gms | German Medical Science

62. Jahrestagung der Deutschen Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie e. V. (GMDS)

Deutsche Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie

17.09. - 21.09.2017, Oldenburg

Introduction of the STRATOS initiative and its main aims

Meeting Abstract

Search Medline for

  • Willi Sauerbrei - Institut für Medizinische Biometrie und Informatik, Universitätsklinikum Freiburg, Freiburg, Deutschland

Deutsche Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie. 62. Jahrestagung der Deutschen Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie e.V. (GMDS). Oldenburg, 17.-21.09.2017. Düsseldorf: German Medical Science GMS Publishing House; 2017. DocAbstr. 321

doi: 10.3205/17gmds191, urn:nbn:de:0183-17gmds1913

Published: August 29, 2017

© 2017 Sauerbrei.
This is an Open Access article distributed under the terms of the Creative Commons Attribution 4.0 License. See license information at



Research questions have become more complex, stimulating continuous efforts to develop new and even more complex statistical methods. Tremendous progress in methodology for analyzing clinical and epidemiological studies has been made, but has it reached researchers who analyze observational studies? Part of the underlying problem may be that even experts (whoever they are) do often not agree on potential advantages and disadvantages of competing approaches. However, many analysts are required de facto to make important modelling decisions and would be delighted to receive help from ‘state-of-the-art’ documents.

In response, the STRATOS (STRengthening Analytical Thinking for Observational Studies) initiative, a large collaboration of experts in many different areas of biostatistical research, was formed in 2013 ([1]. It aims to stimulate a systematic comparison of pros and cons of alternative methodologies, identification of the issues that remain unresolved and promising approaches for tackling them and crucially, communicating the results to researchers spanning the spectrum from methodological development to application. The ultimate objective of the STRATOS initiative is to develop guidance documents for data analysts and researchers with different levels of statistical training, skills and experience. The guidance will cover such practical issues as the awareness of potential pitfalls due to inappropriate use of ‘conventional’ methods, the choice of appropriate, validated analytical methods able to overcome specific challenges, and software to implement these advanced methods.

To demonstrate the necessity of guidance documents we illustrate various ways to investigate the effect of continuous variables on an outcome of interest. As an example we consider the effect of prognostic factors on recurrence free survival time in patients with breast cancer. At least four strategies, with variations, are used in practice. Analysts choose either

step functions (based on categorization)
assume that the functional form is linear
use fractional polynomials or
use one of the many approaches based on spline functions.

In the example the choice of the strategy has a severe influence on the model selected. All approaches for selection of the functional form are criticized, but for different reasons. Key issues will be briefly discussed.

In the example we consider a typical problem of topic group 2 (TG2) ‘Selection of variables and functional forms in multivariable analysis’. Currently STRATOS has nine TGs and ten cross-cutting panels which aim to coordinate the activities of different TGs, to share best research practices, and disseminate research tools and results across TGs. These panels aim to address ‘generic’ issues, such as simulation studies and publications policies, and to develop recommendations and coordinating the efforts of the individual TGs. A brief overview of the STRATOS initiative will be given.


Sauerbrei W, Abrahamowicz M, Altman DG, le Cessie S, Carpenter J; STRATOS initiative. STRengthening Analytical Thinking for Observational Studies: the STRATOS initiative. Statistics in Medicine. 2014;33:5413-5432.