Table 2.

Assessment of risk of bias in the studies reviewed

Study	Concealment of randomization	RCT stopped early	Patients blinded	Healthcare providers blinded	Data collectors blinded	Outcome assessors blinded	Reporting drop-out or withdrawal	Other potential bias	Selective outcome reporting	Jadad score
Lahti et al. 2013 [22]	Yes	No	Yes	Yes	Yes	No	Yes	No	No	4
Rampelli et al. 2013 [23]	Yes	No	Yes	Uncleara	Uncleara	Uncleara	Unclear	Yesc	Yesc	3
Ferrario et al. 2014 [27]	Yes	No	Yes	Uncleara	Uncleara	Uncleara	Yes	No	No	3
Bjerg et al. 2015 [19]	Yes	No	Yes	Yes	Yes	Yes	Yes	Yesb	No	5
Brahe et al. 2015 [20]	Yes	No	Yes	No	No	No	Yes	No	No	4
Hanifi et al. 2015 [21]	Yes	No	Yes	Yes	Yes	Yes	Yes	Yesb	No	5
Simon et al. 2015 [24]	Yes	No	Yes	Uncleara	Uncleara	Uncleara	Yes	No	No	4

Based on PRISMA (and Cochrane)’s tools for assessing risk of bias. The Jadad score is a three-item instrument that evaluates likelihood of bias in terms of quality of randomization, quality of blinding, and reasons for withdrawal/drop-out. It is compiled of the following quality items from the table: Concealment of randomization, Patients blinded, Healthcare providers blinded, Data collectors blinded, Outcome assessors blinded, and Reporting drop-out or withdrawal

^aDouble-blinded study but unclear whether healthcare providers, data collectors, and outcome assessors were all blinded

^bPerformed next-generation sequencing on fecal samples from a subgroup of the included individuals. Bjerg et al. selected 10 (of 32) placebo-treated and 10 (of 32) probiotic-treated participants with the highest triacylglycerol concentration in the blood at week 0. Hanifi et al. selected 20 (all) placebo-treated and 17 (of 20) probiotic-treated participants from the group with the highest dose (10 × 10⁹ colony-forming units) of the probiotic treatment

^cNo direct comparison between treatment groups was made for the age-related dysbiosis