Announcing April BEMC Talk

Dear Berlin-area Epidemiological Methods Enthusiasts,

You are invited to our next BEMC Talk on Wednesday, April 3rd.

BEMC Talk: Wednesday, April 3rd, 2019 @ 4pm

“One Size Does Not Fit All: Teaching Introductory Epidemiology” – Pamela Rist, Boston, USA

  • Please register online: https://goo.gl/forms/pbE7UfFAEX2V1oxG2
  • Description: “As epidemiological methods have advanced, there has been debate over whether introductory epidemiology courses should be updated to include more “modern” concepts or if the material covered in introductory courses should remain constant.  Additionally, introductory courses greatly differ in length and scope, and since they are offered to students with diverse backgrounds and career goals, it is unlikely that one static course will fit all students’ needs. Sometimes, ‘teaching’ epidemiology happens more informally outside the classroom, such as in small consultation sessions or even when coauthoring a manuscript. In this lecture, we will discuss important considerations when developing introductory epidemiology course material, examples of ways to incorporate modern epidemiology concepts, and strategies to tailor the instruction to fit your audience’s needs.”
  • Location: Neurology Seminar Room, Charite Campus Mitte, Bonhoefferweg 3, 1. Etage (look for our BEMC signs)

Upcoming Berlin Epi Events:

  • April 17th – BEMC JClub – Paper will be posted online
  • May 8th – BEMC Talk – “Machine Learning for Population-Based Health Studies” – Christoph Lippert, Potsdam
  • May 15th – BEMC JClub – Paper will be posted online
  • June 5th – BEMC Talk – “Cool applications in R for epidemiologists” – Jochen Kruppa, Berlin
  • June 19th – BEMC JClub – Paper will be posted online
  • July 3rd – BEMC Talk – Tentative title: “Pragmatic trials and lessons from venous thrombosis” – Suzanne Cannigieter, Leiden

Interested in other Institute of Public Health events? Visit our calendar to check out upcoming conferences & short courses!

Follow BEMC on Twitter and leave questions for our speakers: @BEMColloquium

Maarten van Smeden, Regression shrinkage: better answers to causal questions

Dear BEMCers,

We are pleased to be able to share a summary prepared by a student earning credit for participating in the BEMC. A warm thank you to Ana Sofia Oliveira Gonḉalves for letting us share her summary!

Maarten’s slides can be found online: https://www.slideshare.net/MaartenvanSmeden/regression-shrinkage-better-answers-to-causal-questions

As a reminder, we will be having BEMC JClub on Wednesday this week. A link to the article is available under JCLUB.

On Wednesday March 6th, Maarten van Smeden, a senior researcher from the Leiden University Medical Center (NL) shared with the audience valuable insights on coefficient shrinkage in regression, both in a prediction and in a causal research context, with a focus on the latter. The starting point of this lecture was to question the appropriateness of traditional ways of computing the odds ratios (e.g. in a 2×2 table or by standard logistic regressions based on maximum likelihood estimation). Maarten explained that the maximum likelihood estimators for regression coefficients in generalized linear models are biased but consistent.

Throughout the lecture, Maarten used simple logistic regression model as an example. Based on such a model, he provided us with graphical representations of his simulations to show the properties of such estimators. With this simulation he intended to stress the difference between the two concepts of lack of bias and consistency. He then presented us with a solution for the reduction of bias for maximum likelihood estimators: Firth’s correction.

Firth’s correction is a penalized estimation procedure that shrinks regression coefficients, thereby removing a large part of the finite sample bias. The Firth’s correction can be readily implemented in a causal research context and packages in statistical programs already exist. Maarten mentioned other shrinkage estimators, such as Ridge or LASSO can conducted for prediction purposes, since biased coefficients are better suited for this purpose. Nevertheless, he warned the audience against its use in a causal inference context, since these approaches are designed to create bias in coefficient estimators, rather than to remove it.