New Journal Club: starts Weds. Feb 21st

Dear Epidemiological Methods Enthusiasts,

Join us for our first BEMC journal club session next week Wednesday, Feb 21st at 4pm.

This month, we’ll be reading: Chang VW, Langa KM, Weir D, Iwashyna TJ. The obesity paradox and incident cardiovascular disease: A population-based study. PLoS One 2017; 12: e0188636. http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0188636

Location: Seminar room of the Neurology Clinic, first floor (Alte Nervenklinik) – follow our signs, Bonhoefferweg 3, Charité Universitätsmedizin Berlin- Campus Mitte, 10117 Berlin à See: Campus Map

No need to register! To encourage fruitful discussion, we ask everyone to contribute one or more questions/remarks about the paper here: BEMC JClub Contribution Form. These remarks will be compiled and brought to the journal club (hard copy) to help facilitate discussion.

Please feel free to forward this invitation. Hope to see you there!

February BEMC Talk: Rasch Modeling, QA for categorical data

Dear BEMC Community,

We’ve got another great BEMC Talk coming up for you (including a bonus hands-on workshop!)

“How to calibrate a questionnaire: quality-assuring categorical data with psychometric measurement theory” – Prof. Leslie Pendrill (RI:SE Metrology, Gothenburg)

Date: Wednesday, February 7th, 2018 /   Time: 4:00pm

Optional hands-on workshop to follow at 5:15pm, important: bring your laptops with MINISTEPS installed (free!) http://www.winsteps.com/ministep.htm

Space for both events is limited, so please register here in advance: https://goo.gl/forms/ukf6IavdmH04deOs2

We need your help! Please complete these two very short (1 min/each) questionnaires ahead of time so we have some data! Go to www.menti.com

->survey 1 code: 383775 , survey 2 code: 277042

Location: Seminar room of the Neurology Clinic, first floor (Alte Nervenklinik) – follow our signs, Bonhoefferweg 3, Charité Universitätsmedizin Berlin- Campus Mitte, 10117 Berlin–> see Campus Map

Description:

Measurements in the social sciences – with ‘instruments’ such as questionnaires, ability tests, – in education, healthcare and so on, need metrological quality assurance. A patient, for instance, expects the same quality of care wherever and whenever provided. This is a challenge since the usual tools of statistics do not always work on the categorical scales typical of such measurements. Modelling a measurement system where the instrument is a human being, and where the output is a performance metric, i.e., how well the set-up performs an assessment, appears to be a way forward. This BEMC Colloquium will present the necessary tools, such as psychometric Rasch measurement theory, and will be followed by a hands-on workshop where you yourselves can analyse cases such as (i) the Quality of the BEMC Colloquium Series and (ii) a Healthy Lifestyle.

Please share this invite with colleagues. Hope to see you there!


Coming up: we are trying something new – a BEMC journal club. Every third Wednesday of the month, we’ll read a paper together. Want to know more? Check out our website for the full schedule of JClub and Talk events and more info. 

Follow us on Twitter: @bemcolloquium

Invitation: January BEMC Talk

“Prior Beliefs, Posterior Distributions and Frequencies – Basic Concepts of Bayesian and Frequentist Statistics” —Speaker: Dr. Ulrike Grittner (Inaugural lecture/Antrittsvorlesung)

Date: Wednesday, January 10th, 2018 /   Time: 4:00pm followed by small reception

*Note, date and location different from usual schedule!

Space is limited, so please register here in advance: https://goo.gl/forms/25HFixnLvJoeMMFI3

Location: Lecture Hall (Hörsaal) of the Neurology Clinic, first floor (Alte Nervenklinik) – follow our signs, Bonhoefferweg 3, Charité Universitätsmedizin Berlin- Campus Mitte, 10117 Berlin

Campus Map:  https://www.charite.de/service/lageplan/plan/map/ccm_bonhoefferweg_3

Description: Bayesian statistics originates from a posthumously published essay written by the English Reverend Thomas Bayes in 1763. Bayes’ essay proposed a new strategy for making statistical inference by combining the best available knowledge at the current time point with results from new data to create new, “better” evidence. In contrast, most statistical techniques used today originate in the early 20th century and are connected to names such as Ronald A. Fisher, Jerzy Neyman, and Karl Pearson. These “frequentist” approaches do not incorporate prior knowledge (prior beliefs) in the inference statistic, but rather make inferences solely using the data at hand. After several decades of heated discussion and debate in the statistical community, the frequentist concepts were long-time favorites, as they were perceived to be more objective. However, in recent times, technological advances have provided new solutions to the challenging numerical problems of Bayesian models, allowing appropriate weights to be given to prior knowledge and the Bayesian approach a chance to catch up. This lecture will provide insights into relevant concepts and applications of both frequentist and Bayesian statistics.

Please feel free to share this invitation. All are welcome. Hope to see you there!

That’s a wrap!

Thanks for the great turn out yesterday afternoon (50 people!) André Karch gave us an excellent overview of multiple applications for machine learning especially in prediction modeling and touched on some possible uses in causal research. We hope to see you again on January 10th for Ulrike Grittner’s inaugral lecture (Antrittsvorlesung): “Prior Beliefs, Posterior Distributions and Frequencies – Basic Concepts of Bayesian and Frequentist Statistics”. And hey–you can already register here!

Wishing you very happy holidays and einen guten Rutsch!

-Jess, Bob & Tobias

 

December Invite: Machine Learning in Epi

Dear BEMC Community,

We would like to invite you to the upcoming December Berlin Epidemiological Methods Colloquium:

“Machine Learning in Epidemiology”
Speaker: Dr. André Karch (Helmholtz Centre for Infection Research, Braunschweig)
Date: Wednesday, December 6th, 2017 / Time: 4:00pm – 6:00pm

Space is limited, so please register here in advance.

Location: Seminar room of the Neurology Clinic, first floor (Alte Nervenklinik)
Bonhoefferweg 3, Charité Universitätsmedizin Berlin- Campus Mitte, 10117 Berlin
Campus Map

Description: Big data, machine learning, and personalized prediction are the hot topics of biomedical research in the 21th century. But how do these abstract terms translate into everyday practice? How do they affect the way we think in epidemiology? What implications do they have for data analysis strategies and replication studies? In his talk, Dr. Karch will give a short introduction into machine learning algorithms, discuss opportunities and limitations of different concepts, and will show how machine learning can contribute to epidemiological research using examples from microbiome studies.

Please feel free to forward this invitation to anyone who might be interested.
Hope to see you there!

Best wishes,
Jessica Rohmann, Bob Siegerink, and Tobias Kurth

PS: Follow us on Twitter: @bemcolloquium

Wrap up: Nov. BEMC

IMG_20171101_162923429

Hats off to our November presenter Prof. Angela Zink, who shared some valuable wisdom gained through her experience developing and managing an enormous registry (as we learned, terminology can be confusing and this is essentially a large cohort!) of Rheumatology patients in Germany. She provided us with some real-world examples of where randomized controlled trials reach their limits and how well-conducted observational studies can be even more informative under certain circumstances. Her energy, thoughtful methodological approaches, and patience overseeing such a large project certainly left our audience feeling inspired. We hope you enjoyed the discussion following the talk as much as we did.

See you at our next meeting on Weds. December 6th. More detailed information on that to follow in the next week… As always, future meeting dates/topics can be found on our calendar page.

Best wishes, JR

Nov. 1st- Methodological challenges and opportunities in observational research: perspectives from rheumatology

Dear BEMC Community,

You’re invited to join us for the November Berlin Epidemiological Methods Colloquium:

“Methodological challenges and opportunities in observational research: perspectives from rheumatology”

—Speaker: Prof. Dr. Angela Zink

Date: Wednesday, Nov. 1st, 2017

Time: 4:00pm – 6:00pm

Space is limited, so please click here to register in advance.

Location: Seminar room of the Neurology Clinic, first floor (Alte Nervenklinik) – follow our signs

Bonhoefferweg 3, Charité Universitätsmedizin Berlin- Campus Mitte, 10117 Berlin

Campus Map:  https://www.charite.de/service/lageplan/plan/map/ccm_bonhoefferweg_3

Description: Randomized clinical trials are the gold standard for determining the efficacy and short-term safety of new therapies. Their external validity is limited by the strict selection of patients. Longitudinal pharmacoepidemiological cohorts, as they have been run in rheumatology for more than 15 years, enable us to determine the long-term outcomes in large numbers of unselected patients. The lecture will deal with the questions of which evidence on the safety and effectiveness of new therapies can be obtained from observational data and how we deal with the inherent problem of confounding by indication. Examples from the RABBIT cohort study will illustrate the problems and possible solutions.

Rabbit_logo

BEMC calendar:  https://bemcolloquium.wordpress.com/meetings-calendar

Please feel free to forward this invitation to anyone who might be interested.

Hope to see you there!

October 4th, 2017: Mendelian Randomization

Dear BEMC Community,

We would like to extend an invitation to our upcoming October Berlin Epidemiological Methods Colloquium:

“Mendelian randomization: Lessons from studying vitamin D and kidney function”

—Speaker: Dr. Alexander Teumer (Universitätsmedizin Greifswald)

Date: Wednesday, October 4th, 2017

Time: 4:00pm – 6:00pm

Space is limited, so please register here: https://goo.gl/forms/jucAm4NNGvQO79zC2

Location: Seminar room of the Neurology Clinic, first floor (Alte Nervenklinik) – follow our signs

Bonhoefferweg 3, Charité Universitätsmedizin Berlin- Campus Mitte, 10117 Berlin

Campus Map:  https://www.charite.de/service/lageplan/plan/map/ccm_bonhoefferweg_3

Description: Observational studies suggest an association between lower levels of vitamin D and impaired kidney function, although the observed effect direction varies. Confounding might at least partly explain such findings. Mendelian randomization can be used in such observational research settings to examine causal effects. During this colloquium, Dr. Teumer will explain this method using an example from his research showing how we assessed whether circulating vitamin D levels are causally associated with renal function, quantified by the estimated glomerular filtration rate and urinary albumin-to-creatinine ratio.

BEMC calendarhttps://bemcolloquium.wordpress.com/meetings-calendar

Please feel free to forward this invitation to anyone who might be interested.

Hope to see you there!

Insights into quantifying the unmeasurable…

bemcrolf3.jpg

A big thanks to Prof. Rolf Groenwold from Utrecht for his talk on unmeasured confounding. I especially enjoyed the in-depth discussion of pros and cons of popular methods to ‘handle’ this problem.

Methods critically discussed included:

  • Two-stage sampling
  • Cross-over designs
  • Negative controls
  • Instrumental variables
  • Sensitivity analysis

Rolf’s reflection on dealing with multiple sources of bias simultaneously and quantifying the impact of measurement error also got everyone thinking.

Perhaps the highlight was the following quote Rolf shared with us from Mr. Donald Rumsfeld, clearly a scholar in articulating types of unmeasured confounding 😉

quote-there-are-known-knowns-these-are-things-we-know-that-we-know-there-are-known-unknowns-donald-rumsfeld-25-42-14.jpg

   Photo credit: azquotes.com

We hope you will join us again Oct. 4th for our next meeting on Mendelian Randomization. Sign up is already available on our ‘Calendar’ page.

Hope to see you then! -JR

 

 

June 7th, 2017 meeting recap

How does one go about setting up a large cohort study? What is most important to include in the planning phase? How do you cope with problems out of your control (server problems, sample delivery issues, etc.)? How do you successfully collaborate with other centers having different interests? Thanks to Prof. Elke Schäffner, we are all a bit wiser to these questions after hearing her practical examples from the Berlin Initiative Study (BIS). And a big thank you to our members for their continued active participation.

We’re taking a summer break for July and August and look forward to seeing you at our next meeting on Weds. September 6th, 4-6pm. We’re excited to welcome Prof. Rolf Groenwold from Utrecht discussing unmeasured confounding in causal research. Keep an eye out for more information from us sometime in August!

elke1