About

The problem

Statistical flaws impede clinical science

"We currently have a reproducibility crisis, and the credibility of many scientific claims is being questioned."[Goodman, Science 2016] At the root of this problem are flawed study designs, improper understanding and use of statistical tests. Moreover, an unhealthy focus on p-values and the 'magical cut-off value 0.05' has driven a bright-line thinking in which a statistically significant result is used to defend claims of clinical significance. Even the American Statistics Association was forced to issue a statement on how to properly use p-values in clinical studies. [Wasserstein & Lazar, The American Statistician 2016]

The solution

Educational tools are the answer

I love working with clinical researchers and help them excell. GCR Statistics provides online education and research tools that will help you grow a thorough understanding of how to properly design a study, including sample size calculations, perform correct statistical analyses and understand their limitations. With GCR statistics you can focus your efforts on what really matters.

Andor van den Hoven, MD PhD MSc

Founder of GCR Statistics

Hi, my name is Andor van den Hoven. I am a practicing medical doctor, epidemiologist and clinical researcher from the Netherlands. In 2012, I graduated as a medical doctor at the University of Utrecht. In the subsequent four years I conducted full-time PhD research on how to improve a novel minimally-invasive image-guided liver cancer treatment called intra-arterial radioembolization. During the same period, I learned how to properly design and analyze clinical studies during a second master’s study in clinical epidemiology and medical statistics (graduated cum laude). At present, I work as a Nuclear Radiologist at the Sint Antonius Hospital in Nieuwegein. I have learned that many medical doctors struggle with study methodology and statistics, sometimes even unknowingly. Clinical research seems to have shifted from a ‘bed-side pattern evaluation’ practice towards a ‘p-value based numerical association’ practice. Unfortunately, many researchers are unaware of bias in their study design and ignore the assumptions underlying statistical tests. They draw overstated conclusions based on insufficient evidence. As a result, studies cannot be replicated, enormous efforts are waisted, and clinical practice fails to be improved. I like to help colleagues improve their clinical research through the GCR Statistics platform.