JMIR Publications

ABSTRACT

Background:

Therapeutic decision-making in oncology is a complex process. One key reason is due to complexity of the decision process for a physician who needs to consider many relevant forms of medical data about the patient as well as treatment guidelines or protocols. A second key challenge for physicians is to share the decision process with the patient to ensure that they can provide truly informed consent. Computer-based decision tools have the potential to play a valuable role to assist in this complex process, which calls for identification of treatment options, analysis of their efficacy and discussion of these options with patients in the context of the supporting evidence.

Objective:

To investigate the extent to which computer-based decision tools have been successfully used in oncology consultations to improve patient-physician joint therapeutic decision making by conducting a systematic literature review. Therefore, in the context of shared decision making, the following is the review question, how effective are oncology computer-based decision tools in terms of (a) system, information and service quality, (b) system use and user satisfaction, and (c) care quality, access and productivity?

Methods:

This review was carried out in accordance to Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. A literature search was conducted across Cochrane Database of Systematic Reviews (2005 to January 28, 2021), Cochrane Central Register of Controlled Trials December 2020, MEDLINE (1946 to February 04, 2021), EMBASE (1947 to 2021 February 04), Web of Science 1900 to 2021, Scopus 1969 to 2021, and PubMed 1991 to 2021 databases. The reference lists of the included studies, relevant journals, and grey literature sources were snowball and hand searched. Articles were screened by reviewers for quality and inclusion before data extraction.

Results:

There are relatively few studies looking at the use of computer-based decision tools in oncology consultations. From 4431 unique articles obtained from searches, only ten studies satisfied the selection criteria. Out of the ten selected studies, eight computer-based decision tools were identified. The results demonstrated that there were numerous barriers that stopped the adoption of computer-based decision tools in routine oncology practice. Some patients or physicians who used decision tools were satisfied and would recommend their use to others. However, there were also patients who found the language of computer-based decision tools too complex or felt confused by the information presented. Moreover, some patients failed to understand key treatment information. Physicians, on the other hand, did not change their routine practice to take advantage of the decision tools or the information provided, although on average, usage of these decision tools added approximately only five minutes to the total length of their consultations. Some physicians felt that the tools increased patients’ anxiety. No information was found on how computer-based decision tools affected system or service quality such as staff reliability, empathy and responsiveness.

Conclusions:

The results of the study provide a first understanding of the issues that have been confronting developers and implementers of computer-based decision tools in oncology shared therapeutic decision making. From the outcomes of ten studies, five were positive, two were negative, two provided mixed outcomes, and one reported no change. Even though a low level of adoption persists, the results of the study showed that computer-based decision tools improved communication between patients and physicians in some situations. However, challenges remain in the areas of usability and integration with existing electronic health record systems. Moreover, the review did not provide insights about system and service quality. Consequently, further and deeper research is needed to confirm the methods through which computer-based decision tools generally improve patient-physician interactions. Additionally, it is important to know about the system features that promote or obstruct the adoption of computer-based decision tools in oncology shared decision making, and the patient centred care model.