Introduction to Quantitative Bias Analysis

This course has been discontinued

Information is provided for reference purposes only.

Data analyses usually make assumptions (which may be explicit or, more commonly, implicit): for example, “no unmeasured confounding”. When assumptions are untestable their potential importance can only be addressed through a quantitative bias analysis (also known as a sensitivity analysis). This course will introduce you to quantitative bias analysis methods that have been developed to account for unmeasured sources of bias due to confounding, non-random selection into a study, and measurement error/misclassification.

Course profile

Please click on the sections below for more information. 

This 2-day course will be online and consist of live and pre-recorded lectures, live summary plus question & answer sessions, and live computer practical exercises. Each pre-recorded lecture will be followed by a live session which will summarise the main points of the lecture and allow participants to ask questions about the contents of the lecture.

The course is timetabled to start at 09:30 and end at 16:30, including time for coffee breaks and lunch.

By the end of the course participants will be able to understand the concept of conducting a quantitative bias analysis to account for:

  1. unmeasured confounding;
  2. non-random selection into a study;
  3. measurement error/misclassification; and
  4. multiple sources of bias (i.e., multi-bias analysis).

It is assumed that participants will have attended the Causal Inference in Epidemiology: Concepts and Methods short course (or an equivalent course). Participants should be familiar with the concept of unmeasured confounding, selection bias, and information bias; standard regression methods for dichotomous and continuous outcomes beyond the basic introductory level.

The course will include:

  1. a brief revision of unmeasured sources of bias due to unmeasured confounding, non-random selection into a study, and measurement error/misclassification;
  2. introduction to the principles of a quantitative bias analysis;
  3. guidance on conducting a quantitative bias analysis to unmeasured confounding, selection bias, and information bias;
  4. information on available software implementations of quantitative bias analysis methods; and
  5. guidance on conducting a multi-bias analysis.