Background The efficacy of multi-domain interventions for dementia prevention has been investigated in a few large randomized controlled trials (RCT). To link research to community action, we evaluated the implementation and effectiveness of a community-based multi-domain program for older adults at risk of cognitive impairment. Methods Three enterprises implemented a 24-week, bi-weekly, multi-domain program for older adults at risk of cognitive impairment through neighborhood senior centres (SCs). The program comprised a combination of dual-task group physical exercise, pen-and-paper cognitive games, computerised cognitive training, and mobile application-based personalized nutritional guidance. An effectiveness-implementation hybrid design using a RCT design and informed by the RE-AIM (Reach, Effectiveness, Adoption, Implementation, Maintenance) framework was adopted to evaluate the program. Cognition and quality of life were assessed using the Repeatable Battery for the Assessment of Neuropsychological Status (RBANS), and EuroQol EQ-5D-5L (EQ-5D) respectively, at baseline and 24-weeks. Blood lipid panel and physical assessments were also conducted. Questionnaires on implementation outcomes were administered at the participants-, provider- and community-levels to participants, implementers, and SC managers. Program schedules and attendance were obtained from implementers and the research team conducted unannounced, random and non-intrusive observations of the program. Results The intervention program reached almost 50% of eligible participants, had an attrition rate of 22%, and was adopted by 8.7% of the SCs approached. The intervention was implemented as intended, except the nutritional component that was re-designed due to participants’ unfamiliarity with the application. There were no between-group differences in cognition, quality of life, and blood lipid panel, though there was a reduction in quality of life in the control group and improved physical function in the intervention group at 24-weeks. Intervention group participants reported improvement in their physical and cognitive functions. The program was well-received by participants and centre managers, who wanted the intervention to continue post-study. Conclusions A six-month multi-domain program for older adults at risk of cognitive impairment can be implemented through neighborhood SCs and elicit improvement in physical but not cognitive performance. Implementation evaluation highlight areas to improve on the uptake of such community-based interventions. Trial registration Trial registration: ClinicalTrials.gov NCT04440969. Registered 22 June 2020 - Retrospectively registered, https://clinicaltrials.gov/ct2/show/NCT04440969