The population impact number is the projected reduction in the number of cases for a specific health outcome (mental health ED visits, cardiovascular hospitalizations, etc.) if each income group experienced the rate of the highest income group.). As the rate for a specific health outcome goes up in the general population, the population impact number will tend to increase.

The population attributable fraction is the projected reduction in the rate of the health outcome of interest (alcohol attributable hospitalizations, assault-related ED visits, etc.) if each income group experienced the  rate of the highest income group (top 20%). 

The relative index of inequality is the predicted rate of the health outcome of interest (alcohol attributable hospitalizations, assault-related ED visits, etc.) for the least advantaged individual divided by the predicted rate of the specific health outcome for the most advantaged individual. Rates over 1 indicate greater risk of negative outcome for the least advantaged individual.

The relative rate divides the rate of the most marginalized group (lowest income quintile) by the rate for the least marginalized group (highest income quintile). Relative rates greater than 1, therefore, indicate greater risk of negative outcome for the least advantaged individual.