Objective: Health behaviour change interventions (HBCIs), used in health education, health promotion, patient education and psychotherapy areas, are considered complex interventions. The objective of this article is to discuss the value and limitations of using randomised clinical trials (RCTs) to asses HBCIs. Methods: A scoping review of the literature was conducted to identify the main challenges of using RCTs for evaluating HBCIs. The issues were illustrated by case studies selected from research conducted by our multidisciplinary team. Results: In complex interventions, effects are produced not only by the intervention , but are strongly linked to context. Issues relating to transferability of results are therefore critical, and require adjustments to the RCT model. Sampling bias, biases related to the experimental conditions and biases due to the absence of double-blindness were examined and illustrated by case studies. Conclusion: The results underline the importance of a multidisciplinary approach. They call for adapted or alternative evaluation models that overcome the limitations of RCTs.