the experiences of bereaved relatives with palliative ... · the experiences of bereaved relatives...

The experiences of bereaved relatives with palliative sedation and other end-of-life care practices

SophieM.Bruinsma

ISBN978-94-6169-618-2

The experiences of bereaved relatives with palliative sedation and other end-of-life care practicesThesis,ErasmusMC,UniversityMedicalCenterRotterdam

Coverillustration:MarjanNagtegaal(www.marjannagtegaal.nl)Coverdesign:PatrickHesselsLay-outandprint:OptimaGrafischeCommunicatie,Rotterdam,theNetherlands

TheUNBIASEDstudyhasbeenfundedbytheNetherlandsOrganisationforScientificResearchandtheNetherlandsOrganisationforHealthResearchandDevelopment.ThisthesiswasprintedwithfinancialsupportoftheDepartmentofPublicHealth,ErasmusMC,andtheErasmusUniversityRotterdam.

Copyright © 2015 Sophie Marrigje BruinsmaAllrightsreserved.Nopartofthispublicationmaybereproduced,storedinaretrievalsystem,ortransmitted,inanyformorbyanymeans,electronic,mechanical,photocopying,recordingorotherwise,withoutthepriorpermissionoftheauthororthecopyrightowningjournalsforpreviouslypublishedchapters.

The Experiences of Bereaved Relatives with Palliative Sedation and other

End-of-life Care Practices

De ervaringen van naasten met palliatieve sedatie en andere vormen

van zorg rond het levenseinde

PROEFSCHRIFT

terverkrijgingvandegraadvandoctoraandeErasmusUniversiteitRotterdam

opgezagvanderectormagnificusProf.dr.H.A.P.Pols

envolgensbesluitvanhetCollegevoorPromoties.

Deopenbareverdedigingzalplaatsvindenop

14april2015om13.30uur

door

Sophie Marrigje BruinsmageborenteEnschede

Promotiecommissie

Promotor: Prof.dr.A.vanderHeide

Overigeleden: Prof.dr.C.C.D.vanderRijt Prof.dr.I.D.deBeaufort Prof.dr.B.D.Onwuteaka-Philipsen

Copromotor: Dr.J.A.C.Rietjens

contents

1 Generalintroduction 7

2 Theexperiencesofrelativeswiththepracticeofpalliativesedation:Asystematicreview

21

3 Palliativesedation:Afocusgroupstudyontheexperiencesofrelatives 45

4 Makingsenseofcontinuoussedationinendoflifecareforcancerpatients:AninterviewstudywithbereavedrelativesinthreeEuropeancountries

65

5 Nonegativeimpactofpalliativesedationonrelatives’experienceofthedyingphaseandtheirwellbeingafterthepatient’sdeath

85

6 Estimatingthepotentiallife-shorteningeffectofcontinuoussedationuntildeath:Acomparisonbetweentwoapproaches

103

7 Riskfactorsforcomplicatedgriefinolderadults 117

8 Generaldiscussion 137

9 Other 163

SummarySamenvattingDankwoord/AcknowledgementsAbouttheauthorPublicationsPhDportfolio

165169173177179181

Chapter 1General introduction

Chapter 18

1.1 state of the art

end of life care and decision-making

Duringthepastcentury,thecircumstancesinwhichpeoplediehavechangedsubstan-tially.Acutedeathsdue to infectiousdiseaseshavebeengradually replacedbymoreprolonged dying trajectories (1). One third of all deaths in The Netherlands occursuddenly and unexpectedly (2, 3). The increasing importance of chronic diseases asacauseofdeathandtheattentioncurrentlybeingpaidtopatient-centredcareattheend of life have created interest in the role of medicine in the timing and mode ofdeathanddying(1). Inmanyinstances,deathisnotmerelytheresultof thenaturalcourseofa lethaldisease:medicaldecision-makingoftenhasanactiverole (2,4-6).Suchdecision-makingmayconcern theuseofmedical treatment toprolong the lifeofseriouslyillpatients(7).However,thereis increasinglyrecognitionthatextensionoflifemightnotalwaysbethemostappropriategoalofmedicine.Othergoalshavetoguidemedicaldecision-makingattheendoflife,suchasimprovementofqualityoflifeofpatientsandtheirfamiliesbypreventionandreliefofsuffering(8).

Palliative sedation

Sometimes,patientswhoarenearingdeathhavesymptomsthatcannotberelievedwithconventional medical care, such as intractable pain, dyspnoea, and delirium (9, 10).Thissometimesrequiresatreatmentoflastresort:‘palliativesedation’.Palliativeseda-tionisdefinedasthedeliberateloweringofapatient’slevelofconsciousnessinthelaststagesoflife(11).Thetermpalliativesedationmayrefertoseveralsubtypes:temporaryorintermittentsedationandcontinuoussedationuntildeath.Thedegreeofsedationnecessarytorelievesufferingmayvaryfromsuperficialtodeep(11).Continuousseda-tionisalwaysadministeredinthefinalstagesoflifetopatientswhoaredyingandareexperiencingunbearablesuffering.Guidelinesstatethattheaimofpalliativesedationistorelievesuffering;loweringthelevelofconsciousnessisthemeanstothatend.Theaimshouldnotbe to lengthenorcut short thepatient’s life (11).Guidelines furtherrecordthatpalliativesedationcanonlybeusedforpatientswhosedeathwillensueinthereasonablynearfuture,thatis,withinonetotwoweeks(12-14).Studieshaveshownthatpalliativesedationisusedinallsettingswherepatientsdieandforpatientswithallkindsofdiagnoses,butmostofteninhospitalsandforpatientswithcancer(15,16).

how often is continuous sedation until death used?

Findingsfromsurveysofphysicianssuggestthatcontinuoussedationuntildeathhasa rather high frequency of use (15). Within palliative care settings, estimates of theincidenceoftheuseofsedativesrangefrom15tomorethan60%(17-23).However,theseestimatesaredifficulttocompareduetodifferencesinthesettingsstudiedandthe

General introduction 9

definitionsused.In2001,insixEuropeancountrieswithcomparableepidemiologyofterminaldiseases,therewasavariationinprevalenceofcontinuousdeepsedationuntildeathofbetween2.5and8.5%ofdeaths(15).ItalyandBelgiumreportedthehighestpercentagesofcontinuousdeepsedation:8.5and8.2%ofalldeaths,respectively,wereprecededbytheuseofcontinuousdeepsedation.AsurveyintheUKconductedin2007founda frequencyof16.5%ofcontinuousdeepsedationuntildeath(24). Inamorerecent study, itwasestimated that inTheNetherlands in2010,12.3%ofallpatientsreceivedcontinuousdeepsedationuntildeath(25).Thisisaclearincreasecomparedtothefiguresof5.6%for2001and8.2%for2005(25,26).InFlanders(Belgium),asimilarincreaseintheuseofcontinuousdeepsedationwasdemonstrated,from8.2%in2001to14.5%in2007(27).

Guidelines

In several countries,nationalor localguidelineshavebeendeveloped for theuseofsedativesinthelastphaseoflife.Proceduralguidelinesarehelpfulforeducatingmedi-calpractitioners,settingstandardsforbestpractice,andpromotingoptimalcare(28).In2009,theEuropeanAssociationforPalliativeCare(EAPC)publishedaframeworkofrecommendationsfortheuseofsedationinpalliativecarecomparablewithearlierpublishedinternationalrecommendations(28).In2005,inTheNetherlands,theRoyalDutch Medical Association launched a nationwide guideline. This guideline was re-visedin2009(11).AsummaryofthemainrecommendationsoftheDutchguidelineispresentedintextbox1.Guidelineshavebeenpublishedinseveralothercountries,forexample, in2005,aclinicalguidelineforcontinuoussedationwaspreparedinJapan(29),and,in2010inFlanders(Belgium),aguidelinewaspresentedbytheFederationforPalliativeCareFlanders(30).

the debate about continuous sedation until death

Thebenefitsanddrawbacksofpalliativesedationarefrequentlydiscussedbycaregiv-ers,andlegalandethicalexperts.Theuseofcontinuoussedationuntildeathisthemostcontroversialinthisrespect.Ontheonehand,itisoftenpraisedasaneasy,innovativeandindispensabletechniquetoalleviatesuffering,whichisoneofthemostimportantgoalsofendoflifecare.Ontheotherhand,itmaybeappliedtooeasily(31).Loweringapatient’sconsciousnessuntildeathisafar-reachinginterventionthathasanimportantimpactonthepatient,therelativesandthecaregivers.Itdeprivespatientsintheirverylast days of the possibility to communicate and to say goodbye, and many patientsconsider being mentally aware very important at the end of life (32). Relatives andcaregivershavealsobeenfoundtoconsiderthisproceduretobedistressing(33-35).Anotherimportantissuethatisoftendebatediswhethertheuseofcontinuousseda-tionuntildeathmayshortenlife.Itisassumedinguidelinesthatcontinuoussedation

Chapter 110

until death has no life- shortening effect when used for patients with an estimatedlifeexpectancyofatmosttwoweeksandwhensedativesareproperlydosed(14,26).Severalempiricalstudieshavesuggestedthatsedationasusedinclinicalpracticehasno significant life-shorteningeffect (17,23,36-39).However,physiciansmayhaveadifferentperspective.ADutch study in2005 revealed thatphysiciansestimated thatcontinuoussedationuntildeathmighthavehadalife-shorteningeffectin26%ofthecases(26).Further,asubstantialnumberofstudiesshowthatphysiciansdeclaredto

textbox 1: Main recommendations Dutch guideline (11)

• Continuoussedationshouldalwaysbeadministeredinthefinalstagesoflifetopatientswhoaredyingandareexperiencingunbearablesuffering.

• Indicationsforsedationarepresentwhenoneormoreintractableor‘refractory’symptomsarecausingthepatientunbearablesuffering.Thephysicianwillhavetodecidewhetherasymptomistreatableornotonthebasisofacceptedgoodmedicalpractice,bearinginmindthespecificcircumstancesofapatientinthelaststagesoflife.

• Thepatient’slifeexpectancyshouldnotexceedonetotwoweeks.• Incasethepatientiscapableofmakingaconsciousdecision,thepatientmust

agreewithsedation;ifthepatientisnolongercompetenttomakeaninformeddecision,thephysicianmustconsultherrepresentative.

• Theadviceofaconsultantismandatoryiftheattendingphysicianpossessesinsufficientexpertiseand/orisindoubtaboutkeyissuessuchasmedicalindicationsandlifeexpectancy.

• Thesedationisaimedatthereliefofthepatient’ssufferingandnotathasteningorpostponingdeath.

• Theattendingphysicianmustbepresentattheinitiationofthesedation.• Midazolamisthedrugofchoice;theuseofmorphineasasedativeasisregarded

badpractice,morphineshouldonlybegivenorcontinued(alongsidesedatives)torelievepainand/ordyspnoea.

• Incasesofcontinuous,deepsedationuntilthemomentofdeath,thereshouldbenoartificialadministrationoffluids.

• Relativesplayanimportantrole,bothwhensedationisbeingconsideredandwhileitisbeingcarriedout.Relativesshouldbeinvolvedinthedecision-makingprocess,theycanassistinmonitoringandcaringforthepatient,andthattheyshouldbeclearlyinformedandsupported.Further,itisimportantnotonlytoprovidethebestpossibleinformationandemotionalsupportforthepatientandherfamily,butalsotocareforthevariousprofessionalsinvolvedinthecase.


have used continuous sedation until death with a (co)intention to hasten death (16,40-43).Further,whilesomearguethatitshouldbeclearlydistinguishedfromeutha-nasia(11,44,45),othersarguethatitmaybecomesimilartoeuthanasiaorevenbeingconsidereditas‘sloweuthanasia’(46).

international differences

Existing comparative international evidence suggest that there are systematic dif-ferences according to the country in which practice occurs. A study carried out inBelgium, the Netherlands and the UK demonstrated that ‘country’ is an importantfactorinpredictingtheprobabilityofreportinguseofcontinuousdeepsedation(47,48).AqualitativestudyconductedintheUSandtheNetherlandssuggestedthatthejustificationforsedationandtheopennesswithwhichitisdiscusseddifferbetweenthecountries(49).Further,findingsfromaqualitativestudyinBelgium,theNetherlandsand the UK among clinicians and academic researchers working in the field of pal-liative care (50) indicated that Dutch and Belgian respondents position continuoussedation until death as an ‘alternative’ choice to euthanasia (legalized in 2002). Thisalternative should be presented to patients to enhance their autonomy. In contrast,respondents from theUK(where euthanasiahasnotbeen legalized)appeared tobestronglyinfluencedbyadiscourseofpalliativecare,placingmoreemphasisoncarefulmedicalmanagementofsymptoms.Aqualitativestudyin2012demonstratedasystem-aticvariationinend-of-lifecaresedationpracticeanditsconceptualizationintheUK,BelgiumandtheNetherlands(48).UKphysiciansandnursesreportedacontinuumofpracticefromtheprovisionof lowdosesofsedativestocontrolterminalrestlessnesstorarelyencountereddeepsedation.Incontrast,Belgianrespondentspredominantlyreportedtheuseofdeepsedation,emphasizingtheimportanceofrespondingtothepa-tient’srequestforreliefofsuffering.Dutchrespondentsemphasizeda‘formal’medicaldecisiontoinitiatesedationbasedonaprocessofconsultationanddiscussionwiththepatientand/ortheirfamilydependingonthepatient’scapacityandwithprofessionalcolleagues.Sedationwasusedonceitwasestablishedthatarefractorysymptomwaspresentanditwasclearthatapatientwasinthelastdaysoflife(48).

caring for the relatives

Professionals working in palliative care stress the importance of good care for thepatient’s relatives (51). The World Health Organization’s definition of palliative careincorporatesasupportsystemtohelptherelativestocopeduringthepatient’sillnessand during their own bereavement (52). Patient and relatives together are ‘the unitof care’. Being a close relative of someone who is in the final phase of life is oftencomplicated. Relatives must handle both their own sorrow and that of the dyingperson, inadditiontoaddressingamultitudeofpractical issues(53).Relativesoften

Chapter 112

perceivebeinginvolvedinthecareforadyingpersonasburdensome(53).Thiscouldpotentiallyincreasethevulnerabilitytothelossexperienceorslowdownadjustmenttobereavement(54).Althoughthemajorityofadultsrecoverafterthelossofalovedone,aportioncontinuestogrieveforanextendedperiodoftimeanddevelopssymptomsofastateknownascomplicatedgrief(55).

Palliativesedationisafar-reachinginterventionthatmayhaveasignificantimpactontheexperienceofthedyingprocessbothforpatientsandtheirrelatives.Thestartofsedationmaybe the timeatwhich itdawnsonthe family that thepatient’sdeathisimminentandtheintimacyoffamilycaremaybedisruptedbytheintroductionoftechnologiessuchassedation(11).Relativesplayanimportantrole,bothwhenseda-tionisbeingconsideredandwhileitisbeingcarriedout.However,therehasbeenlittleresearchexploringtheperceptionsandexperiencesofbereavedrelativeswithpalliativesedationanditsimpactontheirwellbeingafterthepatient’sdeath.

1.2 this thesis

This thesis aims to provide a comprehensive view of bereaved relatives’ experienceswiththepracticeofpalliativesedation.Because ‘palliativesedation’ is thetermmostcommonlyused inguidelinesandresearchpapers (11,28,29,56), this termwillbeused throughout this thesis. Throughout this thesis, attention will be paid to one ofthesubtypesofpalliative sedation,namelycontinuoussedationuntildeath.Further,weaimtogainmoreinsightinthepotentiallife-shorteningeffectofpalliativesedationandtheriskfactorsforcomplicatedgriefinolderadults.Toachievethesegoals,severalresearchquestionswillbeaddressed.

Research question 1: What are the experiences of bereaved relatives with the practice of continuous palliative sedation?

Research question 2: What is the prolonged impact of palliative sedation on relatives’ experience of the dying phase and their wellbeing after the patient’s death?

Toanswertheseresearchquestions,datawereusedfromtheUNBIASEDstudy.

UnBiaseD study

The‘UNBIASED’study(UK-Netherlands-BelgiumInternAtionalSEDationstudy),whichstartedin2009,isacollaborationbetweenresearchteamsinUK,BelgiumandtheNetherlandswithfundingfromtheEconomicandSocialResearchCouncil(UK),


ResearchFoundationFlanders(BE),theFlemishCancerAssociation(BE),theResearchCouncilofGhentUniversity(BE),theNetherlandsOrganisationforScientificResearch(NL)andtheNetherlandsOrganisationforHealthResearchandDevelopment(NL).The aim of the study is to explore decision-making surrounding the application oftherapeutic (orpalliative) sedation in contemporaryclinicalpractice, experiencesofclinicalstaffanddecedents’companionsofitsuseandtheirperceptionsofitscontribu-tiontothemanagementofdeath.Amixed-methodapproachhasbeenused,consistingofaliteraturereview,focusgroupswithbereavedinformalcaregivers(NL);qualitativeinterviewswithinformalcaregiverscloselyinvolvedinthecareofcancerpatientswhoreceivedcontinuoussedationuntildeath(UK,BE,NL);andasurveyamongbereavedrelatives(NL).

1. Systematic literature reviewToreviewtheexistingevidenceontheexperiencesofrelativeswiththepracticeofpal-liativesedation,asystematicliteraturereviewwasperformedin2010.Severaldatabasesweresearchedforempiricalstudiesonrelatives’experienceswithpalliativesedation.Weinvestigatedrelatives’involvementinthedecision-makingandsedationprocesses,whethertheyreceivedadequateinformationandsupport,andtheirpositiveandnega-tiveemotions(formoredetails,seechapter2).

2. Focus groupsTo explore relatives’ experiences with palliative sedation and to gain more insightin positive and negative elements in their evaluation of palliative sedation, betweenOctober2010andMarch2011threefocusgroupswereheldwithatotalof10relativesofpatientswhoreceivedpalliativesedationinvariouscaresettingsintheNetherlands.Inaddition,fourindividualinterviewswerecarriedout.Thedecision-makingprocess,informationandcommunication, theprocessof sedation,andtheoverallevaluationof the relatives of the use of sedation were addressed during the focus groups andinterviewswiththeuseofanaide-memoire(formoredetails,seechapter3).

3. InterviewsToexplorerelatives’descriptionsandexperiencesofcontinuoussedationinend-of-lifecare forcancerpatientsandto identifyandexplaindifferencesbetweenrespondentsfrom the Netherlands, Belgium and the UK, face-to-face interviews were held withbereavedrelatives.InterviewswereheldbetweenJanuary2011andMay2012with38relativesof32cancerpatientswhoreceivedcontinuous sedationuntildeath inhos-pitals, thecommunity,andhospices/palliativecareunits.Theinterviewsweresemi-structured, supported with the use of aide memoires. The aide memoire focused onrelatives’recollectionofthecareforthepatientandoftheuseofsedationinparticular.

Chapter 114

Relativeswereaskedtodescribethedecision-makingprocess,theinformationreceivedandhowthesedationwascarriedout.Finally,relativeswereaskedhowtheyingenerallookedbackontheuseofsedation.

4. SurveyTostudytheimpactofpalliativesedationonrelatives’experienceofthedyingphaseand their wellbeing after the patient’s death, we conducted an observational studyamongbereavedrelativesofconsecutivepatientswhohaddiedanexpecteddeathintheErasmusMedicalCentre-DanielDenHoedRotterdamorhospiceLaurensCadenzaRotterdam, between 2010 and 2013. We included patients that died after the use ofpalliativesedationordiedwithouttheuseofpalliativesedation.Bereavedrelativesofbothgroupsofpatientswereaskedtofillinaquestionnaireaboutthepossibleuseofpalliativesedation,theirexperienceofthedyingphase,andtheirwellbeingafterthepa-tient’sdeath.Intotal,241bereavedrelativesansweredaquestionnaire(sedationn=151,nosedationn=90).Thetotalresponseratewas44%(formoredetails,seechapter5).

Research question 3: How accurately do physicians’ estimate the potential life-shortening effect of continuous sedation until death?

Togetinsightintheaccuracyofestimatesofthelife-shorteningeffectofcontinuoussedationuntildeath,dataoftheAMROSEstudywereanalysed.

amrose study

Thegoalof theAMROSE-study is todescribe thepracticeofpalliativesedationandto establish to what extent this practice is in accordance with the RDMA-guideline(57, 58). For that purpose 370 physicians were required to answer a questionnaire.DatacollectiontookplacebetweenFebruary2008andSeptember2008.Thepotentiallifeshorteningeffectofcontinuoussedationwasestimatedthroughadirectapproach(question:Didcontinuoussedation,accordingtoyourestimation,hastenthepatients’death?;ifyes:byhowmuchtime?)andanindirectapproach(estimatedlifeexpectancyminusdurationofsedation).Thetwoapproachestoestimatethepotentiallifeshort-ening effect of continuous sedation until death were compared. In total, 370 Dutchphysiciansansweredaquestionnaireandreportedabouttheirlastpatientwhoreceivedcontinuoussedationuntildeath.Thetotalresponseratewas61%(formoredetails,seechapter6).

Research question 4: What are risk factors for complicated grief among older adults?

Toanswerthisresearchquestion,datafromtheRotterdamstudywereused.


rotterdam study

TheRotterdamStudyisaprospectivecohortstudythatstartedin1990inOmmoord,a suburbofRotterdam,among10.994menandwomenaged55andover.ThemainobjectiveoftheRotterdamStudyistoinvestigatetheprevalenceandincidenceofandriskfactorsforchronicdiseasesintheelderly.Thechronicdiseasesofinterestarecar-diovascular,neurological,locomotorandophthalmologicdiseases.ThefindingsfromtheRotterdamStudywillcontribute toabetterpreventionandtreatmentofchronicdiseasesintheelderly.

Theaimof the studywas tofindoutwhetherpersonal characteristicsof thepatientandthebereavedpartner,thepatient’sillness,end-of-lifecareandthenatureofdeathareriskfactorsforcomplicatedgriefinolderadults.Togainmoreinsightintheseriskfactors for complicated grief, a nested case-control study was performed within theRotterdamStudy.100couplesofwhichonepersonhaddeceasedandtheotherpersonexperienced ‘complicatedgrief ’wereselected,and100controlcouplesofwhichoneperson had deceased and the other person experienced ‘normal grief ’. Complicatedgriefwasassessedwitha17-itemInventoryofComplicatedGrief.Determinantswereassessed using several sources of information that were available for all participantsof the Rotterdam Study. Additionally, medical files of the deceased were manuallyscreened(formoredetails,seechapter7).

1.3 oUtline of this thesis

Inchapter 2theresultsofasystematicreviewontheexperiencesofrelativeswiththepractice of palliative sedation are presented in the light of the recommendations inguidelinesonpalliative sedation toprotect thewellbeingof relatives involved in theuseofsedation.Chapter 3describesrelatives’experienceswithpalliativesedationandpositive and negative elements in their evaluation of palliative sedation. Chapter 4exploresrelatives’descriptionsandexperiencesofcontinuoussedationinend-of-lifecareforcancerpatientsandpotentialdifferencesbetweenrespondentsfromtheNeth-erlands,BelgiumandtheUK.Chapter 5examinestheprolongedimpactofpalliativesedationonbereavedrelatives’ satisfactionwith thedyingphaseand theirwellbeingafter the patient’s death. Chapter 6 reports on the accuracy of physicians’ estimatesof the life-shortening effect of continuous sedation until death. Chapter 7 explorespotentialriskfactorsforcomplicatedgriefinolderadults.Finally,inchapter 8,thekeyfindingsofthestudyandthescientificandpolicyconsequencesarediscussed.

Chapter 116

references

1. SealeC.Changingpatternsofdeathanddying.SocSciMed2000;51(6):917-30. 2. vanderHeideA,DeliensL,FaisstK,NilstunT,NorupM,PaciE,etal.End-of-lifedecision-making

insixEuropeancountries:descriptivestudy.Lancet2003;362(9381):345-50. 3. van der Wal G, van der Heide A, Onwuteaka-Philipsen BD, van der Maas PJ. Medische besluit-

vormingaanheteindevanhetleven.Depraktijkendetoetsingsprocedureeuthanasie.Utrecht:DeTijdstroom;2003.

4. DrazenJM.Decisionsattheendoflife.NEnglJMed2003;349(12):1109-10. 5. MurraySA,BoydK,SheikhA.Palliativecareinchronicillness.BMJ2005;330(7492):611-12. 6. QuillTE.Dyinganddecisionmaking--evolutionofend-of-lifeoptions.NEnglJMed2004;350(20):

2029-32. 7. RietjensJAC.Medicaldecision-makingattheendoflife(thesis).Rotterdam:ErasmusMC,Univer-

sityMedicalCentreRotterdam;2006. 8. SepulvedaC,MarlinA,YoshidaT,UllrichA.PalliativeCare:theWorldHealthOrganization’sglobal

perspective.JPainSymptomManage2002;24(2):91-6. 9. QuillTE,ByockIR.Respondingtointractableterminalsuffering:theroleofterminalsedationand

voluntaryrefusaloffoodandfluids.ACP-ASIMEnd-of-LifeCareConsensusPanel.AmericanCol-legeofPhysicians-AmericanSocietyofInternalMedicine.AnnInternMed2000;132(5):408-14.

10. ChernyN,PortenoyRK.Sedationinthemanagementofrefractorysymptoms:Guidelinesforevalu-ationandtreatment.JournalofPalliativeCare1994;10(2):31-8.

11. Royal Dutch Medical Association (RDMA). Guideline for palliative sedation [in Dutch]. 2009.Available from http://knmg.artsennet.nl/Publicaties/KNMGpublicatie/Guideline-for-palliative-sedation-2009.htm.AccessedMarch,2014.

12. de Graeff A, Dean M. Palliative sedation therapy in the last weeks of life: a literature review andrecommendationsforstandards.JPalliatMed2007;10(1):67-85.

13. RietjensJA,vanderMaasPJ,Onwuteaka-PhilipsenBD,vanDeldenJJ,vanderHeideA.TwoDe-cadesofResearchonEuthanasiafromtheNetherlands.WhatHaveWeLearntandWhatQuestionsRemain?JBioethInq2009;6(3):271-83.

14. VerkerkM,vanWijlickE,LegemaateJ,deGraeffA.AnationalguidelineforpalliativesedationintheNetherlands.JPainSymptomManage2007;34(6):666-70.

15. MiccinesiG,RietjensJA,DeliensL,PaciE,BosshardG,NilstunT,etal.Continuousdeepsedation:physicians’experiencesinsixEuropeancountries.JPainSymptomManage2006;31(2):122-9.

16. RietjensJA,vanderHeideA,VrakkingAM,Onwuteaka-PhilipsenBD,vanderMaasPJ,vanderWalG.PhysicianreportsofterminalsedationwithouthydrationornutritionforpatientsnearingdeathintheNetherlands.AnnInternMed2004;141(3):178-85.

17. ChiuTY,HuWY,LueBH,ChengSY,ChenCY.SedationforrefractorysymptomsofterminalcancerpatientsinTaiwan.JPainSymptomManage2001;21(6):467-72.

18. FainsingerRL,LandmanW,HoskingsM,BrueraE.SedationforuncontrolledsymptomsinaSouthAfricanhospice.JPainSymptomManage1998;16(3):145-52.

19. FainsingerRL,WallerA,BercoviciM,BengtsonK,LandmanW,HoskingM,etal.Amulticentreinternational study of sedation for uncontrolled symptoms in terminally ill patients. Palliat Med2000;14(4):257-65.

20. Muller-BuschHC,AndresI,JehserT.Sedationinpalliativecare-acriticalanalysisof7yearsexperi-ence.BMCPalliatCare2003;2(1):2.


21. StoneP,PhillipsC,SpruytO,WaightC.Acomparisonoftheuseofsedativesinahospitalsupportteamandinahospice.PalliatMed1997;11(2):140-4.

22. SykesN,ThornsA.Theuseofopioidsandsedativesattheendoflife.LancetOncol2003;4(5):312-8. 23. SykesN,ThornsA.Sedativeuseinthelastweekoflifeandtheimplicationsforend-of-lifedecision

making.ArchInternMed2003;163(3):341-4. 24. Seale C. End-of-life decisions in the UK involving medical practitioners. Palliat Med 2009;23(3):

198-204. 25. Onwuteaka-PhilipsenBD,Brinkman-StoppelenburgA,PenningC,deJong-KrulGJ,vanDeldenJJ,

vanderHeideA.Trends inend-of-lifepracticesbeforeandafter theenactmentof theeuthanasialawintheNetherlandsfrom1990to2010:arepeatedcross-sectionalsurvey.Lancet2012;380(9845):908-15.

26. RietjensJ,vanDeldenJ,Onwuteaka-PhilipsenB,BuitingH,vanderMaasP,vanderHeideA.Con-tinuousdeepsedationforpatientsnearingdeath in theNetherlands:descriptivestudy.BMJ2008;336(7648):810-3.

27. BilsenJ,VanderSticheleR,BroeckaertB,MortierF,DeliensL.Changesinmedicalend-of-lifeprac-ticesduringthelegalizationprocessofeuthanasiainBelgium.SocSciMed2007;65(4):803-8.

28. ChernyNI,RadbruchL,BoardoftheEuropeanAssociationforPalliativeC.EuropeanAssociationforPalliativeCare(EAPC)recommendedframeworkfortheuseofsedationinpalliativecare.PalliatMed2009;23(7):581-93.

29. MoritaT,BitoS,KuriharaY,UchitomiY.DevelopmentofaclinicalguidelineforpalliativesedationtherapyusingtheDelphimethod.JPalliatMed2005;8(4):716-29.

30. BroeckaertB,MullieA,GielenJ,DesmetM,VandenBergheP.GuidelinePalliativeSedation:Eth-icsSteeringCommitteeoftheFederationforPalliativeCareFlanders[Sedation];2012[cited2012October24].Availablefrom:http://www.pallialine.be/template.asp?f=rl_sedatie.htm.

31. Murray SA, Boyd K, Byock I. Continuous deep sedation in patients nearing death. BMJ 2008;336(7648):781-2.

32. SteinhauserKE,ChristakisNA,ClippEC,McNeillyM,McIntyreL,TulskyJA.Factorsconsideredimportantat theendof lifebypatients, family,physicians,andothercareproviders. JAMA2000;284(19):2476-82.

33. MoritaT,IkenagaM,AdachiI,NarabayashiI,KizawaY,HonkeY,etal.Concernsoffamilymembersofpatientsreceivingpalliativesedationtherapy.SupportCareCancer2004;12(12):885-9.

34. Morita T, Ikenaga M, Adachi I, Narabayashi I, Kizawa Y, Honke Y, et al. Family experience withpalliative sedation therapy for terminally ill cancer patients. J Pain Symptom Manage 2004;28(6):557-65.

35. RietjensJA,HauserJ,vanderHeideA,EmanuelL.Havingadifficulttimeleaving:experiencesandattitudesofnurseswithpalliativesedation.PalliatMed2007;21(7):643-9.

36. InghelbrechtE,BilsenJ,MortierF,DeliensL.ContinuousdeepsedationuntildeathinBelgium:asurveyamongnurses.JPainSymptomManage2011;41(5):870-9.

37. MercadanteS,IntravaiaG,VillariP,FerreraP,DavidF,CasuccioA.ControlledSedationforRefrac-torySymptomsinDyingPatients.JPainSymptomManage2009;37(5):771-9.

38. MoritaT,TsunodaJ, InoueS,ChiharaS.Effectsofhighdoseopioidsandsedativesonsurvival interminallyillcancerpatients.JPainSymptomManage2001;21(4):282-9.

39. RietjensJA,vanZuylenL,vanVeluwH,vanderWijkL,vanderHeideA,vanderRijtCC.Palliativesedationinaspecializedunitforacutepalliativecareinacancerhospital:comparingpatientsdyingwithandwithoutpalliativesedation.JPainSymptomManage2008;36(3):228-34.

Chapter 118

40. BruinsmaSM,RietjensJA,SwartSJ,PerezRS,vanDeldenJJ,vanderHeideA.Estimatingthepoten-tiallife-shorteningeffectofcontinuoussedationuntildeath:acomparisonbetweentwoapproaches.JMedEthics2013.

41. SwartSJ,BrinkkemperT,Rietjens JA,BlankerMH,vanZuylenL,RibbeM,etal.Physicians’andnurses’experienceswithcontinuouspalliativesedationintheNetherlands.ArchInternMed2010;170(14):1271-4.

42. ChambaereK,BilsenJ,CohenJ,RietjensJA,Onwuteaka-PhilipsenBD,MortierF,etal.ContinuousdeepsedationuntildeathinBelgium:anationwidesurvey.ArchInternMed2010;170(5):490-3.

43. VanDeijckRH,KrijnsenPJ,Hasselaar JG,VerhagenSC,VissersKC,KoopmansRT.Thepracticeofcontinuouspalliative sedation inelderlypatients:anationwideexplorative studyamongDutchnursinghomephysicians.JAmGeriatrSoc2010;58(9):1671-8.

44. vanDeldenJ.Terminalsedation:differentpractices,differentevaluations.In:TännsjöT,ed.Terminalsedation:euthanasiaindisguise?Dordrecht:KluwerAcademicPublishers;2004.

45. vanDeldenJJ.Terminalsedation:sourceofarestlessethicaldebate.JMedEthics2007;33(4):187-8. 46. BillingsJA,BlockSD.Sloweuthanasia.JPalliatCare1996;12(4):21-30. 47. AnquinetL,RietjensJA,SealeC,SeymourJ,DeliensL,vanderHeideA.Thepracticeofcontinuous

deepsedationuntildeathinFlanders(Belgium),theNetherlands,andtheU.K.:acomparativestudy.JPainSymptomManage2012;44(1):33-43.

48. SeymourJ,RietjensJ,BruinsmaS,DeliensL,SterckxS,MortierF,etal.Usingcontinuoussedationuntildeathforcancerpatients:Aqualitative interviewstudyofphysicians’andnurses’practice inthreeEuropeancountries.PalliatMed2014(acceptedforpublication).

49. RietjensJA,VoorheesJR,vanderHeideA,DrickamerMA.Approachestosufferingattheendoflife:theuseofsedationintheUSAandNetherlands.JMedEthics2014;40(4):235-40.

50. SeymourJE,JanssensR,BroeckaertB.Relievingsufferingattheendoflife:practitioners’perspec-tivesonpalliativesedationfromthreeEuropeancountries.SocSciMed2007;64(8):1679-91.

51. vanDoorenS,vanVeluwHT,vanZuylenL,RietjensJA,PasschierJ,vanderRijtCC.Explorationofconcernsofrelativesduringcontinuouspalliativesedationoftheirfamilymemberswithcancer.JPainSymptomManage2009;38(3):452-9.

52. WorldHealthOrganization.Definitionofpalliativecare.2008.Availablefromhttp://www.who.int/cancer/palliative/definition/en.AccessedFebruary22,2011.

53. AndershedB.Relatives inend-of-lifecare-part1:asystematicreviewof the literaturethefive lastyears,January1999-February2004.JClinNurs2006;15(9):1158-69.

54. ArandaS,MilneD.Guidelinesfortheassessmentofcomplicatedbereavementriskinfamilymem-bersofpeoplereceivingpalliativecare2000Melbourne:CentreforPalliativeCare.

55. NewsonRS,BoelenPA,HekK,HofmanA,TiemeierH.Theprevalenceandcharacteristicsofcom-plicatedgriefinolderadults.JAffectDis2011;132(1-2):231-8.

56. Claessens P, Menten J, Schotsmans P, Broeckaert B. Palliative sedation: a review of the researchliterature.JPainSymptomManage2008;36(3):310-33.

57. SwartSJ,RietjensJA,vanZuylenL,ZuurmondWW,PerezRS,vanderMaasPJ,etal.Continuouspalliativesedationforcancerandnoncancerpatients.JPainSymptomManage2012;43(2):172-81.

58. SwartSJ,vanderHeideA,BrinkkemperT,vanZuylenL,PerezR,RietjensJ.Continuouspalliativesedationuntildeath:practiceafterintroductionoftheDutchnationalguideline.BMJSupportPalliatCare.2012;2(3):256-63.

Chapter 2The experiences of relatives with the practice of palliative sedation: A systematic review

Bruinsma SM, Rietjens JA, Seymour JE, Anquinet L, van der Heide A

Journal of Pain and Symptom Management 2012; 44 (3): 431-445

Chapter 222

aBstract

Context. Guidelines about palliative sedation typically include recommendations toprotectthewellbeingofrelatives.

Objectives.Theaimofthisstudywastosystematicallyreviewevidenceontheexperi-encesofrelativeswithpalliativesedation.

Methods. PubMed,Embase,WebofSience,PsychInfoandCinahlweresearchedforempiricalstudiesonrelatives’experienceswithpalliativesedation.Weinvestigatedrela-tives’involvementinthedecision-makingandsedationprocess,whethertheyreceivedadequateinformationandsupport,andrelatives’emotions.

Results. Ofthe564studiesidentified,39wereincluded.Thestudies(30quantitative,sixqualitativeandthreemixedmethods)wereconductedin16countries;threestudieswerebaseduponrelatives’reports,26onphysicians’andnurses’proxyreports,sevenon medical records and three combined different sources. The 39 studies yielded acombinedtotalof8791respondentsorstudiedcases.Caregiversinvolvedrelativesinthedecision-makingin69-100%ofallcases(19quantitativestudies)andin60-100%ofallcases,relativeswerereportedtohavereceivedadequateinformation(fivequan-titative studies). Only two quantitative studies reported on relatives’ involvement intheprovisionofsedation.Despitethefactthatthemajorityofrelativeswerereportedtobecomfortablewiththeuseofpalliativesedation(sevenquantitativestudies,fourqualitativestudies),severalstudiesfoundthatrelativesweredistressedduetotheuseofsedation(fivequantitativestudies,fivequalitativestudies).Nostudiesreportedspecifi-callyaboutthesupportprovidedtotherelatives.

Conclusion.Relatives’experienceswithpalliativesedationaremainlystudiedfromtheperspectiveofproxies,mostlyprofessionalcaregivers.Themajorityofrelativesseemsto be comfortable with the use of palliative sedation; however, they may experiencesubstantialdistressduetotheuseofsedation.

Systematic review on relatives’ experiences with PS 23

introDUction

Duringthelastdecades,deathastheresultofacutediseaseslargelyhasbeenreplacedbydeathfromchronicdiseases(1),resultinginanincreasedneedforend-of-lifecare.In some cases, patients who are approaching death experience refractory symptomsthataredifficulttoalleviatedespiteintensivemedicaltreatment(2-3).Thissometimesrequiresatreatmentoflastresort:palliativesedation(3).Thisentailstheuseofsedat-ingdrugstoinduceastateofdecreasedconsciousnessuntildeath(4).

Itisknownthatpalliativesedationisfrequentlyusedinend-of-lifecare.AstudyinsixEuropeancountriesreportedthatitwasusedin2.5%-8.5%ofalldeaths(5).DutchnationwidestudiesshowedthatpalliativesedationisincreasinglyusedintheNether-lands,upto8.2%ofalldeathsin2005(6-7).Palliativesedationisusedinallsettingswherepatientsdie,butmostofteninhospitalsandforpatientswithcancer(5,8-11).Withinpalliativecaresettings,incidenceestimatesoftheuseofsedativespriortodeathrangefrom15%uptomorethan60%ofpatients(12-16).Itisusuallyrecommendedthatfortheuseofpalliativesedation,thepatient’sdiseaseshouldbeirreversibleandadvanced, with a life expectancy of, at most, two weeks; benzodiazepines should bethedrugoffirstchoice;artificialhydrationshouldonlybeofferedtosedatedpatientswhenthebenefitwilloutweightheharm;thesedationshouldnotbeintendedtohastendeath;andadvicefrompalliativecarespecialistsshouldbesoughtbeforeinitiatingtheuseofsedation(4,17).

Toguidecaregivers,severalinternational,nationalandlocalguidelinesfortheuseofpalliativesedationhavebeenpublished(18).Theseguidelinestypicallyalsoincluderecommendationstoprotectthewellbeingofrelatives ofpatientswhoreceivepalliativesedation.In2009,theEuropeanAssociationforPalliativeCareintroduceda10-itemframeworkforthedevelopmentofinstitutionalguidelinesfortheuseofpalliativeseda-tion(17).In2005,theRoyalDutchMedicalAssociation(RDMA)publishedanationalguideline for palliative sedation in the Netherlands, which was revised in 2009 (4).Guidelineshavebeenpublishedinothercountriesalso,forexample,in2005,aclini-calguidelineforpalliativesedationwasconstructedinJapan(19).Accordingtotheseguidelines, relativesshouldbe involved in thedecision-making, forexample,bydis-cussingthedecisiontosedate.Furthermore,relativescanbeinvolvedintheprovisionofthesedation,forexamplebyspendingtimewithandobservingthepatientandtoprovidephysiciansandnurseswithinformationaboutthepatient.Therelativesshouldbekeptinformed,atvariouspointsinthecourseofpalliativesedation,ofthepatient’swellbeingandwhattoexpect;andthecareteamshouldcommunicatewiththerelativesinalanguagetheycanunderstand.Thecareteamalsomustprovidesupportivecaretotherelativesbycomfortingandlendingasympatheticeartohelpthemcopewiththeexperience.

Chapter 224

Howtheserecommendationsrelatetotheactualexperiencesofrelativeshasneverbeensystematicallyinvestigated.Theaimofthisstudyistosystematicallyreviewevi-denceontheexperiencesofrelativeswiththepracticeofpalliativesedation.

methoDs

search strategy

Asearchstrategywasdevelopedforfindingrelevantpublicationsinelectroniclitera-turedatabases.InNovember2010,fiveelectronicdatabasesweresearched(PubMed,Embase,WebofScience,PsychINFO,andCINAHL)usingthefollowingsearchstring(“palliative sedation” OR “terminal sedation” OR “continuous deep sedation” OR“continuous sedation”) AND (“end of life” OR palliat* OR terminal* OR death ORdying*).ThesearchstringwasinitiallydevelopedinPubMedandlateradaptedfortheother databases. Because “experiences of relatives” with palliative sedation were notalwaystheprimaryobjectiveofthestudiesfoundandinformationaboutthistopicwassometimesonlyprovidedintablesortext,theseandotherrelatedkeywordswerenotincluded in thesearchstring.Toretrieveall therelevant literature, thesearchstringwasnotrestrictedbylanguageordateofpublication.Thesearchcoveredtheliteraturepublishedbetween1991and2010.Inaddition,referencelistsoftheeventuallyselectedstudiesweremanuallyscreened.

selection criteria

Studies were included when they met the following inclusion criteria: the studyconcerned empirical research (quantitative or qualitative); the study was about pal-liativesedation,notsedationinthecontextofsurgicalprocedures;thestudyincludedinformationabouttheexperiencesofrelativeswithpalliativesedation;theexperiencesofrelativeswereeitherdirectlymeasuredor foundthroughmedicalrecordsorviaaproxy (e.g., physicians, nurses); and the study was about the provision of palliativesedationinadults(olderthan18).Studieswereexcludedwhentheydidnotmeettheseinclusioncriteria.Reviews,studiesreportingduplicatedata,comments,casestudies,ethicalanalysis,andconferenceabstractswerealsoexcluded.

Relativeswerenotnecessarilyrestrictedtofamilymembers,butcouldalsoincludeothers(friends,etc.).

inclusion and evaluation process

Thestudies identifiedwereentered intoEndNoteandduplicateswere removed.TenpercentoftheofthepublicationswereindependentlyassessedbySMB(firstauthor)andJACR(secondauthor)usingtheinclusioncriteria.Cohen’sKappawascalculated


todeterminethedegreeofagreement:κ=0.78,indicatingasubstantialagreement.TheremainingtitleswereassessedbySMB.Thisprocedurewasrepeatedfortheassessmentoftheabstracts(κ=0.78).Ofallthestudiesthatdidnotpasstheselectionprocess,thereasonsfornoninclusionwerelisted.

Data extraction

Datawereextractedusingastandardformthatincludedasthemes:generalinforma-tion,decision-makingprocess,information/communication,involvementintheseda-tiontherapy,feelings/emotionstowardssedation,andsupport.SMBextractedthedatafromthestudiesanddiscussedtheresultswithJACR.

Quality assessment

Because the review included qualitative, quantitative, and mixed methods studies, amulti-methodsassessmenttool,devisedbyHawkeretal(20)wasusedtoevaluatethequalityofindividualstudies.Anassessmentformwasused,whichcoverednineareas;eachareawasratedona four-pointscale, from1(verypoor) to4(good).Theareascovered were abstract and title; introduction and aims; method and data; sampling;dataanalysis;ethicsandbias; results; transferabilityorgeneralizability;and implica-tionsandusefulness.Foreachpaper,itwaspossibletocalculateatotalscore(9=verypoor to36=good) that indicated itsmethodological rigour.As thestudiesuseddif-ferent methods, outcome measures, and samples, it was not appropriate to combinedataacrossstudiesformeta-analysis(21).Themethodologicalqualityoftheselectedpublications was assessed by SMB; JACR assessed a 10% random sample of studies.Bothauthorsagreedonthequalityassessmentofallthestudies.

resUlts

characteristics of the studies

Searchingtheelectronicdatabases,564studieswereidentified(excludingduplicates).Afterscanningthetitles,abstracts,andfulltexts,36studieswereincluded(6%)(Fig.1).Aftermanuallyscreeningthereferencelistsoftheselectedstudies,threestudieswereadded, resulting in 39 studies. The 39 studies included 30 quantitative studies, sixqualitativestudiesandthreemixed-methodstudies.Thestudiesuseddifferentmethodstogatherdata:questionnaires(23studies),medicalrecords(sevenstudies),interviews(fivestudies),andfocusgroups(onestudy).Threestudiesusedseveralmethods.Thestudies were conducted in different care settings: palliative care unit (eight studies),hospital(fourstudies),home(twostudies),hospice(twostudies),andanursinghome(onestudy).Moststudieswereconductedinmultiplesettings(21studies);andinone

Chapter 226

study,thesettingwasnotrestricted.Threestudiesconcernedrelatives’reportsabouttheirexperiencesand26studiesconcernedproxyreports,mainlyfromphysiciansand/or nurses (23 studies) (in three studies, researchers and pharmacists were includedas respondents). Seven studies concerned reports from content analysis of medicalrecords. Three studies combined several sources. Because the data gathered fromrelatives,proxies,andmedicalrecordsdidnotshowsubstantialdifferences,theresultswill not be broken down for these groups. The studies originated from 16 differentcountries, most often from The Netherlands (10 studies) and Japan (eight studies).Thirteenstudieswerepublishedbetween1999and2005and26studiesbetween2005and2010.The39studiesyieldedacombinedtotalof8791respondentsorstudiedcases(seeTable1forafulldescriptionoftheincludedstudies).

In this study, the concept of relatives was not necessarily restricted to familymembers. Because the included studies did not always provide a clear definition of“relatives”,itremainsuncleartowhomtheconceptofrelativesexactlyrelates.

Decision-making process

Of the 39 studies included, 30 provided information about relatives’ involvement inthedecision-makingprocess.Ofthese,25werequantitative(11,16,22-44),twowerequalitative (18,45)and threeusedmixedmethods (46-48).The30studiesyieldeda

Titles

N= 564 (100%)

Abstracts

N=250 (44%)

Full texts

N= 111 (20%)

Included studies

N= 36 (6%)

- No original empirical researcha: n=128 (51%)

- Not about palliative sedation: n=0 (0%)

- Not about relatives: n=11 (4%)

- Not about adults: n=0 (0%)

- No original empirical researcha: n= 34 (31%)



- Not about adults: n=1 (1%)

- Not able to find: n=6 (5%)

3 studies added after hand search

- No original empirical research a: n=169 (30%)



- Not about adults: n= 5 (1%)

Included studies total N=39

figure 1 Inclusion- and evaluation process. a Including reviews, case studies, ethical analysis, studies re-porting duplicate data, comments, and conference abstracts


tabl

e 1

Char

acte

ristic

s in

clud

ed s

tudi

es

firs

t aut

hor

(ref

eren

ce)

Year

of

publ

icat

ion

coun

try

stud

y de

sign

Dat

a co

llect

ion

met

hod

sett

ing

resp

onde

nts

num

ber o

f re

spon

dent

s/

stud

ied

case

s

met

hodo

logi

cal

appr

aisa

l:to

tal s

core

a

Seal

e (1

1)20

10U.

K.Q

uant

itativ

e st

udy,

re

tros

pect

ive

Que

stio

nnai

reH

ome,

eld

er c

are,

ho

spita

lPh

ysic

ians

51

9 re

spon

dent

s32

Chiu

(16)

2001

Taiw

anQ

uant

itativ

e st

udy,

pr

ospe

ctiv

eM

edic

al re

cord

sb H

ospi

ce a

nd

palli

ativ

e ca

re u

nit

hosp

ital

- 70

cas

es26

Riet

jens

(18)

2007

Uni

ted

Stat

esQ

ualit

ativ

e st

udy,

re

tros

pect

ive

Inte

rvie

ws

Palli

ativ

e ca

re u

nit,

med

ical

inte

nsiv

e ca

re u

nit h

ospi

tal

Nur

ses

16 re

spon

dent

s 33

Alon

so-B

abar

ro (2

2)20

10Sp

ain

Qua

ntita

tive

stud

y,

retr

ospe

ctiv

eM

edic

al re

cord

sH

ome

-29

cas

es25

Cham

baer

e (2

3)20

10Be

lgiu

mQ

uant

itativ

e st

udy,

re

tros

pect

ive

Que

stio

nnai

reH

ome,

hos

pita

l, ca

re h

ome

Phys

icia

ns56

1 ca

ses

31

Clae

ssen

s (24

)20

10Be

lgiu

mQ

uant

itativ

e st

udy,

pr

ospe

ctiv

eQ

uest

ionn

aire

Palli

ativ

e ca

re

units

in h

ospi

tals

an

d ho

spic

e

Nur

ses,

rese

arch

ers

20 c

ases

28

De

Gra

eff (2

5)20

08Th

e N

ethe

rland

sQ

uant

itativ

e st

udy,

re

tros

pect

ive

Med

ical

reco

rdsb

E.g.

hos

pita

l, nu

rsin

g ho

me,

ho

spic

e, h

ome

- 13

8 ca

ses

20

Ecke

rdal

(26)

2008

Swed

enQ

uant

itativ

e st

udy,

re

tros

pect

ive

Que

stio

nnai

rec

Hos

pita

l Ph

ysic

ians

, nur

ses

22 c

ases

19

Ford

e (2

7)20

01N

orw

ayQ

uant

itativ

e st

udy,

re

tros

pect

ieve

Que

stio

nnai

reH

ospi

tal

Phys

icia

ns47

resp

onde

nts

27

Has

sela

ar (2

8)20

08Th

e N

ethe

rland

sQ

uant

itativ

e st

udy,

re

tros

pect

ive

Que

stio

nnai

reH

ospi

tals

, hom

e,

nurs

ing

hom

esPh

ysic

ians

304

case

s29

Mar

in (2

9)20

03Sp

ain

Qua

ntita

tive

stud

y,

pros

pect

ive

Med

ical

reco

rdsb

Hos

pita

l-

36 c

ases

24

Chapter 228

tabl

e 1

Char

acte

ristic

s in

clud

ed s

tudi

es (

cont

inue

d)

firs

t aut

hor

(ref

eren

ce)

Year

of

publ

icat

ion

coun

try

stud

y de

sign

Dat

a co

llect

ion

met

hod

sett

ing

resp

onde

nts

num

ber o

f re

spon

dent

s/

stud

ied

case

s

met

hodo

logi

cal

appr

aisa

l:to

tal s

core

a

Mor

ita (3

0)20

04Ja

pan

Qua

ntita

tive

stud

y,

retr

ospe

ctiv

eQ

uest

ionn

aire

Palli

ativ

e ca

re

units

Phys

icia

ns29

resp

onde

nts

25

Mor

ita (3

1)20

04Ja

pan

Qua

ntita

tive

stud

y,

retr

ospe

ctiv

e Q

uest

ionn

aire

Palli

ativ

e ca

re

units

Phys

icia

ns29

resp

onde

nts

27

Mor

ita (3

2)20

05Ja

pan

Qua

ntita

tive

stud

y,

pros

pect

ive

Que

stio

nnai

rec

Palli

ativ

e ca

re

units

Phys

icia

ns10

2 ca

ses

28

Park

er (3

3)20

08Au

stra

liaQ

uant

itativ

e st

udy,

cr

oss-

sect

iona

lQ

uest

ionn

aire

cH

ospi

tal,

hom

ePh

ysic

ians

1478

resp

onde

nts

28

Pom

eran

tz (3

4)20

04U

nite

d St

ates

Qua

ntita

tive

stud

y,

cros

s-se

ctio

nal

Que

stio

nnai

reN

ot re

stric

ted

Phys

icia

ns

135

resp

onde

nts

29

Porz

io (3

5)20

09Ita

lyQ

uant

itativ

e st

udy,

re

tros

pect

ive

Med

ical

reco

rdsb

Hom

e -

16 c

ases

19

Riet

jens

(36)

2006

The

Net

herla

nds

Qua

ntita

tive

stud

y,

retr

ospe

ctiv

eIn

terv

iew

sH

ome,

nur

sing

ho

me

and

hosp

ital

Phys

icia

ns

211

resp

onde

nts

31

Riet

jens

(37)

2004

The

Net

herla

nds

Qua

ntita

tive

stud

y,

retr

ospe

ctiv

e

Inte

rvie

ws

Hom

e, n

ursi

ng

hom

e an

d ho

spita

lPh

ysic

ians

211

resp

onde

nts

31

Riet

jens

(38)

2008

The

Net

herla

nds

Qua

ntita

tive

stud

y,

retr

ospe

ctiv

eM

edic

al re

cord

sPa

lliat

ive

care

uni

t ho

spita

l-

68 c

ases

29

Van

Doo

ren

(39)

2009

The

Net

herla

nds

Qua

ntita

tive

stud

y,

retr

ospe

ctiv

e M

edic

al re

cord

sbPa

lliat

ive

care

uni

t ho

spita

l-

45 c

ases

28

Van

Dei

jck

(40)

2010

The

Net

herla

nds

Qua

ntita

tive

stud

y,

retr

ospe

ctiv

eQ

uest

ionn

aire

Nur

sing

hom

esPh

ysic

ians

31

6 ca

ses

30

Mer

cada

nte

(41)

2009

Italy

Qua

ntita

tive

stud

y,

pros

pect

ive

Med

ical

reco

rds,

inte

rvie

ws

Palli

ativ

e ca

re u

nit

-, re

lativ

es42

cas

es/

resp

onde

nts

23


tabl

e 1

Char

acte

ristic

s in

clud

ed s

tudi

es (

cont

inue

d)

firs

t aut

hor

(ref

eren

ce)

Year

of

publ

icat

ion

coun

try

stud

y de

sign

Dat

a co

llect

ion

met

hod

sett

ing

resp

onde

nts

num

ber o

f re

spon

dent

s/

stud

ied

case

s

met

hodo

logi

cal

appr

aisa

l:to

tal s

core

a

Mor

ita (4

2)20

04Ja

pan

Qua

ntita

tive

stud

y,

retr

ospe

ctiv

eQ

uest

ionn

aire

Palli

ativ

e ca

re

units

Rela

tives

185

resp

onde

nts

30

Mor

ita (4

3)20

04Ja

pan

Qua

ntita

tive

stud

y,

cros

s-se

ctio

nal

Que

stio

nnai

res

Canc

er c

entr

es,

hosp

itals

pal

liativ

e ca

re u

nits

Nur

ses

2607

resp

onde

nts

28

Swar

t (44

)20

10Th

e N

ethe

rland

sQ

uant

itativ

e st

udy,

re

tros

pect

ive

Que

stio

nnai

reH

ome,

nur

sing

ho

mes

, hos

pice

s,an

d ho

spita

ls

Phys

icia

ns, n

urse

s55

5 re

spon

dent

s26

Blon

deau

(45)

2009

Cana

daQ

ualit

ativ

e st

udy,

retr

ospe

ctiv

eIn

terv

iew

sH

ospi

tals

, ho

spic

es,

long

-ter

m–c

are

faci

litie

s, ho

me

Phys

icia

ns19

resp

onde

nts

28

Blon

deau

(46)

2005

Cana

daM

ixed

met

hods

st

udy,

cro

ss-

sect

iona

l

Que

stio

nnai

rec

Hos

pita

ls,

hosp

ices

, hom

e Ph

ysic

ians

, ph

arm

acis

ts12

4 re

spon

dent

s29

Chat

er (4

7)19

98Ca

nada

, U.K

., Ire

land

, Ita

ly,

Uni

ted

Stat

es,

Aust

ralia

, New

Ze

alan

d, S

outh

Af

rica

Mix

ed

met

hods

stud

y,

retr

ospe

ctiv

e

Que

stio

nnai

reIn

patie

nt p

allia

tive

care

faci

lity,

ho

spita

l, ho

me,

ou

tpat

ient

car

e

Phys

icia

ns, n

urse

s53

resp

onde

nts

32

Venk

e G

ran

(48)

2008

Nor

way

M

ixed

met

hods

st

udy,

cro

ss-

sect

iona

l

Que

stio

nnai

reH

ospi

tals

and

pa

lliat

ive

unit

nurs

ing

hom

e

Nur

ses

73 re

spon

dent

s30

Chapter 230

tabl

e 1

Char

acte

ristic

s in

clud

ed s

tudi

es (

cont

inue

d)

firs

t aut

hor

(ref

eren

ce)

Year

of

publ

icat

ion

coun

try

stud

y de

sign

Dat

a co

llect

ion

met

hod

sett

ing

resp

onde

nts

num

ber o

f re

spon

dent

s/

stud

ied

case

s

met

hodo

logi

cal

appr

aisa

l:to

tal s

core

a

Mor

ita (4

9)19

99Ja

pan

Qua

ntita

tive,

pr

ospe

ctiv

eQ

uest

ionn

aire

cH

ospi

ce

Phys

icia

ns87

cas

es26

Braj

tman

(50)

2003

Isra

elQ

ualit

ativ

e st

udy,

re

tros

pect

ive

Inte

rvie

ws,

focu

s gr

oups

Hos

pice

Rela

tives

, clin

ical

st

aff (n

urse

s, ph

ysic

ians

, soc

ial

wor

ker)

32 re

spon

dent

s17

Mor

ita (5

1)20

04Ja

pan

Qua

litat

ive

stud

y,

retr

ospe

ctiv

eQ

uest

ionn

aire

Palli

ativ

e ca

re

units

Rela

tives

185

resp

onde

nts

32

Ford

e (5

2)20

06N

orw

ayQ

uant

itativ

e st

udy,

re

tros

pect

ive

Que

stio

nnai

res

Hos

pita

l Ph

ysic

ians

12 re

spon

dent

s fir

st q

uest

ionn

aire

, 11

6 re

spon

dent

s se

cond

qu

estio

nnai

re

26

Van

den

Bloc

k (5

3)20

09Be

lgiu

mQ

uant

itativ

e st

udy,

re

tros

pect

ive

Que

stio

nnai

rec

Hom

e, c

are

hom

e (e

lder

ly o

rnu

rsin

g ho

me)

, ho

spita

l, in

patie

nt

palli

ativ

e ca

re u

nit

Phys

icia

ns17

7 ca

ses

33

Mae

ssen

(54)

2009

The

Net

herla

nds

Qua

ntita

tive

stud

y,

retr

ospe

ctiv

eQ

uest

ionn

aire

Hos

pita

l, ou

tpat

ient

car

ePh

ysic

ians

, in

form

al

care

give

rs

31 c

ases

29

Miy

ashi

ta (5

5)20

08Ja

pan

Qua

ntita

tive

stud

y,

retr

ospe

ctiv

eQ

uest

ionn

aire

, m

edic

al re

cord

sCa

ncer

cen

ter ,

pa

lliat

ive

care

uni

tRe

lativ

es32

cas

es28

Riet

jens

(56)

2009

The

Net

herla

nds

Qua

litat

ive

stud

y,

cros

s-se

ctio

nal

Focu

s gro

ups

Hom

e, n

ursi

ng

hom

e an

d ho

spita

lPh

ysic

ians

24

resp

onde

nts

31


tabl

e 1

Char

acte

ristic

s in

clud

ed s

tudi

es (

cont

inue

d)

firs

t aut

hor

(ref

eren

ce)

Year

of

publ

icat

ion

coun

try

stud

y de

sign

Dat

a co

llect

ion

met

hod

sett

ing

resp

onde

nts

num

ber o

f re

spon

dent

s/

stud

ied

case

s

met

hodo

logi

cal

appr

aisa

l:to

tal s

core

a

Seym

our (

57)

2007

U.K.

, the

N

ethe

rland

s, Be

lgiu

m

Qua

litat

ive

stud

y,

cros

s-se

ctio

nal

Inte

rvie

ws

Hos

pice

s, ho

spita

ls,

palli

ativ

e ca

re

unit,

com

mun

ity

sett

ing

Phys

icia

ns, n

urse

s, re

sear

cher

s35

resp

onde

nts

31

a Sc

orin

g sy

stem

: 9=v

ery

poor

, 18=

poor

, 27=

fair,

36=

good

b M

edic

al r

ecor

ds a

lso

refe

r to

ass

essm

ent

form

s, re

cord

ing

form

s, co

nsul

t re

cord

s, cl

inic

al

inve

stig

atio

n re

cord

s, ch

arts

, and

not

atio

ns in

mul

tidis

cipl

inar

y re

cord

sc

Que

stio

nnai

res

also

ref

er t

o su

rvey

s, cl

inic

al v

igne

ttes

, dat

a co

llect

ing

shee

ts a

nd r

egis

trat

ion

form

s.

Chapter 232

combinedtotalof8060respondentsorstudiedcases(quantitative,n=7775;qualitative,n=35;mixedmethods,n=250).

Theinvolvementofrelativesinthedecision-makingprocesswasvariouslydescribedin the studies. Some studies reported about involving the relatives in the decision-making process in general terms, whereas other studies reported specific types ofinvolvement,suchasdiscussingthedecision,obtainingconsent,orinformingtherela-tivesaboutthedecision.Quantitativestudiesfoundthatrelativeswereinvolvedinthedecision-makingprocessin81-100%ofallcasesofpalliativesedation(22,28,32,36,44,47).Specificaspectsoftheuseofpalliativesedation(e.g.,theindication,goal,ortheexpectedcourseofthesedation)werediscussedwiththerelativesin90%-93%(27,32,37-38).Relativesgavetheirconsenttousepalliativesedationin69%-100%(16,23,29-31,35-36,39-41).Consentwassometimesnotobtainedfromrelatives,e.g.becauseitwasalreadyobtaineddirectlyfrompatients(29).Therelativeswereinformedaboutthedecisionin95%-100%(26,44).Further,studiesshowedthattherelativesproposedorrequestedtheuseofpalliativesedationin9%-41%(11,32,34,41).Onestudyprovidedinformationaboutthephasebeforetheproposaltousesedationandshowedthatin,70%,therelativeswereinvolvedintheassessmentofintolerablesuffering(24).Accord-ingtoonestudy,physiciansweremorewillingtoprovidepalliativesedationontheirowninitiativethanattherequestofrelatives(33).Anotherstudyshowedthat38%ofthephysiciansandpharmacistsattributedanimportantroletothefamilyintheprocessofdecidingwhethertochoosesedationornot(46).Further,thewellbeingofrelativeswasanindicationfortheuseofpalliativesedationin12%-22%(25,31).Disagreementabouttheuseofsedationwasfoundamongrelativesin10%-17%,betweenthepatientandrelatives in8%-11%andbetweenrelativesandmedical staff in10%(32,42-43).Accordingtoonestudy,14%ofthephysiciansand3%ofthenursesreportedthattheyfeltpressuretostartsedationfrompatientsand/orrelatives(44)(Table2).

Thequalitativedataadditionallyshowedthatphysiciansacknowledge the impor-tanceofinvolvingtherelativesintheprocessofdecidingwhethertousesedationornot,butthatthepatienttypicallyremainstoppriority(45).Nursessometimesfeltthatpatientsand/ortheirrelativesshoulddecidewhensufferingisintolerableandpalliativesedationisnecessary,insteadofthephysicians(48)(Table3).

information/communication

Eightstudiesreportedspecificallyontheinformationrelativesreceivedaboutpalliativesedationoraboutcommunicationissues.Ofthese,fivewerequantitative(31-32,39,42,49),twowerequalitative(50-51)andoneusedmixedmethods(48).Theeightstudiesyielded a combined total of 738 respondents or studied cases (quantitative, n=448;qualitative,n=217;mixedmethods,n=73).


table 2 Results quantitative data (n=32)

core themes no. (%) of studies reporting

range of answers

references

Decision-making process 27 (84%) (11, 16, 22-44, 46-47)a

Involvement in decision-making process

Relatives involved in decision-making process

6 (19%) 81%-100% (22, 28, 32, 36, 44, 47)b

Relatives consent 10 (31%) 69%-100% (16, 23, 29-31, 35-36, 39-41)

Decision discussed with relatives 4 (13%) 90%-93% (27, 32, 37-38)

Relatives informed of decision 2 (6%) 95%-100% (26, 44)

Other findings

Sedation proposed (requested) by relatives

4 (13%) 9%-41% (11, 32, 34, 41)

Well-being of relatives indication for sedation

2 (6%) 12%-33% (25, 31)

Conflicts about the use of sedation between people involved

3 (9%) 8%-17% (32, 42-43)

information/communicationinformation sufficient for relatives

5 (16%) (31-32, 39, 42, 49)

Information provided to relatives (about indication, goal, expected course; reduction in consciousness, ability to communicate, life-threatening complications; effects on survival, life-supporting functions, treatment options other than PS; physical changes and physical status, and prognosis if PS not induced; risks, benefits)

5 (16%) 60%-100% (31-32, 39, 42, 49)

Relatives understood the information

2 (6%) 89%-100% (39, 42)

Prior discussion about end-of-life issues between relatives and medical staff

2 (6%) 75%-82% (32, 42)

Information sufficient for relatives 1 (3%) 75% (42)

involvement in the sedation process

2 (6%) (52-53)

Monitoring patient by relatives 1 (3%) 42% (52)

Involvement relatives in caring patient

1 (3%) 17% (53)

Chapter 234

table 2 Results quantitative data (n=32) (continued)

core themes no. (%) of studies reporting

range of answers

references

Emotions and evaluation 9 (28%) (16, 39, 41-42, 44, 49, 52, 54-55)

Positive emotions or evaluation

Satisfied (or “fair”) with PS 3 (9%) 78%-93% (42, 44, 49)

PS decreased symptom distress patient

1 (3%) 88% (42)

Decision to start sedation in accordance with families’ wish

1 (3%) 100% (52)

PS is appropriate for relatives 1 (3%) 98% (41)

Timing of PS is appropriate for relatives

1 (3%) 77% (42)

Ethical acceptable (right or might be right to use sedation)

1 (3%) 93% (16)

Peaceful death because of PS 1 (3%) 91% (54)

Negative emotions or evaluation

Relatives experienced distress 1 (3%) 25% (42)

Relatives expressed concerns 1 (3%) 51% (39)

Relatives unsatisfied with PS 1 (3%) 5% (42)

Relatives asked to stop the sedation 1 (3%) 5% (41)

PS neg. associated with good death 1 (3%) - (55)

Delirium, ambivalence of patients’ wishes, and lack of objectivity of distress associated with difficulty in making decision for family members

1 (3%) - (49)

Support 0 (0%)

PS= palliative sedationa Not all the reporting studies are discussed in table, some only in textb Mixed methods studies (46-47)


table 3 Results qualitative data (n=7)

core themes no. (%) of studies reporting (%)

aspects references

Decision-making process 3 (43%) Importance of role relatives, but patient’s top priority

(45)

Well-being relatives was an indication sedation (18)

Relatives should decide whether or not to use PS (48)a

Information/communication

3 (43%) Kind of information relatives received (clinical aspects and physical aspects of dying process)

(48)a

Relatives reported desire to know that the maximum efforts had been made, to prepare for the patient’s death, to tell the patient something important before sedation, to understand the nature of the patient’s suffering, and wishes that medical professionals treat the patient with dignity

(51)

Relatives differ in type of information they need (50)

Involvement in sedation process

0 (0%)

Emotions and evaluation 5 (71%) Positive emotions or evaluation

Sedation made relatives feel more comfortable, because it offered them sense of peace and closure

(18)

Relatives wanted the patient’s suffering to end (50-51, 57)

Relatives are grateful for caregivers who treated patient with respect

(50)

Negative emotions or evaluation

Relatives experienced distress, e.g., anger, frustration, disappointment, concerns, struggles, guilt, helplessness, and physical and emotional exhaustion

(18, 50-51, 56-57)

Distress as a result of: inability to interact with patient, feelings that sedation possibly hastened death, longer duration sedation, well-being of the patient, information not easily obtained or not relevant to needs at that moment

(18, 50-51, 56-57)

Support 0 (0%)

a Mixed methods study: (48)

Chapter 236

Overall, the quantitative data showed that relatives received information fromprofessionalcaregiversin60%-100%(31-32,39,42,49).Inthesestudies,thetypeofinformationrangedfromexplanationsaboutthereductioninconsciousness,patients’inability to communicate, life-threatening complications, physical changes, physicalstatus, and the prognosis of the patient. Relatives were reported to understand theinformationin89%-100%(39,42)andonestudyfoundthatrelativesexperiencedtheprovidedinformationassufficientin75%,slightlyinsufficientin22%,andinsufficientin 2% (42). Further, prior discussions about end-of-life issues and/or the choice ofsedationbetweenmedicalstaffandrelativestookplacein75%-82%(32,42)(Table2).

The qualitative data additionally showed that relatives have needs for specifictypes of information concerning the patient’s symptom distress and treatment, thedyingprocessandwhenthepatientwasexpectedtodie(50).Also,relativesreporteda desire to know that the maximum efforts have been made and that there were noothermethodsavailableforsymptomrelief;toprepareforthepatient’sdeath;totellthepatientsomethingimportantbeforethestartofsedation;tounderstandthenatureofthepatient’ssuffering;andexpressedwishesthatmedicalprofessionalstreatthepatientwithdignity(51)(Table3).

involvement during the provision of sedation

Only twostudies (bothquantitative) reportedon the involvementof relatives in theprovision of sedation (N=305): one study reported that patients were monitored byrelatives in42%(butundersupervisionofprofessionalcaregivers)(52),andanotherthatrelativeswereinvolvedinthecareforthepatientin17%(53)(Table2).

emotions and evaluation

Ofthe39studiesincluded,14studiesprovidedinformationaboutrelatives’emotionsregardingsedation.Ofthese,ninewerequantitative(16,39,41-42,44,49,52,54-55)andfivewerequalitative(18,50-51,56-57).The14studiesyieldedacombinedtotalof2022respondentsorstudiedcases(quantitative,n=1730;qualitative,n=292).

Accordingtosevenquantitativestudies,themajorityoftherelativeswerereportedtohavepositivefeelingsregardingtheuseofpalliativesedation(16,41-42,44,49,52,54).Relativesseemedtobesatisfiedwiththeuseofpalliativesedationin78%-93%(42,44, 49). One study found that 88% of relatives felt that palliative sedation helped todecreasepatient’ssymptomdistress(42).Anotherstudyshowedthatrelativesreportedthatpalliativesedationwasappropriatein93%,becauseitendedthepatients’suffering(41),andonestudyshowedthatthetimingofthesedationwasseenasappropriatein77%(42).Inanotherstudy,relativesdescribedpalliativesedationas“ethicallyaccept-able”in93%(16),andresultsfromanotherstudyshowedthatpalliativesedationwasassociatedwithapeacefuldeathin91%(54).Accordingtoonestudy,thedecisionto


start sedation was in accordance with relatives’ wishes in 100% (52). However, fivequantitative studies showed that relativesalsoexperiencednegativeemotionsdue totheuseofsedation(39,41-42,49,55).Accordingtoonestudy,relativeswereunsatis-fiedwiththesedationtherapyin5%(42).Low-levelsatisfactionwassignificantlyas-sociatedwithpoorsymptompalliationaftersedation,insufficientinformationgiving,concernsthatsedationmightshortenthepatient’slife,andthefeelingthattheremightbeotherwaystoprovidesymptomrelief.Inthesamestudy,relativesexpressedhighlevels of emotional distress about sedation in 25%: 10% of the relatives reported tobe very distressed and 15% to be distressed (42). This distress was significantly as-sociated with poor symptom palliation, feeling the burden of responsibility for thedecisionafter sedation, feelingunprepared forchangesofpatientconditions, feelingthat the physicians and nurses were not sufficiently compassionate with the patient,andashorterintervaltothepatient’sdeath(42).Oneotherstudyfoundthatrelativesexpressedconcernsbetweenthestartofthesedationandthedeathoftheirlovedonesin51%(39).Itconcernedconcernsregardingtheaimofthesedation,thewell-beingof the patient, and the well-being of the relatives themselves (feelings of exhaustionbecauseofsleepdeprivation,orunbearable feelingsofwatchingtheir lovedonedie)(39).Anotherstudyshowedthatrelativesaskedtostopthesedationin5%becausetheywantedtocommunicatewiththepatientbeforedeathandwantedtotakethepatienthome(41)(Table2).

Thequalitativedataprovidedmoreinsightinthetypeofnegativeemotionsrelatives’experiencedduetothesedation.“Distress”wasdescribedintermsofanger,frustration,disappointment, concerns, struggles, guilt, helplessness, and physical and emotionalexhaustion(18,50-51,56-57)(Table3).

support

Nostudiesreportedspecificallyaboutthesupportprovidedtotherelatives.

Quality assessment

ThetotalscoresarepresentedinTable1.Onearticlewasratedbetween“verypoor”and“poor”;11articleswereratedbetween“poor”and“fair”;and27articleswereratedbetween“fair”and“good”.

DiscUssion anD conclUsions

Professionalsworkinginpalliativecarestresstheimportanceofgoodcareforthepa-tient’srelativesbecausetheyaretheoneswhoareoftenpresentduringthelastperiodofthepatient’slife,andobviously,themostcloselyinvolvedwiththepatient(39).The

Chapter 238

WorldHealthOrganization’sdefinitionofpalliativecareincorporatesprovidingasup-portsystemtohelptherelativescopeduringthepatient’sillnessandduringtheirownbereavement(58).Patientandrelativestogetherare“theunitofcare”.Theimportanceofrelativesisalsoreflectedinguidelines,whichstressthatrelativesshouldbeinvolvedinthedecision-makingprocess,thattheycanassistinmonitoringthepatient,andthattheyshouldbeclearlyinformedandsupported(4,17,19).Theresultsfromthisreviewsuggest that themajorityofrelativesareadequately involvedinthedecision-makingandreceiveadequateinformation,althoughthereseemsroomforimprovement.How-ever,hardlyanyinformationisavailableaboutrelatives’involvementintheprovisionofsedationandnostudiesreportspecificallyaboutthesupportprovidedtorelatives.Furthermore,despitethefactthatthemajorityofrelativesreportedtobecomfortablewiththeuseofpalliativesedation,ourreviewshowsthatrelativesmayexpressdistressbeforeorduringtheapplicationofsedation.

Several findings deserve particular attention. The first finding concerns the roleof relatives in the decision making. Guidelines recommend that physicians activelyinvolverelativesinthisprocess,butbecausepalliativesedationisamedicalprocedure,itisthephysicianwhobearsfinalresponsibilityforassessingtheindications(4).Ac-cordingtotheresultsofthisreview,relativessometimesseemtoplayaratherdecisiverole in the decision to use sedation, sometimes even more decisive than guidelinesrecommend.Relativesare,forinstance,ofteninvolvedintheassessmentofintolerablesuffering (24) and quite often propose or request the use of palliative sedation (11,32,34,41).Ontheonehand,caregiverssometimesfeel that itshouldbepossibleornecessaryforrelativestodecidewhensufferingisintolerableandpalliativesedationisnecessary(48).Ontheotherhand,physiciansandnursessometimesfeelpressuredbyrelativestostartsedation(44).However,itisshownthatrelativessometimesfeeltheburdenofresponsibilityforthedecisiontousesedationwhichmayleadtofeelingsofdistress(42).

Secondly,thereisalargevariationinthe“needs”relativesexpress.Relativeswantspecifictypesofinformation;theinformationneedstobeeasilyavailableandrelevanttotheirneedsataparticularmomentintime.Thenatureofthedesiredinformationshowsthatit includesmanyfacetsofthesedationprocess,concerningbothpatients’well-beingasrelatives’well-being,andthatprovisionofinformationisimportantdur-ingthewholeprocessofsedation.

Finally,itwasstrikingthatalthoughthemajorityofrelativesreportedtobecom-fortablewiththeuseofsedation,asubstantialamountofrelativesexpresseddistressasaresultofitsuse.Ontheonehand,relativeswantthepatients’sufferingtoend;ontheotherhand,theyexpressconcernsregardingtheaimofsedation,thepatients’well-beingandtheirownwell-being.Apparently,bothemotionscanexistsimultaneously.Thisisinlinewithfindingsthatrelativesgenerallyreporttobesatisfiedwiththecare


receivedattheendoflife,evenwhentheyhaveunmetneeds(59).Relativesexpressan-ger,frustration,disappointment,concerns,struggles,guilt,helplessness,andphysicalandemotionalexhaustion.Thereasonsforsuchdistressweretheinabilitytointeractwith thepatient, concernsaboutapossiblyhasteneddeath,a longerdurationof thesedation,andthefactthatinformationaboutthesedationwasnoteasilyobtainedorlessrelevanttoneedsoftherelativesatthatmoment.Thefactthatrelativesexperiencedistressduetotheuseofsedationisnotsurprising.Beingacloserelativeofsomeoneinthefinalphaseoflifeingeneralisoftencomplicatedingeneral.Relativesmusthandleboththeirownsorrowandthatofthedyingperson,inadditiontosolvingamultitudeofpracticalproblems(60).Asidefromthesedifficulties,relativesofpatientswhore-ceivepalliativesedationalsofaceissuessuchastheinabilitytocommunicatewiththepatientbecauseof thepatient’sreducedconsciousness;beingawakeforseveraldays,leading to exhaustion; the unfamiliarity with sedation, after sometimes an extendedperiodofseveresufferingandfunctionaldeclineofthepatient.

Ourstudyhassomelimitations.Intheliterature,severaltermsareusedforpallia-tivesedation,forexample,continuousdeepsedationandterminalsedation,potentiallylimitingfullcomparisonandextrapolationofthestudies.Second,“experiences”isnotaclearlydefinableentity.Third,whetherpresentedfactsaboute.g.relativesinvolvementinthedecision-makingcanbeinterpretedas“experiencesofrelatives”canbedebated.Weinterpretedtheconceptofexperiencebroadlyandalsoincluded,forinstance,rela-tives’viewsonpalliativesedation.Fourth,notalltheincludedstudiesappearedtobeof“good”quality.Finally,themajorityofpapersanalyzeddidnothaveasamainaimtoinvestigateonrelatives’experienceswithpalliativesedation.Ifthefocusofresearchhadbeenthis,datacouldhavebeendifferent.

Theresultsofthisreviewshowthatthereseemtoexistsomediscrepanciesbetweentherecommendationsmadeinguidelinesandtheactualexperiencesofrelativeswiththepracticeofpalliativesedation.First,itseemsthatrecommendationsarenotalwaysfollowed. For instance, relatives not always perceive the provision of information assufficient(39).Second,thereobviouslyisalackofevidenceaboutsomeaspectsoftherecommendationsmadeinguidelines.Consideringthefactthatalltheguidelinesaboutpalliativesedationstresstheimportanceofinvolvingrelativesinthesedationprocessandsupportingtherelativesbefore,during,andafterthesedationoftheirlovedones,it isaremarkablefindingthatthereisnoevidenceabouttheseissues.Inconclusion,wefoundthatrelatives’experienceswiththepracticeofpalliativesedationaremainlystudiedfromtheperspectiveofproxies,mostlyprofessionalcaregivers.Studiesshowthatthemajorityofrelativesisinvolvedinthedecision-makingprocess.Themajorityof relatives receivesadequate information, although there is room for improvement.Hardlyanyinformationisavailableintheliteratureaboutrelatives’involvementinthesedationprocessandnostudiesreportspecificallyaboutthesupportprovidedtothe

Chapter 240

relatives. Despite the fact that the majority of relatives seem to be comfortable withthe use of palliative sedation, there are indications that several of them experiencesubstantialdistresswithitsuse.

DisclosUres anD acknowleDGements

Thissystematic review ispartof a larger studyaboutpalliative sedationand its roleinend-of-lifecare: theUNBIASEDstudy.TheUNBIASEDstudy(U.K.-Netherlands-BelgiumInternationalSEDationstudy) isacollaborationbetweenresearch teams intheU.K.,Belgium,andTheNetherlands,withfundingfromtheEconomicandSocialResearchCouncil(U.K.),theResearchFoundationFlanders(BE),theFlemishCancerAssociation(BE),theResearchCouncilofGhentUniversity(BE),TheNetherlandsOr-ganizationforScientificResearch(NL),andTheNetherlandsOrganizationforHealthResearchandDevelopment(NL).Theauthorsconfirmthat therearenofinancialorpersonal relationshipswith individuals,organizations,orcompanies thatcouldhaveinappropriatelyinfluencedthework.

Theauthors thankLouisVolkers forhishelpwiththedevelopmentof thesearchstrategyintheelectronicliteraturedatabases.


references

1. SealeC.Changingpatternsofdeathanddying.SocSciMed2000;51:917-30. 2. ChernyN,PortenoyRK.Sedationinthemanagementofrefractorysymptoms:Guidelinesforevalu-

ationandtreatment.JPalliatCare1994;10(2):31-8. 3. QuillTE,ByockIR.Respondingtointractableterminalsuffering:Theroleofterminalsedationand

voluntaryrefusaloffoodandfluids.AnnInternMed2000;132:402-14. 4. Royal Dutch Medical Association (RDMA). Guideline for palliative sedation [in Dutch]. 2009.

Available from http://knmg.artsennet.nl/Publicaties/KNMGpublicatie/Guideline-for-palliative-sedation-2009.htm.AccessedOctober,2010.

5. MiccinesiG,RietjensJAC,DeliensL,etal.Continuousdeepsedation:physicians’experiencesinsixEuropeancountries.JPainSymptomManage2006;31:122-9.

6. vanderHeideA,Onwuteaka-PhilipsenBD,RurupML,etal.End-of-lifepracticesintheNetherlandsundertheEuthanasiaAct.NEnglJMed2007;356:1957-65.

7. van der Wal G, van der Heide A, Onwuteaka-Philipsen BD, van der Maas PJ. Medical decision-makingat theendof life:practicesandtheeuthanasiareviewprocedure.Utrecht:DeTijdstroom,2003.

8. BilsenJ,VanderSticheleR,BroeckaertB,MortierF,DeliensL.Changesinmedicalend-of-lifeprac-ticesduringthelegalizationprocessofeuthanasiainBelgium.SocSciMed2007;65:803-8.

9. Rietjens J, van Delden J, Onwuteaka-Philipsen B, Buiting H, et al. Continuous deep sedation forpatientsnearingdeathintheNetherlands:descriptivestudy.BMJ2008;336:810-13.

10. SealeC.End-of-lifedecisionsintheUKinvolvingmedicalpractitioners.PalliatMed2009;23:198-204.

11. SealeC.Continuousdeepsedationinmedicalpractice:adescriptivestudy.JPainSymptomManage2010;39(1):44-53.

12. FainsingerRL,WallerA,BercoviciM,etal.Amulticentreinternationalstudyofsedationforuncon-trolledsymptomsinterminallyillpatients.PalliatMed2000;14:257-65.

13. Muller-BuschHC,AndresI,JehserT.Sedationinpalliativecare-Acriticalanalysisof7yearsexperi-ence.BMCPalliatCare2003;2:1-9.

14. StoneP,PhillipsC,SpruytO,WaightC.Acomparisonoftheuseofsedativesinahospitalsupportteamandinahospice.PalliatMed1997;11:140-4.

15. SykesN,ThornsA.Theuseofopioidsandsedativesattheendoflife.LancetOncol2003;4:312-8. 16. ChiuTY,HuWY,LueBH,ChengSY,ChenCY.Sedationforrefractorysymptomsofterminalcancer

patientsinTaiwan.JPainSymptomManage2001;21(6):467-72. 17. ChernyNI,RadbruchL.EuropeanAssociationforPalliativeCare(EAPC)recommendedframework

fortheuseofsedationinpalliativecare.PalliatMed2009;23:581-93. 18. RietjensJA,HauserJ,vanderHeideA,EmanuelL.Havingadifficulttimeleaving:experiencesand

attitudesofnurseswithpalliativesedation.PalliatMed2007;21(7):643-9. 19. MoritaT,BitoS,KuriharaY,UchitomiY.Developmentofaclinicalguidelineforpalliativesedation

therapyusingtheDelphiMethod.JPalliatMed2005;8(4):716-29. 20. HawkerS,PayneS,KerrC,HardeyM,PowellJ.AppraisingtheEvidence:ReviewingDisparateData

Systematically.QualHealthRes2002;12:1284-99. 21. Payne S, Kerr C, Hawker S, Hardey M, Powell J. The communication of information about older

peoplebetweenhealthandsocialcarepractitioners.AgeAgeing2002;31:107-17. 22. Alonso-BabarroA,Varela-CerdeiraM,Torres-VigilI,Rodriguez-BarrientosR,BrueraE.At-home

palliativesedationforend-of-lifecancerpatients.PalliatMed2010;24(5):486-92.

Chapter 242

23. ChambaereK,BilsenJ,CohenJ,etal.ContinuousdeepsedationuntildeathinBelgium:anationwidesurvey.ArchInternMed2010;170(5):490-93.

24. ClaessensP,MentenJ,SchotsmansP,BroeckaertB,onbehalfofthePalsedC.PalliativeSedation,NotSlowEuthanasia:AProspective,LongitudinalStudyofSedationinFlemishPalliativeCareUnits.JPainSymptomManage2010;41(1):14-24.

25. deGraeffA,JobseAP,VerhagenEH,MoonenAA.TheroleofconsultationinpalliativesedationinthecentralregionoftheNetherlands[inDutch].NedTijdschrGeneeskd2008;152(43):2346-50.

26. Eckerdal G, Birr A, Lundstrom S. Palliative sedation uncommon in specialized palliative care [inSweden].Lakartidningen2009;106(15-16):1086-8.

27. FordeR,AaslandOG,FalkumE,BreivikH,KaasaS.Palliativesedationtodyingpatients[inNorway]TidsskrNorLaegeforen2001;121(9):1085-8.

28. HasselaarJG,ReuzelRP,vandenMuijsenberghME,etal.Dealingwithdelicateissuesincontinuousdeepsedation.VaryingpracticesamongDutchmedicalspecialists,generalpractitioners,andnursinghomephysicians.ArchInternMed2008;168(5):537-43.

29. MarinN,KesselH,RodriguezM,etal.Terminalsedation:Thelastresorttoavoidabaddeath.RevEspGeriatrGerontol2003;38(1):3-9.

30. MoritaT.Palliativesedationtorelievepsycho-existentialsufferingofterminallyillcancerpatients.JPainSymptomManage2004;28(5):445-50.

31. Morita T. Differences in physician-reported practice in palliative sedation therapy. Support CareCancer2004Aug;12(8):584-92.

32. MoritaT,ChinoneY,IkenagaM,etal.Ethicalvalidityofpalliativesedationtherapy:amulticenter,prospective,observationalstudyconductedonspecializedpalliativecareunitsinJapan.JPainSymp-tomManage2005;30(4):308-19.

33. ParkerMH,CartwrightCM,WilliamsGM.ImpactofspecialtyonattitudesofAustralianmedicalpractitionerstoend-of-lifedecisions.MedJAust2008;188(8):450-6.

34. PomerantzSC,BhattH,BrodskyNL,etal.Physicians’practicesrelatedtotheuseofterminalseda-tion:moralandethicalconcerns.PalliatSupportCare2004;2(1):15-21.

35. PorzioG,AielliF,VernaL,etal.Efficacyandsafetyofdeep,continuouspalliativesedationathome:aretrospective,single-institutionstudy.SupportCareCancer2009;18:77-81.

36. RietjensJA,vanDeldenJJ,vanderHeideA,etal.Terminalsedationandeuthanasia:acomparisonofclinicalpractices.ArchInternMed2006;166(7):749-53.

37. Rietjens JA,vanderHeideA,VrakkingAM,etal.Physicianreportsof terminalsedationwithouthydrationornutritionforpatientsnearingdeathintheNetherlands.AnnInternMed2004;141(3):178-85.

38. Rietjens JA, van Zuylen L, van Veluw H, et al. Palliative sedation in a specialized unit for acutepalliativecareinacancerhospital:comparingpatientsdyingwithandwithoutpalliativesedation.JPainSymptomManage2008;36(3):228-34.

39. vanDoorenS,vanVeluwHT,vanZuylenL,etal.Explorationofconcernsofrelativesduringcontinu-ouspalliative sedationof their familymemberswithcancer. JPainSymptomManage2009;38(3):452-9.

40. Van Deijck RH, Krijnsen PJ, Hasselaar JG, et al. The practice of continuous palliative sedation inelderlypatients:anationwideexplorativestudyamongDutchnursinghomephysicians.JAmGeriatrSoc2010;58(9):1671-8.

41. MercadanteS,IntravaiaG,VillariP,etal.ControlledSedationforRefractorySymptomsinDyingPatients.JPainSymptomManage2009;37(5):771-79.


42. MoritaT,IkenagaM,AdachiI,etal.Familyexperiencewithpalliativesedationtherapyforterminallyillcancerpatients.JPainSymptomManage2004;28(6):557-65.

43. MoritaT,MiyashitaM,KimuraR,AdachiI,ShimaY.Emotionalburdenofnursesinpalliativeseda-tiontherapy.PalliatMed2004;18(6):550-7.

44. SwartSJ,BrinkkemperT,RietjensJA,etal.Physicians’andnurses’experienceswithcontinuouspal-liativesedationintheNetherlands.ArchInternMed2010;170(14):1271-4.

45. BlondeauD,DumontS,RoyL,MartineauI.AttitudesofQuebecdoctorstowardssedationattheendoflife.Anexplanatorystudy.PalliatSupportCare2009;7:331-7.

46. BlondeauD,RoyL,DumontS,GodinG,MartineauI.Physicians’andpharmacists’attitudestowardtheuseofsedationattheendoflife:Influenceofprognosisandtypeofsuffering.JPalliatCare2005;21(4):238-45.

47. Chater S, Viola R, Paterson J, Jarvis V. Sedation for intractable distress in the dying- a survey ofexperts.PalliatMed1998;12(4):255-69.

48. VenkeGranS,Miller J.Norwegiannurses’ thoughtsandfeelingsregarding theethicsofpalliativesedation.IntJPalliatNurs2008;14(11):532-8.

49. Morita T, Tsunoda J, Inoue S, Chihara S. The decison-making process in sedation for symptomcontrolinJapan.PalliatMed1999;13:262-4.

50. BrajtmanS.Theimpactonthefamilyofterminalrestlessnessanditsmanagement.PalliatMed2002;17:454-60.

51. MoritaT, IkenagaM,Adachi I, et al.Concernsof familymembersofpatients receivingpalliativesedationtherapy.SupportCareCancer2004;12(12):885-9.

52. FordeR,KongsgaardU,AaslandOG.Palliativesedationtothedying[inNorwegian].TidsskrNorLaegeforen2006;126(4):471-4.

53. VandenBlockL,DeschepperR,BilsenJ,etal.Euthanasiaandotherendoflifedecisionsandcareprovidedinfinalthreemonthsoflife:nationwideretrospectivestudyinBelgium.BMJ2009;339:1-8.

54. Maessen M, Veldink JH, Onwuteaka-Philipsen BD, et al. Trends and determinants of end-of-lifepracticesinALSintheNetherlands.Neurology2009;73(12):954-61.

55. MiyashitaM,MoritaT,SatoK,etal.Factorscontributing toevaluationofagooddeath fromthebereavedfamilymember’sperspective.Psycho-Oncology2008;17(6):612-20.

56. RietjensJA,BuitingHM,PasmanHR,etal.Decidingaboutcontinuousdeepsedation:physicians’perspectives:afocusgroupstudy.PalliatMed2009;23(5):410-7.

57. SeymourJE,JanssensR,BroeckaertB.Relievingsufferingattheendoflife:practitioners’perspec-tivesonpalliativesedationfromthreeEuropeancountries.SocSciMed2007;64(8):1679-91.

58. WorldHealthOrganization.Definitionofpalliativecare.2008.Availablefrom:http://www.who.int/cancer/palliative/definition/en.AccessedFebruary22,2011.

59. TenoJM,ClarridgeBR,CaseyV,etal.Familyperspectivesonend-of-lifecareatthelastplaceofcare.JAMA2004;291(1):88-93.

60. AndershedB.Relatives inend-of-lifecare-part1:asystematicreviewof the literaturethefive lastyears,January1999-February2004.JClinNurs.2006;15(9):1158-69.

Chapter 3Palliative sedation: a focus group study on the experiences of relatives

Bruinsma S, Rietjens J, van der Heide A

Journal of Palliative Medicine 2013; 16 (4): 349-55

Chapter 346

aBstract

Background. Most studies that have investigated the practice of palliative sedationhavefocusedonphysicians’practicesandattitudes.

Objective. Theaimof this studywas toexplorerelatives’experienceswithpalliativesedationandtogainmoreinsightinpositiveandnegativeelementsintheirevaluationofpalliativesedation.

Design.Focusgroupsandindividualinterviews.

Setting.VariouscaresettingsintheNetherlands.

Subjects.Atotalof14relativesofpatientswhoreceivedpalliativesedationuntildeathparticipated.

Results. Most relatives evaluated the provision of palliative sedation of their dyingfamilymemberpositively.Positiveexperienceswererelated to: thebeneficial impactof palliative sedation on the patient’s suffering, the opportunity that was offered toprepareforthepatient’sdeath,theirinvolvementinthedecision-makingandcareforthepatient,andthepleasantcareenvironment.However,themajorityoftherelativeswereunsatisfiedwithoneormoreaspectsofhowinformationwasbeingprovidedfor.Somerelativeswerefrustratedaboutthefactthatnurseswerenotauthorizedtomakedecisions about the care for the patient and about the absence of physicians duringweekends.Noneoftherelativesmentionedthelossoftheabilitytocommunicatewiththepatientduringthesedationandthepossibilityof‘‘hasteningdeath’’asdisadvantagesofpalliativesedation.

Conclusion. Relatives tend to evaluate the provision of palliative sedation to theirseverely suffering family member positively because it contributes to a peaceful dy-ingprocess.However,relativesindicateddiscontentwithhowinformationwasbeingprovidedandwiththecommunicationingeneral.

Focus group study on relatives’ experiences with PS 47

introDUction

At theendof life,manypatients suffer fromsevere symptoms.Sometimes, sedationuntildeathisneededtocontrolsymptomsthatcannotberelievedwithconventionalmeasures.Thispracticeisoftenreferredtoaspalliativesedation(1).Thedegreeofseda-tionnecessarytorelievesufferingmayvaryfromsuperficialtodeep.DutchnationwidestudiesshowedthatpalliativesedationisincreasinglyusedinTheNetherlandsinupto8.2%ofalldeathsin2005(2,3).Palliativesedationisusedinallsettingsinwhichpatients die, but most often in hospitals and for patients with cancer (4-8). Withinpalliativecaresettings,estimatesaboutthefrequencyofpalliativesedationvaryfrom15%tomorethan60%ofalldeceasedpatients,dependingonthetypeofsettingandthedefinitionsused(9).

Beingacloserelativeofsomeonewhoisinthefinalphaseoflifeisoftencomplicated.Relativesmusthandleboththeirownsorrowandthatofthedyingperson,inadditiontoaddressingamultitudeofpractical issues(10).Professionalsworking inpalliativecarestress the importanceofgoodcarefortherelativesof thepatient(11).ThiscanalsobeseenthroughtheWorldHealthOrganization’sdefinitionofpalliativecare,inwhichprovidingasupportsystemtohelprelativescopeduringthepatient’sillnessandduringtheirowngriefisincorporated(12).Theimportanceofrelatives’well-beingisalsoreflectedinguidelinesaboutpalliativesedation,whichstressthatrelativesshouldbe involved in the decision-making process, that they can assist in monitoring andcaringforthepatient,andthattheyshouldbeclearlyinformedandsupported(13-15).

Moststudiesthathaveinvestigatedthepracticeofpalliativesedationhavefocusedonphysicians’practicesandattitudes(16-23).Theaimofthisqualitativestudywastoexplore relatives’ experiences with palliative sedation and to gain more insight intopositiveandnegativeelementsintheirevaluationofpalliativesedation.

methoDs

Aspartofalargerstudyaboutthepracticeofpalliativesedationanditsroleinend-of-life care, the UNBIASED study’ (UK–Netherlands–Belgium InternAtional SEDationstudy),weheldfocusgroupsandinterviewswithrelativesofpatientswhoreceivedpal-liativesedationuntildeath.Becauseoftheexploratorynatureofthestudy,focusgroupsweredeemedmostsuitable.Inthefocusgroups,thegroupprocesseswereparticularlyuseful in clarifying relatives’ views on palliative sedation (24). Interviews were heldwhenrelativeswerenotabletoattendafocusgroup.Thetermpalliativesedationwassystematicallyusedthroughoutthestudy,sincethisisthemostcommonlyusedterm

Chapter 348

in guidelines and research papers (1, 13-15).We followed the criteria for qualitativeresearchtoensurerigourinourresearch(25).

recruitment of participants

Becausepalliativesedationisusedinallsettingsinwhichpatientsdie,wefocusedonavarietyofsettings(23).Purposivesamplingwasusedfortheselectionofparticipantstomakesuredifferentsettingswererepresented,andtoachievevariationinrelatives’experiencesandintypesofsedationused.Inclusioncriteriaofthepatientswerethattheywereolder than18andthat theirdeathhadbeenamaximumof1.5yearsago.Inclusion criteria of relatives of patients who had received palliative sedation untildeathwere:frequentcontactwiththepatientinhis/herlastphaseoflife,bothpriortoandduringthesedation,andbeingDutch-speakingandresidentsofTheNetherlands.Relatives were not necessarily restricted to family members, but could also includefriends.Weincludedonerelativeperpatient.Relativeswerecontactednosoonerthan3monthsandnolaterthan12monthsafterthepatient’sdeath.

Relativeswererecruitedusingbothprofessionalandpersonalcontactsofmembersoftheresearchteam.Theprofessionalcontactsconcernedbothnursesandphysiciansworkinginthehome,hospital,orhospicesetting.Theywereaskedtoidentifythemostinvolvedrelativeofapatientwhohadreceivedpalliativesedationandtocontacttherelativebysendinganinformationletter fromtheresearchteam.Inthehospiceandhospitalsetting,25caseswereidentified.Therelativesofthese25caseswereapproachedbythehealthcareprofessionaland8relativesagreedtoparticipate.Tworelativesofallrelatives who have been approached by a general practitioner agreed to participate.It isunknownhowmanyrelativeswereapproachedbygeneralpractitioners intotal.Furthermore, we identified cases through our personal contacts. Five relatives wereapproachedand4agreedtoparticipate.

Tenrelativeswerewillingandabletoparticipateinafocusgroup(twofocusgroupswith threeparticipantsandone focusgroupwith fourparticipants).Fouradditionalinterviewswereheldwithrelativeswhowerewillingtoparticipate,butwerenotabletoattendafocusgroupbecauseofpracticalreasons.ThefocusgroupsandinterviewstookplacebetweenOctober2010andMarch2011.

Procedures

A semistructured questionnaire was developed, based on the recommendationsdescribed in theDutchnationwideguideline forpalliativesedationandasystematicreviewontheexperiencesofrelativeswithpalliativesedation,toensureconsistencyintopicsaddressedacrossgroups(26).Thequestionnairewaspilotedinthreeinterviews.Thisledtosomesmallchangesintheformulationofthequestions.Thequestionnairecoveredseveraltopics:backgroundinformationofpatientandrelatives,thedecision-


making process, information and communication, the process of sedation, and theoverallevaluationoftherelativesoftheuseofsedation.

Allfocusgroupandinterviewparticipantsfilledinashortquestionnaireonsocio-demographicdataandsignedaconsentformbeforethestartofthefocusgroups.Thefocusgroupswereledbyexperiencedmoderators(A.H.,J.R.)andlastedapproximately2hours.Anassistantmoderator(S.B.)waspresenttotakenotes.TheinterviewswereconductedbyS.B.andlastedapproximately60minutes.Participantsconsentedtothediscussionorinterviewbeingaudiotaped.TheErasmusMCMedicalEthicalResearchCommitteeapprovedthestudy.TheprocedurefollowedtheHelsinkiDeclaration.

Data analysis

The audiotaped discussions were transcribed verbatim and analyzed by constantcomparative analysis. We removed names and other information that could lead toidentification.Aftercompletingall focusgroupsand interviews,we(S.B., J.R.,A.H.)read through all transcripts. Data were analyzed by S.B. using the themes from thequestionnaire.Attentionwaspaid to issues thatwerenot explicitly addressed in thequestionnaire.Asanextstep,theresultswerediscussedwithJ.R.andA.H.untilcon-sensuswasreached.Aprofessionaltranslatortranslatedthequotesthatarepresentedintheresults.Eachquoteisfollowedbyinformationonthegenderoftherespondent(F=female,M=male),thenatureoftherelationshipwiththepatient,andthesettingin which the patient died. When there is more than one respondent with the samecharacteristics, thegenderof the respondent is followedbyanumber todistinguishthedifferentrespondents.

resUlts

characteristics patients and relatives

Thecharacteristicsofrelativesandpatientsarepresentedintables1and2.

Decision-making process

Differentissuescameupwhilediscussingthedecision-makingprocess.

Reason(s) for the use of sedationRelativesbelieved that thepatient’s sufferingwas themost importantreason tostartpalliativesedation.Pain,respiratoryproblems,fatigue,restlessness,anxiety,confusion,delirium, inability to drink and eat, and nausea were mentioned as decisive factors.Furthermore,althoughrelativesbelievedthatpatients’sufferingwasthemainreason

Chapter 350

tostartsedation,theirdescriptionsof‘‘suffering’’alsoincludedtheirownsufferingasaresultofthepatients’suffering.

‘My two children and I said, we are going to propose this [sedation], it couldn’t go on like this. And it was the same for him - at least we think it was - but of course you can’t ever really know that. But it was so awful for us to be there: to see how he was fighting to breathe, oh and his hallucinations… it is so hard just to stand by and watch. You don’t know what is going on inside his head’ [F1,partner,home].

table 1 Characteristics relatives

Gender Male 6

Female 8

Age 41-69

Nature of relationship with patient Partner 8

Child 3

Grandchild 1

Daughter in law 1

Parent 1

Time passed since patient passed away 2-16 months

table 2 Characteristics patients

Sex Male 8

Female 6

Age 33-89

Primary diagnose Cancera 12

Alzheimer/dementia 1

No specific diagnoseb 1

Setting Home 7

Institutionc 7

Duration illness 4 weeks-2,5 years

Time spent in setting were patient passed away

4 days-3 months

a Skin cancer (two patients), colon cancer (two patients), lung cancer (two patients), peritoneum cancer, breast cancer, cervical cancer, oesophagus cancer, brain tumour, liver cancer, pleural cancer

b Advanced agec Hospice (4), nursing home (1), psychiatric/geriatric institution (1), hospital (1).


Making the decisionThedecision to start sedationwaseither initiatedby thepatient, the relative,or thephysician.Sometimes itwas thepatientwhospecificallyaskedforsedation.Inthesecases,patientandrelativeshadpreviouslydiscussed theoptionof sedationwith thephysician.Intheend,thephysicianmadethefinaldecisiontostartsedationandtherelativesgavetheirconsent.

‘So the general practitioner explained it to him. Yes (...) that it was possible just to be put to sleep and to let the body itself take over. And he (the patient) said that he would prefer palliative sedation. He really wanted to sleep through the very last bit. He consciously wanted to say goodbye and then go to sleep. He made the choice himself, and we had talked about it a few times. (...) I supported his decision, well rationally I did, but emotionally you would rather it had gone dif-ferently’[F2,partner,home].

Inacoupleofcases,relativesaskedforthesedation.Again,thephysicianmadethefinaldecisiontostartsedation.Therelativeswerepositiveaboutphysician’sinvolvement.

‘Yes it was at our own request. (…) We talked to him (the patient) about sedation, well we talked about it a bit but he didn’t express any particular opinion about it. But he did say that he didn’t want to have to fight for breath, or be in any pain – that sort of thing. (…) He was so thirsty all the time and he was given morphine for it. But then he began to hallucinate and became very restless, and then we got to the point that we said…. (…) to the doctor, I said “well it’s time to give him some strong sedating medication’[F1,partner,home].

Themajorityoftherelativesstated,however,thatthephysicianwastheonewhopro-posedtheuseofsedation.Thisproposalwasinmostcasesprecededbyadiscussionbetweenphysicianandrelativesabouttheneedtorelievethepatients’suffering.

‘The night before he died he was given injections to calm him but they didn’t help at all. (…) .The doctor said that it was inhuman to go on like this. (…) The doctor discussed this with me and the boys. (…) And then the next day they brought a pump…. and then he went to sleep and he didn’t wake up again’ [F3,partner,home].

A few relatives mentioned, however, that the physician made the decision withoutinvolvingthematall.Thereasontostart thesedationwasnotalwaysclear fortheserelatives.

Chapter 352

‘We assumed that if sedation was going to be given, then it would be the patient that made the decision. That the patient would say that he couldn’t stand to go on like this (…). But in reality….we left on Sunday morning, change of shift, and we came back in the evening and we heard that she had been given an injection that had knocked her out and you think, well, you think what’s going on? What on earth has happened? And that’s…, well you think “this just isn’t right”. (…) So it meant my mother wasn’t in pain. And that’s..., but I still don’t understand. Why did it have to happen like that?’[F,child,hospice]

‘And then the telephone call came that the bodily functions were beginning to shut down (…) Then of course you ask “Well, what now?” Is he in pain? No he is not in pain. And, well he is slowly dying. I don’t know what they did to him, no idea. The only thing I am sure of is that the last 2or 3 days, that the well-known ‘pain pump’ was attached. And then you assume that sedation has in fact begun’ [F,child,nursinghome].

Negative emotions regarding the decision-making processSome respondents mentioned their frustration regarding the fact that nurses werenot authorized to make decisions about the care for the patient, whereas they weremostofthetimestheonlyonesthatwereavailable.Anotherpointofcritiquethatwasmentionedwasthatthephysicianwasnotavailableduringtheweekend.

‘So I always had a problem with the fact that medical care is from 9:00 am to 5:00 pm and not during/in the weekend. So I thought ‘‘Here we go again, its the weekend again.’’ And you have to fight so hard to get care; you have to knock on so many doors. So I said to the doctor, I said ‘‘Listen, if you start that and then go home, I want it started properly because nurses aren’t allowed to make decisions on their own’[F,grandchild,hospice].

information and communication

Adequate informationandcommunicationbetween thepatient, relatives,physiciansandotherprofessionalcaregiverswereconsideredasveryimportantbytherelatives.Inallcasesbutone,therelativesmentionedthatprofessionalcaregivershaddiscussedend-of-lifeissueswiththemand(whenpossible)withthepatient.Theseconversationstookplaceatdifferentphasesinthediseaseprocess.

‘We had already talked about euthanasia at an earlier stage, and about pallia-tive sedation—with the general practitioner. He (the patient) also had a written


declaration of intent regarding euthanasia and a do-not-resuscitate statement. We got all that sorted out. Also because we had no idea what was going to happen’[F2,partner,home].

‘Two days after we knew he was sick we went to the general practitioner and we discussed the possibilities (..). And we discussed sedation. And he (the patient) made clear he wanted that’[F4,partner,home].

Nine relatives were, at least at one point in time, dissatisfied with the amount ofinformation they had received. Some of these relatives expressed a lot of criticism,whileothersonlybrieflymentionedthis.Relatives’werediscontentwith(insufficient)informationabout:sedationingeneral,theearlystartofthesedation,thedrugsthatwereused, further treatment (e.g., artificialnutritionandhydration), thewell-beingofthepatientduringthesedation,theexpecteddurationofthesedation,thetimeofdeath,and/orthepossiblesymptomsorreactionsofthepatientduringthesedation.

‘And (they didn’t) tell us what sort of symptoms to expect and what the patient’s reaction could be. There you are then (sighs). All on your own, you have to go through it all on your own’[F5,partner,home].

‘No. Everything was lumped under one heading—pain relief. The words euthanasia or palliation—I can hardly bring myself to say them—they were never mentioned’[M,child,psychiatric/geriatricinstitution].

‘Yes, what’s it really about? Yes...communication....of course. They didn’t say ‘‘We are going to start morphine now and we are going to add some midazolam and then you can expect..’’ because they just don’t think that way…it’s not what they do’[M,child,psychiatric/geriatricinstitution].

Interviewer: ‘How do you think things could have been done better?’ Respondent: ‘Yes - communication. Clearly say what is going to happen, say what you are going to do and what the consequences will be - not only for the person it is being given to, but also for the people around them - say what can happen’ [F,child,nursinghome].

Sometimesthecontentoftheinformationwasunclear.

‘I mean they did not tell us what it practically means. (…) OK. My daughter-in-law knew about these things, yes but that wasn’t really – maybe that is why it was

Chapter 354

not explained, because they automatically assumed that we would know too. And we didn’t ask, didn’t ask what is going to happen because … well, again all the time we were only thinking how much longer does this have to go on. Please let it be over as soon as possible’[M,parent,hospital].

Sometimesrelativesreceivedconflictinginformationfromdifferentprofessionalcare-givers.

‘They weren’t a team. They were all working individually and sometimes against each other. It just makes you so uncertain about everything (…). Sometimes we had to make the decisions about what to do. (…) And I was confronted with more than one way of working; and that shouldn’t be allowed’ [F5,partner,home].

Somerelativeshadthefeelingthattheywerenotbeingtakenseriouslybythephysi-cians.

‘As a family, we just weren’t taken seriously. Information about him and the care and so on, it really wasn’t taken seriously. And I think it is a real shame, and I also think that it is wrong. After all who knows the person better than those who are with him all the time? The nurses took us seriously, but the others didn’t’ [F5,partner,home].

euthanasia

Althoughtheissueofeuthanasiawasnotexplicitlyaddressedinthefocusgroupsorinterviews, almost all respondents referred to it in some way. Some relatives distin-guishedsedationandeuthanasiaasclearlydifferentoptions,whereasothersreferredtosedationas‘‘sloweuthanasia.’’

‘Yes, if you stop that treatment, then they wake up again. In that respect it has nothing to do with euthanasia’ [F,daughter-in-law,home].

‘Actually I think... to me palliative sedation is also a form of euthanasia. You send someone to sleep, and then instead of it all being over in five minutes like it would be with euthanasia, it takes 2 days’ [M,partner,hospice].

Some patients were admitted to a care setting that had a no-euthanasia policy. Theattending physicians had been clear about this and in most cases, the patients andrelativesacceptedoragreedwiththispolicy.


‘I am still a supporter of euthanasia. (…) But we had taken this into consideration (...) and in a very pleasant care environment, then you accept palliative sedation because, well, I’ll put it this way, it is still a humane - you know what I mean - solution’ [M,partner,hospice].

Somerelativesmentionedthatthepatienthadexpressedawishforeuthanasiaandthatthephysicianhadofferedsedationasanalternative.

‘Actually, she (the patient) wanted active euthanasia (…). And the general prac-titioner was willing to perform euthanasia. But then he came with the option to put her to sleep (…) He explained that performing active euthanasia is very burdensome. And she didn’t want that for us, and the children. And then she decided she wanted to be put to sleep’[M1,partner,home].

relatives’ involvement in the process of sedation

The involvement of relatives in the sedation process varied. All relatives visited orstayed with the patient during the sedation. Most relatives provided the physiciansandnursesatsomepointwithinformationaboutthewell-beingofthepatient.Somerelativesprovided(mostof)thecareforthepatientprocessvaried.Allrelativesvisitedorstayedwiththepatientduringthesedation.Mostrelativesprovidedthephysiciansandnursesatsomepointwithinformationaboutthewellbeingof thepatient.Somerelativesprovided(mostof)thecareforthepatientthemselves(e.g.theadministrationofthemedication).

‘We were with her a lot, also when she was receiving professional care. So there came a point when although the carers didn’t see it, we did and we said “She’s in pain’ [M,child,psychiatric/geriatricinstitution].

‘The pump was connected on Thursday and on Wednesday he passed away. All this time at home, we provided all the care’ [F5,partner,home].

overall evaluation of the provision of palliative sedation

Eleven relatives evaluated the provision of sedation to their dying family memberoverallaspositive.Thestartofthesedationwasmostofthetimesseenasarelief,bothforthepatientandfortherelativesthemselves.Duringthesedationthepatientseemedcomfortable.Accordingtomostrelatives,sedationhadbeenthebestsolution.

‘But you have to—otherwise something like this is just unbearable for him. And that would be terrible. We, the people around him, couldn’t have coped otherwise

Chapter 356

because he didn’t know what he was doing any more and he needed to relax. He was so wound up, he kept trying to get out of bed and that just wasn’t possible. That’s not good for you and certainly not for the patient. And this is a very nice way of doing it. Yes, it is’ [F5,partner,home].

‘I think it is a good process in that you actually allow your body to go its own way, that you are no longer interfering with the natural process. And it means that the last battle can be fought unconsciously, that you no longer have to experience it actively. I can see that from the patient’s point of view it can be a relief, but also for us’ [F2,partner,home].

Overall,therelativeswerepositiveaboutprofessionalcaregivers.

‘I am full of praise for them. Both the doctor and the nurse, they did everything, you just had to ask and they did everything in their power’[M,parent,hospital].

Theplaceofdeathwas inall cases inagreementwith thewishesof thepatient.Therelativesweregladtheycouldfulfilthewishofthepatient.

‘Well, then she expressed the wish to die here at home. And that is what I did... I helped. (…) A hospice would have been easier for me of course. But that is not the point, it’s what she wanted’[M2,partner,home].

Somerelativesalsomentionedtheirsatisfactionwithandtheimportanceofthecareenvironment.

‘We are really full of praise about the way it went, (…) and really I mean that. (…) We were very lucky that he was put into a room that was quite large. (…) There were fold-up beds so we could be there day and night. (…) The experience of wanting to and being able to do that was really marvellous. They also provided us with food like it was the most normal thing. I mean, it was really fantastic. (…) They gave us a room where we could go and talk privately. Oh - it was... yes, really. I keep saying it but I can‘t praise them enough’[M,parent,hospital].

Therespondentsknewthatdeathwasnear.Thesedationprovidedanopportunitytoprepare,tosomeextent,forthelossoftheirlovedone.


‘The advantage of something like this, I mean euthanasia - and this as well - is that you know roughly when it is going to happen and you can say goodbye to each other in a conscious way, it is a good way to part’[M1,partner,home].

Afewrelativeshadamorenegativeexperiencewiththeprovisionofsedation.Some-timesthishadtodowithsymptomsorreactionsofthepatientduringthesedation.Therelativeswereconcernedaboutthepatient’swellbeing.

‘The time finally came to start sedation, with a pump. That meant that we knew the time would come (…) when she would no longer regain consciousness. And then the moment came when the last words had been said. We had embraced each other for the last time and were waiting for her to fall asleep. (….). For me that was confirmation that I knew that once she went under, she would indeed go peacefully. But, looking back she woke up twice, oh what a situation was that. When she had to go back to sleep it didn’t work. .(….). She was in so much pain’[F,child,hospice].

Thedurationofthesedationvariedfromfivehourstooneweek.Thedurationofthesedationhadconsequencesfortherelatives’wellbeing.Thelongerthesedationprocess,thehighertheburdenfortherelatives.

‘On Friday she was started on morphine (…), and you just mentioned Dormicum, I think that she received that too at some point (....), and she died on Wednesday. We thought that this was really a very long time to go on - and a time that has had a really big impact. I don’t use the word “traumatic” lightly, but the experience of going through the last phase from say, Friday to Wednesday – it’s a very long time. It was a very difficult time’[M,child,psychiatric/geriatricinstitution].

DiscUssion

Informationabouttheconcernsandneedsofrelativesofpatientswhoreceivepallia-tivesedationcanassisthealthcareprofessionalsinprovidinginformationandsupporttorelativesduringthisdifficult time in their lives(11).Theaimof thisstudywas toexplorehowrelativesevaluatepalliativesedationandtogainmoreinsightintopositiveandnegativeelementsoftheirexperiences.Wefoundthatmanyrelativeshadpositiveexperiences with the provision of sedation for their dying family member. The startofthesedationwasarelief forrelativesbecausethepatient’ssufferingwasfinallyal-leviated.Relativesoftenappreciatedhavinganactiveroleindecision-makingprocess

Chapter 358

andtheprovisionofthesedation.Otherpositiveexperiencesrelatedtothedegreeofinvolvementofprofessionalcaregivers,theappropriatenessoftheplaceofdeath,andthe care environment. On the other hand, several relatives indicated that they weredissatisfiedwiththeinformationtheyreceivedandaboutcommunicationingeneral.Othernegativeexperienceswererelatedtoconcernsaboutthewellbeingofthepatientduringsedation,especiallywhenthesedationprocesslastedlong,thelackofauthorityofnursestomakedecisions,andtheabsenceofphysiciansduringtheweekends.

Afewfindingsdeserveparticularattention.First,althoughrelativesbelievedthatpatients’sufferingwasthemainreasontostartsedation,theirdescriptionsof‘‘suffering’’alsoincludedtheirownsuffering.Accordingtotheguideline,theindicationconsistsofoneormorerefractorysymptomsexperiencedbythepatient(13).Althoughinpallia-tivecarethepatientandtherelativestogetherare‘‘theunitofcare,’’relatives’sufferingisnotmentionedasan indication for sedation.This issueneeds furtherattention inresearch.

Second,itwasstrikingthatthemajorityofrelativesinourstudyexpresseddissatis-factionwithoneormoreaspectsofcommunicationandtheprovisionofinformation.For instance, the physician sometimes made the decision to start sedation withoutinvolvingrelatives,althoughtheguidelineonpalliativesedationrecommendsphysi-cianstoactivelyinvolverelativesinthedecision-making,despitethefactthattheybearfinal responsibility for the decision (13). Many studies about end-of-life care reportthattheprovisionofinformationandcommunicationareinadequateandinsufficient(10).AJapanesestudyonrelatives´experienceswithpalliativesedationshowedthatinsufficientinformationcanleadtolowlevelsofsatisfactionandhighlevelsofdistress(27).Ourresultssupportthefindingsfrompreviousstudies,i.e.,thatrelativesexperi-encedistressduetoalackofinformationorunclearinformation.Suchdistressmaybediminishedbydiscussingthepatient’ssituationopenly;providingfullinformation,e.g.,aboutthemomentofthestartofsedation,thedrugsused,and/orpossiblesymptomsorreactionsofthepatientduringsedation.Furthermore,propercommunicationbetweenthedifferentcaregiverscanenhanceclarityofthesituationfortherelatives.

Although themajorityof relatives seemed toexperiencedistressdue toa lackofinformation or unclear information, they generally evaluated the sedation of theirlovedonespositively.Suchpositiveevaluationofpalliativesedationwasmostlyrelatedtopatient’ssymptomrelief.Beforethestartofthesedation,patientspresentedmultipledistressingphysicalsymptoms,suchasseverepain,fatigue,andrestlessness.Duringthesedationmostpatientsseemedcomfortable.Relativesindicatedthattheyappreciatedthefactthatthepatientcoulddie‘‘naturally,’’withoutburdeningmedicalinterventions.Previousstudieshavealsoshownthatadequatesymptomreliefiskeytotheexperienceofa‘‘gooddeath’’(28-31).Furthermore,palliativesedationmakesrelativesthemselves


feelmorecomfortablebecauseitallowsthemasenseofpeaceandclosureafteradif-ficultperiodofsuffering.9

Additionally,severalrelativesmentionedthattheuseofpalliativesedationprovidedthem with the time necessary to prepare for the loss of their loved one. They knewthat death was near and could say goodbye to the patient in a conscious way. Noneoftherelativesinourstudymentionedlossoftheabilitytocommunicateduringthesedationwith thepatientasadisadvantageofpalliative sedation.This is incontrastwith the frequencywithwhich lossof communication isdiscussedas aproblematicaspect of palliative sedation in the literature (27, 32-33). Findings from the currentstudysuggestthatthebenefitsofpalliativesedationoutweighedthelossoftheabilitytocommunicatewiththepatientbeforedeath.

Almostallrelativestalkedabouttheissueofeuthanasia.Somerelativesconsideredpalliativesedationas‘‘sloweuthanasia,’’othersasaterminalinterventionthatissharplydistinguished fromeuthanasia.Theethicaldebateaboutpalliative sedationoften fo-cusesonitspresumedpotentialtohastendeath(23).However,noneoftherelativesinourstudymentionedthepossibilityof‘‘hasteningdeath’’asadisadvantageofpalliativesedation.Apossibleresemblanceofpalliativesedationtoeuthanasiaseemstobenoissue for these relatives, which may be related to the public and legal acceptance ofeuthanasiainTheNetherlands.

Bothourfocusgroupsandpersonalinterviewsappearedtobesuccessfulmethodsforexploringrelatives’experienceswithpalliativesedation.Therelativesappreciatedtheopportunitytosharetheirstories,andfeltthatithelpedthemtodealwiththeirgrief.Nevertheless,ourstudyhassomelimitations.Thenumberofrespondentswassmall.Therefore,thestudycanonlybeseenasexplorativeandnotasconclusive.However,thenumberofrespondentsincludedwasappropriateforansweringtheresearchquestion.Second,because theanonymityof therespondentshad tobeguaranteed, theresultscouldnotbespecifiedbysetting.Third,tobesensitivetotheearlyphasesofgrieving,relativeswerecontactednosoonerthan3monthsafterthedeathof thepatient.Thedatacouldbeinfluencedbyrecallbias.Furthermore,afocusgroupusuallyconsistsofsix toeightparticipants.Our focusgroups included threeor four relatives.Possibly,thisdiminishedthedynamicinterchangebetweenthegroupmembers.Becauseofthesensitivityoftheissuediscussed,wechosetousesmallerfocusgroups,whichworkedwell. Finally, sedation occurs in all settings in which patients die but most often inhospitals(15).Unfortunately,wewerenotabletorecruitmorerelativesofpatientswhoreceivedsedationuntildeathinahospital.

Inconclusion,therelativesinvolvedinthisstudyseemtoagreethatpalliativeseda-tion is abeneficial intervention forpatientswhoare severely sufferingat theendoftheirlives.Itrelievespatient’ssufferingandprovidesrelativeswithasenseofpeaceandclosureafteradifficultperiod.Also,communicationwithrelativesseemsachallenge

Chapter 360

forphysiciansandnursesinvolvedinthispractice.Thelossoftheabilitytocommuni-catewiththepatientduringthesedationandthepossibilityof‘‘hasteningdeath’’donotseemtobeissuesfortherelatives.

acknowleDGements

Thisfocusgroupstudy ispartofa largerstudyaboutpalliativesedationand itsroleinend-of-lifecare:TheUNBIASEDstudy.The‘UNBIASED’study(UK-Netherlands- Belgium International SEDation study) is a collaboration between research teamsintheUK,BelgiumandtheNetherlandswithfundingfromtheEconomicandSocialResearchCouncil(UK),theResearchFoundationFlanders(BE),theFlemishCancerAssociation (BE), the Research Council of Ghent University (BE), the NetherlandsOrganisationforScientificResearch(NL)andtheNetherlandsOrganisationforHealthResearchandDevelopment(NL).Wethanktherelativeswhoparticipatedinthefocusgroups.


references

1. Claessens P, Menten J, Schotsmans P, Broeckaert B. Palliative sedation: A review of the researchliterature.JPainSymptomManage2008;36(3):310-33.

2. VanderHeideA,Onwuteaka-PhilipsenBD,RurupML,BuitingHM,vanDeldenJJM,Hanssen-deWolfJE,etal.End-of-lifepracticesintheNetherlandsundertheEuthanasiaAct.NEnglJMed2007;356:1957-65.

3. Van der Wal G, van der Heide A, Onwuteaka-Philipsen BD, van der Maas PJ. Medical decision-makingat theendof life:practicesandtheeuthanasiareviewprocedure.Utrecht:DeTijdstroom,2003.

4. MiccinesiG,RietjensJA,DeliensL,PaciE,BosshardG,NilstunT,etal.Continuousdeepsedation:physicians’experiencesinsixEuropeancountries.JPainSymptomManage2006;31:122-9.

5. BilsenJ,VanderSticheleR,BroeckaertB,MortierF,DeliensL.Changesinmedicalend-of-lifeprac-ticesduringthelegalizationprocessofeuthanasiainBelgium.SocSciMed2007;65:803-8.

6. RietjensJ,vanDeldenJ,Onwuteaka-PhilipsenB,BuitingH,vanderMaasP,vanderHeideA.Con-tinuousdeepsedationforpatientsnearingdeathintheNetherlands:descriptivestudy.BMJ2008;336:810-3.

7. SealeC.End-of-lifedecisionsintheUKinvolvingmedicalpractitioners.PalliatMed2009;23:198-204.

8. SealeC.Continuousdeepsedationinmedicalpractice:adescriptivestudy.JPainSymptomManage2010;39:44-53.

9. RietjensJA,HauserJ,vanderHeideA,EmanuelL.Havingadifficulttimeleaving:experiencesandattitudesofnurseswithpalliativesedation.PalliatMed2007;21:643-9.

10. AndershedB.Relatives inendof lifecarepart1:a systematicreviewof the literature thefive lastyears,january1999-febraury2004.JClinNurs2006;15:1158-69.

11. VanDoorenS,vanVeluwHT,vanZuylenL,RietjensJA,PasschierJ,vanderRijtCC.Explorationofconcernsofrelativesduringcontinuouspalliativesedationoftheirfamilymemberswithcancer.JPainSymptomManage2009;38:452-9.

12. World Health Organization. ‘Definition of palliative care’, http://www.who.int/cancer/palliative/definition/en(2008,accessedFebruary2010).

13. Royal Dutch Medical Association (RDMA). Guideline for palliative sedation [in Dutch]. 2009.Available from http://knmg.artsennet.nl/Publicaties/KNMGpublicatie/Guideline-for-palliative-sedation-2009.htm.AccessedOctober,2010.

14. ChernyNI,RadbruchL.EuropeanAssociationforPalliativeCare(EAPC)recommendedframeworkfortheuseofsedationinpalliativecare.PalliatMed2009;23:581-93.

15. MoritaT,BitoS,KuriharaY,UchitomiY.DevelopmentofaclinicalguidelineforpalliativesedationtherapyusingtheDelphiMethod.JPalliatMed2005;8:716-29.

16. FainsingerRL,WallerA,BercoviciM,etal.Amulticentreinternationalstudyofsedationforuncon-trolledsymptomsinterminallyillpatients.PalliatMed2000;14:257–65.

17. Muller-BuschHC,AndresI,JehserT.Sedationinpalliativecare–acriticalanalysisof7yearsexperi-ence.BMCPalliatCare2003;2:2.

18. ChiuTY,HuWY,LueBH,ChengSY,ChenCY.SedationforrefractorysymptomsofterminalcancerpatientsinTaiwan.JPainSymptomManage2001;21:467–72.

19. SykesN,ThornsA.Theuseofopioidsandsedativesattheendoflife.LancetOncol2003;4:312-8. 20. SykesN,ThornsA.Sedativeuseinthelastweekoflifeandtheimplicationsforend-of-lifedecision

making.ArchInternMed2003;163:341–4.

Chapter 362

21. StoneP,PhillipsC,SpruytO,WaightC.Acomparisonoftheuseofsedativesinahospitalsupportteamandinahospice.PalliatMed1997;11:140–4.

22. MiccinesiG,RietjensJA,DeliensL,etal.Continuousdeepsedation:physicians’experiencesinsixEuropeancountries.JPainSymptomManage2006;31:122–9.

23. RietjensJA,vanderHeideA,VrakkingAM,Onwuteaka-PhilipsenBD,vanderMaasPJ,vanderWalG.PhysicianreportsofterminalsedationwithouthydrationornutritionforpatientsnearingdeathintheNetherlands.AnnInternMed2004;141:178–85.

24. KitzingerJ.QualitativeResearch:Introducingfocusgroups. BMJ1995;311:299. 25. PopeC,MaysN.QualitativeResearch:Reachingthepartsothermethodscannotreach:anintroduc-

tiontoqualitativemethodsinhealthandhealthservicesresearch.BMJ1995;311:42-5. 26. BruinsmaSM,RietjensJAC,SeymourJE,AnquinetL,vanderHeideA.Theexperiencesofrelatives

with thepracticeofpalliative sedation:A systematic review. JPainSymptomManage2012;44(3):431-45.

27. Morita T, Ikenaga M, Adachi I, Narabayashi I, Kizawa Y, Honke Y, et al. Family experience withpalliativesedationtherapyforterminallyillcancerpatients.JPainSymptomManage2004;28:557-65.

28. RietjensJAC,vanderHeideA,Onwuteaka-PhilipsenBD,vanderMaasPJ,vanderWalG.Prefer-encesoftheDutchgeneralpublicforagooddeathandassociationswithattitudestowardsend-of-lifedecision-making.PalliatMed2006;20:685-92.

29. PiersonCM,RandallCurtisJ,PatrickDL.Agooddeath:aqualitativestudyofpatientswithadvancedAIDS.AIDSCare2002;1144:587-98.

30. Teno JM, Casey VA, Welch LC, Edgman-Levitan S. Patient-focused, family-centered end-of-lifemedicalcare:viewsoftheguidelinesandbereavedfamilymembers.JPainSymptomManage2001;22:738-51.

31. SteinhauserKE,ChristakisNA,ClippEC,McNeillyM,McIntyreLTulsky JA.Factorsconsideredimportantattheendoflifebypatients,family,physicians,andothercareproviders.JAMA2000;284;2476-82.

32. VenkeGranS,Miller J.Norwegiannurses’ thoughtsandfeelingsregarding theethicsofpalliativesedation.IntJPalliatNurs2008;14:532-538.

33. VerkerkM,WijlickE,vanLegemaateJ,deGraeffA.AnationalguidelineforpalliativesedationintheNetherlands.JPainSymptomManage2007;34:666-670.

Chapter 4

Making sense of continuous sedation in end-of-life care for cancer patients: an interview study with bereaved relatives in three European countries

Bruinsma SM, Brown J, van der Heide A, Deliens L, Anquinet L, Payne SA,

Seymour JE, Rietjens JA, on behalf of UNBIASED

Supportive Care in Cancer 2014; 22(12): 3243-52

Chapter 466

aBstract

Purpose.Thepurposeofthestudywastoexplorerelatives’descriptionsandexperi-encesofcontinuoussedationinend-of-lifecareforcancerpatientsandtoidentifyandexplaindifferencesbetweenrespondentsfromtheNetherlands,Belgium,andtheUK.

Methods. In-depth interviews were held between January 2011 and May 2012 with38 relatives of 32 cancer patients who received continuous sedation until death inhospitals,thecommunity,andhospices/palliativecareunits.

Results. Relatives’ descriptions of the practice referred to the outcome, to practicalaspects,andtothegoalsofsedation.Whilemostrelativesbelievedsedationhadcon-tributedtoa‘gooddeath’forthepatient,yetmanyexpressedconcerns.Theserelatedto anxieties about the patient’s wellbeing, their own wellbeing, and questions aboutwhethercontinuoussedationhadshortenedthepatient’slife(mostlyUK),orwhetheranalternativeapproachwouldhavebeenbetter.Suchconcernsseemedtohavebeenprompted by relatives witnessing unexpected events such as the patient coming toawareness during sedation. In the Netherlands and in Belgium, several relatives re-portedthatthestartofthesedationallowedforaplannedmomentof‘sayinggoodbye’.In contrast, UK relatives discerned neither an explicit point at which sedation wasstartednoraspecificmomentoffarewell.

Conclusions.Relativesbelievedthatsedationcontributedtothepatienthavingagooddeath. Nevertheless, they also expressed concerns that may have been provoked byunexpectedeventsforwhichtheywereunprepared.ThereseemstoexistdifferencesintheprocessofsayinggoodbyebetweentheNL/BEandtheUK.

Making sense of CS: Interview study with bereaved relatives 67

introDUction

Sometimes, patients who are nearing death have symptoms such as dyspnea, pain,nausea,anddeliriumthatcannotberelievedwithconventionaltherapeuticinterven-tions (1, 2). One option for relieving such symptoms entails the continuous use ofdrugswithasedativeeffecttoinduceastateofunconsciousness,sothatthepatient’sperceptionofsymptomsisremoved.Continuoussedationisusedforpatientswithalifeexpectancyof2weeksorless(3–5).Thereissomediscussionabouttheserecommen-dations.Estimatinglifeexpectancyofpatientswithadvanceddiseaseisknowntobeverydifficult(6–8).Moreover,somearguethatpalliativesedationmaybeindicatedforpatientswithalongerlifeexpectancyifthebenefitofpalliationoutweighstheharmofanearlierdeath,iftherearenootheralternatives,andifdosagesaretitratedaccordingto thepatient’sneed(9).Sedation is frequentlyused inend-of-lifecare.AEuropeanstudyundertaken in sixcountries in2001/2 reported thatcontinuousdeepsedationuntildeathwasused in themanagementof2.5–8.5%ofalldeaths(10).Subsequentstudiessuggesttheuseofsedationinthemanagementof15%ofalldeathsinFlanders,Belgium(11),12%ofdeathsinTheNetherlands(12),and17%ofdeathsintheUK(13,14).Ethicalconcernsandpracticalissuessurroundingthepracticehaveresultedinthepublicationofanumberofguidelinesorframeworksforclinicalpractice(3–5,15,16).

Moststudiesinvestigatingtheuseofsedationhavefocusedonphysicians’practicesandattitudes(10,17–23).Therehasbeenlittleresearchexploringtheperceptionsandexperiences of bereaved relatives. Rendering patients unconscious until death is afar-reachinginterventionthatmayhaveasignificantimpactontheexperienceofthedyingprocessbothforthepatientandtheirrelatives.Theintimacyoffamilycaremaybedisruptedby the introductionof technologies suchas sedation (5),with relativesexperiencingalossofcontactwiththepatient.Areviewoftheliteratureconductedin2012showedthatwhatlittleresearchhasbeenconductedtotrytounderstandrelatives’experienceshasbeenprimarilyaccessedfromtheperspectiveofproxies,usuallyclinicalstaff(24).Thelatterreviewshowedthatwhilethemajorityofrelativesperceivepositivebenefitstobeassociatedwiththeuseofsedation,someexperienceitsusetobeassoci-atedwithsubstantialdistress(24).Inthispaper,wereportresultsfromtheUNBIASEDstudy(UK-Netherlands-BelgiumInternationalSedationStudy)(25)conductedintheUK,Belgium,andtheNetherlandstogaininsightsintotheexperiencesofclinicalstaffandrelativeswiththeuseofcontinuoussedationuntildeathforcancerpatientandtheirperceptionsofitscontributiontothedyingprocessindifferentcountries.ApreviousUNBIASEDpapershowedthatthereissystematicvariationinend-of-lifecaresedationpracticebetween the threecountries (26).UKrespondents reportedacontinuumofpracticefromtheprovisionof lowdosesofsedativestocontrolterminalrestlessnesstorarelyencountereddeepsedation.Incontrast,Belgianrespondentspredominantly

Chapter 468

describedtheuseofdeepsedation,emphasizingtheimportanceofrespondingtothepatient’srequest.Dutchrespondentsemphasizedmakinganofficialmedicaldecisioninformedbythepatient’swishandestablishingthatarefractorysymptomwaspresent.UsualpracticeintheNetherlandswasdescribedasinvolvingstartingwithlowdosesofsedativesandthen—ifnecessary—cautiouslyincreasingthedosageuntilthepatientwas in a quiet and peaceful state (26). This paper addresses the following specificquestions:Whatarerelatives’understandingsofthepracticeofcontinuoussedationattheendoflife?Howdorelativesexperiencethedyingprocessofthepatientwhohassedationadministeredcontinuously?Whatarethedifferencesinrelatives’experiencesbetweentheNetherlands,Belgium,andtheUK?

methoDs

Between January 2011 and May 2012, in-depth interviews were held with bereavedrelativesofpatientswithcancerwhodiedafter theuseof continuous sedationuntildeath in the Netherlands, Belgium, and the United Kingdom (UK). In total, 84 de-ceasedpatientswereincludedascasestudies.FulldetailsofthemethodsemployedintheUNBIASEDstudyareavailable inapublished studyprotocol (25);weprovideasummaryhere.

settings

Toreachmaximumvariation(27),thestudyexploredthecareofpatientswhodiedinhospitals(oncologywards),palliativecareunits(PCU)(inBelgium),hospices(intheUKandtheNetherlands),andinthecommunity(athome).

recruitment of relatives

In all countries, senior clinical staff identified eligible decedents: patients aged over18 who had died of cancer and to whom sedating medications (benzodiazepines orpropofol,butnotmorphine),withthe intentiontodecreaseawareness,wereadmin-isteredcontinuously toalleviateotherwiseuncontrollable symptoms (eitherphysicalorpsychological/existential), and forwhomthe sedationwas inplaceat the timeofdeath.Asanext step, they identifiedeligiblerelatives.Theperson identifiedas theirclosestrelativewasinvitedtotakepartinthestudyviaaletterandinformationsheetsent on behalf of the research team by patient’s physician. Relatives were invited atleast3monthsafterthedeathofthepatient(withamaximumof18months).Relativescontactedtheresearchteamiftheywerewillingtotakepartinthestudy.Whenrela-tivesdidnotrespondwithin1month,physicianswereaskedtosendareminder.Most


physicianswerewillingtodothis.Therelativecouldinviteoneortwootherrelativestoparticipateintheinterviewiftheywishedandfivechosetodoso.

Procedure

Theinterviewswereundertakenbytrainedinterviewersandlastedapproximately60min.Allparticipantsconsentedto the interviewbeingaudiotaped.At thebeginningof each interview, socio-demographic information was obtained through a shortquestionnaireandallparticipants signedaconsent form.The interviewswere semi-structured,supportedwiththeuseofaide-memoiretoensurethatallareasofinterestwere explored. The aide-memoire was piloted in focus groups and interviews priortouseintheNetherlandsandsmallchangesintheformulationofthequestionsweremade (28). The aide-memoire focused on relatives’ recollection of the care for thepatientandoftheuseofsedationinparticular.Ifrelativesactivelyreferredtotheuseofsedation,thisbecamethefocusoftheinterview.Ifnot,theywereasked‘Didanyoneevergivehim/hersomethingtohelphim/hertorelax?’Theterm(s)thatrelativesusedtodescribetheuseofsedationwereusedbytheinterviewerthroughouttheinterview.

Data analysis

Allaudiotapedinterviewsweretranscribedverbatim,andalldatathatcouldidentifypatientsorrelativeswereremovedtopreserveanonymity.TheBelgianandDutchinter-viewsweretranslatedintoEnglishbyaprofessionaltranslationbureauandcheckedforaccuracybytheresearchers.SBandJRreadthroughalltranscripts.Qualitativeanalysissoftware(NVIVO9)wasusedtoorganizethedata.Mainthemesrelatingtorelatives’experienceswereidentified.Acodingtreewasdevelopedandagreeduponbyallau-thors.TheinterviewswererereadandcodedwiththeuseofthecodingtreebySB.ThecodeswereindependentlystudiedbySBandJRanddiscusseduntilconsensusaboutthefurtherrefinementandcontentofthecodingtreewasreached.Finally,quoteswereselectedbySBandJRandapprovedbyallauthors.Eachquoteisfollowedbyinforma-tiononthecountryoforiginoftherelative(NL,BE,UK),andthesettingwherethepatientdied.Whenthereismorethanonerespondentwiththesamecharacteristics,thisisfollowedbyanumbertodistinguishthedifferentrespondents.

resUlts

In32of the84cases(13NL;11BE;8UK),oneormorerelativesvolunteeredtobeinterviewed.Intotal,38relativesparticipatedin32interviews(17NL;13BE;8UK).Infivecases,therewasnocontactinformationavailable.Insevencases,thephysicianwasnotwillingtocontacttherelativebecausehe/shewasafraidthattheinvitationwould

Chapter 470

stirupalotofdifficultissues,orthephysiciandidnotrespondtotherequesttoinvitetherelative.In72cases,therelativeswerecontactedbythephysician.Ofthesecases,29didnotreplytotheinvitation;10declinedparticipation,mostlybecausetheyconsid-eredparticipationtooupsetting;andinonecasetherewasnorelative.CharacteristicsofthedeceasedpatientsarepresentedinTable1andoftherelativesinTable2.

Several main themes that were raised during the interviews will be discussed:relatives’understandingsofcontinuoussedationattheendoflife,thecontributionofsedationtothequalityofdying,concernsrelatedtotheuseofsedation,andtheprocessofsayinggoodbye.

table 1 Characteristics patients (n=32)

BE NL UK Total

Sex Male 6 7 6 19

Female 5 6 2 13

Age 30-92 43-83 51-81 30-92

Primary diagnose Lung cancer 1 4 1 6

Pancreatic cancer 2 1 1 4

Prostate cancer 1 1 1 3

Melanoma 1 2 - 3

Other types of cancer 6 5 5 16a

Setting Community 4 5 4 13

Specialist palliative care setting/ hospice

3 3 4 10

Hospital 4 5 - 9

a Leukemia (2), peritoneal (2), renal (2), abdominal/stomach (1), bladder (1), brain (1), colon (1) facial maxillary (1), esophageal (1), unknown primary (2), and uterus (1)

table 2 Characteristics relatives (n=32a)

Be nl Uk total

sex Male 5 4 1 10

Female 6 9 7 22

nature of relationship with patient

Partner 5 9 4 18

Child 3 2 4 9

Sibling 1 1 - 2

Parent - 1 - 1

Grandchild 1 - - 1

Daughter in law 1 - - 1

a Results from the relatives that were identified by the physician as the ‘most involved relative’


relatives’ understandings of continuous sedation at the end of life

Some relatives from the Netherlands and Belgium indicated that they had previousexperienceoftheuseofsedationinthecareofanothercloserelative,whileothershad‘heardaboutit’throughthemedia.Somerelatives,especiallythosefromtheNether-landsandBelgium, indicatedthat thephysiciancaringforthepatienthadexplainedtheconceptofsedationtothem.RelativesfromtheUKwereparticularlyunsureaboutwhatitsuseentailed,althoughthisuncertaintywassharedbysomerelativesfromtheNetherlands and Belgium. This uncertainty sometimes led to feelings of distress forrelatives.

‘Well, I don’t know what they actually,…to be honest, I don’t know what they gave him that last week’ (UK, hospice, 1)

Most of the relatives from the Netherlands and Belgium used the term ‘sedation’ or‘palliativesedation’duringtheinterviews,whileothers,particularlythosefromtheUK,tendedtousephrasessuchas‘makingthepatientmorerelaxed’,‘sleeping’,‘calmingthepatientdown’or‘coma’.

‘She was in like a semi-coma’ (UK, hospice, 2)

Relativesusedthreedistinctwaysofdescribingthepracticeofsedation,theoutcome,thepractical issues,andthegoalsofsedation.Theoutcomeofsedationrelatedto theeffectsthatrelativeswereexpecting;forinstance,termswereusedsuchas‘sleeping’or‘acoma’.

‘To put him to sleep with the knowledge that he couldn’t come out of that by himself ’ (NL, hospice)

Several relatives from all three countries focused on the practical aspects of usingsedation.For instance, they referred to thedrugs thatwereusedand themechanicsof administering the sedatives.Termswereused suchas: ‘the injection’, ‘amorphinepump’, ‘apumpwithasedative’, ‘asyringedriver’, ‘acassettewithsleepingstuff ’,or ‘asleepingpill’.

‘Yes the doctor explained what would happen again (…). That he [the patient] would get a pump with a sedative’ (NL, community, 1)

Relativesalsoreferredtothedurationofthesedation,theuseoffoodandfluids,andtheproportionalityofsedation.

Chapter 472

‘Yes you don’t know how long it would have taken you know. Maybe it might have taken five days, I don’t know. But with the sedation it was a maximum of five days’ (NL, hospital, 1)

Finally,somerelativesreferredtothegoalsofsedation,primarilyintermsofprevent-ing or decreasing the (physical or psychological/existential) suffering of the patient.Sometimesspecificsymptomswerementioned,suchaspainandnausea.Inothercases,relativesdidnotusesuchspecificterms,butdescribedthegoalofsedationas‘treatingthediscomfort’.Othergoalswereassociatedwithallowingthepatienttobepeacefulandtodiewithdignity.

‘So they changed the medication (…) and hopefully it’ll, erm, make him more, more relaxed’ (UK, community, 1).

‘The reason why they gave X those drugs was… yes… doctors shared the opinion that it wasn’t possible anymore and that this was really it for X… this really was not dignified, no one deserves that’ (NL, hospital, 2)

Some Dutch and Belgian relatives compared sedation with other practices such assurgicalanesthesia,andeuthanasia.

‘So it’s just, like an operation, you don’t wake up’ (BE, hospital, 1)

‘In contrast to, to euthanasia, uh, yes, that is, they give an injection and one, two, three, poof and it’s done huh’ (BE, hospital, 2).

contribution to the quality of dying

Relativesfromallthreecountriespositivelyevaluatedtheprovisionofsedationtotheirdyingfamilymember.Manyreportedthatthesedationcontributedtothepatienthav-ingagooddeathanddescribedsedationaseffectiveinachievingsymptomrelief forthepatient,makingthemmorecomfortable,and/orallowingthemtodieinpeaceandwithdignity.Somerelativesuseddescriptionslike‘beautiful’,‘peaceful’,‘wonderful’,or‘dignified’todescribethedeath.

‘Erm … he [the patient] was very reluctant, he was afraid he would have a lot of pain and that he would choke and stuff like that. And because he was just asleep and it was actually painless, I think that for him it had additional advantage…so


he actually had assurance for himself that no crazy things would happen (…) That was the contribution for him’ (NL, community, 1)

Interviewer: ‘Yeah. And how did you feel about that?’ Respondent: ‘Yeah, I think I thought it was better than trying to…seeing him in distress when we’d had to move him and do all of that…to see him sleeping what looked like peacefully… that was fine’ (UK, community, 1)

Severalrelativesalsofeltthatsedationhadcontributedtothepatient’squalityofdyingbecauseitallowedfortheirwishestobehonored.

‘Because of the palliative sedation (…) we knew: look, she still got what she wanted at the very end’ (BE, hospital, 3).

concerns

Inmanycasesrelatives(mostly,althoughnotsolelyfromtheNetherlandsandBelgium)alsoexpressedconcerns,relatingtothewellbeingofthepatientandthemselves,pos-siblealternativestotheuseofsedation,andthepossibilityofhasteningdeath.

Wellbeing of the patient during the sedationConcernswereoftenrelated to thewellbeingof thepatientduring sedation.Severalrelatives felt that when sedation was commenced, the patient did not ‘fall asleep’ asquicklyastheyhadanticipatedandsomerelativesdescribedthepatient‘waking’duringsedation.Othersexpressedconcernsthateitherthepatientwasnotsufficientlysedatedto relieve the suffering, or that the level of sedation led patients to be more deeplyunconsciousthantheyneededtobe(thelatterconcernedmainlyUKrespondents),ortheydiedsoonerthanexpected.

‘And when he was dying he was constantly pulling, always pulling. (…) And he started to vomit.(…) Gosh yes, how should I say this? I’m very content that we had those last 14 days together, but in those final hours, I wouldn’t even wish that on my worst enemy. The agony..’ (BE, community)

Erm… I thought that perhaps the medications maybe had dosed him up too, too much… of course, probably (…) So even, although they put himon some medica-tion that you thought was perhaps making him a lit-, or he was a little bit over-medicated…(UK, community, 2)

Chapter 474

Wellbeing of relativesSomerelativesdescribedconcernsregardingtheuseofsedationthatwerespecificallyrelatedtotheirownwellbeing.Theseweresometimesrelatedtothe(long)durationofthesedation.

‘I thought that was so cruel hey, that was awful (…) because he was in a coma for eleven days, which is really very long (…) and they warned us (…) And we said “yeah we can handle that and we are aware of it and so on, hey, that’s no problem”. But as the days progressed and the sleep, I began to realize it” yes doc-tor, you have indeed warned us, but you cannot know how we feel (…)”. That was something they could not prepare me for’ (BE, PCU, 1)

Alternatives to the use of sedationConsideration of whether an alternative approach to sedation of the patient wouldhavebeenmoreappropriatewasalsoasourceofconcernforsomerelativesinallthreecountries.Severalquestionedwhetheritwouldhavebeenmoreappropriateforseda-tiontohavestartedearlierorwhethereuthanasiawouldhavebeenamoreappropriateoption.Onerelative(UK)feltitwouldhavebeenmoreappropriateforthesedationtohavebeendeeper;another(NL)questionedwhethersedationhadbeennecessaryandreflectedonwhobenefittedfromitsuse.

‘My feeling is that it [the death of the patient] probably would not have gone otherwise, apart from the fact that we now have the feeling that we still have done something for her, (…) because that’s what I just say, seeing suffering is worse than suffering itself. So I wonder if that last hour when she slept on those drugs was it for our benefit or hers’ (NL, hospital, 2)

Hastening deathTherewerefewconcernsaboutwhetherornotsedationhadshortenedthepatient’slifeintheNetherlandsandBelgium.However,intheUK,theissueofhasteningdeathwasaconcernraisedbysomerelatives.

‘I mean, I did think to myself, ‘Have we actually killed him? …Well, it was obvi-ously something very powerful…’ (UK, hospice, 1)

sedation and the process of saying goodbye

Indescriptionsofthedyingprocessthatrelativesgave,thereweremarkeddifferencesbetweenthecountriesinrelationtotheprocessofsayinggoodbye.


Netherlands and Belgium: a planned farewellIn the Netherlands and Belgium, the opportunity to say goodbye was highly appre-ciated by many relatives and patients who knew that death was inevitable and wereable,tosomeextent,toprepareforit.Sayinggoodbyewasfrequentlyaplannedeventoccurringeitherbeforesedationbegan,orasitcommenced.However,somerelativeshad less positive experiences. For example, one relative described the gathering ofeveryonearound thepatients’bed foraplannedgoodbyeasa ‘carnival’.Unexpectedevents,asdescribedabove,suchasthepatientregainingconsciousness,wereperceivedas distressing and interfered with the process of saying goodbye. A further stressorforrelativeswasthetimebetweensayinggoodbyeandthepatients’deathwithsomedescribingthisasa‘vacuum’wheretheywere‘waiting’forthepatienttodie.

‘But they did give us the time to say goodbye. They said…because we had that conversation in the morning, like, we are going to do that [start the sedation] in the afternoon, like they slowly came to prepare her and they kept on saying like, she can still hear you now. So you can still talk to her now. But once we’ve put her to sleep later then contact isn’t possible anymore’ (NL, hospital, 1)

‘It’s difficult to put it into words, but as you said, you have indeed already said goodbye when the sedation is administered. You can say a fancy goodbye hey, right that’s how it should be, what really helps in placing death, huh, but then there…(…) You cannot leave the man there at the palliative unit without going there huh. Just the idea of ‘I’m leaving my father behind, I’m not going… That is not good; my brother said “I’m sorry; I’m not going to go anymore. I’m no longer going, because I already said goodbye…” (BE, PCU, 1)

United Kingdom: gradual processAdiscretemomentofpartingwaslessdominantintheaccountsofrelativesfromtheUK.Here,theuseofsedationwasgenerallydescribedasagradualprocesswithincreas-ingsedativedosesandaccompanyinglossofconsciousnessoccurringovertimeduringsymptom management rather than being a planned single event. Relatives thereforedidnotusuallydescribeaspecificplannedmomentofgoodbye.Somerelativesinallthreecountriesdescribedoccasionswherepatientshadregainedconsciousnessoncesedationhadbeencommenced,however,intheUKsuchsituationswerenottypicallydescribedasdistressing;rather,theywerewelcomedandgreatlyappreciatedbysome.

‘Erm, but, erm…the thing was, they like, when he woke up and you had some communication, these were golden moments, because these were the times…these were all I had left’ (UK, hospice, 3).

Chapter 476

DiscUssion

This was the first study to gain in-depth understanding of relatives’ experiences ofcontinuoussedationuntildeathinthreeEuropeancountries.Mostrelativeswereabletoprovideadescriptionoftheconceptofsedation;nevertheless,severalwereunsureaboutwhat it entailed.Althoughrelativesgenerallybelieved sedationcontributed tothepatienthavingagooddeath,theyalsoexpressedsomeconcernsaboutitsuseandexperiencedsomeunexpectedevents forwhich theywereunprepared.According torelatives from the Netherlands and Belgium, the start of the sedation allowed for aplannedmomentof‘sayinggoodbye’.Incontrast,relativesfromtheUKdescribedtheprocessofsayinggoodbyeasamoregradualandlessexplicitprocess.

relatives’ understandings and descriptions of the concept of sedation

Most relatives in this study were aware of, and used, the term ‘sedation’ or ‘pallia-tivesedation’.Relativeshadthreedistinctwaysofdescribingthepractice,referringtothegoals,practicalaspects,andoutcomesofsedation.Incontrast,astudyamongthegeneralpopulationofDutchcitizensshowedthatthemajoritywereunfamiliarwiththeconceptofsedation,describingitassimilartodehydrationoreuthanasia(29).Whilethisseemsinitiallyincongruentwiththefindingsofthisstudy,thismaybebecausethesamplehadnopersonalexperienceoftheuseofsedation.However,itwasalsoevidentthatsomerelativesfromallthreecountriesinthisstudycontinuedtobeunsureaboutthe use of sedation. Reflecting on the death of their family member, some relativesstillhadquestionsandcertainmisunderstandingsaboutwhat sedationentailed.Forexamplea fewrelatives thoughtmorphinewas thepredominantdrugusedtosedatethepatient,althoughbenzodiazepineswerethesedativesusedforallpatientsincludedinthisstudy.

Althoughrelativesfromallthreecountriesinthisstudygenerallyappearedtobesatisfiedwith theprovisionof information regarding sedation, lackofknowledgeorunderstandingforsomeappearedtoberelatedtoconcernstheyexperiencedasaresultofitsuse.Manystudieshavehighlightedinadequateprovisionofinformationandpoorcommunicationinend-of-lifecarewhichmayresultinfeelingsofisolation,disillusion,anddistress(30)anditisevidentfromthisstudythatgivingofinformation,regularlyovertime,andcheckingofunderstandingcouldbeimportantinfosteringpatientsandrelatives’satisfactionwiththeuseofsedation.

Finally, relatives in our study described the goal of sedation primarily in termsof preventing or decreasing the physical or psychological/existential suffering of thepatient.There isgeneral consensus regarding theuseof continuous sedation for themanagementofphysicalsymptomssuchasdeliriumordyspneathatarerefractorytopalliativetreatments.Theuseofsedationtotreatuncontrolledpsychosocialorexisten-


tialsufferingis,however,consideredcontroversial(31,32).Ithastobekeptinmindthatrelativesmayhavedifferentexperiencesincaseofdifferentrefractorysymptomsneeds.

Quality of dying and relatives’ concerns

Relativesfromallthreecountriesevaluatedtheuseofsedationfortheirdyingfamilymember as being positive overall. This positive evaluation was largely related to thedecreaseinpatients’physicalorpsychological/existentialsuffering(33),andpreviousstudieshaveshownthatadequatesymptomrelief iskeyto theexperienceofa ‘gooddeath’(34–37).

However,ithasbeenarguedintheliteraturethatconsciousnessisafundamentalpartofbeingalive;therefore,deliberatereductionofconsciousnessbytheuseofdeepsedationmaycontributetoalossofdignityforthepatient(15).Nevertheless,accordingtotherelativesinthisstudypatientdignitywasthreatenedbyoverwhelmingsymptomsandsedationcontributedtoadignifieddeathbyofferingsomerelieffromthosesymp-toms.Aswithotherstudies(38-39),thisresearchdemonstratedtheimportanceofcarebeinginlinewithpatient’swishesand,inthisstudy,theuseofsedationwasperceivedascontributingnotonlytothepatient’squalityofdyingbutalsoasameansofhonoringtheirwishesandallowingthemtodiewithdignity.

Althoughthemajorityofrelativesreportedthattheywerecomfortablewiththeuseofsedation,manyalsoexpressedconcernsaboutitsusewhichrelatedtothewellbeingofthepatientandtherelativethemselves,possiblealternativestotheuseofsedation,and the hastening of death. The type of concerns that relatives expressed differedbetween the three countries. Concerns regarding the potential hastening of deathwere for instance mostly expressed by relatives from the UK. Previous results fromtheUNBIASEDstudy showed that in theUK,anoverarchingconcernexistsamongprofessionalcaregiverstoavoidhasteningdeath,whilemostoftheDutchandBelgianhealthcareprofessionaldidnotperceivethatsedationhasteneddeathoracceptedthatitmay,tosomeextent(26).ConcernsregardingthepotentialhasteningofdeathwereforinstancemostlyexpressedbyrelativesfromtheUK.ForrelativesfromtheNether-landsandBelgium,unexpectedevents,suchasthetimeittookthepatienttobecomeunconsciousorpatientscomingtoawarenessduringsedation,wereanimportantcauseofconcern.Thissuggeststwothings,ontheonehandsomerelativesmayhavenothavebeensufficientlyinformedbytheirprofessionalcaregiversaboutwhatcanoccuratthestartof,andduringsedation.Ontheotherhand,itraisesquestionsabouttheeffective-nessof the sedativedrugs, thedosagesused, echoingpreviouslypublishedconcerns(40–42), while at the same time demonstrating the complexities of administratingappropriatedosagetoinduceunconsciousness.

Chapter 478

sedation and saying goodbye

Saying goodbye to the patient as an event or particular moment was an importantthemeintheNetherlandsandBelgium,butlesssointheUK.IntheDutchandBelgiancases,thestartofsedationwasoftenadistinguishableandplannedmoment,inwhichan‘orchestratedfarewell’wascreated.Itisstrikingthatrelatives’descriptionsofsucha planned ‘goodbye’ have many similarities with the practice of euthanasia, (whichis legalized in the Netherlands and Belgium) (43, 44) where relatives gather aroundthebedofthepatientandsaygoodbye,andthephysicianadministersdrugs.Oneofthekeydifferencesthoughseemstobethetrajectorythereafter.Patientswhoreceiveeuthanasiatypicallydiewithinseveralsecondstominutes,incontrast,inpatientswhoreceivesedation,deathtypicallyfollowsafteroneormoredays(45),andassucha‘so-cialdeath’precedesthe‘physical’deathofthepatient(15).Swarteetal.(46)foundthatbereavedrelativesofcancerpatientswhodiedbyeuthanasiacopedbetterwithrespecttogriefandpost-traumaticstressreactionsthanthebereavedofcomparablepatientswhodiedanaturaldeath,andthatthiswaspredominantlyrelatedtotheopportunitytosaygoodbyetothepatient.Theimportanceofsayinggoodbyeandthebeneficialeffectsontheriskofexperiencingtraumaticgriefhavebeendescribedbyothers(35,47).

However, inthisstudy, it tooktimeforsomepatients to loseconsciousness,oth-ersunexpectedlyregainedconsciousness,or lived longerthanexpected,allofwhichprovokedconcernsinsomerelativesandpotentiallycompromisedtheexpectationinBelgiumandtheNetherlandsforplannedmomentofparting,althoughnotnecessarilyforrelativesintheUK.Thishighlightstheinfluenceofculturalnormsandexpectationsininfluencingperceivedexperiencesofcontinuoussedationandthesubsequentdeath.Forexample,whiletheuseofsedationmaysometimesbeconsideredashamperingtheprocessofsayinggoodbye,andasaresultcouldcomplicaterelatives’grievingprocessinonecontext,thismaynotapplyinanother.Itisclearthatsuchculturalexpectationsshouldbetakenintoconsiderationwhenexplainingsedationandgivinginformationtorelatives.

Itwas striking that sayinggoodbyewasa lessdominant theme in theaccountoftheUKrelatives.Ifdescribedatall,theydescribedsayinggoodbyeasamoregradualprocessratherthanaplannedsinglemomentintime.ThisisinlinewiththefindingofapreviousstudythatintheUK,usualsedationpracticeentailsstartingwithalowdoseofsedativeswhichcanbegraduallyincreasedwhennecessary,withlessemphasisplacedonmakinganofficialmedicaldecision(26).

limitations

As this is a qualitative study based on the intense exploration of a relatively smallnumber of cases, it is not possible to generalize in a statistical sense from our find-ings.Secondly,tobesensitivetotheearlyphasesofgrieving,relativeswerecontacted


no sooner than 3 months after the death of the patient, introducing potential recallbias.Thirdly,thedatacouldbeinfluencedbyselectionbias,becauseitispossiblethatrelativeswithrelativelystrongpositiveornegativeexperiencesagreedtoparticipateinthestudy.Fourth,duetothesmallnumberofcasesincludedinthestudy,comparisonbetweencountriesisdifficult.Therefore,ourresultsshouldbeinterpretedwithsomecaution.Finally,theresponseratewasratherlowamongbereavedrelatives.Itishighlylikelythattheimpactofbereavementfollowingadeathwherethepatienthadseveresymptomsmayhave influenced this lowresponse rate,butwedonothaveanyfirmevidenceaboutthis.

recommendations for practice and directions for future research

Understandingrelatives’experienceswithcontinuoussedationuntildeathisofvalueindevelopingeffectivecarestrategies(48).Boththenatureandextentoftheconcernsthatrelativesexperiencesuggestthatrelativesareinneedofcontinuousinformationand professional guidance at the beginning and throughout continuous palliativesedationoftheirfamilymembers(49).Providingfullinformationandregularupdatesabout e.g., the level of consciousness, clinical symptoms, and (the lack of) potentialalternatives is important. It isalso important that there isacommonunderstandingof terms and phrases used for all involved (relatives and health care professionals)andthatchecksaremadetoensurethisunderstandingismaintainedthroughouttheprocess. To ensure that relatives’ diverse concerns are addressed, and to reduce therisk of potential adverse health outcomes for them, the needs of relatives should becontinuouslymonitored.Guidelinesontheuseofsedationinend-of-lifecarestatethatfamiliesshouldbeallowedandencouragedtobewiththepatientbeforeandduringthesedation,andshouldbegivenanopportunitytosaygoodbye(4).Furtherstudiesshouldinvestigateinmoredepthhowtheuseofsedationaffectsgrief,andhowsayinggoodbyecouldbebestincorporatedinthevaryingcontemporarysedationpracticesattheendoflife.

acknowleDGements

Wethanktherelativeswhogavetheirtimetobeinterviewedinthestudy.

Chapter 480

references

1. RietjensJ,vanDeldenJ,Onwuteaka-PhilipsenB,BuitingH,vanderMaasP,vanderHeideA.Con-tinuousdeepsedationforpatientsnearingdeathintheNetherlands:descriptivestudy.BMJ2008;336:810–13.


3. BroeckaertB,MullieA,GielenJ,DesmetM,VandenBergheP.GuidelinePalliativeSedation:Eth-icsSteeringCommitteeoftheFederationforPalliativeCareFlanders[Sedation];2012[cited2013November11].Availablefrom:http://www.pallialine.be/template.asp?f=rl_sedatie.htm.

4. ChernyNI,RadbruchL.BoardoftheEuropeanAssociationforPalliativeCare.EuropeanAssocia-tionforPalliativeCare(EAPC)recommendedframeworkfortheuseofsedationinpalliativecare.PalliatMed2009;23:581–93.

5. Royal Dutch Medical Association (RDMA). Guideline for palliative sedation [in Dutch]. 2009.Available from http://knmg.artsennet.nl/Publicaties/KNMGpublicatie/Guideline-for-palliative-sedation-2009.htm.AccessedOctober,2010.

6. BrandtHE,OomsME,RibbeMW,vanderWalG,DeliensL.Predictedsurvivalvs.actualsurvivalinterminallyillnon-cancerpatientsinDutchnursinghomes.JPainSymptomManage2006;32:560–6.

7. GlareP,VirikK,JonesM,HudsonM,EychmullerS,SimesJ,ChristakisN.Asystematicreviewofphysicians’survivalpredictionsinterminallyillcancerpatients.BMJ2003;327:195–8.

8. RietjensJA,BuitingHM,PasmanHR,vanderMaasPJ,vanDeldenJJ,vanderHeideA.Decidingaboutcontinuousdeepsedation:physicians’perspectives:afocusgroupstudy.PalliatMed2009;23:410–7.

9. JanssensR,vanDeldenJJ,WiddershovenGA.Palliativesedation:notjustnormalmedicalpractice.EthicalreflectionsontheRoyalDutchMedicalAssociation’sguidelineonpalliativesedation.JMedEthics202;38:664–8.

10. Miccinesi G, Rietjens JA, Deliens L, Paci E, Bosshard G, Nilstun T, Norup M, van der Wal G,ConsortiumE.Continuousdeepsedation:physicians’experiencesinsixEuropeancountries.JPainSymptomManage2006;31:122–9.

11. ChambaereK,BilsenJ,CohenJ,RietjensJA,Onwuteaka-PhilipsenBD,MortierF,DeliensL.Con-tinuousdeepsedationuntildeathinBelgium:anationwidesurvey.ArchInternMed2010;170:490–3.

12. Onwuteaka-PhilipsenBD,Brinkman-StoppelenburgA,PenningC,deJong-KrulGJ,vanDeldenJJ,vanderHeideA.Trendsinend-of-lifepracticesbeforeandaftertheenactmentoftheeuthanasialawintheNetherlandsfrom1990to2010:arepeatedcross-sectionalsurvey.Lancet2012;380:908–15.

13. AnquinetL,RietjensJA,SealeC,SeymourJ,DeliensL,vanderHeideA.ThepracticeofcontinuousdeepsedationuntildeathinFlanders(Belgium),theNetherlands,andtheU.K.:acomparativestudy.JPainSymptomManage2102;44:33–43.

14. SealeC.Continuousdeepsedationinmedicalpractice:adescriptivestudy.JPainSymptomManage2010;39:44–53.

15. de Graeff A, Dean M. Palliative sedation therapy in the last weeks of life: a literature review andrecommendationsforstandards.JPalliatMed2007;10:67–85.

16. VerkerkM,vanWijlickE,LegemaateJ,deGraeffA.AnationalguidelineforpalliativesedationintheNetherlands.JPainSymptomManage2007;34:666–70.

17. ChiuTY,HuWY,LueBH,ChengSY,ChenCY.SedationforrefractorysymptomsofterminalcancerpatientsinTaiwan.JPainSymptomManage2001;21:467–72.


18. FainsingerRL,WallerA,BercoviciM,BengtsonK,LandmanW,HoskingM,etal.Amulticenterinternational study of sedation for uncontrolled symptoms in terminally ill patients. Palliat Med2000;14:257–65.

19. Muller-BuschHC,AndresI,JehserT.Sedationinpalliativecare—acriticalanalysisof7yearsexperi-ence.BMCPalliatCare2003;2:2.


21. StoneP,PhillipsC,SpruytO,WaightC.Acomparisonoftheuseofsedativesinahospitalsupportteamandinahospice.PalliatMed1997;11:140-4.

22. SykesN,ThornsA.Sedativeuseinthelastweekoflifeandtheimplicationsforend-of-lifedecisionmaking.ArchInternMed2003;163:341-4.

23. SykesN,ThornsA.Theuseofopioidsandsedativesattheendoflife.LancetOncol2003;4:312–8. 24. BruinsmaSM,RietjensJA,SeymourJE,AnquinetL,vanderHeideA.Theexperiencesofrelatives

withthepracticeofpalliativesedation:asystematicreview.JPainSymptomManage2012;44:431–45. 25. SeymourJ,RietjensJ,BrownJ,vanderHeideA,SterckxS,DeliensL,UNBIASEDstudyteam.The

perspectives of clinical staff and bereaved informal care-givers on the use of continuous sedationuntildeathforcancerpatients:thestudyprotocoloftheUNBIASEDstudy.BMCPalliatCare2011;10:5.

26. SeymourJ,RietjensJ,BruinsmaS,DeliensL,SterckxS,MortierF,etal.Usingcontinuoussedationuntildeath forcancerpatients: aqualitative interviewstudyofphysicians’ andnurses’practice inthreeEuropeancountries.PalliatMed2014(acceptedforpublication).

27. SandelowskiM.Samplesizeinqualitativeresearch.ResNursHealth1995;18:179–83. 28. BruinsmaS,RietjensJ,vanderHeideA.Palliativesedation:afocusgroupstudyontheexperiences

ofrelatives.JPalliatMed2013;16:349–55. 29. vanDeldenJJM,VanDerHeideA,vandeVathorstS,WeyersH,vanTolD.Knowledgeandopinions

ofthegeneralpublicandprofessionalsonmedicaldecisionmakingandtreatmentintheendoflife,the KOPPEL-research [in Dutch]. ZonMw 2011. http://www.zonmw.nl/uploads/tx_vipublicaties/ZonMw_A4_RapportKOPPELDEF_klein.pdf(accessed12March2013)

30. AndershedB.Relativesinend-of-lifecare—part1:asystematicreviewoftheliteraturethefivelastyears,January1999–February2004.JClinNurs2006;15:1158-69.

31. ChernyNI,Commentary:sedationinresponsetorefractoryexistentialdistress:walkingthefineline.JPainSymptomManage1998;16:404-6.

32. Rousseau P. Existential suffering and palliative sedation: a brief commentary with a proposal forclinicalguidelines.AmJHospPalliatCare2001;18:151–3.

33. RausK,BrownJ,SealeC,RietjensJAC,JanssensR,BruinsmaSM,MortierF,PayneSA,SterckxS.Continuous sedation until death: The everyday moral reasoning of physicians, nurses and familycaregiversintheUK,TheNetherlandsandBelgium.BMCMedicalEthics2014;15:14.

34. PiersonCM,CurtisJR,PatrickDL.Agooddeath:aqualitativestudyofpatientswithadvancedAIDS.AIDScare2002;14:587–98.

35. RietjensJA,vanderHeideA,Onwuteaka-PhilipsenBD,vanderMaasPJ,vanderWalG.Preferencesof the Dutch general public for a good death and associations with attitudes towards end-of-lifedecision-making.PalliatMed2006;20:685–92.

36. SteinhauserKE,ChristakisNA,ClippEC,McNeillyM,McIntyreL,TulskyJA.Factorsconsideredimportantattheendoflifebypatients,family,physicians,andothercareproviders.JAMA2000;284:2476–82.

Chapter 482

37. Teno JM, Casey VA, Welch LC, Edgman-Levitan S. Patient focused, family-centered end-of-lifemedicalcare:viewsoftheguidelinesandbereavedfamilymembers.JPainSymptomManage2001;22:738–51.

38. SteinhauserKE,ChristakisNA,ClippEC,McNeillyM,GrambowS,ParkerJ,etal.Preparingfortheendof life:preferencesofpatients, families,physicians,andothercareproviders. JPainSymptomManage2001;22:727–37.

39. StrangP.Mostoftheterminallyillpatientswanttodieathome.Advancedhomecareservicescanoftenreplacepalliativehospitalcare.Lakartidningen2002;99(742–743):746.

40. AnquinetL,RietjensJA,VandervoortA,vanderSteenJT,VanderSticheleR,DeliensL,etal.Con-tinuous deep sedation until death in nursing home residents with dementia: a case series. J AmGeriatrSoc2013;61:1768–76.

41. DeschepperR,LaureysS,Hachimi-IdrissiS,PoelaertJ,DistelmansW,BilsenJ.Palliativesedation:whyweshouldbemoreconcernedabouttherisksthatpatientsexperienceanuncomfortabledeath.Pain2013;154:1505–8.

42. MoritaT,ChinoneY,IkenagaM,MiyoshiM,NakahoT,NishitatenoK,etal.Efficacyandsafetyofpalliativesedationtherapy:amulticenter,prospective,observationalstudyconductedonspecializedpalliativecareunitsinJapan.JPainSymptomManage2005;30:320-8.

43. AdamsM,NysH.ComparativereflectionsontheBelgianEuthanasiaAct2002.MedLawRev2003;11:353–76.

44. LegemaateJ.TheDutchEuthanasiaActandrelatedissues.JLawMed2004;11:312–23. 45. RietjensJA,vanDeldenJJ,vanderHeideA,VrakkingAM,Onwuteaka-PhilipsenBD,vanderMaas

PJ,etal.Terminalsedationandeuthanasia:acomparisonofclinicalpractices.ArchInternMed2006;166:749-53.

46. SwarteNB,vanderLeeML,vanderBomJG,vandenBoutJ,HeintzAP.Effectsofeuthanasiaonthebereavedfamilyandfriends:acrosssectionalstudy.BMJ2003;327:189.

47. SwarteNB,HeintzAP.Euthanasiaandphysician-assistedsuicide.AnnMed1999;31:364–71. 48. Morita T, Ikenaga M, Adachi I, Narabayashi I, Kizawa Y, Honke Y, et al. Family experience with

palliativesedationtherapyforterminallyillcancerpatients.JPainSymptomManage2004;28:557-65. 49. vanDoorenS,vanVeluwHT,vanZuylenL,RietjensJA,PasschierJ,vanderRijtCC.Exploration

ofconcernsofrelativesduringcontinuouspalliativesedationoftheirfamilymemberswithcancer.JPainSymptomManage2009;38:452–9.

Chapter 5No negative impact of palliative sedation on relatives’ experience of the dying phase and their wellbeing after the patient’s death

Bruinsma SM, van der Heide A, van der Lee ML, Vergouwe Y, Rietjens JAC

Submitted

Chapter 586

aBstract

Background. Palliativesedationhamperscommunicationandthismayaffectthequal-ityofthedyingprocessnegatively,bothforpatientsandtheirrelatives.

Objective. Tostudywhetherrelativesofpatientswhoreceivedpalliativesedationdifferintheirexperienceofthedyingphaseandtheirwellbeingafterthepatient’sdeathfromrelativesofpatientswhodiedanon-suddendeathwithouttheuseofsedation.

Methods. We conducted an observational study among relatives of a consecutivesampleofpatientswhodiedanexpecteddeathintheErasmusMCCancerInstituteorhospiceLaurensCadenzainRotterdam,between2010and2013.Relativeswereaskedtofillinaquestionnaireaboutthepossibleuseofpalliativesedation,theirexperienceof the dying phase, and their wellbeing after the patient’s death. The association oftheuseof sedationwithrelatives’experiencesand theirwellbeingwasassessedwithamultivariatelinearregressionanalysis,controllingforsocio-demographicsandcarecharacteristics.

Results. Relativesfilledinquestionnairesfor151patientswhohadbeensedatedand90patientswhohadnotbeensedated.Themediantimesincethepatienthadpassedawaywas 21 months (IQR 14-32). Relatives of sedated patient more often stated that theprofessionalcaregiverscouldhavedonemoretomaketheperiodbeforethedeathofthepatientmorebearable(n=31,20%)thanrelativesofnon-sedatedpatients(n=7,8%)(p=.013).Nosignificantdifferenceswerefoundinrelatives´assessmentsofthequalityofend-of-lifecare(amedianscoreof9inbothgroupsona0-10scale),patients´qualityoflifeinthelastweekbeforedeath(amedianscoreof4and3ona0-10scale),andtheirqualityofdying(amedianscoreof8inbothgroupsona0-10scale),betweenrelativesof patients who did not and did receive sedation. Further, no significant differenceswere found in relatives’ satisfaction with their own life 3 months after the patient’sdeath(amedianscoreof6inbothgroupsona0-10scale)andatthetimeofthesurvey(amedianscoreof8and7ona0-10scale),theirgeneralhealth(amedianscoreof3inbothgroupsona1-5scale)andtheirmentalwellbeing(amedianscoreof11and12ona0-25scale).

Conclusion. Theuseofsedationdoesnotseemtohaveanegativeinfluenceonbereavedpersons’experienceofthedyingphaseoftheirdeceasedrelativeortheirwellbeingafterthepatient’sdeath.

Impact of PS on wellbeing relatives after patient’s death 87

BackGroUnD

Afrequentlyusedlastresortinterventioninend-of-lifecareispalliativesedation,thatis, theadministrationofsedatingdrugs to induceastateofunconsciousness to takeawayadyingpatient’sperceptionof symptoms (1-3).AEuropean studyundertakenin6countriesin2001/2reportedthatcontinuousdeepsedationuntildeathwasusedfor 2.5%-8.5% of all dying patients (2). Subsequent studies suggested that sedationwasusedfor15%ofdeceasedpersonsinFlanders,Belgiumin2007(4),in12%ofalldeceasedpersonsinTheNetherlandsin2010(5),and17%ofalldeceasedpersonsintheU.K.in2007(6,7).Palliativesedationisusedinallsettingswherepatientsdie,butmostofteninhospitalsandforpatientswithcancer(2,8-10).Whilepalliativesedationisoftenseenasanindispensableinterventiontoalleviateseveresufferinginthedyingphase, it is also heavily being criticized for depriving patients of the ability to com-municate,wheremanypatientsconsiderbeingconsciousuntildeathimportant(11),anditsresemblancetoeuthanasia(12).Further,itsusecanbedisturbingforpatients’relatives(13,14).Thestartofsedationmaybethetimeatwhichitdawnsonthefamilythatthepatient’sdeathisimminentandtheintimacyoffamilycaremaybedisruptedbytheintroductionoftechnologiessuchassedation(1).

Therehasbeenlittleresearchexploringtheexperiencesofrelativeswithpalliativesedation.Relatives’experienceshavemainlybeenunderstoodfromtheperspectiveofprofessionalcaregivers.Areviewoftheliteratureshowedthatdespitethefactthatthemajorityofrelativesseemstobecomfortablewiththeuseofsedation,severalstudiesfoundthatrelativesmayexperiencetheuseofsedationdistressing(15).Thereasonsforsuchdistressaree.g.thefeelingthatthepatientstillsufferswhilereceivingsedation,theinabilitytointeractwiththepatientafterthestartofsedation,feelingtheburdenofresponsibilityforthedecisiontostartsedation,concernsaboutapossiblyhasteneddeath,the(long)durationofthesedation,thefactthatinformationaboutthesedationcannoteasilybeobtained,andtheideathattheremightbemoreappropriatewaystoprovidesymptomrelief(14-17).Asaresult,sedationpotentiallynegativelyinfluencesrelatives’experienceofthepatient’sdyingphaseandtheirwellbeingafterthepatient’sdeath.Insightinthesepotentialconsequencesoftheuseofpalliativesedationcansup-portthedevelopmentofevidence-basedcarestrategiestoimprovethedeathexperienceforterminallyillpatientsandtheirrelatives.Althoughmostpalliativecareeffortsfocusonassessingandimprovingqualityoflifeandqualityofcareforpatients,manystudiesalsohighlighttheimportanceoftakinggoodcareofthepatient’srelatives(18-20).

Theaimofthisstudyistoassesswhetherrelativesofpatientswhoreceivedsedationdifferintheirexperienceofthedyingphaseandtheirwellbeingafterthepatient’sdeathcompared to relatives of patients who died a non- sudden death without the use ofpalliativesedation.

Chapter 588

methoDs

study population and recruitment of participants

Weconductedanobservationalstudyamongrelativesofaconsecutivesampleof564patientswhohaddiedanexpecteddeathintheErasmusMCCancerInstituteorhospiceLaurensCadenzaRotterdam,between2010and2013.Therelativewhowasregisteredasthecontactpersonofthepatientinthemedicalfile,wasinvitedtotakepartinthestudyviaaninformationletterthatwassentbyaseniorclinicalstaffmemberoftheparticipatingsettings.Iftherelativewaswillingtotakepart inthestudy,he/shewasaskedtofillinthequestionnaire.Iftherelativewasnotwillingtoparticipate,he/shewasaskedtoinformtheresearchteamviaanansweringsheet,ortoignoretheinvita-tion.Whentherelativedidnotrespondwithinsixweeks,he/shereceivedareminder.

Questionnaire

Thequestionnairewasbasedonseveralvalidatedquestionnaires(21-28).Toassesswhetherornotthepatienthadreceivedsedationpriortodeath,relatives

wereasked:‘Hasyourrelativebeenbroughttosleepwithmedicationpriortodeath?’(yes/no/don’tknow).

Relatives’ experience of the dying process was assessed in three items that werederived from the Quality of Death and Dying questionnaire (21): ratings on a 0-10scaleof(1)thepatient’squalityoflifeinthelastweekbeforedeath,(2)theirqualityofdyingand(3)thequalityofend-of-lifecare.Onthesescales,0representeda‘terribleexperience’and10an‘almostperfectexperience’.

Relatives’ satisfaction with life three months after the patient’s death and at thetimeofthesurveywasmeasuredbythe‘CantrilLadder’(22),i.e.apictureofaladdernumberedfrom0onthebottomrung(worstpossiblelife)to10onthetoprung(bestpossiblelife).Relatives’generalhealthwasassessedwithasingleitemfromtheShortForm36HealthSurvey (SF-36) (23).Ascoreof1 indicatespoorhealthanda scoreof5excellenthealth.Mentalwellbeinginthefourweekspriortothesurveywasas-sessedwiththeSF-36,whichincludes5 itemsaboutmentalhealth(23).Scoreswerelinearlytransformedfroma1-6toa0-5scale.Thesumscorehadaminimumof0andamaximumof25.Alowscoreindicatespoormentalwellbeing(feelingsofnervousnessanddepressionallthetime),andahighscoreimpliesexcellentmentalwellbeing(feelspeaceful,happy,andcalmallthetime).

Questionsaboutthesocio-demographiccharacteristicsofthedeceasedpersonandtherelativewereaddedbytheresearchteam(24,25).Itemsonrelatives’involvementin the care for the patient and the decision- making process, and the provision ofinformationwerebasedon theVOICESquestionnaire (26).Theseverityofpatient’s


symptomsduringthelastweekoflifewasassessedusingitemsfromtheESASandtheSISCquestionnaire(27,28).

Thecontentof thequestionnairewaspilotedamongfiverespondents.This ledtosomesmallchangesintheformulationofthequestions.InJulyandAugust2013,564questionnairesweresenttopotentialparticipants.

Data analyses

Toassesswhetherrelativesofpatientswhoreceivedsedationdifferedintheirexperi-ence of the dying phase and their wellbeing after the patient’s death from relativesofpatientswhohaddiedanon-suddendeathwithouttheuseofpalliativesedation,univariablelinearregressionanalyseswereperformed.Thevarianceofsomedependentvariables varied across the data (heteroscedasticity), therefore the bootstrap methodwas used (29). Bootstrapping is a method for deriving robust estimates of standarderrorsandconfidenceintervalsforpointestimatessuchastheregressioncoefficient.Tobeconsistent,thismethodwasusedforalldependentvariables.Logisticregressionwasusedtoassesstheassociationbetweenpersonalandcarecharacteristics,andtheuseofsedation.Thosevariablesthatshowedap-valueof0.30orlessinthisanalysis,wereincludedinthemainmultivariablelinearregressionmodel(figure1).

OUTCOME MEASURES Relatives’ experience of the dying phase • Quality end of life care • Quality of life last week before death • Quality of dying

Relatives’ wellbeing aer the patient’s death • Satisfaction with life 3 months aer patient’s

death • Current satisfaction with life • Mental health • General health

MAIN DETERMINANT

Use of sedation

FACTORS OF POTENTIAL INFLUENCE ON MAIN ASSOCIATION

Characteristics patients • Sex, age, religion, nationality, cause of death, place

of death, severity symptoms Characteristics relatives• Relationship with patient, sex, age, education,

religion, nationality, time passed between death and survey

Care characteristics • Involvement in care and decision- making;

information and communication, opportunity to say goodbye, presence at death

figure 1 Model analyses

Chapter 590

resUlts

Intotal,243of564relativeswhowereapproachedwerewillingtoparticipate(45%).Tworelativesdidnotanswerthequestiononwhethersedationhadbeenadministered.Therefore,atotalof241caseswereincludedinthestudy,151ofwhomconcernedrela-tivesofpatientswhohaddiedaftertheuseofpalliativesedation.Forafulldescriptionoftheinclusionofcases,aflowchartispresentedinfigure2.

Thecharacteristicsofthedeceasedpatientsandrelativesarepresentedintable1.Themedianageatdeathofnon-sedatedpatientswassomewhathigher(77years(IQR70-86))thantheageofsedatedpatients(71years(IQR62-81))(p=.002).Patientswhoreceivedsedationpriortodeathmoreoftendiedofcancer(92%)thanpatientswhodidnotreceivesedation(77%)(p=.002).Further,themedianscoreforseverityofsymp-tomsduringthe lastweekof lifewasslightlyhigher(worse) forsedatedpatients(42(IQR34-52))thanfornon-sedatedpatients(38(IQR30-46))(p=.008).Nosignificantdifferences were found with regards to sex, religion and place of death. Relatives ofsedatedpatientsweremoreoftenfemale(68%)thanrelativesofnon-sedatedpatients(51%) (p=.007). No significant differences were found between both groups withregardstorelatives’relationshiptothepatient,age,education,religion,andtimesincebereavement.

564 potential participants approached

56 refused to participate

236 no response

29 returned to sender

Deceased person was not sedated

n=91

Deceased person was

sedated n=150

243 participated (45%)

No information about whether or not sedation was administered (n=2)

Total n=241

figure 2 Flowchart inclusion participants


Table 1 Characteristics deceased and relatives (n=241)

no sedation (n=90) sedation (n=151) P value of differencea

Unadjusted

Deceased

sex Female 50 (56%) 71 (47%) .187

Male 39 (44%) 79 (53%)

Missing 1 1

age (median (IQR) (n=228) 77 (70-86) 71 (62-81) .002

religion Catholic 17 (22%) 22 (17%) .338

Christian 21 (28%) 31 (23%)

Otherb 6 (8%) 23 (17%)

No religion 32 (42%) 57 (43%)

Missing 14 18

cause of death Cancer 62 (77%) 133 (92%) .002

Otherc 19 (23%) 12 (8%)

Missing 9 6

Place of death Hospice 83 (92%) 138 (91%) .821

Hospital 7 (8%) 13 (9%)

severity of symptoms during last week of life d (median (IQR)) (n=201)

Overall score 38 (30-46) 42 (34-52) .008

Relatives

relatives relationship to patient

Married/ partner

24 (27%) 51 (36%) .581

Child 42 (48%) 54 (38%)

Parent 5 (6%) 12 (8%)

Othere 17 (19%) 26 (18%)

Missing 2 8

sex Male 44 (49%) 47 (32%) .007

Female 45 (51%) 101 (68%)

Missing 1 3

age (median (IQR)) (n=237) 57 (49-66) 59 (50-65) .465

Missing 1 3

educationf Low 37 (44%) 43 (31%) .400

Intermediate 18 (21%) 55 (39%)

High 29 (35%) 43 (31%)

Missing 6 10

religion Catholic 10 (13%) 15 (12%) .132

Protestant 20 (25%) 33 (25%)

Otherb 5 (6%) 19 (15%)

Chapter 592

Relatives of sedated patients more often stated that the professional caregiverscouldhavedonemoretomaketheperiodbeforethedeathofthepatientmorebearable(n=31, 20%) than relatives of non-sedated patients (n=7, 8%) (p=.013). Such state-mentsincludedsuggestionstoprovide‘moreinformation’,e.g.regardingthedrugsthatwereadministeredandthedurationof thesedation, toprovide ‘morecompassionedcare’,e.g.bybeingmoreattentivetorelativesorofferingalisteningearwhenneeded,andgeneral suggestions toprovide ‘better care’.Otherwise,we foundno statisticallysignificantdifferencesbetweenbothgroupsinthecharacteristicsofthecarethatwasprovidedduringthelastweekoflife(table2).Relativesofnon-sedatedpatientswereinvolvedinthecareforthepatientin74%ofthecases,andrelativesofsedatedpatientsin83%(p=.093).Relativesperceived thedegreeof their involvementas sufficient in84%and88%ofthecases,respectively(p=.357).Incaserelativesdidnotperceivethedegree of their involvement as sufficient, all except one would have preferred moreinvolvement.Relativesofnon-sedatedpatientsperceived theamountof informationtheyreceivedfromprofessionalcaregiversaboutthepatient’ssituationandcareduringthelastweekoflifeassufficientin82%ofthecases,comparedto88%oftherelativesofsedatedpatients(p=.191).Allrelativeswhodidnotperceivetheamountofinforma-tionassufficientwishedtheyhadreceivedmoreinformation.Relativesofnon-sedatedpatients had had the opportunity to say goodbye to the patient in 53% of the casesandtheyhad‘moreorless’hadtheopportunityin30%,versus66%and24%ofrela-tivesofsedatedpatients,respectively.Relativesofnon-sedatedpatientswerepresentat

table 1 Characteristics deceased and relatives (n=241) (continued)

no sedation (n=90) sedation (n=151) P value of differencea

Unadjusted

No religion 44 (56%) 63 (48%)

Missing 11 21

time since bereavement (in months)(median (IQR)) (n=234)

20 (14-31) 22 (15-32) .592

a Difference test is based on logistic regression analysis (univariable). Reference group is no sedation.b E.g. Buddhism, Hinduism, Judaismc CVA, respiratory diseases, dementia, hart failure and other diseasesd Scale consisting of 15 items with range: 1-5. Total scale score with a minimum of 15 (not severe) and a

maximum of 75 (very severe). Symptoms: pain, nausea/ vomiting, dyspnea, confusion, restlessness, de-pression, anxiety, consciousness, fatigue, loss of control, loss of dignity, burden for environment, loss of interest, hopelessness, longing for death.

e Sibling, grandchild, cousin, family in law, aunt/unclef Low= primary education + lower vocational education + lower secondary education; Intermediate= in-

termediate vocational education + upper secondary education; High= higher vocational education or university


table 2 Care characteristics (n=241)

no sedation (n=90)

sedation (n=151)

total P valuea

Relative was involved in the care for the patient by the professional caregivers in last week of life

Yes 67 (74%) 126 (83%) 193 (80%) .093

No 23 (26%) 25 (17%) 48 (20%)

Relative perceived the degree of involvement in care for the patient by the professional caregivers in last week of life as sufficient

Yes 72 (84%) 132 (88%) 204 (86%) .357

No 14 (16%) 18 (12%) 32 (14%)

Missing 4 1 5

Relative perceived the amount of information received from professional caregivers about situation patient and care for patient during last week of life as sufficient

Yes 72 (82%) 132 (88%) 204 (86%) .191

No 16 (18%) 18 (12%) 34 (14%)

Missing 2 1

Relative had the opportunity to say goodbye to the patientb

Yes 48 (53%) 98 (65%) 146 (61%) .200

More or less

27 (30%) 36 (24%) 63 (26%)

No 14 (16%) 15 (10%) 29 (12%)

Missing 1 2 3

Relatives was present at the death of the patient

Yes 50 (56%) 91 (60%) 141 (58%) .473

No 40 (44%) 60 (40%) 100 (42%)

Professional caregivers could have done something to make the period before the death of the patient more bearable for the relative

Yes 7 (8%) 31 (20%) 38 (16%) .013

No 81 (92%) 120 (80%) 201 (84%)

Missing 2 - 2

Professional caregivers could have done something to make the period after the death of the patient more bearable for the relative

Yes 5 (6%) 7 (5%) 12 (5%) .700

No 81 (94%) 143 (95%) 224 (95%)

Missing 4 1 5

a Difference test is based on logistic regression analysis (univariable). Reference group is no sedation.b Variable has been dichotomized for the logistic regression analysis

Chapter 594

thepatient’sdeathin56%,comparedto60%oftherelativesofsedatedpatients.Inallcaseswheretherelativewasnotpresentatthedeathofthepatient,theyindicatedthatanotherrelativehadbeenpresent.Intotal,6%oftherelativesofnon-sedatedpatientsstatedthattheprofessionalcaregiverscouldhavedonesomethingforthemtomaketheperiodafterthedeathofthepatientmorebearable,comparedto5%oftherelativesofsedatedpatients(p=.700).

table 3 Experience of the dying phase and wellbeing after the patient’s death: Differences between rela-tives of patients who received sedation and relatives of patients who had died a non- sudden death with-out the use of palliative sedation (n=241)

no sedation (n=90)

sedation (n=151)

Univariable regressiona multivariable regressionb

Median (IQR) Median (IQR)

p-value β (95% CI) p-value β (95% CI)

Relatives’ experience of the dying phase

Quality end of life carec

(n=241)

9 (8-9) 9 (8-10) .808 0.06 (-0.43-0.52) .888 0.04 (-0.57-0.66)

Quality of life patient last week before deathc

(n=241)

4 (2-7) 3 (2-7) .356 -0.35 (-1.12-0.30) .818 -0.13 (-1.21-0.96)

Quality of dyingc

(n=239)8 (6-8) 8 (6-8) .324 0.31 (-0.32-0.99) .273 0.46 (-0.37-1.28)

Relatives’ wellbeing after the patient’s death

life satisfaction 3 months after the death of the patientc

(n=241)

6 (4-8) 6 (4-7) .082 -0.51 (-1.06-0.09) .970 0.01 (-0.74-0.77)

current life satisfactionc

(n=241)

8 (7-8) 7 (6-8) .268 -0.26 (-0.67-0.19) .581 -0.20 (-0.90-0.51)

General healthd

(n=236)3 (3-4) 3 (3-3) .742 -0.39 (-2.69-1.92) .288 1.62 (-1.38-4.62)

mental healthe

(n=232)11 (9-14) 12 (12-14) .204 0.59 (-0.37-1.50) .371 0.61 (-0.73-1.95)

a Linear regression (univariable)b Linear regression (multivariable). Adjusted for sex patient, age patient, cause of death, severity symp-

toms, religion relative, sex relative, involvement relative in care for patient, satisfaction relative with infor-mation from caregivers, opportunity to say goodbye to patient and care for relative before the death of the patient (p<.30)

c Range 0-10 (0= terrible experience, 10= almost perfect experience)d Range 1-5 (1= poor health, 5= excellent health)e Mental wellbeing four weeks before questionnaire. Scale with 5 items. The sum score had a minimum of

0 (low mental wellbeing) and a maximum of 25 (excellent mental wellbeing)


Table 3 presents relatives’ experience of the dying phase and relatives’ wellbeingafterthepatient’sdeathfornon-sedatedandsedatedpatients.Wefoundnosignificantdifferencesbetweenbothgroupsinunivariableanalysesandmultivariablemodelsthatcontrolledforcharacteristicsofpatients,relativesandcare.Themedianscoreforthequalityofend-of-lifecarewas9inbothgroupsandthemedianscoreforqualityoflifewas4fornon-sedatedand3forsedatedpatients.Themedianscoreforqualityofdyingwas8inbothgroups.Relatives’satisfactionwiththeirownlifethreemonthsafterthepatient’sdeathwasratedatamedianof6inbothgroups,andatthetimeofthesurveythemedianscoreswere8and7,respectively.Themedianscoresonthegeneralhealthscalewere3inbothgroupsandthemedianscoresonmentalhealthwere11and12.

DiscUssion

Theaimofthisstudywastoassesswhetherrelativesofpatientswhoreceivedsedationdifferintheirexperienceofthedyingphaseandintheirwellbeingafterthepatient’sdeathcomparedtorelativesofpatientswhodiedanon-suddendeathwithouttheuseofpalliativesedation.Nosignificantdifferenceswerefoundinrelatives´assessmentsofthequalityofend-of-lifecare,patients´qualityoflifeinthelastweekbeforedeath,andtheirqualityofdying.Further,nosignificantdifferenceswerefoundinrelatives’satisfactionwith theirown life, theirgeneralhealthandtheirmentalwellbeingafterthepatient’sdeath.Apparentlytheuseofsedationdoesnothaveanegativeinfluenceonbereavedpersons’experienceofthedyingphaseoftheirdeceasedrelativeortheirownwellbeingafterthepatient’sdeath,despitethefactthatitisknownthatrelativesmayexperiencetheuseofsedationasdistressing(15,17,30).

Previousstudieshaveshownthatadequatesymptomreliefiskeytotheexperienceofa ‘gooddeath’(11,31-33).Manyrelativeshavebeendemonstratedtoevaluatetheprovision of palliative sedation to their severely suffering family member positivelybecausethepatient’ssufferingisfinallyalleviated(34).Thebenefitofpalliationcouldexplainwhynodifferenceswerefoundinrelatives’experienceofthedyingphaseandwellbeing after the patient’s death, despite the fact that relatives of sedated patientsreportsignificantlymoreseveresymptomsinthelastweekoflifethanrelativesofnon-sedatedpatients,whichhasalsobeenfoundinotherstudies(9).Further,thelossoftheabilitytocommunicatewiththepatientduringthesedation,alsosometimesreferredtoasasocialdeath(35),andapotentiallife-shorteningeffectareoftenconsideredtobekeydrawbacksoftheuseofpalliativesedationbyhealthcareprofessionals,ethicalandlegalexperts(11,12).However,intheevaluationofrelatives,adequatesymptomreliefapparentlyoutweighsthesepossibledrawbacks.Ithasbeenwellestablishedthatthewellbeingofthepatientisacrucialfactorforthehealthandwelfareofthepatient’s

Chapter 596

relatives(36).Adequatesymptomreliefforthepatientmightthereforealsobenefitthewellbeingofrelatives.

Itisknownthatprovidinginformationtofamilymembersandinvolvingthemindiscussionsaboutmedicalcareandinterventionsreducessymptomsofpost-traumaticstress, anxiety, and depression (37, 38). Our study showed that a large majority oftherelativesexperiencedtheamountofinformationtheyreceivedfromcaregiversassufficient,whichmighthavebeenan importantdeterminantof theirwellbeingafterthepatient’sdeath.Thesefindingsarenotfullyinlinewithotherstudieshighlightinginadequate provision of information and poor communication in end of life care ingeneral(19),orwithpalliativesedationinparticular(34,39).Oneexplanationcouldbethatourstudypredominantlyconcernedrelativesofpatientwhowerecaredforinahospice(91%)whichhasbeenfoundtobetheplaceofcarewherebereavedfamilymembershaverelativelyfewunmetneedsforinformation(40).

TheWorldHealthOrganization’sdefinitionofpalliativecareincorporatesasupportsystemtohelptherelativestocopeduringthepatient’s illnessandduringtheirownbereavement (41). Previous studies have shown that relatives need support for theirownwellbeingaswellastoenablethemtobeclosetoandsupportthepatient(42).Themajorityoftherelativesinourstudywassatisfiedwiththesupportthatwasprovidedtothembytheprofessionalcaregiversbeforeandafterthedeathofthepatient.However,one in five relatives of sedated patients stated that the professional caregivers couldhave done something to make the period before the death of patient more bearablefor them.Sincepalliative sedation isa far-reaching intervention thatoften followsatrajectoryof intense suffering for thepatient, caregivers shouldspecifically focusoncomforting,supportingandprovidingcontinuousinformationtothepatient’sfamilywhenpalliativesedationisbeingconsideredandwhileitisbeingadministered.

This study has several strengths. Since we conducted a study among relatives ofconsecutive patients, it is not possible that the senior clinical staff members made aspecificselectionofpatientsorrelatives.Alargesamplewasemployedwhichenhancesthe generalizability of the findings. However, our study also has some limitations.Non-responsemayhaveinfluencedourfindings.Forinstance,peoplewithmoreseverefeelingsofgriefaremorepronetonon-responsethanpersonswithlessfeelingsofgrief(43, 44). Also, relatives may differ in their interpretation of the question “Has yourrelativebeenbroughttosleepwithmedicationpriortodeath”.Nevertheless,wethinkthat this descriptive definition is less ambiguous than the term “palliative sedation”.Previousstudieshaveshownthat“laypersons”havedifferentnotionsaboutwhatthistermentails(45).Further,mostrelativeswererecruitedviaahospice.Thequalityofhospicecareisgenerallyratedhigherthanthequalityofhospitalcare(46,47),alsobyrelatives(26).However,whenwecorrectedtheanalysisforplaceofdeaththeresultsweresimilar.Therefore,itisnotexpectedthattheresultswithregardtothecomparison


between the two groups were influenced. Finally, heterogeneity in the investigatedpopulation,suchasvariabilityincausesofdeath,mayhaveloweredthepowertofinddifferencesbetweenthetwopatientgroups.

Weconclude that theuseof sedationdoesnot, in itself, seem tohaveanegativeinfluenceonbereavedpersons’experienceofthedyingphaseoftheirdeceasedrelativeorontheirownwellbeingafterthepatient’sdeath.Thiscanprobablybeexplainedbythebenefitofrelievingseveresufferingofadyingpatient.Still,palliativesedationisafar-reaching and ethically complex intervention that requires caregivers to focus onprovidingcomfort,supportandcontinuousinformationbothtothepatientandtothepatient’sfamily.

acknowleDGements

Wewouldliketothanktheseniorclinicalstaffmembersoftheparticipatingsettingsfortherecruitementofthebereavedandthebereavedwhoparticipatedinthisstudy.

Chapter 598

references


2. MiccinesiG,RietjensJA,DeliensL,PaciE,BosshardG,NilstunT,etal.Continuousdeepsedation:physicians’experiencesinsixEuropeancountries.JPainSymptomManage2006;31(2):122-9.

3. ChernyNI,RadbruchL,BoardoftheEuropeanAssociationforPalliativeC.EuropeanAssociationforPalliativeCare(EAPC)recommendedframeworkfortheuseofsedationinpalliativecare.PalliatMed2009;23(7):581-93.

4. ChambaereK,BilsenJ,CohenJ,RietjensJA,Onwuteaka-PhilipsenBD,MortierF,etal.ContinuousdeepsedationuntildeathinBelgium:anationwidesurvey.ArchInternMed2010;170(5):490-3.

5. Onwuteaka-PhilipsenBD,Brinkman-StoppelenburgA,PenningC,deJong-KrulGJ,vanDeldenJJ,vanderHeideA.Trends inend-of-lifepracticesbeforeandafter theenactmentof theeuthanasialawintheNetherlandsfrom1990to2010:arepeatedcross-sectionalsurvey.Lancet2012;380(9845):908-15.

6. SealeC.Continuousdeepsedationinmedicalpractice:adescriptivestudy.JPainSymptomManage2010;39(1):44-53.

7. AnquinetL,RietjensJA,SealeC,SeymourJ,DeliensL,vanderHeideA.ThepracticeofcontinuousdeepsedationuntildeathinFlanders(Belgium),theNetherlands,andtheU.K.:acomparativestudy.JPainSymptomManage2012;44(1):33-43.

8. BilsenJ,VanderSticheleR,BroeckaertB,MortierF,DeliensL.Changesinmedicalend-of-lifeprac-ticesduringthelegalizationprocessofeuthanasiainBelgium.SocSciMed2007;65(4):803-8.


10. Seale C. End-of-life decisions in the UK involving medical practitioners. Palliat Med 2009;23(3):198-204.

11. SteinhauserKE,ChristakisNA,ClippEC,McNeillyM,McIntyreL,TulskyJA.Factorsconsideredimportantattheendoflifebypatients,family,physicians,andothercareproviders.JAMA2000Nov15;284(19):2476-82.

12. BillingsJA,BlockSD.Sloweuthanasia.JPalliatCare.1996Winter;12(4):21-30. 13. MoritaT,IkenagaM,AdachiI,NarabayashiI,KizawaY,HonkeY,etal.Concernsoffamilymembers

ofpatientsreceivingpalliativesedationtherapy.SupportCareCancer2004;12(12):885-9. 14. Morita T, Ikenaga M, Adachi I, Narabayashi I, Kizawa Y, Honke Y, et al. Family experience with

palliative sedation therapy for terminally ill cancer patients. J Pain Symptom Manage 2004;28(6):557-65.

15. BruinsmaSM,RietjensJA,SeymourJE,AnquinetL,vanderHeideA.Theexperiencesofrelativeswith the practice of palliative sedation: a systematic review. J Pain Symptom Manage 2012;44(3):431-45.

16. MercadanteS,IntravaiaG,VillariP,FerreraP,DavidF,CasuccioA.ControlledSedationforRefrac-torySymptomsinDyingPatients.JPainSymptomManage2009;37(5):771-9.

17. vanDoorenS,vanVeluwHT,vanZuylenL,RietjensJA,PasschierJ,vanderRijtCC.Explorationofconcernsofrelativesduringcontinuouspalliativesedationoftheirfamilymemberswithcancer.JPainSymptomManage2009;38(3):452-9.


18. SteinhauserKE,VoilsCI,BosworthH,TulskyJA.Whatconstitutesqualityoffamilyexperienceattheendoflife?Perspectivesfromfamilymembersofpatientswhodiedinthehospital.PalliatSupportCare2014;8:1-8.

19. AndershedB.Relativesinend-of-lifecare-part1:asystematicreviewoftheliteraturethefivelastyears,January1999-February2004.JClinNurs2006;15(9):1158-69.

20. HudsonP,ArandaS.TheMelbourneFamilySupportProgram:evidence-basedstrategiesthatpreparefamilycaregiversforsupportingpalliativecarepatients.BMJSupportPalliatCare2014;4(3):231-7.

21. CurtisJR,PatrickDL,EngelbergRA,NorrisK,AspC,ByockI.Ameasureofthequalityofdyinganddeath.Initialvalidationusingafter-deathinterviewswithfamilymembers.JPainSymptomManage2002;24(1):17-31.

22. Cantril H. The pattern of human concern. New Brunswick, New Jersey, USA: Rutgers UniversityPress;1965.

23. Ware JE, Jr., Sherbourne CD. The MOS 36-item short-form health survey (SF-36). I. Conceptualframeworkanditemselection.Medicalcare1992;30(6):473-83.

24. VeerbeekL,vanderHeideA,deVogel-VoogtE,deBakkerR,vanderRijtCC,SwartSJ,etal.UsingtheLCP:bereavedrelatives’assessmentsofcommunicationandbereavement.AmJHospPalliatCare2008;25(3):207-14.

25. ClaessenS,FranckeA,SixmaH,BrandtH,vanderPuttenM,deVeerA,etal.OpwegnaareenCQ-indexPalliatieveZorg:meetinstrumentenontwikkeling[inDutch].Utrecht:NIVEL2009.

26. Addington-HallJM,O’CallaghanAC.AcomparisonofthequalityofcareprovidedtocancerpatientsintheUKinthelastthreemonthsoflifeinin-patienthospicescomparedwithhospitals,fromtheperspectiveofbereavedrelatives:resultsfromasurveyusingtheVOICESquestionnaire.PalliatMed2009;23(3):190-7.

27. BrueraE,KuehnN,MillerMJ,SelmserP,MacmillanK.TheEdmontonSymptomAssessmentSystem(ESAS):asimplemethodfortheassessmentofpalliativecarepatients.JPalliatCare1991;7(2):6-9.

28. Wilson KG, Graham ID, Viola RA, Chater S, de Faye BJ, Weaver LA, et al. Structured interviewassessmentofsymptomsandconcernsinpalliativecare.CanJPsychiatry2004;49(6):350-8.

29. EfronB,TibshiraniR.Introductiontothebootstrap.Hall/CRCC,ed1993. 30. MoritaT,IkenagaM,AdachiI,NarabayashiI,KizawaY,HonkeY,etal.Concernsoffamilymembers

ofpatientsreceivingpalliativesedationtherapy.SupportCareCancer2004;12(12):885-9. 31. PiersonCM,CurtisJR,PatrickDL.Agooddeath:aqualitativestudyofpatientswithadvancedAIDS.

AIDScare2002;14(5):587-98. 32. RietjensJA,vanderHeideA,Onwuteaka-PhilipsenBD,vanderMaasPJ,vanderWalG.Preferences

of the Dutch general public for a good death and associations with attitudes towards end-of-lifedecision-making.PalliatMed2006;20(7):685-92.

33. Teno JM, Casey VA, Welch LC, Edgman-Levitan S. Patient-focused, family-centered end-of-lifemedicalcare:viewsoftheguidelinesandbereavedfamilymembers.JPainSymptomManage2001;22(3):738-51.

34. BruinsmaS,RietjensJ,vanderHeideA.Palliativesedation:afocusgroupstudyontheexperiencesofrelatives.JPalliatMed2013;16(4):349-55.

35. Seale C RK, Bruinsma S, van der Heide A, Sterckx S, Mortier F, Payne S, Mathers N, Rietjens J,onbehalfof theUNBIASEDconsortium.The languageof sedation inend-of-lifecare: theethicalreasoningofcareprovidersinthreecountries.Health2014(acceptedforpublication).

36. Astedt-KurkiP,PaunonenM,LehtiK.Familymembers’experiencesoftheirroleinahospital:apilotstudy.Journalofadvancednursing1997;25(5):908-14.

Chapter 5100

37. LautretteA,DarmonM,MegarbaneB,JolyLM,ChevretS,AdrieC,etal.AcommunicationstrategyandbrochureforrelativesofpatientsdyingintheICU.NEnglJMed2007;356(5):469-78.

38. WrightAA,ZhangB,RayA,MackJW,TriceE,BalboniT,etal.Associationsbetweenend-of-lifediscussions,patientmentalhealth,medicalcareneardeath,andcaregiverbereavementadjustment.JAMA2008;300(14):1665-73.

39. BruinsmaSM,BrownJ,vanderHeideA,DeliensL,AnquinetL,PayneSA,etal.Makingsenseofcontinuoussedationinend-of-lifecareforcancerpatients:aninterviewstudywithbereavedrelativesinthreeEuropeancountries.SupportCareCancer.2014;22(12):3243-52.

40. GallagherR,KrawczykM.Familymembers’perceptionsofend-of-lifecareacrossdiverselocationsofcare.BMCPalliatCare2013;12(1):25.

41. World Health Organization. ‘Definition of palliative care’, http: //www.who.int/cancer/palliative/definition/en(2008,accessedJanuary2015).

42. Andershed B, Ternestedt BM. Involvement of relatives in care of the dying in different care cul-tures:developmentofatheoreticalunderstanding.Nursingsciencequarterly1999;12(1):45-51.

43. StroebeM,StroebeW.Whoparticipatesinbereavementresearch?Areviewandempiricalstudy.JDeathDying1989;20:1-29.

44. SwarteNB,vanderLeeML,vanderBomJG,vandenBoutJ,HeintzAP.Effectsofeuthanasiaonthebereavedfamilyandfriends:acrosssectionalstudy.BMJ2003;327(7408):189.

45. vanDeldenJJM,VanDerHeideA,vandeVathorstS,WeyersH,vanTolD.Knowledgeandopinionsof the general public and professionals on medical decision making and treatment in the end oflife, theKOPPEL-research [inDutch].ZonMw2011.http://www.zonmw.nl/uploads/tx_vipublica-ties/ZonMw_A4_RapportKOPPELDEF_klein.pdf(accessed13Sep2013).

46. GreerDS,MorV,MorrisJN,SherwoodS,KidderD,BirnbaumH.Analternativeinterminalcare:resultsoftheNationalHospiceStudy.JChronicDis1986;39(1):9-26.

47 KaneRL,WalesJ,BernsteinL,LeibowitzA,KaplanS.Arandomisedcontrolledtrialofhospicecare.Lancet1984;1(8382):890-4.

Chapter 6Estimating the potential life-shortening effect of continuous sedation until death: a comparison between two approaches

Bruinsma SM, Rietjens JA, Swart SJ, Perez RS, van Delden JJ, van der Heide A

Journal of Medical Ethics 2014; 40(7): 458-62.

Chapter 6104

aBstract

Context.Insomecases,physiciansestimatethatcontinuoussedationuntildeathmayhavealife-shorteningeffect.Theaccuracyoftheseestimationscanbequestioned.

Aim. The aim of this study is to compare two approaches to estimate the potentiallife-shorteningeffectofcontinuoussedationuntildeath.

MethodsIn2008,370Dutchphysiciansfilledoutaquestionnaireandreportedontheirlastpatientwhoreceivedcontinuoussedationuntildeath.Thepotentiallife-shorteningeffectofcontinuoussedationwasestimatedthroughadirectapproach(question:Didcontinuoussedation,accordingtoyourestimation,hastenthepatient’sdeath?Ifyes:byhowmuchtime?)andanindirectapproach(estimatedlifeexpectancyminusdurationofsedation).Theintrarateragreementbetweenbothapproacheswasdeterminedwithaweightedκ.

Results.Accordingtothedirectapproach,sedationmighthavehadalife-shorteningeffectin51%ofthecasesandaccordingtotheindirectapproachin84%.Theintrarateragreementbetweenbothapproacheswasfair(weightedκ=0.38).In10%ofallcases,thedirectapproachyieldedhigherestimatesoftheextenttowhichlifehadbeenshortened;in58%ofthecases,theindirectapproachyieldedhigherestimates.

Conclusions.Theresultsshowadiscrepancybetweendifferentapproachestoestimatethepotentiallife-shorteningeffectofcontinuoussedationuntildeath.

Estimating the potential life-shortening effect of CSD 105

introDUction

A significant minority of dying people experience serious symptoms that are unre-sponsivetoconventionaltherapies.Insuchcircumstances,palliativesedationmaybeconsidered.Palliativesedationisthedeliberateloweringofapatient’slevelofconscious-nessinthelaststagesoflife(1).Sedationcanbeusedintermittentlyorcontinuouslyuntildeath,and thedegreeof sedationnecessary to relieve sufferingmayvary fromsuperficialtodeep(2).Continuoussedationismostfrequentlyusedinpatientssuffer-ingfromphysicalsymptomssuchasdelirium,dyspnoea,painandnausea(3,4).Themoralstatusofcontinuoussedationuntildeathhasbeenthesubjectoffierceethicaldebate (5). This debate mostly focuses on whether continuous sedation may hastendeath.Itisstatedinguidelinesthatphysiciansshouldusesedationwiththeintentiontorelievesuffering,andnotwiththeintentiontoshortenthepatient’slife(1,6).Further,toprecludeapotentiallife-shorteningeffect,itisrecommendedtorestricttheuseofcontinuoussedationtopatientswithanestimatedlifeexpectancyofatmost2weeks(1, 7, 8). When the patient’s life expectancy is within this limit and when sedativesareproperlydosed,continuoussedationuntildeathhaspresumablynolife-shorteningeffect(1,3,8).

Theunderlyingassumptioninguidelinesisthatphysicianscanestimateapatient’slifeexpectancywithsufficientaccuracy.However,estimatinglifeexpectancyofpatientswithadvanceddiseaseisknowntobeverydifficult(7,9,10).Physicianstendtoover-estimatesurvival:ithasbeenshownthatsurvivalofpatientsistypically30%shorterthanpredictedbyphysicians,butthattheaccuracyofphysicians’predictionsincreaseswhendeathapproaches(10).

Althoughit isassumedinguidelinesthatcontinuoussedationuntildeathhasnolife-shortening effect when used proportionally (1), physicians may have a differentperspective. In a large- scale nationwide follow-up study, performed in 2005 in theNetherlands,physiciansestimatedthatcontinuoussedationuntildeathmighthavehadalife-shorteningeffectin26%ofthecases(3).In20%ofthesecases,itwasestimatedthatsedationhadshortenedthepatient’slifebylessthanoneweek,in4%ofthecasesbylessthanonemonthandin2%bymorethanonemonth.Theseestimatesshouldbeinterpretedcautiously.Ontheonehand,physiciansmightbecautiouswhentheyaredirectly asked to estimate the potential life-shortening effect of continuous sedationuntildeath.Thesensitivityofthetopicmighthaveledtosociallydesirableanswersandphysicians might have been reluctant to state their true opinions (11). For instance,becausetheymightfeel thata life-shorteningeffect isundesirableandmorallycom-plex,andbecauseacknowledgingsuchaneffectmaysuggestthatapatient’sdeathwasactivelyhastened.Ontheotherhand,physiciansareknowntobeinclinedtooveresti-matethelifeexpectancyofpatientswithadvanceddisease(9).Ithasbeenshownthat

Chapter 6106

survivalofpatientsistypically30%shorterthanpredictedbyphysicians,butthattheaccuracy of physicians’ predictions increases when death approaches.9 The fact thatphysicians commonly have difficulty in estimating life expectancy and often are in-clinedtooverestimationmightalsoholdtrueforthelife-shorteningeffectsofsedativemedication inpatientswitha limited lifeexpectancy(12).Obviously,estimating thetruelife-shorteningeffectofcontinuoussedationwouldrequireanexperimentalstudy,whichis,however,notanoptionforthispatientgroup.

Theaimofthisstudythereforeistogetinsightintheaccuracyofestimatesofthelife-shortening effect of continuous sedation until death by comparing two differentapproaches.Wecompareadirectapproach,whereweaskphysicians toestimate thelife-shortening effect of continuous sedation until death, and an indirect approach,whereweaskthephysicianstoestimatethepatient’slifeexpectancyandrelatethattothedurationofthesedation.

methoDs

study Design and Data collection

AsecondaryanalysiswasperformedofdatathatwerecollectedamongphysiciansinastudythatevaluatedthepracticeofpalliativesedationaftertheintroductionoftheRoyalDutchMedicalAssociationguideline(13-14).DatacollectiontookplacebetweenFebruary2008andSeptember2008.Forthisstudy,astructuredquestionnairewassenttoarandomsampleof1580physicians:1128inthenorth-westernandsouth-westernregionsoftheNetherlands(generalpractice,n=466;nursinghome,n=195;andhospital,n=467)and452generalpractitionersinthenorth-easternregion(13).Physicianswereaskedtoreportonthelastpatientforwhomtheyhadbeenresponsibleforprovidingcontinuous sedation until death. One of the issues addressed was the potential life-shorteningeffectofcontinuoussedationuntildeath.Physicianswereaskedwhetherornottheuseofcontinuoussedationuntildeathmighthavehadalife-shorteningeffect(‘Didcontinuoussedation,inyourestimation,hastenthepatients’death?’);toestimatethepatient’slifeexpectancyatthestartofsedation(‘What,inyourestimation,wasthepatient’slifeexpectancy,atthetimecontinuoussedationwasstarted?’);andtoassesstheactualdurationofthesedation(question:‘Howlongafterthestartofcontinuoussedationdidthepatientdie?’)(table1).

analysis

We used a direct and an indirect approach to estimate the potential life-shorteningeffectofcontinuoussedationuntildeath.Thedirectapproachwasbasedonthedirectquestion about the life-shortening effect of sedation. Response categories ranged


tabl

e 1

Two

appr

oach

es o

f est

imat

ing

the

life-

shor

teni

ng e

ffect

of c

ontin

uous

sed

atio

n un

til d

eath

Dir

ect a

ppro

ach

indi

rect

app

roac

h

Life

exp

ecta

ncy

−

Dur

atio

n se

datio

n=

“Did

the

use

of c

onti

nuou

s sed

atio

n in

yo

ur e

stim

atio

n sh

orte

n lif

e?”

“Wha

t was

in y

our e

stim

atio

n th

e pa

tient

’s lif

e ex

pect

ancy

at t

he st

art o

f co

ntin

uous

seda

tion”

“How

long

aft

er th

e st

art o

f the

seda

tion

did

the

patie

nt p

ass a

way

?”li

fe e

xpec

tanc

y M

INU

S du

rati

on

• <

1 da

y sh

orte

ned

• 1-

2 da

ys sh

orte

ned

• 3-

6 da

ys sh

orte

ned

• 1-

2 w

eeks

shor

tene

d•

2 w

eeks

- 1 m

onth

shor

tene

d•

> 1

mon

th sh

orte

ned

• D

on’t

know

• Co

ntra

dict

ory

• <

1 da

y sh

orte

ned

• 1-

2 da

ys sh

orte

ned

• 3-

6 da

ys sh

orte

ned

• 1-

2 w

eeks

shor

tene

d•

2 w

eeks

- 1 m

onth

shor

tene

d•

> 1

mon

th sh

orte

ned

• Co

ntra

dict

ory

• W

eeks

, day

s and

/or h

ours

• >

3 da

ys p

rolo

nged

• 3

days

pro

long

ed

•

2 da

ys p

rolo

nged

• 1

day

prol

onge

d

Co

mbi

ned:

no

life

shor

teni

ng•

< 1

day

shor

tene

d•

1-2

days

shor

tene

d•

3-6

days

shor

tene

d•

1-2

wee

ks sh

orte

ned

• 2

wee

ks- 1

mon

th sh

orte

ned

• >

1 m

onth

shor

tene

d•

Don

’t kn

ow•

Cont

radi

ctor

y

Chapter 6108

between‘nolife-shortening’and‘morethanonemonthshortened’.Withtheindirectapproach,thedurationofthesedationuntildeathwassubtractedfromtheestimatedlife expectancy at the start of sedation. The estimated life expectancy consisted ofcategories varying from ‘less than one day’ to ‘more than one month’. The questionaboutthedurationofthesedationwasansweredinweeks,daysand/orhours.Tomakestatisticalcalculationspossible, lifeexpectancywasrecodedintohoursbytakingthemiddleofcategories:forexample,‘lessthanoneday’wasrecodedas‘12h’.Thediffer-encebetweentherecodedestimatedlifeexpectancyandthedurationofthesedationwascalculated(table1).

A totalof606physicians (38%)filledout thequestionnaire.Responserateswere43%forgeneralpractitioners,50%fornursinghomesphysiciansand24%forclinicalspecialists.Oftherespondingphysicians,370(61%)reportedabouttheirlastcase(11).Physicianswereonaverage49-years-old(range28–64).Amajorityofthephysiciansweremen(64%).Ofallphysicians,15%workedinahospital,17%inanursinghomeor hospice and 68% were general practitioners. Physicians had on average 19 yearsworkexperience(range1–38).Thepatientsphysiciansreportedonwereonaverage70yearsold(range3–99)atthetimeoftheirdeath.Themajorityofthepatients(71%)hadcancerastheirmaindiagnosis.Themajorityofthepatientsweremale(52%)(table2).

Physicianswereexcludedfromtheanalyses if theanswertooneof thequestionswas‘unknown’,‘don’tknow’oruninterpretable(morethanoneanswer).Theintra-rateragreementbetweenthedirectandindirectapproacheswasdeterminedbycalculatingaweightedκ:quadraticweightswereusedtotakeintoaccountthesizeofdifferencesbetweenbothapproaches.ThescoreswereinterpretedusingtheLandisandKochcri-teria:<0.00=pooragreement,0.00–0.20=slightagreement,0.21–0.40=fairagreement,0.41–0.60=moderate agreement, 0.61–0.80=substantial agreement and 0.81–1.00=al-mostperfectagreement(13).Wealsocalculatedweightedκvaluesfortwosubgroupstoseewhetherthelevelofagreementwasdifferentforpatientswitharelativelylongestimatedlifeexpectancyandpatientswitharelativelyshortestimatedlifeexpectancy.Acut-offpointof1weekwasusedforthisanalysis,basedonthemedianvalueoftheestimatedlifeexpectancy.Thesignificancelevelwassetat5%.Fortheanalysis,ExcelandSPSSV.22.0wereused.

resUlts

In269cases(74%),physiciansestimatedthatthepatienthadalifeexpectancyoflessthan1week,in84cases(23%)between1and2weeks,andin10cases(3%)morethan2weeksatthestartofsedation.Thedurationofthesedationvariedbetweenlessthan1day,and1and2weeks(table2).


table 2 Characteristics of responding physicians and patients and sedation characteristics (N=370)

Variable categories mean, range n %

Respondents

Age 49 (28-64)

Sex

Male 237 64

Female 132 36

Work experience 19 (1-38)

Setting

Home 250 68

Nursing home/hospice 64 17

Hospital 56 15

Patients

Age 70 (3-99)

Sex

Male 189 52

Female 175 48

Primary diagnoses Cancer 251 71

Sedation characteristics

Life expectancy at start of sedationa

< 1 day 14 4

1-2 days 116 32

3-6 days 139 38

1-2 weeks 84 23

2 weeks–1 month 9 3

> 1 month 1 0

Duration sedationb

< 1 day 115 32

1-2 days 116 32

3-6 days 115 32

1-2 weeks 15 4

2 weeks – 1 month 0 0

> 1 month 0 0

a Missing: n=7 b Missing: n= 9

Chapter 6110

table 3 Estimated life-shortening effect (direct approach), life expectancy, duration of sedation and the estimated life-shortening effect (indirect approach) (N=370)

estimated life- shortening effect according to the direct approach (n, %)a

estimated life- shortening effect according to the indirect approach (n, %)b

No life-shortening effect 148 (41) 56 (16)

< 1 day 38 (11) 78 (22)

1-2 days 54 (15) 50 (14)

3-6 days 31 (9) 105 (30)

1-2 weeks 23 (6) 58 (16)

2 weeks – 1 month 5 (1) 8 (2)

> 1 month 0 (0) 0 (0)

Don’t know 64 (17) NA

a Missing: n=7b Missing: n=15

table 4 The estimated life-shortening effect of continuous sedation (direct en indirect approach) (n=289a,b)

life shortening effect of continuous sedation until death (direct approach)

Not shortened

< 1 day shortened

1-2 days shortened

3-6 days shortened

1-2 weeks shortened

2 weeks- 1 month shortened

Total

life shortening effect of continuous sedation until death (indirect approach)

Not shortened

35(12%)

7(2%)

3(1%)

0(0%)

1(0%)

1(0%)

47(16%)

< 1 day shortened

40(14%)

11(4%)

9(3%)

2(1%)

0(0%)

0(0%)

62(22%)

1-2 days shortened

21(7%)

8(3%)

9(3%)

2(1%)

0(0%)

1(0%)

41(14%)

3-6 days shortened

32(11%)

11(4%)

17(6%)

17(6%)

6(2%)

0(0%)

83(29%)

1-2 weeks shortened

13(5%)

0(0%)

13(5%)

9(3%)

14(5%)

0(0%)

49(17%)

2 weeks- 1 month shortened

1(0%)

1(0%)

0(0%)

0(0%)

2(1%)

3(1%)

7(2%)

Total 142(49%)

38(13%)

51(18%)

30(10%)

23(8%)

5(2%)

289(100%)

a After excluding cases where the answer to one of the questions was ‘don’t know’, uninterpretable or miss-ing

b Weighted κ (intra-rater agreement): 0.38 (p=0.000)


Whenaskeddirectly,148physicians(41%)statedthatcontinuoussedationdidnothavea life-shorteningeffect,151physicians(42%)statedthattheremighthavebeenalife-shorteningeffectand64physicians(17%)didnotknow.Usingtheindirectap-proach,theuseofsedationmighthavehadalife-shorteningeffectin84%andnosucheffectin16%ofthecases(table3).

Afterexcludingcaseswhere theanswer tooneof thequestionswas ‘don’tknow’,uninterpretable or missing, 289 cases remained available for analysis of the level ofagreementbetweenthedirectandindirectapproachesofassessingtheestimatedlife-shorteningeffectofsedation.Whenaskeddirectly,147physicians(51%)estimatedapotentiallife-shorteningeffectofcontinuoussedation(table4).In13%ofthesecases,itwasestimatedthatsedationmighthaveshortenedthepatient’slifebylessthan1day;in28%ofthecases,between1dayand1week;in8%between1and2weeks;andin2%bymorethan1month.Thedurationofthesedationwasshorterthantheestimatedlifeexpectancyatthestartofsedationaccordingto242physicians(84%).So,followingthisindirectapproach,itwasestimatedthattheuseofsedationmighthavehadalife-shorteningeffectin84%ofthecases(table4).In22%ofthesecases,itwasestimatedthatsedationmighthaveshortenedthepatient’slifebylessthan1day;in43%ofthecases,between1dayand1week;in17%between1and2weeks;andin2%bymorethan1month.In10%ofthesecases,thedirectapproachyieldedhigherestimatesoftheextenttowhichlifehadbeenshortenedthantheindirectapproach;in58%ofthecases, the indirect approachyieldedhigherestimatesand therewasnodifference in31%ofthecases.Thelevelofagreementbetweenthedirectandindirectapproachesasassessedbyweightedκwas0.38,indicating‘fair’agreement(table4).Furtheranalysisshowedthatthelevelofagreementwassomewhathigherforpatientswithanestimatedlifeexpectancyoflessthanaweek(weightedκ=0.26)ascomparedwithpatientswithalifeexpectancyofmorethan1week(weightedκ=0.10).

DiscUssion

In this study, we found that in 51% of the cases Dutch physicians estimate that theuseofcontinuoussedationuntildeathmighthavehada life-shorteningeffectwhenthey are directly asked about such an effect. In contrast, on the basis of physicians’estimations of patients’ life expectancy at the start of sedation and the duration ofsedation until death, it can be estimated that such an effect occurs in 84% of cases.The finding of this substantial discrepancy between the two approaches to estimatethe life-shortening effect of continuous sedation until death confirms the difficultyofpredictingthelifeexpectancyofpatientswithadvanceddisease,andofestimatingthepotential life-shorteningeffectofend-of-life interventions.However,ourfinding

Chapter 6112

thatthereis‘fair’agreementbetweenbothapproachessuggeststhatthisdiscrepancyisnotmerelytheresultofrandominaccuracy.Ingeneral,physiciansmightberelativelycautiouswhen theyaredirectlyasked toestimate thepotential life-shorteningeffectof continuous sedation until death, because they feel that such an effect is undesir-able and morally complex, and because admitting such an effect may suggest that apatient’sdeathwasactivelyhastened.Ontheotherhand,physiciansareknowntobeinclined to overestimate the life expectancy of patients with advanced disease (10),whichmayextendtopatientswhoareprovidedwithcontinuoussedationuntildeath.Suchinclinationcouldexplainthehighproportionofcasesinwhichsedationmighthaveshortenedlife,whentheestimateisbasedonour‘indirect’approach.

Our finding that the agreement between the two approaches to estimate the po-tentiallife-shorteningeffectofcontinuoussedationwashigherforpatientswithalifeexpectancyoflessthan1weekconfirmsfindingsfrompreviousstudiesthatestimationsoflifeexpectancybecomemoreaccuratewhendeathapproaches(10).Animplicationcouldbe,ifphysiciansindeedconsistentlyoverestimatetheirpatients’lifeexpectancy,thatinsomecaseswheresedationwouldbeabeneficialintervention,itisstartedtoolateornotatall,whichwouldinvolveunnecessarysuffering.

It isstrikingthat,accordingtobothapproaches,physiciansoftenthinkthatcon-tinuoussedationuntildeathcanhavea(mostlylimited)life-shorteningeffect.Severalempiricalstudieshavesuggestedthatsedationasusedinclinicalpracticehasnosig-nificantlife-shorteningeffect(15-20).Itcan,ineitheroftheapproaches,bequestionedif physicians tend to overestimate the life-shortening potential of sedation, but ourdataallownofirmconclusionhere.Theimportanceofapotentiallife-shorteningeffectofcontinuoussedationuntildeathcanalsobequestioned.TheRoyalDutchMedicalAssociationguidelinearguesthatthelifeexpectancyofamaximumof2weeksiscondi-tionalforpalliativesedation(1).TheauthorsoftheEuropeanAssociationofPalliativeCareframeworkevenstatedthatsedationshouldonlybecontemplatedifthepatientishoursordaysfromdeath(6).However,onecouldholdthepositionthatpalliationwithlife-shorteningsideeffectsismorallyjustified,aslongasproportionatelyconsequentialreasonsarepresent.The indication for sedationoriginates from thepresenceofoneor more refractory symptoms that lead to severe and unbearable suffering. In suchcircumstances, the life expectancy criterion can be weighed against the severity ofrefractorysymptoms(21).Ifthebenefitofpalliationoutweighstheharmofanearlierdeath, if there are no other alternatives and if dosages are titrated according to thepatient’sneed,palliativesedationmaybeindicated,evenifdeathisnotimminent(22).Insomecaseswheretheconditionsforcontinuoussedationuntildeatharenotmet,brieforintermittentsedationmaybeapossiblealternative(1).

Ourstudyhadsomelimitations.First,thephysiciansinourstudyprovidedinfor-mationretrospectively.Thedatacouldthereforebeinfluencedbyrecallbias.Further,


weaskedaboutthephysician’smostrecentcase,whichmaynotalwaysrepresentphysi-cians’usualpracticesorapproaches(13).Third,theriskthatphysiciansmatchedtheiranswersandestimationsinthequestionnairecannotberuledout.Thetruedifferencesbetweenthetwoapproachesmightthereforebelargerthansuggestedinourstudy.Toimproveaccuracy, itmightusefultoincludepatientsprospectivelyinafuturestudy:lifeexpectancycouldthenbeestimatedbeforethestartofsedation,andthedurationofsedationcouldbetimed.Forthisstudy,asecondaryanalysiswasperformedofdatathat were collected in 2008. Therefore, this approach could not be followed in thepresentstudy.Itcanbeconcludedthatestimatingthelifeexpectancyofpatientswhoare provided with continuous sedation until death is difficult. Recommendations inguidelinesthatcontinuoussedationuntildeathshouldonlybeusedforpatientswithalifeexpectancyoflessthan1or2weeksmaythereforebedifficulttotranslatetoclinicalpractice.Inresearch,thetypeofquestionthatisusedtoestimatethelife-shorteningeffectofsedationandotherend-of-lifeinterventionshastobetakenintoaccountwheninterpretingtheresults.Basedonthisstudy,wecannotconcludewhetheradirectques-tionoramoreveiledapproachispreferable.

acknowleDGements

Asecondaryanalysiswasperformed,ofthedatathatwascollectedamongphysiciansin a study that evaluated the practice of palliative sedation after the introduction oftheRoyalDutchMedicalAssociationguideline(theAmsterdamRotterdamSedationproject).Wewouldliketothankalltherespondentsforfillingoutthequestionnaires.

Chapter 6114

references

1. Royal Dutch Medical Association (RDMA). Guideline for palliative sedation [in Dutch]. 2009.Available from http://knmg.artsennet.nl/Publicaties/KNMGpublicatie/Guideline-for-palliative-sedation-2009.htm.AccessedMay,2012.

2. de Graeff A, Dean M. Palliative sedation therapy in the last weeks of life: a literature review andrecommendationsforstandards.JPalliatMed2007;10(1):67-85.



5. vanDeldenJJ.Terminalsedation:sourceofarestlessethicaldebate.JMedEthics2007;33(4):187-8. 6. ChernyNI,RadbruchL,BoardoftheEuropeanAssociationforPalliativeC.EuropeanAssociation

forPalliativeCare(EAPC)recommendedframeworkfortheuseofsedationinpalliativecare.PalliatMed2009;23(7):581-93.

7. RietjensJA,BuitingHM,PasmanHR,vanderMaasPJ,vanDeldenJJ,vanderHeideA.Decidingabout continuous deep sedation: physicians’ perspectives: a focus group study. Palliat Med 2009;23(5):410-7.


9. BrandtHE,OomsME,RibbeMW,vanderWalG,DeliensL.Predictedsurvivalvs.actualsurvivalinterminallyillnoncancerpatientsinDutchnursinghomes.JPainSymptomManage2006;32(6):560-6.

10. GlareP,VirikK,JonesM,HudsonM,EychmullerS,SimesJ,etal.Asystematicreviewofphysicians’survivalpredictionsinterminallyillcancerpatients.BMJ.2003;327(7408):195-8.

11. AnquinetL,RausK,SterckxS,SmetsT,DeliensL,RietjensJA.Similaritiesanddifferencesbetweencontinuoussedationuntildeathandeuthanasia-professionalcaregivers’attitudesandexperiences:Afocusgroupstudy.PalliatMed2012;27(6):553-61

12. Hasselaar JG, Verhagen SC, Wolff AP, Engels Y, Crul BJ, Vissers KC. Changed patterns in Dutchpalliative sedationpracticesafter the introductionofanationalguideline.ArchInternMed2009;169(5):430-7.

13. SwartSJ,RietjensJA,vanZuylenL,ZuurmondWW,PerezRS,vanderMaasPJ,etal.Continuouspalliativesedationforcancerandnoncancerpatients.JPainSymptomManage2012;43(2):172-81.

14. SwartSJ,RietjensJAC,BrinkkemperT,vanZuylenC,PerezRSG,vanderHeideA.ContinuousPal-liativeSedationuntildeath:PracticeafterintroductionoftheDutchnationalguideline.BMJPalliatSupportCare2012;2:256-63.

15. ChiuTY,HuWY,LueBH,ChengSY,ChenCY.SedationforrefractorysymptomsofterminalcancerpatientsinTaiwan.JPainSymptomManage2001;21(6):467-72.

16. InghelbrechtE,BilsenJ,MortierF,DeliensL.ContinuousdeepsedationuntildeathinBelgium:asurveyamongnurses.JPainSymptomManage2011;41(5):870-9.

17. MercadanteS,IntravaiaG,VillariP,FerreraP,DavidF,CasuccioA.Controlledsedationforrefrac-torysymptomsindyingpatients.JPainSymptomManage2009;37(5):771-9.

18. MoritaT,TsunodaJ, InoueS,ChiharaS.Effectsofhighdoseopioidsandsedativesonsurvival interminallyillcancerpatients.JPainSymptomManage2001;21(4):282-9.


19. RietjensJA,vanZuylenL,vanVeluwH,vanderWijkL,vanderHeideA,vanderRijtCC.Palliativesedationinaspecializedunitforacutepalliativecareinacancerhospital:comparingpatientsdyingwithandwithoutpalliativesedation.JPainSymptomManage2008;36(3):228-34.

20. SykesN,ThornsA.Sedativeuseinthelastweekoflifeandtheimplicationsforend-of-lifedecisionmaking.ArchInternMed2003;163(3):341-4.

21. Swart S. The Practice of Palliative Sedation in the Netherlands after introduction of the NationalGuideline(thesis).Rotterdam:ErasmusMC,UniversityMedicalCentreRotterdam;2013.

22. JanssensR,vanDeldenJJ,WiddershovenGA.Palliativesedation:notjustnormalmedicalpractice.EthicalreflectionsontheRoyalDutchMedicalAssociation’sguidelineonpalliativesedation.JMedEthics2012;38(11):664-8.

Chapter 7Risk factors for complicated grief in older adults

Bruinsma SM, Tiemeier HW, Verkroost- van Heemst J, van der Heide A, Rietjens JAC

Journal of Palliative Medicine (accepted for publication)

Chapter 7118

aBstract

Background. Olderadultsgrievingthedeathofaspousehavebeenfoundtohaveahigherriskofcomplicatedgriefcomparedwithyoungeradults.

Objective.Tofindoutwhetherpersonalcharacteristicsofthepatientandthebereavedpartner, or characteristics of the patient’s illness, end-of-life care and the nature ofdeathareriskfactorsforcomplicatedgriefinolderadults.

Design. Weperformedanestedcase-controlstudywithintheRotterdamStudy.

Subjects. We selected 100 couples of which one person had deceased and the otherpersonexperienced‘complicatedgrief ’,and100controlcouplesofwhichonepersonhaddeceasedandtheotherpersonexperienced‘normalgrief ’.

Measurements.Complicatedgriefwasassessedwitha17-itemInventoryofCompli-catedGrief.DeterminantswereassessedusingseveralsourcesofinformationthatwereavailableforallparticipantsoftheRotterdamStudy.Additionally,medicalfilesofthedeceasedweremanuallyscreened.Logisticregressionanalysiswasperformed.

Results. Only depression at baseline was significantly associated with complicatedgrief.Bereavedpartnerswithdepressionatbaselinehadahigherriskofcomplicatedgriefcomparedtobereavedpartnerswithoutdepression(OR=3.48;95%CI=1.40-8.68).

Conclusions. Ourresultssuggestthatcomplicatedgriefinolderadultsisnotclearlyrelatedtothecircumstancesofdyingofthedeceasedpartner.Pre-existingconditionssuchasdepressionseemtobemoreimportantinexplainingtheoccurrenceofcom-plicatedgrief.

Risk factors for complicated grief in older adults 119

introDUction

Bereavementisacommonexperienceinolderadults.Althoughthemajorityofadultsrecover after the loss of a loved one, a portion continues to grieve for an extendedperiodof timeanddevelops symptomsof a stateknownas complicatedgrief (1)orProlonged Grief Disorder (2). This type of grief is distinct from normal grief as thepersoncannotacceptthedeathandinsteadexperiencesdisbeliefandpreoccupationswith thedeceasedperson(1,3).Complicatedgriefcanbeassociatedwithanumberofnegativehealthoutcomes, suchashypertension, sleep impairmentandsuicidality(4-6).Amongolderadults,complicatedgriefisunderdiagnosed,minimizedasafactoraffectingmentalhealthandfunction,andundertreated(3,7,8).Althoughthebereave-mentexperienceofolderadultshasbeenassociatedwithlessmourning(9),secondaryconsequencessuchassocialisolationmayleadtogriefoflongerdurationandpoorerhealth and mental health outcomes than observed in younger persons. Older adultsgrievingthedeathofaspousehavebeenfoundtohaveahigherriskofcomplicatedgriefcomparedwithyoungeradults(7,8).ADutchstudyshowedthattheprevalenceofcomplicatedgriefinolderadultsinthegeneralpopulationis4.8%(1).

Knowledgeaboutvulnerability tocomplicatedgrief canprovideopportunities totargetcareandresourcesappropriately(10).Theliteratureaboundswithfactorsthatareconsideredusefulinidentifyingwhoisatriskofcomplicatedgrief(10-14).Agreatdeal of attention is paid to the characteristics of the bereaved, such as age, sex andphysicalandmentalillness(10).Anotherareaofinterestisthedeceased’sillness,thecharacteristicsofendoflifecareandthenatureofthepatient’sdeath.(10)Ithasbeenshownthatthereisanincreasedriskofcomplicatedgriefifthedurationofthepatient’sterminalillnesswaseitherveryshortorverylong(12);ifthepatienthadsufferedfromacognitiveimpairment(12);ifthepatientdiedinahospitalsetting(13);ifthepatientdiedaftereuthanasia(14);andifdeathoccurredsuddenlyorunexpectedly(10).Mul-tiplecaretransitionsandhospitalisationsinthelastphaseoflifecouldalsoinfluencegriefexperiences(15).However,littleisknownaboutthesecharacteristicsaspotentialriskfactorsforcomplicatedgriefinolderadults.Sincethebereavementexperienceofolderadultsisdifferentfromyoungeradults,thispotentiallyalsoholdstrueforfactorsthatcontributetothisbereavementexperience.Toexaminewhethercomplicatedgriefinolderadultscanbeexplainedbypredeathinformation,thisstudyincludesbothper-sonalandsituationalfactorsinacomprehensiveanalysis.Theaimofthepresentstudyistofindoutwhetherpersonalcharacteristicsofthepatientandthebereavedpartner,orcharacteristicsofthepatient’sillness,end-of-lifecareandthenatureofdeathareriskfactorsforcomplicatedgriefinolderadults.

Chapter 7120

DesiGn anD methoDs

This study was based upon the Rotterdam Study, an ongoing prospective cohort ofolderadultstoexaminetheoccurrenceandriskfactorsofchronicdiseases(16).

the rotterdam study (1)

TheRotterdamStudycomprisestwocohorts.Thefirststemsfromtheoriginalstudywhichcommencedin1990–1993.Atthistimeallinhabitantsagedover55yearslivingintheOmmoorddistrictofRotterdamwereinvitedtoparticipate;7983persons(78%)participated.In2000,peoplewhohadbecome55yearsofage,orwhowere55yearsoroverandhadmovedintothestudydistrictafterthestartofthestudy,wereaddedasasecondcohort;3011(67%)participated.AdetaileddescriptionofthedesignoftheRotterdamstudyhasbeenpublishedelsewhere(17).

nested case-control

Lossofaspouseisverycommonandaccountsforalargeproportionoflossesamongolderadults (18).Therefore,weselected200(married/partnership) fromtheRotter-damcohorts, inanestedcase-controldesign (19):100couplesofwhichonepersonhaddeceasedandofwhichtheotherexperiencedcomplicatedgrief,and100controlcouplesofwhichonepersonhaddeceasedandtheotherpersonexperienced‘normal’grief.Groupmatchingwasusedtoincreasestatisticalpower.

case definition

Incaseaparticipanthaslostaspouse,griefwasassessedintheoriginalcohortinthefourth follow-up examination (2002– 2004) and in the added cohort in the secondfollow-upexamination(2004–2005)(1).Complicatedgriefwasassessedwitha17-itemDutchversionoftheInventoryofComplicatedGrief(ICG),constructedbyPrigersonetal.(1995)(20).TheICGisthemostwidelyusedinstrumenttomeasurecomplicatedgriefanditemsrepresent thearrayofsymptomsattributedtocomplicatedgrief.Themeasurehashighinternalconsistencyandconvergentandcriterionvalidityanditisconsidered the ‘gold standard’ for measurement of complicated grief in older adults(1).AsummaryscorefortheICGwascalculatedbyaddingupeachindividual itemscore(responsesfrom0=neverto4=always)acrossthe17-items,providingapotentialtotal score rangeof0 to68.Participantswitha total scoregreater than21andwithsymptomsreportedtohavebeenpresentforatleastsixmonthswereconsideredtohavecomplicatedgrief.Intotal,1089(19%)participantsreportedthattheywereexperienc-inggriefatthetimeoftheassessmentandofthese,277participantswereassessedashavingcomplicatedgrief.Adetaileddescriptionof themeasurementofcomplicatedgriefhasbeenpublishedelsewhere(1).


assessment of determinants

Several sources of information were available for all participants of the RotterdamStudy:homeinterviews,examinationsattheresearchcenter,NationwideMedicalReg-istry,andgeneralpractitioners’records.Adetaileddescriptionofthemethodsfordatacollection of the Rotterdam study has been published elsewhere (21). These sourceswere used for the assessment of personal characteristics as well as characteristics ofpatient’s illness, end-of-life care and the nature of death as potential risk factors forcomplicatedgrief.

Personal characteristics at baselineInformation was gathered on sex, age and ethnicity (Caucasian or non-Caucasian).Workingstatuswasrecordedandrecodedintoretiredfromfull-timework/unemployedorworking.EducationwasgroupedaccordingtotheDutchStandardClassificationofEducation(DutchCentralBureauofStatistics(CBS),1989)(22).Theratingsforlow(1)tohigheducation(6)wererecodedintolow,intermediateandhigheducation.Cogni-tivecapacitywasassessedwiththeMiniMentalStateExamination,whichassessessixbroadareasofdailycognitivefunctions(23).ActivitiesofdailylivingperformancewasassessedwiththeStanfordHealthAssessmentQuestionnaire(24)andtheInstrumen-talActivitiesofDailyLivingscale(25).DepressionwasevaluatedwiththeuseoftheCentreforEpidemiologicalStudiesDepressionScale(score≥16)(26).

Patient’s illness, end-of-life care and the nature of deathHealth events were coded according to the tenth edition of the International Clas-sification of Disease (ICD) (27). The underlying causes of death were recorded andclassifiedasneoplasms,diseasesof thecirculatorysystemandothercausesofdeath.Informationwasgatheredonthedateandplaceofdeath.Placeofdeathwasrecordedasathome,inhospital,communityliving(i.e.homefortheelderly/nursinghome)orotherplaces.Further,informationwasobtainedonhospitaladmissionsinthelastyearoflife(numberandmediannumberofdays).

Medical files of the deceased were manually screened for additional informationonthepatient’sillness,characteristicsofendoflifecareandthenatureofthepatient’sdeath.Basedon the literature, a check listwasdevelopedby the research team.Thechecklistwaspilotedbeforeuse:10fileswereindependentlyassessedbySophieBru-insma(SB)andJudithRietjens(JR)anddiscussed.Thisledtosomesmallchangesinthechecklist.Asecondpilotwasperformedtotestthesechanges;nodisagreementwasfound.Asanextstep,allmedicalfileswerecheckedbySBwiththeuseofthischecklist.Severaltopicswerecovered.Thedurationoftheillnessthatwastheunderlyingcauseof death was registered as a continuous variable. The nature of death was classifiedas ‘completely unexpected’, ‘patient was ill, but death occurred unexpectedly’ and

Chapter 7122

‘expected’. ‘Thenumberof transitionsbetweencare settings in the last threemonthsof life’ was registered. End of life decision-making was classified as ‘withholding orwithdrawalofpotential life-prolongingtreatments’, ‘euthanasia’and ‘otherendof lifedecisions’.

Data analysis

Toincreasethecomparabilityofanalysesandreducebias,missingvalues(if>5%)wereimputedwithRandomizedsingleimputation(28).Thiswasthecasefor‘thedurationoftheillness’(30%),‘theexpectednatureofdeath’(9%),and‘thenumberoftransitionsbetweencaresettingsinthelastthreemonthsoflife’(14%).Todeterminetheassocia-tionbetweenthepotentialriskfactors‘personalcharacteristics,andcharacteristicsofdeceasedperson’s illness,endof lifecareandnatureofdeathandthegriefresponse,logistic regression was performed. Those variables that showed a p-value of 0.20 orsmallerintheunivariateregressionanalysiswereincludedinthemultivariateregres-sionmodel.

resUlts

Thecharacteristicsofthedeceasedandthebereavedpartnerarepresentedinappendix1. The majority of the deceased persons were male (66%). They were on average 74years old at the time of death. Almost all were Caucasian (99%). At baseline, mostdeceased persons had been retired or unemployed (85%) and the highest educationattainedwaspredominantly lowor intermediate(91%).Bereavedpartnerswerepre-dominantlyfemale(66%).Theywereonaverage73yearsoldattimeoftheirspouse’sdeath.AllpartnerswereCaucasian(100%).Atbaseline,89%ofthepartnershadbeenunemployedorretired,andthehighesteducationattainedwaspredominantlyloworintermediate(93%).Ofallpartners,19%sufferedfromadepressionatbaseline.Ontheactivitiesofdaily livingscale,partnersscoredonaverage24(=littledifficulties)andonthecognitivestatusscaleonaverage28(=normalcognition).Themedianofthetimepassedbetweenaperson’sdeathandtheinterviewwiththebereavedpartnerwas34months(IQR14-68).

Table1describesthecharacteristicsofdeceasedpersonsandthebereavedpartnersin relation to the occurrence of normal and complicated grief. Very few differenceswerefoundbetweenthetypesofgriefwithregardstopersonalcharacteristics.Ofthebereaved partners with normal grief, 9% suffered from depression at baseline, com-paredto25%ofthebereavedpartnerswithcomplicatedgrief.Themedianofthetimepassedbetweenaperson’sdeathandtheinterviewwiththebereavedpartnerswas39months (IQR 16-67) for normal grief, compared to 33 (IQR 13-68) for complicated


tabl

e 1

Diff

eren

ces

betw

een

olde

r adu

lts e

xper

ienc

ing

norm

al g

rief o

r com

plic

ated

grie

f: Ch

arac

teris

tics

dece

ased

and

ber

eave

d pa

rtne

rs (n

=200

)

nor

mal

gri

ef

(n=1

00)

com

plic

ated

gr

ief

(n=1

00)

Diff

eren

ce b

etw

een

grou

psU

nadj

uste

ddD

iffer

ence

bet

wee

n gr

oups

adju

sted

for s

ex a

nd a

gee

Ove

rall

p-va

lue

OR

( 95%

CI)

Ove

rall

p-va

lue

OR

( 95%

CI)

Dec

ease

d

Sex

Mal

e65

(65%

)67

(67%

).7

651.

00 (r

ef)

.148

1.69

(0.8

3-3.

44)

Fem

ale

35 (3

5%)

33 (3

3%)

0.92

(0.5

1-1.

64)

Age

at d

eath

(mea

n ±

SD)

(n=2

00)

74.8

8 ±

7.11

74.0

8 ±

7.21

.428

0.98

(0.9

5-1.

02)

.064

0.93

(0.8

6-1.

00)

Wor

king

stat

us (b

asel

ine)

W

orki

ng14

(14%

)16

(16%

).6

261.

00 (r

ef)

.702

1.00

Unem

ploy

ed o

r ret

ired

86 (8

6%)

81 (8

1%)

0.82

(0.3

8-1.

80)

0.86

(0.3

9-1.

90)

Miss

ing

-3

Educ

atio

n (b

asel

ine)

Low

49 (4

9%)

51 (5

1%)

.864

1.00

(ref

).3

141.

00

Inte

rmed

iate

40 (4

0%)

38 (3

8%)

0.91

(0.5

1-1.

65)

0.82

(0.4

4-1.

51)

High

11 (1

1%)

8 (8

%)

0.70

(0.2

5-1.

88)

0.62

(0.2

2-1.

70)

Miss

ing

-3

Bere

aved

par

tner

s

Sex

Mal

e36

(36%

)33

(33%

).6

551.

00 (r

ef)

.580

1.00

Fem

ale

64 (6

4%)

67 (6

7%)

1.14

(0.6

4-2.

05)

1.18

(0.6

5-2.

15)

Age

at d

eath

dec

ease

d (m

ean

± SD

) (n=

198)

72.9

8 ±

6.76

72.9

4 ±

6.48

.966

0.99

(0.9

6-1.

04)

.949

1.00

(0.9

6-1.

05)

Wor

king

stat

us (b

asel

ine)

Wor

king

11 (1

1%)

10 (1

0%)

.818

1.00

(ref

).7

501.

00

Unem

ploy

ed o

r ret

ired

89 (8

9%)

90 (9

0%)

0.90

(0.3

6-2.

22)

1.16

(0.4

6-2.

94)

Educ

atio

n (b

asel

ine)

Low

63 (6

3%)

62 (6

2%)

.645

1.00

(ref

).7

871.

00

Inte

rmed

iate

28 (2

8%)

34 (3

4%)

1.23

(0.6

7-2.

27)

1.31

(0.6

8-2.

51)

High

9 (9

%)

4 (4

%)

0.45

(0.1

3-1.

54)

0.48

(0.1

3-1.

74)

Chapter 7124 Risk factors for complicated grief in older adults 125

tabl

e 1

Diff

eren

ces

betw

een

olde

r adu

lts e

xper

ienc

ing

norm

al g

rief o

r com

plic

ated

grie

f: Ch

arac

teris

tics

dece

ased

and

ber

eave

d pa

rtne

rs (n

=200

) (co

ntin

ued)

nor

mal

gri

ef

(n=1

00)

com

plic

ated

gr

ief

(n=1

00)

Diff

eren

ce b

etw

een

grou

psU

nadj

uste

ddD

iffer

ence

bet

wee

n gr

oups

adju

sted

for s

ex a

nd a

gee

Ove

rall

p-va

lue

OR

( 95%

CI)

Ove

rall

p-va

lue

OR

( 95%

CI)

Dep

ress

ion

(CES

D)

(bas

elin

e )

No

depr

essio

n81

(81%

)66

(66%

).0

04**

1.00

(ref

).0

03**

1.00

Dep

ress

ion

9 (9

%)

25 (2

5%)

3.49

(1.4

9-7.

81)

3.86

(1.6

0-9.

33)

Miss

ing

109

ADLa

At b

asel

ine

(n=1

90)

24.4

4 ±

3.03

24.0

8 ±

3.41

.444

0.97

(0.8

8-1.

06)

.441

0.97

(0.8

8-1.

06)

Cogn

itive

stat

usb (M

MSE

) At

bas

elin

e (n

=194

)(m

ean

± SD

)

28.2

4 ±

1.40

27.9

3 ±

1.46

.130

0.85

(0.7

0-1.

05)

.132

0.85

(0.7

0-1.

05)

Tim

e si

nce

deat

h of

par

tner

m

onth

sc (n=1

96) (

med

ian

(IQR)

39 (1

6-67

)33

(13-

68)

.812

0.99

(0.9

9-1.

01)

.678

1.00

(0.9

9-1.

01)

a To

tal s

core

with

a m

inim

um o

f 0 (m

uch

diffi

culti

es w

ith A

DL)

and

27

(litt

le d

ifficu

lties

with

AD

L)

b A

ny s

core

gre

ater

tha

n or

equ

al t

o 27

poi

nts

(out

of 3

0) in

dica

tes

a no

rmal

cog

nitio

n. B

elow

thi

s, sc

ores

can

indi

cate

se

vere

(≤9

poin

ts),

mod

erat

e (1

0-18

poi

nts)

or m

ild (1

9-24

poi

nts)

cog

nitiv

e im

pairm

ent.

c Ti

me

in m

onth

s be

twee

n de

ath

of th

e de

ceas

ed a

nd th

e in

terv

iew

whe

re ty

pe o

f grie

f was

ass

esse

dd

Diff

eren

ce te

st is

bas

ed u

niva

riate

logi

stic

regr

essi

on. R

efer

ence

gro

up is

nor

mal

grie

f (**

p<0

.01)

e D

iffer

ence

test

is b

ased

mul

tivar

iate

logi

stic

regr

essi

on. R

efer

ence

gro

up in

nor

mal

grie

f (**

p<0

.01)

Chapter 7124 Risk factors for complicated grief in older adults 125

tabl

e 2

Diff

eren

ces

betw

een

norm

al g

rief a

nd c

ompl

icat

ed g

rief:

char

acte

ristic

s ill

ness

, end

of l

ife c

are

and

natu

re o

f dea

th

nor

mal

gri

ef

(n=1

00)

com

plic

ated

gri

ef

(n=1

00)

Diff

eren

ce b

etw

een

grou

psU

nadj

uste

deD

iffer

ence

bet

wee

n gr

oups

Adju

sted

for s

ex a

nd a

gef

Ove

rall

p-va

lue

OR

(95%

CI)

Ove

rall

p-va

lue

OR

(95%

CI)

caus

e of

dea

th

(icD

-10)

Neo

plas

ms

35 (3

5%)

45 (4

5%)

.066

1.00

(ref

).0

42*

1.00

Dise

ases

of

the

circ

ulat

ory

syst

em

32 (3

2%)

35 (3

5%)

0.80

(0.4

2-1.

55)

0.76

(0.3

9-1.

49)

Oth

era

32 (3

2%)

21 (2

1%)

0.51

(0.2

5-1.

03)

0.47

(0.2

3-0.

97)

Miss

ing

11

no.

of c

o-m

orbi

diti

es0

7 (7

%)

5 (5

%)

.407

1.00

(ref

).4

381.

00

127

(27%

)28

(28%

)1.

45 (0

.41-

5.14

)1.

51 (0

.42-

5.39

)

218

(18%

)25

(25%

)1.

94 (0

.53-

7.12

)2.

02 (0

.54-

7.59

)

314

(14%

)18

(18%

)1.

80 (0

.47-

6.90

)1.

87 (0

.47-

7.47

)

413

(13%

)10

(10%

)1.

07 (0

.26-

4.43

)1.

22 (0

.29-

5.16

)

>=5

20 (2

0%)

14 (1

4%)

0.98

(0.2

6-3.

73)

1.02

(0.2

6-4.

00)

Plac

e of

dea

thHo

me

36 (3

6%)

37 (3

7%)

.918

1.00

(ref

).9

181.

00

Hosp

ital

42 (4

2%)

40 (4

0%)

0.93

(0.4

9-1.

74)

0.91

(0.4

8-1.

75)

Com

mun

ity

livin

g16

(16%

)16

(16%

)0.

97 (0

.42-

2.23

)1.

00 (0

.42-

2.38

)

Oth

erb

3 (3

%)

3 (3

%)

0.97

(0.1

8-5.

14)

0.91

(0.1

7-4.

86)

Miss

ing

34

no.

of

hosp

ital

izat

ions

la

st y

ear o

f life

020

(20%

)24

(24%

).5

151.

00 (r

ef)

.975

1.00

Chapter 7126

tabl

e 2

Diff

eren

ces

betw

een

norm

al g

rief a

nd c

ompl

icat

ed g

rief:

char

acte

ristic

s ill

ness

, end

of l

ife c

are

and

natu

re o

f dea

th (c

ontin

ued)

nor

mal

gri

ef

(n=1

00)

com

plic

ated

gri

ef

(n=1

00)

Diff

eren

ce b

etw

een

grou

psU

nadj

uste

deD

iffer

ence

bet

wee

n gr

oups

Adju

sted

for s

ex a

nd a

gef

Ove

rall

p-va

lue

OR

(95%

CI)

Ove

rall

p-va

lue

OR

(95%

CI)

126

(26%

)21

(21%

)0.

64 (0

.28-

1.47

)0.

64 (0

.28-

1.48

)

216

(16%

)14

(14%

)0.

64 (0

.27-

1.77

)0.

69 (0

.27-

1.77

)

310

(10%

)11

(11%

)0.

87 (0

.31-

2.48

)0.

88 (0

.31-

2.50

)

45

(5%

)11

(11%

)1.

74 (0

.52-

5.88

)1.

75 (0

.52-

5.96

)

>=5

22 (2

2%)

19 (1

9%)

0.68

(0.2

9-1.

62)

0.70

(0.2

9-1.

65)

Dur

atio

n ho

spit

aliz

atio

ns

last

yea

r of l

ife

(n=1

56)

(med

ian

(IQR)

25 (9

-49)

26 (1

4-50

).8

271.

01 (0

.99-

1.01

).9

501.

00 (0

.99-

1.01

)

Dur

atio

n ill

ness

Ω0-

6 m

onth

s40

(40%

)42

(42%

).2

641.

00 (r

ef)

.247

1.00

6 m

onth

s- 1

year

23 (2

3%)

27 (2

7%)

1.12

(0.5

5-2.

26)

1.14

(0.5

6-2.

31)

1 ye

ar-4

year

s17

(17%

)22

(22%

)1.

23 (0

.57-

2.65

)1.

21 (0

.56-

2.61

)

> 4

year

s17

(17%

)22

(22%

)0.

45 (0

.18-

1.11

)0.

44 (0

.18-

1.10

)

Miss

ing

20

sudd

en d

eath

ΩCo

mpl

etel

y un

expe

cted

20 (2

0%)

14 (1

4%)

.164

1.00

(ref

).1

551.

00

Dec

ease

d ill

, bu

t dea

th

unex

pect

ed

29 (2

9%)

26 (2

6%)

1.28

(0.5

4-3.

04)

1.34

(0.5

6-3.

19)

Expe

cted

51 (5

1%)

60 (6

0%)

1.68

(0.7

7-3.

66)

1.73

(0.7

9-3.

80)


tabl

e 2

Diff

eren

ces

betw

een

norm

al g

rief a

nd c

ompl

icat

ed g

rief:

char

acte

ristic

s ill

ness

, end

of l

ife c

are

and

natu

re o

f dea

th (c

ontin

ued)

nor

mal

gri

ef

(n=1

00)

com

plic

ated

gri

ef

(n=1

00)

Diff

eren

ce b

etw

een

grou

psU

nadj

uste

deD

iffer

ence

bet

wee

n gr

oups

Adju

sted

for s

ex a

nd a

gef

Ove

rall

p-va

lue

OR

(95%

CI)

Ove

rall

p-va

lue

OR

(95%

CI)

num

ber o

f tr

ansi

tion

s las

t 3

mon

thsc

Ω

026

(26%

)25

(25%

).2

431.

00 (r

ef)

.222

1.00

135

(35%

)30

(30%

)0.

93 (0

.45-

1.92

)0.

91 (0

.43-

1.91

)

221

(21%

)19

(19%

)0.

98 (0

.43-

2.23

)0.

96 (0

.42-

2.21

)

>=3

17 (1

7%)

26 (2

6%)

1.65

(0.7

3-3.

74)

1.71

(0.7

4-3.

93)

end

of li

fe

deci

sion

- m

akin

g

With

hold

ing

or w

ithdr

awal

of

pot

entia

l lif

e-pr

olon

ging

tr

eatm

ents

No/

not i

n fil

e76

(76%

)81

(81%

).3

901.

00.3

671.

00

Yes

24 (2

4%)

19 (1

9%)

0.74

(0.3

8-1.

46)

0.73

(0.3

7-1.

45)

Euth

anas

iaN

o/no

t in

file

99 (9

9%)

92 (9

2%)

.044

*1.

00.0

47*

1.00

Yes

1 (1

%)

8 (8

%)

8.61

(1.0

6-70

.17)

8.38

(1.0

2-68

.49)

Oth

erd

No/

not i

n fil

e84

(84%

)90

(90%

).2

111.

00.1

791.

00

Yes

16 (1

6%)

10 (1

0%)

0.58

(0.2

5-1.

36)

0.56

(0.2

4-1.

31)

a O

ther

cau

ses o

f dea

th: D

isea

ses o

f the

resp

irato

ry sy

stem

; dis

ease

s of t

he n

ervo

us sy

stem

s, m

enta

l and

beh

avio

ral d

isor

ders

; cer

tain

infe

ctio

us a

nd

para

sitic

dis

ease

s; d

isea

ses

of th

e di

gest

ive

syst

em; d

isea

ses

of th

e ge

nito

urin

ary

syst

em; s

ympt

oms,

sign

s an

d ab

norm

al c

linic

al a

nd la

bora

tory

fin

ding

s, no

t els

ewhe

re c

lass

ified

; inj

ury,

poi

soni

ng a

nd c

erta

in o

ther

con

sequ

ence

s of

ext

erna

l cau

ses;

ext

erna

l cau

ses

of m

orbi

dity

and

mor

talit

yb

Oth

er=

e.g.

out

side

on

the

stre

et, f

amily

’s ho

me

c Fi

rst t

rans

ition

mai

nly

from

hom

e to

hos

pita

l; se

cond

tran

sitio

n m

ainl

y fr

om h

ospi

tal t

o ho

me

d Pa

lliat

ive

seda

tion,

inte

nsifi

ed a

llevi

atio

n of

sym

ptom

s, D

NR

e D

iffer

ence

test

is b

ased

uni

varia

te lo

gist

ic re

gres

sion

. Ref

eren

ce g

roup

is n

orm

al g

rief (

* p<0

.05)

f D

iffer

ence

test

is b

ased

mul

tivar

iate

logi

stic

regr

essi

on. R

efer

ence

gro

up in

nor

mal

grie

f (* p

<0.0

5)Ω

Impu

ted

resu

lts

Chapter 7128

grief.Univariateregressionanalysisshowedthatbereavedpartnerswithdepressionatbaselinehadahigherriskofcomplicatedgrief(OR=3.86;95%CI=1.60-9.33).

Table 2 describes the characteristics of deceased person’s illness, end of life careand nature of death in relation to the occurrence of normal and complicated grief.For normal grievers, in 35% of the cases the partner had died from cancer, in 32%fromdiseasesofthecirculatorysystem,andin32%fromothercauses.Forcomplicatedgrievers, thiswasrespectively45%,35%and21%.Outof9partnersofpatientswhohaddiedfromeuthanasia,8experiencedcomplicatedgrief.Univariateanalysisshowedthatpartnersofpatientswhohaddiedfromcancerhadahigherriskofcomplicatedgrief thanpartnersofpatientswhohaddiedfromdiseasesof thecirculatorysystem(OR=0.76;95%CI=0.39-1.49)orotherdiseases(OR=0.47;95%CI=0.23-0.97).Furtheritshowedthatifthedeceaseddiedaftertheuseofeuthanasia,complicatedgriefwasmorelikely(OR=8.38;95%CI=1.02-68.49).

Following theunivariateanalyses (table1and2),multivariate logistic regressionwasperformed(forallvariableswithp<.02)(notablepresented).Theunderlyingcauseofdeath; thenatureofdeath; end-of-lifedecisions (euthanasiaandotherendof lifedecisions); sexand theageatdeathof thedeceasedperson,anddepression (CESD)andcognitivestatus(MMSE)atbaselinewereincludedinthemodel.Onlydepressionwas significantly associated with type of grief. Bereaved partners with depression atbaselinehadahigherriskofcomplicatedgriefcomparedtobereavedpartnerswithoutdepression(OR=3.48;95%CI=1.40-8.68).

DiscUssion

Inthepresentstudywefoundthatcomplicatedgriefinolderadultsissignificantlyas-sociatedwithdepression,andnotwithcharacteristicsofthepatient’sillness,end-of-lifecareandthenatureofdeath.

Multivariate analysis showed that only depression at baseline was significantlyassociatedwithan increasedriskofcomplicatedgrief.Apparently,complicatedgriefinolderadultscannotbeexplainedbycircumstancessurroundingthepatient’sdeath(situationalfactors),butpredominantlybyfactorsrelatedtothebereaveditself(per-sonalfactors).Depressionhasbeencloselyassociatedwithgriefintheliterature.Whilestudieshavefocusedondepressionasanoutcomeofgrief(29),orasasyndromefol-lowingaspousaldeath(30),therearefewstudiesthatexaminedtheetiologicrelevanceofdepressionfortheonsetofcomplicatedgrief(31-33).Horowitzetal.,whostudied70individualswhohadexperiencedthedeathofaspousewhentheywerebetweentheagesof21and55years,foundthatthosebereavedwithahistoryofmajordepressivedisorderweremorevulnerabletocomplicatedbereavement(31).Thismaybeexplained


byalackofabilitytocopewithloss(34).Apparently,thisespeciallyholdstrueforolderbereavedpartners.

Socialsupportisimportantinthebereavementperiod,becauseitprotectsagainstphysicalandpsychological illnessandhelps tomaintainqualityof life (35).Astudyperformedamongolderadultswhoexperiencedcomplicatedgrief showed that theyoftenrelyonavailableinterpersonalsupporttohelpthemmanagetheirgrief,butthatsuch support is not always experienced as sufficient (36). Professional support maythereforebeespeciallyimportantforolderpersonsseekingbereavementsupport(35,36).Healthcareprofessionalswhocareforterminalpatientsandtheirpartnersshouldpayparticularattention topartnerswithpre-lossdepression(11).Potentially,physi-cianscouldbe trainedtoperformevidence-basedassessments fordepressionandtolinkthosewhocouldbenefittobereavementormentalhealthspecialists(37).

Inthepresentstudy,8outof9relativesofpatientswhohaddiedfromeuthanasiaexperiencedcomplicatedgrief.ApreviousDutchstudyshowedthatthebereavedfam-ilyofpatientswithcancerwhodiedbyeuthanasiacopedbetterwithrespecttogriefsymptomsandpost-traumaticstressreactionsthanthebereavedofpatientswithcancerwhodiedanaturaldeath(14,38).Ourresultsshouldbeinterpretedwithcautionduetothesmallnumbers,butthedifferencewiththepreviousstudymaybeexplainedbythefactthatpatientsreceivingeuthanasiainourstudymayhavehadarelativelydifficultandprotracteddyingprocess,whereasinthepreviousstudypatientswhoreceivedanddidnotreceiveeuthanasiadiedfromcomparabledisorders.VandenBoom(39)whodescribed the consequences of euthanasia on grief among the bereaved family andfriends, previously found that a complicated euthanasia process was associated withcomplicatedgriefandaddeddistresstothebereavedfamilyandfriends.

No associations were found between socio- demographic characteristics of thebereavedandthetypeofgrieftheyexperience.Whiledemographicfactorsareconsis-tentlyidentifiedasrelatingtobereavementoutcomes,itislikelythattheyareoflittleimportance in determining an individual’s specific risk of complicated bereavementoutcomes (10). Demographic factors such as age, gender and socio-economic statusmay affect health independently of bereavement (10, 40). Age, for instance, may bemoreofanindicatorofdifferencesingrievingstylethanaspecificindicatorofrisk(41).Alsothe(younger)ageofthedeceasedisoftencitedasariskfactorforcomplicatedbereavement in surviving relatives.(10) However, this counts particularly in relationto the death of child (10). As the current study focuses on spousal loss, this couldpotentiallyexplainthelackofasignificantassociation.

Ourstudyhasseveralstrengths.First,itwasconductedwithinapopulation-basedsetting(1).Second,alargesamplewasemployedwhichenhancesthegeneralizabilityofthefindings(1).However,thestudyalsohassomelimitations.First,griefwasdichoto-mized.Dichotomizationofacontinuousoutcomevariablemayleadtoalossofpower.

Chapter 7130

However,complicatedgriefisdesignatedasadisorderwithdistinctcharacteristicsandadverse outcomes. Unfortunately, no information was found in the medical files onrelatives’involvementinthepatient’scare.Intensityofcareprovidedhasbeenfoundtobeariskforcomplicatedgrief(12).Futurestudiesshouldassesswhethercharacteristicsofthecaregivingexperienceresultindistinctiveriskfactorsfordevelopingcomplicatedgriefinolderadults(12).Thisalsoholdstrueforotherbereavedrelatedcharacteristics,suchasahistoryofpreviouslossesandhighpre-deathdistress,orfactorsconcerninginterpersonal relationships, suchas theavailabilityof social supportand the leveloffamilyfunctioning(10).Socialsupporthasbeenshowntobeanimportantprotectivefactoragainstthenegativeeffectsofcomplicatedgrief ’(42-45).Finally,othermentalhealthfactorsthandepressionwerenottakenintoaccountbutmayhavebeenofinflu-ence,suchasanxietyandpanicdisorders.

Inconclusion,ourresultssuggestthatcomplicatedgriefinolderadultsisnotrelatedtothecircumstancesofdyingofthedeceasedspouse.Pre-existingconditionssuchasdepression seem to be more important in explaining the occurrence of complicatedgrief.

acknowleDGements

we would like to thank everyone involved in the data collection process of the Rotterdam Study and the inhabitants of Ommoord for their time and effort.


references

1. NewsonRS,BoelenPA,HekK,HofmanA,TiemeierH.Theprevalenceandcharacteristicsofcom-plicatedgriefinolderadults.JAffectDisord2011;132(1-2):231-8.

2. WorkmanS.Prolongedgriefdisorder:aproblemforthepast,thepresent,andthefuture.PLoSMed.2009;6(8):e1000122.doi:10.1371/journal.pmed.1000122.

3. BoelenPA,vandenBoutJ.Complicatedgrief,depression,andanxietyasdistinctpostlosssyndromes:aconfirmatoryfactoranalysisstudy.AmJPsychiatry2005;162(11):2175-7.

4. PrigersonHG,BierhalsAJ,KaslSV,ReynoldsCF,ShearMK,DayN,etal.Traumaticgriefasariskfactorformentalandphysicalmorbidity.AmJPsychiatry1997;154(5):616-23.

5. PrigersonHG,FrankE,KaslSV,ReynoldsCF,AndersonB,ZubenkoGS,etal.Complicatedgriefandbereavement-relateddepressionasdistinctdisorders:preliminaryempiricalvalidationinelderlybereavedspouses.AmJPsychiatry1995;152(1):22-30.

6. SzantoK,PrigersonH,HouckP,EhrenpreisL,ReynoldsCF.Suicidalideationinelderlybereaved:theroleofcomplicatedgrief.SuicideLifeThreatBehav1997;27(2):194-207.

7. OttCH,LuegerRJ,KelberST,PrigersonHG.Spousalbereavementinolderadults:common,resilient,andchronicgriefwithdefiningcharacteristics.JNervMentDis2007;195(4):332-41.

8. SupianoKP,LuptakM.ComplicatedGriefinOlderAdults:ARandomizedControlledTrialofCom-plicatedGriefGroupTherapy.Gerontologist2014;54:840–856.

9. ParkesCM.Hospiceheritage:Introduction.Omega(Westport)2007;56:1–5. 10. Aranda S, Milne D. Guidelines for the Assessment of Complicated Bereavement Risk in Family

MembersofPeopleReceivingPalliativeCare.Melbourne:CentreforPalliativeCare,2000. 11. ChiuYW,HuangCT,YinSM,HuangYC,ChienCH,ChuangHY.Determinantsofcomplicatedgrief

incaregiverswhocaredforterminalcancerpatients.SupportCareCancer2010;18(10):1321-7. 12. GhesquiereA,HaidarYM,ShearMK.Risksforcomplicatedgriefinfamilycaregivers.JSocWork

EndLifePalliatCare2011;7(2-3):216-40. 13. Bernard L, Guarnaccia C. Husband and adult-daughter caregivers’ bereavement. J Death Dying

2002;45:153–66. 14. Jacobs S, Mazure C, Prigerson H. Diagnostic criteria for traumatic grief. Death Stud 2000;24(3):

185-99. 15. AbarshiE,EchteldM,VandenBlockL,DonkerG,DeliensL,Onwuteaka-PhilipsenB.Transitions

betweencaresettingsattheendoflifeintheNetherlands:resultsfromanationwidestudy.PalliatMed2010;24(2):166-74.

16. HofmanA,BretelerMM,vanDuijnCM,JanssenHL,KrestinGP,KuipersEJ,etal.TheRotterdamStudy:2010objectivesanddesignupdate.EurJEpidemiol2009;24(9):553-72.

17. HofmanA,vanDuijnCM,FrancoOH, IkramMA, JanssenHL,KlaverCC,etal.TheRotterdamStudy:2012objectivesanddesignupdate.EurJEpidemiol2011;26(8):657-86.

18. Federal Interagency Forum on Aging-Related Statistics: Older Americans 2012: Key Indicators ofWell-Being.Washington,DC:FederalInteragencyForumonAging-RelatedStatistics,2012.

19. ErnsterVL.Nestedcase-controlstudies.PrevMed1994;23(5):587-90. 20. Prigerson HG, Horowitz MJ, Jacobs SC, Parkes CM, Aslan M, Goodkin K, et al. Prolonged grief

disorder:PsychometricvalidationofcriteriaproposedforDSM-VandICD-11.PLoSMed2009;6(8):e1000121.doi:10.1371/journal.pmed.1000121

21. LeeningMJ,KavousiM,HeeringaJ,vanRooijFJ,Verkroost-vanHeemstJ,DeckersJW,etal.MethodsofdatacollectionanddefinitionsofcardiacoutcomesintheRotterdamStudy.EurJEpidemiol2012;27(3):173-85.

Chapter 7132

22. DutchCentralBureauofStatistics:Standaardonderwijsindeling1978,codenlijstvanopleidingen,alfabetischgerangschikt.Voorburg:DutchCentralBureauofStatistics,1989.

23. FolsteinMF,FolsteinSE,McHughPR.“Mini-mentalstate”.Apracticalmethodforgradingthecogni-tivestateofpatientsfortheclinician.JPsychiatrRes1975;12(3):189-98.

24. FriesJF,SpitzPW,YoungDY.Thedimensionsofhealthoutcomes:thehealthassessmentquestion-naire,disabilityandpainscales.JRheumatol1982;9(5):789-93.

25. LawtonMP,BrodyEM.Assessmentofolderpeople:self-maintainingandinstrumentalactivitiesofdailyliving.Gerontologist.1969;9(3):179-86.

26. RadloffLS.TheCESDscale: a self-reportdepression scale for research in thegeneralpopulation.ApplPsychMeas1977;1(3):385–401.

27. World Health Organization. International statistical classification of diseases and health relatedhealthproblems,10threvision(ICD-10).Geneva:WHO;1992.

28. DondersAR,vanderHeijdenGJ,StijnenT,MoonsKG.Review:agentleintroductiontoimputationofmissingvalues.JClinEpidemiol2006;59(10):1087-91.

29. ParkesC.Bereavement:StudiesofGriefinAdultLife,3rded.Madison,WI:InternationalUniversi-tiesPress,1998.

30. ZisookS,ShuchterSR.Uncomplicatedbereavement.JClinPsychiat1993;54(10):365-72. 31. HorowitzM,SiegelB,HolenA,BonannoG,MilbrathC,StinsonC.Diagnosticcriteriaforcompli-

catedgriefdisorder.AmJPsychiatry1997;154(7):904-10. 32. Lazare A. Unresolved Grief: Outpatient Psychiatry: Diagnosis and Bereavement. Baltimore, MD:

WiliamsandWilkins,1979,pp.498–512. 33. AllenJY,HaleyWE,SmallBJ,SchonwetterRS,McMillanSC.Bereavementamonghospicecaregivers

ofcancerpatientsoneyear following loss:predictorsofgrief,complicatedgrief,andsymptomsofdepression.JPalliatMed2013;16(7):745-51.

34. Hirsch M, Allison K, Shaw S. Harvard Medical School: Coping with Grief and Loss: A Guide toHealing.AHarvardMedicalSchoolSpecialHealthReport.Boston:HarvardMedicalSchool,2010.

35. Benkel I, Wijk H, Molander U. Family and friends provide most social support for the bereaved.PalliatMed2009;23(2):141-9.

36. GhesquiereA. ‘‘Iwas justtryingtostickitoutuntilIrealizedthatIcouldn’t:’’Aphenomenologi-calinvestigationofsupportseekingamongolderadultswithcomplicatedgrief.Omega(Westport)2013;68:1–22.

37. GhesquiereAR,PatelSR,KaplanDB,BruceML.Primarycareproviders’bereavementcarepractices:recommendationsforresearchdirections.IntJGeriatrPsychiatry2014;29(12):1221-9.


39. VandenBoomF.AIDS,euthanasiaandgrief.AIDScare1995;7:175-85. 40. SandersC.Riskfactorsinbereavementoutcome.In:StroebeM,StroebeW,HanssonR(eds):Hand-

bookofBereavement:Theory,Research,andIntervention.Cambridge:CambridgeUniversityPress,1993.

41. Sanders C. Comparison of younger and older spouses in bereavement outcome. J Death Dying1980–1981;11:217–32.

42. BurkeLA,NeimeyerRA,McDevitt-MurphyME.AfricanAmericanhomicidebereavement:aspectsof social support that predict complicated grief, PTSD, and depression. Omega (Westport) 2010;61(1):1-24.

43. MaciasC,JonesD,HarveyJ,BarreiraP,HardingC,RodicanC.Bereavementinthecontextofseriousmentalillness.PsychiatrServ2004;55(4):421-6.


44. OgrodniczukJS,JoyceAS,PiperWE.Changesinperceivedsocialsupportaftergrouptherapyforcomplicatedgrief.JNervMentDis2003;191(8):524-30.

45. NamI,HyunD.EffectsofSatisfactionwithSocialSupportonComplicatedGriefinBereavedDe-mentiaCaregivers.KoreanJHealthPromot2014;14(2):43-9.

Chapter 7134

appendix 1 Characteristics of deceased and bereaved partners (n=200)

n (%)

Deceased

sex Male 132 (66%)

Female 68 (34%)

age at death (mean ± SD) 74.48 ± 7.15

ethnicity Caucasian 192 (99%)

Non- Caucasian 1 (1%)

Missing 7

working status (baseline) Working 30 (15%)

Unemployed or retired 167 (85%)

Missing 3

educationa (baseline) Low 100 (50%)

Intermediate 78 (40%)

High 19 (10%)

Missing 3

Bereaved partners

sex Male 69 (34%)

Female 131 (66%)

age at death deceased (mean ± SD) 72.96 ± 6.60

ethnicity Caucasian 194 (100%)

Non- Caucasian 0

Missing 6

working status (baseline) Working 21 (11%)

Unemployed or retired 179 (89%)

educationa (baseline) Low 125 (62%)

Intermediate 62 (31%)

High 13 (17%)

Depression (cesD) (baseline) (n=181) No depression 147 (81%)

Depression 34 (19%)

Missing

aDlb(baseline) (n=192) 24.26 ± 3.22

cognitive statusc (mmse) (baseline) (mean ± SD) (n=194)

28.08 ± 1.43

time since death of partnerd (in months) (median (IQR) 34 (14-68)

a Highest education attained (completed or not completed): Low=primary education; primary education, plus higher not completed education; lower vocational education; lower secondary education. Interme-diate= intermediate vocational education; general secondary education. High= higher vocational educa-tion; university

b Total score with a minimum of 0 (much difficulties with ADL) and 27 (little difficulties with ADL) c Any score greater than or equal to 27 points (out of 30) indicates a normal cognition. Below this, scores

can indicate severe (≤9 points), moderate (10-18 points) or mild (19-24 points) cognitive impairment. d Time between the death of deceased and the interview where type of grief was assessed

Chapter 8Discussion

General discussion 139

The aim of this thesis is to provide a comprehensive overview of bereaved relatives’experienceswiththepracticeofpalliativesedation.Further,weaimatgettinginsightinto thepotential life-shorteningeffectofpalliative sedationand the risk factors forcomplicated grief in older adults. First, the key findings of the studies will be sum-marized in this chapter. Next, the strengths and weaknesses of the studies will beexplored.Third, anumberoffindingswillbediscussed inmoredetail.This chapterwillbeconcludedwithimplicationsandrecommendationsforclinicalpractice,policyandforfutureresearch.

8.1 keY finDinGs

Intheintroductionofthisthesis,severalresearchquestionswereformulated.Inthissection,themainanswerstoeachoftheseresearchquestionsaresummarized.

Research question 1: What are the experiences of bereaved relatives with the prac-tice of palliative sedation?

Westudiedtheexperiencesofrelativeswiththepracticeofcontinuouspalliativeseda-tioninasystematicliteraturereview(chapter2),byinterviewingfocusgroupsofDutchbereavedrelatives(chapter3)andbyinterviewingbereavedrelativesofcancerpatientsfromBelgium,theNetherlandsandtheUK(chapter4).Thein-depthinterviewswithbereavedrelativesfromtheNetherlands,BelgiumandtheUKshowedthatmanyrela-tiveswereabletoprovideadescriptionoftheconceptofsedation.Allstudiesshowedthatrelativesingeneralseemtobecomfortablewiththeuseofpalliativesedation.Thiswaspredominantlyrelated to their impression that thepatient’s sufferingwasfinallyalleviated. Relatives generally believed that the sedation contributed to the patienthavinga‘good’death.Positiveexperiencesarealsorelatedtoanactiveroleofrelativesin the decision-making. The results of the focus groups also suggested that relativeswerepositiveabouthavinganactiveroleintheapplicationofthesedation,thedegreeofinvolvementofprofessionalcaregiversincaregivinganddecision-making,andthattheythoughtthattheplaceofdeathwasinagreementwiththewishesofthepatient.According to theparticipantsof the focusgroups, the sedationhadoffered themanopportunitytopreparethemselvesforthepatient’sdeath.TheinterviewstudyprovidedmoreinsightintothewhatseemstoberathertypicallyDutchandBelgianprocessof‘sayinggoodbye’.AccordingtotherelativesfromtheNetherlandsandBelgium,thestartofthesedationallowedforaplannedmomentofsayinggoodbye,whichmostofthemappreciated.Incontrast,sayinggoodbyewasalessdominantthemeintheaccountoftheUKrelatives.Ifdescribedatall,theysaidthatsayinggoodbyewasagradualprocess

Chapter 8140

ratherthanaplannedsinglemomentintime.Relativesseemedtoconsiderthisasanacceptablewayofproceeding.

However,allstudiesalsopointedatrelativesexpressingdistressbeforeorduringtheapplication of sedation. Although our literature review suggests that the majority ofrelativesreceivesadequateinformation,thefocusgroupstudyandtheinterviewstudyrevealed that some relatives indicated dissatisfaction with the received information,andaboutcommunicationwithprofessionalcaregivers ingeneral.Further,concernsexistedaboutpatients’wellbeingduringthesedation.Sometimesrelativesfeltthatthepatient was not sufficiently sedated to relieve his/her suffering. Also, concerns wereexpressedaboutrelatives’ownwellbeing,especiallyifsedationwasofalongerdurationthananticipated.Thefocusgroupssuggestedthatrelativesweresometimesfrustratedaboutthelackofauthorityofnursestomakedecisionsintheabsenceofthephysician.Theinterviewstudyadditionallyshowedthatrelativessometimesquestionedwhethertherewereotheralternativestocontinuoussedation,suchaseuthanasiaoradifferenttype of sedation. Although the loss of the ability to communicate during the seda-tionwiththepatientandapotentiallife-shorteningeffectarefrequentlydiscussedaspossibledisadvantagesoftheuseofsedationbycaregivers, legalandethicalexperts,the relatives in general did not experience these issues as important disadvantages.TheinterviewstudyindicatedthatsomerespondentsfromtheUKexpressedthepo-tential hastening of death as a main concern. In contrast, there were few concernsaboutwhetherornotsedationhadshortenedthepatient’slifeintheNetherlandsandBelgium.

Research Question 2: What is the prolonged impact of palliative sedation on rela-tives’ experience of the dying phase and their wellbeing after the patient’s death?

Inchapter5wereportontheresultsofasurveyamongbereavedrelativesofpatientswhoreceivedanddidnotreceivepalliativesedation.Relativesofsedatedpatientmoreoftenstatedthattheprofessionalcaregiverscouldhavedonemoretomaketheperiodbefore thedeathof thepatientmorebearable thanrelativesofnon-sedatedpatients. Relativesofpatientswhoreceivedanddidnot receive sedationdidnot significantlydifferintheirassessmentofthequalityofend-of-lifecare,patients´qualityoflifeinthelastweekbeforedeathandtheirqualityofdying.Nosignificantdifferenceswerefoundinrelatives’satisfactionwiththeirownlifethreemonthsafterthedeathofthepatientandatthetimeofthesurvey,theirgeneralhealthandtheirmentalwellbeing. Itcanbeconcludedthattheuseofsedationdoesnotseemtohaveanegativeinfluenceonbereavedpersons’experienceofthedyingphaseoftheirdeceasedrelativeortheirwellbeingafterthepatient’sdeath.


Research Question 3: How accurately do physicians estimate the potential life-shortening effect of continuous sedation until death?

Our study indicated a substantial discrepancy between two approaches to estimatethelife-shorteningeffectofcontinuoussedationuntildeathbyphysicians(chapter6).Thedirectapproach(question:Didcontinuoussedation,accordingtoyourestimation,hastenthepatients’death?)ledtothefindingthatsedationmighthavehadalifeshort-eningeffectin51%ofthecasesandtheindirectapproach(thedurationofthesedationwassubtractedfromthepatient’slifeexpectancyasestimatedbythephysician)to84%.Thesevaryingresultsconfirmthedifficultyofpredictingthelifeexpectancyofpatientswithadvanceddisease,andofestimatingthepotentiallife-shorteningeffectofend-of-lifeinterventions.

Research Question 4: What are risk factors for complicated grief in older adults?

Chapter7providesinsight intoriskfactorsforcomplicatedgrief inolderadults.Westudiedtheinfluenceofpersonalcharacteristicsofthepatientandthebereavedpartnerontheonehand,andcharacteristicsofthedyingprocess,suchasthepatient’sillness,end-of-lifecareandthenatureofdeathontheotherhand.Onlypre-existingdepression(thatwasalreadypresentbeforethedeathofthepatient)wassignificantlyassociatedwith complicated grief. No significant associations were found for characteristics ofdeceasedperson’sillness,endoflifecareandthenatureofdeath.Theseresultssuggestthatcomplicatedgriefinolderadultsisnotclearlyrelatedtothecircumstancesofdyingofthedeceasedpartner.Pre-existingconditionssuchasmajordepressionseemtobeofmoreimportanceinexplainingtheoccurrenceofcomplicatedgrief.

8.2 strenGths anD weaknesses

End-of-life care is generally considered tobea complex research topic that requiresmultidimensionalandmultidisciplinarystudies.Yet,themajorityofresearchfocusedonthemedicalaspectssolelyfromperspectivesofhealthcareproviders,whileimpor-tant opportunities for understanding and improving patients’ dying experiences canbefoundamongcareprofessionalsfromotherdisciplinesaswell.Becauseitisknownthatprofessionalcaregiversandrelativesoftenhavedifferentviewsonwhatconstitutesagooddeath(1-3)andhavedifferentviewsontheethicalandmoralconsiderationsinvolvedintheuseofsedation(4),ourstudyincludedtheperspectivesofbothphysi-ciansandnurses,aswellasbereavedrelatives.TheWHOconsiders thewellbeingofrelatives as an explicit aim of end-of-life care (5). So far, however, relatives received

Chapter 8142

surprisinglylittleattentioninend-of-liferesearch.Whenresearchingendoflifecare,it is importanttoconsiderthecomplexinteractionbetweendifferentgroupsandthewidercontextwithinwhichtheyoperate.Thisthesismadeausefulcontributiontothisfieldofresearch.

Further,anotherstrengthofthisthesisisthatitcombinesfindingsfrommultiplecountries (the Netherlands, Belgium and the United Kingdom). Palliative care hasevolvedacrosstheglobeindifferentcontextsandindifferentways,althoughmanyofthechallengesfacedaresimilar(6).Comparisonbetweencountrieshelpstoidentifythebestsolutionsforpatientsandtheirfamilies,whohavecomplexneedsandproblems(6). By aforementioned reasons and by applying multiple methods (both qualitativeandquantitative)andusingmultipledatasources,itwaspossibletogenerateamorecomprehensiveandbalancedpictureoftheexperiencesofbereavedrelativeswithpal-liative sedation at the end of life. To our knowledge, this is the first time that sucha comprehensive approach has been used to understand the experiences of relativesinend-of-lifecare.Further, this thesiscontributes totheethicaldebatebysupplyingevidence-basedargumentsforassessingthebenefitsanddrawbacksofpalliativeseda-tion.

Thedifferentmethodsusedinthisthesiswillbediscussedmoreindepth.

review

Thereviewprovidedthereaderwithanexhaustivesummaryoftheexistingliteratureontheexperiencesofrelativeswiththepracticeofpalliativesedation.Systematicre-viewarewidelyregardedascrediblesourcesofevidence,andareoftenrecommendedasamongthehighestlevelsofevidencethatcanbelocatedonatopic(7).Itisimportantto recognize that there is not a shared language or understanding of what palliativesedation is across or within health and care services. This variety in the terms anddefinitionsused (i.e. variety in the terminology regardingpalliative sedationandanapparent lackofaconsistentoperationaldefinitionofpalliativesedation)potentiallylimited full comparison and extrapolation of the studies. Further, not all studies in-cludedappearedtobeof‘good’quality.

Qualitative studies: focus groups and interview

Byusing focusgroups,weelicited relatives’multiplicityofviews,beliefs,knowledgeaboutandexperienceswiththepracticeofpalliativesedation(8).Thegroupprocesseswereparticularlyusefulinclarifyingrelatives’viewsonpalliativesedation.Inthein-terviews we could examine the issues raised by the relatives in detail and in depth.Inthisway,thefocusgroupsandinterviewscomplementedeachother.Death,dyingandbereavementareallsensitiveandpotentiallyupsettingtopicstodiscuss.However,relativeshighlyappreciatedtheopportunitytosharetheirstories,andfeltthatithelped


them to deal with their grief. Nevertheless, some limitations were encountered. Asthese were qualitative studies based on the in depth exploration of relatively smallnumbersofcases, it isnotpossible togeneralize inastatisticalsensefromourfind-ings. Second, the quality of the research partly depended on the individual skills oftheinterviewerandcouldbeinfluencedbytheinterviewer’spersonaland/orculturalbiases.Theuseof anaide-memoirewithquestionsandnonspecificprompts suchas“tellmemore”wereusedtoavoidinterviewerbias.Further,interviewswererehearsedwithexpertsinresearch,fieldtested,andadjustedbeforefinaluse.Third,tobesensitivetotheearlyphasesofgrieving,relativeswerecontactednosoonerthanthreemonthsafterthedeathofthepatient,introducingpotentialrecallbias.Memoriescanalsohavebeendisturbedby the strongemotions theyexperiencedat that time. However, theeffectofbereavementontheaccuracyofsurrogates’responsesisanunexploredarea(9).Thisisprobablyduetothedifficultiesinconductingthistypeofresearchandinidentifyingwhichfactorsfromthecomplexemotionsassociatedwithgriefimpingeonthememoriesofbereavedrelatives(9).Finally,thedatacouldbeinfluencedbyselec-tionbias,becauseitispossiblethatrelativeswithrelativelystrongpositiveornegativeexperiencesagreedtoparticipate inthestudies.Also,selectionbiascouldhavebeenintroducedbecauseclinicalstaffmemberswereaskedtoidentifyeligibledecedentsandrelatives.

Quantitative studies: the surveys

Thestrengthofquantitativeresearchisthatthedataarestandardizedandnumerical.Statistical analysis allowedus to infer important insights in researchdata, includingdifferences between groups (e.g. sedation versus no-sedation) (chapter 5), and tocontrol for the effects of extraneous variables that might result in confounding in-terpretationsofdataanalyses,morespecificallyofcausality(chapter7).Quantitativestudies nowadays encounter more often that all kinds of information are difficult toobtain through structured survey instruments (by high non-response for instance),particularlyon sensitive topics suchasdeath,dyingandbereavement.Although thequestionnairewasdevelopedusingexistingscales(e.gforqualityofdying)itwasnotexaminedwhetherthescalesperformedwellintermsofconcurrentvalidity,internalconsistency,andreliability(chapter5).Forthedevelopmentofthequestionnaire,vali-datedmeasurementtoolswereusedsuchastheSF-36(chapter5).Howeverthesehavenotbeenvalidatedwithbereavedrelatives.Furthermore,asexplainedearlier,recallbiascouldhaveinfluencedthedataandselectionbiascannotberuledout(chapter5and6).Finally,theresponseratesinbothquestionnairestudiesweremoderate,althoughthisisnotuncommon(10-12).

Chapter 8144

mixed methods

Structuredquestionnairesandsemi-structuredinterviews/focusgroupsareoftenusedinmixedmethodstudiestogeneratemorecompleteknowledge,despitedifferencesinmethodsofdatacollection,analysis,and interpretation.The integrationofquantita-tiveandqualitativedataintheformofamixedmethodsstudyhasgreatpotentialtostrengthen the accuracy and may enrich the analysis and conclusions. Nevertheless,integratingqualitativeandquantitativedataduringanalysisiscomplexandchalleng-ing (13). One of the main problems facing many mixed methods researchers is thequestionofhowtoweighandtointegratethevariousdata,withtheparticularproblemof ‘contradictory’findings (14). Ithas tobekept inmind that suchconflictsmaybemerelytheoutcomeofthefactthatsocialreality(includingmemoriesofrespondents)is complex and can at times be conflicting (14). However, results do not necessarilyhavetoexcludeeachother.Forinstance, inthesurveywefoundthatrelativesareingeneralsatisfiedwiththe informationtheyreceivedfromcaregivers(chapter5),andintheinterviewsandfocusgroups,manyrelativesseemedtobedissatisfiedwiththeprovisionofinformation(chapter3and4).Thesedifferencescouldsimplybetheresultofdifferencesinsample,sinceitispossiblethatrelativeswithrelativelystrongnegativeexperiencesagreedtoparticipateinthequalitativestudies.Differenttechniqueswerealso mixed in the study on complicated grief among older bereaved adults (chapter7).Weuseddataofhome interviewswitholderadults,physicalexaminationsat theresearch center, Nationwide Medical Registry, GPs’ records of the older adults, andmedicalfilesofthedeceased.Thismixtureofmethodshascomplementarystrengthsandnon-overlappingweaknesses,suchasincompletemedicalfiles(chapter7).

8.3 interPretation of the finDinGs

8.3.1 relatives’ evaluation of the practice of palliative sedation

Positive experiencesOneof thekeyfindingsof this thesis is that relatives ingeneral reflectpositivelyonpalliativesedation for theirdyingrelative.Manyrelativesbelieved thesedationcon-tributed to the patient having a ‘good’ death. Some relatives even used descriptionslike‘beautiful’,‘peaceful’,‘wonderful’or‘dignified’todescribethedeathofthepatient.

A ‘ good’ deathA‘good’death isan importantaimforhealthservicesandforusall.Althoughrela-tivesmaydifferintheirunderstandingofa‘gooddeath’,thisthesisshowedthat,intheopinions of bereaved relatives, palliative sedation contributed to the patient having


a ‘good death’. In several studies we found that the most important element of thepositiveevaluationofcontinuoussedationfromtheperspectivesofrelativesisrelatedtothebeneficialimpactofpalliativesedationonthepatient’ssuffering(chapter2-4).Before the start of the sedation, patients suffer from multiple distressing symptoms.After thestartof thesedationthesesymptomsseemtohavebeenrelievedandmostpatientsseemcomfortable.Theresultsfromoursurveysuggestthatadequatesymptomreliefpossiblyexplainswhytherearenodifferencesinrelatives’experienceofthedyingphaseforpatientswhodiedeitherwithorwithoutsedation(chapter5).Anotherele-mentofrelatives’positiveevaluationistheopportunitythatthesedationoffersthemtosaygoodbyetothepatient.ThisespeciallyholdstrueforrelativesfromtheNetherlandsandBelgium,whooftendescribesayinggoodbyeasaplannedevent,occurringeitherbeforesedationbeginsoratthemomentitcommenced(chapter4).

Ourresultsareinlinewithpreviousstudiesthatadequatesymptomreliefiskeytotheexperienceofa‘good’death(2,15-17).Theimportanceof‘providingdesiredphysicalcomfort’ iswidelyrecognizedinexpertguidelinesandconfirmedinresearchamongpatients, families, and health care providers. A study by Steinhauser et al (2) thatfocused on determining important factors at the end of life as reported by patients,theirfamilies,physicians,andothercareprovidersrevealedthatthelargemajorityofrespondentsendorsedtheimportanceofpainandsymptommanagement.AstudybyTenoetal(17)showedthatforbereavedfamilymembers,providingthedesiredlevelofphysicalcomfortwasparamounttotheirperceptionofthequalityofcareprovidedbyhealthcareprofessionals.Familymembersemphasizedtheimportanceofhealthcareprovidersanticipatingtheneedsforphysicalcomfortandrespondingquicklytotheirreportsofpain.Theimportanceofsayinggoodbyehasalsobeendescribedbyothers(16,18).Rietjensetal(16),whostudiedthepreferencesoftheDutchgeneralpublicforagooddeath,foundthatthepossibilitytosaygoodbyetolovedoneswasconsideredimportant for a good death by the large majority of respondents. The reasons whyrelativestendtoevaluatetheprovisionofsedationtotheirdyingoverallaspositiveareconsistentwiththefactorsthatgenerallyareconsideredimportantattheendoflife.

The decision-making processTheimportanceofinvolvementofrelativesofsedatedpatientsinthedecision-makingprocess is stressed in the existing literature. Caregivers reported that relatives areinvolvedinthedecision-makingaboutsedationin81%-100%ofallcases(chapter2).Thequestionnairestudydemonstratedthatalmostallrelativesareactivelyinvolvedinthedecision-makingprocessandappreciatethisactiveroleverymuch(chapter5).Thequalitative studies (chapter 3 and 4) additionally showed that caregivers sometimesdiscussed thedecision to start sedationwith the relatives, sometimes theyasked for

Chapter 8146

theirconsent tobeginsedation,andsometimes they informedthemabout thedeci-sion(chapter3).Althoughthephysicianbearsthefinalresponsibilitybydeterminingwhether the medical indications are present, it is recommended by guidelines that,ifpossible, thepatient,his familyand thehealthcare staff shouldbe involved inallstages of the decision-making process (19). This process is referred to as a shareddecision-making:aninteractive,collaborative,andongoingprocess inwhichhealth-careprofessionals,patientsandrelativesaremutuallyengagedbysharinginformation,wishesanddecisions(20-22).Multipledescriptivestudiesshowthatincreasedshareddecision-makingisassociatedwithgreaterfamilysatisfaction(23).Ourresultssupporttheimplementationandtheimportanceofthisinteractiveandcollaborativeprocess.

Ithastobekeptinmindtoothatrelativesvaryintheextenttowhichtheywanttobeinvolvedindecision-making.Theliteraturereviewindicatedthatrelativessometimesfeel the responsibility for the decision to start sedation as a burden (chapter 2). Aprevious study also concluded that family members sometimes feel that they aloneareresponsiblefordecisionsattheendoflife(17).Relativescanplayadecisiveroleindrawingattentiontoapatient’ssituationandcareneedsbecauseoftheirregularclosecontactwiththepatient.TheDutchnationalguidelineforpalliativesedationstatesthatdeterminingwhetherthereareindicationsforpalliativesedationisamedicaldecision(19).Itshouldbekeptinmindthatrelativesmaysometimesexperiencethisdifferentlyandsometimesassumeresponsibilityforthedecisiontostartsedation.Althoughthedecision-makingshouldbeaninteractive,collaborativeandon-goingprocessinwhichhealthcareprofessionals,patientsand relativesmutually engage (ifpossible), clinicalpracticeinvolvesclearboundariesbetweentherelatives’responsibilitiesandthoseofthemedicalcaretakers.Itshouldnotbepossibleornecessaryforrelativestodecideifpalliativesedationisnecessary.Thisboundaryshouldbeexplicitlydiscussedwiththerelatives.

Other positive elementsSometopicswerementionedlessoftenbytherespondents,buttheycanalsobecon-sideredassignificantelementsinrelatives’evaluationofpatients’endoflifecare.Forinstance,aswithotherstudies(1,24),theinterviewstudyunderlinedtheimportanceofcarebeing in linewith thepatient’swishes(chapter4).Sedationwasperceivedascontributingnotonlytothepatient’squalityofdyingbutalsoasameansofhonoringthe patient’s wishes. Physicians are strongly encouraged to address end-of-life carepreferencesofallpatientswithprogressiveterminalillnesses(25).Whenpossible,themannerofreliefofextremedistressshouldbediscussedbeforeenteringthepalliativesedationprocess.Thisshouldincludediscussionsregardingtheindicationofsedation(25). This thesis highlighted the importance of discussing patient’s preferences and


wishes in advance. It is known that both professionals and relatives fear entering amedicalcrisiswithoutknowledgeofpatientpreferences(3).However,ithastobekeptinmindthatsometimesacutesedationhastotakeplaceformedicalreasons.Insuchcases, theattendingphysicianhas tomake thedecision tousepalliativesedationonthebasisofthepatient’scondition,anditissometimesimpossibletoconsultwithallconcerned(19).

ConcernsRelatives in general reflect positively on palliative sedation for their dying relative,but this thesis also revealed a wide variety in concerns that relatives express due tosedation. These related for instance to anxieties about the patient’s wellbeing, theirownwellbeing,andquestionsaboutwhethercontinuoussedationhadshortenedthepatient’slife(mostlyUK),orwhetheranalternativeapproachwouldhavebeenbetter.Manyoftheaforementionedconcernsseemtoberelatedtoalackofinformationand/or communication with health care professionals. Communication is one of the keyelementsofqualityofendoflifecareforseriouslyillpatientsandtheirfamilymembers(26),andeffectivecommunication is thereforerecognizedasapriority inhealthcare(27).Communicationhasbeenidentifiedasthemostimportantfactortoexplainvari-anceinrelatives’satisfactionwithendoflifecare(28,29).

Information and communicationOurliteraturereview(chapter2)demonstratedthatrelativesingeneralseemtoreceiveadequateinformation,althoughthereisroomforimprovement.Inthesurveywefoundthatrelativesareingeneralsatisfiedwiththeinformationtheyreceivefromcaregiversaboutpatient’ssituationandcareduringthelastweekoflifeandtheyseemtounderstandtheinformationtheyreceive(chapter5).Thepercentagethatwasnotsatisfiedwiththereceivedinformationwasrelativelysmall,namely14%.Theinterviewsandfocusgroupsprovidedmoreinsightinthesenegativeexperiences(chapter3and4).Somerelativesspecificallymentioned the lackof informationorcommunication(chapter3and4).Thenatureoftheconcernsandthequestionsrelativeshadregardingtheconceptofse-dationsuggestthattheserelativesindeedreceivedtoolittleinformationregardingtheuseofsedation,orthattheywerenotabletomakesufficientsenseoftheinformationthatwasprovided to them(chapter3and4). Issues surroundingcommunication inhealthcarearenotuncommon.Manystudieshavehighlightedinadequateprovisionofinformationandpoorcommunicationinendoflifecare(30-36).Forinstance,patientssometimesreceivelessinformationandinvolvementintheircarethantheydesire(36).Further,relativeswerefoundtobedissatisfiedwithcommunicationandinformation(32),relativeshadtomakegreateffortstogetinformation,theyhadtobeobstinateandcontinuetoaskquestions(33)orinformationwasdifficulttoget(32,33).

Chapter 8148

It seems that relatives often agree with the decision to start sedation to relieve thepatient’s severe suffering. They are relieved that there is a solution available (‘some-thinghastobedone’).However,notallrelativesunderstandorareabletooverseetheimplicationsofthisdecision.Intheinterviewstudyitwasshownthatsomerelativesfromallthreecountriescontinuedtobeunsureabouttheuseofsedation(chapter4).Reflectingonthedeathoftheirfamilymember,somerelativesstillhadquestionsandcertainmisunderstandingsaboutwhatsedationhadentailed.It isknownthatphysi-ciansfrequentlyfailtoassesspatients’degreeofunderstanding(12,37).Ontheotherhand, patients sometimes seem to overestimate their own understanding (37). Thesamecouldholdtrueforthepatient’sfamily.

Whereas communication in health care in general is difficult, this especially holdstrueforcommunicationinendoflifecare.Dealingwithdeathinvolvesamultitudeofcomplexissuesforpatients,healthprofessionals,andfamilymembers(38).Researchhasattemptedtoclarifythenatureandproblemsofthesecommunicationissues.Someofthereasonswhydoctorsandnurseshaveproblemswithcommunicationinpallia-tivemedicinearehavinginadequateskills,notknowinghowtohandleanemotionaloutburstofrelativesandpatients,worriesaboutcontainingone’sownemotions,fearofprovokingemotionaldistress,fearofbeingblamedbypatientsandrelativesfor(pos-sible) failure, over identification with certain patients and, having to confront one’sownfearsaboutdeath(39,40).Conversationsaboutcaregoalsareoftenconductedbyphysicianswhodonotknowthepatient,donotroutinelyaddresspatients’nonmedicalgoals,andoftenfailtoprovidepatientswithsufficientinformationaboutprognosistoallowappropriatedecisions;and,inaddition,theytendtooccursolateinthepatient’sillness that their impactoncareprocesses isreduced(38).Also,relatives themselvessometimesexperiencedifficulties.Theycanbereluctanttostarttheconversationwiththephysicianornurseaboutthesensitivesubjectofdeathanddying.Astudyperformedin2005showedthatfamilycaregiverssometimesexpressambivalenceaboutwhattheywanttoknow,andtheysometimeshavedifficultiesincomprehendingandaccepting“badnews” (41).Further, theamountof informationapersonwants candiffer, andthiscanvaryandchangeovertime(30).Ourreviewconfirmedthislargevariationinthe‘needs’ofrelatives.Relativeswantspecifictypesofinformationregardingtheuseofsedation;theinformationneedstobeeasilyavailableandrelevanttotheirneedsataparticularmomentintime(chapter2).Difficultiesincommunicationaboutend-of-lifeissuesarelikelytoresultfrombothphysicians’lackofcommunication,butalsofromfamilycaregivers’difficultywithhearingthenews(41).


It is known in end of life care that inadequate information may result in feelings ofisolation, disillusion and distress for the relatives involved (30). Our results supportthefindingsfrompreviousstudies.Relativesexpressedanger,frustration,disappoint-ment,concerns,guiltandhelplessness,partlydue to the fact that informationaboutthe sedationwasnoteasilyobtainedor less relevant toneedsof the relativesat thatmoment(chapter2).

8.3.2 the potential life-shortening effect of palliative sedation

Ascontinuoussedationuntildeathhasbecomeroutinepartofcommonmedialprac-tice,ithasalsoturnedintoaveryrelevanttopicinmedical-ethicaldiscussions(42,43).A topic thathasoftenbeendebated in the literature iswhethercontinuoussedationshortens life.Theofficialguidelinesassumethatcontinuoussedationuntildeathhasnolife-shorteningeffectforpatientswithanestimatedlifeexpectancyofatmosttwoweeksandwhensedativesareproperlydosed(44,45).Oursurveyindicatedthatphysi-ciansoftenhavetheopinionthatcontinuoussedationuntildeathhasa(mostlylimited)life-shorteningeffect(chapter6).Ithastobenotedthatestimatinglifeexpectancyofpatientswithanadvanceddiseaseisdifficult(46-48).Physicianstendtooverestimatesurvival:ithasbeenassessedelsewherethatsurvivalofpatientsistypically30%shorterthanpredictedbyphysicians,butthattheaccuracyofphysicians’predictionsincreaseswhendeathapproaches(47).Recommendationsinguidelinesthatcontinuoussedationuntil death should only be used for patients with a life expectancy of less than oneor two weeks, may therefore be rather difficult to translate to clinical practice. Butevenifsedationcanshortenthepatient’slife,manyphysiciansdonotconsiderthisasproblematic.Ithasbeenheldthatthisisjustifiableunderwhatisknownasthedoctrineofdoubleeffect(49).Thisdoctrinestressesthataharmfuleffectoftreatment(e.g.theshorteningoflife),evenresultingindeath,ispermissibleifitisnotintendedandoccursasasideeffectofabeneficialaction(thereliefofsuffering)(49).

Relatives’ perspectiveVeryfewrelatives inourstudiesmentionedthepotential life-shorteningeffectasanimportantdisadvantageofpalliativesedation(chapter3and4).Inmostcasesrelativesdidnotraisethis issueatall,butsometimesrelativesspecificallystatedthatthiswasnotseenasaproblematicresultofsedation.Somerelativesinourfocusgroupstudyreferredtosedationas‘sloweuthanasia’(chapter3).However,apossibleresemblanceofpalliativesedationtoeuthanasia,inthesenseofhasteningdeath,seemstobenoissuefortheserelatives.Moreingeneral,wefoundnosignificantdifferencesinrelatives’sat-isfactionwiththedyingphasebetweenrelativesofsedatedandnon-sedatedpatientsinthesurvey(chapter5).Apparently,intheviewsofrelatives,theimportanceofadequatesymptomreliefoutweighsthepotentiallifeshorteningeffectasaresultofsedation.In

Chapter 8150

otherwords,ifthegoalofrelativesisa‘good’death,symptommanagementispursuedvigorously,evenwhenthatpursuithastheunintendedconsequenceofcompromisingsurvival.ADutchstudyfrom2013showedthatmostofthegeneralpublicaccepttheuseofpalliativesedationattheendoflifetoalleviaterefractorysymptoms(50).Likeinourstudies,theyfoundthatthesufferingandwishesofthepatientareconsideredofgreaterimportancethanlifeexpectancyandthepotentialhasteningdeath.Relativesinourfocusgroupsandinterviewstudyperceivedalongdurationofthesedationoftenasburdensome(chapter3and4),andashortintervalbetweenthestartofthesedationandthedeathofthepatientwassometimesevenappreciated.Somerelativesdescribedthe time between saying goodbye and the patients’ death as a ‘vacuum’ where theywere ‘waiting’ for thepatient todie(chapter4).Anotherrecentpublicationrevealedthat during the sedation, family members sometimes become agitated and start toputpressureon thedoctor toend thedyingprocess (51).Somephysicians reportedelsewherethattheyindeedhastenedthedyingprocess,inordertorelievethefamily’ssuffering(51).Thesefindingsareincontrastwiththefrequencywithwhichthepoten-tiallife-shorteningeffectisdiscussedasaproblematicaspectofpalliativesedationintheliterature(43,45,52,53).Whereasthisissuetendstoevokeratherheateddebates,theopinionsofbereavedrelativessuggestthatthebenefitofpalliationshouldperhapshaveamoreprominentplace inthemoralevaluationthantheharmofapotentiallyhasteneddeath(50).

International differencesIntheNetherlandsandinBelgium,apotentiallife-shorteningeffectofpalliativeseda-tiontheresemblanceofpalliativesedationtoeuthanasiaseemstobenoissueformostof the relatives. However, some concerns regarding the potential hastening of deathwere expressed by relatives of the UK. Previous results from the UNBIASED studyfoundthatintheUK,anoverarchingconcernexistsamongprofessionalcaregiverstoavoidhasteningdeath,whilemostoftheDutchandBelgianhealthcareprofessionalshadnoconcerns thatsedationhasteneddeath,oraccepted that itmayhavesuchaneffect(54).BelgianandDutchdoctorsandnursesareworkinginaculturewheredelib-eratelyendingapatient’slifeisaratheracceptableprocedureinlawandinprofessionalcodes,aslongascertainconditionsaremet,andthisisalsopublicallyaccepted,withpatientsandrelativesclearlyusedtoconsideringeuthanasiaasanoption(55).Itseemslikelythatlegalizingeuthanasiabothinfluencescarepracticesandperceptionsofwhatisimportantinethicaldebatesaboutsedation.However,itmightalsobepossiblethatculturalnormsinthesecountriesaffectpeople’spositivejudgmentsoftheacceptabilityoftheseend-of-lifecarepractices.TheUKmedicalcultureofend-of-lifecare,ontheotherhand,isonethatisdeeplyinfluencedbythehospicemovementwhich,althoughit contains secular components, has significant religious overtones (55). This means


thattheviewthatlifehasintrinsicvalue,evenwhenitinvolvessuffering,isperhapsastrongerguidingprinciplethanitisinBelgiumandtheNetherlands.Withtheadditionofa legalprohibitionagainstassisteddying(euthanasiaorphysician-assisteddying),theuseofsedativesinUKend-of-lifecarehasaverydifferentethicalcomplexionforcareproviders.Again,thishighlightstheinfluenceofculturalnormsandexpectationsonexperiencesofcontinuoussedationandthesubsequentdeath.

8.3.3 relatives’ wellbeing after the patient’s death

Bereavementcaninfluenceeveryaspectofwellbeing,fromphysicalandmentalhealthtofeelingsofconnectednessandtheabilitytofunctionatwork(56).Adeathhasahugeeffectonthoseleftbehind.Forinstance,relativeshavetotakeonnewresponsibilities,move,oradjusttodifferentlivingstandards(56).Althoughthelossofsomeonecloseisoneofthemostpainfulexperienceswecanencounter,researchersrarelystudyrela-tivesbeforethedeathofthepatientinordertoassesstheeffectsofbereavement(57).Sincethedeathsofmostpersonsareprecededbychronicdisease,disability,andfamilyinvolvement in caregiving, it is important to assess responses to bereavement whenend-of-lifecareisprovideduptothepatient’sdeath(57).

Studyamongbereavedolderadultssuggestedthatcomplicatedgriefinolderadultsisnotclearlyrelatedtothecircumstancesofdyingofthedeceasedpartner(chapter7).Onlypre-existingconditionssuchasdepressionseemtobeassociatedwiththeoccur-renceofcomplicatedgrief inthispopulation.Thisisasurprisingoutcome,sincethecircumstancesofdyingofapartner(e.g.durationofthelethalillnessandthenatureofdeath)havebeenconsideredusefulinidentifyingwhoisatriskofcomplicatedgriefinotherstudies(59).Arecentstudyamong217familycaregiversofpersonswithdementiademonstratedthatbereavedfamilymembers’perceptionsofthequalityofend-of-lifecare are associated with complicated grief (57). It was made clear that when deathwasprecededbyaprotractedand stressfulperiodof caregiving, caregivers reportedconsiderablereliefatthedeathitself.Theintensityofcaregivingwasnotconsideredinouranalysis,whichcouldpotentiallyexplaintheseconflictingfindings.Understandingbereavementrequirescloseattentiontothecontextinwhichthedeathoccurs.

It isoftenassumed that theuseofpalliative sedation isparticularlyburdensome fortherelativesofdyingpatients,e.g.becauseithamperscommunicationbetweenthemand thepatient in thevery lastphaseof life.Our studies (chapter2-4) clearly showthat relativesmayexperiencedistressdue tosedation.Reasons for thisdistressweree.g.thefeelingthatthepatientstillsufferedduringthesedation,the(long)durationofthesedation,orthefactthatinformationaboutthesedationwasnoteasilyobtained.However,theresultsofoursurveypointedtothefactthattheuseofsedationdoesnot

Chapter 8152

seemtohaveanegativeinfluenceonbereavedpersons’experienceofthedyingphaseoftheirdeceasedrelativeortheirwellbeingafterthepatient’sdeath(chapter5).Relativeshavebeendemonstratedtotendtoevaluatetheprovisionofpalliativesedationfortheirseverely suffering family member positively because the patient’s suffering is finallyalleviated(60).Probably,relativesseemtoperceivesedationasan‘appropriatesolution’,because it managed the distress of the patient and relieved the burden placed uponthemselvesbyhavingtodealwiththesometimesoverwhelmingpatientcareneeds.

We conclude that not the specific circumstances of dying, nor the use of palliativesedationattheendoflife(incomparisontootherend-of-lifepractices)seemtohaveaconsiderablelong-termimpactonrelatives’wellbeingafterthepatient’sdeath.Onetheonehand, theseresults suggest thatother factors seemtobemoresignificantormeaningful for relatives’ long-term experiences with the care for their loved one attheendof life.Thefact that thepatient’swellbeing ispreservedandthat thepatientreceivesthebestcarepossibleconsideringthecircumstancesseemtobeprerequisitesforapositiveevaluation.Further,thisthesisshowedthatpre-existingconditionssuchasdepressionseemtobeofimportanceaswell.Itisthereforenecessarytofindwaystoassesscarerswhoareatriskofdepression.Ontheotherhand,thefactthatrelativessometimesexperiencedconcernsduringtheuseofsedationbutdonotexperiencelongtermconsequencesfortheirwellbeingafterthepatient’sdeathsuggeststhatcliniciansandscientistsshouldviewbereavementnotonlyasaphenomenonthataffectsrelativesafterthedeathofthepatient,butmainlyasonethataffectsmanyrelativesbeforethedeathoccurs(57).Itispossiblethatwhenrelativesknowthatthedeathoftheirlovedoneisapproachingandareawareofthepatient’ssuffering,theygrieveforthelossofthepatientbeforethedeath(57).Thistypeofgriefisalsoknownasanticipatorygrief.Theconceptofanticipatorygrievingcouldplayanimportantroleinpalliativesedation,sincethepatienthasalreadybeenbroughttosleepanddeathisanticipatedwithinshortperiodoftime.Byprovidingasupportiveandsafeenvironment,physiciansandnursescanhelptherelativesunderstandthattheirfeelingsarecommonandareexperiencedbyothersinsimilarsituationsandassistthemwithdevelopingcopingstrategies(61).

8.4 implications for practice, policy and future research

8.4.1 PracticeThisthesisprovidesastrongbasisfordevelopingstrategiesforevidence-baseduseofpalliativesedationandforenhancingthewellbeingofpatients’relatives.To help relatives to cope with their distress, an important task for physicians is toprovidefamilymemberswithappropriateandclearinformation,clear,andtoprovidethisinformationinacompassionatemanner.


Healthcareprofessionalsshouldfocusonprovidingfullinformationtorelativesbeforesedationbeginsandafteritcommencedaboutsedationingeneral,(thelackof)poten-tialalternativestotheuseofsedation,thepatient’slifeexpectancy(althoughestimat-inglifeexpectancyisknowntobeverydifficult),thedrugsthatwillbeused,furthertreatment(e.g.artificialnutritionandhydration),thewellbeingofthepatientduringthesedation,theexpecteddurationofthesedation,and/orthepossiblesymptomsorreactionsofthepatientduringthesedation.

Relatives are, however, not always able to make sufficient sense of the informationthatisprovidedtothem.Thiscouldbebecausetheinformationiscommunicatedtotheminalanguagethattheycannotunderstand,theyreceiveconflictinginformationfromdifferentprofessionalcaregivers,ortheysimplyhavetroubleunderstandingtheinformation due to the fact that is provided to them in a situation where they areextremelydistressed(chapter3).Providingextendedinformationandregularupdatesisimportantandneedsmoreattentioninpractice.Ourdatashowthatitiscrucialthatthereisacommonunderstandingoftermsandphrasesforallactorsinvolved(patient,relativesandhealthcareprofessionals)andthatchecksshouldbemadesystematicallytoensurethisunderstandingismaintainedthroughouttheprocess.

However,thecoordinatedsharingofvitalinformationamongpatients,patients’fami-lies,andtheprofessionalcareteamissometimesdifficulttoaccomplish(62).Severalsuccessfulinterventionshavebeenidentifiedtoimprovecommunicationinendoflifesuchastheteamapproachtocommunication,theformalfamilymeeting,anduseofadvanced practice nurses (63). Also, communication and assessment strategies areavailableforcareprovidersthatfacilitateend-of-lifedecision-making(64).Communi-cationstrategiesinclude:beingclear,avoidingeuphemisms,spellingoutthegoalsandexpectationsoftreatment,usingwordssuchas“death”and“dying,”andbeingspecificwhenusingsuchwordsas“hope”and“better.”Assessmentstrategiesinclude:assessingpatients’ physical conditions and end-of-life wishes, patients’ and family members’understandings of the disease and prognosis, and their expectations and goals (64).Health care professionals should examine which strategies are available and mightbeuseful tosolve thepotentialcommunicationand information issues that theyareexperiencingwithrelativesintheuseofpalliativesedation.

Sincetheprovisionofwell-communicatedcarerequiressignificantstafftimeandeffort,andoneofthemostfrequentlyendorsedbarriersforphysiciansisacontinuouslackoftime,nursestooshouldhaveasignificantroleandtimeinidentifying,observing,measuringandreportingondevelopments.Palliativecarenursingisakeycomponent

Chapter 8154

inthemultidisciplinaryapproachnecessarytomeetthecomplexneedsofindividualsandtheirfamiliesfacingalifethreateningillness.Nurseshavemorefrequentcontactwith patients and their families than physicians, and may be more likely to observechangesinpatientandfamilyneeds.

Physicians seem to be reluctant to consult experts about palliative sedation (65).Whereasmanyphysicianshaveonlylittleexperiencewithprovidingpalliativesedationandcommunicatingitpatientsandtheirrelatives,increasedconsultationofanexpertpalliativecareteammaybehelpfulhere.

Tosupportpatients,physicians,nursesandrelativesininitiatingendoflifecarediscus-sions;advancecareplanningisencouraged(66,67).Earlydiscussionsaboutgoalsofcareareassociatedwithbetterqualityof life,reduceduseofnon-beneficialmedicalcareneardeath,enhancedgoal-consistentcare,positivefamilyoutcomes,andreducedcosts (68).Especially since, lookingback, relativesunderline the importanceof carebeing in line with the patient’s wishes. Ongoing discussions with patients and theirfamiliesareneededtoensuretheircurrentwishesforend-of-lifecarearebothknownand followed.Arecent reviewshowed thatbestpractices indiscussinggoalsof careinclude the sharing prognostic information, eliciting decision-making preferences,understanding fearsandgoals,exploringviewsontrade-offsand impaired function,andwishesforfamilyinvolvement(68).However,inpractice,acutesituationsalwayscanariseinwhichcarryingouteffectiveadvancecareplanningishampered.Advancedirectives represent one method to provide patients with the means to proactivelydeterminetheirfuturecare(69).However,advancedirectivesdonotandshouldnottaketheplaceofongoingdiscussionswithpatientsandtheirfamiliesregardingwishesforend-of-lifecare(69).

Finally, since relatives emphasize the importance of being prepared for the patient’sdeathandparticularlyhighlighttheimportanceofsayinggoodbye,itisstronglyrec-ommendedthatprofessionalsshouldcoordinatetheprocessinawaythatpatientsandtheirfamilymembersalwayshavea‘last’chancetocommunicateabouttheirfeelingsorthoughtsbeforepatientsreceivesedation(70).Whenthepatientishospitalized,everyeffortshouldbemadetoprovideroomforprivacyforemotionalandphysicalintimacytogivepatientsandtheirrelativesanopportunitytosaygoodbye(25).

8.4.2 PolicyThisthesisprovidedinsight intotheexperiencesofrelativeswithpalliativesedation,and the effect of palliative sedation on the wellbeing of relatives after the patient’s


deathandcharacteristicsthatareassociatedwiththis.Implementingthesefindingsinguidelinesisanexcellentmeanstoimprovecareforthedyingandtheirlovedones.

Accordingtoguidelinesonpalliativesedation(19,25,71,72),relativesshouldbein-volvedinthedecision-making,e.g.bydiscussingthedecisiontosedate.Furthermore,relativescanbeinvolvedintheprovisionofthesedation,e.g.byspendingtimewithandobservingthepatientandtoprovidephysiciansandnurseswithinformationaboutthepatient.Therelatives shouldbekept informedatvariouspoints in thecourseofpalliativesedation,aboute.g.patient’swellbeingandwhattoexpect,andthecareteamshouldcommunicatetotherelativesina languagetheycanunderstand.Inlinewithrecommendationsincurrentguidelines,relativesseemtobeadequatelyinvolvedinthedecision-makingprocessofsedationandinthecareforthedyingpatient.Further,rela-tivesseemtoreceivesufficientsupportfromcaregivers.However,thisthesismadeclearthattheinformationprovidedbycaregiversisnotalwayssufficient.Assaid,relativesneedmoreinformationone.g.thepatient’swellbeingduringsedation,possiblealter-natives tocontinuoussedationandthepatient’s lifeexpectancy.However,guidelinesalreadyrecommendthistypeofinformation.Followinguponrelatives’mainconcerns,nomajorchangesormodificationsoftheguidelinesseemtobenecessary.ItshouldbenotedthattheDutchguidelinecanserveasanexampleforotherguidelines,whichareoftenfarlesscomprehensivewithregardstocareforthepatient’srelatives.

However,thefactthatmanyrelativesdoexperienceconcernsduetotheuseofsedationcannotbeignored.Apparentlysomekindoftensionexistsbetweenwhattheguidelinerecommendsandhowthisworksoutinpractice.Thefocusinpolicyshouldthereforemerelybeonimprovingthequalityoftheapplicationoftheguidelinewithrespecttothecareforpatients’relatives.Clinicalauditsofthepracticeanddocumentation(e.g.ofthedecision-makingprocessandconsultationwiththepatient’sfamily)ofpalliativesedationperformedbyhealthprofessionalsorexternalorganizationswillforinstancebeaneffectivetooltorecognizelearningopportunities,andtofosterpracticechange(73).

Guidelines highlight best practices in treating the array of problems patients andtheirrelativesfaceattheendoflife.However,ithastobekeptinmindthatpersonalperspectives often emphasize diverse elements, and those perspectives will never beperfectly integrated inonepolicy.Healthcareprofessionals shouldbeawareof thesediverseneedsandactflexibleinthisrespecttobestpromoteandprotecttheinterestsofpatientsandtheirfamilies.Ultimately,theresponsetoadifficultclinicalsituationisborninpractice,takingintoaccountlegal,medicalandethicalregulations,guidelines

Chapter 8156

or principles (74). By providing a practical and legal framework, guidelines offer astartingpointforsuchresponses.

The main premise of the Royal Dutch Medical Association’s (RDMA) guideline onpalliativesedationisthatpalliativesedation,contrarytoeuthanasia,isnormalmedi-calpractice(19).Oneofthecrucialpropositionsoftheguidelineisthatthepatient’slifeexpectancyshouldnotexceedtwoweeks.Thisthesisshowedthatintheviewsofbereavedrelatives,thebenefitofpalliationshouldhaveamoreprominentplaceinthemoralevaluationofpalliativesedationthantheharmofapotentiallyhasteneddeath.Isthereliefofsufferingthemostimportantgoalofsedationandisitthereforeethicallyacceptablethatguidelinerecommendationsonlifeexpectancyarenotalwaysstrictlyfollowed?Onecouldwonderwhetherrelatives’perspectiveonthisparticularissueisthemostrelevantonetofollow. It isdifficult tobalancethedifferentconsiderationsinvolved indecidingwhether tostartcontinuoussedation.Bothphysicians,patientsandrelativeshaverolestoplayinthedecision-makingprocess.However,determiningwhetherthereareindicationsforpalliativesedationisamedicaldecision.

8.4.3 Future researchImprovingthequalityofhealthcareforpatientsat theendof lifeandtheirrelativesshould be a major clinical and research objective. This should not be limited to thepracticeofpalliativesedation.Reviewsoftheliteraturehaveconsistentlyhighlightedalackofresearchonfamilycaregiverswithinthecontextofpalliativecare(75).Priorityresearchareasinclude:interventiondevelopmentandtesting;underresearchedcare-givergroups;access toservices;unmetneeds;bereavement;experienceand implica-tionsofthecaregiverrole;anddevelopmentofassessmenttools(75).

This thesis made an important contribution to research on family caregivers withinthecontextofpalliativecare.Moreinsightwasgainedinareassuchasrelatives’unmetneeds,bereavementandexperiencesandimplicationsofthecaregiverrole.However,empiricalresearchinotherareasisstilllacking.Forinstance,althoughselectionbiaswas limitedinourstudies,certaingroupswerenot includedinoursamples,suchasyoungcaregiversandethnicminorities(e.g.TurkishandMoroccan).Further,mostofourstudiesfocusedontheexperiencesofrelativesofcancerpatients,althoughfuturestudies should also be conducted among other patient populations, for instance pa-tientswithdementia.Finally,tocontributetotheimprovementofcareforthevulner-ablegroupofsufferingdyingpatientsandtheirrelatives,patients’perspectivescouldpotentiallyalsobeincludedinfutureresearch.Ofcourse,researchamongpatientsthatultimatelyneedpalliativesedationischallengingandtherehasbeencontroversyabouttheappropriatenessofinvolvingpalliativecarepatientsinresearch(76,77).However,


researchtodateinthepalliativecaresettinghassuggestedthatpatientsareinterestedinparticipatinginresearch,andmayactuallybenefitfromdoingso(78).

The findings in this thesis also revealed several new areas of research. First, studiesshouldaimtomapandunderstandtheperceivedbarriersandfacilitatorsaffectingtheprovisionofinformationandcommunicationprocesses.Anotherkeyareathatwarrantscomprehensive attention is ‘bereavement’. Since this thesis showed that pre-existingconditions,suchasdepression,canbeimportant inexplainingcomplicatedbereave-mentafterthepatient’sdeath,studiesshouldbeaimedtoassesscarerswhoareatriskofpoorpsychologicalwellbeing.Further,attentionshouldbepaidtothedevelopmentofcarestrategiesthatdealwithpre-loss/anticipatorygriefamongrelatives.

Oneofthemajorstrengthsofthisthesisisthatcombinesfindingsfrommultiplemeth-ods (both qualitative and quantitative) and multiple data sources. However, severalweaknesses were encountered. For instance, it is not always entirely clear whetherparticipantsrefertothesametypeofsedation,whichpotentiallylimitsfullcomparisonandextrapolationofthestudies.Thereforeacase-basedapproach,suchasusedintheinterviewstudy,wouldbehighlyrecommendedforfutureresearch.Thisdesigncouldbesolidifiedusingelementsfromtheotherdesigns,suchastheuseofvalidatedques-tionnaires.Second,aprospective longitudinaldesignwouldberecommended.Sinceonlyaminorityofpatientreceivespalliativesedationattheendoflife(79),thefocusin this type of study should therefore be on relatives in end-of-life care in general.One of the strengths of this design is that it can be used in order to study changesine.g.communicationprocessesandrelatives’behavior.Thisdesigncouldbefurtherextendedwithanobservationalelement,whichisalsoaverysuitablewaytofindoutmore about relatives’ attitudes, beliefs, expectations, and knowledge. Although theprospective element deals with issues such as recollection bias, there are potentiallymedicalethicalchallengesassociatedwiththisdesign.Effortstoprotecttherightsandwelfareofpatientsandtheirrelativesshouldbalancetheneedforprotectionsagainsttheethicalimperativetoimprovecare.

Chapter 8158

references

1. SteinhauserKE,ChristakisNA,ClippEC,McNeillyM,GrambowS,ParkerJ,etal.Preparingfortheendof life:preferencesofpatients, families,physicians,andothercareproviders. JPainSymptomManage2001;22(3):727-37.

2. SteinhauserKE,ChristakisNA,ClippEC,McNeillyM,McIntyreL,TulskyJA.Factorsconsideredimportantat theendof lifebypatients, family,physicians,andothercareproviders. JAMA2000;284(19):2476-82.

3. SteinhauserKE,ClippEC,McNeillyM,ChristakisNA,McIntyreLM,TulskyJA.Insearchofagooddeath:observationsofpatients,families,andproviders.AnnInternMed2000;132(10):825-32.

4. RausK,BrownJ,SealeC,RietjensJA,JanssensR,BruinsmaS,etal.Continuoussedationuntildeath:theeverydaymoralreasoningofphysicians,nursesandfamilycaregiversintheUK,TheNetherlandsandBelgium.BMCMedEthics2014;15:14.

5. World Health Organization. ‘Definition of palliative care’, http://www.who.int/cancer/palliative/definition/en(2008,accessedJanuary2015).

6. HigginsonIJ.End-of-lifecare:lessonsfromothernations.JPalliatMed2005;8(1):161-73. 7. BaylorC,YorkstonK.Usingsystematicreviewsandpracticeguidelines:Ahow-toguideforclini-

cians.NeurophysiologyandNeurogenicSpeechandLanguageDisorders2007;17:6-10. 8. AnquinetL.Continuoussedationuntildeath:ExperiencesofhealthcareprofessionalsinBelgium,

theNetherlandsandtheUnitedKingdom(thesis).Brussel:VrijeUniversiteitBrussel;2013. 9. Addington-HallJ,McPhersonC.After-deathinterviewswithsurrogates/bereavedfamilymembers:

someissuesofvalidity.JPainSymptomManage2001;22(3):784-90. 10. VanDeijckRH,KrijnsenPJ,Hasselaar JG,VerhagenSC,VissersKC,KoopmansRT.Thepractice

ofcontinuouspalliative sedation inelderlypatients:anationwideexplorative studyamongDutchnursinghomephysicians.JAmGeriatrSoc2010;58(9):1671-8.

11. HasselaarJG,ReuzelRP,vandenMuijsenberghME,KoopmansRT,LegetCJ,CrulBJ,etal.Dealingwithdelicateissuesincontinuousdeepsedation.VaryingpracticesamongDutchmedicalspecialists,generalpractitioners,andnursinghomephysicians.ArchInternMed2008;168(5):537-43.

12. MiyajimaK,FujisawaD,YoshimuraK,ItoM,NakajimaS,ShirahaseJ,etal.AssociationbetweenQuality of End-of-Life Care and Possible Complicated Grief among Bereaved Family Members. JPalliatMed201417(9):1025-31.

13. WisdomJ,CreswellJW.MixedMethods:IntegratingQuantitativeandQualitativeDataCollec-27.tionandAnalysisWhileStudyingPatient-CenteredMedicalHomeModels.Rockville,MD:Agency28.forHealthcareResearchandQuality.February2013.AHRQPublicationNo.13-0028-EF.

14. MayV.Whattodowithcontradictorydata?MorganCentre,UniversityofManchester2010. 15. PiersonCM,CurtisJR,PatrickDL.Agooddeath:aqualitativestudyofpatientswithadvancedAIDS.

AIDScare2002;14(5):587-98. 16. RietjensJA,vanderHeideA,Onwuteaka-PhilipsenBD,vanderMaasPJ,vanderWalG.Preferences

of the Dutch general public for a good death and associations with attitudes towards end-of-lifedecision-making.PalliatMed2006;20(7):685-92.

17. Teno JM, Casey VA, Welch LC, Edgman-Levitan S. Patient-focused, family-centered end-of-lifemedicalcare:viewsoftheguidelinesandbereavedfamilymembers.JPainSymptomManage2001;22(3):738-51.




20. CharlesC,GafniA,WhelanT.Decision-makinginthephysician-patientencounter:revisitingthesharedtreatmentdecision-makingmodel.SocSciMed1999;49(5):651-61.

21. ElwynG,EdwardsA,KinnersleyP,GrolR.Shareddecisionmakingandtheconceptofequipoise:thecompetencesofinvolvingpatientsinhealthcarechoices.BrJGenPract2000Nov;50(460):892-9.

22. WhitneySN.Anewmodelofmedicaldecisions:exploringthelimitsofshareddecisionmaking.MedDecisMaking2003;23(4):275-80.

23. HinkleLJ,BossletGT,TorkeAM.FactorsAssociatedwithFamilySatisfactionwithEnd-of-LifeCareintheICU.Chest2015;147(1):82-93.

24. StrangP.Mostoftheterminallyillpatientswanttodieathome.Advancedhomecareservicescanoftenreplacepalliativehospitalcare.Lakartidningen.2002;99(8):742-3.

25. ChernyNI,RadbruchL.EuropeanAssociationforPalliativeCare(EAPC)recommendedframeworkfortheuseofsedationinpalliativecare.PalliatMed2009;23:581-93.

26. HeylandDK,RockerG,GrollD,GafniA,PichoraD,ShorttS,etal.Whatmattersmostinend-of-lifecare:perceptionsofseriouslyillpatientsandtheirfamilymembers.CMAJ2006;174(5).

27. StewartMA.Effectivephysician-patientcommunicationandhealthoutcomes:areview.CMAJ1995;152(9):1423-33.

28. BrazilK,CupidoC,TaniguchiA,HowardM,Akhtar-DaneshN,FrolicA.Assessingfamiliymembers’satisfactionwithinformationsharingandcommunicationduringhospitalcareattheendof life.JPalliatMed2013;16:82-6.

29. RaijmakersN.Endoflifecareanddecisionmaking:Opinionsandexperiencesofthegeneralpublic(thesis).Rotterdam:ErasmusMC,UniversityMedicalCentreRotterdam;2013.

30. AndershedB.Relativesinend-of-lifecare-part1:asystematicreviewoftheliteraturethefivelastyears,January1999-February2004.JClinNurs2006;15(9):1158-69.

31. DunneK,SullivanK.Familyexperiencesofpalliativecareintheacutehospitalsetting.IntJPalliatNurs2000;6(4):170-8.

32. RogersA,KarlsenS,Addington-HallJ.‘Alltheserviceswereexcellent.Itiswhenthehumanelementcomesinthatthingsgowrong’:dissatisfactionwithhospitalcareinthelastyearoflife.JAdvNurs2000;31(4):768-74.

33. McGrathP.Caregivers’insightsonthedyingtrajectoryinhematologyoncology.CancerNurs2001;24(5):413-21.

34. YurkR,MorganD,FraneyS,StebnerJB,LanskyD.Understandingthecontinuumofpalliativecareforpatientsandtheircaregivers.JPainSymptomManage2002;24(5):459-70.

35. ErikssonE,ArveS,LauriS.Informationalandemotionalsupportreceivedbyrelativesbeforeandafterthecancerpatient’sdeath.EurJOncolNurs2006;10(1):48-58.

36. LermanC,DalyM,WalshWP,ReschN,SeayJ,BarsevickA,etal.Communicationbetweenpatientswith breast cancer and health care providers. Determinants and implications. Cancer 1993;72(9):2612-20.

37. O’Leary KJ, Kulkarni N, Landler MP, Jeon J, Hahn KJ, Englert KM, et al. Hospitalized patients’understandingoftheirplanofcare.MayoClinProc2010;85(1):47-52.

38. NorthouseP,NorthouseL.CommunicationandCancer:IssuesConfrontingPatients,HealthProfes-sionals,andFamilyMembers.JPsychosocOncol1988;5(3):17-46.

39. Fallowfield L, Jenkins V. Communicating sad, bad, and difficult news in medicine. Lancet 2004;363(9405):312-9.

Chapter 8160

40. FallowfieldLJ,JenkinsVA,BeveridgeHA.Truthmayhurtbutdeceithurtsmore:communicationinpalliativecare.PalliatMed2002;16(4):297-303.

41. CherlinE,FriedT,PrigersonHG,Schulman-GreenD,Johnson-HurzelerR,BradleyEH.Commu-nicationbetweenphysiciansandfamilycaregiversaboutcareattheendoflife:whendodiscussionsoccurandwhatissaid?JPalliatMed2005;8(6):1176-85.

42. Rys S, Mortier F, Deliens L, Deschepper R, Battin MP, Bilsen J. Continuous sedation until death:moraljustificationsofphysiciansandnurses--acontentanalysisofopinionpieces.MedHealthCarePhilos2013;16(3):533-42.

43. vanDeldenJJ.Terminalsedation:sourceofarestlessethicaldebate.JMedEthics2007;33(4):187-8. 44. RietjensJ,vanDeldenJ,Onwuteaka-PhilipsenB,BuitingH,vanderMaasP,vanderHeideA.Con-

tinuousdeepsedationforpatientsnearingdeath in theNetherlands:descriptivestudy.BMJ2008;336(7648):810-3.


46. BrandtHE,OomsME,RibbeMW,vanderWalG,DeliensL.Predictedsurvivalvs.actualsurvivalinterminallyillnoncancerpatientsinDutchnursinghomes.JPainSymptomManage2006;32(6):560-6.

47. GlareP,VirikK,JonesM,HudsonM,EychmullerS,SimesJ,etal.Asystematicreviewofphysicians’survivalpredictionsinterminallyillcancerpatients.BMJ2003;327(7408):195-8.

48. RietjensJA,BuitingHM,PasmanHR,vanderMaasPJ,vanDeldenJJ,vanderHeideA.Decidingabout continuous deep sedation: physicians’ perspectives: a focus group study. Palliat Med 2009;23(5):410-7.

49. SykesN,ThornsA.Sedativeuseinthelastweekoflifeandtheimplicationsforend-of-lifedecisionmaking.ArchInternMed2003;163(3):341-4.

50. vanderKallenHT,RaijmakersNJ,RietjensJA,vanderMaleAA,BuevingHJ,vanDeldenJJ,etal.OpinionsoftheDutchpubliconpalliativesedation:amixed-methodsapproach.BrJGenPract2013;63(615):676-82.

51. vanTolDG,KouwenhovenP,vanderVegtB,WeyersH.Dutchphysiciansontheroleofthefamilyincontinuoussedation.JMedEthics2014.doi:10.1136/medethics-2013-101624.

52. BillingsJA,BlockSD.Sloweuthanasia.JPalliatCare1996;12(4):21-30. 53. BillingsJA,BlockSD.Opportunitytopresentourobservationsandopinionsonsloweuthanasia.J

PalliatCare1997;13(2):55-6. 54. SeymourJ,RietjensJ,BruinsmaS,DeliensL,SterckxS,MortierF,etal.Usingcontinuoussedation

untildeathforcancerpatients:Aqualitative interviewstudyofphysicians’andnurses’practice inthreeEuropeancountries.PalliatMed2015;29(1):48-59.

55. Seale C, Raus K, Rietjens J, Bruinsma S, van der Heide A, Sterckx S, et al. The language of seda-tion in end-of-life care: The ethical reasoning of care providers in three countries. Health 2014.pii:1363459314555377.

56. A.Lifeafterdeath:Sixstepstoimprovesupportinbereavement.TheNationalCouncilforPalliativeCare,2014.

57. SchulzR,MendelsohnAB,HaleyWE,MahoneyD,AllenRS,ZhangS,etal.End-of-lifecareandtheeffectsofbereavementonfamilycaregiversofpersonswithdementia.NEnglJMed2003;349(20):1936-42.

58. GhesquiereAR,PatelSR,KaplanDB,BruceML.Primarycareproviders’bereavementcarepractices:recommendationsforresearchdirections.IntJGeriatrPsychiatry2014;29(12):1221-9.


59. Aranda S, Milne D. Guidelines for the Assessment of Complicated Bereavement Risk in FamilyMembersofPeopleReceivingPalliativeCare.Melbourne:CentreforPalliativeCare,2000.

60. BruinsmaS,RietjensJ,vanderHeideA.Palliativesedation:afocusgroupstudyontheexperiencesofrelatives.JPalliatMed2013;16(4):349-55.

61. Norton AS, Talerico KA. Facilitating End-of-Life Decision-Making Strategies for CommunicatingandAssessing.JGerontolNurs2000September;26(9):6–13.

62. AhrensT,YanceyV,KollefM.Improvingfamilycommunicationsattheendoflife:implicationsforlengthofstayintheintensivecareunitandresourceuse.AmJCritCare2003;12(4):317-23.

63. FoxM.ImprovingCommunicationWithPatientsandFamilies intheIntensiveCareUnit.JHospPalliatNurs2014;16(2):93-8.

64. NortonSA,TalericoKA.Facilitatingend-of-lifedecision-making:strategiesforcommunicatingandassessing.JGerontolNurs2000;26(9):6-13.


66. FriedTR,BullockK,IannoneL,O’LearyJR.Understandingadvancecareplanningasaprocessofhealthbehaviorchange.JAmGeriatrSoc2009;57(9):1547-55.

67. Singer PA, Robertson G, Roy DJ. Bioethics for clinicians: 6. Advance care planning. CMAJ 1996;155(12):1689-92.

68. BernackiRE,BlockSD,AmericanCollegeofPhysiciansHighValueCareTaskF.Communicationabout serious illnesscaregoals: a reviewandsynthesisofbestpractices. JAMAInternMed2014;174(12):1994-2003.

69. NortonSA,TalericoKA.Facilitatingend-of-lifedecision-making:strategiesforcommunicatingandassessing.JGerontolNurs2000;26(9):6-13.

70. MoritaT,IkenagaM,AdachiI,NarabayashiI,KizawaY,HonkeY,etal.Concernsoffamilymembersofpatientsreceivingpalliativesedationtherapy.SupportCareCancer2004;12(12):885-9.

71. MoritaT,BitoS,KuriharaY,UchitomiY.DevelopmentofaclinicalguidelineforpalliativesedationtherapyusingtheDelphiMethod.JPalliatMed2005;8(4):716-29.

72. BroeckaertB,MullieA,GielenJ,DesmetM,VandenBergheP.GuidelinePalliativeSedation:Eth-icsSteeringCommitteeoftheFederationforPalliativeCareFlanders[Sedation];2012[cited2012October24].Availablefrom:http://www.pallialine.be/template.asp?f=rl_sedatie.htm.

73. McKinnonM,AzevedoC,BushSH,LawlorP,Pereira J.Practiceanddocumentationofpalliativesedation:aqualityimprovementinitiative.CurrOncol2014;21(2):100-3.

74. Swart S. The Practice of Palliative Sedation in the Netherlands after introduction of the NationalGuideline(thesis).Rotterdam:ErasmusMCUniversityMedicalCentreRotterdam;2013.

75. HudsonPL,ZordanR,TrauerT.Researchprioritiesassociatedwithfamilycaregivers inpalliativecare:internationalperspectives.JPalliatMed2011;14(4):397-401.

76. deRaeveL.Ethicalissuesinpalliativecareresearch.PalliatMed1994;8(4):298-305. 77. MazzocatoC,SweeneyC,BrueraE.Clinicalresearchinpalliativecare:patientpopulations,symp-

toms,interventionsandendpoints.PalliatMed2001;15(2):163-8. 78. WhiteC,HardyJ.Whatdopalliativecarepatientsandtheirrelativesthinkaboutresearchinpallia-

tivecare?-asystematicreview.SupportCareCancer2010;18(8):905-11. 79. Onwuteaka-PhilipsenBD,Brinkman-StoppelenburgA,PenningC,deJong-KrulGJ,vanDeldenJJ,

vanderHeideA.Trendsinend-of-lifepracticesbeforeandaftertheenactmentoftheeuthanasialawin theNetherlands from1990 to2010: a repeatedcross-sectional survey.Lancet2012 ;380(9845):908-15.

Chapter 9

Summary Samenvatting Dankwoord / Acknowledgements About the author Publications PhD portfolio

165Summary

sUmmarY

Palliativesedationisthedeliberateloweringofapatient’sconsciousnessinthelaststageoflife.Itmaybeadministeredcontinuouslyuntildeath,ortemporarily/intermittently.Inthisthesisthefocusisoncontinuoussedationuntildeath.Continuoussedationisadministeredinthefinalstagesoflifetopatientswhoaredyingandareexperiencingunbearablesuffering.

Inthefirstchapter,currentknowledgeofthepracticeofpalliativesedationbasedonempirical research and guidelines is discussed. We present an overview of researchreporting the frequency of the use of continuous sedation until death and discussguidelinesonpalliativesedation,withspecialattentiontotheDutchnationalguideline.Further, we summarize findings regarding clinical characteristics of the practice ofcontinuoussedation,forexampleaspectsofthedecision-making.

Relativesplayan importantrole in the trajectorywhensedation isbeingconsideredandwhile it iscarriedout.This thesisaims toprovideacomprehensiveoverviewofbereavedrelatives’experienceswiththepracticeofpalliativesedation.Further,theaimistogainmoreinsightintothepotentiallife-shorteningeffectofpalliativesedationandintheriskfactorsforcomplicatedgriefinolderadults.Toachievethesegoals,severalresearchquestionswillbeaddressed.

Research question 1: What are the experiences of bereaved relatives with the practice of palliative sedation?

Research question 2: What is the impact of continuous sedation on relatives’ experience of the dying phase and their wellbeing after the patient’s death?

Research question 3: Can physicians’ accurately estimate the potential life-shortening effect of continuous sedation until death?

Research question 4: What are risk factors for complicated grief among older adults?

Toguidecaregivers,severalinternational,national,andlocalguidelinesfortheuseofpalliativesedationhavebeenpublished.Theseguidelinestypicallyalsoincluderecom-mendationstoprotectthewellbeingofrelativesofpatientswhoreceivepalliativeseda-tion.Accordingtotheseguidelines,relativesshouldbeinvolvedinthedecisionmakingandtheycanbeinvolvedintheprovisionofthesedation,forexamplebyspendingtimewithandobservingthepatientandprovidingphysiciansandnurseswithinformationaboutthepatient.Further,relativesshouldbekeptinformed,andthecareteammustprovidesupportivecaretotherelatives.

Chapter 9166

Inchapter2theresultsofasystematicreviewontheexperiencesofrelativeswiththepractice of palliative sedation are presented in the light of the recommendations inguidelines.Moststudiesreport that themajorityof relativesareadequately involvedinthedecision-makingandreceiveadequateinformation,althoughthereseemsroomforimprovement.However,hardlyanyinformationisavailableaboutrelatives’involve-mentintheprovisionofsedationandnostudiesreportspecificallyaboutthesupportprovidedtorelatives.Despitethefactthatthemajorityofrelativeswerereportedtobecomfortablewiththeuseofpalliativesedation,ourreviewshowsthattherelativesmayexpressdistressbeforeorduringtheapplicationofsedation.

Chapter3reportsontheresultsofthefocusgroupstudyconductedamongbereavedrelatives.Atotalof14relativesofpatientswhoreceivedpalliativesedationuntildeathparticipated in focus groups and individual interviews. A semi-structured question-nairewasused.Manyrelativeshadpositiveexperienceswiththeprovisionofsedationfortheirdyingfamilymember.Thestartofthesedationisareliefforrelativesbecausethepatient’ssufferingisfinallyalleviated.Relativesoftenappreciatedhavinganactiverole in decision-making and the provision of the sedation. Other positive experi-encesrelatetothedegreeofinvolvementofprofessionalcaregiversincaregivinganddecision-making,andthattheythoughtthattheplaceofdeathwasinagreementwiththewishesofthepatient.Ontheotherhand,severalrelativesindicatedthattheyweredissatisfiedwiththeinformationtheyreceived,andaboutcommunicationingeneral.Othernegativeexperiencesrelatetoconcernsaboutthewellbeingofthepatientduringsedation,especiallywhenthesedationprocesslastslong,thelackofauthorityofnursestomakedecisions,andtheabsenceofphysiciansduringtheweekends.

Inchapter4wedescribean interviewstudyamongbereavedrelatives fromtheUK,BelgiumandtheNetherlands.Itshowsthatmostrelativesareabletoprovideadescrip-tionoftheconceptofsedation.Neverthelessseveralofthemwereunsureaboutwhatitentails.Althoughrelativesgenerallybelievedsedationcontributestothepatienthavingagooddeath,theyalsoexpressedsomeconcernsaboutitsuseandexperiencedsomeunexpectedeventsforwhichtheywereunprepared.Theserelatetoanxietiesaboutthepatient’swellbeing,theirownwellbeing,andquestionsaboutwhethercontinuousseda-tionhadshortened thepatient’s life,orwhetheranalternativeapproachwouldhavebeenbetter.AccordingtorelativesfromtheNetherlandsandBelgium,thestartofthesedationallowsforaplannedmomentof‘sayinggoodbye’.Incontrast,relativesfromthe UK described the process of saying goodbye as a more gradual and less explicitprocess.

167Summary

It isoftenassumedthattheuseofpalliativesedationisburdensomefortherelativesofdyingpatientsbecauseithamperscommunicationintheverylastphaseoflife.Inchapter5,weexaminedwhetherbereavedrelativesofpatientswhoreceivedsedationandbereavedrelativesofpatientswhohaddiedanon-suddendeathwithouttheuseofpalliativesedationdifferintheirexperienceofthedyingphaseandtheirwellbeingafterthepatient’sdeath.Anobservationalstudywasconductedamongrelativesofconsecu-tivepatientswhodiedanexpecteddeathintheErasmusMCCancerInstituteorhos-piceLaurensCadenzainRotterdam,between2010and2013.Nosignificantdifferenceswerefoundinrelatives´assessmentsofthequalityofend-of-lifecare,patients´qualityof life in the last week before death, and their quality of dying, between relatives ofpatientswhoreceivedanddidnotreceivesedation.Further,nosignificantdifferenceswerefoundinrelatives’ownsatisfactionwithlife,theirgeneralhealth,mentalwellbe-inganddetachmentlevels.Theuseofsedationdoesnot, initself,seemtonegativelyinfluencebereavedpersons’experienceofthedyingphaseoftheirdeceasedrelativeortheirwellbeingafterthepatient’sdeath.

Inchapter6weaimtogetinsightintotheaccuracyofestimatesofthelife-shorteningeffectofcontinuoussedationuntildeathbycomparingtwodifferentapproaches.Wecompare a direct approach, where we ask physicians to estimate the life-shorteningeffectofcontinuoussedationuntildeath,andanindirectapproach,whereweaskthephysicians’toestimatethepatient’slifeexpectancyandrelatethattothedurationofthesedation.Thereisasubstantialdiscrepancybetweenthetwoapproaches.Whendirectlyasked,physiciansestimatedthattheuseofcontinuoussedationuntildeathmighthavehadalifeshorteningeffectin51%ofthecases.Incontrast,onthebasisofphysicians’estimations of patients’ life expectancy at the start of sedation and the duration ofsedationuntildeathsuchaneffecthadoccurredin84%ofcases.Thefindingofthissubstantialdiscrepancybetweentwoapproachesconfirmsthedifficultyofpredictingthelifeexpectancyofpatientswithadvanceddisease,andofestimatingthepotentiallife-shorteningeffectofend-of-lifeinterventions.

Althoughthemajorityofadultsrecoverafterthelossofalovedone,aportioncontin-uestogrieveforanextendedperiodoftimeanddevelopssymptomsofastateknownascomplicatedgrief.Theprevalenceofcomplicatedgriefinolderadultsinthegeneralpopulationisconsiderable(namely4,8%),however,researchoncomplicatedgrief inolderadultsisscarce.Inchapter7,weaimedtoassesswhethercharacteristicsofthepatientandthebereavedrelative,thepatient’sillness,end-of-lifecareandthenatureof death are risk factors for developing complicated grief in older adults. A nestedcase-controlstudywasconductedwithintheRotterdamStudy.100couplesofwhichonepersonhaddeceasedandtheotherpersonexperienced‘complicatedgrief ’and100

Chapter 9168

controlcouplesofwhichonepersonhaddeceasedandtheotherpersonexperienced‘normalgrief ’wereselected.Onlypre-existingdepressionwassignificantlyassociatedwithcomplicatedgrief.Bereavedpartnerswithpre-existingdepressionhadahigherrisk of complicated grief compared to bereaved partners without depression. It canbeconcludedthatcomplicatedgriefinolderadultsisnotrelatedtothecircumstancesofdyingofadeceasedrelative.Pre-existingconditionssuchasdepressionseemtobemoreimportantinexplainingtheoccurrenceofcomplicatedgrief.

Inthefinalchapter(8),wesummarizeanddiscussthemainfindingsofthestudies.Weconcludethat,despitethatrelativesexpressseveralconcernsduetotheuseofsedation,they generally evaluate its use positively. Concerns that relatives express are mostlyrelatedtoalackofinformationorcommunication.Apotentiallifeshorteningeffect,whichisanimportantissueintheethicaldebateonpalliativesedation,doesnotseemtoplayanimportantroleintheirevaluationofsedation,atleastnotintheNetherlandsandBelgium.Thechapterconcludeswithimplicationsforclinicalpracticeandrecom-mendationsforpolicyandfutureresearch.

169Samenvatting

samenVattinG

Palliatievesedatieishetopzettelijkverlagenvanhetbewustzijnvaneenpatiëntindelaatstelevensfase.Erkunnentweeverschillendesituatieswordenonderscheiden:con-tinusederentothetmomentvanoverlijdenenkortdurendofintermitterendsederen.De focus inditproefschrift ligtopcontinusederen totaanhetoverlijden.Continuesedatievindtplaatsbijpatiëntendiestervendezijnenondraaglijklijden.

Inhoofdstuk1wordteenoverzichtgegevenvanonderzoeknaardefrequentiewaarmeesedatie wordt toegepast. Richtlijnen voor palliatieve sedatie worden besproken, metspecifieke aandacht voor de Nederlandse richtlijn die in 2009 door de KNMG werduitgebracht.Daarnaastwordenbevindingenuitonderzoeksamengevattenaanzienvande klinische kenmerken van sedatie, zoals betrokkenheid van patiënten, naasten enpalliatievezorgexpertsbijdebesluitvorming.

Zowel in het traject dat leidt tot palliatieve sedatie als ook gedurende de uitvoeringdaarvan,spelendenaastenvandepatiënteenbelangrijkerol.Doelvanditproefschriftisominzichttegevenindeervaringenvannaastenvanpatiëntendieaanheteindevanhunlevenwordengesedeerd.Daarnaastwordthetmogelijke levensbekortendeeffectvanpalliatievesedatieonderzochtenwordtergekekennaarmogelijkerisicofactorenvoorgecompliceerderouwbijouderen.Deonderzoeksvragenzijn:

Onderzoeksvraag 1: Wat zijn de ervaringen van naasten met palliatieve sedatie?Onderzoeksvraag 2: Wat is de invloed van continue sedatie op de beleving van naasten

van de stervensfase en het welzijn van naasten na het overlijden van de patiënt?

Onderzoeksvraag 3: Hoe nauwkeurig kunnen artsen het mogelijk levensbekortende effect van palliatieve sedatie inschatten?

Onderzoeksvraag 4: Wat zijn de risicofactoren voor gecompliceerde rouw bij ouderen?

Richtlijnen voor palliatieve sedatie bevatten aanbevelingen gericht op het welzijnvandenaasten.Naastenmoetenbetrokkenworden inhetbesluitvormingsprocesennaasten kunnen betrokken worden bij de toepassing van sedatie, door bijvoorbeeldhetmonitorenvandepatiëntenhelpenbijdeverzorging.Daarnaastmoetennaastenduidelijkwordengeïnformeerdenwaarnodigondersteunddoorzorgverleners.

In hoofdstuk 2 worden de ervaringen van naasten met palliatieve sedatie zoals bes-chrevenindeempirischewetenschappelijkeliteratuursystematischinkaartgebracht.Weziendatnaastenoverhetalgemeenadequaatwordenbetrokkeninhetbesluitvorm-

Chapter 9170

ingsprocesenadequateinformatieontvangen,allijkterruimtetezijnvoorverbetering.Erisechterweinigbekendoverdebetrokkenheidvannaasteninhetzorgprocesenerisergeenonderzoekgedaannaardesteundienaastenontvangenvanzorgverleners.Hoeweldemeerderheidvandenaastenaangeeftdetoepassingvansedatiepositieftehebbenervaren,laatdezestudieziendatnaastenvoorafgaandaanoftijdensdesedatieookstresskunnenervaren.

Inhoofdstuk3wordenderesultatengepresenteerdvaneenfocusgroepstudieondernaastenvangesedeerdepatiënten. Tussenoktober2010enmaart2011zijndrie fo-cusgroepen gehouden met in totaal 10 naasten en vier individuele interviews. Voordezegesprekkenwerdgebruikgemaaktvaneensemi-gestructureerdevragenlijst.Destartvandesedatiebleekvaakeenopluchtingvoornaastentezijn,omdathet lijdenvandepatiënteindelijkwerdverlicht.Naastenvondenhetprettigactiefbetrokkentewordenbijhetbesluitvormingsprocesendetoepassingvansedatie.Anderepositieveervaringenbetroffendebetrokkenheidvanzorgverleners,deplaatsvanoverlijdendiein overeenstemming was met de wens van de patiënt en de prettige zorgomgevingmet goede faciliteiten. Sommige naasten waren minder tevreden met de informatiediezijhaddenontvangenenmetdecommunicatieinhetalgemeen.Anderenegatieveervaringenbetroffenzorgenomtrenthetwelzijnvandepatiënttijdensdesedatie,metnamewanneerdesedatielangduurde,hetfeitdatverpleegkundigengeenbeslissingenkondennemenoverdemedicatieendeafwezigheidvanartsenindeweekenden.

Inhoofdstuk4presenterenwederesultatenvaneeninterviewstudieuitgevoerdondernaastenvangesedeerdepatiëntenuitNederland,BelgiëenEngeland.Weziendatdemeestenaastenbegrepenwathetconceptsedatieinhoudt.Hoeweldemeestenaastenmeendendatdesedatiehadbijgedragenaaneen‘goededood’,haddenzijookvragenenhaddenerzichsomsonverwachtesituatiesvoorgedaan(zoalshetwakkerwordenvandepatiënttijdensdesedatie)waaropzijnietwarenvoorbereid.Hunvragenbetroffenhetwelzijnvandepatiëntofhuneigenwelzijn,ofhetmogelijklevensbekortendeeffectvansedatie.Sommigenvroegenzichafof sedatieweldemeestgeschikte interventiewasgeweestomhetlijdenvandepatiëntteverlichten.VolgensveelnaastenuitNed-erlandenBelgiëbetekendedestartvandesedatieeen‘afgebakend’moment,waaropzeafscheidkondennemenvandepatiënt.NaastenuitEngelandhaddendieervaringveelminder:hetprocesvanafscheidnemenwasvoorhenmeereengeleidelijkprocesgeweest.

Erwordtvaakveronderstelddatpalliatieve sedatiebelastend isvoordenaastenvanpatiëntenomdathetdecommunicatiebelemmertindeallerlaatstefasevanhetleven.Inhoofdstuk5isnagegaanofnaastenvangesedeerdepatiëntenennaastenvanniet-

171Samenvatting

gesedeerde patiënten verschillen in hun tevredenheid met de stervensfase en hunwelzijn na het overlijden van de patiënt. Tussen 2010 en 2013 werd een vragenlijstafgenomenbij151naastenvangesedeerdepatiëntenen90naastenvanniet-gesedeerdepatiënten.Hetgebruikvansedatieopzichzelfbleekgeennegatieveinvloedtehebbenopdetevredenheidmetdestervensfaseofophetwelzijnvandenaaste.

In hoofdstuk 6 worden verschillende methoden om het mogelijk levensbekortendeeffectvansedatie in te schattenvergeleken.Bijde ‘directe’benaderingwerdeenartsgevraagd het levensbekortende effect van continue sedatie tot aan het overlijden teschatten.Bijde‘indirecte’benaderingwerdeenartsgevraagdomdelevensverwachtingvandepatiënt te schattenenwerddezevergelekenmetdedaadwerkelijkeduurvande sedatie. Er was een aanzienlijke discrepantie tussen de twee benaderingen. Wan-neerartsendirectwerdengevraagdofcontinuesedatieeenlevensbekortendeffecthadgehad, gaven zij in 51% van de gevallen aan dat dit het geval was geweest. In 84%vandegevallenwasdeduurvandeperiodevanafdestartvandesedatietotaanhetoverlijdenvandepatiëntkorterdandedoordeartsingeschattelevensverwachting.Dediscrepantietussendezetweebenaderingenbevestigtdecomplexiteitvanhetinschat-tenvandelevensverwachtingvanpatiëntenenvanhetmogelijklevensbekortendeffectvanmedischeinterventies.

Er wordt gesproken van gecompliceerde rouw als iemand gedurende ten minste 6maanden intense rouwreacties ervaart die gepaard gaan met ernstige problemen inhetnormalealledaagsefunctioneren.Hoeweldeprevalentievangecompliceerderouwonderouderenvrijhoogis,isernogweinigonderzoeknaargedaan.Inhoofdstuk7isonderzochtofkenmerkenvandepatiëntofdiensziekte,kenmerkenvandenaaste,ofkenmerkenvandezorgrondhetlevenseindeenhetoverlijdeninvloedhebbenophetontwikkelenvangecompliceerderouwbijouderen.BinnendeRotterdamstudiezijn100echtparenwaarvandeeenwasoverledenendeander‘normalerouw’ervoervergelekenmet100echtparenwaarvandeeenwasoverledenendeander‘gecompliceerderouw’ervoer. De aanwezigheid van gecompliceerde rouw bleek niet samen te hangen metkenmerkenvanhetoverlijden.Reedsbestaandeconditieszoalsdeaanwezigheidvandepressiespeeldenwelmeebijhetverklarenvanhetbestaanvangecompliceerderouw.

Inhetlaatstehoofdstuk(8)wordendebelangrijkstebevindingensamengevat.Wecon-cluderendat,hoewelnaastensomsvragenhebbenenzorgenervarenbijdetoepassingvan palliatieve sedatie, zij er over het algemeen positief op terugkijken. Hun zorgenhebbenveelaltemakenmeteengebrekaaninformatieofcommunicatie.Eenmogelijklevensbekortend effect van sedatie (volgens artsen vaak het geval) of het gebrek aancommunicatiemetdegesedeerdepatiëntvoorhetoverlijden,tweebelangrijkeonder-

werpeninhetethischedebatrondomsedatie, lijkengeenbelangrijkeroltespeleninhunbeoordelingvansedatie.Detoepassingvansedatieheeftgeennegatieveinvloedopdetevredenheidvannaastenmetdestervensfasevandepatiëntenhunwelzijnnahetoverlijden.Hethoofdstukwordtafgeslotenmeteenbesprekingvandeimplicatiesvoordeklinischepraktijkenaanbevelingenvoorbeleidenonderzoek.

173Dankwoord/Acknowledgements

DankwoorD/acknowleDGements

Allereerstwil ikmijnbegeleidersAgnesvanderHeideenJudithRietjensbedanken.Het is nu vijf jaar geleden dat we zijn gestart met de UNBIASED studie. Vanaf hetmomentvanmijnsollicitatiewistikhetzeker:vanditfantastischeprojectwilikdeeluitmaken!Agnes,ikhebveelbewonderingvoorjekennisvanzakenendoortastendemaniervanwerken.Jijbehieldhetoverzicht,hielpmijknopendoortehakkenenhieldme op het rechte spoor. Ik heb de afgelopen jaren veel van je mogen leren. Judith:ambitieus,enthousiast,eenhardewerkerenaltijdgeïnteresseerdindevoortgangvanmijnonderzoek.Ontzettenddruk,maartochvolopruimtevooreenintensievebege-leiding.Ikhebhetsamenwerkenmetjoualszeerprettigervaren.Zonderjulliewasditproefschrifternietgeweest!

DearmembersoftheUNBIASEDconsortium.IamgratefulthatIhadthechancetobepartofthisinternationalcollaboration.Palliativesedationhasreceivedgrowingpublicand scientific attention in recent years and it was exciting to work on such a topic.Your professional competence, positive attitudes and warm personalities will alwaysberemembered.MyspecialthankstoLiviaAnquinetandKasperRaus,myfellowPhDcandidates.Itwassoniceworkingwithyou.Goodluckwiththerestofyourcareers!

Inhetbijzonderdankikallenabestaandendiemeehebbengewerktaanhetonderzoek.Ik vond het een eer te mogen delen in een van de meest intieme, emotioneel zwaremomenteninjullieleven.Individueleverhalenengroepsgesprekkendieiknooitmeerzalvergetenendieeenbelangrijkeplaatsinmijnproefschrifthebbengekregen.

JetvanEsch,AstridOosten,HelmavanDijk,MoniqueBooms,JannekeKoningswoudenveleanderen.Zonderrespondentengeenonderzoek,enzonderjulliegeenrespon-denten.Dankzijjulliemedewerkingishetprojectgeslaagd.Bedankt!

Dehoofdstukken inditproefschrift,waarvandemeeste inmiddelszijngepubliceerdininternationalepeer-reviewedjournals,zoudennietvandezekwaliteitzijngeweestzonderdeinbrengvandeco-auteurs.Hartelijkdankvoorjulliekritischeenwaarde-vollefeedbackopmijnmanuscripten.Iwouldalsoliketothankmyforeignco-authorsfortheircriticalandvaluablefeedbackonmymanuscripts.

Graagwilikdeledenvandeleescommissie,prof.dr.C.C.D.vanderRijt,prof.dr.I.D.deBeaufortenprof.dr.B.D.Onwuteaka-Philipsen,bedankenvoordetijddiezehebbengenomenvoorhetlezenenbeoordelenvanmijnproefschrift.

Chapter 9174

Mijnsectiegenoten‘Besluitvormingenzorgrondhetlevenseinde’:Agnes,Judith,Erica,Natasja,Ineke,Arianne,Siebe,Eric,AnoukenLea.EenhechtclubjebinnendeafdelingMaatschappelijkeGezondheidszorgwaarbinnenikmealtijdheelprettighebgevoeld.Stukvoorstukenthousiaste,gedrevenenlievecollega’s.Jammerdatonzesamenwerk-ing‘maar’vierjaarheeftgeduurd!

InekeLokker,mijnkamergenootje.TegelijkertijdbegonnenopMGZenaltijdsameneenkantoorgedeeld.Ikvondhetheelfijnomeen‘palliatievecollega’dichtindebuurttehebben.Altijdkonikbijjeterechtvoorvragenen/ofwerkgerelateerdeproblemen.Maarbovenalvondikhetheelfijnomzo’ndierbarecollegaalskamergenoottehebben.Ikbentrotsdatjemebijstaatalsparanimf!OokjouwtijdbijMGZziternuop.Geluk-kigwerkenwenogwelsteedsbijelkaarindebuurt.Verplichtekoffie/lunchuurtjes!

DavidBlok,deenigemaninae-124.Wathebjijeenhoopgeleerdovervrouwenindeafgelopen vier jaar: gesprekken over onderjurkjes, nagellak, haarkleur etc. Al was jehier inhetbeginwellicht ietsdoorgeïntimideerd,vanafnukan iederedamebij jouadviesinwinnenoverdemeestuiteenlopendezaken.InhetbijzonderbedanktvooraljehulpbijSPSS/computervraagstukken.Hoemoetikhetvanafnuzonderjouzientereddenopdepc?Jewaseenprettigekamergenootenikweetmijnwegnaarde22stenogaltijdtevindenhoor!

LieveRobine.Samenopdemiddelbareschoolensamenstuderen inGroningen.Entotonzegroteverbazingookopdezelfdeplekaanonzecarrièrebegonnen.JijwerktealeenjaartjeopMGZtoenikwerdaangenomen.Watheerlijkomiederedageenbekendengeliefdgezichttekunnenzien.Zullenweindetoekomstnogeenkeerzoveelgelukhebben?Nuvooralgenietenvanje,recentelijkuitgebreide,gezinenhopelijkvindjeindetoekomstdebaanvanjedromen.

Patrick:verantwoordelijkvoordevormgevingvanmijnproefschrift.Ontzettendbe-dankt,hetisprachtiggeworden!Omjetebedankenzullennogveleetentjesvolgen.

Mijnnieuwecollega’svandeafdelingUrologie.FantastischdatiknaafloopvanmijnPhDdirectalspostdoconderzoekerbijjullieaandeslagkongaan.Wederomeenboei-endprojectwaarikmetveelplezieraanbenbegonnen.Bedanktvoorhetvertrouwen.Opeenmooiesamenwerking!

Dan natuurlijk mijn vriendinnen Yasmijn, Robine, Agnes, Ellen, Bach, Vera, Linda,Sanne, Hanneke, Marieke, Evelien en Ilse. Niet direct betrokken geweest bij mijnproefschrift,maarwelgefungeerdalsklankborddeafgelopenjaren.Bedanktvooralle

175Dankwoord/Acknowledgements

werkgerelateerde,maarvooralalleniet-werkgerelateerdesteun.Onderhetgenotvanheerlijkewijntjeskonikaltijdmijnhartbijjullieluchten.

LieveFloor,mijn‘grote’zus.Metwetenschappelijkonderzoekhebjenietveel,maarikweetdatjeenormblijvoormebent!

Eerstmijnmoederdejuristeachterna,daarnamijnvaderdecriminoloog/onderzoeker.Toeval?Bedanktpap,metjouwexpertiseenervaringhebjemevaakkunnenhelpen.Je weet hoe het voelt: deze titel nemen ze je nooit meer af !! Mam, twee zielen ééngedachte.Jouwmentalsupportwasenisonmisbaar.

TotslotVincent,teallentijdeeenluisterendoor.Dankjewellieverd.

177About the author

aBoUt the aUthor

SophieMarrigjeBruinsmawasborninEnschedeonSeptember7,1984.SheobtainedherGymnasiumdiplomaattheJacobusCollegeinEnschedein2002.Aftersecondaryschool,shestartedstudyingDutchLaw,attheUniversityofGroningenandobtainedherpropaedeuticdegree in2005. In2005, sheswitched tostudyingSociologyat theUniversityofGroningen.Shespecializedin‘Health,careandwellbeing’.Herthesiswasaboutthewellbeingofresidentsinanursinghome.SheobtainedherMastersofSciencedegreein2010.In2010,shestartedasaPhDcandidateattheDepartmentofPublicHealthofErasmusMC,theNetherlands.Shewasinvolvedinaninternationalstudy(the UNBIASED study) that explores the perspectives of clinical staff and bereavedinformalcaregiversontheuseofcontinuoussedationuntildeathforcancerpatients.In the same period, she also completed her master’s degree in Epidemiology at theNetherlandsInstituteforHealthSciences inRotterdam.InFebruary2014,shewasavisiting researcher at the Brocher Foundation in Geneva, Switzerland. In that sameyear,shestartedworkingasapostdoctoralresearcherattheDepartmentofUrologyofErasmusMC,theNetherlands,wheresheisinvolvedinaninternationalstudy(GAP3)onactivesurveillanceforlow-riskprostatecancer.

Sophie Marrigje Bruinsma werd op 7 september 1984 geboren. Ze behaalde haarGymnasiumdiplomaaanhet JacobusCollege teEnschede in2002.Nahaarmiddel-bare school is zebegonnenmetde studieNederlandsRechtaandeRijksuniversiteitGroningen.NahetbehalenvanhaarpropedeuseiszeSociologiegaanstuderen,metalsspecialisatieMedischeSociologie.Haarafstudeeronderzoekrichttezichophetwelbev-indenvanbewonersinverzorgingstehuizen.In2010behaaldezedetitel‘MasterofSci-ence’.Vanaf2010werktezealsPhDopdeafdelingMaatschappelijkeGezondheidszorgvan het Erasmus MC te Rotterdam. Ze was betrokken bij een internationale studieover de ervaringen van artsen, verpleegkundigen en naasten met palliatieve sedatie.TijdenshaarPhDperiodebehaaldezijookhaarmastertitelinEpidemiologieaanhetNIHESteRotterdam.Infebruari2014werktezijtijdelijkalsonderzoekerbijdeBrocherFoundationinGeneve,Zwitserland.IndatzelfdejaariszijbegonnenalspostdoctoraalonderzoekeropdeafdelingUrologievanhetErasmusMCeniszijbetrokkenbijeeninternationalestudie(GAP3)over‘activesurveillance’bijprostaatkanker.

Chapter 9178

179List of publications

list of PUBlications

international

BruinsmaSM,TiemeierHW,Verkroost-vanHeemstJ,vanderHeideA,RietjensJAC.Riskfactorsforcomplicatedgriefinolderadults.JPalliatMed2015(inpress).

BruinsmaSM,BrownJ,vanderHeideA,DeliensL,AnquinetL,PayneSA,SeymourJE,RietjensJA,onbehalfofUNBIASED.Makingsenseofcontinuoussedationinendof life care for cancer patients: An interview study with bereaved relatives in threeEuropeancountries.SupportCareCancer2014;22(12):3243-52.

BruinsmaSM,RietjensJA,SwartSJ,PerezRS,vanDeldenJJ,vanderHeideA.Estimat-ingthepotentiallife-shorteningeffectofcontinuoussedationuntildeath:acomparisonbetweentwoapproaches.JMedEthics2014;40(7):458-62.

BruinsmaS,RietjensJ,vanderHeideA.Palliativesedation:afocusgroupstudyontheexperiencesofrelatives.JPalliatMed2013;6(4):349-55.

BruinsmaSM,RietjensJA,SeymourJE,AnquinetL,vanderHeideA.Theexperiencesofrelativeswiththepracticeofpalliativesedation:asystematicreview.JPainSymptomManage2012;44(3):431-45.

SeymourJ,RietjensJ,BruinsmaS,DeliensL,SterckxS,MortierF,BrownJ,onbehalfofUNBIASED.Continuoussedationinendoflifecareforcancerpatients:Aqualitativeinterviewstudyofphysicians’andnurses’practiceinthreeEuropeancountries.PalliatMed2015;29(1):48-59.

Anquinet L, Rietjens JAC, van der Heide A, Bruinsma S, Janssens R, Deliens L,Addington-Hall J, Smithson WH, Seymour J, on behalf of UNBIASED. Physicians’experiencesandperspectivesregardingtheuseofcontinuoussedationuntildeathforcancerpatientsinthecontextofpsychologicalandexistentialsufferingattheendoflife.Psycho-Oncology2014;23(5);539-46.

RausK,BrownJ,SealeC,RietjensJ,JanssensR,BruinsmaS,MortierF,PayneS,SterckxS.Continuoussedationuntildeath:Theeverydaymoralreasoningofphysicians,nursesand family caregivers in the UK, The Netherlands and Belgium. BMC Med Ethics2014;15:14.

Chapter 9180

SealeC,RausK,RietjensJ,BruinsmaS,vanderHeideA,SterckxS,MortierF,PayneS, 1. Mathers N, on behalf of UNBIASED. The language of sedation in end-of-lifecare: The ethical reasoning of care providers in three countries. Health 2014. pii1363459314555377(epubaheadofprint).

BruinsmaSM,vanderHeideA,vanderLeeML,VergouweY,RietjensJAC.Nonega-tiveimpactofpalliativesedationonrelatives’experienceofthedyingphaseandtheirwellbeingafterthepatient’sdeath(submitted).

national

BruinsmaS,RietjensJ,JanssensR,VanZuylenL,VanderHeideA.Eeninternationalestudienaardepraktijkvanpalliatievesedatie.NederlandsVlaamsTijdschriftvanPal-liatieveZorg2011;11(1):22-26.

Book contribution

BruinsmaSM,RietjensJAC,vanderHeideA(2013).Continuoussedationuntildeath:state of the art. In: Sterckx S, Raus K, Mortier F (Eds). Continuous Sedation at theEndofLife:Ethical,ClinicalandLegalPerspectives(pp.29-46).CambridgeUniversityPress.

PhD Portfolio- SUMMARy oF PHD tRAININg

Name SophieBruinsmaPhDperiod April2010-March2014ErasmusMCdepartment PublicHealthPromotor Prof.Dr.A.vanderHeideSupervisor Dr.J.A.C.Rietjens

Year workload (ects/ hours)

1. PhD training

General academic skills

Computer course/ literature search, Erasmus MC 2010 6 hours

Qualitative research in healthcare, Antwerp, Belgium 2010 40 hours

research skills

Master Epidemiology, NIHES 2010-2013 70 ECTS

in depth courses

suffering, Death and Palliative Care, Nijmegen, the Netherlands 2013 1 ECTS

Presentations- international

Oral presentation: The experiences of relatives with palliative sedation. Bioethics conference. Rotterdam, the Netherlands

2012 1 ECTS

Oral presentation: Estimating the potential life-shortening effect of continuous sedation until death: a comparison between two approaches. International Collaborative for End-of-Life Care Research (ICER). Amsterdam, the Netherlands.

2012 1 ECTS

Poster-presentation: Palliative sedation: a focus group study on the experiences of relatives. Conference of the European Association of Palliative Care (EAPC). Trondheim, Norway

2012 1 ECTS

Poster-presentation: The experiences with the practice of palliative sedation: a systematic review. Conference of the European Association of Palliative Care (EAPC). Lisbon, Portugal.

2011 1 ECTS

Presentations- national

Experiences of relatives with the practice of palliative sedation: the UNBIASED study, Erasmus MC, Rotterdam, the Netherlands

2013 1 ECTS

Oral presentation: Estimating the potential life-shortening effect of continuous sedation until death: a comparison between two approaches. Nationaal Congres Palliatieve Zorg. Lunteren, the Netherlands.

2012 1 ECTS

Oral presentation: Care and decision- making at the end of life. Post- EAPC congress. Leiden, the Netherlands.

2012 1 ECTS

Oral presentation: Experiences of relatives with palliative sedation. Flemish Dutch Research Forum Palliative Care. Rotterdam, the Netherlands.

2012 1 ECTS

Oral presentation: Relatives in end-of- life care research. Agora. Nijmegen, the Netherlands.

2011 1 ECTS

(inter) national conferences

Nationaal Congres Palliative Zorg, Lunteren, the Netherlands 2010 1 ECTS

European Association of Palliative Care conference, Lissabon, Portugal 2011 1 ECTS

Conference: Continuous sedation at the end of life: Ethical perspectives, Ghent, Belgium

2011 1 ECTS

Nationaal Congres Palliative Zorg, Lunteren, the Netherlands 2012 1 ECTS

European Association of Palliative Care conference, Trondheim, Norway 2012 1 ECTS

World Congress of Bioethics 2012 1 ECTS

workshops and seminars

Seminars department of Public Health, Erasmus MC 2010-2014 3 ECTS

Flemish Dutch Research Forum Palliative Care, Rotterdam, the Netherlands 2012 1 ECTS

PhD dag (2x) 2011, 2013 8 hours

Symposium ‘Evidentie en beslissen in de gezondheidszorg: Stand van de wetenschap en praktijk’

2010 4 hours

Post EAPC symposium, De Bilt, the Netherlands 2011, 2013 4 hours

Symposium ‘Als botsen raken wordt’ 2010 4 hours

International Collaborative for End-of-Life Care Research (ICER) (2x) 2011, 2013 8 hours

2. teaching activities

Supervising medical students theme 3.C.4 (community project) 2011, 2013 32 hours

Supervising nurses literature review, Sophia, Erasmus MC 2011 20 hours

total 90 ECTS, 126 hours

the experiences of bereaved relatives with palliative ... · the experiences of bereaved relatives...

Documents