Press Briefing Transcripts

CDC News Conference on Influenza

Update on Seasonal Influenza Activity

February 8, 2008

Speakers: Curtis Allen, Spokesman for CDC
Joe Bresee, Branch Chief for Epidemiology, CDC′s Influenza Division

 

OPERATOR: Welcome and thank you for standing by. At this time, all lines have been placed in a listen-only mode until the question-and-answer session.

Today′s conference is being recorded. If anyone has any objections, you may disconnect at this time.

I will now turn the call over to Mr. Curtis Allen, spokesman for CDC.

Sir, you may begin.

ALLEN: Yes. My name is Curtis Allen. I′m in the Division of Media Relations at CDC.

We know that in most years influenza peaks in February or later, and therefore we thought that this would be a good time to bring you up to date on what has been happening so far this year and maybe give you a little view into the future.

With me today is Dr. Joe Bresee, and that is Dr. Joe B-R-E-S-E-E. He is branch chief in the branch of epidemiology and prevention at CDC′s influenza division.

We will make a quick -- a short statement and then we will open it up for questions.

So, Dr. Bresee?

BRESEE: Yes, thank you, Curtis.

This is Joe Bresee. I′m the chief of the epidemiology and prevention branch, as Curtis said. Let me make a couple of opening remarks, and then I′m happy to take questions from everybody.

First, to note that seasonal flu activity was slow to start this year but has increased sharply in recent weeks, consistent with what we see in many years.

As you know, we measure disease activity and influenza activity in the U.S. multiple ways, but let me go through some of the data that we′re looking at this week and how it′s changed over the recent weeks.

This week in the U.S., 31 states are reporting widespread influenza activity, and 17 states are reporting regional influenza activity, and the remaining two states as well as the District of Columbia are reporting local influenza activity. So all states this week are reporting some influenza activity.

We also monitor the proportion of patients who show up at some doctors′ offices that have an influenza illness or influenza-like illness. And again, this week, that proportion is above national baselines and increasing, indicating a trend upward.

Finally, this week, 24 percent of people tested in CDC′s U.S. World Health Organization laboratories are testing positive for influenza, which is a proportion that has gone up in recent weeks as well, indicating increasing activity.

The other thing we monitor -- one other thing we monitor in influenza is deaths that might be associated with influenza, and we do that in two ways.

First, this week, the proportion of deaths that are attributable to pneumonia and influenza in 122 cities′ death registrars′ offices are above the epidemic threshold for the fourth consecutive week.

But again, the trend is what we see often in a typical year and is not atypical if you look back over the previous 20 to 25 years.

So far this year, we have heard about one death among a child that is associated with influenza infection. We have unofficial notification of five additional influenza-related deaths in children that will be added to our report next week, but we′re reporting them as preliminary cases this week, again representing an increase from one to five, and consistent with the increase in influenza activity generally.

Finally, we monitor the viruses that are circulating, the influenza viruses circulating, each year. So far this year, 84 percent of all the viruses that have been typed in our laboratory surveillance system are influenza A viruses. The remaining 16 percent are influenza B viruses.

Of the influenza A viruses, the 84 percent I talked about, slightly over half are influenza A(H1N1) viruses, and slightly less than half are A(H3N2) viruses.

As you know, each year influenza among humans can be caused by three main types or subtypes of influenza virus. Influenza A, there are two types, H1N1 and H3N2, and one type B. Therefore, the vaccine includes a representative strain of all three types.

Of the viruses that influenza division at CDC, as a World Health Organization collaborating laboratory, has characterized this year, the data are the following.

I said that most -- the predominant virus so far this year is an H1N1 A virus. Of those, 96 percent, or virtually all those viruses, look very much like the vaccine strain and should be well covered by the vaccine strain.

Eighty-seven percent of the A(H3N2) viruses have been characterized -- and I′ll repeat this if I need to -- A/Brisbane/10/2007-like viruses, which is a recent antigenic variant of the vaccine strain, A/Wisconsin virus.

And finally, 93 percent of the B viruses are in a lineage that is different from the vaccine strain lineage virus as well.

These data suggest that protection against the H3N2 and B virus strain in the community may not be optimal.

But it′s important to remember that while a less-than-ideal virus match between the viruses in the vaccine and those in the circulating viruses can reduce vaccine effectiveness, we know from past influenza studies that the vaccine can still protect enough to make illness milder or prevent flu-related complications.

This is particularly important in people at high risk for flu complications.

And finally, we know that there are Department of Defense data from last season when this H3N2 virus was predominant in Europe that showed that vaccinating people with the A/Wisconsin strain, which is, again, the virus that we have in our vaccine this year, was 52 percent protective against the drifted A/Brisbane strain.

I′d like to finish this reminding everybody that while vaccine is the best way to prevent influenza and is our primary tool, CDC recommends that people practice other protective measures like hand hygiene and cough etiquette, as well as the appropriate use of antiviral drugs.

I′ll finally spend one minute just updating you all on the issue of antiviral resistance among influenza viruses this year which has come to the forefront in the last one week to two weeks.

This week CDC is reporting that overall the viruses that we′ve tested in the influenza lab from U.S. cases of influenza, 4.5 percent of those viruses are resistant to oseltamivir, or Tamiflu.

All the resistant viruses are among the subtype A(H1N1), and if you just look at that subtype, 8.1 percent of those viruses are resistant to oseltamivir.

We haven′t seen any resistance yet among the A(H3N2) viruses or among the B viruses that we′ve tested this year, and no resistance is found to zanamivir, the other licensed antiviral drug.

So let me end there, and just the closing remarks that what we′re seeing this season is the increasing trend of influenza activity in the United States is, again, typical of influenza seasons in the past 20 seasons and expected.

But it reminds us all that influenza can be a common disease, clearly, and can be a severe disease, particularly among high-risk people for whom vaccine is recommended.

And I′ll end there, and I′m happy to take questions.

ALLEN: OK, Kim, we could open it up for questions, please.

OPERATOR: Thank you.

(OPERATOR INSTRUCTIONS)

One moment please for the first question.

Our first question comes from Will Dunham with Reuters.

Your line is open.

DUNHAM: Hi. This is Will Dunham in the Reuters Washington bureau.

Just a couple small points. How do the child deaths compare in total this year to previous years?

And could you also say -- it doesn′t sound like this year′s flu outbreak is much more severe or any more severe than recent years. Could you just amplify on that?

BRESEE: I′m happy to, Will. Thanks for the question.

DUNHAM: OK. And I′ve got one follow-up after that.

BRESEE: Sure. So I′ll start with those two.

So far, as I said, this year, one official death has been reported among a child to CDC, and we expect five -- five have been reported and will be entered and reported officially next week. And so we have a total of six reported informally or formally to CDC so far this year.

If you compare that with the previous four years for which we′ve done active or passive pediatric death surveillance, this is a relatively low number. But again, we′re just on the upswing of the influenza season, and that certainly may change.

But to give you a frame of reference, in the past four years, the pediatric deaths associated with influenza reported to CDC have varied from 44 to 153.

Since it′s been a reportable condition after the ′03-′04 year, which was particularly severe, between 44 and 74 cases of deaths associated with influenza have been reported to CDC.

I think you′re right in saying that this year so far doesn′t look typical -- doesn′t look atypical of an influenza year.

I would caution to say that if you -- for those of us -- and you all who have been around influenza for a while, it′s hard to predict what the season will look like either before or during the season.

We′re still on the -- have increasing activity in the nation, and so whether this year ends up being a severe season, a moderate or a mild season I think can′t be predicted with certainty at this point.

DUNHAM: And let me just ask, could you explain again how common the Brisbane strain is that you′re seeing in the United States?

BRESEE: Yes. Overall, what we′re seeing now -- let me preface this with -- I′m sorry, 30 seconds with a bit of an explanation.

The way we monitor viral activity in the United States is we have labs throughout the country that report to us two things. They report to us the presence of influenza in their lab and the number of samples they′ve tested that are positive, and they also report the numbers of samples they tested that are A or B, and, if they subtype, the type of subtype of A.

And so that′s where the data come from. We get weekly reports from these laboratories.

So far this season, these labs have reported, again, 84 percent of them are A, and of those 84 percent, 46 percent are H3N2. So if you multiply those two together, about 35 percent to 37 percent -- and you can check my math, and I apologize if I′m -- but around 35 percent of the viruses that we know have been typed are A(H3N2).

And if you assume that about eight out of 10 or nine out of 10 of those are the A/Brisbane strain, you get those numbers. About 30 percent of the circulating strains may be A/Brisbane.

ALLEN: Next question, please?

OPERATOR: Thank you.

Our next question comes from Daniel DeNoon, WebMD.

Your line is open.

DENOON: Thank you.

Could you repeat, please, the number of the B strains that are different from the B strain in the vaccine?

And I have a follow-up question to that.

BRESEE: Happy to. Of the influenza Bs that we have isolated and characterized so far in our lab, 93 percent of the B strains belong to a lineage which is a genetic group of influenza B viruses called the B/Yamagata lineage.

And this is different than the genetic lineage of B viruses where the vaccine representative strain comes from, which is called the B/Victoria lineage.

And so 93 percent of the Bs we′re seeing are in a different lineage than the vaccine strain.

DENOON: Is there any information about how protective the vaccine has been against that strain?

And could you also comment on whether any of these child deaths or any of the flu cases you′re looking at now have been associated with a staph aureus infection?

BRESEE: It′s an excellent question. Thanks. I can′t comment yet on the effectiveness of this year′s vaccine against the drifted B strains or any of the strains.

We are looking at data right now, and we have established study sites where we monitor prevention effectiveness. But those data aren′t yet available, again, because the season is just on the upswing, and so we′re just getting enough cases to make estimates that will come out soon.

The next question was whether we can comment on the number of deaths that are associated with staph aureus infection. We don′t know yet. We are collecting those data now.

ALLEN: The next question, please.

OPERATOR: Your next question comes from Mike Stobbe with the Associated Press.

Your line is open.

STOBBE: Hi, Doctor. Thanks for taking the question.

First, I just want to check those numbers about child deaths. I heard a one, a five, a six and a four. There′s one reported so far, and there is five preliminary which will take the total to six, is that right?

BRESEE: That′s right. Sorry to be so technical about that. I apologize. The reason they′re different is we have a reporting system at CDC by which we collect data from state health departments that report childhood deaths or deaths among people less than 18 years old that have a flu positive test.

That comes into CDC through the mechanism that states report many reportable diseases to CDC. Through that system, one death has been reported so far this year, but we have information on five additional deaths that have occurred and will be added to that system in the next day or two.

STOBBE: Could you comment for just a second about the challenge that health officials face in lining up the vaccine each year so it′ll be protective against the next season?

Because if I followed you, the B strain this year is not a good match, just like it wasn′t last year.

BRESEE: Yes, it′s a good question, and I can comment a bit on the challenges, though I′m not directly involved with the decisions. But let me tell you a couple of things.

It is difficult to predict which strains -- with accuracy which strains of influenza will circulate in the United States or in any country in any given year. We do very well with that generally.

But the decisions for what goes in the vaccine -- because of the time the vaccines need to be produced to be ready for a fall delivery to health care providers, the decisions must be made in February or around February of the year before

And so because influenza viruses are mutating and changing all the time, there may be changes in the virus or viruses that emerge and become predominant during that interval between February and the next flu season that can′t be predicted nine months in advance of the year when the vaccine decisions have to be made.

That said, I would say that -- and I′ve got the numbers here, and I′ll look for them while we talk -- most years in the last 20 years or so -- I′m looking at the numbers now.

In the last 19 -- the last 26 -- here are the numbers. In the last 26 years, in 19 of those years the predominant circulating strain has matched very well the vaccine strains.

ALLEN: Yes.

BRESEE: So most years the prediction is very good. But some years, the strains drift in the meantime or unexpected strains emerge. And I′ll say that while that′s true -- and there′s only been two years where the predominant circulating strain was very poorly matched with the vaccine strains.

That said, again, even when there′s a less-than-ideal match between the circulating strains and the vaccine strains, there is some cross protection, generally speaking, and therefore there′s some effectiveness of giving the vaccine in those years, we would predict.

And this is especially true in people who are at high risk for severe complications, who, if vaccine could lessen the severity of the disease or prevent the disease, may actually decrease their risk of dying or being hospitalized.

ALLEN: Next question, please.

OPERATOR: Your next question comes from Helen Branswell, the Canadian Press.

BRANSWELL: Hi. Thanks very much.

I have two questions, if I could, please. The Brisbane strain, the one that isn′t in the vaccine this year -- is it a case that it materialized after the strain selection meeting, or was it a situation where a good vaccine virus couldn′t be produced in time to make it into the vaccine? That′s the first question.

BRESEE: Thank you. Excellent question. And the truth is that the A/Brisbane/10 strain, which is included in the vaccine for the Southern Hemisphere this year and is similar to the viruses that are circulating now, first emerged and was first identified around February of last year, 2007.

And so your first question is right. The first question is right -- or the first hypothesis is right. The Brisbane strain emerged around the time the vaccine strain committee was meeting to select a vaccine, and therefore the decision not to include it really was -- it was novel, and there was no evidence at that time, at least, that it would emerge into a predominant strain.

Before I leave that, let me update a number I just gave you, because I just found the numbers that I wanted. And this is about how often we′ve had a well matched vaccine. And let me correct that to say that 16 of the last 19 seasons we had a well matched vaccine against circulating strains.

BRANSWELL (ph): Could I ask a follow-up?

BRESEE: Yes.

BRANSWELL: Oh, great. I assume your laboratories have been sequencing viruses, some of the H1N1 viruses, to try to get to the bottom of what′s going on with the H1N1s this year in terms of the Tamiflu resistance issue.

Are you seeing anything? Are there any clues to what might be going on there?

BRESEE: I don′t know yet. And I′m probably not the best person to answer the question. We are looking at these viruses very closely, both doing genetic sequencing and doing other testing on the viruses.

So far, we′ve been monitoring the rate of resistance, but the reasons for the resistance and the reasons for the emergence of this resistance and at a level that was higher than we′ve previously seen are still under investigation.

BRANSWELL: Thanks.

ALLEN: Next question, please?

OPERATOR: Your next question comes from Victoria Elliott, American Medical News.

ELLIOTT: Hi. This is Victoria. I was just wondering if this was the first week that flu had become widespread throughout the country, or it was existing in every single state?

BRESEE: I′ll have to look back at last week′s. I can′t say for certain. We′re certainly seeing increasing numbers of states that have reported widespread disease. It′s certainly not the first week that states have reported widespread disease.

ELLIOTT: OK.

BRESEE: But fewer states reported it last week than this week.

ELLIOTT: OK. Thank you.

ALLEN: Next question, please.

OPERATOR: Your next question comes from Martine Louis (ph) at Reporter newspaper.

LOUIS: Hi. Thank you.

I want to confirm that there was a flu outbreak in Massachusetts, and if so, is it in the Dorchester-Mattapan area?

BRESEE: I don′t know about Massachusetts particularly, but I can confirm that Massachusetts has reported this week -- and I′m looking at -- widespread activity, and so I don′t doubt that there have been flu outbreaks and flu cases in Massachusetts.

LOUIS: Thank you.

ALLEN: Next question, please.

OPERATOR: Your next question comes from Jia-Rui Chong with the Los Angeles Times.

CHONG: Hi. I′ve got a question of clarification. You were talking about deaths attributed to pneumonia and influenza in the 122 cities being above the epidemic threshold for the fourth consecutive week.

I was wondering what an epidemic threshold is, and I was wondering if you had any numbers for deaths of adults associated with influenza.

BRESEE: We don′t routinely report numbers of deaths in adults attributable to influenza. We do monitor deaths, again, in two different ways.

The pneumonia and influenza mortality that I just talked about is through a system by which 122 cities′ death certificate registrars report to us each week the number of deaths they′ve had in that area each week, usually a county or a city, and the proportion of those that list pneumonia or influenza on the death certificate. And so it′s a fraction.

The baseline is set using a model that looks back at previous years′ fractions of deaths and how they change over the season to try to predict what we would see this season.

And so I can refer you to a couple of publications and our Web site if you want the specifics about the model, or I can talk to you after. It′s probably boring to most people on the call.

But basically, when we say above baseline, it′s above what we would expect for that week for that season.

CHONG: OK. Well, because when you said above epidemic threshold, that sounds kind of scary, so I wasn′t...

BRESEE: Yes, the epidemic threshold is a line that′s set slightly higher than the baseline and is, again, a fairly complicated model, and I′m probably not the best one to explain it. I′ll refer you to my smarter colleagues.

CHONG: OK. Thanks.

ALLEN: Next question, please.

OPERATOR: Thank you.

(OPERATOR INSTRUCTIONS)

Our next question comes from Heidi Splete with Pediatric News.

ALLEN: Yes, please.

SPLETE: Hi. Thanks for taking the question.

I was wondering, since it appears that there is plenty of vaccine available this year, do you have any information about whether more people have received the vaccine and also, more specifically, whether more health care workers have received it?

BRESEE: We don′t have that information yet. We do track the number of vaccines distributed, but we don′t during the season have an estimate of the numbers of people who receive those vaccines, and especially among health care workers.

We do get that information through special surveys that are done, but the data aren′t available at this point.

ALLEN: Next question, please.

OPERATOR: Your next question comes from Richard Knox, National Public Radio.

KNOX: Hi. Thanks very much.

A couple of things. Has the vaccine strain selection committee met yet to pick the strains for next year′s vaccine?

BRESEE: They have not yet. They′ll meet in the next week or two.

KNOX: And secondly, we tend to focus on A because it′s the predominant cause of flu cases, but what can you tell us to help us understand the implications of the high degree of mismatch in the B vaccine strain versus what′s happening in the -- 40 of the 43 viruses being Yamagata?

What are the public health or even clinical implications, if there are any?

BRESEE: It′s a good question. I think the implications that the predominant circulating B types that are circulating -- that they aren′t represented well by the vaccine strain probably indicates that the effectiveness of the vaccine may be less than ideal.

But again, I caution to say that while genetically or by the proteins on the surface the viruses may look different than the vaccine strain, the effectiveness of the vaccine may not be completely compromised, and there may be partial effectiveness of the vaccine against those B strains.

We are, again, looking at this question in the U.S., looking at how well a vaccine performs when you give it to people instead of just looking at it in a lab. And those data may be more helpful in understanding what the implications of the Bs are.

KNOX: Is the Yamagata strain a completely different...

BRESEE: I′m sorry, again?

KNOX: Is the Yamagata lineage, you know, really different, so do you expect less cross protection?

BRESEE: We do expect less cross protection. The two B lineages are relatively distinct.

ALLEN: Next question, please.

OPERATOR: Your next question comes from Michael Smith, MedPage Today.

SMITH: Yes, hi, Dr. Bresee. Two questions, really. One is you gave death figures for children which were relatively low compared to previous years. Do you have comparable figures for adults thus far?

And secondly, do you have any kind of incidence numbers to go along with widespread and regional and local, those sorts of descriptors?

BRESEE: That′s a good question. First, about the adults, we don′t have any numbers on the adult deaths because we don′t track those specifically like we do pediatric death.

We can make some assumptions about the adult deaths or the overall deaths based on our information from the death certificate data which I just talked about.

And if you look at those data, the numbers of deaths reported during the season or during the last couple of weeks is not out of the ordinary with what you′d expect for most influenza seasons.

The other question I′ve suddenly forgotten. Can you repeat it, please?

SMITH: It was do you have any incidence numbers? I mean, can you put a little salt on the tail of what widespread flu activity and regional flu activity and local activity means? And can you sort of compare that with...

BRESEE: That′s a great question, too.

SMITH: ... something else?

BRESEE: I can′t, just because the widespread and regional designation of those activities aren′t based on incidence numbers at all.

They′re meant to indicate the geographic spread of influenza within a state, and so they represent, for instance, the number of areas in a state that are reporting influenza or the numbers of outbreaks and where they are.

So it′s meant to be an indicator of geographic dispersion of influenza in a state, not of severity. The definitions of those terms are available on our Web site under the Flu View icon.

SMITH: But do you, in fact, have any incidence numbers at this point?

BRESEE: We do collect incidence numbers but only on pediatric hospitalizations, and those are updated as well each week on our Flu View and again are in line with what we′ve seen previously, and I don′t have them in front of me.

SMITH: Thank you.

ALLEN: Next question, please.

OPERATOR: Your next question comes from Lisa Schnirring with CIDRAP News.

SCHNIRRING: Hi. Thanks for all the excellent information today.

I was wondering -- I know that there was some early warning about the Brisbane strain. Did you notice anything about the B lineage difference elsewhere around the world, any early hints of that?

BRESEE: I don′t know the answer to that. I can look it up and we can talk at a later call, but I don′t have the data in front of me.

SCHNIRRING: OK. Thanks.

ALLEN: And this will have to be our last call. Next question, please.

OPERATOR: Your last question comes from David Brown, Washington Post.

BROWN: Yes. Thank you.

The surveillance data that you gave is obviously more recent than week number four on the Web site, where the fraction of influenza A viruses that were H3 was 34.7, and you said it′s now 46.

Are these updated data available anyplace on the Web site, or can you give them right now?

ALLEN: They should be up on the Web site very shortly. As a matter of fact, they go up on the Web site every Friday, and you could find the surveillance report at www.CDC.gov/flu, F-L-U. And if they′re not up yet, they should be up momentarily.

BROWN: OK. Well, then can I -- I have another unrelated question.

Is the clinical presentation of this year′s influenza at all different or characteristic -- you know, different from the past? You know, is there more pharyngitis or more -- higher fever or some other symptoms?

BRESEE: That′s a good question. We don′t have ways or methods or routinely monitor clinical presentations of influenza, so I don′t know the answer to that.

We certainly have not heard about clusters of illness or reports of illness that are atypical of what we normally see with influenza.

BROWN: OK. Thanks.

ALLEN: OK. Thank you very much for joining us today. Again, the transcript of this briefing will be on the Web site at www.CDC.gov, and it will probably be three or four hours before it is actually up.

Then the surveillance report, again, is on the Web site each Friday and updated each Friday at www.CDC.gov/flu.

If you have additional questions, you can call the media office here at CDC at area code 404-639-3286. That′s area code 404-639-3286. And thank you very much for joining us today.

OPERATOR: Thank you. This concludes today′s conference. You may disconnect at this time. Thank you.

END

####

U.S. DEPARTMENT OF HEALTH AND HUMAN SERVICES