Report a Concern Involving Human Participant Research

UPDATED February 1, 2023

NIH Data Management and Sharing Policy Update: Learn more here.

TRAINING UPDATE: Effective October 1, 2022, all Cornell study personnel involved in Exempt research protocols are now required to complete CITI training in human participant research ethics. See details here. For more information about Cornell IRB training requirements, visit the IRB Training webpage.

This page contains information for researchers and research participants on how and when to report something unexpected in connection with a Cornell research project involving human participants.

For Researchers - how and when to report a protocol deviation or unexpected event

Reporting an Unanticipated Problem/Adverse Event

Principal Investigators (PIs) must promptly report to the IRB any injuries or other unanticipated problems involving risks to human participants or others that occur in connection with a Cornell research project. An unanticipated problem involving risk to human participants or others is one that: (1) was unforeseen at the time of its occurrence, and (2) indicates that participants or others are at an increased risk of harm. One type of unanticipated problem is an "adverse event," referring to an instance where a participant is harmed in the course of participating in the research.

The IRB staff encourages researchers to report such problems as soon as possible. As soon as possible the PI should fill out an Adverse Event report on their RASS-IRB protocol. When in doubt about whether and what information to report to the IRB, please discuss the issue with a member of the IRB staff.

Reporting a Protocol Deviation

A protocol deviation (also referred to as protocol "noncompliance") occurs when research involving human participants is conducted in a manner that disregards or violates federal regulations, the policies or procedures of the Cornell IRB, or a research protocol as it was approved or exempted by the Cornell IRB or IRB staff. Even in the absence of any ill intent, an unapproved or otherwise noncompliant research activity may place a research participant at unnecessary risk.

The IRB may become aware of information indicating noncompliance from a participant, researcher, Cornell employee, or member of the public; from the interpretation of information received during the protocol submission and review process; from the findings of a random or for-cause audit or other quality control activities; or from the PI or member of the research team directly. Depending on the facts and circumstances, the IRB may require the PI to report the Protocol Deviation in RASS-IRB.

See SOPs 4 and 5 for detailed information on Cornell IRB policy and procedure related to protocol deviation and unanticipated problems

For Research Participants - Report a study concern or ask about your rights

The foremost concern of the Cornell IRB is ensuring the safety, rights and welfare of all persons participating in research projects being undertaken by Cornell faculty, staff and students, whether conducted online, in person at Cornell, or at another location or institution.

To resolve a minor issue with a study, it is often best to communicate directly with the research team. You may get in touch with the Principal Investigator for the study using the contact information given during the informed consent process.

If your concern is more grave, or if you prefer to have a third party work with you to investigate and resolve your complaint, contact the IRB office. Please include enough information so that the IRB staff can conduct an investigation and take appropriate action. The title of the study in question and name of the researcher with whom you worked are key information that the IRB will need from you.

By phone:

Contact Myles Gideon, IRB Manager, Office of Research Integrity and Assurance

By email:

Write to the IRB at irbhp@cornell.edu

Anonymously:

Use the Cornell hotline (1-866-293-3077) or online site, both of which are administered by an independent company called EthicsPoint, which protects your anonymity.