations — people experiencing homelessness, undocumented immigrants, people who use drugs — requires careful attention to whether compensation may be coercive, whether disclosure could create legal or social harm, and whether the research is designed with rather than on the community. A Certificate of Confidentiality may be warranted; see the data management guide.

Cross-cutting checklist for special-population protocols

• Applicable subpart identified and cited in the protocol
• Specific regulatory criteria addressed section by section
• Consent, permission, and assent documents appropriate to the population
• Additional safeguards against coercion or undue influence
• Clinical or community expertise on the study team
• Plan for ongoing monitoring of well-being during and after the study

If your study enrolls any of these populations, assume full board review and plan a longer timeline. Work closely with your IRB administrator early — many offices offer pre-submission consultations specifically for special-population protocols.