Skip to main content
v2026.11,610 entries · CC-BY 4.0
LAC HealthLaboratory & ResearchLab & research supplies.Reagents, consumables, PPE & instruments — documented, fast, chain-of-custody shipping.Shop lac.us lac.us
Dictionary termTrack Proposedv2026.1

Central IRB / Single IRB (sIRB)

The model in which one IRB of record -- rather than each participating site's local IRB -- conducts the ethical review for a multi-site human-subjects study under a single approved protocol, with all other participating sites relying on that review through a documented reliance arrangement (typically an IRB Authorization Agreement) instead of each convening a duplicate local review.

ByCASRAI Editorial Board
· Last updated 18 Jul 2026

Examples

Worked examples

  • Is an instance

    A 12-site NIH-funded behavioral intervention trial designates the coordinating center's IRB as the sIRB; all 11 other sites sign an IRB Authorization Agreement (via SMART IRB) relying on that single review instead of each convening its own IRB meeting for the same protocol.

  • Is an instance

    A study team requests an exception to the sIRB requirement from its NIH institute because one participating site is a tribal health organization where tribal law requires local IRB review regardless of the reliance arrangement.

Counter-examples

Looks similar, but isn't

  • Not an instance

    A single-site study with only one participating institution has no reliance question to resolve -- that institution's own IRB reviews it directly, and the central/single-IRB model does not apply.

  • Not an instance

    Two sites each independently obtaining their own local IRB approval for the identical multi-site protocol, with no reliance agreement between them, is the traditional duplicative-review model the sIRB policy was designed to replace, not an instance of it.

Editorial commentary

A Central IRB — also called a Single IRB (sIRB) or IRB of record — is the one Institutional Review Board designated to conduct the ethical review for a multi-site human-subjects study under a single protocol, on behalf of every participating site. Instead of each site’s local IRB independently reviewing and approving the same protocol — historically the default, and a major source of duplicated effort, inconsistent stipulations, and review-timeline delays across multi-site research — participating sites rely on the sIRB’s review through a documented reliance arrangement. Local site staff still handle site-specific responsibilities (informed consent administration, local context assessment, reportable-event tracking at that site), but the actual protocol-level ethical determination is made once, by one board, for all sites.

The NIH single-IRB policy

The central/single-IRB model became a default expectation for federally funded multi-site research through NIH’s single-IRB (sIRB) policy, issued 21 June 2016 and effective for competing grant applications with due dates on or after 25 January 2018. The policy requires domestic sites participating in NIH-funded, multi-site, non-exempt human subjects research that uses the same protocol to rely on a single IRB of record, rather than each site obtaining separate local IRB approval. It applies to qualifying multi-site human subjects research generally — not only to studies that meet NIH’s specific clinical-trial definition, though NIH-defined clinical trials are the paradigm case investigators encounter it in. The policy’s effective date lines up with the broader package of NIH clinical-trial reforms rolled out the same period, including the Good Clinical Practice training requirement and the consolidated Human Subjects and Clinical Trials Information form.

The policy carries two general exception categories: research where a local IRB review is required by state or tribal law, and situations where the funding NIH institute or center determines that use of a single IRB is not appropriate for the particular study. Applicants request the sIRB model in the grant application itself, naming the proposed IRB of record and describing the reliance arrangement.

How reliance actually works: the IRB Authorization Agreement

Designating one IRB as the reviewing body does not, by itself, give that board legal authority to act on another institution’s behalf. That authority is established through a reliance agreement — most commonly an IRB Authorization Agreement (IAA), sometimes called a reliance agreement or IRB cooperation agreement — signed between the institution relying on the review (the “relying site”) and the institution whose IRB is serving as the IRB of record. The agreement documents which institution’s IRB has review authority, what obligations the relying institution retains (informed consent processes tailored to local context, local investigator qualifications, prompt reporting of local unanticipated problems), and how communication flows between the reviewing IRB and each relying site.

In practice, institutions rarely negotiate a bespoke agreement for every study. Master reliance agreements executed once between two institutions (or joined through a shared reliance framework) let subsequent studies invoke the existing agreement rather than drafting new terms each time. The SMART IRB platform is the most widely used shared reliance infrastructure in the US — participating institutions execute one master agreement that then covers reliance for any qualifying study between them, considerably reducing the administrative overhead of standing up a new IAA per protocol. A study’s IRB of record is documented on the relevant regulatory paperwork (including the institution’s Federalwide Assurance filing), and every relying site’s obligations under the reliance agreement remain in force for the life of the study, alongside their own site-specific responsibilities.

What stays local even under a central IRB

  • Informed consent administration — the reviewing IRB approves the master consent form and any site-specific language, but each site’s staff still obtain and document consent locally.
  • Local context review — some reliance models retain a lightweight local review step (sometimes called a “worksheet” or local-context assessment) to confirm the protocol is appropriate given local law, institutional policy, and community considerations, even though the substantive ethical review isn’t repeated.
  • Reportable events at the site — local adverse events, protocol deviations, and unanticipated problems are still identified and reported locally, then routed to the IRB of record per the reliance agreement’s timelines.
  • Investigator qualifications and training — verifying that local site investigators and staff are qualified and appropriately trained remains a site responsibility.

Central IRB vs. the general IRB concept

Every central/single IRB is a fully constituted IRB in the ordinary regulatory sense — membership requirements under 45 CFR 46.107, review authority under 45 CFR 46.109 — there is no separate regulatory category for “central IRBs” distinct from IRBs generally. What distinguishes the central/single-IRB model is purely structural: it describes an arrangement where one such board’s review is extended, via reliance agreement, to cover sites beyond its home institution, replacing what would otherwise be independent, duplicative reviews by each site’s own IRB. A central IRB may be a commercial IRB, an academic medical center’s IRB acting for its network, or any IRB institutions agree to rely on — the term describes the review’s scope and reliance mechanism, not the board’s organizational type.

Related terms

References

  • NIH, Final NIH Policy on the Use of a Single Institutional Review Board for Multi-Site Research (NOT-OD-16-094), effective 25 January 2018
  • 45 CFR 46 (Common Rule), Subpart A
  • SMART IRB reliance platform (smartirb.org)

Machine-readable encodings

Use in your systems

JATS XML <role> element
xml
<role vocab="credit"
      vocab-identifier="https://casrai.org/dictionary/"
      vocab-term="Central IRB / Single IRB (sIRB)"
      vocab-term-identifier="https://casrai.org/dictionary/term/central-irb-single-irb-sirb" />
Schema.org DefinedTerm (JSON-LD)
json
{
  "@context": "https://schema.org",
  "@type": "DefinedTerm",
  "@id": "https://casrai.org/dictionary/term/central-irb-single-irb-sirb",
  "name": "Central IRB / Single IRB (sIRB)",
  "identifier": "https://casrai.org/dictionary/term/central-irb-single-irb-sirb",
  "description": "The model in which one IRB of record -- rather than each participating site's local IRB -- conducts the ethical review for a multi-site human-subjects study under a single approved protocol, with all other participating sites relying on that review through a documented reliance arrangement (typically an IRB Authorization Agreement) instead of each convening a duplicate local review.",
  "inDefinedTermSet": "https://casrai.org/dictionary/domain/clinical-research#set",
  "url": "https://casrai.org/dictionary/term/central-irb-single-irb-sirb",
  "sameAs": [],
  "license": "https://creativecommons.org/licenses/by/4.0/",
  "publisher": {
    "@id": "https://casrai.org/#organization"
  },
  "dateModified": "2026-07-18T02:22:10",
  "inLanguage": "en"
}

Referenced across the research world

University of Cambridge logoColumbia University logoCrossref logoUniversity of Edinburgh logoHarvard University logoUniversity of Oxford logoPrinceton University logoStanford School of Medicine logoUniversity College London logoORCID logoUniversity of Cambridge logoColumbia University logoCrossref logoUniversity of Edinburgh logoHarvard University logoUniversity of Oxford logoPrinceton University logoStanford School of Medicine logoUniversity College London logoORCID logo
  • University of Cambridge logo
  • Columbia University logo
  • Crossref logo
  • University of Edinburgh logo
  • Harvard University logo
  • University of Oxford logo
  • Princeton University logo
  • Stanford School of Medicine logo
  • University College London logo
  • ORCID logo

View CASRAI adoption →