Skip to main content

Analysis of observational studies

Aim: The aim of the course is to obtain knowledge on

  • Analysis of  cohort, case-control and longitudinal studies using existing registers
  • Statistical concepts behind rates, particularly in relation to different time scales
  • Practical Poisson and Cox modelling of rates and applications
  • Logistic regression and repeated measurements analysis
  • Matching in cohort and case-control studies
  • Analysis of epidemiological studies based on health-related registers

The participants are introduced to analysis of studies with different designs and reporting of results from register-based studies. The primary focus in this course will be on analysis of cohort studies, case-control studies and longitudinal studies. Analysis and concepts in follow-up studies will include: Rates, risks and probabilities, Poisson and Cox modelling of rates and the use of these in reporting. Analysis of case-control studies will be based on logistic regression with and without matching (ordinary and conditional logistic regression). Analysis of longitudinal studies will focus on including an autocorrelation structure in the ordinary analysis and analysis using random intercepts and random slopes. The focus will be on application of the concepts through practical exercises using SAS or STATA and practicing these in working with a concrete project.

Course arrangement:
A 5-week course. Five full days in the first week with lectures, group discussions and exercises. Exercises will be with examples from Danish registers. The following 3 weeks participants work with their own project and prepare a course report.

Course credit: 6 ECTS

Teaching arrangement:
In the first week, there will be lectures, group discussions and exercises (full days). The following 3 weeks participants (individually) work with their own project and prepare a course report. Participants will each have a 1 hour consultation with course lecturers regarding their project. In the 5th week the report is handed-in. Finally, each participant presents the report at a seminar by the end of the 5th week.

Course leader: Prof Annette Kjær Ersbøll, ECREPH, National Institute of Public Health, SDU

Course lecturers: Prof Annette Kjær Ersbøll and assoc prof Lau Caspar Thygesen, National Institute of Public Health, SDU

The PhD course Biostatistics I+II or a similar course AND The PhD course Register-based Epidemiology or a similar course. Knowledge about SAS or STATA. Participants should bring their own laptop with SAS version 9.2/9.3 or STATA version 12.

Place: National Institute of Public Health, Copenhagen

Course material: To be announced

Course fee:
The course is free of charge for PhD students enrolled in Universities that have joined the "Open market agreement".
For other participants there is a course fee of DKK 6407,-

Max number of participants: 20 PhD students


Last Updated 26.02.2019