From the time when children enter the preteen years onward, pediatric medical decision-making can entail a complex interaction between child, parents, and pediatrician. When the child and parents disagree regarding medical decisions, the pediatrician has the challenging task of guiding the family to a final decision. Unresolved discord can affect family cohesiveness, patient adherence, and patient self-management. In this article, we outline 3 models for the pediatrician’s role in the setting of decisional discord: deference, advocative, and arbitrative. In the deference model, the pediatrician prioritizes parental decision-making authority. In the advocative model, the pediatrician advocates for the child’s preference in decision-making so long as the child’s decision is medically reasonable. In the arbitrative model, the pediatrician works to resolve the conflict in a balanced fashion. Although each model has advantages and disadvantages, the arbitrative model should serve as the initial model in nearly all settings. The arbitrative model is likely to reach the most beneficial decision in a manner that maintains family cohesiveness by respecting the authority of parents and the developing autonomy of children. We also highlight, however, occasions when the deference or advocative models may be more appropriate. Physicians should keep all 3 models available in their professional toolkit and develop the wisdom to deploy the right model for each particular clinical situation.
- ICC —
- informed consent conversation
“Sarah” is a mature 11-year-old girl admitted to the hospital with an ulcerative colitis flare. Despite intensive outpatient medical management including biologic therapies and immunosuppression, she continues to experience frequent flares. Each flare requires hospitalization and a course of steroids. These hospitalizations led her to miss more than 30 days of school last year. Her gastroenterologist is concerned about her disease trajectory, and he maps out 2 treatment options. The first is to continue with current therapies in hopes of an improvement in the trajectory of the disease course. The second option would be referral to a surgeon for total colectomy. Although this intervention would eliminate any symptoms directly related to ulcerative colitis, she would then have to manage possible long-term complications of the surgery, including frequent loose stools, the need for additional surgeries, and a temporary ostomy. After hearing the options, her parents favor the colectomy. Sarah disagrees and tells the physician, “I don’t want to have surgery.”
Decision-making in pediatrics is a complex interaction involving the pediatrician, the child, and the parents. Most pediatricians adhere to the paradigm of shared decision-making, in which decisional authority and decisional burden are distributed in varying degrees among the child, parents, and pediatrician. The particular nuances of pediatric decision-making, however, pose several challenges to this paradigm.1 Whereas the ethical basis of adult decision-making is derived from the doctrine of informed consent, many children do not have full capacity to provide consent.
The elements of informed consent are2:
Disclosure of information
Understanding of material information
Consent and authorization of a plan
Pediatric decision-making is therefore better understood as a combination of parental permission and child assent.3 Assent does not have a broadly agreed on definition, but generally “child assent refers to affirmative agreement of a minor who is to take part in the informed consent procedure in a way adapted to his or her capabilities, while their legal representative has the formal role of consenting”.4
The American Academy of Pediatrics elements of assent are3:
Helping the patient achieve a developmentally appropriate awareness of the nature of his or her condition.
Telling the patient what he or she can expect with tests and treatment(s).
Making a clinical assessment of the patient’s understanding of the situation and the factors influencing how he or she is responding (including whether there is inappropriate pressure to accept testing or therapy).
Soliciting an expression of the patient’s willingness to accept the proposed care. Regarding this final point, [the committee members] note that no one should solicit a patient’s views without intending to weigh them seriously. In situations in which the patient will have to receive medical care despite his or her objection, the patient should be told that fact and should not be deceived.
This model of permission and assent works well when parents and the child agree, but persistent disagreement between the parent and child can place the pediatrician in a difficult position. Although physicians can support the resolution of many disagreements by helping parents and children to articulate their values and goals, some disagreements will persist.
This decisional discord can go in either direction relative to a given treatment option: the parents may want to pursue a specific treatment and the child patient does not, refusing to provide assent or actively dissenting; or vice versa, the child may want to pursue a treatment option but the parents object. In the setting of either form of discord, the pediatrician should have 2 goals: to help the patient and family reach a decision that is in the best interests of the child, and to reach a decision that parents and the child can agree on and can collaboratively enact. Persistent decisional discord can have a negative impact on family cohesiveness, patient adherence, and patient self-management.5 In this article, we describe 3 models for facilitating decision-making in the setting of persistent decisional discord, highlighting circumstances in which each model may be appropriate. Each model is nested within the paradigm of shared decision-making, describing the degree to which decisional authority is distributed between parents and children.
Three Models of Physician Role in Child-Parent Disagreement
When a pediatrician encounters decisional discord between a child and parents, there are 3 basic models by which the pediatrician can respond: the deference model, the advocative model, and the arbitrative model (Fig 1). These models apply to disagreements along the full spectrum of medical decisions, including the performance of routine procedures (such as injectable immunization), life-changing interventions (such as treatments to cause pubertal delay in a transsexual child), and end-of-life decisions (such as whether to perform cardiopulmonary resuscitation). In this article, we will explore these models by using the initial case of medical versus surgical management of Sarah’s ulcerative colitis, which represents a decision between 2 clinically reasonable options that will nonetheless have significant life-altering consequences. To examine these models more thoroughly, we will explore how key changes to the scenario highlight the value of each model.
In the deference model, the pediatrician prioritizes parental decision-making authority when there is discord. The deference model is supported by the concept of parental authority, which rests on several justifications: (1) Parents are responsible for raising their children, and this responsibility necessitates that parents are allowed a reasonable amount of decision-making prerogative; (2) Parents have to live with the consequences of their decisions and provide support for their children; (3) Parents best know their children and are presumed to make decisions in accordance with their children’s best interests.6,7 When children are young, the deference model is dominant, because the objections of young children to receiving medical treatments are routinely overridden. Some adherents of the deference model also support its application in the care of adolescents, citing data on adolescent impulsiveness, which has been attributed to a “heightened responsiveness to incentives while impulse control is still relatively immature during this period.”8 (Indeed, in Roper v Simmons the Supreme Court specifically identified adolescent impetuousness and susceptibility to outside pressure when making decisions as reasons to abolish the juvenile death penalty.9) Some have argued that adolescents do not have the ability to make informed, forward-thinking decisions, and therefore are incapable of making the best decisions.10,11 From this point of view, increased parental authority is necessary to protect the child from poor decisions and to teach the adolescent to become a better decision-maker in the future.10,12
The deference model, however, has many opponents. The basis for this opposition arises from a respect for the child’s developing autonomy and ability to provide assent.13 Although mandated in clinical research,14–18 assent in medical decision-making is encouraged but not required.19 Many argue, however, that assent remains an appropriate goal for all treatment decisions,3 although honoring the child’s ability to assent leads to the challenging possibility of dissent. Additionally, others argue that if parents make decisions on behalf of a child that have long-lasting effects, they may be limiting the child’s “right to an open future,”20 a right that “protects the child against having important life choices determined by others before she has the ability to make them for herself.”21
The deference model may nevertheless be appropriate when the child is unable to demonstrate capacity to engage in the decision-making process. In our initial case, if Sarah had mild cognitive impairments due to an underlying condition, deferring to parental wishes may be more appropriate. Similarly, the deference model may be warranted in the setting of high acuity illness when the decision is time-sensitive, or when the child’s decision is not medically appropriate. For example, if Sarah were admitted with acute appendicitis and peritonitis, the physician should support the parents’ preferences for surgery despite the child’s dissent.
A pediatrician employing the deference model would initially propose diagnostic or treatment options to both the parent and the child and elicit their preferences. When discord becomes apparent, the pediatrician would work with the parents to explain to the child the necessity of the intervention. For our patient with ulcerative colitis, the pediatrician might start by asking the patient, “What worries you about surgery?” Next, the pediatrician might say, “I understand why you are worried. This type of surgery can be scary, but your parents and I feel strongly that this is best for you. We think surgery will help you to live a healthier life and help to keep you out of the hospital. You missed so much school last year and have been away from your friends for so long. There are some downsides to surgery, but we really believe the downsides are outweighed by the long-term benefits.”
In the advocative model, the pediatrician advocates strongly for the older child or adolescent’s preference in decision-making so long as the patient’s decision is medically reasonable. This model is based on the child’s developing autonomy and the benefits of incorporating children in decision-making.
Involving children in decision-making is beneficial in the following ways (adapted from McCabe)5:
Demonstrates respect for children’s developing capacities, and may foster further development.
Supported by ethical principle of patient self-determination, especially as children further develop decision-making capacity.22
Improves open communication among physicians, children, and parents, and this communication is fundamental to both children’s and parents’ satisfaction with medical care.23 Satisfaction is associated with improved compliance.
Facilitates the child's cooperation with treatment.24
In support of this model, 1 author argued that “we need to support minors’ involvement in decision-making, particularly… where treatment preferences are based upon personal values and ‘quality of life’ issues.”5 Additionally, the Children’s Oncology Group guidelines state that oncologists should encourage and support the involvement of children in decision-making.14,27 Bartholome,28 who pioneered the concept of pediatric assent, went further to state that “children and adolescents should always be included in health care decision-making to the extent of their capacity and willingness to do so.”
Implicit in these arguments is the premise that many children are capable of making informed, meaningful decisions related to their health. Historically, research on decision-making capacity in children has varied widely in methodology and has often resulted in conflicting results with regard to age and previous experiences.19 Some studies have suggested, however, that children as young as 9 years can begin to make informed decisions, and may have developed adequate decision-making capacity by adolescence.29,30 Others have suggested that children reach adult capacity for abstract thought between 14 and 16 years of age.31 Putting the argument into more concrete terms, 1 expert has asserted that “our paradigm should become that adolescents are capable of consent and those wishing to prove otherwise have the burden of proof.”32
A more recent study used a standardized instrument to assess the capacity of children based on their ability to comprehend and make rational decisions.4 These authors concluded that children older than 12 years are generally competent to make rational decisions. They therefore recommended a dual-consent model in which children who demonstrate capacity must provide consent in addition to their parents before a physician can proceed with treatments or procedures. This dual-consent model is used in some European countries,4 but has no legal basis in the United States.
Opponents of the advocative model point out that having capacity to reason abstractly may not be sufficient to mandate the child’s preeminence in decision-making. Although the previously mentioned study found evidence of capacity in children as young as 12 years of age, they also acknowledged that “children may differ from adults by not having developed yet stable long term goals and values in life, meaning that children may procedurally be classified as competent although their decisions are based on values that might change.”4 If children are allowed to make decisions, they may make ill-advised choices that they will regret in the future.
Just as the deference model declines in use as patients grow older, the advocative model is increasingly appropriate as children age and demonstrate expanded capacity to engage in the decision-making process. If Sarah were 16 years old, for instance, physicians would likely be less keen to promote surgery against the patient’s wishes. Even at younger ages, the advocative model may be appropriate for decisions that have minimal consequences. For example, pediatricians would appropriately advocate for Sarah’s ability to dissent if she were refusing medication for seasonal allergies.
In practicing the advocative model, the pediatrician would similarly propose options and solicit insights from the child and parents. When discord is appreciated, the physician would support the child’s decision, assuming the child displays sufficient capacity to provide assent or dissent. The pediatrician might say to Sarah’s parents, “I understand your concerns, and I appreciate that you want what is best for your child. That is what any loving parent would want. But I cannot recommend pursuing surgery if Sarah is against it. This is her body and I truly believe we need to respect her decision. I recommend we continue with medical management for now. We can always see how things develop over time.” The pediatrician can also draw out the negative repercussions of overriding Sarah’s dissent: “Furthermore, Sarah is going to have to grow into the self-manager of her IBD for the rest of her life. If we force her to have the surgery, we might win a short-term struggle but really suffer a setback for our long-term goal of helping her take charge of her own health.”
The arbitrative model resides between the advocative and deference models. In this model, the pediatrician works to resolve the conflict in a balanced fashion, essentially working out a deal that the patient and the parents find acceptable. If an agreement cannot be reached, then the pediatrician would take a tiered approach by (1) further assessing the reasons for discord and facilitating communication between parents and child, (2) discussing the case with a trusted colleague for additional insights, and (3) considering an ethics consultation.
The arbitrative model is based on the principle that parents and children both share decisional authority and both have important insights that should impact the final decision. One way to implement the arbitrative model is by use of “shuttle diplomacy.”33 As described by Bluebond-Langner et al,33 the physician employing shuttle diplomacy first meets with the parents to ascertain what options they are amenable to discussing with the child. The physician subsequently meets with the child, either alone or with the parents, to further understand motives, values, and interest in participating. Through a collaborative process, a final decision is made. As noted by the authors, “physicians also must make clear in the meetings alone with a child and in meetings with the parents and child present that although the child will be listened to by the parents and physician, and the child’s wishes taken into account, the decision is not the child’s alone to make.”33 This diplomatic approach allows all parties to have their voices heard and provides an avenue to deeper understanding of the family unit. Furthermore, this approach protects the child from shouldering the full burden of decisional authority.
Opponents of the arbitrative model question the value of seeking assent if dissent is not an acceptable outcome.34 Some have further argued that not accepting dissent as a potential outcome of decision-making “makes a mockery of the whole idea of assent.”28 Without the ability to say “yes” or “no,” assent becomes “merely a formality without any true respect for the burgeoning autonomy of the minor patient.”34 In response to the challenge of dissent, the AAP Committee on Bioethics stated: “No one should solicit a patient’s views without intending to weigh them seriously. In situations in which the patient will have to receive medical care despite his or her objection, the patient should be told that fact and should not be deceived.”3 If a child’s opinion is sought but not heeded, the child may feel alienated and the relationship in the family may be further strained.
Some of the techniques used in ethics consultations may further the aims of the arbitrative model, including allowing time for adolescents to absorb information (such as information about surgical side effects, which may be initially experienced as threatening) and proposing time-limited trials (such as continuing medical management for another 6 weeks before revisiting the decision).35 Sometimes conditional commitments (“Would you agree to surgery if medical management does not produce improvements in the next 2 months?”) are easier to negotiate than immediate commitments.
We propose that the arbitrative model should serve as the initial approach to decisional discord in nearly all situations, although as highlighted above, physicians may need to deviate from this model at times. This model is especially appropriate when it is uncertain which decision will yield the most medical benefit, or if the potential intervention is invasive or has significant consequences. Furthermore, the arbitrative model is most useful when decisions are less time-sensitive. Our initial case represents such a scenario, because Sarah presumably has shown capacity to provide assent (or dissent) and the nonemergent surgery would be permanently life-altering. This model would not be appropriate, however, for situations in which children have a legal right to confidentiality from their parents. For example, many states have legislation that protects the minor’s privacy related to reproductive health and substance abuse, precluding the pediatrician from engaging the parents in discussion without the minor’s approval.36
The pediatrician applying the arbitrative model would first propose diagnostic or treatment options, then seek out the values and motives that drive the decision-making process for the child and parents. The pediatrician would then try to identify which aspects of the decision are negotiable versus nonnegotiable, saying to the parents, “You seem to be most worried about your child’s quality of life and long-term health. After talking with your daughter, she does not feel that her quality of life is greatly suffering right now, and she is frightened of the surgery. Perhaps we could plan to revisit the option of surgery later? In the meantime, we could arrange for her to meet some of our teenagers who have had similar surgeries. They can help her to see what life is like after a colectomy. She may just need more time to process all of this information.”
The Pediatrician and the Role of Experiential Expertise
Although these 3 models present different notions of how pediatricians should respond to decisional discord between child and parents, pediatricians have a role across all 3 models as expert that warrants consideration. Some pediatricians may strive to base all treatment recommendations solely on objective data, but this is often impracticable given the unfortunate lack of high-quality data to guide many medical decisions. In addition, we suggest that parents and children do not want a strictly objective recommendation. When seeking the expertise of a pediatrician, they are looking not only for technical expertise, but also the pediatrician’s expertise based on breadth of previous experience caring for patients and families in similar situations. In fact, some families will nudge physicians beyond purely objective answers by asking, “What would you do in my situation?” or, “What would you do if this were your child?” The pediatrician should certainly work to elicit the child’s and parents’ values and goals related to the decision, but this should be followed by a recommendation based on the physician’s expertise in combination with the family’s elicited values.
The recommendations for eliciting priorities and values in discussions with patients and families are:
Use open-ended questions, and allow the child or parent sufficient time to talk: “What do you really enjoy doing?” “How has this disease affected you and your family?” “Has your illness prevented you from being able to do things that you want to do?” “What are you hoping for?” “What is your biggest worry about the different treatment options?”
Reflect back the values demonstrated in their responses: “It sounds like you really miss playing on your basketball team.” “It sounds like you are really worried about your friends seeing you with an ostomy.”
Summarize the conversation: “So, what I am hearing is…”
At the conclusion of the discussion, provide a recommendation based on your knowledge and experience, and the values expressed by the child and parents. “In light of what we discussed, I would recommend that we…”
If the decision is not time-sensitive, allow the patient and parents time to process information. Then reengage and help to work toward a decision.
The physician’s combined technical and experiential expertise, therefore, serves as a starting point for decision-making in pediatrics. For parents and patients to make truly informed decisions, the pediatrician must be reflective and transparent about the process by which he or she develops recommendations. Every physician has personal values, preferences, and beliefs that impact the options presented, the manner in which they are presented, and the recommended course of treatment. Such transparency and reflection can help the physician to refocus on the values, goals, and needs of the patient and family. Without this effort, pediatricians run the risk of subverting parental authority and patient autonomy. For example, in 1 study of pediatric gastroenterology and rheumatology patients and parents considering treatment with biologics, providers offered detailed information about their preferred option, but provided less information about other options. “There was minimal elicitation of preferences, treatment goals, or prior knowledge.”37 In one-third of cases, no explicit decision was made, yet the provider took steps to implement his or her preferred treatment. In another recent study, researchers observed physicians during the informed consent conversation (ICC) for enrollment in pediatric Phase I oncology trials with parents and children. In this study, “physicians failed to mention no treatment and/or palliative care as options in 68% of ICCs and that the disease was incurable in 85% of ICCs.”38 Reflective self-management by the pediatrician is essential to creating an environment that supports truly shared decision-making.
Clinical Use of Models
When faced with decisional discord, pediatricians must decide how to respond in a way that leads to the best decision for an individual patient and family. Each of the decision-making models described above has different limitations and raises different concerns. Critics of the deference model argue that it fails to fully support or respect the developing autonomy of the child. Opponents of the advocative model posit that many children are incapable of fully appreciating long-term effects of their decisions. Those who are concerned about the arbitrative model question the validity of asking for assent if the pediatrician is unwilling or unable to honor the child’s dissent. Similarly, some state laws preclude the pediatrician’s ability to engage parents on certain topics without the child’s approval. Reviewing these counterarguments shows the challenging situation that pediatricians face when parent and child disagree. Importantly, several other factors can affect this decision-making process. For example, parental discord and chronic family stress are other important barriers to consider. Similarly, culture, religion, mental health issues, and financial constraints also may affect the role of children in the decision-making process. The impact of these factors should also be taken into account by pediatricians as they navigate decisional discord.
Each decision-making model has value depending on the clinical situation. We recommend that physicians initially employ the arbitrative model when confronted with decisional discord because this model is most likely to lead to a decision that respects the child’s developing autonomy while also supporting the needs of the whole family. Additionally, this model further reinforces the pediatrician’s mandate to more fully understand the motives and beliefs of the child and parent. Even if compromise remains elusive, the child will know that his or her voice was heard and valued by the pediatrician. When the arbitrative model is either inappropriate or has failed, we recommend that physicians reassess the child’s ability to engage in decision-making, the degree to which the treatment options have equivalent benefit-to-risk profiles, and the invasiveness or permanence of the proposed interventions to determine if the advocative or deference models are appropriate. Pediatricians should strive to keep all 3 models in their professional toolkit, employing the most appropriate model for each individual situation.
- Accepted March 9, 2017.
- Address correspondence to Bryan A. Sisk, MD, Department of Pediatrics, St Louis Children’s Hospital, 1 Children’s Place, 3S34, St Louis, MO 63110. E-mail:
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.
FUNDING: Dr DuBois was supported in part by National Center for Advancing Translational Sciences (UL1TR000448). Funded by the National Institutes of Health (NIH).
POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.
- Opel DJ
- Committee on Bioethics
- American Academy of Pediatrics
- Hein IM,
- De Vries MC,
- Troost PW,
- Meynen G,
- Van Goudoever JB,
- Lindauer RJ
- McCabe MA
- Cummings CL,
- Mercurio MR
- ↵Roper v Simmons, 543 U.S. 551 (2005)
- Partridge BC
- Partridge BC
- National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research
- U.S. Department of Health and Human Services
- Feinberg J
- Beauchamp TL,
- Childress JF
- Korsch BM,
- Gozzi EK,
- Francis V
- Nannis ED,
- Susman EJ,
- Strope BE, et al
- Bartholome WG
- Lo B
- Morreale M,
- Stinnett A,
- Dowling EE
- Miller VA,
- Cousino M,
- Leek AC,
- Kodish ED
- Copyright © 2017 by the American Academy of Pediatrics