Evaluate the reliability of the apprenticeship in the first year of medical school: towards a reliable first level ultrasound examination.

ObjectivesOur aim was to motivate apprentices sonographer needs, to appraise their own measurements, to reduce inconsistencies within and between operators. Deep knowledge of ultrasound sectional anatomy is mandatory for an appropriate performance. MethodsIn three different weekdays, 3 sonographer apprentices (rater), randomly selected from a cohort of San Paolo Medical School first year students participated in vertically integrated study of living anatomy through ultrasound examination, repeated lumbar multifidus cross-sections scans on 6 subjects at lumbar level. The Agreement R package 0.8-1 was used to monitored the performances of each apprentice. ResultsThe agreement (CCCintra 0.6749; CCCinter 0.556; CCCtotal is 0.5438) was further from least acceptable CCC of 0.92-0.95. The precision indices (precisionintra 0.6749; inter 0.801; total0.6274) were unsatisfactory, while the accuracy was high (0.9889 to 0.9913). The same occurred for the agreement on rater performances comparisons, where readings were high accurate (0.9537 to 0.9733) but moderately precise (0.7927 to 0.8895), not interchangeable TIR (1.173) but without rater supremacy. IIR (r1 vs r2 1.104, r1 vs r3 1.015, r2 vs r3 0.92) 95% confidence limits. ConclusionsApprentices were not reliable, repeatable, interchangeable. The weak link in the method seemed to be cultural weakness on vivo imaging morphologies, qualitative and quantitative measurement procedure on elementary statistical processing.