BACKGROUND AND STUDY AIMS: Most official training programs in colonoscopy recommend that trainees should carry out a minimum of 100 procedures, but limited data exist on the technical progress of trainees. The aim of our study was to estimate the number of supervised procedures required for obtaining competence in colonoscopy. MATERIALS AND METHODS: Between 1990 and 1997 we have prospectively evaluated the performance of eight consecutive trainees in colonoscopy. The extent of intubated colon was recorded after each endoscopic procedure. Regression analysis was used to study the effect on the trainees' success rates on the number of colonoscopies which they had done. RESULTS: Out of 2,255 colonoscopies carried out over an 8-year period 1,408 were suitable for evaluation, fulfilling the training criteria for diagnostic colonoscopy. Senior staff (n=2) did 430 colonoscopies with a success rate of 91%, which was significantly reduced to 79% (344/434 colonoscopies; X2=20.67, df=l, P<0.001), when taking over colonoscopies which trainees had failed to complete. Regression analysis of trainees' data (y=88.97-(2185/x), r=0.74, F1,21=23.43, P<0.001) showed success rates of 67 % (95 % CI, 59 to 75 %) and 77 % (95 % CI, 66 to 88 %), when 100 and 180 procedures, respectively, had been done. These figures had been attained by the end of the second and third year of training in colonoscopy. CONCLUSIONS: Our regression analysis model shows the technical progress of trainees over a 3-year period as they learned how to carry out colonoscopy. Depending on individual skill, between 100-180 procedures, done over a 2-3-year period, are required before trainees can be considered competent in colonoscopy.
S.Tassios et al. (Mon,) studied this question.