Overwhelming odds: caregiving and the risk of institutionalization.

In this study we explored whether informally or formally provided long-term care services reduce the risk of institutionalization among the impaired elderly and the nature of the relationship between support and risk. We relied principally on the 1982 National Long-Term Care Survey (NLTCS) with supplementary data from the 1984 NLTCS and the national American Housing Survey. Five specific types of effects were tested: direct effects of formal and informal care on institutional risk; buffering effects of support moderating risk factors; supplementation effects of formal support on the risk-reducing impact of informal care; facilitating effects of the housing environment in enabling support to continue; and intervening effects mediating the direct effects. The effects of each type of support differ substantially. The analyses do not confirm the notion that formal, paid care reduces institutional risk. In fact, the use of paid caregivers is associated with greater risk of institutionalization, other things equal. The results for informal care are more mixed. A general measure that accounts for all types of informal support has no effect on institutional risk in any of the formulations tested. But the marginally significant effects of quality and proximate support in the direct effects model are consistent with the hypothesis that caregiving reduces institutional risk. The findings also indicate that a small number of housing or locational features are facilitators of formal support and particular types of informal support.