Analyzing Cancer Outcome Disparities Geographically
Michigan Cancer Surveillance Program
Cancer outcomes such as early detection and length of survival vary geographically and by race and ethnicity. Tools to assess these variations in a statistically valid way are needed to identify true geographic patterns and trends.
Data for breast and prostate cancer stage at diagnosis and survival were analyzed using geospatial analytical tools to identify areas of statistically poorer outcomes at varying scales including counties, legislative districts, neighborhoods, and point-level data. Analyzing outcome disparities between black and white patients demonstrated that the smaller the scale, the smaller the disparity, such that disparities readily apparent at the county or regional level were not present at the tract level.
The work provides new tools for cancer control analysts to identify and track areas of greater need and to monitor progress while focusing on statistically important areas of poorer (or better) outcomes.
The disparities analysis demonstrated that the more finely you focus the geography, the smaller the disparities in outcomes between blacks and whites. Disparities are as much a function of place as race. The factors associated with racial disparities are modifiable and may be amenable to intervention strategies.