Problem representation and diagnostic accuracy during an OSCE

The National Academy of Medicine has called for "identifying opportunities to improve the diagnostic process". We studied the association between problem representation and diagnostic accuracy during an objective structured clinical exam (OSCE).

We conducted a non-randomized controlled trial during a ten-case OSCE. We measured whether a summary statement prompt increased the likelihood that the student listed the correct diagnosis and whether better summary statements were correlated with diagnostic accuracy.

114 students provided 1135 responses. The non-prompted control group yielded 631 responses, listing the correct diagnosis first 73% of the time and within the top three slots 85% of the time. The intervention group exposed to the prompt yielded 453 responses listing the correct diagnosis first 72% of the time (P = 0.617) and within the top three slots 84% of the time (P = 0.760). Summary statements were scored on a 0, 0.5, or 1 rubric. When grouped according to summary statement score, students listed the correct diagnosis first 74%, 70%, and 72% of the time respectively (P = 0.666). The correct diagnosis was included within the top three slots 88%, 82%, and 83% of the time (P = 0.238).

Prompting students to form a summary statement did not improve diagnostic accuracy. Better summary statements were not correlated with diagnostic accuracy.