Fourteen years of progress testing in radiology residency training: experiences from The Netherlands

D. R. Rutgers, F. van Raamt, W. van Lankeren, C. J. Ravesloot, A. van der Gijp, Th. J. ten Cate, J. P. J. van Schaik

To describe the development of the Dutch Radiology Progress Test (DRPT) for knowledge testing in radiology residency training in The Netherlands from its start in 2003 up to 2016. We reviewed all DRPTs conducted since 2003. We assessed key changes and events in the test throughout the years, as well as resident participation and dispensation for the DRPT, test reliability and discriminative power of test items. The DRPT has been conducted semi-annually since 2003, except for 2015 when one digital DRPT failed. Key changes in these years were improvements in test analysis and feedback, test digitalization (2013) and inclusion of test items on nuclear medicine (2016). From 2003 to 2016, resident dispensation rates increased (Pearson's correlation coefficient 0.74, P-value <0.01) to maximally 16 %. Cronbach´s alpha for test reliability varied between 0.83 and 0.93. The percentage of DRPT test items with negative item-rest-correlations, indicating relatively poor discriminative power, varied between 4 % and 11 %. Progress testing has proven feasible and sustainable in Dutch radiology residency training, keeping up with innovations in the radiological profession. Test reliability and discriminative power of test items have remained fair over the years, while resident dispensation rates have increased. • Progress testing allows for monitoring knowledge development from novice to senior trainee. • In postgraduate medical training, progress testing is used infrequently. • Progress testing is feasible and sustainable in radiology residency training.