Some HCI platforms display a significant difference between the overall value for a location and subgroups for the same location. This is designed to better help you find where disparities exist in your community.
When a given indicator has data available for subgroups like race/ethnicity, age or gender and we know not only the indicator value but the confidence intervals, we are able determine if there is a significant difference between the subgroups value and the overall value. A significant difference is defined as non-overlapping confidence intervals. When viewing a chart, you can turn on confidence intervals, to view the confidence interval bars. In this example, you'll see that the red bars do not overlap with the grey overall value and are higher than the value which means this is bad.
Confidence intervals are not available for every indicator. In some cases, a significant disparity may exist but we do not have the data to show the disparity.