Experiments in Epidemiology

Sometime during the end of March or early April 2009, a 5-year-old Mexican boy living down the road from a pig farm got sick. By mid-April, news reports blasted throughout the world press of a new type of influenza outbreak in Mexico City—where half the population of the boy's town commutes to work.

Within weeks, over 80 local people were dead from complications associated with swine flu infection. All schools and public parks were closed. Sunday church services were cancelled. Antiviral drugs were distributed to residents. Facial masks were disseminated to all citizens living within the vast metropolitan area in one of the world's most populated cities. This strain of influenza type A, 2009 H1N1, is remarkable in that it is a virus never before encountered by human kind.

As epidemiologists work tirelessly implementing and tweaking control strategies with each new country to which it spreads, an opportunity arises for prospective, population-level disease transmission studies.

Models for disease transmission facilitate our understanding of how infections spread, taking into account levels of genetic diversity, population density, and mechanisms for their control. New outbreaks such as swine flu provide valuable data with which to test these models and enhance their accuracy, yet scientists must sit and wait until these new data become available. Without natural outbreaks of disease, epidemiological modelers are unable to test their hypotheses.

The field would benefit greatly from the development of a new, financially viable, manipulable model system that mimics human populations with respect to genetic diversity, population density, and disease susceptibility. Ideally, this system should also present analogous behavioral responses to disease and should be amenable to chemical treatments.

The honey bee is a strong candidate.

Each colony of honey bees contains a single queen that has mated with as many as 30 unrelated males. That is as many as 30 patrilines within a single family. Apiaries can be organized to have anywhere from a single hive to hundreds of hives. That's the equivalent of everything from a single family up to multiple major cities. Honey bee hives are susceptible to—and are naturally associated with—arthropod, bacterial, fungal, and viral pathogens. Genetic diversity, population density, and disease susceptibility are all manipulable with artificial insemination, apiary management, and pathogen cultivation. And manipulable at a fraction of the cost necessary for similar experiments with rodents (which provide none of these advantages).

Like humans, honey bees display a variety of disease defense mechanisms. They mount a fever when infected. They groom. They dispose of the dead. We may know more about hygienic behavior in honey bees than in any other potential model system. Like medical doctors, apiculturists have developed disease management strategies. We have chemical treatments for some diseases. As a whole, not only can we manipulate genetic and population characteristics, but the system mimics humans with respect to behavioral responses and the availability of chemical preventatives and treatments.

It is unlikely that the physiological responses of honey bees replicate those of humans. But we are not arguing that honey bee research should replace other methods; rather, disease research would complement existing methods for epidemiological study. In this one experimentally manipulable, behaviorally rich system, we can prospectively study disease transmission and disease suppression within and across multiple populations differing in population density, genetic diversity, and behavioral response in ways directly analogous to those in human populations. And we can do it without breaking the bank.

Philip T. Starks and Noah Wilson-Rich are at the Department of Biology, Tufts University.

As brilliant a hypothesis as this may sound, epidemiologists, would then need to change paradigms as well as basic concepts and methodologies in population-based studies, even with a full understanding of disease pathology and vector biology, transmission patterns. These biases in my opinion, can be expected from the fundamental fact that humans and bees share just one reliable attribute;'social creatures', which simply explains the bees' ability to organize into colonies with defined demographic patterns. However, the burden of risk factors and determinants of disease spread are largely dependent on behaviour, life styles, environmental and genetic factors, all of which are manipulable only by humans.

This article stated that the boy lived down the road from a hog farm. It is my understanding that there is no proven link between this hog farm nor any other hog farm and the boy who got sick, nor has there been any sick animals or workers on this hog farm or any other hog farms in the area.
On NBC news it was ststed that the hog farm is 8 miles from where the boy lives and that there is no link between him and this or any other hog farm. I do not know if this remains true, but if so, stories and comments like this only serve to mislead the public. I understand that historically flu epidemics start in areas where hogs or poultry are in close proximity to humans, but I think that we need to be sure that we do not blame a particular farm or industry as the source of a specific epidemic unless we are sure where it came from. At this point I will disclose that in addition to working in a research lab, I farm.

Bees are having a difficult time these days for reasons that are not entirely clear. It would be difficult to get a good control group. I sure would not want to get them sick and have them flying around with new viruses. Ants are much easier to keep as a colony and also offer some advantages of observable social behavior and lower risk factors. We really need bees, but ants are not critical to our food crops to the same extent.

I would call this article out of the box thinking, except for the fact that bees live in boxes. It is an interesting topic, but I am not seeing any practical proposals in the article, or a correlation to the swine flu.

Since we have little handle on the meaning of genetic diversity at present, (linking to iunfluenza) I'm not sure what this would mean. It seems to me to be putting the cart before the horse.