Skip to main content
. 1999 Sep 11;319(7211):670–674. doi: 10.1136/bmj.319.7211.670

Table 3.

Characteristics of intention to treat analyses

No (%) of trials (n=119)
Deviations from randomised allocation
Did not analyse as randomised 15 (13)
Stated no deviations 5 (4)
Stated analysed as randomised 34 (29)
Appeared to analyse as randomised 65 (55)
Missing outcome
Stated none missing 30 (25)
<10% missing 60 (50)
⩾10% missing 29 (24)
False inclusions
Reported as included 6 (5)
Reported as excluded 19 (16)
None reported 94 (79)