The impending commercial release of affordable VR systems is likely to accelerate both the opportunity and demand for VR applications that specifically target psychological conditions. The aim of this study was to conduct a meta-analysis of outcomes associated with VR psychological interventions and to examine the methodological rigour used in these interventions. Literature search was conducted via Ovid, ProQuest Psychology Journals and ScienceDirect (Psychology) databases. Interventions were required to: be published between 1980 to 2014; use a randomised controlled trial design; be published in a scholarly journal; focused primarily on psychological/behavioural intervention; include validated measures; include reported means and standard deviations of outcome measures; and include one group with clinical/subclinical disorders, syndromes or distressing behaviours. Thirty eligible studies were identified. Random effects meta-analysis found an overall moderate effect size for VR interventions. Individual meta-analyses found an overall large effect size against non-intervention wait-lists and an overall moderate effect size against active interventions. No correlation was found between treatment outcomes and methodological rigour. Limitations may include limited study numbers, the use of a single coder, a need for more in-depth analyses of variation in form VR intervention, and omission of presence as a moderating factor. The current review supports VR interventions as efficacious, promising forms of psychological treatment. Use of reporting guidelines such as the CONSORT and CONSORT-EHEALTH statements should promote greater emphasis on methodological rigour, providing a firm foundation for the further development of clinical VR applications.