Centers for Disease Control and Prevention
CDC Media Briefing: Influenza Vaccine Effectiveness
CDC Media Briefing
April 17, 2008
12:00 p.m. ET
OPERATOR: Welcome and thank you all for standing by. At this time, I would like to remind parties that your lines are in a listen-only mode until the question-and-answer session, at which time you may press star, one to ask a question and star, two to withdraw your question. Today's call is being recorded. If you have any objections, you may disconnect at this time. I will now turn the meeting over to Curtis Allen. Thank you sir, you may begin.
CURTIS ALLEN: Thank you for joining us today and also thank you for your patience because we're running a few minutes late. Today, we have Dr. Dan Jernigan. That's Dr. Dan Jernigan. Dan is Deputy Director for the Influenza Division at CDC. We also have Dr. Jeanne Santoli. That is Jeanne Santoli. She is Deputy Director for the Immunization Services Division at the National Center for Immunization and Respiratory Diseases. Dan will start us off. He will bring us up to date on where we are in the influenza season and also he will discuss the vaccine effectiveness. Following Dan, Jeanne will talk about vaccine supply for last year and also what we can possibly expect for next year. So –
then we will have some Qs and As. So with that, I'll turn it over to Dr. Dan Jernigan.
DAN JERNIGAN, DEPUTY DIRECTOR, CDC INFLUENZA DIVISION, NCIRD: Thank you. Today, I'm going to provide updates on seasonal influenza activities, including what viruses we've seen in circulation this year, and their characteristics based on tests in our laboratories, and then I'll present the findings of an interim study on the effectiveness of this season's inactivated vaccine. These findings that I'll present underscore the importance of three things that can be done to prevent and treat influenza; first, taking time to get vaccinated each year; second, for taking everyday precautions like covering your cough and washing your hands to prevent the spread of germs; and third, for taking anti-viral drugs if your physician thinks you should.
Influenza places a substantial burden on the U.S. and it's an ever-changing and challenging infection to fight. When we talk about the influenza season, we're actually talking about the combined effect of three different strains of influenza viruses that circulate in the community. Each influenza season will differ regarding when each of the strains will start showing up and which of the three will cause the most number of influenza infections that year. In addition, each of these three strains may respond differently to the flu vaccine and may respond differently to anti-viral drugs to treat flu infections. This constantly changing yearly cycle of influenza viruses is why we have to regularly monitor the viruses and regularly update the vaccine's three viruses to best prevent infections.
In terms of this year's influenza activity, the 2007-2008 influenza season has peaked, but we are continuing to see some influenza activity and we know from previous years that influenza viruses can continue to circulate into May. A seasonal update in this week's MMWR summarizes the influenza activity from September 30th to April 5th. The influenza activity in the United States has remained low until January. It peaked in mid-February and has decreased since then. And according to the measurement of influenza activity that we monitor each year, most surveillance indicators this season have been on the high-end compared to the past three seasons.
One of those measurements is pneumonia and influenza deaths that we monitor, and those numbers of death peaked at 9.1 percent. The number of these deaths have exceeded the epidemic threshold for 13 consecutive weeks so far this year. At this point, the 2007/2008 season appears to be most similar to the 2003/2004 season and that season was characterized as moderately severe with the percentage of pneumonia and influenza deaths peaking at 10.4 percent and exceeding the epidemic threshold for nine consecutive weeks.
Data from CDC’s most recent flu weekly influenza surveillance report, which is available on the website, indicates that influenza viruses are continuing to circulate this season with about 13.2 percent of specimens tested during the week ending April 15th being positive for influenza. Although, Influenza A virus is predominantly – predominated this season overall, the most recently isolated viruses are Influenza B viruses in the community.
The CDC also monitors pediatric deaths that are due to influenza, and so far this season 66 deaths in children due to laboratory confirmed influenza related complications have been reported to CDC. Of these 66 pediatric deaths, 52 of them were either not vaccinated or only partially vaccinated or younger than six months so they were not eligible for vaccination. Of the un-vaccinated children, most of the infections were due to Influenza A, however, deaths were due to all three types of the influenza that are circulating. Reporting for this year's season will officially conclude in mid-May and seasonal data will published after that time.
I'd like to next take a few minutes to discuss what we've been finding throughout the year in the laboratory. Most of the circulating influenza viruses this season have been less than optimally matched to the viruses in the vaccine. Influenza A H3N2, one of the two types of Influenza A, and Influenza A H1N1, the other type of Influenza A and Influenza B viruses have been co-circulating in the United States this year. H3N2 viruses, however, are the predominant virus that we have seen this year.
Of the viruses tested, about three quarters were Influenza A and most of those, around another three quarters, were the H3N2 subtype and up those about 70 percent were a string known as A-Brisbane10/ 2007. This particular strain is the one that is the predominant strain this year and is drifted, but it is still somewhat related to the A-Wisconsin strain, which in the 2007 - 2008 vaccine. The majority of the influenza B viruses, and 94 percent were characterized as B Florida, which is belonging to what we call the Yamagada lineage. This is a very different lineage of Influenza B than the Victoria lineage that's in the vaccine this year. No resistance to all Inhibitor has been observe in the other type of influenza AH3N2 or the influenza B viruses. CDC is continuing to monitor the situation closely. Finally, let me now spend a few minutes discussing the vaccine effect of this year's vaccine. Despite a less and optimum match in two of the three vaccine components this year, interim data indicates that the vaccine provided substantial protection against the predominant influenza viruses this season the H3N2 influenza A virus. Interim results of the vaccine effectiveness study carried out at the Marshfield clinic in Wisconsin appeared in this week's MMWR.
These results are from patients enrolled in the Marshfield studies in January 21st to February 8th of 2008. The article summarizes initial vaccine effect in this result for the trivalent inactivated vaccine, flu shot, in preventing medically attended laboratory confirmed influenza among patients living in a 14 zip code areas surrounding Marshfield, Wisconsin during this past influenza season. The study found in overall vaccine effectiveness are 44 percent against all types of influenza both A and B circulating in a community in Wisconsin. However, when you look at the effectiveness of the vaccine at preventing influenza A infections, they found a vaccine effectiveness of 58 percent against circulating influenza A H3N2 viruses, that's the predominant strain this year. There was no vaccine effectiveness against influenza B viruses found. There were no influenza A H1N1 viruses detected, so no vaccine effectiveness is available for the type of influenza A H1N1. While the vaccine effectiveness against the H3N2 viruses is less than might be expected during the season when the viruses and the vaccine, and circulating viruses are well matched, this interim result suggest that vaccination provided substantial protection against H3N2 influenza assorted, associated medically attended illness in the study population. This means that people in the study who are vaccinated were 44 percent less likely to have laboratory diagnosed influenza than those in the study that were not vaccinated. In regarding prevention of the influenza A H3N2, the predominant strain this year, those in the study that were vaccinated were 58 percent less likely to have laboratory diagnosed H3N2 infections than those that were not vaccinated.
The measurable benefit demonstrated in this study, supports existing public health practice of continuing to recommend influenza vaccination even when there is indication of a sub optimal match between the vaccine viruses in circulating influenza viruses. The result of this study also underscore however the influenza vaccine is in perfect and for this reason, health care professionals and the public need to consider the three prong approach to combating influenza, I mentioned at the outset, that is, first, taking time to get vaccinated each year. Second, taking everyday precautions such as covering your cough and washing your hands to prevent the spread of germs. And third, taking anti-viral drugs through physicians are things you should do. If influenza B strains dominate during the remainder of this season, providers can anticipate an increased risk for vaccine failures and should consider early use of anti-viral medications for treatment and prevention of illness in persons at high risk for complications from influenza infection.
CURTIS ALLEN: Thank you, Dr. Jernigan. Now, Dr. Santoli.
JEANNE SANTOLI, DEPUTY DIRECTOR, CDC IMMUNIZATION SERVICES DIVISION, NCIRD: Hello, there. So, as Curtis told you, I wanted to provide you with an end of season look at the vaccines supply and distribution, and also to speak very briefly about what we're thinking about for the upcoming 2008/2009 season. So, for the 2007/2008 season, as you know there were six manufacturers who are licensed to produce vaccine for the U.S., and these manufacturers together produced a record amount of influenza vaccine approximately 140 million doses. This is about 20 million doses more than were produced in the prior season. Now, based on data that were reported to CDC by influenza vaccine manufacturers and distributors throughout last season, we know that approximately a 113 million doses of that vaccine were distributed. And, this is more vaccine that has been distributed in the U.S. in a single season before and it is about 10 million more doses than were distributed in the last season.
Now, looking ahead, we anticipate that next season's vaccine supply will be similar or somewhat increased to what we had available this season. But, we'll be receiving projections to understand that better from influenza vaccine manufacturers some time in the next month. Another thing to keep in mind, of course, about next season is that in February of this year, CDC's advisory committee on immunization practices voted to expand the annual influenza recommendations to include healthy children and adolescents 5 to 18 years of age. This recommendation is to be implemented during the upcoming 2008/2009 season as feasible and no later than the 2009/2010 season.
Now, you probably are well aware, prior recommendations for the vaccination of children and adolescents included all children 6 months through the fifth birthday, children 5 to 18 years of age with high risk medical conditions like asthma, diabetes or heart disease and children 5 to 18 years, who were the household contacts of high risk children and adults. Now this expanded recommendation for next season increases the number of children recommended for vaccination by about 30 million children although we know from experience with other vaccine recommendations that not all of these newly recommended children and adolescents will be vaccinated during the first season following this recommendation. Thank you very much.
CURTIS ALLEN: Rose, we'll open it up for questions.
OPERATOR: Thank you. At this time, if you would like to ask a question, press star followed by one on your touchtone phone. You will be prompted to record your name and I will announce you prior to asking your question. To withdraw your question, you may press star, two. You may queue up at any time and you will hear a tone to indicate your question is pending. Again, press star followed by one. Our first question is from Helen Branswell, Canadian Press. Your line is open.
HELEN BRANSWELL, CANADIAN PRESS: Hi. Thanks very much for taking my question. This efficacy study was done in week 6 which was kind of that around the time when H3 and H3N2 is making its surge and before the late season B surged. With that sort of paint – because the picture was taken at that point, would it paint a nicer picture than actually might have been the case?
DAN JERNIGAN: The study was initiated when they first started seeing circulation in that community. And so, while compared to the overall U.S. population, it target at a relatively earlier point of that – for that community. If that's your – if your concern is that the initiation of the study tended to provide better results to the initiation of the study was done at the time when circulation began in that community.
CURTIS ALLEN: Next question please.
OPERATOR: The next question is from Richard Knox, National Public Radio, your line is open.
RICHARD KNOX, NATIONAL PUBLIC RADIO: Hi, thanks. A couple of things. I've – There is only five people every 65 in the study test that could positive for flu so, I assume that there is no way to calculate vaccine effectiveness amongst elders. Is it likely that there is vaccine effectiveness with conservatively lower than 44 percent or 58 percent for age three and two, and is it possible that it was zero?
DAN JERNIGAN: Yeah, I think this particular study does not have a lot of people in that age category, some studies – other vaccine effectiveness studies do not show a strong in effect in the older individuals. Its a little early, this is still the beginnings of the results being tabulated, there's still part of the season where some of those numbers may increase and we would be able to – with the folks of Marshfield would be able to evaluate that part of this, but at this point, this particular study can't really address that at all.
RICHARD KNOX: Well, I realize that, but is it possible that it’s or likely that it would be lower than 58 percent?
DAN JERNIGAN: It is possible that if you – you really have a larger number of elderly individuals that your vaccine effectiveness will go down.
RICHARD KNOX: And secondly, there was this paper in science actually coming out today, suggesting that the origins of a seasonal flu are in South East Asia, and the authors of that paper suggested that – knowing that may give us a leg up in terms of better vaccine methods of the future. Could you just comment on whether you think that's likely, and whether it may influence the process as soon as the coming season?
DAN JERNIGAN: Yeah, I think historically, the virus is either looked at for being potential vaccine candidate have come from one of the areas of the world where there is good surveillance and where there are laboratories that can adequately characterize those viruses. And also grow viruses in eggs, that allow for us to have those viruses that make good vaccine candidates. I think, as there are more and more laboratories that have the capability to test for influenza viruses and to find those viruses that make good vaccine candidates and that helps everybody. And so, CDC is supporting a lot of efforts in East, Southeast Asia to improve the ability of those laboratories to characterize viruses, to culture them, and to setup surveillance so that those viruses can be detected and characterized.
CURTIS ALLEN: The next question please?
OPERATOR: Thank you. Will Dunham, Reuters, your line is open.
WILL DUNHAM, REUTERS: Hi, this is Will Dunham with Reuters in Washington. Could you say, what is the typical years of vaccine effectiveness, for point of comparison to the 44 percent seen this year. And had there been years with lower vaccine effectiveness than 44 percent?
DAN JERNIGAN: Yes. There are numbers of studies that are actually described in the early part of the MNWR. One of them is a study that was done over four years, where two of those years, the vaccine – the circulating strains in the community were not as well matched to the vaccine as the other two years. But throughout that period, the vaccine effectiveness against culture confirmed the influenza ranged in the 70 percent range. In another study, the Doctor Carolyn Bridges published in JAMA. They were two years that were analyzed there, where one year there was very, very little vaccine effectiveness. And then in another year, were the matched was good there was a six percent. So, we know that each year there are three circulating strains and each of them have a variations in the effectiveness of the vaccine. And so, yes. We definitely see it wide range of effectiveness from year to year.
CURTIS ALLEN: The next question please.
OPERATOR: Dan DeNoon, WebMD, your line is open.
DAN DENOON, WEBMD: Thank you. I guess I'm understanding that the – these strain is one that's predominating now or seems to be predominating now. If so, is there – can you give some sense of the difference in severity between type B illness and type A illness, particularly the type AH3 illness?
DAN JERNIGAN: Right. I think if you – in the graph in the (INAUDIBLE) of ER, you can get a sense about what the magnitude difference is, but even though B is now one of the common strains, it overall will not become the predominant strain for the year. In general, H3N2 is associated with more severity overall and among children, B can also cause a fair amount of severe illness.
DENOON: Is B now a – seems to be predominating in the late part of the season?
DAN JERNIGAN: It's predominating but the overall numbers of viruses that are circulating is much lower than it was earlier in the year.
DAN DENOON: Thank you.
CURTIS ALLEN: The next question please. Next question?
OPERATOR: Denise Grady, New York Times, your line is open.
DENISE GRADY, NEW YORK TIMES: Thank you. I have about three question. I'll try to make them quick. One, is can you explain – do we have any understanding of why there was this mismatch and has this degree of mismatch occurred before? The next thing was you mentioned death percentages and pneumonia and I wonder if you could explain a little bit more of exactly what those percentages are and tell us something more about the increase in pneumonia cases. And then finally, with the vaccine effectiveness, I wonder if there is any other way to express that number. I mean if you say that it's 58 percent does that mean that if you vaccinated 100 people and expose them to the flu 58 would be protected? Thank you.
DAN JERNIGAN: Yeah, so let me talk about the degree of mismatch and so this year the influenza B viruses that have been evaluated in the laboratory have a very – they are very distinct in terms of the antigenic match and so those are just basically laboratory tests that are able to say how well the circulating strains match the strain that is in the vaccine. So, compared to the H3N2 that's circulating this year, the match is not quite as distant and some of that description is available in that Science article that came out today about what it means to have that kind of mismatch. In past years, there have been years where the match is not optimal. The – in the last 20 seasons, 16 have had good matches and there have been four that were less than optimal matches and so it does occur but in general the matches, they are not 100 percent perfect, but they are better than what we have this year.
In terms of the pneumonia and influenza mortality, that is a system where there are 122 cities where individuals report aggregate numbers of people. They report percent of deaths that are due to pneumonia and influenza and a description of this and some of the definitions are available in the MMWR. This information if presented and with that over time, over a five-year period we are able to identify what an expected base line of the percent of deaths due to pneumonia and influenza would be. With that base line then we can estimate what a significant increase above that, that base line would be. So what we are seeing this year is that there has been a significant increase above that base line for 13 weeks. That compares to the nine weeks of 2003/2004. So, in the last several years there have been this kind of increase in 2003/2004, in the year 2000, in 1996-97, and so those years are somewhat similar and that the estimates that we are making about severity are based on that.
In terms of other ways of describing the vaccine effectiveness I think the – one way is to say that the people that were in this study who were vaccinated were 58 percent less likely to have laboratory diagnosed H3N2 than those in the study that were not vaccinated. So, I think that may be the most direct way of saying that.
CURTIS ALLEN: Does that answer your question, Denise? OK. Next question please.
DENISH GRADY: Yeah.
OPERATOR: Mike Stobbe, Associated Press, your line is open.
CURTIS ALLEN: Mike.
MIKE STOBBE, ASSOCIATED PRESS: Hi, thanks for taking the question. Do we know what was the – did you say this already, what was the vaccine effectiveness in '03-'04 and 2000, and in ‘96-'97?
DAN JERNIGAN: In '03-'04, around 50 percent. I think the important point here is that this is the first year that we were able to do this kind of study within the season and so we don't have comparable data to the Marshfield approach from previously, but some studies did show around 50 percent in children.
CURTIS ALLEN: Next question please.
OPERATOR: Again, if you would like to ask a question, press star followed by one. The next is from Will Dunham from Reuters. Your line is open.
WILL DUNHAM: Yes, thanks. I'd like to just follow up from my previous question and ask one other. So, I had asked if there is on record in any of the previous flu vaccines a lower vaccine effectiveness than we have had this season? Is there on record a lower vaccine effectiveness than we have had in this season? And also do you know – is this the most poorly matched vaccine that you guys have seen?
DAN JERNIGAN: For the influenza B component there is no vaccine effectiveness I mean that there have been that in the past, I don't actually know the years it would be, but for the 1997-98 season, the vaccine effectiveness was essentially zero.
WILL DUNHAM: OK, and the other part of the question about is this the most poorly matched vaccine that you guys have had or I guess that one year '97-'98 will be worse.
DAN JERNIGAN: Once again remember for the B component the vaccine effectiveness is zero, but for the influenza A component it is 58 percent. So, it really depends on how you look at the effectiveness in terms of the subtypes that are causing infection.
WILL DUNHAM: Right. You guys use the 44 percent figure, so that's what I am using.
DAN JERNIGAN: But that is overall. But there have been years where in using a different study design they have shown vaccine effectiveness per year.
WILL DUNHAM: Obviously you have had years that are lower than 44 percent effectiveness?
DAN JERNIGAN: Yes, yes.
WILL DUNHAM: And is '97-'98 the most recent year?
DAN JERNIGAN: Yes.
WILL DUNHAM: So, this would be the lowest vaccine effectiveness in a decade?
DAN JERNIGAN: It's really hard. The information that we are providing here is from a study that was done through the season and so there are retrospective studies that have been done in certain populations and so what I can do is we can try and find that information and put that into a way that would be best explained for you.
WILL DUNHAM: But, the data that you have got from Marshfield is suggestive that this is the lowest vaccine effectiveness you have had in a decade.
DAN JERNIGAN: No, it is not. Overall vaccine effectiveness is 44 percent and it has been lower than that in '97-'98.
WILL DUNHAM: But, that's what I am saying, that's a decade ago. I am saying that based on the Marshfield findings that would be suggestive that this year was the lowest vaccine effectiveness since '97-'98, which is a decade ago.
DAN JERNIGAN: Right. In the Marshfield say they have found lower VE in the past, but again I think I don't have the information directed for you now and so we will have to get that back to you later.
CURTIS ALLEN: OK, next question please.
OPERATOR: Denise Grady, New York Times, your line is open.
DENISE GRADY: Thank you. I just wanted to follow up. Do you understand – does anyone understand why there was a mismatch this year? How does that occur when the guess work or the educated guessing is done to figure out what strains to put in? Where did things – it seems like something went wrong this year, is there any understanding of that? And I still would like to ask you again if there is another way to express the vaccine effectiveness? I understand they were 58 percent less likely to get the flu, but I still wonder if we can express that any other way. Could we talk about in terms of a 100 people being vaccinated and X number protected or something like that, that might be a little bit easier for some of us to understand? Thanks.
DAN JERNIGAN: Right. I think in terms of the vaccine mismatch, like I mentioned at the very beginning, the influenza viruses are continually changing and so the virus has to be chosen in terms of what would go into the vaccine much earlier than the season starts and so in order to do that individuals, WHO, and at the CDC and other places identify those viruses that are to be the – likely to be the best vaccine candidates. And that's based on their ability to grow in eggs and their ability to show protection with some of the studies that they do in the laboratory. So, with that information so much earlier in advance than the actual season, there is always a likelihood that you are going to have one of the viruses emerge that was not even available for testing at the time when the vaccine has to be decided. So, finding ways to move the time of decision to the production, we are trying to decrease that, is an important thing that we are working on, but at this point that is a limitation in the overall ability to get the vaccine each year to protect against influenza.
In terms of the way that you can describe this, in the past on the CDC website we had referred to language that the vaccine can reduce the chances of getting influenza by 58 percent. That is a way of saying, but I think it's important to remember that this particular estimate is based on a study population where they have a specific definition of what it means to get influenza infection and in the two group that are compared. So, it is an over-simplification, but to say that the vaccine can reduce the chances of getting the flu in this study population by 58 percent for influenza A is perhaps the most direct way to say it.
CURTIS ALLEN: Next question please.
OPERATOR: Robert Bazell, NBC News, your line is open.
ROBERT BAZELL, NBC NEWS: Hi, thanks for taking my question. I did want to follow-up on what orders Richard Knox – how Richard Knox began the questioning because these – there are these studies in Science and Nature which are large molecular studies about influenza transmission around the world. And they imply that the samples that you take in places other than that large of Asia that includes China seem to be meaningless for predicting the next season's strain types. And it is pretty – this is knowledge, obviously this wasn't something that was known before, if these papers are correct. And is – does it just mean that maybe you have been working in the wrong places I mean I know you've got a go you have got sophisticated labs and analysis, but shouldn't you be concentrating your surveillance on that area since these studies seems to conclude that that is where you are most likely to find next year's strains?
DAN JERNIGAN: Right. I think the conclusion of the study and we clearly agree with that that if this is the source of these infections then surveillance there would be most productive for us. I think one of the findings is that it takes about six to nine months for it to travel from there to other places in the U.S. – in North America. And so having surveillance here is clearly something that can pick up the viruses that are causing disease, but the indication or the suggestion from that study is that doing surveillance in that region may give you additional time to identify those viruses and pick your vaccine candidates and get the vaccine be made.
ROBERT BAZELL: Will that lead to a change in your procedures?
DAN JERNIGAN: Well, right now we have the four WHO collaborating laboratories around the world and so we have been – last few years and the emphasis on studying influenza, we have been able to increase the ability of those laboratories to do appropriate testing and so it is an area that we clearly see as an important place for improving infrastructure in the laboratories and CDC is actively involved in training, in providing personnel, and in providing resources through WHO and other agencies to improve surveillance and laboratory capabilities in those areas.
ROBERT BAZELL: And last, one last – please allow me one last thing, I am sorry. To get back to the 44 percent and 58 percent numbers, since nobody over 65 is included – what's the difference between them – please tell me – tell us again what the difference is between the 44 percent and the 58 percent numbers and if there is nobody over 65 in them aren't they kind of meaningless in terms of the very large spike in mortality that you did see this season?
DAN JERNIGAN: I think the –- and I would direct you to the contacts that are listed for – the folks from Marshfield's have done incredible study here for further follow up on some of those details, but in this particular study, in the population in that region there were both kinds of influenza circulating, influenza A H3N2, and influenza B, and so when you look at the ability for the vaccine to be effective in preventing both types of those where one of them there is zero effectiveness and one has a higher effectiveness, when you combine them you get the 44 percent.
If you look at just the A influenza H3N2 you get the 58 percent. And so the 44 percent is a reflection of having both influenza B and influenza A in the analysis.
CURTIS ALLEN: Does that answer you, Bob? Yeah, we have time for one more question. Next question please.
OPERATOR: Mike Stobbe, Associated Press, your line is open.
MIKE STOBBE: Thanks. Just a clarifying question. Earlier, you said regarding the '97-'98 vaccine that the effectiveness was essentially zero. Where you talking about the entire vaccine or the component that address B?
DAN JERNIGAN: The predominance that year was H3N2 and so it's – H3N2 was the thing that drove the lower vaccine effectiveness.
CURTIS ALLEN: Yeah, well Thank you very much for joining us today. If you have additional questions, you could call the Division of Media Relations at 404-639-3286. A transcript of this briefing will be on the website later this afternoon. Again, thank you very much and if you have questions please call 404-639-3286. Thanks.
END
Page last modified: April 17, 2008
