Improving surgical training : maximising the benefits of surgical simulation
Surgical training is undergoing a period of great change, partly due to the increasingly complex and specialised nature of modern surgical practice. Other factors include working hour restrictions and the need for efficient turnover which restrict trainingtime, and the growing awareness of ethical issues surrounding training on patients. Traditional apprenticeship models of surgical training are increasingly unfeasible and are being replaced by skills lab training, utilising bench models and simulators. A growing body of research supports the value of surgical simulators in surgical training. We aim to investigate ways to improve surgical training, investigating factors which may increase the benefits of simulator-based training programmes.
We trained a total of 80 surgical trainees and medical students to perform a variety of image-guided procedures on virtual reality-based surgical simulators. The full procedures we selected were colonoscopy, hand-assisted laparoscopic colectomy and endovascular renal artery stenting. In addition, we used basic surgical tasks as part of the training programmes. The 3 full procedures and basic tasks were used during the 4 principle studies from which the data was collected. All our subjects underwent training and assessment. Using these procedures, we investigated the impact of proximate feedback on performance, the value of expert versus non-expert feedback, performance although the relationship is not yet clearly defined. In a moreexperienced cohort we found a relationship between specific visuospatial abilities and image-guided procedural performance, and in the intermediate and less experienced cohort, pyschomotor ability was more relevant. Perceptual ability appeared to be less important. Attaining proficiency improves skill retention which otherwise partially declines in the absence of reinforcement. Procedure specific error scores showed better retention than instrument handling scores, which may reflect the value of a cognitive curriculum. A proficiency-based progression training programme for colonoscopy was created, utilising many of these factors and this led to an improved clinical performance. We found high correlations between instrument handling and generic performance metrics on all the simulators, but lower correlations between error scores and these metrics. All our assessment systems demonstrated high inter rater reliability.
We have identified many ways to optimise simulator-based surgical training. Simulator-based training is likely to become as integral to surgery as to the aviation industry and it's potential benefits should be maximised. The findings from this research could be incorporated into curriculum design for future training courses. Improving surgical training should ultimately improve patient care. transferability and retention of surgical skills, trainee self-assessment, the relationship between innate ability and performance and the impact of simulator-based training curricula on real clinical performance. We also examined the internal consistency of performance metrics and outcomes and assessed inter-rater reliability for our assessments.
Our data demonstrate a steep learning curve during repetition of a simulated procedure, highlighting the value of intensive skills lab training. Standardised feedback reduces error commission, and has less effect on more generalised aspects of performance which improve in the absence of feedback. Error commission is aseparate performance metric from the performance assessments which are provided by the simulator. Error scores and more general scores have individual merit but clinically relevant performance metrics should be emphasised during training. Feedback from an expert reduces error commission more than feedback by a non-expert, but has less effect on more generic aspects of performance. However, when errors are objectively assessed, feedback from a non-expert who is familiar with the simulator is equally effective to feedback from an expert. Subject confidence risesafter intensive skills training sessions, and self-assessment reflects overallperformance improvements, although is less accurate for specific performance markers. There is a wide range of inter-subject variability for accuracy of self-assessment, and this self-assessment should be an integral part of all training curricula. Practice on non-surgical tasks such as video games appears to improve performance although our results did not reach statistical significance, likely due to aninsufficient training schedule. Innate ability testing is relevant to surgical skills