The Development of a Valid and Reliable Scale for Rating Anxiety in Dementia (RAID) K

Home , Panic attack

This article was downloaded by: [University College London] On: 23 January 2015, At: 01:59 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Aging & Mental Health Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/camh20 The development of a valid and reliable scale for rating anxiety in dementia (RAID) K. K. SHANKAR , M. WALKER , D. FROST & M. W. ORRELL Published online: 09 Jun 2010.

To cite this article: K. K. SHANKAR , M. WALKER , D. FROST & M. W. ORRELL (1999) The development of a valid and reliable scale for rating anxiety in dementia (RAID), Aging & Mental Health, 3:1, 39-49, DOI: 10.1080/13607869956424 To link to this article: http://dx.doi.org/10.1080/13607869956424

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions Aging &Mental Health 1999;3(1):39±49

ORIGINALARTICLE

Thedevelopmentofavalidandreliablescalef orratinganxiety in dementia (RAID)

K.K.SHANKAR, 1 M . WALKER,1 D. FROST2 &M.W. ORRELL 2

1MentalHealth Unit, PrincessA lexandra Hospital,Harlow & 2University College London MedicalSchool, Department ofPsychiatryandBehaviouralSciences,London,UK

Abstract Aratingscaleto m easurean xietyin dem entiasu fferers wasdevelop edan devaluatedin a sampleof 51 inpatientsan d32 day-hospitalpatient s.A nxietyscores were notrelated to sex, age, accom modationorDSM -IVdiagnosisof thetype of dementia.H owever, bothsubject swithphysica lillnessesan dsubjectswith in sightin totheir m emoryproblem shad signi®can tlyhigher anxietyscores. Thekappavalues fo rinter-raterreliabilityranged from 0.51 to1 andfor test-retestreli- abilityfrom 0.53 to1,which indicatesm oderateto goodreliabil ity.The overallagreem entonind ividualitemsrangedfrom 82± 100% (inter-rater)an d84± 100% (test-retest).The professionalsworkin ginthe care oftheelderly an dcarer groupsfelt thatthe scale was co mprehensiveand all the item sinthe scale were important,therebycon ®rmingthat it hasgood conten t validity.The scalesigni® can tlycorrela tedwith other anxiety scales an dalsowith ind ependentratings both by a consultant psychiatristand also nu rsingstaff, ind icatinggo odconcurrentvalidity.A nxietyscores were signi®cantly higher indem entia patientswhoful® lled m odi®ed DSM -IVcriteriafor anxietyan dclinicaldiagnosisof anxietyd isorder.This showed eviden ce ofgoodcriterio nvalidity.Factor analys isshowed ® vefactors ,includingall item softhescale. S coresof 11andabove o nthe scalein dicatedsign i®cant clin icalan xiety.Overall, the scale had go odreliabilityan dvalidity.It shouldbeausefulclinicaland researchinstrumentforassessingan xietyin dem entiasufferer s.

Introduction DSM-III-Rdementia,o fwhom2 2%had subjective anxiety,1 1%autonomican xiety,3 8%tension,13% Anxietysym ptomsare commonind ementia(A bsher situationalanxietyand 1.8 %panicattacks. Thirty-two &Cummings, 1994),withth eprevalencevar ying (29.4%)had oneo rmorean xietysym ptoms.They from 12to5 0%.Many studieshave focusedon the foundthreemain categoriesof anxietysym ptoms: importantaspectsofdepressionandpsychotic anxietyrelatedtodepression,anxietyrelatedto symptomsindementiasu fferers (Burns,1991;Ballard psychosis andanxietyrelate dtoin terpersonal situa- &Oyebode,1995),butveryfew studieshave focused tions. onanxietysymptoms.However,anxietysubstantially Orrellan dBebbington(1996)foundthatan xiety

Downloaded by [University College London] at 01:59 23 January 2015 reduces thequality of life of those sufferingfrom ind ementiap atients was associatedw ithve ryhigh dementiaan dhas also beenfoundtobe assoc iated levels ofsocial contact,problemsinthep atient±c arer within creasedmortality(O rrell, 1994). relationship andhighphysicaldependency. Independ- Wands et al.(1990)compared 50subjectsw ith entseverethreat life eventswerealso associatedw ith dementiaw ith1 34controlsu bjects.T heyusedthe anxietyindementiap atients,butthis was confounded HospitalAnxietyan dDepressionS caleand fou nd bytherelatio nship betweendepressionandlife events. that16%ofthed ementiag rouphad de®nitiean xiety Theirresu lts suggestedth atanumberofsocial factors andafurther22%possiblean xiety. Therew as no couldle ad toan xietyin d ementiapatien ts. Earlier correlationbe tweenseverity ofdementiaan danxiety studiesusedscales whichwereno tdevelopedfor the scores,althoughthis may bebecause theirg rouphad use intho se sufferingfro mdementia; Konders et al. predominantlym ildc ognitiveim pairment.U singa (1993)usedthe State- TraitA nxietyinven toryand questionnaire,Ballard et al.(1996)lookedinto an xiety Wands et al.(1990)usedtheHospitalA nxietyand symptomsof158consecutivepatien ts attendinga DepressionScaleto m easurean xiety.T hese earlier memoryclinic.One-hundred-and-ninepatientshad scalesmay beinsensitiveto c hanges inco gnitive

Correspondenceto :Dr M.W.Orrell, Readerin Psychia tryoftheA geing,UniversityCo llegeL ondonM edicalS chool, Departmentof PsychiatryandBehavio uralS ciences,WolfsonBuilding,48 RidingHouseStreet, Lo ndonW1N 8AA,UK. E-mail:m [email protected] Receivedfo rpublication7thM arch 1998. Accepted22n dOctober1998.

1360-7863/99/010039-11 $9.00 ½ CarfaxPublish ingLim ited 40 K.K.Shankar et al.

impairments andmay notbeap propriatefor the Worry. ItemsonWorrywerem ainly takenfrom severely cognitivelyim pairedpopulation(P lutchick existingliterature.H ypochondriasis has beenid enti- et al.,1970).Further, thep resentationof symptoms ®edasafeatureofanxietyin the eld erly (Bergmann, may besituationspe ci®c andmay show ¯uctuations 1978).Lader(1982)suggestedhypochondriacal inthe sam eday, soacross-sectionalassessmentmay anxietyco uldbe c lassi® edas aseparaten osological notsh owa truepicture.Im pairments inco ncentra- entity.In theirstud yofphysical health andpsychiatric tion, memory, judgementandlackofinsightin to disorderinthe urban elderly community(G uy’s/Age their illness affecttherespo nses ofdementiapatien ts Concernsurvey), Lindesay (1990)foundth atthe tothequestionnairesandratingscales. Gottlieb et al. highest rateo fcontinuousworrywasassociatedwith (1988)studiedthe re liabilityof psychiatricscale sin generalizedan xiety.Wo rryingaboutfailingmemory patientswithde mentiao fAlzheimertype.They found has also beenreco gnizedindementiasuffe rers (Forsell goodcorrelationbetweenself ratingandthato fa et al.,1993).Yesavage andTaylor(1991)statedthat rater-administeredd epressionscalein patientsw hose theconceptof`worry’ ormentalan xietyin thee lderly Alzheimer’sdisease was of lowseve rity butnoton mustin cluderuminationsaboutcognitivepe rform- thoseo fhighseverity. Reisberg et al.(1987)developed ance.Consideringthepsychiatrics ymptomsin BEHAVE-ADtom easureth ebehavioural symptoms dementiarepo rtedby physiciansandcarers, Forsell inpatien ts sufferingfrom Alzheimer’sdisease.Ithas et al.(1993)identi®e dworryingover tri¯es acom- sevensec tionsanda totalo f25items. Onesection ponentinthe an xietyclu sterofsymptoms. Because was devotedto an xietyan dphobias andhas four theyco nstantlyse ekthe atte ntionof thec aregiver items: (1)anxietyre gardingupcominge vents, (2) over trivial matters,thisanxietyis readily observable. otheran xieties,(3 )fear ofbeingleft aloneand(4) otherph obias. This scaleis notspec i®c for anxiety Apprehension andvigilance. Sleepdisturbances, andn otsufficienttocover thewideran geofpresenta- includedundernon-speci®c symptomsinIC D-10, tionof anxietyinthis population. have beenfoundtoco rrelatew ithanx ietyin the Similarproblemsweree ncounteredinr ating elderly. Peoplew hohave sleepdisturbanceand depressive symptomsind ementiau singquestion- presumably greaterau tonomicaro usal tendtobe naires orpatientinterviews.Fordepression,however, morea nxious,suggestingthatsy mpathetict one scalessuchas theC ornellS cale(A lexopolous et al., heightensinthe e veninghours(Davis et al., 1982, 1988)have beendevelopedwhich useacombination Wagner&Lorion,1984).Othersymptomsofanxiety of clinicalinformationfro mpatientinterviewand inth eelderly includenervoustension,apprehension, otherclin icalin formation.Thisenables aglobal rating irritabilityandpetulantoutbursts (Lader, 1982). of depressive symptomstobe m ade.TheC ornell Scaleh asbeendemonstratedtobevalid,reliablean d Motor tension. Inthe ir reviewof agitatedbehaviour useful inclin icalprac tice(Patterson et al., 1990). inth eelderly, Cohen-Mans® eldan dBilling(1986) Theaim of this studywastod evelopaglobal rating statethat the co nceptofagitatedbehaviouris linked scaleto measurean xietyin de mentiapatien ts. toa varietyof concepts by researchers inthis area. Theirw ork onsu chbehaviours ina nursinghome- Method basedstudyfailedtore veal an`anxiety’facto rlinked Construction ofthescale toa gitation.However, theseconceptsareinter- related(Yesavage&Taylor, 1991).Goudemand et al.

Downloaded by [University College London] at 01:59 23 January 2015 Theitemsofthescalew ered erivedfrom theco ncepts (1994)statean xiety®ndsmoreexpression withm otor of anxietypresen tedin the IC D-10(WorldH ealth agitationthan withspe ech. Organization1 992),DSM-III-R,DSM-IV(American PsychiatricA ssociation,1987;1994),PresentState Autonomichypera ctivity. Symptomsduetoau tono- Examination(PSE:Wing et al.,1974),Geriatric michype ractivityare corec omponentsof anxiety.In MentalState (C opeland et al.,1976),Generalized clinicalpracticeit isrecognizedthat the se symptoms AnxietySc ale(L indesay et al.,1989)andthe literatu re are timeandagainrepo rtedby dementiasuffe rersto onthe p resentationof anxietyin the e lderly andin theirc arers.Thesesym ptomsare groupedto invo lve dementiapatie nts. themajorsystems: cardiovascular (palpitations), Theite msinthe sc alewe rerate daccordingtothe respiratory(shortness ofbreath),centralnervous person’ssymptomsandsignsof anxietyove rthe system (dizziness,lightheaded ness)and others (sweat- previoustwoweek s.This periodwasadequatee nough ing,¯ushesandchills, tinglingandnumbness of for theratin gstob eaffectedby day-to-day ¯uctua- ®ngers). Care was takento restri ctthenumberof tions andtop ickupimportantbehaviours. Onthe itemsinthis sub-groupino rdertoavoid bias ofthe otherh and,itw as sufficientlysho rtforthecarers scaletoward sthis componentofanxiety. generally tobe ab leto rem ember. Eachitemw as ratedac cordingtofourdifferentgrades:Absent,Mild Phobiasandpa nic attacks. Inthe E pidemiological orintermittent,Moderate,an dSevere.Theitemswere CatchmentAreastudy, Reiger et al.(1988)reported dividedintosixsu b-groups. phobias tobecommonfor peopleof all ages, including Anxiety in dementia 41

theelderly. Phobias werethe se condm ost frequent CornellS calefor Depressionin D ementia(Alex o- psychiatricdiagn osis, nextto c ognitiveim pairment polous et al.,1988)was also administered.Further,the forbothm enandw omen,65years of ageor older. ClinicalD ementiaRating scale (H ughes et al., 1982) Panicdisorderwastheleast co mmonanxietyd isorder andtheM ini-MentalS tateE xamination(MMSE: inthis agegroup. However, 11%ofthesampleo n Folstein et al.,1975)were adm inisteredtoassess how lateo nsetagoraphobia had ahistoryof panicattac ks thesc alefu nctionedacross therang eofdementia inLindesay’sstudy(1991). severity.

Administration of thescale Reliability methods

Thescalew asscoredbased onall availableso urcesof Inter-rater reliability. This was testedby tworaters information.First,theclinicianinterviewedthe on33patients. Tworater swerep resentduringthe patient’scarer(u sually aquali® ednurse orcloserela- sameinterview, whichw as conductedbyoneofthe tive)an daskedabouttheitem sinthesc ale.Thecarer raters.Theothe rraterwas allowedto ask questions wasinstructedtobase theirre portontheo bservation for clari®catio nregardingthepatients’ symptoms. ofthepatien t’sbehaviourduringtwow eeks prior to Followingthe inter vieww iththe care randthe patien t, theinterview. Explanationswereg ivento thecarer in therater sscoredthesc aleind ependentlyw ithoutany ordertou nderstandthem eaningofeach item.This furtherco nsultationam ongthem. Test-retestreliability wasfollowedbyinterviewingthepatient.Anyfurther was testedbyoneraterre peatingthein tervieww ith information,includingthepatien t’smedicaln otes, 25patients withino neweekto te ndays ofthe®rst werealso examined.Symptomsthatw erelikely to interview. Internalconsistency ofthescalew as tested, arised uetophysic al illness or medicationweren ot includingall theite msofthesc alee xceptph obias scored.Afterthis process, thesc alew asscoredbase d andpanicattack s.Theinte rnal consistencywas also onthec linician’s®naljudgement.Alltheite msinthe testedfo rthesu b-groups. scalew ered erivedfromthecurrentconceptsof anxiety, andlittlead ditional trainingwas neededto administerth escale. Validitym ethods

Content validity. This was assessedby sendingth e Subjects scalefo rcomments toc onsultants ino ldag epsychiatry, seniorregistrarsino ldag epsychiatryand Eighty-threepatie ntsw ho quali® edforthed iagnosis experiencedprofessionals workingwitheld erly inth e ofdementiabasedontheDSM-IV(American ®eldso f: socialwo rk; nursing;clinical psychology; PsychiatricA ssociation,1994)werein cludedin the occupationaltherapy. Theopinionsofcareranduser study. Inord erto g eta representativesam pleof the groupsincludingtheAlzheimer’sDisease Society, the elderly dementiapo pulation,patientswererec ruited Councilo fRelativesto Assist intheC are ofDementia from acutein patient,d ay hospitalandday centre (CRACDementia),D ementiaR eliefTr ustan dthe patientsandpatientsinthe lon g-stay continuingcare individual carers of thepatients werealso sought. wards. Subjects who had acutem edical illness and Thesepe oplew ereg ivenacopy of theinformation weretoo ill to sit throughthe in terviewwe ren ot sheetaboutthesc ale,theRAIDscale(see A ppendix Downloaded by [University College London] at 01:59 23 January 2015 included.Subjectsw ithc hronicm edical conditions 1)andaquestionnaire tocomplete.Theinformation likelon g-standingdiabetesm ellitus,hypertension,etc. sheetprovidedinformationonreasonsfordeveloping werein cluded.Subjects’ insightintotheir illness was thescale, howth eitemsinth escalew erese lected assessed by askingthequestion,`Do youhave any andthew ay itw as administeredan dscored.The problemswithyo urmemory?’ Those who didn ot questionnairecon sistedof®vequ estions:(1)Arethere acknowledgethe ir memoryproblemswereno tedas anyadditionaltopicswhichyoufeelshouldbe lackingininsig ht. includedin the sc ale?(2 )Doany of theto picsneed moreex planation?(3)Doyou foreseeany speci®c difficultiesin usin gthesc ale?(4)Doyouthinkall Instruments topicsare important?(5)Doyouhave any additional comments? Threeother stand ardizedinstrum entswere ad minis- teredalong w iththe RAID scalefor thepurpose of Concurrent validity. Thepe rformanceof RAIDwas validation.TheC linicalAnx ietyScale (Sn aith et al., compared withthe an xietyscale sCASandASI.The 1982)andA nxietyStatus Inven tory (Zung,1971)are performanceofRAIDwas also comparedwithth e observer-rated anxietyscales. They wereadm inistered CornellSc alefo rDepressioninD ementia. Futher tocompare theperform anceoftheRAID scalein rela- testso fvalidationw erecar riedoutbycomparingth e tionto them .Sinceit was expectedthat an overlap in RAID’sscorew iththe fo llowingtwo m easures: (1) thepresent ationsof anxietyand depressi onexists, the Thecarer’sratingofanxiety.This involvedthe carer s 42 K.K.Shankar et al.

ratingtheleve lofanxietyonthevisu alanaloguescale mentwascalculatedb ythepe rcentageofagreement, below. wherethe rate rsagreedonascoreof zeroo raposi- tivesco re(sc oreo f1,2or 3).Cronbach’salphawas Notatall anxious Extremelyan xious calculatedtoassess internal consistency. Nonpara- 0 100 metrican alyseswerese lectedbecause ratingscales (2)Anxietyratin gby aconsultantpsychiatrist:this yieldedordinaldata (Siegel,1956).Spearmancorrela- involvedthec onsultantinold ag epsychiatry(MO) tionco efficients werecalc ulatedbetweenR AIDand independentlyratin gthepatien t’sanxietyu singthe thecarerrating,consultant’sratingan dtheother samevisualanaloguescale. scales.TheMann-WhitneyUtestw as usedas atest of signi®c ancewherethe rew eretw ogroups and Criterion validity. Therew as no`goldstan dard’for Kruska-Wallis Anova wascalculatedw hentherewe re diagnosinganxietyind ementiasu fferers.Thew idely moreth an twogroups. Thestatist ical analyseswere usedclassi® catorysystems, ICD-10andD SM-IV, carriedo utusingtheSPSSsoftware package(Version didn otallow fordiagnosinganxietyd isorderinth e 6.1.3). presenceof an organicc ondition.This issuewas addressedinthe follow ingways: (1)thec onsultant Results psychiatrist was askedtoco mpletea questionnaire basedonhis clinicalassessmenttoanswerthe Performanceof RAID followingtwo qu estions:Isanxietya signi®cant clin ical featureo fthis patient?Y es/No. Wouldit affe ctth e Thescalewas user-friendly andno signi®can tdifficul- managementofthis patient?Y es/No;and(2)the tiesarose inad ministeringit.T heto taltim efor consultantwas also askedw hetherth epatientsatis- administrationofRAIDwasapproximately20 ® ed the modi® ed DSM-IVcriteria forGeneralized minutes(approximately tenm inuteswiththe carer AnxietyD isorder.T his was basedonthe D SM-IV andte nminutesinterviewwiththe patie nt).It w as criteria for generalizedan xietyd isorder, whereth e anticipatedth atin u sual clinicalpracti ce,staff who restrictionc riteria of anxietyan dworryduetoo ther are familiar withthe patien twouldbeable tocomplete axis Idisorder(criterionD)andduetod irecteffect thescalew ithin® veto te nminutes. Amongst the8 3 ofasubstanceorageneralmedicalcondition patients onw homRAIDwas completed,themean (criterionF)weren otapplied.This was doneto totalsc orew as9.3(S D=7.1;range0to39 ).Figure diagnosean xietybased onthe `co ncept’ ratherthan 1givesthed istributionofRAIDscores. Table 1gives the`cr iteria’ aspresentedinD SM-IV. thefrequencyofindividual itemscores. Theite m `restlessness’ inth escalesc oredmost frequently Construct validity. Aprinciplec omponentan alysis (71.1%).Theitemsinth esub-scale`a utonomic was performedto ex plorethe facto rstructurean d hyperactivity’,andthoseof phobias andpanicattack s constructvalid ity. The18 item sof theR AIDscale tendedtoscoreless frequently. werein cluded.Eigenvalues andtheperce ntageo f variancesexplainedbyeachofthefactorswere Clinical pro® le determined. Themean ageinthe p opulationstudiedwas 79.1 years(SD=7,range62to9 7)an dthem ajority Statisticalanalyses (62%)werew omen. Fifty-one(6 1.4%)werein - Downloaded by [University College London] at 01:59 23 January 2015 patientsand3 2(38.6%)weredayhospital/daycentre Thein ter-raterreliabilityan dtest-retest reliability patients. Informationonphysic al health was avail- werec alculatedu singthekappa statistics(C ohen, able for77p eople.Physical illness included:Parkin- 1960)andoverall agreement(O AG).Overall agree- son’sdisease,ischaemiche artdisease, hypertension,

0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 Total score

FIG.1.Histogramo fRAIDtotalscore ( n =83,mean= 9.3, SD=7.1). Anxiety in dementia 43

TABLE 1. Frequencyofindividualitem scores

Item % scoring % scoring % scoring % scoring 1 2 3 1 or more

1. Worryaboutphysica lhealth 15.7 8.4 1.2 25.3 2. Worryaboutcognit iveperfo rmance 18.1 16.9 2.4 37.3 3. Worryover® nances,fam ilyproblem s 18.1 22.9 7.2 48.2 4. Worryassociatedwith false belief and/orperception 8.4 12.0 1.2 21.6 5. Worryovertri¯ es 12.0 9.6 3.6 25.3 6. Frightenedand anxiou s 22.9 24.1 9.6 56.6 7. Sensitivityto noise 18.1 15.7 2.4 36.1 8. Sleepd isturbance 10.8 19.3 2.4 32.5 9. Irritability 47.0 15.7 2.4 65.1 10. Trembling 14.5 9.6 0.0 24.1 11. Motortensio n 24.1 3.6 2.4 30.1 12. Restlessness 30.1 31.3 9.6 71.1 13. Fatigueability 28.9 13.3 0.0 42.2 14. Palpitations 10.8 0.0 0.0 10.8 15. Drymouth,sin kingfeeling in the stom ach 8.4 2.4 1.2 12.0 16. Shortnessof breath 10.8 1.2 0.0 12.0 17. Dizziness 15.7 3.6 0.0 19.3 18. Sweating,¯ushesan dchills 13.3 1.2 1.2 15.7 19. Phobias 8.4 1.2 1.2 10.8 20. Panicattacks 8.4 2.4 1.2 12.0

chronicob structiveairw ay disease andosteoarthritis. from 82to1 00%.Inth emajorityof itemsboth Forty-foursubjects(5 2%)sufferedfromoneormore interviewersgave ascoreof zero.T hekappa values physical illnesses. BasedontheD SM-IV, subjectsfell rangedfrom0.53to1 andOAGrangedfrom84to intothreediagnosticcategoriesofAlzheimer’s 100%for thetest-re testre liability. Table2 showsthe dementia(66 .3%),vascular dementia(1 3.3%)and kappaandOAGvaluesof individual itemsforboth otherdementias (20.5%). thereliab ilityan alyses. RAIDscorew as notrelatedtoag e(correlation 0.05, p <0.67),sexof theindividual ( p < 0.52), inpatientstatus( p <0.62),typeof dementia( p < Validitya nalyses 0.4),levelofcognitiveim pairment(MMSEscore) (correlation= 0.18, p <0.1)ordementiaseve rity on Content validity of RAID. Atotalof24persons theClinical DementiaR atingScale( p < 0.53). returnedtheir questionnaire tog ivethe ir opinionon However, subjects withphysic al illness had higher thesc ale. Itin cluded®vep sychiatrists, oneclinical scores(Mann-WhitneyU =518, p <0.05).Patients psychologist,threeco mmunitypsychia tricn urses, five within sightin tothe ir illness also had higheranxiety carersandninestaff nurses workingwiththe e lderly (M=14,SD=9)compared withtho se without inw ards andday hospitals andoneoccupational insight(M=7,SD=5)(Mann-WhitneyU=343, p therapist. Fourteenofthemthoughtthatall theite ms < 0.0004). inthe sc alewe reim portant.Onesuggestedthatsleep Downloaded by [University College London] at 01:59 23 January 2015 disturbancem ay notbean importantitem in th e scale.Oneindividual suggestedinclusionof eachof Reliability analyses thead ditionalsymptomslikeloss ofappetite,ag gres- sion,obsessive± compulsive symptomsas an expres- Internalconsistencyof RAID. Cronbach’salpha was sion of anxiety,d ifficultyin c opingwithun familiar 0.83,suggestingthatR AIDhasahighlevelo finternal surroundingsandaseparate sectionfor thesig nsand consistency. Alphawascalculatedfo reachsu b-group symptomsofanxietyth atd onot®tintoa speci®c ofthesc aleto c onsiderwhethertheitemswithina categoryweresug gested.Theexplanationgiveno f sub-groupweree qually affectedbythepatient’s phobias andpanic attac kswerec onsideredunsatisfac- anxietystatu s. Thealph avalues forthesu b-scales torybysevenin dividuals. rangedfrom moderateto high:Worry(alpha=0.65); Theoverlap ofsymptomsof sleepdisturbance, Apprehensionandvigilance(alpha=0.67);Motor tremblingandrestlessness witho thermedical and tension(alpha=0.51);Autonomichy peractivity psychiatricc onditionswerementionedbyfour (alpha =0.74). individuals. Unreliabilityo fthecarer’ saccountwas mentionedby oneCPNandonestaff nurse. Three Inter-rater reliability andtest-re testrelia bility of RAID. individuals questionedthereliabilityof assessing Amongthe33subjects who participatedin inte r- autonomichypersensitivitysymptomsandpanic raterreliability, thekappa valuefortheindividual attacks.Theclinicalpsycho logist andtwopsyc hiatrists itemsrangedfro m0.51to1 andtheO AGranged pointedou tthatsco resofphobias andpanicattac ks 44 K.K.Shankar et al.

TAB LE 2. Inter-raterandtest-retestreliabil ity

Inter-raterreliability Test-retestreliabil ity N = 33 N = 25

Scale kappa OAG% kappa OAG%

1. Worryabo utphysicalhealth 0.71 85 0.81 96 2. Worryabo utcognitiveperform ance 0.54 88 0.53 84 3. Worryover ® nances,fam ilyproblem s 0.68 94 0.84 92 4. Worryassoci atedw ithfalse belief and/orperception 0.69 97 0.68 96 5. Worryover tri¯ es 0.81 100 0.72 88 6. Frightenedan danxious 0.58 82 0.62 80 7. Sensitivityto no ise 0.52 82 0.53 84 8. Sleepd isturbance 0.59 94 0.71 88 9. Irritability 0.53 85 0.69 88 10. Trembling 0.71 94 0.64 92 11. Motorten sion 0.58 82 0.91 96 12. Restlessness 0.53 85 0.83 92 13. Fatigueability 0.51 87 0.58 84 14. Palpitations 0.78 97 1.00 100 15. Drymouth,sinkin gfeelinginthe sto mach 0.84 97 1.00 100 16. Shortnessof breath 0.81 94 0.78 96 17. Dizziness 0.71 94 0.58 92 18. Sweating,¯ushesand chills 0.88 97 0.78 96 19. Phobias 1.00 100 1.00 100 20. Panicattack s 0.65 97 0.57 96

couldn otbeaddedto thetotalsc oreas theyfo rmed DSM-IVcriteria for generalizedan xietyd isorder. aseparate diagnosticc ategory. Threein dividuals felt Thirteenwereconsid eredby theconsulta ntpsychiatrist thatthe sc aleco uldon ly beu sedby professionals tohave clinicalfeatures of anxietythat requ iredtreat- workinginth e®eldan dtrainingw ouldbe re quired ment.T hem eanR AIDscorefor thosew ho ful® lled for moreg eneral use.Therew as also arequestfo r DSM-IVcriteria of generalizedanxietydiso rder(M moreg uidelines. =16.9,SD=7.9)washigherthan those whodidno t (M=7.9,SD=0.5).Similarly, them ean scorew as Concurrent validity ofRAID. TheSpearman’scorre- higherintho se who wereassign edbytheconsultant lationc oefficientwas calculatedbe tweenR AIDand psychiatristtohave clinically signi®c antan xiety(M = thecarer’ srating(8 3subjects)o fthesub ject’sanxiety 15.07,SD=8.9)compared totho se withouthaving andtheconsultant’sratingof anxiety(2 4subjects). signi®c antclinicalan xiety(M =7.55,SD=5.5). Only 38subjectsw ereab leto co mpletethe A SIand Mann-WhitneyUwas calculatedforindependent CAS.TheSpearman’scorrelationc oef® cientwasalso samples based onm odi®ed D SM-IVdiagnosis and calculatedbe tweenR AIDandASI,CASandthe theconsultant’sclinicalimpression.ItshowedRAID CornellS calefo rDepressioninD ementia.T hese was ableto sign i®c antlyd istinguish betweeng roups correlationsare givenin Table 3.Sincesevenitemsin oflowanxietyandhighanxietyw henmodi®e d RAIDandtheCornellS calefo rDepressionin DSM-IVcriteria was applied(U =22.5, p < 0.006) Downloaded by [University College London] at 01:59 23 January 2015 Dementiaare similar, those itemsinb othth escales andalso based ontheco nsultant’sclinicalimpression wered eletedtog etM odi®ed R AID(MRAID)and (U = 31.5, p =0.03).Acut-off scoreo f11andabove Modi®e dCornellS cale(M Cornell).TheSpearman’s had thebest ® tfor sensitivityandspe ci®c ityof the correlationbetweenM RAIDandM Cornellw as0.2. scale. Accordingtom odi®e dDSM-IVcriteria for This indicatesthat R AIDmeasures symptomsother anxiety, atth escoreo f11ormorethe se nsitivityof than depression. thescalewas 90%andspeci®c ity7 8.5%.Thesame cut-off pointhad sensitivityof76.8%andspeci®c ity Criterion validity ofRAID. Tensubjects (of 24rated of 81.8%whenthec onsultant’sclinical impression by theconsult antpsychiatr ist)ful® lledthe m odi®ed was usedto disc riminate.

TAB LE 3. CorrelationbetweenR AIDandother m easures Constructvalidity ofRAID. All18itemsoftheR AID scalewere en teredintoth efactoranalysis. A®ve- Measure Spearmancoefficient* factorstr ucturew as derivedw hichincludedall 18 Consultant’srating( n = 24) 0.66 itemsof thesc alean daccountedfo r63.8%of the Carer’srating( n = 83) 0.73 variance.Thecontentofthefac torstructureisshown ClinicalAn xietyScale ( n = 38) 0.54 inTable 4.Inad dition,Bartlett’stestof sphericity AnxietyS tatusInventory( n = 38) 0.62 rejectedthen ullhypothesis ofanidentitym atrix(ch i CornellScale ( n = 83) 0.69 square =54.63, p =0.0000).TheKaiser MeyerOlkin *Allsigni®can t( p < 0.001). test ofsamplingadequacywasappropriateat 0 .768. Anxiety in dementia 45

TABLE.4. Factoran alysisof RAIDitems

Factor1: Eigenvalue 5.16;28.7%varian ce Q1 Worryoverphysica lhealth Q2 Worryaboutcognitiveperform ance Q3 Worryaboutfamilyproblem s/®nances Q6 Frightenedand an xious Q7 Sensitivityto no ise Q10 Trembling Q11 Tension Q15 Drymouth/sinkingfeeling Q17 Dizziness Q18 Sweating¯ushes

Factor2: Eigenvalue 2.39;13.3%varian ce Q9 Irritability Q12 Restlessness Q14 Palpitations

Factor3: Eigenvalue 1.45;8.0%variance Q8 Sleeplessness Q13 Fatigueability

Factor4: Eigenvalue 1.31;7.3%variance Q5 Worryovertri¯ es Q16 Shortnesso fbreath

Factor5: Eigenvalue 1.18;6.5%variance Q4 Worryassociatedwith false beliefs/ perceptions

Discussion severity of anxietyremains unclear. Panicattac kthat occursinan establishedp hobicsitu ationis regarded RAIDwas easy touse, acceptable tothepatien tsand as an expressionofthese verityo fphobia (ICD-10: popular withthe c arers. Many oftheseve rely disabled WorldH ealth Organization,1992). patients weren otab leto co mmunicatethe ir symp- Subjectsw ithon eormorep hysical illnesses scored tomsreliably( n =45;54%).Howeverthec arerswere higherco mparedwiththo se withoutphysical illness. ableto g ivea detailedaccountof theirb ehaviour. This wasconsistentwiththe stud yofLindesay (1990), Interviewingthe carer ® rst also helpedto in form whofoundthatthepresenceofphysical health questioningaboutc ertainsym ptomsinthe p atient problemswas associatedwithg eneralizeda nxiety interview. Forexam ple,whenpatie ntshad aparticular disorderandagoraphobia.Subjects who retained delusionitc ouldb eenquiredaboutlate roninthe insightintotheirmemoryproblemsweresignificantly interviewafter gettingotherrelevantinfo rmation.It morean xiousas measuredby thesc ale, andthis was also importanttoin terviewthe patien ts lateras ®ndingwasconsistentwiththe stu dyofBallard et al. theywereable tod escribeth eirsym ptomswhichthe (1994). carerfailed ton otice(this wasespecially trueforthe Thestudiesof Reisberg et al.(1985)andBallard et physical symptoms). al.(1994)suggestedthat an xietysym ptomsare more Thoughtheautonomicsym ptomsformacore commoninm ildd ementiasu fferers. However, ina componentof anxiety,the item sof thesu b-scale populationw ithm ildde mentia, Wands et al. (1990) `autonomichyperactivity’tendedtobelessfrequently

Downloaded by [University College London] at 01:59 23 January 2015 foundaslightinc rease inan xietyas cognitivefu nc- scoredthan theotheritems. This may bedueto tiondeclined.Forsell et al.(1993)foundvariationsin coexistingphysical illness andmedicationtak enb y thep hysician’sratingofanxietyand th einformant’s thesubjects overlappingw iththe sym ptomsdueto rating.Thep hysiciansnoteda declineinle velo f anxiety. Sincethe R AIDscaled oesnotallow for ratingsym ptomsrelatedtophysic al illness or side anxietyw ithse verityof dementia, whilethe in form- effects of medication,theymighthave scoredless. ant’sratingshoweda linear increase withse verity. Sincemany oftheseverely disabledpatients were However,in this studylevelofanxietyw asnotassoci- also notable tocommunicateth eirsym ptomsclearly atedwithe itherlevelof cognitiveim pairmentonthe andas many of thesesym ptomsmay notbe read ily MMSEscoreo rthestag eofdementiabased onth e observable by thec aregiverstheytendedtosc ore Clinical DementiaR atingS cale.Earlierstudiesmay less. have haddifficultyin ratin ganxietyin t hem ost Thesco resofphobias andpanicattack w eren ot impairedduetothelac kofadequatesc ales. addedto the to talsco re.Theconceptofphobia TheinternalconsistencyofRAIDwashigh, includedsim plep hobia, socialph obia andagora- suggestingthatR AIDfunctionsasascale.Thealpha phobia.Thiswasconsideredtobetooextensive tobe valuesofthesu b-scalesworry, apprehension and coveredfully inthesc ale.Thevariouspresentationof vigilanceandautonomichype rsensitivitywerealso phobiaswerec overedbyastandardizedphobic high.Thesu b-scaleo fmotorten sion had alower disordersc reeninth eGuy’s/AgeC oncernsurvey alphavaluepe rhaps because theite m`restlessness’ in (Lindesay et al.,1989).Theissueofpanicattacksand thesu b-scalemotortensionscoredm orefre quently 46 K.K.Shankar et al.

thantheotheritemsint hatsub-scale.Cohen- anxietyscale. Also clinically, therew asknowntobea Mans® eld(1 986),inastudyofagitatedbehaviourof signi®c antcomorbiditybe tweenanxietyan ddepres- theeld erly ina nursingh ome,failed toreve al an sion.Inth eGuy’s/AgeConcernsurvey, Lindesay et `anxiety’facto rlinkedto the c oncept.H owever, in al.(1989)foundconsiderablecomorbidityof depres- this study restlessness was showntobe a usefuland sionwithph obias andanxiety. Nearly 40%ofphobic observablesig nof anxietyin d ementiasuffe rers. For subjectsalso had depression andwereab outthree- example,int wopatients whose test-retest score and-a-half timesmorelik elyto have depressionthan changed,itw asassociatedw ithc hangeinthe leve lof thenon-phobicsu bjects. Also, 91%of personswith restlessness. Asubjectwho was calman drelaxed generalizedanxietydiso rderalso had depressivesym p- duringthe®rst interviewwas notedto b emore tomatology.A lexopoulous(1990)foundina seriesof anxiousduringthese condinterviewan dhis restless elderly outpatients withm ajordepressionthat3 8% pacingaroundthew ard was areadily observable of themalso metthe D SM-IIIcriteria for anxiety behavioural changeinh im.Anotherpatientwhowas disorder. extremely restless duringthein itialinte rviewwas Them ean RAIDscorew as higherin tho se who subsequentlypresen tedbyhis wife withan electronic ful® lledthe m odi®ed D SM-IVdiagnosis ofgeneral- organw hich heusedto play.Thisreducedhis restless- izedan xietyd isorderan dthec onsultant’sclinical ness andalso his levelofanxiety. diagnosis.This was expectedas theDSM-IVcriteria RAIDhad moderateto hig hlevels ofbothin ter- was takenin toc onsiderationwhendesigningthe raterreliab ilityan dtest-retestreliability.Thepossible RAIDscale.Theconsultant’sclinicalim pressionwas explanationsfor changesinth etest-retestre liability takenasa`goldstandard’ inthisstudy. Spitzer(1983) may beduetothe fact thatthe rew as agenuine describedasimilarprocedurew hichcouldb eused changeinso mepatientsd uringthetimeinterval (of asanultimatecriterionor`goldstandard’for uptotend ays) inre peatingthescale .This couldalso evaluatingthevalidityofastructureddiagnostic bed uetothe in consistenciesin the c arer’sreport assessmentinstrument.Hedescribeditw iththe duringthe® rst andthesecondinterviews.Hopeand acronym`LEADstandard’.Itinvolvedthreeessent- Fairburn(1992),intheirstu dytodeveloptheP resent ialconcepts:Longitudinal, Expert,a ndAllD ata. BehaviouralExamination(PBE),an investigator- Longitudinal: this meantthatth ediagnostice valu- based interviewtom easurebe havioural abnormali- ationwas notlimitedtoa singlee xaminationdone tiesind ementedsubjectsafter listeningasecond ato nepointint heevolutionoftheillness. Expert: timetoaudio-tapesofinterviews,foundanumberof thecriteriondiagnoseswerem adebyexpertclini- instanceswherethe c arers had givendiffere ntanswers cianswhohave demonstratedtheir abilitytom ake ontwo o ccasionstoexactlythe sam equestions. expertdiagnoses.AllD ata:theexpertcliniciannot Amongtheran geofprofessionals workingw ithth e only systematically evaluatesthesubjects,butwill elderly andcarers who gave their opinionon th e interviewotherinformants, suchasfamily members, scale,thesc alew as feltto be co mprehensive andall andwillh ave access todataprovidedbyother theite mswereco nsideredimportant.Clearerguide- professionals,suchaswardstaff andprevious lineswerene eded,includingbetterexplanationfor therapist. Thepatients inth estudywerem ostly phobias andpanicattack ite ms. However, thecu rrent underthecareo ftheconsultantpsychiatrist who explanationsweretaken fro mtheP SE,whichis a ratedthem.Hence,this studymanagedtoachieve standardizedinstrument.Sincemany staff notedthat theabovec riteria highlightedbytheLEADstand-

Downloaded by [University College London] at 01:59 23 January 2015 phobias andpanicattack swered istinctsyn dromes ard.Itis avalidp rocedurew hichcouldb etakenas thesc aleh as sincebe enm odi®e dwithadd itional a`goldsta ndard’. instructionstatingthatthe sco resofphobias and Thefac toran alysis indicatedthatthe R AIDscale panicattac ksweren ottob eaddedto the to talsc ore. comprised®vefac tors all ofwhich madea contribu- Clear descriptionsfor phobias andpanicattackswere tionto the var iance.Eachof the1 8itemsonthe also added. RAIDscalew asacomponentofthe®vefac tors. This RAIDsigni® cantlycorrelatedwiththevisual suggests thatall itemswerene cessaryandthescale analoguescaleo fthecarer’sratingof anxietyin th e covers agoodrangeofanxietysym ptomsandsigns patient,andthein dependentratingbythec onsultant andhas goodconstructvalidity. onthe leve lof anxiety.It also correlatedw ellw ith RAIDwasnotadiagnosticsc ale. Butinthis study bothCASandASIscores.TheSpearmancorrelation itw as foundthata scoreo f11an dabove had good ofRAIDandtheC ornellD epressionScalewe re sensitivityan dspeci®c ity. These nsitivitysc oreswere higherth an theanx ietysc ales. Thec orrelationo f lowerwhenthe con sultant’sclinicalopin ionw astaken MRAIDandMCornell, however, waslow. This intoc onsideration.RAIDdoes notreplac etheneed suggests thatRAIDmeasurescertainite mswhich are for properclinical assessment.H owever, thescore speci®c for anxietyan dotheritemswhich have some couldbe a helpful guidein assessm entan dmanage- overlap withd epression itemsonthe C ornellSc ale. mentofindividualpatients. Conceptually, theseitemscouldnotbeseparatedfrom Therea ppeartobe n ootherpsychometrically theRA IDscaleas itwouldm ake RAIDanincomplete validatedratingscale sspeci®cally designedfor clinical Anxiety in dementia 47

assessmentofanxietyin de mentia. Previous scales BALLARD , C.G. et al.(1994). Anxietyd isorderin dem entia. had severeclinic al andm ethodologicallim itations IrishJour nalofPsychologicalMedicine , 11(3), 108± 109. whenu sedin d ementiap opulations. Theobserver BALLARD , C.G. & OYEBODE ,F.(1995). Psychoticsym ptoms indem entiasufferer s. InternationalJournal ofG eriatric ratedClinical AnxietySc alean dtheA nxietyS tatus Psychiatry. Inventorywhichwereu sedin th is studycouldon ly BALLARD , C.G., BOYLE, A., BOW LER, C. & LINDESAY, J. (1996). becompletedina minority (38of 83)ofsubjects Anxietydisord ersin dem entiasufferer s. Dementia. In interviewed.Thesescales rely ontheinform ationg iven press. bythesu bjectsan dinvolvean understandingoftheir BERGM AN N,K.(1978). Neurosisan dpersonalityd isorderin oldage. In: A .D.I SAACS & F. POST (Eds), Studies in subjectivesym ptoms.Thesu bjectsalso needtoh ave geriatric psychiatry (pp.41± 75). Chichester:Jo hnWiley. sufficientcomprehension andjudgementtoanswe r BU R NS ,A.(1991). Affectives ymptomsinA lzheimer’s questionsrelated to affectandideatio n.Validinfor ma- disease. InternationalJournal ofG eriatric Psychiatry , 6, tionabout¯uctuatingsymptomscanbe o btained 370± 376. onlyfromsubjectswithin tactm emory. Inthe ex peri- CO HEN,J.(1960). Acoefficientof agreementfornominal enceofadministeringthemin this studythesu bjects scales. Educationaland PsychologicalMeasurement , 20, 37± 46. tendedtogetconfusedwhensymptomswerep robed CO HEN,J.(1968). Weightedkappa: no minalscale agree- into.In addition,certainsym ptomsthatwe resp ecific mentwithprovisi onfo rscaledd isagreementor partia l totheelderly andpatientswithde mentia(lik eworry credit. Psychological Bulletin , 70, 299± 308. over cognitivepe rformance,rep eatedly callingfor COHEN-MANSFIELD,J.(1986). Agitatedbehaviour in the elderly. attentionof caregiversovertrivial matters,etc.)w ere II:Preliminaryresults in the cognitivelydeteriorated. Journal ofthe American Geriatric Society , 34, 722± 727. notincluded. C O H E N -M A N S F IE L D , J. & B IL L IG ,N.(1986).Agitated Thereis apaucityo fresearchonanxietyin behaviourinthe eld erly.I: Aconceptualreview. Jour nal dementiaan dmostoftheex istingliteratured iscusses oftheAmerican Geriatric Society , 34, 711± 721. anxietysym ptomsalongwitho thercognitive,affec- CO PELAND , J.R. et al.(1976). ASemi-structuredclinica l tivean dbehavioural symptoms.Partofthis problem interviewfor theassess mentof diagnosisand m ental statein theelderly. The GeriatricM entalS tateS chedule: is duetolac kofavalid andreliab lescale fo ruse in developmentand reliabi lity. Psychological Medicine , 6, theelderly cogn itively impaired population.Itis hoped 439± 449. thatthe R AIDscalew illbe a usefulinstrumentin D AVIS, M.R. et al.(1982). Anxietyand depress ioninthe clinical practiceto id entify andm easurean xiety. It agedwith sleep d isturbances. JournalofC linical and mayhighlightthene edfor treatmenttoreduced istress Experimental Gerontology , 4, 239± 248. andmeasurere sponse tothe rapeuticin terventions. FO LSTEIN, M.F., FOLSTEIN, S.E. & M CH IGH,P.R.(1975). Mini-MentalState: a practicalm ethodfor gradingthe Wealso hopeitw illb eusefulinre searchstudies, to cognitivestate o fpatientsfo rtheclinicia n. Jour nal of studythe prevalen ceofanxietyin dementia,the course Psychiatric Research , 12, 189± 198. ofanxietysym ptoms,ther isk factorsassociatedw ith FO RSELL, Y., JOR M, A.F. & W INBALD ,B.(1993). Variationin anxiety, andtheevalu ationof treatments foranxiety psychiatrican dbehaviouralsym ptomsindifferen tstages indementia. ofdementia:d atafro mphysicians’exam inationand informantsrepor ts. Dementia, 4, 282± 286. GOTTLIEB, G.L., RAQU EL, E.G. & RUBEN,C.G.(1988). Reli- Acknowledgements abilityof psychiatricscales in patien tswith dem entiaof Alzheimer’stype. American JournalofPsychiatry , 145(7), 857± 860. This researchwassupportedby Camdenand Isling ton GOUDEM AN, M. & T HOM AS,P.(1994). Anxiety,d epression CommunityN HSTrustan dPrioryHospitalsGroup. anddementia. Revue-Du-Praticien , 44,1448± 1452. OPE AIRBUR N Downloaded by [University College London] at 01:59 23 January 2015 Wethan kthepatients,staff andcarersinvolvedinthe H ,T. & F ,C.G.(1992).ThePresentB ehavioural study. Wethan kProfessor JamesLindesay for his Examination(PB E):thed evelopmentofaninterviewto measurecurrentbehaviouralabn ormalities. Psychological helpfulcomments. M edicine, 22, 223± 230. H UGHES, C.P. et al.(1982). Anewscale fo rthestagin gof dementia. British Journal ofPsychiatry , 140, 566± 572. References KOENDER S, M.E. et al.(1993). Traitan xietyan dachieve- A BSHE R, J.R. & CU MM ING S ,J.L.(1994).Cognitivean d mentmotivationare positivelyco rrelatedwith m emory non-cognitiveaspects of dementiasyndrom e:anoverview. performanceinpatient swhovisit a geriatricout-patients In: A. BUR NS & R. LEVY (Eds), Dementia (pp.59± 76). clinicwith am nesiasym ptoms. Psychological Reports , 73, London:Chapman&Hall. 1227± 1231. ALEXO POLOU S,G.S.(1990). Anxiety-depressionsynd romes LADER ,M.(1982). Differentialdiagnosisof anxietyin the in old age. InternationalJournal ofG eriatric Psychiatry , 5, elderly. Journal ofC linicalPsychiatry , 43, 4± 7. 351± 353. LINDESAY,J.(1990)The Guy’s/AgeConcernsurvey:physical ALEXO POLOU S, G.S., ABRAM S, R.C., YOUNG , R.C. & SHAM O- healtha ndps ychiatricd isorder ina nurbanelderly N IAN ,C.A.(1988).CornellS calefo rdepressionin community. InternationalJournal ofG eriatric Psychiatry , dementia. Biological Psychiatry , 23, 271± 284. 5, 171± 178. AMER ICAN PSYCHIATR IC ASSOC IATION (1987). Diagnostic and LINDESAY,J.(1991). Phobicdiso rder inthe elderly. B ritish statisticalmanual ofm entaldisorders ,3rd edition,revised . JournalofPsychiatry , 159, 531± 541. Washington:APA. LINDESAY, J., BRIGGS, K. & M URPHY,E.(1989).TheGuy’s/Age AMER ICAN PSYCHIATR IC ASSOC IATION (1994). Diagnostic and Concernsurvey:prevalen ce ratesof cognitiveim pair- statisticalmanual ofm entaldisorder ,4thedition .Wash- ment,depressionandanxietyina nurbanelderly ington:APA. community. British Journal ofP sychiatry , 155, 317± 329. 48 K.K.Shankar et al.

OR RELL,M.W. (1994). Life eventsand senile dem entia. WAG NER, K.D. & L OR ION,R.P.(1984). Correlatesof death PhD thesis.U niversityof London. anxietyin eld erlyperson s. JournalofC linicalPsychology , OR RELL, M.W. & BEBB ING TON ,P.(1996). Psychosocialstress 40,1234± 1241. andanxietyin senile d ementia. JournalofA ffective WAN D S, K., M E R SK EY, H. et al.(1990). Aquestionnaire Disorders, 39, 165± 173. investigationofanxietyanddepressioninearly PLUTCHIK, R., CONTE, H. et al.(1970). Reliabilityand validity dementia. Journalofth eAmericanGeriatric Society , 38, ofascalefor assessingthefunction ingofgeriatricpatients. 535± 538. Journalofthe AmericanGeriatric Society , 18, 491± 500. W ING , J.K., C OOPER, J.E. & SAR TOR IUS, N. (1974). The REISBERG , B., BO RSTEIN , J., SALOB, S.P., FER RIS, S.H. et al. measurementandclassi® cation ofpsychiatricsymptoms . New (1993). Behaviouralsym ptomsinA lzheimer’sdisease: York:Cam bridge UniversityPress. phenomenologyan dtreatment. Journal ofC linical WORLD H EALTH ORGANIZATION (1992). Thetenth revision of Psychiatry, 48(Suppl.),9± 15. the InternationalClassi® cation ofDiseases and related health SIEGEL, S. (1956). Nonparametric statisticsforbehav ioural problems(ICD-10) .Geneva:WHO. sciences.NewYork:M cGraw-Hill. YESAVAG E, J.A. & TAYLOR ,B.(1991). Anxietyan ddementia. SNAITH , R.P., BAU GH , S.J., CLAYDEN , A.D., H USAIN, A. & In: C. SALZM AN & LEBOW ITZ (Eds), Anxiety in theelderly: SIPPLE,M.A.(1982).TheClinicalAnxiety Scale: an instru- treatment andresea rch .NewYork:Spring er. mentderivedfrom the H amiltonA nxietyS cale. B ritish ZUNG ,W.W.K.(1971). Aratinginstrumentfor anxiety JournalofPsychiatry , 141, 518± 523. disorders. Psychosomatics , 12, 371± 379. Downloaded by [University College London] at 01:59 23 January 2015 Anxiety in dementia 49

Appendix 1

RatingAnxietyInDementiaÐ RAID

Patient’sN ame: DOB: Hospitalno :

Rater’sN ame: Occupation:

Patient’sstatus at evaluation: 1.Inpatient.2. outpatient.3.dayho spital/daycentre patien t.4.Other (specify)......

Scoring system: U.unableto evaluate.0. absent.1.mildor intermittent.2. moderate.3.severe Ratingshouldbe basedo nsymptomsandsigns occurr ingdurin gtwoweeks prio rtotheinter view. Noscoreshou ldbe givenif symptomsresultfro mphysicaldisabilityor illness. Totalsco re isthe su mofitems1to18.Ascoreof 11 ormore suggestssigni® can tclinicalanxiety.

Score

1. Worryabo utphysicalhealth. 2. Worryabo utcognitiveperform ance(failing m emory, gettinglo stwhen goes out, not able to following conversation).

Worry 3. Worryover ® nances,fam ilyproblem s,physicalhealthof relatives. 4. Worryassoci atedw ithfalse belief and/orperception. 5. Worryover tri¯ es (repeatedlycalling for attentionover trivialm atters). 6. Frightenedan danxious(keyed u pandontheed ge). 7. Sensitivityto no ise(exagge ratedstartle respon se). Apprehension andv igilance 8. Sleepd isturbance (troublefalling o rstayingasleep) . 9. Irritability(m ore easilyan noyedthan usu al,sho rttempered andangryoutbursts). 10. Trembling. 11. Motortension (com plainof headache, otherbod yaches and pains). Motor tension 12. Restlessness(® d geting,cannotsitstill, pacing ,wringing hands,picking clothes ). 13. Fatigueability,tiredne ss. 14. Palpitations(complainso fheartracin gorthumping).

Downloaded by [University College London] at 01:59 23 January 2015 15. Drymouth(no tdueto medication),sinkingfeelingin the stomach. 16. Hyperventilating,shortnessof breath(even w henno t Autonomic hypersensitivity exerting). 17. Dizzinessorlight-headedness(co mplainsasif goingto faint). 18. Sweating,¯ushesor chills,tinglin gornumbnessof ®ngers and toes. Phobias:(fears which are excessive,that d onotm akesense and ten dtoavoidÐlike afraid of crowds, goingoutalone,bein gina smallroo m,orbeingfrightenedbysomekindof animals,heights ,etc.) Describe. Panic attacks :(Feelingso fanxietyor dreadthat are sostrongthatthin ktheyare goingtod ieor havea heartattack an dtheysim plyhave to dosomethingto stop them ,likeim mediatelyleavingtheplace, phoningrelatives,etc.) Describe.