Immunisation against chickenpox

Better to confine immunisation to those at high risk

There are three main arguments for universal immunisation against chickenpox in childhood. Firstly, immunisation is good for the children who are immunised; secondly, it is good for immunocompromised children, who will be protected from exposure to children with chickenpox; and, finally, it is cost effective because fewer parents need to take time off to take care of children with chickenpox. In our view, these arguments are not powerful enough to justify universal immunisation.

The natural course of chickenpox is well defined. Most reported cases occur in children under 10, who usually develop a vesicular rash that erupts in clusters and scabs over one week and causes troublesome itching. It is often associated with mild fever and other systemic symptoms. In older patients pneumonia is the most common complication, but bacterial superinfection, meningoencephalitis, and glomerulonephritis may also occur. Death or long term illness from primary chickenpox in immunocompetent children is exceedingly rare. At present, then, chickenpox is a benign illness.

Chickenpox in adults may be much more severe. During the first two trimesters of pregnancy it may result in chickenpox embryopathy. In the last trimester it may result in neonatal chickenpox, which, if severe, may be associated with a mortality as high as 30%.1 Immunocompromised patients are also at risk of serious infection. It is hard to isolate immunocompromised children from community outbreaks of chickenpox because children can transmit the disease several days before they become clinically ill. To protect immunocompromised children, doctors often recommend that healthy recuperating children should be kept out of school until all lesions have scabbed over, even if they do not feel ill.

The main problem with immunisation is that we do not know whether children who are immunised with chickenpox vaccine develop lifelong immunity. In immunocompromised children immunity persists in most of those who have been immunised for at least six years,2 3 but long term immunity is thought to require re-exposure to natural infection or reimmunisation. If the protective effect of immunisation wanes a programme of universal immunisation may create a population of adults who are at risk of serious illness and thus turn a relatively benign childhood illness into a major cause of illness and teratogenicity.

Most childhood immunisations, such as those against Haemophilus influenzae type b infection or pertussis, protect each child as well as promote herd immunity. Universal immunisation programmes benefit all children by protecting them from illnesses that can be severe in those who are young. Even if immunity wanes, infection during adulthood usually leads to less severe disease. By contrast, chickenpox in young healthy children is quite mild, whereas primary infection during adulthood can be severe. Thus the benefits to most children from chickenpox immunisation would be minimal: the benefits accrue only to immunocompromised children.

A programme of universal immunisation to benefit immunocompromised children would require doctors to ask parents to authorise the immunisation of their children not for their own benefit but for the benefit of their less fortunate classmates. Parents would be asked to place their children at potentially increased risk of primary chickenpox as adults. This is compulsory altruism. Given that we do not compel adults to serve as kidney or even blood donors, it seems unfair to require children to be “splendid Samaritans.”4 This also contradicts the “best interest of the child” standard, which is the usual guiding principle for parental decision making.

If the goal of chickenpox immunisation is to protect immunocompromised children other strategies should be used. One option is to immunise high risk children—and the chickenpox vaccine has been given successfully to immunocompromised children.5 6 These children can be further protected by the use of varicella zoster immune globulin and acyclovir. While these children and their parents may desire the increased protection of community immunity, the increased risks that such immunity entails for otherwise healthy children cannot be justified.

The costs of chickenpox infection are partly the medical expenses and partly the days of work lost among families. The medical expenses are generally low. Studies have shown that universal chickenpox immunisation is not cost effective in terms of health costs alone.7 8 These studies may even underestimate the costs, because they do not account for the possible increase in costs if universal immunisation delays disease until adulthood.

The cost of days of work lost by parents because of their children's chickenpox is substantial, and universal chickenpox immunisation would probably be cost effective from this angle.8 9 A large part of the cost, however, is due to policies of isolation. We believe that this cost is avoidable; if it was avoided this might tip the balance of the cost-benefit studies against universal chickenpox immunisation. Children should not have to stay home while asymptomatic but still capable of transmitting the disease. This policy, which is justifiable primarily on the basis of its benefit to immunocompromised children, in fact offers such children false security since they are still exposed to children who are presymptomatic but are capable of transmitting the disease. The best way to protect immunocompromised children is to immunise them, not all their peers.

A policy of mandatory universal immunisation would be justified only if the benefits of participation for each individual outweighed the risks and costs. Given the mild course of chickenpox in healthy children, such a policy is not justified. Chickenpox immunisation should be recommended only to families in which one or more members are at high risk of serious infection.