Div. of Media Relations
1600 Clifton Road
Atlanta, GA 30333
Fax (404) 639-7394
CDC Telebriefing Transcript
Flu Update with Dr. Tim Uyeki, CDC Influenza Expert
January 31, 2002
MR. ALLEN: Thank you for joining us today.
Today's topic is "Influenza Surveillance Update."
Joining us is Dr. Tim Uyeki, and that is spelled U-y-e-k-i. He is a Medical Epidemiologist in our Influenza Branch. He will make a few brief
remarks and then turn it over for your questions.
I'll turn it over to Dr. Uyeki now.
Thank you very much.
DR. UYEKI: Before I get into the points of the MMWR article that will be published tomorrow, February 1, I wanted to give a little background to
help understand our surveillance and how we conduct it in the United States for influenza activity.
We have four components to our surveillance, and we look at them collectively; we don't look at them really individually on their own, but with
a whole picture of influenza activity. We look at the combined data.
So briefly, one of the four components is what we call the Sentinel Physician Reporting System. Another component we call the 122 Cities
Mortality Reporting System that reports mortality due to pneumonia and influenza. We also have another component which is weekly reports by
State and Territorial epidemiologists throughout the U.S. and Puerto Rico. And finally, we have weekly reporting that is done by World Health
Organization collaborating laboratories and State public health laboratories, as well as what we call the National Respiratory and Enteric Virus
Surveillance System collaborating laboratories.
I will just go over these a little bit more individually.
Our Sentinel Physicians Reporting System consists of volunteer physicians in 47 States that report weekly from October to May to CDC the
percentage of patients in their practices that they have seen each week with influenza-like illness. And the definition of influenza-like
illness is fever greater than or equal to 100.0 degrees Fahrenheit and either cough or sore throat. This is a nonspecific definition, and we
don't call this influenza, but influenza-like illness activity. And the reason why it is called influenza-like activity is because there are
many other infectious organisms such as respiratory viruses and bacteria that can cause similar influenza-like symptoms that influenza viruses
also can cause.
In terms of the 122 Cities Mortality Reporting System, these are 122 jurisdictions in the United States, primarily cities, that report on a
weekly basis, actually throughout the year, but for our system, we are primarily looking at October through May. They are reporting the
percentage of total deaths in their cities or jurisdictions that were due to pneumonia and influenza. Again, it is important to realize that
this is not just influenza, but pneumonia and influenza deaths, or deaths that are attributed to pneumonia and influenza.
We take this historical data that has been reported over the years, and we put it into a logarithmic model, and basically, if the model predicts
what we call a baseline or expected baseline of activity, we also predict what is called an epidemic threshold. In this model, the graph looks
like a sinusoidal curve. I think that many of you may have seen this graph on our web pages or in MMWR publications. We use this to sort of
gauge the severity of the influenza season. When reported deaths due to pneumonia and influenza exceed the epidemic threshold for a number of
weeks or greatly exceed the expected epidemic threshold, that usually correlates with a more severe influenza season.
In terms of the weekly report by laboratories, by World Health Organization and National Respiratory and Enteric Virus Surveillance System
laboratories, these are weekly reports we are receiving from laboratories for respiratory specimens from patients across the United States that
have been placed into viral culture. So we get the total number of specimens that were tested as well as the number and percentage that test
positive for influenza viruses. This is done weekly from October to May as well.
Finally, the fourth component is the State and Territorial epidemiologists' weekly assessments of influenza activity in their States or
Territories. There are various levels of activity, and we define these--and it is important to realize that this is a combination of both
laboratory-confirmed influenza as well as influenza-like illness, which may not always represent influenza activity; it may represent cold
viruses such as rhinoviruses, and other respiratory pathogens that can cause influenza-like illness. But we have the four levels as "no
activity," "sporadic activity," "regional activity," or "widespread activity," and you can see in the MMWR
publication for tomorrow or from our website or previous publications exactly what those definitions represent.
So with that background on the surveillance system, I guess I'll go into some of the major points of the article that is going to come out
I think one of the main points is that so far this season at the national level, when we look at all four of our influenza surveillance
components collectively, to this point in the season influenza activity has been low on a national basis.
However, influenza activity has increased in recent weeks, and we expect that influenza activity is going to increase in the coming weeks.
Of the influenza viruses that have been isolated throughout the United States and sent to the Centers for Disease Control and Prevention for
specific testing that we call "antigenic characterization"--this is a very specific method of testing to identify the specific strains
of viruses that are circulating.
Of the isolates of influenza virus that we have received and antigenically characterized so far this season, all of these isolates are
well-matched by the current 2001-2002 influenza vaccine. So that is very good news.
The predominant viruses that we are seeing in the United States circulating so far this season are Influenza A, subtype H3N2 viruses.
We estimate that there are approximately 10 million doses of influenza vaccine that are still available from manufacturers and additional doses
may also be available from distributors.
Since there is plenty of vaccine that is still available, and because influenza activity has not yet peaked, and we expect influenza activity to
increase in the coming weeks, and because the circulating influenza viruses in the United States are well-matched by the current 2001-2002
influenza vaccine, we recommend that health care providers continue to offer and vaccinate patients for influenza vaccination this season.
Unvaccinated persons can still benefit from influenza vaccine even if influenza activity has been detected in their community, and we would
recommend that health care providers continue to offer influenza vaccine to patients during February.
I think those are the main points that I wanted to get across, and I would be happy to take some questions at this point.
MR. : Please open it up for questions.
MR. : Ladies and gentlemen, if you wish to ask a question, please press the "1" on your touch-tone phone. You will hear a tone
indicating you have been placed in queue, and you may remove yourself from queue at any time by pressing the "pound" key.
If you are using a speaker phone, please pick up your handset before pressing the numbers.
Our first question will come from the line of Susan Ferraro [ph.], with New York Daily News.
Please go ahead.
MS. FERRARO: Good morning. Thank you for having this teleconference.
I am with New York's home town newspaper. How is New York City doing and New York State?
DR. UYEKI: The most recent reports that CDC has received from New York State and New York City are that they are experiencing widespread
influenza activity. As I said earlier, that type of assessment is a combination of laboratory-confirmed influenza as well as just influenza-like
illness activity, which we define as fever of [greater than or equal to] 100 degrees Fahrenheit and either cough or sore throat. So it is a
combination of that. But they have been reporting, at least the most recent week, widespread activity.
But I would recommend that you speak to people in the New York City Health Department and the New York State Health Department for the most
MS. FERRARO: Thank you.
MR. : The next question comes from the line of Emma Hitt [phonetic], with Reuters Health.
Please go ahead.
MS. HITT: Hi. Thanks for having this.
I wanted to know how does the number of doses remaining, the 10 million, compare to that of previous years, and will the manufacturers suffer a
financial loss if it is an excessive number of doses.
DR. UYEKI: Your question is how does the number of doses available at this time compare with, say, last season or previous seasons in terms of
how many doses are available?
MS. HITT: Right. Is this an excessive number of doses, I am trying to determine.
DR. UYEKI: Right. It is a little hard to make that comparison for a few reasons. Prior to last season, typically, most influenza vaccine in the
United States is administered between September and the end of November. Any remaining doses of vaccine are basically returned to the
manufacturers or are simply not used.
We don't have--I don't have available data on exactly how many were returned or available at this point last year or the previous year. But last
season, as you know, we had less total doses of influenza vaccine produced, and we also had a delay in the availability of vaccine.
So influenza vaccine was still being distributed and administered during December and January of last year.
It is a little hard to compare this season with last season and the previous season. This season, we have also experienced a slight delay in the
availability of vaccine, so during this season and the last season, more vaccine was distributed during December and January than in prior
But our information at this time is that there are at least 10 million doses of influenza vaccine available.
MS. HITT: Okay. So that's enough.
MR. : The next question will come from the line of A.J. Hofstedler [ph.], with Richmond Times Dispatch.
Please go ahead.
MS. HOFSTEDLER: Hi, Dr. Uyeki. I won't complain that I came down with the flu last week after getting the flu shot, because I know that
sometimes happens, but I was glad that it lasted only a couple of days.
I have a couple of questions. Do you have any thoughts on why we had a slightly slow start, and can you comment at all on the severity of the
cases, with the predominant strain being Panama or Panama-like?
And then, my other question is has there been any fallout or is anybody seeing anything--I know there was a lot of concern both here in Virginia
among physicians and nationwide that they would get a rush of people with flu-like illness, in part because of the anthrax scare. I just
wondered if there has been any sign of that.
DR. UYEKI: Okay. I'll try to address all three points or questions.
The first point, you are reporting that you came down with the flu, and you were vaccinated, but I would have several responses to that, and I
think it is useful to make this point to those who are on the call.
MS. HOFSTEDLER: Okay.
DR. UYEKI: My first response would be that if that is indeed true that you clearly did have influenza, then I am sorry that that occurred; that
would represent a vaccine failure. However, what often happens is that people who get vaccinated, particularly healthy people, people who are
not elderly, report that they got the flu even though they got vaccinated, or they got it three or four times during the season.
And my question always is: How do you know you got the flu? There are two possibilities--you either did, or you didn't.
So because many, many different other infectious organisms can cause similar symptoms, one possibility is that you did not have influenza, but
you had some other infection such as rhinovirus infection, the common cold virus; there are other viruses such as adenoviruses, coronaviruses,
respiratory syncytial [ph.] viruses, parainfluenza viruses, et cetera, as well as bacterial infections and related organisms that can cause
So one possibility you didn't. So when people always tell me this, my first response is did you get laboratory confirmation to document that in
fact this was true--and most people don't.
It is possible that you did have the flu, and your case would represent a vaccine failure, because the vaccine is not 100 percent effective.
However, because this season so far, we know from our testing at CDC that the circulating viruses are well-matched by the vaccine, and you are
presumably a healthy person who is not elderly, we would expect the vaccine to be very effective in you. But it is not 100 percent effective. It
is approximately 70 to 90 percent effective when there is a good match. So it is possible that you did have the flu.
Anyway, I'm not going to question whether you did or not. I just wanted to explain that this is a common point that many, many respiratory
infections causing influenza-like illness do occur in the winter months, that not everything is caused by influenza viruses, and that influenza
vaccine only protects against influenza virus infection and not against other viruses or bacteria.
In terms of your question about the slow start to the season, the way I would respond to that is that if you look back at the last two years,
that indeed would be true. Last season peaked approximately in mid to late January, and the 1999 to 2000 season peaked approximately mid to late
However, if you look back over the past 25 influenza seasons, the data suggests that during 15 of those, of the past 25 seasons, influenza
activity actually peaked during February or later. In some seasons, it peaks in March and actually into early April.
So that, sure, compared to the last two seasons, this is a later-peaking season, and perhaps got off slowly, at least from the national picture,
but if you do look historically over the past 25 seasons, in fact this is not unusual at all, and--well, that's all I would say there.
In terms of your question about the fear in the early fall that this winter, we might see lots and lots of people flooding emergency rooms
because they had influenza-like illness symptoms, and they were concerned that they might have inhalation anthrax, we have just not heard about
any of this occurring.
And one possibility is that in fact, this has been a mild season, so there just hasn't been the typical overloading of emergency rooms,
overwhelming of the emergency medical system in major American cities that we do see at the peak of influenza activity during severe seasons.
Another reason is probably that we have not had inhalation anthrax cases occurring for several months now. That outbreak is over.
I think those two factors probably are good explanations of why we have not observed this massive rush to emergency rooms and overwhelming of
Again, as I mentioned earlier, we do expect influenza activity to increase in the coming weeks. If you look at the weekly percentage of
respiratory specimens that test positive for influenza viruses at the national level, in recent years, that has peaked at between 24 and 33
percent of specimens submitted in a week--or, tested in a week, testing positive for influenza viruses.
Our most recent week was the week ending January 19, and that was 13.9 percent. So I guess from that surveillance component, we would say that
influenza season has not peaked yet.
MS. HOFSTEDLER: Thank you.
MR. : The next question comes from the line of Tom Kelter [ph.], with the Baltimore Sun.
Your line is open.
MR. KELTER: If you could give me an estimate of roughly how many influenza cases there are a year in the United States; how many there were
roughly last flu season, last year; and whether you expect this year to break last year's number.
DR. UYEKI: We do not have any estimates about that, and I guess the most important point to realize is that influenza activity, or the severity
of influenza seasons, varies from year to year. It is dependent on many different factors, and one of the most important factors would be what
influenza viruses are the predominant strains.
For example, last season, the predominant strains or the predominant types of influenza viruses that were in the United States were Influenza A
viruses and Influenza B viruses, and in particular, the subtype of Influenza A virus was AH1N1.
In general, seasons that are predominated by Influenza AH1N1 and Influenza B viruses are milder or less severe than seasons predominated by
Influenza AH3N2 viruses.
Now, if you go back two, three, four seasons ago, where Influenza AH3N2 viruses predominated, these were more severe seasons; we would consider
them moderately severe seasons.
So one factor is what viruses in fact are circulating out there, and we really can't compare apples to oranges because every season is
Another factor is what is the level of immunity or immune protection in the American population or in a population to the circulating influenza
strain, and that is going to be dependent upon several factors which would include vaccine-derived immunity from prior year vaccination or
current year vaccination as well as people who have been infected with influenza viruses previously. They may have some antibody to protect them
from the current strain. It will also vary on the individual. Elderly people may have very, very short, limited duration of immunity, whereas
healthy young people may have a much longer level of immunity of antibody protection.
So in terms of your question about an estimate of how many infections, the first answer would be that we have no data on that. The main reason
is that few physicians ever test patients for influenza viruses to document that they indeed were infected. So most of the data for any kind of
projections would have to come from very old community studies and some that are ongoing. But unlike other infectious agents, such as HIV or
syphilis, et cetera, where there are specific tests, and most patients who have those are--well, I shouldn't say most patients--I would say that
we get reporting data of patients with those infections, and a high proportion of patients who are symptomatic with those illnesses and other
reportable diseases will get tested, and we can have that data--we have that data available, and we can generate incidences and compare from
year to year.
But with influenza, most patients either are sick, they stay at home, they do not go to a medical care provider, or if they do, and even if they
get hospitalized, they may never be tested. So we do not know the total number of influenza infections that occur from year to year.
MR. : The next question will come from the line of Aaron McLamb [ph.] with Associated Press.
Please go ahead.
MR. McLAMB: Hi. Thanks. I just have a couple of questions.
First of all, are you able to tell specifically when this season might peak?
Secondly, why is it that this year's batch of vaccine appears to be so well-matched to fight the flu?
Also, I would like to know whether you go by "Tim" or "Timothy" in print.
DR. UYEKI: The second question about predicting when the season will peak, I think we can--I mean, the best response is that we can never
predict the severity or when the season is going to peak, but generally, the season will peak sometime between November and April, and typically
during January or February.
But at this point in time, we could not predict the week, but the surveillance data suggests that the season will peak in February or perhaps
later. But we cannot predict if the season will continue to be mild, if the season will then be severe, and that is because all we can report is
data received to date. We could not predict whether or not there may be other influenza viruses that will predominate later in the season.
And the reason why there is such unpredictability is that influenza viruses are constantly undergoing change, genetic change, and that is why it
is always a very big challenge to us to try to figure out our best estimate of what viruses may be predominant in the United States for the
coming season. So we are already collecting and analyzing surveillance data, not just from the United States but worldwide, to help us and the
World Health Organization in selecting the strains for the influenza vaccine for next winter. And it is a very long process that takes 6 to 8
months to actually produce vaccine, and therefore, these decisions are made between January and March for the coming season.
I go by "Tim." I can't remember if I answered all of your questions.
MR. McLAMB: You did. Thank you.
DR. UYEKI: Okay.
MR. : Our next question comes from the line of Susan Shackman [phonetic], from CBS News.
Please go ahead.
MS. SHACKMAN: Oh--my question was already answered about when we thought the epidemic threshold would be reached and what is the epidemic
DR. UYEKI: I think you are referring to our 122 Cities Mortality Reporting System and the mortality attributed to pneumonia and influenza. And
so far, since the beginning of our surveillance period in October, the mortality attributed to pneumonia and influenza has been below the
epidemic threshold, has been basically below baseline, and we could not predict if it is going to increase above the epidemic threshold or not,
but so far, it has been a mild season compared to other influenza seasons, period.
MS. SHACKMAN: Okay.
MR. : We have a question from the line of Lee Bauman [ph.] from Scripps News Service.
Please go ahead.
MR. BAUMAN: Hi, Doctor.
Just real quickly, if the figure of 87 million doses produced is approximately right, and you say there are about 10 million left, does that
mean that we can project that approximately 77 million people have received the vaccine, and is that the most that have ever been vaccinated?
DR. UYEKI: I think what we can tell you is that information that we have received from the three influenza vaccine manufacturers is that they
have produced 87.7 million doses this season and that there are at least 10 million doses still available from manufacturers. In other words,
they have distributed 77.7 million doses of vaccine.
At this point in time, we could not tell you if all--and the manufacturers I do not believe could tell you--how much of that 77.7 million doses
of vaccine that have been distributed have been administered.
MR. BAUMAN: Okay.
DR. UYEKI: I'll just give you some comparison data. Last season, 2000-2001, the total amount of doses that was produced was 70.4 million. During
1999 and 2000, the total number of influenza vaccine doses produced was 76.8 million.
So this season, the three manufacturers have produced a substantial increase in total doses of influenza vaccine produced. That's all I would
say. I think for further information about how much has actually been administered, you might want to contact the manufacturers as well as the
Food and Drug Administration, but I think it is very difficult at this time for anyone to know. Basically, they can tell you how much vaccine
has been produced and how much has been distributed.
MR. : Next question?
MR. : Once again, ladies and gentlemen, if there are any questions, press "1" at this time.
MR. : Sir, at this time, there are no other questions in queue.
MR. ALLEN: Okay. Thank you very much for participating today.
The transcript of this briefing will be on the CDC website later this afternoon.
And again, thank you very much.
MR. : Ladies and gentlemen, this does concluded our conference for today. Thank you for your participation, and thank you for using AT&T