Humanism: is its evaluation captured in commonly used performance measures?

There is an increased awareness of the importance of humanistic behavior and its education in the medical school curriculum. Relatively little is known about correlations between humanism and other performance measures.

To determine the correlation between humanism and other commonly used performance measures, and to determine if more humanistic students perform better, the same, or worse than less humanistic students.

During the Family Medicine clerkship, standardized patients (SPs) used the Physician Humanism Scale to assess 428 students for humanism. Clinical preceptors, SPs, written assignments, and a national knowledge examination also assessed student performance. The humanism scores were correlated with the SP and non-SP performance measures.

Humanism scores were significantly and positively correlated to all of the performance measures, but the correlation coefficients were low, ranging from .12 to .31. Students in the lowest quartile for humanism consistently scored lower for all of the performance measures, including both local and national exams (two-tailed significance < .018). Students with very low humanism scores still passed other performance evaluations.

The correlation between humanism and other performance measures is quite low, indicating that a separate measure for humanism provides different and additional information that current performance measures do not include. More humanistic students perform better than their less humanistic peers, but current performance measures do not identify students with the lowest scores on humanistic behavior. This study supports the inclusion of humanism as an additional, independent performance measure.