Aging societies, urban migration, and changing lifestyles are contributing to unprecedented rises in cancer burden worldwide, making global cancer control a leading public health challenge of our time. Effective global cancer control policy should incorporate knowledge about local contextual factors – features of the places where people live, work and play – that determine who develops cancer and who gets treatment. Decisions regarding where to place oncology services must balance considerations of equity and efficiency. Geospatial data science, a novel methodological field that draws from geography, statistics, and computer science, provides analytic tools that can help us consider the role of these important geographic factors in our work towards reducing the global burden of cancer.

This eight-week online course will introduce students to the essential concepts and analytic tools they need to integrate geospatial data science into their work in global cancer control. Students will learn about different geospatial data types, where to acquire geographic data, and how to analyze these data using open source software and coding platforms. In addition, students will learn from experts at the forefront of this new field, who are applying geospatial data science approaches in their research to understand how geographic features and accessibility influence cancer risk and survival. Finally, students will independently prepare a geospatial data science project, helping them gain confidence in applying these new methods in their own work.

Intended audience:

This course is designed for medical and public health professionals who wish to incorporate geospatial data science into their work. We assume familiarity with basic training in epidemiology and biostatistics. Prior programming experience will be helpful but is not required.

Though the course has a public health focus, we expect students from a wide range of backgrounds will be interested in this topic. Current master’s students or recent graduates in public health, economics, or computer/data science may be interested in developing skills in this area to advance their careers. Mid-career policy professionals may also wish to learn about these novel geospatial data sources and analytic approaches. Non-profit and government workers may seek to incorporate geographic data science approaches to better target cancer services to populations in need.

Learning Goals:

• Learn vocabulary and concepts related to geographic data analysis
• Learn how to search for and acquire geographic data
• Apply geospatial analytic software to perform the following tasks:
  • Make maps using geographic data
  • Link spatial factors to tabular datasets
  • Estimate geographic accessibility
  • Test for spatial autocorrelation
• Learn about applications of geospatial data science for cancer control and prevention
  • Environmental exposures (e.g. air pollution, chemicals)
  • Neighborhood contextual exposures (e.g. deprivation, segregation, green space, light at night)
  • Geographic accessibility to cancer services (e.g. estimating geographic accessibility in low-resource settings)

Required Materials:

All students are required to use their own computers. Students will also be required to download R and R Studio to complete practical exercises. Students must also create accounts to access data from Google Earth Engine. For the first time, we have made certain lectures available to the public through a YouTube channel. These videos provide examples of the types of content that we include in our course.

Register for the course here or email cgcp@hshp.harvard.edu for more information.