In the fall of 1999 the mosquito-borne West Nile virus-a virus never be seen in this hemisphere-killed seven people in the New York City area made dozens more very sick. It was initially misidentified as a differen related mosquito-borne virus often found in the United States. The incorrect diagnosis did not significantly affect the appropriateness of spraying and other mosquito control steps quickly begun in response. so, the outbreak can serve as a source of lessons about how public hea officials can be better prepared in responding to potential crises invol uncertain causes. And although the outbreak is considered to have bee natural occurrence-possibly introduced by international travelers, migrating birds, or mosquitoes accidentally brought from abroad—it c also provide lessons about detecting and responding to an act of biolo terrorism.

You asked us to review this outbreak and the response to it. We focused efforts on the following:

• establishing a thorough chronological account of the significant eve and communications that occurred, from doctors and others who fi saw the symptoms and from the officials mounting a response, and

identifying lessons learned for public health and bioterrorism preparedness.

interviewed key officials, reviewed existing studies and reports,
sembled a detailed chronology of what occurred, and discussed the
plications of our findings with a wide range of agency and area officials.
asked officials to provide documentation to the extent possible for key
cisions or events; however, to some extent the chronology of events is
sed on officials' recollections of the dates that particular events
curred. Appendix I describes our methodology in more detail. We
nducted our work from May through August 2000 in accordance with
nerally accepted government auditing standards.

e analysis of the West Nile virus outbreak began-and for several weeks
ntinued-as two separate investigations: one of sick people, the other of
ing birds. On the human side, the investigation began quickly after a
ysician at a local hospital reported the first cases, and the original
gnosis, while incorrect, led to prompt mosquito control actions by New
rk City officials. The ongoing investigation involved the combined efforts
many people in public health agencies and research laboratories at all
els of government. A consensus that the bird and human outbreaks were
ked, which was a key to identifying the correct source, took time to
velop and was initially dismissed by many involved in the investigation.
en the bird and human investigations converged several weeks after the
tial diagnosis, and after laboratory research was launched independently
several of the participants to explore other possible causes, the link was
de and the virus was correctly identified.

y lessons that emerge from the investigation and response to this tbreak are as follows:

The local disease surveillance and response system is critical. In public
health, surveillance is the ongoing collection, analysis, and
interpretation of health-related data. In this outbreak, many aspects of
the local surveillance system worked well, in that the outbreak was
quickly spotted and immediately investigated. Assessments of the
infrastructure for responding to outbreaks suggest that surveillance
networks in many other locations may not be as well prepared.
Better communication is needed among public health agencies. As the
investigation grew, lines of communication and decision-making were
often unclear, and efforts to keep everyone informed were awkward

Background

The Viruses: W

St. Louis Encep

1

(such as conference calls that lasted for hours and involved dozens people). Many officials reported problems in this area.

• Links between public and animal health agencies are becoming mo
important. Many emerging diseases, including West Nile, affect bot
animals and humans. So do many viruses or other disease-causing
agents that might be used in bioterrorist attacks. The length of time
took to connect the bird and human outbreaks of the West Nile vir
signals a need for better coordination among public and animal hea
agencies.

• Ensuring adequate laboratory capabilities is essential. Even though
was a relatively small outbreak, it strained resources for several mo
Officials said there is a need for broadening laboratory capabilities
ensuring adequate staffing and expertise, and improving the ability
deal with surges in testing needs. These concerns point out the
importance of ensuring adequate networks between public health a
other types of laboratories, and of completing assessments of what
laboratory capacity is needed and what capacity is available.
• Because a bioterrorist event could look like a natural outbreak,
bioterrorism preparedness rests in large part on public health
preparedness. While the West Nile virus outbreak is considered by
Centers for Disease Control and Prevention (CDC) and others to ha
been a naturally occurring event, at one point there was speculation
it might have had an unnatural (bioterrorist) origin. The ensuing
investigation and post-outbreak assessments illustrate the challeng
identifying the source of an outbreak, supporting public health offic
views that public health preparedness is a key element of bioterror
preparedness.

West Nile and phalitis

Identified in 1937 and named after the Ugandan province where its discovery took place, West Nile virus has a widespread distribution in Africa, West Asia, and the Middle East, occasionally causing epidemics in Europe. Many people infected with the virus do not become ill or sh symptoms, and even when they do, symptoms may be limited to a headache, sore throat, backache, or fatigue. Because no effective antiv drugs have been discovered, treatment for those who do become serio ill can only attempt to address symptoms such as swelling of the brain (encephalitis) and other complications such as bacterial pneumonia.

ality rates the percentage of people with confirmed infections who e died-have ranged from 3 to 15 percent for West Nile and are highest he elderly.

‣ virus that was originally misidentified as the cause of the New York break is called St. Louis virus. Both West Nile and St. Louis encephalitis ises are in a group called "flaviviruses," and can be spread when squitoes bite birds (often a natural host for the virus), acquire the virus, I then pass it on to humans (see fig. 1). St. Louis encephalitis is found in ure through much of the lower 48 states and is the most common squito-borne virus causing outbreaks of human disease in the United tes. About 30 confirmed cases occur on average each year during nonbreak years. St. Louis encephalitis is also similar to West Nile in that st people infected with it show no symptoms, but for those who become iously ill, no effective antiviral drugs are available. St. Louis encephalitis a slightly higher fatality rate than West Nile, ranging from 3 to 30 cent of confirmed cases.

pid and accurate diagnosis of disease outbreaks is essential for many sons. It can help contain an outbreak quickly by allowing health officials mplement appropriate control or prevention measures and provide the st effective treatment for those who are affected. Rapid and accurate gnosis is essential not only for the public at large, but also for health e workers and others who work with patients and laboratory samples. curate diagnosis is also important in providing information that could

help determine whether the outbreak could have been deliberateof bioterrorism. Public health officials use the term "surveillance" t denote the ongoing effort to collect, analyze, and interpret health-re data so that public health actions can be planned, implemented, and evaluated.

Local health personnel are likely to be the front line of response. Lo state health departments might be the first to recognize unusual pat illnesses. For example, an epidemiologist (a health official trained t investigate diseases of unknown origin) in a city health department receive phone calls from nurses, doctors, or emergency room perso about increasing numbers of patients with similar symptoms. If the problem is thought to be widespread or suspicious in origin, the loc health department is likely to involve the state health department, w responsible for statewide surveillance and investigations involving n locations. The local and state response may also involve emergency management personnel. Current protocols recommend that law enforcement officials be notified if a case or series of cases have a suspicious origin.

Local, state, and federal laboratories also play a vital role. Initially, t may be to determine whether the unusual cases have the same path (the specific causative agent for the disease), and if so, to identify it an outbreak is established, laboratories may be called upon to test s such as blood or spinal fluid from persons with similar symptoms, t determine who has the illness and the extent of the problem.

At the federal level, CDC, an agency of the Department of Health an Human Services (HHS), is available upon request to help state and 1 officials investigate the nature and origin of disease outbreaks. For example, CDC maintains several laboratories that identify unusual exotic viruses and other pathogens when other laboratories are una do so. One such laboratory, at the Division of Vector-Borne Infectio Diseases in Fort Collins, Colorado, deals with viral and bacterial dis transmitted by vectors such as mosquitoes and ticks. It is part of CD National Center for Infectious Diseases. Besides providing laborato services, this division also develops ways to diagnose vector-borne pathogens more quickly and helps develop and evaluate approaches preventing and controlling outbreaks.