Jump to page content

A comprehensive assessment of changes to prespecified trial outcomes, selective reporting, and spin in clinical trials 

Clinical trials play a large role in medical decision-making. Although they are normally preregistered in clinical trial databases, they can still be prone to reporting biases, e.g., when changes to registered outcomes are not reported transparently.

Objectives

In this project, we aim to provide a thorough overview of the state of changes to prespecified outcomes, from the first trial registration to the results publication, and spin related to primary outcomes in results publications.

Methodology

We will assess publicly available registry information from ClinicalTrials.gov and the Deutsches Register Klinischer Studien (DRKS; German WHO primary trial registry), including historical records, of a sample of clinical trials conducted in Germany between 2009 and 2017 (IntoValue). Specifically, we will determine the extent to which trials show changes of outcomes in the registry entry (within-registration changes), and between the last registry entry before publication and the respective publication. We will assess how changes to the outcome are reported in the respective publication. Additionally, we will look at spin in the reporting of statistically nonsignificant primary outcomes.

Implications

The nature of these changes are important to know for the integrity of the scientific process, and have not been assessed at a larger scale until now. The results of this project will give an overview of what happens to presprecified outcomes in clinical research and how transparently theese changes are reported.

Preregistriation on OSF.

Contacts

  • Martin Haslberger, MSc

    Research fellow

  • Dipl.-Psych. Martin Holst

    Research fellow