about Epidemiology & the department

Epidemiology academic information

Epidemiology faculty

Epidemilogy resources

sites of interest to Epidemiology professionals

Last Updated

18 Apr 2003

Source: Washington Post, April 18, 2003


Unprepared For a Plague

By Philip Alcabes

Smallpox has been on the minds of public health officials for the past year, as Americans worry about a bioterrorist attack. But the disease that has struck isn't smallpox, nor is it somebody's deliberate attempt to spread germs. Rather, it is nature's newest surprise: severe acute respiratory syndrome (SARS). SARS has gone from a few cases of pneumonia in southern China to a worldwide infection in just four months, with more than 150 deaths so far. This should remind us about the basics of disease control.

In his State of the Union address, President Bush called for a Project BioShield. Millions of dollars in federal grants are going to researchers developing electronic "biopreparedness" systems. The Centers for Disease Control and Prevention (CDC) has opened a high-tech "war room" to keep tabs on SARS and other new infections. Many cities are setting up expensive electronic surveillance systems in hopes of better watching for bioterrorist events. But while it always sounds exciting to put cutting-edge technology to use, Americans would be better off if the public health apparatus simply dealt well with the problems we know how to manage instead of trying to foresee the unlikely ones.

Take the early-warning monitoring for bioterrorism -- what the epidemiology techies call "syndromic surveillance." Syndromic surveillance doesn't work. The first anthrax events of 2001 would have escaped attention but for the savvy of a Florida clinician; databases were no help. The anthrax cases later that year happened despite the CDC's complete knowledge of the method of infection and willingness to share data to prevent it. Another example: All through the '90s, good infectious-disease epidemiologists knew that mosquito-borne disease was going to break out in the New York City area after years of poor vector control. Yet the emerging-infections surveillance system of the CDC couldn't predict the coming of West Nile encephalitis (or even identify it when it did occur) in 1999.

The crux of the problem with syndromic surveillance is that it relies on statistics instead of science. Most of the effort in establishing bioterror early-warning systems goes into estimating the probability, based on what has been seen beforehand, that some cluster of cases might occur naturally. If that case cluster seems improbable, it is deemed an "aberration" -- and it merits action. The problem is that it's only when we have seen an event before that we can calculate its probability. When we are looking for the unprecedented, probability is of no use at all. The nearly 3,000 dead in the World Trade Center disaster attest to that.

It is the syndromes we haven't thought to watch for that will signal the next plague. We should have learned that lesson a quarter-century ago. Then, smallpox had just been eradicated, and with vaccines and good antibiotics, it seemed likely the other epidemic infections were on their way to a similar fate. But along came AIDS. Who, back in the early '80s, thought that uncommon infections in a few otherwise healthy young men were the first signs of what was to become the global pandemic of HIV/AIDS? Maybe SARS will be the new plague, or maybe it will be something we can't yet imagine.

Besides chasing the chimera of the unforeseeable, biopreparedness systems are based on an erroneous premise. Our public health authorities presume that bioterrorism is a serious threat to public health. They're wrong. The number of deaths attributable to willfully produced epidemics, ever, pales by comparison with the toll taken by natural ones. In 1918-19 an influenza pandemic killed more people in just 16 months than World War I had killed in six years. Smallpox killed 10 times as many people in the first half of the 20th century as did both world wars combined. Even today malaria kills 2 million people each year; so does tuberculosis. By contrast, deliberate epidemics in the past 100 years, mostly through the actions of armies at war, have been responsible for only a few thousand deaths.

So what is to be done? Public health officials should stop tinkering with electronics and get down to basics. What America needs is not more people mining more data, but better public health. We already know how to do this: Provide good primary care, track disease outbreaks by counting cases, run effective preventive programs, vaccinate, and keep the food safe, the air breathable and the water drinkable. We can make people healthier and save more lives if we resist the lure of databases. And if we succeed at maintaining good public health, we can protect ourselves against most eventualities better than if we waste time and resources looking for "aberrant clusters."

The writer, an infectious-disease epidemiologist, is an associate professor in the Program in Urban Public Health at the Hunter College School of Health Sciences, City University of New York.