About | Search | Report Feedback  

Even Better Together.

The quality and quantity of tools such as workshops, presentations and curriculum support is magnified when we work together.

Internal Medicine In-Training Examination

Developed by the American College of Physicians (ACP) and AAIM, the Internal Medicine In-Training Examination (IM-ITE) is a well-validated and reliable self-assessment instrument. The examination was first offered in 1988 and is now administered annually to more than 92% of internal medicine residents in the United States, Canada, Lebanon, Japan, Jordan, Saudi Arabia, Ireland, Barbados, Jamaica, Iceland, and Qatar. Residents at all levels of training in categorical, primary care, and medicine-pediatrics programs participate in IM-ITE.

Learn More

Visit the Internal Medicine In-Training Examination website to review frequently asked questions, view the examination blueprint, and learn how to register.

ACP Announces No Evidence of Impact from Testing Abnormalities for 2016 IM-ITE

Due to a problem with the web-based testing platform, test takers of the 2016 Internal Medicine In-Training Examination experienced repeated technical difficulties during days one through four of the test period. To ensure that programs and residents are provided with scores they can continue to rely upon, ACP--co-sponsor of the IM-ITE with AAIM--asked NBME (testing vendor) to perform a series of statistical analyses.

NBME conducted a multistep analytical process to determine whether multiple restarts occurring during the 2016 IM-ITE administration were associated with lower scores than would be expected. To do this, they first performed a series of statistical analyses to predict IM-ITE scores for PGY1s, PGY2s, and PGY3s based on previous test data that included USMLE Step 2 Clinical Knowledge (CK) scores for PGY1s, and USMLE Step 2 CK scores plus the prior year’s IM-ITE scores for PGY2s and PGY3s.

Next, the NBME compared these previous test scores (the predicted scores) with 2016 IM-ITE scores from the first 8 days of testing (the actual scores). Days 1 to 4 included test administrations characterized by multiple restarts and days 5 to 8 reflected restart rates that were minimal and occurred at usual baseline levels. These data allowed NBME to assess whether actual versus predicted scores were different for affected (days 1 to 4) and unaffected (days 5 to 8) residents. If the multiple restarts had an impact, any differences between actual and predicted scores should be greater for residents completing the ITE on days 1 to 4.

View analysis results (PDF)
After careful analysis, the NBME found no evidence that the testing abnormalities during days 1 to 4 had a meaningful impact on 2016 IM-ITE scores. Affected and unaffected residents performed as well as or better than predicted.  ACP and NBME continue to investigate these issues, will monitor scores through the end of this testing period, and are working diligently to implement measures to ensure that major disruptions to testing experiences do not occur in the future.
For more information about the analyses or the exam, please contact IM-ITE@acponline.org.