Undue Influence and Vulnerable Populations in Research

“Undue influence” is a specific term of art in human subjects research ethics, distinct from — and often confused with — coercion and ordinary incentive. This guide covers how the Common Rule (45 CFR 46) defines and regulates it, which populations receive additional Subpart-level protections because of heightened vulnerability, and what actually triggers extra IRB scrutiny during protocol review. It assumes you already know what informed consent requires in general; for that foundation, see Informed Consent in Research, The 4 Principles of Informed Consent, and The Components of Informed Consent. This page picks up where those leave off, on the narrower question of when a consent process is compromised by influence, incentive, or a participant’s diminished capacity to refuse.

Coercion, Undue Influence, and Undue Inducement: Three Different Things

These three terms are frequently used interchangeably in casual conversation but describe different failure modes in the consent process, and IRBs review them differently:

• Coercion is a threat of harm for refusing to participate, or a threat of withdrawing a benefit the person is already entitled to. It is the narrowest and most serious of the three — a credible threat, explicit or implied.
• Undue influence is the broader category: any excessive, unwarranted, or inappropriate overture offered to obtain compliance, whether or not it rises to a threat. It can be monetary or non-monetary — for example, a supervisor recruiting subordinates, or a physician recruiting their own patients, where the relationship itself creates pressure to say yes regardless of any payment involved.
• Undue inducement is the payment-specific instance of undue influence: compensation structured or sized in a way that leads a reasonable person to accept a risk they would otherwise decline, or to withhold accurate information (about symptoms, prior participation, eligibility) in order to qualify for or complete the study.

There is no fixed dollar threshold in federal regulation for when a payment becomes an undue inducement — it is a case-by-case IRB judgment that weighs payment size against the level of research risk, how the payment is structured (see below), and the vulnerability of the population being recruited. For the full glossary treatment of the payment-specific concept, see Inducements in Research (Undue Inducement); for how this plays out in trial budgeting and payment-schedule design specifically, see Clinical Trial Participant Payments.

The Regulatory Basis: 45 CFR 46.111(b)

The Common Rule’s IRB approval criteria, at 45 CFR 46.111(b), require the IRB to determine that “when some or all of the subjects are likely to be vulnerable to coercion or undue influence, such as children, prisoners, individuals with impaired decision-making capacity, or economically or educationally disadvantaged persons, additional safeguards have been included in the study to protect the rights and welfare of these subjects.” Two things about this text matter for how it is actually applied:

• The “such as” list is illustrative, not exhaustive. An IRB can and should treat other populations as vulnerable under 46.111(b) even though they aren’t named in the regulation or covered by a dedicated Subpart — see the discussion of deferential vulnerability below.
• Only three of the categories named in 46.111(b) — pregnant women/fetuses/neonates, prisoners, and children — have a dedicated regulatory Subpart spelling out mandatory additional protections. The others (impaired decision-making capacity, economic or educational disadvantage) are addressed through the same general “additional safeguards” requirement, evaluated by the IRB on a study-by-study basis, without a separate numbered Subpart.

Vulnerable Populations With a Dedicated Subpart: B, C, and D

45 CFR 46 has three additional-protection subparts beyond Subpart A (the Common Rule’s general provisions), each adding requirements on top of the baseline consent and IRB-review rules:

• Subpart B — pregnant women, human fetuses, and neonates. Adds requirements around risk to the fetus, the father’s role in consent for certain fetal research, and restrictions on research involving nonviable neonates.
• Subpart C — prisoners. Restricts the categories of research permitted, and requires a prisoner representative to sit on the IRB reviewing the study, given the inherently constrained, custodial setting that limits a prisoner’s ability to decline freely.
• Subpart D — children. Requires parental or guardian permission under §46.408, plus the child’s own assent where the IRB determines the child is capable of it — assent and parental permission are both required, not one or the other, unless the IRB waives assent under specified conditions.

What these three subparts share is a common design logic: each addresses a situation where the subject’s capacity to give voluntary, uncoerced consent is structurally limited — by developmental stage, by legal custody, or by physical dependency on the pregnancy — rather than by circumstance alone. That is what separates them from the broader, non-Subpart vulnerability categories covered next.

Vulnerability Without a Subpart: Deferential Vulnerability and Other 46.111(b) Categories

Not every population an IRB treats as vulnerable has a dedicated regulatory Subpart. A recurring category is what research-ethics literature calls deferential vulnerability — a situational vulnerability arising from a power or authority relationship between researcher and subject, rather than from an inherent trait of the subject. Students recruited by their own instructor, employees recruited by their employer or supervisor, and patients recruited by their treating physician all fall into this category: the person may feel obligated to agree, or fear a negative consequence for declining, even absent any explicit threat. Economically or educationally disadvantaged persons and individuals with impaired decision-making capacity are named directly in 46.111(b) but likewise have no dedicated Subpart — IRBs address them through study-specific additional safeguards rather than a fixed regulatory checklist. See Vulnerable Populations in Research for the fuller glossary treatment of how these categories are defined and reviewed.

What Actually Triggers Extra IRB Scrutiny

In practice, protocols get flagged for closer review of undue-influence and vulnerability concerns when one or more of the following is present:

• A pre-existing power relationship between the investigator (or study team) and the pool of potential subjects — instructor/student, supervisor/employee, treating clinician/patient.
• Recruitment from a captive or dependent setting — a correctional facility, an inpatient unit, a residential program — where declining is logistically or socially harder than it would be in an ordinary community setting.
• Payment structures that create pressure to stay enrolled or to conceal disqualifying information — large lump-sum or heavily back-loaded payments contingent on study completion are scrutinized more closely than modest, prorated per-visit payments, because they can discourage a participant from exercising their right to withdraw. See Clinical Trial Participant Payments for how IRBs evaluate payment schedules specifically.
• Subject-matter overlap with a named Subpart population — pregnancy, incarceration, or minors automatically routes the protocol through the applicable Subpart B/C/D requirements in addition to standard review.
• Diminished capacity to understand or weigh the consent information — cognitive impairment, acute illness, or a language/literacy barrier that isn’t adequately addressed by the consent process, which can independently trigger 46.111(b) review even without a payment or power-relationship element.

For a general walkthrough of what a compliant consent process needs to document regardless of population, see the Informed Consent Checklist for Researchers and When Should Informed Consent Be Obtained?.

Coercion vs. Legitimate Incentive: Where the Line Actually Falls

Reasonable compensation for a participant’s time, travel, and inconvenience is not, on its own, coercive or unduly influential — federal guidance treats fair payment for burden as legitimate and consistent with respect for persons. The concern is specifically about structure, not the mere existence of payment:

• Compensation tied to time and burden (per-visit, prorated across the study) is generally viewed as appropriate.
• Compensation tied to completion or paid as a large lump sum at the end is more likely to be scrutinized, because it can discourage withdrawal — which must remain available to the participant at any point, regardless of payment structure, under 45 CFR 46.116(a)(8).
• Non-monetary “incentives” are subject to the same undue-influence analysis as payment — free clinical care, waived fees, or access otherwise unavailable to the person can just as easily create excessive pressure to enroll or remain enrolled.

The regulatory anchor for minimizing undue influence in the consent process itself is 45 CFR 46.116(a)(1), which requires that consent be sought only under circumstances that provide the prospective subject sufficient opportunity to decide whether to participate and that minimize the possibility of coercion or undue influence.

How This Page Relates to Other CASRAI Consent Pages

Because CASRAI covers informed consent from several angles, here is exactly what each related page addresses, so you can go to the right one:

• The Components of Informed Consent — the required disclosure elements under 45 CFR 46.116 (what must be in the consent document/process).
• The 4 Principles of Informed Consent — the bioethics principles (autonomy, beneficence, non-maleficence, justice) underlying consent generally.
• Informed Consent in Research — the full picture of what consent requires: voluntariness, disclosure, comprehension, capacity, and withdrawal.
• This page — the narrower question of when a consent process is compromised by coercion, undue influence, or inducement, and which populations get additional regulatory protection because of heightened vulnerability.

Frequently Asked Questions

Is paying research participants coercive?

Not inherently. Reasonable payment for time, travel, and burden is treated as legitimate compensation, not coercion or undue influence. The concern is about payment structure — large completion-contingent or lump-sum payments can create undue influence by discouraging withdrawal, whereas prorated, per-visit payment is generally viewed as appropriate.

What is the difference between coercion and undue influence?

Coercion is a threat of harm for refusing to participate or for withdrawing. Undue influence is broader: any excessive or inappropriate overture — monetary or not — that could compromise a genuinely voluntary decision, even without a threat involved.

Which populations get additional protections under 45 CFR 46?

Three populations have a dedicated regulatory Subpart: pregnant women, human fetuses, and neonates (Subpart B), prisoners (Subpart C), and children (Subpart D). Other populations named in 46.111(b) — individuals with impaired decision-making capacity, and economically or educationally disadvantaged persons — are addressed through the same “additional safeguards” requirement without a dedicated Subpart, evaluated case-by-case by the IRB.

Does a power relationship between researcher and subject count as undue influence?

It can. This is often referred to as deferential vulnerability — the concern that a student, employee, or patient may feel obligated to agree to a request from an instructor, employer, or treating clinician even without an explicit threat. IRBs typically require additional safeguards, such as recruitment by someone outside the direct authority relationship, when this dynamic is present.

Is there a dollar threshold for when payment becomes an undue inducement?

No. Federal regulation does not set a fixed dollar amount. IRBs make a case-by-case judgment weighing the payment amount and structure against the level of research risk and the vulnerability of the population being recruited.