It was easier to interpret the results of rigorously designed studies that focused on a single behaviour or single intervention tailored to the needs of individuals and carers. Future studies should seek to replicate the findings outlined here, improving methodologies where necessary and including outcome measures that encompass the interests of people with dementia, family caregivers and health professionals.

Background: This article examines some of the factors responsible for older patients' decision to report current depressive symptoms to their general medical practitioner. A companion article considers factors contributing to general practitioners' (GPs') recognition of major depressive episode when it was present. Methods: A survey was conducted of a stratified sample of 1,021 patients aged 70+ years of 30 GPs in Melbourne, Australia, to gauge the prevalence of depressive symptoms, the frequency with which patients had informed GPs of their symptoms, and GPs' recognition of major depressive episodes. Patients and informants were questioned using the Canberra Interview for the Elderly, which generates rigorous ICD-10 research diagnoses. Results: Logistic regression analysis showed that symptom disclosure was associated in descending order of importance with higher depressive scores, previous contact with a psychiatrist, and female gender. Even so, 48% of persons with ICD-10 moderate or severe depressive episode had not reported any current complaints to their doctor at the time of interview. Conclusion: Older patients often do not report depressive symptoms to their medical practitioner. Men and patients lacking “psychological mindedness” may be at special risk.

Background: It is important that serious depressive illness be recognized and treated appropriately by primary care practitioners. This and the preceding article in this issue examine (a) factors responsible for older patients' decision to report depressive symptoms to their doctor and (b) general practitioners' (GPs') recognition of depression when it was present. Methods: A survey was conducted of a stratified sample of 1,021 patients aged 70+ years of 30 GPs in Melbourne, Australia, using the Canberra Interview for the Elderly, which generates rigorous ICD-10 research diagnoses. Results: GPs' ratings of depression were best predicted in descending order of importance by patients' past contact with a psychiatrist, the doctor's view that a patient did not have dementia, the number of current depressive symptoms, patients' disclosure of these symptoms, and current physical pain. Physicians' assessments of patients' mood concurred with research diagnoses in 23 of 35 (66%) cases of ICD-10 mild depressive episode and 23 of 26 (88%) cases of moderate or severe depressive episode. GPs were unaware, however, of many depressive symptoms and often rated patients as being depressed when they were not. Conclusion: The use of a simple checklist of depressive symptoms would lead to a dramatic improvement in doctors' knowledge of patients' current psychiatric status.

Objective. To examine the effectiveness of dementia programmes and report factors related to programme outcomes. To describe the characteristics which placed hostel residents at risk for nursing home placement and to measure changes in dependencies and impairments over 2 years. Design. Longitudinal, quasi‐experimental using in situ resident groups matched on resident and facility characteristics. Setting. Australian hostels for the elderly. Subjects. 587 residents (programme group N=184, comparison group N=162, frail groups N}=241). Measures. Mini‐Mental State Examination, Geriatric Depression Scale and staff‐rated indices of functioning, including activities of daily living, problem behaviours, psychiatric symptomology and health status, were used to monitor changes in resident characteristics. Time to nursing home placement was another outcome measure. Results. Residents in hostel dementia programmes remained significantly longer than those in the comparison group (2.5 months over 2 years) before exit to a nursing home. Quality of life for residents in dementia programmes was enhanced through higher levels of social contact with relatives and lower reported levels of depressive symptoms. Conclusions. Dementia programmes worked, but the reasons why were more difficult to establish. The programmes did not appear to modify the capacities of residents by slowing rates of decline. Dementia programmes provided specialist (non‐personal care) staff focusing on the social and emotional needs of residents. These staff provided appropriate, targeted activities for residents with dementia, had a clearly defined role directed exclusively to these residents and felt directly responsible for them. Dementia programmes produced a system effect. They increased the capacity of hostels to care for residents with dementia for longer periods, before admission to a nursing home. © 1997 John Wiley & Sons, Ltd.

This paper begins with an account of the structure of Australia's residential long-term care program, which was divided into two distinct levels of hostel and nursing home care until 1997. In response to changed policy objectives, a number of measures were then taken to create an integrated residential care system. The main measures were the development of a single scale for classification of resident care need and associated funding to replace two previous separate scales, and the implementation of a new quality assurance system, which included new standards for buildings as well as revised standards for care. I give accounts of these measures and the extent to which they have achieved their intended outcomes before proposing some further developments that could see closer links among pre-admission assessment, resident classification, and quality assurance.