Existenceofbenefitfindingand ...some general cancer review papers published, but none which have focused on people with head and neck cancer (Stanton, Bower, & Low, 2006; Sumalla,

Page 1

Existence of benefit finding andposttraumatic growth in people treated forhead and neck cancer a systematic review

Submitted 9 October 2013Accepted 12 January 2014Published 11 February 2014Corresponding authorSam Harding shardingjbgmailcom

Academic editorGerhard Andersson

Additional Information andDeclarations can be found onpage 18

DOI 107717peerj256

Copyright2014 Harding et al

Distributed underCreative-Commons CC-BY 30

OPEN ACCESS

Sam Harding1 Fatimeh Sanipour1 and Timothy Moss21Department of Psychology University of the West of England Bristol United Kingdom2Centre for Appearance Research University of the West of England Bristol United Kingdom

ABSTRACTBackground The impact of head and neck cancer (HNC) in long-term survivorsdiffers widely among individuals and a significant number of them suffer from thenegative effects of disease whereas others report significant positive effect This sys-tematic review investigated the evidence the implications of treatment for HNC andsubsequent development of Benefit Finding (BF) or Posttraumatic Growth (PTG)Purpose To understand how differing medical psychological and social characteris-tics of HNC may lead to BFPTG and subsequently inform post-treatment interven-tions to encourage positive outcomesMethod In February 2012 five databases including Pubmed and Psych Info weresearched for peer-reviewed English-language publications Search strings includedkey words pertaining to HNC BF and PTG One thousand three hundred and sixtythree publications were identified reviewed and reduced following Cochrane guide-lines and inclusionexclusion criteria specified by a group of maxillofacial consul-tants and psychologists Publications were then quality assessed using the CASPCohort Critical Appraisal toolFindings Five manuscripts met the search and selection criteria and were sourcedfor review All studies were identified as being level IIb evidence which is a mediumlevel of quality The majority of studies investigated benefit finding (80) and weresplit between recruiting participant via cancer clinics and postal survey Theyfocused on the medical psychological and social characteristics of the patient fol-lowing completion of treatment for HNCConclusion Demographic factors across the papers showed similar patterns of re-lationships across BF and PTG that higher educationqualification and cohabita-tionmarriage are associated with increased BFPTG Similarly overlap with diseasecharacteristics and psychosocial factors where hope and optimism were both posi-tively correlated with increased reported BFPTG

Subjects Oncology Psychiatry and PsychologyKeywords Posttraumatic growth Benefit finding Head and neck Cancer Silver lining questionnairePosttraumatic growth inventory Quality of life Systematic review

How to cite this article Harding et al (2014) Existence of benefit finding and posttraumatic growth in people treated for head and neckcancer a systematic review PeerJ 2e256 DOI 107717peerj256

Page 2

INTRODUCTIONA great deal of evidence has accumulated over the past thirty years for the negativesequelae of trauma Traumatic events can include a range of experiences including healththreats The literature on coping with health difficulties has documented a variety ofnegative consequences including depression (egMoyer amp Salovey 1996 Timberlakeet al 1997 posttraumatic stress disorder (PTSD) (eg Alter et al 1996 Andrykowskiet al 1998 and adjustment difficulties (eg Schulz et al 1995) These models havetended to work towards a clinical diagnosis for which treatment may be prescribed

By contrast models of positive illness recovery have been informed by a range of moregeneral theories of life change (Horowitz 1986 Park amp Ai 2006 Paton 2006) Thesehave tried to understand the mechanisms that may underpin the positive sequelae ofhealth-related trauma Since these models are not working towards a diagnosis forprescription there has been no imperative to coalesce around a common agreedunderstanding against which a diagnosis can be made

Morse (1997) conceptualises coping with life-threatening illness as incorporating fivedistinct stages The first stage is uncertainty or vigilance during which patients suspect acondition and attempt to maintain emotional control whilst trying to understand theircondition and its severity The second stage is disruption a time when individuals realisethat they are affected by what they perceived to be a serious disease and may experiencehigh levels of stress In the third stage striving for recovery individuals may try to gaincontrol over their illness with the help of personal and environmental resources Thefourth stage is striving to restore onersquos self and making sense of altered reality The fifthand final stage is learning to live with the altered self in which patients attain a newequilibrium as a result of accepting the illness and its consequences In chronic illness areturn to a prior state of health may not be a realistic outcome This and subsequentmodels suggest that it is the time of diagnosis and the disruption stage especially whenthis involves news of a life-threatening illness that patients are the most likely toexperience trauma (Morse amp Johnson 1991) This is also the stage during whichindividuals are most likely to confront existential issues posed by the diagnosis (Doka2008)

Brennan (2001) proposes that social cognitive transition (SCT) model builds onprevious theories of coping traumatic stress social-cognition and cognitive theories ofemotion This theory hinges on the central components of the cognitive models of PTSDexcept it allows for both positive and negative psychological outcomes after a traumaBrennan (2001) proposes that all individuals have mental models of the world made up ofassumptions As an individual interacts with their world these assumptions are eitherconfirmed or disconfirmed by experience If we consider Leventhalrsquos model ofSelf-regulation (Leventhal Nerenz amp Steele 1984) then his stimulus is a disruption orchallenge to the Assumptive World The arising representations map to an expectationand the coping behaviours to new experiences The subsequent outcomes either confirmor disconfirm the mental model of the Assumptive World In this way Brennanrsquos medical

Harding et al (2014) PeerJ 107717peerj256 223

Page 3

model encompasses Leventhalrsquos broader psychosocial framework and provides anaccount for the diverse psychosocial outcomes experienced by cancer patients

This model would propose that PTSD is the negative result of an extremely troublingevent that is highly incongruent with the individualrsquos assumptions about the worldBrennan (2001) indicates that denial and avoidance are the first responses of atraumatised individual which create more stress and potentially lead to the developmentof new assumptions about the world assumptions that may be dysfunctional and lead toheightened levels of distress or PTSD However avoidance and denial can also serve apositive roll by diluting lsquolsquothe absorption of lsquotraumaticrsquo informationrsquorsquo (Brennan ampMoynihan 2004 p 9) Conversely Brennan ampMoynihan (2004) proposes that an adaptiveresponse to traumatic experiences requires worry It is hypothesised that worry is a partof the cognitive attempt to anticipate and prepare for future threat (Brennan ampMoynihan2004 Eysenck 1992) By imagining and confronting worst case scenarios bylsquolsquodecatastrophisingrsquorsquo them the individual can appraise the realistic nature of the eventBrennan ampMoynihan (2004) proposes that positive outcomes from traumatic experiencescan then occur as unrealistic goals or outcomes are discarded and implicit long-standinglife goals become clear and distinct

Benefit finding (BF) and posttraumatic growth (PTG) describe similar outcomesfollowing adversity yet there are clear differences Both describe a positive outcome withBF being described as the acquisition of benefit from adversity (Collins Taylor amp Skokan1990 Tennen amp Affleck 2002) and PTG growth being the success with which individualscoping with the aftermath of trauma reconstruct or strengthen their perceptions of selfothers and the meaning of events (Tedeschi amp Calhoun 1996) Examples of BF findinginclude a positive change in relationships a greater appreciation of life and a change inlife priorities PTG is also described as lsquothe experience of significant positive changearising from the struggle with a major life crisisrsquo with examples of increased sense ofpersonal strength changed priorities and richer existential and spiritual life being cited inthe literature (Calhoun et al 2000)

Despite these similarities there is emerging evidence that there are critical differencesfor example Sears Stanton amp Danoff-Burg (2003) showed that BF was predicted bypersonal characteristics (ie education optimism and hope) but PTG was not Benefitfinding may start immediately after diagnosis and results from challenges to theindividualrsquos cognitive representations that is they have the same personalrepresentations but have positive ways of coping By contrast PTG is a re-assembly of theassumptive world in a new way following trauma and develops as a result of therumination and restructuring of the selfworld relationship that occurs in the weeksmonths and even years following trauma and is focussed on changes in onersquos capacity todeal with adverse events (Calhoun amp Tedeschi 1998) So PTG results from challenges todeeper cognitive representations than BF and result in changed lsquorules for livingrsquo and lsquocoreschemarsquo whereas BF may be more superficial and transient in nature This difference mayalso lead one to expect more PTG growth with increasing time post-trauma becausemore time is available for cognitive processing (Sears Stanton amp Danoff-Burg 2003)

Harding et al (2014) PeerJ 107717peerj256 323

Page 4

However this hypothesis has yet to be tested and given that PTG has no diagnosticperiod of onset unlike PTSD (American Psychiatric Association 2013) this systematicreview has aggregated BF and PTG and will search for both of these concepts andwordsphrase used synonymously such as lsquostress-related growthrsquo and lsquoexistential growthrsquoThe authors will refer to these concepts throughout the remainder to this manuscript asBFPTG unless making specific reference to information from research where onetheoretical perspective has been purposely selected

Recent studies have provided evidence that these positive processes also take place inchronically ill patients including individuals suffering from cancer (Affleck amp Tennen1996 Carver amp Antoni 2004 Petrie et al 1999 Schulz amp Mohamed 2004 Sears Stantonamp Danoff-Burg 2003 Tomich amp Helgeson 2004) The bulk of this research has beenundertaken on females with breast cancer (Carver amp Antoni 2004 Petrie et al 1999Sears Stanton amp Danoff-Burg 2003 Tomich amp Helgeson 2004) There have also beensome general cancer review papers published but none which have focused on peoplewith head and neck cancer (Stanton Bower amp Low 2006 Sumalla Ochoa amp Blanco2009) In the United Kingdom 1259 females in every 100000 will suffer from breastcancer and 10 males For oral cancer the figures are 55 and 124 respectively (CancerResearch UK 2013) Additionally Cancer Research UK (2013) statistics indicate thatpeople with oral cancer are older at diagnosis than those with breast cancer These twofactors combined with the location of the tumour may impact the development ofBFPTG and it is for this reason that a systematic review of this cancer site is needed

This systematic review investigates the literature on BFPTG in the patients treated forcancer in the region of the Head and Neck (HNC) The aim is to collate the currentquantitative data to understand how differing medical psychological and socialcharacteristics of HNC may lead to BFPTG and subsequently may inform diagnosis andfuture post-treatment interventions to encourage sustained positive outcomes

METHODSThe review strategy was adapted from the Cochrane Collaboration systematic reviewmethodology and uses a narrative synthesis (The Cochrane Collaboration 1999) andguidance from Petticrew amp Roberts (2006)

Identification of selection criteriaThe Booth amp Fry-Smith (2004) PICO model (population intervention comparisonoutcome) guided the development of the search strategy

The lsquoPopulationrsquo of interest was defined as adults (gt18 years) of either sex with HNCChildren and adolescents can develop HNC but due to high relevance of developmentalstage and cognitive maturity they are excluded from the review Terminal patients andthose with recurrent metastatic disease on entry to the study were excluded as theywould currently be experiencing significant on-going challenging and potentiallytraumatic experiences

This systematic review is not investigating an lsquoInterventionrsquo in the sense of lsquoCognitiveBehavioural Therapyrsquo as an example The interventions of interest that may affect

Harding et al (2014) PeerJ 107717peerj256 423

Page 5

Table 1 ICD10 codes related to cancer sites and incidence

Cancer site ICD10 code Number of regis-trations 2000

Incidence crude rate per 100000 2000

Men WomenMouth lip amp oral cavity C00-06 2329 59 37Salivary glands C07-8 422 1 08Pharynx C09-14 1339 4 16Nasal cavity ear amp sinuses C30-31 352 08 06arynx C32 1903 66 13Thyroid C73 1131 13 33

outcome is the treatment for the malignant tumour ie surgery radiotherapychemotherapy and any combination of these treatments or specifically named variationssuch as photodynamic therapy In relation to lsquocomparisonsrsquo no limitations were put on thesearch strategy However it was noted that comparison may be possible by simplycomparing intervention groups cancer sites (Table 1) or measure pre and postintervention

When considering the relevance of lsquooutcomersquo measures to the development of thesearch strategy this review focused purely on quantitative studies The studies mustinclude lsquopaper and pencilrsquo or lsquocomputer basedrsquo psychometrically sound measures of BFandor PTG This will allow comparison of statistical analysis of the relationship betweenBFPTG and categorical medical and social variables as well as other psychologicalcharacteristics collected via validated measures Data collected via studies reportingqualitative data only were excluded

Search strategyThe search strategy was designed in consultation with a senior librarian and the searchterms following a review of the literature and discussion with a Maxillofacial Consultant(Supplemental Information A) A combination of lsquofree textrsquo terms with Boolean operatorsand truncations were used Five separate searches were conducted in electronic databasesPubmed Psych Info (CSA) Psyc Articles (CSA) OVID Medline and PILOTS (PublishedInternational Literature on Traumatic Stress) to identify appropriate studies in articlespublished from the earliest entries of any of the databases until February 2012 No limitswere placed on the electronic search in relation to age range of participants studied orlanguage of publication The PRISMA checklist was followed and a flow chart (Fig 1)details the process of article selection

The citations retrieved from each database were exported to lsquoReference Manager 11rsquobibliographic management software (Thomson ResearchSoft 2000) Duplicates wereremoved and article screened for relevance removing animal studies and medical andpsychological studies which had been retrieved as they contained one or more of thesearch terms eg Squamous Cell or Benefit (Supplemental Information B) To this pointin the review process no limits or restrictions had been placed on lsquocancer sitersquo while

Harding et al (2014) PeerJ 107717peerj256 523

Page 6

Figure 1 PRISMA flowchart

searching the electronic databases or retrieved articles This enabled papers reporting onmultiple cancer sites to be identified and integrated for patterns between tumourlocations Supplemental Information B provides the list of search terms used to identifyappropriate tumour locations within the head and neck region We did not limit thesearch to include or exclude any type of intervention within this participant cohort Inthis review an intervention would be the type of cancer treatment they received Cancerlocation and treatment were specific factors that were identified as potentialconfoundersvariables within the selected papers but this did not require additionalterminology to be added to the research strings or strategies The 514 abstracts of theremaining articles related to BF PTG andor cancer were screened by SH and twentypercent randomly sampled were reviewed by TM and FS

Guidelines dissertations and theses greater than 5 years old handbookscommentaries review articles expert opinions and case reports as well as trials withfewer than ten participants were excluded as were qualitative studies Disagreementbetween the review authors was resolved by consensus through discussion This identified

Harding et al (2014) PeerJ 107717peerj256 623

Page 7

lsquopotentially relevant articlesrsquo (n = 155) and these were obtained and appraised criticallyThree articles (Harrington McGurk amp Llewellyn 2008 Ho et al 2011 Llewellyn et al

2011) were identified from this search strategy After completing the literature searchreferences from these articles review articles thesis and books were examined to identifyadditional grey literature and the author (SH) contacted researchers identified Twoprojects were identified but no responses were received when the authors were contactedTwo of the authors of this Systematic Review (SH amp TM) have two manuscripts inpreparation for submission and these were included in this review as grey literature (SHarding amp T Moss 2013a unpublished data S Harding T Moss 2013b unpublisheddata)

The five identified manuscripts were summarised separately including a description ofthe study design sample size measurement and time since diagnosis or treatment ofHNC and are presented in Tables 2 and 3

One of the five identified papers did not provide sufficient data to extract as part of thisreview The authors of that article were approached and subsequently provided anadditional publication that enabled a fuller understanding of their data and greatercomparison with other published work (Horney et al 2011)

Quality assessmentThis review has identified a very limited number of studies it is therefore insufficient tolimit the assessment of papers to those with the lsquobestrsquo methodology The studies identifiedin this review all represented lsquolevel IIbrsquo evidence (Supplemental Information C NationalInstitute for Clinical Excellence 2004) or those at a medium level of quality where highlevels would refer to studies in the top of the hierarchy of evidence (eg systematicreviews randomised controlled trials) and lsquolowrsquo refers to those near the bottom of thehierarchy (case series case reports expert opinion) Given this assessment of quality theremaining assessment of quality reflects variation within that small banding

Quality was assessed using the Critical Appraisal Skills Programme (CASP) CohortStudy appraisal tools (Critical Appraisal Skills Programme 2011) This tool provides a 12point check list of study validity risk of bias in recruitment exposure outcomemeasurement confounding factors reporting of results and the transferability of findings(maximum score of 12) The key questions from CASP were taken as a template for thequality appraisal (Supplemental Information D) The appraisal questions were answeredwith lsquoyesrsquo lsquocanrsquot tellrsquo and lsquonorsquo Where lsquoyesrsquo was used the study was felt to fill the criteria forthat question Where lsquocanrsquot tellrsquo was used the study was considered to meet some of thecriteria for the question but not others Where lsquonorsquo was used the study was considered toexplicitly not meet the criteria for the question CASP does not provide cut-offs forquality levels however no studies were ruled out on the basis of the quality appraisal sincequality levels were similar between studies

All identified manuscripts were checked for quality against the appraisal toolindependently by SH and FS and confirmed by TM Consensus was immediate betweenthe reviewers Each of the scales used within the studies were also assessed and reported

Harding et al (2014) PeerJ 107717peerj256 723

Page 8

Table 2 Study descriptors

Study Author(s) Aim of the study Study design Study measures Demographicfactors

Medical factors Time of measure-ment

1 HarringtonMcGurk ampLlewellyn(2008)

(1) to determine theextent to which patienttreated for HNCexperience positiveconsequences of theirillness (2) to identifyfactors associated withbenefit finding amongthis patient group

Cross-sectionalpostal survey

Benefit findingscale (BFS)Hospital Anxietyand Distress Scale(HADS) LifeOrientationTest-Revised(LOT-R) BriefCOPE

Age GenderEthnicityEducationEmploymentMarital status

Type of treatmenttime since lasttreatmentdiagnosis offurther illnesssince treatmentsite type of cancerand stage of cancer

0ndash6mths = 16ndash12mths = 313ndash24mths = 725ndash47mths = 2048ndash72mths = 1973ndash121mths = 26

2 Llewellynet al (2011)

(1) to determine theextent to which patienttreated for HNCexperience positiveconsequences of theirillness (2) to establish therelationship between BFother patient-reportedoutcomes and predictivefactors such as copingstrategy and level ofoptimism

Repeatedmeasuresprospectivestudy usingself-completionquestion-naires

Benefit findingscale (BFS)Hospital Anxietyand Distress Scale(HADS) LifeOrientation Test(LOT-R) BriefCOPE MedicalOutcomes ShortForm 12 (SF-12)Two-item measurederived from TheEuropeanOrganization forResearch andTreatment(EORTC) ofCancer Quality ofLife Questionnaire(QLQ-C30)

Age GenderEthnicityEducationEmploymentMarital status

Type of treatmentsite and stage ofcancer

T1 = Betweendiagnosis and startof treatment T2 =6 months aftercompletion oftreatment

3 Ho et al(2011)

Investigate if PTG occursin oral cancer patientsand if hope andoptimism showssignificant positivecorrelation with PTG

Cross-sectionalpostal survey

ChinesePosttraumaticGrowth Inventory(PTGI) Hopescale (HS) LifeOrientation Test -Revised (LOT-R)

Age GenderReligionEducation levelincome

Time sincediagnosis stage ofdisease andtreatment type

Mean time was36yrs (SD 034)

4 S Harding ampT Moss(2013aunpublisheddata)

Investigate therelationship between BFdemographicalbiomedical and HRQoLfollowing the treatmentfor HNC

Cross-sectionalpostal survey

Silver LiningQuestionnaire(SLQ) Universityof WashingtonHead and NeckCaner Quality ofLife (UoW)Medical OutcomesShort Form 12(SF-12)

Age at diagnosisAge at time ofcompletingquestionnaireGender EthnicityIndex of MultipleDeprivationOccupationFamily Status

Tumour site Stageof diseaseLocation oftumour Treatment

Mean time fromcompletingtreatment tocompletingquestionnaires2730mths (Range3ndash76 SD 218)

Harding et al (2014) PeerJ 107717peerj256 823

Page 9

5 S Harding ampT Moss(2013bunpublisheddata)

Investigate thelongitudinal relationshipbetween BFdemographicalbiomedical and HRQoLfollowing the treatmentfor HNC

Repeatedmeasuresprospectivecross-sectionalstudy usingself-completionquestion-naires

Silver LiningQuestionnaire(SLQ) Universityof WashingtonHead and NeckCaner Quality ofLife (UoW)Medical OutcomesShort Form 12(SF-12)

Age at diagnosisAge at time ofcompletingquestionnaireGender EthnicityIndex of MultipleDeprivationOccupationFamily Status

Tumour site Stageof diseaseLocation oftumour Treatment

(Supplemental Information E) Upon reviewing the studiesrsquo data collection tools andstatistical analysis it became apparent that there was too great a variation between themand thus it was not appropriate to conduct additional analysis such as a meta-analysisusing the reported findings

RESULTSQuality Assessment FindingsThe fashion in which data is collected may affect the results Two of the included studiescollected the data during patientsrsquo clinic visits (Ho et al 2011 Llewellyn et al 2011) Thismay have increased the potential sample size but it may also have caused the respondentsto report positive outcomes due to feelings of appreciation for medical treatment or as ameans of thanking the clinical team for treatment The other three studies posted themeasures to the participants which is less likely to elicit socially desirable responses (SHarding amp T Moss 2012a unpublished data S Harding amp T Moss 2012b unpublisheddata Harrington McGurk amp Llewellyn 2008) Postal surveys can result in a low returnrate although those reviewed here received 53ndash55 (respectively S Harding amp T Moss2013a unpublished data Harrington McGurk amp Llewellyn 2008) and can be argued to bereasonable A separate consideration is that they may be biased through participantsself-selecting and subsequently call into the question the generalisability of the findings

All the studies included in this review were quantitative in nature and used previouslyconstructed measures (Supplemental Information E) Measures such as the MedicalOutcomes Short Form 12 (SF-12) have normative date that allows findings to becompared with general population (S Harding amp T Moss 2013a unpublished data SHarding amp T Moss 2013b unpublished data Llewellyn et al 2011) Other measures haveonly been used in other disease populations such as hospital anxiety and depression scale(Harrington McGurk amp Llewellyn 2008 Llewellyn et al 2011) An exception to this wasone of the measures used in Llewellyn et al (2011) In this study two items were derivedfrom the EORTC QLQ-C30 which were used to assess cancer specific global Quality ofLifehealth status

In medical population studies the confounding factors such as stage or exact locationof tumour may be predictive factors and it is therefore important to ensure that these areappropriate selected and analysed (Bellizzi amp Blank 2006 Brunet et al 2010Gallagher-Ross 2012) Similar factors were used across all studies included in this review

Harding et al (2014) PeerJ 107717peerj256 923

Page 10

Table3Pa

rticipan

tsan

dvariables

Stud

yAutho

r(s)

Participan

ts(gender

age)

Timeof

measurement

Non

-respon

dents

drop

outs

Exclusioncrite

ria

Can

cersite

Can

cerstaging

Can

certreatments

Timesince

completionof

treatm

ent

1Harrin

gton

McG

urkamp

Llew

ellyn

(2008)

N=7

6(55

respon

serate37Male

39Female

MeanAge

669SD126

Range32ndash97

71White)

0ndash6m

ths=

16ndash12mths=

313ndash24m

ths=

725ndash47m

ths=

20

48ndash72m

ths=

19

73ndash121mths=

26

Sign

ificant

diffe

rence

betweengend

erin

respon

ders

and

non-respon

ders

(morefemales

respon

ding

)

Und

er18

yearso

fage

Havingpalliative

treatm

entRe

current

diagno

sism

etastatic

diseasein

otherp

arts

ofthebo

dy(excluding

neck

nodes)a

diagno

sisof

lymph

omamentalto

cogn

itive

impairm

ents

orinsufficient

understand

ingof

English

Not

stated

Stage1ndash

2-N

=53Stage

3ndash4-

N=

23

Surgeryon

ly-N

=35

Radiotherapy

only-

N=

10Surgery

and

Radiotherapy

-N=

30Surgery

radiotherapy

and

chem

otherapy

-N=

1

0ndash6m

ths=

16ndash

12mths=

313-24m

ths=

725ndash4

7mths=

20

48ndash7

2mths=

19

73ndash1

21mths=

26

2Llew

ellyn

etal

(2011)

T1N

=103

(73Males30

Females

MeanAge

63SD139

Range23ndash91

93White)

T2N

=68

(Gender

Age

Ethn

icity

data

provided)

T1=

Between

diagno

sisand

starto

ftreatm

entT2

=6mon

thsa

fter

completionof

treatm

ent

Therewereno

significant

diffe

rences

betweenpatie

nts

includ

edand

notincluded

with

respectto

gend

erstage

ofcancer35

peop

ledidno

tcompletethe

second

time

pointNo

inform

ationis

givenabou

tthey

comparedatT1

Und

er18

yearso

fage

Havingpalliative

treatm

entRe

current

diagno

sism

etastatic

diseasein

otherp

arts

ofthebo

dy(excluding

neck

nodes)a

diagno

sisof

lymph

omamentalto

cogn

itive

impairm

ents

orinsufficient

understand

ingof

English

OralC

avity

-N

=68

Pharyn

x-N

=8Larynx

-N

=19O

ther

-N=

8

Stage1-N

=34

Stage2-N

=25

Stage3-N

=23

Stage4-N

=17

Missingdata-

N=

4

Surgeryon

ly-N

=36

Radiotherapy

only-

N=

25

Chemotherapy

only-

N=

3Surgeryand

Radiotherapy

-N=

17R

adiotherapyand

chem

otherapy

-N=

13Surgery

radiotherapy

and

chem

otherapy

-N=

9

Sixmon

thsa

tT2

3Hoetal

(2011)

N=5

0(21

Male29

Female

MeanAge

60SD1306)

Meantim

ewas

36yrs(SD034)

Noinform

ation

isrepo

rted

Non

-nativeCantonese

speakerslessthan

6mthsp

osttreatment

completion

recurrence

OralC

avity

Oroph

aryn

xging

ivalfloor

ofmou

th

tong

uesalivary

glandsbuccal

mucosapalate

Num

bersat

each

siteno

tstated

Stage1ndash

2-N

=41Stage

3ndash4-

N=

5Missing

inform

ation-N

=4

Surgeryon

ly-N

=34

Surgeryand

Radiotherapy

-N=

16

Meantim

ewas

36yrs(SD034)

Harding et al (2014) PeerJ 107717peerj256 1023