OBJECTIVES:

While mentoring has been associated with research productivity, the specific characteristics of successful mentoring have not been well studied. Thus, we performed a case-control study to identify characteristics of successful mentoring programs.

METHODS:

Institutions were divided based on number of plenary research presentations at an annual society meeting over 6years. Case institutions (Group A) had more presentations vs. controls (Group B). A survey of professors and research fellows assessed characteristics of their mentoring program. Chi-square and logistic regression analyses were performed.

RESULTS:

Of 159 surveyed, response rates were 46% for professors and 51% for fellows. Compared to Group B, Group A was more likely to have: an additional year of protected fellowship research training (62% vs. 24%; p=0.003), an established program to connect a mentor and mentee with similar research interests (52% vs. 27%; p=0.049), methods to provide feedback to mentors (62% vs. 29%; p=0.01), require mentee research progress reports (45% vs. 21%; p=0.047), and report ease of identifying a mentor (90% vs. 69%; p=0.046). On multivariate analyses, the additional year of research training (OR=7.53, 95% CI: 2.10-27.09; p=0.002) and ease at identifying a research mentor (OR=7.45, 95% CI: 1.44-38.6; p=0.017) remained as independent factors associated with higher research productivity.

CONCLUSIONS:

Our data suggest that programs can enhance research productivity with the incorporation of accountability features including formalized reports of progress and mentorship feedback in fellowship training. Facilitating the identification of a mentor and providing an additional year of research may be independent factors associated with research productivity.

Copyright © 2012. Published by Elsevier Inc.