Full topic list
This is a session of Topic 6: Statistics education, training and the workplace

(Tuesday 13th, 16:30-18:00)

Medical statistics



According to the editorial in the New England Journal of Medicine (Vol. 342, No.1) the application of statistics to medicine counts among eleven most important medical developments of the past millennium. And while the roots of statistical thought can be traced back for many centuries, Bradford Hill’s randomised clinical trial of streptomycin (1948) and Doll and Hill’s case-control study of smoking (1950) are generally viewed as landmarks of quantitative observational research in medicine. Statistics have since invaded the medical research literature.

There have been many critical reviews of the use of statistics in medical research, but the estimated proportion of around 50% of papers showing some deficiency does not seem to decrease with time. There is then ample space for teachers to convey knowledge to the end users, particularly those with lesser statistical background.

In this session we will focus on three areas of the use of statistics in medicine:
  • Errors and controversies in statistical modelling.
  • Use of controls in randomised double-blind controlled clinical trials.
  • Challenges to statisticians brought about by recent scientific evolutions in medicine and neighbouring professions.
The talks will be non-technical, focusing on the concepts and stressing the areas where teachers could play the most important role.


PaperTitlePresenter(s) / Author(s)
6D1Continuous variables: to categorise or to model?Willi Sauerbrei (Germany)
Patrick Royston (United Kingdom)
6D2Control in clinical trialsStephen Senn (United Kingdom)
6D3The applied statistical scientist in a high-profile academic environmentGeert Molenberghs (Belgium)