Rapid communications Effectiveness of seasonal 2010/11 and pandemic

This study provides mid-season estimates of the effectiveness of 2010/11 trivalent influenza vaccine and previous vaccination with monovalent influenza A(H1N1)2009 vaccine in preventing confirmed influenza A(H1N1)2009 infection in the United Kingdom in the 2010/11 season. The adjusted vaccine effectiveness was 34% (95% CI: -10 - 60%) if vaccinated only with monovalent vaccine in the 2009/10 season; 46% (95% CI: 7 - 69%) if vaccinated only with trivalent influenza vaccine in the 2010/11 season and 63% (95% CI: 37 - 78%) if vaccinated in both seasons.


Introduction
Following the emergence of pandemic influenza A(H1N1)2009 virus and the development of several monovalent pandemic influenza A(H1N1)2009 vaccines, a number of observational studies have since demonstrated the clinical effectiveness of these vaccines in various settings during the 2009/10 influenza A(H1N1)2009 pandemic [1][2][3].Uncertainty exists, however, about their duration of protection.
Vaccination with the 2010/11 northern hemisphere seasonal trivalent influenza vaccine, which includes the influenza A(H1N1)2009 strain, was started in autumn 2010.The United Kingdom (UK) target populations for vaccination were individuals aged six months to under 65 years in clinical risk groups at elevated risk of severe disease (including pregnant women) and individuals aged 65 years and over [4].Approximately 35% of those under 65 years of age in a clinical risk group had already received monovalent pandemic influenza vaccine in 2009/10 [4].
In the period December 2010-January 2011, the UK experienced widespread influenza A(H1N1)2009 transmission.Using the established swab-negative casecontrol approach in primary care [5,6], this study sets out to provide in-season interim estimates of the effectiveness of the 2010/11 seasonal influenza vaccine in preventing confirmed influenza infection in the UK in 2010/11 and the potential effect of previous vaccination with monovalent A(H1N1)2009 vaccine.

Study population and period
This study uses data from four influenza sentinel surveillance schemes in England, Scotland and Wales.Details of the Royal College of General Practitioners (RCGP), Health Protection Agency (HPA) Regional Microbiology Network (RMN) and Health Protection Scotland (HPS) swabbing schemes have been described previously [3].Public Health Wales operates a sentinel general practitioner (GP) swabbing scheme with 44 practices covering a population of 355,705, 12 per cent of the population in Wales.
This study covers samples collected in the period from 1 September 2010 to 11 January 2011.Cases were individuals presenting with an acute influenza-like illness (ILI) in a participating practice in the study period who were swabbed and tested positive for influenza regardless of type or subtype.ILI was defined as an acute respiratory illness with fever or complaint of feverishness.Controls were individuals presenting with ILI in the same period that were swabbed and tested negative for influenza.A standard specimen request form provided demographic and clinical information on cases and controls including date of birth, sex, risk group, date of onset of illness, date of specimen collection, influenza vaccination status for the current and previous season and vaccination dates.

Laboratory methods
Samples in England were sent to the HPA Microbiology Services (RCGP scheme) or one of the local HPA Regional laboratories (RMN scheme).Samples in Wales were sent to the Public Health Wales Specialist Virology Centre and in Scotland to the West of Scotland Specialist Virology Centre (HPS scheme) for molecular testing.Laboratory confirmation was undertaken using reverse transcription polymerase chain reaction (RT-PCR) assays for circulating influenza A viruses, influenza B viruses and other respiratory viruses [7,8].

Statistical methods
In order to assess vaccine effectiveness (VE) against influenza A(H1N1)2009 infection, a four-level variable was defined with the following four categories: Individuals were considered vaccinated if their date of seasonal or pandemic influenza A(H1N1)2009 vaccination was 14 days or more before the date of onset of illness.Persons for whom the interval between vaccination and onset of illness was less than 14 days were excluded, as their immunity status was considered unknown.If a person's trivalent vaccination status was known but not their pandemic influenza A(H1N1)2009 vaccination status or vice versa, they were excluded from the estimation of VE for influenza A(H1N1)2009 vaccine.For the estimation of VE for influenza A(H3) or B, pandemic vaccination status was not considered of interest.If the date of trivalent vaccination was missing, it was assumed that the person was vaccinated more than 14 days before the onset date, and for pandemic influenza A(H1N1)2009 vaccine it was assumed the person was vaccinated in 2009/10.
The same approach was used if date of onset was missing in a vaccinated individual.Respiratory samples with a delay greater than 29 days between onset of illness and sample collection were excluded as the sensitivity of the PCR test reduces for long intervals between onset and sampling.A sensitivity analysis was undertaken censoring at seven days between onset of illness and sample collection.
Vaccine effectiveness was estimated as 1-[odds ratio] using multivariable logistic regression models with influenza A(H1N1)2009 or influenza B PCR results as outcomes and seasonal or pandemic vaccination status

Results
This report has information on 4,554 individuals from whom samples were collected during the study period.Of these, 3,204 samples were collected through the RCGP surveillance scheme, 469 through the RMN scheme, 743 through the HPS scheme and 138 through the Public Health Wales scheme.
Those excluded from the study because of missing information (including PCR results and available vaccination history) are summarised in Table 1.Date of onset of illness was missing for 521 persons (11.4%): these were still included in the analyses.In the analyses evaluating VE in preventing influenza A(H1N1)2009 infection, samples positive for influenza A(H3) or influenza B were excluded and vice versa.There were therefore 2,891 persons for whom data on both vaccination status (for both vaccines) and pandemic influenza A(H1N1)2009 infection was available.Similarly, there were 2,487 persons included in the estimation of trivalent vaccine for prevention of influenza B or A(H3).
Table 2 shows the distribution and completeness of the baseline characteristics of the study participants according to whether they were influenza A (H1N1)2009 cases or controls.Age group, surveillance scheme and time period were found to be significantly associated with a confirmed influenza A(H1N1)2009 infection (Table 2).).Censoring samples taken more than seven days after symptom onset did not significantly change the VE estimates: the adjusted VE for those vaccinated last season was 44% (95% CI: 0 -68%), for those vaccinated only this season was 63% (95% CI: 32 -79%) and for those vaccinated both seasons was 64% (95% CI: 36 -80%).

Vaccine effectiveness in prevention of influenza A(H1N1)2009 infection
The adjustment for month had a large effect on the VE point estimate for the group vaccinated in 2009/10; it decreased from 62% (crude) to 34% after adjustment.This is because the number of people vaccinated in 2009/10 only decreases across months (whilst influenza A(H1N1)2009 incidence is increasing), whereas the number of people vaccinated in 2010/11 is increasing over time.
There was no evidence of significant effect modification of vaccine by age group (using the same five age groups, likelihood ratio test p=0.21),although some of the vaccine-age sub-groups did not have any PCR positive results among them.

Vaccine effectiveness in prevention of H3 or influenza B infection
Twenty-one of 216 persons vaccinated with trivalent influenza vaccine (9.7%) were positive for influenza B or A(H3) compared to 478 of 2,271 persons unvaccinated with trivalent influenza vaccine (21%).This gives a crude VE of 60% (95% CI: 36 -75%).If adjusted for age group, surveillance scheme and time period (month), adjusted VE was reduced to 50% (95% CI: 17 -70%).There was no evidence of significant age-vaccine interaction (likelihood ratio test p=0.37).

Discussion
The swab-negative case -control study design is an established approach to estimate influenza vaccine effectiveness.A number of studies have recently been published on the methodology [9,10].The potential limitations of the approach presented in this paper have been outlined previously and relate to convenience sampling; the potential for selection bias; missing data items and lack of information on risk status.The likely impact of each of these on VE estimates has been addressed earlier [3].
This This study provides some of the first evidence that this season's trivalent influenza vaccine is effective in reducing confirmed influenza A(H1N1)2009 and B infection in persons consulting in primary care.This level of protection is consistent with several studies undertaken with trivalent influenza vaccines in the prepandemic era and is congruent with moderately good matching between the vaccine and the circulating influenza strain [5,6].We found no evidence that protection was significantly different by age group; however it is likely that the study size was not sufficiently large to address this point specifically.
Although recently published work has demonstrated in several geographical settings, that the pandemic influenza A(H1N1)2009 vaccine was highly effective last season in preventing confirmed influenza A(H1N1)2009 infection that season [2,3], this study indicates that pandemic vaccine protection may not last across seasons.This corroborates recent findings from a longitudinal sero-epidemiological survey, which suggests that population A(H1N1)2009 antibody levels may start to reduce in the post-pandemic period, particularly in the 5-14-years old age-band [11].Further work needs to be undertaken in this area.Our paper does suggest that within the data available at present there is a doseresponse relationship and, that vaccination with this In conclusion, this study undertaken mid-season provides evidence that this season's trivalent influenza vaccine does provide protection against infection to both strains of influenza circulating this season (A(H1N1)2009 and influenza B) in Europe.It is important to note that more precision in this estimate will be available at the end of the season.The findings seem to provide some of the first published evidence that protection might wane following vaccination with influenza A(H1N1)2009 vaccine after 12 months and reinforces the recommendation that annual re-immunisation of target groups is required regardless of vaccination the previous season (including those vaccinated with an adjuvanted vaccine).
study demonstrates three key findings: vaccination with this current season's trivalent influenza vaccine provides protection against both confirmed influenza A(H1N1)2009 and influenza B infection and immunisation with A(H1N1)2009 vaccine in 2009/10 followed by trivalent influenza vaccine this season provides better protection against confirmed influenza A(H1N1)2009 infection.Finally vaccination only last season with A(H1N1)2009 vaccine, seems to provide the least protection against confirmed influenza A(H1N1)2009 infection.
season's trivalent influenza vaccine of individuals who have already received monovalent A(H1N1)2009 vaccine last season produced the highest effectiveness compared to vaccination only in the 2010/11 season or vaccination with A(H1N1)2009 vaccine alone in the 2009/10 season.This reinforces the importance of the UK policy for vaccination of those who had received the monovalent vaccine in the previous season.

Table 1
Inclusion and exclusion criteria of participants for specimens submitted, United Kingdom, 1 September 2010 -11 January 2011 a Including eight people with sample taken later than 29 days after onset of illness.PCR: Polymerase chain reaction.

Table 2
Details for pandemic influenza A(H1N1)2009 cases and controls, United Kingdom, September 2010 -January 2011 (n=3,480) a HPS: Health Protection Scotland; RCGP: Royal College of General Practitioners' surveillance scheme; RMN: Health Protection Agency Regional Microbiology Network.a Includes those with missing vaccination history and/or interval from onset of illness to sample longer than 29 days.

Table 3
Number and proportion of samples positive for influenza A(H1N1)2009 according to vaccination status, United Kingdom, September 2010 -January 2011 a Chi-square test p<0.001 on three degrees of freedom.