Intention to treat / Per protocol

You must cite this article if you use its information in other circumstances. An example of citing this article is:
Ronny Gunnarsson. Intention to treat / Per protocol [in Science Network TV]. Available at: https://science-network.tv/statistics/special-situations-in-statistics/intention-to-treat-per-protocol/. Accessed September 18, 2019.
Suggested pre-reading What this web page adds
  1. Study design
This web page clarifies briefly the different options to analyse data in randomized controlled trials. Reading this page will make you understand more when reading a published randomized controlled trial.

Overview

There are a few different pathways to statistically analyse the effect size investigated in randomized trials:

  1. Per Protocol (PP) analysis
    Only patients with a complete data set and who actually received and complied with the allocated intervention/control are analyzed.
  2. Complete Case (CC) analysis
    Only patients with a complete data set is included in the analysis irrespective if they received or complied to the allocated intervention/control.
  3. Intention To Treat (ITT) analysis
    ITT means all patients randomized are included in the statistical analysis even if they did not receive the allocated treatment and they are also included even if they do not have a complete data set. In the latter case it means any missing data, if there is any, needs to be managed in some way and the most common options are: :
    a) Simple imputation methods such as “Last Observation Carried Forward (LOCF)”.
    b) Multiple imputation methods (MI).
    c) Mixed effects model analysis (MM).

ITT is the preferred analysis in all randomized trials. MI or MM are suitable to compensate for any missing data. However, MI or MM might be unsuitable if the available data are of poor quality . CC is likely a better choice if data is suspected of not being missing purely at random .

Per Protocol analysis (PP)

(This section is still under construction. Sorry for the inconvenience.)

Complete Case analysis (CC)

(This section is still under construction. Sorry for the inconvenience.)

Intention To Treat (ITT) analysis

(This section is still under construction. Sorry for the inconvenience.)

References

1.
Desai M, Esserman D, Gammon M, Terry M. The use of complete-case and multiple imputation-based analyses in molecular epidemiology studies that assess interaction effects. BMC: Epidemiologic Perspectives & Innovations [Internet]. 2011;8(5). Available from: https://epi-perspectives.biomedcentral.com/articles/10.1186/1742-5573-8-5
1.
Ranstam J, Turkiewicz A, Boonen S, Van Meirhaeghe J, Bastian L, Wardlaw D. Alternative analyses for handling incomplete follow-up in the intention-to-treat analysis: the randomized controlled trial of balloon kyphoplasty versus non-surgical care for vertebral compression fracture (FREE). BMC Medical Research Methodology [Internet]. 2012 Mar 24 [cited 2019 Jul 23];12(1):35. Available from: https://doi.org/10.1186/1471-2288-12-35
You must cite this article if you use its information in other circumstances. An example of citing this article is:
Ronny Gunnarsson. Intention to treat / Per protocol [in Science Network TV]. Available at: https://science-network.tv/statistics/special-situations-in-statistics/intention-to-treat-per-protocol/. Accessed September 18, 2019.

Comments are closed.