PURPOSE: We analyzed multi-institutional data to understand the relationship of ACR Diagnostic Radiology In-Training Examination (DXIT) scores to ABR Core Examination performance.

METHODS: We collected DXIT rank scores and ABR Core Examination outcomes and scores for anonymized residents from 12 different diagnostic radiology residency programs taking the ABR Core Examination between 2013 and 2019. DXIT scores were grouped into quintiles based on rank score for residency year 1 (R1), residency year 2 (R2), and residency year 3 (R3) residents. Core outcome was scored as fail when conditionally passed or failed. Core performance was grouped using SD from the mean and measured by the percent of residents with scores below the mean. Differences between DXIT score quintiles for Core outcome and Core performance were statistically evaluated.

RESULTS: DXIT and Core outcome data were available for 446 residents. The Core Examination failure rate for the lowest quintile R1, R2, and R3 DXIT scores was 20.3%, 34.2%, and 38.0%, respectively. Core performance improved with higher R3 DXIT quintiles. Only 2 of 229 residents with R3 DXIT score > /= 50th percentile failed the Core Examination, with both failing residents having R2 DXIT scores in the lowest quintile.

CONCLUSIONS: DXIT scores are useful evaluation metrics to identify a subgroup of residents at significantly higher risk for Core Examination failure and another subgroup of residents at significantly lower risk for Core Examination failure, with increasing predictive power with advancing residency year. These scores enable identification of approximately one-half of R3 residents whose risk of Core Examination failure is negligible.