Machine learning is burgeoning in the clinical decision support domain, with the potential to bolster the power of decision support systems, improving data-informed clinical decision making. However, barriers persist to the adoption and regular use of machine-learning based clinical decision support systems (ML-CDSS), including the fact that many systems lack model transparency and understandability, precluding clinicians’ trust. One strategy for increasing model understandability and subsequent trust is incorporating clinical expertise in development of the machine learning model. However, clinician requirements for trusting ML-CDSS and evidence on the impact of involving clinical experts in model development for the purposes of facilitating trust are lacking.