Effect of Having, but Not Consulting, a Computerized Diagnostic Aid

Area of Science:

• Medical Informatics
• Cognitive Psychology
• Human-Computer Interaction

Background:

• Physicians exhibit reluctance towards using computerized diagnostic aids (CDAs).
• Limited experimental research exists on the impact of *not* consulting readily available CDAs.

Purpose of the Study:

• To experimentally investigate the effects of physicians’ non-consultation of CDAs on perceived competence.
• To explore factors influencing algorithm aversion and CDA adoption in clinical settings.

Main Methods:

• Two experiments were conducted involving participants evaluating diagnostic scenarios.
• Experiment 1 compared competence ratings when a CDA was available but unconsulted versus when no CDA was mentioned.
• Experiment 2 manipulated reasons for non-consultation and CDA accuracy, assessing competence and negligence.

Main Results:

• Failure to consult a CDA did not significantly affect physician competence ratings.
• Non-consultation of CDAs did not lower physician ratings, even when the CDA was superior.
• Physician ratings decreased when they distrusted the CDA but surprisingly increased if they believed they knew the CDA's output.

Conclusions:

• Physicians' failure to consult CDAs does not necessarily reduce perceived competence.
• Algorithm aversion and trust dynamics play a crucial role in CDA adoption.
• Findings have implications for improving clinical practice and CDA integration.