Bla ordlisten:
There is no clear definition of what a modified intention-to-treat analysis is, and descriptions of these analyses vary greatly from study to study.
Studies that report modified intention-to-treat analyses often have industry funding and authors' conflicts of interest, and might have a higher risk of bias than studies that report an intention-to-treat analysis, due to disruption of the baseline equivalence established by random allocation and reporting bias.
Because modified intention-to-treat analysis is an ambiguous term, we recommend against using it.
If you feel that this definition hasn't helped you to understand the term, click on our monkey to let us know.