METHODS A patient-randomised controlled trial (RCT) and a cluster RCT of the same intervention were conducted independently of each other. METHODS The evidence-based intervention (a powered toothbrush and behavioural advice on timing, method and duration of toothbrushing) was framed to target oral hygiene self-efficacy (Social Cognitive Theory) and action plans (Implementation Intention Theory) to influence oral hygiene behaviour and therefore clinical outcomes. The content and the delivery of the intervention were standardised as a series of steps-altogether taking approximately 5 min. The control groups received routine care, even if that included oral hygiene advice. METHODS The primary outcome measures were behavioural (timing, duration and method of toothbrushing) matching the advice given in the intervention. Secondary outcomes were cognitive measures of self-efficacy and planning, and clinical measures of plaque and gingival bleeding. RESULTS The study included 87 dental practices and 778 patients (patient RCT, 37 dentists and 300 patients; cluster RCT, 50 dentists and 478 patients). Controlled for baseline differences, pooled results showed that trial participants who experienced the intervention had better behavioural (timing, duration, method), cognitive (confidence, planning), and clinical (plaque, gingival bleeding) outcomes. Clinical outcomes were only significantly better in the cluster RCT, however. CONCLUSIONS A simple, theory-based intervention delivered within the constraints of a primary care environment was more effective than routine care in influencing patients' oral hygiene cognitions, behaviour and health. As clinical outcomes were significantly better only in the cluster RCT, the impact of trial design on results needs to be further explored.