A systematic review of medical skills laboratory training: where to from here? Marita Lynagh, 1 Robert Burton 2 & Rob Sanson-Fisher 1 OBJECTIVES The aim of this review was to evaluate the effectiveness of medical skills laboratories or simulators. In particular, it aimed to determine if performance in medical skills laboratories is transferable to actual clinical performance and maintained over time. METHODS A range of databases was utilised to search for relevant papers published from 1998 to June 2006. Articles were included in the review if they met a number of criteria that included the evaluation of a skills laboratory or simulator for the purpose of procedural skills training, that participants were either undergraduate medical students or post- graduate medical trainees, and that the study used a randomised, controlled trial (RCT) research design in evaluation. RESULTS A total of 44 RCTs were identified for inclusion in the review. Overall, 32 (70%) studies reported that simulator training significantly improved procedural skills performance in compari- son with standard or no training. Twenty (45%) RCTs assessed the transfer of simulator performance to clinical skills performance; however, 8 of these used animal models, not real patients. Only 2 studies assessed the maintenance of skills post-intervention, both at 4-month follow-up periods. CONCLUSIONS Medical skills laboratories do lead to improvement in procedural skills compared with standard or no training at all when assessed by simulator performance and immediately post- training. However, there is a lack of well designed trials addressing the crucial issues of transferability to clinical practice and retention of skills over time. Further research must be carried out to address these matters if medical skills laboratories are to remain an integral component of medical education. KEYWORDS review [publication type]; clinical com- petence ⁄ *standards; laboratory techniques and pro- cedures ⁄ *standards; education, medical ⁄ *standards; teaching ⁄ *standards; randomised controlled trials. Medical Education 2007: 41: 879–887 doi:10.1111/j.1365-2923.2007.02821.x INTRODUCTION For over half a century since its origination in a 1929 amusement park ride which gave its passengers the impression of flying a plane, the aviation and aerospace industries have used simulation as a teaching tool. 1 Simulators are now widely used in education and training in a variety of professions and disciplines, including the military, commercial air- lines, nuclear power plants, business and medicine. 2,3 Denson and Abrahamson 4 first introduced human patient simulation to the medical community in 1969, when they used a patient anaesthesia simulator to augment resident training. Since then, increas- ingly sophisticated developments have occurred worldwide to improve the learning of individual and team reasoning, communication, and technical and other skills through the development of medical skills laboratories that employ various levels of simulation. 5,6 Early prototypes included bench models such as Ôhit the holeÕ manikins and other models that simulate human anatomy, and were commonly used to teach basic psychomotor skills, such as cannulation of the clinical skills 1 School of Medicine and Public Health, University of Newcastle, New South Wales, Australia 2 Department of Epidemiology and Preventive Medicine, Monash University, Victoria, New South Wales, Australia Correspondence: Dr Marita Lynagh, School of Medicine and Public Health, University of Newcastle, Locked Bag No. 10, Wallsend, New South Wales 2287, Australia. Tel: 00 61 2 4924 6328; Fax: 00 61 2 4924 6208; E-mail: Marita.Lynagh@newcastle.edu.au Ó Blackwell Publishing Ltd 2007. MEDICAL EDUCATION 2007; 41: 879–887 879