Assessment of the relationship between entomologic indicators of Aedes aegypti and the epidemic occurrence of dengue virus 3 in a susceptible population, São José do Rio Preto, Sao Paulo, Brazil

(1)

ActaTropica142(2015)167–177

ContentslistsavailableatScienceDirect

Acta

Tropica

j ou rn a l h o m epa g e :w w w . e l s e v i e r . c o m / l o c a t e / a c t a t r o p i c a

Assessment

of

the

relationship

between

entomologic

indicators

of

Aedes

aegypti

and

the

epidemic

occurrence

of

dengue

virus

3 in

a

susceptible

population,

São

José

do

Rio

Preto,

São

Paulo,

Brazil

Francisco

Chiaravalloti-Neto

a,∗

_,

_Mariza

_Pereira

b

_,

_Eliane

_Aparecida

_Fávaro

c

_,

Margareth

Regina

Dibo

d

_,

_Adriano

_Mondini

e

_,

_Antonio

_Luiz

_{Rodrigues-Junior}

f

_,

Ana

Patrícia

Chierotti

d

_,

_Maurício

_Lacerda

_Nogueira

c

a_Departamento_de_{Epidemiologia,}_Faculdade_de_Saúde_Pública,_Universidade_de_São_Paulo,_Av._Dr._Arnaldo,_715,_01246-904_São_Paulo,_SP,_Brazil b_Diretoria_de_Combate_a_Vetores,_{Superintendência}_de_Controle_de_Endemias,_Rua_Paula_Souza,_166,_01027-000_São_Paulo,_SP,_Brazil

c_Laboratório_de_Virologia,_Faculdade_de_Medicina_de_São_José_do_Rio_Preto,_Av._Brigadeiro_Faria_Lima,_5416,

15090-000SãoJosédoRioPreto,SP,Brazil

d_Laboratório_de_Vetores_de_São_José_do_Rio_Preto,_{Superintendência}_de_Controle_de_Endemias,_Av._Brigadeiro_Faria_Lima,_5416,_15090-000_São_José_do_Rio

Preto,SP,Brazil

e_Faculdade_de_Ciências_{Farmacêuticas,}_Universidade_Estadual_Paulista,_Rodovia_Araraquara_–_Jaú,_Km_1,_14801-902_Araraquara,_SP,_Brasil

f_Departamento_de_Medicina_Social,_Faculdade_de_Medicina_de_Ribeirão_Preto,_Universidade_de_São_Paulo,_Av._{Bandeirantes,}₃₉₀₀_Hospital_{Universitário,}

2◦_andar,_14049-900_Ribeirão_Preto,_SP,_Brazil

a

r

t

i

c

l

e

i

n

f

o

Articlehistory: Received28June2014 Receivedinrevisedform 24November2014 Accepted27November2014 Availableonline4December2014

Keywords: Dengue Aedesaegypti Entomologicindicator Spatialanalysis Temporalanalysis

a

b

s

t

r

a

c

t

Theaimsofthisstudyweretodescribetheoccurrenceofdengueinspaceandtimeandtoassessthe rela-tionshipsbetweendengueincidenceandentomologicindicators.Weselectedthedengueautochthonous casesthatoccurredbetweenSeptember2005andAugust2007inSãoJosédoRioPretotocalculate inci-denceratesbymonth,yearandcensustracts.Themonthlyincidenceratesofthecitywerecomparedto themonthlyBreteauindices(BI)oftheSãoJosédoRioRegion.BetweenDecember2006andFebruary 2007,anentomologicalsurveywasconductedtocollectimmatureformsofAedesaegyptiinJaguaré,a SãoJosédoRioPretoneighborhood,andtoobtainentomologicalindices.Theseindiceswererepresented usingstatisticalinterpolation.TorepresenttheoccurrenceofdengueintheJaguaréneighborhoodin2006 and2007,weusedtheKernelratioandtoevaluatetherelationshipbetweendengueandthe entomolog-icalindices,weusedageneralizedadditivemodelinaspatialcase–controldesign.BetweenSeptember 2005andAugust2007,theoccurrenceofdengueinSãoJosédoRioPretowasalmostentirelycaused byDENV3,andthemonthlyincidenceratespresentedhighcorrelationcoefficientswiththemonthlyBI. InJaguaréneighborhood,theentomologicalindicescalculatedbyhectarewerebetterpredictorsofthe spatialdistributionofdenguethantheindicescalculatedbyproperties,butthepupaequantificationdid notshowbetterpredictionqualitiesthantheindicesbasedonthecontainerpositivity,inrelationtothe riskofdengueoccurrence.Thefactthatthemunicipality’spopulationhadahighsusceptibilitytothe serotypeDENV3beforethedevelopmentofthisresearch,alongwiththealmosttotalpredominanceof theoccurrenceofthisserotypebetween2005and2007,facilitatedtheanalysisoftheepidemiological situationofthediseaseandallowedustoconnectittotheentomologicalindicators.

Abbreviations: BI,Breteau Index(cis);GIS,geographicinformationsystem; SINAN,notifiablediseasesinformationsystem.

∗Correspondingauthor.Tel.:+551130617920/982081602. E-mailaddresses:[email protected](F.Chiaravalloti-Neto),

[email protected](M.Pereira),[email protected](E.A.Fávaro), [email protected](M.R.Dibo),[email protected](A.Mondini), [email protected](A.L.Rodrigues-Junior),[email protected](A.P.Chierotti), [email protected](M.L.Nogueira).

1. Introduction

Dengue infection is caused by four antigenically distinct

serotypesofthedenguevirus(DENV1,DENV2,DENV3,andDENV4),

anditsmainvectoristheAedesaegyptimosquito.Eachserotype

infectionprovidesspecificlong-termimmunitybutonlytransient

immunitytotheotherserotypes(WHO,2010).Dengueis

consid-eredthemostsignificantarbovirusthataffectshumans.Dengue

is estimatedtoannuallycause390millioninfections, including

http://dx.doi.org/10.1016/j.actatropica.2014.11.017

(2)

96millioncasesofclassicaldengueand20,000deathscausedby

dengue(Bhattetal.,2013).

InBrazil,thenumbersofcasesanddeathsrelatedtodenguehave

beenincreasingovertimebecauseoftheintroductionor

reintro-ductionofvirusserotypesandthesimultaneouscirculationofthe

fourserotypes(Drumondetal.,2012,2013;Figueiredoetal.,2008;

Teixeiraetal.,2013,2005).InthestateofSãoPaulo,thenumber

ofdenguecaseshavefollowedthesamepattern,increasingfrom

2001(51,668cases)to2007(92,345cases),2010(189,330cases),

and2013(201,498cases)(CVE,2014).

Trackingtheviralserotypesthat arecirculatingisan

impor-tanttoolforepidemiologicalresearch,particularlyinrelationto

thespatialand temporaldynamics ofthediseaseand infection

incidence(Mondinietal.,2009).Moleculartechniquesbasedon

PCRprovidebothadvanceddiagnosticsofviralinfections,mainly

dengue and yellow fever, and quick and precise answers for

disease-surveillancesystems(Bronzonietal.,2005;Teixeiraetal.,

2005).

Regardingentomologicalindicators,many authorshave

sug-gestedthequantificationofthepupaeofAe.aegyptiasanadequate

way to evaluate the risk of dengue transmission (

Arredondo-Jimenezand Delgado-Valdez, 2006; Barrera et al.,2006; Focks, 2003;FocksandChadee,1997;NathanandFocks,2006; Romero-VivasandFalconar,2005).Incontrast,Sanchézetal.(2010,2006)

showedthatinareaswithlowinfestationlevels,theBreteauIndex

(BI)wassuitabletoindicateareasthathadahigherriskofdengue

occurrence,andMorrisonetal.(2008)statedthatotherstudies

arenecessarytovalidatethoseresultsbeforetheirapplicationto

surveillanceactivitiesandvectorcontrolofthedisease.

Authors,suchasFocks(2003),Focksetal.(2000)andNathan

andFocks(2006)havesuggestedthatindicatorscalculatedbythe

areaofpropertiesandbyinhabitantswouldhaveabetter

predic-tioncapacityforthedisease-occurrenceriskthansolelyindicators

calculatedbyproperty,recommendingthedevelopmentofstudies

thataddresstheseissues.

Geographicinformationsystems(GIS),inadditiontobeingtools

forspatialanalysis,enablethespatialcomponenttobeincludedin

studiesofmetaxenicdiseasesandallowtheassociatedvariablesto

bemappedandspatiallyanalyzed.SomeexamplesofGIS

applica-tionsarethestudiesdevelopedbyBarreraetal.(2000),Mondini

andChiaravalloti-Neto(2008)andTranetal.(2004).

Theaimsofthisstudyweretodescribetheoccurrenceofdengue

inspaceandtimeandtoassessthepotentialuseofentomologic

indicatorsthatwerebasedonthequantificationofpupaeandon

thequantificationofcontainerswithimmatureformsofAe.aegypti

calculatedbypropertyandbyareatopredictdengueoccurrence.

2. Materialsandmethods

ThemunicipalityofSãoJosédoRioPretoislocatedinthe

north-westregionofthestateofSãoPaulo(Fig.1A),Brazil(20◦₄₉′₁₁′′_S

and49◦₂₂′₄₆′′_W)._The_altitude_of_this_municipality_is₄₇₅_m_above

sealevel,anditissituatedinaregionoftropicalclimatewithan

averageannualtemperatureof25◦_C_and_an_average_annual

pre-cipitationof1410mm.SãoJosédoRio Pretoisthemaincityin

anareaconsistingof102municipalities.Thecityhasmorethan

400,000inhabitants,analmostuniversalwatersupplysystem,and

developmentindicescomparabletothoseofdevelopedareas.

AfterthereintroductionofAe.aegyptiinthecityin1985,the

firstautochthonouscasesofdengueoccurredin1990,whenthe

presenceofDENV-1wasdetected.In1998,DENV-2wasdetected,

in2005DENV-3wasdetected(Mondinietal.,2009),andin2011,

DENV-4wasdetected(Roccoetal.,2012).Sincethe2000s,thecity

hasbeenconsideredendemicfordengue(Mondinietal.,2005).

Alloftheautochthonousdenguecasesthatoccurredintheurban

areaofSãoJosédoRioPretobetweenSeptember2005andAugust

2007wereconsideredinthisstudy.Thecaseswereconfirmedby

serologictesting(ELISA-IgM),RT-PCRorclinical–epidemiological

criterionandwereregisteredintheNotifiableDiseasesInformation

System(SINAN).Duringthisperiod,serologictestingwasusedto

confirmthedenguecasesuntiltheincidencehadreached

approx-imately300casesper100,000inhabitants.Afterthat,accordingto

thegovernmentcasedefinition,theclinical–epidemiologic

crite-rionwasusedtoconfirmthedenguecases.

BetweenSeptember2005andAugust2007,aviraldiagnostic

testwasperformedusingtechniquesof molecularbiology with

blood samples frompatients withclinical diagnosesof dengue

fever(Bronzoniet al.,2005)who wereselectedin healthunits.

Theirserawerestoredat−80◦C. ViralRNAwasextractedfrom

140␮lofserumwiththeQIAampViralRNAMinikit(QIAGEN,Inc.)

asdescribedbythemanufacturer.Themethodologydescribedby

Bronzonietal.(2005)wasusedtodetectthedenguevirusserotypes present.

Theannualandmonthlyincidenceratesofdenguewere

cal-culatedfor themunicipality duringtheperiod fromSeptember

2005toAugust2007.Themonthlyrateswerecompared tothe

BI(numberofcontainerswithAe.aegyptiper100premises)

esti-matedmonthlyratesfortheregionofSãoJosédoRioPreto,based

onthedatarecordedintheSystemSISAWEBoftheEndemic

Con-trolSuperintendence.Wecalculated,fortheannualperiodsfrom

September2005toAugust2006andSeptember2006toAugust

2007,thePearsoncorrelationcoefficients betweenthemonthly

incidencesandBI,laggedtwoandthreemonthsinrelationtothe

rates,builtscatterplotsandfittedLOESScurves.

Ageo-codification ofdengue cases wasperformedbased on

thepatients’residenceaddressesbymeansofmetric

interpola-tionusingthecartographicbaseofthestreetaxesofSãoJosédo

RioPreto.Thematicmapswerethenproducedshowingthedengue

incidencerateinthemunicipalityaccordingtoitsurbancensus

tracts,andthecasesforwhichtheserotypeswereidentifiedwere

mappedfortheperiodfromSeptember2005toAugust2007.

Anevaluationoftherelationshipbetweendengueand

ento-mologicalindicatorswasperformedintheJaguaréneighborhood,

which is located on the northern zone of the city (Fig. 1B).

Thisneighborhoodhasapproximately37,000inhabitants,

approx-imately11,000propertiesandanareaof397.5ha.Thecriteriafor

selectionwere:thehighestlevelsofinfestationofAe.aegyptiin

relationtotheotherneighborhoods,ahighincidenceofdengue,

thepresenceofresidentswithdifferentsocioeconomicstatuses,

andresidentialandnon-residentialproperties.

AnentomologicalsurveywasconductedintheJaguaré

neigh-borhoodbetweenDecember2006andFebruary2007.Thisperiod

waschosenbecausethesemonthsarethemostfavorablemonths

forthedevelopmentofthevectorinthisregionandincludean

infestationpeak (Barbosaet al.,2010).Inthissurvey, ateamof

fieldresearcherswerecontractedforthisactivityandaccordingly

trained;theyvisitedallofthepropertiesintheneighborhoodand

collectedalloftheobservedlarvaeduringthe3rdand4thstages

andallofthepupaeincontainerswiththepresenceofculicids.The

collectedspecimenswereidentifiedandquantified.Tominimize

thenumber ofpropertiesthatcouldnotbeinspected,thevisits

wereperformedduringtheworkingdaysoftheweekuntil7pm

andonSaturdays.Othervisitswereperformedtotargetproperties

thatwerenotinspectedduringthefirstvisit.Moredetailsabout

thissurveyareprovidedbyFavaroetal.(2013).

TheentomologicalindicatorsconsideredforAe.aegyptiincluded

twoindicescalculatedperproperty(theBIandthenumberofpupae

perproperty)andtwoindicescalculatedbyarea(thenumberof

containerswithimmatureformsperhectareandthenumberof

pupaeperhectare).Thepropertyareasinhectareswereobtained

(3)

F.Chiaravalloti-Netoetal./ActaTropica142(2015)167–177 169

Fig.1.(A)MunicipalityofSãoJosédoRioPreto,stateofSãoPaulo,Brazil.(B)JaguaréneighborhoodandurbanperimeterofSãoJosédoRioPreto.

Theseindicatorswerecalculatedforthewholeneighborhoodand

byblocks(blockswithlessthanfivepropertiesweregroupedwith

adjacentblocks).Inaddition,thenumberofpupaeperinhabitant

wasestimatedfortheentireneighborhood.

Basedontheentomologicalindicatorscalculatedperblock,

the-maticmapswerebuilttovisualize theirspatialdistributions in

theJaguaréneighborhood.Weassumedthatthecentroidsofthe

geometricrepresentationsof theblocks couldbeconsideredas

representativesofpolygonobjectsand,therefore,assamplesofa

continuousgeographicphenomenoninphysicalspacethatcould

beusedtomodelastochasticGaussianpunctualprocess(Bailey

andGatrell,1995).

Theordinarykrigingmethodwasusedtoobtainamathematical

modelandtoconstructthematicmapsbasedonstatistical

inter-polationusingthemethodofgeneralizedweightedleastsquares

estimation,withtheweightsdefinedbythesemi-variancefunction

(semivariogram).Theisotropicsemivariogramwasdefinedwith

asphericalfunctionwiththeparametersnugget,contributionand

amplitude.WeusedthepackagegeoRfromtheprogramRversion

3.0.2toperformthestatisticalinterpolation(RibeiroandDiggle,

2001).

RegardingdenguecasesintheJaguaréneighborhood,we

con-sideredjusttheonesthatoccurredbetweenJanuaryandMay2006

and2007.Thesebothtimeperiodsrepresentedmorethan90%of

thetotalnumberofcasesrecordedduringthesetwoyearsandthey

includethepeaksofincidenceandvectorinfestation.Injustifying

theinclusionofthe2006periodintheanalysisoftherelationship

betweenentomologicindicatoranddengueoccurrence,we

con-sideredthattheinfestationpatternsduringthesummersof2005/6

and 2006/7 didnot have expressive variations; there were no

atypicalclimaticvariations,noalterationsofthecontrolactivities

developedand nostructuralalterations intheneighborhood.In

addition,thiscomparisonwasopportunisticbecausetheepidemic

thatoccurredin2006exposedthecompletelysusceptible

popu-lationtotherecentlyintroducedserotypeDENV3(Mondinietal.,

2009).

Torepresenttheautochthonouscasesofdenguethatoccurred

intheJaguaréneighborhoodin2006and2007,weusedtheKernel

ratio obtainedfromthedivision of theKernel function applied

tothesitesofdengueoccurrencebytheresultoftheapplication

ofthisfunctiontothecentroidsofthelotscorrespondingtothe

residencesintheneighborhood,weightedbythenumberof

inhab-itantsper residence(Câmara and Carvalho, 2004).Thisprocess

generated measurements that corresponded to the incidence

rates.Theserateswerepresentedinthematicmapsaccordingto

quintiles.

We used a generalized additive model (GAM), in a

spa-tialcase–controldesigned,toevaluatetherelationshipbetween

dengueandentomologicalmeasures(HastieandTibshirani,1986).

The autochthonous dengue cases were considered the cases

themselvesandthecontrolstorepresenttheinhabitantsof the

neighborhood(thesourcepopulationofthedenguecases)were

generatedrandomly,inaproportionof1:1withthecases.We

cal-culatedtheamountofcontrolsfor eachcensustract,weighting

byitspopulation.ArcGIS10.2softwarewasusedtogeneratethe

randomspatialdistributionofthecontrols.Valuesofthe

entomo-logicalmeasuresforeachcaseandeachcontrolwereobtainedusing

themaps ofthe entomologicalindicators generatedby kriging,

consideringthegeographicalcoordinatesofthecasesandcontrols

(Barbosaetal.,2014).

TheGAMwereadjustedconsideringtheoccurrenceofdengue

(4)

(January toMay of 2006and of 2007) and each entomological

indexasapossibleexplanatoryvariable(e.g.,webuiltamodelfor

eachstudyperiodandeachentomologicalindex).Thegeographic

coordinatesofthecasesandcontrolswereusedasthespatial

com-ponentofthemodels.Itwasconsideredassmoothingfunctions

ofthecoordinateswithabandwidthof100mrepresenting,in

thissituation,themeanflightradiousofAe.aegypti(Barbosaetal.,

2014;HastieandTibshirani,1986).Theentomologicalmeasures

werescaled(usingtheirmeansandstandarddeviations)allowing

tocomparewiththeresultsofthemodels.Weusedthesoftware

Rversion3.0.2 withthe epigamlibrary toadjust thesemodels

(Barbosaetal.,2014).

ThisstudywasapprovedbytheEthicalReviewBoardofthe

FaculdadedeMedicinadeSãoJosédoRioPreto.

3. Results

TheincidenceratesofdenguefeverinSãoJosédoRioPretofor

theannualperiodsfromSeptember2005toAugust2006(2005/6)

andSeptember2006toAugust2007(2006/7)were3014.6and

2370.1casesper 100,000inhabitant-years, respectively(12,526

autochthonouscasesin2005/6and10,051in2006/7).Fig.2Aand

Bshowstheseratesaccordingtotheurbancensussectorsandin

quintiles,in whichwenotedthatthepresenceofheterogeneity

variedbetween0and13,184.3casesper100,000inhabitant-years

duringthefirstyearandbetween0and8206.3duringthesecond

year.

Regardingthecharacterizationofthecirculatingserotypes,in

2005/6,theviralserotypespresentwereidentifiedbyPCRin336of

718serumsamples,ofwhich331containedDENV3,fourcontained

DENV2,andonecontainedbothoftheseserotypes.In2006/7,94

sampleswereprocessed,ofwhich46werepositivelyidentified:

45asDENV3andoneasDENV2.Thedenguecasesforwhichthe

viralserotypewasidentifiedforthetwoannualperiodsanalyzed

areshowninFig.2CandD,respectively.

Fig. 3 shows the partial monthly incidence rate (including

only the dengue cases confirmed by either serological testing

orRT-PCR)and thetotal monthlyincidencerate(including the

denguecasesconfirmedbyserologicaltesting,RT-PCRandbythe

clinical–epidemiologic criterion). During 2005/6, the total

inci-denceratewasalmostequaltothepartialincidenceratesuntil

March,and,inthenextyearperiod,betweenJanuaryandMarchof

2007,approximatelyhalfoftheconfirmeddenguecaseswere

con-firmedbyeitherserologicaltestingorRT-PCR.Theincidencerates

weregreaterin2005/6thanin2006/7,althoughshowedsimilar

behaviorduringthetwoannualperiodsstudied.Moreover,both

demonstratedincidencepeaksinAprilandconcentrationsofmore

than90%ofthecasesbetweenJanuaryandMay.

Fig.3alsopresentsthemonthlyBIbetweenSeptember2005

andAugust2007fortheregionofSãoJosédoRioPreto;wenoted

similarinfestationlevelsandseasonalpatternsinthetwoyears,

withhighervaluesbetweenDecemberandMarch.Acomparisonof

thetwocurvesshowedthattheincidencefollowedtheinfestation

curve,withtheinfestationpeaksprecedingtheincidencepeaks

bytwotothreemonths.Thehighestcorrelationcoefficients,for

bothannualperiods,wereobtainedwhenwelaggedtheBIfortwo

monthandtheirvalueswere0.94(p=0.000)and0.84(p=0.001),

respectively,for2005/6and2006/7.ThescatterplotsandtheLOESS

curveswerebetterfittedforthefirstannualperiodcomparedtothe

second(Fig.4).

IntheJaguaréneighborhood,1528and563autochthonouscases

ofdenguefeverwererecordedfortheperiods2005/6and2006/7,

respectively,ofwhich96.2%and90.9%ofcasesoccurredbetween

JanuaryandMay.Fig.5AandBshowstheincidenceratesresulting

fromtheKernelratiointhatperiodforthetwoyearsstudied,and

wenotethedisseminationofthediseasethroughouttheentire

neighborhood,withhighervaluesin2006.

Fig.6AandBshowsthespatialriskmapswiththeoddsratio(OR)

isolinesfordengueandFig.6CandDshowstherespectivepvalue

mapsfor2006and2007,asaresultofapplyingGAMconsidering

onlythespatialdistributionofthecasesandcontrols.Theareas

withORover1andsignificantarebroaderin2006thanin2007,

withisinagreementwiththeincidenceratemapspresentedin

Fig.5AandB.

IntheentomologicalsurveyperformedintheJaguaré

neighbor-hood,9875propertieswereinspected(89.8%oftheneighborhood’s

properties).Totalsof36,119larvaeand4178pupaeofAe.aegypti

werefoundin1051propertiesand1867containers.Thelarvaeand

pupaewerefound,respectively,in1015and442propertiesand

in1788and647containers.Thefollowingentomological

indica-torsofAe.aegyptiwereobtainedintheJaguaréneighborhood:18.9

containerswithimmatureformsper100properties(BI)and0.42

pupaeperproperty,5.2containerswithimmatureformsand11.6

pupaeperhectare,and0.07containerswithimmatureformsand

0.15pupaeperinhabitant.

Fig.7showsthemapsoftheentomologicalindicatorsobtained

bykrigingbasedontheinformationaggregatedbyblocks.TheBI

andthenumberofpupaeofAe.aegyptiperpropertyareshownin

Fig.7AandB,andthenumberofcontainersharboringAe.aegyptiper

hectareandthenumberofAe.aegyptipupaeperhectareareshown

inFig.7CandD.Bothindicatorscalculatedperpropertyshowed

thesamepattern, which alsooccurredfor theindicators

calcu-latedperhectare.However,thecomparisonbetweentheindicators

calculatedperpropertyandthosecalculatedperhectarerevealed

differences,especiallyinthesouthernandwesternregionsofthe

Jaguaréneighborhood:alowinfestationbasedontheindicators

calculatedperhectareandahighinfestationbasedontheindicators

calculatedperproperty(Fig.7).

Table1presentsthemodelsshowingtherelationshipbetween

occurrenceof dengue,for bothstudyperiods,and foreach one

oftheentomologicalindicatorsevaluated,adjustedforthespatial

distributionofthecasesandcontrols.In2006,allentomological

indicatorswereconsideredriskfactorstotheoccurrenceofdengue

(OR>1andsignificant),howeverthosecalculatedperhectare

pre-sentedhighervaluesofORthantheindicesperproperty.In2007,

onlytheindicescalculatedperhectarewereconsideredriskfactors

fortheoccurrenceofdengue,neverthelesswithlowervaluesthan

in2006.Indicatorscalculatedperhectareshowed,inbothstudy

periods,tobebetterpredictorsoftheoccurrenceofthediseasethan

thosecalculatedperproperty,furthermore,theydemonstratedto

haveasimilarpredictionpower.

Theresultspresentedaboveareinconsonancewiththe

entomo-logicalindexmaps(Fig.7A–D),theincidenceratemaps(Fig.5Aand

Table1

Dengueriskestimatedmodels*_for_each_one_of_the_scaled**_{entomological}_indices evaluated,adjustedforthespatialdistributionofthecasesandcontrols,Jaguaré neighborhood,municipalityofSãoJosédoRioPreto,stateofSãoPaulo,Brazil.

Oddsratio 95%Confidenceinterval pvalue***

JanuarytoMayof2006

BreteauIndex 1.06 1.03–1.09 0.000

Pupaeperproperty 1.09 1.06–1.13 0.000 Containerperhectare 1.35 1.31–1.39 0.000

Pupaeperhectare 1.31 1.27–1.35 0.000

JanuarytoMayof2007

Breteauindex 0.98 0.92–1.04 0.434

Pupaeperproperty 1.01 0.95–1.07 0.735 Containerperhectare 1.17 1.11–1.23 0.000

Pupaeperhectare 1.17 1.11–1.24 0.000

(5)

(6)

Fig.3.Partialandtotalmonthlyincidenceratesofdengue,municipalityofSãoJosédoRioPreto,andmonthlyBreteauindices(BI),regionofSãoJosédoRioPreto,September 2005toAugust2007,stateofSãoPaulo,Brazil.

B)andthespatialriskmaps(Fig.6AandB).Itcanbeseenthatthe

indicesperhectare,inbothperiods,hadahigherdegreeof

concord-ancewithdenguethanthosecalculatedperproperty.Inaddition,

itisworthnotingthatthedegreeofconcordanceforindicesper

hectarewithdenguewashigherin2006thanin2007,whichisin

accordancewiththelowervaluesofORobtainedin2007.

4. Discussion

AccordingtoTeixeiraetal.(2005),thedenguefeverepidemics

inBrazilhaveoccurredaftertheintroductionorreintroductionof

certainvirusserotypeswhenahighproportionofsusceptible

indi-vidualsexistsinthepopulation.Thispatterncouldbeverifiedin

SãoJosédoRioPreto,wheretheresultsobtainedthroughPCR

sug-gestedthatalmostallofthecasescouldbeattributedtoDENV3,a

viruswhoseintroductionoccurredin2005(Mondinietal.,2009).In

additiontothehighdegreeofsusceptibilityofthepopulation,this

scenariogeneratedhighincidencesofdenguefeverin2005/6and

2006/7,withcasesthroughouttheurbanareaofthemunicipality.

Tocomparetheentomologicalindicatorsandtheincidenceof

denguefeverinagivenneighborhood,itwouldbeidealtoperform

entomologicmeasuresinaperiodwhenthedenguefever

trans-missionoccurredinacompletelysusceptiblepopulation(Mondini

etal.,2009).Inthisstudy,thecomparisonperformedbetweenthe

indicatorsmeasuredinDecember2006toFebruary2007andthe

incidencesofdenguefeverin2007hadthedisadvantageofpartof

thepopulationintheneighborhoodhavingbeenimmunized

dur-ingthepreviousperiodbytheserotypeDENV3.Thisimmunization

alteredthepotentialoftheindicatorsofinfestationtoestimate

thetransmissionrisk ofdenguefeverbecausefurther

transmis-sionnowalsodepended,amongotherfactors,ontheproportionof

(7)

Fig.5.IncidenceratesofdenguefromJanuarytoMayof2006(A),andfromJanuarytoMayof2007(B),Jaguaréneighborhood,municipalityofSãoJosédoRioPreto,stateof SãoPaulo,Brazil.

thepopulationsusceptibletoDENV3.Manyauthorsnotethatthe

entomologicalthresholdforthetransmissionofdenguedepends,

amongotherfactors,ontheimmunologicstatusofthepopulation

(Focks,2003;Focksetal.,2000;Ghouthetal.,2012;Halstead,2000; Morrisonetal.,2008).

Thedecisiontouseentomologicmeasuresobtainedinthe

sum-mer2006/7tocomparewiththeincidenceratesofdenguefeverin

both2006and2007intheJaguaréneighborhoodwasbasedon

thepatternof infestationobserved between2005and 2007for

theregionofSãoJosédoRioPreto,whichshowedthesame

sea-sonalbehaviorandsimilarlevelsofinfestation,aswasnotedinthe

methodology.

Regardingthecapacityoftheentomologicalindicatorsto

indi-catethespatialriskofdengueoccurrencein2006and2007,both

indicatorscalculatedperhectareshowedbetterresultsthanthose

calculatedperproperty,throughoutthoseperiods.Moreover,the

entomologicalindicatorsbasedonthenumberofcontainerswith

Ae.aegyptiandonthenumberofpupaeidentifiedsimilarpatterns

ofinfestation,andthelatterdidnotprovideadditionalinformation

inrelationtothefindingsprovidedbytheclassicindicators.

For authorssuchas Focks(2003), Focksand Chadee(1997),

andNathanandFocks(2006),theindicatorsbasedoncontainer

positivitywouldhaveweakrelationshipswiththeriskofdengue

occurrence, while theindicators based onpupae quantification

wouldbemoreadequatetoestimatethatrisk.However,inthis

study,themaindifferenceregardingthecapacityofthe

entomo-logicalindicatorstoindicatespatialdengueriskwasnotrelated

totheuseofpupaequantificationbuttotheapplicationof

indi-catorscalculatedperhectare.Bothoftheindicatorscalculatedper

hectare,notonlytheonebasedonthequantificationofpupae,were

inagreementwiththedengueriskmaps.

Thisresultisinagreementwiththeargumentsputforwardby

severalauthorswho noted thatindices calculated byareamay

bemore appropriatemeasuresfor assessing theriskof dengue

(Barreraetal.,2006;Focks,2003;Focksetal.,1993,2000;Focks andChadee,1997;Getisetal.,2003;Morrisonetal.,2004;Nathan andFocks,2006;Romero-VivasandFalconar,2005).

Oneexplanationforthedifferencesfoundbetweenthe

indica-torscalculatedperpropertyandthosecalculatedperhectarewas

thepresence ofpropertiessuchasschools, churches,clubsand

parks withlargeareasin theJaguaréneighborhood;becauseof

thoselargeareas,therewasahigherprobabilityofhosting

con-tainerswithimmatureformsofAe.aegypticomparedtosmaller

properties.Usingthoselargeareastocalculateindicesper

prop-erty ledto artificialincreases in those values compared to the

indicescalculatedperhectare(Favaroetal.,2013).Theindicators

calculatedperpropertymighthavesimilarperformancetothose

calculatedperhectareinareaswherethepropertieshad

homo-geneoussizes,butthishomogeneityisnotacommonsituationin

intermediate-sizeandlargemunicipalitiesinBrazil.

Therefore,inSãoJosédoRioPretoandinotherBraziliancities

withsimilarcharacteristics,changingthecalculationof

entomo-logicalindicesbysubstitutingthenumberofpropertieswiththeir

areascouldprovidetheseindiceswithagreatercapacityto

pre-dictthespatialriskofdenguefeveroccurrence.Eventhen,theuse

oftheseindicatorsisrestrictedbecausetheindicatorsshouldnot

beconsideredalone.Theindicatorsmake senseonlyiftheyare

analyzedwithothervariables relatedtothedynamicsof

trans-mission,suchasthedegreeofimmunityofthepopulationforthe

differentserotypesofthedenguevirus;theserotypes,genotypes

andcirculatingclades;vectorbehavior;andweather,

environmen-talandsocioeconomiccharacteristics(Drumondetal.,2013,2012;

Focksetal.,1995;Ghouthetal.,2012;Kuno,1977;Leeetal.,2012; Morrisonetal.,2008;Teixeiraetal.,2005;Yamanakaetal.,2011).

Eventhoughtheindices calculatedforpropertyhaveproved

worsepredictorsofthespatialriskofoccurrenceofdenguethan

thosecalculatedperhectare,BI,laggedtwomonths,wasagood

predictoroftemporalriskofdengue,asanotherstudyconducted

intheregionofSãoJosédoRioPretohadalreadyshown(Diboetal.,

2008).

The degree interference of immunity in the relationship

betweenentomologicalindicesanddenguebecameevidentinthis

study,sinceitwasfoundhigherORvaluesfor2006than2007,in

(8)

Fig.6.Spatialriskmapfordenguewithoddsratio(OR)fromJanuarytoMayof2006(A)andfromJanuarytoMayof2007(C)andpvaluemapforthefirstperiod(C)andfor thesecondperiod(D),Jaguaréneighborhood,municipalityofSãoJosédoRioPreto,stateofSãoPaulo,Brazil.

entirecityin2005/6thanin2006/7.Thesediscrepanciesare

proba-blyduetothehigherdegreeofimmunityagainstDENV3in2006/7

thanin2005/6.Ghouthet al.(2012)andMorrisonetal.(2008)

pointedoutthedecreaseintheentomologicalindicator’saccuracy

topinpointdengueriskwouldbeexpectedinpartiallyimmunized

areas.

AccordingtotheresultsofthisworkandoftheworkofNi ˜no

(2008),theuseofgeostatisticaltechniques,suchassemivariograms

andkriging,toestimatethevaluesofentomologicalindicatorsin

ageographicareaallowsfortheidentificationofsiteswithgreater

infestationbyAe.aegyptiwithoutbeingrestrictedtoregionswhere

anoutlinehasbeenpreviouslyestablished.TheuseofGIStogether

withspatialanalysesalsoallowsfortheassessmentofthe

relation-shipbetweendengueandtheentomologicalindicators,whichwas

observedinboththisstudyandthestudybySanchézetal.(2010,

2006).

However,toconductstudiesidentifyingareaswithhigher

prob-abilitiesof theoccurrence of diseaseor ofAe. aegypti withthe

goalofassessingtherelationshipbetweendengue,riskfactors,and

measuresforvectorcontrolandtoidentifyandselectpriorityareas

forvector-controlactivities(Eisenetal.,2009;Resendesetal.,2010;

Thaietal.,2010),itisnecessarytohavegeo-referencedmapsofthe

assessmentunitsofthesitesstudied,whichispresentlyenabled

by the use of increasingly user-friendly GIS and by the

avail-ability of geo-referenced databases (Eisen and Lozano-Fuentes,

2009).

Anotheraspecttobeconsideredin thediscussionaboutthe

relationship between entomological indicators and the risk of

dengue is thelevel of vectorinfestation. São José do Rio Preto

showedlowerlevelsofinfestation(Barbosaetal.,2010;Ferreira

andChiaravalloti-Neto,2007)thansitesusedinstudiesinwhich

otherindicatorsbasedonpupaequantificationwerebetter

pre-dictors of dengue occurrence (Barbazan et al., 2008; Barrera

et al., 2006; Focks, 2003). The lower levels of infestation are

related to the almostuniversal and adequatewater provisions

and collectionof solid residuesand thelow frequency of large

water-holdingcontainers(Favaroetal.,2013)thatresultfromthe

(9)

Fig.7. BreteauIndex(BI)(A),numberofAe.aegyptipupaeperproperty(B),numberofcontainersharboringAe.aegyptilarvaeandpupaeperhectare(C),andnumberofAe. aegyptipupaeperhectare(D)intheJaguaréneighborhood,municipalityofSãoJosédoRioPreto,stateofSãoPaulo,Brazil,fromDecember2006toFebruary2007.

intermissioninthewatersupply.Despitethelowlevelsof

vec-torinfestation,denguehasendemicbehaviorinthemunicipality

(Mondinietal.,2005)withtheoccurrenceofepidemicpeaks(CVE, 2014).

Inadditiontothelackofdifferencesamongindicesbasedon

pupae quantification and container positivity, theuse of

ento-mological indicators that employ pupae quantification in cities

withcharacteristics suchas theones foundin São José do Rio

Preto would require sampling plans that use a larger

num-berof propertiesthan thesamplingplansusedtomeasurethe

indices based on container quantification, as finding

contain-ers with pupae can be considered a rare event (Favaro et al.,

2013).

Thepotentiallimitationsofthepresentstudyincludetheuse

entomologicalindicatorsthatweremeasuredforthesummerof

2006/7,inJaguaré,toberelatedwithdengueoccurrenceinboth

studyperiods;theuseofthedenguecasesthatwerenotifiedtothe

surveillancesystem,whichisknowntonotincludealloccurrences;

theconsiderationofthepatient’sresidenceastheprobablesiteof

infection;andtheuseoftheclinical–epidemiologicalcriterionto

confirmsomeofthedenguecases.However,atthetime ofthis

outbreak,theclinical–epidemiologicalcriterionwasofficiallyused

bythegovernmenttoquantifydenguecases.Asawaytoaddmore

concretedata,wealsousedPCRtoconfirmdenguecasesinour

analysis.

Despitetheselimitations,someofthepositivecharacteristicsof

thisstudyincludetheperformedanalyses,suchasthe

accomplish-mentof viraldiagnosticsusingtechniquesofmolecularbiology

onsamplesfrom patientswithsuspecteddengue;thescopeof

thesurveillancesystemofSãoJosédoRioPreto,whichincludes

unitswithprimary,secondaryandtertiarypublicandprivatecare

andisdistributedthroughoutthemunicipality;andthe

achieve-ment of conducting the entomological survey through the use

of a census format and not by sampling. Even the use of the

clinical–epidemiologicalcriteriontoconfirmdenguecasescould

beminimizedasapotentialbiasbecauseinepidemicsituations

(highprevalence),thepositivepredictivevalueofthistypeoftest

couldbeconsideredhigh.

Anaspecttobehighlightedwasthefactthatthemunicipality’s

populationhadahighsusceptibilitytotheDENV3serotypebefore

theperiodofdevelopmentofthisresearchandthealmost

com-pletepredominanceoftheoccurrenceofthis serotypebetween

2005and 2007,which facilitated theanalysisof the

(10)

theentomologicalindicators.Suchaspectaddingtheextentand

quality of theentomological survey and the plausibility of the

resultscaninpartminimizethecitedlimitations.

5. Conclusions

It is possible to conclude that, in the study neighborhood

(Jaguaré),theentomologicalindicescalculatedbyhectarewere

bet-terpredictorsofthespatialdistributionofdenguethantheindices

calculatedbyproperty,andthepupaequantificationdidnotexhibit

betterpredictionqualitiesthantheindicesbasedonthecontainer

positivity,inrelationtothespatialriskofdengueoccurrence.We

alsoconcludedthattheBI,laggedtwomonths,wasagood

predic-torofthetemporaldistributionofdengueinthecityofSãoJosédo

RioPreto.Evenwiththepopulationpartiallyimmunizedin2007,it

waspossibletofindassociationbetweenhigherinfestationlevels

andanincreasedriskofdengue.

Acknowledgments

ToMarlene CG Souza,EdmilsonRodrigues deOliveira,Célio

Vamberto,NivaldoFirminodaSilva,FernandoRobertodeOliveira

and Cícero Alves da Silva for their assistance with the field

work;toBeatrizACBelini,NeuzaFASantanaandPerpétuaMM

Sereno of the Laboratorio de Vetores of Superintendência de

Controle de Endemias for their assistance with the laboratory

work.

ToFundac¸ão deAmparo àPesquisado Estado deSão Paulo

(FAPESP) for the grants that supported this work (Process no.

2005/02800-8andProcessno.2006/02473-0).

References

Arredondo-Jimenez, J.I., Delgado-Valdez, K.M., 2006. Aedes aegypti pupal/demographic surveys in southern México: consistency and practi-cality. Ann. Trop. Med. Parasitol. 100, 17–32, http://dx.doi.org/10.1179/ 136485906X105480.

Bailey,T.C.,Gatrell,A.C.,1995.InteractiveSpatialDataAnalysis.PrenticeHall, Har-low.

Barbazan, P., Tuntaprasart, W., Souris, M., Demoraes, F., Nitatpattana, N., Boonyuan, W.,Gonzalez, J.P., 2008. Assessmentof a newstrategy, based onAedesaegypti(L.)pupalproductivity,forthesurveillanceandcontrolof denguetransmissioninThailand.Ann.Trop.Med.Parasitol.102,161–171, http://dx.doi.org/10.1179/136485908X252296.

Barbosa,A.A.C.,Fávaro,E.A.,Mondini,A.,Dibo,M.R.,Chiaravalloti-Neto,F.,2010. Evaluationof oviposition traps as an entomological surveillance method for Aedes aegypti (Diptera, Culicidae). Rev. Bras. Entomol. 54, 328–331, http://dx.doi.org/10.1590/S0085-56262010000200017.

Barbosa,G.L.,Donalísio,M.R.,Stephan,C.,Lourenc¸o,R.W.,Andrade,V.R.,Arduino, M.B.,Lima,V.L.C.,2014.Spatialdistributionoftheriskofdengueandthe ento-mologicalindicatorsinSumaré,StateofSãoPaulo,Brazil.PLoSNegl.Trop.Dis. 8,e2873,http://dx.doi.org/10.1371/journal.pntd.0002873.

Barrera,R.,Amador,M.,Clark,G.G.,2006.Useofthepupalsurveytechniquefor mea-suringAedesaegypti(Diptera:Culicidae)productivityinPuertoRico.Am.J.Trop. Med.Hyg.74,290–302,http://dx.doi.org/10.1590/S1020-49892000000900001. Barrera,R.,Delgado,N.,Jiménez,M.,Villalobos,I.,Romero,I.,2000.Estratificaciónde umaciudadhiperendémicaemdenguehemorrágico.Rev.Panam.SaludPublica 8,225–233.

Bhatt, S., Gething, P.W., Brady, O.J., Messina, J.P., Farlow, A.W., Moyes, C.L., Drake,J.M.,Browntein,J.S.,Hoen,A.G.,Sankoh,O.,Myers,M.F.,George,D.B., Jaenisch,T., Wint, G.R.W.,Simmons,C.P., Scott,T.W., Farrar,J.J., Hay, S.I., 2013.Theglobaldistributionandburdenofdengue.Nature496,504–507, http://dx.doi.org/10.1038/nature12060.

Bronzoni,R.V.,Baleotti,F.G.,Nogueira,R.M.R.,Nunes,M.,Figueiredo,L.T.M.,2005. Duplexreversetranscription-PCRfollowedbynestedPCRassaysfordetection andidentificationofBrazilianalphavirusesandflaviviruses.J.Clin.Microbiol. 43,696–702,http://dx.doi.org/10.1128/JCM.43.2.948-950.2005.

Câmara,G.,Carvalho,M.S.,2004.AnáliseEspacialdeEventos.In:Druck,S.,Caravalho, M.S.,Câmara,G.,Monteiro,A.M.V.(Eds.),AnáliseEspacialdeDadosGeográficos. Embrapa,Planaltina,pp.53–76.

CVE–CentrodeVigilânciaEpidemiológica ProfessorAlexandreVranjac,2014. SecretariadeSaúdedoEstadodeSãoPaulo.Dengue.Dadosestatísticos.São Paulo,Availablefrom:http://www.cve.saude.sp.gov.br/htm/cvedengue.html (accessed7.02.14).

Dibo, M.R., Chierotti, A.P., Ferrari, M.S., Mendonc¸a, A.L., Chiaravalloti-Neto, F., 2008. Study of the relationship between Aedes (Stegomyia)

aegypti egg and adult densities, dengue fever and climate in Miras-sol, state of São Paulo, Brazil. Mem. Inst. Oswaldo Cruz 103, 554–560, http://dx.doi.org/10.1590/S0074-02762008000600008.

Drumond,B.P.,Mondini,A.,Schmidt,D.J.,Bosch,I.,Nogueira,M.L.,2012. Pop-ulation dynamics of DENV-1 genotype V in Brazil is characterized by co-circulationandstrain/lineagereplacement.Arch.Virol.157,2061–2073, http://dx.doi.org/10.1007/s0075-012-1393-9.

Drumond,B.P.,Mondini,A.,Schmidt,D.J.,Bronzoni,R.V.M.,Bosch,I.,Nogueira, M.L., 2013. Circulation of differentlineages of dengue virus2, genotype American/AsianinBrazil:dynamicsandmolecular andphylogenetic char-acterization. PLoS ONE 8, e59422, http://dx.doi.org/10.1371/journal.pone. 0059422.

Eisen,L.,Beaty,B.J.,Morrison, A.C.,Scott, T.W.,2009.Proactivevectorcontrol strategiesandimprovedmonitoringandevaluationpracticesfordengue pre-vention.J.Med.Entomol.46,1245–1255,http://dx.doi.org/10.1603/033.046. 0601.

Eisen,L.,Lozano-Fuentes,S.,2009.Useofmappingandspatialandspace-time mod-elingapproachesinoperationalcontrolofAedesaegyptianddengue.PLoSNegl. Trop.Dis.3,e411,http://dx.doi.org/10.1371/journal.pntd.0000411.

Favaro, E.A., Dibo, M.R., Pereira, M., Chierotti, A.P., Rodrigues-Junior, A.L., Chiaravalloti-Neto,F.,2013.Aedesaegyptientomologicalindicesinanendemic areafordengueinSaoPauloState,Brazil.Rev.SaúdePública47,588–597, http://dx.doi.org/10.1590/S0034-8910.2013047004506.

Ferreira, A.C., Chiaravalloti-Neto, 2007. Infestac¸ão da área urbana por Aedes aegyptierelac¸ãocomníveissocioeconomicos.Rev.SaúdePública41,915–922, http://dx.doi.org/10.1590/S0034-89102007000600005.

Figueiredo,R.M.,Naveca,F.G.,Bastos,M.S.,Melo,M.N.,Viana,S.S.,Mourão,M.P., Costa,C.A.,Farias,I.P.,2008.Denguevirustype4,Manaus,Brazil.Emerg.Infect. Dis.14,667–669,http://dx.doi.org/10.3201/eid1404.071185.

Focks,D.A.,2003.AReviewofEntomologicalSamplingMethodsandIndicatorsfor DengueVectors.WorldHealthOrganization,Gainsvill.

Focks,D.A.,Brenner,R.J.,Hayes,J.,Daniels,E.,2000.Transmissionthresholdsfor dengueintermsofAedesaegyptipupaeperpersonwithdiscussionoftheirutility insourcereductionefforts.Am.J.Trop.Med.Hyg.62,11–80.

Focks,D.A.,Chadee,D.D.,1997. Pupalsurvey: anepidemiologically significant surveillancemethodforAedesaegypti:anexampleusingdatafromTrinidad. Am.J.Trop.Med.Hyg.56,159–167.

Focks,D.A.,Haile,D.G.,Daniels,E.,Mount,G.A.,1993.Dynamiclifetablemodelfor Aedesaegypti(Diptera:Culicidae):analysisoftheliteratureandmodel develop-ment.J.Med.Entomol.30,1003–1017.

Focks,D.A.,Daniels,E.,Haile,D.G.,Keesling,J.E.,1995.Asimulationmodelofthe epidemiologyofurbandenguefever:literatureanalysis,modeldevelopment, preliminaryvalidation,andsamplesofsimulationresults.Am.J.Trop.Med.Hyg. 53,489–506.

Getis,A.,Morrison,A.M.Y.C.,Gray,K.,Scott,T.W.,2003.Characteristicsofthespatial patternofthedenguevector,Aedesaegypti,inIquitos,Peru.Am.J.Trop.Med. Hyg.69,494–505.

Ghouth,A.S.B.,Amarasinghe,A.,Letson,G.W.,2012.DengueOutbreakinHadramout, Yemen,2010: an epidemiological perspective.Am. J. Trop.Med.Hyg.86, 1072–1076,http://dx.doi.org/10.4269/ajtmh.2012.11-0723.

Halstead,S.B.,2000.Successesandfailuresindenguecontrol–globalexperience. DengueBull.24,60–70.

Hastie,T., Tibshirani,R.,1986. Generalizedadditivemodels.Stat.Sci. 1,297– 318.

Kuno,G.,1977.Factorsinfluencingthetransmissionofdengueviruses.In:Gubler, D.J.,Kuno,G.(Eds.),DengueandDengueHaemorrhagicFever.CABInternational, London,pp.61–87.

Lee,K.S., Lo, S.,Tan, S.S., Chua, R., Tan,L.K., Xu,H., Ng, L.C., 2012. Dengue virus surveillance in Singaporereveals high viral diversity through mul-tiple introductions and in situ evolution. Infect. Genet. Evol. 12, 77–85, http://dx.doi.org/10.1016/j.meegid.2011.10.012.

Mondini, A., Bronzoni, R.V.M., Nunes, S.H.,Chiaravalloti-Neto, F., Massad, E., Alonso,W.J., Lázzaro, E.S.M., Ferraz, A.A., Zanotto, P.M.A., Nogueira, M.L., 2009.Spatio-temporaltrackingandphylodynamicsofanurbandengue out-breakinSãoPaulo,Brazil. PLoSNegl. Trop.Dis.3,e448,http://dx.doi.org/ 10.1371/journal.pntd.0000448.

Mondini, A., Chiaravalloti-Neto, F., Sanches, M.G., Lopes, J.C.C., 2005. Análise espacial da transmissão de dengue em cidade de porte médio do inte-rior paulista. Rev. Saúde Pública 39, 444–451, http://dx.doi.org/10.1590/ S0034-89102005000300016.

Mondini, A., Chiaravalloti-Neto, F., 2008. Spatial correlation of incidence of denguewithsocioeconomic,demographicandenvironmentalvariablesina Braziliancity. Sci.TotalEnviron. 393,241–248,http://dx.doi.org/10.1016/j. scitotenv.2008.01.010.

Morrison,A.C.,Gray,K.,Getis,A.,Astete,H.,Sihuincha,M.,Focks,D.,etal.,2004. TemporalandgeographicpatternsofAedesaegypti(Diptera:Culicidae) produc-tioninIquitos,Peru.J.Med.Entomol.41,1123–1142,http://dx.doi.org/10.1603/ 0022-2585-41.6.1123.

Morrison,A.C.,Zielinski-Gutierrez,E.,Scott,T.W.,Rosenberg,R.,2008.Defining chal-lengesandproposingsolutionsforcontrolofthevirusvectorAedesaegypti.PLoS Med.5,e68,http://dx.doi.org/10.1371/journal.pmed.0050068.

Nathan,M.B.,Focks,D.A.,2006.Pupal/demographicsurveystoinformdengue-vector control.Ann.Trop.Med.Parasitol.100,1–3.

(11)

Resendes,A.P.C.,Silveira,N.A.P.R.,Sabroza,P.C.,Souza-SantosR,2010. Determina-tionofpriorityáreasfordenguecontrolactions.Rev.SaúdePública44,274–282, http://dx.doi.org/10.1590/S0034-89102010000200007.

RibeiroJr.,P.J.,Diggle,P.,2001.geoR:apackageforgeostatisticalanalysis.R-NEWS 1,15–18.

Rocco,I.M.,Silveira,V.R.,Maeda,A.Y.,Silva,S.J.S.,Spenassatto,C.,Bisordi,I.,Suzuki,A., 2012.Firstisolationofdengue4inthestateofSãoPaulo,Brazil,2011.Rev.Inst. Med.Trop.54,49–51,http://dx.doi.org/10.1590/S0036-46652012000100009. Romero-Vivas,C.M.,Falconar,A.K.I.,2005.Investigationofrelationshipsbetween

Aedesaegyptiegg,larvae,pupae,andadultdensityindiceswheretheirmain breedingsiteswerelocatedindoors.J.Am.Mosq.ControlAssoc.21,15–21, http://dx.doi.org/10.2987/8756-971X(2005)21[15:IORBAA]2.0.CO;2. Sanchéz,L., Cortinas, J.,Pelaez, O., Gutierrez,H., Concepción, D., Stuyft,P.V.,

2010. BreteauIndex thresholdlevels indicatingriskfor dengue transmis-sioninareaswithlowAedesinfestation.Trop.Med.Int.Health15,173–175, http://dx.doi.org/10.1111/j.1365-3156.2009.02437.x.

Sanchéz,L.,Vanlerberghe,V.,Afonso,L.,Marquetti,M.C.,Guzman,M.G.,Bisset,J., etal.,2006.Aedesaegyptilarvalindicesandriskfordengueepidemics.Emerg. Infect.Dis.12,800–806.

Teixeira,M.G.,Costa,M.C.,Barreto,M.L.,Mota,E.,2005.Dengueanddengue hem-orrhagicfeverepidemicsinBrazil:whatresearchisneededbasedontrends,

surveillance,and controlexperiences?Cad. SaúdePública 21,1307–1315, http://dx.doi.org/10.1590/S0102-311X2005000500002.

Teixeira, M.G., Siqueira-Junior,J.B., Ferreira, G.L.C., Bricks, L.,Joint, G., 2013. EpidemiologicaltrendsofdenguediseaseinBrazil(2000–2010):a system-atic literature search and analysis. PLoS Negl. Trop. Dis. 7 (12), e2520, http://dx.doi.org/10.1371/journal.pntd.0002520.

Thai,K.T.D.,Nagelkerke,N.,Phong,H.L.,Nga,T.T.T.,Giao,P.T.,Hung,L.Q.,etal., 2010. Geographical heterogeneity of denguetransmission in two villages insouthernVietnam. Epidemiol.Infect.138,585–591,http://dx.doi.org/10. 1017/S095026880999046X.

Tran,A.,Deparis,X.,Dussart,P.,Morvan,J.,Rabarison,P.,Remy,F.,etal.,2004. Denguespatialandtemporalpatterns,FrenchGuiana,2001.Emerg.Infect.Dis. 10,615–621.

Yamanaka,A.,Mulyatno,K.C.,Susilowati,H.,Hendrianto,E.,Ginting,A.P.,Sary, D.D.,Rantam,F.A.,Soegijanto,S.,Konishi,E.,2011.Displacementofthe pre-dominantdenguevirusfromtype2totype1withasubsequentgenotype shiftfromIVtoIinSurabaya,Indonesia2008–2010.PLoSONE6,e27322, http://dx.doi.org/10.1371/journal.pone.0027322.