Fit as a Fiddle or Sick as a Dog: Effects of Subjective Patient Reports on Uptake of Clinical Decision Support

This paper reports research on improving decisions about hospital discharge - a critical healthcare quality and cost determinant identified by the Centers for Medicare and Medicaid Services. We report an experiment on effects of subjective information about patients' health status on discharge decisions as well as uptake of recommendations from a clinical decision support system (CDSS). Subjective information about readiness for discharge was obtained during examinations of standardized patients, who are regularly employed in medical education, but in our experiment had been given scripts developed for the experimental treatments. The CDSS presents evidence-based discharge recommendations obtained from econometric analysis of data from de identified electronic health records (EHR) of hospital patients. Subjects in the experiment were third and fourth year medical students. We find that the CDSS decreases hospital stay by one day while decreasing readmissions of high-risk patients. Subjects are responding appropriately to information conveyed by standardized patients when such information is consistent with the EHR. Compared to patient discharge from the hospital absent patient reports, Eager patients when also EHR-Fit are at least seven times more likely to be discharged whereas Reluctant patients when also EHR-Sick are about four times less likely to be discharged.