Objectives: To systematically review cost benefit studies of telemedicine. Design: Systematic review of English language, peer reviewed journal articles. Data sources: Searches of Medline, Embase, ISI citation indexes, and database of Telemedicine Information Exchange. Studies selected: 55 of 612 identified articles that presented actual cost benefit data. Main outcome measures: Scientific quality of reports assessed by use of an established instrument for adjudicating on the quality of economic analyses. Results: 557 articles without cost data categorised by topic. 55 articles with data initially categorised by cost variables employed in the study and conclusions. Only 24/55 (44%) studies met quality criteria justifying inclusion in a quality review. 20/24 (83%) restricted to simple cost comparisons. No study used cost utility analysis, the conventional means of establishing the "value for money" that a therapeutic intervention represents. Only 7/24 (29%) studies attempted to explore the level of utilisation that would be needed for telemedicine services to compare favourably with traditionally organised health care. None addressed this question in sufficient detail to adequately answer it. 15/24 (62.5%) of articles reviewed here provided no details of sensitivity analysis, a method all economic analyses should incorporate. Conclusion: There is no good evidence that telemedicine is a cost effective means of delivering health care.