A longitudinal study of distress (depression and anxiety) up to18 months after radiotherapy for head and neck cancer

Kate Neilson1*, Annabel Pollard1, Ann Boonzaier1, June Corry2,3, David Castle4,5, David Smith6, Tom Trauer3,4,5

and Jeremy Couper1,4,51Psycho-Oncology Research Unit, Peter MacCallum Cancer Centre, Melbourne, Victoria, Australia2Head and Neck Cancer Service, Peter MacCallum Cancer Centre, Melbourne, Victoria, Australia3Monash University, Melbourne, Victoria, Australia4Department of Psychiatry, St Vincent’s Hospital, Melbourne, Victoria, Australia5University of Melbourne, Melbourne, Victoria, Australia6Department of Psychology, RMIT University, Melbourne, Victoria, Australia

*Correspondence to:Clinical Psychology, PeterMacCallum Cancer Centre,Locked Bag 1 A'Beckett Street,Melbourne, Victoria 8006,Australia. E-mail: Kate.Neilson@petermac.org

Received: 2 February 2012Revised: 26 October 2012Accepted: 27 October 2012

AbstractObjective: The aim of the study was to assess symptoms of depression and anxiety in patients with headand neck cancer up to 18 months after radiotherapy.

Methods: Prospective observational study of consecutive head and neck outpatients was conductedat a tertiary cancer centre (n= 101). Eligibility included diagnosis of cancer in the head and neckregion, where the patient agreed to radiotherapy with curative intent. Data were collected beforecommencement of radiotherapy and 3 weeks and 18 months after completion. Symptoms of depressionand anxiety were assessed by the Hospital Anxiety and Depression Scale. Tumour/treatment-relatedphysical symptoms were assessed using the ‘Additional Concerns’ subscale of the Functional Assessmentof Chronic Illness Therapy for Head and Neck Cancer.

Results: The prevalence of identified probable cases of depression was 15% at baseline, increasingto 29% 3 weeks post-treatment, falling to 8% at 18-month follow-up. The number of probable casesof anxiety was 20% at baseline, 17% at 3 weeks post-treatment and 22% at 18-month follow-up.Depression scores significantly increased from baseline to 3 weeks post-treatment and decreased at18-month follow-up. Variability in depression scores was accounted for by tumour/treatment-relatedphysical symptoms. Anxiety scores significantly decreased between baseline and 3-week post-treatmentand increased at 18-month follow-up. Younger age and more tumour/treatment-related physicalsymptoms predicted anxiety scores.

Conclusions: The rates of depression in head and neck cancer patients increase following cancertreatment and are related to tumour/treatment-related physical symptoms. Anxiety levels are higherpre-treatment, lower immediately following cancer treatment but rise to near pre-treatment levelsmore than a year after completion of cancer treatment.Copyright © 2012 John Wiley & Sons, Ltd.

Introduction

Head and neck cancer (HNC) patients are required toundergo some of the most distressing and disfiguringtreatments among all cancers. Rates of 5-year survivalrange between 23% and 64% depending on the site andtype of cancer [1]. Patients often live with chronic func-tional impairment and disfigurement. Functional changesassociated with the disease, resulting from the acuteeffects of radiation can include difficulties with speaking,swallowing and breathing, as well as pain [2–4]. For thesereasons, HNC has been described as more emotionallytraumatic than any other form of cancer [5].Studies have shown that patients diagnosed with an

HNC display clinically significant levels of psychologicaldistress. Psychological distress can be described as acombination of symptoms including anxiety, depression,cognitive and behavioural impairments [6]. Psychologicaldistress extends along a continuum ranging from transient,normal feelings of vulnerability, sadness and fear toproblems that can become disabling such as symptomsof depression and anxiety. More severe symptoms of

psychological distress may meet the criteria for apsychiatric diagnosis. Psychiatric co-morbidity involvesa diagnosis of a mental disorder such as mood (e.g.depression) or anxiety disorders. Symptoms of depressioninclude less enjoyment of usual activities, feeling sad andless energy, whereas symptoms of anxiety include feelingtense, frightened, worried, unable to relax and/or panic[7,8]. Percentages of HNC patients with significant levelsof distress range from 11% to 52% [9–13].Changes in distress levels over time have been demon-

strated in a few studies of HNC patients, suggesting thatthese patients may experience a reduced quality of life(QOL) and increased symptoms of distress at differentpoints in the treatment process. Studies have shownincreased distress at the time of diagnosis, as well asduring and immediately following radiotherapy and/orchemotherapy [12,14–19]. Kohda et al. [18] found thatthere was a significant decline in QOL and increase insymptoms of depression during radiotherapy, and thatdepressive symptoms persisted after completion of radio-therapy. Some studies have demonstrated that a proportionof patients with HNC continue to struggle with lower

Copyright © 2012 John Wiley & Sons, Ltd.

Psycho-OncologyPsycho-Oncology (2012)Published online in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/pon.3228

QOL and more symptoms of depression up to 1 yearafter treatment [20,21].Undetected and untreated psychological needs are

associated with reductions in patient QOL, extendedhospital stays, malnutrition, more complications withtreatment and treatment non-compliance [22–24]. Onestudy has also found that the suicide rate for HNCpatients is more than four times the rate in the generalUS population [25]. Recognition of anxiety and depres-sive disorders in cancer patients is therefore importantbecause effective psychotherapeutic and pharmacologictreatments exist which can improve QOL and reducepsycho-morbidity [26–28].Investigating predictive factors for psychological distress

helps to identify vulnerable patient groups early, in order totarget these groups with services/interventions to preventsevere and persistent symptoms of distress and facilitatelong-term adjustment and recovery. Demographic factorsthat have been reported to be predictive of psychologicaldistress in HNC include living alone, being male, lackingsocial support, employed at start of treatment, youngerage and having a lower educational level [9,12,15,29–31].Cancer-related physical symptoms and advanced stagecancer have also been associated with increased distressand lower QOL [9,23,30,32].Although this past research contributes to understanding

the experiences of HNC patients, many studies have includedonly short follow-up periods (most ≤3 months followingtreatment) [14,15,17–19], focused only on depression[9,11,12], did not examine predictors [14,19] and/orincluded mixed HNC treatment groups [9,11,16,21].The aim of this longitudinal study was to follow the

psychological trajectory of an Australian HNC populationup to 18 months after receiving radiotherapy. A previouspaper has summarised the 3-week post-treatment follow-up data [33]. This paper extends our findings with thispatient group up to 18 months after radiotherapy. Thisstudy specifically aimed to: identify rates of depressionor anxiety cases; identify trends in depression and anxietysymptom before and up to 18 months after radiotherapy;and explore what bio-psychosocial factors predict depressionand anxiety scores up to 18 months following radiotherapy.On the basis of past research, factors such as the physicalsymptoms associated with HNC tumour and treatments(e.g. chemotherapy), age, gender, living alone and othermedical treatments for cancer were expected to predictdepression and anxiety post-treatment.

Method

Procedure

This was a prospective longitudinal observational study ofHNC outpatients attending the Peter MacCallum CancerCentre (Melbourne, Australia) between May 2008 andMay 2009. Clinic oncologists assessed patient eligibility.Eligibility criteria were a diagnosis for the first time ofHNC, age >17 years, where the patient agreed to under-take cancer treatment involving radiotherapy with curativeintent. Exclusion criteria included inability to give informedconsent (including active psychosis, intellectual disabilityor dementia), receiving treatment for another cancer, or

minimal understanding of English. Participants wererecruited by two student researchers and a researchofficer. All follow-up contact, including posting outquestionnaires and contacting participants to remind tocomplete questionnaires were completed by the researchofficer. Each patient gave written consent to participate.Ethics approval was obtained from the hospital humanresearch ethics committee.Questionnaire measures were collected at three time

points: (i) pre-treatment (before radiotherapy); (ii) 3 weekspost-treatment (approximately 3 weeks after completingradiotherapy); and (iii) 18 months post-treatment.Eighteen months after treatment was chosen as the finalfollow-up as it was expected that the majority of patientswould have recovered from most of the physical effectsof radiotherapy.

Measures

• A Demographic Questionnaire asked participants torecord their gender, age, marital status, employment,education, current tobacco use and mental healthhistory.

• Cancer information such as diagnosis, tumour his-tology and treatment were obtained from reviewingpatients’ medical records.

• The Hospital Anxiety and Depression Scale (HADS)[7] is a well validated and reliable self-report mea-sure designed to detect the presence and severity ofanxiety and depression [34,35]. The questionnaireitems exclude somatic symptoms and thereforeavoid symptom overlap between somatic illnessesand mood disorders. Respondents are asked to ratetheir symptoms over the previous week. Higherscores on the two subscales (depression and anxiety)indicate more severe symptoms. Scores >7 can beused as a threshold for identifying probable casesof depression. Normative data for the HADS isavailable in a non-clinical sample and show a meanscore of 6.14 for the anxiety subscale and 3.68 forthe depression subscale [36]. A large study of oncol-ogy patients (n= 3035) demonstrated a mean anxietysubscale score 6.76 and depression 4.3 [37].

• The Functional Assessment of Chronic IllnessTherapy-Head and Neck Version [38] is a 39-itemscale developed to measure quality of life inpatients undergoing cancer treatment for HNC. An“Additional Concerns” subscale assesses physicalsymptoms specific to HNC tumours and treatment(ability to eat, dry mouth, difficulty breathing,difficulty speaking or communicating, unhappinesswith physical appearance). Respondents are askedto report their symptoms over the previous week.The subscale score is the sum of the respondent’sratings for the nine items. Higher subscale scoresdenote higher QOL and less physical symptoms. Itis a valid and reliable measure [38].

Statistical analyses

Baseline characteristics of the study sample were sum-marised usingmeans and standard deviations for continuousvariables, and frequencies and percentages for categorical

K. Neilson et al.

Copyright © 2012 John Wiley & Sons, Ltd. Psycho-Oncology (2012)DOI: 10.1002/pon

variables. Differences in baseline characteristics betweenstudy completers and non-completers were analysed usingt-tests.Multilevel mixed-effects linear regressions estimated

the association between predictors and symptoms of (i)depression and (ii) anxiety. This approach recognises themultilevel nature (assessments clustered within patients)of the data and accommodates missing data from patientswho withdrew or became ineligible to continue the study.Predictors included in the models were time (measured inmonths), age, sex, living alone, received chemotherapy,pain and physical symptoms associated with HNC tumoursand treatment.Analyses were performed using STATA Version 12

(StataCorp, Texas, USA) and SPSS Version 16 (SPSSInc, Chicago, Ill USA).

Results

Participants

In total, 172 patients were referred for the study; of these,71 declined to participate and 101 patients completedbaseline measures.Pre-treatment demographic characteristics and cancer

type and treatment of participants are presented in detailin Table 1. Participants consisted of 85 men and 16women. The average age of participants was 63 years(range 37–85 years). Most participants were diagnosed withsquamous cell carcinoma, and all received radiotherapy aspart of treatment. The average duration of radiotherapy +/� chemotherapy was 6 weeks. Most participants hadsurgery and radiotherapy (35%). One participant reporteda pre-treatment psychiatric condition (depression).Of the 101 eligible patients who completed pre-treatment

questionnaires, 75 completed the 3-week post-treatmentquestionnaires (26 withdrew from the study at 3 weekspost-treatment). Of the post-treatment participants whocompleted questionnaires, 23 patients were excluded fromthe 18-month follow-up because of diagnosis of recurrenceor metastases during the first 18 months after treatment(n=10), diagnosis and current treatment for another cancer(n=3), or death (n=10). Patients with recurrent disease orcurrent treatment for other cancers were excluded to reducethe influence of these events on self-reported symptoms. Afurther n=15 withdrew at 18-month follow-up. Thus, atotal of 37 participants completed 18-month follow-up.Participants who completed the study did not differ fromnon-completers (either excluded or withdrew) on age,sex or baseline anxiety. Completers did tend to be lessdepressed at baseline compared with non-completers(t(20) = 2.19, p = .04).

Rates of depression and anxiety cases

The proportion of participants scoring above the clinicalthreshold (>7) on HADS depression and anxiety sub-scales is shown in Figure 1. In terms of HADS depressionscores, 15% of participants reported symptoms above thecut-off for depression pre-treatment, rising to nearly 30%3 weeks post-treatment and dropping to 8% at 18-monthfollow-up. On the anxiety scale, nearly 20% were above

the cut-off pre-treatment, 17% at 3 weeks post-treatmentand nearly 22% at 18-month follow-up.

Changes in depression and anxiety over time

A multilevel mixed-effects linear regression analysed pre-dictors of depression scores. Depression scores did changesignificantly over time; however, this effect disappearedwhen physical symptoms associated with HNC tumoursand treatment was included. The severity of physicalsymptoms associated with HNC tumours and treatment

Table 1. Participants’ pre-treatment demographic characteristics(n= 101)

Characteristic n %

Age35–44 5 545–54 24 2455–64 29 2865–74 26 2675–84 15 1585+ 2 2

Cancer histology typeSCC 82 81Other 19 19

Primary tumour siteOral cavity 33 32Oropharynx 29 29Larynx/hypopharynx 23 23Parotid gland 7 7Paranasal sinus/nasal cavity 4 4Nasopharynx 3 3Thyroid 2 2

Cancer treatment/sRadiation therapy only 13 13Chemo+ radiation therapy 29 28Surgery + radiation therapy 35 35Surgery + chemo+ radiation 24 24

Marital statusMarried 65 64De facto 7 7Never married 10 10Divorced 11 11Widowed 8 8

Lives aloneAlone 25 25Not alone 76 75

EducationPrimary 8 8Secondary 44 44Apprenticeship 19 19Undergraduate 9 9Postgraduate 18 18

Employment statusFull-time 30 30Part-time 6 6Casual 4 4Unemployed 7 7Retired 46 46Pension/sickness benefits 6 6

Income (annual household)$0–$15,599 23 23$15,600–$36,399 22 22$35,400–$78,000 28 28$78,001 or more 20 20

Current smokingNo 89 88Yes 10 10Did not respond 2 2

SCC, squamous cell carcinoma.

Distress 18 months after radiotherapy for head and neck cancer

Copyright © 2012 John Wiley & Sons, Ltd. Psycho-Oncology (2012)DOI: 10.1002/pon

was the only significant predictor of depression scores(b=�.24; 95% CI: (�.30, �.17); p< .01) with morereported physical symptoms associated with higherdepression scores. Change in physical symptoms over timeis presented in Figure 2. No other predictors (time, pain,chemotherapy, age, sex or living alone) were significantlyrelated to depression scores.Anxiety scores also changed significantly over time,

with both time (b=�.86; 95% CI: (�1.19,�.53); p< .01)and time squared (b= .05; 95% CI: .03, .07; p< .01) beingstatistically significant, indicating a decrease in anxietyscores between baseline and 3 weeks post-treatment andan increase in anxiety between 3 weeks post-treatmentand 18-month follow-up. Although mean anxiety scoressignificantly change over time, these means appear tovary less dramatically than depression (refer to Figure 1).Age of participants also predicted anxiety, where anxiety

scores were higher for younger participants (b=�.54;95% CI: (�.10, �.01); p= .03). Physical symptoms pre-dicted anxiety (b=�.09; 95% CI: (�.15, �.02; p= .01),with more physical symptoms also associated with higheranxiety scores. Other predictors were not significantlyrelated to anxiety scores (pain, chemotherapy, age, sex andliving alone).

Discussion

This is the first prospective (up to 18 months) study of pre-dictors of depression and anxiety symptoms for HNCpatients treated with radiotherapy. This longitudinal studydemonstrated risk factors and the change in symptomsof psychological distress (depression and anxiety) overthe course of treatment and recovery for HNC, up to18 months after radiotherapy.Participants’ anxiety levels were lower at 3 weeks

post-treatment compared with baseline and 18 monthspost-treatment. A number of risk factors for anxiety overthe course of treatment were identified, with younger ageand more severe physical symptoms associated withhigher anxiety levels.Depression, when analysed alone, increased signifi-

cantly between baseline and 3 weeks post-treatment, andthen decreased at 18-month follow-up. However, furtheranalysis revealed that variability in depression scoreswas accounted for by participants’ experience of physicalsymptoms associated with HNC tumours and treatment(ability to eat, dry mouth, difficulty breathing, difficultyspeaking or communicating, unhappiness with physicalappearance). Results suggest that as physical symptomsassociated with their cancer and treatment increased, sodid symptoms of depression. This highlights the importantrelationship between HNC patients’ tumour/treatment-related physical symptoms and psychological distress.This finding is supported by previous studies of HNCpatients [9,23].The pattern of change observed in depression scores

over treatment is similar to other studies [15,17–19].

19.817.3

21.6

14.9

29.3

8.1

5.32

4.04

5.16

4.02

5.49

3.54

0

5

10

15

20

25

30

35

40

45

50

Pre-treatment 3-weeks post-treatment

18-months post-treatment

Time

% o

f p

arti

cip

ants

ab

ove

cu

t-o

ff (

bar

s)

0

1

2

3

4

5

6

Mea

n s

ub

scal

e sc

ore

(lin

es)

HADS Anxiety % abovecut off

HADS Depression %above cut-off

HADS Anxiety mean

HADS Depression mean

Figure 1. Percentage of participants scoring above threshold (>7) and means for Hospital Anxiety and Depression Scale (HADS) depressionand anxiety subscales

26.29

19.14

26.15

0

5

10

15

20

25

30

Pre-treatment 3-weeks Post-treatment

18-months Post-

treatment

Time

Mea

n s

ub

scal

e sc

ore

Figure 2. Mean Functional Assessment of Chronic Illness Therapy(FACT)-Additional Concerns subscale scores at each time (lowerscores indicate more physical symptoms)

K. Neilson et al.

Copyright © 2012 John Wiley & Sons, Ltd. Psycho-Oncology (2012)DOI: 10.1002/pon

However, contrary to the current study, many previousstudies have found no significant change in anxiety overtime [17,19]. Other risk factors (such as pain, receivingchemotherapy, sex and living alone) identified by previ-ous studies for distress were not supported as risk factorsin the current study [10,18].Compared with previous studies using the HADS,

patients in the current study reported fewer symptomsof anxiety across time than the general populationand other oncology samples [36]. However, theHNC sample in the current study scored higher onthe depression subscale 3 weeks post-treatment whencompared with a general population [36] and a generaloncology sample [37], with one third of patients report-ing symptoms above the cut-off threshold for depressionat 3 weeks following radiotherapy. This indicates thatHNC patients experience more symptoms of depressionfollowing radiotherapy than other oncology and generalpopulations.Future studies in this area should investigate the role of

other risk factors to identify vulnerable patients, such ascoping style and social support that may also providevaluable information. It would also be useful to gain abroader understanding of participants’ experiences duringtreatment and recovery using qualitative methods. Thiswould further elucidate the relationship between physicalsymptoms and mood symptoms reported on the quantita-tive measures in the current study.This high incidence of symptoms of depression associ-

ated with physical symptoms in this HNC patient grouppoints to the importance for ongoing and routine surveil-lance to monitor psychological distress (particularlysymptoms of depression) in this patient group as theyproceed through medical treatments and rehabilitation,and to provide appropriate interventions. Furthermore,designing interventions aimed at supporting patientsas they develop physical symptoms from treatments andpreventing additional suffering from symptoms of depres-sion would presumably assist this patient group: this is anarea worthy of further scientific study.

A number of limitations of our study should be keptin mind when interpreting results. Over the course ofthe study, 41.5% of participants either dropped out orwere not contactable at follow-up. Furthermore, 22.8%were excluded after completing baseline measures dueto change in eligibility (e.g. diagnosed with metastaticdisease or another cancer, or died). Although this is a lim-itation of the study, these participant retention difficultiesreflect the challenges of working with this patient group.Participants who completed the study were less depressedat pre-treatment compared with participants that droppedout or who became ineligible. This may indicate a self-selection bias, where more depressed patients were lesslikely to complete the study, and therefore 3 weeks and18 months post-treatment results may underestimate ratesof depression in this patient group. However, the currentstudy does include a larger sample than many longitudi-nal studies of HNC patients, particularly given that thisstudy attempted to follow up participants beyond 1 yearpost-treatment.In conclusion, the results of this study indicate that

patients with HNC experience significant changes inpsychological distress, specifically symptoms of depres-sion post-radiotherapy, and these changes are related tothe physical symptoms associated with HNC treatments.One third of HNC patients reported symptoms above thecut-off threshold for depression following radiotherapy.Therefore, ongoing monitoring to identify patients atrisk is important. Patients undergoing radiotherapy forHNC are seen regularly (almost daily) by medical staff,providing opportunities to monitor patients mood andphysical symptoms, and to deliver interventions as needed.Psychosocial interventions need to be developed andevaluated to meet the changing needs of this patientgroup as they undergo treatment for HNC.

Acknowledgements

This research was funded by beyondblue: The National Depression Ini-tiative. Thank you also to KarenMead andMarcelle Gray, who assistedthis project as part of their postgraduate (psychology) dissertations.

References

1. English E, et al. Cancer Survival Victoria:Estimates of Survival in 2004, VictorianCancer Registry, Cancer Epidemiology Centre,Cancer Council Victoria: Melbourne, 2007.

2. Allison P, et al. Teaching head and neckcancer patients coping strategies: results ofa feasibility study. Oral Oncol 2004;40(5):538–44.

3. Fischer D, et al. Anxiety, depression, andpain: differences by primary cancer. SupportCare Cancer 2010;18(7):801–810.

4. Glavassevich M, McKibbon A, Thomas S.Information needs of patients who undergosurgery for head and neck cancer. Can OncolNurs J 1995;5:9–11.

5. Koster M, Bergsma J. Problems and copingbehaviours of facial cancer patients. Soc SciMed 1990;30:569–578.

6. Kallay E, Degi L, Vincze A. Dysfunctionalattitudes, depression and quality of life in asample of Romanian Hungarian cancerpatients. J Cogn Behav Psychot 2007;7(1):97–108.

7. Zigmond A, Snaith R. The hospital anxietyand depression scale. Acta Psychiatr Scand1983;67(6):361–370.

8. American Psychiatric Association. Diagnosticand Statistical Manual of Mental Disorders -Text Revision. American Psychiatric Associa-tion: Washington, 2000.

9. de Leeuw J, et al. Prediction of depression6 months to 3 years after treatment of headand neck cancer. Head Neck 2001;23(10):882–898.

10. Haisfield-Wolfe ME, et al. Prevalence andcorrelates of depression among patients withhead and neck cancer: a systematic review ofimplications for research. Oncol Nurs Forum2009;36(3):E107–25.

11. Karnell LH, et al. Persistent posttreatmentdepressive symptoms in patients with head andneck cancer. Head Neck 2006;28(5):453–461.

12. Sehlen S, et al. Depressive symptoms duringand after radiotherapy for head and neckcancer. Head Neck 2003;25(12):1004–1018.

13. Singer S, et al. Predictors of emotionaldistress in patients with head and neck cancer.Head Neck 2012;34(2):180–187.

14. Chawla S, et al. Temporal assessment ofquality of life of head and neck cancerpatients receiving radical radiotherapy. QualLife Res 1999;8:73–78.

15. Chen A, et al. Prospective study of psychoso-cial distress among patients undergoing radio-therapy for head and neck cancer. Int J RadiatOncol Biol Phys 2009;73(1):187–193.

16. Infante-Cossio P, et al. Impact of treatment onquality of life for oral and oropharyngeal car-cinoma. Int J Oral Maxillofac Surg 2009;38(10):1052–1058.

17. Kelly C, et al. Deterioration in qualityof life and depressive symptoms duringradiation therapy for head and neck cancer.Otolaryngol Head Neck Surg 2007;136(1):108–111.

18. Kohda R, et al. Prospective studies onmental status and quality of life in patientswith head and neck cancer treated by radia-tion. Psycho-Oncology 2005;14:331–336.

19. Rose P, Yates P. Quality of life experiencedby patients receiving radiation treatment forcancers of the head and neck. Cancer Nurs2001;24(4):255–263.

Distress 18 months after radiotherapy for head and neck cancer

Copyright © 2012 John Wiley & Sons, Ltd. Psycho-Oncology (2012)DOI: 10.1002/pon

20. Derks W, et al. Elderly patients with head andneck cancer: physical, social and psychologi-cal aspects after 1 year. Acta Otolaryngol2004;124(4):509–514.

21. Ronis D, et al. Changes in quality of life over1 year in patients with head and neck cancer.Arch Otolaryngol Head Neck Surg 2008;134(3):241–248.

22. Britton B, et al. Baseline depression predictsmalnutrition in head and neck cancer patientsundergoing radiotherapy. Support Care Cancer2012;20(2):335–342.

23. de Leeuw J, et al. Prediction of depressivesymptomatology after treatment of head andneck cancer: the influence of pre-treatmentphysical and depressive symptoms, coping,and social support. Head Neck 2000;22(8):799–807.

24. Hawton K, van Heeringen K. Suicide. Lancet2009;373:1372–1381.

25. Zeller J. High suicide risk found for patientswith head and neck cancer. J Am Med Assoc2006;296(14):1716–1717.

26. Massie M, Popkin M. Depressive disorders.In Psycho-Oncology, Holland JC (ed.). OxfordUniversity Press: New York, 1998.

27. Newport D, Nemeroff C. Assessment andtreatment of depression in the cancer patient.J Psychosom Res 1998;45:215–237.

28. Pirl W, Roth A. Diagnosis and treatment ofdepression in cancer patients. Oncology1999;9:1293–1306.

29. Katz M, et al. Psychosocial adjustment inhead and neck cancer: the impact of disfigure-ment, gender and social support. Head Neck2003;25(2):103–112.

30. Kugaya A, et al. Prevalence, predictivefactors, and screening for psychologicdistress in patients with newly diagnosedhead and neck cancer. Cancer 2000;88(12):2817–2823.

31. Tromp D, et al. Medical care-seeking andhealth-risk behavior in patients with headand neck cancer: the role of health value, con-trol beliefs and psychological distress. HealthEduc Res 2005;20:665–675.

32. Bilodeau B, Degner L. Informational needs,sources of information, and decisional rolesin women with breast cancer. Oncol NursForum 1996;23:691–696.

33. Neilson K, et al. Psychological distress(depression and anxiety) in people with head

and neck cancers. Med J Aust 2010;193(5):s48–s51.

34. Hermann C. International experiences withthe Hospital Anxiety and DepressionScale—a review of validation data andclinical results. J Psychosom Res 1997;42:17–41.

35. Luckett T, et al. A review and recommenda-tions for optimal outcome measures ofanxiety, depression and general distress instudies evaluating psychosocial interventionsfor English-speaking adults with heterogeneouscancer diagnoses. Support Care Cancer2010;18(8):1241–1262.

36. Crawford J, et al. Normative data for theHADS from a large non-clinical sample. BrJ Clin Psychol 2001;40:429–434.

37. Sellick S, Edwardson A. Screening newcancer patients for psychological distressusing the hospital anxiety and depressionscale. Psycho-Oncology 2007;16:534–542.

38. List M, et al. The performance status scale forhead and neck cancer patients and the func-tional assessment of cancer therapy-head andneck scale. A study of utility and validity.Cancer 1996;77:2294–2301.

K. Neilson et al.

Copyright © 2012 John Wiley & Sons, Ltd. Psycho-Oncology (2012)DOI: 10.1002/pon