Injury Prevention

Overview

RENCI helped Dr. Andres Villaveces (UNC department of Epidemiology and the Injury Prevention Research Center) to show the impacts of injury intervention techniques on the prevention and control of injuries and violence, especially traffic, alcohol, and firearm-related injuries and violence. The first visualization shows the ratio of injury events over a time line at different times of the day for the general population. The user drags the mouse up and down the time bar to look at different time periods. Small color bars shown at the bottom of the window indicate if a specific day is a holiday or payday, or if a specific intervention technique was applied.

The second visualization shows the statistical impact of different laws on alcohol-related fatalities in a 3-dimensional parallel coordinate plot, which allows simultaneous one-to-one relational analysis. This 3-dimensional parallel coordinate plot maps alcohol-related fatalities with eight different laws: speed limits (spdlaw), seatbelt law (sbtlaw), blood alcohol concentration law (baclaw), sobriety checkpoint law (soberlaw), mandatory jail for first degree offender law (mjail1), repeat offender law (roffend), administrative license revocation law (alrlaw), and helmet law (helmet). Each cylinder axis is scaled individually to show the full range of its attribute value. A value of 1 indicates passage of the mapped corresponding law and a value of 0 at the end of the axes indicates failure to pass the law. These 3D parallel coordinate plots show that simultaneously passing of each of these eight laws all reduces alcohol-related fatalities statistically.

The third visualization shows a table view of the filtered data records with selected attributes on the left and a parallel coordinate plot on the right. The parallel coordinate plot maps the administrative license revocation law (alrlaw) to the left vertical axis, blood alcohol concentration law (baclaw) to the right vertical axis, and the number of alcohol-related fatalities (acount) to the middle vertical axis, so each data record is mapped to a polyline that connects the points on the vertical axes. Each axis is scaled individually to show the full range of its attribute value. A value of 1 (top left and right axes for HealthViz5.png and bottom of the axes for HealthViz6.png) indicates passage of the mapped corresponding law and a value of 0 on the other end of the axes indicates failure of passage. In addition, for image HealthViz6.png, the black lines represent states with each corresponding law passed at a specific time, and the green lines represent those states with each corresponding law not passed at a specific time. This parallel coordinate plot shows that laws that revoke driver’s licenses and blood alcohol concentration laws reduce alcohol-related fatalities. The two data records selected and highlighted in the table view are also selected and highlighted in red in the parallel coordinate plot, showing alcohol-related fatalities for North Carolina in two consecutive months after passage of the license law. HealthViz7 shows the same data but with a black background and parallel coordinate lines shown in white.

Project Team

Hong Yi

Partner

Dr. Andres Villaveces, UNC department of epidemiology