BACKGROUND: Although the "learning curve" is commonly analyzed by splitting the data into arbitrary chunks of experience, this does not allow for precise estimation of where the curve plateaus or the rate at which learning is achieved. Our objective was to describe a simple way to characterize the learning curve for a fundamental laparoscopic task. METHODS: Sixteen medical students performed 40 repetitions of the Fundamentals of Laparoscopic Surgery (FLS) pegboard task and were scored using validated metrics. A learning curve was plotted and nonlinear regression was used to fit an inverse curve (Y = a - b/X), yielding an estimate of a (asymptote) and b (slope) for each subject. Two values were derived from these estimates: "learning plateau," defined as the theoretical best score achievable (when X = infinity, Y = a) and the "learning rate," defined as the number of trials required to reach 90% of potential (Y = 0.9a when X = 10 *b/a). Analysis of variance (ANOVA) was used to compare subjects reporting an interest in a surgical career (n = 4) to those not interested (n = 4) or undecided (n = 8). Data expressed as mean values +/- standard deviations. RESULTS: The raw starting score was 48 +/- 24, increasing to 94 +/- 8 for the 40th trial. The curve-fitting estimated "learning plateau" was 90 +/- 10 (range, 61-99), whereas the "learning rate," or the number of trials to 90% of potential, was 6 +/- 2 (range, 2-11). Subjects not interested in a surgical career had lower starting scores and learning plateau and slower learning rate compared with subjects interested in surgery or undecided (ANOVA; P < .05). CONCLUSION: Fitting an inverse curve allowed for estimation of learning plateau and learning speed for this fundamental laparoscopic task. These parameters allowed for comparisons to be made within subgroups of subjects and may have utility as an outcome for educational interventions designed to impact the learning curve.
BACKGROUND: Although the "learning curve" is commonly analyzed by splitting the data into arbitrary chunks of experience, this does not allow for precise estimation of where the curve plateaus or the rate at which learning is achieved. Our objective was to describe a simple way to characterize the learning curve for a fundamental laparoscopic task. METHODS: Sixteen medical students performed 40 repetitions of the Fundamentals of Laparoscopic Surgery (FLS) pegboard task and were scored using validated metrics. A learning curve was plotted and nonlinear regression was used to fit an inverse curve (Y = a - b/X), yielding an estimate of a (asymptote) and b (slope) for each subject. Two values were derived from these estimates: "learning plateau," defined as the theoretical best score achievable (when X = infinity, Y = a) and the "learning rate," defined as the number of trials required to reach 90% of potential (Y = 0.9a when X = 10 *b/a). Analysis of variance (ANOVA) was used to compare subjects reporting an interest in a surgical career (n = 4) to those not interested (n = 4) or undecided (n = 8). Data expressed as mean values +/- standard deviations. RESULTS: The raw starting score was 48 +/- 24, increasing to 94 +/- 8 for the 40th trial. The curve-fitting estimated "learning plateau" was 90 +/- 10 (range, 61-99), whereas the "learning rate," or the number of trials to 90% of potential, was 6 +/- 2 (range, 2-11). Subjects not interested in a surgical career had lower starting scores and learning plateau and slower learning rate compared with subjects interested in surgery or undecided (ANOVA; P < .05). CONCLUSION: Fitting an inverse curve allowed for estimation of learning plateau and learning speed for this fundamental laparoscopic task. These parameters allowed for comparisons to be made within subgroups of subjects and may have utility as an outcome for educational interventions designed to impact the learning curve.
Authors: J B Pagador; J Uson; M A Sánchez; J L Moyano; J Moreno; P Bustos; J Mateos; F M Sánchez-Margallo Journal: Int J Comput Assist Radiol Surg Date: 2010-08-11 Impact factor: 2.924
Authors: Ezra N Teitelbaum; Nathaniel J Soper; Fahd O Arafat; Byron F Santos; Peter J Kahrilas; John E Pandolfino; Eric S Hungness Journal: J Gastrointest Surg Date: 2013-09-04 Impact factor: 3.452
Authors: Daniel A Hashimoto; Pramudith Sirimanna; Ernest D Gomez; Laura Beyer-Berjot; K A Ericsson; Noel N Williams; Ara Darzi; Rajesh Aggarwal Journal: Surg Endosc Date: 2014-12-25 Impact factor: 4.584