One of the most divisive issues in pediatrics today concerns the proper response by pediatricians to parents who refuse routine childhood immunizations for their children. Many pediatricians refuse to care for such families. Others continue to provide care and continue to try to convince parents that the benefits of immunizations far outweigh the risks. Two of the most powerful arguments in favor of dismissing such parents are as follows: (1) their refusal suggests such lack of trust in the physicians’ recommendations that it undermines the basis for a meaningful physician–patient–parent relationship; and (2) unimmunized children present an unacceptable risk to other children in the physicians’ waiting rooms. This article examines those arguments.
- AAP —
- American Academy of Pediatrics
- MMR —
One of the most divisive issues in pediatrics today concerns the proper response by pediatricians to parents who refuse routine childhood immunizations for their children. Many pediatricians refuse to care for such families. Others continue to provide care and continue to try to convince parents that the benefits of immunizations far outweigh the risks. In this Ethics Rounds, we present a case of a pediatric practice trying to decide what their policies should be regarding immunizations. We asked experts in pediatric infectious disease to offer their advice.
A group of pediatricians is trying to decide what their policy should be with regard to parents who refuse immunizations for their children. Some physicians in the group think that it is important to continue to care for the children and to keep trying to convince the parents to immunize their children. They cite the policy of the American Academy of Pediatrics (AAP),1 which states, “In general, pediatricians should avoid discharging patients from their practices solely because a parent refuses to immunize his or her child.” Other physicians in the group cite another line from that same policy, “When a substantial level of distrust develops, significant differences in the philosophy of care emerge, or poor quality of communication persists, the pediatrician may encourage the family to find another physician or practice.” The group calls the hospital ethicist to ask what course of action is best and why.
Angela Myers comments:
Physicians are obligated by oath to “do no harm.”2 Although this obligation sometimes conflicts with other obligations (eg, when we prescribe chemotherapy for a patient with cancer), it is the spirit of this oath that we carry with us every day. In addition, physicians are expected to provide beneficence to patients while minimizing maleficence.3
In the pediatric setting, immunization is second only to hand-washing in terms of “doing good” by promoting the health and well-being of children.4–7 However, this idea falls flat when a parent refuses to follow our recommendation to immunize their children. Of course, we recommend many things, and parents often do not follow our recommendations. Generally, we continue to care for their children and encourage them to lead healthier lives. I think immunizations are different. Let me explain why.
The difference is in the implications that vaccine refusal has on other children. If we tell people to wear their seat belts and bicycle helmets, and they do not do so, they are only hurting themselves. If, as a result of not following our advice, something terrible happens, such as being thrown from a vehicle during a crash or experiencing an intracranial bleed from hitting their head on a curb, it happens only to that person. There is no risk to others conferred by their reckless decision.
The lack of risk to others is not the case with immunizations. When a nonimmunized child gets measles or an infection with Haemophilus influenzae type b, they have exposed other children to these life-threatening diseases. The exposure begins in the days before the illness even began, when they were completely asymptomatic. Secondary cases can be seen in closed communities of nonimmunized and underimmunized populations8 but can also be seen in larger populations.9,10
We recently cared for 2 children with H influenzae type b disease from a nonimmunized community. Substantial morbidity was seen in both cases, and 1 of the children died. This outcome led to a community-wide immunization effort through the county health department.11 Similar patterns occurred, and on a much larger scale, in 2015. A large measles outbreak occurred in southern California, attributed to families with nonimmunized children visiting Disneyland. Sixteen percent of cases occurred in infants who were too young to be immunized. An additional 12% of cases occurred in children between the ages of 1 and 4 years, who would have had only 1 dose of measles-containing vaccine.7
Transmission of vaccine-preventable infections could occur in the community office setting in which a nonimmunized child with a communicable disease inadvertently transmits their infection to a young infant or a child who is immunosuppressed. This eventuality is even more likely when one considers the rate at which families are traveling internationally. In 2014 alone, >68 million people traveled from the United States to an international destination.12 A significant portion of these trips are to places in which immunization is much less common and therefore the risk of contracting a communicable disease is considerably higher. As in the Disneyland outbreak, the index case is often suspected or proven to be a nonimmunized individual who contracts the disease while traveling overseas and then becomes ill after arrival back home.7 Thus, the risk of exposure to a communicable disease in the waiting room of the physician’s office, while low, is not zero.
The presence of such risks creates a difficult dilemma for primary care pediatricians. All children deserve safe, high-quality medical care. Pediatricians, overall, are committed to the care of all children. However, that commitment to safe, high-quality care means that pediatricians have an obligation to make their offices and clinics as safe as they can possibly be. No child should be exposed to a potentially life-threatening communicable disease in the physician’s office.
The 2012 policy statement from the AAP1 recognized the tension between pediatricians’ general obligations to care for children and their specific obligations to protect children from risks. The statement begins, “In general, pediatricians should endeavor not to discharge patients from their practices solely because a parent refuses to immunize a child.” I agree with this statement. In my view, pediatricians should make sincere and strenuous efforts to answer parents’ questions and address their concerns about immunizations.13 They should provide the evidence in a nonjudgmental way to convince families of the benefits of immunization. They should acknowledge the reality of a low, but not zero, risk of adverse reactions to immunization.14 In addition, pediatricians should consider a focused attempt to gain trust by providing the potentially most important of the recommended vaccines (eg, pneumococcal conjugate) per the recommended schedule, while allowing for delays of others (eg, polio) if it means that the child will eventually be fully immunized.
I am not advocating for an alternative immunization schedule. However, the AAP policy states that “As respect, communication, and information build over time in a professional relationship, parents may be willing to reconsider previous vaccine refusals.”15 Allowing some compromises up front may increase trust, and therefore willingness to immunize, later. Indeed, some of the strategies described here and detailed elsewhere have, in fact, been shown to increase immunization rates.16,17
However, pediatricians can and should only go so far. Not all families are open to discussion of this important issue. Some categorically refuse all preventive measures from birth. Families who refuse vitamin K, erythromycin ointment, and the initial hepatitis B vaccine are more likely to continue to refuse all vaccines in the future.18 The pediatrician has a perfect opportunity in the first few days of life to convince a parent that these preventive measures are important and, in rare instances, lifesaving. If the pediatrician is unable to gain trust during this initial period, what hope do they have of developing a respected partnership with a family? Partnership implies shared decision-making and a give-and-take relationship. This bond cannot exist when one party refuses to consider recommendations made by the other.
In situations in which it is clear that further discussion will not be fruitful, it is appropriate for the pediatrician to refer the family to a different health care provider, one who shares their values and with whom they can establish a trusting and nonconflictual relationship.18
Pediatricians do not want to dismiss any patients or families from their practice. The AAP understands this objective and endorses it. However, the AAP policy also appropriately recognizes that there are uncommon circumstances in which the pediatrician and family are at an impasse, and the only logical conclusion is to sever the relationship. The primary reason why it is sometimes appropriate to sever the pediatrician-family relationship is that the child may get higher quality care from a pediatrician who shares the parents' values. A pediatrician whose concerns about transmission of infectious disease in the waiting room leads to anxiety about other patients and frustration with parents and may not be able to provide quality care.
Kenneth Alexander and Thomas Lacy Comment:
We are all concerned that the presence of unvaccinated children in our offices poses a small but real risk of transmitting vaccine-preventable diseases to our other patients, especially to those who are immunocompromised, those who cannot be immunized, and those who, by no fault or choice of their own, face the prospect of significant morbidity or mortality should they acquire a vaccine-preventable disease. We often believe that nonvaccinating parents, by not immunizing their own children, knowingly (albeit unintentionally) place our vulnerable patients at risk. Given that nonimmunizing parents knowingly place other peoples’ children at risk, it is understandable that many pediatric practices view nonimmunization as a breach of both physician–family trust and patient safety that merits a severance of the relationship with the family. In practices who choose to exclude unvaccinated families, parents are generally given a period of time to get their child's immunizations up to date, after which time they are advised to seek alternative care. Such deadlines often lead families to get their children immunized. Unfortunately, some families (no one knows how many) simply disappear from the practice to an unknown quality of care. Thus, although exclusion may be a successful means of encouraging nonimmunizing parents to accept vaccination for their children, and may also increase the safety of our vulnerable patients by reducing their exposure to nonimmunized children, at least in the office waiting room, what is the cost of this action to excluded children?
Countering the argument that nonvaccinated families should be dismissed from practices is the belief that children are still best cared for by pediatricians, even children in families who refuse to immunize. This argument is powerful. We believe that an nonimmunized child will fare better in the care of a pediatrician (who, even if a family refuses to immunize, will still advocate on behalf of the child, recommending bicycle helmets, bathwater safety, healthy diet, and safe storage of firearms) than in the care of a less qualified practitioner (who may recommend disproven or unproven and potentially dangerous remedies, such as homeopathy or naturopathy). By continuing to care for the child, the pediatrician can continue promoting immunization and will continue providing other recommended health and safety guidance and intervention.
When addressing the question of exclusion of families for failing to immunize their children, our ethicist colleagues and the AAP provide us with mixed messages.1 On the one hand, they advocate nonabandonment. On the other hand, they permit physicians to choose whom to serve. When debating the matter of excluding nonimmunizing families from pediatric practices, the principles of beneficence and nonmaleficence are often cited. Unfortunately, applying these principles is challenging and gives us opposing answers. On the one hand, pediatricians can justify excluding nonimmunizing families from their practices because pediatricians believe that nonimmunized children place other patients at risk, especially in the medical office. By excluding nonimmunizing families, pediatricians believe that they are protecting their most vulnerable patients. Thus, excluding nonimmunized children from a practice can be viewed as form of beneficence toward the vulnerable, protecting them from harm. On the flip side, excluding a nonvaccinated child from a practice could be viewed as detrimental to that child's health, provided we believe that pediatricians do good beyond simply immunizing children. From this vantage, excluding nonvaccinating families from our practices denies children the benefits of our nonimmunization-related health promotion efforts. From this view, exclusion of nonimmunizing families from our practices can be viewed as a maleficent action. Furthermore, we doubt that exclusion of nonvaccinating families would have the desired effect of protecting our vulnerable children. Insofar as it is unlikely that a family excluded from a practice for refusing to immunize their children is unlikely to seek immunization elsewhere, exclusion itself probably does nothing to decrease the proportion of nonimmunized children. Because excluded parents are not likely to seek vaccination elsewhere, thus leaving excluded children unvaccinated, exclusion may well maintain the possibility that our at-risk patients will be exposed to vaccine-preventable diseases out in the world, if not in our office waiting rooms; another maleficent act. Thus, ethical analysis based on beneficence and nonmaleficence points us in 2 different directions; excluding unimmunized children from practice is a beneficent act performed for the benefit of the vulnerable, but excluding children is a maleficent act toward the nonimmunized child because we are denying them the benefits of our nonvaccination-related care, potentially with the added cost of continued nonimmunization. Applying the principles of beneficence and nonmaleficence to the question of excluding nonimmunizing families from our practices leaves us in an ethical fog of opposing conclusions. Maybe we need a tool beyond beneficence and nonmaleficence.
Perhaps the ethical principle we should use to address the exclusion question is that of double effect.19 The principle of double effect asserts that when an intended positive action carries with it a smaller and unintended negative effect, the action is still positive overall. The principle of double effect requires that:
The intended effect is positive
The positive effect is intended
The negative effect is not intended as a means to the positive effect, nor as an end in itself
The positive effect outweighs the negative effect
Continual effort is made to minimize the negative effect
Applying the principle of double effect to the question of excluding nonimmunizing families from our practices, I am confident that good pediatric primary care brings to children and their families frequent substantial benefits that extend beyond vaccination. Good primary care is a definite and substantial positive, even without immunization. To assess the degree of unintended negative effect (the risk that unimmunized children pose to vulnerable children in our waiting rooms), we must keep in mind that the rates of vaccine-preventable diseases (with the exceptions of human papilloma virus infection and influenza) are low compared with the rates of nonimmunization.) For example, in 2014 in California, there were 9 157 390 people aged <18 years.20 According to the National Immunization Survey–Teen, 8.6% of people aged <18 years in California had received <2 doses of measles-mumps-rubella (MMR) vaccine.21 This scenario means that in 2014, a total of 787 536 Californians aged <18 years had received 0 or 1 dose of MMR. If it is assumed that all 667 measles cases in California22 in 2014 occurred in patients aged <18 years who had received 0 or 1 dose of MMR, the measles attack rate among that population in California was then 0.085%.
Thus, even among nonimmunized children, the rates of vaccine-preventable diseases are low; transmission of vaccine-preventable disease to a vulnerable child is an even rarer event.
Combining our confidence in the good that primary care pediatricians do every day (the intended positive effect) with the observation that risk of transmission of a vaccine-preventable disease to a vulnerable child in our offices is low (the lesser and unintended negative effect), we assert that the care we give by keeping unimmunized children in our practices creates more intentional positive effect than the unintended negative consequences of increased risk to our vulnerable children. In addition, because the principle of double effect requires that we continually strive to mitigate the unintended negative consequence (the risk of transmission of a vaccine-preventable disease from an unimmunized patient to a vulnerable patient), we keep nonimmunizing families in our practices, unceasingly educating and encouraging them to accept immunization for their children. Taken together, nonexclusion with continued encouragement to vaccinate satisfies the principle of double effect.
In our final assessment, we believe that by keeping nonimmunizing families in our practices, the very small risks brought to vulnerable children are exceeded by the large benefits brought to children in our nonimmunizing families. If all we did was immunize, and if we only offered immunization once, we would vote to exclude nonimmunizing families. However, this scenario is not the case. We as pediatricians do so much more than immunize. We teach, we advocate, we role model, and even when parents say no to immunization the first 10 times we bring it up, we bring it up an 11th and 12th time. Therein lies at least part of the answer: perhaps the real solution to the problem of nonimmunizing families lies with us as we cultivate relationships with families and as we continue in our unceasing efforts to persuade parents to vaccinate. We keep families with us, and we keep advocating for immunization. Thus, our vote in the end is to keep nonimmunized children in our offices, while continuing (and perhaps increasing) our zeal for immunization.
John D. Lantos Comments:
The debate about vaccines is an excellent example of a situation in which people on opposing sides disagree about the most fundamental facts. As a result, arguments that are based on those contested facts cannot possibly persuade those who do not accept the facts as a basis for argument. The result is that arguments become more and more strident, the conclusions more and more extreme, and the ethical implications that follow from those conclusions more and more questionable.
This outcome happens both at the impersonal level of policy making (eg, which vaccines should be mandated and what sorts of exemptions should be permitted), as well as at the deeply personal level of the individual physician–patient–parent relationship.
The fallacies of parental arguments against immunizations are well rehearsed and well understood by pediatricians; thus, I will not re-analyze them here. Instead, I will highlight the fallacies or inconsistencies in some of the more common arguments made by pediatricians in support of decisions or policies not to care for vaccine-refusing parents.
The first and most powerful argument that is often put forward by pediatricians to justify the dismissal of such parents is that the parents’ unwillingness to follow the pediatricians’ recommendations regarding immunization leads to a breakdown in trust and thus a fatal blow to an ongoing physician–parent relationship. However, parents disagree with many of the things that their physicians recommend, often with consequences more dire than vaccine refusal. Many parents continue to feed unhealthy diets to their obese children, despite constant exhortations by physicians to curtail the child’s junk food consumption. I know of no pediatricians who suggest that they should refuse to care for obese children. Many parents continue to smoke, putting their children at risk for infections and asthma. I know of no pediatricians who refuse to care for children whose parents smoke. Parents do not insist on bike helmets, they do not apply sunblock, they keep firearms in their homes, and they let their children play football. These actions are all associated with childhood injuries and deaths. If disagreement or nonadherence with pediatricians’ recommendations about any one thing suggested broken trust on all things, pediatricians would then be left with very few patients indeed. Instead, a centerpiece of medical ethics from the time of Hippocrates until today has been that physicians should care for patients regardless of the patients’ beliefs, values, or practices. Physicians care for the rape victim and the rapist, the enemy soldier as well as their own troops, the virtuous patient and the sinner. Why, of all transgressions, should vaccine hesitancy be the one that obliterates professional obligations?
The other powerful argument for dismissing unimmunized children is the risk to other children in the waiting room. It is entirely appropriate for physicians to be concerned about all their patients. But the key question is not whether the nonimmunized child presents some risk but how the magnitude of that risk compares with the other risks present in the waiting room. Pediatricians’ waiting rooms are very dangerous places to be; it would be unwise for an immunocompromised child to ever sit in a pediatrician’s waiting room. By doing so, he or she would be exposed to a frightening variety of pathogens. The only way to truly protect that immunocompromised child would be to refuse to care for any child who had a communicable disease or was at risk for getting one (because, of course, children are infectious before they are symptomatic).
Given the concentration of pathologic conditions in most waiting rooms, the incremental risk posed by an asymptomatic nonimmunized child is very small indeed. Oddly, the greatest risk probably comes from parents who do not annually immunize their child against influenza. But that lapse is only rarely cited as a reason to dismiss parents. If, in fact, a physician truly is primarily concerned about other children in the waiting room, created by an unimmunized child, then they should treat the unimmunized child as if he or she were always contagious. They should insist that the child wear a mask at all times, or wait in the car until an examination room is ready. That would greatly reduce the risk to other children. It might also send an explicit message to parents about the doctors' concerns. It might even induce them to voluntarily seek another provider.
There may be good reasons why pediatricians should not care for children whose parents refuse immunizations. However, neither a breakdown in the physician–patient relationship nor the risks to other children in the waiting room are good reasons to do so.
- Accepted May 12, 2016.
- Address correspondence to John D. Lantos, MD, Children’s Mercy Hospital, 2401 Gillham Rd, Kansas City, MO 64108. E-mail:
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.
FUNDING: No external funding.
POTENTIAL CONFLICT OF INTEREST: Dr Alexander is a paid speaker and consultant for Merck Vaccines; the other authors have indicated they have no potential conflicts of interest to disclose.
- American Academy of Pediatrics Clinical Report
- Sokol DK
- Pittet D. Improving adherence to hand hygiene practice: a multidisciplinary approach. Emerg Infect Dis. 2001;7(2):234–240
- Myers AL,
- Jackson MA,
- Zhang L,
- Swanson D,
- Gilsdorf J
- US Department of Commerce, International Trade Administration, National Travel and Tourism Office (NTTO)
- Diekema DS; American Academy of Pediatrics Committee on Bioethics
- Sahni V,
- Lai FY,
- MacDonald SE
- American Medical Association; Council on Ethical and Judicial Affairs
- Aquinas T
- ↵United States Census Bureau. Quick facts, California. Available at: www.census.gov/quickfacts/table/PST045215/06#. Accessed 4 May, 2016
- ↵Centers for Disease Control and Prevention. For immunization managers. Available at: www.cdc.gov/vaccines/imz-managers/coverage/nis/teen/data/tables-2014.html#overall. Accessed May 4, 2016
- ↵Centers for Disease Control and Prevention. Measles (rubeola). Available at: www.cdc.gov/measles/cases-outbreaks.html. Accessed May 4, 2016
- Copyright © 2016 by the American Academy of Pediatrics