Figure 3.

Overlaying "new" diabetes patient addresses on the background neighbourhood rate grid. In order to make further comparison between the new clinic's patients and the neighbourhood disease burden, a test of statistical significance was calculated for each node. Each of the 29 patients home address was assigned to the closest node, these being displayed in Figure 3 which shows both the grid and the residential locations. All other geographic information is removed to eliminate the possibility of reverse engineering patient addresses [37].