A one-year audit of specialist psycho-oncology
services in an Irish tertiary referral centre
Brian Hallahan, Malcolm Garland
Ir J Psych Med 2004; 21(4): 128-130
Table 1: Cancer diagnosis in psycho-oncology referrals at Beaumont
Objectives: Psychological services to patients with cancer
Hospital compared to most recently available National Cancer Registry2
are very limited in Ireland. A dedicated psycho-oncology
service was established at our Institution 18 months ago.
The aim of this study was to examine referral rates,
Cancer n % Deaths from cancer
psychiatric diagnoses and treatment interventions in this
Methods: The clinical and psychiatric characteristics of Breast 12 19.0 8.5
consecutive patients (n = 63) referred to a recently Lymphoma 12 19.0 3.3
established specialist psycho-oncology service in an Irish
university teaching hospital over a one-year period were Colorectal 8 12.7 12.6
examined prospectively. International diagnostic criteria Lung 7 11.1 18.9
were applied. Forms of treatment instituted and clinical
Ovarian 6 9.5 2.9
response to same were also audited.
Results: Over half (54%) of patients presented with Oesophagus 6 9.5 4.0
affective disorder spectrum illnesses. The remaining
Melanoma 4 6.3 1.1
referrals represented a diverse group of psychiatric
diagnoses, but included a large proportion of patients with Brain 2 3.2 2.4
delirium. Approximately 20% of patients did not meet Bladder + Kidney 2 3.2 4.0
criteria for a psychiatric diagnosis. There was an over-
representation of breast cancer and lymphoma. Thirty-eight Prostate 1 1.6 6.5
per cent of patients were on prescribed corticosteroids at Pancreatic 1 1.6 4.8
the time of referral. Eighty-two per cent of patients required
*Cancer classified as miscellaneous accounted for 31% of cancer deaths in 1999. Death rates
some form of psychiatric intervention, and 86% of these rather than incidence data are shown, as the former data is biased by high rates of benign skin
benefited clinically from our intervention. cancers.
Conclusions: The data suggest that even in oncology
services with good existing support networks, such as this bedded university teaching hospital with a single-practice 20-
one, the role of an additional input from liaison psychiatry bedded in-patient oncology ward with an attached day
is considerable. hospital. Approximately 1,000 new patients are seen per
annum, with almost the same number of admissions; there
Key words: Psycho-oncology; Liaison psychiatry; are approximately 5,000 day hospital patient visits per annum
Psychological morbidity; Audit. (Dr L Grogan, Dept. Head, pers comm).
The psycho-oncology service was commenced in July
Introduction 2002 and comprises a consultant liaison psychiatrist (MG)
The provision of psychological services to cancer patients and an independently funded research fellow in psychiatry
is an important component of liaison psychiatry. This is with sessional commitment to oncology (BH). This team
reflected in the growth of the sub-specialty of psycho-oncol- attends the weekly psycho-social (‘dry’) ward round, provides
ogy worldwide. Despite this, to our knowledge, only two teaching, both formal and informal, to all staff, and assesses,
oncology units in this country receive dedicated input from a treats and follows-up referred patients. For resource reasons
liaison psychiatry service. Beaumont Hospital is a 660- this service was made available to medical oncology only;
haematology and surgical oncology services were not
provided with such input.
Brian Hallahan, DPM, MRCPsych, Research Fellow,
*Malcolm Garland, MD, MRCPI, MRCPsych, DPM, Consultant Aims and methods
Psychiatrist, Department of Psychiatry, Beaumont Hospital, The aims were to characterise the clinical status of referred
Beaumont Road, Dublin 9 and the Royal College of Surgeons, patients to our service and to broadly compare this with the
Ireland. international data. Data refers to those consecutive patients
(n = 63) seen over a one-year period (Jan-Dec 2003). Site of
SUBMITTED: FEBRUARY 27, 2004. ACCEPTED: NOVEMBER 8, 2004.
primary tumour and extent of disease were the principal
Ir J Psych Med 2004; 21(4): 128-130
Table 2: Cancer disease status of referred patients Table 3: Psychiatric diagnoses in referred patients
n % Diagnosis n %
Early/localised 10 15.9 Depression 31 49.2
Advanced 32 50.8 Delirium 5 7.9
Terminal 21 33.3 Adjustment disorder 4 6.3
Total 63 100.0 Psychosis 3 4.8
Anxiety disorder 3 6.3
oncological data collected. Patients’ psychological status
was defined using International Classification of Disease- Bipolar affective disorder 3 4.8
(ICD-) 10 operational criteria.1 We also defined what treat- Alcohol misuse 2 3.2
ments were used, both psychotherapeutic and
Dementia 1 1.6
pharmacological and ascertained which patients improved
clinically (assessed by informal clinical means only). No active psychiatric diagnosis 11 17.5
Total 63 100.0
Table 1 classifies referred patients in terms of primary
tumour. Both lymphoma and breast cancer were somewhat psycho-stimulants for lethargy/amotivation, although few in
over-represented in our study compared to the national number, responded well to treatment. Only small amounts of
cancer registry of new case cancer incidence and deaths other drug groups, including benzodiazepines, were used.
from cancer.2 Eighty-six per cent of patients improved clinically with phar-
Table 2 describes the oncological disease status of the macotherapy. Two of the six patients unimproved on the initial
patients. Most patients were in the later stages of disease antidepressant subsequently improved upon augmentation
advancement. Table 3 describes the psychiatric diagnoses or switching.
seen. Of note, more than 50% of patients were diagnosed
with affective-spectrum disorders. Of the 39 patients referred Discussion
by the oncology team for ‘depression’, only 24 (61.5%) There is a strong international bank of literature on the
fulfilled ICD-10 criteria for a current depressive episode, the epidemiology of psychiatric disorder in cancer patients. Much
remaining patients being diagnosed with either no diagnosis of this has now been summarised in reviews, which facilitates
or adjustment disorder. comparison with our data, the first, to our knowledge, exam-
When all referrals were considered, the accuracy of the ining an Irish population. Although greatly biased by disease
perceived psychiatric diagnosis by the oncologists fell to status, diagnostic criteria, referral procedures and pre-exist -
50.1%. Identifiable causes for such misdiagnosis included, ing support networks, as well as by socio-cultural factors,
for example, misclassifying delerium as psychosis and vice Massie and Holland estimate overall prevalence rates to be
versa. Almost 20% of patients had no formal psychiatric 25%.3 This is significantly higher than the prevalence rate for
disorder. Approximately one third of patients had a past general hospital inpatients, which in turn is about twice that
psychiatric diagnosis (Table 4). of the general population.4
It was noted that 24 patients (38.1%) were on prescribed The majority of diagnoses are adjustment disorders. This is
corticosteroids at the time of referral. Fourteen (58.3%) of confirmed in an Irish population by data from an earlier study
these patients fulfilled criteria for depression, two for hypo- of 80 patients with breast cancer, which found (using struc-
mania (8.3%) and eight (33.3%) for psychosis with or without tured interviews) prevalence rates of psychological morbidity
delirium. Twenty-seven patients (42.8%) were on of 38% – almost half of which were adjustment disorders.5
psychotropic medication (mainly low-dose benzodiazepines) The predominance of more hierarchically serious disorders
prior to consultation. such as depression in our data (Table 3) suggest that the
Fifty-two patients (82.5%) required some form of clinical former are being dealt with by patients, their families and the
intervention. One patient (with moderate depression) refused nursing and social work component of the oncology service.
treatment. Treatments administered included supportive Nonetheless, even only considering new oncology patient
psychotherapy, psycho-education, referral for specific coun- throughput, a referral rate of 63/1000 (6.3%) is very low and
selling (grief or alcohol), as well as pharmacotherapy suggests the need for greater effort in detection of psycho-
(detailed in Table 5). One patient was treated with interper- logical morbidity, including the possible use of screening
sonal therapy. instruments.
Nine patients (14.3%) received psycho-education or Moreover, that the referring physician had a psychiatric
supportive psychotherapy as the principal component of their diagnostic ‘hit rate’ of only 50%, suggests a role for more
psychiatric input. Four (6.3%) patients were referred for educational input. The low referral data may also suggest
specific counselling. there are strong stigma factors militating against disclosure
As regards pharmacotherapy (Table 5), almost 40% were by the patient or alternatively, and more optimistically, the
commenced on anti-depressant therapy. An additional 10% existence of greater support networks for Irish patients.
required only dosage adjustment. Patients commenced on Screening has its advocates but the ideal instrument is
Ir J Psych Med 2004; 21(4): 128-130
Table 4: Past psychiatric history of referred patients Table 5: Description of pharmacotherapy instituted
Diagnosis n % Treatment Total Improved on treatment
Depression 12 60.0 Improved No benefit
Psychosis 2 10.0 Antidepressants 25 19 6
Alcohol misuse 2 10.0 Citalopram 6 4 2
BPAD 2 10.0 Escitalopram 7 5 2
Anxiety-spectrum disorder 1 5.0 Mirtazapine 9 7 2
Adjustment disorder 1 5.0 Sertraline 1 1 0
Total 20 100.0 Venlafaxine 2 2 0
Methylphenidate 4 4 0
hard to arrive at as many of the somatic symptoms of depres- Antipsychotics 6 6 0
sion (fatigue, insomnia, weight loss, anorexia etc.) mimic
Hypnotic/BZ D 3 3 0
those of systemic carcinomatosis.
The psychological sequelae of being diagnosed with, and Mood stabiliser 1 1 0
enduring, cancer are obviously enormous and have been
dealt with extensively in the literature. 6 Each stage of the
process carries with it different potential psychological crisis The offering of alternative agents (eg. mirtazapine for
points – the disbelief, anger and grief of initial diagnosis, the anorexia, stimulants for fatigue), although not documented
body image disturbance involved with surgery, the potential herein, has been an important part of our service.
rigours of medical treatment, the anxiety and unknowing that
surrounds recurrence, the acceptance of disease progres- Conclusion
sion and finally, of death. We conclude that even in oncology services with well-
Psychological morbidity rates generally follow a tri-modal developed multi-disciplinary support services, the addition of
distribution – the peaks representing the initial diagnosis and specialist liaison psychiatry services can be an important
treatment, recurrence and finally the consequences of further contribution to patient care.
advanced disease.7 Severity of the illness has been shown to
be the single most important variable associated with depres- Declaration of interest: None
sion8 and our data would accord with this (Table 2).
Certain cancers, such as pancreatic and breast carcinoma
are associated with disproportionally high rates. However, 1. The ICD-10 classification of mental and behavioural disorders. Diagnostic criteria for
even in those with terminal disease, modern treatments do research. Geneva, World Health Organisation, 1993.
2. National Cancer Registry Ireland. Cancer in Ireland 1994 – 2002; Incidence,
not make pain and physical suffering an inevitability. Mortality, Treatment and Survival.
Even more importantly, the experience of having cancer 3. Massie MJ, Holland JC. Depression and the cancer patient. J Clin Psychiatry 1990;
can be an opportunity to confront one’s assumptions about 4. Rodin G, Voshart K. Depression in the medically ill: an overview. Am J Psychiatry
life and grow spiritually.9 Also, episodes of depressed mood, 1986; 143: 696-705.
5. Garland M, Lavelle E, Doherty D, Golden-Mason L, Fitzpatrick P, Hill ADK, Walsh N,
anxiety and social withdrawal need not necessarily be seen O’Farrelly C. Psychological Medicine 2004 (InPress).
as negative: almost 20% of our referrals were for patients 6. Weisman AD, Worden JW The existential plight in cancer: significance of the first
100 days. Int J Psychiatry Med 1976; 7(1): 1-15.
with no formal psychiatric diagnosis, many of whom were 7. Sellick SM, Crooks DL. Depression and cancer: an appraisal of the literature for
reacting appropriately to bad news, or a difficult disease. prevalence, detection and practice guideline development for psychological
intervention. Psycho-oncology 1999; 8(4): 315-33.
The use of corticosteroids in oncology is common for the 8. Cassileth PA, Gerson SL, Bonner H, Neiman RS, Lusk EJ, Hurwitz S. Identification of
early relapsing patients with adult acute nonlymphocyte leukaemia by bone marrow
symptomatic treatment of a wide range of symptoms includ- biopsy after initial induction chemotherapy. J Clin Oncol 1984; 2(2): 107-11.
ing anorexia, fatigue, tumour oedema and vomiting.10 We 9. Petry JJ. Psychological response and survival in breast cancer. Lancet 2000;
report an alarming rate of diverse psychological morbidity 10. Levy MH, Rosen SM, Ottery FD, Hermann J. Supportive care in oncology. Curr
associated with these agents. Probl Cancer 1992; 16(6): 329-418.