Continued Validation of Ultrasound Guidance Targeting Tasks: Relationship with Procedure Performance

To determine if deliberative practice with novel ultrasound guidance targeting tasks improves simulated procedural skill. In a nonrandomized interventional trial first year medical students practiced the previous described dowel and straw targeting tasks 1 hour a week for 4 weeks (training group) or had no training (controls). Afterward, they each performed a simulated amniocentesis (AMN) and chorionic villus sampling (CVS) procedure. Procedures were scored using a global rating scale (GRS) and compared between groups with Mann-Whitney U tests. Two-way random effects intraclass correlation coefficients for the inter- and intra-rater variability were calculated for each item in both GRS's. The training group (n = 22) had higher scores on several aspects and overall performance of AMN compared to controls (n = 15). There were no differences between groups for CVS. The inter-rater and intra-rater reliability of the GRS's for both AMN and CVS ranged from 0.16 to 0.89 with most values demonstrating good to excellent agreement. This study demonstrates validity evidence in the content and internal structure domains for the AMN and CVS simulators and their accompanying GRS's. Repetitive practice of the targeting tasks improved student performance in simulated AMN, but modifications are needed for it to be relevant to other procedures such as CVS.