The aim of this study was to examine the effectiveness of a new, integrated service delivery (ISD) network of health and social services for frail elderly living in a semi-urban community. A quasi-experimental study was conducted from 1997 to 2000, with measures taken before implementation (T0) and every 12 months after implementation for a 3-year period (T1, T2, T3); 482 people aged 75 years or older from 2 communities, 272 in the experimental and 210 in the control group, were followed. Analyses were conducted using parametric and non-parametric statistics, a generalized linear model, and Cox regression for survival analysis. A declining trend in institutionalization was observed and the desire to be institutionalized was lower in the study group. When absence of deterioration on follow-up was analysed in terms of the level of autonomy, the ISD network produced positive effects on the frailer clientele at T1, effects that tended to be maintained at T2. In regard to caregivers' burden, the ISD network seemed to have been effective, with caregivers' burden being less at T1 and T2. The ISD network did not have a marked effect on the utilization of services. The ISD network did, in fact, produce some effects that should be seen as important for the elderly population.