Hotspot Analysis Example - Georgia, USA

Follow the steps below to conduct a hotspot analysis in the USDSS Analysis Module. Here we use an example of identifying potential hotspots of newly diagnosed diabetes for the state of Georgia in the year 2019.

1) Select the Hotspots Analysis option

  1. Click on the Hotspots tab.

  2. Check out the default settings: the default Geography Level is “National”, the default Surveillance Indicator is “Diagnosed Diabetes”, and the default year represents the most recently available dataset for that indicator.

2) Select the Geography Level

Arrows indicate what options to change to move from the national hotspot analysis page to a state page for a different indicator.

  1. Change the Geography Level from “National” to “State” and choose “Georgia” from the drop-down list.

  2. In the Surveillance Indicator menu, change the drop-down menu from “Diagnosed Diabetes” to “Newly Diagnosed Diabetes”. Check that the bottom drop-down menu is set to 2019.

  3. Hotspot analysis for Newly Diagnosed Diabetes in the state of Georgia in 2019:

A hotspot analysis of newly diagnosed diabetes in Georgia in 2019 showing a coldspot in northeast Georgia and hotspots in south central Georgia.

Note

Possible hotspots (or coldspots) are based on \(z\)-scores (see Calculations page) and matched to a continuous color gradient ranging from blue (coldspots) to orange (hotspots). The darker the color, the more likely it is that the county is a hotspot or coldspot because of higher absolute values of the \(z\)-score.

3) Use the slider bar to modify displayed results

The color scale for the hotspot analysis showing the threshold sliders, the toggle to show data between the sliders or outside, and the reset button.

  1. Adjust the sliders to display \(z\)-scores with values above, below, or between certain thresholds.
Note

The default thresholds for the color gradient are set at -2.00 and 2.00, and only counties with \(z\)-score values outside these numbers are shown (i.e. values greater than or equal to 2.00 and less than or equal to -2.00).

  1. To display only counties that could be possible coldspots, toggle the switch from “SHOW OUTSIDE RANGE” to “SHOW IN-BETWEEN RANGE”.

    Diagram showing what would be inside the range of the sliders and what would be outside.

  2. View the map showing only those counties with a \(z\)-score less than zero.

Moving the left slider all the way to the left and the right slider near 0, then toggling to show "In-Between Range" will display only coldspots.

  1. Changing the toggle back to “SHOW OUTSIDE RANGE” will display only those counties with a \(z\)-score greater than zero.

Using the same slider settings from the previous image but toggling to "Show outside range" displays only the hotspots.

  1. Identify the 4 counties with the highest \(z\)-scores:

    i. Click the Reset button next to the slider

    ii. Toggle to “SHOW IN-BETWEEN RANGE”

    iii. Move the right slider all the way to the right and the left slider to the right until it reaches a value near 3.14

The sliders can be positioned to display counties with z-scores within a particular range, such as the highest four z-score values.

4) Select other features

  1. Click the Reset button to return to the default display for “Newly Diagnosed Diabetes” and “Georgia”.
Note

We can see that a possible coldspot is found in northeast Georgia, where the newly diagnosed diabetes values are disproportionately lower than what is found across the rest of the state. Additionally, there appear to be pockets in south-central Georgia that may have disproportionately high values of newly diagnosed diabetes compared to the rest of the state.

A hotspot analysis of newly diagnosed diabetes in Georgia in 2019 showing a coldspot in northeast Georgia and hotspots in south central Georgia.

  1. Expand the left slide all the way to the left and the right slider all the way to the right to view the pattern across the whole state without any filtering:

Moving both sliders to the far left and right side will show the pattern across the whole state which highlights how patterns change across the state.

Note

In addition to highlighting the areas that are most likely to be hotspots or coldspots, this map also shows how the patterns vary across the entire state and which counties are proportionally similar (e.g. neither hot nor cold; pale colors) to the counties across the rest of the state.