Attempts to explore the mix between the provision of support by informal and formal sources typically involve evaluations of a single sample of seniors. This study examined the sources of support (i.e. informal, formal, mixed) for instrumental activities of daily living (IADL) received by seniors at two points in time, approximately 12 years apart. Sources of support were compared to consider whether those used by men and women in 1983 (n = 1479), were similar to those used by their counterparts more than a decade earlier (n = 2512). For women, the proportions receiving assistance from various sources were identical in 1983 and 1971. For men, they were not. Among seniors receiving some IADL assistance, the probability of receiving it from primarily informal sources in 1983, relative to 1971 was significantly lower for men than for women. Moreover, the probability of receiving support from primarily formal sources in 1983, relative to 1971 was significantly higher for men. These findings might imply that, for men, but not for women, the receipt of support has moved toward primarily formal sources and away from primarily informal sources. Several potential explanations for these findings are explored.

Using variables derived from Attachment Theory, gender differences in several motivational and structural factors on the levels of care provided by adult offspring to their parents are explored. Data from the Winnipeg Parent Support Project on informal helpers who were sons and daughters between the ages of 30 and 64 (n = 79) were analysed. Consistent with previous research, adult daughters reported greater caregiving than sons. The structural variables of Employment, Marriage and Presence of Children did not emerge as significant predictors of caregiving involvement for sons or daughters. Considerable support was found for Attachment Theory. Perceived Parental Dependence, followed by Frequency of Visits, emerged as the strongest predictors of sons' and daughters' caregiving involvement. Perceived Parental Ability accounted for additional variation in daughters', but not sons', caregiving. No support was found for Exchange Theory; reciprocity did not predict caregiving involvement. Implications for understanding the helping behaviour of adult sons and daughters are discussed.

This paper describes patterns of caring for people aged 65 or over with dementia in Canada, and the relationship this has with the well-being of caregivers. Data were drawn from a representative sample of elderly people, selected from the community and from institutions in the ten provinces of Canada as part of the Canadian Study of Health and Aging (CSHA). Interviews were held with the caregivers of people diagnosed with dementia, and with a comparison group of caregivers for non-demented persons. About half of all people with dementia in Canada are living in the community, and our results show that over 98 per cent of them have a caregiver. The caregivers in 94 per cent of these cases are unpaid family members, relatives or friends: the spouse in 37 per cent of cases and a daughter in 29 per cent. Caregivers rarely use community support services. Among those caring for someone in the community, those caring for a person with dementia are more likely to experience chronic health problems and depressive symptoms than are those caring for a non-demented elderly person. Those caring for a person in the community are much more likely to feel burdened than those whose loved one is in an institution, even though those who are institutionalized are more likely to suffer from severe dementia. These national figures support many of the findings reported from smaller, localized studies. The findings suggest that long-term care institutions are serving a role for Canada's seniors, particularly for the most demented, and that caregivers often provide care for loved ones even when suffering from chronic health conditions, depression, and burden themselves.

Previous work on the association between total aluminum (Al) concentrations and a measure of cognitive impairment is extended by considering the associations between this measure of cognitive impairment, not only with total Al, but also with turbidity and DOC both of which also contain Al. The multivariate, but not bivariate, analyses show that the odds of showing indications of mental impairment increase with turbidity but decrease with DOC, and this effect is most pronounced at a medium pH. It is suggested that the two variables considered could account for some of the unexplained features of the association between Al concentrations and measures of cognitive impairment. The other associations, namely those with F, pH, source of drinking water (ground or surface) remain similar to those reported previously in this series of papers.

The interest in preventive approaches to the health problems of older persons has increased substantially in recent years, partly in response to escalating health care costs. The present paper involves a critical examination of the potential usefulness of preventive home visits in the Canadian context in light of demonstration projects in other jurisdictions. The paper concludes with recommendations for Canadian research on the use of preventive home visits.

In this study we examine the impact of parent status and marital status on (1) the foundations of social support networks and (2) who one turns to if help is required with activities of daily living. Data are taken from structured interviews with 678 community-dwelling persons aged 55 and over in London, Canada. This sample overrepresents the childless, single (never married), and divorced, thus ensuring adequate numbers in each category for data analysis. Our findings indicate that marital status is more influential than parent status regarding differences in the foundations of social support. However, for more specific forms of assistance, the divorced and single are more likely than the married, and the childless are more likely than parents to rely on formal support and paid help and less likely to rely on family. As well, for both marital and parent status, there are significant interactions with gender.

Self-determination theory (Deci & Ryan, 1985a) has recently been applied to the motivation behind daily activities in old age (Vallerand & O'Connor, 1989). According to this theory, a primary determinant of motivation is the degree of self-determination in the living environment: actual opportunities for self-determination are assumed to affect the experience of self-determination, which in turn affects motivation. This prediction was tested among 129 residents from 11 nursing homes. Both actual opportunities for autonomy and experienced autonomy were associated with motivational styles. Furthermore, the effects of the objective environment on motivation were mediated by perceptions of the environment. The results provide suggestive support for a causal sequence that has proven difficult to establish in previous research.

The relationship between the MMSE as a measure of cognitive deficit and two procedures for assessing medical-legal competence is explored. The findings on 60 patients assessed for financial capacity by a multidisciplinary panel, and 41 published cases assessed using the HCAT for capacity to consent to treatment are analysed using logistic regression and ROC curves based on the binary outcome capable/incapable. Cognitive deficit is not a good indicator of the results for mental capacity obtained either by the multidisciplinary panel or the HCAT. The relationship between cognitive deficits and procedures for the allocation of decisional authority is unclear. Some discussion is given to account for this discrepancy.

This study assessed the reliability and validity of the Activities Checklist (Reid & Ziegler, 1981) as well as its utility for predicting cognitive functioning in the elderly. Data were drawn from two longitudinal projects where elderly subjects had been tested on two occasions approximately five years apart. Test-retest reliabilities were .54 and .43 for the larger (N = 158) and smaller (N = 76) samples respectively. Validity was demonstrated by an age shift towards more passive activities and by positive correlations with education, occupational status, health and well-being. Weighted versions of activity scores were not superior to unweighted versions as predictors of cognitive functioning. For both samples activity levels significantly predicted performance on some cognitive measures. It was concluded that the Activities Checklist is a potentially useful instrument to employ in the study of individual differences in cognitive aging.