BACKGROUND: Laparoscopic simulation has become a standard component of surgical training, but there is limited knowledge regarding skills transfer between procedural tasks. The objective was to investigate the specificity of procedural simulator training.

METHODS: This was randomized single-centre educational superiority trial. Surgical novices practised basic skills on a laparoscopic virtual reality simulator. On reaching proficiency, participants were randomized to proficiency-based training. The intervention group practised two procedures on the simulator (appendicectomy followed by salpingectomy), whereas the control group trained on only one procedure (salpingectomy). The main outcomes were number of repetitions and time to proficiency for the second procedure.

RESULTS: Ninety-six participants were randomized, of whom 74 per cent were women, with a median age of 26 years. The intervention group needed significantly fewer attempts than the control group to reach proficiency in the second procedure: median (i.q.r.) 22 (17-34) versus 32 (26-41) attempts, which corresponded to 24·1 per cent fewer attempts as assessed by multivariable analysis (P = 0·004). The intervention group required significantly less time than the control group to reach proficiency: median (i.q.r.) 88 (63-127) versus 131 (101-153) min respectively, corresponding to a difference of 31·1 min as assessed by multivariable analysis (P = 0·001).

CONCLUSION: Practising two procedures, compared with only one, reduced the number of attempts and time to reach proficiency in the second procedure. Skills transfer is seen between two tasks in laparoscopic simulator training; however, task specificity is still present when practising procedures.

REGISTRATION NUMBER: NCT02069951 (http://www.clinicaltrials.gov).