Background: Two-arm parallel group randomized controlled trial to measure the cost-effectiveness of caries prevention in caries-free children aged 2-3 years attending general practice. Methods: The setting was 22 dental practices in Northern Ireland. Participants were centrally randomized into intervention: 22,600 ppm fluoride varnish; toothbrush; 50 ml tube of 1,450 ppm fluoride toothpaste; and standardised prevention advice, and control: advice-only, both provided at 6-monthly intervals during 3-year follow-up. The primary outcome measure was conversion from caries-free to caries-active states assessed by calibrated and blinded examiners; secondary outcome measures included dmfs, pain and extraction. Cumulative costs were related to each of the trial’s outcomes in a series of incremental cost effectiveness ratios (ICERs). Sensitivity analyses examined the impact of using dentist’s time as measured by observation rather than that reported by the dentist. The costs of applying topical fluoride were also estimated assuming the work was undertaken by dental nurses or hygienists rather than dentists. Results: A total of 1248 children (624 randomised to each group) were recruited and 1096 (549 in the intervention group and 547 in the control group) were included in the final analyses. The mean difference in direct healthcare costs between groups was £107.53 (£155.74 intervention, £48.21 control, P<0.05) per child. When all healthcare costs were compared the intervention group’s mean cost was £212.56 more than the control group (£987.53 intervention, £774.97 control, P<0.05). Statistically significant differences in outcomes were only detected in respect of carious surfaces. The mean cost per carious surface avoided was estimated at £251 (95% CI £454.39 - £79.52). Sensitivity analyses did not materially affect the study’s findings. Conclusions: This trial raises concerns about the cost-effectiveness of a fluoride-based intervention delivered at the practice level in the context of a state-funded dental service.