Apples and oranges? A review of evaluations of community-based long-term care.

This article synthesizes the contradictory findings of the community-based long-term care evaluation literature by grouping 13 studies into three models of care tested. All studies are reviewed according to tenets of internal and external/construct validity to ascertain what is "known" and "not known" about the effectiveness of this new type of care, and to specify areas needing further research. Findings suggest that increased preoperational specification of underlying theory, increased sophistication in targeting services to high-risk groups, use of multivariate analysis, and the development of more relevant outcome measures will improve the quality of future study findings, thereby contributing to theory and model building in this field.