Assessment of asthma control using CARAT in patients with and without Allergic Rhinitis: A pilot study in primary care.

(1)

Please cite this article in press as: Domingues M, et al. Rev Port Pneumol. 2015.

http://dx.doi.org/10.1016/j.rppnen.2015.10.014

ARTICLE IN PRESS

+Model

RPPNEN-1119; No.ofPages4

RevPortPneumol.2015;xxx(xx):xxx---xxx

www.revportpneumol.org

BRIEF COMMUNICATION

Assessment of asthma control using CARAT in patients with and without Allergic Rhinitis: A pilot study in primary care

M. Domingues

^a

, R. Amaral

^b

, J.A. Fonseca

^b^,^c^,^d^,∗

, P. Azevedo

^e

, J. Correia-de-Sousa

^f^,^g

aSãoPedrodaCovaFamilyHealthUnit,Gondomar,Porto,Portugal

bCINTESIS--- CentreforResearchinHealthTechnologiesandInformationSystems---FacultyofMedicine,UniversityofPorto, Porto,Portugal

cAllergyUnit,CUFPortoInstitute&Hospital,Porto,Portugal

dInformationandDecisionSciencesDepartment---FacultyofMedicine,PortoUniversity,Porto,Portugal

eLaSaletteFamilyHealthUnit,OliveiradeAzeméis,Aveiro,Portugal

fLifeandHealthSciencesResearchInstitute(ICVS),SchoolofHealthSciences,UniversityofMinho,Braga,Portugal, ICVS/3B’s---PTGovernmentAssociateLaboratory,Braga/Guimarães,Portugal

gHorizonteFamilyHealthUnit,Matosinhos,Porto,Portugal

Received23August2015;accepted21October2015

KEYWORDS Asthma;

Control;

Portugal;

Questionnaires;

Rhinitis;

Allergic/diagnosis;

Treatmentoutcome

Abstract

Background: AsthmaandAllergicRhinitis(AR)aretwochronicinﬂammatorydiseasesthatare often concomitant.TheControlofAllergic RhinitisandAsthmaTest(CARAT)was developed toevaluatethecontrolofthesediseasesfromthepatients’perspective.Itsperformancein asthmapatientswithoutARhasnotbeenpreviouslystudied.

Aim: TotestthehypothesisthatCARATcanbeusedtoassessasthmacontrolinpatientswith asthmaandwithoutAR.

Methods:A cross-sectionalstudy wasconducted in3primaryhealthcarecentresinNorthern Portugal.AdultpatientsidentiﬁedintheElectronicPatientRecordwithadiagnosisofasthma were invited toparticipate. CARAT was used toassess asthma controland AsthmaControl Test(ACT)asacomparator.TheassociationsbetweenasthmapatientswithoutAR(AsAR)and with AR(AwAR)were analyzed withSpearman correlation.Additionally, ReceiverOperating Characteristic(ROC)curveanalysis,summarizedbyAreaUndertheCurve(AUC),wasusedto assessperformanceofCARATforscreeningasthmathatwasnotwell-controlled.

Results:A total of 103 asthma patients completed the study, 64 (62%) had AwAR and in 87 (85%) asthma was not well-controlled. We observed a strong correlation between CARAT andACT scores(r=0.734)inallasthmapatients andinbothgroups:AsAR(r=0.737)

∗Correspondingauthor.

E-mailaddress:fonseca.ja@gmail.com(J.A.Fonseca).

(2)

ARTICLE IN PRESS

+Model

2 M.Dominguesetal.

andAwAR(r=0.843).ROCcurvedemonstratedCARATashavingagooddiscriminativepower forbothAsARandAwARgroups(AUC=0.894and0.946,respectively).

Conclusion:These initialresultssuggestthatCARAT hasagooddiscriminative performance, similartootherasthmacontrolassessmenttools,forasthmapatientswithandwithoutAR.

licenses/by-nc-nd/4.0/).

Introduction

The ‘‘Allergic Rhinitis and its Impact on Asthma’’ (ARIA) highlightsa close relationship between asthma and Aller- gicRhinitis(AR)andrecommendsacombinedapproachto evaluateandmanagethesetwoconditions.¹

TheControlofAllergicRhinitisandAsthmaTest(CARAT) wasthe first questionnaire to assess the level of control of both asthma and AR using a single tool.^2,3 CARAT was developedusingathoroughformalmethodologicalapproach andthevalidationstudiesshowedithasthegoodmeasure- ment properties observed fulfilling the ten items on the COnsensus-basedStandardsfortheselectionofhealthMea- surementInstruments (COSMIN).³ It is a Patient-Reported Outcomethatquantifiesthedegreetowhichtherapygoals are met in patients with a previous diagnosis of AR and asthmaandcanbeusedatalllevelsofcare.^4---6As CARAT’s propertieshave beentested in patientswithboth asthma andrhinitis,theaimofthisstudywastotestthehypothesis thatCARATcanbeusedtoassessasthmacontrolinpatients withasthmaandwithoutAR.

Methods

Across-sectionalstudywasconductedinCaldasdasTaipas, a sub-urbanarea of Northern Portugal. All adult patients caredfor inoneof thethreeprimarycare centres(PCC), between November and December 2010, with an asthma diagnosisregisteredontheElectronicPatientRecord(EPR), wereeligibleforthestudy.Theexclusioncriteriawere:mis- diagnosis/codingerror conﬁrmed by both the patientand thefamilydoctor;physicalorcognitivedisabilitiesthatpre- ventedthecompletionofthe questionnaire;patientsthat couldnotattendanappointment.Aletterwassenttoalleli- giblepatients,explainingtheaimoftheprojectandinviting themtoparticipate.

The Ethics Committeeof the Regional HealthAdminis- trationapprovedthe study.Written informedconsentwas obtainedfromeachparticipant.

ParticipantswereclassiﬁedhashavingARbasedonthe presenceoftwoormoreofﬁvesymptoms,asrecommended byARIAguidelines.¹

CARAThas10questionsandaratingscalefrom0to30, witha cut-off value of 24 pointsto classifypatients: not well-controlled (≤24) and controlled asthma (>24).⁴ The AsthmaControl Test (ACT)⁷ wasusedasa comparatorfor asthmacontrolusingacut-offof>19,wellcontrolled,and

≤19, notwell-controlled. ForassessingARsymptoms, the VisualAnalogScale(VAS)wasused,rangingfrom0cm(good) to10cm(bad),withacut-offof5.5cmformildversusmod- erate/severerhinitis.

Statistical analysis was performed using IBM SPSS v21 (Armonk,NY;IBM Corp.). The associations between CARAT withACTscores inasthmapatientswithoutAR(AsAR)and with AR (AwAR) were analyzed by Spearman correlation.

ReceiverOperatingCharacteristic(ROC)curveanalysiswas usedtoassesstheperformanceofCARATscoreagainstACT, summarizedusingtheAreaUndertheCurve(AUC).Asignif- icancelevelof˛<0.05wasconsidered.Apoweranalysisfor 2independentproportionswasconductedposthoc.

Results

Ofthe192 patientsidentiﬁed103participated ---29could not be contacted, 5 refused to participate,20 missed >1 appointments, 7 were misdiagnosed and 28 (15%) were unable to complete the questionnaire. The participation rateofeligiblepatientswas72%.

From the 103 patients included, 64 (62%) had AwAR (Table1).Eighty-sevenparticipants(85%)wereclassiﬁedas havingnotwell-controlledasthmabasedontheCARATscore and56(54%)bytheACTscore.InthegroupofAwARahigher proportion were classiﬁed as having not well-controlled asthmabothbyCARATandACT,97%and61%,respectively, whichcompareswith64%and44%inAsARgroup(Table1).

Therewasastrongandsigniﬁcant correlation between ACT and CARAT scores in all asthma patients (r=0.734) andinbothAsARandAwARgroups(r=0.737andr=0.843, respectively)(Fig.1).

The performance of CARAT for detecting not well- controlledasthma,usingACTascomparatorwasplottedas ROCcurvesandtherespectiveAUCwere0.885forallasthma patients,0.894forAsARgroupand0.946forAwAR(Fig.1).

Discussion

ThisstudywastheﬁrsttotestCARATforevaluatingasthma controlinpatients withoutAR.Overallourresultssuggest thatCARATperformedwellinassessingthelevelofasthma controlregardless ofwhetherpatients hadrhinitisor not.

ThecorrelationsbetweenCARATandACTwerehigherthan 0.73 and the AUC higher than 0.88. These were slightly lower than those between Asthma Control Questionnaire (ACQ)andACT(r=−0.87to−0.89;AUCs=0.85---0.90)and

(3)

ARTICLE IN PRESS

+Model

AssessmentofasthmacontrolusingCARATinpatientswithandwithoutAllergicRhinitis 3

Table1 Characterizationofallasthmapatientsandinbothgroups:AsARand(AwAR).

Total n=103

AsAR n=39

AwAR n=64 Gender,n(%)

Male 26(25) 11(28) 15(23)

Female 77(75) 28(72) 49(77)

Age,years

Mean(SD) 49.51(18.07) 56.87(17.4) 45.03(17.07)

Min---Max 18---88 22---88 18---88

Eversmoked,n(%)

Yes 19(18) 8(20) 11(17)

No 84(82) 31(80) 53(83)

CARATscore

Median(IQR) 17(11---22) 23(18---26) 13(10---18)

AsthmaoverallcontrolbyCARAT,n(%)

Notwell-controlled 87(85) 25(64) 62(97)

Controlled 16(15) 14(36) 2(3)

VAS---rhinitis,median(IQR) 4.8(0---7.8) 0(0---0) 5.95(4.8---8.4)

Mildrhinitis,n(%) 65(64) 35(92) 30(47)

Moderate/severerhinitis,n(%) 37(36) 3(8) 34(53)

ACTscore

Median(IQR) 19(14---22) 20(17---22) 18(13---22)

AsthmacontrolbyACT,n(%)

Notwell-controlled 56(54) 17(44) 39(61)

Controlled 47(46) 22(56) 25(39)

AsAR:asthmawithoutallergicrhinitis;AwAR:asthmawithallergicrhinitis;CARAT:ControlofAllergicRhinitisandAsthmaTest;ACT:

AsthmaControlTest;VAS:VisualAnalogScale;SD:standarddeviation;Min---Max:Minimum---Maximum;IQR:interquartilerange.

1.0

0.8

0.6

Sensitivity^0.4

0.2

0.0

1.0

0.8

0.6

Sensitivity^0.4

0.2

0.0

1.0

0.8

0.6

Sensitivity^0.4

0.2

0.0

0.0 0.2 0.4 0.6 0.8 1.0

0.0 0.2 0.4

1-specificity 1-specificity 1-specificity

AUC=.885

r=.734* AUC=.894

r=.737* AUC=.946

r=.843*

0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0

Total AsAR AwAR

Figure1 ReceiverOperatingCharacteristic(ROC)curvesandtherespectivecorrelationofCARATscoreagainstACT,forscreening notwell-controlledasthma,inallpatientsandbothgroups:AsARandAwAR.AsAR:asthmawithoutallergicrhinitis;AwAR:asthma withallergicrhinitis.AUC:AreaUndertheCurve;rdeSpearman;*p<0.05.

similartothosebetween ACQandAsthma ControlScoring System(ACSS).^8---10Thedifferenceobservedintheproportion ofcontrolmeasuredwithCARATandACTwasalsoreported inastudyconductedinPortuguesepharmacies,suggesting thatCARATmayhavemoresensitivityasanasthmacontrol screening test in the community.⁶ A possible explanation for the differences inthe classiﬁcation of asthma control betweenCARATandACTmaybetheinclusioninACTofthe question‘‘Howwouldyourateyour asthmacontrolduring the past 4weeks?’’ Inpatients whodo notperceivepoor control,thisquestionmaydecreasethesensitivitytodetec- tingpoorcontrol.Infact,inthePortugueseAsthmasurvey

weobservedthat88%ofnotwell-controlledasthmapatients statedtheirdiseasewascontrolledwhenansweringaques- tionsimilartotheACTquestion.¹¹

Themainlimitationofthisstudyisnothavingthephysi- cian’s assessment of asthma control to compare withthe CARATscore,however,thissamplecanpotentially leadto generalizable results which can answer our study aim, as these patients were followed in primary care, by family doctors. ACT is a well-known asthma control assessment tool,usedfrequently in Portugal,even though noclinical validationstudy of thePortuguese (European) versionhas beenpublished.Thestudyhasanadequatestatisticalpower

(4)

ARTICLE IN PRESS

+Model

4 M.Dominguesetal.

(1−ˇ=0.662)andtheparticipationratewasgoodforthis typeof study.However,thesample size is alimitation as itdoes notallowfor furthersubgroup analysis.The inclu- sionofpatientswitharegistereddiagnosisofasthmainthe EPRisbothastrengthandlimitationofthestudy,because it reduces the probability of including patients with dis- easesotherthanasthma(speciﬁcity)andatthesametime, many patients with asthma, perhaps less severe asthma, wereprobablynotregistered intheEPR. Finally, thehigh levelof notwell-controlled asthmaobserved maybe par- tially explainedby the study design. There is a need for studiesconducted in differenthealthcare settings,in primary,secondary,tertiarycare toassesstheprevalence of uncontrolledasthmaindifferentsituations.

Inconclusion,theresultsobtainedfromthisstudysup- porttheuseofCARATtoevaluatethecontrolofasthmanot onlyinpatientswithrhinitisbutalsoinpatientswithoutAR.

Furtherstudiesareneeded toconﬁrm theapplicabilityof CARATinasthmapatientswithoutrhinitis.

Ethical disclosures

Protection of human and animal subjects.The authors declarethatnoexperimentswereperformedonhumansor animalsforthisstudy.

Conﬁdentialityofdata.Theauthorsdeclarethattheyhave followedtheprotocolsoftheirworkcenteronthepublica- tionofpatientdata.

Right to privacy and informed consent.The authors declarethatnopatientdataappearinthisarticle.

Funding

The study required no additional external funding. The statistical analysis was conducted with the support of the CARAT Project CINTESIS belonging to the Faculty of Medicine,UniversityofPorto.

Conﬂicts of interest

JoãoA.Fonseca,MD,PhD isthecoordinatoroftheCARAT project.Theauthorsdeclarethattheyhavenoconﬂictsof interestinrelationtothisstudy.

Acknowledgements

The authors wouldlike toacknowledgetoCláudia Camila Diasforhelpinthestatisticalanalyses.

References

1.BousquetJ,KhaltaevN,CruzA.AllergicRhinitisanditsImpact on Asthma (ARIA).Allergy. 2008;63 April (Suppl. 86):8---160, http://dx.doi.org/10.1111/j.1398-9995.2007.01620.x.

2.Nogueira-Silva L, Martins SV, Cruz-Correia R, Azevedo LF, Morais-AlmeidaM,Bugalho-AlmeidaA,etal.Controlofaller- gic rhinitis and asthma test --- a formal approach to the developmentofameasuringtool.RespirRes.2009;17(10):52, http://dx.doi.org/10.1186/1465-9921-10-52.

3.FonsecaJA,Nogueira-SilvaL,Morais-AlmeidaM,AzevedoL,Sa- SousaA,Branco-FerreiraM,etal.Validationofaquestionnaire (CARAT10)toassessrhinitisandasthmainpatientswithasthma.

Allergy.2010;65:1042---8.

4.AzevedoP,CorreiadeSousaJ,BousquetJ,Bugalho-AlmeidaA, DelGiaccoSR,DemolyP,etal.ControlofAllergicRhinitisand AsthmaTest(CARAT):disseminationandapplicationsinprimary care.PrimCareRespirJ.2013;22(1):112---6.

5.vanderLeeuwS,vanderMolenT, DekhuijzenPNR,Fonseca JA,vanGemertFA,GerthvanWijkR,etal.Theminimalclini- callyimportantdifferenceofthecontrolofallergicrhinitisand asthmatest(CARAT):cross-culturalvalidationandrelationwith pollencounts.NPJPrimCareRespirMed.2015:25.

6.Lourenc¸o O, Calado S, Sá-Sousa A, Fonseca JA. Evalua- tion of allergic rhinitis and asthma control in a Portuguese community pharmacy setting. J Manage Care Pharm: JMCP.

2014;20(5):513---22.

7.RoxoJP,PonteEV,RamosDC,PimentelL,D’OliveiraA,CruzA.

Portuguese-languageversionoftheAsthmaControlTest.JBras Pneumol.2010;36(2):159---66.

8.SchatzM,SorknessCA,LiJT,MarcusP,MurrayJJ,NathanRA, etal.AsthmaControlTest:reliability,validity,andresponsive- nessinpatientsnotpreviouslyfollowedbyasthmaspecialists.

JAllergyClinImmunol.2006;117(3):549---56.

9.ZhouX,DingFM,LinJT,YinKS.Validityofasthmacontroltest forasthmacontrolassessmentinChineseprimarycaresettings.

Chest.2009;135(4):904---10.

10.LeblancA,RobichaudP,LacasseY,BouletLP.Quantiﬁcationof asthmacontrol:validationoftheAsthmaControlScoringSys- tem.Allergy.2007;62:120---5.

11.Sá-Sousa A, AmaralR, Morais-Almeida M,Araújo L, Azevedo LF, Bugalho-Almeida A, et al. Asthma control in the Por- tugueseNationalAsthmaSurvey.RevPortPneumol.2015;21(4):

209---13.