We conducted a parallel group randomized controlled trial of children initially aged 2 to 3 y who were caries free, to prevent the children becoming caries active over the subsequent 36 mo. The setting was 22 dental practices in Northern Ireland, and children were randomly assigned by a clinical trials unit (CTU) (using computer-generated random numbers, with allocation concealed from the dental practice until each child was recruited) to the intervention (22,600-ppm fluoride varnish, toothbrush, 50-mL tube of 1,450 ppm fluoride toothpaste, and standardized, evidence-based prevention advice) or advice-only control at 6-monthly intervals. The primary outcome measure was conversion from caries-free to caries-active states. Secondary outcome measures were number of decayed, missing, or filled teeth (dmfs) in caries-active children, number of episodes of pain, and number of extracted teeth. Adverse reactions were recorded. Calibrated external examiners, blinded to the child’s study group, assessed the status of the children at baseline and after 3 y. In total, 1,248 children (624 randomized to each group) were recruited, and 1,096 (549 intervention, 547 control) were included in the final analyses. Eighty-seven percent of intervention and 86% of control children attended every 6-mo visit (P = 0.77). A total of 187 (34%) in the intervention group converted to caries active compared to 213 (39%) in the control group (odds ratio, 0.81; 95% confidence interval, 0.64–1.04; P = 0.11). Mean dmfs of those with caries in the intervention group was 7.2 compared to 9.6 in the control group (P = 0.007). There was no significant difference in the number of episodes of pain between groups (P = 0.81) or in the number of teeth extracted in caries-active children (P = 0.95). Ten children in the intervention group had adverse reactions of a minor nature. This well-conducted trial failed to demonstrate that the intervention kept children caries free, but there was evidence that once children get caries, it slowed down its progression (EudraCT No: 2009-010725-39; ISRCTN: ISRCTN36180119).