A cross-sectional investigation of communication in Do-Not-Resuscitate orders in Dutch hospitals

Clinical

paper

A

cross-sectional

investigation

of

communication

in

Do-Not-Resuscitate

orders

in

Dutch

hospitals

M.

Schluep

*

,

S.E.

Hoeks

,

H.

Endeman

,

S.

IJmkers

,

T.M.M.

Romijn

,

J.

Alsma

,

F.H.

Bosch

,

A.D.

Cornet

,

A.H.M.

Knook

,

A.W.M.M.

Koopman-van

Gemert

,

T.

van

Melsen

,

R.

Peters

,

K.S.

Simons

,

E.J.

Wils

,

R.J.

Stolker

,

M.

van

Dijk

a

DepartmentofAnaesthesiology,ErasmusUniversityMedicalCenter,Rotterdam,theNetherlands b_{DepartmentofIntensiveCareMedicine,ErasmusUniversityMedicalCentre,Rotterdam,theNetherlands} c_{DepartmentofAnaesthesiology,IkaziaHospital,Rotterdam,theNetherlands}

d_{DepartmentofAnaesthesiology,AlbertSchweitzerHospital,Dordrecht,theNetherlands} e

DepartmentofInternalMedicine,ErasmusUniversityMedicalCentre,Rotterdam,theNetherlands f

DepartmentofIntensiveCareMedicine,RijnstateHospital,Arnhem,theNetherlands g

DepartmentofIntensiveCareMedicine,MedischSpectrumTwente,Enschede,theNetherlands h_{DepartmentofIntensiveCareMedicine,ReinierdeGraafgasthuis,Delft,theNetherlands} i_{DepartmentofIntensiveCareMedicine,HaaglandenMedicalCentre,TheHague,theNetherlands} j_{DepartmentofCardiology,TergooiHospital,Hilversum,theNetherlands}

k

DepartmentofIntensiveCareMedicine,JeroenBoschHospital,‘ s-Hertogenbosch,theNetherlands l

DepartmentofIntensiveCareMedicine,FranciscusGasthuis&Vlietland,Rotterdam,theNetherlands

Abstract

Background:The decisionto attempt orrefrainfrom resuscitationispreferably basedon prognosticfactorsfor outcomeandsubsequently communicatedwithpatients.Bothpatientsandphysiciansconsidergoodcommunicationimportant,howeverlittleisknownaboutpatientinvolvementin andunderstandingofcardiopulmonaryresuscitation(CPR)directives.

Aim:TodeterminetheprevalenceofDoNotResuscitate(DNR)-orders,todescriberecollectionofCPR-directiveconversationsandfactorsassociated withpatientrecollectionandunderstanding.

Methods:Thiswasatwo-weeknationwidemulticentrecross-sectionalobservationalstudyusingastudy-specificsurvey.Thestudypopulation consistedofpatientsadmittedtonon-monitoredwardsin13hospitals.Datawerecollectedfromtheelectronicmedicalrecord(EMR)concerning CPR-directive,comorbidityandat-homemedication.PatientsreportedtheirperceptionandexpectationsaboutCPR-counsellingthroughaquestionnaire. Results:Atotalof1136patientscompletedthequestionnaire.Patients’CPR-directivesweredocumentedintheEMRasfollows:63.7%fullcode, 27.5%DNRandin8.8%nodirectivewasdocumented.DNRwasmostoftendocumentedforpatients>80years(66.4%)andinpatientsusing>10 medications(45.3%).Overall,55.8%ofpatientsrecalledhavinghadaconversationabouttheirCPR-directiveand48.1%patientsreportedthesame CPR-directiveastheEMR.MostpatientshadagoodexperiencewiththeCPR-directiveconversationingeneral(66.1%),aswellasitstiming(84%)and location(94%)specifically.

Conclusions:TheaverageDNR-prevalenceis27.5%.CorrectunderstandingoftheirCPR-directiveislowestinpatientsaged80yearsand multimorbidpatients.CPR-directivecounsellingshouldfocusmoreonpatientinvolvementandtheircorrectunderstanding.

Keywords:DNR,CPR-directive,Do-Not-Resuscitate,Decision-making,Patientsurvey

* Correspondingauthorat:DepartmentofAnaesthesiology,ErasmusUniversityMedicalCentre,P.O.Box2040,3000CA,Rotterdam,theNetherlands. E-mailaddress:m.schluep@erasmusmc.nl(M. Schluep).

m

Prof.StolkerandProf.vanDijkhavemadeequalcontributionsand,ifpossible,sharedfinalauthorshipispreferred.

https://doi.org/10.1016/j.resuscitation.2020.04.004

Received12November2019;Receivedinrevisedform30March2020;Accepted2April2020 Availableonlinexxx

Available

online

at

www.sciencedirect.com

Resuscitation

Introduction

Cardiopulmonaryresuscitation(CPR)forin-hospitalcardiacarrest hasalowone-yearsurvivalrateof13%(95%CI:11% 15%).1The decision to attempt or refrain from CPR is preferably based on prognostic factors for outcome and established through shared decision-making.2 5_{Althoughpatientsandphysiciansconsidergood}

communicationon thissubjecttobeimportant, thisisnotalways achieved.6 Evidence concerning optimal timing, location of and specificcommunication strategies islacking.7 Expertsstress that decisionsshouldbepatient-centredandthatCPR-directivesshould bepartofdiscussionsregardingfuturecareplanning.8,7

Communicationbetweenpatientsandphysiciansseems subopti-mal whilemost patientswant tobe actively involvedin decision-makingwithregardtoCPR.9 11TwodecadesofBritishnewspaper coverageonthesubjectlargelypertainstomiscommunicationand insufficientinformationgivenbyphysicians,sometimesevenleading tolegalcases.4_{Patientshavelimitedknowledgeaboutcardiacarrest}

andtheytendtooverestimatetheprobabilityofsurvivalafterCPR.12 Moreover,DNR-ordersareoftenmistakenforwithdrawaloftreatment, euthanasiaorthoughtsubjecttoageism.4,10,13,14,8

AninternationalsurveyonCPR-directivepracticesreportedlarge heterogeneityinapproachesduetodifferingculturesandeconomic status.15_{Themajorityofrespondentsindicatednationalguidanceon}

CPR-counsellingiswarranted,butcurrentlyoftenlacking.Although CPR is not specificallymentioned in Dutch legislation, itis stipulatedthat

patients are informed and provide consent for any proposed

treatment.16_{AnationalguidelineondiscussingDNRinfrailelderly}

patientsisavailableforgeneralpractitioners;nosuchguidelineexists forhospitalcare.17ItisproposedthattheDutch“openculture”facilitates CPR-counselling.15Still,themostrecentDutchstudy(2005)reported that90%ofpatientfileslackedaCPR-directive.18_{LiteratureonDNR}

prevalenceandpatientperceptionisscarce.Toachievebetterpatient counselling andto implement therightcommunication interventions,we must identify which patients need information, when they should receiveitandhowmuchisremembered.19Theobjectiveofthisstudy wastoprovideanexaminationofpatients’perceptionsofCPR-directive counselling.Theprimaryaimwastoassesstheprevalenceof DNR-orders.The secondaryaims weretoestablishhowmanypatients recollectedaconversationaboutaCPR-directive,whatCPR-directive the patients then reportedand if this was in agreement with the electronicmedicalrecord.Furthermorepatientswereaskedabouttheir experienceswiththeconversationandexpectationstowardssurvival ratesafterIHCA.Lastlyanassociationbetweentheaforementioned outcomesandpatients_’age,morbidity,familiaritywithCPRandtypeof admissionwasassessed.

Methods

Studydesign

A nationwide multicentre cross-sectional observational study was conductedin13participatinghospitals.Weusedagroupofpeopleto interviewpatientspresentateachlocationatoneday.Inthiscasethe groupofpeopleconsistedofourlocalinvestigatorsandstudentteam, andthelocationswerehospitalsites.Thishasbeenusedinsimilar previousinvestigations.20Participatinghospitalswererecruitedfrom the19hospitalsparticipatinginastudyassessinglong-termoutcomes

of in-hospitalcardiacin theNetherlands.21 _{Thecurrentstudy was}

registeredatclinicaltrials.gov(NCT03807206).Astructured question-nairewascreatedthroughfocusgroupsessionswithanaesthetists, intensivists,internists,anursingscientist,anepidemiologist,aclinical ethicist,andalinguisticconsultant.Thequestionnairewas assessed for legibility,clinimetricvalueandwaspilot-testedtoassessreadability. Patientpopulation

Thestudypopulationconsistedofalladulthospitalizedpatientswho wereatriskforsufferingin-hospitalcardiacarrestandwhowereable toprovideinformedconsentforthestudy.Asmentionedthereisno protocolforCPR-directiveconversations.Inourclinicalexperience patients whoare admitted totheward orwhoare scheduled for surgeryhaveaCPR-directiveenteredintheelectronicmedicalrecord. Noguidelineorprotocolexistsdictatingthisbediscussedwiththe patient.Weexcludedpatientsfromtheintensive/cardiac/strokecare unit,becausemostpatientsarenotabletoprovideconsentoranswer thequestions.Weexcludedpatientsfromdaytreatmentcentres(e.g. day-caresurgery,outpatientdialysis),becausetheirhospitalstayis very short,and patientswith cognitive impairmentor alanguage barrier without interpreter available. Furthermore we excluded patients fromtheemergency room,because theywerelikely not havespokentoaphysicianpriortooursurveyandparticipationwould betoostrenuous.Toprotectourstudentspatientswithcontagious disease(influenza,norovirus)wereexcluded.Cognitiveimpairment wasgenerallydefinedasaCognitivePerformanceCategory(CPC) score4orCPC3andunabletoprovideconsent.22_Caseswere

reviewedbylocalinvestigators.Ifpatientsornursingstaffrefused participation,thereasonwasnotedanonymously.

Ethicalconsiderations

Studyparticipantsprovidedconsentforparticipationinthestudyand were given the possibility to opt-out. The study protocol was considered not to be subject to the Dutch Medical Research in HumanSubjectsAct(WMO)duetoitsnon-interventionaldesign.This studywasregisteredasMEC2018 1344withtheErasmusUniversity MedicalCentreMedicalEthicsCommittee.

Datacollection

DatawerecollectedbetweenJanuary21st2019andFebruary7th 2019.Eachhospitallocationwasvisitedforonedayfrom09:30amto 6:00pm,leadingto13planneddatacollectiondays.Eachhospitalhad beeninformedabouttheplanneddatacollectiondatebeforehand.On thedayitself,theprincipalinvestigator(MS)andlocalinvestigators informedthewardnursesandtheheadnursewasaskedtoprovidea listofpatientswhomettheexclusioncriteria.Alleligiblepatientswere askedtoparticipateinthestudy.Afterprovidingconsent,thepatient completedastructuredquestionnaireonatabletcomputer,aidedbya student if necessary. These students had medical, nursing or psychology backgrounds and were instructed to obtain consent and helpwith thequestionnaire. Studentswereinstructedhowto clarifyquestionstoavoidmisclassificationbias.

Outcomemeasures

Demographicdatawerecollectedviathequestionnaire,includingthe natureofthehospitalstayandhealth-relatedqualityoflifeusingthe

EuroQoLdescriptivesystemwith5healthdimensionsand3response levels(EQ-5D-3L).23_{SecondlypatientswereaskedifaCPR-directive}

had been discussed with them. They were asked how they

experiencedtimingandlocationofthisconversationandwhatthey thoughttheirCPR-directivewas.Lastly,theywereaskedtoestimate theone-yearsurvivalprobabilityofCPRforin-hospitalcardiacarrest (0 100%).Aresearcher, blindedfromtheinterview, collectedthe following data from the electronic medical record (EMR): CPR-directive, CharlsonComorbidity Index diseases24 and numberof medicationsusedathome(excludingfoodsupplementsandlotions). TheCPR-directivefromtheEMRwasdividedintothreecategories: fullcode(FC),donotattemptcardiopulmonaryresuscitation(DNR) and not documented (ND). Patient responses yielded two more categories:codeunknowntopatient(CU)ornotdiscussedwiththe

patient (NDP). The data were pseudonymized. The translated

questionnaireandcasereportformsareprovidedinSupplement1. Openanswerswithregardtopatientexperienceswerecategorized bytheinvestigators(TR,SIJ,MS)intofourcategories:positive,neutral,

negativeandself-determined.Findingitusefulorappreciatinghaving had aCPR-directiveconversationwas codedas‘positive’; having thoughtaboutaCPR-directivebeforehandandexpressingthisthought wascodedas‘self-determined’.Withregard totimingandlocationofthe conversationpatientsrespondedonatwoorthreepointLikert-scale. WecomparedtheCPR-directivesfromtheEMRwithpatientrecallof havingaCPR-directiveconversationandwhetherpatientswereaware of their CPR-directive (patient understanding). Correct patient understanding wasassessed forpatients whohad adocumented CPRdirective.Correctunderstandingconsistedof:(1)recollectionof havingspokentoahealthcareprofessionalabouttheCPR-directive, and(2)reportingthesamedirectiveasdocumentedintheEMR. Statisticalanalysis

Descriptivestatisticswereusedaccordingly.Subgroupanalyseswere doneforpre-specifiedsubgroupsonthebasisof(1)age(perdecade), (2)Age-CombinedCharlsoncomorbidityIndex(ACCI),24_(3)number

Fig.1 – Study flowdiagram. *Not including:intensive andcritical care units, emergency andoperating rooms,

obstetrics,paediatrics,outpatienthaemodialysis;**Nursesreservedtherighttorefuseaccesstopatientsiftheyfelt

ofmedicationsusedathome(asaproxyofchronicillness),25₍₄₎

familiaritywithCPRand(5)beingaCPR-survivorand(6)admission specialty.FortheCharlsoncomorbidityindex(CCI)acut-offpointof7 pointswaschosenasitisassociatedwithreducedoutcomeinseveral cohorts.24AlsoanACCIwasstratifiedforlow(0 4points),medium(5 7points)orhigh(8+points)burdenofageanddisease.Ahighscore was previously associated with lower survival.26,27 _{Data were}

analyzedusingSPSSstatisticsv25.0(IBM,Chicago,IL,USA)and R.(RFoundationforStatisticalComputing,Vienna,Austria).

Results

Thirteenhospitalswerevisited.Intotal3409patientswerepresentin thenursing wards,subsequently1884patientswerescreened for eligibility,1699patientswereeligibleforinclusionand1136patients completedthequestionnaire.Thisyieldsaresponserateof67.0%. TheflowchartforinclusionissummarizedinFig.1.Includedpatients hadamedianageof70years(IQR59 78),halfofthepopulationwas maleandmostwerebornwiththeDutchnationality(87.0%).Patient characteristicsareshowninTable1.

CPR-directivesandpatientrecollection

TheCPR-directivesfromtheElectronicMedicalRecord(EMR)forthe included1136patientsweredistributedasfollows:63.7%fullcode (FC), 27.5% do not attempt resuscitation (DNR) and 8.8% not documented.ThedistributionofCPR-directivesandpatient recollec-tionisdepictedinFig.2.Ofallquestionedpatients,634/1136(55.8%) recalledaconversationregardingaCPR-directive.Ofpatientswitha full code, 384/724 (53.0%) recalled speaking to a health care professional,ofpatientswithaDNR-orderthiswas228/312(73.1%) (p<0.001).OfpatientswithadocumentedCPR-directiveofeitherFC orDNR499/1036(48.1%)reportedknowingtheirstatusandreported itinaccordancewiththeEMR.ForpatientswithFCthisresultwas330/ 724(45.6%)andforpatientswithDNR169/312(54.2%)(p=0.01).For 81/1136(7.0%)patientsthedirectivetheymentionedwasnottheone registeredintheEMR.

Subgroupanalyses

Results on subgroups were stratified by (1) DNR-prevalence

accordingtotheEMR,(2)CPR-directiveconversationpatientrecall and(3)correctpatientunderstanding.ResultsareshowninTable2. Whilenoneofthepatientsbelow40yearshadaDNR-status,the proportionofpatientswithaDNR-statusincreasedto66.4%inover 80-year-olds(p<0.001).FortheAge-CombinedCharlson Comor-bidityIndex(ACCI)amajorincreasewasseeninDNR-prevalencefor 5points(49.6%)comparedtolowerscores(13.4%)(p<0.001).The DNR-prevalenceincreasedwiththenumberofmedicationsusedat homefrom7.2%(zeromedications)to45.3%(_10medications)(p <0.001). DNR-prevalencewas higherin cancerpatients(37.3%) thaninnon-cancerpatients(24.2%)(p<0.001).

CPR-directiveconversationrecall

In total634/1136 (55.8%) recalleda CPR-directiveconversation. Recallwas28.4%forpatients39years,50.1%for40 64years, 58.9%for 65 79 yearsand 65.9% in patients80 years old (p <0.001).Patientsusinglessversus10medicationshadarecall

percentageof53.2%and68.9%(p_<0.001)respectively.Inverselya lower correct understanding was seen in patients using more medications from73.3% (10medications) vs. 83.8%(9 medi-cations)(p=0.006).

Table1_–Characteristicsofthepatientpopulation(_n

=1136).IQR,interquartilerange;SD,standard

deviation;EQ-5D,EuroQol5dimensionquestionnaire;

CPR,cardiopulmonaryresuscitation.

Characteristicsa

Age(median,IQR) 70(59 78)

Sex,male 567(49.9) Admissionspecialty Medical 449(39.5) Generalsurgery 217(19.1) Cardiology/cardiacsurgery 193(17.0) Neurology/neurosurgery 101(8.9)

Othersurgicalspecialties 176(15.5)

Bornnationality

Dutch 989(87.0)

ofwhichsecondgenerationimmigrant 75(6.5)

Surinam 40(3.5) Moroccan 12(1.1) Turkish 9(0.8) Other 86(7.6) Religion None 560(49.3) Christian 406(35.7) Islamic 44(3.9) Other 126(11.0) Levelofeducation

Primaryschoolornone 170(15.0)

Secondaryschool prevocational 275(24.2)

Secondaryschool higherlevel 75(6.6)

Vocationaleducation 337(29.7)

Univ.ofappliedsciences 215(18.9)

University 64(5.6)

CharlsonComorbidityindex(median,IQR) 1(0 2)

Numbermedicationsusedathome

None 139(12.2)

1 5 476(41.9)

6 9 331(29.1)

>10 190(16.7)

EQ-5Dself-reportedhealthstate(mean,SD)b 62.1(18.8) FamiliarwithCPR?

Yes,witnessedinthestreetorathome 163(14.3)

Yes,witnessedin-hospital 64(5.6)

SeenonTVorinternet 332(29.2)

No 456(40.1)

Didnotrespond 121(10.6)

CPR-survivor 54(4.8)

Estimatedone-yearsurvivalin%;(med,IQR) 55(40 75)

HowwasyourreactiontotheCPR-directiveconversation?c

Positive 219(34.4) Neutral 103(16.3) Negative 159(25.0) Self-determinedd _98(15.4) Noresponseentered 57(8.9) a

Allvaluesaredisplayedas(n,%),unlessotherwisespecified.

b

Rangesfrom0(worstimaginablehealth)to100(bestimaginablehealth).

c_{PatientswereonlyabletoreplyifaCPR-directiveconversationhadtaken}

place(n=636).

d_{Self-determinedmeanspatientshadalreadythoughtabouttheirstatus}

prior to the conversation and felt confident and/or prepared for this conversation.

Patients’experienceswithCPR-directiveconversations Withregardtopatientexperiencespatientswereaskedtoprovidean openanswer.Mostpatientswerepositive(34.4%),neutral(16.3%)or self-determined (15.4%) about the CPR-directive conversation (Table1). Ofthe25.0%ofpatients whohadanegative reaction abouttheconversation, themajoritywasoverwhelmed oraghast,

whereas the rest found themselves unprepared to answer the

questionatthattime.Whenspecifiedforlocationandtimingona Likert-scale,84%waspositiveaboutthetimingand94%waspositive orneutralaboutthelocation.ThisisdisplayedinmoredetailinFig.3

andSupplementalFig.2.Patientsreportedfewernegative experi-encesonaverageiftheCPR-directiveconversationhadtakenplacein theoutpatientclinicorathome(1.7%)comparedtotheERorward (7.4%) (p=0.016). No major differences were observed for the predefinedsubgroups(SupplementalFig.3).Theone-yearsurvival rateafterCPRforin-hospitalcardiacarrestwasestimatedatamedian of 55%(IQR 40 75%. When stratifiedfor patient-reported CPR-statusestimatedsurvivalwaslowestintheDNR(median50.0%,IQR

30.0 62.5%) and CU group (median 50.0%, IQR 30.0 80.0),

followedbyNDP(median57.5%,IQR41.3 80.0%)andFC(median 60.0%,IQR50.0 80.0%)(p<0.001).Nosignificantdifferencesin survivalestimationwerefoundbetweenpatientswhowereorwerenot familiarwithCPRorbetweenagegroups.

Discussion

Ofthehospitalizedpatientsincluded inthisstudy 27.5%had a DNR-order.Of all patientswhoparticipated inthe study55.8%

recalledspeakingtoahealthcareprofessionalabouttheir CPR-directive.TheprevalenceofDNR-statusincreasedwithageand withthenumberofmedicationsusedathome.Theprevalenceof

DNR also increased with a higher Age-Combined Charlson

comorbidity Index (ACCI). The most striking discrepancy we

foundwasthat7.0%ofpatientsrecalledadifferentCPR-directive thantheoneintheEMR.

InourstudyaCPR-directivewasdocumentedin91.2%ofmedical recordsversus9.8%inaDutchsinglecentrestudyfrom2005.18

DNR-prevalence in our study is higher than reported previously. Two studiesfromtheUSAreporteda15%DNR-prevalenceamongtrauma patientsand11.7%prevalenceinanintensivecaresetting.28,29 DNR-ordersweremoreprevalentinpatientsaged>80years,withanACCI >5pointsorusing>10medicationsathome.AhigherACCIhasbeen previouslyassociatedwithpooroutcome.26,27_Inaprevious

meta-analysisonthissubjectagewasassociatedwithahigherprevalence ofDNR,howeverotherimportantfactorsthatmighthaveaffected DNRdecisions,suchaspatients’premorbidstatus,functionalstatus, andprobabilityofsurvivalwerenotuniformlyincludedinallstudies.30 TheauthorsdidsuggestthesefactorscouldinfluenceDNR-decisions. Thepresentstudyconfirmstheinfluenceofageandseverityofillness (byACCIanduseofmedications).

Patientsestimatedone-yearsurvivalafterIHCA2.5timeshigher thantheactualsurvivalratefoundinourretrospectivestudyand meta-analysis.1,26_{Patients withaFChavehigherexpectations ofCPR}

survivalinourstudy,asopposedtopatientwithothercodes.Thisisin linewithfindingsfromaquestionnaireinpatients(admittedtomedical wards)fromtheUSA.31Noassociationwasobservedbetweenthe expectedsurvivalrateafterIHCAandpatient'sownexperience,TVor Internetexposure,norwithage,comorbidityorthenumberofused

Fig.2–CPR-statusprevalence,patients’recallofaconversationandtheCPR-statuspatientsrecollected.*Intwo

medications.Thisimpliesthereisroomforbettereducationonthe prognosisofcardiopulmonaryresuscitation.

Tounderstandthediscrepancieswefoundbetweendocumented CPR-directive and patients recollection, we must consider the possiblesituationsinwhichpatientsareadmittedtohospital.The firstwouldbeelectiveadmissionthroughtheoutpatientclinic(mostly surgical),inwhichtheCPR-directiveisdocumentedintheoutpatient clinicandmaynotbecommunicatedwiththepatients.Reasonsfornot doingthismaybethatthereislittletimedodiscussallaspectsof surgery/treatmentandthegoaloftheadmissionisfullcuration.Most patientswillhaveafullcodedocumented.Moreover,evenifthe CPR-directivewasdiscussed,patientscouldhaveforgottenbythetimethey areadmitted.Thesecondpossiblesituationisunplannedadmission, in which patients may not always receive adequate information becauseoftheemergencysetting;meaningtheyare(considered)to sicktodiscussthisinformationwith,orbecauseoftheirseverityof diseasetheycannotrecollectlateron.Lastlythesituationremainsthat patientshaveapriordocumentedCPR-directiveandthisstatusisnot confirmedoralteredwhenpatientsareadmittedonanextoccasion.In thisstudywecouldnotpinpointtheexactscenarios,asrecollection was similar throughout admission types. We therefore think the discrepancies in recollection are surely in part attributable to admissiontype,howeverevenmoresotopatients’characteristics.

Thecross-sectionalresearchdesignofcollectingdataonedayper site,usingagroupofstudents,isausefulmethodforassessingpoint prevalenceandgatheringinformationinashortperiodoftime.The response rate the study was 67.0%, which is relatively high.32

Furthermorethereasonsfor non-inclusionwereclearlydescribed (Fig.1).OurstudycanbeconsideredrepresentativeofDutchsociety withregardtoethnicity,educationallevelandreligiousbackground.33 35_{WithregardtorepresentationoftheDutchhealthcaresystemour}

samplecontained1(outof8)academichospital,8(outof37)large regionalhospitalsand4(outof57)smallorruralhospitals.Although thisstudypertainsto theDutch medicalsystem,we considerour resultstobeapplicabletoabroadrangeofWesterncountries. Limitations

Certainlimitationsshouldbetakenintoaccount.Firstlywehaveonly assessed dataatonedayperhospital,andnumberscanchange throughouttheyear.Ourstudydesignhashoweverenabledusto collectalargeamountofdatain ashortperiodoftime.Secondly inclusionwaslimitedbypatientswhowerenotabletoparticipate,i.e. cognitivelyimpaired orseverelyill, whereasthese patientsareof special interest for our researchobjectives. Weencountered this limitationbecausethesepatientscouldnotprovideinformedconsent.

Table2_–SubgroupanalysisofDNR-prevalence,codestatus/CPRconversationrecallandcorrectpatient

understanding.Pre-specifiedsubgroupswereused.DNR,DoNotResuscitate;EMR,ElectronicMedicalRecord;

Othersurgicalspecialties,e.g.orthopaedics,plasticsurgery,otorhinolaryngology.

Subgroupn/groupn(%) n DNR-prevalenceinEMR CPRconversationpatientrecall Correctpatientunderstandinga

Allpatients 1136 312/1136 (27.5) 634/1136 (55.8) 499/612 (81.5) Age Youngadults(18 39) 67 0/67 (0) 19/67 (28.4) 16/18 (88.9) Olderadults(40 64) 349 51/349 (14.6) 175/349 (50.1) 141/170 (82.9) Seniors(65 79) 494 111/494 (22.5) 291/494 (58.9) 232/282 (82.3) Elderly(80) 226 150/226 (66.4) 149/226 (65.9) 110/142 (77.5)

CharlsonComorbidityIndex(CCI)

0 6points 1089 289/1089 (26.5) 605/1089 (55.6) 479/583 (82.2)

7points 47 23/47 (48.9) 29/47 (61.7) 20/29 (69.0)

Age-CombinedCharlsonIndex(ACCI)

0 4points 695 93/695 (13.4) 342/695 (49.2) 275/325 (84.6)

5 7points 313 149/313 (47.6) 208/313 (66.5) 160/204 (78.4)

8+points 128 70/128 (54.7) 84/128 (65.6) 64/83 (77.1)

Numberofmedicationsusedathome

0 139 10/139 (7.2) 61/139 (43.9) 49/57 (86.0) 1 5 476 102/476 (21.4) 246/476 (51.7) 191/229 (83.4) 6 9 331 114/331 (34.4) 196/331 (59.2) 163/195 (83.6) 10 190 86/190 (45.3) 131/190 (68.9) 96/131 (73.3) FamiliaritywithCPRb No 456 293/456 (29.8) 254/456 (55.7) 202/246 (82.1)

Yes,seeninreallife 227 41/227 (18.1) 125/227 (55.1) 96/118 (81.4)

Yes,seenonTVorinternet 332 95/332 (28.6) 189/332 (56.9) 145/185 (78.4)

CPR-survivor No 1082 293/1082 (27.1) 603/1082 (55.7) 475/581 (81.8) Yes 54 19/54 (35.2) 31/54 (57.4) 23/30 (76.7) Admissionspecialty Internalmedicine 449 171/449 (38.1) 269/449 (39.6) 208/262 (79.4) Generalsurgery 217 37/217 (17.1) 105/217 (51.2) 79/100 (79.0) Cardiology/cardiacsurgery 193 55/193 (28.5) 123/193 (63.8) 104/121 (86.0) Neurology/neurosurgery 101 27/101 (26.7) 53/101 (47.5) 39/51 (76.5)

Othersurgicalspecialties 176 27/176 (15.3) 84/176 (47.7) 67/76 (88.2)

a

OnlyappliesifpatientsrecalledaCPR-directiveconversation.CorrectpatientunderstandingmeansthatiftheEMRreads“FullCode”,thepatientsprovidedthe sameanswer,idemforotherdirectives.Patientswithrecollection,butnodocumenteddirectivewereexcluded(n=22).

Weconsiderthiseffecttobenegligiblebecausethesecaseswere specificallyreviewedineachhospitalandthereforethenumberof exclusionsonthesegroundsislow.Theseexclusionsmightleadtoa slightunderestimationoftheDNR-prevalence,asperhapsthesickest patientswerenotincluded.Duetoprivacylegislationwedonothave specificdataregardingageormorbidityofthenon-includedpatients. Moreover misclassification bias couldexist as we did not use a validatedquestionnaire.Wehopethiseffecthasbeenminimizedby expert-reviewandpilot-testing ofthequestionnaireandbyhaving trainedstudentspresentattheinterview.Thethirdpossiblelimitation isbiasbyrefusalof5/18hospitalorganizationstoparticipateinthis study.Weexpectbiastobeminimalduetoalargesamplesizeandthe distributioninhospitaltypes,sizesandpatientcharacteristics.The distributionofcodeswasdifferentbetweenhospitals.Wedidnothave sufficientdatahowevertoexplainthisfinding,asitwasnotinour primaryaims.Lastly,wedidnotenrolpatientsfromoutpatientclinics, intensivecareandpalliativecareunits.Thismighthaveresultedinan underestimationoftheincidenceofDNRorders.

Themajorityofpatientsstatedtheyrecalledaconversationabout theirCPR-directive.Howeverspecificsubgroupsmightwarrantmore attentionforbetterunderstanding,as7.0%of patientsmentioned anotherdirective than was registered in the EMR.This situation shouldbeavoidedatallcost.Patientsweregenerallynotopposedto discussingCPR-directives andwere more than willingto answer questionsonthesubject.ThelowCPR-directiveconversationrecallin youngpatientsmightbeduetothisgroupbeinggenerallyhealthyand thereforebydefaultCPRwillbeattempted.Growingapplicationof e-healthmight proveuseful, asthisgroupisapttobe informedvia multimediaandifnecessaryalongerconversationmayfollow.36,37

Forpatientswhoarepronetoforgetwhathadbeendecided,repetition ofthisconversationoralongerfirstCPR-directiveconversationcould aidinrecollectionandunderstanding.19

Resuscitationpolicyshouldbetailoredtothepatients_’situation andpatientsshouldbeawareoftheirCPR-directive.Weshouldspeak toourpatientsaboutwhatisimportanttothemandwhatlimitations modernmedicinehas.Initiativessuchastherecommendedsummary planforemergencycareandtreatment(ReSPECT)fromtheUKgives patients and physicians the possibility to talk about advanced directives.Thisway manymisunderstandingscan beavoided.4,38

DNR-orders can becomea part of advanced care planning and

emergencycaretreatmentplans.7Wesupportrecommendationsfor nationalguidelinesandtrainingofCPR-counsellingtohelpphysicians guidetheirpatientsinshareddecision-making.8_{Asmentionedthereis}

noprotocolforCPR-directiveconversations.Inmosthospitalsitis commonpracticetoenterthisintheEMRuponadmissiontotheward. Howoftenthisisjustanadministrativetask,ratherthanaconscious decisionisnotclear.Thecurrentstudygivesrisetothesuggestionthat inyoungandhealthypatientsitismostlyadministrative.Weenvision threepossiblescenariosforCPR-counselling.First:CPRislikelytobe successful,andCPRwillbeattemptedincaseofIHCAifthepatient agreestothis.Second:noclearpredictioncanbemade,inwhichcase thedecisionwillbemadebasedonthebestavailableevidenceandin agreement with thepatient. Third: the potential burdens of CPR outweighthebenefits,thepatientshouldbeinformedoftheseburdens andaDNRorderisdiscussedwiththepatient.Inallthreescenarios thefocusshouldbelayonthebenefitsandshareddecision-making.38 Weconcludefromthisstudythatpatientsshouldbemoreinvolved inCPR-counsellingandphysiciansshouldfocusoncorrectpatient Fig.3–Patients’opinionsregardingtheCPR-directiveconversationwithregardtolocation(n=622)ortiming(n=631)

oftheconversation,specifiedforthespecificlocationormomentthatitwasdiscussed.*Otherlocationswere:

understandingofthedirectivethatwillbedocumented.Inthisprocess physiciansshouldpayattentiontopatientunderstandinginspecific subgroups,suchaselderlyandmultimorbidpatients.Weproposethat theemphasisinfutureresearchlayonfindingoptimaltimingfor CPR-counsellingandpossibleincorporationinearlyadvanceddirective conversations.

Funding

Thisstudywasfundedbydepartmentalfundsandbyanunrestricted research grant from the European Society of Anaesthesiology (Brussels, Belgium) & Air Liquide (Paris, France). The funding sources hadno role in studydesign, data collection, analysis of interpretation,norinwritingofthereportorthedecisiontosubmitfor publication.Allauthorshadfullaccesstoallthedata.

Authors

and

contributors

M. (Marc) Schluep: principal investigator, study design, data collection,dataanalysis,writingofmanuscript.

S.E. (Sanne) Hoeks, PhD: study design, data analysis and interpretation,datacollection,methodologicalcounsellor,writingof manuscript.

H. (Rik) Endeman, MD, PhD: supervising investigator, data interpretation,writing.

S.(Susanne)IJmkers,MD:datacollection,dataanalysis,project logistics.

T.M.M.(Tessa)Romijn,MD:datacollection,dataanalysis,project logistics.

J.(Jelmer)Alsma,MD:methodologicalcounsellor,studydesign. F.H.(Frank)Bosch,MD,PhD&A.D.(Alex)Cornet,MD,PhD&A. H.M.(Marco)Knook,MD,MBA&A.W.M.M.(Ankie)Koopman-van Gemert,MD,PhD&T.(Trudy)vanMelsen,MD&R.(René)Peters, MD,PhD&K.S.(Koen)Simons,MD,PhD&E.J.(Evert-Jan)Wils,MD, PhD-datacollection,writingofmanuscript.

R.J.(RobertJan)Stolker,MD,PhD&M.(Monique)vanDijk,RN, PhD- supervising investigators,studydesign,datainterpretation, writingofmanuscript.

Conflicts

of

interest

Theauthorsdeclarenoconflictsofinterest.

Acknowledgements

Wewouldliketothankallparticipatinghospitalorganizations,local investigatorsandnursingstaff.Wewouldliketothankourstudent team for their work. We thank Isabel van den Boogert for her exceptionalworkinmanagingthisproject.Andwewouldliketothank ErwinKompanjeandKoHagoortfortheiradvice.

Appendix

A.

Supplementary

data

Supplementarymaterialrelatedtothisarticlecanbefound,inthe onlineversion,athttps://doi.org/10.1016/j.resuscitation.2020.04.004.

REFERENCES

1.SchluepM,GravesteijnBY,StolkerRJ,EndemanH,HoeksSE. One-yearsurvivalafterin-hospitalcardiacarrest:asystematicreviewand meta-analysis.Resuscitation2018,doi:http://dx.doi.org/10.1016/J. RESUSCITATION.201809001.

2.BossaertLL,PerkinsGD,AskitopoulouH,etal.European ResuscitationCouncilGuidelinesforResuscitation2015Section11. Theethicsofresuscitationandend-of-lifedecisionsonbehalfofThe ethicsofresuscitationandend-of-lifedecisionssectionCollaborators 1.Resuscitation2015;95:302 11.

3.PitcherD,SmithG,NolanJ,SoarJ.Trainingisneededtodispel confusionaroundDNAR.BMJ2009;338:b2021.

4.BeedM,deBeerT,BrindleyPG.TwodecadesofBritishnewspaper coverageregardingdonotattemptcardiopulmonaryresuscitation decisions:lessonsforclinicians.Resuscitation2015;86:31 7.

5.SudoreRL,LumHD,YouJJ,etal.Definingadvancecareplanningfor adults:aconsensusdefinitionfromamultidisciplinarydelphipanel.J PainSymptomManage201753:821.e1 832.e1.

6.PetterssonM,HöglundAT,HedströmM.PerspectivesontheDNR decisionprocess:asurveyofnursesandphysiciansinhematology andoncology.PLOSONE2018;13:e0206550.

7.PerkinsGD,FritzZ.Timetochangefromdo-not-resuscitateordersto emergencycaretreatmentplans.JAMANetwOpen2019;2:e195170.

8.HallCC,LugtonJ,SpillerJA,CarduffE.CPRdecision-making conversationsintheUK:anintegrativereview.BMJSupportPalliat Care2019;9:1 11.

9.GolinCE,WengerNS,LiuH,etal.Aprospectivestudyof patient-physiciancommunicationaboutresuscitation.JAmGeriatrSoc 2000;48:S52 60.

10.HeylandDK,FrankC,GrollD,etal.Understandingcardiopulmonary resuscitationdecisionmaking.Chest2006;130:419 28.

11.RobinsonC,KolesarS,BoykoM,BerkowitzJ,CalamB,CollinsM. Awarenessofdo-not-resuscitateorders:whatdopatientsknowand want?CanFamPhysician2012;58:e229 33.

12.JonesGK,BrewerKL,GarrisonHG.Publicexpectationsofsurvival followingcardiopulmonaryresuscitation.AcadEmergMed2000;7:48

53.

13.FanS-Y,WangY-W,LinI-M.Allownaturaldeathversus do-not-resuscitate:titles,informationcontents,outcomes,andthe considerationsrelatedtodo-not-resuscitatedecision.BMCPalliat Care2018;17:114.

14.FritzZ,FuldJ.Ethicalissuessurroundingdonotattemptresuscitation orders:decisions,discussionsanddeleteriouseffects.JMedEthics 2010;36:593 7.

15.GibbsAJO,MalyonAC,FritzZBM.Themesandvariations:an exploratoryinternationalinvestigationintoresuscitation decision-making.Resuscitation2016;103:75 81.

16.HendriksAC.Moetikbijopnamevragennaarreanimatiewensen? (ShouldIaskforDNR-codepreferencesuponadmissionto hospital?.).NedTijdschrGeneeskd2017;161:D1831.

17.vanDeldenJJM,VanDerER,GraafED,etal.Multidisciplinaire RichtlijnBesluitvormingoverreanimatie(MultidisciplinaryGuideline onDecisionsconceringCPR).2013(Accessed28January2019,at

https://www.verenso.nl/kwaliteit-en-richtlijnen/richtlijnendatabase/ reanimatie-1).

18.MeilinkM,vandeWeteringK,KlipH.Discussinganddocumenting(do notattempt)resuscitationordersinaDutchHospital:adisappointing reality.Resuscitation2006;71:322 6.

19.BeckerC,LechelerL,HochstrasserS,etal.Associationof communicationinterventionstodiscusscodestatuswithpatient decisionsfordo-not-resuscitateorders.JAMANetwOpen2019;2: e195033.

20.AlsmaJ,VanSaaseJLCM,NanayakkaraPWB,etal.Thepowerof flashmobresearchconductinganationwideobservationalclinical studyoncapillaryrefilltimeinasingleday.,doi:http://dx.doi.org/ 10.1016/j.chest.2016.11.035.

21.SchluepM,StolkerRJ,HoeksSE,etal.Outcomesafterresuscitation atthehospital:theRoutinestudy.NedTijdschrGeneeskd2017;161.

22.PerkinsGD,JacobsIG,NadkarniVM,etal.Cardiacarrestand cardiopulmonaryresuscitationoutcomereports:updateoftheutstein resuscitationregistrytemplatesforout-of-hospitalcardiacarrest:a statementforhealthcareprofessionalsfromataskforceofthe internationalliaisoncommittee.Resuscitation2015;96:328 40.

23.EuroQolGroup.EuroQol—anewfacilityforthemeasurementof health-relatedqualityoflife.HealthPolicy1990;16:199 208.

24.QuanH,LiB,CourisCM,etal.Updatingandvalidatingthecharlson comorbidityindexandscoreforriskadjustmentinhospitaldischarge abstractsusingdatafrom6countries.AmJEpidemiol2011;173:676

82.

25.CharlesworthCJ,SmitE,LeeDSH,AlramadhanF,OddenMC. Polypharmacyamongadultsaged65yearsandolderintheunited states:1988 2010.JGerontolSerABiolSciMedSci2015;70:989.

26.SchluepM,RijkenbergS,StolkerRJ,HoeksS,EndemanH.One-year mortalityofpatientsadmittedtotheintensivecareunitafterin-hospital cardiacarrest:aretrospectivestudy.JCritCare2018;48:345 51.

27.PiscatorE,HedbergP,GoranssonK,DjarvT.Survivalafterin-hospital cardiacarrestishighlyassociatedwiththeAge-combinedCharlson Co-morbidityIndexinacohortstudyfromatwo-siteSwedish Universityhospital.Resuscitation2016;99:79 83.

28.QuillCM,RatcliffeSJ,HarhayMO,HalpernSD.Variationindecisions toforgolife-sustainingtherapiesinUSICUs.Chest2014;146:573 82.

29.SalottoloK,OffnerPJ,OrlandoA,etal.Theepidemiologyof do-not-resuscitateordersinpatientswithtrauma:acommunitylevelone traumacenterobservationalexperience.ScandJTraumaResusc EmergMed2015;23:9.

30.CookI,KirkupAL,LanghamLJ,MalikMA,MarlowG,SammyI.Endof lifecareanddonotresuscitateorders:howmuchdoesageinfluence

decisionmaking?Asystematicreviewandmeta-analysis.Gerontol GeriatrMed20173:2333721417713422.

31.KimJ,ElliottJO,WallS,SaulE,ShethR,CoffmanJ.Associationswith resuscitationchoice:donotresuscitate,fullcodeorundecided.Patient EducCouns2016;99:823 9.

32.SaundersCL,ElliottMN,LyratzopoulosG,AbelGA.Dodifferential responseratestopatientsurveysbetweenorganizationsleadtounfair performancecomparisons?:Evidencefromtheenglishcancerpatient experiencesurvey.MedCare2016;54:45 54.

33.CentraalBureauvoordeStatistiek(CBS).Bevolkingnaar

migratieachtergrond.2016(Accessed31March2019,athttps://www.

cbs.nl/nl-nl/achtergrond/2016/47/bevolking-naar-migratieachtergrond).

34.CentraalBureauvoordeStatistiek(CBS).Meerdandehelft Nederlandersnietreligious.2018(Accessed31March2019,athttps:// www.cbs.nl/nl-nl/nieuws/2018/43/meer-dan-de-helft-nederlanders-niet-religieus).

35.CentraalBureauvoordeStatistiek(CBS).Onderwijs Cijfers Maatschappij:TrendsinNederland2018 CBS.2018(Accessed31 March2019,athttps://longreads.cbs.nl/trends18/maatschappij/ cijfers/onderwijs/).

36.SchenkerY,FernandezA,SudoreR,SchillingerD.Interventionsto improvepatientcomprehensionininformedconsentformedicaland surgicalprocedures:asystematicreview.MedDecisMaking 2011;31:151 73.

37.VandenBulckJJ.Theimpactoftelevisionfictiononpublic expectationsofsurvivalfollowinginhospitalcardiopulmonary resuscitationbymedicalprofessionals.EurJEmergMed 2002;9:325 9.

38.FritzZ,SlowtherA-M,PerkinsGD.Resuscitationpolicyshouldfocus onthepatient,notthedecision.BMJ2017;356:j813.