JMIR Publications

ABSTRACT

Background:

A large quantity of data is collected during the delivery of cancer care. However, once collected this data is difficult for clinical teams to access to support their clinical decision making or reviewing performance. There is a need for interfaces that make clinical data more accessible to support clinicians in these activities. One approach for enabling health professionals with easy access to clinical data is to create the infrastructure and interface for a clinical dashboard to make data accessible in a timely and relevant manner.

Objective:

The objective of this research study was to develop and evaluate two prototype dashboards for displaying data on the identification and management of lymphoedema.

Methods:

The study used a co-design framework to develop two prototype dashboards, one for displaying cohort data and another for displaying single patient data. This began with group meetings with diverse stakeholders (n=34) to establish core needs and goals for such dashboards. A series of fortnightly meetings over six months with an Advisory Committee occurred. Based on these, an interactive mockup dashboard was developed. This mockup was then presented to end users to get preliminary feedback about the design and use of such dashboards. Feedback from the presentations was reviewed and used to inform the development of the final prototypes. A structured evaluation was conducted, using Think Aloud Protocol (TAP) and semi-structured interviews with diverse stakeholder groups among the potential end-users.

Results:

A set of data visualization requirements was created from the 34 member multi-disciplinary team was reviewed and grouped to create an individual and a cohort dashboard. A qualitative evaluation is reported for five health professionals. These participants were selected from three specialties; surgery (n=1), radiation oncology (n=2) and occupational therapy (n=2). Participants were able to complete the majority of tasks on the dashboard. Semi-structured interviews themes fell in to three categories: engagement/enthusiasm for the dashboard, user experience and data quality and completeness.

Conclusions:

Findings from this study constitute the first report of a co-creation process for creating a dashboard for cancer-care clinicians. Health professionals are interested in the use of data visualization tools to make routinely collected clinical data more accessible. To be used effectively dashboards need to be reliable and sourced from accurate and comprehensive enough data sets to meet the needs of clinicians. Whilst the multi-disciplinary process used to develop the visualization tool proved effective for designing a single patient dashboard, the more complex data set required for a cohort view remained a challenge.