Collaborative Visual Analytics: A Health Analytics Approach to Injury Prevention.
Samar Al-HajjBrian FisherJennifer SmithIan PikePublished in: International journal of environmental research and public health (2017)
Background: Accurate understanding of complex health data is critical in order to deal with wicked health problems and make timely decisions. Wicked problems refer to ill-structured and dynamic problems that combine multidimensional elements, which often preclude the conventional problem solving approach. This pilot study introduces visual analytics (VA) methods to multi-stakeholder decision-making sessions about child injury prevention; Methods: Inspired by the Delphi method, we introduced a novel methodology-group analytics (GA). GA was pilot-tested to evaluate the impact of collaborative visual analytics on facilitating problem solving and supporting decision-making. We conducted two GA sessions. Collected data included stakeholders' observations, audio and video recordings, questionnaires, and follow up interviews. The GA sessions were analyzed using the Joint Activity Theory protocol analysis methods; Results: The GA methodology triggered the emergence of 'common ground' among stakeholders. This common ground evolved throughout the sessions to enhance stakeholders' verbal and non-verbal communication, as well as coordination of joint activities and ultimately collaboration on problem solving and decision-making; Conclusions: Understanding complex health data is necessary for informed decisions. Equally important, in this case, is the use of the group analytics methodology to achieve 'common ground' among diverse stakeholders about health data and their implications.
Keyphrases
- big data
- mental health
- pet ct
- decision making
- public health
- healthcare
- artificial intelligence
- machine learning
- health information
- electronic health record
- health promotion
- randomized controlled trial
- working memory
- quality improvement
- human health
- deep learning
- high resolution
- social media
- study protocol
- mass spectrometry
- data analysis