Analysis and reporting

Participation changes in clinical trials can pose challenges for statisticians and others doing trial analyses. It’s important that trial analyses are done in ways that maximise the chance of getting reliable trial results, even when there have been participation changes. Reports about trials should be clear about what participation changes have happened, and how they were managed.

Use the links in the menu to find out more about:

Click ‘Read More’, below, to see a summary of the recommendations in this section.

Read More

Study analysis:

  • Study analysis should be conducted in ways that give the best chance that the study will still have reliable results despite any participation changes that have occurred. This is facilitated by careful statistical planning and collection of useful data. Any deviations from the analysis plan should be declared in the final study report.

End of study reporting:

  • We recommend approaches to ensure clear reporting of information about participation changes in addition to the minimum standard defined by the CONSORT statement.
  • When reporting the ‘participant flow,’ use clear terminology to describe the participation changes.
  • It is unlikely to be necessary to report every detail of how participation has changed, but enough detail should be reported to allow correct and complete understanding and interpretation of the study results.
  • At the least, it is likely to be useful to separate the major categories of participation change.
  • Modifications to the CONSORT flow diagram could help distinguish between participation changes that led to no further study follow-up from those where the participant continued to provide data for the study.
  • Consider how to ensure that reported information on ‘reasons’ for ‘losses and exclusions after randomisation’ can be as useful as possible.
  • In some cases, information about the timing of participation changes might be important to include, for example if many participants in one treatment group stopped or changed their participation soon after randomisation.
  • We refer to best practice recommendations elsewhere about reporting on ‘missing data’ in studies, and reporting analysis methods.