RAND study: Give health care workers the smallpox vaccine

Study says inoculations are prudent despite the risk

Widespread smallpox vaccination of the general population is too dangerous to justify unless the likelihood of a major biological attack on the United States is substantial. But it is prudent to vaccinate health care workers now against the disease, says a new study1 by the nonprofit RAND Center for Domestic and International Health Security.

The study, published in the Jan. 30, 2003, issue of the New England Journal of Medicine, estimates that if 60% of the U.S. population were immunized, there would be about 500 deaths — a price the RAND researchers say is too high to pay if there is little chance of a widespread attack against America. It also found that if nearly all 10 million health care workers in the United States were vaccinated against smallpox, an estimated 25 people would die. The study says the risk is justified, though, because health care workers could come in close contact with the sick before the disease is recognized. The article is available at www.nejm.org.

"After 9/11, we all wanted to do something, and decision making over smallpox could have an immediate impact," notes Samuel A. Bozzette, MD, PhD, a senior natural scientist at RAND Health Care in Santa Monica, CA, director of the health services team at VA San Diego, and lead author of the article. "RAND founded the Center for Domestic and International Health Security to turn its attention to the issues a little more vigorously. As both an infectious disease doc and a policy analyst, this was a natural for me."

How did the researchers determine if a certain level of risk was acceptable?

"In this case, the net benefit, or the bottom line, was determined by a set of thresholds that showed where the expectation was that you would save net lives," says Bozzette. "Our policy goal was to minimize the expectation of life loss — to help demonstrate which policy can be expected to minimize the lives lost due to smallpox."

A number of different scenarios were modeled, with varying levels of deaths predicted. Prior vaccination of health care workers provided a net benefit in the building and airport attack scenarios, "while prior vaccination of the public provided a benefit only in the airport attacks and greatly increased the number of deaths from vaccination."1

In addition, among the various scenarios, health care workers accounted for 19% of all infected people in the airport attacks and for 57% of those infected in the laboratory-release scenario.

Accounting for variables

The researchers took a number of variables into account — for example, a population is assumed to have a higher percentage of immune-compromised individuals than the population 50 years ago.

"We assumed that only 90% of the health care workers would actually be immunized," says Bozzette. "Secondly, we decided to apply historical complication and death rates. Historically, it has always been the case that a big chunk of the complications were from secondary complications. Even in the old days, half the complications were transmitted from one vaccinated person to another. In addition, adults are much less likely to transmit complications than children, and everyone should be counseled severely about precautions to take for covering of the site."

In addition, he says, in some sense the number of immunocompromised people is overblown.

"How many HIV-infected people are there?" he poses. "In the U.S., probably 800,000 or so in 290 million. Even if it came to several million, it’s a relatively small proportion of the population. It doesn’t mean we don’t have to check people, but it’s not like huge swaths of the population will be immune-compromised."

Another concern raised by critics of the government’s plan to inoculate health care workers is that a large number of individuals will have to be furloughed due to complications, thus limiting the ability to care for patients.

"Even in primary infections from the initial vaccination, the rates of serious complication run about 50-60 per million," Bozzette responds. "It’s extremely unlikely that any single facility will have enough people affected to impact the staff in any serious way. The numbers will be tiny, relative to the absenteeism you would normally see during the flu season." 

Increased preparedness is key

The authors note that their model "supports increased preparedness."1 Some of the keys to preparedness include:

Efficient delivery of the vaccine. Occupational health workers do not have direct control over this process, Bozzette observes, but forward positioning of vaccine supplies should be practiced — in other words, moving the vaccine into an area that can be easily reached if it is needed quickly.

Being well positioned to make a diagnosis. "For example," he notes, "the Scripps hospital chain is preparing kits, just as you would for other occupational exposures." These kits indicated key steps such as patient isolation, lesion testing, what tests to order, who to contact in the local health department, and so forth.

Being involved in your local community’s plan to respond. "Every health department in the country has been required to forward a plan, so your health department has a plan," says Bozzette. "You ought to know what it is and who does what. You need ready access to people who are experienced in response techniques."

Should you volunteer?

Given that vaccination is currently voluntary, would Bozzette say his group’s findings argue for universal vaccination of health care workers?

"What our model shows is that it is in the national interest for health care workers with patient contact to be vaccinated," he replies. "This includes individuals such as security guards. This is a population health issue."

On the individual level, he concedes, the odds are that people are not going to benefit, because the odds are there’s not going to be an outbreak in your town. "But from a public health and national security point of view, being vaccinated is the right thing to do," he says.

However, Bozzette says he understands it is a matter of individual choice. "Most of those who disagree with us don’t argue with the model per se, but with the conclusions we draw," he says. "I can tell you the feedback we got from health care professionals indicates they think the president’s recommendations are overly aggressive, but policy analysts and defense analysts think that it’s not aggressive enough. We believe inoculation will have a strong deterrent effect."

[For more information, contact:

  • Samuel A. Bozzette, MD, PhD, Senior Natural Scientist, RAND Health Care, 1700 Main St., Santa Monica, CA 90407-2138. Telephone: (619) 543-0620. E-mail: sam_bozzette@rand.org.]


1. Bozzette SA, Boer R, Bhatnagar V, et al. A model for a smallpox-vaccination policy. N Eng J Med 2003; 348:5.

Smallpox vaccination handbook available now

Call today for your copy of Hospital Handbook on Smallpox Vaccination, a facilitywide resource containing all the information you need about smallpox vaccination in most health care facilities and environments. Learn about possible side effects and adverse reactions and how to protect your staff, patients, and family members against this most controversial vaccine. Read about the challenges of implementing a smallpox care team plus critical staffing and screening issues that will affect your entire facility as well as crucial questions about liability and other legal concerns.

This newly released handbook confronts the implications that departments face throughout the hospital, including infection control, the ED, critical care, employee health, and risk management.

Order your copy today of this comprehensive resource guaranteed to provide you with the knowledge you need to manage this unprecedented response to a bioterrorism threat. The cost is just $99 for subscribers who will receive a discount off the original price of $149. To order, call customer service at (800) 688-2421. When ordering, please refer to the effort code: 76991.