Background: Assistive technology and telecare have been promoted to manage the risks associated with independent living for people with dementia, but there is limited evidence of their effectiveness. Objectives: This trial aimed to establish whether or not assistive technology and telecare assessments and interventions extend the time that people with dementia can continue to live independently at home and whether or not they are cost-effective. Caregiver burden, the quality of life of caregivers and of people with dementia and whether or not assistive technology and telecare reduce safety risks were also investigated. Design: This was a pragmatic, randomised controlled trial. Blinding was not undertaken as it was not feasible to do so. All consenting participants were included in an intention-to-treat analysis. Setting: This trial was set in 12 councils in England with adult social services responsibilities. Participants: Participants were people with dementia living in the community who had an identified need that might benefit from assistive technology and telecare. Interventions: Participants were randomly assigned to receive either assistive technology and telecare recommended by a health or social care professional to meet their assessed needs (a full assistive technology and telecare package) or a pendant alarm, non-Monitored smoke and carbon monoxide detectors and a key safe (a basic assistive technology and telecare package). Main outcome measures: The primary outcomes were time to admission to care and cost-effectiveness. Secondary outcomes assessed caregivers using the 10-item Center for Epidemiological Studies Depression Scale, the State-Trait Anxiety Inventory 6-item scale and the Zarit Burden Interview. Results: Of 495 participants, 248 were randomised to receive full assistive technology and telecare and 247 received the limited control. Comparing the assistive technology and telecare group with the control group, the hazard ratio for institutionalisation was 0.76 (95% confidence interval 0.58 to 1.01; p = 0.054). After adjusting for an imbalance in the baseline activities of daily living score between trial arms, the hazard ratio was 0.84 (95% confidence interval 0.63 to 1.12; p = 0.20). At 104 weeks, there were no significant differences between groups in health and social care resource use costs (intervention group - control group difference: mean -£909, 95% confidence interval -£5336 to £3345) or in societal costs (intervention group - control group difference: mean -£3545; 95% confidence interval -£13,914 to £6581). At 104 weeks, based on quality-adjusted life-years derived from the participant-rated EuroQol-5 Dimensions questionnaire, the intervention group had 0.105 (95% confidence interval -0.204 to -0.007) fewer quality-adjusted life-years than the control group. The number of quality-adjusted life-years derived from the proxy-rated EuroQol-5 Dimensions questionnaire did not differ between groups. Caregiver outcomes did not differ between groups over 24 weeks. Limitations: Compliance with the assigned trial arm was variable, as was the quality of assistive technology and telecare needs assessments. Attrition from assessments led to data loss additional to that attributable to care home admission and censoring events. Conclusions: A full package of assistive technology and telecare did not increase the length of time that participants with dementia remained in the community, and nor did it decrease caregiver burden, depression or anxiety, relative to a basic package of assistive technology and telecare. Use of the full assistive technology and telecare package did not increase participants’ health and social care or societal costs. Quality-adjusted life-years based on participants’ EuroQol-5 Dimensions questionnaire responses were reduced in the intervention group compared with the control group; groups did not differ in the number of quality-adjusted life-years based on the proxy-rated EuroQol-5 Dimensions questionnaire.