IRB Review and Research Oversight

Human subjects research in genetics is governed by a specific regulatory structure, and genetic counselors encounter it from both sides: enrolling patients in studies, and having their own clinical questions turn into research. The practical skill is recognizing when an activity has crossed from clinical care or quality improvement into research requiring oversight, because that boundary is where people get into trouble.

• The Common Rule (45 CFR 46) is the federal regulation governing federally supported human subjects research. Research is defined as a systematic investigation designed to develop or contribute to generalizable knowledge, and a human subject is a living individual about whom an investigator obtains data through intervention or interaction, or obtains identifiable private information.
• The generalizable knowledge test is the crux. Reviewing your own clinic's outcomes to improve your own practice is quality improvement. Doing the same review intending to publish it as findings others should apply is research. Intent at the outset determines this, and deciding to publish afterward does not retroactively make oversight optional.
• Three review pathways: exempt (minimal risk categories, still requires a determination by the IRB rather than by the investigator), expedited (minimal risk, reviewed by the chair or a designated member rather than the full board), and full board (greater than minimal risk, or vulnerable populations).
• The IRB decides, not the investigator. An investigator may not self-certify that their own project is exempt. This is one of the most common procedural errors.
• Single IRB review is now required for most federally funded multi-site studies, so a site relies on a central IRB rather than each site's own board.
• The Belmont Report supplies the ethical foundation: respect for persons, beneficence, and justice, which map onto informed consent, risk-benefit assessment, and equitable subject selection.
• Elements of research informed consent differ from clinical consent. Consent must state that the activity is research, describe its purpose and duration, disclose foreseeable risks and any benefits, describe alternatives, explain confidentiality protections, and state clearly that participation is voluntary and may be withdrawn without penalty. Since the 2018 Common Rule revisions, consent must open with a concise summary of the key information a reasonable person would need to decide.
• Therapeutic misconception is the specific and pervasive risk in this setting: the participant believes a study is designed to benefit them personally when it is designed to answer a question. It is heightened in genetics, where families with no treatment option often understand enrollment as care. Naming it directly is part of honest consent.
• Certificates of Confidentiality protect identifiable research data from compelled disclosure and are automatically issued for NIH-funded research.
• Vulnerable populations receive additional protection: children, prisoners, pregnant women and fetuses, and people with impaired decisional capacity. Pediatric research requires parental permission plus the child's assent where developmentally appropriate.

• A counselor notices an unusual phenotype pattern across six patients and plans a case series for publication. Because the intent from the outset is generalizable knowledge, this is research and requires IRB review or a formal exempt determination before data collection, not after the manuscript is drafted.
• The same counselor reviews clinic wait times to shorten them locally, with no plan to publish. This is quality improvement and does not require IRB review. If the results later look publishable, the correct step is to consult the IRB rather than assume.
• A family with an undiagnosed condition is offered enrollment in a research sequencing study. They say, "So this will finally get us the answer." The counselor addresses the therapeutic misconception explicitly: the study may return a result, may return nothing, and is designed to advance knowledge rather than to provide their clinical care, which continues regardless.
• A 15-year-old is enrolled in a natural history study. Parental permission is obtained and the adolescent's assent is documented separately. When she turns 18, the study re-consents her as an adult in her own right.

• Do not decide for yourself that a project is exempt. Exempt is an IRB determination, not an investigator's judgment.
• Do not assume "we were going to do it anyway" removes the research question. The distinction is the intent to produce generalizable knowledge, not whether the activity would have happened.
• Do not let a research consent read like a clinical consent. Voluntariness, the right to withdraw, and the absence of promised personal benefit are the elements that distinguish it, and they are the ones most often blurred.
• Do not ignore therapeutic misconception because the family seems well informed. Educated and highly motivated families are, if anything, more susceptible, because they most want the study to be the answer.
• Do not confuse a Certificate of Confidentiality with absolute protection. It blocks compelled disclosure, but it does not prevent voluntary disclosure required for reporting obligations such as suspected abuse.