Table 5.

Risk of bias assessment for reviewed studies.

Study	Confounding biasa	Selection biasb	Misclassification bias	Performance bias	Attrition bias	Detection bias	Reporting bias
Bazzano et al. (2015)	Serious	Low	Moderate	Unclear	Low	Serious	Moderate
Bögels et al. (2008)	Serious	Low	Moderate	Low	Low	Serious	Moderate
Bögels et al. (2014)	Serious	Low	Moderate	Low	Low	Serious	Moderate
Corthorn (2018)	Serious	Low	Unclear	Unclear	Moderate	Serious	Moderate
Chan and Neece (2018)#	–	Low	Unclear	Low	Low	Serious	Moderate
Chaplin et al. (2018)#	–	Unclear	Unclear	Unclear	Low	Serious	Moderate
De Bruin et al. (2015)	Serious	Low	Unclear	Low	Low	Serious	Moderate
Eames et al. (2015)	Serious	Low	Low	Unclear	Serious	Serious	Moderate
Ferraioli and Harris (2013)#	-	Unclear	Unclear	Low	Moderate	Serious	Moderate
Haydicky et al. (2015)	Serious	Low	Moderate	Low	Moderate	Serious	Moderate
Jones et al. (2018)	Serious	Low	Unclear	Unclear	Moderate	Serious	Moderate
Lewallen and Neece (2015)	Serious	Low	Unclear	Unclear	Moderate	Moderate	Moderate
Lo et al. (2017a)#	–	Unclear	Unclear	Low	Low	Serious	Moderate
Lo et al. (2017b)#	–	Low	Unclear	Low	Low	Moderate	Low
Maloney and Altmaier (2007)	Serious	Low	Unclear	Unclear	Unclear	Serious	Critical
Mann et al. (2016)#	–	Low	Moderate	Low	Moderate	Serious	Low
Meppelink et al. (2016)	Serious	Low	Unclear	Unclear	Moderate	Serious	Moderate
Neece (2014)#	–	Low	Unclear	Low	Low	Serious	Moderate
Potharst et al. (2017)	Serious	Low	Unclear	Low	Moderate	Serious	Moderate
Potharst et al. (2018)	Serious	Low	Moderate	Low	Moderate	Serious	Moderate
Racey et al. (2017)	Serious	Low	Moderate	Moderate	Critical	Critical	Moderate
Ridderinkhof et al. (2017)	Serious	Low	Unclear	Moderate	Moderate	Serious	Moderate
Short et al. (2017)	Serious	Low	Unclear	Moderate	Low	Serious	Moderate
van de Weijer-Bergsma et al. (2012)	Serious	Low	Unclear	Low	Moderate	Moderate	Serious
van der Oord et al. (2012)	Serious	Low	Low	Low	Low	Serious	Moderate
Voos (2017)	Serious	Low	Moderate	Unclear	Moderate	Serious	Moderate
Xu (2017)	Serious	Low	Unclear	Unclear	Serious	Serious	Moderate
Zhang et al. (2017)	Serious	Low	Unclear	Moderate	Low	Serious	Moderate

^#RCT. For all RCTs in this table, the terms used to describe the level of bias have been changed from “Low,” “High,” and “Unclear” (used in the RoB tool), to “Low,” “Moderate,” “Serious,” “Critical,” and “Unclear,” to reflect the terms and judgment guidelines used in ROBINS-I;

^anot relevant for RCTs;

^bFor RCTs, the assessment of selection bias asks (1) whether there was random sequence generation and (2) whether there was allocation concealment. In this table, only one risk assessment is reported for RCTs under this bias domain, as the level of risk assessed for these two aspects of selection bias was equal for each of the reviewed RCTs.