North American Butterfly Association

Zebra Swallowtail

The following article recently appeared in the Summer 2000 issue of American Butterflies

Public Panic

over West Nile Virus

         by Michael Gochfeld, Professor of Environmental and Community Medicine at the Robert Wood Johnson Medical School and School of Public Health

NABA members, even those living far from New York, could hardly have missed the news coverage of the great encephalitis outbreak (or "scare"). The discovery in September, 1999 that a new viral encephalitis had appeared in the New York Metropolitan area, fostered a media campaign as well as some ill-advised control efforts. The virus, genetically very similar to the West Nile Virus which is widespread in the Old World, caused more than 50 documented cases of disease and about 7 deaths of humans (all of people over the age of 60), as well as widespread mortality in crows and sporadic mortality in other birds. West Nile Virus belongs to a large group of arthropod-borne viruses (nicknamed ARBO viruses) which includes many viruses that cause more familiar forms of encephalitis as well as dengue fever and yellow fever.

As do other virus, ARBO viruses require a host, in which to live. The primary hosts for the virus are referred to as reservoirs. Viruses are sometimes spread from host to host by other animals, which are then referred to as vectors. For West Nile Virus, mosquitoes are the vectors and birds are the primary hosts, while humans are an alternate host. In areas such as Africa, where the virus has occurred for years, most bird species have developed resistance to the virus through the selective elimination of susceptible individuals and the survival of those who could resist the infection. This evolutionary process is similar to the process by which mosquito populations build up resistance to pesticides, or by which bacteria develop resistance to antibiotics.

Having studies ARBO viruses briefly in Trinidad many years ago, I have a healthy respect for their potential impact. Nonetheless, it is relatively difficult to contract an ARBO virus disease. These diseases are not spread directly from person to person. Rather, a person has to be bitten by an arthropod (usually a mosquito) that has recently bitten an infected animal in whose blood the virus was already circulating.

Not only are humans not the primary host for this virus, but human are probably not even a very good host. Based upon experience with other West Nile Virus epidemics, we know that typically less than one tenth of one percent of people bitten by infected mosquitoes develop any clinical signs of disease, and of those who do develop disease symptoms most do not develop the serious encephalitis manifestations. However, up to 10% of those who actually develop encephalitis may die of the disease.

It is clear that there is something particularly terrifying about encephalitis that conjures up more frightful images than other more common infectious diseases, such as influenza, which each year kills more people than does encephalitis. Thus it is easy to understand why the arrival of this West Nile-like virus caused such consternation and alarm in the media and the public. However, in a city such as New York, where more people die each day of respiratory diseases such as asthma and tuberculosis than died in the entire two month encephalitis "epidemic," it is surprising that there was such a dramatic response among politicians and public health workers who expressed an urgency to eradicate mosquitoes. Particularly fearsome was the prospect that infected birds would soon be migrating and spreading the virus.

Often, when people think of mosquito control, their first thought is pesticide spraying -- what I call "quick Henry -- the Flit" mentality (after a popular ad of 50 years ago). This is obviously what politicians thought of first when West Nile Virus was discovered in New York. Indeed, pesticide spraying can reduce local outbreaks of mosquitoes temporarily, but populations recover quickly and over the long term pesticide-resistance emerges. However, serious mosquito-borne diseases, such as malaria, encephalitis and dengue, continue to occur in many parts of the world, requiring public health officials to develop effective means of reducing risks to humans.

The well-established public health techniques for preventing exposure to mosquito-borne diseases include eliminating mosquito breeding places or treating them to eliminate the larval mosquitoes, the use of screens or netting in homes, the use of mosquito repellants, reducing exposed areas of the body and reducing activities that expose people to mosquitoes. The use of pesticides to eliminate adult mosquitoes is a last resort and should be used only as an interim measure when and where there are large infestations of mosquitoes in close proximity to human populations.

Despite these recommendations from the Centers for Disease Control and the World Health Organization, widespread spraying (termed "broadcast" spraying) was conducted in the New York metropolitan area -- even at a time when mosquito populations were declining naturally due to cool weather. Pesticide advocates asserted, incorrectly, that the pesticides used were innocuous to humans and pets and ignored outcries about the harm to non-target species. Advocates even pointed to the Monarch migration - the largest in years - as evidence that the spray was harmless to butterflies. Although reports of dead birds were widely publicized, no attempt was made to document dead butterflies or other non-target insects.

Television coverage emphasized the spraying and reported where spraying was being conducted, but rarely described all the above-mentioned steps that homeowners could take to reduce their contact with mosquitoes. There were impressive exceptions. Union County, New Jersey took out a full-page advertisement in the Newark Star Ledger explaining why it was not spraying and clarifying the recommendations that empowered readers to act in a way that reduced their risk of mosquito bites. Experts in environmental medicine at our Environmental and Occupational Health Sciences Institute concluded that the media coverage of deaths and spraying actually distracted the public from these effective measures and created the mistaken impression that broadcast spraying was the first choice, the preferred effective solution to West Nile Virus disease, and the only thing that needed to be done.

In some counties, the finding of a single dead crow became a sufficient stimulus for county-wide aerial spraying, ignoring several important features of viral disease ecology, avian ecology and pesticide science, and particularly ignoring the rather significant risks of broadcast pesticide spraying not only to humans but to non-target species that are of economic, aesthetic and ecological importance. Moreover, the investment in spraying diverted scarce funds from other public health problems.

Although malathion (an organophosphate) and the synthetic pyrethroids sumithrin (Anvil) and resmithrin (Scourge) have relatively low toxicity to humans and other mammals and birds, they are not innocuous or harmless. People can become sick from exposure to these pesticides as well as from the so-called "inert ingredients" in which they are applied. This is particularly true when spraying is repeated. Organophosphates such as malathion and pyrethroids reduce the activity of the enzyme acetylcholinesterase which is essential for normal nervous system function. This is the mode of action by which these pesticides kill insects and harm humans. For example, malathion is in the same class of chemicals as the nerve gases such as sarin, and workers who produce malathion or blend it into final products as well as those who apply it, if not well-protected, can suffer agitation, sleepy difficulty and weakness, as well as anxiety, forgetfulness and depression.

Pesticides, even those with relatively low acute toxicity to adults may be more problematic in young children, with immature nervous systems, and in the elderly. It is also a problem for those who have, or believe that they have, unusual sensitivities to pesticides or other chemicals. TV broadcasts cautioned listeners to remain indoors during spraying, but the time of spraying at any one location could not be predicted and many people could not remain indoors all day on the announced day of spraying. Moreover, it is easier to make a house mosquito-proof than spray proof. Indeed, it is well-established that outdoor air pollutants tend to accumulate at higher levels indoors than out.

The reason that my colleagues and I argued against broadcast spraying are 1) in broadcast application most of the spray falls on areas where the likelihood of mosquito-human contact is low. Thus areas with few mosquitoes but many other non-target species are sprayed, as are densely populated residential areas where mosquito populations are low to begin with. 2) It is not made clear to the public that these broad-spectrum insecticides kill many other insects besides mosquitoes. These include economically valuable insects such as honeybees, praying mantids and ladybird beetles as well as conspicuous and attractive species such as butterflies. Such insecticides also destroy innumerable less conspicuous insects that are important components of biodiversity and are the food for birds and small mammals. This, after all, was the message of Rachel Carson's "Silent Spring" published in 1962, that even her detractors recognize as one of the most influential books of the 20th century.

In weighing the risks and benefits of mosquito control, we should also consider the disease itself and the risk to the human population. The media always paired the words "lethal" or "deadly" with "West Nile" or "encephalitis," reinforcing in the public’s mind the danger from the disease. But it would be equally appropriate to characterize West Nile Virus infection as "inapparent," "usually asymptomatic," or "occasionally serious." Seven deaths in a population of over 10 million people over a one month period is certainly tragic, but pales besides the number of deaths from many other diseases that are addressed less aggressively.

The only human epidemic of West Nile Virus infection that has been well-studied occurred in Romania in the late summer of 1996. The U.S. Centers for Disease Control and Prevention, certainly the premier infectious disease control agency in the world, assisted in the evaluation and control of that epidemic and recently published a report in The Lancet, the leading British medical journal.

In that epidemic an estimated 94,000 people were infected by the virus, of whom about 400 developed clinically apparent encephalitis confirmed by virological studies. Fifteen of those people, almost all over the age of 65, died. Thus, even if one is bitten by an infected mosquito, the risk of suffering disease is very low and the risk of dying much lower. Moreover, in Africa where West Nile Virus has been recognized for more than sixty years and where it is widespread, there have been very few human epidemics. In fact, West Nile Virus infection is characterized by its sporadic outbreak in humans, even in areas where it is endemic in birds. This is likwise true of related infections, such as St. Louis encephalitis and Easter equine encephalitis, where 30 or more years may pass between human outbreaks. A knowledge of these numbers is crucial in assessing the risk-risk tradeoffs essential to public health decisions in this area.

Finally, emphasis has focused on dead birds, particularly dead crows. Crows are among the most susceptible species to West Nile Virus in the Old World and are obviously going to be a major reservoir for the virus in the New World. The fact that West Nile Virus does not usually kill the birds that carry it is one piece of evidence supporting the hypothesis that it recently arrived in North America, where our crows and other birds have yet to develop any immunity. However, in order for the disease to be sufficiently widespread in the birds to cause a number of bird deaths, and in order for there to be enough infected birds to infect enough mosquitoes to produce an epidemic, the virus had to be present in the bird population for many months, perhaps even for years. In southern Europe the virus was present for several years before the Romanian outbreak.

This means that the virus is probably already fairly widespread in birds, not just in crows. Birds, including crows, range widely and are not cognizant of county barriers. Thus a bird that dies in county A today may have been infected in county B last week. Thus, to base county-wide control programs on a single dead crow is inappropriate.

It is interesting to contrast the experience and response in New York versus that in New Jersey, which has a statewide network of County Mosquito Control Commissions, established long ago to control New Jersey’s state "bird," the mosquito. New Jersey has extensive experience in conducting surveillance programs for viral encephalitis, including the strategic placing of sentinel chickens in cages in area where mosquitoes are prevalent. The chickens are then regularly tested for virus activity. This surveillance program successfully reduced human and equine infection by Easter equine encephalitis and St. Louis encephalitis even though these viruses continued to be active in bird populations.

Although New Jersey abandoned this program years ago because the yield seemed to be low, it quickly re-instated surveillance when news of West Nile Virus arrived. Moreover, the county Commissions have been vigilant in keeping infestations of mosquitoes in check through the use of Integrated Pest Management which reduces breeding areas in proximity to human activities, introduces natural predators to keep mosquito populations in check and uses judicious spot-applications of larvicides. Spraying to control adult mosquitoes is considered a last resort to be used only when local infestations become a serious nuisance or health threat.

Local health officials, the media, and the public, need to be aware of these facts and need to be reminded lest our vigilance is reduced and serious disease outbreaks occur. The seductiveness of aerial pesticide spraying, devastating to butterflies and other non-target species, and frightening and potentially harmful to many humans, should be the last, not the first, approach to controlling disease spread.

West Nile Virus is likely to surface again this year, heralded by dead crows and great fanfare, and the pressure to use broadcast spraying may be powerful indeed. Improved surveillance will show that the virus occurs over a much wider area than just the New York metropolitan region. Spraying is likely to begin much earlier in the season - June rather than August. If this happens, it is likely that the impact on butterflies and other non-target species will be severe.

Alternatively, public health agencies can focus on the less flashy, but more effective, strategies for reducing mosquito populations. Integrated Pest Management has been very successful in reducing mosquitoes in New Jersey and elsewhere where it has been tried. It can virtually eliminate the risk of an ARBO-virus disease, without eliminating many other valuable or aesthetically pleasing creatures and without jeopardizing human health.


Brevard County, Florida's Health Risk