Objective: To use a model of health learning to examine the role of health-learning capacity and the effect of a school-based oral health education intervention (Winning Smiles) on the health outcome, child oral health–related quality of life (COHRQoL). Setting: Primary schools, high social deprivation, Ireland/Northern Ireland. Design: Cluster randomised controlled trial. Method: A total of 383, 7- to 8-year-old children were invited to participate and randomly allocated into intervention and control conditions. Baseline and 12-month follow-up assessments of COHRQoL, self-esteem, toothbrushing–fluoride toothpaste knowledge and unstimulated saliva samples were made. An 18-hour post-brushing, saliva fluoride concentration was used to assess toothbrushing with fluoride toothpaste (behaviour). The data were entered onto SPSSv22. Structural equation modelling was applied using AMOSv22 to test for the role of health-learning capacity (baseline self-esteem and COHRQoL) and simultaneous effects of Winning Smiles upon knowledge, behaviour and COHRQoL (at follow-up). Results: A total of 238 children participated at baseline and follow-up. A partial latent hybrid model fitted the data reasonably well (χ² = 65.6, df = 50, p = .07) as shown in addition by a Comparative Fit Index of .97 and a Root Mean Square Error of Approximation (RMSEA) value of .042 (90% confidence interval [CI]: .00, .06). The intervention had a significant effect on toothbrushing–fluoride toothpaste knowledge ( p < .03) and an effect on COHRQoL at the 6% level ( p < .06). Knowledge was strongly associated with saliva fluoride concentration ( p < .002). Conclusion: The model of health-learning capacity assisted in explaining the effect of a school-based intervention upon knowledge, toothbrushing behaviour and tentatively on COHRQoL.