Evaluating an automated mental health care system: making meaning of human-computer interaction

  • Authors:

  • Ramesh Farzanfar;Sophie Frishkopf;Robert Friedman;Kevin Ludena

  • Affiliations:

  • Medical Information Systems Unit, Boston University Medical Center, 720 Harrison Avenue, Suite 1102, Boston, MA 02118, USA;Medical Information Systems Unit, Boston University Medical Center, 720 Harrison Avenue, Suite 1102, Boston, MA 02118, USA;Medical Information Systems Unit, Boston University Medical Center, 720 Harrison Avenue, Suite 1102, Boston, MA 02118, USA;Medical Information Systems Unit, Boston University Medical Center, 720 Harrison Avenue, Suite 1102, Boston, MA 02118, USA

  • Venue:
  • Computers in Human Behavior
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

Objectives:: To qualitatively evaluate the response of patients with unipolar depression who used a computer telephony system designed to monitor their disease severity and support self-care, principally adherence to medication regimen and clinical office visit attendance. Methods:: Weekly in-depth interviews were conducted with 15 patients who used the computer telephony system for 4 weeks. Users had a diagnosis of unipolar depression and took at least one antidepressant. All interviews were audio-taped and immediately transcribed. The transcripts of the interviews were subsequently coded and analyzed thematically by two qualitative researchers. Results:: The patients spoke about the automated system as if it was a social actor. They did not, however, have an illusion that there was a health professional communicating through the system. Instead, they felt that it was designed to appear human-like. The majority offered suggestions intended to make the system behave and sound more like a ''human professional'' and less like a ''machine''. They believed that the system would be more usable, acceptable and effective if these changes were made. Conclusions:: These results do not support the ''anthropomorphism'' construct which posits that users of computer-mediated systems who attribute human qualities to the system are under an illusion that the system is human.