Childhood immunisation against varicella zoster virus

BMJ 2008; 337 doi: http://dx.doi.org/10.1136/bmj.a1164 (Published 19 August 2008) Cite this as: BMJ 2008;337:a1164
  1. Andrew Farlow, research fellow in economics
  1. 1Oriel College, Oxford, Oxford OX1 4EW
  1. Andrew.Farlow{at}oriel.ox.ac.uk

    Universal adoption depends on science, politics, and society's attitude to risk

    When childcare centres, schools, and colleges in the United States reopen after the summer break, attendees will need to prove that they have been immunised against the varicella zoster virus—which causes varicella (chickenpox) and herpes zoster (shingles)—or to demonstrate naturally acquired immunity.1 In the United Kingdom (and most of Europe), the debate intensifies as to whether we should follow suit and instigate universal childhood immunisation against this virus.

    Immunisation has reduced the incidence of varicella in the US since the first vaccine became available in 1995.2 3 4 So why have more countries not adopted universal immunisation?

    Firstly, there is a concern that childhood immunisation for varicella zoster virus may shift the burden of disease to adults. Most children get varicella at some time, and in most cases the consequences are not life threatening. Getting varicella as an adult, when pregnant, or when immunologically compromised generally leads to more severe complications. Adults are 23-29 times more likely to die from primary infection.5 Once a high proportion of children are routinely immunised it is increasingly difficult for non-immunised children to contract varicella naturally, which exposes them to greater risk from varicella in adult life. This is the main reason that the school and college requirement is being enforced.

    Secondly, getting varicella as a child confers lifelong immunity, but it is not yet clear how long vaccine induced immunity will last. Early data indicated that the duration of immunity from a one dose regimen was at least 20 years.6 However, because varicella was so endemic, it was not until most children had been immunised that children in trials were not re-exposed to varicella through friends and siblings—a phenomenon that would potentially extend their immunity.7 Recent analysis has shown that efficacy wanes more quickly than had been hoped.8 In 2005, the US adopted a two dose regimen.1 Current evidence indicates that this will substantially reduce disease outbreaks and severity of disease.9

    Thirdly, the value of universal childhood immunisation rests heavily on the relation between varicella and herpes zoster. The virus lies dormant in the sensory nerve ganglia of people who have had varicella until a decline in cell mediated immunity, usually in old age, is thought to trigger reactivation as herpes zoster. Herpes zoster, especially if it leads to neuralgia, is difficult to relieve and can seriously reduce the quality of life of elderly people. One way to protect against reactivation is thought to be to boost immunity via occasional re-exposure to wild-type (naturally occurring) varicella, often through contact with children and grandchildren.10 If this is correct, vaccine induced herd immunity will reduce circulating wild-type varicella and increase the prevalence of herpes zoster. When the varicella vaccine was licensed, this possibility was mostly based on mathematical modelling.11

    Monitoring could check whether immunisation has increased the prevalence of zoster, but population sizes in sites where varicella surveillance occurs have often not been large enough to monitor the age specific incidence of herpes zoster. It will also take some time for the effects of universal immunisation to be seen. As children become adults, if a vaccine yields life long immunity and if coverage is universal then the risk of herpes zoster in those who are immunised will decline.

    Policy makers in the UK have several different immunisation strategies to choose from. In the absence of universal childhood immunisation, immunising adolescents who have not acquired natural immunity will have little effect on childhood varicella but will reduce cases of adult varicella and help reduce cases of herpes zoster. Immunising high risk groups (to the extent that they can be identified) and those who come into contact with high risk groups, such as hospital staff, has been effective in preventing cases in adults and fetuses. In the case of universal childhood immunisation, coverage would have to be very high and should aim eventually to eliminate circulation of the wild-type virus; this would be even more important if the vaccine induces insufficient immunological memory. If a second dose does not provide immunity into old age, booster shots will be needed as populations age. In addition, to counter any decline in boosting for people who were not immunised but who were infected with wild-type varicella, booster zoster vaccines would probably have to be used. In the US, the Advisory Committee on Immunization Practices (ACIP) recommended in October 2006 that people aged 60 and over—including those who had had previous episodes of herpes zoster—should receive a recently licensed zoster vaccine.

    To complicate matters, some UK parents have opted not to immunise their children after rumours about the dangers of the combined measles, mumps, and rubella vaccine, so a compulsory programme is politically problematic. If compulsory, one solution would be to offer a choice of how a child is immunised—either via a combined measles, mumps, rubella, and varicella vaccine or a stand alone varicella vaccine. Indeed, in the US, the ACIP recently reversed its preference for the combined vaccine to a neutral position. A high proportion of adults are also unlikely to adhere to a voluntary programme of adult immunisation and boosters are unlikely to be as protective as natural boosting.12 At the very least, arguments in favour of universal childhood immunisation against varicella must also contain strategies to immunise adult populations.

    Policy makers face a dilemma. Must we always wait for all the evidence we need before acting so that we will be judged, with hindsight, as having made the right decision? What if we can only know by taking actions that give us the natural experiment that will tell us the best policy? Whether or not we should have universal childhood immunisation for varicella in the UK hinges on a complex mix of society’s attitude towards different risks, politics, and scientific understanding.


    Cite this as: BMJ 2008;337:a1164


    • Competing interests: None declared.

    • Provenance and peer review: Commissioned; not externally peer reviewed.