j our na l ho me page: www.elsev ier .com/ locate /vacc ine
he effectiveness of seasonal trivalent inactivated influenza vaccine inreventing laboratory confirmed influenza hospitalisations inuckland, New Zealand in 2012
ikki Turnera,∗, Nevil Pierseb, Ange Bissieloc, Q Sue Huangc, Michael G. Bakerb,arc-Alain Widdowsond, Heath Kellye,f, on behalf of the SHIVERS investigation team1
The University of Auckland, Private Bag 92019, Victoria St West, Auckland, New ZealandThe University of Otago, PO Box 7343 Wellington South 6242, Wellington, New ZealandInstitute of Environmental Science and Research, PO Box 40-158 Upper Hutt 5140, Wellington, New ZealandCenters for Disease Control and Prevention, Atlanta, GA 30333, USAThe Australian National University, Canberra 0200, ACT, AustraliaVictorian Infectious Diseases Reference Laboratory, 10 Wrecklyn St., North Melbourne, 3051 Melbourne, VIC, Australia
r t i c l e i n f o
rticle history:eceived 5 November 2013eceived in revised form 24 March 2014ccepted 2 April 2014vailable online xxx
eywords:nfluenza vaccineaccination
mmunisationaccine effectiveness
a b s t r a c t
Background: Few studies report the effectiveness of trivalent inactivated influenza vaccine (TIV) in pre-venting hospitalisation for influenza-confirmed respiratory infections. Using a prospective surveillanceplatform, this study reports the first such estimate from a well-defined ethnically diverse population inNew Zealand (NZ).Methods: A case test-negative design was used to estimate propensity adjusted vaccine effectiveness.Patients with a severe acute respiratory infection (SARI), defined as a patient of any age requiring hos-pitalisation with a history of a fever or a measured temperature ≥38 ◦C and cough and onset within thepast 7 days, admitted to public hospitals in South and Central Auckland were eligible for inclusion in thestudy. Cases were SARI patients who tested positive for influenza, while non-cases (controls) were SARIpatients who tested negative. Results were adjusted for the propensity to be vaccinated and the timingof the influenza season.Results: The propensity and season adjusted vaccine effectiveness (VE) was estimated as 39% (95% CI16;56). The VE point estimate against influenza A (H1N1) was lower than for influenza B or influenza
A (H3N2) but confidence intervals were wide and overlapping. Estimated VE was 59% (95% CI 26;77) inpatients aged 45–64 years but only 8% (−78;53) in those aged 65 years and above.Conclusion: Prospective surveillance for SARI has been successfully established in NZ. This study for thefirst year, the 2012 influenza season, has shown low to moderate protection by TIV against influenzapositive hospitalisation.
Please cite this article in press as: Turner N, et al. The effcine in preventing laboratory confirmed influenza hospitalisathttp://dx.doi.org/10.1016/j.vaccine.2014.04.013
1 Authors in the SHIVERS investigation team: Don Bandaranayake, Jazmin Duque,ameron C. Grant, Diane Gross, Lyndsay LeComte, Graham Mackereth, ColincArthur, Sarah Radke, Sally Roberts, Ruth Seeds, Susan Taylor, Paul Thomas, Mark
hompson, Adrian Trenholme, Richard Webby, Deborah A. Williamson, Conroyong, Tim Wood.
Influenza continues to cause a significant burden of illness inadults and children [1,2] despite vaccines having been used inter-nationally for more than 60 years and being recommended by theWorld Health Organization [3]. Estimates of efficacy (from trials)and effectiveness (from observational studies) for seasonal triva-lent inactivated vaccine (TIV) have been variable. An umbrellareview of meta-analyses of community studies from 2005 to 2011
ectiveness of seasonal trivalent inactivated influenza vac-ions in Auckland, New Zealand in 2012. Vaccine (2014),
concluded that protection against laboratory-confirmed influenza(largely mild disease) by TIV ranged from 59 to 65% with esti-mates being similar in working age adults and children aged 2 yearsand above [4]. There have been too few trials in children under
years for accurate estimates of efficacy in this age group [5,6].bservational studies provide a range of effectiveness estimates
rom zero to approximately 60% protection in young children [7,8].hile studies specifically of older adults are less common, vaccine
ffectiveness (VE) has been reported to be as high as 57% in adultsver 70 years [6], there are significant concerns over bias in stud-es in this age group [9] and other studies report much lower orven null estimates [10]. However significant variability by sea-on is acknowledged [6] and increasing immunosenescence andhe presence of comorbidities are likely to reduce effectiveness [6].
Results are more limited when reviewing protection againstnfluenza-confirmed hospitalisation. No trials address this out-ome. Estimates from observational studies include no protectiony TIV against laboratory-confirmed influenza [11] to a protectiveange of 49% to 61% in adults [12–14]. Pooled European data for VEgainst A (H3N2) during 2011/2012 gave a point estimate for thearget groups for vaccination of 29% with wide confidence intervals15].
The antigenic composition of influenza vaccines is reviewednnually to predict the best match for a constantly evolving virus.he impact of vaccination is expected to be higher in the presencef a good antigenic match, although significant effectiveness haseen shown even in seasons when the circulating strain is not aood match [16,17].
In New Zealand (NZ) seasonal unadjuvanted TIV is offered annu-lly free of charge to all adults aged 65 years and over, pregnantomen and all those over 6 months of age with chronic medical
onditions that are likely to increase severity of infection. The vac-ines are also available from early March on the private market forll others over 6 months of age. The influenza season usually occursomewhere between early May and late September.
Using a case test-negative design (a modification to the case-ontrol study design [18]), we aimed to estimate the effectivenessf seasonal TIVs in preventing hospitalised laboratory-confirmednfluenza in persons aged at least 6 months who were admitted
ith an acute respiratory illness to public hospitals in Central, Southnd East Auckland between April 2012 and February 2013. Thetudy reports results from the first year of a five year SHIVERSSouthern Hemisphere Influenza Vaccine Effectiveness, Researchnd Surveillance) project.
. Methods
Ethics approval for the study was obtained from the Northern Aealth and Disability Ethics Committee (NTX/11/11/102 AM02).
. Study design
We used the standard case test-negative design [19] and a sim-lar analytic approach to a previous study of hospitalised patients,
ith adjustment for the propensity to be vaccinated [13]. From0 April 2012 to 28th February 2013 we attempted to enrol all
ndividuals aged 6 months and older who were hospitalised with severe acute respiratory infection (SARI). Based on the Worldealth Organization definition, this was defined as a patient requir-
ng hospitalisation with a patient-reported history of a fever or aeasured temperature ≥38 ◦C, cough and onset within the past 7
ays [20].A confirmed case of hospitalised influenza was defined as a SARI
atient with a positive laboratory result for any influenza virusetected by real time reverse transcription polymerase chain reac-
Please cite this article in press as: Turner N, et al. The effcine in preventing laboratory confirmed influenza hospitalisathttp://dx.doi.org/10.1016/j.vaccine.2014.04.013
ion (RT-PCR) or viral isolation, while non-cases (controls) werehose who tested negative to all influenza viruses.
Eligible patients were those admitted to the public hospitalsiddlemore, Kidz First Children’s, Auckland City and Starship
PRESSxx (2014) xxx–xxx
Children’s which together serve a population of approximately838,000 people in Central, South and East Auckland. Recruitmentwas undertaken by trained research nurses. The nurses recruitedpatients during the day and screened all overnight admissions offebrile patients with respiratory symptoms daily from Monday toSaturday. Sunday admissions were captured on Mondays if thepatients were still hospitalised.
All identified SARI cases who gave verbal consent completed acase report form, administered by a research nurse, and provideda nasopharyngeal swab or aspirate for influenza testing by RT-PCRand/or viral isolation.
Excluded from the analysis were patients transferred fromanother hospital, those see outside the influenza season, childrenunder 6 months of age, patients who had not provided consent,patients with incomplete data for vaccination status or age, orpatients who were swabbed more than 7 days after the onset ofsymptoms. At the end of the season, people with multiple SARIhospitalisations were excluded if their vaccination status differedbetween hospitalisations, otherwise the first influenza positiveadmission was used. Only the first in season hospital admission wasused if a person had multiple admissions but no influenza positiveadmission.
4. Participant information
Demographic data collected for all cases and non-cases includedage; sex; ethnicity (Maori, Pacific, Asian, NZ European or other);and income, with low income defined as a household that receivedeither a government benefit or held a community services card. Theage data were cross validated with hospital held electronic data.Clinical information was obtained from both the case report formand electronic data extraction from hospital databases. These dataincluded clinical symptoms and signs; influenza vaccination statusrecorded on the case report form; smoking status; body mass indexbased on either measured weight and height or a visual estimationby the research nurses (using the categories obese, overweight,normal weight, underweight or unsure); a patient or caregiverreported measure of dependence, assessing requirement for assis-tance with normal activity or full dependency on nursing care;a simple frailty measure based on use of long term oxygen; anychronic medical conditions; and a self-defined health status scoreusing the general health question from the SF36 [21] and combiningfair or poor versus all others (the SF-36 is a generic, multi-purpose,short-form health survey that generates a functional health andwell-being score).
The chronic medical conditions examined were the following:asthma, with the need for preventative therapy; diabetes; chronicobstructive pulmonary disease (COPD); other chronic lung disease;cardiac disease; cerebrovascular disease; moderate to severe cog-nitive impairment; other chronic neurological disease; psychiatricdisorder (psychotic or major affective disorder); current alcohol ordrug dependence; active cancer (excluding non-invasive skin can-cer); immune deficiency condition (including asplenia, HIV/AIDS);immune suppressive treatment; chronic renal disease; or chronicliver disease (including cirrhosis, chronic hepatitis, transplant).
Vaccination status was recorded as fully vaccinated if the patientor caregiver reported influenza vaccination given during the cur-rent season at least 14 days prior to the onset of symptoms forwhich they were hospitalised. All children less than 9 years of agewere recorded as fully vaccinated if they had received a vaccinationin the season at least 14 days prior to onset of symptoms, and had
ectiveness of seasonal trivalent inactivated influenza vac-ions in Auckland, New Zealand in 2012. Vaccine (2014),
ever received an earlier vaccine at least one month prior to the cur-rent vaccine. Children under 9 years of age who had received onlyone dose of vaccine in the season and no previous vaccine were con-sidered partially vaccinated but were analysed as non-vaccinated.
Two commercial vaccine products were available on the NZarket in 2012: Fluarix® (GlaxoSmithKline) approved for used in
eople 6 months and over, and Fluvax® (bioCSL) approved for usen people aged 5 years and over, but recommended to be used
ith caution in children aged 5–8 years. Both vaccines contained/California/7/2009 (H1N1)-like strain, A/Perth/16/2009 (H3N2)-
ike strain and B/Brisbane/60/2008-like strain (belonging to the/Victoria lineage).
. Laboratory methods
Nasopharyngeal swabs were collected with a COPAN flockedwab and transported in viral transport medium (VTM) at 4 ◦C.asopharyngeal aspirates and other respiratory samples wereollected according to hospital standard operational procedures.espiratory samples were tested using the United States Center forisease Control and Prevention real time RT-PCR protocol [22] atuckland District Health Board Laboratory and the AusDiagnosticCR protocol at the Counties Manukau District Health Board labo-atory [23]. All influenza positive PCR cases were forwarded to theational Influenza Centre on a weekly basis for antigenic typing.T-PCR assays detected influenza virus types A and B and subtypingas performed for A subtypes. A small proportion of cases (10.4%)ere subjected to viral isolation by inoculation into Madin–Darby
anine kidney (MDCK) cells.
. Statistical analysis
The influenza season was defined to start when there were twoonsecutive weeks of four or more cases and to end when thereere no consecutive weeks of four or more cases. Patients at higher
isk of an adverse outcome from influenza infection, and for whomhe vaccine is provided at no charge, may be more likely to be vac-inated. We allowed for this by using a multivariate logistic modelo calculate the propensity to get vaccinated for the controls, givenhe range of available patient characteristics. The results from theropensity model are presented as odds ratios (OR) and were usedo adjust the VE estimate.
In the primary analysis VE% was defined as 100 times (1-R) with the OR derived from logistic regression models where
nfluenza was the outcome variable. We calculated the crude VE%djusting only for the timing of the admission relative to the peakf the influenza season and the adjusted VE which included bothhe timing of the admission and the estimated propensity for vac-ination. This approach has previously been used by Talbot et al.13].
The season was defined to include continuous weeks with ateast two laboratory-confirmed influenza cases. It began on week
of the study (May 29, 2012) and ended on week 24 (October 22,012). We present a summary VE estimate against all influenza
nfections and separate stratified estimates for six age ranges (6–5 years, 6–17, 18–44, 45–64, 65–79, 80+ years) and influenza A
H1N1) pdm09, A (H3N2) and influenza B. VE estimates for typesnd subtypes used cases positive for the specific type or subtypeompared with influenza negative non-cases.
For all patient characteristics, other than age and vaccinationtatus, missing values were imputed as the modal baseline casef non-Maori, non-Pacific ethnicity, female, not low income, notregnant, non-smoker, without chronic disease, not obese, withelf-rated health average or better, not on long term oxygen usend living without assistance. Sensitivity analyses were performed
Please cite this article in press as: Turner N, et al. The effcine in preventing laboratory confirmed influenza hospitalisathttp://dx.doi.org/10.1016/j.vaccine.2014.04.013
xcluding individuals with missing data and samples tested onlyy viral isolation. We also compared results from the propen-ity adjusted model with an epidemiological model where theame covariates were forced into a logistic regression model with
PRESSxx (2014) xxx–xxx 3
influenza as the outcome and vaccination as the primary exposurevariable. We further constructed a statistical model where only thevariables in the epidemiological model that were significant at thelevel of P < 0.05 in univariate analysis were included in the logisticregression model. Assuming VE was 50% and vaccine coverage inthe controls was 20%, we estimated 207 cases gave 80% power todetect a significant protection due to vaccine.
7. Results
The influenza season ran from the 28th of May 2012 until the10th of October 2012. Case selection and exclusion/inclusion crite-ria are shown in Fig. 1. Of the 6373 admissions screened, 2682(42%) met the definition of SARI. After exclusions for lack of consent(n = 224), no record of vaccination history (n = 300), no recordeddate of birth (n = 1), aged less than 6 months (n = 226), outside ofinfluenza season (n = 484) or no laboratory results available (n = 88),a total of 1359 admissions remained, of whom 382 (21%) wereinfluenza positive (Fig. 2). Excluding multiple admissions, 1329SARI patients were included in this analysis. Of the 353 (27%) whowere influenza positive, 112 (32%) were vaccinated and of the976 who tested negative for influenza, 385 (39%) were vaccinated(Fig. 1). There were 88 samples where no laboratory results wereavailable because we were unable to obtain or analyse the sample.Most (n = 49) were unvaccinated children under the age of 5 yearsand, of the remaining 39 more than five years of age, 14 (36%) werevaccinated.
The 353 influenza positive cases and 976 influenza negativenon-cases were compared across a range of patient characteristics.Patients who were influenza positive were more likely to be unvac-cinated, younger, of Pacific ethnicity, not on long term oxygen, torequire assistance with daily living and to be admitted during theinfluenza season. There were no statistically significant differencesby gender, income, pregnancy, smoking, presence of a chronic dis-ease, obesity or self-rated health (Table 1). The adjusted odds ratiosfor the association of various patient characteristics with likeli-hood of vaccination showed that older age groups, those with otherlung diseases (excluding COPD and asthma), cardiovascular diseaseor on long term oxygen were more likely, and those with cancerwere less likely, to be vaccinated (Table 1). In contrast, there wasno statistically significant difference in the likelihood of vaccina-tion by sex, asthma, diabetes, cognitive impairment, psychiatricdisorder, COPD, other neurological disease, immune deficiency,immune suppression, renal, liver, alcohol or drug dependency,pregnancy, smoking, obesity, self-rated health and living in a resthome (Table 1).
8. Vaccine effectiveness
The VE against any influenza infection and adjusted only forthe timing relative to the peak of season was 27% (95% confidenceinterval 6;44). After adjusting for the propensity to be vaccinatedthe estimated VE was 39% (16;56). In the sensitivity analysis, theVE against any influenza infection estimated from the epidemi-ological model was 42% (19;58) and from the statistical modelwas 41% (21;58). There was no significant change to these esti-mates when omitting patients with missing values or for whominfluenza was tested only by viral isolation (data not shown). VE forpatients aged 45 to 64 years was 59% (26;77) but only 8 (−78;53)in patients aged 65 to 79 years. The VE estimate against influenzaA (H1N1) was lower than the point estimates for influenza B and
ectiveness of seasonal trivalent inactivated influenza vac-ions in Auckland, New Zealand in 2012. Vaccine (2014),
influenza A (H3N2) but confidence intervals were wide and over-lapping (Table 2).
The vaccine formulations used in New Zealand in 2012 includeda strain that matched the circulating 2009 A (H1N1) pdm09 strain
Please cite this article in press as: Turner N, et al. The effectiveness of seasonal trivalent inactivated influenza vac-cine in preventing laboratory confirmed influenza hospitalisations in Auckland, New Zealand in 2012. Vaccine (2014),http://dx.doi.org/10.1016/j.vaccine.2014.04.013
ARTICLE IN PRESSG ModelJVAC-15294; No. of Pages 7
4 N. Turner et al. / Vaccine xxx (2014) xxx–xxx
Fig. 1. Flowchart of all selected, recruited, eligible, complete and unique admissions for VE analysis.
0
20
40
60
80
100
120
140
18
20
22
24
26
28
30
32
34
36
38
40
42
44
46
48
50
52 2 4 6
May Jun Jul Aug Sep Oct Nov Dec Jan Feb
Nu
mb
er o
f S
AR
I c
as
es
Week of admissi on
SARI cases - inf luenza po sitive
SARI cases -all oth ers
Fig. 2. Weekly SARI cases including influenza PCR positive cases during 30 April 2012 to 28 February 2013 (NB pre-season is before week 5; 29 May 2012; Post season isafter week 24; 22 October 2012).
NB 1. Adjusted odds ratio compared to referent group: female, aged 46 to 64 years, non-Maori non-Pacific ethnicity, not low income, not pregnant, non-smoker, withoutchronic disease, not obese, with self-rated health average or better, not on long term oxygen use and living without assistance.N nsity
w her ned and l
w2rcMewcat(
9
2dswlitscAc
TE
fl
B 2. The only factors listed in the table are those that were significant for the propeere sex, asthma, diabetes, cognitive impairment, psychiatric disorder, COPD, otisease, alcohol or drug dependency, pregnancy, smoking, obesity, self-rated health
hich represented 10% (247/2425) of all viruses detected in NZ in012. VE against this strain was 29% (−26;60). Influenza B virus rep-esented 13% (306/2425) of all viruses detected with both B lineageso-circulating. The 2012 vaccine strain was a B/Victoria lineage.ore B/Yamagata lineage viruses (84%, 99/118) than B/Victoria lin-
age viruses (16%, 19/118) were detected. VE against influenza Bas 47% (1;72). There were too few data to calculate B lineage spe-
ific VE estimates. Influenza A (H3N2) viruses represented 65% ofll viruses detected. Most A (H3N2) viruses had drifted away fromhe A/Perth/16/2009 vaccine strain (data not shown). VE against AH3N2) was 46% (16;66).
. Discussion
This study has four important findings. First, we found that1% of SARI admissions in Auckland during the study period wereue to influenza. Second, VE against hospitalisation for the 2012eason was relatively low with a point estimate of 39%. Third,hile underpowered to show age effects, vaccination appeared
ess effective in patients aged at least 65 years although this find-ng may reflect response to the circulating influenza type ratherhan an age effect, and lastly VE varied by influenza type and
Please cite this article in press as: Turner N, et al. The effcine in preventing laboratory confirmed influenza hospitalisathttp://dx.doi.org/10.1016/j.vaccine.2014.04.013
ubtype. Specifically, older people did not appear to be signifi-antly protected against hospitalisation for infection with influenza
(H3N2). Studies of hospitalised patients using a laboratory-onfirmed endpoint have reported a wide range of results for VE.
able 2stimated vaccine effectiveness, overall by age range and by sub-strain, for the crude and
6 m to 5 1 18 64 256 to 17 2 8 20 318 to 44 16 53 73 1345 to 64 25 100 54 964 to 79 42 133 18 580+ 26 73 12 2
* Flu+ = influenza detected, Flu− = influenza not detected. LCL = 95% lower confidence lu season.
to be vaccinated; other factors included in the propensity model but not significanturological disease, immune deficiency, immune suppression, renal disease, liveriving in a rest home.
Our point estimate of 39% is within this range. Using a case test-negative design over two seasons, Puig-Barbera and colleagueswere unable to show a significant protective effect from seasonalvaccine for adults, although adjusted VE for monovalent A (H1N1)pandemic provided higher protection [11]. Talbot et al. studiedadults 50 years and older hospitalised in the US and found a VEestimate of 61% (18;82) [13]. A case test-negative study in the2010/2011 season in Spain showed VE of 58% (20;89) against hos-pitalised laboratory-confirmed influenza in adults aged 52-to-84years, although VE rose to 68% (13;77) for those who had receivedboth the seasonal and pandemic A (H1N1) monovalent vaccine[24]. A more recent Spanish study of hospitalised patients usingthree different control groups showed similar vaccine effective-ness of 60% for all ages but higher effectiveness (89%) againstsevere cases [25]. In the southern hemisphere, Cheng et al. reportedVE against influenza hospitalisation of 49% (13;70) for adultsaged 18 years and older in the 2010 Australian influenza season[14].
The lower VE results from our study are not dissimilar to apopulation-based study by Baxter et al. to determine the associa-tion between hospitalisation and prior vaccination over an 11 yearperiod. The study examined VE within the season, compared to out-
ectiveness of seasonal trivalent inactivated influenza vac-ions in Auckland, New Zealand in 2012. Vaccine (2014),
side the season, using a “difference-in-differences” approach [26].Estimated reduction in influenza-attributable hospitalisations frominfluenza vaccination were 28% (9;30) for adults 50–64 years and48% (12;26) for those 65 years and over.
Similar to our study findings, low to moderate effectivenessnd variations by age and type/subtype have been identified ashallenges with seasonal influenza vaccines [27]. In the 2012anish influenza season no significant protection against influenza-onfirmed hospitalisation for patients aged 65 years and older waseen for A (H3N2), whereas VE in the same age group for influenza
was 69% (26;87) [28].In light of the suggestion that the vaccine may be less effective
n the elderly the current policy of individual protection for thelderly may be a limited strategy; other strategies may need to beurther considered such as ring protection and herd immunity toetter protect the vulnerable elderly.
0. Strengths and limitations
The strengths of this study include the establishment of anffective influenza surveillance system in New Zealand, and usingT-PCR-confirmed hospitalised influenza as the outcome measure.stimating VE against the severe outcome of hospitalisation mayave more relevance for public health policy than estimation ofrotection against mild disease that is managed in the commu-ity. The recruitment process selected potential SARI cases from a
ull range of respiratory illness categories including both acute andhronic illnesses. However we would not have captured influenzaelated hospitalisations that presented as a non-respiratoryllness such as patients with exacerbation of cardiovascularisease.
We collected a large range of variables that could be includedn the statistical model. However adjustment for the differencesetween vaccinated and unvaccinated study participants may haveiffered in ways that had not been considered or were difficult toeasure. Furthermore, little is known about the 220 individualsith SARI who did not give informed consent. This was mostly due
o language barriers. It is possible that people for whom English isot a first language may have greater challenges in accessing healthare services and therefore more likely to be under-vaccinated. Aurther study weakness was the record of vaccination status, basednly on patient recall. However self-report has been shown to beenerally accurate in hospitalised elderly [29] and when comparingelf-report with documented status VE estimates are very similar30]. Future studies will use verified vaccinations based on providerecords. The influenza laboratory test has some limitations. Thelinical sensitivity for all targets (influenza A, influenza B, A (H1N1)dm09 and A (H3N2) is greater than 93% and the clinical specificityor all targets is greater than 90% [31]. Any resulting misclassifica-ion can be expected to have biased our estimates towards the null19] and this could be more likely in hospitalised patients where anpper respiratory tract specimen is used to diagnose lower respi-atory tract disease.
Case test-negative designs may have other biases that weave not accounted for in this study, in particular differences inealth care-seeking behaviour between cases and non-cases affect-
ng vaccine status, possible transient non-specific immunity (viralnterference) following an acute respiratory tract infection and pos-ible modification of symptomatic illness by vaccination status mayll bias the VE estimates. Furthermore, this study design is not ableo make any comment on the effectiveness of vaccination on hos-italisation based on complications of influenza rather than acuteresentations [18].
In summary, we have established prospective surveillance forARI in New Zealand and have shown low to moderate protection
Please cite this article in press as: Turner N, et al. The effcine in preventing laboratory confirmed influenza hospitalisathttp://dx.doi.org/10.1016/j.vaccine.2014.04.013
rom TIV in influenza-positive SARI patients in the 2012 influenzaeason. In future years, surveillance will include patients seen inentinel general practices as well as those admitted to hospital ande will be able to provide a more complete picture of the burden of
PRESSxx (2014) xxx–xxx
influenza in a well-defined ethnically diverse community. In par-ticular, we will be able to compare the effectiveness of influenzavaccines in preventing attendance at a general practitioner forlaboratory-confirmed influenza with effectiveness against the, pos-sibly more policy relevant, outcome of admission to hospital.
Acknowledgements
The SHIVERS (Southern Hemisphere Influenza and VaccineEffectiveness Research and Surveillance) project is funded byU.S. Department of Health and Human Services, Centers for Dis-ease Control and Prevention (CDC) (1U01IP000480-01). The SARIsurveillance is a key component of the SHIVERS project. The projectis a five year research cooperative agreement between Institute ofEnvironmental Science and Research and US CDC’s National Centerfor Immunization and Respiratory Diseases (NCIRD) Influenza Divi-sion. The SHIVERS project is a multi-centre and multi-disciplinarycollaboration. Special thanks go to these collaborating organisa-tions for their commitment and supports: ESR, Auckland DistrictHealth Board, Counties Manukau District Health Board, Universityof Otago, University of Auckland, the US Centres for Disease Controland Prevention and WHO Collaborating Centre at St Jude Children’sHospital in Memphis, USA.
Special thanks to: The research nurses at Auckland DistrictHealth Board (ADHB): Kathryn Haven, Debbie Aley, Pamela Mupon-isi, Bhamita Chand, Yan Chen, Laurel Plewes, Frann Sawtell;The research nurses at Counties Manukau District Health Board(CMDHB): Jo Smith, Shirley Lawrence, Franie Gravidez, Mandy Ma,Shona Chamberlin, Kirstin Davey, Tania Knowles, Jo-Ann McLeish;The WHO National Influenza Centre, Institute of Environmental Sci-ence and Research (ESR): A. Todd, R. Hall, D. de Joux, J. Bocacao,J. Ralston, W. Gunn, P. Kawakami, S. Walker, N. Moore, B. Waite;The Health Intelligence Team, ESR: L. Lopez; The ADHB Laboratory:Fahimeh Rahnama, Naeem Amiry, Wikke Koopmans; The CMDHBLaboratory: Helen Qiao, Fifi Tse, Mahtab Zibaei, Tirzah Korrapadu,Louise Optland, Cecilia Dela Cruz. Special thanks to IT staff in var-ious organisations and SARI surveillance participants. Also to Dr.Dean Erdman from Gastroenteritis and Respiratory Viruses Labo-ratory Branch, the U.S. Centers for Disease Control and Preventionwho provided the real-time PCR assay for non-influenza respiratoryviruses. Support in kind is provided by the Ministry of Health.
Disclaimer: The findings and conclusions in this report are thoseof the authors and do not necessarily represent the views of theCenters for Disease Control and Prevention.
References
[1] Nair H, et al. Global burden of respiratory infections due to seasonalinfluenza in young children: a systematic review and meta-analysis. Lancet2011;378(9807):1917–30.
[2] Carrat F, et al. Influenza burden of illness: estimates from a national prospectivesurvey of household contacts in France. Arch Intern Med 2002;162(16):p1842.
[3] World Health Organization. WHO vaccine use. World Health Organization;2013. Available from: 〈http://www.who.int/influenza/vaccines/use/en/〉 [cited2013 March 28].
[4] Manzoli L, et al. Effectiveness and harms of seasonal and pandemic influenzavaccines in children, adults and elderly: a critical review and re-analysis of 15meta-analyses. Hum Vaccine Immunother 2012;8(7):851–62.
[5] Jefferson T, et al. Vaccines for preventing influenza in healthy children.Cochrane Database Syst Rev 2008;2(2.).
[6] Osterholm MT, et al. Efficacy and effectiveness of influenza vaccines: a system-atic review and metaanalysis. Lancet Infect Dis 2011;12(1):36–44.
[7] Hoberman A, et al. Effectiveness of inactivated influenza vaccine in preven-ting acute otitis media in young children: a randomized controlled trial. JAMA2003;290(12):1608–16.
ectiveness of seasonal trivalent inactivated influenza vac-ions in Auckland, New Zealand in 2012. Vaccine (2014),
[8] Shuler CM, et al. Vaccine effectiveness against medically attended: laboratory-confirmed influenza among children aged 6 to 59 months, 2003–2004.Pediatrics 2007;119(3):e587–95.
[9] Baxter R, Lee J, Fireman B. Evidence of bias in studies of influenza vaccineeffectiveness in elderly patients. J Infect Dis 2010;201(2):186–9.
10] Jefferson T, et al. Vaccines for preventing influenza in the elderly.Cochrane Database of Systematic Reviews 2010, http://dx.doi.org/10.1002/14651858.CD004876.pub3. Issue 2. Art. No.: CD004876.
11] Puig-Barbera J, et al. Effectiveness of seasonal 2009–2009, 2009–2010 and pan-demic vaccines, to prevent influenza hospitalizations during the autumn 2009influenza pnademic wave in Castellon, Spain. A test-negative, hospital-based,case-control study. Vaccine 2010;28(47):7460–7.
12] Eisenberg KW, et al. Vaccine effectiveness against laboratory-confirmedinfluenza in children 6 to 59 months of age during the 2003–2004 and2004–2005 influenza seasons. Pediatrics 2008;122(5):911–9.
13] Talbot H, et al. Effectiveness of seasonal vaccine in preventing confirmedinfluenza-associated hospitalizations in community dwelling older adults. JInfect Dis 2011;203(4):500–8.
14] Cheng AC, et al. Effectiveness of H1N1/09 monovalent and trivalentinfluenza vaccines against hospitalization with laboratory-confirmed H1N1/09influenza in Australia: a test-negative case control. Vaccine 2011;29:7320–5.
15] Rondy M, et al. 2011–12 Seasonal influenza vaccines effectiveness against con-firmed A (H3N2) influenza hospitalisation: pooled analysis from a EuropeanNetwork of Hospitals. A pilot study. PloS One 2013;8(4):e59681.
16] Kelly HA, et al. Moderate influenza vaccine effectiveness with variable effec-tiveness by match between circulating and vaccine strains in Australianadults aged 20–64 years, 2007–2011. Influenza and other respiratory viruses2013;7(5):729–37.
17] Tricco AC, et al. Comparing influenza vaccine efficacy against mismatchedand matched strains: a systematic review and meta-analysis. BMC Med2013;11(1):153.
18] Foppa IM, et al. The case test-negative design for studies of the effectiveness ofinfluenza vaccine. Vaccine 2013;31(30):3104–9.
Please cite this article in press as: Turner N, et al. The effcine in preventing laboratory confirmed influenza hospitalisathttp://dx.doi.org/10.1016/j.vaccine.2014.04.013
19] Orenstein EW, et al. Methodologic issues regarding the use of three observa-tional study designs to assess influenza vaccine effectiveness. Int J Epidemiol2007;36(3):623–31.
20] World Health Organization. Interim epidemiological surveillance standardsfor influenza. World Health Organization; 2012. Available from: 〈http://www.
21] Jenkinson C, Coulter A, Wright L. Short form 36 (SF36) health survey question-naire: normative data for adults of working age. BMJ: Br Med J 1993;306(6890):1437.
22] Shu B, et al. Design and performance of the CDC real-time reverse transcriptasePCR swine flu panel for detection of 2009 A (H1N1) pandemic influenza virus.J Clin Microbiol 2011;49(7):2614–9.
23] Szewczuk E, et al. Rapid semi-automated quantitative multiplex tandem PCR(MT-PCR) assays for the differential diagnosis of influenza-like illness. BMCInfect Dis 2010;10(1):113.
24] Castilla JF, et al. Effectiveness of trivalent seasonal and monovalent influenza A(H1N1) 2009 vaccines in population with major chronic conditions of Navarre,Spain: 2010/2011 mid-season analysis. Age 2011;11 (152,585).
25] Castilla J, et al. Influenza vaccine effectiveness in preventing outpatient: inpa-tient, and severe cases of laboratory-confirmed influenza. Clin Infect Dis2013;57(2):167–75.
26] Baxter R, Ray GT, Fireman BH. Effect of influenza vaccination on hospitalizationsin persons aged 50 years and older. Vaccine 2010;28(45):7267–72.
27] Kelly H, Steffens I. Complexities in assessing the effectiveness of inactivatedinfluenza vaccines. Euro Surveillance: European Communicable Disease Bul-letin 2013:18.
28] Bragstad K, et al. Low vaccine effectiveness against influenza A (H3N2) virusamong elderly people in Denmark in 2012/13—a rapid epidemiological andvirological assessment. Euro Surveill 2013;18(6).
29] Skull SA, et al. Validity of self-reported influenza and pneumococcal vac-cination status among a cohort of hospitalized elderly inpatients. Vaccine2007;25(25):4775–83.
30] Ohmit SE, et al. Influenza vaccine effectiveness in the 2011–2012 season: pro-
ectiveness of seasonal trivalent inactivated influenza vac-ions in Auckland, New Zealand in 2012. Vaccine (2014),
tection against each circulating virus and the effect of prior vaccination onestimates. Clin Infect Dis 2014;58(3):319–27.
31] CDC. 510(K) Summary for Centers for Disease Control and Prevention HumanInfluenza Virus Real time RT-PCR Detection and Characterization Panel. CDC;2008.
