Physician training requirements in sonography: how many cases are needed for competence?

Physician competence in the performance of sonographic studies was assessed after their involvement in predetermined increments of cases to determine whether the case volumes currently required by the American Institute of Ultrasound in Medicine and the American College of Radiology for training in sonography can be lowered substantially.

Sonographic competence tests were administered to 10 first-year diagnostic radiology residents after their involvement in increments of 50 cases, up to a total of 200 cases (four competency tests). Each competency test consisted of the resident's independently scanning and interpreting 10 clinically mandated studies that were scored in comparison with the examination performed by the sonographer and interpreted by an attending radiologist. Trainee studies were graded on the percentage of anatomic landmarks depicted, the number of reporting errors, the number of clinically significant reporting errors, and the percentage of cases receiving a passing score.

Although resident performance improved progressively with increasing experience for all parameters assessed, performance of the group was poor even after their involvement in 200 cases. At this testing level, the mean percentage of anatomic landmarks depicted successfully was 56.5%; the mean total reporting errors per case was 1.2; the mean clinically significant errors per case was 0.5; and the mean percentage of cases receiving a passing score was 16%. Impressive performance differences were observed among residents for all parameters assessed, and these differences were not explained by the number of months of radiology training the resident had taken before the sonography rotation.

Involvement in 200 or fewer cases during the training period is not sufficient for physicians to develop an acceptable level of competence in sonography.