This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Stark, D. Articles by Selby, P. Search for Related Content PubMed PubMed Citation Articles by Stark, D. Articles by Selby, P. Social Bookmarking                            What's this?

Anxiety Disorders in Cancer Patients: Their Nature, Associations, and Relation to Quality of Life

From the Cancer Research UK Clinical Centre at Leeds, St James’s University Hospital, and Academic Unit of Psychiatry and Behavioural Sciences, School of Medicine, University of Leeds, Leeds, United Kingdom.

Address reprint requests to D. Stark, MD, Cancer Research UK Clinical Centre at Leeds, St James’s University Hospital, Beckett St, Leeds LS9 7TF, United Kingdom; email: csjds{at}cancermed.leeds.ac.uk

	ABSTRACT

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
PURPOSE: We aimed to estimate the prevalence and types of anxiety disorders diagnosed according to standardized criteria in cancer patients, to compare screening tools in detecting them, and to examine their demographic, oncologic, and psychosocial associations.

METHODS: In this cross-sectional observational study of 178 subjects with lymphoma, renal cell carcinoma, malignant melanoma, or plasma cell dyscrasia, we related responses to questionnaires (administered by computer touch-screen) measuring psychological symptoms, quality of life (QOL), and social support to standardized psychiatric interviews and cancer management.

RESULTS: Forty-eight percent of subjects reported sufficient anxiety for anxiety disorder to be considered. At subsequent diagnostic interview, 18% fulfilled International Classification of Disorders, 10th Revision criteria for anxiety disorder, including 6% of patients who reported low levels of anxiety by questionnaire. When subjects reported anxiety by questionnaire, if disruptive somatic anxiety was present, this increased the probability of diagnosable anxiety disorder from .31 to .7. The most accurate screening questionnaires were the trait scale of the State-Trait Anxiety Inventory and the Hospital Anxiety and Depression scale. Female sex and negative aspects of social support were associated with anxiety disorder in multivariate analyses. Anxiety disorder was independently associated with a deficit in QOL, particularly insomnia.

CONCLUSION: Anxiety symptoms are common in cancer patients. Screening by questionnaire seems to assess anxiety symptoms adequately but discriminates abnormal anxiety inadequately. To improve this, we may need to use criteria such as disruption from anxiety, as illustrated by the impact of anxiety disorders on QOL. There seem to be few oncologic variables that could target screening for anxiety disorders.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
ANXIETY IS A response to a threat, cancer is threatening, and so many patients are anxious. In one study, 77% of 913 patients within 2 years of treatment recalled experiencing anxiety.¹ However, anxiety after cancer diagnosis is not necessarily abnormal, may not present a problem, or may even be a constructive part of dealing with problems.² An understanding of the nature of the anxiety in cancer patient populations is important because abnormal anxiety is disruptive³ and amenable to pharmacologic and psychological treatment.⁴

Our limited understanding of anxiety in cancer care is illustrated by the wide range of prevalence estimates of abnormal anxiety in cancer patient populations. This varied from 0.9% to 49% in one review of the literature,⁵ although in large studies using standardized psychiatric interviews and applying research diagnostic criteria the range is narrower, from 10% to 30%.⁶ The potential explanations for this range include the definition of morbid anxiety, the range of self-report measures applied without a clear understanding of their relation to morbid anxiety, social or demographic factors, and aspects of cancer diagnosis and treatment that may be associated with morbid anxiety.

Anxiety is characterized by a number of typical symptoms and signs. Autonomic overactivity is important, with symptoms such as palpitation or tremor. The World Health Organization International Classification of Disorders (ICD-10) and other classification systems used in psychiatry require a core of anxiety symptoms manifesting autonomic overactivity to be present, and that anxiety also be abnormal, to fulfill a diagnosis of anxiety disorder. Abnormality is defined by symptoms that are out of proportion to the level of threat, which persist or deteriorate without intervention and are causing disruption, such as emotional distress or disruption of functioning.

These criteria are hard to apply in cancer care. There is usually a very real threat. There are few longitudinal studies of the natural history of anxiety disorders. Disruption and functional impairment may therefore be areas that need to be emphasized if we wish to identify abnormal anxiety more reliably. Because one of the purposes of the measurement of self-reported quality of life (QOL) is to measure health in terms of function, a specific QOL deficit associated with anxiety disorders might contribute to our ability to characterize abnormal anxiety in cancer patients.

Anxiety disorders take several forms. The anxious adjustment disorder is a quantitatively excessive response starting within 1 month of a stressful event. Generalized anxiety disorder (GAD) requires more symptoms than anxious adjustment disorder and persistence of symptoms over 6 months. Often the anxiety seen in these disorders is free-floating, meaning that it occurs without a particular trigger or a crescendo pattern. In panic disorder, anxiety builds up to a rapid crescendo. Phobic anxiety only occurs in relation to a provoking stimulus, so anticipatory avoidance is common. Phobias may arise in relation to hospitals or treatments, or simple phobias of animals or social settings may predate the cancer. Most anxiety disorders are grouped descriptively based on these criteria, but when abnormal anxiety is clearly linked to an organic trigger, it is called organic anxiety, whatever its features. Organic anxiety in cancer care can be caused by drug treatments including interferon.⁷ Anxiety may also be present in association with depression.⁸

The primary responsibility for the recognition of anxiety disorders in cancer patients remains with cancer care specialists. Cancer care clinicians remain poor at detecting their patients’ psychological problems.⁹ Many questionnaires have been studied in the detection of psychological distress¹⁰ and depression in cancer patients.¹¹ They all perform only moderately or poorly when standardized psychiatric interviews are used as the gold standard,¹² and their use does not in itself improve the outcome for depressed or anxious patients.¹³ There is limited explanation for these disappointing results. A number of self-report questionnaires are available to measure anxiety specifically, but we do not know clearly how they compare in the detection of abnormal anxiety.

One way to improve identification of anxiety disorders may be to identify subgroups with increased risk. Younger people, women, and people from lower socioeconomic groups in the general population are at high risk of anxiety.¹⁴ Anxiety symptom levels are high soon after the onset of cancer but reduce over time.¹⁵ Cancer treatments are associated with anxiety, but this is highly dependent on the specific circumstances.^16-18 Studies in cancer have usually measured anxiety as a continuum rather than identifying morbid levels, so it is uncertain whether anxiety disorders or adaptive normal anxiety are influenced by such circumstances.

Despite this literature on anxiety in cancer patients, many important questions remain unanswered, and in this study we aim to characterize the anxiety reported by cancer patients using standardized diagnostic criteria for anxiety disorder; to estimate the prevalence and types of anxiety disorders found in cancer patients; to compare screening tools in detecting these anxiety disorders; and to examine the demographic, oncologic, and psychosocial associations of anxiety disorders in cancer patients, including any deficit in QOL.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
Measures
Subjects completed questionnaires presented in electronic format by using computers with touch-screen monitors before medical consultation in the outpatient clinic. These assessed their own reports of anxiety and depression with the Hospital Anxiety and Depression Scale, past psychiatric history, QOL with the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire C30 (EORTC QLQC30), cancer-related concerns using the cancer concerns checklist, and social support with the Close Persons Questionnaire. Within 3 weeks, in their own home or a hospital consulting room (at subjects’ preference), further measures were made, including further anxiety measurement with the State-Trait Anxiety Inventory and a formal assessment of psychiatric problems by Present State Examination. All consenting subjects completed all measures and were interviewed, as summarized in Table 1.

Sociodemographic data were assessed by patient report for all patients as shown in Table 2. Socioeconomic status was estimated from postcode by using the superprofile system, a classification of households into one of 10 categories according to the neighborhood (approximately 200 houses) where they are located. The 10 categories were generated from cluster analyses of data from the 1991 United Kingdom census and ranked by affluence. The groups demonstrate features such as more deprived districts in inner city than rural areas and consistency with other measures of poverty such as the Carstairs Index.²⁸ The classification is associated with morbidity and mortality data including that from psychiatric problems in Yorkshire and can be applied through a postal code directory.^29,30 Clinicians reported the World Health Organization performance status of subjects and the presence and severity of treatment toxicity after consultation. We measured the timing and nature of events in the cancer diagnosis and treatment by analysis of electronic patient records, a process which we have previously demonstrated to have good intra- and interrater reliability.²⁶ Timelines of events were generated, although for many patients exact dates of events are not available, so uncertain dates were all rounded to the first of the month.

Setting
The Leeds Cancer Centre manages cancers from the surrounding area and also acts as a referral center for several less common cancers from a wider area of Northeast England. The unit operates cancer-site–specialized clinics. To allow all patients to be offered entry onto the study, we examined clinic attendance lists the day before the clinics ran, and eligible subjects were identified then from case records. When the numbers were too high, a deferred strategy was brought into operation. The strategy was based solely on the interval of time between the current and the previous clinic attendance, with patients with longer follow-up intervals being approached first. Those not approached were noted and approached at a later visit. Patients were approached before medical consultation, the study was discussed and written information was given, but consent was not sought at that attendance. All identified subjects were approached again on returning to the clinic, to seek informed and written consent to enter the study. Several wished to defer study entry to a later date. This was permitted, but recorded, and is discussed later. Ethical approval was obtained from the appropriate institutional ethical review committees.

Analysis
Statistical analyses were carried out using SPSS Version 9. Receiver operating characteristic (ROC) analysis was performed to compare self-report questionnaire responses with psychiatric diagnoses. They illustrate the sensitivity and specificity of the instrument for detecting a case of anxiety disorder at each cutoff value. Efficient screening instruments are indicated by ROC curves that have a high area under the curve (AUC). A graph at the diagonal line with an AUC of 0.5 is screening by chance alone. To maximize sensitivity and specificity (give the lowest overall rate of misclassification), the optimum cutoff score corresponds to the point nearest to the top and left of the graph if false-positive and false-negative results are of equal importance, but the choice of cutoff depends on the role of the questionnaire. The most accurate cutoff might be selected if the questionnaire response is being used to directly offer treatment. Alternatively, questionnaire responses may be used to alert clinicians to a problem, which they then appraise. In this case, high sensitivity and minimum false-negative responses are required.

Logistic regression analyses were performed to identify univariate associations of anxiety disorders from among the sociodemographic and oncologic parameters. Age, superprofile, cancer diagnosis, time since diagnosis, time since informed of last relapse, extent of the cancer, number of organ sites involved with metastatic disease, number of previous relapses, current treatment, clinical trend in disease growth or shrinkage, and performance status were entered into this regression, whatever the extent of any univariate association, because they were of interest a priori. All other variables associated at a significance level of less than .2 were entered simultaneously into appropriate multivariate models.³¹ Statistical significance of associations in logistic regression was assessed by Wald statistic and likelihood ratio test.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
Over the 9-month period, 305 eligible patients considered entering the study; 210 of those consented to screening for anxiety. Six patients were ineligible due to physical disability and six due to dementia. Thirty-two who consented were unable or unwilling to comply with the home interviews within 3 weeks. Complete data are presented on 178 patients. The mean age was 54.9 years, 60% were male, and 14% had attended further education after age 18. Half had experienced a drop in their household income since cancer diagnosis. The subjects were a mean of 2.9 years since diagnosis (range, 0.1 to 23.7 years) (Table 2).

To assess recruitment bias, we conducted an analysis of demographic data in an externally validated administrative database containing all 498 patients with these cancers managed within our unit over the last 3 years. There was no statistically significant difference between our cohort and the administrative database sample by sex (60.1% v 56.2% male, ² = 0.81, P = .37) or socioeconomic distribution (t = 0.36, P = .72), but the recruited cohort was younger than the group from the administrative database (54.9 years v 58.1 years, t = 2.67, P = .01). Nineteen subjects wished to defer study entry as described above. There was no statistically or clinically significant difference between this group and the others in the cohort on the sociodemographic, oncologic, or psychological variables measured (data not presented). None of the results or conclusions of the analyses presented are significantly altered when these individuals’ data are removed, so the data presented includes all 178 consenting subjects. Table 2 describes the cohort, along with the distribution of the main variables.

Eighty-five subjects (48%) reported HAD-A scores over 7. Thirty-two (18%) fulfilled ICD-10 criteria at SCAN interview for anxiety disorder and 27 (15%) fulfilled criteria for depression. Of the 32 fulfilling anxiety disorder diagnoses, 26 were in the group with HAD-A scores over 7. Examination of the SCAN interview data for the six subjects with anxiety disorders but low levels of HAD-A–reported anxiety indicated that three had longstanding diagnoses of phobic anxiety related to simple exposures, such as animals or heights, and did not experience anxiety in medical situations. Of the other three, one had GAD, one had phobic anxiety related to medical situations, and one had panic disorder. Figure 1 illustrates the anxiety described at interview by the low and high self-reported anxiety groups. Reasons for failure to fulfill ICD-10 psychiatric criteria differed in the two groups. In the high self-reported anxiety group, the commonest reason for failure to fulfill ICD-10 diagnostic criteria was that anxiety was not disruptive enough to be rated clearly abnormal, often because there was no report that the subject viewed the anxiety as undesirable or interfering in life. In contrast in the low self-reported anxiety group, most subjects had too few anxiety symptoms to fulfill psychiatric diagnostic criteria.

View larger version (16K): [in this window] [in a new window]   Fig 1. The comparison of self-reported anxiety to ICD-10 criteria for anxiety disorder.

The onset of the current anxiety disorder was assessed at the SCAN interview. For 31% of subjects with anxiety disorders, the current episode began more than 6 months before the date of cancer diagnosis, with 80% of these episodes being over 2 years long. For 38% of those with anxiety disorders, the episode began during the 1-year period running from 6 months before to 6 months after their cancer diagnosis, when symptoms may have been developing and being investigated. For 31%, the episode began more than 6 months after the cancer was diagnosed.

Table 3 summarizes the responses to the self-report questionnaires measuring anxiety. The scores on the State Trait Anxiety Inventory (STAI) for cancer patients with anxiety disorders were comparable to prior data from subjects without cancer.³² When those with anxiety disorders were compared with those without, the emotional function scores of the QLQC30 and both subscales of the STAI demonstrate a statistically significant difference (two-sided t test; P < .05). This is not the case for the cancer concerns checklist responses.

View larger version (22K): [in this window] [in a new window]   Fig 2. ROC curves relating self-report questionnaire responses to ICD-10 psychiatric diagnosis of anxiety disorder. ROC curves illustrate the efficacy of a screening technique. Better screening tools have curves further to the left measured by a greater AUC. The optimum cutoff for a screening measure is the point nearest the top left of the curve, if false-positive and false-negative results are of equal importance. A tool screening by chance alone would have a ROC curve along the diagonal line. The 95% confidence intervals are based on asymptotic assumptions, for which the sample size is borderline.33 The diagonal segments are produced by ties. *The interviewer was aware of the anxiety score on the HAD-A scale when the SCAN interview to ascertain anxiety disorders took place.

Within the group with HAD-A scores higher than 7, demographic, social, and oncologic parameters and responses to alternative questionnaires were examined as independent discriminators of morbid anxiety. Neither emotional functioning nor insomnia from the QLQC30 improved discrimination of morbid anxiety over and above scores on the HAD-A questionnaire.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
Almost half of cancer patients report sufficient anxiety by questionnaire for a diagnosis of anxiety disorder to be considered. When validated semistructured interviews and standardized ICD-10 diagnoses were used, for 70% this anxiety did not meet criteria for anxiety disorder. Sometimes the number of symptoms was insufficient, but also disruption caused by the anxiety frequently distinguished normal anxiety from anxiety disorder. This may be due to features of the questionnaires. A diagnosis of anxiety disorder requires symptoms of autonomic overactivity to be present and that anxiety to be abnormal, causing disruption such as emotional distress or disruption of functioning. This may be insufficiently assessed by the questionnaires. The HAD-A scale covers seven symptoms, but only one specific to autonomic overactivity in anxiety. Four items cover tension and worry, which are also assessed during the SCAN interview but as nonspecific symptoms common to anxiety, depression, and other psychiatric problems. Within the HAD-A, only one question measures the disruption from a symptom as well as its presence and frequency. For the patients completing the questionnaire, therefore, symptoms may be present but not causing problems or disruption. They report these symptoms by questionnaire, but this may be normal anxiety. Similar observations apply to the STAI questionnaires.

Anxiety disorders were present in 18% of cancer patients in this study. This confirms the work of others using similar methods.⁶ We knew little previously about the types of abnormal anxiety seen. The relative frequencies of the different types of anxiety disorder are broadly similar to those in the general population.⁸ For those with a single type of anxiety, phobic anxiety was the most prevalent, although the largest group was those with several forms of anxiety occurring together. Although 41 subjects (23%) were receiving biologic therapies such as interferon and interleukin-2, only two of 32 anxiety disorders were attributable to the psychological effects of these agents.

Twelve (38%) of 32 subjects with anxiety disorders had coexistent depression. In this group, an even greater majority of the subjects had multiple types of anxiety together. In primary care, 45% of subjects with anxiety disorders have comorbid depression,³⁴ and in the general population, 26%.³⁵ Therefore, anxiety disorders in cancer patients represent a substantial problem in their own right, over and above that which would be treated were depressed patients being referred to specialized services and treated. Over two thirds of the anxiety disorders detected had a point of onset that suggests a relation to the processes of cancer diagnosis or treatment.

None of the self-report questionnaires performed well in detecting anxiety disorders. In particular, several had very low specificity when applying sensitive cutoff values that are needed to identify these important and treatable problems.⁴ The trait scale of the STAI demonstrated marginally better efficacy at its most accurate cutoff than the anxiety scale of the HAD. The HAD-A is, however, substantially shorter at seven questions rather than 20 and designed as a screen rather than a measurement instrument.

There were few sociodemographic, psychological, or oncologic associations of anxiety disorder that seemed strong enough to be clinically useful predictors. Anxiety was only weakly related to oncologic parameters and only weakly related to demographic parameters, unlike in the general population. A weaker association between demographic parameters and psychiatric morbidity has previously been noted in the presence of physical illness. It has been suggested the demographic differences that predispose to anxiety disorders after life events in the general population become less relevant when a very severe stressor occurs, such as cancer.³⁶ The association with sex that is observed remains in multivariate analyses. Therefore, this may help clinicians to identify individuals with anxiety disorder. The specific relation of anxiety disorder to negative aspects of social support (inadequate support or problems emanating from those providing support) is also potentially useful in improving identification and might be included within the oncologic consultation. The model these analyses generated predicts the status correctly for a large proportion of the patients, but these predictions are derived from and applied to the same population, so should be viewed with caution until further examined.

Anxiety disorders were associated with a functional deficit in cancer-specific QOL, as has been observed by others.³⁷ Much of that deficit is probably not caused by the anxiety disorder itself because several associations were not present after adjustment for sex, depression, and social support. Insomnia was independently associated with anxiety disorder, even though sleep disturbance was not part of the diagnosis of anxiety disorder within the SCAN interview. Insomnia is an important and disruptive problem, relatively little explored in detail, a symptom of anxiety and other forms of psychological distress, but not previously observed to be closely related.³⁸ Any causal inference from this association should be cautious and await at least the results of our longitudinal study of anxiety (currently undergoing analysis). The association with insomnia was not strong enough for its inclusion to improve the ability of multivariate models to identify morbid anxiety.

The design of this study has both strengths and weaknesses. The proportion of patients consenting and completing the study was disappointing, in keeping with previous experience.^39,40 Unmeasured biases in this cohort due to the consent rate cannot be excluded. The patients studied were marginally younger than the whole population under the care of our unit with the same cancer types. However, age was not a factor in predicting anxiety disorder, so this had minimal impact on estimates of the prevalence of anxiety disorder. The interviewer was aware of the scores on the HAD questionnaire before conducting the SCAN interview, which may have inflated the ROC results for all measures.

There are limitations to the generalization of these results. The cancers chosen for study are not the most common, although the non-Hodgkin’s lymphomas are a rapidly increasing problem.⁴¹ In previous investigations in cancer care, the predictors of psychological morbidity have not been cancer site itself, but demographic, social, psychiatric, or related to cancer through symptoms or disability.^5,42 We chose these cancer sites because they include a broad range of natural histories, social groups, ages of presentation, and treatment intent from the palliative to the curative. They include the use of cytotoxics and the important issue of biologic agents with their documented psychological adverse effects. Despite this broad range of parameters, the independent associations of anxiety disorders that remain in this multivariate analysis are social and demographic, with a weak effect of cancer type only. There were only a small number of patients with severe treatment toxicity, a reflection that the subjects were outpatients.

The study used computerized techniques for measurement of psychological symptoms, QOL, and social support. As we have previously reported, this is feasible, is acceptable to patients, produces results that are similar to paper-based questionnaires, and reflects clinical events.⁴³ Our results include subjects who deferred study entry at their own choice. External measurable criteria were used to sequence study entry, whereas every effort was made to ensure anxiety level did not influence for or against recruitment. This doesn’t seem to have introduced bias into the results, although we cannot be certain. We used a two-stage technique for case finding with a screening self-report questionnaire followed by a validated interview. This has the advantage of being comparable to large international community-based studies of psychiatric morbidity.³⁴

The new findings in these results have implications for detecting anxiety and hence for the care of cancer patients. The HAD scale was able to exclude the presence of abnormal anxiety for 52% of subjects at the expense of six cases missed, of which three were simple phobias. Formal psychiatric assessment of the remaining 48% would present a logistic problem in clinical practice. If we are to improve screening for anxiety disorders, we may need a different approach. The STAI and HAD measure anxiety by the number of symptoms present and their frequency. One approach may be to improve the questionnaires. Analyses of the HAD scale using item-response theory demonstrate that the certain items within the HAD-A, such as "I get sudden feelings of panic," are good discriminators between people with different levels of anxiety. However, items such as "I can sit at ease and feel relaxed" are much less useful (Smith et al, manuscript submitted for publication). It may be that a branching questionnaire, which provides the most discriminating items where necessary based on earlier answers, can increase the specificity of self-report instruments without such loss of sensitivity. An alternative approach is for questionnaires to assess whether anxiety symptoms are abnormal. This study suggests disruption from anxiety may be a useful parameter to explore. This approach was in the development of the Anxiety Screening Questionnaire, although the efficacy of this technique of self-report screening has yet to be demonstrated.⁴⁴

Alternatively, clinicians can discriminate abnormal anxiety. A problem occurring in around one in five patients has a strong case for inclusion in the routine training of oncologists, but there are many competing pressures on limited training time. The data presented here suggest that a combination of these two approaches may be most useful. Clinicians could receive training in which areas of simple clinical inquiry the evidence suggests are most discriminating when cancer patients report anxiety on questionnaires, perhaps the presence of disruption from somatic anxiety or problems with social support. Specific guidance as to the physician’s assessment after a positive screening test for psychiatric morbidity has been used previously, notably using the PRIME-MD screen.⁴⁵ This involves an initial self-report questionnaire, followed by a structured interview that the physician uses when the questionnaire is positive. When used in primary care, this technique proved highly accurate (accuracy 88%) and most consultations in those with positive screening tests took under 20 minutes. In a small study in cancer patients, this has also proven feasible and accurate.⁴⁶

Our data illustrate the potential for a very simple clinical assessment of the presence of abnormal somatic anxiety when a short self-report questionnaire suggests anxiety is present. In this study, the probability of a patient having an anxiety disorder was 18 in 100. After screening with the HAD-A, a score above 7 increased the posttest probability of anxiety disorder being present to 31 in 100. Six cases among 32, 19%, are missed if this is the only screening approach. If the patient confirmed disruption from somatic anxiety, the posttest probability increased to 70 in 100, with still 19% of cases missed. Although these data would likely be overestimates of the benefits if prospective studies were in place, they compare very favorably with current oncologic practice, where misclassification rates are from 26% to 36%.^9,47 Such inquiry would require relatively simple clinical training and has the potential to substantially improve detection of anxiety disorders in cancer patients.

	ACKNOWLEDGMENTS

 
Supported by grants from the St James’s and Seacroft Hospitals Special Trustees, Cancer Research UK, and the National Lotteries Charities Board.

	NOTES

 
The software was designed by A.S., who can be contacted for further information at the Cancer Research UK Clinical Centre at Leeds, Leeds, United Kingdom.

	REFERENCES

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
1. Ashbury FD, Findlay H, Reynolds B, et al: A Canadian survey of cancer patients’ experiences: Are their needs being met? J Pain Symptom Manage 16: 298-306, 1998[CrossRef][Medline]

2. Andersen BL, Tewfik HH: Psychological reactions to radiation therapy: Reconsideration of the adaptive aspects of anxiety. J Pers Soc Psychol 48: 1024-1032, 1985[CrossRef][Medline]

3. Sherbourne CD, Wells KB, Judd LL: Functioning and well-being of patients with panic disorder. Am J Psychiatry 153: 213-218, 1996[Abstract/Free Full Text]

4. Sheard T, Maguire P: The effect of psychological interventions on anxiety and depression in cancer patients: Results of two meta analyses. Br J Cancer 80: 1770-1780, 1999[CrossRef][Medline]

5. van’t Spijker A, Trijsburg RW, Duivenvoorden HJ: Psychological sequelae of cancer diagnosis: A meta-analytical review of 58 studies after 1980. Psychosom Med 59: 280-293, 1997[Abstract/Free Full Text]

6. Stark DP, House A: Anxiety in cancer patients. Br J Cancer 83: 1261-1267, 2000[CrossRef][Medline]

7. Trask PC, Esper P, Riba M, et al: Psychiatric side effects of interferon therapy: Prevalence, proposed mechanisms, and future directions. J Clin Oncol 18: 2316-2326, 2000[Abstract/Free Full Text]

8. Jenkins R, Bebbington P, Brugha TS, et al: British psychiatric morbidity survey. Br J Psychiatry 173: 4-7, 1998[Free Full Text]

9. Fallowfield L, Ratcliffe D, Jenkins V, et al: Psychiatric morbidity and its recognition by doctors in patients with cancer. Br J Cancer 84: 1011-1015, 2001[CrossRef][Medline]

10. Razavi D, Delvaux N, Farvacques C, et al: Screening for adjustment disorders and major depressive disorders in cancer in-patients. Br J Psychiatry 156: 79-83, 1990[Abstract/Free Full Text]

11. Sellick SM, Crooks DL: Depression and cancer: An appraisal of the literature for prevalence, detection, and practice guideline development for psychological interventions. Psycho-Oncology 8: 315-333, 1999[CrossRef][Medline]

12. Hall A, Ahearn R, Fallowfield L: Are we using appropriate self-report questionnaires for detecting anxiety and depression in women with early breast cancer? Eur J Cancer 35: 79-85, 1999[CrossRef][Medline]

13. Gilbody SM, House AO, Sheldon TA: Routinely administered questionnaires for depression and anxiety: Systematic review. Br Med J 322: 406-409, 2001[Abstract/Free Full Text]

14. Mayou R, Hawton K: Psychiatric disorder in the general hospital. Br J Psychiatry 149: 172-190, 1986[Abstract/Free Full Text]

15. Fallowfield LJ, Hall A, Maguire P, et al: Psychological effects of being offered choice of surgery for breast cancer. Br Med J 309: 448, 1994 (comment)[Free Full Text]

16. Schag CA, Heinrich RL: Anxiety in medical situations: Adult cancer patients. J Clin Psychol 45: 20-27, 1989[Medline]

17. Hughes JE: Psychological and social consequences of cancer. Cancer Surv 6: 455-475, 1987[Medline]

18. Fallowfield L, Fleissig A, Edwards R, et al: Tamoxifen for the prevention of breast cancer: Psychosocial impact on women participating in two randomized controlled trials. J Clin Oncol 19: 1885-1892, 2001[Abstract/Free Full Text]

19. Herrmann C: International experiences with the Hospital Anxiety and Depression Scale: A review of validation data and clinical results. J Psychosom Res 42: 17-41, 1997[CrossRef][Medline]

20. Osoba D, Rodrigues G, Myles J, et al: Interpreting the significance of changes in health-related quality-of-life scores. J Clin Oncol 16: 139-144, 1998[Abstract/Free Full Text]

21. Harrison J, Maguire P, Ibbotson T, et al: Concerns, confiding and psychiatric disorder in newly diagnosed cancer patients: A descriptive study. Psycho-Oncology 3: 173-179, 1994

22. Harrison J, Maguire P: Predictors of psychiatric morbidity in cancer patients. Br J Psychiatry 165: 593-598, 1994[Abstract/Free Full Text]

23. Stansfeld S, Marmot M: Deriving a survey measure of social support: The reliability and validity of the close persons questionnaire. Soc Sci Med 35: 1027-1035, 1992[CrossRef][Medline]

24. World Health Organization Schedules for Clinical Assessment in Neuropsychiatry, Version 2. Geneva, Switzerland, World Health Organization, 1994

25. Aaronson NK, Ahmedzai S, Bergman B, et al: The European Organization for Research and Treatment of Cancer QLQ-C30: A quality-of-life instrument for use in international clinical trials in oncology. J Natl Cancer Inst 85: 365-376, 1993[Abstract/Free Full Text]

26. Velikova G, Wright EP, Smith AB, et al: Self-reported quality of life (QL) of individual cancer patients: Concordance of results with clinical course and medical records. Eur J Cancer 35: 998-1107, 1999

27. Zigmond AS, Snaith RP: The hospital anxiety and depression scale. Acta Psychiatr Scand 67: 361-370, 1983[Medline]

28. Carstairs V, Morris R: Deprivation and health. Aberdeen, United Kingdom, Aberdeen University Press, 1991

29. Yorkshire Public Health: A Census Based View of the Population and Its Health. Harrogate, United Kingdom, Yorkshire Health, 1994

30. Forman D, Rider L: Cancer in Yorkshire: Cancer Registry Report—Cancer Statistics 1989-93. Leeds, United Kingdom, Leeds Graphic Press, 1996

31. Altman D: Practical Statistics for Medical Research. London, United Kingdom, Chapman and Hall, 1996

32. Spielberger CD: The Manual for the State-Trait Anxiety Inventory. Palo Alto, CA, Consulting Psychologists Press, 1983

33. Altman DG: ROC curves and confidence intervals: Getting them right. Heart 83: 236, 2000 (letter)[Free Full Text]

34. Sartorius N, Ustun TB, Lecrubier Y, et al: Depression comorbid with anxiety: Results from the WHO study on psychological disorders in primary health care. Br J Psychiatry 38-43, 1996 (suppl)

35. Regier DA, Rae DS, Narrow WE, et al: Prevalence of anxiety disorders and their comorbidity with mood and addictive disorders. Br J Psychiatry 24-28, 1998 (suppl)

36. Noyes RJ, Holt C, Massie M: Anxiety disorders, in Holland JC (ed): Psycho-Oncology. New York, NY, Oxford University Press, 1998, pp 548-563

37. Skarstein J, Aass N, Fossa SD, et al: Anxiety and depression in cancer patients: Relation between the Hospital Anxiety and Depression Scale and the European Organization for Research and Treatment of Cancer Core Quality of Life Questionnaire. J Psychosom Res 49: 27-34, 2000[CrossRef][Medline]

38. Savard J, Morin CM: Insomnia in the context of cancer: A review of a neglected problem. J Clin Oncol 19: 895-908, 2001[Abstract/Free Full Text]

39. Hopwood P, Stephens RJ: Depression in patients with lung cancer: Prevalence and risk factors derived from quality-of-life data. J Clin Oncol 18: 893-903, 2000[Abstract/Free Full Text]

40. Cella DF, Orofiamma B, Holland JC, et al: The relationship of psychological distress, extent of disease, and performance status in patients with lung cancer. Cancer 60: 1661-1667, 1987[CrossRef][Medline]

41. Cartwright R, Brincker H, Carli PM, et al: The rise in incidence of lymphomas in Europe 1985-1992. Eur J Cancer 35: 627-633, 1999[CrossRef][Medline]

42. Aass N, Fossa SD, Dahl AA, et al: Prevalence of anxiety and depression in cancer patients seen at the Norwegian Radium Hospital. Eur J Cancer 33: 1597-1604, 1997[CrossRef][Medline]

43. Velikova G, Wright EP, Smith AB, et al: Automated collection of quality-of-life data: A comparison of paper and computer touch-screen questionnaires. J Clin Oncol 17: 998-1007, 1999[Abstract/Free Full Text]

44. Wittchen HU, Boyer P: Screening for anxiety disorders: Sensitivity and specificity of the Anxiety Screening Questionnaire (ASQ-15). Br J Psychiatry 10-17, 1998 (suppl)

45. Spitzer RL, Williams JB, Kroenke K, et al: Utility of a new procedure for diagnosing mental disorders in primary care: The PRIME-MD 1000 study. JAMA 272: 1749-1756, 1994[Abstract/Free Full Text]

46. Leopold KA, Ahles TA, Walch S, et al: Prevalence of mood disorders and utility of the PRIME-MD in patients undergoing radiation therapy. Int J Radiat Oncol Biol Phys 42: 1105-1112, 1998[CrossRef][Medline]

47. Cull A, Gould A, House A, et al: Validating automated screening for psychological distress by means of computer touchscreens for use in routine oncology practice. Br J Cancer 85: 1842-1849, 2001[CrossRef][Medline]

Submitted August 24, 2001; accepted April 23, 2002.

CiteULike    Complore    Connotea    Del.icio.us    Digg    Facebook    Reddit    Technorati    Twitter    What's this?

This article has been cited by other articles:

				P. P. Roy-Byrne, K. W. Davidson, R. C. Kessler, G. J.G. Asmundson, R. D. Goodwin, L. Kubzansky, R. B. Lydiard, M. J. Massie, W. Katon, S. K. Laden, et al. Anxiety Disorders and Comorbid Medical Illness Focus, September 1, 2008; 6(4): 467 - 485. [Abstract] [Full Text] [PDF]

				A. J. Mitchell Pooled Results From 38 Analyses of the Accuracy of Distress Thermometer and Other Ultra-Short Methods of Detecting Cancer-Related Mood Disorders J. Clin. Oncol., October 10, 2007; 25(29): 4670 - 4681. [Abstract] [Full Text] [PDF]

				A. F. W. van der Steeg, J. De Vries, F. W. C. van der Ent, and J. A. Roukema Personality Predicts Quality of Life Six Months after the Diagnosis and Treatment of Breast Disease Ann. Surg. Oncol., February 1, 2007; 14(2): 678 - 685. [Abstract] [Full Text] [PDF]

				A Lienard, I Merckaert, Y Libert, N Delvaux, S Marchal, J Boniver, A-M Etienne, J Klastersky, C Reynaert, P Scalliet, et al. Factors that influence cancer patients' anxiety following a medical consultation: impact of a communication skills training programme for physicians Ann. Onc., September 1, 2006; 17(9): 1450 - 1458. [Abstract] [Full Text] [PDF]

				M. J. Ott, R. L. Norris, and S. M. Bauer-Wu Mindfulness Meditation for Oncology Patients: A Discussion and Critical Review Integr Cancer Ther, June 1, 2006; 5(2): 98 - 108. [Abstract] [PDF]

				A. D. Purushotham, S. Upponi, M. B. Klevesath, L. Bobrow, K. Millar, J. P. Myles, and S. W. Duffy Morbidity After Sentinel Lymph Node Biopsy in Primary Breast Cancer: Results From a Randomized Controlled Trial J. Clin. Oncol., July 1, 2005; 23(19): 4312 - 4321. [Abstract] [Full Text] [PDF]

				M. Okamura, S. Yamawaki, T. Akechi, K. Taniguchi, and Y. Uchitomi Psychiatric Disorders Following First Breast Cancer Recurrence: Prevalence, Associated Factors and Relationship to Quality of Life Jpn. J. Clin. Oncol., June 1, 2005; 35(6): 302 - 309. [Abstract] [Full Text] [PDF]

				K. Ell, K. Sanchez, B. Vourlekis, P.-J. Lee, M. Dwight-Johnson, I. Lagomasino, L. Muderspach, and C. Russell Depression, Correlates of Depression, and Receipt of Depression Care Among Low-Income Women With Breast or Gynecologic Cancer J. Clin. Oncol., May 1, 2005; 23(13): 3052 - 3060. [Abstract] [Full Text] [PDF]

				S. D. Fossa, A. A. Dahl, and J. H. Loge Fatigue, Anxiety, and Depression in Long-Term Survivors of Testicular Cancer J. Clin. Oncol., April 1, 2003; 21(7): 1249 - 1254. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Stark, D. Articles by Selby, P. Search for Related Content PubMed PubMed Citation Articles by Stark, D. Articles by Selby, P. Social Bookmarking                            What's this?