SUMMARYThe present study describes the development of a measure of stress in the family caregivers of institutionalized dementing elders using a brief two-dimensional scale. Univariate and multivariate analysis of the responses of 257 French adult caregivers to the scale suggest that a year after the admission to care unhappiness is expressed by a third of the population, principally due to exclusion from the caregiving role, financial burden, problems with family and poor contact with staff. Responses appear to be independent of the degree of dementia, the presence of associated pathology in the institutionalized elder and recency of admission. Significantly greater satisfaction with care and lower levels of stress were reported by caregivers with a relative in non-medical communal care as opposed to a long-stay hospital.KEY wom-Senile dementia, long-stay care evaluation.A great deal of attention has been given in recent years to the caregivers of elderly demented persons who still live in the community, as these persons not only relieve pressure on formal health services but are also themselves at high risk of physical and mental ill-health (Poulshock and Deimling, 1984;Robinson, 1983;Zarit et al., 1980;Morris et al., 1988). The wellbeing of the families of institutionalized persons with dementia has, on the other hand, been largely ignored on the assumption that institutionalization resolves the problem of caregiver burden. While research in a number of countries has indeed demonstrated a significant reduction in stress for family caregivers as a result of institutionalization (Levin et al., 1983;Wells and Jorm, 1987;Grad and Sainsbury, 1965;Gilleard et al., 1984b), a longitudinal study conducted in the United Kingdom Sainsbury, 1965, 1968;Sainsbury and Grad, 1970) revealed that 21% of family caregivers continued to experience stress despite the institutionalization of the dementing family member. This observation suggests that there are sources of stress in caregiving which are not met by long-term institutionalization and/or that institutionalization itself gives rise to new forms of stress which are not found in community care. Controlled studies of caregiver stress following institutionalization are presently needed in order (a) to identify the needs of this group and (b) to assess to what extent institutionalization of a dementing elderly person actually modifies morbidity rates in caregivers. The present article aims to address the first question by attempting to construct a measure of caregiver stress which targets sources of stress experienced by the families of institutionalized elderly persons, thereby complementing the extensive work which has been done in the area of domiciIiary caregiving. While a number of questionnaires exist for the study of domiciliary caregivers, these are not adequate for institutionalized persons, as institutionalization introduces new elements into the situation, such as satisfaction with care, while removing others, such as the need for surveillance.A review of the liter...