Methodological issues in the development of the Canadian Cancer Incidence Atlas

Area of Science:

• Epidemiology
• Biostatistics
• Geographic Information Systems (GIS) in public health

Background:

• National cancer atlases are increasingly utilizing incidence data over mortality data.
• The Canadian Cancer Incidence Atlas represents a significant advancement in cancer surveillance.
• Understanding spatial patterns of cancer incidence is crucial for public health interventions.

Purpose of the Study:

• To describe the statistical methodologies employed in the Canadian Cancer Incidence Atlas.
• To assess the significance and spatial correlation of age-standardized rates (ASIRs) across Canadian census divisions.
• To ensure the quality and reliability of the cancer incidence data presented.

Main Methods:

• Mapping cancer sites based on the expected number of cases per area.
• Assessing ASIR significance using Poisson distribution simulations.
• Examining observed case distributions against the Poisson distribution.
• Utilizing bootstrap confidence intervals (CI) for ASIRs.
• Evaluating spatial correlation with Moran's I/I(max) and simulation, accounting for area population variation.
• Presenting data quality indicators and registry questionnaire data.

Main Results:

• Established methods for mapping cancer incidence across 290 Canadian census divisions.
• Validated the statistical significance and spatial clustering of cancer incidence rates.
• Incorporated robust data quality measures for enhanced reliability.
• The atlas provides a comprehensive spatial overview of cancer incidence in Canada.

Conclusions:

• The methodologies employed ensure the validity and reliability of the Canadian Cancer Incidence Atlas.
• The atlas serves as a valuable resource for cancer research and public health planning in Canada.
• Spatial analysis of cancer incidence provides critical insights into geographic disparities and potential environmental or social determinants.