DMAP is a Windows compatible program that produces disease rates using variable spatial filters and tests for their statistical significance using Monte Carlo simulations. The program computes values that are inputs to GIS software that will produce disease rate maps and maps of statistical significance. Input data are either individual disease records and individual at risk records or are aggregates of the above. DMAP produces disease rate maps from two types of data
1) Address matched records of individuals with the disease (known as 'numerator' events) and individuals at risk for having the disease (known as 'denominator' events). In this case weights equal one.
2) Address matched data for spatial aggregates of people with the disease ('numerator' events) and people at risk of having diseases ('denominator' events). An example of spatial aggregation might be census blocks. When spatial aggregates are used, the weights are greater than one.
DMAP Process
- Step 1: Make a grid that covers your study area.
- Step 2: Calculate the distance from each grid point to all health events within the filter area.*
*This calculation has to be done twice: once for the disease events which is the numerator, and once for the population at risk which is the denominator used for calculating rates. 
- Step 3: Calculate observed disease rates for each
grid point using the numerator and denominator
files produced in step 2.
This is an example of a typical file containing disease rates. The rate file produced by DMAP is called RATE.DAT. The first column contains the grid ID, the second column contains the disease rate, the third column contains the total number of disease events within the specifies filter size at that grid point and the fourth column contains the total number of events at risk within the specifies filtersize at that grid point. For example, the first line should be interpreted as saying that at grid 118, the disease rate is 23.258, the number of disease events within the specifies filter size is 1 and the number of events at risk is 43.
- Step 4: Use GIS software to map the rates.
(See related information in section 2 of this module.
- Step 5: Test the observed rates for statistical significance
using Monte Carlo SImulations.
This is an example of a typical file containing statistical significance of observed disease rates using Monte Carlo simulations. The significance file produced by DMAP called SIGNIF.DAT. The first column is the grid ID and the second is a measure of the statistical significance of the observed disease rates at that grid point. Statistical significance of the observed disease rate at any grid point is computed as the proportion of the simulated rate at that grid point. For example, the first line identifies grid ID 769 and the statistical significance of the observed rate at that grid point is 0.1770. This means that about 18% of the time, the simulated rates are less than the observed rate at that grid point. - Step 6: Make a map of statistical significance of observed rates.
- The DMAP Program and a test data set are available for download.