Little research has focussed on nursing home residents with a history including a non-dementia related chronic mental illness (HCMI) that manifested before placement in a nursing home. Data from the 1993 Minimum Data Set for Nursing Home Resident Assessment and Care Screening (MDS) on over 70,000 residents in Kansas, Maine, Mississippi, and South Dakota were used to investigate differences in characteristics and care between these residents and more traditional residents. HCMI residents were more likely to be under 65 years of age, male, Medicaid recipients, less functionally impaired, to exhibit higher levels of problem behaviours, and to receive psychoactive medications and psychological therapy, though therapy prevalence rates were low. These data may indicate that appropriate care for HCMI residents is a concern.

Antidepressant use was examined with samples from long-term care facilities in Toronto (Canada), Sapporo and Naie (Japan), Reykjavik (Iceland) and Prague (Czech Republic). Only in Iceland did the majority of residents with depression receive an antidepressant. Rates of depression and antidepressant use were uniformly low in Japan, and there was a great discrepancy between diagnosed depression and behavioural signs of depression in the Czech Republic. In all countries, about half the recipients of antidepressants have no clear indication of depression present. For some countries, antidepressant use was lower among residents who were female, older and more disabled. Depression is clearly under-diagnosed in the Czech Republic, but low rates of depression in Japan are somewhat more difficult to interpret. Given the widespread consensus that depression is under-detected and under treated, these results suggest that responses to depression could be improved through instruments like the MDS.

Cross-sectional data (n = 6,487) from four Nordic countries, Denmark, Finland, Iceland and Sweden, were utilized to determine the prevalence of daily pain and its association with disability in institutional long-term care. Every resident in each of the participating institutions was assessed with Minimum Data Set version 1.0. The sample was representative of institutional long-term care in Copenhagen and Reykjavik. In addition, the data collected from Stockholm and Helsinki provided substantial information on the residents in these capitals. The results showed that 22–24 per cent of the residents experienced daily observable pain and this was most evident in the most disabled subjects. In addition to disability and female gender, diseases or conditions independently associated with pain were terminal prognosis, osteoporosis, pneumonia, arthritis, depression, anaemia, peripheral vascular disease, cancer and cardiac heart failure. The association between pain and severe cognitive impairment was negative. The results strongly indicate that daily pain in long-term care has a complex association with disability, the latter acting together with underlying diseases as a source and/or result of pain. Thus, a vicious circle between pain and disability can be anticipated.

While a substantial body of literature has emerged on the relationships between stress, social support and well-being in community-based older persons, few studies have examined institutionalized populations. The present study used data from pilot studies of the MDS 2.0 in three hospitals, along with supplementary surveys related to psychological well-being. It was found that patients with higher social engagement scores tended to have higher levels of well-being, and this effect was greatest among those with more stable physical health. Pain was an important predictor of lower well-being. Since the MDS 2.0 provides a comprehensive approach to identifying and responding to social, psychological and physical problems in institutionalized older persons, it can have an important impact on well-being when used to support decision making and clinical intervention.

We establish the potential for outcomes evaluation of a long-term care population-based dataset. Data come from the Health Care Financing Administration's case-mix and quality demonstration project including all Medicare/ Medicaid certified homes of five U.S. states during 1992–1994. Using the Minimum Data Set, we identified nearly 70,000 residents over 65 years of age with congestive heart failure. We provide a preliminary analysis of pharmacotherapy of congestive heart failure and its effects on decline in physical functioning. Functional outcome, expressed as the rate of ADL decline, was improved for patients on combination therapy relative to those on digoxin or ACE-inhibitors alone. The availability of a population-based data set provides a means of evaluating current policies and practices.

Prior to the past decade, much research examining outcomes of home care programs, including efforts at delaying institutional placement, maintaining function, and supporting independence, was atheoretical in character. Outcomes hoped for were often unobserved. New policy developments require comprehensive assessment of need and aggregation of this assessment information. As more and more patients leave hospitals with complex clinical problems and extensive rehabilitative goals there has been a corresponding explosion of home care services. Social care models, while they still exist, are becoming a smaller component of the overall home care market. In this changing environment, questions are now being asked concerning the appropriateness of the care programs in home care and other post-acute care settings. There are also concerns that need to be addressed about movement of clients between post-acute settings. In this paper, we describe a set of proposed functional, behavioural, and social outcome measures that are germane to evaluating the efficacy of programmatic efforts within the post-acute continuum. Data were collected with a standardized data collection instrument, the Resident Assessment Instrument for Home Care (RAI-HC). We provide data summarizing these proposed outcomes and evidence of known groups validity in a cross-national sample of home care clients. Data highlight the differing characteristics of clients across these agencies and provide evidence that this standardized data collection instrument can capture data that is reliable and valid for describing populations and evaluating program effectiveness.

The nursing home case-mix classification system, Resource Utilization Groups Version III (RUG-III), has been tested and refined for long-term home care clients. The study sample included 804 individuals seeking home care through the Michigan Care Management Program or the Home and Community Based Waiver for the Elderly and Disabled. Clients were classified, and RUG-III models were derived using the Minimum Data Set for Home Care (MDS-HC). A refined home care model, RUG-III/HC, was developed incorporating Instrumental Activities of Daily Living (IADLs) to the nursing home RUG-III classification. The model explained 33.7 per cent of the variance of per diem cost, using cost weighted formal and informal care as the dependent variable. Resource use within groups was relatively homogeneous. The case-mix index (CMI) of weighted formal and informal care time spanned an eight-fold range. Further analysis is suggested regarding the inclusion of informal care as a cost in case-mix classification for long-term home care clients.