A human factors investigation of medication alerts: barriers to prescriber decision-making and clinical workflow

AMIA Annu Symp Proc. 2009 Nov 14;2009:548-52.


Computerized medication alerts (e.g., drug-drug interaction alerts), which are intended to protect patient safety, should also be designed to support prescriber workflow. However, relatively few studies have examined the use of medication alerts during patient care processes. To assess barriers associated with the use of medication alerts, we directly observed medication prescribing during routine patient care. Prescribers (physicians, pharmacists, and nurse practitioners) were recruited from five outpatient primary care clinics at a major Midwestern Veterans Affairs Medical Center (VAMC). A total of 199 alerts were observed across 91 patients and 20 prescribers during normal patient care tasks. Through inductive qualitative analysis, we identified 15 barriers associated with medication alerts; herein, we describe five of the key barriers in detail. Results may be used to create alert redesigns, which have the potential to more fully support clinical workflow, prescriber decision-making, and patient safety.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Attitude of Health Personnel
  • Decision Making
  • Drug Interactions
  • Drug Therapy, Computer-Assisted / statistics & numerical data
  • Hospitals, Veterans
  • Humans
  • Medical Order Entry Systems* / statistics & numerical data
  • Medication Errors / prevention & control
  • Midwestern United States
  • Patient Care Management
  • Reminder Systems
  • User-Computer Interface
  • Workflow*