Abstract
Medical errors occur daily, and may rank among the highest causes of inpatient death in the US. Our study modeled modes of communication between patients and doctors to discover the best way to communicate to reduce error. We used narratives to model interview or text-box formats, and descriptor lists to model symptom checklists. We found that narratives produced a significant increase in accuracy (score_paper; F(27)=15.70, p<0.001, score_verbal; F(27)=22.26, p<0.001, score_computer; F(27)=4.82, p<0.05) with no significant increase in time compared to lists of descriptors (average lists time= 485.77s, average sentences time=479.50). We also found that sentences were less mentally demanding (F(1)=24.49, p<0.001) and difficult (F(1)=16.43, p<0.001) than lists, and that participants could effectively estimate their success at completing tasks (B=0.13, t(1)=5.94, p<0.001).
Get full access to this article
View all access options for this article.