Background: Evidence suggests that the management of depression and anxiety in primary care is not consistent with current NICE guidance. Continuing medical education (CME) programmes attempt to address the gap between research and practice. This review aims to systematically identify controlled process evaluations of CME interventions to improve the management of depression and anxiety in primary care, in order to identify the effective components of educational/training interventions. Methods: Systematic searches were conducted in Medline, Embase, PsychInfo and Cinahl from 1950 to March 2011. Controlled studies evaluating CME interventions with primary care professionals to improve the management of depression or anxiety on process outcomes (practitioner behaviour, knowledge, attitudes) were included in the review. A narrative synthesis aimed to identify effective components of training interventions. Findings: Forty Papers (35 studies) were included in the review. The papers reviewed provide a mixed picture of the effective characteristics of training interventions. Two components: the inclusion of skills practice and the use of theory to inform intervention content are associated with positive outcomes. Limitations of the outcome measures used, a lack of theory and poor descriptions of the interventions makes it difficult to ascertain definitive effective characteristics of the training interventions evaluated. Discussion: The limitations of the papers reviewed support the development of a framework or taxonomy for the reporting, design and evaluation of health professional training interventions. Further research is needed to develop valid and reliable outcome measures of practitioner behaviour. Future studies should incorporate theory into the design and the evaluation of interventions in this area.