Vol. - No. - - 2014 Journal of Pain and Symptom Management 1

Review Article

How Are the Patient PopulationsCharacterized in Studies InvestigatingDepression in Advanced Cancer? ResultsFrom a Systematic Literature ReviewElene Janberidze, MD, Marianne Jensen Hjermstad, PhD,Dagny Faksv�ag Haugen, MD, PhD, Katrin Ruth Sigurdardottir, MD,Erik Torbjørn Løhre, MD, Hanne Cathrine Lie, PhD, Jon H�avard Loge, MD, PhD,Stein Kaasa, MD, PhD, and Anne Kari Knudsen, MD, PhD, on behalf of EUROIMPACTEuropean Palliative Care Research Centre (PRC) (E.J., M.J.H., D.F.H., K.R.S., E.T.L., S.K.,

A.K.K.), Department of Cancer Research and Molecular Medicine, Faculty of Medicine, Norwegian

University of Science and Technology (NTNU), Trondheim; Department of Oncology (E.J., E.T.L.,

S.K., A.K.K.), St. Olavs Hospital, Trondheim University Hospital, Trondheim; Regional Centre for

Excellence in Palliative Care (M.J.H.), South Eastern Norway, Oslo University Hospital, Oslo;

Regional Centre of Excellence for Palliative Care (D.F.H., K.R.S.), Western Norway, Haukeland

University Hospital, Bergen; Sunniva Centre for Palliative Care (K.R.S.), Haraldsplass Deaconess

Hospital, Bergen; Department of Behavioural Sciences in Medicine (H.C.L., J.H.L.), Institute of Basic

Medical Science, Faculty of Medicine, University of Oslo; and National Resource Centre for Late Effects

after Cancer Treatment (J.H.L.), Oslo University Hospital, Oslo, Norway

Abstract

Context. Prevalence rates of depression in patients with advanced cancer vary

considerably. This may be because of heterogeneous samples and use of differentassessment methods. Adequate sample descriptions and consistent use ofmeasures are needed to be able to generalize research findings and apply them toclinical practice.

Objectives. Our objective was twofold: First, to investigate which clinicallyimportant variables were used to describe the samples in studies of depression inpatients with advanced cancer. Second, to examine the methods used for assessingand classifying depression in these studies.

Methods. PubMed, PsycINFO, Embase, and CINAHL were searched combiningsearch term groups representing ‘‘depression,’’ ‘‘palliative care,’’ and ‘‘advancedcancer’’ covering 2007e2011. Titles and abstracts were screened, and relevant full-text articles were evaluated independently by two authors. Information on 32predefined variables on cancer disease, treatment, sociodemographics,

Address correspondence to: Elene Janberidze, MD, Euro-pean Palliative Care Research Centre (PRC), Depart-ment of Cancer Research and Molecular Medicine,Faculty ofMedicine, NorwegianUniversity of Scienceand Technology (NTNU), Kunnskapssenteret 4.etg.,

St. Olavs Hospital, N-7006 Trondheim, Norway.E-mail: elene.janberidze@ntnu.no

Accepted for publication: December 12, 2013.

� 2014 U.S. Cancer Pain Relief Committee.Published by Elsevier Inc. All rights reserved.

0885-3924/$ - see front matterhttp://dx.doi.org/10.1016/j.jpainsymman.2013.11.013

2 Vol. - No. - - 2014Janberidze et al.

depression-related factors, and assessment methods was extracted from thearticles.

Results. After removing duplicates, 916 citations were screened of which 59articles were retained. Age, gender, and stage of the cancer disease were the mostfrequently reported variables. Depression-related variables were rarely reported,for example, antidepressant use (17%) and previous depressive episodes (12%).Only 25% of the studies assessed and classified depression according to a validateddiagnostic system.

Conclusion. Current practice for describing sample characteristics and assessingdepression varies greatly between studies. A more standardized practice isrecommended to enhance the generalizability and utility of findings.Stakeholders are encouraged to work toward a common standard for sampledescriptions. J Pain Symptom Manage 2014;-:-e-. � 2014 U.S. Cancer Pain ReliefCommittee. Published by Elsevier Inc. All rights reserved.

Key Words

Advanced cancer, palliative care, depression, generalizability, assessment

IntroductionDepression is probably the most studied psy-

chiatric disorder in advanced cancer patients,1

with reported prevalence rates ranging from3% to 58%.2,3 The great variability in preva-lence rate estimates reflects in part the hetero-geneity of the populations studied and in partthe lack of agreed-on standards for definingand assessing depression in this patient group.Thus, clear descriptions of the study sampleand of the assessment methods are necessaryto judge the generalizability of study findingsand their relevance for clinical practice.4

Common symptoms of advanced cancer dis-ease, such as fatigue, lack of appetite, and sleepproblems, are also used as diagnostic criteria fordepression (e.g., Diagnostic and Statistical ManualofMental Disorders, FifthEdition [DSM-V]; Interna-tional Statistical Classification of Diseases and RelatedHealth Problems, Tenth Revision [ICD-10]).5,6 De-pending on which symptoms are included inthe different depression assessment methods,the extent of the cancer disease may to varyingdegrees inflate the number of false-positivedepression cases and consequently threaten thevalidity of the depression assessment and influ-enceprevalencerateestimates.7e12Furthermore,a systematic literature review published in 2009,covering the period from 1966 to 2007, demon-strated that 106 different assessment tools fordepression were applied in studies in palliativecancer care and that a validateddiagnostic system

was used only in a minority of the studies.13 Ifthese diverse assessment practices still dominate,is not known. However, to reduce the problem ofthe great variation in sample descriptions anddepression assessment methods as presented inthe literature, theEuropeanUnion-fundedEuro-pean Palliative Care Research Collaborative(www.epcrc.org) worked toward developing astandardizedassessment andclassification systemfor common symptoms in palliative care cancerpatients.14 This work has been continued withinthe European Palliative Care Research Centre(www.ntnu.edu/prc), an international researchcollaborative with the overall aim to improvesymptom management and research quality inpalliative care.Adding to the problem of a valid assessment

of depression, is the heterogeneous nature ofadvanced cancer populations with regard toage, diagnosis, extent of the cancer disease, sur-vival, symptom burden, comorbidity, physicalfunctioning, and need for treatment andfollow-up.15e17 A precise characterization of thestudy sample is needed to be able to compare re-sults across studies and transfer researchfindingsto clinical practice.4,18,19 An international expertgroup recently emphasized poor and unsystem-atic reportingof sample characteristics in clinicalstudies in palliative care as an important barrierfor conducting high-quality research.20 Stan-dardized descriptions of patient samples havebeen advocated for clinical studies in general.

Vol. - No. - - 2014 3Patient Characterization in Depression Studies

The Consolidated Standards of Reporting TrialsStatement was developed 20 years ago and in-cludes a 25-item checklist and a participantflow diagram to guide the reporting from ran-domized controlled trials (RCTs), also includingspecifications onhow to describe the populationstudied.21 Similarly, the Strengthening the Re-porting of Observational Studies in Epidemi-ology statement recommends a checklist of 22items, including sample characteristics, for re-porting studies in epidemiology.22 For clinicalstudies in palliative care, Currow et al.19,23 haveproposed a similar checklist of core variables todescribe populations and service characteristics,also applicable for advanced cancer populationsbecause these terms overlap. The list includes in-formation related to five domains: ‘‘individualparticipant’s demographics,’’ ‘‘caregiver,’’ ‘‘ser-vice,’’ ‘‘health and social policy,’’ and ‘‘research.’’Demographics cover age, gender, socioeco-nomic status, ethnicity, life-limiting illness, per-formance status, and days from referral untildeath.23 However, despite these statements andinitiatives, a recent literature review investigatingthe description of palliative care cancer patientsamples included in RCTs concluded that veryfew demographic or disease-related variableswere consistently registered and reported.24

Previous research has identified several riskfactors for development of depression in can-cer patients:25 female gender,26 poor perfor-mance status,27 and certain cancer diagnosessuch as pancreatic,28 lung,29,30 and breast can-cer.31 Furthermore, a prior psychiatric history,previous depressive episodes, and alcoholdependence are known risk factors for devel-oping depression during the cancer diseasetrajectory.32,33 In addition to these risk factors,the duration of depressive symptoms, theirfunctional consequences, and ongoing treat-ment in terms of drugs and/or psychotherapyare relevant variables to include in a report ondepression in advanced cancer.32,34e36

To the best of our knowledge, no studieshave systematically examined the extent of re-porting relevant sample characteristics instudies on depression in advanced cancer.Thus, the overall aim of the present study isto identify candidate variables to includewhen reporting clinical studies of depressionin advanced cancer patients as well as informa-tion on how depression is assessed and classi-fied. The following research questions were

addressed in the present systematic literaturereview:

� How were patient samples described in re-ports on clinical studies of depression inadvanced cancer?

� How was depression assessed and classi-fied in these studies?

MethodsThe databases PubMed, CINAHL (through

EBSCOhost), PsycINFO, and Embase (throughOvidSP) were searched for relevant studies.The search strings were designed for each data-base using both free text and controlled vocabu-lary (e.g., MeSH, CINAHL Headings, APAThesaurus, EMtree) and consisted of combina-tions of terms representing ‘‘depression,’’ ‘‘palli-ative care,’’ and ‘‘advanced cancer’’ (Appendix I;available at jpsmjournal.com). The searchescovered the years 2007e2011with the last searchdate being January 3, 2012. Search details andthe specific search strings for all bibliographicaldatabases will be provided on request by the cor-responding author.

Titles, abstracts, and keywords of the 916 iden-tified citations were independently screened bytwo authors (E. J. and A. K. K. or M. J. H. andD. F. H. or K. R. S. and E. T. L.) to judge their po-tential relevance. The following inclusioncriteria were applied: clinical study with adultadvanced cancer patients (18 years and older)and primary study outcome including the termdepressionordepressive disorder. The exclusioncriteria were non-English publications, reviews,commentaries, case reports, publications ad-dressing tool development (e.g., validationstudies), and studies with noncancer patientsincluded in the sample.

The term ‘‘advanced cancer patients’’ wasdefined to include patient samples describedwith the following terminology: ‘‘palliative,’’ ‘‘met-astatic,’’ ‘‘terminally ill,’’ ‘‘end-of-life,’’ and/or‘‘not curable.’’

References deemed eligible for inclusion inthe screening process were then independentlyexamined in full text by two of the authors. Anydiscrepancies between the reviewers’ selectionswere discussed to obtain consensus, or a thirdauthor (M. J. H.) was consulted.

A predefined checklist (Appendix II; avail-able at jpsmjournal.com) was used to extractdata from the identified full-text articles. The

Fig. 1. Flowchart presenting the study selection pro-cess. *The study addresses other symptoms of can-cer patients and/or cancer treatment.

4 Vol. - No. - - 2014Janberidze et al.

list was partially based on the content of theEuropean Association for Palliative Care(EAPC) Basic Dataset,37 holding a set of itemsrecommended for use when reporting resultsfrom studies in palliative cancer care,including sociodemographic, disease-related,and patient-reported variables. Specific vari-ables of relevance for diagnosing and treatingdepression were also included in the list, suchas psychiatric history, previous depressive epi-sodes, duration of the depressive episode, his-tory of alcohol and/or drug abuse, andongoing treatment for depression (pharmaco-logical and nonpharmacological). The check-list also contained information on methodsused for assessment and classification (i.e.,diagnostic system) of depression.13

This was intended as a descriptive study ofsample descriptors and assessment methods,thus quality rating of the included studieswas beyond the scope of the present report.Standard statistical procedures were used foranalysis of frequencies. Extracted data werestored and analyzed in PASW, version 18.0(SPSS, Inc., Chicago, IL).

ResultsCitations and Publications

The search yielded 1669 potentially relevantcitations, of which 753 were duplicates. Afterthe initial screening of titles, keywords, andabstracts, 125 (14%)publications of the remain-ing 916 were retained for full-text examination.These 125 articles were assessed according tothe predefined inclusion and exclusion criteria,resulting in the inclusion of 59 publications.The main reasons for exclusion were publica-tions not reportingon advanced cancer patients(13%) and publications not having depressionas a primary outcome (18%) (Fig. 1).

Overview of the Included StudiesAll 59 publications (Table 1) addressed

advanced cancer patients having depression asa study endpoint. Most studies (23 [39%])were cross sectional, 32% were prospective,and 12% were RCTs. The sample size rangedfrom 23 to 1439 patients. The most commonstudy locations were general hospital depart-ments, such as oncology departments (26[44%]), followed by hospice/palliative careunits (14 [24%]). Outpatient services (36%)

were the most common setting, followed byinpatient care (30%); 17% of the studies pro-vided no information about the setting.

Description of the Patient PopulationsTable 2 shows results for description of the

patient samples. Age and gender were themost frequently reported sociodemographicvariables, reported in 93% and 90% of thestudies, respectively. Twenty-nine (49%) studiesreportedmarital status. Informationon religionwas reported in six (10%) and social networkalso in six (10%) studies. Performance statuswas reported in 38 (65%) studies; nine (15%)of these studies used performance status as aninclusion criterion. Information about cogni-tive function was reported in 23 (38%) studies,being an inclusion criterion in 21 of these 23.The stage of the cancer disease was

described as locally advanced, advanced, ormetastatic in 44 (75%) of the studies. Differentterminology to describe the sample for diseasestatus was applied as follows; ‘‘advanced’’ in 26

Table 1Assessment and Classification of Depression in the 59 Studies Included in the Review

First Author,Ref Year,Country N

TerminologyUsed,a Setting Design Main Outcome

DiagnosticSystem

Assessment Methods

ResultsType ofInterview

Depression-SpecificQuestionnaires

QoLQuestionnaires

Classification of depression based on a diagnostic system (n ¼ 15)

Akechi,67 2010,Japan

728 Metastatic,inpatients

Retrospective Gender differences forsuicide

DSM-IV Structured clinicalinterview

d d 100% with depression.Poor physical functioningand advanced stage weresignificant risk factorsamong male patients

Akechi,53 2009,Japan

122 Not operable,no information

Prospective,two-yearfollow-up

Psychological factors andsurvival

DSM-III-R Structured clinicalinterview

HADS POMS 7% with depression. Noneof the psychosocial factorswere associated with survival

Capozzo,58 2009,Italy

50 Terminal,inpatients

Intervention,two-weekfollow-up

Efficacy of ADs DSM-IV No information HADS d 100% with depression. ADs areeffective for reducing depressivesymptoms on HADSsubscale

Inagaki,38 2007,Japan

21 No information,no information

Cross-sectional Regional cerebral glucosemetabolism

DSM-IV Structured clinicalinterview SCID-I

HDRSb MDASI 2% with depression. Higherglucose metabolism indepressed patients

Jacobsen,46

2010, USA123 Advanced, no

informationProspective Distinguish grief from

depressionDSM-IV and

Endicottcriteria(substitute)

Structured clinicalinterview SCID-I

d BRC, ICG-R 38% with depression. Grief wasdistinct from depression andwas associated with wish to die

Lichtenthal,47

2009, USA289 Advanced, no

informationCross-sectional If prevalence of mental

disorders increases atthe end of life

DSM-IV andEndicottcriteria(substitute)

Structured clinicalinterview SCID-I

d MQOL, ICG-R 11% with mental disorders.No association was found

Okamura,59

2008, Japan60 Advanced,

inpatientsRetrospective If treatment algorithm was

followed in patientswith MDD

DSM-IV No information d d 100% with depression.Applicability rates werehigh, but several problemswere identified

Rabkin,44 2009,USA

58 Advanced, in-and outpatients

Prospective To determine if new clinicaldepression emerges overtime

DSM-IV Semistructuredinterview byPHQ-9

d PANAS, VAS 7% with depression, 3% werealways depressed, 14% weredepressed for the first time.MDD is not an inevitable partof dying process

Reeve,45

2007, UK70 Terminal,

outpatientsProspective Prevalence of depression ICD-10 Structured clinical

interview CIS-rEDS d 4% with depression

Ruijs,68

2011, TheNetherlands

64 End-of-life,outpatients

Prospective,two-monthfollow-up

Prevalence of depressionand relationship toeuthanasia

DSM-IV Semistructuredinterview SCAN

HADS, single-itemdepressionquestion(from ESAS)

d 2% with depression. 27%requested euthanasia.No association found

Schillani,61 2008,Italy

23 Terminal,inpatients

Intervention,two-weekfollow-up

Examine depression, anxiety,and mental adaptation tocancer in relation to ADtreatment

DSM-IV Structured clinicalinterview

HADS d 100% with depression. Scores ofanxiety and depression onHADS scale were reduced aswell as the scores of subscale ofMini-Mac

Schillani,62 2011,Italy

46 Advanced,inpatients

Intervention,two-weekfollow-up

Effect of ADs on depression,anxiety, and mentaladaptation to cancer

DSM-IV No information HADS d 100% with depression.Anxiety subscale scoreson HADS were decreased

(Continued)

Vol.

-No.

--

2014

5Patien

tCharacterization

inDepression

Studies

Table 1Continued

First Author,Ref Year,Country N

TerminologyUsed,a Setting Design Main Outcome

DiagnosticSystem

Assessment Methods

ResultsType ofInterview

Depression-SpecificQuestionnaires

QoLQuestionnaires

Shimizu,39

2007, Japan20 End-of-life,

inpatientsRetrospective,

three-monthfollow-up

Effect of AD treatment andintervention

DSM-IV Structuredclinical interview

d d 100% with depression. Noresponse achieved whentime was less than threeweeks

Warmenhoven,2

2012, TheNetherlands

64 Advanced,outpatients

Prospective Prevalence of depression DSM-IV Structured clinicalinterview SCAN

HADS d 3% with depression. Lowprevalence rates

Wilson,52 2007,Canada

381 Palliative, in-and outpatients

Prospective Prevalence and comorbidityof depression

DSM-IV Semistructuredinterview PRIMEMD modified

d SISCb 13% with depression, 10% withcomorbidities

Classification of depression independent of diagnostic system (n ¼ 44)

Chen,40 2011,China

90 Advanced, in-and outpatients

Prospective Effect of depressive symptomson survival

d d HADS-D d 22% with depression. Survival timewas shorter in depressedpatient group

Delgado-Guay,72

2009, USA216 Advanced, in-

and outpatientsRetrospective Relationship between

frequency of physicalsymptoms and depression

d d HADS ESAS 37% with depressed mood. Patientswith depressed mood had higherfrequency of symptoms

Du-Quiton,73

2010, USA84 Advanced, in-

and outpatientsProspective Measuring depression and

anxietyd d HADS d 25% with depression, 40%

with anxietyJacobson,97

2008, USA73 Advanced,

inpatientsCross-sectional Relationship between

depression andinterleukin-6

d d HDRSb d Low levels of depression.No correlation found

Kandasamy,41

2011, India50 Advanced,

inpatientsCross-sectional Association of SpWB with

distressing symptoms,anxiety, and depression

d d HADS MDASI,FACT-pal,FACT-G,FACIT-Sp

No percent presented. SpWB isrelated to physical andpsychological symptoms ofdistress

Lloyd-Williams,88

2009, UK87 Advanced,

outpatientsProspective, four-

and eight-weekfollow-up

Association of depressionwith survival

d d EDS Seven-item VRS 29% with depression.No association was found

Lo,83 2010,Canada

342 Advanced,outpatients

Prospective Whether age correlates toattachment securityand SpWB

d d BDI-II MSAS, FACIT-Sp 36% younger age with depression.Age was inversely associatedwith depression

Lo,27 2010,Canada

365 Metastatic,outpatients

Prospective Course and predictors ofdepressive symptoms

d d BDI-II BHS, FACIT-Sp,MSAS

35% with mild depression, 16%moderate to severe. Predictorsof depressive symptomswere identified

Maric,96 2010,Serbia

100 Advanced, in-and outpatients

Cross-sectional Frequency of symptoms ofdepression and anxiety

d d HDRSb SF-36, EORTCQLQ-C30

25% with mild, 19% moderate,and 2% severe depression

McMillan,89

2009, USA275 No information, in-

and outpatientsCross-sectional Relationship between

depressive symptomsand symptom distress

d d CES-D MSAS 38% with depression. Moderatecorrelation was found betweenCES-D scores and totalnumber of symptoms

Miller,84 2011,Canada

569 Advanced,outpatients

Cross-sectional Prevalence of depressionin males and females

d d BDI-II MSAS 22% with depression. Nosignificant genderdifferences were found

Moorey,64

2009, UK80 Advanced,

outpatientsRCT, 6-, 10-,

and 16-weekfollow-up

To reduce symptoms ofdepression and anxiety

d d HADS d 17% in CBT and 19% in TAUgroups with depression. Groupof CBT had lower anxiety scores

6Vol.

-No.

--

2014

Janberidze

etal.

Mystakidou,74

2008, Greece94 Advanced,

outpatientsCross-sectional Relationship between

preparatory grief withhopelessness, depression,and anxiety

d d HADS, PGAC BHS Severe 3%, possible 32%, andprobable 34% depression.Significant relation was found

Mystakidou,48

2009, Greece102 Terminal,

inpatientsCross-sectional Relationship between sleep

quality, depression, andhopelessness

d d BDI BHS 10% mild to moderate and 20%moderate to severe depression.Hopelessness was highlycorrelated with depression

Mystakidou,54

2007, Greece102 Advanced,

inpatientsCross-sectional Prevalence of clinical

characteristics forhastened death

d d BDI BHS 10% mild to moderate and 20%moderate to severe depression.Strong association foundbetween desire for death,depression, and hopelessness

Mystakidou,49

2007, Greece102 Terminal,

inpatientsCross-sectional Sleep quality and its

relationship with pain,depression, andhopelessness

d d BDI BHS, MDASI No percent presented.Hopelessness, interferenceof pain, and opioids seemed toinfluence patients’ sleep quality

Mystakidou,75

2009, Greece94 Multiple terms,

outpatientsCross-sectional Relationship of hopelessness,

anxiety, distress, andpreparatory grief

d d HADS, PGAC BHS 34% with probable and 3% withsevere depression. Depression,preparatory grief, and age werepredictors of hopelessness

Mystakidou,55

2008, Greece100 Multiple terms, in-

and outpatientsCross-sectional Relationship between

posttraumatic growthand psychological distress

d d HADS d No percent presented. Negativeassociation was found betweenPTGI-II (‘‘new possibilities’’)and HADS-depression

Mystakidou,85

2008, Greece102 Multiple terms,

inpatientsCross-sectional Relationship of QOL

and psychologicalmorbidities

d d BDI BHS, SF-12 No percent presented. Associationwas found between gender, PS,opioids, and depression

Mystakidou,56

2007, Greece82 Multiple terms,

inpatientsCross-sectional Relationship between

depression, hopelessness,cognitive status, pain,and spirituality

d d BDI BHS 20% with moderate to severe and12% with severe depression.Association found betweenhopelessness, depression, andcognitive status

Neron,76 2007,Canada

49 Multiple terms,outpatients

Prospective, threetime follow-up

Prevalence and incidenceof disease-specificdepression

d d HADS, MADRSb d 49% depressed with MADRSand 18% with HADS

O’Connor,42

2010, Australia266 Palliative, in-

and outpatientsProspective Prevalence and

predictors of anxietyand depression

d d HADS d 46% with possible and 20%with probable depression.Past anxiety in the familypredicted probable depression

Okuyama,77

2009, Japan60 Advanced,

outpatientsRCT If patients reluctance

to discuss psychologicaldistress was associated withunder recognition ofdepression by physicians

d d HADS, single-item depressionscreenerb

d 30% with depression.No association was found

Pirl,60 2008,USA

52 Advanced, noinformation

RCT, six- and12-monthfollow-up

Development of depressivesymptoms and fatigue

d d BDI d 10e16% with depression.No difference was found

Pirl,43 2008,USA

43 Advanced,outpatients

Prospective,two-monthfollow-up

Association betweendepression and survival

d d HADS d 23% with depression. Patients withdepression had poor survival

Popa-Velea,65

2010, Romania60 Advanced, no

informationRCT, follow-up

every twomonths

Effect of personalizedapproach on psychiatricmorbidity

d d HADS SF-36 No percent presented.QOL, depression, andanxiety scores were lowerin personalized care group

(Continued)

Vol.

-No.

--

2014

7Patien

tCharacterization

inDepression

Studies

Table 1Continued

First Author,Ref Year,Country N

TerminologyUsed,a Setting Design Main Outcome

DiagnosticSystem

Assessment Methods

ResultsType ofInterview

Depression-SpecificQuestionnaires

QoLQuestionnaires

Rodin,86 2009,Canada

406 Metastatic,outpatients

Cross-sectional Relationship betweendepression andhopelessness

d d BDI-II BHS,FACIT-Sp-12,MSAS

No percent presented.Depression andhopelessness weremutually reinforcingbut distinct constructs

Rodin,87 2007,Canada

326 Multiple terms,outpatients

Cross-sectional Relationship between diseaseburden and depression

d d BDI-II MSAS 23% with depression.Attachment securitybuffered the effect ofdisease-related factorson depression

Rolke,78 2010,Norway

479 No information,no information

Intervention,three-weekfollow-up

Change of QOL, depression,and anxiety

d d HADS EORTCQLQ-C30,EORTCQLQ-LC 13

No percent presented. Depressedpatients experienced improvedsocial functioning and appetite

Salvo,70 2012,Canada

1439 Advanced, in-and outpatients

Retrospective Prevalence of depressionand anxiety by ESAS

d d d ESAS 55% mild and 11% with severedepression

Sela,90 2007,Canada

132 Palliative,outpatients

Prospective Prevalence of depressionby BZSRDS

d d BZSRDS, VAS(single question)

d 32% moderate and 19% with severedepression

Sephton,91

2009, USA125 Metastatic, no

informationCross-sectional Depression, stress, and

immune functionresponses

d d CES-D d 29% women with depression.Women reported moredepressive symptoms

Slovacek,92

2009, Czechrepublic

41 Palliative,inpatients

Cross-sectional Prevalence of depressionsymptoms and levelsof HRQOL

d d ZSRDS EQ-5D 12% with severe depression.Low level of HRQOL

Slovacek,93

2009, Czechrepublic

30 Palliative,inpatients

Cross-sectional Incidence and prevalenceof depressive symptoms

d d ZSRDS d 30% with severe depression

Slovacek,94

2009, Czechrepublic

64 Metastatic,inpatients

Cross-sectional Screening of depression infemales

d d ZSRDS d 12% with severe depression

Steel,57 2007,USA

28 Multiple,outpatients

RCT, three monthfollow-up

To improve HRQOL anddecrease psychologicaldistress

d d CES-D d No percent presented. Interventiongroup reported reductionin depressive symptoms

Stockler,63

2007, Australia189 Advanced, no

informationRCT, four- and

eight-weekfollow-up

Effect of ADs d d HADS, CES-D,SPHERE

FACT-G, UBQ-C,Pt DATA Form,SQLIb

25% consistent with clinicaldepression. No effectfound

Teunissen,79

2007, TheNetherlands

79 Multiple terms,inpatients

Prospective Depressed mood, anxiety, andphysical symptoms

d d HADS, single-itemdepressionscreener

ESAS 56% depressed mood by HADS,37% by single question, and 44%depressed by ESAS

Tian,95 2009,China

233 Advanced, noinformation

Prospective Association between coping,nutrition, psychologicalstatus, and PS

d d DSI d No percent presented. Depression,low scores of confrontation, andHB may be risk factorsof poor PS

Tsai,80 2007,Taiwan

77 Terminal,inpatients

Cross-sectional Examine levels of fatigueand its related factors

d d HADS Modified SDS 47% with depression. Fatigue wasassociated with depression,anxiety, overall symptomdistress, and PS

8Vol.

-No.

--

2014

Janberidze

etal.

Turner,81

2007, UK191 Palliative,

outpatientsProspective,

two-monthfollow-up

Psychological distress andpatients’ concerns

d d HADS Concernschecklist,EORTCQLQ-C30

21% with depression. All agegroups had similar concerns andlevels of depression and anxiety

Uchida,50

2011, Japan85 Advanced,

outpatientsRCT Frequency of unmet needs d d HADS EORTC

QLQ-C30,SCNS-SF34

14% with depression. Medianof 10 unmet needs identifiedincluding psychological domain

Ugalde,51 2012,Australia

108 Not curableand not operableoutpatients

Cross-sectional Levels of anxiety, depression,global distress, and unmetneeds

d d HADS NA-ALCP, DT 19% with clinical andsubclinical depression.40% reported distress.Unmet needs wererelated to medicalcommunication

Waller,66 2011,Australia

219 Multiple terms,outpatients

Intervention Impact of using guidelinesand assessment tools onanxiety and depression

d d HADS EORTC-QLQ-C30,SCNS-SF34

11% with depression.Moderate to high needsacross all domainswere found

QOL ¼ quality of life; DSM-IV ¼ Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition; DSM-III-R ¼ Diagnostic and Statistical Manual of Mental Disorders, Third Edition Revised; HADS ¼ HospitalAnxiety and Depression Scale;71 POMS ¼ Profile of Mood States;122 ADs ¼ antidepressants; SCID-I ¼ Structured Clinical Interview for the DSM-IV Axis I Disorders; HDRS ¼ Hamilton Depression RatingScale;98 MDASI ¼ M.D. Anderson Symptom Inventory;124 BRC ¼ Pargament’s Brief RCope;125 ICG-R ¼ Inventory of Complicated Grief-Revised;126 MQOL ¼ McGill Quality of Life Questionnaire;127

MDD ¼ major depressive disorder; PHQ-9 ¼ Patient Health Questionnaire-9;114 PANAS ¼ Positive and Negative Affect Schedule;128 VAS ¼ Visual Analogue Scale;129 ICD-10 ¼ International Statistical Classi-fication of Diseases and Related Health Problems, Tenth Revision; CIS-R ¼ Revised Clinical Interview Schedule;154 EDS ¼ Edinburgh Depression Scale;130 ESAS ¼ Edmonton Symptom Assessment System;131 PRIMEMD ¼ Primary Care Evaluation of Mental Disorders;155 SISC ¼ Structured Interview of Symptoms and Concerns;133 FACT-pal ¼ Functional Assessment of Cancer Therapy-Palliative Care;135 FACT-G ¼ Func-tional Assessment of Cancer Therapy Scale-G;117 FACIT-Sp ¼ Functional Assessment of Chronic Illness Therapy Spiritual Well-Being Scale;134 SpWB ¼ spiritual well-being; VRS ¼ Visual Rating Scale; BDI ¼Beck Depression Inventory;82 BDI-II ¼ Beck Depression Inventory-II;136 MSAS ¼Memorial Symptom Assessment Scale;137 BHS ¼ Beck Hopelessness Scale;123 SF-36 ¼ Short Form Health Survey-36;138 EORTCQLQ-C30 ¼ The European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30116 CES-D ¼ Clinical Epidemiological Scale-Depression;139 RCT ¼ randomized controlledtrial; CBT ¼ cognitive behavioral therapy; TAU ¼ treatment as usual; PGAC ¼ Preparatory Grief in Advanced Cancer Patients;140 PTG ¼ posttraumatic growth; SF-12 ¼ Short Form Health Survey-12;141 PS ¼performance status; MADRS ¼ Montgomery-�Asberg Depression Rating Scale;142 FACIT-Sp-12 ¼ Functional Assessment of Chronic Illness Therapy Spiritual Well-Being Scale-12;117 EORTC QLQ-LC 13 ¼ TheEuropean Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Lung Cancer-13;143 BZSRDS ¼ Brief Zung Self-Rating Depression Scale;144 ZSRDS ¼ Zung Self-Rating DepressionScale;145 EQ-5D ¼ European Quality of Life Questionnaire;146 HRQoL ¼ health-related quality of life; SPHERE ¼ Somatic and Psychological Health Report;147 UBQ-C ¼ Utility-Based Questionnaire;148 PtDATA Form ¼ questionnaire designed for the study assessing 17 symptoms and seven aspects of well-being;63 SQLI ¼ Spritzer’s Quality of Life Index;149 DSI ¼ Depression Status Inventory;150 Modified SDS ¼Modified Symptom Distress Scale;151 SCNS-SF34 ¼ Short-Form Supportive Care Needs Survey Questionnaire;152 NA-ALCP ¼ Needs Assessment for Advanced Lung Cancer Patients;153 DT ¼ DistressThermometer.132aTerminology used to describe the sample of advanced cancer/palliative care population. Wording used throughout the articles, especially in title, introduction, and aims.bInterview conducted by health care provider.

Vol.

-No.

--

2014

9Patien

tCharacterization

inDepression

Studies

Table 2Reporting of Patient Characteristics in the 59 Included Studies

Characteristics

No Informationin the Publication,

N (%)Information in thePublication, N (%)

Assessed in the Study,But Not Reported in the

Publication, N (%)

Sociodemographic variablesAgea 1 (2) 55 (93) 3 (5)Gendera 1 (2) 53 (90) 5 (8)Ethnicitya 47 (80) 9 (15) 3 (5)Marital status 28 (48) 29 (49) 2 (3)Educationa 29 (49) 25 (42) 5 (9)Religion 52 (88) 6 (10) 1 (2)

Medical information (disease-specific variables)Principal diagnosisa 2 (3) 57 (97)Time since diagnosisa 43 (73) 16 (27)Stage of cancer diseasea 15 (25) 44 (75)Metastases 36 (61) 23 (39)Site of metastasesa 53 (90) 6 (10)Additional diagnoses (comorbidity)a 51 (86) 8 (14)Expected survival 51 (86) 8 (14)Real survival 44 (75) 15 (25)Weight lossa 57 (97) 2 (3)Performance statusa 19 (32) 38 (65) 2 (3)Cognitive functiona 28 (48) 23 (38) 8 (14)

TreatmentPresent anticancer treatmenta 24 (41) 35 (59)Medicationa 43 (73) Opioids 13 (22);

neuroleptics 3 (5)Caregiver issues

Living situationa 49 (83) 10 (17)Social network 53 (90) 6 (10)

SettingPlace of carea 5 (9) 54 (91)Provision of carea 10 (17) 49 (83)

aVariables included in the European Association for Palliative Care Basic Dataset.37

10 Vol. - No. - - 2014Janberidze et al.

(44%) studies, ‘‘palliative’’ in six (10%)studies, ‘‘metastatic’’ in five (8%),‘‘terminallyill’’ in six (10%), ‘‘end-of-life’’ in two studies(3%), and ‘‘not curable’’ in two (3%).

The following eight of the 17 variablesincluded in the EAPC Basic Dataset37 weremost often reported: age (93%), gender(90%), principal cancer diagnosis (97%),stage of the cancer disease (75%), perfor-mance status (65%), anticancer treatment(59%), place of care (91%), and provision ofcare (83%). The remaining nine variables ofthe EAPC Basic Dataset were less often re-ported: education (42%), cognitive function(38%), time since diagnosis (27%), medica-tion (opioids in 22% and neuroleptics in5%), living situation (17%), ethnicity (15%),additional diagnoses (comorbidities) (14%),site of metastasis (10%), and weight loss(3%). Expected survival was recorded in eight(14%) of the studies, and real survival was re-corded in 15 (25%) of the studies (Table 2).

Depression-Related VariablesClinical information related to depression

was rarely reported (Table 3). The durationof the present depressive episode was reportedin two studies (3%).38,39 Information on previ-ous depressive episodes and psychiatric historywas provided in seven (12%)39e45 and 10(17%)39,41,45e52 studies, respectively. Historyof alcohol and/or drug abuse was reportedin eight (13%)41,48,49,53e57 studies. The use ofantidepressant medication was reported in 10(17%) studies,27,39,44,52,58e63 whereas differenttypes of psychotherapies were reported in four(7%) studies.57,64e66

Assessment and Classification of DepressionDepression was classified according to a

diagnostic system in 15 (25%) studies(Table 1). The DSM-IV criteria were used in13 studies,2,38,39,44,46,47,52,58,59,61,62,67,68 theICD-10 in one,45 whereas one study53 used

Table 3Reporting of Depression-Specific Clinical Information in the 59 Included Studies

CharacteristicsNo Information,

N (%)Information,

N (%)

Assessed in theStudy, But Not Reportedin the Publication, N (%)

Use of antidepressants 49 (83) 10 (17) dDifferent types of psychotherapy 55 (93) 4 (7) dPsychiatric history 40 (68) 10 (17) 9 (15)Duration of present depressive episode 57 (97) 2 (3) dPrevious depressive episodes 48 (81) 7 (12) 4 (7)History of alcohol and/or drug abuse 50 (85) 8 (13) 1 (2)

Vol. - No. - - 2014 11Patient Characterization in Depression Studies

the revised DSM-III criteria.69 Different assess-ment methods were applied in these 15studies: structured clinical interviews wereused in nine,2,38,39,45e47,53,61,67 semistructuredinterviews were used in three studies,44,52,68

whereas the remaining three studies58,59,62

gave no information on the assessmentmethods used for establishing a diagnosis ofdepression.

Forty-four (75%) of the studies did not use adiagnostic system for classifying depression(Table 1). However, all but one study70 hadused depression-specific assessment tools.Among these, patient-reported tools were mostcommonly used: the Hospital Anxiety andDepression Scale71 was used in 21 (48%)studies40e43,50,51,55,63e66,72e81 and different ver-sions of the Beck Depression Inventory82 in 11(25%),27,48,49,54,56,60,83e87 whereas nine studiesused other depression-specificquestionnaires.57,88e95 Two studies96,97 used theinterviewer-administered Hamilton DepressionRating Scale.98 In most studies, more than oneassessment method was used.

DiscussionThis systematic literature review included 59

full-text articles on depression in patients withadvanced cancer published during the five-yearperiod 2007e2011. The description of the pa-tient samples and how depression was assessedand classified varied considerably across studies.Our findings confirm that the reporting ofimportant characteristics of advanced cancer pa-tient samples is unsystematic. Age, gender, andcancer diagnosis were the most frequently re-ported sociodemographicandmedical variables,confirming the previous findings by Sigurdar-dottir et al.24 As a novel finding, clinically impor-tant information related to depression, such aspsychiatric history, duration of the present

depressive episode, and treatment (drug treat-ment with antidepressants and/or psychologicaltherapy), was rarely reported. In 25% of thestudies, depression was classified according to avalidated diagnostic system. This result is similar(23%) to the finding in an earlier review ofstudies published before 2007.13

The prevalence rate estimates of depressionin the included publications varied between2% and 56%, except for six studies39,58,59,61,62,67

having depression as an inclusion criterion andthus a prevalence rate of 100%. This variabilityconfirms previous findings13,99,100 and under-lines the need for standardization of how tocharacterize patient samples and of how toassess and classify depression in advanced can-cer patients. The European Clinical Guidelinesfor the Management of Depression in PalliativeCare stress the importance of assessingdepression-related variables such as use of anti-depressants, psychiatric history, and duration ofthedepressive episodewhendiagnosingdepres-sion.101 Information regarding medication is acrucial part of anymedical history, and informa-tion on previous and ongoing treatment withantidepressants is part of a complete descrip-tion of a patient with depressive symptoms.However, only 11 (17%) of the studies includedin the present work reported information abouttreatment with antidepressants. A history ofdepressive disorder (e.g., major depressive dis-order or dysthymia) and long duration of thedepressive episode heighten the risk for devel-oping new or subsequent depressive epi-sodes.102,103 Despite this knowledge, data onthese variables were rarely reported in the re-viewed publications. If research is to influenceclinical practice, the assessments performed inresearch should mirror best clinical practice.Thus, one might question the external validityof many of the published studies on depressionin advanced cancer patients.

12 Vol. - No. - - 2014Janberidze et al.

Classificationofdepression is basedoncriteriaof diagnostic systems such as DSM-V.5 To diag-nose adepressivedisorder in anadvanced cancerpatient may be challenging as these patientscommonly experience similar somatic symptomsfrom the cancerdisease itself. Itmay beproblem-atic to decide which of the somatic symptomsidentified in the DSM-V criteria are attributableto depression and which are because of cancerdisease.8 Possible confoundingeffects of the can-cer disease should thus be taken into consider-ation when investigating depression in patientswith advanced cancer. This can be done byincludingmeasures of disease load, for example,time to death, performance status, and how thecancer disease influences symptoms of depres-sion, particularly the somatic symptoms.104

Most investigated studies did in fact report in-formation about the cancer disease (97%) andits stage (75%).However, despite thewidespreaduse of the tumor, node, metastasis classificationsystem in oncology in general,105 it might notbe expected that staging information can bepresented similarly for all different types ofcancer. Anticancer treatment should be takeninto consideration in studies including cancerpatients, especially with advanced disease. How-ever, information on present anticancer treat-ment was only reported in 59% of the studies.The terminology related to describing the sam-ple as palliative care patients vs. patients withadvanced disease differed across studies. Forexample, the term ‘‘palliative’’ was used in 10%of the studies, whereas ‘‘advanced’’ was used in44%. This may reflect different existing views ofwhat ‘‘palliative care’’ is. In some settings, pallia-tive care is viewed as only the last few weeksbefore death, whereas others define palliativecare from the day the patient is beyond cure.Integration of palliative care into oncology hasin recent years been in focus.106e108

Furthermore, it is important to take into ac-count information on other symptoms, asmost patients with advanced cancer report mul-tiple coexisting distressing symptoms72 thathave been shown to be interdependent.109

Even if the direction of causality of these rela-tionships is unclear,110 the presence of symp-toms common in advanced cancer patients,such as uncontrolled pain or cachexia,111,112

may influence the presence of depression. Assuch, information about pain treatment is alsoimportant to report. However, only 22% of

the studies provided information about treat-ment with, for example, opioids.Ethnicity may be of clinical relevance, for

example, in a patient with depression whohas moved from one country to another, andit is well known that depressive feelings andsymptoms are reported in different ways indifferent cultures. However, there is no stan-dardized manner in how to address this, asfar as we know, other than acknowledging cul-ture as a relevant factor.Wasteson et al.13 demonstrated that depres-

sion was frequently not classified according to adiagnostic system and that more than 100different assessment methods were used. Thus,our systematic literature review confirms thatthe unsystematic use of assessment methods fordiagnosing depression in advanced cancer pa-tients continues. Among the included studies,only 15 used structured or semistructured inter-views based on an approved classification systemfor diagnosing depression. A structured clinicalinterview is currently the reference standard fordiagnosing depression in clinical practice.113

However, a structured interview is timeconsuming and may be difficult to conduct infrail and fatiguedcancerpatientswithapoorper-formance status. Several self-reporting screeningtools have been designed to assess depression,butmost often they assess the intensity of variousdepressive symptoms and are insufficient fordiagnostic purposes.When deciding on which assessment method

to use, a clear understanding of what is to be as-sessed is paramount: depressive disorder or psy-chological distress/depressive symptoms. Ourrecommendation is that for the assessment ofdepressive disorders, a self-report tool based onthe standardized diagnostic criteria of the DSMshould be applied, when a structured clinicalinterview cannot be undertaken. The PatientHealth Questionnaire-9 is such a tool,114 whichhas been validated in patients with cancer.115 Us-ing the same tool across depression studies canfacilitate more accurate comparisons.For the assessment of psychological distress/

depressive symptoms, several more generalassessment tools are available, either in theform of general symptom assessment tool,quality-of-life tools, and symptom-specific tools.Examples of frequently used tools are the Euro-pean Organization for Research and Treatmentof Cancer Quality of Life Questionnaire-C30,116

Vol. - No. - - 2014 13Patient Characterization in Depression Studies

Functional Assessment of Cancer TherapyScale,117 and the Distress Thermometer.118

Among specific depression tools, the HospitalAnxiety and Depression Scale for self-report ofboth depression and anxiety71 and Beck Depres-sion Inventory82 are well-known instruments.Furthermore, the new Patient-Reported Out-comes Measurement Information System(PROMIS)-Depression instrument from thePROMIS119 is currently being developed.120,121

This tool is intended for use by computers anduses a computer-adaptive testing algorithm aspart of the software, which makes it easier totailor the questions to the individual patient.However, both the methods and tools itselfneed to be validated against structured clinicalinterviews, in cancer and other patient groups.

A potential limitation in all reviews is whetherall relevant publications are identified. In thissystematic literature review, we only includedpublications from studies investigating advancedcancer patients and explicitly defining depres-sion as a primary outcome. This was donebecause we hypothesized that the descriptionof the patient samples and the assessment ofdepression would be as optimal as possible in aclearly defined population. An extension of theinclusion criteria to cover cancer patients in gen-eral and/or depressive symptoms in a broadersense may of course have given other results. Apredefined relatively comprehensive checklistwas used for data extraction; however, it cannever be ruled out that the use of an even moreextensive checklistmay haveprovidedadditionalinformation. The limitations of the selectionprocedure regarding period and language implythat we might have missed publications of inter-est. The search covered the five-year period2007e2011. This was chosen because an exten-sive systematic literature review by Wastesonet al.13 on assessment and classification ofdepression in palliative care covered the perioduntil 2007. The search was confined to studiespublished in English language according to thelanguage skills of our research group andbecause this is a common procedure for mostreviews. Our selection of the well-known majordatabases representing various disciplines, andthe use of recognized search strategy to identifyrelevant literature in advanced cancer wasapplied,13 make us think that the restriction tothe English language is not a major study limita-tion. Despite the limitations, we think that the

present review presents valuable new informa-tion onhow patients are characterized in clinicalstudies having depression as a primary outcome.

Further work toward a consensus on how tocharacterize study samples in advanced cancerpatients with depression should in our opinioninclude patients’ sociodemographic character-istics, disease-related information, and infor-mation on depression-specific risk factors. Inaddition, information on common symptomsand conditions such as pain and depressionshould be reported. Our general recommen-dation for clinical studies investigatingadvanced cancer samples is to include theEAPC Basic Dataset37 as well as informationabout the cancer disease and anticancer treat-ment. In addition, for depression-specificstudies, we recommend to assess duration ofthe depressive episode, psychiatric history,and ongoing treatment for depression as aminimum of depression-related variables.

Stakeholders such as the European Associa-tion for Palliative Care Research Network(EAPC RN) in collaboration with core journalsand other relevant bodies could bring thiswork forward by agreeing on a common setof core variables that precisely describe a studypopulation. Furthermore, editors of medicaljournals could be requested to use the agreedcommon data set as a checklist.

ConclusionsOur systematic literature review demon-

strates unsystematic and inconsistent reportingof core sociodemographic and medical samplecharacteristics in populations with advancedcancer and depression. The major findingwas the lack of reporting of importantdepression-related variables in studies investi-gating depression in advanced cancer patients.Validated diagnostic tools for depression wererarely used. Assessment methods for depres-sion continued to vary greatly. There is anobvious need for a more stringent character-ization of study populations to increase thegeneralizability of results from studies ondepression in patients with advanced cancer.We hope our results will inspire journals andother central stakeholders to produce recom-mendations for sample descriptions and assess-ment methods in future studies.

14 Vol. - No. - - 2014Janberidze et al.

Disclosures and AcknowledgmentsThis article is part of the European Intersec-

torial and Multidisciplinary Palliative CareResearch Training (EURO IMPACT) project.EURO IMPACT is funded by the EuropeanUnion Seventh Framework Programme (FP7/2007e2013, under grant agreement number264697). EURO IMPACT aims to develop amultidisciplinary, multiprofessional, and inter-sectorial educational and research trainingframework for palliative care research in Eu-rope. EURO IMPACT is coordinated by ProfLuc Deliens and Prof Lieve Van den Block ofthe End-of-Life Care Research Group, GhentUniversity and Vrije Universiteit Brussel, Brus-sels, Belgium. Other partners are VU Univer-sity Medical Center, EMGO Institute forHealth and Care Research, Amsterdam, TheNetherlands; King’s College London, CicelySaunders Institute, London, Cicely SaundersInternational, London, and International Ob-servatory on End-of-Life Care, Lancaster Uni-versity, Lancaster, United Kingdom; EAPCRN, Trondheim, Norway; Fondazione IRCCSIstituto Nazionale dei Tumori, Milan, Italy,and Cancer Research and Prevention Institute,Florence, Italy; EUGMS European Union Geri-atric Medicine Society, Geneva, Switzerland;Springer Science and Business Media, Houten,The Netherlands.

The authors declare no conflicts of interest.In recognition of the collaborative nature of

EURO IMPACT, the authors thank thefollowing EURO IMPACT members for theircontribution, namely: Van den Block Lieve(End-of-Life Care Research Group, Ghent Uni-versity and Vrije Universiteit Brussel, Brussels,Belgium), De Groote Zeger (End-of-Life CareResearch Group, Ghent University and VrijeUniversiteit Brussel, Brussels, Belgium), Brear-ley Sarah (International Observatory on End-of-Life Care, Lancaster University, Lancaster,United Kingdom), Caraceni Augusto (EAPCRN, Trondheim, Norway; Fondazione IRCCS Is-tituto Nazionale dei Tumori, Milan, Italy), Co-hen Joachim (End-of-Life Care ResearchGroup, Ghent University and Vrije UniversiteitBrussel, Brussels, Belgium), Francke Anneke(VUUniversityMedical Center, EMGOInstitutefor Health andCare Research, Amsterdam, TheNetherlands),HardingRichard (King’s CollegeLondon, Cicely Saunders Institute, London;

Cicely Saunders International, London), Hig-ginson Irene J (King’s College London, CicelySaunders Institute, London; Cicely SaundersInternational, London), Linden Karen(Springer Science andBusinessMedia,Houten,The Netherlands), Miccinesi Guido (CancerResearch and Prevention Institute, Florence,Italy),Onwuteaka-PhilipsenBregje (VUUniver-sity Medical Center, EMGO Institute for Healthand Care Research, Amsterdam, TheNetherlands), Pardon Koen (End-of-Life CareResearch Group, Ghent University and VrijeUniversiteit Brussel, Brussels, Belgium), Pas-man Roeline (VU University Medical Center,EMGO Institute for Health and Care Research,Amsterdam, The Netherlands), Pautex Sophie(EUGMS European Union Geriatric MedicineSociety, Geneva, Switzerland), Payne Sheila(International Observatory on End-of-LifeCare, Lancaster University, Lancaster, UnitedKingdom), and Deliens Luc (End-of-Life CareResearch Group, Ghent University and VrijeUniversiteit Brussel, Brussels, Belgium; VUUni-versity Medical Center, EMGO Institute forHealth and Care Research, Amsterdam, TheNetherlands). The authors thank Ingrid I. Ri-phagen, Unit for Applied Clinical Research,Faculty of Medicine, Norwegian University ofScience and Technology, for her help withsearching the computerized bibliographicdatabases.

References1. Chochinov HM. Depression in cancer patients.

Lancet Oncol 2001;2:499e505.

2. Warmenhoven F, van Rijswijk E, van Weel C,Prins J, Vissers K. Low prevalence of depressive dis-order in ambulatory advanced cancer patients usingthe Schedules for Clinical Assessment in Neuropsy-chiatry (SCAN 2.1). J Affect Disord 2012;136:1209e1211.

3. Meyer HA, Sinnott C, Seed PT. Depressivesymptoms in advanced cancer. Part 2. Depressionover time; the role of the palliative care profes-sional. Palliat Med 2003;17:604e607.

4. Burchett H, Umoquit M, Dobrow M. How dowe know when research from one setting can be use-ful in another? A review of external validity, applica-bility and transferability frameworks. J Health ServRes Policy 2011;16:238e244.

5. American Psychiatric Association, DSM-5 TaskForce. Diagnostic and statistical manual of mental

Vol. - No. - - 2014 15Patient Characterization in Depression Studies

disorders: DSM-5, 5th ed. Washington: AmericanPsychiatric Association, 2013.

6. World Health Organisation. The ICD-10 classi-fication of mental and behavioral disorders: diag-nostic criteria for research. Geneva: World HealthOrganisation, 1993.

7. Trask PC. Assessment of depression in cancerpatients. J Natl Cancer Inst Monogr 2004;80e92.

8. Rayner L, Lee W, Price A, et al. The clinicalepidemiology of depression in palliative care andthe predictive value of somatic symptoms: cross-sectional survey with four-week follow-up. PalliatMed 2011;25:229e241.

9. Endicott J. Measurement of depression in pa-tients with cancer. Cancer 1984;53:2243e2249.

10. Kroenke K, Zhong X, Theobald D, et al. So-matic symptoms in patients with cancer experi-encing pain or depression: prevalence, disability,and health care use. Arch Intern Med 2010;170:1686e1694.

11. Akechi T, Nakano T, Akizuki N, et al. Somaticsymptoms for diagnosing major depression in can-cer patients. Psychosomatics 2003;44:244e248.

12. Bukberg J, Penman D, Holland JC. Depressionin hospitalized cancer patients. Psychosom Med1984;46:199e212.

13. Wasteson E, Brenne E, Higginson IJ, et al.Depression assessment and classification in pallia-tive cancer patients: a systematic literature review.Palliat Med 2009;23:739e753.

14. Kaasa S, Loge JH, Fayers P, et al. Symptomassessment in palliative care: a need for interna-tional collaboration. J Clin Oncol 2008;26:3867e3873.

15. Kaasa S, Torvik K, Cherny N, Hanks G, deConno F. Patient demographics and centre descrip-tion in European palliative care units. Palliat Med2007;21:15e22.

16. Clark D, ten Have H, Janssens R. Commonthreads? Palliative care service developments inseven European countries. Palliat Med 2000;14:479e490.

17. Laugsand EA, Jakobsen G, Kaasa S, Klepstad P.Inadequate symptom control in advanced cancerpatients across Europe. Support Care Cancer 2011;19:2005e2014.

18. Borgsteede SD, Deliens L, Francke AL, et al.Defining the patient population: one of the prob-lems for palliative care research. Palliat Med 2006;20:63e68.

19. Currow DC, Wheeler JL, Glare PA, Kaasa S,Abernethy AP. A framework for generalizability inpalliative care. J Pain Symptom Manage 2009;37:373e386.

20. Sigurdardottir KR, Haugen DF, van derRijt CC, et al. Clinical priorities, barriers and solu-tions in end-of-life cancer care research across

Europe. Report from a workshop. Eur J Cancer2010;46:1815e1822.

21. Schulz KF, Altman DG, Moher D, CONSORTGroup. CONSORT 2010 statement: updated guide-lines for reporting parallel group randomized trials.Ann Intern Med 2010;152:726e732.

22. von Elm E, Altman DG, Egger M, et al. TheStrengthening the Reporting of ObservationalStudies in Epidemiology (STROBE) statement:guidelines for reporting observational studies.J Clin Epidemiol 2008;61:344e349.

23. Currow DC, Tieman JJ, Greene A, et al.Refining a checklist for reporting patient popula-tions and service characteristics in hospice and palli-ative care research. J Pain Symptom Manage 2012;43:902e910.

24. Sigurdardottir KR, Oldervoll L, Hjermstad MJ,et al. How are palliative care cancer populationscharacterized in randomized controlled trials? Aliterature review. J Pain Symptom Manage 2013.http://dx.doi.org/10.1016/j.jpainsymman.2013.06.005. (Epub ahead of print).

25. Bottomley A. Depression in cancer patients: aliterature review. Eur J Cancer Care (Engl) 1998;7:181e191.

26. Kathol RG, Mutgi A, Williams J, Clamon G,Noyes R Jr. Diagnosis of major depression in cancerpatients according to four sets of criteria. Am J Psy-chiatry 1990;147:1021e1024.

27. Lo C, Zimmermann C, Rydall A, et al. Longitu-dinal study of depressive symptoms in patients withmetastatic gastrointestinal and lung cancer. J ClinOncol 2010;28:3084e3089.

28. Jia L, Jiang SM, Shang YY, et al. Investigation ofthe incidence of pancreatic cancer-related depres-sion and its relationship with the quality of life ofpatients. Digestion 2010;82:4e9.

29. Hopwood P, Stephens RJ. Depression in pa-tients with lung cancer: prevalence and risk factorsderived from quality-of-life data. J Clin Oncol2000;18:893e903.

30. Zabora J, BrintzenhofeSzoc K, Curbow B,Hooker C, Piantadosi S. The prevalence of psycho-logical distress by cancer site. Psychooncology2001;10:19e28.

31. Love AW, Grabsch B, Clarke DM, Bloch S,Kissane DW. Screening for depression in womenwith metastatic breast cancer: a comparison of theBeck Depression Inventory Short Form and theHospital Anxiety and Depression Scale. Aust N Z JPsychiatry 2004;38:526e531.

32. Burcusa SL, Iacono WG. Risk for recurrence indepression. Clin Psychol Rev 2007;27:959e985.

33. Webber K, Davies AN. An observational studyto determine the prevalence of alcohol use disor-ders in advanced cancer patients. Palliat Med2012;26:360e367.

16 Vol. - No. - - 2014Janberidze et al.

34. Judd LL. The clinical course of unipolar majordepressive disorders. Arch Gen Psychiatry 1997;54:989e991.

35. Spijker J, de Graaf R, Bijl RV, et al. Duration ofmajor depressive episodes in the general popula-tion: results from The Netherlands Mental HealthSurvey and Incidence Study (NEMESIS). Br J Psy-chiatry 2002;181:208e213.

36. NICE.Depression in adults with a chronic healthproblem: treatment andmanagement. 2009. Availableat: http://www.nice.org.uk/nicemedia/pdf/CG91NICEGuideline.pdf. Accessed August 30, 2013.

37. Sigurdardottir KR, Kaasa S, Rosland JH, et al.The European Association for Palliative Care basicdataset to describe a palliative care cancer popula-tion: Results from an international Delphi process.Palliat Med 2014. http://dx.doi.org/10.1177/0269216314521264. (Epub ahead of print).

38. Inagaki M, Yoshikawa E, Kobayakawa M, et al.Regional cerebral glucose metabolism in patientswith secondary depressive episodes after fatalpancreatic cancer diagnosis. J Affect Disord 2007;99:231e236.

39. Shimizu K, Akechi T, Shimamoto M, et al. Canpsychiatric intervention improve major depressionin very near end-of-life cancer patients? Palliat Sup-port Care 2007;5:3e9.

40. Chen ML, Chen MC, Yu CT. Depressive symp-toms during the first chemotherapy cycle predictmortality in patients with advanced non-small celllung cancer. Support Care Cancer 2011;19:1705e1711.

41. Kandasamy A, Chaturvedi SK, Desai G. Spiritu-ality, distress, depression, anxiety, and quality of lifein patients with advanced cancer. Indian J Cancer2011;48:55e59.

42. O’Connor M, White K, Kristjanson LJ,Cousins K, Wilkes L. The prevalence of anxietyand depression in palliative care patients with can-cer in Western Australia and New South Wales.Med J Aust 2010;193:S44eS47.

43. Pirl WF, Temel JS, Billings A, et al. Depressionafter diagnosis of advanced non-small cell lung can-cer and survival: a pilot study. Psychosomatics 2008;49:218e224.

44. Rabkin JG, McElhiney M, Moran P, Acree M,Folkman S. Depression, distress and positive moodin late-stage cancer: a longitudinal study. Psychoon-cology 2009;18:79e86.

45. Reeve J, Lloyd-Williams M, Dowrick C. Depres-sion in terminal illness: the need for primary care-specific research. Fam Pract 2007;24:263e268.

46. Jacobsen JC, Zhang B, Block SD,Maciejewski PK, Prigerson HG. Distinguishingsymptoms of grief and depression in a cohort ofadvanced cancer patients. Death Stud 2010;34:257e273.

47. Lichtenthal WG, Nilsson M, Zhang B, et al. Dorates of mental disorders and existential distressamong advanced stage cancer patients increase asdeath approaches? Psychooncology 2009;18:50e61.

48. Mystakidou K, Parpa E, Tsilika E, Galanos A,Vlahos L. Does quality of sleep mediate the effectof depression on hopelessness? Int J Psychol 2009;44:282e289.

49. Mystakidou K, Parpa E, Tsilika E, et al. Sleepquality in advanced cancer patients. J PsychosomRes 2007;62:527e533.

50. Uchida M, Akechi T, Okuyama T, et al. Pa-tients’ supportive care needs and psychologicaldistress in advanced breast cancer patients in Japan.Jpn J Clin Oncol 2011;41:530e536.

51. Ugalde A, Aranda S, Krishnasamy M, Ball D,Schofield P. Unmet needs and distress in peoplewith inoperable lung cancer at the commencementof treatment. Support Care Cancer 2012;20:419e423.

52. Wilson KG, Chochinov HM, Skirko MG, et al.Depression and anxiety disorders in palliative can-cer care. J Pain Symptom Manage 2007;33:118e129.

53. Akechi T, Okamura H, Okuyama T, et al. Psy-chosocial factors and survival after diagnosis ofinoperable non-small cell lung cancer. Psychooncol-ogy 2009;18:23e29.

54. Mystakidou K, Parpa E, Tsilika E, et al. Depres-sion, hopelessness, and sleep in cancer patients’desire for death. Int J Psychiatry Med 2007;37:201e211.

55. Mystakidou K, Tsilika E, Parpa E, et al. Per-sonal growth and psychological distress in advancedbreast cancer. Breast 2008;17:382e386.

56. Mystakidou K, Tsilika E, Parpa E, et al.Exploring the relationships between depression,hopelessness, cognitive status, pain, and spiritualityin patients with advanced cancer. Arch PsychiatrNurs 2007;21:150e161.

57. Steel JL, Nadeau K, Olek M, Carr BI. Prelimi-nary results of an individually tailored psychosocialintervention for patients with advanced hepatobili-ary carcinoma. J Psychosoc Oncol 2007;25:19e42.

58. Capozzo MA, Schillani G, Aguglia E, et al. Sero-tonin transporter 5-HTTLPR polymorphism andresponse to citalopram in terminally ill cancer pa-tients: report of twenty-one cases. Tumori 2009;95:479e483.

59. Okamura M, Akizuki N, Nakano T, et al. Clin-ical experience of the use of a pharmacologicaltreatment algorithm for major depressive disorderin patients with advanced cancer. Psychooncology2008;17:154e160.

60. Pirl WF, Greer JA, Goode M, Smith MR. Pro-spective study of depression and fatigue in menwith advanced prostate cancer receiving hormonetherapy. Psychooncology 2008;17:148e153.

Vol. - No. - - 2014 17Patient Characterization in Depression Studies

61. Schillani G, Capozzo MA, Aguglia E, et al. 5-HTTLPR polymorphism of serotonin transporterand effects of sertraline in terminally ill cancer pa-tients: report of eleven cases. Tumori 2008;94:563e567.

62. Schillani G, Capozzo MA, Era D, et al. Pharma-cogenetics of escitalopram and mental adaptationto cancer in palliative care: report of 18 cases. Tu-mori 2011;97:358e361.

63. Stockler MR, O’Connell R, Nowak AK, et al. Ef-fect of sertraline on symptoms and survival in pa-tients with advanced cancer, but without majordepression: a placebo-controlled double-blind rand-omised trial. Lancet Oncol 2007;8:603e612.

64. Moorey S, Cort E, Kapari M, et al. A cluster ran-domized controlled trial of cognitive behaviourtherapy for common mental disorders in patientswith advanced cancer. Psychol Med 2009;39:713e723.

65. Popa-Velea O, Cernat B, Tambu A. Influenceof personalized therapeutic approach on quality oflife and psychiatric comorbidity in patients withadvanced colonic cancer requiring palliative care.J Med Life 2010;3:343e347.

66. Waller A, Girgis A, Johnson C, et al. Improvingoutcomes for people with progressive cancer: inter-rupted time series trial of a needs assessment inter-vention. J Pain Symptom Manage 2012;43:569e581.

67. Akechi T, Okamura H, Nakano T, et al. Genderdifferences in factors associated with suicidal idea-tion in major depression among cancer patients.Psychooncology 2010;19:384e389.

68. Ruijs CD, Kerkhof AJ, van der Wal G, Onwutea-ka-Philipsen BD. Depression and explicit requestsfor euthanasia in end-of-life cancer patients in pri-mary care in the Netherlands: a longitudinal, pro-spective study. Fam Pract 2011;28:393e399.

69. American Psychiatric Association, Work Groupto Revise DSM-III. Diagnostic and statistical manualof mental disorders: DSM-III-R, 3rd ed. Washington:American Psychiatric Association, 1987.

70. Salvo N, Zeng L, Zhang L, et al. Frequency ofreporting and predictive factors for anxiety anddepression in patients with advanced cancer. ClinOncol (R Coll Radiol) 2012;24:139e148.

71. Zigmond AS, Snaith RP. The hospital anxietyand depression scale. Acta Psychiatr Scand 1983;67:361e370.

72. Delgado-Guay M, Parsons HA, Li Z, Palmer JL,Bruera E. Symptom distress in advanced cancer pa-tients with anxiety and depression in the palliativecare setting. Support Care Cancer 2009;17:573e579.

73. Du-Quiton J, Wood PA, Burch JB, et al. Acti-graphic assessment of daily sleep-activity pattern ab-normalities reflects self-assessed depression andanxiety in outpatients with advanced non-smallcell lung cancer. Psychooncology 2010;19:180e189.

74. Mystakidou K, Parpa E, Tsilika E, et al. Prepara-tory grief, psychological distress and hopelessness inadvanced cancer patients. Eur J Cancer Care (Engl)2008;17:145e151.

75. Mystakidou K, Tsilika E, Parpa E, et al. Illness-related hopelessness in advanced cancer: influenceof anxiety, depression, and preparatory grief. ArchPsychiatr Nurs 2009;23:138e147.

76. Neron S, Correa JA, Dajczman E, et al.Screening for depressive symptoms in patientswith unresectable lung cancer. Support Care Cancer2007;15:1207e1212.

77. Okuyama T, Endo C, Seto T, et al. Cancer pa-tients’ reluctance to discuss psychological distresswith their physicians was not associated with under-recognition of depression by physicians: a prelimi-nary study. Palliat Support Care 2009;7:229e233.

78. Rolke HB, Bakke PS, Gallefoss F. HRQoLchanges, mood disorders and satisfaction after treat-ment in an unselected population of patients withlung cancer. Clin Respir J 2010;4:168e175.

79. Teunissen S, de Graeff A, Voest E, de Haes J.Are anxiety and depressed mood related to physicalsymptom burden? A study in hospitalized advancedcancer patients. Palliat Med 2007;21:341e346.

80. Tsai LY, Li IF, Lai YH, et al. Fatigue and its asso-ciated factors in hospice cancer patients in Taiwan.Cancer Nurs 2007;30:24e30.

81. Turner N, Muers M, Haward R, Mulley G. Psy-chological distress and concerns of elderly patientstreated with palliative radiotherapy for lung cancer.Psychooncology 2007;16:707e713.

82. Beck AT, Ward CH, Mendelson M, Mock J,Erbaugh J. An inventory for measuring depression.Arch Gen Psychiatry 1961;4:561e571.

83. Lo C, Lin J, Gagliese L, et al. Age and depres-sion in patients with metastatic cancer: the protec-tive effects of attachment security and spiritualwellbeing. Ageing Soc 2010;30:325e336.

84. Miller S, Lo C, Gagliese L, et al. Patterns ofdepression in cancer patients: an indirect test ofgender-specific vulnerabilities to depression. SocPsychiatry Psychiatr Epidemiol 2011;46:767e774.

85. Mystakidou K, Tsilika E, Parpa E, et al. Therelationship between quality of life and levels ofhopelessness and depression in palliative care.Depress Anxiety 2008;25:730e736.

86. Rodin G, Lo C, Mikulincer M, et al. Pathwaysto distress: the multiple determinants of depression,hopelessness, and the desire for hastened death inmetastatic cancer patients. Soc Sci Med 2009;68:562e569.

87. Rodin G, Walsh A, Zimmermann C, et al. Thecontribution of attachment security and social sup-port to depressive symptoms in patients with meta-static cancer. Psychooncology 2007;16:1080e1091.

18 Vol. - No. - - 2014Janberidze et al.

88. Lloyd-Williams M, Shiels C, Taylor F, Dennis M.Depressiondan independent predictor of earlydeath in patients with advanced cancer. J Affect Dis-ord 2009;113:127e132.

89. McMillan SC, Rivera HR Jr. The relationshipbetween depressive symptoms and symptom distressin patients with cancer newly admitted to hospicehome care. J Hosp Palliat Nurs 2009;11:41e51.

90. Sela RA. Screening for depression in palliativecancer patients attending a pain and symptom con-trol clinic. Palliat Support Care 2007;5:207e217.

91. Sephton SE, Dhabhar FS, Keuroghlian AS,et al. Depression, cortisol, and suppressed cell-mediated immunity in metastatic breast cancer.Brain Behav Immun 2009;23:1148e1155.

92. Slovacek L, Slovackova B, Slanska I, et al.Depression symptoms and health-related quality oflife among patients with metastatic breast cancerin programme of palliative cancer care. Neoplasma2009;56:467e472.

93. Slovacek L, Slanska I, Slovackova B, et al.Screening for depression in survivors of metastaticovarian cancer in a programme of palliative cancercare. Bratisl Lek Listy 2009;110:655e659.

94. Slovacek L, Slovackova B, Slanska I, et al. Can-cer and depression: a prospective study. Neoplasma2009;56:187e193.

95. Tian J, Chen ZC, Hang LF. Effects of nutri-tional and psychological status of the patients withadvanced stomach cancer on physical performancestatus. Support Care Cancer 2009;17:1263e1268.

96. Maric DM, Jovanovic DM, Golubicic IV, et al.Psychological well-being in advanced NSCLC pa-tients in Serbia: impact of sociodemographic andclinical factors. Neoplasma 2010;57:1e7.

97. Jacobson CM, Rosenfeld B, Pessin H,Breitbart W. Depression and IL-6 blood plasma con-centrations in advanced cancer patients. Psychoso-matics 2008;49:64e66.

98. Hamilton M. A rating scale for depression.J Neurol Neurosurg Psychiatry 1960;23:56e62.

99. Mitchell AJ, Chan M, Bhatti H, et al. Prevalenceof depression, anxiety, and adjustment disorder inoncological, haematological, and palliative-care set-tings: a meta-analysis of 94 interview-based studies.Lancet Oncol 2011;12:160e174.

100. Hotopf M, Chidgey J, Addington-Hall J, Ly KL.Depression in advanced disease: a systematic review.Part 1. Prevalence and case finding. Palliat Med2002;16:81e97.

101. Rayner L, Price A, Hotopf M, Higginson IJ.The development of evidence-based Europeanguidelines on the management of depression inpalliative cancer care. Eur J Cancer 2011;47:702e712.

102. Karsten J, Hartman CA, Smit JH, et al. Psychi-atric history and subthreshold symptoms as

predictors of the occurrence of depressive or anxi-ety disorder within 2 years. Br J Psychiatry 2011;198:206e212.

103. Patten SB, Williams JV, Lavorato DH,Bulloch AG, MacQueen G. Depressive episode char-acteristics and subsequent recurrence risk. J AffectDisord 2012;140:277e284.

104. Byar KL, Berger AM, Bakken SL, Cetak MA.Impact of adjuvant breast cancer chemotherapy onfatigue, other symptoms, and quality of life. OncolNurs Forum 2006;33:E18eE26.

105. UICC. Union for International Cancer Con-trol. TNM classification for malignant disease.2013. Available at: http://www.uicc.org/resources/tnm. Accessed November 15, 2013.

106. Levy MH, Adolph MD, Back A, et al. Palliativecare. J Natl Compr Canc Netw 2012;10:1284e1309.

107. Kaasa S. Integration of general oncology andpalliative care. Lancet Oncol 2013;14:571e572.

108. WHO. WHODefinition of Palliative Care. 2011.Available at: http://www.who.int/cancer/palliative/definition/en/. Accessed November 15, 2013.

109. Kroenke K, Theobald D, Wu JW, et al. The as-sociation of depression and pain with health-relatedquality of life, disability, and health care use in can-cer patients. J Pain Symptom Manage 2010;40:327e341.

110. Laird BJA, Boyd AC, Colvin LA, Fallon MT. Arecancer pain and depression interdependent? A sys-tematic review. Psychooncology 2009;18:459e464.

111. Giannousi Z, Gioulbasanis I, Pallis AG, et al.Nutritional status, acute phase response and depres-sion in metastatic lung cancer patients: correlationsand association prognosis. Support Care Cancer2012;20:1823e1829.

112. Laird BJ, Scott AC, Colvin LA, et al. Pain,depression, and fatigue as a symptom cluster inadvanced cancer. J Pain Symptom Manage 2011;42:1e11.

113. First MB, Spitzer RL, Gibbon M, Williams JB.Structured Clinical Interview for the DSM-IV Axis IDisorders, Clinical Version (SCID-CV). Washington:American Psychiatric Press, 1996.

114. Spitzer RL, Kroenke K, Williams JB. Validationand utility of a self-report version of PRIME-MD: thePHQ primary care study. Primary Care Evaluation ofMental Disorders. Patient Health Questionnaire.JAMA 1999;282:1737e1744.

115. Thekkumpurath P, Walker J, Butcher I, et al.Screening for major depression in cancer outpa-tients: the diagnostic accuracy of the 9-item patienthealth questionnaire. Cancer 2011;117:218e227.

116. Aaronson NK, Ahmedzai S, Bergman B, et al.The European Organization for Research andTreatment of Cancer QLQ-C30: a quality-of-life in-strument for use in international clinical trials inoncology. J Natl Cancer Inst 1993;85:365e376.

Vol. - No. - - 2014 19Patient Characterization in Depression Studies

117. Cella DF, Tulsky DS, Gray G, et al. The Func-tional Assessment of Cancer Therapy scale: develop-ment and validation of the general measure. J ClinOncol 1993;11:570e579.

118. Holland JC, Andersen B, Breitbart WS, et al.Distress management. J Natl Compr Canc Netw2010;8:448e485.

119. PROMIS. Patient-Reported Outcomes Measure-ment Information System (PROMIS). 2013. Availableat: www.nihpromis.org. Accessed November 15, 2013.

120. Pilkonis PA, Choi SW, Reise SP, et al. Itembanks for measuring emotional distress from thePatient-Reported Outcomes Measurement Informa-tion System (PROMIS(R)): depression, anxiety, andanger. Assessment 2011;18:263e283.

121. Olino TM, Yu L, McMakin DL, et al. Compari-sons across depression assessment instruments inadolescence and young adulthood: an itemresponse theory study using two linking methods.J Abnorm Child Psychol 2013;41:1267e1277.

122. Mc Nair DM, Lorr M, Droppelman LF. Manualfor the Profile of Mood States. San Diego, CA:Educational and Industrial Testing Services, 1971.

123. Beck AT, Weissman A, Lester D, Trexler L. Themeasurement of pessimism: the hopelessness scale.J Consult Clin Psychol 1974;42:861e865.

124. Cleeland CS, Mendoza TR, Wang XS, et al. As-sessing symptom distress in cancer patients: the M.D. Anderson Symptom Inventory. Cancer 2000;89:1634e1646.

125. Pargament KI, Koenig HG, Perez LM. Themany methods of religious coping: developmentand initial validation of the RCOPE. J Clin Psychol2000;56:519e543.

126. PrigersonHG,Maciejewski PK, ReynoldsCF 3rd,et al. Inventory of Complicated Grief: a scale to mea-sure maladaptive symptoms of loss. Psychiatry Res1995;59:65e79.

127. Cohen SR, Mount BM, Bruera E, et al. Validityof the McGill Quality of Life Questionnaire in thepalliative care setting: a multi-centre Canadian studydemonstrating the importance of the existentialdomain. Palliat Med 1997;11:3e20.

128. Watson D, Clark LA, Tellegen A. Developmentand validation of brief measures of positive andnegative affect: the PANAS scales. J Pers Soc Psychol1988;54:1063e1070.

129. Cohen SR, Mount BM. Quality of life in termi-nal illness: defining and measuring subjective well-being in the dying. J Palliat Care 1992;8:40e45.

130. Cox JL, Holden JM, Sagovsky R. Detection ofpostnatal depression. Development of the 10-itemEdinburgh Postnatal Depression Scale. Br J Psychia-try 1987;150:782e786.

131. Bruera E, Kuehn N, Miller MJ, Selmser P,Macmillan K. The Edmonton Symptom Assessment

System (ESAS): a simple method for the assessmentof palliative care patients. J Palliat Care 1991;7:6e9.

132. Roth AJ, Kornblith AB, Batel-Copel L, et al.Rapid screening for psychologic distress in menwith prostate carcinoma: a pilot study. Cancer1998;82:1904e1908.

133. Wilson KG, Graham ID, Viola RA, et al. Struc-tured interview assessment of symptoms and con-cerns in palliative care. Can J Psychiatry 2004;49:350e358.

134. Peterman AH, Fitchett G, Brady MJ,Hernandez L, Cella D. Measuring spiritual well-beingin people with cancer: the functional assessment ofchronic illness therapydSpiritual Well-being Scale(FACIT-Sp). Ann Behav Med 2002;24:49e58.

135. Lyons KD, Bakitas M, Hegel MT, et al. Reli-ability and validity of the Functional Assessment ofChronic Illness Therapy-Palliative Care (FACIT-Pal) scale. J Pain Symptom Manage 2009;37:23e32.

136. Beck AT, Steer RA, Brown GK. Manual of theBeck Depression InventorydII. San Antonio: Psy-chological Corporation, 1996.

137. Portenoy RK, Thaler HT, Kornblith AB, et al.The Memorial Symptom Assessment Scale: an in-strument for the evaluation of symptom prevalence,characteristics and distress. Eur J Cancer 1994;30A:1326e1336.

138. Ware JE, Sherbourne CD. The MOS 36-itemshort-form health survey (SF-36). I. Conceptualframework and item selection. Med Care 1992;30:473e483.

139. Radloff LS. The CES-D scale: a self-reportdepression scale for research in general population.Appl Psychol Meas 1977;1:385e401.

140. Mystakidou K, Tsilika E, Parpa E, et al. Lifebefore death: identifying preparatory grief throughthe development of a new measurement inadvanced cancer patients (PGAC). Support CareCancer 2005;13:834e841.

141. Ware J Jr, Kosinski M, Keller SD. A 12-ItemShort-Form Health Survey: construction of scalesand preliminary tests of reliability and validity.Med Care 1996;34:220e233.

142. Muller MJ, Himmerich H, Kienzle B, Szegedi A.Differentiatingmoderate and severe depression usingthe Montgomery-Asberg Depression Rating Scale(MADRS). J Affect Disord 2003;77:255e260.

143. Bergman B, Aaronson NK, Ahmedzai S, Kaasa S,Sullivan M. The EORTC QLQ-LC13: a modular sup-plement to the EORTCCoreQuality of LifeQuestion-naire (QLQ-C30) for use in lung cancer clinical trials.EORTC Study Group on Quality of Life. Eur J Cancer1994;30A:635e642.

144. Dugan W, McDonald MV, Passik SD, et al. Useof the Zung Self-Rating Depression Scale in cancerpatients: feasibility as a screening tool. Psychooncol-ogy 1998;7:483e493.

20 Vol. - No. - - 2014Janberidze et al.

145. Zung WW. A self-rating depression scale. ArchGen Psychiatry 1965;12:63e70.

146. Brooks R. EuroQol: the current state of play.Health Policy 1996;37:53e72.

147. Hickie IB, Davenport TA, Hadzi-Pavlovic D,et al. Development of a simple screening tool forcommon mental disorders in general practice.Med J Aust 2001;175:S10eS17.

148. Martin AJ, Greatorex V, Simes RJ. P81 towardsa utility-based assessment for cancer patients: reli-ability and validity of the UBQ-C(ancer) items. Con-trol Clin Trials 1997;18:S160.

149. Spitzer WO, Dobson AJ, Hall J, et al.Measuring the quality of life of cancer patients: aconcise QL-index for use by physicians. J ChronicDis 1981;34:585e597.

150. Zung WW. The Depression Status Inventory:an adjunct to the self-rating depression scale.J Clin Psychol 1972;28:539e543.

151. McCorkle R, Young K. Development of a symp-tom distress scale. Cancer Nurs 1978;1:373e378.

152. Boyes A, Girgis A, Lecathelinais C. Brief assess-ment of adult cancer patients’ perceived needs:development and validation of the 34-item Support-ive Care Needs Survey (SCNS-SF34). J Eval ClinPract 2009;15:602e606.

153. Schofield P, Gough K, Ugalde A, et al. Valida-tion of the needs assessment for advanced lung can-cer patients (NA-ALCP). Psychooncology 2012;21:451e455.

154. Lewis G, Pelosi AJ, Araya R, Dunn G.Measuring psychiatric disorder in the community-da standardized assessment for use by lay inter-viewers. Psychol Med 1992;22:465e486.

155. Spitzer RL, Williams JB, Kroenke K, et al. Util-ity of a new procedure for diagnosing mental disor-ders in primary care. The PRIME-MD 1000 study.JAMA 1994;272:1749e1756.

Appendix I

Search Strategy PubMed(Depressive disorder[mh] OR Depression[mh] OR depression[tiab] OR depressive[tiab]) AND

(Palliative care[mh] OR palliat* [tiab] OR Terminal care[mh] OR terminal care[tiab] OR end-of-life care[tiab] OR supportive care[tiab] OR comfort care[tiab] OR Hospices[mh] OR hospice*[tiab] OR Terminally ill[mh] OR (terminal*[tiab] AND (ill[tiab] OR illness*[tiab])) OR (advanced[tiab] AND cancer[tiab])) AND cancer[sb] e Limits: Publication Date from 2007.

Categories Alternatives

Author Name of the first authorYear Publication yearCountry If the study is multinational please, write ‘‘multinational’’Journal Name of the journalStudy design 0 ¼ Prospective/longitudinal, 1 ¼ Cross-sectional,

2 ¼ Retrospective, 3 ¼ Intervention, 4 ¼ RCTSample size Indicate number at inclusionTerminology used to describe the sample 0 ¼ No information, 1 ¼ Advanced, 2 ¼ Palliative,

3 ¼ Metastatic, 4 ¼ Terminally ill, 5 ¼ End-of-life,6 ¼ Not curable/not operable, 7 ¼ Multipleterminology used

Follow-up Provide time of the follow-upStudy main outcomeAssessment and classification of depression

Is depression classified according to a diagnostic system? 0 ¼ No information, 1 ¼ Yes, 2 ¼ NoIf ‘‘yes,’’ which system was used 0 ¼ DSM-IV, 1 ¼ DSM-III, 2 ¼ ICD-10, 4 ¼ OtherIf ‘‘other,’’ please specifyWhich type of interview was used for diagnosingdepression?

0 ¼ No information, 1 ¼ Structured, 2 ¼ Unstructured

Please provide the name of the interviewPercentage of patients with depression 0 ¼ No information, 1 ¼ InformationSpecify percentage of patientsIs depression assessed by using depression-specificinstruments?

0 ¼ No, 1 ¼ Yes

If ‘‘yes,’’ please, specifyAre QoL instruments used for assessment of depression 0 ¼ No, 1 ¼ YesIf ‘‘yes,’’ please specify

Depression-related variablesUse of antidepressants 0 ¼ No information, 1 ¼ InformationDifferent types of psychotherapy 0 ¼ No information, 1 ¼ InformationDuration of present depressive episode 0 ¼ No information, 1 ¼ Information, 2 ¼ Assessed but

not reportedPrevious depressive episodes 0 ¼ No information, 1 ¼ Information, 2 ¼ Assessed but

not reportedPsychiatric history 0 ¼ No information, 1 ¼ Information, 2 ¼ Assessed but

not reportedHistory of alcohol and/or drug abuse 0 ¼ No information, 1 ¼ Information, 2 ¼ Assessed but

not reportedDescription of patient samples

Sociodemographic variablesAge/date of birtha 0 ¼ No information, 1 ¼ Information, 2 ¼ Assessed but

not reportedGendera 0 ¼ No information, 1 ¼ Information, 2 ¼ Assessed but

not reportedEthnicitya 0 ¼ No information, 1 ¼ Information, 2 ¼ Assessed but

not reportedMarital status 0 ¼ No information, 1 ¼ Information, 2 ¼ Assessed but

not reportedEducationa 0 ¼ No information, 1 ¼ Information, 2 ¼ Assessed but

not reportedReligion 0 ¼ No information, 1 ¼ Information, 2 ¼ Assessed but

not reportedMedical information (disease-specific variables)Principal diagnosisa 0 ¼ No information, 1 ¼ InformationTime since diagnosesa 0 ¼ No information, 1 ¼ InformationStage of cancer diseasea 0 ¼ No information, 1 ¼ Described as locally advanced,

advanced, or metastatic/disseminated (informationprovided)

(Continued)

Appendix II

Extraction Log

Vol. - No. - - 2014 21Patient Characterization in Depression Studies

Continued

Categories Alternatives

Metastases 0 ¼ No information, 1 ¼ InformationSite of metastasesa 0 ¼ No information, 1 ¼ InformationPatients’ additional diagnoses (comorbidities)a 0 ¼ No information, 1 ¼ InformationExpected survival 0 ¼ No information, 1 ¼ InformationReal survival 0 ¼ No information, 1 ¼ InformationWeight lossa 0 ¼ No information, 1 ¼ InformationPerformance statusa 0 ¼ No information, 1 ¼ Inclusion criterion, 2 ¼ Specified,

3 ¼ Both 1 and 2, 4 ¼ Assessed but not reportedCognitive functiona 0 ¼ No information, 1 ¼ Inclusion criterion, 2 ¼ Specified,

3 ¼ Both 1 and 2, 4 ¼ Assessed but not reportedTreatmentPresent anticancer treatment 0 ¼ No information, 1 ¼ InformationMedication

Opioidsa 0 ¼ No information, 1 ¼ InformationNeurolepticsa 0 ¼ No information, 1 ¼ Information

Caregiver issuesLiving situationa (information about caregiver issues

or household/living with, more than marital status)0 ¼ No information, 1 ¼ Information

Information about social network 0 ¼ No information, 1 ¼ InformationInformation about settingPlace of carea 0 ¼ No information, 1 ¼ Home, 2 ¼ Long-term care

facilities, 3 ¼ Hospice/palliative care unit, 4 ¼ Hospital,general departments including oncology department,5 ¼ Other, 6 ¼ Two/multiple locations

Provision of carea 0 ¼ No information, 1 ¼ Inpatients, 2 ¼ Outpatients,3 ¼ Both 1 and 2

Comments

RCT ¼ randomized controlled trial; DSM-IV¼ Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition; DSM-III ¼ Diagnostic and Statis-tical Manual of Mental Disorders, Third Edition; ICD-10 ¼ International Statistical Classification of Diseases and Related Health Problems, Tenth Revision;QoL ¼ quality of life.aVariables included in the EAPC Basic Dataset.

Appendix II

22 Vol. - No. - - 2014Janberidze et al.