Applied regression analysis with examples from health sciences

Target audience:

The course is aimed at those who have basic knowledge of statistics and want to learn more about statistical modeling with regression models.

Content:

The course introduces various regression models commonly used in health sciences. The presentation begins with linear regression, where the inclusion of multiple covariates, interactions, and non-linear relationships, such as through splines, is introduced. These concepts are revisited when discussing logistic regression, Poisson regression, and Cox regression (Cox proportional hazards model). Before modeling of hazards is introduced, survival data is explained and how it can be summarized through Kaplan-Meier curves and cumulative incidence curves.

Furthermore, students learn about methods for assessing model fit and strategies for building models. In addition, the difference between predictive modeling and modeling for drawing conclusions about causal relationships from the analysis is discussed.

Computer exercises with statistical software are regularly offered during the course and provide students with the opportunity to acquire practical skills in adapting the discussed regression models and in reporting and interpreting the outcome of the analysis. Special emphasis is placed on understanding what is important when documenting and reporting technical details in an analysis to make the results understandable and reproducible for others.

After the course, you will be able to:

Formulate a model and a formula describing the model based on a research question. Apply regression models to data using statistical software. Evaluate and interpret the results.

Research questions from health sciences are used in the presentation of the course's methods. However, the methods can be used for any type of research question.

Study forms:

The course is an online course, consisting of a mixture of pre-recorded materials and live online lectures, seminars, and computer exercises. The teaching takes place on digital platforms. The course also involves individual work and group work.