A number of countries are beginning to implement plans to offer smallpox vaccination to sub-groups of healthcare workers in advance of a deliberate release (1, 2). At the same time there have been calls for mass vaccination of the public, or vaccination of the public on a voluntary basis. In the United States (US), while concentrating on selective vaccination, the government has decided to make the vaccine ‘available to all’ from 2004 (3). The European Commission Taskforce on Bioterrorism (BICHAT) has commented that a number of complex issues and present uncertainties would make a one-to-one translation of the US plan to the European situation an unjustified oversimplification (4). A series of articles was published electronically ahead of print publication in the New England Journal of Medicine on 19 December 2002 and aims to provide informed guidance on the current public health question ‘whom should we be vaccinating?’ By highlighting the adverse consequences of a mass vaccination policy, the articles support current plans for selective smallpox vaccination, and provide physicians with information to help them educate their patients when the inevitable requests for vaccination are received.

Mack counsels against widespread pre-emptive vaccination, either of healthcare workers or the public, because historical records indicate that a smallpox outbreak could be successfully and rapidly controlled (5). The disease is likely to be easily distinguishable, the virus is not highly contagious, and the interval between successive cases (about 14 days) makes transmission of infection open to intervention.

A new mathematical model from Bozzette et al favours prior vaccination of healthcare workers (6). The balance in favour of vaccination is biased towards the likelihood of an attack, which is difficult to assess. Their results show that pre-emptive vaccination of ten million healthcare workers produces net benefits in terms of lives saved when the probability of a ‘large’ attack (350 initial cases of smallpox) exceeds 0.22. Pre-emptive mass vaccination of the general public saves lives on a net basis only when a ‘massive’ attack (5000 people initially infected) is expected (probability greater than 0.22).

Sepkowitz considers adverse effects due to secondary spread of vaccinia, a factor that has not yet been taken into account in mathematical models (7). Complications from transmission of vaccinia have been described in 112 cases in the twentieth century, including 12 deaths. Most of these occurred in hospitals. If healthcare workers are to be vaccinated, it is essential that adequate precautions are taken to prevent spread of vaccinia to patients, especially now that many of them are immunosuppressed due to illness or treatment.

Voluntary vaccination of the public has its advocates both among health professionals and in the media. However a paper from Blendon et al shows that the public may not yet know enough about smallpox to be able to make an informed decision (8). In a questionnaire survey of 1006 US adults with a response rate of 65%, most people had misconceptions about the disease. Despite 25% saying it was likely they would die from the vaccine (6% thought this was very likely), over 60% said they would choose to receive vaccination as a precaution against a terrorist attack.

An issue that is not mentioned in these articles is the opportunity cost of smallpox vaccination. It may cause harm not only directly, but through the potentially very large resources consumed that might be better used to prevent or treat other diseases, and also by having a negative effect on public attitudes to other components of national immunisation programmes. It is essential to ensure that current smallpox vaccination policies are developed and implemented in the context of overall public health priorities and strategies, and that they are underpinned by sound science and risk assessment.