Recognition of distress and psychiatric morbidity in cancer patients: a multi-method approach

M. Keller1,*, S. Sommerfeldt1, C. Fischer1, L. Knight1, M. Riesbeck2, B. Löwe3, C. Herfarth4 and T. Lehnert4

1 Psychosocial Care Unit, 4 Department of Surgery, University Hospital, Heidelberg; 2 Rhine Clinics, Heinrich Heine University Hospital, Düsseldorf; 3 Department of Internal Medicine II (General Internal and Psychosomatic Medicine), Medical University Hospital, Heidelberg, Germany

* Correspondence to: Dr M. Keller, Psychosocial Care Unit, Department of Surgery, University Hospital, Im Neuenheimer Feld 155, D-69120 Heidelberg, Germany. Tel: +49-6221-562723; Fax: +49-6221-565250; Email: monika_keller{at}med.uni-heidelberg.de


    Abstract
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 Conclusions
 References
 
Background: This study aimed to determine the prevalence of psychiatric morbidity and distress among 189 consecutively recruited cancer patients upon admission to surgical oncology wards, and to investigate the recognition of distressed patients by medical staff.

Patients and methods: Assessment consisted of a diagnostic psychiatric interview (SCID, DSM-IV), patient-reported distress using a standardised questionnaire (Hospital Anxiety and Depression Scale), and physicians' and nurses' estimates of patients' distress. Twenty-eight per cent of patients were assigned a psychiatric diagnosis, with adjustment disorder predominating.

Results: Surgeons accurately recognised marked distress in 77% of patients with a psychiatric disorder and nurses did so in 75%. Because of low specificity, the positive predictive value was only 39% in surgeons and 40% in nurses. However, recognition of distress translated into referral to the psychosocial liaison service for only a minor proportion of distressed patients.

Conclusions: Since a remarkable proportion of distressed patients remained unrecognised by the medical staff, only systematic screening of patients upon admission allows timely support to those who are most in need.

Key words: cancer surgery, prevalence, psychiatric morbidity, psychosocial liaison service, recognition of distress, referral


    Introduction
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 Conclusions
 References
 
A substantial proportion of cancer patients suffer from psychological distress. Across a variety of studies, clinically relevant distress is reported in ~25% of patients; however, this figure ranges from 5% to 50% [1Go–6Go]. The considerable variation in prevalence rates is due to disease-related characteristics and treatment modalities, but also to different methods of assessment.

Psychosocial distress can be attributed to critical events during the course of the disease, with peaks immediately after diagnosis and during primary treatment, when recurrence occurs and during terminal stages [7Go]. Psychosocial interventions, provided timely and properly, effectively reduce distress, anxiety and depression, and can also prevent psychological morbidity during the course of disease. Early detection of relevant distress during these critical stages is therefore crucial, especially with more widespread managed care and shorter hospital stays [6Go–9Go].

Medical staff caring for cancer patients play a central role in providing patients with basic psychosocial support, in identifying patients with psychosocial morbidity, and in referring them for specialist evaluation and treatment. Several studies have investigated oncologists' ability to recognise psychosocial distress in cancer patients [4Go–6Go, 10Go–12Go]. When oncologists' estimates of patient distress were compared to probable cases identified by patient-administered screening measures such as the Hospital Anxiety and Depression Scale (HADS) and the General Health Questionnaire (GHQ), the overall rate of correct recognition across several studies was less than optimal.

Newell et al., using the HADS, found that oncologists' sensitivity in detecting clinical anxiety and depression was 17% and 6%, respectively [4Go]. In a radio-oncology setting, oncologists detected only 37% of patients with severe distress [12Go]. Fallowfield et al. investigated doctors' detection rates in a large sample of cancer patients [11Go]. When patient-reported distress indicated by the GHQ-12 was compared to doctors' ratings of patient distress, sensitivity ranged from 25% to 33% and specificity from 83% to 89%. Similar results were reported regarding nurses' ability to detect distress in patients (reviewed in [13Go]).

Although brief screening instruments allow rapid assessment in busy clinics, they will, by definition, have limited sensitivity and specificity to detect psychological morbidity [11Go, 12Go, 14Go]. Possibly, the overall performance of the screening could be improved by comparing medical staff estimates to a standardised diagnostic evaluation of psychiatric morbidity. To date, this issue has rarely been addressed and, if so, small sample sizes and almost no data on the reliability of the diagnostic procedure have been reported [6Go, 10Go, 14Go]. For practical reasons it is essential to evaluate whether appropriate recognition of distressed patients eventually translates into referral to mental health services [6Go].

The general purpose of this trial was to explore how distressed patients can be identified in a specific setting in order to provide appropriate support. The study aimed: (i) to determine the prevalence of concurrent psychiatric morbidity and psychological distress among cancer patients treated in a surgical unit; (ii) to assess doctors' and nurses' recognition of patients' psychological morbidity compared to a diagnostic psychiatric interview; (iii) to evaluate patient needs for psychosocial support from doctors' and nurses' viewpoints; and (iv) to assess whether recognition of distress eventually translates into referral to a psychosocial liaison service.


    Patients and methods
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 Conclusions
 References
 
Procedure
The Psychosocial Care Unit has been providing a liaison service for all cancer patients at the Department of Surgery, University of Heidelberg, for >20 years. Usually, nurses and physicians initiate referral. During a 6-month period, patients were recruited consecutively upon admission to the Department of Surgery. Eligible patients were aged ≥18 years, were admitted for suspected or proven malignancy, including all tumour sites and stages, and were physically and mentally able to complete a short questionnaire and to participate in a diagnostic psychiatric interview. Patients were approached during the first days after admission, before undergoing surgery, and were asked for written informed consent following thorough information about the purpose of the study. Due to limited resources, we were unable to conduct a diagnostic interview with all patients. The aim was therefore to interview a subsample of 50%. To avoid sampling bias, patients were randomly assigned to the interview or no interview condition after stratification for age and gender. The study protocol was approved by the Medical Ethics Committee of the University Hospital Heidelberg.

Data collection
A multi-level approach was used, with patients, medical staff and the clinical interview serving as sources of data.

Structured clinical interview
The Structured Clinical Interview (SCID-I) for DSM-IV, Axis 1, was administered to assess current psychiatric morbidity [15Go, 16Go]. The interview provides a diagnosis of depressive and anxiety disorder, post-traumatic stress disorder (PTSD) and adjustment disorder, and screens for substance abuse, somatoform and psychotic disorder. To establish a diagnosis, certain criteria have to be met, including severity and duration of symptoms and their impact on everyday activities. For instance, symptoms need to persist for ≥2 weeks to be considered relevant for the diagnosis of a current psychiatric condition. Acute distress related primarily to the impending surgery is not taken into account when diagnosing a psychiatric disorder.

After intensive training on the use of the SCID interview and diagnostic assessment, an experienced clinical psychologist who was unaware of the questionnaire results conducted the interview. Consenting patients were interviewed in a separate room on the ward, 1–3 days before surgery [mean time to surgery 2.0 days, standard deviation (SD) 2.3 days]. No interview was conducted on the day of surgery. Time to complete the interview ranged from 30 to 80 min.

Inter-rater reliability was assessed by double rating of 49 (63%) audiotaped interviews by a second trained psychologist. Overall agreement between raters on whether or not a diagnosis was assigned was 98%, with 88% agreement on the specific type of diagnosis.

Patients
Patients rated the degree of general psychological distress using the HADS [17Go, 18Go]. The HADS is a valid and reliable instrument widely used to screen for psychological distress in cancer patients [3Go, 19Go, 20Go]. It consists of 14 items resulting in a score ranging from 0 to 21 both on the anxiety and depression subscale. The impact of physical symptoms and functional impairment on daily activities was assessed by two items derived from the Medical Outcome Survey (SF-36) [21Go] using a 5-point Likert format (1 = no impact to 5 = very strong impact). Demographic data included age, gender, marital status, living arrangements and level of education. Medical data were extracted from the patients' records.

Medical staff
Medical staff rated patients' psychosocial condition within 3 days of admission (mean 2 days, SD 2.3 days). Nurses' estimates were based on an initial assessment conducted with each patient upon admission to the ward. Surgeons usually judged the patient's situation after taking the medical history. The surgeon and the nurse in charge of the individual patient each rated his/her psychosocial distress and need for psychosocial support on an author-derived, 10-item, Likert-scale questionnaire (0 = not at all to 5 = very much). Items covered anxiety, depression, illness-related distress, and the patient's current physical and emotional condition. The total score of eight items (range 0–40) addressing patient distress was used for analysis (Cronbach's {alpha}: surgeons = 0.85, nurses = 0.89). Referral for psychosocial consultation was derived from the patients' consultation form.

Statistical analyses
Standard statistics for parametric and non-parametric data were applied to calculate prevalence of psychosocial morbidity/distress. Correlations were calculated using Spearman's rank-order correlation coefficients. To establish threshold values for probable caseness, the sensitivity, specificity, positive predictive value (PPV) and misclassification rate (MR) for the HADS global score were calculated, according to the presence or absence of a diagnosis of current DSM-IV psychiatric disorder serving as the ‘gold standard’. Similarly, surgeons' and nurses' ability to detect psychosocial distress in patients was assessed first by calculating sensitivity and specificity of the composite eight-item distress score compared to the DSM-IV diagnoses, and secondly compared to the HADS threshold value established for the sample. Concordance between the patients', surgeons' and nurses' assessment was measured using kappa statistics. All statistical comparisons were two-sided, using P ≤0.05 as the level of significance. All analyses were performed using the Statistical Package for Social Sciences (SPSS, release 11.0).


    Results
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 Conclusions
 References
 
From February to July 2001, 279 newly admitted patients with proven or suspected diagnosis of cancer were recruited. Out of 240 eligible patients, 190 (79%) consented to participate in the study. Declining patients were on average 10 years older than participants (66.9 compared with 57.4 years, respectively; P<0.001), and more of them suffered from concomitant physical diseases (61% compared with 39%, respectively; {chi}2=4.213, df=1, P=0.04). Regarding tumour site and stage of disease, declining patients did not differ from those who consented. One hundred and seventy-four patients agreed to take part in a SCID interview. Eight consenting patients could not be reached in time before surgery, therefore 78 patients (45%) had an SCID interview. Questionnaire data are available for 189 patients.

Twenty-six physicians and 45 nurses rated 181 and 165 patients, respectively. Most physicians were registered surgeons and 83% were male. The majority of the nurses were registered and 82% were female. The number of patients rated by any one physician ranged from 1 to 29 (mean 7.0), and those rated by any one of the nurses from 1 to 33 (mean 3.6).

Demographic and medical characteristics of the sample are presented in Table 1.


View this table:
[in this window]
[in a new window]
 
Table 1. Characteristics of patients: total sample and Hospital Anxiety and Depression Scale (HADS) high-scorers

 
Prevalence of psychiatric morbidity and distress
Patients assigned to the interview condition (n=78) did not differ from those without interview regarding demographic and medical characteristics. A trend was observed towards a higher level of psychological distress assessed by the HADS among patients with interview (Table 4).


View this table:
[in this window]
[in a new window]
 
Table 4. Comparison of patients with versus without interview

 
Twenty-two patients were diagnosed with a current psychiatric disorder (28%). Two patients were assigned two diagnoses, resulting in a total of 24 psychiatric diagnoses. Adjustment disorder was diagnosed in 17 patients (22%) (Table 2), major depression in four patients (5%), anxiety disorder in two (3%), and psychosis in one patient. With respect to adjustment disorders, depressive affect predominated in three patients, anxious affect in two patients, whereas mixed mood was diagnosed in 12 patients. There were no gender differences between patients with and without a current psychiatric diagnosis.


View this table:
[in this window]
[in a new window]
 
Table 2. Prevalence of psychiatric disorder and distress

 
Patient-rated distress, assessed by the HADS, was on average 6.5 (SD 4.6) on the anxiety subscale and 5.1 (SD 4.3) on the depression subscale. For the total score, the value was 11.6 (SD 8.5). According to the established cut-off values, 64% of the patients scored within normal ranges (<8) on the anxiety scale, and 77% did so on the depression scale. Using the cut-off of ≥11, suggesting probable clinical caseness, the proportion was 19% for anxiety and 14% for depression (Table 2). Applying the cut-off value of ≥16 for the total score, the proportion was 26%.

Using the DSM-IV diagnoses as the ‘gold standard’, a cut-off value of ≥16 for the HADS total score was calculated. Applying this cut-off, 19 out of 22 patients with a current psychiatric disorder were correctly identified as cases (sensitivity 86%), whereas 48 out of 55 patients with no current DSM diagnosis were correctly identified as non-cases (specificity 87%). The PPV of the HADS in this sample was 73% and misclassification occurred in 10 patients (MR 13%).

High-scorers (above the cut-off) were more likely to be female ({chi}2=4.89, df=1, P=0.027), and to suffer from physical symptoms ({chi}2=5.08, P=0 024) and from impaired daily life activities due to reduced performance ({chi}2=4.60, P=0.032) (see Table 1). Illness-related characteristics, such as site of tumour and stage of disease did not differ among high- and low-scorers.

Doctors' and nurses' ratings of patients' psychosocial distress
Surgeons' estimates of patient distress ranged from 9 to 38 [mean 22.4 (SD 5.8)] and nurses' ratings ranged from 8 to 36 [mean 21.5 (SD 6.4)]. A cut-off of ≥22 was chosen for the total score of the medical staff's ratings. This was based on the calculation of the sensitivity and specificity of these scores compared to DSM-IV diagnosis. Using this cut-off score, surgeons and nurses rated 55% and 50% of patients as highly distressed, respectively (Table 2). Doctors' and nurses' ratings were only moderately interrelated, as shown by a correlation coefficient of r=0.43 for the total score.

A higher proportion of female patients were judged as distressed by both nurses ({chi}2=11.95, P=0.001) and doctors ({chi}2=4.86, P=0.033). Apart from that, no significant association was found between patients' demographic and illness-related characteristics, and staff-rated distress.

Recognition of psychiatric morbidity by doctors and nurses
Using a cut-off of ≥22 for doctors' composite scores, 17 out of 22 patients with a DSM-IV psychiatric disorder were correctly identified by their doctors as highly distressed (sensitivity 77%; Table 3). Likewise, 29 out of 55 patients with no DSM-IV diagnosis were correctly estimated as not or only mildly distressed (specificity 53%; Table 3).


View this table:
[in this window]
[in a new window]
 
Table 3. Surgeons' and nurses' recognition of distress and need for psychosocial support

 
Nurses correctly recognised 15 out of 20 patients with a DSM psychiatric disorder (sensitivity 75%; Table 3), whereas 29 out of 51 patients with no DSM diagnosis were correctly identified as not or mildly distressed (specificity 57%; Table 3).

Recognition of psychological distress by doctors and nurses
Doctors correctly identified 30 out of 47 high-scorers on the HADS total score (sensitivity 64%) and 63 out of 131 low-scorers (specificity 48%). Concordance between patients' and doctors' ratings was low ({kappa}=0.088; Table 3).

Nurses correctly identified 31 out of 43 high-scorers (sensitivity 72%) and 69 out of 122 low-scorers (specificity 57%). Concordance between patients' and nurses' ratings was {kappa}=0.219 (Table 3).

Patient characteristics and recognition of distress
We further analysed whether recognition of distress by the medical staff was related to patients' demographic and medical characteristics, by comparing patients who were correctly classified to those who were misclassified. With respect to surgeons' ratings, the rate of accurate recognition depended neither on patients' demographic nor on illness-related characteristics.

Identical results were found for nurses' recognition of distressed patients, with the exception that female patients were more frequently misclassified than male patients (females 51% compared with males 32%; {chi}2=6.089, P=0.016).

Altogether, the accuracy of recognition by the medical staff could not be explained by any of the patient characteristics that were assessed in this study.

Need for professional psychosocial support evaluated by surgeons and nurses
Surgeons identified a need for professional psychosocial support in 15 out of 19 patients who received a DSM-IV diagnosis (sensitivity 79%) and in 29 out of 47 patients without a psychiatric diagnosis (specificity 62%). With respect to the HADS, psychosocial support was recommended in 25 out of 36 high-scorers (sensitivity 69%) and in 59 out of 107 low-scorers (specificity 55%; Table 3). Notably, the results were restricted to those patients for whom surgeons made a statement as to whether support was needed or not. In 32 (18%) cases, surgeons were undecided and felt unable to give a firm recommendation.

Nurses recognised a need for psychosocial support in 11 out of 16 patients with a current DSM diagnosis (sensitivity 69%), and also in 13 out of 39 patients with no diagnosis (specificity 67%). With respect to the HADS high-scorers, nurses identified a need for psychosocial support in 23 out of 36 patients (sensitivity 64%), and also in 40 out of 88 low-scorers (specificity 54%; Table 3). In 40 cases (24%), nurses were undecided whether or not the patient needed support.

Referral for psychosocial support
As documented by the patients' consultation forms, nine of the 22 patients with a current DSM-IV diagnosis (41%) and 15 out of 49 (31%) high-scorers on the HADS were referred to the psychosocial liaison service by the medical staff during their hospital stay.


    Discussion
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 Conclusions
 References
 
With more widespread managed care and shorter hospital stays, early and accurate recognition of patients with psychosocial morbidity is warranted. This cross-sectional study aimed to explore how distressed patients can best be identified in a surgical oncology setting. Various methods of assessment were compared. As evidenced by a high response rate, the sample is representative of cancer patients undergoing surgery at various stages of the disease. Reliability of the diagnostic procedure according to DSM-IV diagnoses was confirmed by a high concordance among experts' ratings.

Prevalence of psychiatric morbidity and distress
Across various measures, a quarter of the patients were identified as severely distressed. Twenty-eight per cent of the patients were assigned a psychiatric diagnosis, predominantly adjustment disorder, and 26% were high-scorers on the HADS. Prevalence rates found in our sample are consistent with findings across various oncological settings [4Go–6Go, 11Go, 12Go], thereby confirming our assumption that patients undergoing surgery for either recently diagnosed or recurrent cancer are particularly vulnerable. As evidenced by psychiatric assessment, adjustment disorder typically presents with mixed effect, including anxious as well as depressive symptoms [5Go, 22Go].

Using the HADS as a screening instrument: cut-off and performance
The performance of self-assessment screening measures, such as the HADS and the GHQ, to reliably detect patients with high levels of distress, has been questioned by some authors. Both sensitivity and specificity were low when established cut-off values were used [23Go, 24Go]. Defining an HADS total score threshold above 15, based on DSM-IV diagnosis and specific to the sample under study, yielded a favourable performance in identifying patients with psychiatric morbidity. This was confirmed by an acceptable misclassification rate of 13%, and thus proved a practicable procedure.

Surgeons' and nurses' recognition of patient distress
Several authors report a low detection rate of distressed patients by doctors and nurses when their distress estimates are compared with screening measures with limited sensitivity and specificity [4Go–6Go, 10Go–12Go]. We expected to achieve a more favourable detection rate by comparison with psychiatric morbidity, assessed by structured diagnostic interview. In fact, surgeons' ability to detect current psychiatric disorder in patients was superior to their accuracy in recognising patient-reported distress. This discrepancy could be explained by surgeons' medical education, which is primarily oriented towards diagnostic classification of signs and symptoms. Nurses' judgement of the patient's psychosocial condition appeared closer to patient-reported distress, as evidenced by a nearly identical recognition rate, whether based on DSM-IV diagnoses or on patient-reported distress. However, a gender effect cannot be excluded, since most surgeons were male and most nurses female.

In contrast with other authors, patient characteristics assessed in our study were unrelated to medical staff's awareness of distress [12Go]. Maybe the similarities at that particular stage, when patients hope for cure through surgery, outweigh the differences that might emerge at another stage of illness and treatment.

In our sample, cancer patients with psychiatric morbidity had a fairly high chance of being accurately recognised by doctors and nurses. However, this clearly happens at the expense of a high false-positive rate, thereby confirming findings of several authors [4Go, 5Go, 12Go]. From a clinical point of view it is highly desirable that no patient with a psychiatric disorder should go unrecognised and, hence, untreated [7Go, 13Go]. This implies, however, that a considerable proportion of patients who appear distressed to the medical staff need to be evaluated for (absent) morbidity. Considering the limited resources of professional psychosocial treatment in the majority of institutions, this is hardly a realistic option.

Controversy persists in the literature as to whether patients' distress is under- or overestimated by doctors and nurses. It appears primarily to be a question of the threshold, and how sensitivity and specificity are determined [4Go, 5Go, 7Go, 13Go, 23Go]. For instance, sensitivity necessarily will decrease with a higher cut-off, resulting in a considerable underestimation of distressed patients. On the contrary, lowering the threshold will result in overestimation, with many patients with low to moderate distress being judged as severely distressed. Both under- and overestimation occur as a result of poor diagnostic awareness by doctors and nurses.

The reasons why patients' distress is frequently overlooked have been broadly discussed by several authors: some patients will be reluctant to share their concerns and emotions. Since doctors frequently communicate in a rational manner, where little space is left for sharing emotional concerns, patients possibly feel discouraged from disclosure [25Go]. On the other hand, many doctors are afraid of eliciting strong emotions that increase their workload further and make them feel helpless about how to manage those emotions. Newell et al. found oncologists' ability to recognise distress in their patients was associated with their pressure of workload [4Go]. As stated by Fallowfield et al. [11Go], the majority of doctors in oncology are under-resourced, overstretched and were never trained in communication skills.

Actual referral
A wide discrepancy was observed between medical staff's estimate of patients' distress and need for professional psychosocial support on one hand, and the low rate of actual referral on the other: only 40% of patients with a DSM-IV disorder, and 31% of the HADS high-scorers were eventually referred to the psychosocial liaison service. Solely relying on surgeons' and nurses' referral results in substantial under-treatment of patients in need.

Our results confirm those published by several authors, with respect to both prevalence rates of psychiatric distress and morbidity, and also to poor recognition rate. Although it might be expected that medical oncologists are more empathetic and skilful in recognising distress in their patients compared with surgeons, this assumption is not supported by results from studies that investigated oncologists' and radiotherapists' recognition of distress [4Go, 11Go, 12Go]. Therefore, the findings from this study can probably be generalised to other disciplines involved in the care of cancer patients.

Some limitations of the study need to be mentioned. Since situational anxiety and distress inadvertently precede every surgical intervention, assessment before surgery might not appropriately predict the need for psychological support during the hospital stay. Moreover, psychological distress due to unexpected complications and delayed recovery after surgery is hardly predictable. To address this issue appropriately, initial diagnostic evaluation needs to be followed by subsequent re-evaluation of patients' psychological status whenever needed during the hospital stay.

It is worth noting that the performance of the medical staff's recognition is possibly overestimated in our study, because the threshold for our instrument was chosen for optimal sensitivity and specificity in the same sample that was used to assess the test's performance. Ideally, the cut-off point should have been determined beforehand in a different sample.

It cannot be concluded that our findings apply exclusively to cancer patients. Tension and distress are likely to precede every surgical intervention to a certain degree, irrespective of the underlying disease. Although it was beyond the scope of the present study, it would be interesting to investigate distress in patients undergoing surgery for a non-malignant condition. We are currently addressing this issue by applying a screening questionnaire to all patients newly admitted to surgical wards. Preliminary results from 260 patients indicate a higher degree of distress in cancer patients compared with non-cancer patients.


    Conclusions
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 Conclusions
 References
 
Given the considerable prevalence of psychiatric morbidity observed in a surgical oncology setting, systematic screening immediately upon admission is warranted to assure early evaluation and management of psychosocial distress. Because of the moderate detection rate of surgeons and nurses, initial diagnostic evaluation using self-administered screening measures is recommended. A threshold specific to the sample should be established first. Further, training aimed at improving communication and diagnostic skills of doctors and nurses is warranted. There is convincing evidence that training is effective in enhancing performance and communication skills, and also in reducing distress among medical staff [25Go–28Go].


    Acknowledgements
 
Support from the Medical Faculty of the University of Heidelberg (grant no. 175/2000) is gratefully acknowledged.

Received for publication January 20, 2004. Revision received April 8, 2004. Accepted for publication April 15, 2004.


    References
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 Conclusions
 References
 
1. Hopwood P, Howell A, Maguire P. Psychiatric morbidity in patients with advanced cancer of the breast: prevalence measured by two self-rating questionnaires. Br J Cancer 1991; 64: 349–352.[ISI][Medline]

2. Ford S, Lewis S, Fallowfield L. Psychological morbidity in newly referred patients with cancer. J Psychosom Res 1995; 39: 193–202.[CrossRef][ISI][Medline]

3. Ibbotson T, Maguire P, Selby P et al. Screening for anxiety and depression in cancer patients: the effects of disease and treatment. Eur J Cancer 1994; 30A: 37–40.[ISI][Medline]

4. Newell S, Sanson-Fisher R, Girgis A, Bonaventura A. How well do medical oncologists' perceptions reflect their patients' reported physical and psychosocial problems? Data from a survey of five oncologists. Cancer 1998; 83: 1640–1651.[CrossRef][ISI][Medline]

5. Passik S, Dugan W, McDonald M et al. Oncologists' recognition of depression in their patients with cancer. J Clin Oncol 1998; 16: 1594–1600.[Abstract]

6. Payne DK, Hoffman RG, Theodoulou M et al. Screening for anxiety and depression in women with breast cancer. Psychosomatics 1999; 40: 64–69.[Abstract/Free Full Text]

7. Watson M. Is ‘bedside medicine’ returning to oncology? Acta Oncol 2000; 39: 1–3.[CrossRef][ISI][Medline]

8. Meyer TJ, Mark MM. Effects of psychosocial interventions with adult cancer patients: a meta-analysis of randomized experiments. Health Psychol 1995; 14: 101–108.[CrossRef][ISI][Medline]

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

10. Hardman A, Maguire P, Crowther D. The recognition of psychiatric morbidity on a medical oncology ward. J Psychosom Res 1989; 33: 235–239.[CrossRef][ISI][Medline]

11. Fallowfield L, Ratcliffe D, Jenkins V, Saul J. Psychiatric morbidity and its recognition by doctors in patients with cancer. Br J Cancer 2001; 84: 1011–1015.[CrossRef][ISI][Medline]

12. Söllner W, DeVries A, Steixner E et al. How successful are oncologists in identifying patient distress, perceived social support, and need for psychosocial counselling? Br J Cancer 2001; 84: 179–185.[CrossRef][ISI][Medline]

13. Lampic C, Sjoden P. Patient and staff perceptions of cancer patients' psychological concerns and needs. Acta Oncol 2000; 39: 9–22.[CrossRef][ISI][Medline]

14. Hopwood P, Keeling F, Long A et al. Psychological support needs for women at high genetic risk of breast cancer: some preliminary indicators. Psychooncology 1998; 7: 402–412.[CrossRef][ISI][Medline]

15. First M, Gibbon M, Spitzer R, Williams J. Users Guide for Structured Clinical Interview for DSM-IV Axis 1 Disorders—Research Version. Washington DC: American Psychiatric Press, Inc., 1996.

16. Wittchen H-U, Wunderlich U, Gruschwitz S, Zaudig M. SKID-I Strukturiertes Klinisches Interview für DSM-IV, Achse I. Göttingen: Hogrefe: Psychische Störungen 1997.

17. Zigmond A, Snaith R. The hospital anxiety and depression scale. Acta Psychiatr Scand 1983; 67: 361–370.[ISI][Medline]

18. Herrmann C. International experiences with the hospital anxiety and depression scale—a review of validation data and clinical results. J Psychosom Res 1997; 42: 17–41.[CrossRef][ISI][Medline]

19. Hopwood P, Howell A, Maguire P. Screening for psychiatric morbidity in patients with advanced breast cancer: validation of two self-report questionnaires. Br J Cancer 1991; 64: 353–356.[ISI][Medline]

20. Aass N, Fossa S, Dahl A, Moe T. Prevalence of anxiety and depression in cancer patients seen at the Norwegian radium hospital. Eur J Cancer 1997; 33: 1597–1604.[CrossRef][ISI][Medline]

21. Ware JJ, Kosinski M, Keller S. A 12-item short-form health survey: construction of scales and preliminary tests of reliability and validity. Med Care 1996; 34: 220–233.[CrossRef][ISI][Medline]

22. Sellick SM, Crooks DL. Depression and cancer: an appraisal of the literature for prevalence, detection and practice guideline development for psychological interventions. Psychooncology 1999; 8: 315–333.[CrossRef][ISI][Medline]

23. Hall A, A'Hern RP, Fallowfield LJ. Are we using appropriate self-report questionnaires for detecting anxiety and depression in women with early breast cancer? Eur J Cancer 1999; 35: 79–85.[CrossRef][ISI][Medline]

24. Maguire P. Improving communication with cancer patients. Eur J Cancer 1999; 35: 2058–2065.[CrossRef][ISI][Medline]

25. Maguire P, Booth K, Elliott C, Jones B. Helping health professionals involved in cancer care acquire key interviewing skills—the impact of workshops. Eur J Cancer 1996; 32A: 1486–1489.[CrossRef][ISI][Medline]

26. Fallowfield L, Jenkins V, Farewell V et al. Efficacy of a Cancer Research UK communication skills training model for oncologists: a randomised controlled trial. Lancet 2002; 359: 650–656.[CrossRef][ISI][Medline]

27. Roter D, Hall J, Kern D et al. Improving physicians' interviewing skills and reducing patients' emotional distress. A randomized clinical trial. Arch Intern Med 1995; 155: 1877–1884.[Abstract]

28. Razavi D, Delvaux N. Communication skills and psychological training in oncology. Eur J Cancer 1997; 33 (Suppl 6): S15–S21.[ISI][Medline]





This Article
Abstract
Full Text (PDF)
E-letters: Submit a response
Alert me when this article is cited
Alert me when E-letters are posted
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in ISI Web of Science
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Search for citing articles in:
ISI Web of Science (2)
Disclaimer
Request Permissions
Google Scholar
Articles by Keller, M.
Articles by Lehnert, T.
PubMed
PubMed Citation
Articles by Keller, M.
Articles by Lehnert, T.