A pilot program based on the substitution of domiciliary care for traditional institutional care was established by the Commonwealth of Pennsylvania to provide community-based care for individuals who could no longer live independently (aging, mentally retarded, mentally ill). To ascertain the cost saving potential of this program, 190 participants and a comparable pool of non-participants were followed for an average 10-month follow-up period. Within each subgroup, participants were disaggregated into two categories: those residing in a community setting at pretest, and those residing in an institutional setting. An analysis of medical care and social support service utilization profiles indicated that program savings exceeded program cost for five of the six study subsamples; the one exception was the mental retardation cohort residing in a community setting at pretest. Net savings were greatest for the three subsamples residing in an institutional setting at pretest. For all the subsamples, over 90 per cent of the program saving stemmed from a lower use of institutional placements.