Programme overview

The course will run for 4 days (Tuesday-Friday) during the following times. Click here for a detailed time schedule.

Tues 19/1 9:00-18:00 (registration from 8:00, lunch from 12-14)
Wed 20/1 9:00-18:00 (lunch from 12-14)
Thur 21/1 9:00-18:30 (lunch from 12-14)
Frid 22/1 9:00-17:00 (lunch from 12-13)

Following is an overview of the preliminary programme. The name of the primary teacher for each day is listed, but multiple faculty members will contribute to each day. The course will consist of lectures, group discussions, and demonstrations of how methods can be applied using statistical software but not hands-on computing sessions.

Day 1: Introduction and concepts (Max Parkin)

  • Goals of descriptive epidemiology
  • Sources of data - strengths and weaknesses
  • Outcome measures used in cancer control - incidence, mortality, prevalence, survival
  • Linkage with other data sources
  • Analysis of classical descriptive data
  • Statistical measures - probabilities, rate, ratio, proportional methods
  • Conceptual framework for descriptive studies
  • Methods for standardisation
  • Concept of time scale
  • Online exercise with WHO mortality / cancer incidence database

Day 2: Time trends (Freddie Bray)

  • The rationale for time trend analyses of cancer, with examples from the literature
  • The graphical description of trends in cancer rates
  • The role of statistical models in trend analyses
  • The theory and applications of age-period-cohort models
  • Predicting the future cancer burden
  • Demonstration of time trend facilities within NORDCAN

Day 3: Spatial studies (Eero Pukkala)

  • Methods of spatial descriptive cancer epidemiology
  • Traditional and more attractive ways to show cancer frequencies as maps
  • What could or should not be studied with spatial-temporal illustrations; incidence, prevalence, mortality, predictions, survival, familial aggregation, screenings, registration practices, prevalence of risk factors and other background data…
  • Examples of studies linking spatial exposures to cancer outcome

Day 4: Survival studies (Paul Dickman)

  • What is 'population-based cancer survival analysis' and what makes it special compared to other applications of survival analysis?
  • Net survival; cause-specific survival; relative survival; relative merits of cause-specific survival and relative survival for population-based cancer registry data;
  • Interpreting relative survival estimates; statistical cure;
  • Cohort, complete, period and hybrid approaches to estimation;
  • Modelling excess mortality (relative survival) using Poisson regression;
  • Non-proportional hazards and how to adjust for them;
  • Cure models for relative survival - estimating and modelling the cure proportion;
  • Impact of data quality, completeness, stage migration, screening and lead-time bias.

Special topics (Max Parkin) included in days 3 and 4

Approximately 75% of day 3 will be spatial studies (presented by Eero Pukkala) and 75% of day 4 will be survival studies (presented by Paul Dickman). Max Parkin will present aditional topics on these 2 days, including the following:

  • Migrant studies
  • Ecological studies methods / fallacy / confounding
  • Ecological studies: gapminder exercise and discussion of pros, cons