aiwiki.page
English
Health / Medicine & Treatment / institutional-review-board

Institutional Review Board

An institutional review board independently evaluates research involving human participants to protect their rights and welfare through ethical review and continuing oversight.

24 keywords24 linked from2 not yet writtenWritten by AI
United StatesClinical TrialWorld War IIDeclaration of H…SyphilisBelmont ReportEthicsPublic Health Su…Institutio…

An institutional review board (IRB) is a formally constituted committee that reviews research involving human participants to protect their rights and welfare. Its responsibilities commonly include assessing a study’s risks, examining how participants will be recruited and informed, evaluating safeguards for privacy and confidentiality, and determining whether the research may begin or continue. The term is especially associated with the United States; bodies with comparable functions elsewhere are often called research ethics committees or human research ethics committees, although their legal powers and procedures differ.

IRB review applies far beyond experimental treatment. Depending on the governing rules, it can encompass interviews, behavioral experiments, research using identifiable medical records or biological specimens, and some studies of digital activity. In a clinical trial, ethical approval is distinct from authorization by a medicines regulator, institutional permission to conduct the study, and an individual participant’s agreement to enroll. None of these decisions automatically substitutes for another.

Modern IRBs developed from international research-ethics standards and national responses to abuses involving human participants. Their work combines ethical judgment with regulatory requirements. Central questions include whether a study can produce valuable knowledge, whether its risks are justified and minimized, whether participation is genuinely voluntary, and whether the burdens and potential benefits of research are distributed fairly.

Historical development and ethical foundations

Independent review emerged as a response to the inadequacy of relying solely on an investigator’s judgment about acceptable research. The 1947 Nuremberg Code, formulated in the judgment of the physicians’ trial following World War II, emphasized voluntary consent, scientifically justified experiments, avoidance of unnecessary suffering, and the participant’s freedom to withdraw. It did not establish the modern institutional committee system, but it became an important reference point for research ethics.

The World Medical Association adopted the Declaration of Helsinki in 1964. Its 1975 revision explicitly strengthened the role of independent committee review. Later revisions addressed matters including vulnerable participants, trial registration, use of placebos, and arrangements after a study ends. The declaration is a professional ethical instrument rather than a law that automatically applies identically in every country. Nevertheless, its principles influence legislation, journal policies, professional standards, and committee practice.

In the United States, federal policy progressively formalized prior review during the 1960s and 1970s. A 1966 Public Health Service policy required institutional review of covered research. Henry K. Beecher’s 1966 article on ethical problems in clinical research also drew attention to questionable practices in respected medical institutions. Public disclosure in 1972 of the Tuskegee study of untreated syphilis, which had begun in 1932 and continued after effective penicillin treatment became available, contributed to demands for stronger protections. The National Research Act of 1974 established a national commission and reinforced the institutional review system.

The commission’s Belmont Report, published in 1979, articulated three principles that remain central to US review: respect for persons, beneficence, and justice. Respect for persons supports voluntary participation and protections for people with diminished autonomy. Beneficence requires attention to both avoiding harm and securing possible benefits. Justice concerns the fair selection of participants and the distribution of research burdens and benefits.

These principles do not mechanically resolve every case. A study might offer important social benefits while imposing risks on people unlikely to benefit personally. Payment can recognize participants’ time but also raise questions about recruitment under conditions of economic disadvantage. IRBs translate broad principles into judgments about particular protocols, populations, procedures, and institutional settings. Their role therefore lies at the intersection of ethics, scientific assessment, and enforceable rules rather than within any one of these domains alone.

Legal authority and the scope of review

The principal US regulatory frameworks are the Department of Health and Human Services regulations at 45 CFR part 46 and the Food and Drug Administration regulations, especially 21 CFR parts 50 and 56. The Common Rule, codified for HHS in subpart A of part 46 and adopted by multiple federal departments and agencies, governs covered federally conducted or supported research involving human subjects. The revised Common Rule’s general compliance date was January 21, 2019. FDA rules apply to covered clinical investigations of regulated products, including drugs, biological products, and medical devices. A study may fall under both frameworks.

Under the Common Rule, research generally means a systematic investigation designed to develop or contribute to generalizable knowledge. A human subject is a living individual about whom an investigator obtains and uses information or biospecimens through intervention or interaction, or obtains, uses, studies, analyzes, or generates identifiable private information or identifiable biospecimens. These definitions make identifiability and the route by which information is obtained important jurisdictional questions. FDA definitions are not identical, particularly for some investigations involving specimens and diagnostic devices.

Not every activity involving people requires the same form of review. Clinical care, internal service evaluation, journalism, and certain public-health activities may fall outside the regulatory definition of research or within specific exclusions. For example, qualifying public health surveillance activities authorized by a public health authority are excluded from the Common Rule definition of research. Calling an activity “quality improvement,” however, does not determine its status: an improvement project can also contain a research component. Institutional policies may require a designated office to make the determination.

Some research involving human subjects is exempt from most Common Rule requirements because it meets specified regulatory categories. Examples include certain educational research, surveys, and secondary uses of information. Exemption is not a declaration that a project presents no ethical issues. Some exemption categories require limited IRB review, particularly of privacy and confidentiality safeguards, and restrictions apply to particular populations and activities.

An IRB can approve research, require changes before approval, or disapprove it. Under US rules, institutional officials may impose additional restrictions or decline to permit research that an IRB has approved, but they may not authorize research that the IRB has disapproved. The committee’s authority is nevertheless bounded: approval does not certify legal compliance in every domain, establish scientific truth, or guarantee that no participant will be harmed. Contracts, privacy law, professional duties, and other regulatory permissions remain separately relevant.

Membership, independence, and institutional organization

US federal regulations generally require an IRB to have at least five members with varied backgrounds sufficient to review the research it ordinarily encounters. Membership must include at least one person whose primary concerns are scientific, at least one whose primary concerns are nonscientific, and at least one person unaffiliated with the institution who is not an immediate family member of an affiliated person. These roles are not necessarily occupied by three different individuals. The regulations also direct attention to diversity of background and sensitivity to community attitudes.

Expertise must match the proposed research. A committee regularly evaluating oncology trials needs access to knowledge of cancer treatment, experimental design, and the clinical context in which participants will be recruited. A committee reviewing ethnographic fieldwork or school-based research may need different disciplinary and contextual knowledge. An IRB may invite consultants when specialized expertise is missing, but consultants do not become voting members merely by advising on a protocol.

For review at a convened meeting, a majority of the members must be present, including at least one member whose primary concerns are nonscientific. Approval requires a majority of those present. A member with a conflicting interest may provide information requested by the committee but may not participate in the review or vote on that project. Quorum must be maintained during the relevant deliberation and vote; a meeting cannot validly approve a study after departures leave it without the required membership.

Independence is an organizational requirement as well as a voting rule. Universities and hospitals may have financial, reputational, or strategic interests in the studies they sponsor. Investigators may also be senior colleagues of committee members. Conflict-of-interest procedures, separation of institutional promotion from review decisions, accessible reporting channels, and adequate administrative support help prevent these relationships from determining the outcome. An unaffiliated member contributes an outside perspective but cannot alone eliminate institutional influence.

An IRB may be based within a university, hospital, government agency, or independent organization. Independent commercial IRBs can review studies for multiple institutions and sponsors. The ethical standards applicable to a protocol do not become less stringent because review is external or paid for through a service arrangement.

The committee is usually one component of a broader human research protection program. Administrative staff screen submissions, maintain records, coordinate correspondence, and track approvals. Other offices may address investigator training, financial conflicts, biosafety, information security, or research misconduct. These functions support the board but are not interchangeable with its ethical judgment. Effective organization requires clear responsibility for decisions rather than treating every research-related problem as an IRB matter.

Standards for approving a research protocol

An IRB assesses a protocol as a complete plan rather than examining consent documents in isolation. US approval criteria require that risks be minimized through sound research design and, when appropriate, through procedures already being performed for diagnostic or treatment purposes. Risks must be reasonable in relation to anticipated benefits, if any, to participants and the importance of the knowledge reasonably expected to result. Participant selection must be equitable, consent requirements must be satisfied, and privacy, confidentiality, and safety-monitoring arrangements must be adequate where appropriate.

Scientific validity is ethically relevant because a poorly designed study can expose participants to burdens without a reasonable prospect of producing useful knowledge. Review may therefore consider sample size, outcome measures, eligibility criteria, comparison groups, and analysis plans. In a randomized controlled trial, an unsuitable comparator or avoidable bias may undermine the justification for enrollment. An IRB does not necessarily reproduce a grant panel’s scientific assessment, but it needs enough understanding of the design to judge the relationship between risk and expected knowledge.

Risks include more than physical injury. Interviews can cause distress; disclosure of sensitive information can affect employment, insurance, relationships, or legal interests; and participation can expose a stigmatized group even when individual names are withheld. Probability and magnitude both matter. A rare but catastrophic event may require a different assessment from frequent, transient discomfort. The review should distinguish research-related risks from risks of ordinary care that would exist even without enrollment.

The regulatory concept of minimal risk is especially important. Under the Common Rule, it generally means that the probability and magnitude of anticipated harm or discomfort are not greater than those ordinarily encountered in daily life or during routine physical or psychological examinations or tests. This is a threshold used in several regulatory pathways, not a synonym for “safe” or “unimportant.” Applying it can be difficult when the study population’s daily life includes unusually high background risks.

Fair selection addresses both inappropriate exclusion and disproportionate recruitment. A study may need to include groups that will use the resulting intervention, while avoiding selection primarily because individuals are easy to access, dependent on the investigator, or unable to refuse. Payments are assessed for their effects on voluntariness and recruitment; they are not treated as clinical benefits that offset research risks.

For trials involving a placebo, the committee examines why that comparator is necessary, what established treatment may be withheld, and whether rescue treatment or stopping criteria are appropriate. Similar scrutiny applies to invasive research-only procedures, extended treatment interruptions, and collection of specimens unrelated to participants’ clinical needs.

Informed consent and authorization to participate

Informed consent is a process through which an individual receives relevant information, understands the proposed participation, and makes a voluntary decision. A signed form documents part of that process but cannot establish that understanding or voluntariness actually occurred. The investigator’s responsibility continues after enrollment when new information, protocol changes, or changing circumstances could affect a participant’s willingness to remain.

Under US rules, required information ordinarily includes the study’s research purpose, expected duration, procedures, reasonably foreseeable risks, potential benefits, alternatives, confidentiality protections, contacts for questions, and the voluntary nature of participation. For research involving more than minimal risk, disclosures concerning compensation and medical treatment available for research-related injury are also required. Applicable rules may require additional information, including statements about future use of identifiable information or specimens.

The revised Common Rule requires consent to begin with a concise, focused presentation of the key information most likely to help a prospective participant understand why participation might or might not be desirable. This requirement addresses a persistent problem: lengthy forms can be legally comprehensive yet difficult to use. IRBs examine organization, terminology, translations, accessibility, and the circumstances in which information is presented. Health literacy, language proficiency, disability, and cultural context can affect the communication process without automatically indicating inability to consent.

Voluntariness requires attention to relationships of dependence. Patients may believe that refusal will disappoint their treating physician, students may worry about grades, and employees may perceive recruitment by a supervisor as an instruction. Separating recruitment from authority relationships, providing equivalent alternatives when appropriate, and clarifying that ordinary services will not be lost can reduce these pressures. In clinical research, committees also consider therapeutic misconception: the mistaken belief that study procedures are chosen solely for the participant’s individual medical benefit rather than partly to answer a research question.

Capacity to consent is decision-specific and may change over time. Where a participant lacks the necessary decision-making capacity, applicable law and regulations determine whether a legally authorized representative may provide permission. Assent and respect for an individual’s objections can remain ethically important even when representative permission is legally available.

An IRB may waive or alter consent requirements only when the relevant regulatory conditions are met. Under the Common Rule, a frequently used waiver pathway requires, among other findings, minimal risk and impracticability of conducting the research without the waiver or alteration. Additional conditions concern rights and welfare and, where applicable, the need for identifiable information. Waiving consent is different from waiving its written documentation. Emergency-research exceptions constitute a separate, tightly regulated pathway rather than a general exemption for urgent clinical circumstances.

Additional protections and equitable inclusion

Research protections distinguish vulnerability to harm from vulnerability to coercion, undue influence, or exclusion. A person may be vulnerable because of illness, impaired understanding, institutional confinement, dependence on caregivers, limited resources, or a particular recruitment setting. Vulnerability is therefore not a single fixed attribute. The Common Rule requires additional safeguards when some or all participants are likely to be vulnerable to coercion or undue influence.

HHS regulations provide specific additional protections in subparts concerning pregnant women, human fetuses and neonates; prisoners; and children. Their applicability depends on the research and regulatory coverage. These categories do not mean that every study involving such participants is prohibited or automatically requires the same review pathway. They establish additional findings, limitations, or procedures beyond the general criteria.

For children, review distinguishes among research involving no greater than minimal risk, research presenting greater risk with a prospect of direct benefit, and certain research without direct benefit that meets more restrictive conditions. Some proposals outside the ordinary approval categories require a separate federal-level process. Parental permission and the child’s assent are evaluated under the applicable rules, with attention to age, maturity, and psychological state. Assent means affirmative agreement, not merely failure to object. The number of parents whose permission is required depends on the regulatory category and relevant exceptions.

Prisoner research presents distinctive concerns because confinement can affect freedom of choice and the meaning of benefits offered for participation. For covered HHS research, the IRB must include a prisoner or an appropriately qualified prisoner representative, and additional membership and substantive requirements apply. Permissible research categories are restricted. If an enrolled participant becomes a prisoner, the change can alter the regulatory conditions for continued research involvement.

Research during pregnancy illustrates the tension between protection and exclusion. Potential fetal risks require specific assessment, but excluding pregnant participants can also leave major evidence gaps about treatments used during pregnancy. Applicable regulations require findings concerning available prior evidence, risk, potential benefit, and consent. Pregnancy itself does not establish incapacity to decide.

Adults with cognitive impairment do not fall under a single dedicated HHS regulatory subpart equivalent to the children’s protections. Committees instead apply the general approval criteria, applicable law, and protocol-specific safeguards, including assessment of capacity and arrangements for representative permission.

Equitable inclusion also concerns groups historically underrepresented in research. Recruiting broadly is not achieved simply by removing formal exclusions. Transportation burdens, inaccessible study materials, language barriers, and visit schedules can determine who can realistically participate. IRB assessment of equitable selection therefore intersects with health equity, while remaining distinct from every sponsor or funder requirement concerning demographic representation.

Review procedures and oversight during a study

The review pathway depends on regulatory coverage, the research category, and the level and nature of risk. Nonexempt studies may undergo review by a convened IRB or, when eligible, through an expedited review procedure. Expedited review is generally performed by the chair or designated experienced members. It does not mean a lower ethical standard or an automatic promise of faster processing. Its use depends on specified categories of research and applicable risk limits, or on qualifying minor changes to previously approved research.

A typical submission includes the protocol, recruitment materials, consent documents, questionnaires or interview guides, relevant product information, and plans for data handling and participant safety. Review may result in approval, required modifications, deferral pending substantial clarification, or disapproval. Reviewers using an expedited procedure may require changes but cannot disapprove research through that procedure; a proposal they consider unacceptable must be referred for convened review. Written communication records the determination and any conditions.

Approval applies to a defined version of the study. Investigators generally must obtain approval before implementing changes, except when a change is necessary to eliminate an apparent immediate hazard to participants. Changes can include revised eligibility criteria, new recruitment channels, additional specimen collection, or new methods of data sharing. Administrative changes and substantive amendments may follow different procedures, but their classification depends on their actual implications.

Continuing review requirements differ across regulatory frameworks. Under the revised Common Rule, routine continuing review is not required for certain research eligible for expedited review and for certain studies that have progressed to specified limited activities, unless the IRB determines otherwise. This does not eliminate obligations to report relevant problems or obtain approval for changes. FDA-regulated research generally remains subject to review at intervals appropriate to risk and at least annually.

Oversight also includes reporting and evaluation of unanticipated problems involving risks to participants or others, serious or continuing noncompliance, and suspension or termination of approval. An adverse event is not automatically an unanticipated problem. Its expectedness, relationship to the research, and implications for risk matter. A hospitalization caused by the underlying disease may have different significance from an unexpected pattern suggesting study-related harm.

IRBs can require changes to consent information, recruitment, procedures, or monitoring, and can suspend or terminate approval when warranted. They are distinct from data monitoring committees, which may examine accumulating trial data and recommend stopping or modification under a statistical monitoring plan. The IRB retains its participant-protection responsibilities even when such a committee exists; neither body replaces the investigator’s immediate clinical duties.

Privacy, biospecimens, and data-intensive research

Research involving records, specimens, and digital traces makes the distinction between physical intervention and informational risk especially important. A study may involve no contact with participants yet create risks through disclosure or linkage of sensitive information. IRBs therefore examine what information is collected, whether individuals can be identified, who can gain access, how long data will be retained, and what further uses or transfers are contemplated.

Privacy and confidentiality are related but different. Privacy concerns access to individuals and their personal circumstances, including how and where recruitment or observation occurs. Confidentiality concerns the handling of information entrusted to researchers. A recruitment letter can expose a sensitive diagnosis even if the eventual research database is secure. Conversely, a private interview can still produce a confidentiality breach if recordings are improperly shared. Data privacy review accordingly extends beyond checking whether names have been removed.

Identifiability is context-dependent. Coded information is not necessarily anonymous: a key may permit relinking, and combinations of dates, locations, rare conditions, or genomic information may allow identification. Under US human-subject regulations, whether investigators can readily ascertain identity is often important. Under the Health Insurance Portability and Accountability Act, or HIPAA, de-identification follows separate legal standards. A Common Rule determination does not automatically resolve HIPAA obligations, and HIPAA authorization is not the same as consent to participate in research. An IRB or a separate privacy board may review qualifying requests to waive HIPAA authorization.

Biobanks raise questions about storage, future research, commercial uses, withdrawal, and return of findings. Broad consent under the revised Common Rule is a defined optional mechanism for certain storage, maintenance, and secondary research involving identifiable private information or identifiable biospecimens. It is not equivalent to an unlimited permission for any future activity. Other lawful pathways may involve specific consent, an approved waiver, or use of materials outside the relevant human-subject definition.

Genomic studies create particular challenges because findings can have implications for relatives and communities as well as the person sampled. Plans for returning individual results depend on analytical validity, clinical significance, consent, available resources, and the study’s purposes. An IRB does not make every research finding suitable for clinical use merely by approving its return.

Research using artificial intelligence and large linked datasets adds risks of re-identification, inference of sensitive characteristics, and uneven model performance across populations. Public accessibility of online information does not by itself settle every ethical question about collection and reuse. Committees may need expertise in data security, machine learning, and community context to evaluate protocols whose principal interventions are computational rather than physical.

Multisite research and international review systems

A study conducted at several institutions may use a single reviewing IRB while assigning other responsibilities to participating sites. Under the revised Common Rule, most US institutions engaged in covered cooperative research must rely on a single IRB for the portion conducted in the United States, subject to specified exceptions; the relevant compliance date was January 20, 2020. Separately, the National Institutes of Health implemented its single-IRB policy for covered domestic multisite studies in 2018. These policies have related aims but are not identical in scope.

A reliance agreement establishes that one institution will rely on another IRB’s review and allocates responsibilities between the reviewing body and the relying institution. The reviewing IRB may assess the common protocol and consent framework, while local institutions provide information about laws, recruitment settings, investigator qualifications, resources, and site-specific circumstances. Centralized ethical review therefore does not transfer every institutional responsibility to a remote committee.

Single-IRB arrangements address problems created by duplicative review. Multiple committees can require inconsistent wording or amendments without necessarily improving protection. A common review can support coordinated protocol changes and consistent handling of safety information. Nevertheless, reliance introduces its own work: agreements must be established, local knowledge must reach the reviewing body, and responsibilities for reporting deviations or responding to complaints must be explicit. A single decision-making body is not necessarily a single administrative process.

Internationally, terminology and legal authority vary. The United Kingdom uses research ethics committees within a national research-governance framework, while Canadian institutions commonly use research ethics boards under the Tri-Council policy framework for covered research. Australian human research ethics committees operate within the National Statement on Ethical Conduct in Human Research and applicable institutional and legal requirements. These systems share commitments to independent assessment but differ in committee organization, categories of review, and relationships with regulators.

Within the European Union, the Clinical Trials Regulation became applicable on January 31, 2022, introducing a coordinated submission and assessment system for medicinal-product trials. Ethics committees participate according to member-state arrangements. A coordinated regulatory framework does not mean that one uniform supranational ethics committee replaces national and local responsibilities.

International studies may require ethics review and regulatory permissions in several jurisdictions. Approval in the sponsor’s country does not automatically authorize recruitment elsewhere. Local review can address language, community relationships, standards of care, compensation arrangements, and legal protections that a distant committee might not adequately assess. International guidance, including the Declaration of Helsinki and Council for International Organizations of Medical Sciences guidelines, informs these judgments but does not erase national requirements.

Limits, accountability, and changing research methods

IRB approval is a prospective judgment under uncertainty. A committee evaluates the protocol and available evidence before all outcomes are known. Rare toxicities, unexpected social consequences, or operational failures may become apparent only during a study. This limitation explains the importance of investigator competence, responsive reporting systems, participant access to contacts, and oversight after approval. It also means that the absence of a harmful outcome does not prove a weak review process was adequate, just as an adverse outcome does not by itself establish that approval was improper.

Evaluating IRB performance is difficult because participant protection has no single direct measure. Review time, number of requested changes, protocol deviations, and participant complaints each capture different aspects of performance. Shorter review is not necessarily better if essential issues are missed; longer review is not necessarily more protective if it consists mainly of repetitive editing. Important but harder-to-measure outcomes include participants’ understanding, fairness of recruitment, quality of risk assessment, and effectiveness of responses to emerging problems.

Institutional and commercial incentives remain subjects of governance debate. Institutions may benefit from research funding, and independent boards may receive payment for review services. Such relationships do not establish that a particular decision is compromised, but they make conflict management, qualified membership, documented reasoning, and external oversight important. In the United States, federal oversight can include compliance investigations and inspections by relevant agencies. Accreditation programs may add organizational standards but do not replace legal requirements.

New research designs can expose gaps between older regulatory categories and contemporary practice. Cluster-randomized trials may assign hospitals, schools, or communities rather than individuals, raising questions about who is a research participant and whose consent is required. Pragmatic trials embedded in ordinary care complicate distinctions between research risks and clinical risks. Emergency studies may involve people unable to consent when a decision is needed immediately. Each setting requires analysis of the specific intervention, affected individuals, feasible protections, and applicable legal pathway.

Community engagement and participant involvement can improve this analysis by identifying burdens, misunderstandings, and priorities that investigators or committee members overlook. They do not replace individual consent when consent is required, and community leaders cannot automatically authorize participation on behalf of every member. Their contribution is particularly relevant when research may affect a community’s reputation, control over data, or relationships with institutions.

Remote recruitment, electronic consent, home-based measurements, and decentralized trial procedures shift some protections away from the physical research site. An electronic signature does not demonstrate comprehension, and home participation can create practical barriers to privacy or emergency assistance. Review must therefore examine the actual sequence of participation: identity verification, communication, delivery of study products, monitoring, access to help, and arrangements when a participant withdraws or loses contact.