Patient and physician delay in the diagnosis and treatment of non-small cell lung cancer in Turkey

Patient

and

physician

delay

in

the

diagnosis

and

treatment

of

non-small

cell

lung

cancer

in

Turkey

Ahmet

Selim

Yurdakul

*,

Celalettin

Kocatu¨rk

,

Hu¨lya

Bayiz

,

Soner

Gu¨rsoy

,

Ahmet

Bircan

_,

_Aysenaz

_O

_{¨ zcan}

_,

_Atilla

_Akkoc¸lu

_,

_Funda

_Uluorman

_,

_Pinar

_C¸elik

_,

Deniz

Ko¨ksal

,

Bahar

Ulubas¸

,

Eylem

Sercan

,

O

¨ mer

O

¨ zbudak

,

Tuncay

Go¨ksel

,

Tug˘ba

O

¨ nalan

,

Esra

Yamansavci

,

Figen

Tu¨rk

,

Go¨khan

Yuncu

,

C¸ig˘dem

C¸opuraslan

,

Tug˘ba

Mardal

,

Esin

Tuncay

,

Altemur

Karamustafaog˘lu

,

Pinar

Yildiz

,

Funda

Sec¸ik

,

Muhammet

Kaplan

,

Emel

C¸ag˘lar

,

Mediha

Ortako¨ylu¨

,

Mine

O

¨ nal

,

Akif

Turna

,

Evlin

Hekimog˘lu

,

Levent

Dalar

,

Sedat

Altin

,

Meral

Gu¨lhan

,

Eylem

Akpinar

,

I˙smail

Savas

,

Nalan

Firat

,

Gu¨ngo¨r

C¸amsari

,

Gu¨lc¸ihan

O

¨ zkan

,

Erdog˘an

C¸etinkaya

,

Emine

Kamilog˘lu

_,

_Bu¨lent

_C¸elik

_,

_Yavuz

_havlucu

a

PulmonaryDepartment,GaziUniversitySchoolofMedicine,Ankara,Turkey

b

PulmonaryandThoracicSurgeryDepartment,YedikuleChestDiseasesandChestSurgeryEducationandResearchHospital,Istanbul,Turkey

c_{PulmonaryDepartment,Atatu¨rkChestDiseasesandChestSurgeryEducationandResearchHospital,Ankara,Turkey}

d_{ThoracicSurgeryDepartment,I˙zmirSuatSerenChestDiseasesandChestSurgeryEducationandResearchHospital,I˙zmir,Turkey} e

PulmonaryDepartment,SuleymanDemirelUniversitySchoolofMedicine,Isparta,Turkey

f

PulmonaryDepartment,DokuzEylu¨lUniversitySchoolofMedicine,Ankara,Turkey

g

PulmonaryDepartment,CelalBayarUniversitySchoolofMedicine,Manisa,Turkey

h

PulmonaryDepartment,MersinUniversitySchoolofMedicine,Mersin,Turkey

i

PulmonaryDepartment,AkdenizUniversitySchoolofMedicine,Antalya,Turkey

j_{PulmonaryDepartment,EgeUniversitySchoolofMedicine,I˙zmir,Turkey} k_{PulmonaryDepartment,PamukkaleUniversitySchoolofMedicine,Denizli,Turkey} l

PulmonaryDepartment,AnkaraOncologyHospital,Ankara,Turkey

m

ThoracicSurgeryDepartment,TrakyaUniversitySchoolofMedicine,Edirne,Turkey

n

MedicalOnologyDepartment,DicleUniversitySchoolofMedicine,Diyarbakir,Turkey

o

ThoracicSurgeryDepartment,CerrahpasaUniversitySchoolofMedicine,Istanbul,Turkey

p

PulmonaryDepartment,UfukUniversitySchoolofMedicine,Ankara,Turkey

q_{PulmonaryDepartment,AnkaraUniversitySchoolofMedicine,Ankara,Turkey} r_{StatisticsDepartment,GaziUniversitySchoolofHealthSciences,Ankara,Turkey}

ARTICLE INFO

Articlehistory: Received25July2014

Receivedinrevisedform23December2014 Accepted30December2014

Availableonline7February2015 Keywords: Lungcancer Delay Patient Doctor ABSTRACT

Aim:Theearlydiagnosisandtreatmentoflungcancerareimportantfortheprognosisofpatientswith lungcancer.Thisstudywasundertakentoinvestigatepatientanddoctordelaysinthediagnosisand treatmentofNSCLCandthefactorsaffectingthesedelays.

Materialsandmethods:Atotalof1016patients,including926(91.1%)malesand90(8.9%)femaleswitha meanageof61.510.1 years,wereenrolledprospectivelyinthisstudybetweenMay2010andMay 2011from17 sitesinvariousTurkishprovinces.

Results:Thepatientdelaywasfoundtobe49.996.9days,doctordelaywasfoundtobe87.799.6days, andtotaldelaywasfoundtobe131.3135.2 days.Thereferraldelaywasfoundtobe61.6127.2 days, diagnosticdelaywasfoundtobe20.444.5 days,andtreatmentdelaywasfoundtobe24.454.9 days. Whenthemajorfactorsresponsibleforthesedelayswereexamined,patientdelaywasfoundtobemore frequent inworkers, whilereferraldelaywasfound tobemorefrequent inpatientslivinginvillages (p<0.05).Wedeterminedthatreferraldelay,doctordelay,andtotaldelayincreasedasthenumberofdoctors whowereconsultedby patientsincreased (p<0.05).Additionally, wedeterminedthatdiagnosticand treatmentdelaysweremorefrequentattheearlytumourstagesinNSCLCpatients(p<0.05).

* Correspondingauthorat:2144sok.No:11/9MustafaKemalMahallesi,Ankara06620,Turkey.Tel.:+903122026135. E-mailaddress:ayurdakul@gazi.edu.tr(A.S.Yurdakul).

ContentslistsavailableatScienceDirect

Cancer

Epidemiology

The

International

Journal

of

Cancer

Epidemiology,

Detection,

and

Prevention

j o urn a lhom e pa g e :ww w . ca nc e re pi d e mi ol o gy . ne t

http://dx.doi.org/10.1016/j.canep.2014.12.015

1. Introduction

Assmokingrateshaveincreased,lungcancerhasbecomethe mostfrequentlyobservedcancerinbothmenandwomen,andit hasthehighestrateofmortality.Itisalsothemostfatalcancer— lungcancerisresponsiblefor1.3milliondeathsannuallyacross theworldandcontinuestobeamajorhealthproblem[1].The age-standardizedincidenceoflungcancerinTurkeywasfound tobe75.8/100,000populationin menand9.6/100,000 popula-tionin women[2].Lung cancerincidenceis 30–35/100,000in the world, 48/100,000 in EU region among men and 13–14/ 100,000 in both EU region and world among women. The incidenceis7–10/100,000inourcountry[3].InTurkey,smoking rateswas27%inTurkishpeople,itwas46%inmenand13%in women[3].

Tumourstageistheleadingfactoraffectingprognosisinlung cancer.Approximately49%ofcaseswithlungcancerhaveadistant metastasisatonset.Mediastinallymphaticinvolvementisfoundin 26%ofthesepatients[4].Duringconsultation,80%ofcaseswith lungcancer were foundto be inoperable, leaving only 20% as candidates for surgical treatment [5]. NSCLC cases involving a distantmetastasisdemonstrateda mediansurvivaltimeof 4–5 monthswithnotreatment,andonly10%ofsuchcaseslivedfor 1year[6].Afive-yearsurvivalrateof67%wasreportedatstageIA, 57%wasreportedatstageIB,55%wasreportedatstageIIA,39% wasreported at stage IIB, and 23% wasreported stageat IIIA. Cancersurvivalisakeymeasureoftheeffectivenessofhealth-care systems. Verdecchia et al. analysed survival data for patients diagnosedwithcancer,collectedfrom47oftheEuropeancancer registriesparticipatingintheEUROCARE-4studyandfoundthat age-adjusted 5-year period survival was 10.9% in lung cancer

[7].Fiveyearsurvivalwaslowat9–11%intheUKandDenmark versus15–20%inAustralia,Canada,SwedenandNorway[8].

Becausepatientanddoctordelayswillalterthetumourstagein lungcancer,many patientslosetheirchanceforsurgerydue to diagnosticdelays,althoughtheywouldhavebeenresectablewhen their initialsymptomsappeared. Althougha 30-dayperiod has beenconsideredtobeanimportantcriterioninpatientdelayin previousstudies,nodefiniteperiodhasbeenspecifiedfordoctor delayanditssubdomains.TheBritishThoracicSociety(BTS)has madevariousrecommendationsconcerningtheonsetofdiagnosis andtreatmenttimesforpatientswithlungcancer[9].Veryfew studieshavebeenreportedintheliteratureregardingdiagnostic andtreatmentdelaysinlungcancer.

Inthepresentstudy,weinvestigatedpatientanddoctordelays atdiagnosisandtreatmentstagesinpatientswithNSCLCandthe factorsaffectingsuchdelays.

2. Materialsandmethods

Seventeensitesparticipatedinthepresentstudyfromvarious provincesofTurkey(fivedifferentsitesinAnkara,twodifferent sitesinIstanbul,threedifferentsitesinIzmir,andonesiteeachin Diyarbakir, Denizli, Tekirdag˘, Antalya, Manisa, Mersin, and Isparta). Thisstudy was conducted prospectively withpatients capableofgivingtheiranamnesis,whoconsultedbetweenMay 2010andMay2011,andwhohadanewdiagnosisofNSCLCatone of the seventeen institutions. A questionnaire was completed

through personal interviews with the patients. The patients’ clinical files were also reviewed. Cases without a diagnosis of primary lung cancer, cases with a history other than NSCLC, patientswhowerenotwillingtocompletethequestionnaire,and cases fromwhominsufficientsurvey informationwasobtained wereinitiallyexcludedfromthestudy.Atotalof 1016patients from 17 differentsites who werediagnosed withNSCLC were includedinthepresentstudy.

Theage,sex,occupation,education,smokingstatus, socioeco-nomicstatus,socialsecuritystatus,andplaceofresidenceofeach patientwererecorded.Theirfirstsymptoms,thehealthinstitution first visited,thespecialisation ofthephysicianfirst visited,the number of non-pulmonary diseasespecialist physicians visited, themethodof finaldiagnosis,historical diagnoses, thestage of theirdisease,thedateatwhichthefirstsymptomsappeared,the timepassedbetweenthefirstappearanceofasymptomandthe firstpresentationtoanon-pulmonarydiseasespecialistphysician, andthereasonsforanydelaywerealsorecorded.Thetimepassed fromthefirstvisittoanon-pulmonarydiseasespecialistphysician untilwritingareferraltoapulmonarydiseasespecialist,thetime passed from seeing a pulmonary disease specialist until the diagnosis,andthetimepassedfrombeingdiagnosedwithlung cancertosubsequenttreatmentwererecorded.Relevantperiods and delays werecalculated based on thesedates. The possible reasonsfordiagnosticandtreatmentdelayswereassessedincases wheredelayswerediscovered.

The time between the onset of the first complaint and presentation to a non-pulmonary disease specialist physician was defined as the Patient Presentation Time; if this period exceeded30days,itwasacceptedasbeingaPatientDelay[10– 12].DoctorDelaywasdefinedasthetimepassedfromthefirst visitofapatientuntiltreatment.Doctordelaywasexaminedinthe following threesubdomains. Thetime passedbetween thefirst appointment of the patient with a non-pulmonary disease specialist physician until seeing a pulmonary disease specialist was defined as the Patient Referral Time, the time passed betweenseeingapulmonarydiseasespecialistandthe pathologi-cal diagnosiswasdefinedastheDiagnosis Time,and thetime passed from the pathological diagnosis until treatment was definedastheTreatmentTime[11,13,14].

BasedontheperiodsdefinedbytheBTSandSimunovicetal.

[15],apatientreferraltimeexceeding2weekswasacceptedasa criterionforReferralDelay,adiagnosistimeexceeding2weeks wasacceptedasacriterionforDiagnosticDelay,atreatmenttime exceeding 2 weeks wasaccepted as a criterion for Treatment Delay,andthetimebetweenthefirstpresentationtoaphysician andtreatmentthatexceeded6weekswasacceptedasacriterion forDoctorDelay.Thetime passedfromthefirstcomplaintofa patientuntiltreatmentwasdefinedasTotalDelay. Considering thetimeswedefinedforpatientdelayanddoctordelay,atotal periodin excessof72days (6weeks+30 days)wastakenasa criterionforTotalDelay[10].

InTurkey,allhealthcareandrelatedsocialwelfareactivitiesare coordinatedbytheMinistryofHealth.TheMinistryisresponsible to provide health care for thepeople and organise preventive healthservices,buildandoperatestatehospitals,privatehospitals, train medical personnel, regulate the price of medical drugs nationwide,controldrugproductionandallpharmacies. Discussion: Theextendedlengthofpatientdelayunderscoresthenecessityofeducatingpeopleabout lungcancer.Todecreasedoctordelay,educationisacrucialfirststep.Additionally,tofurtherreducethe diagnosticandtreatmentdelaysofchestspecialists,multidisciplinarymanagementandalgorithmsmust beusedregularly.

Health services are now financed through a social security scheme covering the majority of the population, the General HealthInsuranceScheme.TheMinistryofHealthisthemainactor and provides primary, secondary and tertiary care through its facilitiesacrossthecountry.Universitiesarealsomajorproviders oftertiarycare. Theprivate sectorhasincreaseditsrange over recentyears.

Afamilypractitionersystemwasextendedtocoverthewhole country.FamilypractitionersareGeneralPractitionersandfamily physicianspecialistsprovidingprimarycaretothepopulationon theirlists.Themajordrawbackofthesystemisthelackofareferral systembetween primary, secondary and tertiary care.In other words,patientsarefreetoenterthesystematwhateverpointthey prefer.Mostofthehospitalsanddoctorsareconcentratedinthe citiesandbigtowns,meanwhilethereislittlehealthserviceinthe countrysideandruralareas[16,17].

2.1. Statisticalanalyses

Data analysis was performed using SPSS 15.0 (Statistical Package for the Social Sciences, Chicago, IL, USA). Continuous variableswerepresentedasmeanandstandarddeviation,whereas categoricalvariablesasfrequenciesandpercentages.Multivariate logisticregressionanalysiswasusedtoevaluateriskfactorsfor delay,usingselectionoffactorsassociated(p0.10)withdelayin univariateanalysisorthoseknowntohaveclinicalsignificance. Thegoodnessoffitofthemultiplelogisticregressionmodelswas assessedusingtheHosmer–Lemeshowtest.Oddsratios(ORs)and 95% confidence intervals (CIs) were presented. A two-sided pvalue<0.05wasconsideredsignificantforallanalyses.

3. Results

Intotal,926(91.1%)ofthe1016casesincludedinthepresent studyweremaleand90(8.9%)werefemale.Theirmeanagewas 61.510years.Atotalof880ofthecasesweresmokerswhosmoked anaverageof50.328packs/year(Table1).

Additionally,35.9%ofthecaseswereretiredand98%ofthem receivedsocialsecurity(Table2).Themostfrequentcomplaintsof thepatientswerecough(55.4%)andshortnessofbreath(40.7%). Thephysicianmostfrequentlyvisitedfirstbypatientswithlung cancerwasachestspecialist(41.5%).

Byassessingthenumberof(different)non-pulmonarydisease specialistphysiciansvisitedbythepatientsandtheirtotalnumber ofnon-pulmonarydiseasespecialistphysicianvisits,wefoundthat 261(28%)patients visited twoor morenon-pulmonarydisease

specialist physicians, and thenumber ofvisits rangedbetween 1and15.

Whenthecaseswerereviewedintermsofdiagnosticmethod,it wasobservedthat diagnosesweremademostlyusing broncho-scopicmethods.Non-smallcellcarcinomaofsquamoustypewas foundmostfrequentlyatarateof39.7%amongpatientswithlung cancer.

18.6% ofpatients weretreated bysurgery, 20.4% ofpatients withradiotherapyorradiotherapyand chemotherapy,and7.1% patientswithsupportivecaretreatment.Theothers(53.9%)were treatedwithchemotherapy.

Whenthepatientswithlungcancerwereassessed,themean patient presentation time was found to be 49.996.9 days. Althoughthepatientpresentationtimewas30daysorshorterin 63.9%ofthepatients,thisperiodwaslongerthan30daysin36.1%of thecases.

No statistically significant relationship was found between patientdelayandage,sex,socialsecurity,education,incomelevel, presenceofsymptoms,orplaceofresidence(p>0.05).However, patientdelaywaslonger inworkersthaninotheroccupational categories(oddsratio(OR):1.8,p<0.027).

In366cases(36.1%)wherethepatientpresentationtimewas longerthan30days,themostfrequentreasonforpatientdelaywas ‘‘patients’disregardoftheircomplaints,’’followedbyassociating complaintswithanadditionaldisease,fear,andeconomicreasons (Table3).

Theaveragetimebetweenthefirstpresentationtoadoctorand treatmentwas87.799.6days.Theaveragetimesbetweenthefirst presentationtoadoctorandvisittoapulmonarydiseasespecialist, betweenthevisittoapulmonarydiseasespecialistanddiagnosis,and betweenthediagnosisandtreatmentwere61.6127.2,20.444.5, and24.454.9days,respectively.Doctordelaywasexperiencedby 67.3%ofthepatients;therewasreferraldelayin65.1%ofthecases, diagnostic delay in 37.6% ofthem, andtreatment delay in42.8% (Table4).

When themajor factorsfor thesedelays wereexamined by multivariateanalyses,itwasfoundthatasthenumberofdoctors Table 1

Demographiccharacteristicsofthepatients.

n % Sex Male 926 91.1 Female 90 8.9 Age(years) Meanage 61.510 Youngest–oldest 19–87 Education Noeducation 51 5.5 Literate 82 8.8 Primaryschool 484 52 Secondaryschool 142 15.3 Highschool 116 12.5 University 55 5.9 Smokinghabits Smokers 880 91.7 Formersmokers 18 1.9 Non-smokers 62 6.4 Table 2

Occupation,socialsecuritystatus,andplaceofresidenceofthepatients.

Occupation n % Retired 327 35.9 Governmentemployee 41 4.5 Farmer 172 18.9 Housewife 63 6.9 Worker 79 8.7 Self-employed 216 23.7 Unemployed 14 1.5

Socialsecuritystatus

SGK(SocialSecurityInstitution) 883 88.8

Greencard 91 9.2 Onpayroll 20 2.0 Placeofresidence Province 488 50.6 Borough 321 33.3 Village 155 16.1 Table 3

Possiblereasonsforpatientdelay.

Reasonfordelay(n=366) n %

Disregardofcomplaints 254 69.4

Associatingdiseasewithanadditionaldisease 84 22.9

Fear 41 11.2

Economicreasons 38 10.4

Lackofhealthinsurance 12 3.3

Other(e.g.,socio-culturalorfamilyreasons,

workload,orassociatingtheircomplaintswithsmoking)

visitedincreased,referraldelay,doctordelay,andtotaldelayalso increased(p<0.05).Additionally,diagnosticdelayandtreatment delaywerelonger at theearlystages of thetumour(p<0.05). Referraldelaywaslongerinthosewholivedinvillages,andtotal delaywaslongerincaseswithnoeducation(p<0.05).

Anassessmentofcaseswithadiagnosisdelayshowedthatthe mostfrequentreason forthedelay was‘‘physician’sopinionof anotherdiagnosis’’(Table5).

The major reasons for treatment delays were prolonged examinationsfortumourstaging,patientrefusaloftreatment,a largevolumeofpatients,and extratimerequiredforadditional diseaseassessmentsbeforetreatment(Table6).Theaveragetotal delayfromtheonsetofthefirstcomplaintuntiltreatment was 131.3135.2days.Thetotaldelaywaslongerthan72daysin62.6% ofthepatients.

4. Discussion

Themostimportantissueinthetreatmentofcancertodayisto diagnosepatients at anearly stageand enablethem tohave a chanceforsurgery.Inlungcancer,thetimeittakesforatumourto

doubleitsvolumeisbetween4and56weeks[18,19].Delaysin diagnosisinparticularcancauseachangeinthetumourstagein many patients, resulting in a loss of their chance for surgery. Periodsover30dayshavebeentakenascriteriaforpatientdelayin previousstudies;however,nodefiniteperiodhasbeenspecified fordoctordelayanditssubdomains.TheBTShassuggestedthat diagnosticproceduresshouldnot exceed2 weeks,and thatthe timebetweenpresentationtoachestspecialistandthoracotomy shouldnotbelongerthan8weeks[9].TheSwedishLungCancer StudyGroupsuggestedthatdiagnostictestsshouldbecompleted within4weeksfollowingconsultationwithachestspecialist,and that treatment shouldbegin within 2 weeks thereafter[20]. A studyconductedinCanadasuggestedthatthetimebetweenthe first presentation toa doctor and diagnosis shouldnot exceed 4weeks,andthatthetimebetweenthecompletionofdiagnostic testsandsurgeryshouldnotexceed2weeks[15].

Therearegrowinginternationaleffortstodescribeandmeasure patientjourneyspriortoacancerdiagnosis.Accuratedescriptions of these patientjourneysand validmeasurement of diagnostic intervalsareessentialtodeterminetheeffectivenessof interven-tions to reduce them. Weller et al. suggested that the Aarhus checklistwouldfacilitatethestandardisedanduniformdefinition andreportingofstudiesinthisarea[21].

Few studies in the literature have reported on patient and doctordelaysinlungcancer. Importantly,ourstudyreflectsthe dataforTurkeyasawholebecauseitinvolved17differentsites fromvarious Turkishprovinces andwe exploredthelengths of diagnosticandtreatmentdelaysandthepossiblereasonsforthem inanNSCLCpatientgroup.Themeanpatientdelaywasfoundtobe 49.996.9daysinourstudy,and36.1%ofourcaseshadpatient delays.

In lung cancer, studies have reported diverse results with respecttopatientanddoctordelays.Webelievethatthereasonfor such diverse resultsincludes the assessment of patient groups from different countries, different socioeconomic levels, and different health policies. Milleron et al. [22] reported in their seriesof72casesthattheaveragetimewas103daysforpatient delay,88daysfordoctordelay,and155daysfortotaldelay.Intheir prospectivestudyinvolving134caseswithlungcancerandchest tumours,Koyietal.[13]reportedtheaveragepatientdelaytobe 43 days,andthetotaldelaywasreportedtobe203days.They further reported that there were cases where the total delay exceeded 2 years. Salomaa et al. [11] found in their study of 132casesthatthepatientdelaywas14days,firstdoctordelaywas 16days,seconddoctordelaywas15days,andtreatmentdelaywas 15days.

Keebleetal.indicatedthatpatientintervalforlungcancerin 2009–2010inEnglandwas12days[23]andBaughanetal.founda mediandelayof9daysinlate2000sinScottishpatientswithlung cancer[24].

InaretrospectivestudybyO¨ zlu¨ etal.inTurkey[25],thepatient delay was foundto be64 days, doctor delay wasfound tobe 48days,andtotaldelaywasfoundtobe102days.Patientdelay wasshorter inourstudy,whereasdoctordelay andtotaldelay werelonger.

Suluetal.reportedthatthemeantimeswas59.9daysforthe applicationdays,40.3daysforthereferralinterval,16.4daysfor thediagnosticinterval,and24.7daysforthetreatmentinterval

[26].Amedianvalueof14daysfordoctordelayinpatientswith lungcancerwasfoundbyLyratzopoulosetal.[27]andBaughan etal.reporteddoctordelaywas12days(median)[24].

A previous study reported that the most important part of doctordelayswasthereferraldelay[11].Ourresultsshowthat only34.9%ofourpatientshadnoreferraldelay.62.4%ofthemwere diagnosedwithinaperiodof2weeks,and57.2%ofthemhadtheir treatmentstartedwithinaperiodof2weeks.Salomaaetal.[11]

Table 4

Doctordelayandthedelayineachsubgroup.

n % Doctordelay 6weeks 252 32.7 >6weeks 519 67.3 Referraldelay 2weeks 328 34.9 >2weeks 612 65.1 Diagnosticdelay 2weeks 576 62.4 >2weeks 347 37.6 Treatmentdelay 2weeks 429 57.2 >2weeks 321 42.8 Table 5

Reasonsforthedelayinpatientdiagnosis.

Reasonfordelay(n=347) n %

Physician’sopinionofanotherdiagnosis 135 38.9 Delaysinpathologicexamination 126 36.3

Additionaldiseaseassessment 100 28.8

Delaysinradiologicexamination 87 25.1

Patientrefusaltoundergoprocedure 44 12.7 Reasonsrelatedtothehealthsystem 43 12.4 Delaysinbronchoscopicexamination 20 5.8 Delaysinconsultativeexamination 45 12.9 Longwaitingtimeforhospitalisation 18 5.2 Alargepatientvolume/lackofbeds 43 12.4

Referraltoanotherunit 36 10.4

Other 108 31.1

Table 6

Reasonsfordelayinpatienttreatment.

Reasonfordelay(n=321) n %

Longwaitingtimefortestsrequiredforstaging 40 12.5

Patientrefusalofthetreatment 38 11.9

Alargevolumeofpatients 36 11.2

Additionaldiseaseassessmentbeforetreatment 34 10.6 Reasonsrelatedtothehealthsystem 25 7.8 Delaysinconsultativeexamination 12 3.7

Referraltoanotherunit 20 6.2

Longtimeforradiotherapyappointment 12 3.7

Other 12 3.7

reportedthatonly38%ofpatientswereabletoreceivetreatment within1monthfollowingtheirpresentationatthecentre.Yılmaz etal.[10] reported,conversely,that63.6%ofthepatientshad a diagnosis within 2 weeks, 30.4% of them had their treatment startedwithin 2 weeks,and only 26.1% ofthe patients had no doctor delay. Thepercentage of patients in ourstudy with no doctordelaywas32.7%.

The major factors that can affect patient delay are age, sex, socioeconomicconditions,andeducationlevel[28].Inourstudy,no relationshipwasfoundbetweenthelengthofdelayandpatients’ sex,socioeconomic conditions,socialsecurity,education level,or placeofresidence.However,patientdelaywasfoundtobelongerin workers.A study conducted by Gonzales et al. [29] concerning patientdelaydemonstratedthatnoneoftheabovefactorsaffected patientpresentationtimeother thanliving in ruralareas versus urbanareas.Milleronetal.[10]reportedthatdelayswerelongerin cases where the first presentation was made to a general practitionerthaninthosewherethevisitwastoachestspecialist. The mean age our lung cancer patients (61.510.1) was younger,inmostcountriestheaverageageofalungcancerpatients was70–73.Itmaybeinterestedintheoverallaverageageatwhich smokersbegansmokingcigarettesregularlywasbetweentheagesof 15and17amongTurkishpeople.ItindicatesthatsmokersinTurkey aresmokingregularlyatanearlierage.Inaddition,therewasno relationshipwasfoundbetweenthelengthofdelayandpatients’age. Andalso,Raineetal.investigatedsocialvariationsinaccessto hospitalcareforpatientswithcolorectal,breast,andlungcancer andfoundthatsocialfactorsstronglyinfluenceaccesstoandthe provisionofcare[30].

AnotherstudyperformedinBrazilreportedthattwosignificant factors that might affect patient delay were people’s level of knowledgeaboutcomplaints and therisksof smoking and the difficultiestheyfacedinreachingahealthcarecentreoraphysician

[12].

Smithetal.foundthatpatientdelaywas99days(median)and theysuggestedthatlongtermsmokers,thosewithCOPDand/or thoselivingaloneareatparticularriskoftakinglongertoconsult withsymptomsoflungcancerandpractitionersshouldbealert this[31].Breakdownsinvolvedinthesocialsecuritysystemand deficienciesin thehealthcare systemmay alsoadverselyaffect patientdelay.Otherpossiblereasonsforpatientdelaymayinclude patients’ ignorance about their complaints or associating them with their concomitant diseases, including smoking patients linkingtheir complaintssuchas cough or sputum tosmoking. Wealsofoundthatthemostimportantreasonsforpatientdelays werepatients’ignoranceabouttheircomplaintsandassociating themwithadditionaldiseases.Crawfortetal.foundthatpatients withtheshortestdelayhadmoreadvanceddiseaseandsurvival wasleastlikelyforthesepatients.Forthisreason,theysaidthat delayisaconfoundingfactor[32].

Significant event audit (SEA) is a quality improvement techniquethatiswidelyusedinUKprimarycarepractice.Mitchell etal.analysisofSEAsfrom92generalpracticesintheNorthof EnglandCancerNetwork.TheyfoundthatmostSEAs demonstrat-edtimelyrecognitionandreferral[33].

Koyietal.[13]suggestedthatpatientdelayscouldbereduced byeducatingpeople andmakingit easierforpatientstoaccess healthcareinstitutions.Patientdelayinseekingacancerdiagnosis is an important problem, and in addition, it is a behavioural problemamenabletopsychologicalanalysis[34].Intheirstudies, Silvaetal.[12]andPereiraetal.[35]heldinadequacyofmedical services,delaysinreferrals,andlowperformanceofdiagnosticand supplementarytestsresponsiblefordoctordelays.Itwasstated that an important reason for doctor delays was insufficient knowledgeoflungcancerbythephysicianswhowereinvolved inmonitoringofpatientswithlungcancer,particularlyprimary

carephysicians.Itisacommon problemnottoadministerlung radiography,particularlyincaseswherechroniclungcomplaints are involved [28]. Another problem is that 25–90% of lung radiographsareassessedincorrectly,and28%ofthelesionsthat areassessedincorrectlyhaveadiameterof1–3cm[36].Another significant reason for delays is that doctors focus on other diagnoses,notevenconsideringadiagnosisoflungcancer.In astudyinvolvingcaseswithlungcancer,itwasreportedthat other diagnoseswere considered by doctors 40%of the time

[37].Themostfrequentreasonforadiagnosisdelaywasalso ‘‘doctor’s consideration of another diagnosis’’ in our study. Another important reason for doctor delays is performing unnecessarydiagnosticproceduresorimproperperformanceof diagnosticmethods.Patients’refusalofdiagnosticprocedures mayalsocausediagnosticdelays[38,39].Otherreasonsleading to doctor delay include problems in the health system and insufficienciesofmedicalservicesandlaboratories.Considering these reasons, major approaches to reducing doctor delays should be educating doctors, correcting the deficiencies in clinicalsignsandsymptomsthatmayindicatethepresenceof lungcancerandinlaboratorysystems,andavoiding unneces-sarymedicalexaminations[40].

Nealetal.foundthattheoverallmeandiagnosticintervalfold by 5.4 days by implementation of the 2005 NICE Guidelines between2001–2002and2007–2008[41].Inourstudy,themajor reasonfortreatmentdelayswasprolongedexaminationsforthe stagingoftumours.Anotherremarkableresultfoundinourstudy wasthat65.5%ofthepatientshadconsultedtwoormoredifferent non-pulmonarydiseasespecialistphysicians.Gonzalesetal.[42]

reported that 80% of patients consulted at least two different doctors,andthatthenumberofvisitstoadoctorwasmorethan 1 in 83% of cases. Cancer patients had twice as many GP consultations,10to11timesmorediagnosticinvestigationsand fivetimesmorehospitalcontactsthanthereferencepopulation

[43].Asthenumberofdoctorsconsultedincreases,referral,doctor, andtotaldelaysalsoincreased.Lyratzopoulosetal.indicatedthat lungcancerpatientshaveahighproportion(30%)of3ormore pre-referralconsultations[44].

Toeliminatethisproblem,deficienciesinthesystemshouldbe corrected,andadditionaltimeshouldbespentoneducationand high-risk patients should be directed without delay to health centresutilisingmultidisciplinaryapproachesonthesubject.

Inthepresentstudy,wefoundthatdiagnosticandtreatment delayswerelongerinearly-stagecases.Webelievethecausemay bethatmostpatientswithlungcancerhadnocomplaintsatthe earlystagesortheyignoredtheircomplaintsandassociatedthem withsmoking.

Wefoundthattotaldelayswereexperiencedby62.6%ofour patients. Yılmaz et al. [11] reported a longer average period, 176.2days,forthetotaldelay,butasimilarmediantime,98days, andmentionedthat71.8%oftheirpatientshadtotaldelays.The mediantimewasmeasuredas71.5daysinanotherstudy[45].

Hansenetal.reportedthatmediantotaldelaywas108daysin lung cancer and suggested that system delay accounted for a substantialpartofthetotaldelayexperiencedbycancerpatients

[46].

Walteretal.systematicallyreviewedtheliteraturereporting theapplicationofAndersen’sModeloftotalpatientdelay(delay stages:appraisal, illness, behavioural,scheduling, treatment)in studieswhichassesscancerdiagnosis.Theyfoundthattherewas strong evidence to support the existence and importance of appraisalandtreatmentdelayasdefinedintheAndersenModel, althoughtreatmentdelayrequiresexpansion[47].

In conclusion, very long patientdelays indicate theneed to informpatientsaboutlungcancerandtheriskfactorsinvolved, and the need to warn them not to omit periodic check-ups.

To reduce doctor delays, education shouldbe increased at the primarycarelevel,andat-riskpatientsshouldbedirectedwithout delaytohealthcentresutilisingmultidisciplinaryapproacheson the subject. It is important that health centres dealing with diagnosisandtreatmentusemultidisciplinaryapplicationsbased oncurrentalgorithmsandacommonlanguagetofurthershorten thetimerequiredtomakeadiagnosisandinitiatetreatment. Conflictofintereststatement

Wehavenoprovidedanyfinancialandpersonalrelationships with other people or organisations that could inappropriately influence(bias)ourwork.

Authorshipcontribution

Thismanuscript(Patientanddoctordelaysinthediagnosisand treatmentofnon-smallcelllungcancerinTurkey)hasbeenseen andapprovedbyallco-authorsandthismanuscripthasnotbeen publishedorsubmittedforpublicationelsewhereandwehaveno provided any financial and personal relationships with other peopleororganisationsthatcouldinappropriatelyinfluence(bias) ourwork.

References

[1]CohenV,KhuriFR.Progressinlungcancerchemoprevention.CancerControl 2003;10:315–24.

[2]TurkishThoracicSociety,LungandPleuralMalignanciesStudyGroup.The projectofmapoflungcancerinTurkey,lungcancerincidenceofTurkey. Istanbul,Turkey:TTSAnnualCongress,2010.

[3]http://www.kanser.gov.tr/index.php, http://www.saglik.gov.tr/TR/belge/ 1-15486.

[4]JemalA,ThomasA,MurrayT,ThunM.Cancerstatistics,2002.CACancerJClin 2002;52:23–47.

[5]ScagliottiG. Symptoms,signsandstagingoflungcancer. Eur RespirMon 2001;17:86–119.

[6]SchillerJH,HarringtonD,BelaniCP,LangerC,SandlerA,KrookJ,etal. Comparisonoffourchemotherapyregimensforadvancednon-small-cell lungcancer.NEnglJMed2002;346:92–8.

[7]VerdecchiaA,FrancisciS,BrennerH,GattaG,MicheliA,MangoneL,etal. RecentcancersurvivalinEurope:a2000–2002periodanalysisofEUROCARE-4 data.LancetOncol2007;8:784–96.

[8]ColemanMP,FormanD,BryantH,ButlerJ,RachetB,MaringeC,etal.Cancer survivalinAustralia,Canada,Denmark,Norway,Sweden,andtheUK,1995– 2007(theInternationalCancerBenchmarkingPartnership):ananalysisof population-basedcancerregistrydata.Lancet2011;377:127–38.

[9]BTS.BTSrecommendationstorespiratoryphysiciansfororganisingthecareof patientswithlungcancer.Thorax1998;53:1–8.

[10]YılmazA,DamadogluE,SalturkC,OkurE,TuncerLY,HalezerogluS.Delays inthediagnosisandtreatmentofprimarylungcancer:arelongerdelays associatedwithadvancedpathologicstage?UpsalaJMedSci2008;113:287– 96.

[11]SalomaaER,SallinenS,HiekkanenH,LiippoK.Delaysinthediagnosisand treatmentoflungcancer.Chest2005;128:2282–8.

[12]SilvaPPA,PereiraJR, IkariFK,MinamotoH.Cancerdepulmaoeretardo diagnostico:analisede300cases.RevAssMedBrasil1992;38:145–9.

[13]KoyiH,HillerdalG,BrandenE.Patient’sanddoctor’sdelaysinthediagnosisof chesttumors.LungCancer2002;35:53–7.

[14]ChristensenED,HarvaldT,JendresenM,AggestrupS,PettersonG.Theimpact ofdelayeddiagnosisoflungcanceronthestageatthetimeofoperation.EurJ CardiothoracSurg1997;12:880–4.

[15]SimunovicM,GagliardiA,McCreadyD,CoatesA,LevineM,DePetrilloD.A snapshotofwaitingtimesforcancersurgeryprovidedbysurgeonsaffiliated withregionalcancercentersinOntario.CanMedAssocJ2001;165:421–5.

[16]http://www.tuik.gow.tr. [17]http://www.euro.who.int.

[18]WeissW.Peripheralmeasurablebronchogeniccarcinoma.AmRevRespirDis 1971;103:198–208.

[19]SteeleJD,BuellP.Asymptomaticsolitarypulmonarynodules-hostsurvival, tumoursiteandgrowthrate.JThoracCardiovascSurg1971;65:140–51.

[20]MyrdalG,LambeM,HillerdalG,LambergK,AgustssonTH,StahleE.Effectof delaysonprognosis inpatients withnon-small celllung cancer.Thorax 2004;59:45–9.

[21]WellerD,VedstedP,RubinG,WalterFM,EmeryJ,ScottS,etal.TheAarhus statement:improvingdesignandreportingofstudiesonearlycancer diagno-sis.BrJCancer2012;106:1262–7.

[22]MilleronB,MangiapanG,TerriouxPH,RosencherL,GuigayJ,MayaudC.Delays inthediagnosisandtreatmentoflungcancer.Thorax1997;52:398.

[23]KeebleS,AbelGA,SaundersLC,McPhailS,WalterFM,NealRD,etal.Variation inpromptnessofpresentationamong10,297patientssubsequentlydiagnosed withoneof18cancers:evidencefromaNationalAuditofCancerDiagnosisin PrimaryCare.IntJCancer2014;135:1220–8.

[24]BaughanP,O’NeillB,FletcherE.Auditingthediagnosisofcancerinprimary care:theexperienceinScotland.BrJCancer2009;101:S87–91.

[25]O¨ zlu¨ T,Bu¨lbu¨lY,O¨ ztunaF,C¸anG.Howlongthediagnosticandtherapeutic processtakesinlungcancers.TubercThorax2002;50:288–91.

[26]SuluE,TasolarO,BerkTakirH,YagciTuncerL,KarakurtZ,YilmazA.Delays in the diagnosis and treatment of non-small-cell lung cancer. Tumori 2011;97:693–7.

[27]LyratzopoulosG,AbelAG,McPhailS,NealRD,RubinGP.Measuresof prompt-nessof cancerdiagnosisinprimarycare:secondaryanalysisofnational auditdata onpatients with 18common andrarer cancers.Br J Cancer 2013;108:686–90.

[28]YılmazA,Bog˘aS,SuluE,DurucuM,YılmazD,BaranA,etal.Delaysinthe diagnosisandtreatmentofhospitalizedpatientswithsmear-positive pulmo-narytuberculosis.RespirMed2001;95:802–5.

[29]GonzalezJM,deCastroFJ,BarruecoM,CordovillaR,FermandezJL,GomezFP, etal.Delaysinthediagnosisoflungcancer.ArchBronconeumol2003;39: 437–41.

[30]RaineR,WongW,ScholesS,AshtonC,ObichereA,AmblerG.Socialvariations inaccesstohospitalcareforpatientswithcolorectal,breast,andlungcancer between1999and2006:retrospectiveanalysisofhospitalepisodestatistics. BMJ2010;340:b5479.

[31]SmithS, CampbellNC,MacleodU, LeeAJ,RajaA,WykeS, etal.Factors contributing to timeto consult withsymptoms of lung cancer: a cross sectionalstudy.Thorax2009;64:523–31.

[32]CrawfordSC,DavisJA,SiddiquiNA,deCaesteckerL,GillisCR,HoleD,etal.The waitingtimeparadox:populationbased retrospectivestudyoftreatment delayand survival ofwomen with endometrialcancer inScotland.BMJ 2002;27(325):196.

[33]MitchellED,RubinG,MacleodU.Understandingdiagnosisoflungcancerin primarycare:qualitativesynthesisofsignificanteventauditreports.BrJGen Pract2013;63:37–46.

[34]AndersenBL,CacioppoJT,RobertsDC.Delayinseekingacancerdiagnosis: delaystagesandpsychophysiologicalcomparisonprocesses.BrJSocPsychol 1995;34:33–52.

[35]PereiraJR,IkariFK,MinamotoH,CassioliJC.Delayfactorsinthediagnosisof lungcancer:apublichealthproblem.RevPaulMed1991;109:109–12.

[36]QuekelLG,KesselsAG,GoeiR,vanEngelshovenJM.Missrateoflungcanceron thechestradiographinclinicalpractice.Chest1999;115:720–4.

[37]RachtanJ,PawlickiM.Causesofdelayedtreatmentofpatientswithsmallcell lungcancer.PneumonolPol1990;58:615–9.

[38]AragonesesFG,MorenoN,LeonP,FontanEG,FolqueE.Influenceofdelayson survivalinthesurgicaltreatmentofbronchogeniccarcinoma.LungCancer 2002;36:59–63.

[39]YoshimotoA,TsujiH,TakazakuraE,WatanabeT,HaratakeJ,KasaharaK,etal. Reasonsforthedelaysinthedefinitivediagnosisoflungcancerformorethan one yearfrom therecognition of abnormal chest shadows. InternMed 2002;41:95–102.

[40]YılmazA,AybatlıA.Delaysinthediagnosisandtreatmentoflungcancer.Turk ThoracJ2005;6:68–72.

[41]NealRD,DinNU,HamiltonW,UkoumunneOC,CarterB,StapleyS,etal. Comparisonofcancerdiagnosticintervalsbeforeandafterimplementationof NICEguidelines:analysisofdatafromtheUKGeneralPracticeResearch Database.BrJCancer2014;110:584–92.

[42]GonzalezPorJM,deCastroFJ,BarruecoM,CordavillaR,FermandezJK,Gomez FP, et al. Delays in the diagnosis of lung cancer. Arch Bronconeumol 2003;39:437–41.

[43]ChristensenKG,Fenger-GrønM,FlarupKR,VedstedP.Useofgeneralpractice, diagnosticinvestigationsandhospitalservicesbeforeandaftercancer diag-nosis–apopulation-based nationwideregistrystudyof127,000incident adultcancerpatients.BMCHealthServRes2012;12:224.

[44]LyratzopoulosG,NealRD,BarbiereJM,RubinGP,AbelGA.Variationinnumber ofgeneralpractitionerconsultationsbeforehospitalreferralforcancer: find-ingsfromthe2010NationalCancerPatientExperienceSurveyinEngland. LancetOncol2012;13:353–65.

[45]O¨ zlu¨ T,Bu¨lbu¨lY,O¨ ztunaF,C¸anG.Timecoursefromfirstsymptomtothe treatmentoflungcancerintheEasternBlackSeaRegionofTurkey.MedPrinc Pract2004;13:211–4.

[46]HansenRP,VedstedP,SokolowskiI,SøndergaardJ,OlesenF.Timeintervals fromfirstsymptomtotreatmentofcancer:acohortstudyof2,212newly diagnosedcancerpatients.BMCHealthServRes2011;11:284–92.

[47]WalterF,WebsterA,ScottS,EmeryJ.TheAndersenmodeloftotalpatient delay:asystematicreviewofitsapplicationincancerdiagnosis.JHealthServ ResPolicy2011;1–11.