Table 2.

Quality assessment of non-randomized studies (ROBINS-I Checklist)

	Kenny, 2001	Parry, 2009	Ryan, 2010
Was the allocation of the intervention to the patients randomized?	?*	+	+
The person who includes patients should not be aware of the randomization sequence. Was that the case here?	?	+	?
Were the patients and the practitioners blinded for the treatment?	-	+	+
Were the effect assessors blinded for treatment?	?	?	?
Were the groups comparable at the beginning of the trial? If not: has this been corrected in the analyzes?	+	+†	+
Is a complete follow-up available from a sufficient proportion of all participants? If not: is selective loss-to-follow-up sufficiently excluded?	+	−‡	+
Have all the included patients been analyzed in the group in which they were randomized?	+	+	-
Have the groups been treated equally, apart from the intervention?	?	+	?
Is selective publication of results sufficiently excluded?	?	?	?
Is unwanted influence of sponsors sufficiently excluded?	?	+	+

* block randomization, in blocks of eight; † Crossover study; ‡ >20% dropout