Literature Review: Rand Corp. identifies risks of smallpox vaccinations

Bozzette SA, Boer R, Bhatnagar V, et al. A model for a smallpox-vaccination policy. N Engl J Med 2003; 348(5):416-425.

How great is the risk of a smallpox attack? That question underlies the current campaign to vaccinate health care workers and military personnel — and to offer the vaccinia vaccine to those who want it in the general public.

The benefit of those vaccines can’t be calculated without an estimate of the risk both of smallpox and of vaccine-related adverse events. Researchers at the RAND Corp. in Santa Monica, CA, have attempted to do just that. They identified six possible scenarios, from a hoax to a high-impact attack at a major airport. Their conclusion: Vaccination of health care workers is justified. Widespread vaccination of the public is not.

"In our judgment, the probability of a release of variola virus may exceed the thresholds for prior vaccination of health care workers," the researchers stated. "We endorse a policy of vaccinating all eligible health care workers and first responders before an attack."

Such a policy is even more justifiable when the risk/benefits are calculated on a local basis, the authors said. "Local officials should welcome such a program, which should include appropriate monitoring and evaluation," they wrote. "In contrast, we cannot endorse a public vaccination campaign at this time, because the certainty of harm outweighs the small chance of a net benefit. Nonetheless, we acknowledge the distinction between this position and the argument for allowing access to vaccination on demand."

In the model, the RAND researchers estimated that vaccination would lead to 25 deaths of health care workers and 482 deaths nationally if the general public received the vaccine. It assumed that about 9 million health care workers would receive the vaccine.

No one knows whether any terrorist group or nation hostile to the United States has access to variola. In fact, the authors noted, "The known supplies of variola virus are limited. Rogue states with the virus would probably fear boomerang effects or devastating retaliation, and terrorists are unlikely to be capable of successfully handling a lethal mammalian virus."

Nonetheless, they assumed that such an event is possible, and they create a range of scenarios: A hoax using monkeypox virus via a mailed threat (monkeypox is not transmitted via person-to-person contact); a release in which a vaccinated laboratory worker unknowingly transmits variola to his or her children due to sabotage of a biosafety hood; attacks in which previously vaccinated terrorists infect themselves, then ride mass transit; aerosolized virus sprayed into a building vent; and attacks with aerosolized virus in terminals of 10 major U.S. airports, with either low- or high impact (depending on success of aerosolization).

The authors then considered how many cases would occur before the smallpox outbreak could be halted. The number of initial infections in the scenarios range from two to 100,000. Including secondary spread of the virus, the expected number of deaths range from seven to 54,728.

"We used public sources of data and expert opinion to develop detailed, realistic, and feasible scenarios for smallpox attacks, given access to variola, historical tendencies, and methods of terrorists, and known domestic vulnerabilities," they stated.

"We considered several control strategies, including vaccination of contacts of infected persons [household members, health care workers, and others] and isolation of patients, as well as pre- or post-attack vaccination of 60% of 290 million U.S. residents, 90% of 10.1 million health care workers, or both."

The authors noted that most outbreaks of smallpox in Europe and North America after World War II involved a single index case, and that the highest number of secondary cases occurred in hospital transmission.

"Prior vaccination of health care workers is expected to save lives at lower threshold probabilities of an attack and is expected to cause relatively few deaths, which will be concentrated among workers whose professional ethic includes acceptance of a risk of personal harm for the public good," the authors stated. "In addition, prior vaccination sharply reduces the disproportionate burden of disease among health care workers in the event of an attack and, by eliminating a major route of exposure, helps protect their families — effects that should help maintain staffing levels at health care facilities," they wrote.