The development and validation of a dashboard prototype for real-time suicide mortality data.

Area of Science:

• Public Health
• Data Science
• Epidemiology

Background:

• Data visualization is crucial for data-driven decision-making but is under-explored in suicide surveillance.
• Real-time monitoring systems are essential for timely intervention in public health crises.

Purpose of the Study:

• To develop an interactive dashboard prototype to enhance a real-time suicide surveillance system.
• To facilitate emerging cluster detection, risk profiling, and trend observation.
• To establish data sharing connections with stakeholders via an intuitive interface.

Main Methods:

• Analysis of individual-level demographic and circumstantial data for suicide cases (2008-2017).
• Application of retrospective and prospective space-time scan statistics using a discrete Poisson model.
• Utilized R software with "rsatscan" and "shiny" packages for cluster analysis and dashboard visualization.

Main Results:

• Retrospective analysis identified emerging non-significant clusters over a 10-year period.
• Prospective analysis demonstrated the model's predictive capability for emerging trends.
• Outputs visualized via geographical maps and timelines of cluster occurrences.

Conclusions:

• Integration of geo-visualization, space-time scan statistics, and predictive modeling enables early detection of suicide clusters and at-risk populations.
• The dashboard prototype offers real-world applicability as a proactive monitoring tool for suicide prevention.
• Facilitates informed planning and preparedness for timely responses to emerging suicide trends.