Dr Stevens-Simon raises some excellent points. Her first argument is that the consent process may assist children's cognitive development and may even have some therapeutic benefit; therefore, allowing children to opt out of the assent process may be inappropriate. This line of reasoning raises 2 issues. First, in pediatric practice we frequently advise parents on preferred methods of child rearing, and we often persuade parents when we believe they are making choices that are not in their child's best interest. For example, pediatricians often counsel parents to vaccinate their children when parents initially prefer to forgo immunizations on the basis of fear or misinformation. In the case of immunizations, we generally feel confident that our recommendations are based on good science and, therefore, believe we are justified in persuading parents accordingly.In the case of assent, however, we are on more shaky ground. Although many of us agree with the position of Dr Stevens-Simon (ie, that encouraging children to make their own choices will assist them in developing the ability to make well-reasoned choices as adults), there is little scientific evidence for this belief. Therefore, if parents choose to shield their child from certain conversations and decision-making, there is less support for us to convince them otherwise. Certainly, without solid evidence that participation in a decision-making process will benefit the child, it is not appropriate for the pediatrician to overrule the preferences of the family and require the child's assent, particularly when the child prefers to have the parents speak for him or her.Second, it is questionable whether a researcher should provide anticipatory guidance in the context of research. Clearly, such guidance is a necessary component of physician-child and physician-parent interactions; however, whether such conversations are appropriate in the context of research is less clear. Certainly, researchers have significant obligations to their research subjects, and if providing specific information or an intervention is clearly in the subject's best interest, then researchers have an obligation to act. However, particularly when the benefits to the child are more intangible and when we lack data demonstrating clear benefit, it may be inappropriate for researchers to convince families to act in the way that the researcher believes is best. Dr Stevens-Simon writes that allowing a child to relinquish his or her right to assent conflicts with the basic precepts of preventative pediatrics; however, it is not the role of the researcher to provide preventative care. Whereas in clinical care one may support a pediatrician persuading families to include the child in the decision-making process, there are substantial ethical problems with a researcher requiring the child's assent when the child prefers to abstain. When clinicians act as researchers and enroll their own patients in research projects, this distinction can become muddled. Clinicians who are researchers ought to ensure that they do not blur the line between pediatrician and researcher. Otherwise, they may find themselves performing both functions at the same time and run the risk of encouraging a therapeutic misconception in both families and themselves.