SFOK15012U Survival and Event History Analysis

Volume 2015/2016
Education

Full Degree Master

Content

The course introduces statistical concepts and methods for analyzing time-to-event (survival) data obtained from following individuals until a particular event occurs, or they are lost to follow-up. We will illustrate the use of modern tools for time-to-event analysis and discuss interpretation and communication of results. The course provides practical experience with real health science data.

The content includes: censoring and truncation; Kaplan-Meier estimation; log-rank tests; Cox regression; Aalen's additive regression model; competing risks; clustered data.

Learning Outcome

Knowledge

  • Distinguish methods for analysis of time-to-event data from other types of measurements.
  • Understand the concepts of censoring and truncation.
  • Explain central survival analysis concepts such as hazard, survival and cumulative incidence and their relationships.
  • Describe and compare models for time-to-event data. Illustrate how the models can be applied to epidemiological or public health data.
  • Understand and recognize common pitfalls specific to time-to-event data.

 

Skills

  • Determine the proper statistical method to address a specific scientific question from a given time-to-event data set. This includes understanding the underlying assumptions of the method and identifying violations of these.
  • Perform time-to-event analysis with modern techniques using statistical software. Assess the fit of the model.
  • Interpret the results reported by statistical software. Communicate the results and conclusions of a time-to-event analysis in a clear and precise way.
  • Write a statistical report with a well-defined problem statement, method and result presentations, discussion and conclusion.
  • Critically read and review reports or articles addressing epidemiological and public health questions by time-to-event analysis.

 

Competencies

  • Identify and conceptualize questions encountered in the professional life of a public health researcher that can be addressed by time-to-event studies.
  • Independently design, carry out and communicate time-to-event studies.
  • Give advice and take active part in collaborations where decisions are based on the statistical analysis of time-to-event data.
The statistics course from the MSc in Public Health Science (or a similar course).
Class instruction and home assignments.
  • Category
  • Hours
  • Class Instruction
  • 10
  • Exam
  • 25
  • Lectures
  • 10
  • Preparation
  • 93
  • Total
  • 138
Credit
5 ECTS
Type of assessment
Continuous assessment
Written assignment
Project report at the end of the course.
Exam registration requirements

Participants are required complete a minimum of 80% of the home assignments.

Aid
All aids allowed
Marking scale
passed/not passed
Censorship form
No external censorship
Exam period

See the exam plan http://sund.ku.dk/uddannelse/vejledning-information/eksamensplaner/folkesundhedsvidenskab/

Re-exam

See the exam plan http://sund.ku.dk/uddannelse/vejledning-information/eksamensplaner/folkesundhedsvidenskab/

Criteria for exam assesment

Knowledge

  • Distinguish methods for analysis of time-to-event data from other types of measurements.
  • Understand the concepts of censoring and truncation.
  • Explain central survival analysis concepts such as hazard, survival and cumulative incidence and their relationships.
  • Describe and compare models for time-to-event data. Illustrate how the models can be applied to epidemiological or public health data.
  • Understand and recognize common pitfalls specific to time-to-event data.

 

Skills

  • Determine the proper statistical method to address a specific scientific question from a given time-to-event data set. This includes understanding the underlying assumptions of the method and identifying violations of these.
  • Perform time-to-event analysis with modern techniques using statistical software. Assess the fit of the model.
  • Interpret the results reported by statistical software. Communicate the results and conclusions of a time-to-event analysis in a clear and precise way.
  • Write a statistical report with a well-defined problem statement, method and result presentations, discussion and conclusion.
  • Critically read and review reports or articles addressing epidemiological and public health questions by time-to-event analysis.