Abstract
Major depression is a serious health problem in older persons, occurring in 1–3% of the elderly population [1]. It results in significant disability, cognitive impairment, suicide, medical illness and increased mortality [2,3]. As a consequence of the mix of medical, cognitive, neurological and psychosocial factors which are likely to affect outcome, the assessment and treatment of older patients with depression requires a comprehensive approach [4,5]. Consequently, the discipline of old age psychiatry has developed to focus research and clinical service efforts on older persons [6].
In Australia, the first stage of a National Mental Health Policy (1993–1998) [7] focused attention on the reorganisation of specialist mental health services to improve community-based assessment and continuity of care. The second stage (1998–2003) [8] of the policy emphasises improved assessment and treatment of depressive disorders, particularly within community and primary care-based systems. The impact of these policy directives on the treatment and outcome of depression in older persons is of special interest. In response to service deficits, and the specific medical and psychiatric needs of older patients, specialised psychogeriatric services have focused heavily on patients over 65 years of age and/or patients with neurocognitive disorders. Nonetheless the provision of high-quality care to those with severe depressive disorders is also a priority area [9].
Unfortunately, the restriction of specialised psychogeriatric services largely to those over 65 years of age is not necessarily relevant to the needs of other older patients with depressive disorders. Patients who present with affective disorders for the first time after 50 years of age are at high risk of other neurological (e.g. cerebrovascular disease) [3,10,11] and medical (e.g. hypothyroidism [12], postmenopausal status) risk factors. Those patients over 50 years of age at onset who have clinical symptoms (or neuroimaging evidence) of structural brain changes are also at high risk of side-effects from treatment, impaired short-and long-term response to treatment and development of dementia [10,13–15].
In the geriatric medicine and psychiatry literature particular emphasis has been placed on the value of comprehensive and multidisciplinary assessment systems. Such systems result in improved detection of depression and cognitive impairment [16], improved recovery from depression [17], delay in the development of disability and reduction of long-stay nursing home admissions [18]. Such systems also emphasise the key roles of medical, psychiatric and psychosocial assessment. Although increased involvement of general practitioners in the ongoing care of the elderly may improve the continuity of care, this needs to be balanced by the provision of appropriate physician education and support to ensure that the quality of the service is maintained [19–22].
The extent to which such comprehensive assessments and continuity of care occur outside specialised services is rarely evaluated. This report examines the extent to which assessments of older patients with depression took place within the different clinical teams of a district mental health service (i.e. specialised psychogeriatric services, hospitalbased services and community-based adult mental health teams). Additionally, we examined whether there was any major variation in the long-term illness outcomes of patients as a function of the clinical service they attended for initial assessment.
Method
Services
The St George Hospital and Community Health Service provides publicly funded health care to a catchment population of 220 000 persons. Fourteen per cent of the population are over 65 years of age (compared with the New South Wales state average of 11%), and 34% of residents speak a language other than English at home. The psychiatric unit of a general hospital is the principal admission service for clients of the St George district and has 18 beds for general adult psychiatry patients including patients over 65 years of age. (A second inpatient unit is located outside the catchment area and accepts patients who cannot be accepted as urgent admissions in our unit.) The psychogeriatric service is a specialised outpatient and community service, which treats persons aged over 65 years at the onset of their neuropsychiatric disorders. By contrast, community-based adult mental health teams provide services to persons who develop psychiatric disorders before the age of 65 years and may then continue to see patients with psychosis and/or depression who grow old. These teams work in an integrated fashion so that they supervise directly the care of patients admitted to the hospital unit. Typically, however, the community-based adult mental health teams deal mostly with younger persons with acute and chronic psychotic disorders. There is no specialist mood disorders service in the district.
Sample
Subjects were patients aged 50 years and over who presented to any of the public mental health services of the St George district for treatment of a major depressive episode (DSM-IV) [23] during the calendar year of 1995. Patients were excluded if their depression was clearly secondary to another medical or psychiatric disorder (e.g. schizophrenia, dementia of the Alzheimer's type), or substance abuse.
Assessments
Medical records were reviewed for demographic (age, sex, living circumstances) and clinical information documented within the first month following presentation. Clinical notes were reviewed for medical diagnoses (diabetes, hypertension, cerebrovascular disease), age of onset of affective disorder, other psychiatric diagnoses, psychiatric assessment by a mental health worker, assessment by a psychiatrist or trainee psychiatrist, physical examination, notation of potential risks of medical treatment, notation of suicidal or homicidal ideation, conduct of the Mini-Mental State Examination (MMSE) [24], clinical haematology and clinical chemistry investigations and relevant brain imaging studies.
Clinical outcome was assessed initially by the treating physician (clinical psychiatrist, trainee psychiatrist or general practitioner) between 20 and 38 months following initial presentation. Measures of outcome included:
Final global outcome of depression, rated continuously from 1 (worse) through to 5 (complete recovery).
Patient illness course since 1995, rated 1 (recovered and continuously well) through 8 (worse, including suicide). For analyses this scale was collapsed to 1 (recovered or subsequent attacks and recovered), 2 (subsequent or no subsequent attacks and not yet recovered) and 3 (chronic course or worse). If the patient had died this was recorded.
Cognitive impairment, rated from 1 (no cognitive impairment) through to 3 (probable or definite dementia).
Current living circumstances, rated initially in terms of location (home vs other) and degree of dependency, but collapsed for analyses to: 1 (living at home) and 2 (living in a dependent arrangement).
Medical morbidity, rated from 1 (no symptoms) though 5 (deceased).
Ongoing medical and psychiatric care.
Statistical analyses
The clinical outcome scales generated were treated as continuous measures for the purpose of some analyses but were also split into meaningful subgroups for the purpose of some categorical comparisons. Categorical data were analysed with chi-squared tests. ANOVAS were used for comparison of means across relevant service subgroups.
Results
Of the 99 patients included in the study, 44% presented to the specialised psychogeriatric service, 35% presented to community-based adult mental health teams and 21% to the inpatient services (including the medical services). Of patients treated by inpatient services, 48% were admitted to medical units and 52% to the psychiatry unit (where 33% were managed by psychogeriatricians and 19% by the integrated adult mental health teams).
The subjects identified represented 22% of all patients aged over 50 years who presented to psychiatric services during the 12-month period. Older patients presenting with primary neurocognitive disorders were managed by a separate clinical team. The majority of patients treated were female (69%) and the mean age was 68.9 (SD ± 11.6) years. Sixty per cent were living independently at home with a spouse and/or family member. The mean age of onset of depression was 57.1 (SD ± 18.8) years. The mean number of past episodes of depression was 2.9 (SD ± 3.8). Patients with first episodes of illness (34/99 or 34%) were spread equally across the service sectors. The proportion of patients with first episodes within each service sector (inpatients, 43%; psychogeriatric services, 32%; community mental health teams, 34%) did not differ (χ2 = 0.9, df = 2, not significant). Ten per cent of patients had chronic depression, seven per cent of patients had bipolar disorder and 13% of patients also had an Axis II diagnosis. Ten per cent of patients had a concurrent diagnosis of probable early dementia, while 45% also had a moderate or severe medical illness. Of particular interest in patients with late-onset depression [3], 32% of patients had hypertension, 14% had clinical evidence of cerebrovascular disease and 12% had diabetes.
Sociodemographic and illness characteristics of the 99 patients treated at the three service sites are summarised in Table 1. Patients who were admitted as inpatients and those treated within specialised psychogeriatric services were of comparable age. As expected, patients seen by the community-based adult mental health teams were significantly younger. Importantly, patients seen by the community-based teams had the earliest age of onset of illness and tended to have the highest rate of chronic illness. While all patients had primary affective disorders, the psychogeriatric services assessed the largest percentage of patients with concurrent cognitive impairment. The inpatient services were more likely to assess patients with moderate or severe medical illness.
Demographic characteristics, psychiatric status and medical illness history of patients treated by the psychiatric services
With regard to the adequacy of the assessments (see Table 2), patients admitted as inpatients were most likely to receive a physical examination. By contrast, the rate of physical examination by the community-based adult mental health teams was low (15%). If biomedical assessments were conducted by the general practitioner, rather than community-based adult mental health teams, then these findings would be less significant. There was little or no evidence, however, of this activity in the community records. Patients seen by the psychogeriatric services and at inpatient treatment centres more frequently received blood tests and a MMSE. These patients were also more frequently assessed by a psychiatrist or trainee psychiatrist. Further, communication to the general practitioner was more likely to flow from the inpatient and the psychogeriatric services. Possible risks of treatment were most frequently noted by the specialised psychogeriatric services.
Percentages of specific clinical assessments or investigations provided by psychiatric services at initial assessment
In 84% (83/99) of cases outcome ratings were completed (40 by psychiatrists and 43 by general practitioners). Nine patients (of 83 assessed) were deceased at follow up, though none by suicide. Overall there was no change in living status as 63% of patients were still residing at home either alone or with a spouse or family member, 16% were supported partially at home, 7% were at home, but completely dependent on another person, 9% were in hostels, 4% in nursing homes and 1% were in hospital. With regard to depression outcomes (59/83) 73% demonstrated at least moderate recovery at the follow-up assessment of which 49% had remained well throughout the period or had had further episodes from which they had made a full recovery. Consistent with the severity and chronicity of depression in the initial sample, 14% had run a chronic or deteriorating course. With regards to cognitive impairment at outcome, 51% had no evidence of cognitive changes, 25% were considered to have possible cognitive impairment, 19% probable cognitive impairment and 5% had definite dementia syndromes.
For the key clinical outcomes, patients treated by the community mental health teams were more likely to be depressed at follow up (F = 5.8, p < 0.01), while the patients of the specialised psychogeriatric services were most likely to be cognitively impaired (F = 4.4, p < 0.05) (Table 3). Similar percentages of patients treated at each service were receiving medical care from a general practitioner and/or a psychiatrist at follow up.
Psychiatric and medical outcome of patients (n = 81) treated by psychiatric services at 20–38-month follow up
Discussion
In the setting of a district mental health service, there were clear differences in the quality of medical and behavioural assessments provided to patients with depression over 50 years of age. Standard community-based adult mental health teams (who largely assess patients with an onset of depression before age 65 and/or patients with chronic disorders) were least likely to provide comprehensive assessment. The service deficits were not confined to quality of assessments, but also extended to communication with other key agencies. By contrast, the specialised psychogeriatric service (which has a mandate to provide care to patients with later age of onset, higher rates of cognitive impairment and concurrent medical illness) placed more emphasis on comprehensive medical and psychiatric assessment and provision of continuity of care with family practitioners. As this retrospective study did not assess the quality of treatments provided to patients, we are not able to comment on possible differences in the utilisation of available treatments.
The long-term outcome of depression for patients over 50 years of age treated by the community-based adult mental health teams was particularly poor. While this is likely to be largely a consequence of the inherent characteristics of this patient group (i.e. earlier age of onset and more chronic illnesses prior to assessment), we are not able to comment on whether it might also reflect deficits in the treatments provided. While patients managed by community teams did receive basic behavioural assessments they were much less likely to have been assessed cognitively, to have undergone physical examination or to have been subject to relevant laboratory investigations (including thyroid function, serum calcium or liver function tests). This lack of thorough medical assessment was compounded by the lack of adequate communication between the community-based adult mental health team and the family practitioner. Our study suggests that such patients were not perceived to be a high risk group by community-based adult mental health service staff. Assessment practice appeared to be driven more by the standards of clinical practice that are relevant to younger patients with severe mental disorders (i.e. less emphasis placed on concurrent medical, laboratory and/or cognitive factors). This is of particular concern because these older patients actually had poor illness outcomes. In older persons with depression, earlier age of onset is typically associated with more past episodes of depression [25], greater likelihood of relapse [26] and more severe episodes [27]. The poor outcome of older patients with an earlier age of onset in this study is consistent with other reports [26,28].
Conclusions
Current limitations of specialised assessment services for older patients with depression, to those over 65 years of age at onset, is arbitrary and inconsistent with current knowledge of risk factors [3]. To correct this situation, we need either to extend such services to younger patients or greatly improve the quality of assessment provided by community-based adult services. An improvement in quality within general services would require specific management and/or academic strategies that focus staff attention on the actual medical and behavioural needs of this patient group.