An assessment instrument capable of measuring the wellbeing of the aged in a number of significant domains is described. This Philadelphia Geriatric Center Multilevel Assessment Instrument (MAI) systematically assesses behavioral competence in the domains of health, activities of daily living, cognition, time use, and social interaction and in the sectors of psychological wellbeing and perceived environmental quality. Determination of the psychometric qualities of measures of different length in each of these domains and sectors was made. The performance of 590 older people in groups composed of independent community residents, in-home services clients, and people awaiting admission to an institution was determined. The MAI is seen as useful for both research and for assessment in service-giving situations.

Groups of spouse (N = 285) and adult child (N = 244) caregivers of elderly parents suffering from Alzheimer's disease were interviewed regarding their caregiving behaviors, evaluations of caregiving, and general psychological well-being. A model of caregiving dynamics where the objective stressor, caregiver resources, and subjective appraisal of caregiving (operationalized as caregiving satisfaction and burden) were studied as they affected both positive affect and depression was tested. For spouses, caregiving satisfaction was not related to aspects of the stressor, but was a significant determinant of positive affect. Among adult children, high levels of caregiving behavior resulted in both greater caregiving satisfaction and burden. Burden, in turn, was related to depression in both groups but, among adult child caregivers, positive affect was not affected by caregiving satisfaction. Limited support was found for the hypothesis that the positive and negative aspects of caregiving contributed to analogous aspects of generalized psychological well-being but not to the opposite-valence outcomes.

Caregivers of disabled older people were studied in terms of their appraisal of the caregiving process. A conceptual approach based on stress theory suggested that such appraisal was broader than the traditional term, "caregiving burden." An item pool was constructed using traditional and new items to represent dimensions of subjective caregiving burden, caregiving satisfaction, caregiving impact, caregiving mastery, and traditional caregiving ideology. Component analysis of responses of 632 caregivers in a respite research project yielded factors that corresponded with those hypothesized; the content of similar factored responses from 239 caregivers in another study was quite similar. A series of exploratory and confirmatory factor analyses resulted in acceptance of the subjective burden, caregiving satisfaction, and caregiving impact factors. Used as composite item scores, evidence of their psychometric quality is presented.

Daughters and daughters-in-law of presently unmarried elders were studied longitudinally, and the data were analyzed to determine how two transitions in caregiving status affected the women of the younger generation. One transition compared noncaregivers who had become caregivers 1 year later ("caregiving entrants," n = 33) with continuing noncaregivers (n = 56) and with veteran continuing caregivers (n = 78) over the same period. The second transition followed Time 1 new caregivers as they became "new veteran" caregivers (n = 69), comparing them with "old veteran" caregivers (n = 189) over the same year. The transition to caregiving was marked by a decrease in the care receiver's competence and an increase in the amount of care received, but caregiving entrants' quality of life did not change significantly over 1 year, as compared with either continuing noncaregivers or veteran caregivers. Although longitudinal study shows little positive evidence for the wear-and-tear model of caregiving, methodological improvements are needed before discarding the hypothesis that caregiving erodes mental health.

The purposes of this research were first to describe how a group of urban elderly spend their days and, second, to examine the responsiveness of the behavioral day to rather marked environmental and personal differences. Detailed time budgets for one full day (including the content, duration, social and environmental context, and the amount of liking for each activity) were obtained from 535 persons, from four groups: independent community residents, public housing tenants, recipients of intensive in-home services, and an institutional waiting list group. Consistent patterns appeared across groups, particularly in the amount of time spent in the various social contexts and in most discretionary activities. However, the independent groups spent more time than the service groups away from home and in obligatory tasks such as housework, cooking, and shopping. Independent groups tended to express greater liking for many activities but not for more novel activities which the service groups liked equally as well.