EPIET/EUPHEM Multivariable analysis moduleArchived
The training course provided EPIET and EUPHEN fellows with basic definitions and concepts for the different types of regression models, including an approach to survival analysis, and with the skills needed to perform a multivariable analysis using both conditional and non-conditional logistic regression.
The objectives of the training course were to:
- Understand the principles of multivariable analysis and its role in field epidemiology
- Distinguish different types of multivariable analysis (linear, logistic, poisson...)
- Understand in which analysis situations it is adequate to use which multivariable analysis
- Interpret correctly the results of the different regression models: the meaning of the parameters, and the corresponding inferences.
- Deal with confounding and interactions in a logistic regression analysis, interpret correctly the coefficients of each term.
- Understand how to build an optimal regression model
- Understand the need, the advantages and the disadvantages of conditional logistic regression and how to use it.
- Understand the principles of Poisson Regression.
- Get familiar with the concepts of survival analysis
The training course was held at the Spanish National School of Public Health in Madrid. Lectures and exercises were given in a plenary room and included discussions on the main results.