Welcome to the Section for Clinical Biometrics

Clinical Biometrics is the methodology of empirical clinical research.

The section, founded in 1991, provides biostatistical support to clinical research and develops innovative methodology for biostatistical applications.
Biostatistical consultation for clinical projects is available by appointment during regular business hours.

Current lectures of the section are announced by the Wiener Biometrische Sektion.

 

New impact factors: two papers of our Institute on top!

In the new journal citation report 2019 (https://jcr.clarivate.com/JCRLandingPageAction.action), our review papers on variable selection (Heinze and Dunkler, 2017; Heinze, Dunkler and Wallisch, 2018), were the top cited impact factor relevant publications in the scientific journals Transplant International and the Biometrical Journal. While ‘Five myths about variable selection (Heinze and Dunkler, 2017), published in 2017, has accumulated 23 citations in 2018 and 2019, ‘Variable selection – a review and recommendations for the practicing statistician’ was published in 2018 and has obtained 42 citations in 2019. These citation numbers were the highest in each journal and are major drivers for the journals’ impact factors. Considering all 18,184 impact factor relevant articles published in journals of the category ‘Statistics & Probability’ in 2017 and 2018, only 15 (0,08 %) exceeded the citation count of ‘Variable selection – a review and recommendations for the practicing statistician’.

Heinze G, Dunkler D. Five myths about variable selection. Transpl Int. 2017;30(1):6-10. doi:10.1111/tri.12895

Heinze G, Wallisch C, Dunkler D. Variable selection - A review and recommendations for the practicing statistician. Biom J. 2018;60(3):431-449. doi:10.1002/bimj.201700067

 

COVID-19: Potentially flawed models may lead to inappropriate decisions

In an international collaboration, Christine Wallisch, Maria Haller and Georg Heinze from the Institute of Clinical Biometrics reviewed the evidence about prediction models used in COVID-19 for diagnosis and prognosis. They found high risk of bias in 27 out of 27 published studies. This alarming result may lead to wrong diagnosis of COVID-19 and inappropriate predictions for patients with COVID-19. The study was initiated by researchers at the Universities of Maastricht and Utrecht and involved prognosis researchers from the Netherlands, Belgium, the United Kingdom and Austria. The study was published on April 7, 2020 in BMJ. The review is a ‘living review’ and will be periodically updated.

Wynants L, van Calster B, Bonten MMJ, Collins GS, Debray TPA, de Vos M, Haller MC, Heinze G, Moons KGM, Riley RD, Schuit E, Smits LJM, Snell KIE, Steyerberg EW, Wallisch C, van Smeden M. Prediction models for diagnosis and prognosis of covid-19 infection: systematic review and critical appraisal. BMJ 2020; 369:m1328. doi:10.1136/bmj.m1328

 

 

 
 

Featured