45 CFR 46 Subpart B is the federal rule that adds extra protections for pregnant women, fetuses, and newborns whenever they are involved in research that the Department of Health and Human Services conducts or funds. It sits on top of the Common Rule (Subpart A) and tightens the consent requirements, caps the allowable risk to the fetus or neonate, and separates research decisions from clinical care. Different rules apply depending on who benefits from the study and, after delivery, whether the newborn is viable, nonviable, or of uncertain viability.
When Subpart B Applies
The subpart covers all research involving pregnant women, fetuses, neonates of uncertain viability, or nonviable neonates when the work is conducted or supported by HHS.1U.S. Department of Health and Human Services. 45 CFR 46 Subpart B – Additional Protections for Pregnant Women, Human Fetuses and Neonates Involved in Research That includes research done inside HHS facilities by anyone and research done anywhere by HHS employees, at any performance site, domestic or international.
Any institution engaged in nonexempt human subjects research with HHS funding must hold a Federalwide Assurance filed with the Office for Human Research Protections. The FWA commits the institution to comply with Subparts A through E of 45 CFR 46, including everything in Subpart B.2U.S. Department of Health and Human Services. Terms of the Federalwide Assurance for the Protection of Human Subjects
A private institution running research without HHS funding is not bound by Subpart B unless it has voluntarily extended its FWA to cover all of its human subjects research. Studies that qualify for one of the eight exemption categories in 45 CFR 46.104(d), such as certain educational or secondary-data research, also fall outside Subpart B’s substantive requirements as long as the exemption conditions are met.3eCFR. 45 CFR 46.104 – Exempt Research
Definitions That Decide Which Rules Apply
Subpart B uses precise terminology in 45 CFR 46.202, and those definitions control every downstream rule.4eCFR. 45 CFR 46.202 – Definitions
- Fetus. The product of conception from implantation until delivery.
- Dead fetus. A fetus showing no heartbeat, spontaneous breathing, voluntary muscle movement, or pulsation of the umbilical cord.
- Pregnancy. The period from implantation to delivery. A woman is presumed pregnant if she shows common signs such as missed menstrual periods, until a pregnancy test comes back negative or delivery occurs.
- Delivery. The complete separation of the fetus from the woman.
- Neonate. A newborn.
- Viable neonate. A newborn able to survive, with available medical treatment, to the point of independently maintaining a heartbeat and breathing.
- Nonviable neonate. A newborn that is alive after delivery but cannot survive.
A fetus is governed by the rules for research on pregnant women. Once complete separation occurs, the newborn becomes viable, nonviable, or of uncertain viability, and each carries different consent and risk rules.
Subpart B does not define “minimal risk” separately. It relies on the Subpart A definition: the probability and magnitude of harm anticipated are no greater than what a person would ordinarily encounter in daily life or during routine physical or psychological exams.5eCFR. 45 CFR 46.102 – Definitions for Purposes of This Policy This threshold controls any study that offers no direct benefit to the fetus, because in that case fetal risk cannot exceed minimal.
Rules for Research on Pregnant Women and Fetuses
A study may include pregnant women or fetuses only if it satisfies all ten conditions in 45 CFR 46.204.6eCFR. 45 CFR 46.204 – Research Involving Pregnant Women or Fetuses The IRB must document that each condition is met before approving the research.
Preliminary Data and Risk Limits
Where scientifically appropriate, researchers must first conduct preclinical studies (including on pregnant animals) and clinical studies on nonpregnant women to generate data on potential risks before enrolling pregnant participants. The qualifier means it is not a blanket rule, but the IRB must be satisfied that available data adequately addresses fetal and maternal risk.
Fetal risk runs on two tracks. If the procedures offer a prospect of direct benefit to the woman or the fetus, they are permissible so long as the risk is the least possible for achieving the study’s objectives. If there is no direct benefit, fetal risk cannot exceed minimal, and the study must aim to develop important biomedical knowledge that cannot be obtained any other way.
Who Must Consent
Consent requirements depend on who stands to benefit:
- If the research offers direct benefit to the woman, to both the woman and the fetus, or no direct benefit but only minimal fetal risk with important knowledge at stake, only the woman’s informed consent is required.
- If the direct benefit is solely to the fetus, both the woman and the father must give informed consent.
The father’s consent is waived when he cannot be located, lacks capacity to consent, is temporarily incapacitated, or the pregnancy resulted from rape or incest. The rule does not spell out what evidence is required to establish unavailability, and IRBs evaluate this case by case.
When the pregnant participant is a minor, Subpart D (the pediatric research rules) applies simultaneously, and its consent and assent requirements must also be satisfied. Every person providing consent must be fully informed about the reasonably foreseeable impact of the research on the fetus or neonate.
Keeping Research Separate From Clinical Care
Three bright-line prohibitions in 46.204 keep research interests from influencing medical care. No financial or other inducements may be offered to terminate a pregnancy. Researchers may not have any part in decisions about the timing, method, or procedures used to end a pregnancy. And researchers may not take part in determining whether a neonate is viable. That last prohibition matters because the viability determination controls which regulatory track governs the newborn.
Rules for Research on Neonates
After delivery, the newborn falls into one of three categories, and the rules differ meaningfully for each. All neonatal research under Subpart B shares a few baseline requirements: preclinical and clinical data must be gathered first when scientifically appropriate, every consenting individual must be told about the foreseeable impact on the newborn, and researchers cannot participate in viability determinations.7eCFR. 45 CFR 46.205 – Research Involving Neonates
Neonates of Uncertain Viability
Until viability is established, research is allowed only if the study offers a prospect of improving the newborn’s chance of survival with the least possible risk, or the study aims to develop important biomedical knowledge unobtainable by other means and adds no risk to the newborn.
Consent from either parent is sufficient. If neither parent can consent because of unavailability, incompetence, or temporary incapacity, a legally authorized representative of either parent may consent instead. The father’s consent is not required when the pregnancy resulted from rape or incest.
Nonviable Neonates
The protections here are the strictest in Subpart B. Researchers may not artificially maintain the newborn’s vital functions, and the study may not terminate the newborn’s heartbeat or breathing. The research must aim to produce important biomedical knowledge unobtainable any other way, and it may add no risk to the newborn.
Both parents must give informed consent. If one parent is unavailable, incompetent, or temporarily incapacitated, the other parent alone may consent, and the father’s consent is not required if the pregnancy resulted from rape or incest. A legally authorized representative cannot stand in for either parent in this category.
Viable Neonates
Once a newborn is determined to be viable, Subpart B’s neonatal provisions no longer apply. The newborn is treated as a child under the Common Rule and Subpart D, which has its own risk categories, parental permission requirements, and assent rules.
Placenta, Dead Fetus, and Fetal Material After Delivery
Research using the placenta, a dead fetus, fetal tissue, or cells and organs removed from a dead fetus is governed by 45 CFR 46.206, and the treatment is significantly lighter. These studies must comply with applicable federal, state, and local laws, but Subpart B imposes no independent consent or risk requirements on them.8eCFR. 45 CFR 46.206 – Research Involving, After Delivery, the Placenta, the Dead Fetus or Fetal Material
One exception matters. If the researcher records information linked to the material in a way that could identify a living person (the mother or anyone else), that person becomes a research subject, and all of the standard human-subjects protections under 45 CFR Part 46, including informed consent and IRB review, apply. Stripping identifiers before recording data avoids triggering this provision.
The Secretary’s Narrow Approval Pathway
Section 46.207 creates a route for research that does not fit within 46.204 or 46.205 but could shed light on a serious health problem affecting pregnant women, fetuses, or neonates.9eCFR. 45 CFR 46.207 – Research Not Otherwise Approvable Which Presents an Opportunity to Understand, Prevent, or Alleviate a Serious Problem Affecting the Health or Welfare of Pregnant Women, Fetuses, or Neonates The procedural hurdles are steep.
The IRB must first find that the research offers a reasonable opportunity to understand, prevent, or alleviate a serious problem in this population despite not satisfying the standard conditions. The Secretary must then convene a panel of experts in fields such as science, medicine, ethics, and law, and there must be a public comment period that includes a public meeting announced in the Federal Register. After that process, the Secretary may approve the research only if it either satisfies the standard conditions after all, or meets three criteria: it presents a reasonable opportunity to address a serious health problem, it will be conducted according to sound ethical principles, and informed consent will be obtained under the applicable Common Rule provisions. The mechanism is a safety valve, used sparingly.
What Noncompliance Costs
The Office for Human Research Protections monitors compliance with all of 45 CFR 46 and has a range of enforcement tools when it finds violations.10U.S. Department of Health and Human Services. OHRP Compliance Oversight Assessments OHRP can attach conditions to or limit an institution’s FWA, halting some or all HHS-supported human subjects research at that institution. Studies covered by a restriction may be suspended, and already-enrolled participants may continue only if the investigator, IRB, and institutional official determine that continuation is in their best interests. OHRP can require new monitoring mechanisms as a condition of resuming research, and it can recommend that an institution or individual investigator be temporarily suspended, permanently removed from specific projects, or debarred from federal funding entirely.
Because these actions can reach the institution’s entire portfolio of federally funded human subjects research, a single noncompliant study involving pregnant women or neonates can put an entire research program at risk.