CDC based its optimistic assessment on “influenza-like illness” (ILI) reported at outpatient clinics in its surveillance network and found a drop in seven of the country’s nine regions (the two exceptions include New York City and Boston). But ISDS collects its own data from a dozen sites around the country, and, as Olson explains, is able to “drill down” and analyze more subtle measurements like fever and severe pneumonia. Their analysis, he says, indicates that the first few weeks of the swine flu outbreak led many people who had respiratory illness and diarrhea—the “worried ill”—to seek medical care. “In normal times, these people would have stayed home sick,” says Olson.
As fear of swine flu has subsided, Olson suspects that fewer of these people have gone to the doctor, leading to drops in reports of ILI that are not accurate indicators of drops in the spread of the novel H1N1 virus that causes swine flu.
Although CDC had also noted increases in Boston and New York City, Olson says his more carefully parsed data show “massive increases” that look “mild” in the CDC regional data. CDC also shows ILI in Seattle dropping down at week 20, which is the opposite of what ISDS finds. Unfortunately, says Olson, no one has systematically collected data about the actual presence of the new H1N1 virus, as it would overtax testing labs. “We’re dealing with so much uncertainty, and we have imperfect measures,” says Olson. “But we need to know what they’re weaknesses are.”
Olson says it’s possible that the epidemic exploded, quickly infected the susceptible population, and will now fade out, but he doubts that, given that New York City has seen the disease peak, drop, and then peak again. “If what New York City has seen in last 10 days is any indication, we’re going to be seeing that everywhere else,” says Olson.
Image Credit: ISDS