Shelly McNeil, MD 2018 Adult Immunization Forum Adelaide ... · 1.05 1.34 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 2.2 2.4 Odds Ratio Referent =Antivirals

ShellyMcNeil,MD2018AdultImmunizationForum

Adelaide,AustraliaJune8,2018

Disclosures� Researchgrants:GSK,Merck,SanofiPasteur,Pfizer

�  TheSOSNetworkisfundedbyCIHR,PHACandbycollaborativeresearchagreementswithGSK(influenza)andPfizer(CAP/IPD)

� Clinicaltrials:(all)vaccinemanufacturers� Chair,ImmunizeCanada�  FormermemberofNationalAdvisoryCommitteeonImmunization(NACI)

2010Site2009Site

ThePCIRN/CIRNSOSNetwork:• 2009:8hospitalsin5provinces,5000beds• 2010:10hospitalsin6provinces,6000beds

• 2011:40hospitalsin6provinces,15,000beds• 2012:45hospitalsin7provinces,18,000beds• 2013:45hospitalsin7provinces,18,000beds• 2014:15hospitalsin5provinces,9000beds

•  2015:15hospitalsin5provinces,9000beds•  2016:14hospitalsin4provinces,8200beds

Vancouver

Edmonton

OttawaToronto

Hamilton

QuebecCity Halifax

Sherbrooke

SaintJohnWinnipeg

2011Site

Montreal

Toronto-TIBDN

Sudbury

Moncton

TroisRivieres

2012Site

Brampton

2013Site

2014Site2015Site2016Site

Calgary

SOSObjectives� Toassesstheeffectivenessofinfluenzavaccinationinthepreventionofinfluenza-relatedhospitalizationinolderCanadianadults(≥65y)

� ToassesstheburdenofinfluenzadiseasesamongolderCanadianadults

�  Indoingso,considermeasuresandoutcomesthatareimportantforolderpeople:frailty,mobility,function,socialvulnerability

�  SubsequentlyevolvedtoenablesurveillanceforCAP/IPDandseroepidemiologyofS.pneumoniae

Methods�  15-45academicandcommunityhospitalsacrossCanada

� activesurveillanceforinfluenzainfectioninadults(≥16yearsofage)(Nov.15)� NPswabobtainedfromallpatientswithanadmittingdiagnosisofCAP,exacerbationofCOPD/asthma,unexplainedsepsis,anyrespiratorydiagnosisorsymptom

�  AllNPswabstestedforinfluenzaA&BbyPCR�  InfluenzatypingandBlineagecharacterizationperformedatCIRNSOSCentralLab,CCfV

McNeilSAetal.EuroSurveill.2014March6;19(9).

Methods� Case:

�  Adultpatientswithpositivetestforinfluenzawhoseadmissionisattributabletoinfluenzaoracomplicationofinfluenza

� Control:�  consentingadultpatientsatsamesitewith:

�  diagnosiscompatiblewithinfluenza(i.e.eligibleforNPswabatadmission)

�  NPswabobtainedwithin7daysofonsetofsymptoms,andtestnegativeforinfluenza

�  Admissiondatewithin14dofDOAofcase�  Sameagestrataascase(≥65yor<65y)

VaccineEffectiveness�  VEestimatedas(1-matchedORofinfluenzainvaccinatedvsunvaccinated)*100�  assumingprotectionfromvaccinefrom14dayspostvaccination

�  Unadjusted&Adjusted(conditionallogisticregressionwithbackwardstepwiseselection;p≤0.1)

�  VEpointestimatesand95%CIpresented�  OverallVEandVEinagesubgroups(<65y,≥65y)assessed�  FortheassessmentofVEagainstdeathorneedformechanicalventilationorintensivecareunitadmission,onlymatchedsetsinwhichthecaseexperiencedtheoutcomewereconsideredfortheanalysis

�  VEbyinfluenzatype/subtypeassessed

CasesandControlsperseasoninSOSNetwork

Season #ofCases

#ofControls

Dominantcirculatingstrain(s)

NotesonSeason/Vaccine

2011/2012 528 835 InfluenzaB(Yamagata)

B-lineagestrainincludedintheTIVmismatchedtoBstraincirculating

2012/2013 1292 1573 H3N2 DominantH3N2season2013/2014 1574 2152 H1N1/Influenza

B(Yamagata)MixedH1N1,influenzaBseason

2014/2015 1262 1538 H3N2 MismatchofH3N2includedinTIVtoH3N2straincirculating

2015/2016 1161 NA H1H1/B GoodMatch2016/2017 687 807 H3N2/B H3N2matched/Bmismatchedto

TIVcomponent2017/2018 1872 TBD H3N2 25%mismatch

Total:8,277

Immune function and influenza Incidenceofseriousoutcomesofinfluenzaé

Mostinfluenzadeathsoccurinolderpeople(andotherhighriskgroups)

Foreveryinfluenzadeath,thereare3–4influenzahospitalizations(mostare≥65)

ResponsetovaccinationêCURRENTINFLUENZAVACCINE

Effectivenessinpreventingrespiratoryillnessislowerinolderpeople(andmanyhighriskgroups)thaninhealthyadults

BUThasbenefitinpreventionofpooroutcomes

Sowhatdoesfrailtyhavetodowithinfluenza?

Figure credit: Janet McElhaney

DefinitionofFrailty

Cleggetal.,TheLancet,2013Frailtyisastateofincreasedvulnerabilityto

poorresolutionofhomoeostasisafterastressorevent,whichincreasestheriskofadverseoutcomes.

Frailty:itcomesdownto

Vulnerability

Insults Reserve

0 10 20 30 40 50 60 700

0.2

0.4

0.6

0.8

1

AfrailtyindexbasedonaComprehensiveGeriatricAssessment(FI-CGA)betterstratifies70-monthsurvivalthandoesage

Age FI-CGA

70

80

90

100

0.1

0.2

0.3

0.40.5

Survivaltime(months)

Survivalproba

bility

0 10 20 30 40 50 60 700

0.2

0.4

0.6

0.8

1

Rockwood,Rockwood,Mitnitski.,JAmGeriatricsSoc,2010;58:318-323.

Functionallossiscommonwhenolderpeopleareinhospital

CovinskyJAGS2003

Catastrophicdisabilityv  Definedasalossofindependencein≥3activitiesofdailylivingv  72%whoexperiencecatastrophicdisabilityhavebeenhospitalizedv  Leadingcausesofcatastrophicdisability

1.  Strokes2.  CHF3.  Pneumoniaandinfluenza4.  Ischemicheartdisease

5.  Cancer6.  Hipfracture

Ferrucci et al. JAMA 277:728, 1997 Barker et al. Arch Int Med 158:645, 1998 Falsey et al. N Engl J Med. 2005;352:1749 Andrew et al, IDWeek 2016

VaccinePreventableDisability

15%of65+hospitalized

withinfluenza

FigurecreditDr.JanetMcElhaney

TheproblemofBIAS:howdovaccinatedandunvaccinatedpeoplediffer?� Biasisanyfactorindependentlyassociatedwithriskofdiseaseandvaccinationstatus� Healthyuserbias-personsmorelikelytobevaccinatedarelesslikelytodevelopdisease-�  OVER-estimatesVE

�  Indication(frailty)bias-personsmorelikelytobevaccinated(e.g.frailelderlypeople)aremorelikelytohavesuboptimalvaccineresponseandexperienceadversemoreinfluenzaoutcomes�  UNDER-estimatesVE

QuantifyingFrailty� TheFrailtyIndex(FI)iscalculatedbyaddingthenumberofdeficitsapersonhasdividedbythetotalpossibledeficits FI=#deficits/totalpossibledeficits

� Canincludeasmanypossibledeficitsasavailabledataallows;ideally~40

RockwoodCMAJAug2005SearleBMCGeriatrics2008

Function•  BarthelIndex:collectedat3timepoints,baseline(priorto

onsetofcurrentillness),duringadmission,and30daypostdischarge.

•  AssessmentofindependenceinADL.•  Scorebetween0-10foreachindividualsection:bowels,

bladder,grooming,toiletuse,feeding,transfer,mobility,dressing,stairs,andbathing,givingascorebetween0-100(100indicatingcompleteindependenceinADL).

Distributionofhospitaladmissionsbyweekandinfluenzastrain(2011/12)

A/H1

A/H3

B-Vic

B-Yam

AndrewMKetal.BMCInfectDis.2017;17(1):805

ClinicalCharacteristic Casesn=528(%)

Controlsn=835(%)

p-value

Agemean(range) 67.1(18-104) 69.2(18-99) 0.73

Age≥65y 80.6(65-104) 78.8(65-99) 0.001

Female 288(54.5) 469(56.2) 0.58

Obese(BMI>=30) 103(19.5) 229(27.4) 0.016

PregnantMeangestwks

10(1.9)27.94

1(0.1)27.22

0.006

Aboriginal 2(0.4) 3(0.4) NS

AdmittedfromLTCF 50(9.5) 38(4.5) <0.05

CurrentsmokerPastsmoker

61(29.3)30(14.4)

83(30.6)70(25.8)

0.900.004

CardiacdiseasePulmonarydisease

210(39.8)231(43.8)

415(49.7)426(51.0)

0.0050.021

Currentseasonvaccine 262(49.6) 529(63.4) <0.001

Priorseasonvaccine 248/481(51.6) 515/793(64.9) <0.001

BLFrailtymean(SD) 0.22(0.13) 0.20(0.11) 0.006

BarthelIndexmean(SD) 81.7(28.8) 88.1(21.0) 0.003

ClinicalCharacteristics(2011/12)Variable Vaccinated

(n=792)%Unvaccinated(n=602)%

p-value

Agemean(range) 73.7(18-104) 61.4(18-98) <0.001Age≥65ymean65-75>75

80.0(65-104)32.4%67.6%

78.2(65-98)39.6%60.4%

0.003

0.041Gender NSObese(BMI>30) NSAdmittedfrLTCF 9.0% 2.6% <0.001Currentorpastsmoker

54.5% 49.2% 0.02

BLFrailtymean(SD) 0.20(0.11) 0.17(0.11) <0.001≥1comorbidity 98.1% 87.9% <0.001≥4medications 77.3% 49.4% <0.001

AgeandBurdenofDiseaseAge 16 – 49 N = 128

Age 50-64 N = 118

Age 65-75 N = 109

Age >75 N = 237

% vaccinated

BOD by strain

Death 1 (0.8%) 3 (2.5%) 6 (5.5%) 36 (15.2%)

ICU 16 (12.5%) 20 (16.9%) 17 (15.6%) 22 (9.3%)

FrailtyandBurdenofDiseaseLow Frailty (FI < 0.2) N = 92

Med Frailty (FI 0.2-0.45) N = 84

High Frailty (FI ＞0.45) N = 14

% vaccinated

BOD by strain

Death 5 (5.4%) 11 (13.1%) 5 (35.7%)

ICU 7 (7.6%) 11 (13.1%) 1 (7.1%)

Outcomesbytype/subtype(2011/12)

* P ≤0.05

Variable Influenza A n = 161

Influenza B n = 299

A/H1N1 n=99

A/H3N2 n=61

B/Vic n=89

B/Yam n=204

Mean LOS (SD) 10.0 (10.4) 10.4 (11.9)

9.3 (9.1) 11.0 (12.4) 11.3 (13.2) 10.1 (11.4)

Admit to ICU 22 (13.7%) 30 (10.0%)

15 (15.2%) 7 (11.5%) 12 (13.5%) 18 (8.8%)

30d mortality 10 (6.2%) 23 (7.7%)

3 (3.0%) 7 (11.5%)* 3 (3.4%) 20 (9.8%)*

UnadjustedandAdjustedVEinOlderAdults

45.0

58.0

43.3

58.7

0

10

20

30

40

50

60

70

80

VaccineEffectiven

ess(%)

Adjusted(FullModel)*

Adjusted(FullModelwithoutFrailty)*

Adjusted(FrailtyOnly)*

*Post-HocAnalysis

Unadjusted

AndrewMKetal.JInfectDis.2017;216(4):405-414

AdjustedVEestimatesbyfrailtylevel-Notallolderadultsarealike!*

77.6

51.059.6

-24.8

-40

-20

0

20

40

60

80

100

VaccineEffectiven

ess(%)

FrailMostFrail

Non-Frail

Pre-Frail

*Post-HocAnalysis

0

50

100

150

200

250

300

350 -1

00

-90

-80

-70

-60

-50

-40

-30

-20

-10 0 10

20

30

40

50

60

70

80

90

100

Num

ber

of P

atie

nts

Changes in Barthel from Baseline to Day 30

Case

Control

BarthelchangesfrombaselinetoDay30post-discharge

15%of65+hospitalized

withinfluenza

Summary�  TIVdemonstratedmoderateyetsignificantprotectionagainstinfluenza-relatedhospitalizationsinolderadults≥65y(VE:58.0%).

�  FrailtywasthemostsignificantcontributorinthefullyadjustedVEmodel.NotadjustingforfrailtymayunderestimatetrueVEestimates.

�  VEdemonstratesatrendofdecreasingasleveloffrailtyincreases.Givenmostolderadultsarenotfrail(frailtyprevalenceestimatedat~24%incommunity-dwellingolderadults),thebenefitofthevaccineinnon-frailolderadultsshouldnotbeunderestimated.

�  14.6%ofpatientslostbetween20and100pointsontheBarthelIndex,indicatingcatastrophicdisabilityfollowinghospitalization.

Conclusions� EvaluatingtheimpactoffrailtyonVEandseriousoutcomesiscriticallyimportantforfullyunderstandingthehealthbenefitsoftheinfluenzavaccineinolderadults

� TheTIVremainsaneffectivetoolforpreventinginfluenza-relatedhospitalizationsinanolderadultpopulationandshouldbecontinuedtobeusedtopreventseriousoutcomesassociatedwithinfluenza

Conclusions�  Understandingtheimpactofinfluenzaonfrailty(andoffrailtyoninfluenza)iscriticaltounderstandingitstrueburden

�  OurdatasuggestsafrailtybiasinobservationalstudiesofVE�  Indicationbias(ratherthanhealthyuserbias)

�  ObservationalstudieswhichdonotquantifyandadjustforfrailtywillsystematicallyUNDERESTIMATEtheestimatedvaccineeffectivenessinthispopulation

�  VEestimatestendtoincreasewhenadjustedforfrailty;thishasimportantimplicationsfortargetingvaccinationcampaignsandunderstandingthetruebenefitsofvaccination

CaoimheMcParland,BScH,MDCandidateonbehalfoftheSOSNetworkofthe

CanadianImmunizationResearchNetwork

Season #ofinfluenzaACases

#ofinfluenzaBcases



2012/2013 1891 148 H3N22013/2014 1384 844 H1N1/InfluenzaB

(Yamagata)Total 3484 1375

SeasonOverview

InfluenzaA InfluenzaB p-value

Age(mean) 65.8 71.2 <0.01

Gender(male) 48.7% 46.0% 0.12

Admissionfromlong-termcare

5.5% 12.1% <0.01

Numberofmedications(>4)

59% 64.6% <0.01

Priormedicalcomorbidities(Yes)

88.3% 90.2% 0.05

Results:Demographics

Results:FrailtyInfluenzaA InfluenzaB p-value

Priortoillnessonset

0.21

0.22

0.02

Worstbetweenadmissionandenrolment

0.28 0.29 0.11

30-dayspostdischarge

0.20 0.21 0.12

*FrailtyIndexisonascaleofincreasingfrailtyfrom0to1

Results:ClinicalOutcomesInfluenzaA InfluenzaB p-value

Durationofhospitalization

11.1days 10.27days 0.07

ICUadmission 18.05% 12.22% <0.01

Mechanicalventilation

11.77% 7.27% <0.01

Antiviralusepriortoadmission

11.45% 12.80% 0.19

Antiviraluseduringadmission

94.32% 91.49% <0.01

Mortality30dayspost-discharge

9.01% 9.45% 0.63

�  CurrentperceptionconsidersinfluenzaAtobeofmoresignificancethaninfluenzaB�  InfluenzaAissignificantlymorelikelytobeadmittedtotheICUorrequiremechanicalventilation

�  However,thereisnodifferenceindurationofhospitalizationormortalityrates

�  InfluenzaBhasamoresignificanteffectonthefrailelderly,particularlythosecomingfromalong-termcarefacility�  Carefulconsiderationshouldbegiventothedevelopmentofhigh-doseandadjuvantedQIVtoenhanceinfluenzaBprotection

Conclusion

PooledVE:Anassessmentofaveragebenefitofvaccinationovertime�  Influenzavaccineremainsourbestmethodofprotectionfrominfluenzainfectionandassociatedseriousoutcomes

� Seasonalinfluenzavaccineeffectiveness(VE)variesyeartoyeardependingonvaccine-strainmismatch,circulatingstrains,andhostfactorsandisgenerallynotpredictable

�  InfluenzaVEinolderadults,likelyduetoacombinationoffactorsincludingimmunosenescence,increasedcomorbidities,andfrailty,isgenerallyshown

tobelowerthanVEinworking-ageadults

NicholsMKetalVaccine36(16);2018:2166-2175

Clinicalcharacteristicsofcasesandcontrols(11/12,12/13,13/14pooled)

Characteristics Cases (n=3394) n (%)

Controls (n=4560) n (%)

p value

Age mean (range) 16-49y 50-64y 65-75y >75 y

67.6 (16-105) 611 (18.0) 705 (20.8) 674 (19.9)

1404 (41.4)

68.8 (16-104) 626 (13.7) 995 (21.8)

1063 (23.3) 1876 (41.1)

0.193

Female 1805 (53.2) 2436 (53.4) 0.94

≥1 comorbidities 3025 (89.1) 4234 (92.9) 0.00 Pregnant 87 (2.6) 13 (0.3) 0.00 Smoker 1669 (49.2) 2702 (59.3) 0.00

Antiviral use PTA 33 (1.0) 32 (0.7) 0.33 Current season vaccine

1585 (46.7) 2806 (61.5) 0.00

Prior season vaccine

1588/2957 (53.7) 2360/3758 (62.8) 0.00

Overallstraindistribution(11/12,12/13,13/14pooled)

72.4%

27.6% %total(subtypeknown=3489)

InfluenzaA

InfluenzaB

A/H1N1(41.3%)

A/H3N2(58.7%)B(Vic)(11.6%)

B(Yam)(88.4%)

AdjustedVEestimatebyinfluenzaseason

42.8

58.0

33.2 34.8

25.6

47.751.3

46.3

57.3

-20

-10

0

10

20

30

40

50

60

70

80

VaccineEffectiven

ess(%)

2011/2012 2012/2013 2013/2014

Overall

Overall

Overall

Age≥65

Age≥65

Age≥65

Age<65

Age<65

Age<65

41.7 39.348.0

58.2 58.7 60.8

31.324.2

44.4

17.6

49.5

-11.5

29.2

40.8

29.0

-60

-40

-20

0

20

40

60

80

100

VaccineEffectiven

ess(%)

Overall

Overall

Overall

Overall

Age≥65

Age≥65

Age≥65

Age≥65

Age<65

Age<65

Age<65

Age<65

Age<65


AdjustedVEestimatebyinfluenzasubtype(11/12,12/13,13/14pooled)

Overall

AnyStrain H1N1 H3N2 B(Victoria) B(Yamagata)Ag

e≥65

AdjustedVEestimatebyseverity(11/12,12/13,13/14pooled)

41.7 39.348.0

54.1 49.658.9

74.580.6

95.4

54.171.0

-120

-100

-80

-60

-40

-20

0

20

40

60

80

100

VaccineEffectiven

ess(%

)

Overall

Overall

Overall

Influ

enzaA

H1N1

H3N2

Influ

enzaB

DeathinPatients≥65ICU/MechanicalVentilation

HospitalizationAg

e≥65

Age<65

Age≥65

Age<65


Summary� Whileinfluenzavaccineeffectivenessvariesyear-to-yearduetofactorssuchasvirulenceofthecirculatingstrainandmatchbetweencirculatingandvaccinestrains,wedemonstrateastatisticallyandclinicallyimportantbenefitofvaccinationinadultsspanningthreeinfluenzaseasons(overallVE42%)

�  Over3seasons,TIVeffectivenessforthepreventionofhospitalizationduetoinfluenzaA(H3N2)was24%inolderadults

�  StatisticallysignificantprotectionagainstsevereoutcomesincludingneedforICUadmissionormechanicalventilationanddeathwasdemonstratedinolderadults(VEestimate54%and75%,respectively),andthisprotectionincreasedwiththeseverityoftheoutcome

Conclusion�  Theindividualandpublichealthbenefitsofinfluenzavaccinesshouldnotbeunderstatedandpublicmessagingshouldaddressoverallbenefitsovertimewhileacknowledgingyear-to-yearvariability

AdjuvantedVaccines� Adj-influenzavaccine(Fluad®)wasfirstapprovedforuseinCanadain2011andwasrecommendedforuseinolderadultsbyNACI

� Notallprovinces/territoriesfundadj-influenzavaccine�  Inclinicaltrials,adj-influenzavaccinehasbeenshowntoelicitastrongerimmuneresponseinolderadultsthannon-adjuvantedinfluenzavaccines

�  Itisdifficulttodemonstrateadj-vaccine’sbenefitovernon-adjvaccinesinobservationalstudies

WhyEvaluateAdj-InfluenzaVE?•  Thebulkofseriousoutcomesassociatedwithinfluenza(forexample:hospitalization,ICUadmission,functionaldeclineordeath)comesfromolderadults(≥65)

• Understandingifthereisanadditionalbenefittovaccinatingolderadultswithadjuvantedvaccinescomparedtonon-adjuvantedvaccinesisimportanttoinformvaccinepolicyandpotentiallyoptimizeuseofinfluenzavaccinationinolderadults(≥65)

Clinicalcharacteristicsofcasesandcontrols≥65y(pooled2011-2014)

VariableCases

2078(41.42%)Controls

2939(58.58%)

PValue1

Sex Male 1000(48.12%) 1355(46.10%) 0.15

Age 65-7475andolder

674(32.44%)1404(67.56%)

1063(36.17%)1876(63.83%)

0.006*

Wasvaccinatedincurrentseason

Yes

1244(59.80%)

2126(72.34%)

<0.001*

BMI2

Underweight<18.5 117(5.63%) 190(6.46%) 0.0006*Normalweight18.5-24.99 740(35.61%) 1073(36.51%)

Overweight25-29.99 598(28.47%) 808(27.49%)Obese30-40 327(15.74%) 561(19.09%)

Veryobese>40 39(1.88%) 108(3.67%)Pastorcurrent

smoker2Yes 1029(49.52%) 1744(59.34%) <0.001*

Medicalcomorbidities

Yes 2010(96.73%) 2864(97.45%) 0.13

#ofmedicationspriortoadmission2

0-4 529(25.46%) 506(17.22%) <0.001*

Clinicalcharacteristicsofpatients≥65ywhoreceivednon-adjTIVvsadjTIV(2011-2014)

Variable

Adjuvanted(Fluad®)1N=284

Non-Adjuvanted1N=2049 PValue2

Sex Male 127(44.72) 1002(48.90%) 0.18Age MeanAge

MedianRange

83.5885.0

65-105

79.8580.0

65-102

<0.0001*

Wasvaccinatedinpreviousseason3

Yes 197(69.37%) 1646(80.33%) 0.17

Pastorcurrentsmoker3 Yes 133(46.83%) 1205(58.81%) 0.002*

Medicalcomorbidities Yes 277(97.54%) 2007(97.95%) 0.64

Antiviralusepriortoadmission

Yes 11(3.87%) 17(0.83%) <0.0001*

Numberofmedicationspriortoadmission3

0-44+

45(15.85%)237(83.45%)

332(16.20%)1704(83.16%)

0.88

Admissionfromalong-termcarefacility3

Yes 163(57.39%) 116(5.66%) <0.0001*

Frailtyindexpriortoadmission3

Non-Frail 11(3.87%) 311(15.18%) <0.0001*Pre-FrailFrail

MostFrail

43(15.14%)126(44.37%)56(19.72%)

857(41.83%)716(34.94%)55(2.68%)

VEofvaccinetypesforpreventinginfluenza-relatedhospitalizationsinpatients≥65y,2011-2014

39.3 32.5

61.3

0

10

20

30

40

50

60

70

80

90

100

Vac

cine

Eff

ecti

vene

ss (

%)

AllVaccines Non-Adjuvanted Adjuvanted

VEofvaccinetypesforpreventinginfluenza-relatedhospitalizationsbyleveloffrailty

26.9

52.6

26.1

56.5

-40

-20

0

20

40

60

80

100

Non-AdjuvantedAdjuvanted

Non-AdjuvantedAdjuvanted

Non-FrailandPre-Frail FrailandMost-Frail

Summary•  OverallVEofallinfluenzavaccineswas~39%forpreventinginfluenza-relatedhospitalizationsinpatients≥65yenrolledintheSOSNetworkbetween2011-2014

•  VEofadj-influenzavaccinewas61.3%inpatients≥65y;representinganincreaseof~30%overnon-adjinfluenzavaccine(differencewasnotstatisticallysignificant)

•  VEofadj-influenzavaccineforpreventinginfluenza-relatedhospitalizationswasgood(61.3%)inthiselderly,frailpopulation,withalargeproportionofpatientsadmittedfromlong-termcare

•  Appearstobeatrendofincreasedprotectionfromadj-influenzavaccinebutshouldbeinterpretedcautiously-95%CIswerewideandoverlappedthenon-adjandall-vaccinesVEestimates

Conclusions� OurfindingsdemonstrateapossibletrendofincreasedVEofadjuvantedinfluenzavaccinerelativetonon-adjuvantedvaccinesinanelderly,hospitalized,andfrailpopulation

� ContinuedmonitoringofVEforadjuvantedaswellashighdoseinfluenzavaccinesinfuturestudyyearsisnecessarytoinforminfluenzaimmunizationpolicyinCanada

MichaelaNichols-Evans,MScEpidemiologist,CIRNSOSNetwork

Overview

�  Variabilityininfluenzavaccineeffectiveness(VE)estimatesbetweenseasonsandstrainsmaynotbefullyexplainedbythesefactors

�  PriorvaccinationhasemergedasafactorthatmayimpactsubsequentVE

�  AntigenicDistanceHypothesis(1):Couldbenegativeinterferencefrompriorimmunizationwhentheantigenicdistanceissmallbetweensuccessivevaccinecomponentsbutlargebetweenvaccineandcirculatingstrains

HumanFactors(Age,Comorbidities)

InfluenzaStrainFactors(Mismatch)

VaccineFactors(immunogenicity)

VaccineEffectiveness

(1)Smith,D.J.,Forrest,S.,Ackley,D.H.,&Perelson,A.S.(1999).Variableefficacyofrepeatedannualinfluenzavaccination.ProceedingsoftheNationalAcademyofSciencesoftheUnitedStatesofAmerica,96(24),14001–14006.

Overview�  Severalrecentobservationalstudies(2,3,4)haveshownanimpactofpriorseasonalinfluenzavaccinationonsubsequentinfluenzavaccineeffectiveness

�  TherewasaneedtoassessifthisimpactwaspresentwithinCanada’sinfluenzahospitalizationnetwork

� Welookedatthisimpactover4influenzaseasonsinCanada,whichenabledseasonalcomparisons

2.SkowronskiDM,ChambersC,SabaiducS,DeSerresG,WinterAL,DickinsonJA,etal.Aperfectstorm:Impactofgenomicvariationandserialvaccinationonlowinfluenzavaccineeffectivenessduringthe2014-15season.ClinInfectDis.2016Mar29.3.SkowronskiDM,deSerresG,CrowcroftNS,JanjuaNZ,BoulianneN,HottesTS,etal.Associationbetweenthe2008-09seasonalinfluenzavaccineandpandemicH1N1illnessduringspring-summer2009:FourobservationalstudiesfromCanada.PLoSMed.2010/;7(4).4.McLeanHQ,ThompsonMG,SundaramME,MeeceJK,McClureDL,FriedrichTC,etal.ImpactofrepeatedvaccinationonvaccineeffectivenessagainstinfluenzaA(H3N2)andBduring8seasons.ClinInfectDis.2014Nov15;59(10):1375-85.

VEcalculations� Casesandcontrolswerethendividedinto4categoriesofvaccinationstatus:�  (1)Vaccinatedinneitherseason(REFERENT)�  (2)Vaccinatedincurrentseasononly�  (3)Vaccinatedinbothcurrentandpriorseason�  (4)Vaccinatedinpriorseasononly

�  VE=1-ORx100% �  Unadjusted&Adjusted(conditionallogisticregressionwithbackwardstepwiseselection;p≤0.1)

�  VEpointestimatesand95%CIpresented*AllVEanalysesarepost-hoc

CasesandControlsperseasoninSOSNetwork

Season #ofCases

#ofControls


NotesonSeason/Vaccine


B-lineagestrainincludedintheTIVdidnotmatchingtheBstrain

circulating2012/2013 1292 1573 H3N2 DominantH3N2season(73%of

SOSsubtypedcases)2013/2014 1574 2152 H1N1/

InfluenzaB(Yamagata)

MixedH1N1,influenzaBseason

2014/2015 1262 1538 H3N2 MismatchofH3N2includedinTIVtoH3N2straincirculating

2011/2012Season-Overall

-90 -80 -70 -60 -50 -40 -30 -20 -10 0 10 20 30 40 50 60 70

VaccineEffectiveness(%)

2011/2012 Adjusted Vaccine Effectiveness for Preventing Influenza-Associated Hospitalizations

PriorSeasonOnly BothSeasons CurrentSeasonOnly

-90 -80 -70 -60 -50 -40 -30 -20 -10 0 10 20 30 40 50 60 70




-13.2

40.7

18.3

-100 -80 -60 -40 -20 0 20 40 60 80


CurrentSeasonOnlyBothSeasonsPriorSeasonOnly

-22.8

28.2

62.5

-80 -60 -40 -20 0 20 40 60 80 100



-90 -80 -70 -60 -50 -40 -30 -20 -10 0 10 20 30 40 50 60 70




-90 -80 -70 -60 -50 -40 -30 -20 -10 0 10 20 30 40 50 60 70




-120 -100 -80 -60 -40 -20 0 20 40 60 80 100



in Patients >= 65


-120 -100 -80 -60 -40 -20 0 20 40 60 80VaccineEffectiveness(%)


in Patients <65


-80 -70 -60 -50 -40 -30 -20 -10 0 10 20 30 40 50 60 70 80 90


2012/2013 Adjusted Vaccine Effectiveness for Preventing Influenza-Associated

Hospitalizations



-13.7

20.2

64.9

-80 -60 -40 -20 0 20 40 60 80 100


2012/2013Season-Age≥65

-90 -80 -70 -60 -50 -40 -30 -20 -10 0 10 20 30 40 50 60 70




-90 -80 -70 -60 -50 -40 -30 -20 -10 0 10 20 30 40 50 60 70




-120 -100 -80 -60 -40 -20 0 20 40 60 80 100



in Patients >= 65




in Patients <65


-80 -70 -60 -50 -40 -30 -20 -10 0 10 20 30 40 50 60 70 80 90



Hospitalizations




-90 -80 -70 -60 -50 -40 -30 -20 -10 0 10 20 30 40 50 60 70




-90 -80 -70 -60 -50 -40 -30 -20 -10 0 10 20 30 40 50 60 70




-120 -100 -80 -60 -40 -20 0 20 40 60 80 100



in Patients >= 65




in Patients <65


-80 -70 -60 -50 -40 -30 -20 -10 0 10 20 30 40 50 60 70 80 90



Hospitalizations


25.9

52.2

63.4

-20 0 20 40 60 80VaccineEffectiveness(%)


-12.7

11.1

24.7

-80 -60 -40 -20 0 20 40 60 80



-90 -80 -70 -60 -50 -40 -30 -20 -10 0 10 20 30 40 50 60 70




-90 -80 -70 -60 -50 -40 -30 -20 -10 0 10 20 30 40 50 60 70




-120 -100 -80 -60 -40 -20 0 20 40 60 80 100



in Patients >= 65




in Patients <65


-80 -70 -60 -50 -40 -30 -20 -10 0 10 20 30 40 50 60 70 80 90



Hospitalizations


-70 -60 -50 -40 -30 -20 -10 0 10 20 30 40 50 60VaccineEffectiveness(%)


Hospitalizations

PriorSeasonOnly BothSeasons


Summary2011/2012Season

2012/2013Season

2013/2014Season

2014/2015Season

DominantStrain

InfluenzaB H3N2 H1N1 H3N2

VaccineCompositioninrelationtopreviousyear

Same UpdatedBUpdatedH3N2SameH1N1

UpdatedBUpdatedH3N2SameH1N1

Same

Mismatch YES-Bcomponent

No No YES-H3N2component

Effect Non-signifPositive

Negative Non-signifNegative

Non-signifNegative

Conclusions� TherewasvariedimpactofpriorvaccinationonsubsequentVEobservedfromseasontoseasonandbetweenagegroups�  Largestimpactswereseeninthe2012/2013seasonwhereinfluenzaAH3N2wasthedominantcirculatingstrain

� Unmeasuredbiasbyindicationcannotberuledout� Current-onlyandboth-seasonsVEwasalwaysbetterthanprior-onlyVE,indicatingreceivingannualinfluenzavaccinationisstillprovidingaddedprotectionovernotreceivingannualinfluenzavaccination

ZacharyShaffelburg,MDcandidate2018

Doestreatmentwithantiviralsimproveoutcomes?Whatabouttiming?

� WHOandothersrecommendthattreatmentwithneuraminidaseinhibitorsshouldbeinitiatedasearlyaspossibleforanypatientwithconfirmedorsuspectedinfluenzawhoishospitalized,hassevereillness,oramongtheriskgroupstargetedforvaccination.

� Cliniciansoftenhesitatetouseantivirals,especially>2daysaftersymptomonset.

Clinicalcharacteristicsofhospitalizedpatientswithlaboratory-confirmedinfluenza(11/12,12/13,13/14pooled)

ClinicalCharacteristics

Characteristics Allpatients(n=4862)n(%)

Agemedian(range)16-49y50-64y65-75y>75y

70(16-105)892(18)1061(22)928(19)1981(41)

Female 2535(52)≥1comorbidities 4319(89)

Pregnant 118(2)Smoker 2318(48)

Antiviralusepriortooutcome(PTO)Timefromsymptomonsettoantiviralstart

2642(54)Mean:4.21d,Range:0-21d

Currentseasonvaccine 1850(38)InfluenzaA 3484(72)InfluenzaB 1375(28)

ORsofriskfactorsforanoutcomeofICUadmissionormechanicalventilationinhospitalizedpatientswithlaboratory-confirmedinfluenza

0.7

1.0

1.3

0.1

1.4

1.7

0.1

0.5

0.8

0.6

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8

Odds Ratios (ORs)

Age >75 Years

Age 65-75 Years

Age 50-64 Years

Age 16-49 Years [Referent]

Pregnancy

Current or Prior Smoker

Enrolled in 13/14 season

A/H1N1 [referent]

Antiviral Use PTO

B/ Victoria

B/ Yamagata

A/H3N2

0.09

0.09

0.11

1.05 1.34

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 2.2 2.4

Odds Ratio

Referent =Antivirals <2 Days

Evenafter5+days,antiviraluseisstillbeneficialinreducingICU/mechanicalventilation

<2 Days

2 Days- <5 Days

5+ Days

Referent = No Antivirals

2 Days- <5 Days

5+ Days

*Post-Hoc

Variable OR (95% CI) P value

Pregnant 0.1 (0.0-0.4) 0.006

Smoker 1.4 (1.1-1.8) 0.018

Antiviral use PTO 0.10 (0.08-0.14) <0.001

Hospitalized in 13/14 Season 1.9 (1.1-3.4) 0.020

ORsofriskfactorsforanoutcomeofICUand/orMechVentilationinhospitalizedpatientswithlaboratory-confirmedinfluenzaA

Variable OR (95% CI) P value

Smoker 1.8 (1.1-2.7) 0.012

Antiviral use PTO 0.14 (0.1-0.2) <0.001

ORsofriskfactorsforanoutcomeofICUand/orMechventilationinhospitalizedpatientswithlaboratory-confirmedinfluenzaB

Sowhatdoesfrailtyhavetodowithinfluenza?

Adjustingforfrailtyisimportantinmeasuringinfluenzavaccineeffectiveness(FrailtyBias)Understandingtherelationshipbetweeninfluenzaandfrailtyiscriticaltounderstandingthetrueburdenofinfluenza

Figurecredit:JanetMcElhaney

NOTAddingLifetoYears

4

56

78 9

321

Healthinsults

AddingLifetoYears:canfrailtybeprevented?

4

56

78 9

321

Candidates:• Exercise• Socialintegration• Physiologicalinterventions:nutrition,inflammation,immune,drugs?• Goodcare?

*Atleastwecanpreventsomeconsequencesandcomplicationsoffrailty!*Avoidableillness&hospitalizations

Acknowledgements

ManythankstotheSOSNetworkteam!MelissaAndrew,JanetMcElhaney,ArdithAmbrose,DonnaMacKinnon-Cameron,ChristinaWang,PeterYe&thededicatedSOSNetworksurveillancemonitors

76

QUESTIONS&DISCUSSION

Documents

Shelly McNeil, MD 2018 Adult Immunization Forum Adelaide ... · 1.05 1.34 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 2.2 2.4 Odds Ratio Referent =Antivirals