*2.12. Risk of Bias Assessment*

The grade of bias risk was independently considered, and in duplicate by the two independent reviewers at the moment of data extraction process. According to Moher and Higgins, this revision followed the Cochrane Collaboration's two-part tool for assessing risk of bias and PRISMA statement [14,15].

Potential causes of bias were investigated:


In this way, the possible random sequence generation, the possible allocation concealment, the possibility of blinding of participants and personnel, the possible presence of having incomplete outcome data and other biases were all considered and evaluated.

This method applied by the two reviewers was valuable for giving to each study a level of bias. Then, the selected papers were classified with low, moderate, high and unclear risk.
