Symposium sponsored by:

Johns Hopkins Center for Civilian Biodefense Studies

Department of Health and Human Services

Infectious Diseases Society of America

Outbreak Surveillance and Management at the State and Local Level: Current Realities
Marcelle Layton, MD, MPH

DR. HENDERSON: Our last speaker in this group is Dr. Marcie Layton, who is the Director of Communicable Diseases of the New York City Department of Health, and like it or not, a national expert on West Nile virus disease.

(Laughter.)

In addition to her concerns with West Nile, she is also responsible for all other infectious disease activities in the city and for disaster planning for the threat of bioterrorism. She's a medical graduate of Duke University with clinical and epidemiological training in infectious diseases at Syracuse, Yale, and CDC, and she will speak on the realities of outbreak surveillance and management at the state and local level. Marcelle.

(Applause.)

DR. LAYTON: Thanks. As was pointed out earlier, the technology gap exists at our local health department level, and to prove that point, I've been told that I'm the first speaker to use Kodachromes as opposed to Power Point. So on that note, can I have the first slide?

I've been impressed at how many times West Nile has come up this morning. I was last told about ten times it's been mentioned with respect to bioterrorism preparedness, and what I'm going to try and do over the next 20 minutes is rapidly overview our recognition and response to the West Nile outbreak in New York City last summer, and although it was a big story, it was actually a relatively small outbreak, and it doesn't address all of the issues that would arise during bioterrorism. But I do think that there are some lessons to be learned with respect to how our existing public health infrastructure can be strengthened to improve our nation's response for bioterrorism.

Just to give you some background, viral meningitis and encephalitis are two of 70 reportable conditions in New York City and New York State, but unlike many of the diseases that we track, the diagnosis is primarily clinical based on a characteristic syndrome of signs and symptoms and often not a positive laboratory test because full viral laboratory testing is not done. Because of that we're very dependent on physician reporting, and unfortunately, like many areas of the country, physician reporting is incomplete. In New York City we averaged about nine cases of encephalitis per year prior to this outbreak and 170 of viral meningitis.

We recognize that data is not complete, but despite those limitations, there is some value to looking at our historical data. It's hard for me to see, but every year in August and September, we do see a seasonal peak primarily in viral meningitis. This is primarily occurring among young children, and when there is a diagnosis, it's primarily enteroviruses, a family of viruses that's transmitted person to person through infected stool. So when this outbreak was first detected last summer, that's what we expected it to be, common things being common. The other thing to notice on this slide is the 1999 data, which was our provisional data that was in the database the day the outbreak was first recognized, and I show this just to make the point that based on routine surveillance data alone, there really was no indication that anything unusual was going on in New York City based on historical trends.

At the city Health Department over the past several years we've been actively promoting the importance of physician reporting of unusual disease clusters or manifestations, both because of our concerns about emerging infectious diseases, but also because of our concerns about bioterrorism. And there's really no better example of how powerful a single physician's report can be than this particular outbreak. I was actually the one on the receiving end of this phone call last August 23rd, a day I'll never forget, when I was contacted by Dr. Debbie Asmis, an infectious disease physician in northern Queens, who was concerned about two patients that she was seeing both of whom she thought probably had viral encephalitis, but one of them more than the other had severe muscle weakness associated with their presentation. And she was also concerned about botulism, and the neurologist seeing the case was also concerned about the possibility of Guiamme Barre.

She presented the cases to me, and for a number of reasons, I did not think these patients had botulism, did not recommend botulism testing, but encouraged her to send spinal fluid and sera up to our state health department lab for viral testing. We sent staff out to review these charts and kept in contact with her, and on August 27th, she called again because she actually only had a third case at this point, but all three cases had developed, in addition to the viral encephalitis symptoms, severe muscle weakness, and while I was talking to her on the phone, the neurologist at the hospital just happened to walk in, heard that she was talking to the health department, and mentioned that he was seeing a very similar case at a nearby community hospital.

So when I got off the phone, that phone call, I had four cases of an unusual encephalitis, when normally I only see nine cases city-wide in a year. So by definition, this required an outbreak investigation. This was more than expected, and it is usually my bad luck at the Health Department. August 27th was a Friday, and this phone call came around 4:30 in the afternoon. So a colleague of mine and myself spent the weekend at these two community hospitals, and through some active case finding, by Sunday morning had identified eight suspect cases, all in the same community.

Just to give you a sense of my initial concern and why it concerned us, these were all relatively healthy, older adults living at home in a small area of the city, a 16 square mile area of northern Queens, and their clinical presentations were basically identical. They all had a febrile illness associated with some mild GI symptoms, followed by the onset of altered mental status, and seven of the eight had this very severe, diffuse muscle weakness to the point of paralysis in over half of the patients, and it was that unusual characteristic that concerned us the most because it didn't really fit any known viral cause of encephalitis. And the laboratory parameters were very suggestive of an infectious and specifically a viral etiology as opposed to a bacteria.

For those who don't know New York City, this is the White Stone Peninsula of northern Queens, the same area blown up closer to me, and it just shows how close these patients lived to each other. None of them lived more than a mile or two from each other.

That weekend at the hospital, we played medical detective, interviewing these patients' families extensively to try and figure out what they had in common, if they had been in the same place at the same time or if they had any common associations. The first thing we were able to tell is that these patients did not know each other prior to this hospitalization, and then no matter what we asked, we really couldn't find any common links. They had not traveled anywhere in common, had not attended a common social event, eaten at the same restaurant or shopped at the same store.

As I mentioned, enteroviruses are what we'd expect to be in New York City at this time of year. Those are viruses that are transmitted person to person, and we were struck by the absence of any secondary illness among these patients' immediate household and social circle, and also since enteroviruses are carried by children in any community, that none of these patients had had recent contact with young children.

The only thing we could find in common was that all of these patients when we asked how they spent their days in the week or two before illness reportedly spent time outdoors in their backyards in their neighborhoods, especially in the evening hours doing things like gardening or smoking on their porch deck.

Well, that first week of the investigation was the most difficult and as far as raising acid levels in our stomach because we really weren't sure of what was going on. And similar to what we do for many outbreaks, our first step was notifying those who knew more than we did, and that Sunday morning, we actually notified both the arboviral and the enteroviral experts at CDC to get expert opinions.

We also notified our neighboring public health agencies to see if anyone was seeing similar cases, and they were not. Everyone we talked to agreed with us that the most important thing was getting a diagnosis. The same would be true for a suspect bioterrorist event. We really can't make public health or medical recommendations until we have a specific diagnosis.

So we continued to prioritize, obtaining spinal fluid and sera and sending it to our state health department lab for viral testing. We also began much more active surveillance city-wide by alerting the New York City medical community both by broadcast fax and E-mail. That was passive on our part. We just described this initial cluster and asked physicians to call us if they were seeing similar cases in other parts of the city because we were concerned that although the outbreak was recognized in northern Queens, that's only because one physician just happened to call from that part of town. And because we didn't want to completely trust physicians to call us, we also began very active surveillance by having our staff call infectious disease and neurology doctors at the 70 hospitals in New York City to see if anyone else was seeing similar cases.

And as this week unfolded, we were actually up to about 30 suspect cases by the end of the week, and because the other calls of encephalitis that tends to occur in outbreaks in the summertime are arboviruses or mosquito borne viruses, we did send an entomology team -- we actually had to borrow an entomologist from the Museum of Natural History because we didn't have one on staff at the time. We've since hired her -- but --

(Laughter.)

-- we used her and our staff to assess the area for potential mosquito breeding sites around where each of these patients lived, and it was the evening that this team came back and reported their findings that we became much more convinced that we were dealing with an arbovirus because patient after patient, they described something very concerning in the area around where each of these patients lived, whether it was a pile of old tires or a partially excavated swimming pool.

Well, our suspicions were first confirmed actually on September 2nd, when we got a preliminary report from our state health department lab that an antibody test in their lab was most consistent with what at the time was thought to be St. Louis encephalitis. We did not go public that day because we wanted to wait for CDC confirmation, but we used those 24 hours to develop contingency plans for mosquito control in the event that CDC confirmed it, which they did on September 3rd at two o'clock in the afternoon, and again, my bad luck. September 3rd was the Friday before Labor Day weekend. So I was very glad for that 24 hours to be able to mobilize a response. It would have been very difficult for us to do at the start of a three-day holiday weekend, and we were able to implement mosquito control in northern Queens over Labor Day weekend.

We continued active surveillance, had our first positive case in Brooklyn less than a week later, and it was that case, along with numerous suspect cases throughout the city that led to the unprecedented decision to begin mosquito control city-wide.

The identification of a human outbreak of encephalitis in the late summer requires, the public health response requires rapid mobilization of mosquito control measures. This was a public health emergency, and similar to other public health emergencies required a great deal of ongoing communication and coordination between a large number of agencies at the local, state, and federal level. This was done both by group E-mails and conference calls.

New York City, as I mentioned, did not have an existing mosquito surveillance and control program. So it required us to pull together emergency contracts for mosquito surveillance and control. We also got a chance to practice our plans at least on a small scale for mass prophylaxis. Jerry Hauer purchased 400,000 cans of mosquito repellant, and we distributed them free of charge through various city venues because we were concerned about a run on city pharmacies when those headlines hit.

The biggest challenge was public health education. It's always going to be the biggest challenge. This was easiest that first weekend when the outbreak was localized to one small part of town. We were able to focus our efforts in northern Queens and actually send health department staff, hundreds of them, door to door with information sheets translated into eight different languages about the virus itself, the mosquito control protective measures that were needed, and also address concerns about the pesticides. But when we went city-wide, we really needed to rely on a public hot line. That was stationed at our Office of Emergency Management, but staffed by the health department 24 hours a day, ran for almost seven weeks, received 150,000 calls during that time, was the busiest hot line the city has ever set up, and in setting this up actually at its peak, we had 75 staff there per shift, including physicians and toxicologists as back-up, required a great deal of training of staff and oversight to maintain that.

Obviously there was a need to communicate and coordinate with the media. We did this through daily press conferences. They were almost always attended by Mayor Guiliani and our Commissioners of Emergency Management and Health.

And then similar to a bioterrorist event, there was a need to have direct communication between us as a public health agency and the medical community, to be providing more specific medical information, and we continued to use our broadcast fax and E-mail system that goes to several different offices in each hospital, and also set up a separate hot line for providers.

There also was a need to coordinate with a number of agencies. This is our Department of Sanitation helping to clean up a tire pile. We worked with our police department using their helicopters to map an amazing number of backyard swimming pools that existed in this part of the city, many of which had not been well maintained because of the drought that occurred last summer or two summers ago, and we would identify these by air and then send health department sanitarians door to door to clean them up, and this is just one of the many deep distribution centers that was set up with EMS.

Well, what was already unexpected, having SLE in your city, became unprecedented with the recognition of West Nile, and the recognition of West Nile was primarily due to the recognition that there was a simultaneous bird die off, and, no, this bird flying around isn't a prop that I brought for my talk, but it primarily last year and this year affected crows, and that this outbreak or epizootic was due to the same virus. And it's really thanks to the veterinary pathologists who recognized that these birds were also dying of encephalitis and sent these tissues for viral testing, leading to the recognition of West Nile as the cause of both the human and avian outbreaks.

There are a number of reasons for the delayed recognition of West Nile, and I'm not going to go into detail about them. Partly the cross-reactivity among these, this close cousins of West Nile and St. Louis, which I'm very grateful for, and also the difficulty that neither CDC nor the state were ever able to culture the virus from human specimens. But the biggest error in retrospect is that the bird die off was initially felt to be unrelated since arboviruses do not normally kill birds. It's their natural reservoir host. It was unprecedented to have either West Nile or SLE cause simultaneous outbreaks of human and avian disease.

Similar to what we would need to do for a bioterrorist event, our response, once we identified the cause, was to define the geography of the outbreak and also to continue active surveillance to evaluate the need for control measures. In New York City, we implemented a multi-faceted surveillance program with respect to human surveillance. We did both passive and active surveillance, passive surveillance through those almost weekly fax and E-mail updates that we sent to the city medical community that updated them on what was going on with the outbreak, but also reminded them out to report and not to send specimens to us for testing.

We continued to do active surveillance, actually expanded it during the peak of the outbreak, and had our staff calling nine different specialists in each of our 70 hospitals, adult and pediatric, general wards, infectious disease, neurology and intensive care, to make sure that we were hearing about every case.

We also had two back-up systems, a laboratory based system where we collected spinal fluid and also a retrospective system collecting hospital discharge data to be sure that we picked up cases that might not be reported by clinicians. And our environmental staff at the same time needed to implement both mosquito and bird surveillance systems.

We tried to provide physicians with clear criteria for reporting. Obviously the same thing would be needed for bioterrorism. For West Nile, we emphasized that a syndrome of the initial Queens cluster encephalitis was severe muscle weakness; also looked for any case of encephalitis, fever with paralysis, or the milder syndrome of viral meningitis.

This slide shows, as is always true with active surveillance, that you always pick up more cases than you do passively. Actually during the three months of this investigation, we picked up three times as many cases as we normally hear about in a year. Each suspect case needed to be evaluated and prioritized for laboratory testing, and then similar to what we would do for a bioterrorist event, all positive cases were interviewed to determine potential sites of exposure. Last year's outbreak, there were 62 cases primarily in the New York area, and there were seven deaths, and we actually worked closely with our Medical Examiner's Office to insure autopsies in all cases. I can't see how well that slide shows, but this just shows that the human outbreak was most intense in the area of the city where it was first recognized in northern Queens.

And as I mentioned, the avian outbreak was much larger than the human outbreak, actually extended into both New Jersey and Connecticut, and there was a horse outbreak in eastern Long Island that was not recognized as being due to West Nile until some time in October. Well, everyone involved in last year's outbreak agrees that this represented a natural introduction of a virus into New York City, with the most likely explanations being that an infected bird, infected human, or possibly infected mosquitoes themselves traveled probably by airplane or potentially by ship.

However, the possibility of bioterrorism needed to be considered. It would have been negligent and also naive of us not to consider bioterrorism, as there were characteristics of this outbreak that met the criteria which should always prompt public health authorities to think bioterrorism. This was a cluster of unexplained serious illness with unusual manifestations, that severe muscle weakness that had not been reported before for either St. Louis or West Nile as a predominant feature. Initially there was no obvious common exposure. It was an unusual location for an arbovirual outbreak. We had not had an arbovirus in New York City in over 100 years, since the yellow fever epidemics back in the 1800s, and there was also a simultaneous bird outbreak.

Although none of us involved, again, think that this was bioterrorism, the outbreak itself and our response raised some concerns regarding the implications for our nation's preparedness for bioterrorism, and in New York City, there's no question that our four years or more than four years of planning for a bioterrorist event definitely helped us both recognize and respond to this outbreak, but there are lessons to be learned. And I think, again, the biggest lesson to be learned with respect to both emerging infectious diseases and bioterrorism is the need for all of us, both the clinical and public health community to remain open minded to the unexpected as initially, again, those avian deaths were thought to be unrelated.

On a positive note, West Nile was detected because an astute infectious disease doctor saw something unusual and did the right thing. She reported it to us, and a responsive public health agency acted on that call, and public health agencies can't be aware of what's going on in their jurisdictions unless someone calls us, and we need to recognize that and actively foster strong relationships with our local medical communities. During last year's outbreak, the veterinary investigation unfortunately occurred largely in isolation from the human investigation and wasn't brought to our attention at the city health department, and we were the lead agency investigating the human outbreak until two days before the West Nile virus was identified, and that taught me that I also need to engage non-traditional public health partners, clinical veterinarians, wildlife experts and, as others have mentioned, epizootics where animal outbreaks may be early warning signs for human disease outbreaks, and we all need to recognize that many potential BT agents are also zoonotic diseases, such as anthrax and smallpox.

And although I'm proud that we detected the outbreak as soon as we did, I also recognize how lucky we were. At the time that our epi. investigation started last year, there were 19 patients hospitalized with West Nile in New York City hospitals, and 15 or 80 percent of them had not yet been reported to us. So if this one ID doctor hadn't reported the initial cluster, it's unclear when or if this outbreak might have been detected.

On a similar note, we, again, first became aware of the crow die-offs on Labor Day weekend when our hot line began to receive hundreds of calls, but we didn't hear that these birds were actually dying of encephalitis until three weeks later, and in retrospect, by going back and looking at some community newspaper reports in northern Queens, we found out or we now know that there were a lot of reports about bird die-offs in that area as early as late June, five weeks before the human outbreak began. So if the veterinary investigation had begun earlier and had been more aggressively investigated earlier, it's possible that the human outbreak might have been mitigated or even averted completely.

As far as surveillance, ongoing surveillance, similar to what would be expected of us as a public health agency during a bioterrorist event, there was a tremendous demand for accurate, up to date information obviously to guide control decisions and also to inform key partners, and this required the capacity on our part to simultaneously track and map a great deal of data, human, avian, and mosquito reports, both positive and negative, laboratory results, and there's no question that this was our biggest challenge, data management. It required the establishment of a sophisticated database to more easily track and analyze our surveillance data, and unfortunately we really didn't have a significant amount of information technology expertise in house. This was not easily available to us, and we actually relied on one EIS officer, who fortunately was assigned to New York City at the time to set this up.

During the '99 outbreak in the tri-state area, we all relied on a single lab at CDC, which happened to be in Colorado, so half a country away, to do all of our testing. They tested 2,000 specimens over this three-month investigation, and were open and actively working seven days a week, and this illustrates the need to both improve and regionalize our public health laboratory capacity so that we're not in this situation again. CDC and the Association of Public Health Labs has recognized this with the establishment of a national laboratory response network, and I think there's been significant strides in enhancing our diagnostic capability for potential BT agents at the state and local level.

A number of speakers had mentioned the importance of surge capacity, and during the West Nile outbreak, our response at the communicable disease program was maintained from basically the beginning or even late August through the end of November. Our staff worked more than 12-hour days seven days a week for most of that time, and we did get some help. We got some reassigned staff from other internal programs, TB, STD, and AIDS, and also several CDC staff were sent to assist. No one stayed for more than three weeks, and it required us to continuously train new help as they arrived, and there's no question that from the surveillance and epi. perspective, the bulk of the response was carried out by the communicable disease program staff, and even with the demands of West Nile, we were still needed and expected to carry out our core public health activities, such as managing cases of meningococcal meningitis and other outbreaks, which unfortunately continued to occur. At the time that this outbreak occurred last year, our program had only 13 surveillance staff, including field epidemiologists and public health nurses and three medical epidemiologists. But we are lucky compared to many programs in this country, especially at the local level. We're much bigger than most, and people need to recognize that public health capacity varies widely across the country.

Similar to a bioterrorist attack in a large metropolitan area, there's obviously going to be the need or will involve more than one jurisdiction, and there are challenges to being a part of a multi-jurisdictional outbreak investigation. At the peak of West Nile, there were 18 different local, state, and federal agencies involved. Coordination was not easy. It was primarily done via conference calls that lasted several hours, tended to be chaotic, and initially at least, it was unclear who was in charge. And this illustrates the need ahead of time for us to work out issues related to centralized coordination and decision making, the need for secure and efficient means for information sharing, and to have consistent methods and a common outbreak database for surveillance and laboratory tracking. During West Nile, each jurisdiction kept their own outbreak database, as did the CDC lab, a separate database, which made it very difficult to link results and summarize our findings. During West Nile, the public health response, again, required the rapid implementation of mosquito control in a city that did not have an existing control program, and our ability to do this was mostly due to our preexisting and very positive relationship with our local Emergency Management Agency.

Under the leadership of Jerry Hauer, we were able to facilitate an immediate and effective coordination of the response. Fortunately, our public health agency had some experience with the incident command system and emergency operations centers through numerous table talk exercises, anthrax hoaxes, and real city emergencies, such as ice storms and blackouts. And these clear lines of responsibility and authority allowed the city to rapidly mobilize equipment and supplies, establish emergency contracts and oversee the logistics of their response. Okay. I'll talk even faster.

The most difficult aspect of any emergency is obviously the need for communication to all those who need to know, and though I think our press office and the mayor's press office did an excellent job of making key officials available, it's easy for the media with our dailies to both sensationalize the news a little bit, and also politicize it. And obviously there's a need to insure effective communication not just with the media, but with a number of partners. The medical community, we need good links with the medical community. They need to get their information from us.

The public with the need to mobilize hot lines, have pre-prepared materials translated into different languages, and also to insure that key government officials are informed. That was a problem during West Nile. We really need to work with them to make sure they're supportive and help relay our message to their constituents. And, again, it goes without saying the capacity to handle the media effectively and with the unified voice is key. Luckily during West Nile, there was a small number of cases. There was no need for mass care, mass prophylaxis, such as during the 1918 influenza pandemic or during the last smallpox outbreak in New York City in 1947.

So to conclude, similar to what occurred with West Nile, during a bioterrorist attack, there's no question the front lines of defense will be the clinical and public health community, and insuring our nation's preparedness will require enhancing awareness and training of clinicians, including veterinarians, to recognize the characteristic features of potential BT agents, and also the importance of reporting to their local health authorities. They need to know who to call, what to report, and how to reach us.

Also, we need to rebuild or build our public health infrastructure so that state and local health departments can be responsive when physicians call. The solutions are not high tech. We do not need fancy trucks or space suits. What we need is much more basic. We need staff expertise in infectious disease and epidemiology and information technology, reference lab capacity, the ability to respond 24 hours a day, electronic communication links, and good relationships with emergency management and local law enforcement.

This is my last slide, and I think this point has been made, that all of these funds for bioterrorism preparedness really have sort of a double value, the double Green Stamps value, as coined at the JFK School of Government executive session on domestic preparedness that I'm fortunate enough to participate in, and that all of these funds will make us better prepared, as happened with West Nile to respond to natural infectious disease outbreaks. I just want to end by saying in the past funding for public health infrastructure tends to be crisis driven, as occurred with West Nile, and unfortunately it's not maintained in the absence of an ongoing threat, and that there really is a need to emphasize the need for a commitment to continue and enhance these funds at the federal, state, and especially the local level. Every little of this money has crept down into local health agencies, and that this investment will allow us to cope with both natural and deliberate emerging infectious diseases.

And on that note, I apologize for going over, and thank you.
(Applause.)