Demonstrating Diagnostic Error in the Care of Simulated Pediatric Inpatients

Academic Article


  • Background and Objectives: Diagnostic errors in medicine result in substantial harm to patients. Premature diagnostic closure, the acceptance of a diagnosis before its full verification, is a common source of error. We hypothesized that premature closure could be demonstrated during simulation with pediatric residents. Methods: Resident teams participated in one of two pairs of simulated patient cases. A text page to the residents briefly describing the patient started each case. The description contained either a symptom (Sx) or an incorrect diagnosis (iDx). The remainder of each case was scripted and presented identically. Data regarding evaluation, treatment, and correct diagnosis was collected in real time on a standardized recording tool. Comparisons between case descriptions (Sx vs. iDx) were analyzed with Fisher’s exact test and the nonparametric Wilcoxon test for equal location. Results: Sixty-one residents participated. In all four cases, residents seemed to more frequently consider the correct diagnosis when they started with a symptom than when they started with the incorrect diagnosis (67, 75, 88, 93 % vs. 24, 44, 87, 88 % for cases 1–4, respectively). In three cases, groups who began with the symptom may have considered the correct diagnosis early in the case (22, 50, 25, 67 % vs. 0, 22, 47, 50 %). Symptom-receiving residents also obtained more or equivalent information and more frequently began appropriate therapy. These results did not reach statistical significance. Conclusion: Simulated cases provide an opportunity to evaluate cognitive error and may allow teaching avoidance strategies.
  • Published In

    Digital Object Identifier (doi)

    Author List

  • Nassetta LB; Tofil NM; Kim YI; Eschborn S; White ML
  • Start Page

  • 623
  • End Page

  • 629
  • Volume

  • 26
  • Issue

  • 4