In unvaccinated individuals, tetanus remains a constant threat. In developing countries, neonatal tetanus is common because of unhygienic practices, particularly with regard to care of the umbilical cord stump.1-3 Such cases usually occur in children born to unvaccinated mothers who have no antibodies to confer protection on their infants. In developed countries, where immunisation coverage is often higher and hygiene is better, neonatal tetanus is almost unheard of. Older, previously vaccinated adults (> 60 years) in whom immunity has waned form the bulk of cases of tetanus1 (seven notified cases in adults over 55 years in 1998, giving a notification rate of 0.2 per 100 000).4

In countries where there are high rates of childhood immunisation, tetanus is rare among children. For example, in South Australia, where more than 85% of children are immunised, the last known case of childhood tetanus occurred in 1969.4

A child presenting to the Women's and Children's Hospital, Adelaide, with tetanus (see Box) raised various important ethical questions.

Because this case raised difficult ethical issues, a peer-group discussion was convened under the auspices of the hospital's Patient Care Ethics Group. This is a group of clinicians with a largely supportive and advisory role and without authority to arbitrate on the hospital's behalf. This forum provided diverse points of view, with the aim of achieving the best decision possible.

Deliberations covered all aspects of child protection, the rights of the child and the rights of the parents. The clinicians involved in the care of this child had to decide how to respond when they considered the parents to be making the wrong decision for their child and to judge what should be the limits to the right of parents to make decisions on behalf of their children.

When a child succumbs to a vaccine-preventable disease for which the parents have refused vaccination, can such parents be said to be acting truly in their child's best interests? Based on toxin neutralisation assays in mice,3 and evidence of repeat episodes of tetanus arising from re-infection with Clostridium tetani,5-10 we hold the view that immunity to tetanus is not acquired by natural infection.3 On this basis, we were concerned that the patient remained at risk (albeit very low) of subsequent re-infection.

Do clinicians, acting in the interests of the wider community, have the right to use cases such as this to raise public awareness, with the hope of improving levels of vaccination? Should community pressure on parents to vaccinate their children be increased, for example by limiting access to child care services or school funding, or even by refusing school entry for unvaccinated children?11 How should we think about the economic cost of treating a child whose illness could have been prevented at minimal expense?

In treating this patient, we felt our duty to the child was compromised. However, vaccination in Australia is not compulsory and the decision to vaccinate, regardless of the clinical setting, remains with the parent. The doctor has a responsibility to give appropriate advice to the family: to provide comprehensive and up-to-date information on the benefits and unwanted effects of vaccination, as well as to offer a considered opinion on the advisability of vaccinating a particular person.

Our society (and most others) recognises that parents, generally speaking, are in the best position to make decisions for their children, for several reasons.12

• There is an identity-of-interests argument: the values parents hold are likely to be similar to their children's values, both now and in the future, as parents contribute significantly to their children's moral, psychological and social development. Therefore, asking parents what they think is a reasonable substitute for asking the children themselves.

• We tend to think that parents are likely to be in the best position to judge the best interests of their children. This is so because the "caring love" that most parents exhibit toward their children has at its core the promotion of the child's best interests.13

• Finally, and particularly important at the policy level, someone must speak for children and act on their behalf when decisions need to be made. At a societal and policy level, it is hard to imagine an alternative to parental responsibility for children that would not be incredibly burdensome for all involved.

How should we act when apparently thoughtful, caring and reasonable parents reach decisions that are not in accord with those of healthcare professionals? There is a range of reasons why healthcare professionals and patients do not always agree.14 Many "irrational" decisions by competent people reflect ways of thinking which others do not share. For example, some people may have an unreasonable bias toward the present, may give undue consideration to the risk of suffering or pain, or fear may get in the way when they are thinking about treatment. Alternatively, they may assess risk in unusual, but thoughtful, ways, or their belief systems may mean that choices that appear irrational to others are reasonable for them. In some cases, there may be a role for a psychiatric assessment to help clarify whether an apparently irrational view indicates that the parent is not competent to make the decision, or whether it forms part of a belief system that is against the best interests of the child. Menahem and Halasz15 provide recommendations on ways to reduce the risk of parental non-compliance, including building trust, eliciting the aid of a parental partner, and organising a second opinion, thereby improving the chances of a successful outcome.

In this case, it is unlikely that the parent was ill-informed or confused about the possible implications of not vaccinating the child, if only because of the experience of the child's serious illness and hospitalisation. The parent may not have understood the seriousness of tetanus before the child's stay in hospital, but surely did after the event. The clinicians involved also expended considerable effort to help the parent understand the implications of not having the child vaccinated. Hard as it may be for the healthcare professionals involved, we are forced to accept that this case is most likely an example of incommensurate value systems. If the clinicians shared this person's values and beliefs, they would likely reach the same conclusion. If this person shared the clinicians' beliefs, this child would have been vaccinated.

Faced with this conundrum, what could we do? We limited our response to a strong recommendation in favour of vaccination. Because parents are prima facie the principal decision-makers for their children, to override this responsibility requires that healthcare professionals be confident that the child will indeed be harmed, in both the short and long term, if the parents' decision is allowed to stand. In this case, the risk of significant harm because of immunisation status alone is quite small, given that tetanus requires both an injury and the introduction of Clostridium tetani spores and is not a communicable disease. In addition, the child's long term need for care in a loving and supportive environment could be jeopardised if vaccination were forced on an unwilling family. There is no reason, other than perhaps unusual beliefs about vaccination, to question the parent's capacity to provide an environment that served the child's best interests.

None of this suggests that healthcare professionals should not attempt to influence parents. Healthcare professionals have an obligation to try to persuade parents to do what the healthcare professionals consider to be in a child's best interests, provided this persuasion does not shade into manipulation or coercion.16 In this case, we made repeated attempts to persuade the parent while the child was in hospital. In addition, a letter was sent to the parents conveying the views of the medical team with respect to tetanus vaccination. The letter clearly reflected that its writing had been motivated by a duty of care on the part of the doctors towards the patient's future health. A copy of the National Health and Medical Research Council (NHMRC) recommendations1 for tetanus vaccination was sent with the letter.

Doctors have an obligation to promote and protect the health of the whole community. Communities can expect to be protected from infectious diseases, particularly when such protection is safe, relatively inexpensive and easy to administer. For infectious diseases other than tetanus, the effect on individual and herd immunity by dissenting families can be a major issue. However, this does not necessarily translate into a requirement that all members of the community be vaccinated whether they wish it or not. The best way to balance a community's right to be protected from infectious diseases with respect for individuals' autonomous decisions not to be vaccinated (or to have their children vaccinated) is by making vaccinations effective and accessible to all.17

Not only does the community have a right to protection from easily preventable infectious disease, but the community must surely have a say in how their tax dollars are best spent. There remains, therefore, the right to question whether it is a misuse of the healthcare system for individuals to seek cure for a largely self-inflicted disease, or, as in this case, a disease brought about by parental omission or choice.