Math May Distinguish Flukes from Flu Epidemic Outbreaks


Home / Math May Distinguish Flukes from Flu Epidemic Outbreaks

Domestic fowl may harbor Avian Flu : Image by Martin Cathrae

A research paper released today demonstrates a quicker and more cost-effective way for clinical research organizations to determine whether a new disease can be spread from person to person, or whether it only infects people exposed to sick livestock. This should save time and money in correctly evaluating the threat of a new epidemic, such as swine flu or the dreaded avian flu.

This article uses influenza as the example of a disease outbreak, but the math could apply to other animal-borne diseases as well.

Flu Outbreak: Need to Quickly Assess the Risk

When a new strain of influenza virus moves from animal hosts to humans, it may be named “avian flu” or “swine flu” depending on the preferred host animal. If the new influenza virus cannot be transmitted from human to human, there is no risk of a flu epidemic. Yet, with some new strains, one person can infect others.

Tracing the source of infections is costly and time-consuming, but very important in determining the correct medical response. Is a small group at risk, or might the disease spread as an epidemic, or even a pandemic? What is research finding about assessing the risk that a new disease may become an epidemic? Simple math, based on statistical sampling, has provided a new option for evaluation.

In Using Routine Surveillance Data to…, Dr Simon Cauchemez (et al) report a mathematical breakthrough. By analyzing the statistics only from specific cases in an outbreak, it is possible to determine the rate of transmission among humans as reliably as tracing the source of each patient’s infection. The information gathered by this approach can help health officials come to a quick decision: Whether to launch a broad assault against the start of a potential epidemic, or to contain a limited situation affecting mainly people who come in close contact with the infected animals.

Outbreak Evaluation: When can Routine Sentinel Surveillance Say “It’s not an Epidemic”?

The first key public health decision in the face of the outbreak of a new disease, is whether to launch an outbreak investigation.

If it is clear that “0<R<<1” (R, the reproduction number for humans infecting humans, is above zero but much less than one), then the risk of human-to-human transmission is low, and there is no need for an outbreak investigation. Instead, the task is to protect the people exposed to the animal reservoir.

Routine sentinel surveillance estimates the value of ‘G‘, the proportion of cases that came from the original reservoir of cases, by learning whether the patient had been exposed to the animal reservoir within the incubation period for the disease. For example, a number of people had fallen ill shortly after visiting an agricultural fair. With that common factor, these cases were found to be the heads of very short chains of transmission. In this example, “0<<G<1”: nearly every patient had been infected from the animal reservoir and the disease was not being transmitted from one person to another.

The research paper demonstrates that, if routine sentinel surveillance estimates ‘G‘ to be close to one, then “R=1-G” reliably indicates no need for an outbreak investigation. ‘R‘ will be close to zero; and there are few human-to-human transmissions.

Leave a Comment