Background: Forming implementation intentions (specifying when, where and how to act) has been proposed as a potentially effective and inexpensive intervention, but has mainly been studied in controlled settings for straightforward behaviors. Purpose: To examine if forming implementation intentions (II) could be used in large-scale, population-based interventions that aim to promote more complex and clinically relevant behavior change, we tested the impact of different II on increasing daily physical activity (PA) aimed at weight maintenance among 709 Dutch adults. Methods: At T0, participants were randomly allocated to a control group or to form II for 1) a prescribed action (walking), 2) self-selected activities, 3) self-selected activities and repeat making these II two times. All participants were asked to increase PA by at least two hours a week (15-20 minutes per day). Post-tests took place two weeks (response 85%), three months (response 78%) and six months (response 79%) post-intervention. Results: No main effects of II formation on BMIor physical activity were found. Intention to increase physical activity moderated the effects of repeated II, but not of the other II conditions. Forming repeated II had a positive effect on total PA and number of active days for respondents with strong intentions. Conclusion: Implementation intention interventions may not yet be ready for implementation on its own for large-scale obesity prevention in the general public. Future research should test strategies for optimal II formation in both initiating and maintaining behavioral change. © 2009 De Vet et al; licensee BioMed Central Ltd.