Background: Behavioural medicine is characterised by findings for the effectiveness and efficacy of complex behaviour change interventions. Comparatively, scant attention has been paid to key intervention components or mechanisms of action. Evaluating relationships between process variables (fidelity) and intervention effects is central to addressing this imbalance. Accordingly, in the current study, we sought to explore the magnitude and direction of effect between fidelity predictors (dietitian adherence and competence) and intervention effects (patient nutritional status) during the intervention phase of a real-world, stepped-wedge evaluation of ‘EAT: Eating As Treatment’. Methods: The EAT clinical trial was conducted within five major Australian hospitals located in Queensland, Western Australia, Victoria and South Australia between 2013 and 2016. EAT is a dietitian-delivered health behaviour change intervention designed to reduce malnutrition in head and neck cancer (HNC) patients undergoing radiotherapy. Dietitian adherence and competence ratings were derived from a 20% random sample of audio-recorded dietetic consultations (n=194) conducted after dietitians (n=18) were trained in EAT. Sessions were coded by trained, independent, coders using a study checklist, the Behaviour Change Counselling Index (BECCI) and items from the Cognitive Therapy Scale-Revised (CTS-R). Patient nutritional status was measured using the Patient-Generated Subjective Global Assessment (PGSGA). Results: Dietitian adherence to a written nutrition plan (β=7.62, 95% CI=0.65 to 14.58, p=0.032), dietitian adherence to behaviour change counselling (β=0.69, 95% CI =0.02 to 1.38, p=0.045) and competence in delivering behaviour change counselling (β=3.50, 95% CI =0.47 to 6.53, p=0.024) were significant predictors of patient nutritional status. Dietitian adherence and competence ratings were higher during consultations with intervention patients at greater risk of malnutrition. Conclusions: This study contributes new insights into the relationship between fidelity and treatment outcome by demonstrating that dietitian adherence and competence is greater when working with more challenging patients. This is likely central to the demonstrated success of the EAT intervention in reducing malnutrition and highlights the importance of ensuring that providers are adequately equipped to flexibly integrate intervention elements according to patient need. Trial registration: This study is a process analysis of a stepped-wedge randomised controlled trial prospectively registered on the Australian New Zealand Clinical Trials Registry (ACTRN12613000320752; Date of registration 21/03/2013).