Combining two forms of simulation to predict the potential impact of interface design on technology-induced error in healthcare

  • Authors:

  • Andre Kushniruk;Elizabeth Borycki;James G. Anderson;Marilyn M. Anderson

  • Affiliations:

  • University of Victoria, Victoria, BC, Canada;University of Victoria, Victoria, BC, Canada;Purdue University, West Lafayette, IN;Anderson Consulting, West Lafayette, IN

  • Venue:
  • Proceedings of the 2008 Spring simulation multiconference
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper will describe how simulations of clinical activity (involving human subjects carrying out clinical tasks) and mathematical computer-based simulations can be linked to forecast the impact of interface design features upon medical errors in healthcare information technology. The paper describes our approach in two phases. In Phase 1 a clinical simulation was conducted involving 10 physicians who were asked to use a hand-held prescription writing application to enter and record medications administered during a simulated clinical interaction. In this phase of the study data arising from the clinical simulation was collected and then analyzed using qualitative approaches to assess the relationship between aspects of interface design (i.e. usability problems) and medication error prescribing rates. In Phase 2, the base rates for error associated with specific types of usability problems (from Phase 1) formed the input into a computer-based mathematical simulation. Using this approach, comparative graphs of total mistakes and slips from Phase 1 were forecasted over time. The work described in this paper is unique in health care as it directly connects two distinct forms of simulations: (1) clinical simulations of user behavior and (2) mathematical simulation to forecast error rates over time. This approach links clinical simulations with computer simulations and demonstrates the impact of aspects of interface design upon medical error.