Centers for Disease Control (CDC). 1981. Pneumo cystis pneumonia—Los Angeles. MMWR Morb Mortal Wkly Rep 30:250–252.
METHODS
Sentinel animals
Underground coal mines are dangerous places to work, in part because toxic, even fatal, levels of carbon monoxide can build up in caves with poor ventilation. Because carbon monoxide is odorless, miners would often keep a bright yellow canary in the mines with them; if the canary remained alive, no carbon monoxide was present, but if the canary died, the miners were forewarned. The “canary in a coal mine” is perhaps the best-known case of using animals as sentinels or harbingers, though this approach has been used to identify viral infections in the wild as well. For ex ample, sentinel species, such as monkeys, are placed inside cages near the entrance to caves where bats reside. Epidemiologists periodically check on the health of the monkey; if the animal became sick, not only would this indicate a health concern, but virus could be isolated from the affected host to determine the nature of the pathogen, and perhaps to learn something about how it was spread. Methods that do not require the incarceration of the sentinel organism are also in use, including collection of feces and urine in the wild for subsequent laboratory analysis. In fact, in 1947, scientists conducting routine surveillance for yellow fever virus in the Zika forest of Uganda recovered a novel virus, later named Zika virus.
Publications and websites help to distribute consistent and timely information to health care workers across the globe. The Morbidity and Mortality Weekly Report, published by the CDC, provides a central clearinghouse for health care providers in the United States to communicate individual cases of infectious diseases or to report unusual observations. ProMED (Program for Monitoring Emerging Diseases), sponsored by the International Society for Infectious Diseases, is a world-wide effort to promote communication among members of the international infectious disease community. Reporting of individual cases, when considered by epidemiologists in the aggregate, may catch an epidemic in its earliest days, when intervention is most effective.
More-informal “crowdsourced” approaches have recently gained attention for their power to share data, educate the public, and rapidly identify a potential outbreak. Real-time data-gathering tools, such as Google Flu Trends and Google Dengue Trends, are Web-based applications that survey search queries from more than 25 countries to predict epidemics. The predictions made from these applications have been generally consistent with more traditional surveillance data-gathering approaches. The innovative use of keyword collection to monitor viral outbreaks underscores how collaboration between distinct fields (e.g., epidemiology and search engine design) can lead to creative ways to detect incipient epidemics. Social media monitoring also is an excellent way to gauge the impact of public education efforts in understanding viral infections (Fig. 1.9).
Figure 1.9 Twitter as a tool in viral epidemiology. Between May 1 and December 31, 2009, the relative proportion of tweets using “H1N1” increased in an almost linear fashion, indicating a gradual adoption of the WHO-recommended H1N1 terminology as opposed to “swine flu.” Blue = use of the term “swine flu”; red = use of the term “H1N1”; green = combined use of “swine flu” and “H1N1.” Adapted from Chew C, Eysenbach G. 2010. PLoS One 5:e14118, under license CC BY 4.0. © 2010 Chew et al.
Network Theory and Practical Applications
How many people do you encounter each day in conversation, in the classroom, or passing on the street? Dozens to hundreds of interactions—some long-lasting, others fleeting—can occur in a day, and each of these individuals has a personal “network” as well. The science of social networks as a tool to understanding spread of pathogens within communities has revolutionized epidemiology. Such networks define potential transmission routes; for example, contact tracing identifies likely transmission network connections from known infected cases and then applies this information to treat or contain their contacts, thereby reducing the spread of infection. Contact tracing is a highly effective public health tool, as it uses the underlying transmission dynamics to target control efforts and does not rely on a detailed understanding of the etiology of the infection.
Network analysis has been used most effectively when considering viruses that are spread via sexual activity. In contrast to airborne infections, sexually transmitted viruses, such as human immunodeficiency virus type 1 and some herpesviruses, have transmission routes that should be easily identified, provided one can recall recent sexual partners. In these cases, an individual identifies their sexual partners over a given period, these partners are then contacted and asked for their partners, and so on; this process is known as snowball sampling, and is used by many public health officials to contact individuals who may be at risk for infection.
Developing methods to trace viruses’ spread via aerosols is far more challenging. One successful strategy took advantage of the fact that most people carry mobile phones; in this study, data were collected using Bluetooth to sense other mobile phones in the vicinity. These data gave a highly detailed account of an individual’s behavior and contact patterns, and allowed highly detailed interaction maps to be developed.
Parameters That Govern the Ability of a Virus to Infect a Population
One often hears that a virus is “going around,” and such comments usually correlate with particular times of year (flu season). The seasonal appearance of some viruses, especially those that cause respiratory and gastrointestinal disease, raises the question of what parameters facilitate seasonal or temporal spread in a population. This question is relevant both to viruses that cause widespread epidemics and to more mundane infections, such as the common cold. Identifying the variables associated with increased risk in a population has obvious value in clinical and educational efforts to prevent or limit outbreaks. As discussed below, multiple aspects of both the host and the environment contribute to maintaining a virus in a community.
Geography and Population Density
Some viruses are found only in specific geographical locations. The regional occurrence of viral infections may be due to the restriction of a vector or animal reservoir to a limited area. For example, most insect vectors are restricted to a specific region or ecosystem; unless this vector “escapes” its natural habitat, the viruses that it harbors will also be geographically constrained. Changes in migration routes or territory of a reservoir species may therefore influence the distribution of a virus and lead to new interactions with other species, increasing the risk of zoonotic transmission. A striking example of how a vector can change the location in which a virus is found is provided by the global spread