FRM9915 - NFIF - Introduction to and Application of Propensity Scores

Schedule, syllabus and examination date

Choose semester

Course content

The course will focus on current applications of propensity score methods for bias reduction in pharmacoepidemiology studies. The emphasis of this course will be on practical applications of propensity score methods, backed by theoretical lectures.

Learning outcome

After completing the course the student should:

  • Understand the role of the propensity score as a method for controlling confounding in observational studies
  • Be able to build a propensity score model and use the resulting propensity score in a variety of applications (matching, adjustment, stratification, weighting)
  • Perform diagnostics and balance checks on propensity score models
  • Communicate the results of a propensity score analysis in an oral presentation

Admission

All PhD students within the Norwegian PhD School in Pharmacy, open to non-NFIF PhD students or researchers, but with 75% of seats reserved for NFIF students.

PhD candidates who are admitted to other faculties at UiO or education institutions than UiO must at the same time apply for visiting status. They must submit, by the registration deadline, the same documentation as regular visiting status applicants, but the application and documentation (Official documentation of your admission to a PhD programme and confirmation from your supervisor/administration that the course/courses you wish to follow will be included as a part of the training component in your Phd programme) is to be sent by e-mail to The School of Pharmacy, halvor.aandal@farmasi.uio.no and not to the Faculty point of contact.

Applicants must be able to present original documentation on request.

The course will be offered to maximum 12 participants. The course will not be offered if there are fewer than 5 participants.

Prerequisites

Formal prerequisite knowledge

Students should have previous experience in epidemiology (equivalent to MF9230) and biostatistics (equivalent to MF9560) and should be comfortable with programming in either SAS, Stata or R. Prior experience with multivariable logistic regression is required.

Teaching

The course is centered on a four day obligatory session held in Oslo; students will be expected to have done substantial reading prior to the course meeting.

The individual program during the intensive four-day course includes:

  • 8 50-minute lectures (students expected to do 1-2 hours of preparation for each lecture)
  • 3 journal club-style discussions of current literature (each student will be responsible for presenting 1 article and will be expected to have read the other articles discussed - 1-2 hours prep for each session)
  • 3 3.5-hour practical skills data labs (lab assignments will be completed within the lab session; final presentation may require additional preparation, depending on student progress in lab)
  • 1 15-minute presentation on the students mini-project

The intensive four-day course will include: lectures by local, national and foreign invited speakers within the field of pharmacoepidemiology who are experienced in propensity score methods; journal club-style discussions of important literature on propensity score methods (1-2 students will be assigned to present each article); and practical lab sessions for skill practice and development.

Additional reading will complete work hours requirements according to the ECTS Users' Guide 2015. Reading list to be determined at a later point.

Mandatory activities:

Four-day course in Oslo which includes

  • 8 lectures
  • 3 lab sessions
  • 3 student-led discussions of assigned reading
  • 1 presentation for each candidate

Prelimnary time schedule:

A prelimnary time schedule for the course can be found here.

Examination

At the beginning of the course, students will choose from one of several research questions. These topics will be designed in advance to highlight important topics in observational epidemiology, with a particular focus on confounding, as this is the main source of bias that propensity scores aim to address. Students may apply in advance of the course to use their own data for the lab session. Students will apply knowledge from the morning lecture sessions during afternoon lab sessions, and will receive formative feedback on lab work (e.g. identifying potential confounders, fitting propensity score models, diagnostics and balance checks, etc). On the last day of the course, each student will give a 15-minute presentation on his or her mini-Project.

Examination support material

No examination support material is allowed.

Language of examination

The examination text is given in English, and you submit your response in English.

Grading scale

Grades are awarded on a pass/fail scale. Read more about the grading system.

Explanations and appeals

Resit an examination

If students are unable to give this final presentation due to illness or other valid reasons for absence, the alternative assignment is to complete a written report of their findings. The report should be approximately 5000 words in length, structured as a scientific manuscript and follow standard IMRAD layout including appropriate tables and figures, with special attention paid to the propensity score methods. This assignment will be due 2 weeks after the course ends, and will be graded pass/fail.

Re-scheduled examinations are not offered to students who withdraw during, or did not pass the original examination.

Special examination arrangements

Application form, deadline and requirements for special examination arrangements.

Facts about this course

Credits

5

Level

PhD

Teaching

Every spring

Examination

Every spring

Teaching language

English


Registration deadline

3 April 2017

Registration