What Extra Protections Are Needed When Research Involves Children?

What changes ethically when the research participant is a child?

Children are not simply smaller adult research participants. Their ability to understand research develops over time, adults ordinarily hold legal responsibility for many decisions on their behalf, and the acceptable balance between research risk and benefit may be governed by additional protections.

Yet “children need parental consent” is an incomplete summary. Children who cannot provide legally effective informed consent may still be capable of understanding what will happen and expressing whether they want to participate. Different levels of research risk can trigger different requirements. In some circumstances, parental permission can even be waived under the governing rules.

Ethical research with children therefore involves several connected questions: Who legally authorizes participation? What role should the child have? How much risk is permissible? Does the research offer direct benefit? And is including children scientifically necessary rather than merely convenient?

Children generally need protections that combine adult permission with respect for the child's developing autonomy

Protecting children means considering permission, assent, risk, and developing autonomy together

“Child” is partly a legal definition, not simply another word for anyone under 18

Researchers should first determine how the governing framework defines a child.

Under U.S. HHS regulations, children are persons who have not attained the legal age for consent to the treatments or procedures involved in the research under the applicable law of the jurisdiction where the research occurs. This means the regulatory answer is not always produced simply by checking whether someone is younger than 18.

For certain treatments or procedures, applicable law may allow some minors to consent independently. OHRP explains that if a research activity involves solely treatments or procedures for which the minor has reached the legal age of consent under applicable law, that person may not meet the Subpart D definition of a child for that research.

Parental permission and child assent are not the same thing

Under HHS Subpart D, parental permission and child assent replace the ordinary adult informed-consent model for children within the regulation's scope. “Permission” means the agreement of a parent or guardian to the child's participation, while “assent” means the child's affirmative agreement to participate.

The distinction preserves an important ethical idea: adults may hold legal authority to permit participation, but a child capable of understanding and expressing a preference should not necessarily be treated as a passive object of that decision.

Assent should reflect what the child can actually understand

HHS regulations require the IRB to consider children's ages, maturity, and psychological state when determining whether they are capable of assent. That determination can apply to a group of children in a protocol or be made individually.

OHRP further explains that the assent process should reflect the child's capabilities. Adolescents whose understanding approaches that of adults may receive information resembling an adult consent discussion. Younger children who cannot grasp the full research framework may still be able to understand what will happen, how long it will take, and whether procedures may hurt or feel uncomfortable.

An assent form written at a lower reading level is therefore not the entire solution. The process should help the particular child understand participation to the degree they are capable.

Silence or compliance is not assent

This distinction is especially important when children are accustomed to following adult instructions.

Under HHS regulations, assent requires affirmative agreement. Mere failure to object should not be interpreted as assent.

A child who quietly follows a parent into a research room has not necessarily agreed to research. Neither has a student who completes a questionnaire because everyone else in class was handed one.

Researchers should therefore consider how willingness will actually be expressed and how children will be given a meaningful opportunity to decline when assent is required.

A capable child's dissent can matter even when a parent agrees

Under HHS Subpart D, if the IRB determines that a child is capable of assent and requires assent to be obtained, the child generally cannot participate without providing it. OHRP explains that in such circumstances a child's dissent prevails even when a parent or guardian has given permission.

There are exceptions. The regulations allow the IRB to determine that assent is not a necessary condition in certain circumstances, including when the child's capability is so limited that the child cannot reasonably be consulted or when the research intervention offers an important prospect of direct benefit available only in the research context. Assent requirements can also be waived in specified circumstances.

This is why “the parent signed” is not a complete pediatric consent analysis.

Research risk determines which additional rules apply

Under HHS Subpart D, pediatric research is not governed by one universal risk threshold. The IRB must determine which regulatory category applies.

These categories are specific to research governed by the cited U.S. HHS regulations. Other jurisdictions and regulatory systems may use different structures.

Direct benefit changes the ethical analysis but does not erase risk limits

Under §46.405, more-than-minimal-risk research can be permissible when an intervention or procedure offers a prospect of direct benefit to the individual child, but the risk must be justified by the anticipated benefit and the relationship of anticipated benefit to risk must be at least as favorable as available alternatives.

This is different from saying that any possibility of benefit justifies any level of research risk. The regulatory conditions remain substantive.

Researchers should also distinguish direct benefit to the participant from the broader social value of producing knowledge. Learning something useful for future children is important, but it is not the same thing as a prospect of direct benefit to the child currently being enrolled.

Research without direct benefit faces tighter limits

Under §46.406, HHS-regulated research involving more than minimal risk and no prospect of direct benefit can be approved only under specified conditions. Among them, the risk must represent a minor increase over minimal risk, the experiences must be reasonably commensurate with those inherent in the children's actual or expected medical, dental, psychological, social, or educational situations, and the research must be likely to yield generalizable knowledge of vital importance about the children's disorder or condition.

Research outside these categories is not automatically impossible, but §46.407 establishes a substantially different review pathway for certain research presenting a reasonable opportunity to understand, prevent, or alleviate a serious problem affecting children's health or welfare.

One parent or two parents may be required depending on the research category

Under HHS Subpart D, an IRB may find permission from one parent sufficient for research under §46.404 or §46.405. For research under §46.406 or §46.407, permission from both parents is generally required when parental permission is required, subject to specified exceptions such as when one parent is deceased, unknown, incompetent, not reasonably available, or only one parent has legal responsibility for the child's care and custody.

This is another reason researchers should avoid writing “parental consent will be obtained” as though it were a single procedure independent of risk classification.

Parental permission can sometimes be waived

Parental permission is not an absolute requirement in every study governed by HHS regulations.

Subpart D allows waiver under specified circumstances. It also provides a particular mechanism when the research concerns conditions or a participant population for which parental or guardian permission is not a reasonable requirement to protect the children, with neglected or abused children given as an example. An appropriate substitute protection must be provided, and the waiver must be consistent with applicable federal, state, or local law.

This matters because requiring parental involvement can sometimes create rather than reduce risk. The ethical and legal analysis, however, is highly context-dependent and belongs with the IRB and applicable law, not unilateral investigator preference.

Children can also be vulnerable to authority outside the family

Parental permission does not address every possible pressure on a child.

School-based research, for example, may involve teachers who control grades or classroom relationships. Children may assume that an activity presented during class is mandatory simply because adults distribute it alongside ordinary schoolwork.

Researchers should therefore consider dependency on teachers, clinicians, institutions, and other authority figures as well as parental involvement.

Growing older during a study can change the consent arrangement

Longitudinal research can create a particularly interesting consent problem: a child may reach the legal age for consent while still enrolled.

OHRP treats informed consent as an ongoing process and explains that, when a child enrolled with parental permission subsequently reaches the legal age of consent, the participant's own legally effective informed consent generally must be obtained for continued interactions or interventions unless the IRB has approved an applicable waiver.

In other words, yesterday's parental permission does not necessarily remain the appropriate authorization once the participant becomes legally able to decide independently.

A school survey shows why parental permission is only one part of child protection

A signed parental form alone would not answer whether the child's own participation is ethically respectful.

Children's research protections are more nuanced than “get the parent's signature”

Design the protection around the child's legal status, capabilities, and the study's risk

When preparing research involving children, avoid treating pediatric protection as an extra signature added to an adult protocol. The research design itself should account for children's developmental and legal circumstances.

Children also differ substantially from one another. Age, maturity, health, dependency, family circumstances, and the study setting can change what protection means. The broader principle remains the same as for research involving participants who may require additional safeguards: identify the actual vulnerability and respond proportionately rather than relying on the label alone.

Before enrolling children, verify more than parental permission

Frequently asked questions about children in research

Protecting children means giving adults responsibility without making children voiceless

How to Cite This Guide

This guide is intended to be read, shared, and used in research, teaching, and academic work. If you draw on its ideas, explanations, or other content, please acknowledge the source by citing the guide. Doing so gives appropriate credit and helps your readers locate the original resource.