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

Farzanfar, Ramesh; Frishkopf, Sophie; Friedman, Robert H.; Ludena, Kevin

doi:10.1016/j.chb.2004.11.015

Cited by 26 publications

(32 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In particular, Farzanfar et al [41] found that depressed adults needed more empathy and that they found the system to be insensitive in a suicide-risk situation. There were some results that were not predicted in advance.…”

Section: Resultsmentioning

confidence: 99%

Active Assistance Technology for Health-Related Behavior Change: An Interdisciplinary Review

Kennedy¹,

Powell²,

Payne³

et al. 2012

J Med Internet Res

View full text Add to dashboard Cite

BackgroundInformation technology can help individuals to change their health behaviors. This is due to its potential for dynamic and unbiased information processing enabling users to monitor their own progress and be informed about risks and opportunities specific to evolving contexts and motivations. However, in many behavior change interventions, information technology is underused by treating it as a passive medium focused on efficient transmission of information and a positive user experience.ObjectiveTo conduct an interdisciplinary literature review to determine the extent to which the active technological capabilities of dynamic and adaptive information processing are being applied in behavior change interventions and to identify their role in these interventions.MethodsWe defined key categories of active technology such as semantic information processing, pattern recognition, and adaptation. We conducted the literature search using keywords derived from the categories and included studies that indicated a significant role for an active technology in health-related behavior change. In the data extraction, we looked specifically for the following technology roles: (1) dynamic adaptive tailoring of messages depending on context, (2) interactive education, (3) support for client self-monitoring of behavior change progress, and (4) novel ways in which interventions are grounded in behavior change theories using active technology.ResultsThe search returned 228 potentially relevant articles, of which 41 satisfied the inclusion criteria. We found that significant research was focused on dialog systems, embodied conversational agents, and activity recognition. The most covered health topic was physical activity. The majority of the studies were early-stage research. Only 6 were randomized controlled trials, of which 4 were positive for behavior change and 5 were positive for acceptability. Empathy and relational behavior were significant research themes in dialog systems for behavior change, with many pilot studies showing a preference for those features. We found few studies that focused on interactive education (3 studies) and self-monitoring (2 studies). Some recent research is emerging in dynamic tailoring (15 studies) and theoretically grounded ontologies for automated semantic processing (4 studies).ConclusionsThe potential capabilities and risks of active assistance technologies are not being fully explored in most current behavior change research. Designers of health behavior interventions need to consider the relevant informatics methods and algorithms more fully. There is also a need to analyze the possibilities that can result from interaction between different technology components. This requires deep interdisciplinary collaboration, for example, between health psychology, computer science, health informatics, cognitive science, and educational methodology.

show abstract

Section: Resultsmentioning

confidence: 99%

Active Assistance Technology for Health-Related Behavior Change: An Interdisciplinary Review

Kennedy¹,

Powell²,

Payne³

et al. 2012

J Med Internet Res

View full text Add to dashboard Cite

show abstract

“…For example, qualitative evaluations have shown that in some programs users anthropomorphize TLC applications although they are fully aware that TLC is only a computer [48][49][50]. The experience of anthropormorphization is enhanced by the use of a recorded human voice rather than a computerized voice.…”

Section: Treating Impulsivitymentioning

confidence: 94%

“…Indeed, in the mental health assessment study, the less human the voice sounded, the more users felt that they were talking to a computer and were thus less worried about confidentiality and privacy issues [52]. On the other hand, other users have told us that they generally prefer interaction with a person and not with TLC because they can have open-ended conversations with a human counsellor; they experience TLC as less adaptive than a human counselor, who would be able to address particularities of different situations [49]. To conclude, it seems that user perceptions of technologies such as TLC could mostly depend on specific personality orientations and users' likes or dislikes of a system might have more to do with users' predisposition than system characteristics.…”

Section: Treating Impulsivitymentioning

confidence: 95%

“…In fact, we have modulated the TLC voice to either enhance the human-like aspects of TLC or diminish them. In a prestudy regarding a TLC-system for advice-giving to depressed patients regarding adherence to their medication regimen, most users preferred a female voice, considering it more nurturing, and advised us to make sure that the voice sounded compassionate and reassuring; we thus needed to enhance human-like attributes of the voice [49]. On the other hand, in two other studies of TLC, one a physical activity promotion program and the other a mental health assessment system, we were told by users that enhancing human-like characteristics in the voice should be avoided as users prefer to talk to a computer and not a human being and any similarities could be unsettling [51,52].…”

Section: Treating Impulsivitymentioning

confidence: 98%

“…To conclude, it seems that user perceptions of technologies such as TLC could mostly depend on specific personality orientations and users' likes or dislikes of a system might have more to do with users' predisposition than system characteristics. Given the above, it certainly seems possible that patients in any treatment environment could well respond positively to a TLC system TLC applications have been developed for targeted health-related behaviors such as diet [53], physical activity [54][55][56], medication adherence [57], smoking cessation [58], chronic obstructive pulmonary disease [COPD; 59] and problem drinking [43], as well as for psychiatric conditions such as depression [52], and other mental health disorders [49], including obsessive compulsive disorder [60] and abstinence from alcohol and drugs (RCT underway). A TLC system for the assessment of psychiatric disorders in the workplace was recently designed and evaluated; in comparison to a control group, the intervention group showed a significantly greater reduction in depression at 3 months, a significantly greater improvement in mental health at 6 months, and a significantly greater improvement on two Work Limitations Questionnaires subscales, the Mental-Interpersonal Scale at 3 months and the Time-Scheduling Scale at 6 months (paper under preparation).…”

Section: Treating Impulsivitymentioning

confidence: 99%

See 2 more Smart Citations

Design and Development of a Telephone-Linked Care (TLC) System to Reduce Impulsivity among Violent Forensic Outpatients and Probationers

et al. 2010

Self Cite

View full text Add to dashboard Cite

Forensic services face the challenge of reducing relapse among clients with a history of violent crime. An automated interactive voice response (IVR) service of the complex Telephone-Linked Care (TLC) type, with a focus on reducing impulsivity, could improve the adequacy of service responses to client needs. Theoretically based in Dialectical Behavior Therapy (DBT), Cognitive Behavior Therapy (CBT) and Motivational Interviewing (MI), the forensic TLC system offers interactive conversations on coping with the emotions of anger, shame and loneliness; activities of daily life such as getting out of bed, asking for help, visiting social services and taking medication; and other areas such as hearing voices, drinking alcohol and self-critical thoughts. We describe the user's flow through the system, with an in-depth synopsis of the hearing voices intervention. Issues regarding voluntary versus mandatory use of the system are addressed in connection with prospective introduction of the system in forensic settings.

show abstract

Experiences of computer‐based and conventional self‐help interventions for eating disorders: A systematic review and meta‐synthesis of qualitative research

Yim

Schmidt

2019

Intl J Eating Disorders

View full text Add to dashboard Cite

Objective Self‐help interventions have been demonstrated to be effective in treating bulimic‐type eating disorders (EDs). In particular, computer‐based interventions have received increasing attention due to their potential to reach a wider population. This systematic review aimed to synthesize findings from qualitative studies on users' experiences of self‐help interventions for EDs and to develop an exploratory framework. Method A systematic review and meta‐synthesis on seven peer‐reviewed qualitative studies on structured computer and book‐based self‐help interventions for EDs was conducted using Noblit and Hare's (1988) 7‐phase meta‐ethnography. Four of the selected studies investigated computer‐based self‐help programs, and three of the studies investigated book‐based guided self‐help programs. Results Six concepts were synthesized. They included intervention‐related factors (anonymity and privacy; accessibility and flexibility; guidance) and user‐related factors (agency/autonomy; self‐motivation; and expectations/attitudes). The study revealed the “machine‐like” and relational properties of the computer; the expansion of treatment time and space in psychological interventions, the changing role of the medical health professional from a “therapist” to a “guide,” and a change from understanding interventions as a conclusive treatment plan to a starting point or stepping stone toward recovery. Discussion Computer‐based self‐help interventions should take advantage of the “machine‐like” properties of a computer (neutrality, availability, etc.) as well as its ability to facilitate human interactions. Users should also be facilitated to have a realistic understanding of the purpose of self‐help interventions and the place of self‐help interventions in their broader treatment plans to moderate expectations and attitudes.

show abstract

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

Cited by 26 publications

References 18 publications

Active Assistance Technology for Health-Related Behavior Change: An Interdisciplinary Review

Active Assistance Technology for Health-Related Behavior Change: An Interdisciplinary Review

Design and Development of a Telephone-Linked Care (TLC) System to Reduce Impulsivity among Violent Forensic Outpatients and Probationers

Experiences of computer‐based and conventional self‐help interventions for eating disorders: A systematic review and meta‐synthesis of qualitative research

Contact Info

Product

Resources

About