Background: Despite interest in using virtual humans (VHs) for assessing health care ­communication, evidence of validity is limited. We evaluated the validity of a VH application, MPathic-VR, for assessing performance-based competence in breaking bad news (BBN) to a VH patient.
Methods: We used a two-group quasi-experimental design, with residents participating in a 3-hour seminar on BBN. Group A (n=15) completed the VH simulation before and after the seminar, and Group B (n=12) completed the VH simulation only after the BBN seminar to avoid the possibility that testing alone affected performance. Pre- and postseminar differences for Group A were analyzed with a paired t -test, and comparisons between Groups A and B were analyzed with an independent t -test.
Results: Compared to the preseminar result, Group A’s postseminar scores improved significantly, indicating that the VH program was sensitive to differences in assessing performance-based competence in BBN. Postseminar scores of Group A and Group B were not significantly different, indicating that both groups performed similarly on the VH program.
Conclusion: Improved pre–post scores demonstrate acquisition of skills in BBN to a VH patient. Pretest sensitization did not appear to influence posttest assessment. These results provide initial construct validity evidence that the VH program is effective for assessing BBN performance-based communication competence.
Keywords: verbal behavior, health communication, informatics, clinical competence, empathy

摘要视频链接：Initial construct validity evidence of VH communication assessment