Background: Earlier research showed that multi-component dyadic interventions - including a combination of intervention strategies and addressing both the person with dementia and caregiver - have a beneficial impact on the mental and physical health of people with dementia and their family caregivers. A randomized controlled trial (RCT) of a multi-component dyadic intervention, which is a translated and adapted version of an intervention that has been shown to be effective in the US by Teri et al. (2003), was performed. The effects on caregivers' mood (primary outcome), burden, general health, and salivary cortisol levels (secondary outcomes) were studied. Methods: Community-dwelling people with dementia and their family caregivers (N = 111 dyads) were randomly assigned. The experimental group received eight home visits during three months, combining physical exercise and support (psycho-education, communication skills training, and planning of pleasant activities). Both the physical exercise and support component were directed at both the person with dementia and the caregiver. The comparison group received monthly information bulletins and phone calls. There were three measurements at baseline (prior to the intervention), at three months, and at six months into the intervention. Data were analyzed with Generalized Estimating Equations (GEE) based on an intention-to-treat analysis of all available data. Results: All analyses showed no benefits of the intervention over time on any of the outcomes. Conclusion: The negative results might be explained by the translation and adaptation of the intervention that has been shown to be effective in the US: the intervention was shortened and did not include cognitive reframing. However, only the health effects on people with dementia and not on caregivers were studied in the US. Several other factors might also have played a role, which are important for future studies to take into account. These are: the usual health care in the country or region of implementation; the wishes and needs of participants for specific intervention components; the room for improvement regarding these components; the inclusion of positive outcome measures, such as pleasure, and the quality of the relationship.