Constructing geographic areas for cancer data analysis: A case study on late-stage breast cancer risk in Illinois

► The small numbers (population) problem occurs in analysis of rare disease (including cancer). ► A GIS-based automated method is proposed to construct areas with above a threshold population. ► Cancer rates in the new areas have sufficiently large base population, and are thus more reliable. ► This permits direct mapping, exploratory spatial data analysis, and even simple OLS regression.