Aim To establish the effectiveness of a restorative home support service on institutional-free survival in frail older people referred for needs assessment. Method A randomised controlled trial of Community Flexible Integrated Responsive Support Team versus usual care in 113 older people at risk of institutionalisation in New Zealand. Primary outcome was death or permanent institutional care evaluated in a Cox proportional hazards model adjusted for baseline prognostic variables. A variety of secondary outcomes including the health of informal caregivers were also evaluated. Results The primary outcome was lower, but not statistically significant, in the intervention group (adjusted hazard ratio 0.69, 95% confidence interval 0.4 to 1.16). Informal caregivers of interventional participants had significantly better physical and mental component scores on the 36-item Short-Form questionnaire. Conclusion Restorative home support may reduce the risk of death and institutionalisation in frail older people without affecting the health of informal caregivers.