The effects of a computerized clinical decision aid on clinical decision-making in psychosis care

OBJECTIVE

Clinicians in mental healthcare have few objective tools to identify and analyze their patient's care needs. Clinical decision aids are tools that support this process. This study examines whether 1) clinicians working with a clinical decision aid (TREAT) discuss more of their patient's care needs compared to usual treatment, and 2) agree on more evidence-based treatment decisions.

METHODS

Clinicians participated in consultations (n = 166) with patients diagnosed with psychotic disorders from four Dutch mental healthcare institutions (research registration number 201700763). Primary outcomes were measured with the modified Clinical Decision-making in Routine Care questionnaire and combined with psychiatric, physical and social wellbeing related care needs. A multilevel analysis compared discussed care needs and evidence-based treatment decisions between treatment as usual (TAU) before, TAU after and the TREAT condition.

RESULTS

First, a significant increase in discussed care needs for TREAT compared to both TAU conditions (β = 20.2, SE = 5.2, p = 0.00 and β = 15.8, SE = 5.4, p = 0.01) was found. Next, a significant increase in evidence-based treatments decisions for care needs was observed for TREAT compared to both TAU conditions (β = 16.7, SE = 4.8, p = 0.00 and β = 16.0, SE = 5.1, p = 0.01).

CONCLUSION

TREAT improved the discussion about physical health issues and social wellbeing related topics. It also increased evidence-based treatment decisions for care needs which are sometimes overlooked and difficult to treat. Our findings suggest that TREAT makes sense of routine outcome monitoring data and improves guideline-informed care.

Qualitative analysis of clinicians' perspectives on the use of a computerized decision aid in the treatment of psychotic disorders

Background

Clinical decision aids are used in various medical fields to support patients and clinicians when making healthcare decisions. Few attempts have been made to implement such tools in psychiatry. We developed Treatment E-Assist (TREAT); a routine outcome monitoring based computerized clinical decision aid, which generates personalized treatment recommendations in the care of people with psychotic disorders. The aim of this study is to investigate how TREAT is used and evaluated by clinicians and how this tool can be improved.

Methods

Clinicians working with TREAT during a clinical trial were asked to participate in semi-structured interviews. The Unified Theory of Acceptance and Use of Technology (UTAUT) was used as a sensitizing theory to structure a part of the interview questions. The transcripts were analyzed using inductive thematic analysis to uncover the main themes.

Results

Thirteen clinicians (mean age: 49) of which eight psychiatrists and five nurse practitioners, participated in this study. Eight clinicians experienced TREAT as beneficial, whereas five experienced no additional benefits. Thematic analysis revealed five themes surrounding usage and evaluation of TREAT, views on TREAT’s graphic representation of routine outcome monitoring results, guideline based treatment recommendations, contextual factors, effects on patients and effects on shared decision-making. Performance and effort expectancy were perceived as high by clinicians. The facilitating conditions were optimal and perceived social influence was low.

Conclusion

This article presents a qualitative evaluation by clinicians of a computerized clinical decision aid in psychosis care. TREAT was viewed by most clinicians as beneficial during their consultations. The graphic representation of routine outcome monitoring results was well-appreciated and provided input to discuss treatment planning with patients. The treatment recommendations did not change most treatment decisions but supported clinical reasoning. However, some clinicians were unconvinced about TREAT’s benefits. The delivery, applicability and the availability of resources require improvement to increase TREAT’s efficacy. Not all patients responded well to TREAT but the observed facilitation of shared decision-making is promising. All four predictors of the Unified Theory of Acceptance and Use of Technology were positively evaluated by the majority of clinicians.

[Decision aid for the treatment of psychotic illness: development of TReatment E-AssisT (TREAT)]

Use of decision aids in mental health care is increasing and will also be introduced in the Dutch mental healthcare system. We describe the context of this development and discuss how decision aids could facilitate evidence-based psychiatry.<br/> AIM: To describe the development of the decision aid TReatment E-Assist (TREAT) in the Dutch mental healthcare system that aims to optimize treatment of people with a psychotic illness.<br/> METHOD: We describe how the TREAT application works and discuss its potential contribution to the treatment of people with a psychotic illness.<br/> RESULTS: In a pilot study TREAT was judged as user friendly and useful. TREAT seemed to increase the integration of ROM-results in treatment and the advice offered new view points for practioners.<br/> CONCLUSION: TREAT is a novel application which combines routine outcome monitoring results with current treatment guidelines and standards of care in order to generate personalised treatment recommendations in the context of a psychiatric treatment trajectory. A multicentre study is being conducted in different provinces in the Netherlands to investigate the effectiveness of TREAT.

The development and evaluation of a computerized decision aid for the treatment of psychotic disorders

BACKGROUND

Routinely monitoring of symptoms and medical needs can improve the diagnostics and treatment of medical problems, including psychiatric. However, several studies show that few clinicians use Routine Outcome Monitoring (ROM) in their daily work. We describe the development and first evaluation of a ROM based computerized clinical decision aid, Treatment-E-Assist (TREAT) for the treatment of psychotic disorders. The goal is to generate personalized treatment recommendations, based on international guidelines combined with outcomes of mental and physical health acquired through ROM. We present a pilot study aimed to assess the feasibility of this computerized clinical decision aid in daily clinical practice by evaluating clinicians' experiences with the system.

METHODS

Clinical decision algorithms were developed based on international schizophrenia treatment guidelines and the input of multidisciplinary expert panels from multiple psychiatric institutes. Yearly obtained diagnostic (ROM) information of patients was presented to treating clinicians combined with treatment suggestions generated by the algorithms of TREAT. In this pilot study 6 clinicians and 16 patients of Lentis Psychiatric Institute used the application. Clinicians were interviewed and asked to fill out self-report questionnaires evaluating their opinions about ROM and the effectiveness of TREAT.

RESULTS

Six clinicians and 16 patients with psychotic disorders participated in the pilot study. The clinicians were psychiatrists, physicians and nurse-practitioners which all worked at least 8 years in mental health care of which at least 3 years treating patients with psychotic illnesses. All Clinicians found TREAT easy to use and would like to continue using the application. They reported that TREAT offered support in using diagnostic ROM information when drafting the treatment plans, by creating more awareness of current treatment options.

CONCLUSION

This article presents a pilot study on the implementation of a computerized clinical decision aid linking routine outcome monitoring to clinical guidelines in order to generate personalized treatment advice. TREAT was found to be feasible for daily clinical practice and effective based on this first evaluation by clinicians. However, adjustments have to be made to the system and algorithms of the application. The ultimate goal is to provide appropriate evidence based care for patients with severe mental illnesses.