red diagonal background pattern

Et Al: The Uncertainty Principle

By Alfred Sommer, MD, MHS ’73

The four anthrax letters that followed 9/11 killed five people, disrupted mail service, and closed Congress. Admirably, Americans didn’t panic. There was, however, pervasive uncertainty. We don’t like uncertainty, particularly when it relates to our lives.

We were uncertain whether there would be more anthrax. Federal officials were uncertain how best to treat Congressional staffers and postal workers in potentially contaminated facilities; in fact, they were uncertain about what constituted a “contaminated facility” (one anthrax spore on the carpet or 100 in the air?).

As a society, we expect more from medicine. Few doctors tell their patients they are uncertain how best to treat their flu, headache, or heart disease. As former Hopkins cardiologist and American Red Cross President Bernadine Healy recently told a Congressional committee: “I’m a physician. I see a problem, I diagnose it, and I treat it.”

By contrast, public health decisions affecting thousands or millions of lives must often be made without such “confidence.” An individual can take the advice of his or her physician to accept a low overall risk of a serious adverse event (say one in 1,000), but the same advice offered to millions might well result in hundreds or thousands of serious (even fatal) side effects. Public health officials constantly deal with uncertainty, even as they calculate the potential tradeoffs between risk and benefit.

In 1972 we abandoned routine smallpox vaccination in the United States even though we still faced the threat of imported cases. The risk of hundreds of cases of serious disease and five to ten deaths per year from the vaccine exceeded the potential risk of death and disease from imported smallpox. We had sufficient certainty from available data to make definitive policy, and sufficient confidence the public would understand and accept this balancing of risks until such time as the entire world was free of smallpox (a public health triumph achieved in 1978). 

Anthrax presented a different problem. We had insufficient experience upon which to make confident decisions. What evidence we did have suggested that 60 days of antibiotic treatment should prevent all clinical disease. Then doubts set in, largely because spores were found in the lungs of a few animals as long as 100 days after experimental exposure. Whether these long-hibernating spores could cause disease, or even whether this animal model was at all relevant to humans, was entirely unknown. But to play it safe, some in the medical community recommended an additional 40 days of antibiotic treatment, for a total of 100 days. They also advised making the anthrax immunization available to anybody potentially exposed who wanted it, despite any evidence that the vaccine was either safe or effective when administered after human infection. 

In the absence of sufficient data for calculating the potential “risk to benefit” tradeoff, scientists at the Centers for Disease Control would only advise patients about alternative options: Report to your physician at the first sign of illness; or take an additional 40 days of antibiotics (with or without concurrent immunization). For not having taken a decisive stance, in the absence of anything like decisive data, the CDC was pilloried by the press. 

The situation might have been different had we possessed a stronger public health system. Every person could have been closely followed; at the first sign of the delayed onset of disease in a single patient, all those presumed to have been exposed could have been vigorously re-treated. 

Withholding additional treatment and conducting close surveillance might have resulted in a few additional deaths. But, placed in perspective, this may have been acceptable: In the same 50 days that five people died of anthrax, roughly 4,000 Americans died in automobile accidents, 60,000 succumbed to their tobacco use, and 2,500 died from homicide. As it turned out, few subjects chose additional treatment or immunization, and no additional cases occurred.

Our recent brush with anthrax terrorism has taught us much—hopefully, more than we will ever need to use in the future. But we’ve not yet come to grips with the reality of risk and the necessary uncertainty of response. Life is filled with risk and uncertainty. The most we can hope to do is minimize both. President George W. Bush’s promise of vast new funding to strengthen our country’s flagging public health capacity could do just that, but only if it is sustained over decades, and not just the political “dollop du jour.”